From dalke at bioreason.com  Tue Jun 22 21:19:29 1999
From: dalke at bioreason.com (Andrew Dalke)
Date: Tue, 22 Jun 1999 19:19:29 -0600
Subject: It's just magic, and I hate it ...
References: <slrn7mss5c.f90.mgm@unpkhswm04.bscc.bls.com> <14190.35530.772547.166720@buffalo.fnal.gov> <slrn7mvlhr.4is.mgm@unpkhswm04.bscc.bls.com>
Message-ID: <37703621.B29B72CB@bioreason.com>

Mitchell Morris <mmorris at mindspring.com> asked:
> Are you [Charles G Waldman] saying that this worked for you? Or
> was this a guess?

I'm going to guess that that was a guess because I was
guessing the same thing... I guess.

What you described is a classic indication of buffered output.
There are three places I can think of where your output could
be buffered:
  1) in python
 stdout can be buffered, unbuffered, or line buffered, depending
on the situation, eg, if output is a tty or a pipe.  That's why
you would want to start python with -u since running from the
command line is running from a tty while the program talks to
the web server using a pipe.  To be on the safe side, try calling
flush().

  2) in the web server.  From the Apache FAQ
http://www.apache.org/docs/misc/FAQ.html#nph-scripts
 that's likely not the case because versions 1.3 and later don't
do an intermediate buffer, but you didn't say what web server
you are using nor the version number.  That documentation may
describe what to do.

  3) in the client.  I don't recall if you have to give a
special header to specify that the data should be read and parsed
straight up.  I've never done this so I don't know what to do
next.

						Andrew
						dalke at acm.org



From jeff at ollie.clive.ia.us  Wed Jun 16 10:21:11 1999
From: jeff at ollie.clive.ia.us (Jeffrey C. Ollie)
Date: Wed, 16 Jun 1999 09:21:11 -0500
Subject: Python and Netware
References: <3764D7EE.4C87@sightline.nl>
Message-ID: <slrn7mfcmv.su1.jeff@worf.netins.net>

-----BEGIN PGP SIGNED MESSAGE-----

On 14 Jun 1999 10:34:30 GMT, Marc R. Poppenk <m.r.poppenk at sightline.nl> wrote:
>
>I'm looking for some one who can give me the answers to these questions:
>- Is a Python version for Novell Netware available?

By "a Python version for Novell Netware" do you mean a version of
Python that will run on the NetWare server itself or do you just mean
a version of Python that runs on a workstation running Windows and can
access information stored in NDS?

In either case, using JPython to access the Novell Java interfaces is
likely to be the closest you'll get right now.  NetWare 5 includes an
implementation of Java that runs on the server, so JPython should
theoretically run there since JPython is 100% Pure Java, although I
don't know of anyone that has tried it.

However the Novell Java interfaces are rather baroque and I found them
difficult to use, although I suppose that with JPython you could create
some Python wrapper classes.

- - Can it be downloaded somewhere?

You can find JPython at <http://www.jpython.org/>

You can find information on Novell's Java interfaces at
<http://developer.novell.com>.

>- Do I have to port Python for Netware myself?

I would love to see a module for Python that can access information
stored in NDS without using JPython to go through the Novell Java
interfaces.

Python running natively on NetWare would be cool too, but using JPython
with a better interface to NDS would be acceptable to me.

I don't have the time or experience to tackle these projects myself,
but I'd be more than willing to participate in a group project.

>- Does any one have experience with this?

I have no experience whatsoever with this, I'm just speculating.


[A copy of the headers and the PGP signature follow.]

Date: Wed, 16 Jun 1999 09:21:11 -0500
Newsgroups: comp.lang.python
Organization: netINS, Des Moines, IA, USA
References: <3764D7EE.4C87 at sightline.nl>
Subject: Re: Python and Netware

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2
Comment: AnySign 1.4 - A Python tool for PGP signing e-mail and news.

iQCVAwUBN2ey25wkOQz8sbZFAQGf2wP8Cwy9adBsUL+y1j9VqkYLHs9PD9+9EiqF
SZAqbR5cxnFIdLlCIvz0gR5hEBq/0UO0F0fsvH6Y6sTBt8NRrP1/6nyDShFbXLL8
HhR9M3eGFH4ffMc14pVjEQHuCNLBbyp3dabAdUouo/ds1cDNuWMgVVMLuKnekLAr
VD2eUWVBlpA=
=iLgu
-----END PGP SIGNATURE-----
-- 
Jeffrey C. Ollie                     |            Should Work Now (TM)




From ke at gnu.franken.de  Wed Jun 23 15:23:15 1999
From: ke at gnu.franken.de (Karl Eichwalder)
Date: Wed, 23 Jun 1999 19:23:15 GMT
Subject: glint?
References: <376F48B9.133D37C6@phoenix-edv.netzservice.de>
Message-ID: <shlndavhzw.fsf@tux.gnu.franken.de>

Holger Jannsen <holger at phoenix-edv.netzservice.de> writes:

|   I'm working on an installation-tool for linux/win without any
|   graphical interface but file-controlled.

How do you define "installation-tool"?  There's "xrpm" that's able to
(de)install RPM packages nicely, and more.  But "xrpm" has a GUI, too.

It comes with SuSE Linux.

-- 
Karl Eichwalder




From jam at newimage.com  Thu Jun  3 09:55:14 1999
From: jam at newimage.com (jam)
Date: Thu, 3 Jun 1999 09:55:14 -0400
Subject: Python & IRC?
In-Reply-To: <7j4frp$4e9@journal.concentric.net>; from Christopher Petrilli on Wed, Jun 02, 1999 at 04:48:09PM -0700
References: <rNHvjEdhm5Pp-pn2-tyCQ80XVscVt@d185d18f8.rochester.rr.com> <7U853.1589$nn.478377@news.shore.net> <7j4frp$4e9@journal.concentric.net>
Message-ID: <19990603095514.D21168@toast.internal>

On Wed, Jun 02, 1999 at 04:48:09PM -0700, Christopher Petrilli wrote:
> 
> I'll throw in my two cents which is that at some point I hacked out a quick
> implementation of the protocol using asynchat, and probably could do so
> again, but would rather just say use it :-)  It's really a great way to
> do this, since IRC has no pipelining, there's really not a concern there,
> and async sockets are REALLY low overhead---at least on any real OS :-)
> 
> Chris
> 

ok.. how about a python and asycchat based implementation of a *server* as
well? I'd be interested in the client you mentioned.. could you mail
it to me, or send along a url?

regards,
J
-- 
|| visit gfd <http://quark.newimage.com/> 
|| psa member #293 <http://www.python.org/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>



From skip at mojam.com  Sun Jun  6 15:57:29 1999
From: skip at mojam.com (Skip Montanaro)
Date: Sun, 06 Jun 1999 19:57:29 GMT
Subject: Deleting characters from a string
References: <87n1ye3g4v.fsf@pc-hrvoje.srce.hr>
Message-ID: <375AD334.6E45A6C3@mojam.com>

> What I would really like to be able to do is write:
> 
>   host = string.translate(host, None, ' "')
> 
> Am I missing a more elegant solution?

Probably less elegant, but you can use string.replace.  You'll have to
call it twice to get the spaces and the double quotes:

    host = string.replace(string.replace(host, '"', ""), " ", "")

-- 
Skip Montanaro	| Mojam: "Uniting the World of Music"
http://www.mojam.com/
skip at mojam.com  | Musi-Cal: http://www.musi-cal.com/
518-372-5583



From alltube at netscape.net  Thu Jun 24 15:33:59 1999
From: alltube at netscape.net (alltube at netscape.net)
Date: Thu, 24 Jun 1999 19:33:59 GMT
Subject: Python & Tkinter problems...
Message-ID: <7ku177$f42$1@nnrp1.deja.com>

Hi there,

I am trying to run Tkinter (tcl/tk 8.1 I compiled and installed it..)
on Solaris 2.6.  Python version 1.5.2.  The tcl/tk seems to work
fine, I was able to run a couple of demos, the TCL_LIBRARY is also set
ok, my environment stuff:

/usr/local/lib/tcl8.1 $ls
auto.tcl      http2.0       msgcat1.0     safe.tcl
encoding      init.tcl      opt0.4        tclAppInit.c
history.tcl   ldAix         package.tcl   tclIndex
http1.0       ldAout.tcl    parray.tcl    word.tcl
/usr/local/lib/tcl8.1 $echo $TCL_LIBRARY
/usr/local/lib/tcl8.1
/usr/local/lib/tcl8.1 $

The error I get from Python is:

/usr/local/lib/tcl8.1 $python
Python 1.5.2 (#5, Jun 24 1999, 14:52:16)  [GCC 2.8.1] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import Tkinter
>>> Tkinter._test()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 1947, in _test
    root = Tk()
  File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 886, in
__init__
    self.tk = _tkinter.create(screenName, baseName, className)
TclError: Can't find a usable init.tcl in the following directories:




This probably means that Tcl wasn't installed properly.

>>>


The init.tcl is in fact in /usr/local/lib/tcl8.1 and it appears to be
ok, I have run a number of tk demo programs...

Thanks for any help...

Alex.
Toronto


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From pduffin at mailserver.hursley.ibm.com  Mon Jun 28 08:52:10 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Mon, 28 Jun 1999 13:52:10 +0100
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <37682DD3.43923243@compaq.com> <7ka2be$qja$1@nntp6.u.washington.edu>
Message-ID: <37776FFA.2D85@mailserver.hursley.ibm.com>

Donn Cave wrote:
> 
> Greg Ewing <greg.ewing at compaq.com> writes:
> ...
> | In my experience, I find that I can do with Python everything
> | that Tcl was designed for, do it more easily, do a lot
> | more besides, and have more fun in the process. I believe
> | the reason for this is rooted in some fundamental design
> | features of these languages, which I have sketched above.
> | I think that's about as close as one can get to providing
> | an objective argument as to whether one language is better
> | than another for a given purpose.
> 

Actually you find that you can do everything that YOU want to
in Python and YOU find it easier than doing it in Tcl. I do not
believe that you can do everything in Python that you can do
in Tcl (at least as regards extending the language itself).

> How about Expect as an example?  I don't mean to criticize the
> existing Python Expect implementation(s), have no idea where the
> state of the art is on that.  I just remember trying to think of
> a natural Python idiom that would replace the "expect" verb's case
> switch flow of control.
> 
> My experience is more or less the same as yours - rewrote Tcl
> software in Python and was henceforth a convert.  But I think
> Tcl is syntactically more adaptable, where Python kind of makes
> a virtue of its fixed ways.
> 

Correct. With Tcl you can create new control structures which
are indistinguishable from the built in ones, you cannot do
that in Python. Python is much more rigid in its syntax although
it does have a lot of nice hooks to allow objects to behave
in different ways.


-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From matz at netlab.co.jp  Wed Jun  2 21:19:09 1999
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 03 Jun 1999 10:19:09 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
 <87pv3iq6ss.fsf@ev.netlab.co.jp> <7iv86a$iud@news.dns.microsoft.com>
 <u3r9nwtd9q.fsf@ares.sys1.ngo.okisoft.co.jp>
 <87so8bnsd6.fsf@ev.netlab.co.jp>
 <u37lpndko6.fsf@ares.sys1.ngo.okisoft.co.jp>
 <87pv3eorg4.fsf@ev.netlab.co.jp>
 <u3aeuichff.fsf@ares.sys1.ngo.okisoft.co.jp>
Message-ID: <87n1yinks2.fsf@ev.netlab.co.jp>

Hisao Suzuki <suzuki611 at okisoft.co.jp> writes:

|In article <87pv3eorg4.fsf at ev.netlab.co.jp>,
|Yukihiro Matsumoto <matz at netlab.co.jp> wrote:
|> BTW, did I say Ruby can invoke arbitrary destructors specified by
|> arbitrary users?  It can't.
|
|Is this your mental premise in your opinion about garbage
|collection?  Then, the unpredictability might not be matter at
|all...

I was talking about real GC in general.  I was saying real GC does not
fall into unpredictable automatically.  As for Python, I agree with
your statement in <u37lpndko6.fsf at ares.sys1.ngo.okisoft.co.jp>:

|Even if Python has "real" GC someday, IMHO, the GC should be
|used only to imitate infinite resources.  For the sake of
|conceptual simplicity, the future Python should still use
|reference counting both to invoke __del__ methods and to free
|storages; the GC routine should be used as the last resort.

						matz.



From meowing at banet.net  Mon Jun 14 23:03:32 1999
From: meowing at banet.net (greg andruk)
Date: 14 Jun 1999 23:03:32 -0400
Subject: Created a Python module for ID3 MP3 tag manipulation
References: <yttaeu7z4o8.fsf@gilgamesh.cse.ucsc.edu> <c753dzv8tel.fsf@pc142.cosc.canterbury.ac.nz> <87g13vy05d.fsf@slip-32-100-55-159.ma.us.ibm.net> <yttaeu2b2w4.fsf@gilgamesh.cse.ucsc.edu>
Message-ID: <87lndmrwqj.fsf@banet.net>

Ben Gertzfield <che at debian.org> writes:

{parsing MPEG audio}

> Is there a simple way to do this? What kind of scanning does it
> involve?

Well, it's in that other P language, but here's an example:

<URL:http://members.xoom.com/meowing/programs/mpd.zip>  (About 3K)




From aahz at netcom.com  Thu Jun  3 14:14:23 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Thu, 3 Jun 1999 18:14:23 GMT
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net> <aahzFCpHw8.Gxt@netcom.com> <19990602132331.A281155@epa.gov>
Message-ID: <aahzFCrK00.128@netcom.com>

In article <19990602132331.A281155 at epa.gov>,
Randall Hopper  <aa8vb at yahoo.com> wrote:
>
>If I was still locked onto vi, it'd be a pain, and I can see that being an
>issue with some editors.  Though with emacs, there's rectangular kill
>(^Xrk) and block indent/dedent (^X^I) -- key sequences I use without even
>thinking anymore.

I am still locked into vi.  ;-)  But even that aside, it'd still be
annoying to make sure to [in|out]dent precisely the correct number of
lines.
-- 
                      --- Aahz (@netcom.com)

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

"NRE is a lot like cocaine. Feels great while you're on it, makes you
feel very special and talented, and you become highly annoying to those
who aren't on it."  -- Stef



From davecook at home.com  Fri Jun 25 23:09:41 1999
From: davecook at home.com (David M. Cook)
Date: Sat, 26 Jun 1999 03:09:41 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net>
Message-ID: <slrn7n8h6g.j7.davecook@sputnik.escnd1.sdca.home.com>

On Fri, 25 Jun 1999 18:00:22 GMT, Mikael Lyngvig <mikael at pobox.com> wrote:

> [...] Unfortunately, TkInter appears to run sluggish [...]

That and the widget selection just doesn't cut it anymore.  I'm using pygtk
these days, but it only runs on unix right now.

Dave Cook



From larsga at ifi.uio.no  Thu Jun 10 12:17:19 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 10 Jun 1999 18:17:19 +0200
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com>
Message-ID: <wk909syqr4.fsf@ifi.uio.no>

* Fernando Mato Mira
|
| I'm trying to collect a list of all the _current_ issues making tcl
| a bad language choice. 

You can find three interesting articles on this at

<URL: http://itrc.uwaterloo.ca/~papresco/tcl/>

--Lars M.



From fdrake at cnri.reston.va.us  Thu Jun 17 12:47:56 1999
From: fdrake at cnri.reston.va.us (Fred L. Drake)
Date: Thu, 17 Jun 1999 16:47:56 GMT
Subject: Python Webserver
In-Reply-To: <wkaeu08cyf.fsf@ifi.uio.no>
References: <3767A69D.50DC7FBF@serop.abb.se>
	<wkaeu08cyf.fsf@ifi.uio.no>
Message-ID: <14185.9916.984708.453488@weyr.cnri.reston.va.us>

Lars Marius Garshol writes:
 > There's a very simple one in the standard distribution (actually, more

  Thanks to Moshe Zadka, these will be documented in the next release
of the documentation.
  Thanks, Moshe!


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From avv at quasar.ipa.nw.ru  Wed Jun 30 20:03:12 1999
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Wed, 30 Jun 1999 17:03:12 -0700
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773c3a0.102825445@news.oh.verio.com> <xPPc3.535$or1.22866@typhoon-sf.snfc21.pbi.net> <xJRc3.292$Of3.67094@news.shore.net> <7laovo$c4s$1@Starbase.NeoSoft.COM> <377A24B7.AD6FAA1A@cfdrc.com> <377A9796.176DA97E@quasar.ipa.nw.ru>
Message-ID: <377AB03F.B55DF35B@quasar.ipa.nw.ru>

"Alexander V. Voinov" wrote:

> I installed it on Red Hat 6.0. I failed to recompile it without Mesa, but
> even with it, it complains as follows:
>
> [avoinov at dhcp101 tests]$ python window.py
> Traceback (innermost last):
>   File "window.py", line 1, in ?
>     from FXPy import *
>   File "/usr/lib/python1.5/site-packages/FXPy/__init__.py", line 1, in ?
>     from fox import *
>   File "/usr/lib/python1.5/site-packages/FXPy/fox.py", line 2, in ?
>     import foxc
> ImportError: /usr/lib/python1.5/site-packages/FXPy/foxcmodule.so: undefined
> symbol: glXQueryExtension
>
> It happens both with Mesa-3.0 (Nov 98) and some Mesa-3.1beta, found in
> RedHat contrib.

Sorry, it was my fault. If you have Mesa installed it compiles out of box.

Cross-platform transparency is fantastic: I have two windows on the same
screen:
1. Native win32 FXPy example
2. The same run on a Linux box and displayed via eXceed.

The only differences I see and feel are cosmetical: the icons in the left
corner and the style of the default system font (medium/bold).

Alexander





From tim_one at email.msn.com  Mon Jun 14 20:54:30 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 14 Jun 1999 20:54:30 -0400
Subject: Solution to finalisation problem [Re: fork()]
In-Reply-To: <7k3mtg$65t$1@cronkite.cc.uga.edu>
Message-ID: <000301beb6c9$a01d2960$b2a22299@tim>

[Greg Ewing]
> Indeed, which leads me to believe that the idea of giving objects
> a __del__ method is fundamentally flawed in the first place.

[Tim]
> Except it works so nice in "the usual" case (no cycle, no resurrection):
> everyone agrees on what should be done then, and it's remarkably
> surprise-free!

[Graham Matthews]
> The problem with this argument

Thought of it as an observation; unless there's some argument about what to
do in the absence of cycles that everyone has been too shy to make here
<wink>.

> Tim is that is is very difficult for the programmer to recognise when
> they are in "the usual case", since circular references are easy to
> create. Hence it is very difficult for the programmer to verify the
> correctness of their code.

"is" is too strong in both cases.  "can be" I'll buy.  In the majority of
classes I've written or seen, it's obvious that instances can never be part
of a cycle -- or, more rarely, obvious that they can be.  Does get muddier
over time with "big classes", though (e.g., the implementation of IDLE
appears to be growing more cyclic relations over time, and that's not
unusual).

> ...
> No cycles are not the problem. The problem is that finalisers are run in
> "incomplete contexts".

*A* problem I buy; not "the".

> These incomplete contexts arise because implementations of GC delete
> or partially delete collected objects before calling finalisers.
> That is a mistake.

Insert "some" after "because" & I agree.

> If you run *all* finalisers of *all* collected objects before you
> actually delete or partially delete any of those objects then all
> finalisers run in well defined contexts and there are no problems
> with finalisers.

Sure there are:  finalizers in cycles remain largely useless in practice
without a way to specify the order in which they're invoked.  Go back to the
Java part of the thread.  The language can't know whether a cycle needs to
be finalized in a particular order, or, if it does, what that order may be.
In Java this has nothing to do with partial deletion:  the spec is very
clear that the implementation must guarantee that all objects reachable from
a finalizer are wholly intact when the finalizer is run.  What it doesn't
say-- because it can't --is that all objects reachable from a finalizer will
be in a "useful" state.  That's left on the programmer's head, and when e.g.
child and parent can be finalized in either order or even simultaneously,
correct non-trivial finalizers are jokingly difficult to write.

The Scheme guardian gimmick at least attempts to address this, by never
finalizing anything by magic, leaving it up to the programmer to explicity
ask for an enumeration of the currently-finalizable objects and nuking them
or not however the programmer sees fit.  If a cycle needs special treatment,
it's up to the programmer to realize that a cycle exists and "do the right
thing" by their own lights.  The language doesn't help, but at least it
doesn't get in the way by doing a wrong thing either; it simply doesn't do
anything at all.

> That is the point of two pass collection.

Understood the first time; but if finalizers are to be invoked on cycle
members, that's just necessary, not sufficient.

a-problem-isn't-solved-until-the-users-agree-ly y'rs  - tim





From harryt at gte.net  Fri Jun  4 10:41:59 1999
From: harryt at gte.net (Harry Tanovich)
Date: Fri, 04 Jun 1999 14:41:59 GMT
Subject: pyGTK on Windows
References: <3757D453.A302E757@interar.com.ar> <037b01beadf5$8b5d1a80$f29b12c2@pythonware.com>
Message-ID: <3757E7A4.F89DE04E@gte.net>

Fredrik Lundh wrote:

> Nahuel <nahuel at interar.com.ar> wrote:
> > Hi, im about to do a cross-plataform GUI application in Linux and
> > Windoze... i must
> > choose the GUI toolkit.. my question is .. can i use pyGTK under
> > Windows???
> >
> > (I need a gui that can display a table with about 4000 lines without
> > crashing (not like Tk :))
>
> you do mean 4,000,000 lines, don't you?
>
> or did you use a seriously broken version of Tk(inter)?
>
> after all, if it's a bug in your program, changing the UI
> toolkit won't help...
>
> </F>

What grid widget is in question?




From news at dorb.com  Fri Jun 25 17:31:15 1999
From: news at dorb.com (Darrell)
Date: Fri, 25 Jun 1999 17:31:15 -0400
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <JWQc3.2620$U6.10996@newsr2.twcny.rr.com> <3773EE44.F231F1E4@otenet.gr>
Message-ID: <_ASc3.2630$U6.11110@newsr2.twcny.rr.com>

Is this more trouble than learning tkinter and or wxPython ?
I just installed and gave the jpi thing a try. It's not ready for prime time
yet.

--
--Darrell






From sjoshi at ingr.com  Thu Jun 24 09:27:07 1999
From: sjoshi at ingr.com (sjoshi at ingr.com)
Date: Thu, 24 Jun 1999 13:27:07 GMT
Subject: Question - Returning exit codes
Message-ID: <7ktbmu$5os$1@nnrp1.deja.com>

Hello All
   I'm calling my python program from a perl script using the system
command. I was wondering what would be the best way to return exit
codes from my python script so that they can be intercepted in the perl
program.

thanks
Sunit


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From tim_one at email.msn.com  Thu Jun  3 01:30:17 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 3 Jun 1999 01:30:17 -0400
Subject: Idle comments
In-Reply-To: <7j3u7h$45j$2@hub.org>
Message-ID: <000d01bead82$2a2564c0$309e2299@tim>

[Ragica, zzz at NO-MAIL.com, which means he'll probably never see this since
 the mail->news gateway is down]

> I too have posted before asking about how to change the blinding
> (to me) IDLE background colour... and have seen your previous post
> also (and watched avidly for a response to it, as my message got no
> responses... but neither did yours).
>
> So hopefully someone anwsers this mystery this time! Just to let
> you know that there are others with the same problem who want to
> use IDLE but can't figure out how to change the background colour
> to something tolerable....

Well, it's *Python* code:  you look in the source, find where it sets the
colors, and change 'em to whatever you want.  If you grab the development
IDLE off the CVS repository at python.org, you'll find that it now has a
IdlePrefs.py file containing a ColorPrefs class you can fiddle more directly
(but nothing in the development tree is ready for general use -- that's why
it's a development tree <wink>).

start-changing-one-color-and-you'll-end-changing-'em-all-ly y'rs  - tim





From graham at sloth.math.uga.edu  Tue Jun  1 11:37:05 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 1 Jun 1999 15:37:05 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7iqhkb$d7s@wiscnews.wiscnet.net> <7iu7ma$d33$1@cronkite.cc.uga.edu> <007901beac05$5c964b00$f29b12c2@pythonware.com>
Message-ID: <7j0un1$2dm$2@cronkite.cc.uga.edu>

Gang Li (gangli at msn.com) wrote:
: Python is a real programming language.  We should not turn it in to a Java
: extension. I hate someone just want to take Python to Java mass and say 'me 
: too'.  Perl, Tcl/Tk, and Smalltalk have no plan to turn themselves in to a 
: Java extension.  Why Python?  If you love Jave to death, spent all your 
: time to help sun conquer the world. Don't mess with Python.
Graham Matthews <graham at sloth.math.uga.edu> wrote:
> You are envangelising (a not uncommon response when someone remotely
> criticises Python).
Fredrik Lundh (fredrik at pythonware.com) wrote:
: read his message again.  he's saying that Python
: shouldn't be tied to Java.  sure looks like a tech-
: nical argument to me.

I suggest you read his post again Fredrik. The only "reasons" he offers
for why Python shouldn't be "tied to Java" are all hyperbole and
evangelism. No technical content whatsoever.

Graham Matthews <graham at sloth.math.uga.edu> wrote:
> Stop evangelising and start considering the technical
> issues involve (read (!) other posts for what those issues
> are).
Fredrik Lundh (fredrik at pythonware.com) wrote:
: exactly.  and since you seem to imply that you have con-
: sidered those issues, maybe you should follow up on some
: of the "non-envangelising" posts instead of miscrediting
: people who don't happen to agree with you?

I have followed up Fredrik -- cf. posts on classes of references (some
compactable, some not), colouring mark and sweep collectors, JNI, etc.
But then again I should have known better than to start such a thread.

graham

-- 
      This mountain is not visible from any inhabitable place. 
       It has no name, and since it's discovery in 1856 by a 
      British survey team, it has mysteriously resisted every
    attempt to name it. The mountain is known simply and starkly
     by it's original designation on the surveyor's map ... K2.



From mnot at pobox.com  Tue Jun 22 08:48:21 1999
From: mnot at pobox.com (Mark Nottingham)
Date: Tue, 22 Jun 1999 12:48:21 GMT
Subject: htmllib: CR in CDATA
References: <00d201bebc5a$adb129a0$0301a8c0@mnot.net> <012601bebc91$f3ded760$0301a8c0@mnot.net> <014401bebc9a$1963a9e0$0301a8c0@mnot.net> <000b01bebca0$8a022c20$f29b12c2@pythonware.com>
Message-ID: <017501bebcad$830a3180$0301a8c0@mnot.net>

> well, htmllib doesn't claim to be HTML 4.0 compliant...

But it does claim 2.0:
"""HTML 2.0 parser.

See the HTML 2.0 specification:
http://www.w3.org/hypertext/WWW/MarkUp/html-spec/html-spec_toc.html
"""

Fine. Now, if we have a look at
http://www.w3.org/MarkUp/html-spec/html-spec_9.html#SEC9.1
we'll see that attributes are marked as CDATA.

Unfortunately, I don't have a copy of the SGML specification, so I can't
definatively say that this is the proper way to treat CDATA; all I have is
the description in the HTML 4.0 docs, as previously referenced. So, it's
probably not a good idea to patch this in sgmllib.py (as I did). However, it
is IMHO reasonable to conclude that, since both 2.0 and 4.0 refer to SGML
for the definition of CDATA, we can apply what we know about it from one to
the other (SGML being a fairly stable spec AFAIK).

I'm certainly willing to admit that this isn't directly specified behaviour
for a 2.0 parser, but I still think it's the Right Thing. A a practical
level, I'm parsing HTML with these constructs in it; if I pass off an HREF
to httplib that has a newline in it, all sorts of bad things happen.

I've ended up calling a cleaning function each time I parse attributes in my
subclassed parser; this does the job nicely. However, IMHO this sort of
lexical processing/second guessing shouldn't be necessary by the user of a
parser.


> ...and it doesn't claim to be a "user agent", either...

*sigh*
Do we _really_ want to take a trip down this semantic rabbit warren? In HTML
2.0-land, user agent is:
A component of a distributed system that presents an interface and processes
requests on behalf of a user; for example, a www browser or a mail user
agent.

Now, htmllib certainly:
* is a component
* part of a distributed system (i.e., the Web)
* presents an interface (programmatic)
* processes requests on behalf of a user

I'm curious... if it's not a user agent in the quoted context, what is it?






From pduffin at mailserver.hursley.ibm.com  Fri Jun 11 07:59:27 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Fri, 11 Jun 1999 12:59:27 +0100
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <wk909syqr4.fsf@ifi.uio.no> <375FF969.C879B4CD@iname.com> <37601668.D9105225@mayo.edu> <3760E130.7456447@iname.com>
Message-ID: <3760FA1F.7BCD@mailserver.hursley.ibm.com>

Fernando Mato Mira wrote:
> 
> Bob Techentin wrote:
> 
> > These flame wars are more than five years old.  Tcl has had namespaces
> > and a byte-code compiler for years.
> 
> I know that. But I see some serious things there (the `call by name' issue,
> for example) that give me the impression must be still around unless
> the semantics have gone through a major change.
> 
> I would cut this thread right here before someone comes up
> with some myth. There's enough work for debunkers there.
> 
> Thanks,

Tcl does not have 'call by name' semantics.
In fact its calling semantics are remarkably like C's.

    C only supports call by value
    Tcl only supports call by value

    In C if you want to pass a "reference" you pass a pointer
    In Tcl if you want to pass a "reference" you pass a variable name

"Call by name" has some very particular characteristics the main one
being that you cannot implement "swap". I can assure you that you can
write a swap function in Tcl.

	proc swap {&a &b} {
	    upvar ${&a} a
	    upvar ${&b} b

	    set t $a
	    set a $b
	    set b $t
	}

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From matz at netlab.co.jp  Tue Jun  1 02:59:59 1999
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 01 Jun 1999 15:59:59 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
 <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>
 <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net>
Message-ID: <87hfosquc0.fsf@ev.netlab.co.jp>

Hi.

Paul Prescod <paul at prescod.net> writes:

|> >As many people have noted before, you would not be able to use
|> >destructors to free up resources like memory or file descriptors.
|> 
|> which, as I understand, means destructors can not be used with real
|> GC.  That's just not true.
|
|Is there a language with real GC where destructors are called as
|predictably as they are in Python?

I said nothing about the destructor invocation predictability.

I don't understand and really interested in why you guys require
to predict destructor invocation so much, but it's another story.

|> >From my experience with Scheme and Ruby, both with real GC, the real
|> GC is mostly very fast.  
|
|Speed wasn't my concern. I asked about references in C code. Do we use
|PyObject **s instead of PyObject *s now?

Sorry, I don't understand.  Do we have to use PyObject **s instead of
PyObject *s, if we choose real GC?

|> Do you really mean Python the object-oriented programming language
|> rarely treat complicated data structures?  
|
|No, I said that Python is optimized for basic users, not for the experts.
|The experts can figure out how to work around problems. The beginning
|users cannot. "I'm opening all of the files in a directory in a loop and
|in large directories I get out of file handle errors."

First, it's not the straight result from real GC.  For exapmle, Ruby
uses real GC, and starts GC automatically on that situation.  No
problem arise.  It's totally a implementation matter.

Second, do you think that optimization for basic users compensates the
risk of potential memory leak, by cyclic structures or missing DECREF
in extension modules?  Do you mean it's easy for experts to find
cyclic structure and cut their reference to destruct?  For me, it's
not.  I hate that.  Is this because I'm not a Python expert?

						matz.



From tim_one at email.msn.com  Sat Jun 26 13:48:15 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 26 Jun 1999 13:48:15 -0400
Subject: total idiot question: +=, .=, etc...
In-Reply-To: <slrn7n8qiv.kl.behrends@allegro.cse.msu.edu>
Message-ID: <000001bebffc$12060200$da9e2299@tim>

[Tim]
> ... any OO language with declarations has to allow local vars to
> shadow instance vars, else method implementations can be broken by
> trivial changes in far-removed superclasses

[Reimer Behrends]
> To the contrary; Eiffel, under no circumstances allows shadowing of
> instance variables by formal parameters or by local variables.

Let me ammend that, to "any OO language with declarations <that isn't
Eiffel> has to ..." <wink>.  Eiffel allows inherited features to be accessed
under local alias names as if the original name never existed (its "rename"
clauses), so it has *a* way to recover locally from trivial superclass name
changes.  Java (& most other languages) take a different approach.

> Yes, changing stuff in a superclass can thus break code in a subclass; but
> this can always happen (like if you add a new method that also exists in
> a subclass).  Doing a major internal change to a superclass requires at
> least rechecking subclasses, anyway.

I did say "trivial".  A purely local way of addressing this is vital:  the
subclass author may not be able to change the superclass code, or may not
even have read access to its source code; in the other direction, a
superclass author may have no idea who subclasses from them.

Eiffel and Java both "work" in this respect, albeit thanks to different
gimmicks.

> (And I prefer compilers that are too picky rather than too lenient.)

Excellent!  Me too.  So we're both strongly in favor of Python making you
explicitly identify your instance vars with a "self." prefix <wink>.

[Java shadowing]
> Can you point me to the place in the Java language specification where
> this is said?  I'm asking, because I've heard that, too, but haven't ever
> been able to find it and because the JDK happily allows to me shadow
> instance variables left and right.

I'm not sure what happened after the various forms of inner/nested classes
were introduced.  The full story in the original JLS is spread out all over
the place, most of which can be tracked down by following the index's
"hiding" entry.  Curiously, that *doesn't* point you to section 14.3.2,
which prohibits shadowing of locals by other locals or exception parameters.

It does generally allow shadowing of class and instance names, but sometimes
in the sense of allowing *potential* ambiguity provided that an ambiguous
name isn't actually referenced.

In the latter cases, Eiffel requires local renaming of ambiguous inherited
names, while Java requires explicit qualification.  You can probably guess
which I generally prefer <wink>.

>> Readability is more important to me over the long haul.

> Huh? Where have I said otherwise? I just happen to think that these two
> goals are not mutually exclusive. And there should be a better solution
> than the language ordaining a strange kind of Hungarian notation.

The C++ conventions I mentioned were a strange kind of Hungarian; "x.y" is
just the way Python *always* spells "give me the value of attribute y in x's
namespace", whether x is a module, class or instance object.  Consistent,
explicit & obvious, given Python's modeling of modules, classes and
instances *as* namespaces.

>> Say what you like, but which vars belong to the instance is screamingly
>> obvious the instant you see
>>
>>     self.x = self.y + k
>>
>> while in
>>
>>     x = y + k
>>
>> it's at best a minor puzzle and at worst a long & tedious grep project.

> I tell you what--I have this great idea for an addition to Python. You
> know that you often can't tell what type a variable is? Well, we could
> introduce the following convention:
>
> 	Require variables to be prefixed with "$" to denote scalar
> 	values--strings and integers, "@" for lists, and "%" for
> 	dictionaries.
>
> Given the precedent of having to use self to access instance variables,
> this is entirely consistent with the current language, and is going to
> eliminate all those nasty problems where you can't tell from the context
> what the type of a variable or function is. :)
>
> 	Tongue-firmly-in-cheek-ly yours, Reimer

Seriously, languages rarely start out doing things that are insane within
their own worldview.  Perl would be *much* harder to work with in the
absence of those $@%& thingies!  Array vs scalar is an ubiquitous magical
distinction in Perl that affects semantics in crucial ways, and without the
$@ decorations you'd be left with no way to determine the intent of a line;
e.g.,

    x = split;

That does wildly different things depending on whether x is a scalar or an
array, so "$x" vs "@x" is a crucial clue.  Python doesn't have that kind of
context sensitivity, and doesn't build up complex data structures out of
explicit chains of scalar references either (and again you need the $%@
thingies to dope out what a chain of Perl *dereferences* is doing).

So the decorations solve a problem in Perl that Python doesn't have, so they
make good sense in Perl but not in Python.  Python's object.attr notation
solves Python's problem:  with respect to which namespace is attr to be
resolved?  Christian Tismer once suggested that Python take that another
step, and require that e.g.

    def f(x, y):
        global sum1, sum2, total
        sum1 = x
        sum2 = y
        total = x + y

instead be written

    def f(x, y):
        global.sum1 = x
        global.sum2 = y
        global.total = x + y

I think that's a fine idea too.  I'd even go a step toward Perl and make
mapping objects use "{}" for indexing instead of "[]".  Stop me before I sin
<wink>.

not-eager-to-trade-instant-certainty-for-potential-confusion-unless-
    the-confusion-is-helpful-ly y'rs  - tim





From akuchlin at mems-exchange.org  Thu Jun 10 10:10:22 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: Thu, 10 Jun 1999 10:10:22 -0400 (EDT)
Subject: fork()
In-Reply-To: <5laeu8zmza.fsf@eric.cnri.reston.va.us>
References: <000601beb1c7$682c6220$559e2299@tim>
	<7jjghd$9il$1@cronkite.cc.uga.edu>
	<375D5826.23197692@inrialpes.fr>
	<375D6C60.FF9998DE@appliedbiometrics.com>
	<5laeu8zmza.fsf@eric.cnri.reston.va.us>
Message-ID: <14175.51022.577016.265867@amarok.cnri.reston.va.us>

Guido van Rossum writes:
>(You can create a cycle using only lists, but I bet that happens only
>for demonstration purposes, as in ``a = []; a.append(a)''.  So I can
>live with not freeing such cycles.)

      I'm doubtful of that argument; I've written classes which track
all instances that get created, and this can be done with a class
attribute containing a list, that gets mutated as instances are
created.  You reconsider this decision later in your post.

	  In this solution, I don't understand how Python would keep
track of all the dictionaries that get created, in order to determine
which ones are left unreachable, and hence can be garbage collected.
>From my memory of the Jones-Lins book, this is usually handling by
doing all the allocation yourself by slicing memory into chunks, and
then sweeping through it.  It's not obvious (to me, at least) how
you'd do this with objects created by malloc().  Perhaps lists and
dictionaries could contain prev/next pointers that are used to
maintain them in a doubly-linked list, and a linear-time traversal
would be done to find unmarked objects, but there has to be a better
way.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
Should I not have changed either the day for carrying out my scheme, or the
scheme itself -- but preferably only the day?
    -- Ovid, _The Metamorphoses_



From tismer at appliedbiometrics.com  Sun Jun 27 19:41:41 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Sun, 27 Jun 1999 23:41:41 GMT
Subject: GC In Python: YAS (Yet Another Summary)
References: <000301bebffc$184fcf60$da9e2299@tim> <3776A025.53B35103@inrialpes.fr>
Message-ID: <3776B6B5.D74616F8@appliedbiometrics.com>


Vladimir Marangozov wrote:
> 
> A not very modest plug
> 
> Tim Peters wrote:
> >
> > [Andrew Dalke, worries about Boehm collection vs Fortran]
> > ...
> > Despite all that, it would probably work fine for you.  I think it's a dead
> > issue here regardless, though, because Guido isn't going to tell anyone they
> > have to use a particular routine for memory allocation.  Python could use
> > its own malloc for its own objects (& Vladimir Marangozov has written a very
> > nice PyMalloc, available from his Starship page), but
> 
> Thanks for the compliment, but ...
> 
> > that breaks down as soon as any extension has its own view of how to get memory.
> 
> I can't let you say this (heh, it's too late ;-) because you seem to
> overlook one of the nicest features of pymalloc. If the extension module
> works in current Python + libc malloc, it will work with Python +
> pymalloc
> because of the dynamic block management. IOW, if one recompiles the core
> with pymalloc instead of the usual libc malloc, everything should work
> as before, even *without* recompiling any extension or 3rd party
> modules.

It is a great concept to catch just one aspect
(as in you case, small blocks) and leave the rest as it was.

> The concept is simple, but useful. I told you that pymalloc is tricky
> ;-)

Welcome in the club. You know what me thinks of tricky code ;-)

ciao - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
    About to publish true continuations which are very tricky. Pah!



From dkuhlman at netcom.com  Sun Jun 13 16:36:58 1999
From: dkuhlman at netcom.com (G. David Kuhlman)
Date: Sun, 13 Jun 1999 20:36:58 GMT
Subject: Module magic methods
References: <WgP83.7966$Yr2.129670@news2.rdc1.on.home.com> <m3g13wt71h.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <dkuhlmanFDA99M.6AH@netcom.com>

Michael Hudson (mwh21 at cam.ac.uk) wrote:
> 
> You know that, in general, modules aren't unloaded, except at
> interpreter shutdown time? And then the execution state is so
> unpredictable you can't really do anything?
> 
> HTH
> Michael
> 

Witticisms by Dave: "Questions on comp.lang.python are like clothes
hangers in your closet. You try to pull out one, and twenty come
tangled to it."

Maybe, in general, modules aren't unloaded.  But what if I do:

    import mymodule
    #
    # Do some stuff with mymodule
    #
    mymodule = None

Won't reference counting cause the space taken up by mymodule to be
garbage collected?  I hope so, because, where I work, this is what
I'm going to tell them to do when they want to reclaim space in
long running programs.

[Pause for thought and experimentation with Python ... ]

Oops.  Wait a minute.  What if I create an instance on of class
defined in mymodule:

    anInstance = mymodule.MyClass()
    mymodule = None
    anInstance.show()

Something must stay around so that I can call the methods defined
MyClass.

[Pause for more heavy thinking in a light-weight brain ...]

OK, then what about after all the reference counts for instances of
mymodule go to zero?

    anInstance = None

  - Dave



From greg.ewing at compaq.com  Thu Jun 24 18:25:55 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 25 Jun 1999 10:25:55 +1200
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0rr2.2fi.behrends@allegro.cse.msu.edu>  <8L_b3.1213$4s.136166@typhoon-sf.snfc21.pbi.net> <slrn7n3gi0.44d.behrends@allegro.cse.msu.edu>
Message-ID: <3772B073.B0C637F6@compaq.com>

Reimer Behrends wrote:
> 
> To remind you, pretty much every single OO language other
> than Python does not require you to add a self;

And in some of those languages, some people have
resorted to naming conventions to keep the various
scopes under control.

In the Object Pascal version of MacApp, for example,
Apple adopted the convention of prefixing the names
of all instance variables with "f" (for "field").

Mind you, Apple's Object Pascal is a somewhat
pathological case, since it's scope rules for
instance variables and local variables are,
shall we say, rather unintuitive...

Greg



From cgw at fnal.gov  Fri Jun 25 16:46:30 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Fri, 25 Jun 1999 20:46:30 GMT
Subject: faking HAVE_LARGEFILE_SUPPORT 
Message-ID: <14195.60070.164177.1422@buffalo.fnal.gov>

I am building and testing software on Linux which will eventually be
deployed onto a variety of systems, including OS's with large
filesystem support.  I need to be sure that all of the code (some of
which is C extension modules) is prepared to handle python Long
integer objects as file sizes and offsets - for instance some of the C 
code naively does a PyArg_ParseTuple(args, "i", &file_size) which
works fine as long as the file size is a plain int.

For testing purposes,  I've modified config.h after running
./configure, so that the symbol HAVE_LARGEFILE_SUPPORT is #defined.
This allows me to continue testing on Linux, and be fairly sure that
there will be no surprises when I run the code on an SGI or Solaris
machine.

Since this feature is useful,  I've started building Python this way
all the time.  It seems to run fine,  I haven't run into any trouble.
I'm wondering if there's some hidden pitfall that I haven't thought
of.




From MHammond at skippinet.com.au  Wed Jun  2 08:35:41 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Wed, 2 Jun 1999 22:35:41 +1000
Subject: need help checking for stack overflow in Win32
References: <5lk8to3su3.fsf@eric.cnri.reston.va.us> <7j1pcc$256$1@m2.c2.telstra-mm.net.au> <375502C7.39C4FD0E@appliedbiometrics.com>
Message-ID: <7j4d2i$7jf$1@m2.c2.telstra-mm.net.au>

Christian Tismer wrote in message
<375502C7.39C4FD0E at appliedbiometrics.com>...

>Nope. The trap is:
>You compiled with VC++ 6.0 which has better register spilling.
>Therefore, recusion costs only 100 bytes, not 104.
>Not calculate 104 * 10000 and you see why 1 MB stack
>lets it crash.

Cool.  Thanks.

But doesnt the fact remain that counting the recursion depth is dangerous?
Im thinking of embedded environments, where a) the stack size of the hosting
.exe can't be known and b) the stack used by the app before Python is even
called can't be determined.  Add to this callbacks in GUI and COM type
applications, and we seem to be applying more band-aids.

The idea of counting space used seems better, but still assumes a known
stack size.

>I submitted a patch program for this kind of
>problems, so I think there is no need to change
>Python 1.5.2 at all.

Huh?  Are you suggesting people continue to build from sources, then run
your script to get an "official" version?  Wouldnt it be better to get Guido
to apply these changes to the build process?

You have provided an excellent interim fix, but IMO the problem remains.

Mark.






From fredrik at pythonware.com  Mon Jun 14 06:04:46 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 14 Jun 1999 10:04:46 GMT
Subject: <language> broken
References: <1282792849-95706075@hypernet.com> <3764D7C3.718E125E@iname.com>
Message-ID: <002f01beb64d$58e7b5f0$f29b12c2@pythonware.com>

(posted and mailed)

> Don't go into a flame war. I don't know ANY language that 
> is not `broken' somehow, even my favorite ones or the ones
> that are not my favorite but that are also `good'.

as far as I can tell, you haven't even mentioned Python in any
of your posts to comp.lang.python and the python language
mailing list.  sure looks like the usual cross-posting time-wasting
usenet troll to me, but maybe someone is abusing your user
account?

...

if you really are interested in Python, why not read the news-
group and language FAQ's, and come back when you have some-
thing relevant to ask or say.  you'll find the necessary resources
at:

http://www.deja.com/getdoc.xp?AN=488749782&fmt=text
http://www.python.org/doc/FAQ.html

you'll find Python source code, documentation, and lots of
other community resources at:

http://www.python.org

why not give it a try?

...

on the other hand, if you're not interested in Python, maybe you
could stop wasting our time?

</F>




From Nathan.Clegg at p98.f112.n480.z2.fidonet.org  Tue Jun 29 21:24:51 1999
From: Nathan.Clegg at p98.f112.n480.z2.fidonet.org (Nathan Clegg)
Date: Wed, 30 Jun 1999 02:24:51 +0100
Subject: printing with print
Message-ID: <000015ba@bossar.com.pl>

From: Nathan Clegg <nathan at islanddata.com>

> Angus MacKay wrote:
>> foo = 2
>> print "hi" + foo

How about:

print "hi" + `foo`


----------------------------------
Nathan Clegg
 nathan at islanddata.com






From alex at monsoon.swcp.com  Sun Jun 13 01:51:59 1999
From: alex at monsoon.swcp.com (alex at monsoon.swcp.com)
Date: 12 Jun 1999 23:51:59 -0600
Subject: High performance Python on NT
References: <000301beb3b7$4f973810$6eba0ac8@kuarajy.infosys.com.ar>
Message-ID: <m3r9ngve9s.fsf@monsoon.swcp.com>

"Bruno Mattarollo" <bmatt at acm.org> writes:

> 	I am about to start a project where I will need to do some high
> performance Web CGI scripts on WinNT ... I have no choice ... I know
> it's not the best combination. It's not my decision ;-) I am looking
> for some advice. Is it better if I use some kind of Active Python
> Scripting Page (ASP but with Python)? Or just doing CGI with threads
> it will be fine... ?

CGI is simply not high performance. If you throw enough hardware at it,
I can be OK, but if you are getting a large amount of hits, the overhead
of starting Python interpreter (or Perl or whatever) every time a script
is head, is going to kill you.

PHP as an Apache module is a speed demon.

As was suggested already, look closely at Zope (www.zope.org) or the
components Zope is built from. It uses a daemon for the python
interpreter, thus elimiting the overhead of CGI. 

I was doing a test last night on Zope, and it easily handled several
hits per second on my old P133 linux box.

-- 
Alex Rice    |    alrice at swcp.com    |    http://www.swcp.com/~alrice
    Current Location: N. Rio Grande Bioregion, Southwestern USA




From phd at emerald.netskate.ru  Tue Jun 15 04:32:03 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Tue, 15 Jun 1999 12:32:03 +0400 (MSD)
Subject: Looking for modules?  Config files?  Ini files?  in python....
In-Reply-To: <rNHvjEdhm5Pp-pn2-WD108jh22jtO@d185d18f8.rochester.rr.com>
Message-ID: <Pine.LNX.4.04.9906151230120.30587-100000@emerald.netskate.ru>

Hello!

On Tue, 15 Jun 1999, Benjamin Schollnick wrote:
> 	I'm trying to find a simple method to deal with *.INI or other styles
> of
> configuration files....
> 
> 	I haven't seen anything, and would hate to re-invent the wheel....
> 
> 	Any suggestions?

   Goto
http://memebers.xoom.com/phd2/Software/Python/
http://www.fortunecity.com/skyscraper/unix/797/Software/Python/

   Look into "Flat ASCII Database and Config files modules".

> 				Benjamin

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From greg.ewing at compaq.com  Sun Jun 13 18:19:04 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Mon, 14 Jun 1999 10:19:04 +1200
Subject: Newbie class questions
References: <tN13wIASNEY3EwAl@puckish.demon.co.uk>
Message-ID: <37642E58.737D5A39@compaq.com>

Toby Kelsey wrote:
> 
> (1) Is there a reason why I cannot call MyClass.q()?  Why require a
> meaningless instance arg to a class-static method?

There is no such thing as a class-static method in Python.
(There used to be, but Guido noticed that we were using
them and put some code in to stop them from working :-)

There are workarounds, but most of the time it's simpler
and clearer just to use a module-level function instead.

Greg



From gmcm at hypernet.com  Thu Jun 10 09:54:22 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 10 Jun 1999 08:54:22 -0500
Subject: global vars?
In-Reply-To: <375E4332.645E6BBF@phoenix-edv.netzservice.de>
Message-ID: <1283099378-77267406@hypernet.com>

Holger Jannsen writes:

> Why do I have to declare that variable again as global in that
> class-procedre (with :	global PREDEFINED_PROGDIR)?
> If I don't do that the value of that variable outside of that 
> procedure is the old one.
> It seems that python takes a global variable inside a procedure in a
> class but not outside.

Without some code to demonstrate your confusion, it's difficult to 
guess exactly why you're feeling confused.

In both cases, changing the binding of a global variable requires a
"global" statement within the function / method. Without that 
statement, Python takes an assignment to the variable (a binding) to 
be local in scope (that is, it's a new variable that exists only 
within the function / method).

This does not apply to reading a variable - Python will search first 
the locals, then the globals. So you can modify a mutable variable 
(eg, append a string to a global list of strings) without a "global" 
statement. The binding of the global variable (the name -> object 
binding) is not changed, so the rule doesn't apply.

So in your case, the different behaviors are due to something else 
you're doing, not to the fact that one is in a function and the other 
in a method.

- Gordon



From clarence at silcom.com  Thu Jun 24 10:28:30 1999
From: clarence at silcom.com (Clarence Gardner)
Date: Thu, 24 Jun 1999 14:28:30 GMT
Subject: It's just magic, and I hate it ...
References: <slrn7mss5c.f90.mgm@unpkhswm04.bscc.bls.com> <14190.35530.772547.166720@buffalo.fnal.gov> <slrn7mvlhr.4is.mgm@unpkhswm04.bscc.bls.com> <37703621.B29B72CB@bioreason.com> <slrn7n1r6p.keg.mgm@unpkhswm04.bscc.bls.com> <slrn7n2aj0.n0c.mgm@unpkhswm04.bscc.bls.com>
Message-ID: <igrc3.658$dR.56613@newsfeed.avtel.net>

Mitchell Morris (mgm at unpkhswm04.bscc.bls.com) wrote:
: As much as I hate to admit I'm an idiot, I got caught by the simplest of
: things. When the child forks, it inherits all the open file descriptors of
: the parent ... which includes sys.stdin and sys.stdout. Since these two (at
: least) are attached to the socket from the browser, the browser doesn't
: follow the redirect until *ALL* the fds close.
: 
: My new question is how do I close all the connections to the browser after
: the fork? The obvious:
: 	sys.stdin.close()
: 	sys.stdout.close()
: 	sys.stderr.close()
: didn't get them all, since the browser continues to wait. Is there a
: mechanism to determine all opened files?

I had the same problem a while ago, which resulted in FAQ #4.87:

4.87. Why doesn't closing sys.stdout (stdin, stderr)
really close it?

Python file objects are a high-level layer of abstraction on top of C
streams, which in turn are a medium-level layer of abstraction on top
of (among other things) low-level C file descriptors. 

For most file objects f you create in Python via the builtin "open"
function, f.close() marks the Python file object as being closed from
Python's point of view, and also arranges to close the underlying C
stream. This happens automatically too, in f's destructor, when f
becomes garbage. 

But stdin, stdout and stderr are treated specially by Python, because
of the special status also given to them by C: doing 

    sys.stdout.close() # ditto for stdin and stderr

marks the Python-level file object as being closed, but does not close
the associated C stream (provided sys.stdout is still bound to its default
value, which is the stream C also calls "stdout"). 

To close the underlying C stream for one of these three, you should
first be sure that's what you really want to do (e.g., you may confuse
the heck out of extension modules trying to do I/O). If it is, use
os.close: 

    os.close(0)   # close C's stdin stream
    os.close(1)   # close C's stdout stream
    os.close(2)   # close C's stderr stream


-- 
-=-=-=-=-=-=-=-=
Clarence Gardner
AvTel Communications
Software Products and Services Division
clarence at avtel.com



From jtravs at debian.org  Mon Jun 21 02:53:00 1999
From: jtravs at debian.org (John Travers)
Date: Mon, 21 Jun 1999 07:53:00 +0100
Subject: pyOpengl for Debian
Message-ID: <99062107535200.03236@Vegus>

Has anyone made Debian/GNU Linux packages of pyOpenGl?
--
John Travers

"Sometimes I think the surest sign that intelligent life exists elsewhere
in the universe is that none of it has tried to contact us!"



From gmcm at hypernet.com  Sat Jun 12 00:23:54 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 11 Jun 1999 23:23:54 -0500
Subject: Broken 1.5.2 installation? Missing cw3215mt.dll
In-Reply-To: <37611D32.47BED973@callware.com>
Message-ID: <1282960807-85602620@hypernet.com>

Ivan Van Laningham wrote:

> [snort]
> 
> I've just been installing Python/Tcl/PyWin/Win32All/blah blah on
> several Win machines (95/98/NTW/NTS) over the past several days. 
> It's a long story (soon to be a major motion picture starring Mel
> Gibson), but suffice it to say that in *every* case IDLE refused to
> run out of the box.  

> ...  I've also installed Python/Tcl on my three Linux
> machines--RH5.2 and RH6.0-SMP--but I didn't have to fiddle with
> anything at all in order to get IDLE to run just fine.  Of course,
> Python and Tcl etc. have been in my path permanently for a long time
> on Linux;-)

[double snort]

A few months ago you were running some auto-port from a PDP-8. Did
RH finally convince you no warm-blooded animals were used in user
testing?



- Gordon



From graham at sloth.math.uga.edu  Tue Jun  8 15:43:25 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 8 Jun 1999 19:43:25 GMT
Subject: fork()
References: <000601beb1c7$682c6220$559e2299@tim> <7jjghd$9il$1@cronkite.cc.uga.edu> <375D5826.23197692@inrialpes.fr>
Message-ID: <7jjrot$h16$2@cronkite.cc.uga.edu>

Graham Matthews wrote:
> But the reality is that copying memory can in fact speed applications
> up! Think about locality of reference after you have copied all live
> blocks to the start of memory.
Vladimir Marangozov (Vladimir.Marangozov at inrialpes.fr) wrote:
: True enough. At least, it sounds true and promising ;-)
: However, locality of reference (LOR) is something quite hard to measure
: in practice. One has to tweak the kernel to collect some stats on this,
: and even if one achieves LOR improvements on a partuclar system, say
: Solaris, it could result in a LOR degradation on Linux or Windows.
: Not to mention that nowadays nobody wants to play this game...

Sure I agree with all this. I am just saying that it's quite simplistic
nowdays to assume that copying objects when collecting is a bad idea.
It used to be the case that copying was bad -- that's why mark and sweep
collectors were invented before copying collectors. But on modern CPUs
and modern OS's it's no longer uniformly true.

Vladimir Marangozov (Vladimir.Marangozov at inrialpes.fr) wrote:
: Graham, you said that you won't contribute a RC+GC scheme to Python, despite
: your positive experience. If you change your mind and consider
: devoting some spare time to give it a try, I'll devote some spare
: time to help you with Python's internals and we'll see whether we
: could come up with something viable, which could compete with the
: actual RC scheme alone. Does this sound constructive enough? :-)

Very constructive indeed! I am indeed sorry that I don't have the time
to contribute code for a collector for Python. I think it would be greatly
beneficial. But between working on a language of my own, and working on
my PhD, well there is only a finite amount of time in a week (damn shame
that!). 

I was actually wondering if anyone had done any work putting the Boehm
collector under Python. The Boehm collector was designed to run in a
C environment.

graham
-- 
           As you grow up and leave the playground
       where you kissed your prince and found your frog
          Remember the jester that showed you tears
                   the script for tears



From greg.ewing at compaq.com  Sun Jun 27 23:43:33 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Mon, 28 Jun 1999 15:43:33 +1200
Subject: Python for large projects?
References: <3774b5d1.690922@news.force9.net>  <T6bd3.2883$U6.12014@newsr2.twcny.rr.com> <slrn7naup9.jk.thantos@brimstone.mecha> <RGid3.2957$U6.12374@newsr2.twcny.rr.com> <87n1xm85mf.fsf@lisboa.i-did-not-set--mail-host-address--so-shoot-me> <Jhsd3.3090$U6.12410@newsr2.twcny.rr.com>
Message-ID: <3776EF65.A07CBEFC@compaq.com>

Darrell wrote:
> 
> I'll give that a look, but they can always get to it if they want. Maybe
> that's ok, just worries me.

If you clearly warn them in the docs not to do
it, and they do it anyway and get into trouble,
it's not your fault.

Greg



From Vladimir.Marangozov at inrialpes.fr  Sun Jun 27 18:05:25 1999
From: Vladimir.Marangozov at inrialpes.fr (Vladimir Marangozov)
Date: Mon, 28 Jun 1999 00:05:25 +0200
Subject: GC In Python: YAS (Yet Another Summary)
References: <000301bebffc$184fcf60$da9e2299@tim>
Message-ID: <3776A025.53B35103@inrialpes.fr>

A not very modest plug

Tim Peters wrote:
> 
> [Andrew Dalke, worries about Boehm collection vs Fortran]
> ...
> Despite all that, it would probably work fine for you.  I think it's a dead
> issue here regardless, though, because Guido isn't going to tell anyone they
> have to use a particular routine for memory allocation.  Python could use
> its own malloc for its own objects (& Vladimir Marangozov has written a very
> nice PyMalloc, available from his Starship page), but

Thanks for the compliment, but ...

> that breaks down as soon as any extension has its own view of how to get memory.

I can't let you say this (heh, it's too late ;-) because you seem to
overlook one of the nicest features of pymalloc. If the extension module
works in current Python + libc malloc, it will work with Python +
pymalloc
because of the dynamic block management. IOW, if one recompiles the core
with pymalloc instead of the usual libc malloc, everything should work
as before, even *without* recompiling any extension or 3rd party
modules.
The concept is simple, but useful. I told you that pymalloc is tricky
;-)

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



From gerrit at nl.linux.org  Sun Jun  6 15:12:59 1999
From: gerrit at nl.linux.org (Gerrit Holl)
Date: 6 Jun 1999 19:12:59 GMT
Subject: Suggestion: Dutch Python mailinglist
Message-ID: <slrn7lli2q.ia.gerrit@gerrit.palga.uucp>

This e-mail is in two languages. Look below for the Dutch version.
Deze e-mail is in twee talen. Kijk beneden voor de Nederlandse versie.

Hello all,
I'm a 13-years-old Dutch boy, and my English is not good. Now I have the
chance to set up a Dutch python mailinglist. I know there are many Dutch
python-users, but most of them know English very well. So will there be
any users who'll subscribe to such a mailinglist, or will it remain very quiet?

regards,
Gerrit.



Ik ben een Nederlandse jongen van 13 jaar, en mijn Engels is niet goed. Ik
heb nu de mogelijkheid om een Nederlandse Python mailinglist op te zetten. Ik
weet dat er veel Nederlandse Python-gebruikers zijn, maar de meeste van hen
kennen genoeg Engels. Zullen er gebruikers zijn die zich zouden abboneren op
zo'n mailinglist, of zal het doodstil blijven?

groeten,
Gerrit.

-- 
The Dutch Linuxgames website.		De Nederlandse Linuxgames pagina.
Everything about games on Linux.	Alles over spelletjes onder Linux.
Site address:				http://linuxgames.nl.linux.org
Personal homepage:			http://nl.linux.org/~gerrit/



From gmcm at hypernet.com  Wed Jun  2 23:27:23 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 2 Jun 1999 22:27:23 -0500
Subject: Newbie seeks help with COM? DLLs?
In-Reply-To: <7j4iiv$l1r$1@autumn.news.rcn.net>
Message-ID: <1283741805-38623148@hypernet.com>

Sam Hunting writes:

> Background: I'm operating in a VB/NT shop, but I want to introduce a
> Python solution for the sake of doing *extremely* rapid development
> of GUI widgets.

> We need to incorporate the widgets into an existing software product
> so that, for example, a Python-developed dialog will appear, into
> which the user will enter data, which is then passed back to the
> application.
> 
> Now here is where my newbie nature takes over, since I am very new
> to Python, and even newer to the VB/NT/Windows worlds -- how
> "incorporate"?
> 
> The software product understands DLLs, and also understands COM.
> I've had it suggested to me that the way to get where I need to do,
> is to turn the dialogs into COM servers, and then call them from the
> software application, which is the client.
> 
> (1) I worked through the "Quick Start to Server Side COM and
> Python", and registered the Hello World server successfully. Can
> anyone point me to some VB code that would enable me to test the
> test server?
> 
> (2) If the COM approach is not the right approach, can someone
> suggest a better one? And point me to some sample code that I can
> test, run, and improve to help me solve the problem?

You need http://www.robanal.demon.co.uk/demos/pyvb/index.html. (What 
you *really* need is Andy & Mark's book, but that'll take awhile.)

Assuming your client is VB, this seems the natural way to go. If it 
were C or C++, it might be easier through the Python's C API.

- Gordon



From arcege at shore.net  Sat Jun 19 19:34:32 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Sat, 19 Jun 1999 23:34:32 GMT
Subject: Tkinter & socket question
References: <19990618215411.23898.qmail@nwcst277.netaddress.usa.net> <sogic333t.fsf@foreman.ac.rwth-aachen.de>
Message-ID: <cOVa3.642$7X1.149581@news.shore.net>

Gregory A. Landrum <landrum at foreman.ac.rwth-aachen.de> wrote:

:>2) I need to poll the connection periodically, say, every second. This would
:>not have been a problem in console mode... I don't know how to do this in a
:>Tkinter environment though. So, how can I do something every second (or every
:>n milliseconds, or whatever the time span may be)? Are threads the way to go?
:>I haven't used them before.

: Wow, a question I can answer.... amazing.
: Obligatory-Mandatory-Necessary-Essential Disclaimer:  I bear no
: resemblance whatsoever to a python wizard, I just happen to have done
: something like what you want to do.  There's almost definitely a
: better way of doing this, but my method seems to work.

: I assume that you have already opened a nonblocking socket connection,
: and that that is what you are polling.  Your main Tkinter window has a
: method called "after" which can be used to do the polling.
: Documentation can be found at:
: http://www.pythonware.com/library/tkinter/introduction/x8550-alarm-handlers-and-other.htm

: An example, in my stupid little networked python game, I have the
: following polling function:
:     def ListenForMessages(self):
:         msg = None
:         theConn = None
:         
:  # Check to see if we are the client or the server
:         if self.serverActive == 1:
:             if self.clientConn is None:
:                 print "No Client!"
:                 return
:             theConn = self.clientConn
:         elif self.clientActive == 1:
:             theConn = self.socket

:         if theConn is None: return

:  # see if there is anything there to read
:         try:
:             msg = theConn.recv(1024)
:      # Got it!  Dispatch the message.
:             self.Dispatch(self.message.Type(msg),msg)
:         except error:
:      # nope... no one loves us.
:             pass

:  # re-register the after callback, with a 200msec delay
:         self.listenID = self.Game.rootWin.after(200,self.ListenForMessages)

: The last line is important, because the after callback is removed
: after being called.  So it needs to be re-registered with Tk.

: The callback is first registered immediately upon establishing the
: socket connection.

You can also deal directly with Tcl/Tk's filehandler:

  from Tkinter import tkinter
  self.root.tk.createfilehandler(socket, tkinter.READABLE,
    self.handle_msg
  )
  ...
  self.root.tk.deletefilehandler(socket)

The mainloop() method handles the rest for you.

  -Arcege




From fysx at spots.ab.ca  Mon Jun 14 21:54:53 1999
From: fysx at spots.ab.ca (Maxwell Sayles)
Date: Tue, 15 Jun 1999 01:54:53 GMT
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com>
	 <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com>
	 <375FDCA7.62FF@mailserver.hursley.ibm.com>
	 <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com>
	 <3762988E.C94AA252@iname.com>
	 <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu>
	 <3763C746.5CA8FE78@iname.com> <m3u2sby4gf.fsf@soma.andreas.org>
	 <87pv2zjhhl.fsf@home.ivm.de> <3764F3FE.F7CBE6BA@acm.org> <0Sgs3BA5FPZ3Ew6E@jessikat.demon.co.uk>
Message-ID: <3765B3B8.E45F373B@spots.ab.ca>


Robin Becker wrote:

> postfix, prefix, infix etc etc are isosemantic

for myself and the others who might not know... can we get an example of
each?
and i remember someone mentioned parenthesized vs non-parenthesized...

Maxwell Sayles




From jefftc at leland.Stanford.EDU  Fri Jun 25 20:59:12 1999
From: jefftc at leland.Stanford.EDU (Jeffrey Chang)
Date: Fri, 25 Jun 1999 17:59:12 -0700
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
In-Reply-To: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net>
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net>
Message-ID: <Pine.GSO.3.96.990625170056.28322B-100000@saga3.Stanford.EDU>

On Fri, 25 Jun 1999, Mikael Lyngvig wrote:
> I'd like some discussion on whether or not it is a good idea to
> abandon TkInter, before it gets too widely used, and instead adopt
> wxPython as the primary GUI framework for Python.

Abandoning Tkinter seems a little bit extreme, since there are some
applications for which its event model is well-suited. 

A few weeks ago I needed to build a viewer that displays binary trees
whose branches have variable lengths (actually, phylogenetic trees for
biological data).  Since the trees are large, I need to be able to slide
it around the window.  In addition, I needed each node to be clickable so
that the user can query for more data. 

Tkinter is very good for that type of thing.  All I needed to do was
create a Canvas and then for each node, create an Oval with a callback to
a function that handles node clicks.  Tk keeps track of the coordinates of
each Item so that I don't have to.  Also, whenever I wanted to move the
tree, all I had to do was:
for widget in self._canvas.find_all():
    self._canvas.move(widget, xoff, yoff)

Of course, Tk incurrs an overhead by making everything you draw on the
canvas its own object, but for this case, it made thigns a lot easier.

Although I haven't used wxWindows before, but the documentation suggests
that it would be a lot more difficult for this kind of application.  In
wxWindows, I would need to draw each of my nodes into a window using a
device context.  Then, I'd have to write a function that would handle
mouse clicks in the window.  That function would need to get the
coordinates and then figure out whether it intersects with any of my
nodes, which Tk would've handled for me.  In addition, whenever I wanted
to move the tree, I'd have to erase the window and redraw each node.  If I
wanted that optimized, I'd need to write routines that would only redraw
the uncovered parts of the tree, which means figuring out which nodes are
newly revealed.

>From what other people have said about wxPython, it seems like it may be
appropriate for other types of applications.  However, I'd hate to see
Tkinter dumped because it's not appropriate for everything!

Jeff




From paul at digicool.com  Wed Jun  2 20:19:50 1999
From: paul at digicool.com (Paul Everitt)
Date: Wed, 02 Jun 1999 20:19:50 -0400
Subject: Is pcgi usable without Zope?
References: <ltd7zf6wtv.fsf@asfast.com>
Message-ID: <3755CA26.BCB7828A@digicool.com>

Lloyd Zusman wrote:
> 
> I have to admit that I'm a bit confused about pcgi these days.  The
> most recent documentation for it was last updated in August, 1998, and
> it still refers to Bobo and Principia.

We're certainly not helping -- I agree that stuff is strewn around all
over the place.
 
> Now that we have Zope, is pcgi now intimately bundled with it, never
> to emerge on its own again?  Or is there a way to run pcgi without
> Zope?

I don't think PCGI is intimately bundled with Zope.  Neither is the
object publisher, object database, the template system, etc.  All can be
used independently.

--Paul



From phd at emerald.netskate.ru  Thu Jun 10 04:40:29 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Thu, 10 Jun 1999 12:40:29 +0400 (MSD)
Subject: os.mkdir()
In-Reply-To: <375E5DCE.7005D71A@phoenix-edv.netzservice.de>
Message-ID: <Pine.LNX.4.04.9906101239340.6646-100000@emerald.netskate.ru>

On Wed, 9 Jun 1999, Holger Jannsen wrote:
> Look:
> >>>os.mkdir("c:\\blabla")
> OSError: [Errno 17] File exists: 'C:\\blabla'
> 
> or
> >>>os.makedirs("c:\blabla\nana\gugu")
> produces 
> OSError: [Errno 2] No such file or directory: 'c:\blabla\nana'

os.makedirs("c:\\blabla\\nana\\gugu")
               ^^      ^^    ^^

   Double your backslashes.

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From zigron at jps.net  Fri Jun 25 14:23:32 1999
From: zigron at jps.net (Zigron)
Date: Fri, 25 Jun 1999 11:23:32 -0700
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773c3a0.102825445@news.oh.verio.com>
Message-ID: <xPPc3.535$or1.22866@typhoon-sf.snfc21.pbi.net>

Kevin Dahlhausen <morse at harborcom.net> wrote in message
news:3773c3a0.102825445 at news.oh.verio.com...
> m
> >I'd like some discussion on whether or not it is a good idea to
> >abandon TkInter, before it gets too widely used, and instead adopt
> >wxPython as the primary GUI framework for Python.
    I'm thinking this is too late, actually. :) From my understanding, its
the de-facto standard or whatever, and plenty of people use it already.
Not that that's a good thing.. I find it rather unpleasant that the de-facto
GUI for Python requires you to install a /different/ scripting language
just to run,.. and, does seem to be rather slower then wxPython from the
uses i've had for both... which doesn't surprise me, since Tkinter /does/
have to run through a second scripting-interpreter and all.

> >Having said this, I have to say that I'm in no way a TkInter or a
> >wxPython expert.  I've evaluated both of them and found wxPython to be
> >the winner - and TkInter without a real chance of getting used by me.
    Awhile back I was using Tkinter for a program i'm doing, and was so
incredibly frustrated with it, and my inability to get it to make apps that
look like I want to -- /native/ and still functional. ( I tried to make a
wxListCtrl, ugh!)... Then I ran into wxPython, and have been happy ever
since :)

> I agree with your observations on TK.  I looked at wxPython and some other
> GUIs.  I liked the Fast Light Toolkit API the best.  At least Python
allows us
> to use the best GUI, with 'best' being defined by people according to
their own
> preferences.
    True, 'tis a good thing that people can pick and choose beween GUI's,
but I
still think TK is unsavory. .. to make an app that you want to distribute,
you'd not
only have to install Python on their machine, but TCL too. Yuck. :)

> >P.S. Let's not make a war out of this; I just want to know if there's
> >anybody else who feels wxPython is performing better than TkInter.
>
> As long as we don't talk about politics,tabs, spaces, and indentation, we
should
> be fine.

    Do'nt forget garbage collection and perl in those lists of disallowed
topics. :)

--S





From erenouf at opticominc.com  Wed Jun  9 11:30:42 1999
From: erenouf at opticominc.com (Eric Renouf)
Date: Wed, 09 Jun 1999 11:30:42 -0400
Subject: getting memory information under NT
References: <002201beab54$66218d30$f29b12c2@pythonware.com> <003301beab56$88cb3af0$f29b12c2@pythonware.com>
Message-ID: <375E88A2.68C1F118@opticominc.com>

Does anyone know of an easy way to get the the total physical memory
and/or available memory on an NT system using Python 1.5.2?  Is there a
module that has functions for things like this already?

Thanks,
Eric Renouf



From zeitlin at seth.lpthe.jussieu.fr  Tue Jun 29 10:23:45 1999
From: zeitlin at seth.lpthe.jussieu.fr (Vadim Zeitlin)
Date: 29 Jun 1999 14:23:45 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <7l4nht$jl7$1@nnrp1.deja.com> <9DDBCC52A7B26002.649D772F7232D418.CE2709D4B08E7401@lp.airnews.net> <ui6747qhvz.fsf@bacon.ethz.ch>
Message-ID: <slrn7nhlor.q40.zeitlin@seth.lpthe.jussieu.fr>

On 29 Jun 1999 15:02:40 +0200, Ernst-Udo Wallenborn <wallenborn at phys.chem.ethz.ch> wrote:
>Would this be any different with a different GUI kit?

 Hi,

 doing GUI calls from more than one thread is a recipe for disaster IMHO. I
don't say that it's impossible (in fact, the threads sample in wxWindows
distribution which I (co)wrote does just this), but it's not a good idea.

 OTOH, there is no problem in having as many of worker threads as you wish and
one "main" GUI thread. This design is much more robust and easier to implement
too...

 Regards,
VZ



From amackay at starvision.com  Mon Jun 21 17:08:52 1999
From: amackay at starvision.com (Angus MacKay)
Date: Mon, 21 Jun 1999 14:08:52 -0700
Subject: chomp()
Message-ID: <376EA9E4.5B3D38EC@starvision.com>

is there an equivilent of perl's chomp() in python or
do I have to write the function myself?

cheers, Angus.



From ronERASEjeffries at ERASEacm.org  Fri Jun 18 21:27:58 1999
From: ronERASEjeffries at ERASEacm.org (Ronald E Jeffries)
Date: Sat, 19 Jun 1999 01:27:58 GMT
Subject: Extreme Python Testing
References: <375C99FE.E61E511@pdq.net>
Message-ID: <8D1E40EBACC2A6F5.29A2B8B9FA3EDD18.CD1734958179DCBB@lp.airnews.net>

On Mon, 07 Jun 1999 23:20:14 -0500, John Dell'Aquila <dellaq2 at pdq.net>
wrote:

>I admire the Smalltalk "Extreme Programming" methodology enough that
>I'm applying it in my Python development. XP is impossible without an
>automated test framework, so the first thing I did was write a Python
>version of Kent Beck's Testing Framework.

Way ta go John! Glad to see another entry in the Testing Framework
collection. Would you mind if I post it on the armaties XP site?

I'm trying to get JPython going, as soon as that happens, I'll try out
your framework.

Regards,

Ron Jeffries
Extreme Programmer
http://www.armaties.com



From alex at somewhere.round.here  Fri Jun  4 13:51:44 1999
From: alex at somewhere.round.here (Alex)
Date: 04 Jun 1999 13:51:44 -0400
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net> <aahzFCpHw8.Gxt@netcom.com> <7j4e2u$8rh$1@m2.c2.telstra-mm.net.au> <37564132.2D01B7C3@pop.vet.uu.nl> <3756BAFC.2066A636@home.com> <3756F071.E717D0FA@pop.vet.uu.nl> <37573C16.D0233810@home.com>
Message-ID: <etdr9nrx39r.fsf@m2-225-11.mit.edu>

> I'm rambling about Python these days. Yesterday I read a paper from
> Guido about how he was thinking about bringing Python (or some
> successor language) to *billions* of users. :)

Could someone point me at this paper?

Alex.



From MHammond at skippinet.com.au  Wed Jun 16 03:01:06 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Wed, 16 Jun 1999 17:01:06 +1000
Subject: do anonymous pipes normally work on NT?
References: <7juatc$fef$1@nnrp1.deja.com> <7k43sk$6vd$1@nnrp1.deja.com>
Message-ID: <7k7i1g$87d$1@m2.c2.telstra-mm.net.au>

Im afraid I can't be of much help here.  I have had a good play, and nothing
seems obvious.

Interestingly, if you change cat.py to:

    f.write("Stdin fileno is %s\n" % (sys.stdin.fileno(), ))
    f.write("Stdin handle is %s\n" %
(msvcrt.get_osfhandle(sys.stdin.fileno()), ) )

You will notice that when run "normally", stdin has a fileno of 0, and a
valid handle.  However, when run by runproc.py, the fileno is -1, and
attempting to get back the handle fails.  If I had more time I would dig
into where fileno() gets its value from, and see if this points to why we
are failing.

Adding
except:
    traceback.print_exc(file=f)

Also helps see this stuff.  I am afraid I am all out of time with this
stuff.  If you are really keen, the next step would probably be to recode
this small sample in C, and see if that works.  If it does, Python (or the
win32 extensions) has a problem we should fix as a matter of priority.  If
not, then we need to find the magic words to chant to make it work from C,
and move that back to Python...

Mark.





From gmcm at hypernet.com  Tue Jun  8 11:38:39 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 8 Jun 1999 10:38:39 -0500
Subject: fork()
In-Reply-To: <m3aeuac0en.fsf@pcrm.win.tue.nl>
Message-ID: <1283265923-67249193@hypernet.com>

Stephan Houben wrote:
...
> Don't know if anyone wrote a Python wrapper around the POSIX shared
> memory stuff. However, you probably don't want/need to do that. The
> reason is that there's basically no (portable) way to get malloc()
> to allocate new objects in the shared memory, so you can't create
> Python objects in the shared memory. So you can probably only access
> the shared memory as a special array a la the "array" module.

Vladimir Marangozov has a SysV shared memory / semphore module at 
http://sirac.inrialpes.fr/~marangoz/python/shm/.

Like Mark Hammond's Win32 shared memory extension it uses a file like 
API.

Probably not what's needed in the context of this thread, but handy 
nonetheless.

- Gordon



From behrends at cse.msu.edu  Wed Jun 30 08:30:49 1999
From: behrends at cse.msu.edu (Reimer Behrends)
Date: 30 Jun 1999 12:30:49 GMT
Subject: Typing [Re: total idiot question: +=, .=, etc...]
References: <000d01bebeda$1b7430a0$d29e2299@tim> <slrn7n8qiv.kl.behrends@allegro.cse.msu.edu>  <3778C2C7.3B101FD0@geekspace.com>
Message-ID: <slrn7nk3gv.18i.behrends@allegro.cse.msu.edu>

Joshua Rosen (rozzin at geekspace.com) wrote:
> Reimer Behrends wrote:
> >         Require variables to be prefixed with "$" to denote scalar
> >         values--strings and integers, "@" for lists, and "%" for
> >         dictionaries.
> > 
> > Given the precedent of having to use self to access instance variables,
> > this is entirely consistent with the current language, and is going to
> > eliminate all those nasty problems where you can't tell from the context
> > what the type of a variable or function is. :)

> 	+There is no difficulty in getting the program to know what type
> 	some piece of data is, which lessens the need for the
> 	type-declaring prefixes--use type(), and then check
> 	obj.__class__ if obj is a class-instance, and you need to -know-
> 	what class it is (if you want to -force- it to be a certain
> 	type, keep reading).

Ahem. I happen to know Python fairly intimately after I don't know how
many years of constant use. What I wrote above was mostly a tongue in
cheek comment (heck, I even said so). The point I was making was that
the very same argument that justifies a self prefix can also be used to
justify type prefixes--in fact, I pretty much paraphrased it from an old
posting by Tom Christiansen on 10/26/98 (the message id is
<710kg1$mr0$1 at csnews.cs.colorado.edu>)

> 	+Requiring variables to bear a prefix denoting type sortof makes
> 	Python into a statically-typed language, where one of the big
> 	reasons for using it is that it's a dynamically-typed language.

I never said otherwise. But the same argument can be used mutatis
mutandum to support a "self"-less notation. :)

> 	+There is no such thing as a `scalare', in Python--strings and
> 	numbers are different data-types, numeric data can actually be
> 	one of many different classes.

I know. It was a facetious argument.

> 	+Forcing data to be of a specific type is already possible--see
> 	documentation on int(), str(), list(), tuple(), float(), long(),
> 	complex(), string.atoi(), et al. This functionality also
> 	lessens, if not completely eliminates the need for typing
> 	prefixes; the program knows what type the data is being coered
> 	into, and so does the programmer. In fact, all of the
> 	functionality that you're proposing (and then some) is already
> 	there--it just looks different.

And I just happen to be aware of that, too.

> 	+If you really want a feature that -really- isn't offered by a
> 	language, the best decision may be to just use another language
> 	that -does- offer it. Try Pike, for one (I suggest it because it
> 	has nifty typing which you may like: you can declare a variable
> 	to be one or -more- types, or use a mixed type). Choose your
> 	languages based upon what you need them to do.

And I know Pike, too. In fact, I'm pretty familiar with it.

By the way, would you mind using shorter lines next time?

				Reimer Behrends



From matz at netlab.co.jp  Mon Jun  7 01:57:17 1999
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 07 Jun 1999 14:57:17 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
 <7j90va$6lk$1@cronkite.cc.uga.edu>
 <u3g1443iht.fsf@ares.sys1.ngo.okisoft.co.jp>
Message-ID: <87pv384koy.fsf@ev.netlab.co.jp>

Hisao Suzuki <suzuki611 at okisoft.co.jp> writes:

|In article <7j90va$6lk$1 at cronkite.cc.uga.edu>,
|graham at sloth.math.uga.edu (Graham Matthews) wrote:
|> I am completely confused by this statement. The two people that have
|> been advocating GC the hardest are myself and the author of Ruby. Since
|> Ruby has GC the author has clearly implemented a collector. ...
|
|Regrettably, I'd have to say that Ruby is (was?) suffering from
|its insufficient portability.  Once I have used a certain i860
|SVR4 box.  On that box, Ruby did not run but dumped immediately
|with its twiddling the machine stack, while Python ran perfectly
|with no difficulty.  As far as I know, the collector of Ruby has
|not changed so much since then.

It's because conservative-ness of Ruby's GC, you know.  There should
be no portability problem with Graham's proposal, mark and sweep GC
with ref counting, unless implementor try to make it conservative.

I've never claimed Ruby is as portable as Python.  Definitely Python
runs on the machines where Ruby cannot run, say on small devices.

|(I have before told this to the author of Ruby on a Japanese
|Pythoneers' mailing-list against his advocacy of Ruby, and now
|he appeared here again with just the same opinion before...)

What opinion of mine do you have in mind?

I claimed in summary:

  * I don't like mere ref counting; it cannot reclaim cyclic data, and
    easy to forget INCREF/DECREF.

  * Many among Python users misunderstood about real GC.
    (slow, hangs, non portable, etc.)

  * ref counting with GC may be good for Python 2.0.

Did I do any advocacy on Ruby?  I was trying constructive discussion.

						matz.



From dkhchan at my-deja.com  Sun Jun 13 21:25:06 1999
From: dkhchan at my-deja.com (dkhchan at my-deja.com)
Date: Mon, 14 Jun 1999 01:25:06 GMT
Subject: Serial programming
References: <7jp1jt$8hg$1@imsp009a.netvigator.com> <i5a83.2876$nn.847925@news.shore.net>
Message-ID: <7k1llb$cae$1@nnrp1.deja.com>

I try but not success.  Modem receive my command (I try to dail out)
but I have not get any respone.  Anything I missed??

My program:
#!/usr/bin/python

modem = open('/dev/ttyS0', 'w+')
modem.write('ATZ\r')
print modem.read()  # no respone here
modem.close()

DORAEMON.HK

In article <i5a83.2876$nn.847925 at news.shore.net>,
  "Michael P. Reilly" <arcege at shore.net> wrote:
> DORAEMON.HK <abc at def.ghi.jkl> wrote:
> : I am looking for some example in python about send and receive data
from
> : serial port.  Like send "ATZ" and receive "OK".  Any help?? Thanks
a lot!!
>
> You can open the device file directly and handle the information
> yourself.  UNIX devices are handle just like files.  You can change
> device attributes (speed, parity, etc.) with the termios module or
> fcntl.ioctl function.
>
>   modem = open('/dev/cua0', 'r+')
>   modem.write('AT\r')
>   f = modem.read()
>   if f[:2] != 'OK':
>     raise ValueError, "modem not okay"
>   ...
>
> There are some Expect modules/extensions around that could help you as
> well;  my ExpectPy extention is at
> http://starship.python.net/~arcege/ExpectPy/.  You can find others
> through the Python website.
>
>   import ExpectPy
>   modem = ExpectPy.spawnfile(open('/dev/cua0', 'r+'))
>   modem.send("AT\r")
>   try:
>     f = modem.expect(
>       (ExpectPy.EXACT, "OK\r\n", "OK"),
>       (ExpectPy.GLOB, "*\r\n", "Not OK")
>     )
>     if f == "Not OK":
>       raise ValueError, "modem not okay"
>   except (ValueError, ExpectPy.TimeoutError), value:
>     raise ValueError, value
>
> I hope this helps.
>
>   -Arcege
>
>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From paul at prescod.net  Mon Jun  7 09:23:52 1999
From: paul at prescod.net (Paul Prescod)
Date: Mon, 7 Jun 1999 13:23:52 GMT
Subject: Suggestion: Dutch Python mailinglist
References: <slrn7lli2q.ia.gerrit@gerrit.palga.uucp>
Message-ID: <375BC7E8.BC53EFD@prescod.net>

Gerrit Holl wrote:
> 
> I know there are many Dutch
> python-users, but most of them know English very well. 

Dutch people use Python? Since when?

> Ik ben een Nederlandse jongen van 13 jaar, en mijn Engels is niet goed. 

Actually, I can pretty much read that without learning Dutch. It looks
like English with some spelling mistakes. And no more typing mistakes than
is typical for mailing lists.

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for only himself
 http://itrc.uwaterloo.ca/~papresco

"Silence," wrote Melville, "is the only Voice of God." The assertion,
like its subject, cuts both ways, negating and affirming, implying both
absence and presence, offering us a choice; it's a line that the Society
of American Atheists could put on its letterhead and the Society of
Friends could silently endorse while waiting to be moved by the spirit
to speak. - Listening for Silence by Mark Slouka, Apr. 1999, Harper's



From Timothy.R.Evans at p98.f112.n480.z2.fidonet.org  Wed Jun 30 19:05:42 1999
From: Timothy.R.Evans at p98.f112.n480.z2.fidonet.org (Timothy R Evans)
Date: Thu, 01 Jul 1999 00:05:42 +0100
Subject: Arbitrary precision calculations
Message-ID: <0000159e@bossar.com.pl>

From: Timothy R Evans <tre17 at pc142.cosc.canterbury.ac.nz>

catlee at my-deja.com writes:

> Hi,
> 
> I'm looking for a package that will let me do arbitrary precision
> arithmatic in python.  I've come across the real.py module written by
> Jurjen N.E. Bos, and it seems like a really nice module except for the
> fact that you can't compare two numbers for equality,inequality,etc.
> 
> Cheers,
> Chris
> 
> P.S. I found the real.py package by doing a search on www.deja.com for
> arbirary precision in comp.land.python
> 
> 
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.

If I'm reading this right you want arbitrary precision integers right?
Just use the Python `Long' integer, created using the `long' function
or by placing an L after a literal number (e.g. 10L).  All standard
numerical operators ork on long intergers (if you're doing crypto
stuff then don't forget that `pow(X,Y,Z)' calucates X to the power of
Y, modulo Z efficiently.

--
Tim Evans



From kiki at pixar.com  Thu Jun 24 22:52:19 1999
From: kiki at pixar.com (kiki)
Date: Thu, 24 Jun 1999 19:52:19 -0700
Subject: Tkinter: radiobutton
References: <3771B52E.80F3FA46@pixar.com> <qjsc3.77$Of3.12274@news.shore.net>
Message-ID: <3772EEE3.2DCCE7C0@pixar.com>

"Michael P. Reilly" wrote:

>
> No, it's not an obvious question at all.  Unfortunately, you need to
> track the index when you create the menu.  By default, the first call
> starts at 1, because the "tear-off" is 0.  If the menu is created with
> the -tearoff=0 option, then the index starts at 0
>

Bleh.  Oh well.  Kind of a shame, since I'd rather just have the color value
put into the color variable, not an index with my own list I have to maintain
myself.

Well, cool, at least I have a solution.

Thanx so much for your help!

Kiki




From doughellmann at mindspring.com  Fri Jun 18 14:16:58 1999
From: doughellmann at mindspring.com (Doug Hellmann)
Date: Fri, 18 Jun 1999 18:16:58 GMT
Subject: error? inconsistent dedent
References: <376A6796.814CBA24@phoenix-edv.netzservice.de>
Message-ID: <376A8DB5.3B008C8C@mindspring.com>

Holger Jannsen wrote:
> 
> What does 'inconsistent dedent' mean?
> Why does that error occur?

You could post the code for a more specific answer, but usually when I
see that error it means that I'm mixing tabs and spaces for indention
and the count of "whitespace" on the line in question is off.  I'd check
that, first.

Doug



From aa8vb at ipass.net  Fri Jun 25 13:32:08 1999
From: aa8vb at ipass.net (Randall Hopper)
Date: Fri, 25 Jun 1999 13:32:08 -0400
Subject: Python / VTK problem
In-Reply-To: <37738721.E2B17705@nevex.com>; from Greg Wilson on Fri, Jun 25, 1999 at 09:41:53AM -0400
References: <37738721.E2B17705@nevex.com>
Message-ID: <19990625133208.A755922@epa.gov>

Greg Wilson:
 |Hi, folks.  I've just downloaded, compiled, and installed VTK 2.3, and
 |for the first time, I've built with Python extensions enabled.
 |Unfortunately, while I can run the Tcl examples, I can't get the Python
 |examples to run.  The command 'python mace.py' (or similar) produces:
 |
 |  Traceback (innermost last):
 |    File "./mace.py", line 2, in ?
 |      from libVTKGraphicsPython import *
 |ImportError: /home/gvwilson/lib/libVTKGraphicsPython.so: undefined
 |symbol: XtStrings
 |
 |The symbols XtStrings is defined in /usr/X11R6/lib/libXt.so (I'm on Red
 |Hat Linux).  The directory /usr/X11R6/lib is in my LD_LIBRARY_PATH and in
 |my PYTHONPATH, as is /home/gvwilson/lib (which contains symbolic links to
 |the VTK .so's).  I'd be grateful to anyone who can tell me what I'm doing
 |wrong.

It sounds like libVTKGraphicsPython.so wasn't implicitly linked with
libXt.so when you compiled VTK.  No problems here on IRIX; it appears that
the linker pulled it in automatically.

To check:

   > cd vtk/graphics
   > ldd libVTKGraphicsPython.so
     [you should see libXt.so in the list; but I'd wager that you won't]
   > rm libVTKGraphicsPython.so
   > make
     [VTK doesn't explicitly link in Xt, but it probably should]

A quick nm of the source files linked into libVTKGraphicsPython reveals
that vtkXRenderWindowInteractor.o depends on Xt calls.

Try adding a:

   X_EXTRA_LIBS = -lXt -rpath /usr/X11R6/lib

to your user.make file.

Randall




From pduffin at mailserver.hursley.ibm.com  Fri Jun 11 08:36:25 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Fri, 11 Jun 1999 13:36:25 +0100
Subject: Python as a shared library on Unix
Message-ID: <376102C9.5648@mailserver.hursley.ibm.com>

Is it possible to build Python itself as a small executable and a
shared library, rather than an archive library which of
course results in a monolithic executable. 

I know how to get the modules built as shared libraries but cannot get
Python built as a shared library.

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From bwinton at tor.dhs.org  Thu Jun 17 00:07:44 1999
From: bwinton at tor.dhs.org (Blake Winton)
Date: Thu, 17 Jun 1999 04:07:44 GMT
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <37682DD3.43923243@compaq.com> <37684B76.AA8D0CD5@hex.net>
Message-ID: <slrn7mgpll.p7.bwinton@tor.dhs.org>

On Wed, 16 Jun, Peter A. Koren <pkoren at hex.net> wrote:
snip
>In the spirit of abductio ad absurdum, I can do everything you can do
>in python or Tcl with a Turing machine.  I don't.  I am also not into
>whips and chains.

And, since there is a Turing machine implementation for vi, it can do
anything Python can do.  Hmmm...  Perhaps I'll switch from using vi to
write Python code to using Python to write vi code...[0]

Later,
Blake.

[0] Of course, I'm using vim with Python support, so I've already used
    Python to write vi code, but that's beside the point.

-- 
One Will.  One Dream.  One Truth.  One Destiny.  One Love.



From phd at emerald.netskate.ru  Wed Jun 23 06:54:00 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Wed, 23 Jun 1999 14:54:00 +0400 (MSD)
Subject: PIL and cgi link problem
In-Reply-To: <7kqcd3$3at$1@nnrp1.deja.com>
Message-ID: <Pine.LNX.4.04.9906231452210.8942-100000@emerald.netskate.ru>

Hi!

On Wed, 23 Jun 1999 entree at my-deja.com wrote:
> We want to use the library from a python cgi

   Can you use it from command-line programs?

> script. Here we encounter a strange problem with
> the Image module. The cgi script tries to import
> it in response to a web-query. but it failes with
> an ImportError Exception: _imaging C module is
> not installed. further investigation of the
> exception showed the following (link) error:
> undefined symbol PyExc_IOError

   What OS? Compiler?
   For Linux, sometimes it need to export symbols from main module (-Wl,-E
compiler option)...

> Does anybody has an idea?.
> 
> We are using Python 1.5.1 and LibImaging 1.0b1
> 
> Thanx in advance,
> 
> Robbert
> Entree

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From greg.ewing at compaq.com  Sun Jun 27 22:27:00 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Mon, 28 Jun 1999 14:27:00 +1200
Subject: RuntimeError?
References: <19990627112432.A29097@optiplex.palga.uucp>
Message-ID: <3776DD74.4C1B2342@compaq.com>

Gerrit Holl wrote:
> 
> can someone please give me an example when a RuntimeError is raised?

At a guess I'd say it's a base class from which
other runtime error classes are derived, and
isn't actually raised itself.

Greg



From larsga at ifi.uio.no  Wed Jun 16 03:41:34 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 16 Jun 1999 09:41:34 +0200
Subject: XML-toolkit for Windows - binaries available?
References: <376701a3.72288277@newssrv>
Message-ID: <wkzp208udt.fsf@ifi.uio.no>

* Lee Borkman
| 
| Can anyone help me out in installing from the standard distribution?
| I don't have a C-compiler on my system, but I do have nmake/dmake.

I think the web page is out of date, since the /xml/windows directory
in the CVS does contain the .dll files for pyexpat and sgmlop (and has
done for quite a while). So just download the ZIP file and they should
be there.

If not, send me an email and I'll send them to you.

--Lars M.



From rceci at my-deja.com  Mon Jun  7 15:50:07 1999
From: rceci at my-deja.com (rceci at my-deja.com)
Date: Mon, 07 Jun 1999 19:50:07 GMT
Subject: Question about envvars
References: <7jgl9n$p13$1@nnrp1.deja.com> <375BFE38.3FB3E83C@ix.netcom.com>
Message-ID: <7jh7pc$as$1@nnrp1.deja.com>

Stephen,

Boy I was excited to get a reasonable explaination so quickly, but alas,
it still does not work.  Here is my test script:

#!/usr/local/bin/python

import MDIreg

reg = MDIreg.FileRegistry(1,1)
reg.exportEnvVars()

import os
print "child has it : "
os.system("env | grep MDI_REG_VERSION")
print "but do I have it? :"
print os.environ['MDI_REG_VERSION']

---

I am interested in this "MDI_REG_VERSION" envvar which gets exported by
my library.  Here is the output:

child has it :
MDI_REG_VERSION=10.0
but do I have it? :
Traceback (innermost last):
  File "regtest.py", line 12, in ?
    print os.environ['MDI_REG_VERSION']
  File "/usr/local/lib/python1.5/UserDict.py", line 12, in __getitem__
    def __getitem__(self, key): return self.data[key]
KeyError: MDI_REG_VERSION

Any other ideas?  Any way to tell the os/posix module to "refresh"?

By the way, thanks for the library implementation suggestion, but I must
keep it the way it is, because other c, and c++ (at the class level)
programs access this library.

Thanks

Rob



In article <375BFE38.3FB3E83C at ix.netcom.com>,
  "Stephen J. Turner" <sjturner at ix.netcom.com> wrote:
> Hi Rob,
>
> > I have built a shared module that reads flat files for environment
> > variable information, and makes subsequent putenv calls for those
> > pertinent ones.  This functionality is encapsulated in a c++ class,
> > FileRegistry, which I think I have working correctly, using shadow
> > classes under python.  All I expose in the class is the ctor, the
dtor,
> > and the method called exportEnvVars():
> >
> > %module MDIreg
> > %{
> > #include "fileregistry.h"
> > %}
> >
> > class FileRegistry
> > {
> > public:
> >   FileRegistry(bool, bool);
> >  ~FileRegistry();
> >   void exportEnvVars();
> > };
> >
> > ...
> >
> > However, I can't seem to get the envvars to persist into the current
> > process space.  If I fork off a child, then I can see the envvars
that I
> > expect.
>
> The problem is that, although putenv does indeed modify the current
> process's environment, Python only reads the environment when the
posix
> module (imported by os) is first imported, so as to create the environ
> dictionary.  (Try confirming this behavior by postponing "import os"
> until after your call to MDIreg.exportEnvVars.  You should see your
> environment changes reflected in os.environ.)
>
> The proper solution is, I think, to avoid the direct use of the putenv
> function, and instead to use the Python C API to modify the os.environ
> mapping (which calls putenv as a side effect).  Here is a crude C
module
> to do just such a thing:
>
> -------------------- START modenv.c --------------------
> #include <Python.h>
>
> #include <stdio.h>
> #include <string.h>
>
> static PyObject *modenv_load(PyObject *self, PyObject *args)
> {
>   FILE *fp;
>   PyObject *os, *environ, *v;
>   char buf[256];
>   char *file;
>   char *value;
>
>   if (!PyArg_ParseTuple(args, "s", &file))
>     return NULL;
>
>   os = PyImport_ImportModule("os");
>   environ = PyObject_GetAttrString(os, "environ");
>
>   if (fp = fopen(file, "r")) {
>     while (fgets(buf, sizeof(buf), fp), !feof(fp)) {
>       /* skip lines that do not contain equals sign */
>       if (!(value = strchr(buf, '=')))
> 	continue;
>
>       /* remove trailing newline */
>       buf[strlen(buf)-1] = '\0';
>
>       *value++ = '\0';
>       v = PyString_FromString(value);
>       PyMapping_SetItemString(environ, buf, v);
>       Py_DECREF(v);
>     }
>     fclose(fp);
>   }
>
>   Py_DECREF(environ);
>   Py_DECREF(os);
>
>   Py_INCREF(Py_None);
>   return Py_None;
> }
>
> static struct PyMethodDef modenv_methods[] = {
>   {"load", (PyCFunction)modenv_load, METH_VARARGS},
>   {NULL, (PyCFunction)NULL, 0, NULL}
> };
>
> void initmodenv()
> {
>   if (!Py_InitModule4("modenv", modenv_methods, NULL, NULL,
> 		      PYTHON_API_VERSION))
>     Py_FatalError("can't initialize module modenv");
> }
> -------------------- END modenv.c --------------------
>
> Hope it helps.
>
> Regards,
> Stephen
>
> --
> Stephen J. Turner <sjturner at ix.netcom.com>
>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From donn at u.washington.edu  Thu Jun 17 01:50:38 1999
From: donn at u.washington.edu (Donn Cave)
Date: 17 Jun 1999 05:50:38 GMT
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <37682DD3.43923243@compaq.com>
Message-ID: <7ka2be$qja$1@nntp6.u.washington.edu>

Greg Ewing <greg.ewing at compaq.com> writes:
...
| In my experience, I find that I can do with Python everything
| that Tcl was designed for, do it more easily, do a lot
| more besides, and have more fun in the process. I believe
| the reason for this is rooted in some fundamental design
| features of these languages, which I have sketched above.
| I think that's about as close as one can get to providing
| an objective argument as to whether one language is better
| than another for a given purpose.

How about Expect as an example?  I don't mean to criticize the
existing Python Expect implementation(s), have no idea where the
state of the art is on that.  I just remember trying to think of
a natural Python idiom that would replace the "expect" verb's case
switch flow of control.

My experience is more or less the same as yours - rewrote Tcl
software in Python and was henceforth a convert.  But I think
Tcl is syntactically more adaptable, where Python kind of makes
a virtue of its fixed ways.

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



From john at micrsoft.com  Fri Jun 11 01:53:16 1999
From: john at micrsoft.com (John)
Date: Thu, 10 Jun 1999 22:53:16 -0700
Subject: WinNT: Corrupt installation detected
References: <7joq9h$qd2$1@schbbs.mot.com>
Message-ID: <7jq8b6$fa4@dfw-ixnews4.ix.netcom.com>

Check for a virus, it could be the CIH virus.





From tacitus at home.com  Sun Jun 20 17:05:51 1999
From: tacitus at home.com (TREVOR HUGHES)
Date: Sun, 20 Jun 1999 21:05:51 GMT
Subject: Python programming / Medford NJ
Message-ID: <376D5880.BD548A0F@home.com>

Programming positions available
Full time and part time


Full time position: Project leader/programmer
Skills:
     Highly skilled in Python/JPython
     Experience with PostgreSQL, Informix or Oracle, etc.
     Database design and normalization, pattern and workflow modeling
     GUI design experience
     Documentation stickler
     Familiarity with Linux


Part time positions: Programmer
Python, Java and HTML programmers with the following backgrounds are
needed.
     GUI
     Database
     Application/Module 'Glueing'
     Accounting



Please send your resume and portfolio to mailto:tacitus at home.com
  Include a method to view sample applications you have written (URL,
  executable, etc.) and excerpts of documented python code of which you
  are particularly proud.

The positions are based in Medford, New Jersey.

Contact:
Trevor Hughes
609.714.8900 x 251
tacitus at home.com



From proteus at cloud9.net  Tue Jun  1 13:09:44 1999
From: proteus at cloud9.net (Michael Muller)
Date: Tue, 01 Jun 1999 13:09:44 -0400
Subject: Windows Stack Size
References: <3752C225.65C98B52@cloud9.net> <3752F154.6C444928@ricochet.net>
Message-ID: <375413D8.C3B02D2@cloud9.net>

Guido van Rossum wrote:
[snip]
> > I've experienced some traps creating some deeply nested Tkinter controls
> > under windows and I'm wondering if it could be due to a stack overflow.
> 
> If it is indeed a stack overflow, the cause is more likely infinite
> recursion on
> your part.
> 
> --Guido van Rossum

AFAIK, the code is not recursive (sometimes with GUI callbacks, you can 
get a lot deeper into the stack than you might expect :-).  It's been 
running fine under Linux for months, but under Win 95 it traps
consistantly in some windows, _but not in all of them_, which is why I 
thought there might be a tiny-stack issue. 1 meg should be plenty of
room.  
Here is the control in question:

class ScrollingText(Frame):
   
   def __init__(self, parent = None, **kw):
      self.text = None
      if parent:
         Frame.__init__(self, parent)
      else:
         Frame.__init__(self)
      self.text = apply(Text, (self,), kw)
      self.text.grid(sticky = N+S+E+W)
      
      self.vsb = Scrollbar(self, command = self.text.yview)
      self.vsb.grid(row = 0, column = 1, sticky = N+S)
      
      self.hsb = Scrollbar(self, command = self.text.xview,
                           orient = 'horizontal'
                           )
      self.hsb.grid(sticky = W+E)
      
      self.text.config(yscrollcommand = self.vsb.set,
                       xscrollcommand = self.hsb.set,
                       wrap = 'none'
                       )
      self.rowconfigure(0, weight = 1)
      self.columnconfigure(0, weight = 1)
   
   def __getattr__(self, attr):
      if self.text:
         return getattr(self.text, attr)
      else:
         raise AttributeError(attr)

Upon further consideration, I wonder if the getattr could have something
to do with this.

I'd debug it myself except that I'm happily Windows-free (only creates
problems when I have to write code for Windows users).  If there is no
obvious answer, I have a work-around.

=============================================================================
michaelMuller = proteus at cloud9.net | http://www.cloud9.net/~proteus
-----------------------------------------------------------------------------
In this book it is spoken of the Sephiroth, and the Paths, of Spirits
and
Conjurations; of Gods, Spheres, Planes and many other things which may
or
may not exist.  It is immaterial whether they exist or not.  By doing
certain things certain results follow. - Aleister Crowley
=============================================================================



From jbauer at rubic.com  Wed Jun  2 09:14:27 1999
From: jbauer at rubic.com (Jeff Bauer)
Date: Wed, 2 Jun 1999 13:14:27 GMT
Subject: Is pcgi usable without Zope?
References: <ltd7zf6wtv.fsf@asfast.com> <3755CA26.BCB7828A@digicool.com>
Message-ID: <37552E33.C3CA673B@rubic.com>

Paul Everitt wrote:
> 
> Lloyd Zusman wrote:
> >
> > I have to admit that I'm a bit confused about pcgi these days.  The
> > most recent documentation for it was last updated in August, 1998, and
> > it still refers to Bobo and Principia.
> 
> We're certainly not helping -- I agree that stuff is strewn around all
> over the place.
> 
> > Now that we have Zope, is pcgi now intimately bundled with it, never
> > to emerge on its own again?  Or is there a way to run pcgi without
> > Zope?
> 
> I don't think PCGI is intimately bundled with Zope.  Neither is the
> object publisher, object database, the template system, etc.  All can be
> used independently.
> 
> --Paul

Argh.  I responded to Lloyd and CC'd the Zope list, but
neglected to include the Python list.  My reply follows.

-Jeff
----------------

Lloyd,

PCGI is a tiny, yet integral part of Zope for some set of
Zope applications, esp. where a web server front-end 
(e.g. Apache) is desired.

It is possible to run Zope without PCGI (e.g. internal network,
a.k.a. buzzword: 'intranet').  It is also possible to use PCGI 
without Zope.  The pcgitest example is provided to test and 
demonstrate a free-standing PCGI application.

And yet another alternative is to use PCGI with Zope components
(ZPublisher, DocumentTemplate, BoboPOS) to build custom
solutions.

the-tyranny-of-choice-prevails-ly yr's

Jeff Bauer
Rubicon, Inc.

P.S.  A couple weeks ago I posted a message describing a new
version of pcgi_publisher that would include an inactivity
timer, i.e. would shutdown the LRP automatically after a 
specified period of inactivity.  I think some people may
find such a feature useful for custom PCGI apps.  -JB



From tre17 at pc142.cosc.canterbury.ac.nz  Tue Jun 29 19:10:51 1999
From: tre17 at pc142.cosc.canterbury.ac.nz (Timothy R Evans)
Date: 30 Jun 1999 11:10:51 +1200
Subject: printing with print
References: <377925EF.1CBB640F@starvision.com>
Message-ID: <c75iu8661s4.fsf@pc142.cosc.canterbury.ac.nz>

Angus MacKay <amackay at starvision.com> writes:

> is there any other way to print that with print?
> perhaps with a real function.
> 
> I want to print two things without a space being added between them:
> foo = "there"
> print "hi" + foo
> 
> works fine but: 
> foo = 2
> print "hi" + foo
> 
> does not. I like the printing of any object ability of "print" so I
> do not want to use "" % style printing, I just want to be able to
> govern the whitespace that gets output.
> 
> cheers, Angus.

import string

def print_without_space(*args):
    print string.join(map(str, args), '')


>>> foo = 2
>>> print_without_space('foo', '2', foo)
foo22

This does what you want, but wouldn't the % operator be easier?

--
Tim Evans




From Lyle.Johnson at p98.f112.n480.z2.fidonet.org  Wed Jun 30 00:07:51 1999
From: Lyle.Johnson at p98.f112.n480.z2.fidonet.org (Lyle Johnson)
Date: Wed, 30 Jun 1999 05:07:51 +0100
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
Message-ID: <000015b6@bossar.com.pl>

From: Lyle Johnson <jlj at cfdrc.com>

Just to add fuel to the fire, we now have a Python interface to FOX, a
cross-platform GUI toolkit:

    http://www.cfdrc.com/pub/FOX/fox.html

The latest version of the Python wrappers can be picked up at:

    ftp://ftp.cfdrc.com/pub/FOX/FXPy-0.99.42.tar.gz

I don't know how the speed of FXPy would compare with some of the others
mentioned (Tk, wxWindows, Qt, GTK, ...) but it would be worthwhile for
some independent opinions.

Lyle

Cameron Laird wrote:

> In article <xJRc3.292$Of3.67094 at news.shore.net>,
> Michael P. Reilly <arcege at shore.net> wrote:
>                         .
>                         .
>                         .
> >There are other graphical packages for Python out there too, it would
> >be appropriate to contrast them as well.
>                         .
>                         .
>                         .
> I occasionally work on such a comparison, currently available at <URL:
> http://starbase.neosoft.com/~claird/comp.lang.python/python_GUI.html>.
> I welcome updates from readers.
> --
>
> Cameron Laird           http://starbase.neosoft.com/~claird/home.html
> claird at NeoSoft.com      +1 281 996 8546 FAX



From doughellmann at mindspring.com  Fri Jun 11 10:11:41 1999
From: doughellmann at mindspring.com (Doug Hellmann)
Date: Fri, 11 Jun 1999 14:11:41 GMT
Subject: tkinter and events
References: <7oX73.898$q3.114540@news21b.ispnews.com>
Message-ID: <376119F3.F13759C9@mindspring.com>

Paul Stillwell wrote:
> 
> So, when I run the code and double click on o1, I get:
> 
> >>> Clicked on domain 20
> 
> which is the last object I created even though I double clicked on the first
> object.  What am I doing wrong?

The reason you're always seeing the same value is that the call in
MyTestObject.__init__ to bind <Double-1> on the canvas always changes
the existing binding on the *canvas*, but what you want is to bind
events on the *rectangle*.

Try changing these two lines:

self.Box = Rectangle(Parent, self.Begx, self.Begy, self.Endx,
                             self.Endy)
...
self.Canvas.bind('<Double-1>', self.DoubleClickHandler)

to:

self.Box = Rectangle(Parent, self.Begx, self.Begy, self.Endx,
                             self.Endy,
                             fill='white')
...
self.Box.bind('<Double-1>', self.DoubleClickHandler)

I added the fill to the Rectangle so that I could double click on it
more easily.  Then, the important change is to use the bind method of
the Rectangle object, instead of the Canvas.

Doug



From skip at mojam.com  Tue Jun 15 22:54:53 1999
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 16 Jun 1999 02:54:53 GMT
Subject: Cut down on writing repetitive code
References: <3766f6a8.19474000@news2.mainnet.net>
Message-ID: <3767128E.E049636C@mojam.com>

Tim Gallati wrote:

> If you do a lot of repetitive typing, check out Perfect Keyboard Lite.
> I use it everyday and consider it my most useful program.

Gee, I thought the usual OO way to cut down on repetitive typing was to
abstract common code into base classes.  I guess the original OO folks
(Simula? Smalltalk?) got it all wrong 20- to 30-odd years ago.  They
should have invested in Mavis whats-her-name instead!

No disclaimers in the original post.  I guess it was an off-charter
advertisement.  Another company hits my bit bucket.

-- 
Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
518-372-5583



From mgm at unpkhswm04.bscc.bls.com  Mon Jun 21 13:04:44 1999
From: mgm at unpkhswm04.bscc.bls.com (Mitchell Morris)
Date: 21 Jun 1999 17:04:44 GMT
Subject: It's just magic, and I hate it ...
Message-ID: <slrn7mss5c.f90.mgm@unpkhswm04.bscc.bls.com>

Following is a piece of Python that is making me pull out my hair. The idea
was to have a long-running process started by the user (via browser), and
have the results trickle back in without timing out the request. My strategy
was to define a per-connection file to hold results, fork off a child to
generate results and put them into the file, and have the parent return a
redirect to the results file. That way everyone is satisfied.

It works if I run it from the command line as myself. It works if I run it
from the command line as user 'WWW' group 'WWW' which is how the server will
see it. The results file is created, the redirect is printed and a shell
prompt returns, and the child has forked off and in fact updates the file
every so often until it finishes at which time it replaces the contents with
the simulated results.

When run from the browser, however, the redirect doesn't happen until the
child finishes. To recap, this is exactly the situation I need to avoid.

Please help.

+Mitchell



#! /usr/bin/env python

import sys
import cgi
import time
import os
import string
import socket

form = cgi.FieldStorage()
if form.has_key("title"):
	hostname = socket.gethostbyaddr(socket.gethostname())[0]
	uniq = hex(os.getpid())[2:] + hex(int(time.time()))[2:]
	addr = "http://%s/~mgm/results/%s.html" % (hostname, uniq)
	fn = "/home/mgm/public_html/results/%s.html" % uniq

	# Generate the results file
	open(fn, "w").write("""<HTML>
<HEAD>
<META HTTP-EQUIV=REFRESH CONTENT=5>
<TITLE>Results for [%s]</TITLE>
</HEAD>
<BODY>
<P>Beginning search at %s:
</BODY>
</HTML>
""" % (form["title"].value, time.asctime(time.localtime(time.time()))))

	# Parent returns a redirect, child pretends to work for a while
	# accumulating results
	if os.fork() == 0:
		for i in (1, 2, 3):
			txt = open(fn, 'r').readlines()
			txt.insert(len(txt)-2, "<P>Still searching\n")
			open(fn, 'w').write(string.join(txt, ''))
			time.sleep(10)
		open(fn, 'w').write("""<HTML>
<HEAD>
<TITLE>Finished</TITLE>
</HEAD>
<BODY>
<H1>Finished</H1>
<P>If you had results, they'd be here
</BODY>
</HTML>
""")
	else:
		print "Status: 302 Moved"
		print "Location:", addr
else:
	print """Content-type: text/html

<HTML>
<HEAD>
<TITLE>Sample Form</TITLE>
</HEAD>
<BODY>
<P>
<H1>Sample Form</H1>
<FORM METHOD="POST">
Title: <INPUT TYPE="TEXT" NAME="title"><P>
<INPUT TYPE="SUBMIT" NAME="submit">
</FORM>
</BODY>
</HTML>"""



-- 
Mitchell Morris

A man said his credit card was stolen but he decided not to
report it because the thief was spending less than his wife
did.



From dalke at bioreason.com  Sun Jun 27 20:41:33 1999
From: dalke at bioreason.com (Andrew Dalke)
Date: Sun, 27 Jun 1999 18:41:33 -0600
Subject: GC In Python: YAS (Yet Another Summary)
References: <000301bebffc$184fcf60$da9e2299@tim> <3776A025.53B35103@inrialpes.fr>
Message-ID: <3776C4BD.F266EC1@bioreason.com>

Vladimir Marangozov <Vladimir.Marangozov at inrialpes.fr> said:
> you seem to overlook one of the nicest features of pymalloc. If
> the extension module works in current Python + libc malloc, it
> will work with Python + pymalloc because of the dynamic block
> management.

Ahh, but as I found out trying 1.5.2beta, the library we use
requires libmalloc, and not the libc malloc; it wants mallopt.

To make things worse, it doesn't actually *use* that function.
It contains an internal function which references mallopt but the
header files don't list the internal function so none of our code
uses it.  We only need the -lmalloc for creating shared libs with
1.5.2.  I believe their current version removed this requirement.


						Andrew
						dalke at bioreason.com



From marilena_kolia at my-deja.com  Sun Jun 27 06:13:58 1999
From: marilena_kolia at my-deja.com (marilena_kolia at my-deja.com)
Date: Sun, 27 Jun 1999 10:13:58 GMT
Subject: sms tests
Message-ID: <7l4th5$kub$1@nnrp1.deja.com>

Hi,

I am trying to test the SMS performance and get some results about the
time it takes a message to be transfered, but also the delays of each
part of this procedure. I am planning to use Tcl programming. Does
anybody know how to do it or have some suggestions about it?

Thank you



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From johann at physics.berkeley.edu  Fri Jun 11 17:47:25 1999
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 11 Jun 1999 14:47:25 -0700
Subject: Problem with Lists & Index command
References: <rNHvjEdhm5Pp-pn2-zOTjsVjHebb4@d185d18f8.rochester.rr.com>
Message-ID: <mthfoetno2.fsf@astron.berkeley.edu>

Benjamin Schollnick writes:

> It appears that INDEX doesn't work on a multiple
> dimension list????

Yes, index won't work.

Why?  Because the list is a list of tuples,

  lst = [("key1", "val1"), ("key2", "val2")...]

so "key1" isn't an element of the list.  The tuple ("key1", "val1")
is, however, but to find that you'd need to know the value, which
defeats the purpose.

Basically, the list could have been

   lst = [("key1", "val1"), ("key2", "val2"), "key1"]

Here, it seems much more plausible that lst.index("key1") will return
2, not 0.  Since list can contain any type, it doesn't automatically
know that it should look inside sublists to find elements.

-- 
Johann Hibschman                           johann at physics.berkeley.edu



From kontiki3136 at my-deja.com  Sat Jun 26 01:07:55 1999
From: kontiki3136 at my-deja.com (kontiki3136 at my-deja.com)
Date: Sat, 26 Jun 1999 05:07:55 GMT
Subject: is there a Gantt chart toolkit?
References: <7in1y2imtt.fsf@betty.artisan.com>
Message-ID: <7l1n76$o9g$1@nnrp1.deja.com>

In article <7in1y2imtt.fsf at betty.artisan.com>,
  alt at artisan.com wrote:
>
> subject says it all.
>
> Thanks!
> Albert

Try PlanBee (http://www.guysoftware.com/planbee.htm)


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From bb296_nospam_ at ecs.soton.ac.uk  Tue Jun  8 17:03:26 1999
From: bb296_nospam_ at ecs.soton.ac.uk (Ben)
Date: Tue, 8 Jun 1999 22:03:26 +0100
Subject: Problem with embedded python
Message-ID: <7jk0cm$4hk@wapping.ecs.soton.ac.uk>

Greetings,
    I have embedded python into a win32 c++ application using
PyRun_SimpleFile. This executes a python script at regular (25 times a
second-ish) intervals. Also included in this C++ application is a procedure
which handles the window messages, which I have also tried to script.
Unfortunately, running any script in this function crashes windows with a
nasty protection fault.
Do I need to run my scripts on seperate threads... or in different
interpreters. or is this genuinely a bug..

Yours confusedly

Sean (srd196 at ecs.soton.ac.uk)





From pduffin at mailserver.hursley.ibm.com  Fri Jun 11 08:02:41 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Fri, 11 Jun 1999 13:02:41 +0100
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <375FD703.C6D@mailserver.hursley.ibm.com> <375FEC11.FD20EDE6@iname.com>
Message-ID: <3760FAE1.373B@mailserver.hursley.ibm.com>

Fernando Mato Mira wrote:
> 
> Paul Duffin wrote:
> 
> > Whether the choice of language is a good or bad idea depends entirely on
> > the context. i.e. environment and job to be done.
> >
> > You cannot avoid taking into account the context even if you only look
> > at the language from a semantic / syntactic view point.
> 
> OK. Let's drop the word `choice'. Why do people think tcl is a bad
> language then, strictly from a _design_ viewpoint? Let's assume
> imperative languages are OK (responses like "it's not declarative"
> are obvious and uninteresting and do not help comparing it to other
> imperative languages).

Obviously I don't it has a few holes in it and some things could be
improved but I don't consider it a "bad" language.

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From akuchlin at mems-exchange.org  Tue Jun  1 16:13:12 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: Tue, 1 Jun 1999 16:13:12 -0400 (EDT)
Subject: Browsers
In-Reply-To: <dkuhlmanFCnxzH.72@netcom.com>
References: <37530E1B.CAC2991C@cwcom.net>
	<dkuhlmanFCnxzH.72@netcom.com>
Message-ID: <14164.16088.585496.742187@amarok.cnri.reston.va.us>

>Daniel Faulkner <m01ymu00 at cwcom.net> wrote:
>> Is there a basic browser some where that I can look at to see how it
>> works? (not grail)
>> As I can't understand much of the python internet software and don't
>> understand how to parse the HTML once I've got it.

G. David Kuhlman writes:
>Lynx is a text mode browser:
>    http://lynx.browser.org/
>For fancier stuff, look at Mozilla:
>      http://www.mozilla.org/

       Note, however, that an HTML parser capable of coping with all
the invalid HTML on the Web is a complicated beast.  For example, Lynx
currently has an SGMLish style parser that has been brain damaged in
various ways to cope with invalid HTML.  I don't know how much error
correction Grail includes, but it might actually be a simpler parser
if it hasn't been complicated with various error recovery hacks.
Another good option might be to look at the test code in htmllib.py,,
which does simple HTML-to-text formatting.  (When trying to figure out
a module, always look in the module's code first, since authors will
often include simple examples or test scripts inside an 'if
__name__=='__main__'" block.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
Time, place, and action may with pains be wrought, / But Genius must be born;
and can never be taught.
    -- John Congreve




From wtanksle at dolphin.openprojects.net  Wed Jun 30 20:46:28 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Thu, 01 Jul 1999 00:46:28 GMT
Subject: Language extensibility (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <37682DD3.43923243@compaq.com> <7ka2be$qja$1@nntp6.u.washington.edu> <37776FFA.2D85@mailserver.hursley.ibm.com> <lw7looxx7m.fsf@copernico.parades.rm.cnr.it> <876748mcxg.fsf@home.ivm.de> <37788BB9.ABD@mailserver.hursley.ibm.com> <lw67471fyi.fsf@copernico.parades.rm.cnr.it> <3779ED0F.2C67@mailserver.hursley.ibm.com> <nkjzp1hizvb.fsf@tfeb.org> <377A3383.D453468@iname.com> <377aa228.31380423@news.jpl.nasa.gov>
Message-ID: <slrn7nlej3.5j1.wtanksle@dolphin.openprojects.net>

On Wed, 30 Jun 1999 23:07:23 GMT, William H. Duquette wrote:
>On Wed, 30 Jun 1999 17:10:59 +0200, Fernando Mato Mira
>>Tim Bradshaw wrote:

>>> But the language of Lisp macros is Lisp, that's really the whole
>>> point!  Without knowing TCL, I find it hard to see how you can
>>> introduce new constructs to the language *without* a macro language,
>>> even if that language is TCL.

>>Err.. What about Forth keywords?

>Tcl is a lot like Forth, too. :-)

That's how it initially stuck me -- Tcl looks like it was, at least in
concept, designed by someone attempting to apply Forth's looseness to a
language with a parser and a syntax.  This is something I've always been
curious about, and Tcl satisfied that curiosity -- I don't think it's a
good mix.

>You can do more interesting things in Lisp with readmacros,
>from what I'm told, including making it look like an entirely
>different language.  Tcl code always looks like Tcl code.

Now Lisp is a worthwhile improvement on Forth.  ;-)

(Yes, I know -- the ancestry goes the other way around: Lisp -> Scheme ->
Forth.)

>Will

-- 
-William "Billy" Tanksley



From news at dorb.com  Tue Jun 22 13:23:22 1999
From: news at dorb.com (Darrell)
Date: Tue, 22 Jun 1999 13:23:22 -0400
Subject: Python scope and parameter passing method
References: <Pine.LNX.3.96.990622173938.8191A-100000@alunos.uevora.pt>
Message-ID: <YIPb3.1885$U6.6756@newsr2.twcny.rr.com>

What your asking is more than a five minute explanation. And looks like a
home work assignment.

Check out the python docs, search for these topics with dejanews and check
out the FAQ at python.org

Don't forget books like "Learning Python" and others from O'REILLY.

Good luck.
--
--Darrell





From morse at harborcom.net  Wed Jun 16 23:30:57 1999
From: morse at harborcom.net (Kevin Dahlhausen)
Date: Wed, 16 Jun 1999 23:30:57 -0400
Subject: Undefined symbol when freezing?
Message-ID: <37686BF0.942532CF@harborcom.net>

I'm trying to freeze a simple app w/an extension created with Swig.
Freeze generates the code and it compiles fine, but at runtime I get the
following:

Traceback (innermost last):
  File "./hello.py", line 4, in ?
    from fltk import *
  File "fltk.py", line 1, in ?
  File "fltk_pre.py", line 2, in ?
ImportError: /usr/local/lib/libswigpy.so: undefined symbol: PyType_Type

I'm not sure where to start looking to solve this.
Does anyone have any ideas?





From olipt at mayo.edu  Wed Jun 16 16:41:06 1999
From: olipt at mayo.edu (Travis Oliphant)
Date: Wed, 16 Jun 1999 15:41:06 -0500
Subject: extention modules.
In-Reply-To: <99061620265802.00254@Vegus>
References: <99061620265802.00254@Vegus>
Message-ID: <Pine.LNX.4.10.9906161535170.5336-100000@us2.mayo.edu>

> Should I write it as two seperate modules (the compiled C one and
> the native python one) or is there a way to combine them in one. I am
> using the SWIG program to wrap my C library.

It is often the practice to put make a Python module that imports a C
module.  That's what I'm doing with Multipack
http://oliphant.netpedia.net/multipack.html
You can construct C-modules that don't need a Python wrapping, however, it
is easier to handle the user-end API in Python.

If you are doing math intensive calculations, have you looked at the
Numeric extensions?

Good luck,

Travis





From bernhard at alpha1.csd.uwm.edu  Sat Jun 19 19:37:12 1999
From: bernhard at alpha1.csd.uwm.edu (Bernhard Reiter)
Date: 19 Jun 1999 23:37:12 GMT
Subject: BiBTeX module?
References: <joe-160619991358444033@chinacat.salk.edu> <7k950a$hsb@cs.vu.nl> <slrn7miqcg.tpi.bernhard@alpha1.csd.uwm.edu> <slrn7mk9ui.dic.gobry@blanc.idiap.ch> <wk1zf9wo5y.fsf@ifi.uio.no>
Message-ID: <slrn7moad8.4l5.bernhard@alpha1.csd.uwm.edu>

On 18 Jun 1999 16:58:49 +0200, Lars Marius Garshol <larsga at ifi.uio.no> wrote:
>| I've also started such a project as free software. 

>BTW: an XML version of BibTeX would be very useful. 
What do you mean with XML version?

> People keep asking
>for it in XML/SGML newsgroups, and it just plain makes sense, since
>XML is much better suited for this sort of thing than TeX is.

I am not sure. 
Somebody has to write a good DTD first and then
XML doesn't have TeX's typesetting capabilities. 
Often I find myself using them to really typeset the References precisely.

If you want something like a .bib format in XML you might have to invent the
needed typesetting rules, too.

But, of course, if would be good, if the new program can handle different
Bibliographic databases.

	Bernhard


-- 
Research Assistant, Geog Dept UM-Milwaukee, USA.         (www.uwm.edu/~bernhard)
Funding/Promoting Association for a Free Informational Infrastructure (ffii.org)



From robin at jessikat.demon.co.uk  Thu Jun  3 07:04:17 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Thu, 3 Jun 1999 12:04:17 +0100
Subject: Threaded Tkinter
References: <m3btex4vxu.fsf@tiana.kli.prv>
Message-ID: <07u8dGAxEmV3EwoL@jessikat.demon.co.uk>

In article <m3btex4vxu.fsf at tiana.kli.prv>, Gerald Gutierrez
<gutier at intergate.bc.ca> writes
>
>Hi all.
>
>I have two questions and I'd really appreciate it if someone could
>help me sort it out.
>
>My first one -- I'm starting out with Tkinter. I wrote the code below,
>which will supposedly create a button with an image, that image
>changing every second. If I set the button to call swapIcons() when
>pressed, the image changes fine, but using the SpinThread, the icons
>don't change. Is this because Tkinter is not thread-safe and I cannot
>safely change the button's image property from within a thread other
>than the main one?
>
... code & Q2 elided
...
I tried this with 1.5.2 + win95 + local gifs and it worked fine for me.
Should I leave it running for a while?
-- 
Robin Becker



From andreas.schoerk at sap-ag.de  Mon Jun  7 07:43:46 1999
From: andreas.schoerk at sap-ag.de (Andreas Schoerk)
Date: Mon, 7 Jun 1999 13:43:46 +0200
Subject: Newbie-q: Debugging tkinter(and other GUI)-Programs
Message-ID: <7jgbd4$sse$1@mailusr.wdf.sap-ag.de>

Hello,
I have have problems using idle, pydebug.py or pythonwin-debugger for the
debugging of programs
which use tkinter-widgets in NT. Sometimes python hangs, sometimes it
doesn`t show any widgets of the
debugged program.
Is it right, that using pdb.py is the only suitable way to debug GUI-Progs ?
Perhaps, I encountered only configuration problems. Does anybody know how to
solve this.
I tried to solve this by myself by controling pdb like xxgdb controls gdb,
but I don?t want
to create a new debugger on my own.
Andreas





From bwarsaw at cnri.reston.va.us  Fri Jun  4 11:01:54 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Fri, 4 Jun 1999 11:01:54 -0400 (EDT)
Subject: "The Python Way"
References: <14167.3323.707498.300295@anthem.cnri.reston.va.us>
	<99060320192502.01312@quadra.teleo.net>
Message-ID: <14167.60002.783156.583193@anthem.cnri.reston.va.us>

>>>>> "PP" == Patrick Phalen <pphalen at teleo.net> writes:

    PP> For that matter, what makes me think that Guido and Timbot are
    PP> two different entities?

Well, the one time I was in Boston and had the chance to meet Timbot,
"he" feigned illness.  Guido and Gordon both claim to have met him,
but somehow I think they have vested interests in maintaining the
illusion of Tim-as-not-bot.  Way I see it, making the Timbot code Open
Source is the only moral thing to do.  Besides, wouldn't it be cool to
check out the Python translation of the old Emacs Lisp RMS AI code?

-Barry



From eugene.leitl at lrz.uni-muenchen.de  Fri Jun 11 04:07:40 1999
From: eugene.leitl at lrz.uni-muenchen.de (eugene.leitl at lrz.uni-muenchen.de)
Date: Fri, 11 Jun 1999 02:07:40 -0600
Subject: CCL:Announcing PyDaylight 0.5
Message-ID: <3760C3CC.9FF7A619@bioreason.com>

From: Andrew Dalke <dalke at bioreason.com>

PyDaylight 0.5

  Bioreason <www.bioreason.com>, a leading provider of
chemoinformatics knowledge discovery services. has developed an
interface between the Daylight toolkit and the Python programming
language and is making this package freely available under the
LGPL license.  Additional information is available from:
   http://starship.python.net/crew/dalke/PyDaylight/

and from my MUG'99 presentation at:
   http://www.daylight.com/meetings/mug99/

  The end result is that new chemical analysis algorithms can be
developed in a much shorter time and with fewer programming errors.



  Python is an interpreted, interactive, object-oriented programming
language with modules, classes, exceptions, very high level dynamic
data types, and dynamic typing.  It is a superb tool for rapid
application prototyping and development.  For more information, see
http://www.python.org/ .

  The Daylight toolkit is sold by Daylight Information Systems (see
http://www.daylight.com/).  The toolkit includes C/FORTRAN libraries
for manipulating molecular topologies, depicting the results,
conducting subgraph searches, and more.  (We don't get any money from
them for promoting their software :)


  The PyDaylight interface builds on Roger E. Critchlow Jr.'s DaySWIG
package but adds a true object-oriented interface layer above the
underlying C libraries as well as iterators, automatic garbage
collection, and exception handling.


  We develop all of our applications using PyDaylight and it has been
very stable and useful.  Since it has been an in-house project there
is very little end-user documentation or tutorials, which is why
it is labeled 0.5 instead of 1.0.  Thus, this release shouldn't be
used by someone who doesn't like reading source code to figure out
what's going on.

  However, I would like to know who might be interested in
introductory material, so if you plan on using the Daylight toolkit
and would like to try out the PyDaylight interface, please drop me a
message so I can better gauge what to work on next.

                                                Andrew Dalke
                                                dalke at bioreason.com
-= This is automatically added to each message by mailing script =-
CHEMISTRY at ccl.net -- To Everybody    |   CHEMISTRY-REQUEST at ccl.net -- To Admins
MAILSERV at ccl.net -- HELP CHEMISTRY or HELP SEARCH
CHEMISTRY-SEARCH at ccl.net -- archive search    |    Gopher: gopher.ccl.net 70
Ftp: ftp.ccl.net  |  WWW: http://www.ccl.net/chemistry/   | Jan: jkl at osc.edu








From jhouchin at texoma.net  Mon Jun 21 15:22:34 1999
From: jhouchin at texoma.net (Jimmie Houchin)
Date: Mon, 21 Jun 1999 19:22:34 GMT
Subject: IPwP books available, 99cents
References: <376a479b.131911984@news.texoma.net> <376e7e32.727764@news.demon.co.uk>
Message-ID: <376e8fa9.161102353@news.texoma.net>

I don't think you've misread Python's direction.

This bookstore is not normal bookstore. I don't know what there
sources are, but all of their books are 75% off minimally and none
over $4.99 regardless of original price.

I would suspect the average consumer in this bookstore has never heard
of Python so they end up on the 99 cent shelf simply because they
aren't moving.

Most of their computer books aren't worth buying. 
"Using Netscape 1.0", etc.... :)

However, IPwP has maintained it's value much better than most computer
books. :)

Jimmie

On Mon, 21 Jun 1999 18:03:53 GMT, andy at robanal.demon.co.uk (Andy
Robinson) wrote:

>jhouchin at texoma.net (Jimmie Houchin) wrote:
>
>>Hello,
>>
>>I was visiting a bookstore in Plano, TX yesterday and saw two new
>>copies of Internet Programming with Python on their shelves for $.99,
>>yes 99 cents.
>>
>Well, here in the U.K. they are appearing back on the shelves in the
>good bookstores after a long absence, and still at full price.  I
>really hope your bookstore has misread Python's direction and not
>mine!
>
>- Andy Robinson
>




From chalaoux at cybercable.fr  Sat Jun  5 17:42:23 1999
From: chalaoux at cybercable.fr (FR chalaoux)
Date: Sat, 05 Jun 1999 23:42:23 +0200
Subject: Looking for help to compile Python
Message-ID: <375999BE.38F30059@cybercable.fr>

Hi All,

I'm trying to compile Python 152 on a Intel machine with the standard
RH51 and I obtain this error (cf error make below). I joined the
config.log file in attachement. Some explanation about failed in
config.log are welcome ...

Thanks for help.

FR

error from make:
=============

gcc  -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H -c ./timemodule.c
./timemodule.c: In function `time_strptime':
./timemodule.c:429: warning: assignment makes pointer from integer
without a cast
gcc  -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H -c ./operator.c
gcc  -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H -c ./fcntlmodule.c
gcc  -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H -c ./pwdmodule.c
gcc  -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H -c ./grpmodule.c
gcc  -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H -c ./selectmodule.c
gcc  -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H -c ./socketmodule.c
./socketmodule.c: In function `initsocket':
./socketmodule.c:2212: `uint32_t' undeclared (first use this function)
./socketmodule.c:2212: (Each undeclared identifier is reported only once

./socketmodule.c:2212: for each function it appears in.)
./socketmodule.c:2212: parse error before `0xe0000000U'
./socketmodule.c:2217: parse error before `0xe0000001U'
./socketmodule.c:2222: parse error before `0xe00000ffU'
make[1]: *** [socketmodule.o] Error 1
make[1]: Leaving directory `/usr/local/Python-1.5.2/Modules'
make: *** [Modules] Error 2




--
FR Chalaoux
ADVL (Au Del? de Vos Limites)
chalaoux at cybercable.fr
http://www.advl.org/


-------------- next part --------------
This file contains any messages produced by compilers while
running configure, to aid debugging if configure makes a mistake.

configure:608: checking MACHDEP
configure:653: checking CCC
configure:669: checking for --without-gcc
configure:746: checking for gcc
configure:859: checking whether the C compiler (gcc  ) works
configure:875: gcc -o conftest    conftest.c  1>&5
configure:901: checking whether the C compiler (gcc  ) is a cross-compiler
configure:906: checking whether we are using GNU C
configure:915: gcc -E conftest.c
configure:934: checking whether gcc accepts -g
configure:997: checking LINKCC
configure:1014: checking LDLIBRARY
configure:1043: checking for ranlib
configure:1076: checking for ar
configure:1150: checking how to run the C preprocessor
configure:1171: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1230: checking for AIX
configure:1255: checking for minix/config.h
configure:1265: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1261: minix/config.h: No such file or directory
configure: failed program was:
#line 1260 "configure"
#include "confdefs.h"
#include <minix/config.h>
configure:1304: checking whether gcc accepts -OPT:Olimit=0
configure:1318: gcc -OPT:Olimit=0 -o conftest -g -O2   conftest.c  1>&5
cc1: Invalid option `-OPT:Olimit=0'
configure: failed program was:
#line 1314 "configure"
#include "confdefs.h"
int main() { return 0; }
configure:1338: checking whether gcc accepts -Olimit 1500
configure:1352: gcc -Olimit 1500 -o conftest -g -O2   conftest.c  1>&5
gcc: 1500: No such file or directory
cc1: Invalid option `-Olimit'
configure: failed program was:
#line 1348 "configure"
#include "confdefs.h"
int main() { return 0; }
configure:1375: checking for C preprocessor type
configure:1388: gcc -c -g -O2  conftest.c 1>&5
configure:1405: checking for ANSI C header files
configure:1418: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1485: gcc -o conftest -g -O2   conftest.c  1>&5
configure:1516: checking for dlfcn.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for fcntl.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for limits.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for locale.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for ncurses.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for pthread.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for signal.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for stdarg.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for stddef.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for stdlib.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for thread.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1522: thread.h: No such file or directory
configure: failed program was:
#line 1521 "configure"
#include "confdefs.h"
#include <thread.h>
configure:1516: checking for unistd.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for utime.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for sys/audioio.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1522: sys/audioio.h: No such file or directory
configure: failed program was:
#line 1521 "configure"
#include "confdefs.h"
#include <sys/audioio.h>
configure:1516: checking for sys/file.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for sys/lock.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1522: sys/lock.h: No such file or directory
configure: failed program was:
#line 1521 "configure"
#include "confdefs.h"
#include <sys/lock.h>
configure:1516: checking for sys/param.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for sys/select.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for sys/time.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for sys/times.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for sys/un.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for sys/utsname.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for sys/wait.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1557: checking for dirent.h that defines DIR
configure:1570: gcc -c -g -O2  conftest.c 1>&5
configure:1595: checking for opendir in -ldir
configure:1614: gcc -o conftest -g -O2   conftest.c -ldir   1>&5
ld: cannot open -ldir: No such file or directory
configure: failed program was:
#line 1603 "configure"
#include "confdefs.h"
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char opendir();

int main() {
opendir()
; return 0; }
configure:1681: checking for clock_t in time.h
configure:1710: checking for mode_t
configure:1743: checking for off_t
configure:1776: checking for pid_t
configure:1809: checking return type of signal handlers
configure:1831: gcc -c -g -O2  conftest.c 1>&5
configure:1850: checking for size_t
configure:1883: checking for uid_t in sys/types.h
configure:1918: checking size of int
configure:1937: gcc -o conftest -g -O2   conftest.c  1>&5
configure:1957: checking size of long
configure:1976: gcc -o conftest -g -O2   conftest.c  1>&5
configure:1996: checking size of void *
configure:2015: gcc -o conftest -g -O2   conftest.c  1>&5
configure:2036: checking for long long support
configure:2046: gcc -c -g -O2  conftest.c 1>&5
configure:2060: checking size of long long
configure:2079: gcc -o conftest -g -O2   conftest.c  1>&5
configure:2102: checking size of off_t
configure:2122: gcc -o conftest -g -O2   conftest.c  1>&5
configure:2144: checking whether to enable large file support
configure:2170: checking for --with-next-framework
configure:2187: checking for --with-dyld
configure:2213: checking SO
configure:2227: checking LDSHARED
configure:2267: checking CCSHARED
configure:2289: checking LINKFORSHARED
configure:2324: checking for dlopen in -ldl
configure:2343: gcc -o conftest -g -O2   conftest.c -ldl   1>&5
configure:2371: checking for shl_load in -ldld
configure:2390: gcc -o conftest -g -O2   conftest.c -ldld  -ldl  1>&5
ld: cannot open -ldld: No such file or directory
configure: failed program was:
#line 2379 "configure"
#include "confdefs.h"
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char shl_load();

int main() {
shl_load()
; return 0; }
configure:2424: checking for t_open in -lnsl
configure:2443: gcc -o conftest -g -O2   conftest.c -lnsl  -ldl  1>&5
/tmp/cca058591.o: In function `main':
/usr/local/Python-1.5.2/configure:2439: undefined reference to `t_open'
configure: failed program was:
#line 2432 "configure"
#include "confdefs.h"
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char t_open();

int main() {
t_open()
; return 0; }
configure:2464: checking for socket in -lsocket
configure:2483: gcc -o conftest -g -O2   conftest.c -lsocket -ldl  -ldl  1>&5
ld: cannot open -lsocket: No such file or directory
configure: failed program was:
#line 2472 "configure"
#include "confdefs.h"
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char socket();

int main() {
socket()
; return 0; }
configure:2504: checking for socket in -lnet
configure:2523: gcc -o conftest -g -O2   conftest.c -lnet -ldl  -ldl  1>&5
ld: cannot open -lnet: No such file or directory
configure: failed program was:
#line 2512 "configure"
#include "confdefs.h"
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char socket();

int main() {
socket()
; return 0; }
configure:2547: checking for --with-libs
configure:2561: checking for --with(out)-readline
configure:2576: checking for --with-dec-threads
configure:2592: checking for --with-threads
configure:2606: checking for --with-thread
configure:3113: checking for --with-sgi-dl
configure:3136: checking for --with-dl-dld
configure:3169: checking for alarm
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for chown
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for clock
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for dlopen
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for execv
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for flock
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for fork
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for fsync
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for fdatasync
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for ftime
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for ftruncate
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for getpeername
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for getpgrp
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for getpid
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for getpwent
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for gettimeofday
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for getwd
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca061601.o: In function `main':
/usr/local/Python-1.5.2/configure:3191: the `getwd' function is dangerous and should not be used.
configure:3169: checking for kill
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for link
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for lstat
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for mkfifo
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for mktime
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for nice
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for pause
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for plock
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca062881.o: In function `main':
/usr/local/Python-1.5.2/configure:3191: undefined reference to `plock'
configure: failed program was:
#line 3174 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char plock(); below.  */
#include <assert.h>
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char plock();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_plock) || defined (__stub___plock)
choke me
#else
plock();
#endif

; return 0; }
configure:3169: checking for pthread_init
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca063011.o: In function `main':
/usr/local/Python-1.5.2/configure:3191: undefined reference to `pthread_init'
configure: failed program was:
#line 3174 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char pthread_init(); below.  */
#include <assert.h>
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char pthread_init();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_pthread_init) || defined (__stub___pthread_init)
choke me
#else
pthread_init();
#endif

; return 0; }
configure:3169: checking for putenv
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for readlink
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for select
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for setgid
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for setlocale
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for setuid
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for setsid
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for setpgid
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for setpgrp
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for setvbuf
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for sigaction
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for siginterrupt
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for sigrelse
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca065061.o: In function `main':
/usr/local/Python-1.5.2/configure:3191: undefined reference to `sigrelse'
configure: failed program was:
#line 3174 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char sigrelse(); below.  */
#include <assert.h>
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char sigrelse();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_sigrelse) || defined (__stub___sigrelse)
choke me
#else
sigrelse();
#endif

; return 0; }
configure:3169: checking for strftime
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for strptime
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for symlink
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for tcgetpgrp
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for tcsetpgrp
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for timegm
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for times
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for truncate
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for uname
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for waitpid
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3226: checking for fseek64
configure:3254: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca066791.o: In function `main':
/usr/local/Python-1.5.2/configure:3248: undefined reference to `fseek64'
configure: failed program was:
#line 3231 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char fseek64(); below.  */
#include <assert.h>
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char fseek64();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_fseek64) || defined (__stub___fseek64)
choke me
#else
fseek64();
#endif

; return 0; }
configure:3226: checking for fseeko
configure:3254: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca066921.o: In function `main':
/usr/local/Python-1.5.2/configure:3248: undefined reference to `fseeko'
configure: failed program was:
#line 3231 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char fseeko(); below.  */
#include <assert.h>
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char fseeko();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_fseeko) || defined (__stub___fseeko)
choke me
#else
fseeko();
#endif

; return 0; }
configure:3226: checking for fstatvfs
configure:3254: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca067051.o: In function `main':
/usr/local/Python-1.5.2/configure:3248: undefined reference to `fstatvfs'
configure: failed program was:
#line 3231 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char fstatvfs(); below.  */
#include <assert.h>
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char fstatvfs();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_fstatvfs) || defined (__stub___fstatvfs)
choke me
#else
fstatvfs();
#endif

; return 0; }
configure:3226: checking for ftell64
configure:3254: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca067181.o: In function `main':
/usr/local/Python-1.5.2/configure:3248: undefined reference to `ftell64'
configure: failed program was:
#line 3231 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char ftell64(); below.  */
#include <assert.h>
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char ftell64();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_ftell64) || defined (__stub___ftell64)
choke me
#else
ftell64();
#endif

; return 0; }
configure:3226: checking for ftello
configure:3254: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca067311.o: In function `main':
/usr/local/Python-1.5.2/configure:3248: undefined reference to `ftello'
configure: failed program was:
#line 3231 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char ftello(); below.  */
#include <assert.h>
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char ftello();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_ftello) || defined (__stub___ftello)
choke me
#else
ftello();
#endif

; return 0; }
configure:3226: checking for statvfs
configure:3254: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca067441.o: In function `main':
/usr/local/Python-1.5.2/configure:3248: undefined reference to `statvfs'
configure: failed program was:
#line 3231 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char statvfs(); below.  */
#include <assert.h>
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char statvfs();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_statvfs) || defined (__stub___statvfs)
choke me
#else
statvfs();
#endif

; return 0; }
configure:3282: checking for dup2
configure:3310: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3282: checking for getcwd
configure:3310: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3282: checking for strdup
configure:3310: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3282: checking for strerror
configure:3310: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3282: checking for memmove
configure:3310: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3337: checking for getpgrp
configure:3387: gcc -c -g -O2  conftest.c 1>&5
configure: In function `main':
configure:3383: too many arguments to function `getpgrp'
configure: failed program was:
#line 3380 "configure"
#include "confdefs.h"
#include <unistd.h>
int main() {
getpgrp(0);
; return 0; }
configure:3403: checking for setpgrp
configure:3453: gcc -c -g -O2  conftest.c 1>&5
configure: In function `main':
configure:3449: too many arguments to function `setpgrp'
configure: failed program was:
#line 3446 "configure"
#include "confdefs.h"
#include <unistd.h>
int main() {
setpgrp(0,0);
; return 0; }
configure:3469: checking for gettimeofday
configure:3519: gcc -c -g -O2  conftest.c 1>&5
configure:3538: checking whether time.h and sys/time.h may both be included
configure:3552: gcc -c -g -O2  conftest.c 1>&5
configure:3573: checking whether struct tm is in sys/time.h or time.h
configure:3586: gcc -c -g -O2  conftest.c 1>&5
configure:3607: checking for tm_zone in struct tm
configure:3620: gcc -c -g -O2  conftest.c 1>&5
configure:3678: checking for time.h that defines altzone
configure:3690: gcc -c -g -O2  conftest.c 1>&5
configure: In function `main':
configure:3686: `altzone' undeclared (first use this function)
configure:3686: (Each undeclared identifier is reported only once
configure:3686: for each function it appears in.)
configure: failed program was:
#line 3683 "configure"
#include "confdefs.h"
#include <time.h>
int main() {
return altzone;
; return 0; }
configure:3712: checking whether sys/select.h and sys/time.h may both be included
configure:3725: gcc -c -g -O2  conftest.c 1>&5
configure:3741: checking whether char is unsigned
configure:3805: checking for working const
configure:3859: gcc -c -g -O2  conftest.c 1>&5
configure:3882: checking for working volatile
configure:3891: gcc -c -g -O2  conftest.c 1>&5
configure:3908: checking for working signed char
configure:3917: gcc -c -g -O2  conftest.c 1>&5
configure:3934: checking for prototypes
configure:3943: gcc -c -g -O2  conftest.c 1>&5
configure:3958: checking for variable length prototypes and stdarg.h
configure:3977: gcc -c -g -O2  conftest.c 1>&5
configure:3993: checking for bad exec* prototypes
configure:4002: gcc -c -g -O2  conftest.c 1>&5
configure:4019: checking for bad static forward
configure:4040: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:4059: checking whether va_list is an array
configure:4074: gcc -c -g -O2  conftest.c 1>&5
configure:4090: checking for gethostbyname_r
configure:4118: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:4138: checking gethostbyname_r with 6 args
configure:4159: gcc -c -g -O2     conftest.c 1>&5
configure:4333: checking for __fpu_control in -lieee
configure:4352: gcc -o conftest -g -O2   conftest.c -lieee  -ldl  1>&5
configure:4382: checking for --with-fpectl
configure:4405: checking for --with-libm=STRING
configure:4426: checking for --with-libc=STRING
configure:4450: checking for hypot
configure:4478: gcc -o conftest -g -O2   conftest.c -lieee -ldl  -lm 1>&5
configure:4505: checking for hypot
configure:4563: checking for genuine getopt
configure:4583: gcc -o conftest -g -O2   conftest.c -lieee -ldl  1>&5
configure: failed program was:
#line 4571 "configure"
#include "confdefs.h"
#include <stdio.h>
extern int optind, opterr, getopt();
extern char* optarg;
int main() {
	char* av[] = { "testprog", "parameter", "-fFlag", NULL };
	opterr = 0;
	if (getopt(3, av, "f:") == 'f') { exit(1); }
	exit(0);
}
configure:4601: checking what malloc(0) returns
configure:4628: gcc -o conftest -g -O2   conftest.c -lieee -ldl  1>&5

From mark_removethis_ at _removethis_btweng.krakow.pl  Thu Jun 10 13:59:39 1999
From: mark_removethis_ at _removethis_btweng.krakow.pl (MK)
Date: Thu, 10 Jun 1999 17:59:39 GMT
Subject: dict vs kjBuckets vs ???
References: <000e01beb21f$a4c5a860$2fa22299@tim>
Message-ID: <375ffb1d.26189164@news.cyf-kr.edu.pl>

On Tue, 8 Jun 1999 22:27:39 -0400, "Tim Peters"
<tim_one at email.msn.com> wrote:
>>  A million?

>No.

>It all depends on how much memory you have and how well the hash function is
>working.  Note that dicts work hard to "randomize" the addresses at which
>they store keys, so memory locality can be terrible.  But if there's no
>pattern to customer ID accesses, locality would be terrible in a 1M array
>too.

In some book on algorithms I've read that after inserting limited
number of items performance of operating on hash tables 
drops dramatically. I plan to write a program that would store
lots (in range of 10M or even more) of relatively small objects
(a few hundred bytes at most), so what do you think I should use? I
thought about dictionaries, kjBuckets, or maybe even library called
Metakit for Python
(http://www.equi4.com/metakit/info/README-Python.html).

what-do-you-think-ly y'rs



--------------------------------------------------
Reality is something that does not disappear after
you cease believing in it - VALIS, Philip K. Dick
--------------------------------------------------

Delete _removethis_ from address to email me



From MHammond at skippinet.com.au  Sat Jun 19 05:07:56 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Sat, 19 Jun 1999 19:07:56 +1000
Subject: NT
References: <376A6D64.DF881D27@phoenix-edv.netzservice.de>
Message-ID: <7kfmj5$ksm$1@m2.c2.telstra-mm.net.au>

Holger Jannsen wrote in message
<376A6D64.DF881D27 at phoenix-edv.netzservice.de>...

>C:\myPython>installation15.py /? |more
>Traceback (innermost last):
...

>If I start with 'C:\myPython>python installation15.py /? |more'
>there's no problem and it works allright.

This is a known bug with the NT command processor.  Every application has
this problem, not just Python.

Mark.






From schorsch at schorsch.com  Thu Jun 17 17:44:11 1999
From: schorsch at schorsch.com (Georg Mischler)
Date: Thu, 17 Jun 1999 21:44:11 GMT
Subject: do anonymous pipes normally work on NT?
References: <4D0A23B3F74DD111ACCD00805F31D8100DB90DE4@RED-MSG-50>
Message-ID: <7kbq78$vgm$1@nnrp1.deja.com>

Bill Tutt wrote:
>
> Patch info:
>
> Alter the PyHANDLE_Check() branch of
win32processmodule_win32.cpp:sethandle
> to:
> 		*pobHandle = v;
> 		if (PyWinObject_AsHANDLE(v, ph))
> 		{
> 			Py_INCREF(v);
> 		}
> 		else
> 		{
> 			rc = -1;
> 		}
> The cast that was there before was setting *ph equal to the
> Python reference count on v.


Ok, I'll have to build this stuff myself some day anyway...

But then, I just downloaded win32_src.zip and Pythonwin_src.zip
and didn't find anything resembling a win32processmodule in there.
Guess I am looking in the wrong place. Where is this module defined?

Thanks a lot

-schorsch

--
Georg Mischler  --  simulation developper  --  schorsch at schorsch.com
+schorsch.com+  --  lighting design tools  --  http://www.schorsch.com/


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From paul at prescod.net  Tue Jun  1 10:41:37 1999
From: paul at prescod.net (Paul Prescod)
Date: Tue, 1 Jun 1999 14:41:37 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
		<7imr6b$1s0$1@nnrp1.deja.com>
	 <87pv3iq6ss.fsf@ev.netlab.co.jp>
		<3752C76D.604CD474@prescod.net>
	 <87n1ykr0z9.fsf@ev.netlab.co.jp>
		<375372F6.A874554@prescod.net> <87hfosquc0.fsf@ev.netlab.co.jp>
Message-ID: <3753F121.E4A9BE8B@prescod.net>

Yukihiro Matsumoto wrote:
> 
> |Is there a language with real GC where destructors are called as
> |predictably as they are in Python?
> 
> I said nothing about the destructor invocation predictability.

I consider that a part of the definition of destructor.

> I don't understand and really interested in why you guys require
> to predict destructor invocation so much, but it's another story.

What if I am holding GUI "device context" objects and the GUI has a
limited number of them. I don't think that there is a way to tell the GC
that there is a limited number of them. It doesn't know that device
contexts are a limited resource, just as memory is.

> |Speed wasn't my concern. I asked about references in C code. Do we use
> |PyObject **s instead of PyObject *s now?
> 
> Sorry, I don't understand.  Do we have to use PyObject **s instead of
> PyObject *s, if we choose real GC?

Imagine I have some C code. It stores away a pointer to a Python object:

PyObject *myObject=someObject;

Now a compacting GC moves someObject. My pointer is out of date, right? If
I want to be able to move someObject around in a mark and sweep style
operation then I need the pointer I hold to be static.

What GC algorithm does Ruby use and how much overhead does it add?

> Second, do you think that optimization for basic users compensates the
> risk of potential memory leak, by cyclic structures or missing DECREF
> in extension modules?  Do you mean it's easy for experts to find
> cyclic structure and cut their reference to destruct?  For me, it's
> not.  I hate that.  Is this because I'm not a Python expert?

It's pretty much a toss-up for me. In general I prefer optimizations that
make the language easy for new users to those that are for complex, long
running programs but full garbage collection would make complex programs a
fair bit easier.

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for only himself
 http://itrc.uwaterloo.ca/~papresco

"Silence," wrote Melville, "is the only Voice of God." The assertion,
like its subject, cuts both ways, negating and affirming, implying both
absence and presence, offering us a choice; it's a line that the Society
of American Atheists could put on its letterhead and the Society of
Friends could silently endorse while waiting to be moved by the spirit
to speak. - Listening for Silence by Mark Slouka, Apr. 1999, Harper's



From shunting at goSPS.com  Wed Jun  2 20:39:43 1999
From: shunting at goSPS.com (Sam Hunting)
Date: Wed, 2 Jun 1999 20:39:43 -0400
Subject: Newbie seeks help with COM? DLLs?
Message-ID: <7j4iiv$l1r$1@autumn.news.rcn.net>

Background: I'm operating in a VB/NT shop, but I want to introduce a Python
solution for the sake of doing *extremely* rapid development of GUI widgets.

(When you have a layout manager as in Tkinter, you don't spend a lot of time
manually tweaking layouts in VB's IDE. In fact, you don't even need to worry
about entering coordinate points, so it would be very easy to create a
data-(XML) driven widget factory.)

We need to incorporate the widgets into an existing software product so
that, for example, a Python-developed dialog will appear, into which the
user will enter data, which is then passed back to the application.

Now here is where my newbie nature takes over, since I am very new to
Python, and even newer to the VB/NT/Windows worlds -- how "incorporate"?

The software product understands DLLs, and also understands COM. I've had it
suggested to me that the way to get where I need to do, is to turn the
dialogs into COM servers, and then call them from the software application,
which is the client.

(1) I worked through the "Quick Start to Server Side COM and Python", and
registered the Hello World server successfully. Can anyone point me to some
VB code that would enable me to test the test server?

(2) If the COM approach is not the right approach, can someone suggest a
better one? And point me to some sample code that I can test, run, and
improve to help me solve the problem?

I thought the hard part would be learning enough of Python to construct the
widgets -- and now the widgets are constructed, and even some factory
machinery. But the hard part turned out to be the easy part.

Thanks in advance for any help you can give.

Sam Hunting





From faria at ppgia.pucpr.br  Tue Jun 22 14:31:31 1999
From: faria at ppgia.pucpr.br (faria)
Date: Tue, 22 Jun 1999 15:31:31 -0300
Subject: Unbound method
Message-ID: <001701bebcdd$73d3b080$656211c8@miro.lami.pucpr.br>

Hi everyone!!!

    I'm new in Python and in Brasil we don't wave any book of Python (and
it's very expensive to import!), so I have to ask you all this boring
questions!

    Why it doens't work ?

--- test.py
|
|class A:
|    a = 'Greetings from Brasil!!!'
|    def pa(self):
|        print self.a


Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> from test import *
>>> x = A
>>> x.a
'Greetings from Brasil!!!'
>>> x.pa
<unbound method A.pa>
>>>



Henrique Faria
Desenvolvedor Multim?dia
LAMI - Laborat?rio de M?dias Interativas Puc-PR
faria at ppgia.pucpr.br

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Henrique Faria.vcf
Type: text/x-vcard
Size: 436 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/19990622/f7400051/attachment.vcf>

From claird at Starbase.NeoSoft.COM  Thu Jun 10 16:03:05 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 10 Jun 1999 15:03:05 -0500
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <mtU73.2731$nn.808830@news.shore.net>
Message-ID: <7jp5lp$jo1$1@Starbase.NeoSoft.COM>

In article <mtU73.2731$nn.808830 at news.shore.net>,
Michael P. Reilly <arcege at shore.net> wrote:
			.
			.
			.
>I believe that current releases of Tcl can handle null bytes, but I
>have not verified that.
Yes.  Tcl8.0 and after handle null bytes.
			.
			.
			.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From wtanksle at dolphin.openprojects.net  Wed Jun 30 13:51:57 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Wed, 30 Jun 1999 17:51:57 GMT
Subject: Language extensibility (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <37682DD3.43923243@compaq.com> <7ka2be$qja$1@nntp6.u.washington.edu> <37776FFA.2D85@mailserver.hursley.ibm.com> <lw7looxx7m.fsf@copernico.parades.rm.cnr.it> <876748mcxg.fsf@home.ivm.de> <37788BB9.ABD@mailserver.hursley.ibm.com> <lw67471fyi.fsf@copernico.parades.rm.cnr.it> <3779ED0F.2C67@mailserver.hursley.ibm.com> <nkjzp1hizvb.fsf@tfeb.org> <377A3383.D453468@iname.com>
Message-ID: <slrn7nkm9t.579.wtanksle@dolphin.openprojects.net>

On Wed, 30 Jun 1999 17:10:59 +0200, Fernando Mato Mira wrote:
>Tim Bradshaw wrote:

>> But the language of Lisp macros is Lisp, that's really the whole
>> point!  Without knowing TCL, I find it hard to see how you can
>> introduce new constructs to the language *without* a macro language,
>> even if that language is TCL.

>Err.. What about Forth keywords?

Forth doesn't have keywords, so I'm guessing you mean immediate words.
They're written in the same language, but they tend to use words that
wouldn't be used otherwise (source access and such).  So they really are a
seperate language; after all, Forth is all about building new languages.

Another possible meaning of 'keyword' might be creating words.  Those are
cool, and don't require or generally use new vocabulary, but do require
grasping some pretty new concepts -- compile-time and runtime take on new
meanings.

-- 
-William "Billy" Tanksley



From mwh21 at cam.ac.uk  Tue Jun 15 21:21:51 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 16 Jun 1999 02:21:51 +0100
Subject: ANN: Stackless Python 0.2
References: <37628EAA.C682F16C@appliedbiometrics.com> <7k4fd4$c04$1@brick.cswv.com> <376636EA.84240129@appliedbiometrics.com> <7k6ttr$i5g$1@brick.cswv.com>
Message-ID: <m3909lszww.fsf@atrus.jesus.cam.ac.uk>

neelk at brick.cswv.com (Neel Krishnaswami) writes:

> In article <376636EA.84240129 at appliedbiometrics.com>,
> Christian Tismer  <tismer at appliedbiometrics.com> wrote:
> >
> >
> >Neel Krishnaswami wrote:
> >
> >[Stackless Python 0.2]
> >
> >> This is very neat, and you are completely deranged. I know just
> >> enough to know that I should cheer you on, mind, but I'll try to
> >> cheer loudly. Feel free to make an announce when you add coroutine
> >> support. Please? :)
> >
> >Thanks :-)
> >
> >Sure I will. I'd like to know wether my binary works ok for you.
> >You might also try a recursive function, it should raise an
> >exception after 29999 recursions (just to have *a* limit).
> 
> Unfortunately, I couldn't get it to build (I'm on a Linux machine).
> I tried copying the files from the stackless distribution into the
> appropriate places in the Python source distribution, but I get 
> the following error:
> 
> $ gcc -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H   -c myreadline.c -o myreadline.o
> In file included from ../Include/Python.h:38,
>                  from myreadline.c:42:
> ../Include/patchlevel.h:69: parse error before `5'
> ../Include/patchlevel.h:69: stray '\' in program
> ../Include/patchlevel.h:70: stray '\' in program
> ../Include/patchlevel.h:71: stray '\' in program
> 
> [plus more cascading errors as a result of this]
> 
> Apparently gcc is not liking this:
> 
> /* Version as a single 4-byte hex number, e.g. 0x010502B2 == 1.5.2b2.
>    Use this for numeric comparisons, e.g. #if PY_VERSION_HEX >= ... */
> #define PY_VERSION_HEX ((PY_MAJOR_VERSION << 24) | \   
> 			(PY_MINOR_VERSION << 16) | \   
> 			(PY_MICRO_VERSION <<  8) | \
> 			(PY_RELEASE_LEVEL <<  4) | \
> 			(PY_RELEASE_SERIAL << 0))
> 
> The '5' in the "parse error before '5'" is the PY_MICRO_VERSION; the
> macro looks ok to me, but I haven't used C to any great extent in
> literally years. Am I making some very naive mistake? 

Ah, got this one; Naughty Chris's archive contains files with CRLF
line endings. This doesn't bother gcc in most places, except where a
backslash is used to continue a line, when it gets the screaming
heebie-jeebies. Eliminate the nasty ^M's and it all gets better. Also
diff stops reporting that every single line in the file as changed,
which was confusing me.

It doesn't seem to link yet. Oh well, it's late and I'm going to bed.

I can send you a diff against cvs's latest if that would help.

Yours,
Michael
 
> >I'm just undecided on the design. There are a couple of
> >different coroutine interfaces. You find them for instance in
> >Scheme, in Modula II, some Oberons, Icon, and in Tim Peter's
> >example which I added to my stackless archive.
> >Do you have a proposal?
> 
> Not really; all I know is a little Scheme, and that part of my brain
> is telling me that "coroutines are just closures over continuations." 
> Fortunately, I'm still able to realize that that's probably the
> wrong interface. :)
> 
> 
> Neel



From thstr at serop.abb.se  Thu Jun 17 02:50:42 1999
From: thstr at serop.abb.se (Thomas S. Strinnhed)
Date: Thu, 17 Jun 1999 08:50:42 +0200
Subject: GNU python? (GNUbe question)
References: <3765B8B4.18DE0257@spots.ab.ca> <37667309.F9BBAE9A@mojam.com> <022c01beb7ec$f9617070$f29b12c2@pythonware.com>
Message-ID: <37689AC2.B8F6579A@serop.abb.se>

Hi

Fredrik Lundh wrote:
> 
> Skip Montanaro <skip at mojam.com> wrote:
> > > Looking to try out python... is there a GNU python for Win9x?  and a
> > > tutorial maybe?
> >
> > Not sure just how a GNU Python would differ from the real thing.
> 
> a bit more like GNUscript, I suppose:
> 
> http://art.net/Studios/Hackers/Hopkins/Don/text/gnuscript.html
> 
> </F>

Sounds like a good thing this GNUscript, it'll do almost anything.
Is there a download site ;-)

/ The GNUbe



From tuna at delirious.fsnnet  Mon Jun 21 03:12:28 1999
From: tuna at delirious.fsnnet (Lloyd Weehuizen)
Date: 21 Jun 1999 07:12:28 GMT
Subject: Linking an Interpreter dynamically
Message-ID: <slrn7mrpes.94.tuna@delirious.fsnnet>

Hi!

I was just wondering if it is possible to create a dynamically linkable 
python interpreter? I've looked in the docs, but can't seem to find
anything, and "make sharedinstall" doesn't seem to give me any linkable
libraries??

Any help appreciated.
Thanks
Lloyd
Lloyd




From dave_kirby at eur.3com.com  Mon Jun 21 11:51:12 1999
From: dave_kirby at eur.3com.com (Dave Kirby)
Date: Mon, 21 Jun 1999 15:51:12 GMT
Subject: wierdness with pythonwin dialogs, threads & refcounts
References: <1282457297-115891134@hypernet.com>
Message-ID: <376e53a0.24292290@nnrp.uk.insnet.net>

On Thu, 17 Jun 1999 19:15:42 -0500, "Gordon McMillan" >
>I played with this enough to find that he's onto something. Without
>any threading, the dialog gets a WM_DESTROY and goes thru __del__.
>When executed from a threading.Thread, it never goes thru OnDestroy.
>I tried doing DoModal myself, then un-hooking the messages myself.
>No luck. Perhaps because there's no parent window, and the trhead
>does not have a Windows message queue?

Yep, I have got it to work correctly by replacing the
thread.start_new_thread function with my own version using the
WinThread class:

#------- 8< -----------------------------
from pywin.mfc.thread import WinThread

class MyThread(WinThread):
  def __init__(self, fn, args ):
    WinThread.__init__(self)
    self.fn = fn
    self.args = args
    
  def InitInstance(self):
    apply( self.fn, self.args )    
    self.PumpIdle()
    self.close()

def start_win_thread( fn, args ):
    thread = MyThread( fn, args )
    thread.CreateThread()
#------- 8< -----------------------------

I am still climbing the learning cliff of Win32 threads so I dont know
if it is the approved way of doing it, but it seems to work well
enough for my purposes.

	Dave K




From tim_one at email.msn.com  Fri Jun 25 03:12:38 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 25 Jun 1999 03:12:38 -0400
Subject: total idiot question: +=, .=, etc...
In-Reply-To: <slrn7n66t8.rh7.behrends@allegro.cse.msu.edu>
Message-ID: <000d01bebeda$1b7430a0$d29e2299@tim>

[Reimer Behrends]
> While I'd be the first to agree that C++ is not exactly the epitome of
> language design, I would think that at least the local variables should
> be easy to identify. The only real problem is to distinguish global
> variables from instance variables. And this is made worse in C++ by
> spreading all the information related to a class over more than one
> file. It should be noted that other languages do not necessarily have
> the same problem and that careful design can easily avoid it.

But even with declarations up the wazoo, distinguishing local from instance
vars is a major problem in real-life C++:  any OO language with declarations
has to allow local vars to shadow instance vars, else method implementations
can be broken by trivial changes in far-removed superclasses (e.g., Java
flatly prohibits shadowing in most cases -- but specifically allows this
one).

Any reasonably careful programmer will avoid such shadowing when the code is
first written, but under time & modification that never lasts; perhaps just
because the names most natural to a problem domain are limited.

So I've never worked on a (successful <0.9 wink>) C++ project where coding
conventions didn't require prefixing instance vars with some magical
sequence reserved for that purpose.  The one I'm working on now requires
member var names to match the case-sensitive regexp

    ^m[A-Z]\w*

and similarly starting with "g" for extern data and "s" for file-static data
(local names are "anything else").  Other projects enforced other
conventions -- it's a real relief to come back to Python and see it always
spelled "self."!

> But I do not think that the most basic operations of an OO language
> should be complicated by cumbersome syntax (this includes instance
> variable access and local method calls). The language should encourage
> use of these elements, not get in your way.

Readability is more important to me over the long haul.  Say what you like,
but which vars belong to the instance is screamingly obvious the instant you
see

    self.x = self.y + k

while in

    x = y + k

it's at best a minor puzzle and at worst a long & tedious grep project.

explicitly-favoring-explict-ly y'rs  - tim





From akuchlin at mems-exchange.org  Thu Jun 17 11:36:42 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: Thu, 17 Jun 1999 11:36:42 -0400 (EDT)
Subject: New release of mmapfile module
Message-ID: <199906171536.LAA03274@amarok.cnri.reston.va.us>

A new release of the mmapfile module is available from:

  ftp://starship.python.net/pub/crew/amk/new/mmapfilemodule.c

A simple test program is at:
  ftp://starship.python.net/pub/crew/amk/new/mmapdemo.py

The module is based on Sam Rushing's code for Win32, and has been
hacked to work on Unix as well.  It allows treating a chunk of a file
as a sequence object that behaves mostly like a mutable string; for
example, you can open a file, map a large chunk of it, and then use
the string and re modules to search through it. 

It's been suggested that mmapfile be added to the distribution, and
before doing that I'd like to get the portability kinks ironed out --
hence this new release.  Please try it and let me know if there are
compilation problems on your favorite Unix variant.  If other OSes
such as MacOS, BeOS, or whatever, also support memory-mapped files
through another mechanism, patches would be welcome.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
All I want is the answer to one simple question before I run screaming back to
the bughouse. Is this real or isn't it?
    -- Cliff Steele in DOOM PATROL #21




From gmcm at hypernet.com  Sat Jun 12 00:23:53 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 11 Jun 1999 23:23:53 -0500
Subject: Embedding Question from a Newbie
In-Reply-To: <00b801beb43d$b3efe240$a515bfd1@EVILDEAD>
Message-ID: <1282960805-85602773@hypernet.com>

CyRiuS writes:
> i have a question. i am new to python and am trying to use it as a
> scripting language for a game i am working on. I have the
> interpreter embedded and running scripts but i have a few questions.
> 
> 1) can i return values from the python script to the C/C++ program?
> for example say at the end of my script, it returned 'character' and
> character was equal to say 'hi' now is there anyway my C/C++ program
> could find out that the script just returned the value 'hi'?
> 
> 2) is there anyway I can pass values from the C/C++ program into the
> python script?
> 
> and
> 
> 3) say i had a python script that had several functions (defs). is
> there anyway i could call just one of these defs from the C/C++
> program without calling the entire script?

The answer to all 3 is yes. The details will be found in the 
embedding docs and the C API docs. I'd guess so far you've only used 
the Very High Level layer (PyRun_ etc). You want to start exploring 
the Abstract layer, where you work with PySequences and such. At that 
level it's still pretty straight forward. Drop down further only if 
you really have to. Almost anything you'll need to do is done 
somewhere in the Modules or Objects subdirectories.

grep-is-your-best-friend-ly y'rs

- Gordon



From bmcd at es.co.nz  Mon Jun 21 17:45:59 1999
From: bmcd at es.co.nz (Ben Caradoc-Davies)
Date: 21 Jun 1999 21:45:59 GMT
Subject: chomp()
References: <376EA9E4.5B3D38EC@starvision.com>
Message-ID: <slrn7mtcst.je.bmcd@ripley.sulaco.net>

On Mon, 21 Jun 1999 14:08:52 -0700, Angus MacKay <amackay at starvision.com> wrote:
>is there an equivilent of perl's chomp() in python or
>do I have to write the function myself?

string.rstrip(s) strips all trailing whitespace characters from s. If you
really, really, want your trailing whitespace, try

if s[-1]=='\n': s = s[:-1]

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



From gherman at my-deja.com  Tue Jun 15 10:23:08 1999
From: gherman at my-deja.com (Dinu C. Gherman)
Date: Tue, 15 Jun 1999 14:23:08 GMT
Subject: Python Paterns
Message-ID: <7k5nk0$nc0$1@nnrp1.deja.com>

jblake at eskimo.com (Jonathon) posted the following about
a week ago:

> Subject: Python Paterns
> Date: 9 Jun 1999 23:35:22 GMT
> From: jblake at eskimo.com (Jonathon)
> Organization: Pi Psi
> Newsgroups: comp.lang.python
>
>
> Searching Python.org, I didn't come across
> anything _recent_ about
> patterns in Python.  Did that SIG die, get swallowed up
> or did I miss their home?
>
> The only documentation I could find was
> http://starship.python.net/crew/gandalf/DNET/classes/index.html
> which was for Python 1.4 on an NT machine, last updated in
> 1997 and a paper given at the sixth conference on Python.


I tried to reply to this before using Deja.com, but somehow
Deja.com was unable to locate this posting... Let me try again.

Well, I guess the question can be answered like this: the SIG
died as you correctly noticed, the reasons being a lack of
contributions as well as an ill-defined (or at least vague)
purpose of the SIG itself.

Occasionally, there are pattern-related postings to c.l.p.
but it's rather hard to locate them later, especially if
the authors mis-spell crucial words inside... ;-)

Cheers,

Dinu

---
Dinu C. Gherman       :  Women inspire us to great things and
                      :  prevent us from achieving them. (Dumas)
................................................................
LHS International AG  :  http://www.lhsgroup.com
8050 Zurich           :  http://www.zurich.ch
Switzerland           :  http://pgp.ai.mit.edu
                      :  mobile://49.172.3060751
......................:.........................................
Vote against SPAM on http://www.politik-digital.de/spam


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From Klaus.Schilling at home.ivm.de  Wed Jun 16 11:34:20 1999
From: Klaus.Schilling at home.ivm.de (Klaus Schilling)
Date: 16 Jun 1999 17:34:20 +0200
Subject: OO (was: Why is tcl broken?)
References: <000801beb745$02768220$979e2299@tim> <871zfdp73f.fsf@home.ivm.de> <wklndk8ly6.fsf@ifi.uio.no>
Message-ID: <873dzsma6b.fsf@home.ivm.de>

Lars Marius Garshol <larsga at ifi.uio.no> writes:

> * Klaus Schilling
> | 
> | Steele just rediscovered Scheme.
> 
> What, do you mean recently? 

No, in the sense `Steele merely rediscovered scheme, not bring it
into being'

Klaus Schilling



From scott at chronis.pobox.com  Wed Jun  2 21:15:49 1999
From: scott at chronis.pobox.com (scott cotton)
Date: Wed, 2 Jun 1999 21:15:49 -0400
Subject: metaclasses and setting __class__
In-Reply-To: <1283761142-37459855@hypernet.com>; from Gordon McMillan on Wed, Jun 02, 1999 at 05:05:01PM -0500
References: <19990602133443.A3846@chronis.pobox.com> <1283761142-37459855@hypernet.com>
Message-ID: <19990602211549.A5987@chronis.pobox.com>

On Wed, Jun 02, 1999 at 05:05:01PM -0500, Gordon McMillan wrote:
| scott cotton writes:
| 
| > Last night i was playing with meta classes, and found out
| > that we are not able to assign an instance's __class__
| > attribute to a metaclass instance.  was just wondering if
| > this is probably always going to be the case, or if there
| > are any plans to make .__class__ assignable to such beasts.
| 
| It wouldn't do any good. By the time you have an instance, it's too 
| late to add metaclass behaviors. All the magic happens at the time 
| the "class MyClass(MyMetaClassInstance):" is encountered. At this 
| point, MyMetaClassInstance can do all kinds of evil things to the way 
| MyClass works.

It seems like all this could happen when __class__ is
assigned to as well, provided the underlying mechanism knew
what to do -- seems like somehow checking interface rather
than type for that assignment would do the trick, but all
that is conjecture - i don't even know if it's possible - or
desirable but it did seem intuitive.

| Assigning to an instance's __class__ will alter the instance's 
| behavior and appearance, but not to the same extent that using a 
| metaclass will. 

challenge: make a metaclass that mucks with it's own class
hierarchy at run time on a per-instance basis.  for example:

class A(Meta): pass
class B(A): pass
class C(A): pass

have instances of A become instances of C on instantiation,
and instances of B become instances of C with B replaced for
A in all the __bases__ (and all __bases__ of all __bases__
...) of C.   Never mind why it might be considered useful,
or even worse why it could actually make things more clear. 

I think I found that metaclasses are the wrong tool for
this, but only after hours of playing with the madness and
nearing a state of raving lunacy. at one point early in the
game, the following:

class MyRealClass(MetaClassInstance): pass
print MyRealClass, type(MyRealClass)

yielded "__getitem__ <'string'>".

I swear - and no I didn't redefine the builtin 'type'.

needless to say, i gave up after a while, and used the
following function instead:

def rebuild(leaf_class, oldbase, newbase):
    if leaf_class == oldbase:
        return newbase
    class foo: pass
    oldbases = leaf_class.__bases__
    newbases = []
    for base in oldbases:
        newbases.append(rebuild(base, oldbase, newbase))
    foo.__bases__ = tuple(newbases)
    for k, v in leaf_class.__dict__.items():
        if k != '__bases__':
            foo.__dict__[k] = v
    return foo

It was used sortof like this:

class A:

    def __init__(self):
        if self.__class__ != A:
            self.__class__ = rebuild(C, A, B)
	else:
            self.__class__ = C
class B(A):

      def extrastuff(self): ...

class C(A): 
      
      def other_type_of_extrastuff(self): ...

scott




From michaelb at gold.net.au  Fri Jun 11 06:39:53 1999
From: michaelb at gold.net.au (Michael Bell)
Date: Fri, 11 Jun 1999 10:39:53 GMT
Subject: Scripting Telnet
References: <a6U73.1265$VL2.195835@news.direcpc.com> <37606A0D.6332A9C2@horvath.com> <7jpv0q$90v$1@Starbase.NeoSoft.COM>
Message-ID: <19990611.10395300@belfry>

Once upon a time there was a Python solution called PIPE (for Python 
Interactive Process Enabler), put together by John Croix. I've got 
version 0.1, but it's dated 1997.

Did this project go anywhere?  I did a quick search of python.org but 
didn't find anything.

On 6/11/99, 11:15:38 AM, claird at Starbase.NeoSoft.COM (Cameron Laird) 
wrote regarding Re: Scripting Telnet:

> Bob Horvath  <bob at horvath.com> wrote:
>                       .
> >Well, the tcl based expect is probably the most appropriate tool for 
such a
> >thing, but there is a pythonized version(s) of it as well.  The one I 
can
> >think of is ExpectPy, 
(http://www1.shore.net/~arcege/python/ExpectPy/), but
> >I seem to remember seeing at least one other one.
> >

> You're probably thinking of <URL:
ftp://ftp.bbn.com/pub/timo/python/exploop.py>.







From rceci at adams.com  Mon Jun  7 10:34:31 1999
From: rceci at adams.com (rceci at adams.com)
Date: Mon, 07 Jun 1999 14:34:31 GMT
Subject: Question about envvars
Message-ID: <7jgl9n$p13$1@nnrp1.deja.com>

swig 1.1p5
Python 1.5.1
SGI-Irix 6.5/MIPspro 7.2.1

Hello,

I have built a shared module that reads flat files for environment
variable information, and makes subsequent putenv calls for those
pertinent ones.  This functionality is encapsulated in a c++ class,
FileRegistry, which I think I have working correctly, using shadow
classes under python.  All I expose in the class is the ctor, the dtor,
and the method called exportEnvVars():

%module MDIreg
%{
#include "fileregistry.h"
%}

class FileRegistry
{
public:
  FileRegistry(bool, bool);
 ~FileRegistry();
  void exportEnvVars();
};

...


However, I can't seem to get the envvars to persist into the current
process space.  If I fork off a child, then I can see the envvars that I
expect.  Here is a pretty explanatory log:

[rceci_latest]{rcecil}/vobs/srg/src/registry> python
Python 1.5.1 (#3, May 21 1999, 16:24:24) [C] on irix646-32-mips2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import os
>>> import MDIreg
>>> m = MDIreg.FileRegistry(1,1)
>>> m.exportEnvVars()
>>> os.system("env | grep MDI_REG_VERSION")
MDI_REG_VERSION=10.0
0
>>> os.environ['MDI_REG_VERSION']
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/local/lib/python1.5/UserDict.py", line 12, in __getitem__
    def __getitem__(self, key): return self.data[key]
KeyError: MDI_REG_VERSION
>>> os.environ.keys()
['WINEDITOR', 'TS', 'MAIL', 'NOMSGLABEL', 'PWD', 'MU',
'MDI_PRINT_UCO_SIZ', 'topdir', 'MDI_ACAR_SITE', 'VOB_ROOT', 'QTDIR',
'LM_LICE
NSE_FILE', 'NPX_PLUGIN_PATH', 'PRINTER', 'HOST', 'PYTHONPATH',
'CLASSPATH', 'MOZILLA_HOME', 'REMOTEUSER', 'NOMSGSEVERITY', 'LIB_TOOL
S', 'CVSROOT', 'SHELL', 'TCLLIBPATH', 'LIB_VAR', 'MANPATH', 'EDITOR',
'CLEARCASE_BLD_UMASK', 'HZ', 'HOME', 'CLEARCASE_ROOT', 'SHLVL'
, 'LIBRARY_PATH', 'TZ', 'BUILDMEISTER', 'TERM', 'LANG', 'VOB_TOP',
'LD_LIBRARY_PATH', 'MSGVERB', 'MAGIC_PATH', 'CVINSTRLIB', 'PATH',
 'T', 'USER', 'LOGNAME', 'CPLUS_INCLUDE_PATH', 'SCRIPTS_DIR',
'HOSTTYPE', 'REMOTEHOST', 'BL', 'CLEARCASE_CMDLINE', 'LIB_ARCH']
>>>
>>> os.system("/usr/local/bin/python")
Python 1.5.1 (#3, May 21 1999, 16:24:24) [C] on irix646-32-mips2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import os
>>> os.environ['MDI_REG_VERSION']
'10.0'
>>> ^D
0
>>> ^D

---

As you can see I can fork of either a simple csh command, or a
completely new python interpreter, and either way, the children can see
the result of the exportEnvVars() method call.  By the way,  when I
build my module in debug mode, I can see my "destroyed" messages I
specifically coded in the dtor of the class, to verify that my
FileRegistry object is being destroyed when I expect it.

Any ideas or suggestions?

Thanks

Rob Cecil



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From john.michelsen at gte.net  Tue Jun 29 17:23:40 1999
From: john.michelsen at gte.net (John Michelsen)
Date: Tue, 29 Jun 1999 21:23:40 GMT
Subject: Python GUIs: Summary and Conclusion
References: <99FA4A1A1280BEBD.0A1ABA62CC989479.4C2C6022F695BEAC@lp.airnews.net> <377805C8.C6BBD4C1@unifix-online.com>
Message-ID: <wPae3.439$4K6.21052@dfiatx1-snr1.gtei.net>

Philippe, they look very nice. How did you use them with Python
and Tkinter? Any Python code that you would like to share
demonstrating BWidgets?  I would be especially interested
to see use of the tree widget.

Thanks,

John


Philippe Chassany wrote in message <377805C8.C6BBD4C1 at unifix-online.com>...
>> ....
>> More widgets should be added to Tk(Inter).
>> ....
>
>Our company has made a 100% Tcl/Tk Toolkit named "BWidget".
>BWidgets are "well known" in the Tcl community because
>they are free, flexible, powerful and completly interpreted (no C).
>More than 25 widgets are available.
>
>We have tested BWidget with Tkinter the result is quite fine :
>Quick and good looking. It's a good alternative to GTk or Qt
>compiled programming to have professional application GUIs.
>
>Try it to make your own idea (free download, screenshots,
>full HTML manuals)
>
>http://www.unifix-online.com/BWidget
>
>--
>Philippe Chassany
>UNIFIX - http://www.unifix-online.com
>--
>
>





From herzog at online.de  Sat Jun 19 14:05:16 1999
From: herzog at online.de (Bernhard Herzog)
Date: 19 Jun 1999 20:05:16 +0200
Subject: An Un-optimization anecdote with text-processing
References: <376B1AAC.19FE8BCE@mysolution.com> <d%Ga3.498$dR.35932@newsfeed.avtel.net> <376B8C51.5D59C774@vision.arc.nasa.gov>
Message-ID: <m3yahgm5gi.fsf@greebo.nodomain.de>

Chad Netzer <chad at vision.arc.nasa.gov> writes:

> Well, I was in the process of starting to write a Perl program
> for a simple text processing task I have to do, when this post
> steered me back to my language of choice.  As I wrote my script,
> which is called by the Unix "find" command, I got to thinking
> about how it was anti-optimized...  Here is what I wrote:
> 
> import string
> import sys
> 
> # Way stupid algorithm, but I'm running it overnight :)
> filename = sys.argv[1]
> f = open(filename, 'r')
> text = f.read()
> while 1:
>     i = string.find(text, r'/"')
>     if i == -1: break
>     text = text[:i+1] + "index.html" + text[i+1:]
> 
> f.close()
> f = open(filename, 'w')
> f.write(text)
> f.close()
>
> 
> 
> As you can see, the core loop scans the entire file for a string
> match, builds a new string by concatenation, then starts scanning from
> the beginning yet again.  It almost doesn't seem like it could be made
> much worse, unless one deliberately tried.
> 
> Which brings me to my new game...  Rather than finding a nice hack
> that would speed things up and use less lines of code, how about
> finding the hack that uses the least amount of code to make this as
> slow as possible?  Adding pointless operations, or other obfuscations
> don't count.  Try to make this operation as elegantly short, and as
> painfully slow as possible...
> 
> For example:
> 
> I had originally coded this with string.index() using a try/except
> pair to break out of the loop, which took more lines of code but is
> probably faster than my loop (it avoids the 'if' statement in the
> loop, for the overhead of one exception).  So I changed it to be
> shorter and (possibly) slower.  Can anyone think of other elegant
> unoptimizations?

My first attempt at a shortest implementation for the string replacement
part is this:


Python 1.5.2 (#3, Apr 21 1999, 16:51:54)  [GCC 2.7.2.3] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> from string import * 
>>> s = ' "www.python.org/" xyzzy/plugh "fee/fie/foe/foo/" '
>>> join(split(s, '/"'), '/index.html"')
' "www.python.org/index.html" xyzzy/plugh "fee/fie/foe/foo/index.html" '
>>> 

But I fear it's also one of the fastest...

Or how about:

>>> reduce(lambda x,y: x[-1] == '/' and y == '"' and x+'index.html"' or x+y, s)
' "www.python.org/index.html" xyzzy/plugh "fee/fie/foe/foo/index.html" '

you don't even need the string module. This is veerrryyyy
sssssllllllooooooowwwwwwww because the result string is copied over and
over again and you call a python function for every char.


To give you an impression of the speed difference:

>>> def test1(s):
..     reduce(lambda x,y: x[-1] == '/' and y == '"' and x+'index.html"' or x+y, s)
.. 
>>> def test2(s):
..     join(split(s, '/"'), '/index.html"')
.. 
>>> import time 
>>> from string import *
>>> s = open("data.html").read()
>>> t = time.clock(); test1(s); print time.clock() - t 
69.49
>>> t = time.clock(); test2(s); print time.clock() - t
0.01
>>> 



-- 
Bernhard Herzog	  | Sketch, a python based drawing program
herzog at online.de  | http://www.online.de/home/sketch/



From aahz at netcom.com  Thu Jun 24 14:37:37 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Thu, 24 Jun 1999 18:37:37 GMT
Subject: Thanks for Python...
References: <376ead64@anonymous.newsfeeds.com> <7ktfn0$5q$1@aurwww.aur.alcatel.com> <14194.18324.582362.94739@amarok.cnri.reston.va.us> <37726458.C8CC9623@zipzap.ch>
Message-ID: <aahzFDuH2p.9C5@netcom.com>

In article <37726458.C8CC9623 at zipzap.ch>,
Boris Borcic  <zorro at zipzap.ch> wrote:
>Andrew M. Kuchling wrote:
>>
>> Once you accept that the world is a giant computer run by white mice, all
>> other movies fade into insignificance.
>>     -- Mutsumi Takahashi, on _The Hitch-hiker's Guide to the Galaxy_
>
>Should this be taken as a rebuttal of "Matrix" ?

_THHGttG_ was written so long before _Matrix_ that if anything it's the
other way 'round.
-- 
                      --- Aahz (@netcom.com)

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

Javascript: The 90s incarnation of the Commodore 64 Sprite -- SJM



From Klaus.Schilling at home.ivm.de  Fri Jun 18 07:05:48 1999
From: Klaus.Schilling at home.ivm.de (Klaus Schilling)
Date: 18 Jun 1999 13:05:48 +0200
Subject: Problem with Curses
References: <ppgmj7.s7r.ln@solling.seismic.de> <jpeck7.1661.ln@solling.seismic.de> <ve1dk7.1e02.ln@solling.seismic.de>
Message-ID: <87ogidzs37.fsf@home.ivm.de>

kjwolf at online.de (Klaus-Juergen Wolf) writes:
> It depends on when the executable is linked to libmytinfo, which only
> takes place on platforms that don't know the terminfo database, like
> BSD and Solaris; there are two competing declarations of tputs().

Why doesn't Freebsd know the terminfo database? Can't the one that is 
integrated into ncurses be used on FreeBSD? 

Klaus Schilling



From davecook at home.com  Fri Jun  4 01:13:44 1999
From: davecook at home.com (David M. Cook)
Date: Fri, 04 Jun 1999 05:13:44 GMT
Subject: pyGTK on Windows
References: <3757D453.A302E757@interar.com.ar>
Message-ID: <slrn7leo4r.fh.davecook@sputnik.escnd1.sdca.home.com>

On Fri, 04 Jun 1999 10:27:48 -0300, Nahuel <nahuel at interar.com.ar> wrote:

>Hi, im about to do a cross-plataform GUI application in Linux and
>Windoze... i must
>choose the GUI toolkit.. my question is .. can i use pyGTK under
>Windows???

As someone has mentioned, even if it is made to work under windows, pygtk
would not have a windows look and feel.  wxPython is probably your best bet
if tkinter is not acceptable.

Dave Cook



From k_mcdermott at my-deja.com  Fri Jun 11 05:37:05 1999
From: k_mcdermott at my-deja.com (k_mcdermott at my-deja.com)
Date: Fri, 11 Jun 1999 09:37:05 GMT
Subject: Help: MIME Attachments
Message-ID: <7jqlc1$bk0$1@nnrp1.deja.com>

Hi,

I am trying to send a plain text file, as an attachment through e-mail.

This is a hacked version of my test code below, basically the
program interprets a dump from a diary system, and then writes
Lotus Notes importable Calendar entries.  I want to then mail
them to users with an accompanying instruction header to allow them
to import them into their calendars...

The only problem is the mailing them to users bit, I either get
the plain text in the file (which I don't want, I just want them
to detach the text file and then import) or I get the Base64 encoded
text, either way there is no attachment...

My Problem is that the MIME headers indicating Base64 encoding
appears a line after the content type, and I believe that this blank
line terminates the headers...

So my question is, at what point should I be introducing these
headers...?

TIA

Kevin

ps Python has brought a project that was originally quoted at "tens of
thousands of pounds and hundreds of man-hours" down to around 4 hours
so far, and if I can crack this, ZERO outlay :-)

import base64
import sys
import StringIO
import MimeWriter
from smtplib import SMTP

# Mails the Diary Import file to the user in MIME format
# Based upon code by GVR
def mailFileToUser(_userName, _fileName):
	outputfp = StringIO.StringIO()
	w = MimeWriter.MimeWriter(outputfp)
	w.addheader("subject", "Diary Entries from Mainframe")
	w.flushheaders()
	w.startmultipartbody("mixed")
	subwriter = w.nextpart()
	f = subwriter.startbody('application/octet-stream;
name="DiaryFile.notes"')

>>> This is the puzzling bit...

	subwriter.addheader("Content-Transfer-Encoding", "base64")
	subwriter.addheader("Content-Disposition", 'attachment;
filename="DiaryFile.notes"')
>>>

	subwriter.flushheaders()
	base64.encode(open('./DiaryFile.notes', 'r'), f)
	w.lastpart()
	# s = SMTP("localhost")
	# s.sendmail("diary at system.com", ["my at email.address.com"],
outputfp.getvalue())
	# s.close()
	print outputfp.getvalue()
if __name__=='__main__':
	mailFileToUser('Kevin McDermott', 'TestFile.out')


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From arcege at shore.net  Wed Jun  2 07:50:27 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 02 Jun 1999 11:50:27 GMT
Subject: Python & IRC?
References: <rNHvjEdhm5Pp-pn2-tyCQ80XVscVt@d185d18f8.rochester.rr.com>
Message-ID: <7U853.1589$nn.478377@news.shore.net>

Benjamin Schollnick <rfreedm1 at nospam.rochester.rr.com> wrote:
: Folks,

: Has anyone done anything in regards to making a IRC
: daemon/server in Python?

: I'm thinking about sitting down and doing some socket
: work with Python, and I wouldn't want to reproduce the wheel again...

I don't think anyone has, but I could be wrong.  There are some
IRC client modules out there that might aid you (mine follows the
protocol as close as I can get it: http://www.shore.net/~arcege/python/
Pyirc-1.5.1.tar.gz).

You might also want to look at the SocketServer module (which includes
ThreadingTCPServer and ThreadingUDPServer classes).

  -Arcege




From bernhard at alpha1.csd.uwm.edu  Wed Jun  2 19:56:48 1999
From: bernhard at alpha1.csd.uwm.edu (Bernhard Reiter)
Date: 2 Jun 1999 23:56:48 GMT
Subject: Why won't this work???
References: <7j0rkr$1eeu$1@rtpnews.raleigh.ibm.com> <7j1qc5$3rj$1@m2.c2.telstra-mm.net.au> <7j3e5q$nrq$1@rtpnews.raleigh.ibm.com>
Message-ID: <slrn7lbh5v.j9o.bernhard@alpha1.csd.uwm.edu>

On Wed, 2 Jun 1999 10:13:59 -0400, Jr. King <n at n.com> wrote:
>Uhhhh, excuse me but, I did point out the problem.  The PyEval_CallObject
>call is failing everytime.  I have it commented in the code, I guess that
>wasn't good enuff.  
I guess so, too.
Some of the experts here have very limited time and nobody wants to waste it,
because we all want them to produce more of the work only they can do fast and
so good.  ;-)

So you are not only wasting Mark's, but also my time.
He probably get's hunderets of email messages and newsgroup questions
and still stays friendly. And he was still answering friendly.
Note that I also try to be friendly and explain the situation for you.

>Excuse me, but that is the way I have done and seen it
>done on c.l.c and the questions are usually answered minus the attitude.

What is c.l.c? comp.linux...?
In my eyes Mark was very friendly, I would probably have ignored you.
And I think you seem to believe in having a right to get an answer from 
someone, which is not the case.

>And I will be doggoned if I give out my darn email address on newsgroups
>again, I don't need all that trash about porno.
Okay, then you probably don't need the answer to the question from Mark.

It is IMO unfriendly to not include an email address.
It is understandable, if you encode your email a bit, but still the
chances of an answer are falling then.

If you want an answer, try to raise your chances.
	Bernhard



From ajung at sz-sb.de  Wed Jun 23 02:08:22 1999
From: ajung at sz-sb.de (Andreas Jung)
Date: Wed, 23 Jun 1999 06:08:22 GMT
Subject: popen2/zlib bug in Python 1.5.2/Solaris !
Message-ID: <19990623080822.A15920@sz-sb.de>

Subject: Some zlib/popen bugs in 1.5.2/Solaris !

After the change from 1.5.2b2 to 1.5.2 I encountered serveral problems
in the official release. The test suite fails in the test_zlib and
the test_popen2 module.

Traceback from test_zlib:

(gdb) run
Starting program: /ojs/home/ojs/src/Python-1.5.2/python Lib/test/test_zlib.py
warning: Unable to find dynamic linker breakpoint function.
warning: GDB will be unable to debug shared library initializers
warning: and track explicitly loaded dynamic code.
0xe5c1a120 0x43b6aa94
0xbd602f7 0xbd602f7
expecting Bad compression level
expecting Invalid initialization option
expecting Invalid initialization option
normal compression/decompression succeeded
compress/decompression obj succeeded
decompress with init options succeeded
decompressobj with init options succeeded

Program received signal SIGBUS, Bus error.
0xef5c63bc in t_delete ()
(gdb) bt
#0  0xef5c63bc in t_delete ()
#1  0xef5c6080 in realfree ()
#2  0xef5c687c in cleanfree ()
#3  0xef5c5ccc in realloc ()
#4  0x4b0b4 in _PyString_Resize (pv=0xeffff228, newsize=3850) at stringobject.c:607
#5  0x87268 in PyZlib_flush (self=0x104620, args=0x8000) at ./zlibmodule.c:618
#6  0x54700 in call_builtin (func=0x1129f8, arg=0xe6070, kw=0x0) at ceval.c:2359
#7  0x545cc in PyEval_CallObjectWithKeywords (func=0x1129f8, arg=0xe6070, kw=0x0) at ceval.c:2324
#8  0x53018 in eval_code2 (co=0x110520, globals=0x0, locals=0x70, args=0x0, argcount=1124856, kws=0xffffffff, kwcount=0, 
    defs=0x0, defcount=0, owner=0x0) at ceval.c:1654
#9  0x50924 in PyEval_EvalCode (co=0x110520, globals=0xf1a98, locals=0xf1a98) at ceval.c:324
#10 0x3210c in run_node (n=0xf4fb0, filename=0x110520 "", globals=0xf1a98, locals=0xf1a98) at pythonrun.c:887
#11 0x320bc in run_err_node (n=0xf4fb0, filename=0x110520 "", globals=0xf1a98, locals=0xf1a98) at pythonrun.c:872
#12 0x32090 in PyRun_File (fp=0xe36e0, filename=0xeffffa0e "Lib/test/test_zlib.py", start=257, globals=0xf1a98, locals=0xf1a98)
    at pythonrun.c:860
#13 0x3159c in PyRun_SimpleFile (fp=0xe36e0, filename=0xeffffa0e "Lib/test/test_zlib.py") at pythonrun.c:570
#14 0x31178 in PyRun_AnyFile (fp=0xe36e0, filename=0xeffffa0e "Lib/test/test_zlib.py") at pythonrun.c:451
#15 0x21c9c in Py_Main (argc=2, argv=0xeffff8bc) at main.c:287
#16 0x216bc in main (argc=2, argv=0xeffff8bc) at python.c:12
(gdb) 



test_popen2:

bonnie@/ojs/home/ojs/src/Python-1.5.2(90)%  ./python Lib/test/test_popen2.py
testing popen2...
testing popen3...
Traceback (innermost last):
  File "Lib/test/test_popen2.py", line 16, in ?
    main()
  File "Lib/test/test_popen2.py", line 14, in main
    popen2._test()
  File "./Lib/popen2.py", line 95, in _test
    assert not _active
AssertionError

All tests are reproduceable using egcs-1.1.2 and egcs-2.95 under Solaris 2.6.

Any idea ?

Thanks,
Andreas



-- 
                                _\\|//_
                               (' O-O ')
------------------------------ooO-(_)-Ooo--------------------------------------
   Andreas Jung, Saarbr?cker Zeitung Verlag und Druckerei GmbH
   Saarbr?cker Daten-Innovations-Center
   Gutenbergstr. 11-23, D-66103 Saarbr?cken, Germany
   Phone: +49-(0)681-502-1528, Fax: +49-(0)681-502-1509
   Email: ajung at sz-sb.de (PGP key available)
-------------------------------------------------------------------------------



From chalaoux at cybercable.fr  Mon Jun  7 12:15:07 1999
From: chalaoux at cybercable.fr (FR chalaoux)
Date: Mon, 07 Jun 1999 18:15:07 +0200
Subject: Looking for C Guru to compile Python
Message-ID: <375BF00B.B1E19B54@cybercable.fr>

Hi All,

I'm trying to compile Python 152 on a Intel machine with the standard
RH51 ( configure and make ) and I obtain this error (see below make
error ). I joined the
config.log file in attachement. Some explanation about failed in
config.log are welcome ...

Thanks for help.


error from make:
=============

gcc  -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H -c ./timemodule.c
./timemodule.c: In function `time_strptime':
./timemodule.c:429: warning: assignment makes pointer from integer
without a cast
gcc  -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H -c ./operator.c
gcc  -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H -c ./fcntlmodule.c
gcc  -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H -c ./pwdmodule.c
gcc  -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H -c ./grpmodule.c
gcc  -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H -c ./selectmodule.c
gcc  -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H -c ./socketmodule.c
./socketmodule.c: In function `initsocket':
./socketmodule.c:2212: `uint32_t' undeclared (first use this function)
./socketmodule.c:2212: (Each undeclared identifier is reported only once

./socketmodule.c:2212: for each function it appears in.)
./socketmodule.c:2212: parse error before `0xe0000000U'
./socketmodule.c:2217: parse error before `0xe0000001U'
./socketmodule.c:2222: parse error before `0xe00000ffU'
make[1]: *** [socketmodule.o] Error 1
make[1]: Leaving directory `/usr/local/Python-1.5.2/Modules'
make: *** [Modules] Error 2



--
FR Chalaoux
ADVL (Au Del? de Vos Limites)
chalaoux at cybercable.fr
http://www.advl.org/


-------------- next part --------------
This file contains any messages produced by compilers while
running configure, to aid debugging if configure makes a mistake.

configure:608: checking MACHDEP
configure:653: checking CCC
configure:669: checking for --without-gcc
configure:746: checking for gcc
configure:859: checking whether the C compiler (gcc  ) works
configure:875: gcc -o conftest    conftest.c  1>&5
configure:901: checking whether the C compiler (gcc  ) is a cross-compiler
configure:906: checking whether we are using GNU C
configure:915: gcc -E conftest.c
configure:934: checking whether gcc accepts -g
configure:997: checking LINKCC
configure:1014: checking LDLIBRARY
configure:1043: checking for ranlib
configure:1076: checking for ar
configure:1150: checking how to run the C preprocessor
configure:1171: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1230: checking for AIX
configure:1255: checking for minix/config.h
configure:1265: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1261: minix/config.h: No such file or directory
configure: failed program was:
#line 1260 "configure"
#include "confdefs.h"
#include <minix/config.h>
configure:1304: checking whether gcc accepts -OPT:Olimit=0
configure:1318: gcc -OPT:Olimit=0 -o conftest -g -O2   conftest.c  1>&5
cc1: Invalid option `-OPT:Olimit=0'
configure: failed program was:
#line 1314 "configure"
#include "confdefs.h"
int main() { return 0; }
configure:1338: checking whether gcc accepts -Olimit 1500
configure:1352: gcc -Olimit 1500 -o conftest -g -O2   conftest.c  1>&5
gcc: 1500: No such file or directory
cc1: Invalid option `-Olimit'
configure: failed program was:
#line 1348 "configure"
#include "confdefs.h"
int main() { return 0; }
configure:1375: checking for C preprocessor type
configure:1388: gcc -c -g -O2  conftest.c 1>&5
configure:1405: checking for ANSI C header files
configure:1418: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1485: gcc -o conftest -g -O2   conftest.c  1>&5
configure:1516: checking for dlfcn.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for fcntl.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for limits.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for locale.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for ncurses.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for pthread.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for signal.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for stdarg.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for stddef.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for stdlib.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for thread.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1522: thread.h: No such file or directory
configure: failed program was:
#line 1521 "configure"
#include "confdefs.h"
#include <thread.h>
configure:1516: checking for unistd.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for utime.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for sys/audioio.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1522: sys/audioio.h: No such file or directory
configure: failed program was:
#line 1521 "configure"
#include "confdefs.h"
#include <sys/audioio.h>
configure:1516: checking for sys/file.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for sys/lock.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1522: sys/lock.h: No such file or directory
configure: failed program was:
#line 1521 "configure"
#include "confdefs.h"
#include <sys/lock.h>
configure:1516: checking for sys/param.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for sys/select.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for sys/time.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for sys/times.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for sys/un.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for sys/utsname.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1516: checking for sys/wait.h
configure:1526: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:1557: checking for dirent.h that defines DIR
configure:1570: gcc -c -g -O2  conftest.c 1>&5
configure:1595: checking for opendir in -ldir
configure:1614: gcc -o conftest -g -O2   conftest.c -ldir   1>&5
ld: cannot open -ldir: No such file or directory
configure: failed program was:
#line 1603 "configure"
#include "confdefs.h"
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char opendir();

int main() {
opendir()
; return 0; }
configure:1681: checking for clock_t in time.h
configure:1710: checking for mode_t
configure:1743: checking for off_t
configure:1776: checking for pid_t
configure:1809: checking return type of signal handlers
configure:1831: gcc -c -g -O2  conftest.c 1>&5
configure:1850: checking for size_t
configure:1883: checking for uid_t in sys/types.h
configure:1918: checking size of int
configure:1937: gcc -o conftest -g -O2   conftest.c  1>&5
configure:1957: checking size of long
configure:1976: gcc -o conftest -g -O2   conftest.c  1>&5
configure:1996: checking size of void *
configure:2015: gcc -o conftest -g -O2   conftest.c  1>&5
configure:2036: checking for long long support
configure:2046: gcc -c -g -O2  conftest.c 1>&5
configure:2060: checking size of long long
configure:2079: gcc -o conftest -g -O2   conftest.c  1>&5
configure:2102: checking size of off_t
configure:2122: gcc -o conftest -g -O2   conftest.c  1>&5
configure:2144: checking whether to enable large file support
configure:2170: checking for --with-next-framework
configure:2187: checking for --with-dyld
configure:2213: checking SO
configure:2227: checking LDSHARED
configure:2267: checking CCSHARED
configure:2289: checking LINKFORSHARED
configure:2324: checking for dlopen in -ldl
configure:2343: gcc -o conftest -g -O2   conftest.c -ldl   1>&5
configure:2371: checking for shl_load in -ldld
configure:2390: gcc -o conftest -g -O2   conftest.c -ldld  -ldl  1>&5
ld: cannot open -ldld: No such file or directory
configure: failed program was:
#line 2379 "configure"
#include "confdefs.h"
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char shl_load();

int main() {
shl_load()
; return 0; }
configure:2424: checking for t_open in -lnsl
configure:2443: gcc -o conftest -g -O2   conftest.c -lnsl  -ldl  1>&5
/tmp/cca058591.o: In function `main':
/usr/local/Python-1.5.2/configure:2439: undefined reference to `t_open'
configure: failed program was:
#line 2432 "configure"
#include "confdefs.h"
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char t_open();

int main() {
t_open()
; return 0; }
configure:2464: checking for socket in -lsocket
configure:2483: gcc -o conftest -g -O2   conftest.c -lsocket -ldl  -ldl  1>&5
ld: cannot open -lsocket: No such file or directory
configure: failed program was:
#line 2472 "configure"
#include "confdefs.h"
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char socket();

int main() {
socket()
; return 0; }
configure:2504: checking for socket in -lnet
configure:2523: gcc -o conftest -g -O2   conftest.c -lnet -ldl  -ldl  1>&5
ld: cannot open -lnet: No such file or directory
configure: failed program was:
#line 2512 "configure"
#include "confdefs.h"
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char socket();

int main() {
socket()
; return 0; }
configure:2547: checking for --with-libs
configure:2561: checking for --with(out)-readline
configure:2576: checking for --with-dec-threads
configure:2592: checking for --with-threads
configure:2606: checking for --with-thread
configure:3113: checking for --with-sgi-dl
configure:3136: checking for --with-dl-dld
configure:3169: checking for alarm
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for chown
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for clock
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for dlopen
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for execv
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for flock
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for fork
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for fsync
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for fdatasync
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for ftime
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for ftruncate
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for getpeername
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for getpgrp
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for getpid
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for getpwent
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for gettimeofday
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for getwd
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca061601.o: In function `main':
/usr/local/Python-1.5.2/configure:3191: the `getwd' function is dangerous and should not be used.
configure:3169: checking for kill
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for link
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for lstat
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for mkfifo
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for mktime
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for nice
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for pause
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for plock
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca062881.o: In function `main':
/usr/local/Python-1.5.2/configure:3191: undefined reference to `plock'
configure: failed program was:
#line 3174 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char plock(); below.  */
#include <assert.h>
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char plock();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_plock) || defined (__stub___plock)
choke me
#else
plock();
#endif

; return 0; }
configure:3169: checking for pthread_init
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca063011.o: In function `main':
/usr/local/Python-1.5.2/configure:3191: undefined reference to `pthread_init'
configure: failed program was:
#line 3174 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char pthread_init(); below.  */
#include <assert.h>
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char pthread_init();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_pthread_init) || defined (__stub___pthread_init)
choke me
#else
pthread_init();
#endif

; return 0; }
configure:3169: checking for putenv
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for readlink
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for select
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for setgid
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for setlocale
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for setuid
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for setsid
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for setpgid
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for setpgrp
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for setvbuf
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for sigaction
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for siginterrupt
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for sigrelse
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca065061.o: In function `main':
/usr/local/Python-1.5.2/configure:3191: undefined reference to `sigrelse'
configure: failed program was:
#line 3174 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char sigrelse(); below.  */
#include <assert.h>
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char sigrelse();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_sigrelse) || defined (__stub___sigrelse)
choke me
#else
sigrelse();
#endif

; return 0; }
configure:3169: checking for strftime
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for strptime
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for symlink
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for tcgetpgrp
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for tcsetpgrp
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for timegm
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for times
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for truncate
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for uname
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3169: checking for waitpid
configure:3197: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3226: checking for fseek64
configure:3254: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca066791.o: In function `main':
/usr/local/Python-1.5.2/configure:3248: undefined reference to `fseek64'
configure: failed program was:
#line 3231 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char fseek64(); below.  */
#include <assert.h>
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char fseek64();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_fseek64) || defined (__stub___fseek64)
choke me
#else
fseek64();
#endif

; return 0; }
configure:3226: checking for fseeko
configure:3254: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca066921.o: In function `main':
/usr/local/Python-1.5.2/configure:3248: undefined reference to `fseeko'
configure: failed program was:
#line 3231 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char fseeko(); below.  */
#include <assert.h>
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char fseeko();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_fseeko) || defined (__stub___fseeko)
choke me
#else
fseeko();
#endif

; return 0; }
configure:3226: checking for fstatvfs
configure:3254: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca067051.o: In function `main':
/usr/local/Python-1.5.2/configure:3248: undefined reference to `fstatvfs'
configure: failed program was:
#line 3231 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char fstatvfs(); below.  */
#include <assert.h>
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char fstatvfs();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_fstatvfs) || defined (__stub___fstatvfs)
choke me
#else
fstatvfs();
#endif

; return 0; }
configure:3226: checking for ftell64
configure:3254: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca067181.o: In function `main':
/usr/local/Python-1.5.2/configure:3248: undefined reference to `ftell64'
configure: failed program was:
#line 3231 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char ftell64(); below.  */
#include <assert.h>
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char ftell64();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_ftell64) || defined (__stub___ftell64)
choke me
#else
ftell64();
#endif

; return 0; }
configure:3226: checking for ftello
configure:3254: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca067311.o: In function `main':
/usr/local/Python-1.5.2/configure:3248: undefined reference to `ftello'
configure: failed program was:
#line 3231 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char ftello(); below.  */
#include <assert.h>
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char ftello();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_ftello) || defined (__stub___ftello)
choke me
#else
ftello();
#endif

; return 0; }
configure:3226: checking for statvfs
configure:3254: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
/tmp/cca067441.o: In function `main':
/usr/local/Python-1.5.2/configure:3248: undefined reference to `statvfs'
configure: failed program was:
#line 3231 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char statvfs(); below.  */
#include <assert.h>
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char statvfs();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_statvfs) || defined (__stub___statvfs)
choke me
#else
statvfs();
#endif

; return 0; }
configure:3282: checking for dup2
configure:3310: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3282: checking for getcwd
configure:3310: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3282: checking for strdup
configure:3310: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3282: checking for strerror
configure:3310: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3282: checking for memmove
configure:3310: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:3337: checking for getpgrp
configure:3387: gcc -c -g -O2  conftest.c 1>&5
configure: In function `main':
configure:3383: too many arguments to function `getpgrp'
configure: failed program was:
#line 3380 "configure"
#include "confdefs.h"
#include <unistd.h>
int main() {
getpgrp(0);
; return 0; }
configure:3403: checking for setpgrp
configure:3453: gcc -c -g -O2  conftest.c 1>&5
configure: In function `main':
configure:3449: too many arguments to function `setpgrp'
configure: failed program was:
#line 3446 "configure"
#include "confdefs.h"
#include <unistd.h>
int main() {
setpgrp(0,0);
; return 0; }
configure:3469: checking for gettimeofday
configure:3519: gcc -c -g -O2  conftest.c 1>&5
configure:3538: checking whether time.h and sys/time.h may both be included
configure:3552: gcc -c -g -O2  conftest.c 1>&5
configure:3573: checking whether struct tm is in sys/time.h or time.h
configure:3586: gcc -c -g -O2  conftest.c 1>&5
configure:3607: checking for tm_zone in struct tm
configure:3620: gcc -c -g -O2  conftest.c 1>&5
configure:3678: checking for time.h that defines altzone
configure:3690: gcc -c -g -O2  conftest.c 1>&5
configure: In function `main':
configure:3686: `altzone' undeclared (first use this function)
configure:3686: (Each undeclared identifier is reported only once
configure:3686: for each function it appears in.)
configure: failed program was:
#line 3683 "configure"
#include "confdefs.h"
#include <time.h>
int main() {
return altzone;
; return 0; }
configure:3712: checking whether sys/select.h and sys/time.h may both be included
configure:3725: gcc -c -g -O2  conftest.c 1>&5
configure:3741: checking whether char is unsigned
configure:3805: checking for working const
configure:3859: gcc -c -g -O2  conftest.c 1>&5
configure:3882: checking for working volatile
configure:3891: gcc -c -g -O2  conftest.c 1>&5
configure:3908: checking for working signed char
configure:3917: gcc -c -g -O2  conftest.c 1>&5
configure:3934: checking for prototypes
configure:3943: gcc -c -g -O2  conftest.c 1>&5
configure:3958: checking for variable length prototypes and stdarg.h
configure:3977: gcc -c -g -O2  conftest.c 1>&5
configure:3993: checking for bad exec* prototypes
configure:4002: gcc -c -g -O2  conftest.c 1>&5
configure:4019: checking for bad static forward
configure:4040: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:4059: checking whether va_list is an array
configure:4074: gcc -c -g -O2  conftest.c 1>&5
configure:4090: checking for gethostbyname_r
configure:4118: gcc -o conftest -g -O2   conftest.c -ldl  1>&5
configure:4138: checking gethostbyname_r with 6 args
configure:4159: gcc -c -g -O2     conftest.c 1>&5
configure:4333: checking for __fpu_control in -lieee
configure:4352: gcc -o conftest -g -O2   conftest.c -lieee  -ldl  1>&5
configure:4382: checking for --with-fpectl
configure:4405: checking for --with-libm=STRING
configure:4426: checking for --with-libc=STRING
configure:4450: checking for hypot
configure:4478: gcc -o conftest -g -O2   conftest.c -lieee -ldl  -lm 1>&5
configure:4505: checking for hypot
configure:4563: checking for genuine getopt
configure:4583: gcc -o conftest -g -O2   conftest.c -lieee -ldl  1>&5
configure: failed program was:
#line 4571 "configure"
#include "confdefs.h"
#include <stdio.h>
extern int optind, opterr, getopt();
extern char* optarg;
int main() {
	char* av[] = { "testprog", "parameter", "-fFlag", NULL };
	opterr = 0;
	if (getopt(3, av, "f:") == 'f') { exit(1); }
	exit(0);
}
configure:4601: checking what malloc(0) returns
configure:4628: gcc -o conftest -g -O2   conftest.c -lieee -ldl  1>&5

From parkw at better.net  Sat Jun 12 01:27:58 1999
From: parkw at better.net (William Park)
Date: Sat, 12 Jun 1999 01:27:58 -0400
Subject: force sequenz with only one item?
In-Reply-To: <375FA9AE.6299E9DD@phoenix-edv.netzservice.de>; from Holger Jannsen on Thu, Jun 10, 1999 at 12:03:58PM +0000
References: <375FA9AE.6299E9DD@phoenix-edv.netzservice.de>
Message-ID: <19990612012758.A1196@better.net>

On Thu, Jun 10, 1999 at 12:03:58PM +0000, Holger Jannsen wrote:
> High,
> 
> I think it's easy to solve, but I couldn't get it right now.
> Take a look at the command-line source below:
> 
> >>> A = 0
> >>> B = 1
> >>> seq=raw_input("?")
> ?A,B
> >>> print seq
> A,B
> >>> print eval(seq)
> (0, 1)
> >>> seq=raw_input("?")
> ?A
> >>> print eval(seq)
> 0
> >>> A in seq
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> TypeError: string member test needs char left operand
> >>> A in eval(seq)
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> TypeError: 'in' or 'not in' needs sequence right argument
> >>>
> 
> Ok? Check it out?
> That's what I want to do with it:
> There's a file I'm parsing over special keywords per line, e.g.:
> Action = CREATE, VERBOSE, APPEND
> When read the line I put the right term to a variable, e.g:
> anAction = eval(rightTerm)
> Then I want to test for that keywords (global integer variables in 
> source, e.g.: CREATE = 0):
> if CREATE in anAction:
> 	pass #do something
> 
> But how could I force python to accept one-item sequences (see below!)?
> 
> Thanx for comin' with me...
> 
> Ciao,
> Holger

One item list is '[A]'
One item tuple is '(A,)'



From jmaziarz at linus.mitre.org  Mon Jun 14 15:56:24 1999
From: jmaziarz at linus.mitre.org (Jeremy Maziarz)
Date: Mon, 14 Jun 1999 15:56:24 -0400
Subject: beginners Q: CGI and file paths
References: <7ijcve$o4o$1@towncrier.cc.monash.edu.au>
Message-ID: <37655E68.EA2BEAF6@linus.mitre.org>

John,

$ pwd
/afs/rcf/user/jmaziarz

$ python
>>> outFile = '/tmp/temp.out'
>>> output = open(outFile, 'w')
>>> output.write('This is my first test...\n')
>>> output.close()

$ more /tmp/temp.out
This is my first test...

$ python
>>> outFile = '/tmp/temp.out'
>>> output = open(outFile, 'a')
>>> output.write('This is my second test...\n')
>>> output.close()

$ more /tmp/temp.out 
This is my first test...
This is my second test...

$ python
>>> outFile = '/tmp/temp.out'
>>> output = open(outFile, 'w')
>>> output.write('This is my third test')
>>> output.close()

$ more /tmp/temp.out 
This is my third test 

Try using single quotes? 
-Jer


Mr J Littler wrote:
> 
> Hi,
> I've written a cgi script that outputs a couple of files.
> The default placement is in the cgi-bin directory and
> I want them to go elsewhere. I've tried ...
> 
> filename = "/proper_path_here/c_list.html"
> append = open(filename, 'a')
> 
> but putting the path in front of the filename makes no
> difference (and doesn't generate an error)
> I read thro cgi.py but can't see what to do.
> Any help appreciated...
> Cheers
> John
> --
> jlittler at silas.cc.monash.edu.au
> http://jlittler.cjb.net
> http://linuxmusic.cjb.net



From matomira at iname.com  Thu Jun 17 04:28:03 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Thu, 17 Jun 1999 10:28:03 +0200
Subject: OO (was: Why is tcl broken?)
References: <000801beb745$02768220$979e2299@tim>
	 <871zfdp73f.fsf@home.ivm.de> <wklndk8ly6.fsf@ifi.uio.no>
	 <873dzsma6b.fsf@home.ivm.de> <wk3dzs87mr.fsf@ifi.uio.no>
	 <87lndkkmda.fsf@home.ivm.de> <87k8t3opm4.fsf@oak.treepeople.net>
Message-ID: <3768B193.209464A4@iname.com>


Graham Hughes wrote:

> trivial things, *precisely* because the damn thing is so small.  Fer
> crissakes, you don't even have the ability to create an aggregate type
> in R5RS!  You can fake it; but it sucks, and it doesn't integrate well
> with the built in type system...

I have a somewhat similar opinion. I can understand the open philosophy of
the standard, but the real problem is that almost every implementation goes
with their favorite OO way (maybe sometimes because of a lack of a CLOS
exposure. I didn't know about CLOS until I started to do CL, and many
schemers are plainly _disgusted_ about CL [Disclaimer: my students called me
"the Schemeboy" [oops, IYKWIM ;-)] back at UBA] ).
Anyway, I find it strange that tinyclos is not more popular as a ready-to-go
and optimized object system.

So then, you're faced with the choice of STk `beign right', MzScheme having
the best plan (IDE, static debugger (what about the dynamic debugger now?)
but no compilation (is it there yet?).
Of course I wouldn't choose STk just because it has a nativized tinyclos. I
could have used the tinyclos
source as it was, but I just went and integrated it into the whole type
hierarchy of MzScheme (adding metaclasses for the original MzScheme object
system (not for me, to give away), nativized primitives.
Unfortunately, that stuff didn't get merged into the distribution, as it
adds to support burden (and probably it's not interesting for them), and
they have dev plans to get some better implementation.

I even went and added a lot of CL stuff on top of that implementation
(source code from CMUCL).
It's possible to move around. This is very important when at some point you
get stuck (thread-unsafe implementation, inefficiency, requirement to run on
JVM, whatever).
I now use a portable way of writing Lisp, at least until CLers and Schemers
decide what can we do about
#f and nil (err.. did I say `comet' before?)






From emmanuel.viennet at lipn.univ-paris13.fr  Sat Jun 26 10:14:57 1999
From: emmanuel.viennet at lipn.univ-paris13.fr (Emmanuel Viennet)
Date: Sat, 26 Jun 1999 16:14:57 +0200
Subject: buffer API and pickle
Message-ID: <3774E061.A15E060D@lipn.univ-paris13.fr>

Hi,
I saw the new "buffer" APi in the python 1.5.2 sources,
but was unable to find much more information about it.

It seems that the buffer mecanism would be ideal
to avoid conversion to string (duplicating data in memory)
when writing an object to a stream.

- Is the next version of pickle supposed to make use of
  this buffer API ?

- What types of objects will support it: 
   strings, arrays, NumPy arrays, PIL images, ...


Emmanuel

-- 
Emmanuel Viennet: <viennet at lipn.univ-paris13.fr>
LIPN - Institut Galilee - Universite Paris-Nord       
93430 Villetaneuse -  France
http://www-lipn.univ-paris13.fr/~viennet/



From frehe491 at student.liu.se  Sun Jun 20 16:45:30 1999
From: frehe491 at student.liu.se (Fredrik Henbjörk)
Date: Sun, 20 Jun 1999 22:45:30 +0200
Subject: nerdnews.py
Message-ID: <376D52EA.41C6@student.liu.se>

Creates a htmlpage with todays Dilbert strip and 
headlines from Linux Today and Slashdot.

-- 
Fredrik Henbjork

Email: frehe491 at student.liu.se
WWW: http://o112.ryd.student.liu.se
-------------- next part --------------
#! /usr/bin/python
#
# Fredrik Henbjork <frehe491 at student.liu.se> 1999
# License: GNU GPL <http://www.gnu.org/copyleft/gpl.html>
#
# Why not take a look at the PyApache homepage <http://www.msg.com.mx/pyapache/>
#

import re
import urllib

# don't ask why it's done this way ...
# return url to today's strip
def get_dilbert():
	url = ''
	tries = 1
	while url == '' and tries < 10:
		try:
			dilbert = urllib.urlopen('http://www.dilbert.com/comics/dilbert/index.html')
			html = dilbert.read()
			dilbert.close()
			url = 'http://www.dilbert.com/comics/dilbert/archive/images/dilbert%s.gif' % (re.split('.gif\"> <IMG SRC=\"/comics/dilbert/archive/images/dilbert', re.split('<A HREF=\"/comics/dilbert/archive/images/dilbert', re.sub("\'", '', html), 1)[1], 1)[0])
		except:
			url = ''
		tries = tries + 1		
	return url

# get headlines from linuxtoday.com
# return list of triples (headline, url, date)
def get_linux_today():
	try:
		linux_today = urllib.urlopen('http://linuxtoday.com/lthead.txt')
		raw_news = re.split('&&', linux_today.read())[1:]
		linux_today.close()
		news = []
		for item in raw_news:
			news.append(re.split('\n', item)[1:4])
		return news
	except:
		return []

# get headlines from slashdot.org
# return list of triples (headline, url, date)
def get_slashdot():
	try:
		slashdot = urllib.urlopen('http://slashdot.org/ultramode.txt')
		raw_news = re.split('%%', slashdot.read())[1:-1]
		slashdot.close()
		news = []
		for item in raw_news:
			news.append(re.split('\n', item)[1:4])
		return news
	except:
		return []

def nerd_news():
	print '<html>'
	print '  <head>'
	print '    <title>Nerd News</title>'
	print '  </head>'
	print '  <body bgcolor=\"#ffffff\" text=\"#000000\" link=\"#64284c\" alink=\"#64284c\" vlink=\"#2c507c\">'
	print '    <center>'
	print '      <br><br>'
	print '      <table border=\"0\" cellspacing=\"0\" cellpadding=\"0\">'
	
	# news from Linux Today
	linux_today_news = get_linux_today()
	if linux_today_news != []:
		print '        <tr><td>Linux Today</td></tr>'
		print '        <tr><td><small>(<a href=\"http://linuxtoday.com\">http://linuxtoday.com</a>)</small></td></tr>'
		for item in linux_today_news:
			print '        <tr><td><a href=\"%s\">%s</a></td></tr>' % (item[1], item[0])
		print '        <tr><td>&nbsp;</td></tr>'

	# news from Slashdot
	slashdot_news = get_slashdot()
	if slashdot_news != []:
		print '        <tr><td>Slashdot</td/tr>'
		print '        <tr><td><small>(<a href=\"http://slashdot.org\">http://slashdot.org</a>)</small></td></tr>'
		for item in slashdot_news:
			print '        <tr><td><a href=\"%s\">%s</a></td></tr>' % (item[1], item[0])
		print '        <tr><td>&nbsp;</td></tr>'

	# today's Dilbert strip
	dilbert_url = get_dilbert()
	if dilbert_url != '':
		print '        <tr><td>Dilbert</td></tr>'
		print '        <tr><td><small>(<a href=\"http://www.dilbert.com/comics/dilbert/\">http://www.dilbert.com/comics/dilbert/</a>)</small></td></tr>'
		print '        <tr><td><img src=\"%s\"></td></tr>' % (dilbert_url)
		print '        <tr><td>&nbsp;</td></tr>'

	print '      </table>'
	print '    </center>'
	print '  </body>'
	print '</html>'

# main
if __name__ == '__main__':
# Uncomment the two lines below if you are going to use this as a cgi script
#	print "Content-type: text/html"
#	print
	nerd_news()


From jordi.inglada at enst-bretagne.fr  Fri Jun 11 14:01:42 1999
From: jordi.inglada at enst-bretagne.fr (Jordi Inglada)
Date: Fri, 11 Jun 1999 20:01:42 +0200
Subject: Prolog module
Message-ID: <99061120061800.01147@pruneta.titinet.enst-bretagne.fr>

Hi,

I am new to Python Programming (just did the tutorial and played
a little with it). It is really a great language!

I have seen that Python can interact with other languages. My question
is: can Python access a Prolog kernel? Or even, is there a module that
implements in Python a Prolog kernel? (in the same way one can use
Java from Python).

Thanks.
--
---------------------------------------------
Jordi INGLADA                              
Departement ITI - ENST Bretagne           
BP 832 - 29285 Brest Cedex - FRANCE      
Tel. +33 (0)2 98 00 13 55 / 10 62                 
http://www-iti.enst-bretagne.fr/~inglada/ 

"Computers are like air conditioners: they stop working properly when you open Windows"
"Linux, parce que je le vaux bien..."



From brainwav at hotmail.com  Sun Jun 13 17:38:30 1999
From: brainwav at hotmail.com (Denis Lamarche)
Date: Sun, 13 Jun 1999 21:38:30 GMT
Subject: Python as script language?
Message-ID: <37644DC7.DE422A3B@hotmail.com>

How can I set up Python so that I can call the scripts from within an
application (that I wrote) so that it will run and interface with the
application (that I wrote)?  Is this possible?

Denis Lamarche



From faassen at pop.vet.uu.nl  Mon Jun  7 13:32:49 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Mon, 07 Jun 1999 19:32:49 +0200
Subject: Suggestion: Dutch Python mailinglist
References: <slrn7lli2q.ia.gerrit@gerrit.palga.uucp> <375BC7E8.BC53EFD@prescod.net>
Message-ID: <375C0240.AA0C9B2B@pop.vet.uu.nl>

Paul Prescod wrote:
> 
> Gerrit Holl wrote:
> >
> > I know there are many Dutch
> > python-users, but most of them know English very well.
> 
> Dutch people use Python? Since when?

Me me me! Of course, there's this guy whose name starts with a G too. I
forget.. Gedo..Guddo? something like that. :)

> > Ik ben een Nederlandse jongen van 13 jaar, en mijn Engels is niet goed.
> 
> Actually, I can pretty much read that without learning Dutch. It looks
> like English with some spelling mistakes. And no more typing mistakes than
> is typical for mailing lists.

Ja, maar zonder handige vertaling erbij is het veel moeilijker te
volgen.

Yes, but without useful translation it-with is it much difficulter to
follow.

Still, Dutch is one of the languages closest to English, especially
grammar-wise.

Regards,

Martijn



From frankn=news at cs.vu.nl  Wed Jun 23 10:35:49 1999
From: frankn=news at cs.vu.nl (Frank Niessink)
Date: Wed, 23 Jun 1999 14:35:49 GMT
Subject: How to obtain an instance's name at runtime?
References: <7kqpu8$7u1$1@nnrp1.deja.com>
Message-ID: <7kqrc5$9fs@cs.vu.nl>

Dinu C. Gherman <gherman at my-deja.com> wrote:

>>>> class C:
> ...    pass
> ...
>>>>
>>>> c = C()
>>>> C.__class__.__name__ # ok
> 'C'
>>>> c.__name__           # not ok
> Traceback (innermost last):
>   File "<interactive input>", line 0, in ?
> AttributeError: __name__
>>>>

> Question: How to obtain an instance's name at runtime?

That instance doesn't have a name. Take for example:

>>> a = b = c = C()

That creates one instance of C, with three references to it. 
Now what is the `real' name of the instance?

Cheers, Frank

-- 
The suit into which the man's body had been stuffed looked as if it's only
purpose in life was to demonstrate how difficult it was to get this sort of
body into a suit.
	-- Douglas Adams, 'The Restaurant at the End of the Universe'



From air at apex.net  Thu Jun  3 14:51:16 1999
From: air at apex.net (Todd Neal)
Date: Thu, 3 Jun 1999 13:51:16 -0500
Subject: Re example
Message-ID: <18A53.61$G2.553366@news.apex.net>

Looking for some examples on how to use re, had a script made with sed ,awk,
and lynx to download and parse webpages for viewing on a palmpilot, I would
like to convert it to Python but I cant find any good examples of how to use
the re modules.
                                                                Thanks, Todd





From tim_one at email.msn.com  Mon Jun 21 00:02:43 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 21 Jun 1999 00:02:43 -0400
Subject: fork()
In-Reply-To: <u3u2s2mphd.fsf@ares.sys1.ngo.okisoft.co.jp>
Message-ID: <000001bebb9a$eabc01a0$619e2299@tim>

[Hisao Suzuki, still likes his anlaogy between explicitly delete'ing
 in C++, and explicitly removing all references in CPython]

You say my position "sounds too particular about superficial differences",
and I say yours reaches too far from superficial similarities.  We're never
going to agree on this one, so I'll leave it there.

> ...
> All in all, each idea is a valid model for the current Python.
> The definitive difference lies in the semantics of language it
> relies on.  Regarding the favor of destruction predictability
> such as seen in this newsgroup, the current actual semantics is
> perhaps as important as the official one.

Trying to nudge this in a useful direction <wink>, the "current actual
semantics" don't suggest anything about what "should be" done in the
presence of cycles with destructors in a GC world, except for the obvious
"reclaim a cycle the instant it becomes unreachable".  That is doable but--
I think --not cheaply enough to be practical.  So what then?  BTW, I'm told
that most of the C++ committee disagrees with Stroustrup on this one -- the
notion that an object can be destroyed without its destructor getting
invoked is distasteful at first bite, and leaves a bad taste even after
three helpings of creative rationalization <wink>.

If there's a good argument to made for it, it should rest where Guido left
it:  the behavior falls out of the simplest implementation, and that's
really all there is to it:  you *put up* with the semantics that fall out,
hoping they're not too bad in practice.  Lots of things work like that
(floating-point arithmetic is a particularly miserable example <0.3 wink>);
sometimes it's just the best you can get.  Let's not pretend it's a good
thing for "deep reasons", though!

>| Further the C++ 3rd Ed. says:
>|    "It is possible to design a garbage collector to invoke the
>|     destructors for objects that have been specifically
>|     registered' with the collector.  However, there is no
>|     standard way of `registering' objects.  Note that it is
>|     always important to destroy objects in an order that ensures
>|     that the destructor for one object doesn't refer to an
>|     object that has been previously destroyed.  Such ordering
>|     isn't easily achieved by a garbage collector without help
>|     from the programmer."

[Tim sez Java solved that one]

[Hisao]
> Even if the storage of such objects are not reclaimed yet, their
> finalizer()s may have been already invoked.  This may be a real
> problem for aggregation objects (if you once rely on finalizers
> in Java).  Note that Stroustrup's "destroy" implies "call the
> destructor".

We don't have an argument about Java (at least none that I've been able to
detect <wink>), and Stroustrup wasn't writing about Java.  His "destroy"--
being about C++ --*also* implies "nuke the memory", which isn't the case in
Java.  Java does guarantee that the memory of every object reachable from a
finalizer is wholly intact, so-- again --Java addresses the point Stroustrup
is making (which was in support of his view that GC not invoke destructors
at all).  GC simply doesn't need any help from the programmer to guarantee
that much; although Stroustrup is right (Graham notwithstanding <wink>) that
it's not easily achieved.

[quotes from the JLS, and a Java aggregation example where the order
 finalizers run in is important]
> ...
> You see, to implement such a class needs _help_ from the
> programmer.

Certainly.  It's a different issue, though:  GC doesn't need help to avoid
referencing bogus memory, it needs help to implement the intent of the
algorithm.  If Stroustrup were concerned about the latter, he would not have
said "such ordering isn't easily achieved ... without help", he would have
said such ordering is *impossible* to achieve without programmer
intervention.  He's a careful writer.

Even in the absence of cycles, CPython today doesn't (and can't!) guarantee
to run finalizers in the order your code may need; e.g., if you run this
today under 1.5.2:

class Parent:
    def __init__(self):
        self.kids = []
    def add_child(self, name):
        self.kids.append(Child(name))
    def __del__(self):
        global estate
        estate = 1000000
        print "parent died, leaving", estate, "dollars"
        for k in self.kids:
            print k.name, "wants a share of the cash!"

class Child:
    def __init__(self, name):
        self.name = name
    def __del__(self):
        global estate
        me = estate / 2
        estate = estate - me
        print self.name, "died of grief with", me, "of those bucks"

p = Parent()
p.add_child("Nancy")
p.add_child("Tim")
del p

It prints:

parent died, leaving 1000000 dollars
Nancy wants a share of the cash!
Tim wants a share of the cash!
Nancy died of grief with 500000 of those bucks
Tim died of grief with 250000 of those bucks

But if you run it under the current CVS Python snapshot, Tim gets the half
million and poor Nancy gets 250000 (lists get destroyed in reverse order in
the development version).

The order of finalizer invocation certainly matters a lot here (well, to me
and my sister, or at least to the charities in our wills <wink>), and it's
impossible for Python-- or any other language --to guess the intent.

All that can be said is that refcount rules *usually* avoid such surprises;
in the end, though, it's still up to you to force ordering when you need it.

BTW, that does suggest one "natural" extension of CPython semantics to
cycles:  since CPython today doesn't guarantee anything about the order in
which finalizers are called when more than one object loses its last
reference at the same time, it would be consistent to run finalizers for
members of GC'ed cycles in an arbitrary order too; and if your program
relies on a particular order, tough.

One bad aspect is that it's a rule the compiler+runtime can't enforce, in
the sense that violations can't be detected.  Python tries (much harder than
C++) to avoid rules like that.

it's-a-good-game-that-can't-be-won<wink>-ly y'rs  - tim





From ivanlan at callware.com  Tue Jun  1 11:17:01 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Tue, 1 Jun 1999 15:17:01 GMT
Subject: proposal: new mailing list
References: <37519172.55C86D95@acm.org>
Message-ID: <3753F96D.9E784D90@callware.com>

Hi All--

"Max M. Stalnaker" wrote:
> 
> I have taught some programming classes.  It seems to me python is a good
> choice for an introductory CS class since it has minimal key words, a
> clean syntax, and lots of CS structures available.  It also has good
> object orientation..  No texts, though.
> 
> It is also starting to attract a little attention from non-programmers
> who want to be programmers.  It seems to me, in an evangelical mode,
> that it is worthwhile addressing their needs, which are different than
> the typical programmer who is discovering Python.  The newsgroup may not
> be appropriate for these people?  Maybe python.org should start a
> mailing list just for them?  Python-eggs?
> 
> Over time, this mailing list, if archived, would develop a good
> reference set for the clientele.   The clientele would be partially
> self-servicing.
> 
> Mandate:
> 
> Discussion of programming issues and techniques in Python for people for
> whom Python is a first programming language.
> 

[N.B.:  I've not finished reading all of my email this morning, so have
undoubtedly missed some others' replies on this topic.]

1)  We already have the Python Tutor mailing list, which really is aimed
at people just getting started with Python.  Too many Python lists would
dilute our energies, I think.

2)  Re ``no texts, though.''  That is a situation which I am attempting
to remedy.  I've just signed a contract with SAMS to write _Teach
Yourself Python in 24 Hours_.  I am deliberately aiming the book at
people who do not know how to program, as I believe that Python is a
very nearly ideal first programming language.  The existing books are
targetted at those who know at least something about programming, but
mine will be for the true programming beginner.

Naturally, I can't speak to SAMS' scheduling or production, so do not
take what I say here as legally binding in any way (I disclaim
everything!), but my feeling is that the book *might* be in the stores
this year.

<yet-more-totally-shameless-self-promotion>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From florent.heyworth at radbit.com  Fri Jun 25 18:33:24 1999
From: florent.heyworth at radbit.com (Florent Heyworth)
Date: Sat, 26 Jun 1999 00:33:24 +0200
Subject: mxCGIPython question (was: Build the Python library into the binary)
References: <376F9518.D3252CEE@interet.com> <7kolnv$gp6$1@nnrp1.deja.com> <015e01bebe1f$202e9560$4500a8c0@thomasnotebook>
Message-ID: <377403B4.21319894@radbit.com>

Thomas Heller wrote:

> > Hi
> >
> > did you have a look at mxCGIPython at
> > http://starship.skyport.net/~lemburg/mxCGIPython.html ?
> > This is a single executable not shared library with the
> > standard Python library embedded.
> >
> > Cheers
> > --
> > Florent Heyworth
> >
> Works great, except for one problem (under Windows):
> you cannot use any extension modules, because they do
> not find python15.dll (or, worse, they find the wrong
> python15.dll).
>
> Is it possible to build a python15.dll containing the
> standard library or is it only possible to freeze executables?
>
> Thomas Heller

Hi Thomas

yes it is possible to build a python15.dll containing the standard
library. In fact it's  easier than making an excutable out of it as the
whole build process on Windows is geared to producing dlls. I haven't
had time to make notes or automate the build process yet but email me if
you want some info and I'll try to scrape something together.

Marc Lemburg's mxCGIPython package is a good starting point.

Cheers

Florent Heyworth




From ivnowa at hvision.nl  Wed Jun  2 15:37:46 1999
From: ivnowa at hvision.nl (Hans Nowak)
Date: Wed, 2 Jun 1999 19:37:46 GMT
Subject: A newbie learns... [Part 1]
In-Reply-To: <syai32j7g.fsf@foreman.ac.rwth-aachen.de>
References: <syai32j7g.fsf@foreman.ac.rwth-aachen.de>
Message-ID: <199906021936.VAA27111@axil.hvision.nl>

On 2 Jun 99, Gregory A. Landrum wrote:

> Howdy,
> 
> I'm in the process of learning python, Tkinter and all that good
> stuff. I'm doing this by writing an application which, hopefully, will
> actually be useful.  Along the way, I'm experimenting and playing and
> trying to figure out how to make things go faster/better.
> 
> What I figure is that as I find tricks which I think might be useful
> to other people trying to learn python, I'll post them.  Mabye other
> newbies will also find them useful.  
> 
> If people find this a waste of time/space/electrons, let me know and I
> won't continue.

On the contrary, I think it's very useful. I'll try to extract pieces 
of it and put them on my snippets page. 
(http://www.hvision.nl/~ivnowa/snippets) Of course, if you happen to 
have little example scripts which run out of the box, it would be 
very nice if you'd like to send me some.  ^_^

(The snippets site should be updated Real Soon Now; I have a new PC 
now, but everything (snippets, site, Python-DX, etc) is still on the 
old one. It might take a while before the updated site is up.)

Veel liefs,

--Hans Nowak (ivnowa at hvision.nl)
Homepage: http://fly.to/zephyrfalcon



From jeremy at cnri.reston.va.us  Mon Jun  7 15:57:17 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Mon,  7 Jun 1999 15:57:17 -0400 (EDT)
Subject: Python 2.0
In-Reply-To: <wkd7z858c3.fsf@turangalila.harmonixmusic.com>
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7iqhkb$d7s@wiscnews.wiscnet.net>
	<7iu7ma$d33$1@cronkite.cc.uga.edu>
	<02hfonzsx2.fsf@rtp.ericsson.se>
	<02909zzou0.fsf@rtp.ericsson.se>
	<87zp2f4qrz.fsf@pc-hrvoje.srce.hr>
	<wkd7z858c3.fsf@turangalila.harmonixmusic.com>
Message-ID: <14172.6652.550882.854800@bitdiddle.cnri.reston.va.us>

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

  DS> Hrvoje Niksic <hniksic at srce.hr> writes: | Kumar Balachandran
  DS> <kumar*xspam*@*xspam*rtp.ericsson.se> writes: | | > What you
  DS> said is quite wrong. Since Python provides closures, I can | >
  DS> change my programming style to be completely different from the
  DS> | > standard C or Fortran style and use a functional interface.
  DS> | | Since when does Python provide closures?

  DS> The bytecodehacks module allows the creation of closures as I
  DS> understand them.

Yes and no.  The bytecodehacks closures work right most of the time,
but there some problems with them.  Hence, I would guess, the name
"hacks."  A "BCH closure" does what you want most of the time, but it
does not work for immutable types or for assignments to mutable types.

The Scheme world view has a different world view, and a closure in
Scheme works a bit differently.  I suspect it may be a little
confusing for people expect BCH to make Python's variable binding work 
like Scheme's variable binding.  In Scheme, a variable is bound to a
location that stores a value.  You can store a new value in that
location using set!.  If two different functions share the same
binding for a variable, a set! performed in one function is visible in 
the other.

>>> (let ((sum 0))
       (let ((add (lambda (x) (set! sum (+ sum x))))
             (sub (lambda (x) (set! sum (- sum x)))))
                (add 2)
                (add 3)
                (sub 1)
             sum))
4

The BCH model is different, because it doesn't allow two functions to
share the same "location."  It only allows them to share the same
value.  This is a limitation of the way BCH implements closures;
non-local references are turned into references to constants when the
bind call is made.  This approach often works the constant is an
object and you're referencing one of its attributes or calling a
method.  It doesn't work for assignment.

The natural translation of the above Scheme code in Python+BCH would
probably be:
from bytecodehacks import closure

def notQuite():
    sum = 0
    def add(x):
        sum = sum + x
    def sub(x):
        sum = sum - x
    cadd = closure.bind_locals(add)
    csub = closure.bind_locals(sub)
    cadd(2)
    cadd(3)
    csub(1)
    return sum

But this doesn't work.  You need to bind sum to a mutable object (not
a string or tuple) and that operator on that object indirectly.  The
following works because it contains the value of interest inside a
list (and list operations are methods on the list object).
Effectively, you need to create the shared location that Scheme
creates for free.

def works():
    sum = [0]
    def add(x):
        sum[0] = sum[0] + x
    def sub(x):
        sum[0] = sum[0] - x
    cadd = closure.bind_locals(add)
    csub = closure.bind_locals(sub)
    cadd(2)
    cadd(3)
    csub(1)
    return sum[0]

Jeremy



From jam at newimage.com  Wed Jun 23 00:17:26 1999
From: jam at newimage.com (jam)
Date: Wed, 23 Jun 1999 00:17:26 -0400
Subject: simple 'binfmtload' script in python
Message-ID: <19990623001726.Q5115@toast.internal>

greetings,

in an effort to move forward with another project I've been working on, I
wrote up a simple script that tells a linux 2.2.x kernel how to execute a
java binary. 

I based this script on information in /usr/src/linux/Documentation/java.txt
and experimentation.

it seems to work for me. I'd like to know if anyone else finds it useful, or
has ideas on how it can be improved. it's terribly simple, but it get's the
job done.

regards,
J
-- 
|| visit gfd <http://quark.newimage.com/> 
|| psa member #293 <http://www.python.org/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>
-------------- next part --------------
#!/usr/bin/env python

fp = open("/proc/sys/fs/binfmt_misc/register", "w+b")
fp.write(":Java:M::\xca\xfe\xba\xbe::/usr/local/java/jdk1.2/bin/java:\n")
fp.close()
fp = open("/proc/sys/fs/binfmt_misc/register", "w+b")
fp.write(":Applet:M::<!--applet::/usr/local/java/jdk1.2/bin/appletviewer:\n")
fp.close()

From gerrit.holl at pobox.com  Wed Jun 16 15:08:45 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Wed, 16 Jun 1999 21:08:45 +0200
Subject: how to play waves?
Message-ID: <19990616210844.A8571@optiplex.palga.uucp>

Hello all,

is it possible to _play_ a wave file with python? So not importing a wave
module which is much to large and searching hours for the way to do it... I
just want to give the name of a file and the module plays it for me... is it
possible?

regards,
Gerrit.

--
The Dutch Linuxgames website.		De Nederlandse Linuxgames pagina.
Everything about games on Linux.	Alles over spelletjes onder Linux.
Site address:				http://linuxgames.nl.linux.org
Personal homepage:			http://nl.linux.org/~gerrit/



From fdrake at cnri.reston.va.us  Tue Jun 29 14:31:50 1999
From: fdrake at cnri.reston.va.us (Fred L. Drake)
Date: Tue, 29 Jun 1999 18:31:50 GMT
Subject: Tkinter slide question
In-Reply-To: <3778EC8B.833DEBC4@hti.com>
References: <3778EC8B.833DEBC4@hti.com>
Message-ID: <14201.4374.294974.254135@weyr.cnri.reston.va.us>

David Stokes writes:
 > Code snipit:
 > Scale(label="Rescan",variable=self.yspeed,orient=HORIZONTAL,).grid(column=0)
 > 
 > currently returns PY_VAR12 instead of the slider value.

David,
  Appearantly, you're refering to self.yspeed to get the value of the
variable, although you seem to have initialized self.yspeed with an
instance of Tkinter.StringVar or one of it's friends.
  To get the value, use self.yspeed.get(); the object itself is simply 
a reference to a Tcl variable.  The get() method will return the value 
as the data type indicated by the name of the class you instantiated
to create self.yspeed.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From rfreedm1 at nospam.rochester.rr.com  Fri Jun 11 17:12:00 1999
From: rfreedm1 at nospam.rochester.rr.com (Benjamin Schollnick)
Date: Fri, 11 Jun 1999 21:12:00 GMT
Subject: Problem with Lists & Index command
Message-ID: <rNHvjEdhm5Pp-pn2-zOTjsVjHebb4@d185d18f8.rochester.rr.com>

Hi All!

I've doing some python programming, and while I'm still learning
I'm getting better every day >g<

But this has me puzzled....

I'm trying to use a list that contains multiple pieces of 
information...
Basically a list that contains the equivalent of a "Array"....

And I'm not getting a error, but INDEX doesn't seem to want
to work.... Here's a "test program" made out of a clip from the
program....

import string

entry   = 'ABCD'

tcmlist = [('EFGHI',1),('JKLMN',2),('ABCD',3)]
found   = 0
entry   = string.upper(string.strip(entry))

try:
        found = tcmlist.index (entry)
        tcmlist[found][1] = tcmlist[found][1] + 1
        print 'found'
except ValueError:
        tcmlist.append (entry,1)
        print 'Sorry could not find it'
else:
        pass

print tcmlist

I don't see anything that looks wrong, and getting 
no syntax errors, but....

It appears that INDEX doesn't work on a multiple
dimension list????

I've worked around it for now, but it would be sooooo
much neater (for my program code, etc) if it worked..

Suggestions?

			- Benjamin



From tom at hrmanager.com  Thu Jun 17 14:37:11 1999
From: tom at hrmanager.com (tom at hrmanager.com)
Date: Thu, 17 Jun 1999 18:37:11 GMT
Subject: MySQL module compiling c in win32
Message-ID: <7kbf8l$qtn$1@nnrp1.deja.com>

I have been trying to compile the MYSQL module for python with MSVC++
5.0. Every time i try compiling it the same error appears.
   IT cannot open the python.lib file.
 I have changed the nessecary location of the file in the module. The
error is still present. Does anyone know how to fix this problem?
Or does any one know of another module to use?


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From planders at mail.utexas.edu  Thu Jun 24 11:02:22 1999
From: planders at mail.utexas.edu (Preston Landers)
Date: 24 Jun 1999 10:02:22 -0500
Subject: Emacs? (was Re: total idiot question: +=, .=, etc...)
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0o3q.toe.bernhard@alpha1.csd.uwm.edu> <7kr7ii$dss$1@nnrp1.deja.com>
Message-ID: <87pv2lk5fl.fsf_-_@pbl.ddns.org>

catlee at my-deja.com writes:

> I've got the following in my .vimrc file:
> au BufReadPost * if b:current_syntax == "python"
> au BufReadPost * 	map! += ^[:s/\(\S*\)$/\1=\1+/^M:noh^MA
> au BufReadPost * endif

I'm thinking of doing a similar thing in (X)Emacs.  It's my editor of
choice though I've never actually *programmed* something into emacs
itself.  I'm wondering if anyone's already done a similar thing so I
can spare the trouble (and the learning exercise, bah.)  

I'm thinking the best way might be to hook in to abbrev mode or
something... I guess I have some homework to do in this area.  I was
just wondering if someone's already done something like this. 

In case you're just joining this thread, the general idea is to have
your editor expand this:

foo += 1

to this:

foo = foo + 1

In theory it sounds easy enough.  It's just a simple matter of
programming.

Thanks,

---Preston




From rfreedm1 at nospam.rochester.rr.com  Mon Jun 28 00:05:38 1999
From: rfreedm1 at nospam.rochester.rr.com (Benjamin Schollnick)
Date: Mon, 28 Jun 1999 04:05:38 GMT
Subject: ConfigParser & STRPTIME ????
Message-ID: <rNHvjEdhm5Pp-pn2-OUBSngdMEQZP@d185d18f8.rochester.rr.com>

Folks,

	Sorry to repost this, but I haven't had any real answers in
regards to these questions....

	1) Can someone show me a example for STRPTIME?
		I'm following the language reference, and getting
		errors with STRPTIME.  I suspect that either the manual
		is in error, or that it doesn't work.

	2) Can you use CONFIGPARSER to write a *.INI file that
		you have made, or read into config parser, back
		out to disk? From reading Dejanews, I believe that
		it can't be?!?!?!?!?!? If so, why?  It seems silly to support
		read but not write.

			- Benjamin



From da at ski.org  Tue Jun 15 16:47:11 1999
From: da at ski.org (David Ascher)
Date: Tue, 15 Jun 1999 13:47:11 -0700 (Pacific Daylight Time)
Subject: Newbie: Truth values (three-valued logic)
In-Reply-To: <Pine.OSF.4.05.9906152037050.23333-100000@pinax.Mathematik.Uni-Bielefeld.DE>
Message-ID: <Pine.WNT.4.04.9906151331390.296-100000@rigoletto.ski.org>

On Tue, 15 Jun 1999, Olaf Delgado wrote:

> This behaviour is completely consistent in a two-valued logic, but not
> what I actually want, because maybe should neither be true nor false, nor
> should its negation. As far as I know, the logical negation operator can
> not be overloaded, so I have no chance to change this, right? No big deal,
> really, but I'd like to know if I missed something here. So, is there some
> magical trickery to achieve the behaviour I want?

I don't believe so, due to the following bit in ceval.c:cmp_outcome:

	case IS:
	case IS_NOT:
		res = (v == w);
		if (op == (int) IS_NOT)
			res = !res;
		break;


In other words, the behavior of 'not' is based on the operation of C's !
operator on an int.  

Of course, it could be changed (e.g. by adding a __not__ special hook),
but the added code complexity and speed hit would have to be justified.

Actually, after looking at the documentation for the operator module, I
think that there's a documentation problem -- reading it, one would think
that there is a __not__ special method, but it is not supported by the
core.  I think a special comment should be added to the doc to that
effect.

--david ascher





From digitome at iol.ie  Sun Jun 13 13:17:33 1999
From: digitome at iol.ie (Sean Mc Grath)
Date: Sun, 13 Jun 1999 17:17:33 GMT
Subject: Variations on a theme by "pound bang"
Message-ID: <3763e75f.3265255@news.iol.ie>

Can anyone shed some light on the benefits of:-

# !/usr/bin/env python

as opposed to

# !/usr/bin/python

Sean





From jon at rdt.monash.edu.au  Wed Jun 23 20:08:28 1999
From: jon at rdt.monash.edu.au (Jonathan Giddy)
Date: 24 Jun 1999 00:08:28 GMT
Subject: currentThread(): no current thread for ... using threading and Events?
References: <7kqivi$cli$1@m2.c2.telstra-mm.net.au>
Message-ID: <7krsts$uff$1@towncrier.cc.monash.edu.au>

srn at fn.com.au (Stephen Norris) writes:

>I'm seeing 

>currentThread(): no current thread for XXXXX

>I'm also using thread style threads. Is this what's going wrong? Does this
>confuse things? The lib ref doesn't say it will, but it seems possible,
>especially now that I look at the code. Hmmm....

Yes, this is the reason you are getting the messages.  If you can tolerate 
the messages, there's no problem.  It doesn't cause any harm.  If you can't
tolerate the messages, you could set sys.stdout to somewhere else, or you
could remove the line that prints this message from threading.py

Jon.



From fredrik at pythonware.com  Fri Jun 11 04:10:57 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 11 Jun 1999 08:10:57 GMT
Subject: Free Sex Links  7546
References: <3CN73.93097$pl3.57568@newreader.ukcore.bt.net> <375fd775@nemo.idirect.com> <19990610211744.A7465@optiplex.palga.uucp>
Message-ID: <016001beb3e1$efad7b50$f29b12c2@pythonware.com>

Gerrit Holl <gerrit.holl at pobox.com> wrote:
> > From: "Snoopy :-\)\)" <genius at acme.ca>
> True thing, I totally agree to that.
> kvrekp, do whatever you want, but don't bother us please...

just fyi, those of us who are using spam-filtered
ISP's only sees your followups, not the original
rubbish.  this might also be true for the python-
list (egroups.com doesn't have the original post,
only the followups).

in other words, simply ignoring usenet spam is
the best way to minimise their impact.  if you
find them annoying, consider finding a better
ISP.

</F>




From aahz at netcom.com  Fri Jun  4 13:18:55 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Fri, 4 Jun 1999 17:18:55 GMT
Subject: "The Python Way"
References: <14167.60002.783156.583193@anthem.cnri.reston.va.us> <Pine.A41.4.10.9906041155510.41420-100000@barney.mcit.com>
Message-ID: <aahzFCtC3J.LKr@netcom.com>

In article <Pine.A41.4.10.9906041155510.41420-100000 at barney.mcit.com>,
Robert Meegan  <Robert.Meegan at wcom.com> wrote:
>On Fri, 4 Jun 1999, Barry A. Warsaw wrote:
>> 
>> Way I see it, making the Timbot code Open Source is the only moral thing
>> to do.  Besides, wouldn't it be cool to  check out the Python
>> translation of the old Emacs Lisp RMS AI code?
>
>Actually, he's just a version of ELIZA that's stuck in a particularly
>cranky loop.

Why do you think he's just a version of ELIZA that's stuck in a
particularly cranky loop?

{And, no, I did *not* change my .sig for this post.}
-- 
                      --- Aahz (@netcom.com)

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

"Why do you like my boyfriend to tie you up and beat me.?"
    -- ELIZA generates a poly moment



From redsand at dsuper.net  Sun Jun 27 16:42:00 1999
From: redsand at dsuper.net (Dave)
Date: Sun, 27 Jun 1999 16:42:00 -0400
Subject: Advice for Newbie
References: <377588E9.82E8F567@dsuper.net> <7l4255$c22@dfw-ixnews13.ix.netcom.com> <7l49q5$p5u$1@nntp2.atl.mindspring.net>
Message-ID: <37768C98.4BD8674A@dsuper.net>

Thanks, guys. I'll try the on-line tutorial and then Learning Python.
Without getting in a religious war, it seems Python has more potential than Java
does.

Wesley Phoa wrote:

> _Learning Python_ is much better than _Programming Python_, which is a bit
> disorganized and hard to follow, and also a bit dated. And if you're
> interested in the win32 extensions, or in JPython, then _Learning Python_ is
> definitely the place to start. It also has some useful sections on 'gotchas'
> that beginners (like me) often run into.
>
> But, I have to say, I found that after a glance through the books, it was
> pretty easy just to start writing Python code with the occasional glance at
> the online documentation. And the code just works. Totally different from my
> experience learning C.
>
> Wesley.
>
> Aahz Maruch wrote in message <7l4255$c22 at dfw-ixnews13.ix.netcom.com>...
> >In article <377588E9.82E8F567 at dsuper.net>, Dave  <redsand at dsuper.net>
> wrote:
> >>
> >>I read the Python homepage. I sounds promising. What books do you
> >>commend for a beginner?
> >
> >If you already have experience with other programming languages, some
> >combination of the following: _Programming Python_, _Learning Python_,
> >and the on-line tutorial/documentation.  If you're not a programmer and
> >you're not in a hurry, wait for _Teach Yourself Python in 24 Hours_.
> >--
> >                      --- Aahz (@netcom.com)
> >
> >Hugs and backrubs -- I break Rule 6       <*>
> http://www.rahul.net/aahz/
> >Androgynous poly kinky vanilla queer het




From herzog at online.de  Tue Jun 29 06:49:33 1999
From: herzog at online.de (Bernhard Herzog)
Date: 29 Jun 1999 12:49:33 +0200
Subject: total idiot question: +=, .=, etc...
References: <000001bebffc$12060200$da9e2299@tim> <slrn7ngj6b.g9.behrends@allegro.cse.msu.edu>
Message-ID: <m3so7buvr6.fsf@greebo.nodomain.de>

behrends at cse.msu.edu (Reimer Behrends) writes:

> For instance, I noticed that I much preferred the way Ruby denotes
> attributes, even though it uses a superficially similar approach (Ruby
> prefixes attributes with an "@" symbol rather than with "self."). So,
> for instance, instead of
> 
> 	self.last = self.lines[-1]
> 	self.lines = self.lines[:-1]
> 
> we have something like:
> 
> 	@last = @lines[-1]
> 	@lines = @lines[:-1]
> 

I think a better solution that better fits the python syntax is to
simply omit the self but not the dot:

	.last = .lines[-1]
 	.lines = .lines[:-1]

That is, if one really wanted to have a shorthand for instance variables
in methods.


-- 
Bernhard Herzog	  | Sketch, a python based drawing program
herzog at online.de  | http://www.online.de/home/sketch/



From cwebster at math.tamu.edu  Mon Jun 28 20:47:50 1999
From: cwebster at math.tamu.edu (Corran Webster)
Date: 29 Jun 1999 00:47:50 GMT
Subject: ANN: Stackless Python 0.2
References: <37628EAA.C682F16C@appliedbiometrics.com> <7l6gc8$du4$1@news.tamu.edu> <377757D5.25A2AFF6@appliedbiometrics.com>
Message-ID: <7l953m$2he$1@news.tamu.edu>

In article <377757D5.25A2AFF6 at appliedbiometrics.com>,
Christian Tismer  <tismer at appliedbiometrics.com> wrote:
>
>Corran Webster wrote:
>> 
>> I managed to get this to build eventually (thanks to the hints from Michael
>> Hudson).  Looking at it, I wonder whether there's the potential here for
>> more than coroutines and to write an implementation of threading within
>> Python.  Each thread would presumably need its own Python stack, and a
>> queueing and locking system would need to be added somehow, but because
>> the Python and C stacks are no longer intertangled, switching between
>> threads should be easy (as opposed to impossible <wink>).
>
>Please be a little patient. :-)

Apologies if this read as a request for someone else, such as yourself,
to impliment this.  This was at the idle speculation stage for me, and
I hadn't seen any previous discussion of it, so I thought it worth asking
about.  Apparently it has been talked about, which is cool.

>> This wouldn't be quite as flexible as the current implimentations of
>> threads at the C level - C extensions would be called in a single block
>> with no way to swap threads until they return.  On the other hand, this
>> would allow every platform to have some sort of threading available,
>> which would be a boon.
>
>Not really. C extensions can be written in a stackless
>manner. Whenever they need to call into the interpreter,
>they can do it in a conformant manner. They do so by just
>using their own frame with their own executor.
>You need to change your thinking here: There is no return
>in that sense any longer. We just have frames in some chains,
>and all local state is kept in frames.

Okay - from your reply and Tim Peters', it seems as though my understanding
of the way the Python stack works was somewhat wrong.

>> Unfortunately I'm not familiar enough with threads to go out there and
>> implement it right away, but I thought I'd at least raise it as a
>> possibility and see what people think and what the pros and cons are.
>
>It will be available soon, and it will be nothing more than
>a Python module which tames continuations to behave as threads.

Looking forward to see what you come up with.

Corran




From gerrit.holl at pobox.com  Mon Jun 21 16:39:50 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Mon, 21 Jun 1999 22:39:50 +0200
Subject: Check for valid date
In-Reply-To: <7klq51$etp$1@nnrp1.deja.com>; from Remi Delon on Mon, Jun 21, 1999 at 04:44:20PM +0000
References: <7klq51$etp$1@nnrp1.deja.com>
Message-ID: <19990621223950.A7885@optiplex.palga.uucp>

On Mon, Jun 21, 1999 at 04:44:20PM +0000, Remi Delon wrote:
> From: Remi Delon <rdelon at my-deja.com>
> Newsgroups: comp.lang.python
> Subject: Check for valid date
> Date: 	Mon, 21 Jun 1999 16:44:20 GMT
> To: python-list at python.org
> 
> Hi, I'm trying to use the time module to check whether a date is
> actually a valid day or not.
> Unfortunately, it looks like the strptime function only performs a basic
> range check. (for example, it will accept 02/30/199)
> Here is a little test I did:
> 
> >>> strptime('02/30/1999','%m/%d/%Y)
> (1999, 2, 30, 0, 0, 0, 61, 0)

Right. strptime does what it has to do, it parses the given string.

> >>> mktime((1999, 2, 30, 0, 0, 0, 61, 0))
> 920361600.0

But this should raise an error, in my opinion. Hmm... what about
DateError?

> >>> ctime(920361600.0)
> 'Tue Mar  2 00:00:00 1999'
> 
> Is there a way to check test whether the day actually exists or not ?
> 
groeten,
Gerrit.

-- 
The Dutch Linuxgames homepage:	http://linuxgames.nl.linux.org
Personal homepage:		http://www.nl.linux.org/~gerrit/

Discoverb is a python program in Dutch and English which tests the words you
learned by asking it. Homepage: http://www.nl.linux.org/~gerrit/discoverb/



From fredrik at pythonware.com  Mon Jun 14 10:47:03 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 14 Jun 1999 14:47:03 GMT
Subject: <insert favourite flame-bait here> broken
References: <1282792849-95706075@hypernet.com> <3764D7C3.718E125E@iname.com> <002f01beb64d$58e7b5f0$f29b12c2@pythonware.com> <37651951.ED54894E@acm.org>
Message-ID: <005501beb674$c5575890$f29b12c2@pythonware.com>

> If the subject does not interest you, you don't need to read.
> Myself, I find it very enriching to meet people from other
> `cults'. That's how you can learn new things.

> Beware. That does not do Python any good. Suddenly, I have
> the impression the group is populated with very excitable and
> all-knowing `Linux kiddies'.

> And no, I don't see why I should use Python instead of
> other things.

pretty much speaks for itself, doesn't it...

btw, while we're on this topic, here's an interesting
article by kent pitman:

http://world.std.com/~pitman/pfaq/cross-posting.html

    "Why are crossposts generally antisocial?
    - Physical Burden of Reading Crossposted News
    - Consumption of Community Resources
    - Targeted Audience
    - Inclusivity and Exclusivity"

well worth thinking about!

(and that concludes my weekly broadcast to c.l.python ;-)

</F>




From greg.ewing at compaq.com  Sun Jun 27 21:34:56 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Mon, 28 Jun 1999 13:34:56 +1200
Subject: Windows - Drag & drop on Python script?
References: <000101bebeb2$f9c75940$d29e2299@tim>
Message-ID: <3776D140.587F6EF8@compaq.com>

Tim Peters wrote:
> 
> Note too that parts of Dale's method require that you're running
> NT

Which is not any help to me. Thanks for the
pointer, though.

Seems to me the -x option really needs an
argument specifying how many line to skip.

Greg



From kevinsl at yahoo.com  Mon Jun 21 13:47:45 1999
From: kevinsl at yahoo.com (kevinsl)
Date: Mon, 21 Jun 1999 09:47:45 -0800
Subject: choosing random numbers with weights/probability?
Message-ID: <929987267.6227@www.remarq.com>

I've been using the whrandom.choice routine and it's very
useful. But is there any way to add weights or
probabilities of being chosen to the items in the list?

example:

list=['one','two','three']
item=whrandom.choice(list)

Is there any way to say that 'one' and 'two' have a 25%
chance of being chosen, and 'three' has a 50% chance?

I'm hoping there's already a module to do this... or else
I'll be writing my own..

thanks,

Kevin



**** Posted from RemarQ - http://www.remarq.com - Discussions Start Here (tm) ****



From mss at kepler.de  Mon Jun  7 02:40:32 1999
From: mss at kepler.de (Michael S. Schliephake)
Date: Mon, 07 Jun 1999 08:40:32 +0200
Subject: [pickle] Different version of a class
References: <375a9760@nntp.server.uni-frankfurt.de>
Message-ID: <375B6960.9299611E@kepler.de>

Hallo -

> my application uses pickle to store and load data.
> 
> During developing cycle, many variables are added
> to the classes which are pickled. Whenever I load
> a file pickled with a previous version, some
> variables are not existent (naturally, because
> __init__ is not called). Is there an easy way
> to initialize the "new" variables or must I
> call __init__ - which would overwrite some of
> my saved variables.

In the following the method I use for program options an so on.
It's easy to maintain backward compatibility too.

Michael
---

import pickle


class aOptionDict :
  DVersion = 0              # no version traced
  def __init__( self ):
    self.DVersion = self.__class__.DVersion

  def __getinitargs__(self):
    return ()

  def __setstate__(self, dict):
    self.__dict__.update(dict)


"""
# The class in version 1:

class AppOptions (aOptionDict):
  DVersion = 1
  def __init__( self ):
    aOptionDict.__init__(self)
    self.inpData  = 'in Version 1 added'

  def __setstate__( self, dict ):
    # Not so much to do, transfer data into self
    aOptionDict.__setstate__(self, dict)

"""

# The class in version 2:

class AppOptions (aOptionDict):
  DVersion = 2
  def __init__( self ):
    aOptionDict.__init__(self)
    self.inpData  = 'in Version 1 added, in 2 initialized'
    self.newData  = 'in version 2 added'
    self.calcData = ''  # in version 2 added, depends from version 1 too

  def __setstate__( self, dict ):
    if dict['DVersion'] < self.DVersion:
      print 'Preprocessing older object ' , \
                  self.DVersion, ' <- ', dict['DVersion']
      # delete unnecessary elements from dict (but only if no
      # compatibility for backward use will be required).

    # transfer data into self
    aOptionDict.__setstate__(self, dict)

    if self.DVersion < self.__class__.DVersion:
      print 'Postprocessing older object ' , \
                  self.__class__.DVersion, ' <- ', self.DVersion
      # Do some calculations if necessary
      # Default values are already correct
      # (For backward compatibility do not change the semantic
      # of existing class members)
      self.calcData = self.inpData + ' ' + self.newData
      
      # Update the version number
      self.DVersion = self.__class__.DVersion
    

def saveOptions(od):
  fp = open('xxx.txt', 'w')
  pck = pickle.Pickler(fp)
  pck.dump(od)

def readOptions()
  fp = open('xxx.txt', 'r')
  pck = pickle.Unpickler( fp )
  return pck.load()



From andrew at andrewcooke.free-online.co.uk  Sun Jun 20 11:15:13 1999
From: andrew at andrewcooke.free-online.co.uk (Andrew Cooke)
Date: Sun, 20 Jun 1999 15:15:13 GMT
Subject: Cookie Filter (Was: Proxying HTTP - end of message?)
References: <7kgkbk$uso$1@nnrp1.deja.com>
Message-ID: <7kj0hr$j0c$1@nnrp1.deja.com>

Thanks for the help - if anyone is interested the finished script
is a minimal filter for HTTP cookies, available from
http://www.andrewcooke.free-online.co.uk/jara/junk/index.html

Thanks again,
Andrew

PS It was indeed trying to make a persistent connection.

In article <7kgkbk$uso$1 at nnrp1.deja.com>,
  Andrew Cooke <andrew at andrewcooke.free-online.co.uk> wrote:
> But my browser continues to wait.  If I press "stop" I have all the
> page, but the browser thinks the download has been interrupted.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From rozzin at geekspace.com  Mon Jun 28 22:17:02 1999
From: rozzin at geekspace.com (Joshua Rosen)
Date: Mon, 28 Jun 1999 22:17:02 -0400
Subject: file.tell problems (1.5.2/Win32)
References: <7krd1r$2oa$1@holly.csv.warwick.ac.uk> <7l3aao$65c$1@nnrp1.deja.com> <3776ED63.35D87F89@compaq.com>
Message-ID: <37782C9E.3061AF25@geekspace.com>

Greg Ewing wrote:
> 
> Bill Wilkinson wrote:
> 
> The problem is, tell() returns 0 at this point
> even if the file being appended to is NOT EMPTY!
[...] 
> Only after something has been written does tell()
> return the correct file position.

Note the following text from the Python documentation for the open() function:

"...(which on some Unix systems means that all writes append to the end of the
file, regardless of the current
     seek position)."

This leads me to believe that `append mode' is intended to `append' at the
*current* point in the file, which should, for consistency's sake, start at 0.

I'm guessing that the idea is: start at 0, for consistency's sake. Try to
append to the current position--if the OS believes that the append should be
done at the end, move to the end first.

If you want to make sure that you've got file.tell() returning the position of
the end of the file, just use file.seek(0, 2) immediately after open()ing the
file.

			-Rozzin.




From suzuki611 at okisoft.co.jp  Wed Jun  9 04:15:21 1999
From: suzuki611 at okisoft.co.jp (Hisao Suzuki)
Date: 09 Jun 1999 17:15:21 +0900
Subject: Python 2.0
References: <000b01beb21f$a17ff340$2fa22299@tim>
Message-ID: <u37lpdssbq.fsf@ares.sys1.ngo.okisoft.co.jp>

| >[Salvador]
| > I have been using Java since it was invented and I haven't programed a
| > finaliser ever. They should run under so unknown and inconsistent
| > environment that they are not usable at all (except for freeing external
| > resources).

I have not ever programmed any finalizer, too, while I have
programmed a lot of codes in Java.

| [Graham Matthews]
| > I am not sure I understand this. Can you explain to me how a finaliser
| > can run in an unknown environment. Surely everything the finaliser needs
| > is referenced by the finaliser (either directly or indirectly) and hence
| > is still there.

| The caveats are explained well in the Java Language Spec, section 12.6.
| ...

| All Java defines is enough so that the Java runtime itself never chokes on
| e.g. an internal null pointer, i.e., your "still there" guarantee.  Say you
| have a mail server holding a ref to a mail client, which in turn holds a ref
| back to the mail server.  Can the client object tell the server it's going
| away at finalization time?  Not if the server object shut down the
| communications link beforehand because the server's finalizer happened to
| get called first.  Etc.  Yes, the references are "still there", but without
| a lot of pain there's no guarantee they're still in a *useful* state.

It would be an issue of distributed garbage collection.  And it
is a difficulty for automatic GC with finalization.  In
addition, perhaps we should also take a (close) look at
ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz

It summarizes other (or rather more general) issues to consider
in the design of a finalization mechanism (and presents one
solution for Scheme---actually it is adapted by gnu guile).  I'd
think anyone who is planning improvements on Python's storage
management has to _notice_ what are summarized there first.
(The solution is another story.  It is not so obvious for
Python.)

# Certainly the current reference counting system of Python
# reduces the issues to fairly (though _not_ fully) tolerable
# ones...

--===-----========------------- Sana esprimo naskas sanan ideon.
SUZUKI Hisao            suzuki611 at okisoft.co.jp, suzuki at acm.org.



From rfreedm1 at nospam.rochester.rr.com  Thu Jun 24 22:23:01 1999
From: rfreedm1 at nospam.rochester.rr.com (Benjamin Schollnick)
Date: Fri, 25 Jun 1999 02:23:01 GMT
Subject: Python Webserver
References: <3767A69D.50DC7FBF@serop.abb.se> <37729F69.6F93B78E@boi.hp.com>
Message-ID: <rNHvjEdhm5Pp-pn2-DhedpYc3Rxg6@d185d18f8.rochester.rr.com>

On Thu, 24 Jun 1999 21:13:13, Bill Anderson <banderson at boi.hp.com> 
wrote:

> > I don't know if I've dreamt, but I think I saw something about
> > a webserver written in Python. (It's not the embedded-in-Apache-thing
> > I'm thinking of)
> > Now I just can't seem to find it anywhere, so does someone know anything
> > about this. Hipshots are welcome. I thought it'd be something fun to
> > play with.
> 
> Is Zope the 'embedded-in-apache-thing' you refer to?
> If so, you should note that it doesn't *need* apache, and can be used
> seperately (I do it this way).
> 
> Zope 1.1x uses the Python httpserver mentioned in some of the other
> posts, whereas zope 2.0.x(alpha!) uses Zserver, based on Medusa, as
> mention in the other posts. :-)

There is also a web server included in the python source library
(basehttpserver, simplehttpserver, cgihttpserver)

I'm also using those files as a basis for a slightly different 
webserver
design.

		- Benjamin



From mgutierrez at dss.es  Mon Jun 28 03:57:18 1999
From: mgutierrez at dss.es (mgutierrez at dss.es)
Date: Mon, 28 Jun 1999 07:57:18 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net>
Message-ID: <7l79sr$b4i$1@nnrp1.deja.com>

In article
<52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162 at lp.airnews.net>,
  mikael at pobox.com (Mikael Lyngvig) wrote:
> Hi,

> Also, can anyone comment on the features available in TkInter and
> wxPython - perhaps even make a feature comparison chart?
>
> [1] By sluggish I mean: You can see, or at least sense, the window
> update operations.  Windows are not drawn instantly, but in visually
> detectable steps.
>

Tkinter , or tk in general has got several advantages:
 - it's reasonably extended in the free soft community.
 - there's some packages : tix, blt

If you think a commercial system should be faster, take a look at
Java Swing running in Netscape, I've seen windows scrollings where the
windows
, buttons and combo boxes blink like crazy horses. ha-ha-ha, a
commercial
system fast ?
And if you want to switch why not to gtk ?
Anyway tk seems far more extensible than any other thing .

Regards
Manolo



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From robin at jessikat.demon.co.uk  Mon Jun 14 08:11:37 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 14 Jun 1999 13:11:37 +0100
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com>
 <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com>
 <375FDCA7.62FF@mailserver.hursley.ibm.com>
 <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com>
 <3762988E.C94AA252@iname.com>
 <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu>
 <3763C746.5CA8FE78@iname.com> <m3u2sby4gf.fsf@soma.andreas.org>
 <87pv2zjhhl.fsf@home.ivm.de> <3764F3FE.F7CBE6BA@acm.org>
Message-ID: <0Sgs3BA5FPZ3Ew6E@jessikat.demon.co.uk>

In article <3764F3FE.F7CBE6BA at acm.org>, Fernando D. Mato Mira
<matomira at acm.org> writes
>Klaus Schilling wrote:
>
>> Prefix rocks, infix is crap.
>
>Wow, wow, wow. Let's not go there..
postfix, prefix, infix etc etc are isosemantic


-- 
tree searching-ly yrs Robin Becker



From jeremy at cnri.reston.va.us  Fri Jun  4 17:41:46 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Fri,  4 Jun 1999 17:41:46 -0400 (EDT)
Subject: threads
In-Reply-To: <87emjr67ff.fsf@pc-hrvoje.srce.hr>
References: <1283593620-47537497@hypernet.com>
	<87emjr67ff.fsf@pc-hrvoje.srce.hr>
Message-ID: <14168.18308.709890.544779@bitdiddle.cnri.reston.va.us>

>>>>> "HN" == Hrvoje Niksic <hniksic at srce.hr> writes:

  [Somebody wrote:]
  [Since the mailing list to news gateway is working again, Gordon's
   messages are probably in DejaNews again.  So you can check there.]
  >> > If Greg's patches haven't been applied to the base, surely mine
  >> > would not be either.
  >> 
  >> As I recall, Greg's patches didn't make it because (1) they were
  >> late and (2) they were slow.

  HN> Interesting.  What was the source of slowness?  (If you _remove_
  HN> locks, I'd expect things to go faster, even in a single-threaded
  HN> world.)  Maybe the patches simply introduced a different set of
  HN> problems?

If I recall correctly, the problem is that taking away one big lock
usually requires introducing lots of little locks.  (There's some kind 
of conservation of concurrency going on.)  Right now, there is no need 
to have locks on access to builtin types like dictionaries, because
only a single thread can hold the interpreter lock anyway.  When that
goes away, you need to add lots of fine-grained locking on the builtin 
types.

Jeremy



From yopen at my-deja.com  Sat Jun 26 16:58:36 1999
From: yopen at my-deja.com (Bill Wilkinson)
Date: Sat, 26 Jun 1999 20:58:36 GMT
Subject: Drag and dropable scripts on NT
References: <376F5312.C404B9E0@arch.sel.sony.com>
Message-ID: <7l3etp$7ch$1@nnrp1.deja.com>

There are several ways to do this, here is one:

make a .bat file that says this:
   "c:\program files\python\python.exe" c:\temp\drag.py %1
use the correct path to your python.exe. replace drag.py with your
script name.

make a shortcut on your desktop that points to your .bat file.

use sys.argv[1] to get the name of the file to be acted on.

For example, this script:
import sys
a = sys.argv[1]
print a

will return the name of the file dragged onto the icon.

Hope this helps
Bill Wilkinson


In article <376F5312.C404B9E0 at arch.sel.sony.com>,
  Peter <peterb at arch.sel.sony.com> wrote:
> I'm interested in packaging up some Python code such that I can drop a
> file
> onto the script icon and have it do it's thing...
>
> I have the script working fine. Don't know where to turn to figure out
> the
> next step.
>
> Thanks -
>
>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From sdm7g at Virginia.EDU  Fri Jun 11 22:06:38 1999
From: sdm7g at Virginia.EDU (Steven D. Majewski)
Date: Fri, 11 Jun 1999 22:06:38 -0400 (EDT)
Subject: Problem with Lists & Index command
In-Reply-To: <rNHvjEdhm5Pp-pn2-XcIGYOCMcDxI@d185d18f8.rochester.rr.com>
References: <rNHvjEdhm5Pp-pn2-zOTjsVjHebb4@d185d18f8.rochester.rr.com>
	<37618424.8B8F3116@mindspring.com> <rNHvjEdhm5Pp-pn2-XcIGYOCMcDxI@d185d18f8.rochester.rr.com>
Message-ID: <Pine.A32.3.90.990611215835.15407D-100000@elvis.med.Virginia.EDU>

On Sat, 12 Jun 1999, Benjamin Schollnick wrote:

> But, the most obvious question, is.....
> 
> Can dictionaries be made & manipulated from
> variables?
> 
> > tcmvalues = { 'EFGHI':1,
> >               'JKLMN':2,
> >               'ABCD':3,
> >             }
> 
> i.e.
> 	tcmvalues.append ( entry:entry3)
> 

   tcvalues[entry]=entry3 


 
> Please keep in mind, that I don't know anything about
> dictionaries... The book I'm using hasn't hit 'em yet....
> (I'm not done reading yet)


Linear search is usually the slowest algorithm. 
That's why they put indexes and tables-of-contents in books,
so they can be accessed non-sequentially! ;-) 



---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---

 "IA-64 looks just about like what you would expect from a PA-RISC
  and IA-32 train wreck with a little VLIW thrown in for spice."
 * Thomas J. Merritt  <tjm at spam.codegen.com> in <news:comp.arch> *




From che at debian.org  Mon Jun  7 17:06:54 1999
From: che at debian.org (Ben Gertzfield)
Date: 07 Jun 1999 14:06:54 -0700
Subject: waste of resources ?
References: <375C0F91.1C73E4E3@icrf.icnet.uk> <375C16A9.5ADC4990@icrf.icnet.uk> <375C2003.B01BEC3@arlut.utexas.edu> <87yahvzqz3.fsf@pc-hrvoje.srce.hr>
Message-ID: <ytt1zfnen4h.fsf@gilgamesh.cse.ucsc.edu>

>>>>> "Hrvoje" == Hrvoje Niksic <hniksic at srce.hr> writes:

    Hrvoje> Except that when a signal arrives during an IO operation,
    Hrvoje> it will be canceled due to the EINTR brain-damage, and
    Hrvoje> Python will throw an exception.  So if you expect your
    Hrvoje> program to accept SIGCHLD signals at random intervals,
    Hrvoje> you'd better encase all your IO operation in a loop like
    Hrvoje> this:

<snip>

    Hrvoje> This problem with signals has been my long-standing gripe
    Hrvoje> with Python.

If I recall correctly, Perl has just about the exact same problem.
Is there a good way to fix it?

-- 
Brought to you by the letters Z and I and the number 1.
"Well, I think Perl should run faster than C.  :-)"
Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/



From ngps at madcap.dyn.ml.org  Fri Jun 25 12:06:48 1999
From: ngps at madcap.dyn.ml.org (Ng Pheng Siong)
Date: 25 Jun 99 16:06:48 GMT
Subject: SWIG and binary strings?
References: <al8pv2nwnny.fsf@myntti.helsinki.fi>
Message-ID: <3773a918.0@news.cyberway.com.sg>

In article <al8pv2nwnny.fsf at myntti.helsinki.fi>,
Markus Stenberg  <mstenber at cc.Helsinki.FI> wrote:
>I've swig'ed OpenSSL library for my Python needs. 

Me too! 

Well, I've only done RSA and block ciphers, since that's
what my application needs for now. (Ok, one block cipher. 
Adding more block ciphers looks like cut-&-paste to me.)


>Is there any elegant way to make SWIG generate s# automatically (as opposed
>to s, and requiring unneccessary length parameter and much whining if
>string has 0's?)
>
>My (ugly) solution was following:

Dunno if my solution qualifies as ugly, but here's what I do:

typedef struct _blob {
        unsigned char *data;
        int len;
} Blob;

%typemap(python, in) Blob * {
        if (!PyString_Check($source)) {
                PyErr_SetString(PyExc_TypeError, "expected Blob*");
                return NULL;
        }
        $target=(Blob *)malloc(sizeof(Blob));
        if (!$target) {
                PyErr_SetString(PyExc_MemoryError, "malloc Blob*");
                return NULL;
        }
        $target->data=PyString_AsString($source);
        $target->len=PyString_Size($source);
}

Blob *rsa_private_encrypt(RSA *rsa, Blob *from, int padding) { 
	Blob *to;
	...
	to->len=RSA_private_encrypt(from->len, from->data, to->data, ...)
	...
}


Cheers.
-- 
Ng Pheng Siong <ngps at post1.com>




From mfletch at vrtelecom.com  Mon Jun 14 10:38:01 1999
From: mfletch at vrtelecom.com (Mike C. Fletcher)
Date: Mon, 14 Jun 1999 14:38:01 GMT
Subject: TKinter modal dialog
Message-ID: <018301beb673$8fdd6ca0$a9807018@yec1.on.wave.home.com>

Okay, here's the brief:

File processing task (adds structure to VRML files according to
heuristic rules determined by the user) needs to query the user every
so often if there is a missing/ambiguous rule.  I would like to make
this a GUI dialogue using TKinter.  I have said dialogue written using
the simpledialog class, and have looked at
http://www.pythonware.com/fredrik/tkdraft/ which suggests I might use
widget.wait_window, however, this doesn't really get the effect I'm
looking for, as it returns control to a widget, whereas I just want
the dialog to exit (and be destroyed) and control to return to the
processing run.

The win32 extension dialogs have a "doModal", which is exactly what
I'm looking for.  Anyone care to point out what section of the docs I
didn't read that describes a TKinter doModal :) .

Enjoy,
Mike

______________________________________
   M i k e   C .  F l e t c h e r

       mcfletch at vrtelecom.com

        Designer, Researcher





From avv at quasar.ipa.nw.ru  Sun Jun 13 23:19:38 1999
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Sun, 13 Jun 1999 20:19:38 -0700
Subject: Giving talk on Python. Need references.
References: <375BFAF2.C543178B@spirit.gcrc.upenn.edu>
Message-ID: <376474CA.3EBB8059@quasar.ipa.nw.ru>

Tom Fenn wrote:

> I will be giving a talk on using Python at the ASMS Conference on Mass
> Spectrometry next week.  I'm discussing a "Cross Platform solution to
> data analysis applications" with an emphasis on chromatography and mass
> spectrometry.  I'd like some references to other scientific/numerical
> applications that I can reference.  If anyone has a relevent document
> they'd like cited, please
> send me the reference. (And a copy I can look at if it's not readily
> available.) Thanks.

A.V.Voinov (1999) GPS data processing automation with the use of Python
scripting language. In: J.Kakkuri (ed.)  Final Results of the Baltic Sea
Level 1997 GPS Campaign. Reports Finn. Geod. Inst., 99:2,
pp. 75-84.

It will appear at my home page at the Starship soon (with .ps).

The abstract is:

  An set of Python scripting language library modules aimed at automation
and more flexible customization of the GPS data processing with the use of
Bernese software v. 4.x is described. The motivation for the development
of  this package is explained.  A brief introduction into the Python
scripting  language and its application to the high level control of
numeric data  processing is given. A synopsis of the particular modules,
including those  which help dealing with RINEX files, GPS dates, etc,
together with those which
  encapsulate the Bernese functionality is presented. The strategy of GPS
data  preprocessing, called ``optimal baseline selection loop'' is
described.  The  examples of data processing are related to the BSL
campaign, discussed in  more detail in a separate paper found in this
issue.

Regards

Alexander





From yourname at massey.ac.nz  Sun Jun  6 00:56:19 1999
From: yourname at massey.ac.nz (yourname)
Date: Sun, 06 Jun 1999 16:56:19 +1200
Subject: Newbie question
Message-ID: <3759FF72.E393CAEA@massey.ac.nz>

Hi

I'm getting this error when I try to call a function

Traceback (innermost last):
  File "ftpclient", line 229, in ?
    FTPLogin(oFtp,site,port,sUserID,sUserPassword) #site,port)
NameError: FTPLogin

This is my first project in Python and I'm doing a really simple ftp
client that uses a UNIX server as a file repository.  All I have is a
set of functions that log me in send and recieve files and a whle loop
that sits by itself at the end of the document that spins and asks the
user for input to choose an option.

Where on the python site are their docs for error codes?

TIA

Jason




From stianh at ifi.uio.no  Wed Jun  2 12:07:48 1999
From: stianh at ifi.uio.no (Stian Husemoen)
Date: Wed, 02 Jun 1999 18:07:48 +0200
Subject: All this self dot is making me...
References: <Vn653.468$HD.99317@typhoon-sf.snfc21.pbi.net> <m33e0a52x2.fsf@chinon.cnrs-orleans.fr> <37551fa6.90445994@news.oh.verio.com> <8Rb53.1622$nn.483094@news.shore.net>
Message-ID: <375556D4.10E3@ifi.uio.no>

Michael P. Reilly wrote:
 
> The convention is to use "self".  It was the adoption of the original
> object-oriented language (SmallTalk) in the early '70s.  Why did C++
> change it to "this"?  I don't know.. probably just to be different.

SmallTalk was _not_ the first object-oriented language. Simula
introduced OO as early as 1967... Simula uses "this".

-- 

Stian Husemoen <stianh>



From mstenber at cc.Helsinki.FI  Mon Jun 28 00:57:04 1999
From: mstenber at cc.Helsinki.FI (Markus Stenberg)
Date: 28 Jun 1999 07:57:04 +0300
Subject: SWIG and binary strings?
References: <al8pv2nwnny.fsf@myntti.helsinki.fi> <3773a918.0@news.cyberway.com.sg>
Message-ID: <al8bte0udlr.fsf@myntti.helsinki.fi>

ngps at madcap.dyn.ml.org (Ng Pheng Siong) writes:
> In article <al8pv2nwnny.fsf at myntti.helsinki.fi>,
> Markus Stenberg  <mstenber at cc.Helsinki.FI> wrote:
> >I've swig'ed OpenSSL library for my Python needs. 
> Me too! 
> Well, I've only done RSA and block ciphers, since that's
> what my application needs for now. (Ok, one block cipher. 
> Adding more block ciphers looks like cut-&-paste to me.)

Ah, I've concentrated on SSL itself (TLS actually, but still; full TLS/SSL
functionality, no ciphers whatsoever wrapped). I wish someone wrapped whole
OpenSSL and maintained it.. ;-) (I'd do it, but I'm too lazy by half)

> >Is there any elegant way to make SWIG generate s# automatically (as opposed
> >to s, and requiring unneccessary length parameter and much whining if
> >string has 0's?)
> >
> >My (ugly) solution was following:
> Dunno if my solution qualifies as ugly, but here's what I do:

It does; I'd prefer solution that didn't sacrifice any speed (although,
admittedly, I guess your approach's overhead isn't very high) because
esp. SSL_write is called fairly often ;-) Also, being transparent to the
user wouldn't hurt (.. count length of strings? isn't this Century of the
Fruitbat?)

I thought of that as well.. Oh well. Hoped there was some really elegant
solution I was just missing ;-)

> Cheers.
> -- 
> Ng Pheng Siong <ngps at post1.com>

-Markus Stenberg

-- 
"Schizophrenia, like the Unicorn, is described in various ways by
various people. And its elimination, like the capture of the Unicorn,
appears to require some special conditions, not all of which have yet
been specified. Finally, whether or not those conditions exist, belief
in their existence has had significant consequences."
	- Schizophrenia, Behavioural Aspects (Salzinger, 1972)



From mwh21 at cam.ac.uk  Wed Jun 16 06:39:36 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 16 Jun 1999 11:39:36 +0100
Subject: ANN: Stackless Python 0.2
References: <37628EAA.C682F16C@appliedbiometrics.com> <7k4fd4$c04$1@brick.cswv.com> <376636EA.84240129@appliedbiometrics.com> <7k6ttr$i5g$1@brick.cswv.com> <m3909lszww.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <m37lp4tonr.fsf@atrus.jesus.cam.ac.uk>

Michael Hudson <mwh21 at cam.ac.uk> writes:

> neelk at brick.cswv.com (Neel Krishnaswami) writes:
> 
> > In article <376636EA.84240129 at appliedbiometrics.com>,
> > Christian Tismer  <tismer at appliedbiometrics.com> wrote:
> > >
> > >
> > >Neel Krishnaswami wrote:
> > >
> > >[Stackless Python 0.2]
> > >

[stuff]

> It doesn't seem to link yet. Oh well, it's late and I'm going to bed.

Got this one; PyEval_Frame_Dispatch is defined static in
Python/ceval.c. Take the keyowrd static out, and it builds and runs
fine (in my limited testing so far).

Michael
 
> I can send you a diff against cvs's latest if that would help.
> 
> Yours,
> Michael



From matomira at acm.org  Tue Jun 15 20:09:22 1999
From: matomira at acm.org (Fernando D. Mato Mira)
Date: Wed, 16 Jun 1999 02:09:22 +0200
Subject: Good news
Message-ID: <3766EB32.1B347B64@acm.org>

Guess what? The previous discussion inspired me for some paper
regarding something I've got dormant for years
(nothing talking about `broken' stuff).

So now I'll have to study Python after all!

Cheers,

--
Fernando D. Mato Mira
Real-Time SW Eng & Networking
Advanced Systems Engineering Division
CSEM
Jaquet-Droz 1                   email: matomira AT acm DOT org
CH-2007 Neuchatel                 tel:       +41 (32) 720-5157
Switzerland                       FAX:       +41 (32) 720-5720

www.csem.ch      www.vrai.com     ligwww.epfl.ch/matomira.html






From adjih at technologist.com  Thu Jun  3 08:16:42 1999
From: adjih at technologist.com (adjih at technologist.com)
Date: Thu, 03 Jun 1999 12:16:42 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <87n1ykr0z9.fsf@ev.netlab.co.jp> <7ivs9q$dc5@news.dns.microsoft.com> <87k8toqv70.fsf@ev.netlab.co.jp> <aahzFCnx28.1Jp@netcom.com> <7j44ql$908$1@cronkite.cc.uga.edu>
Message-ID: <7j5rn5$dvs$1@nnrp1.deja.com>

In article <7j44ql$908$1 at cronkite.cc.uga.edu>,
  graham at sloth.math.uga.edu (Graham Matthews) wrote:
> Aahz Maruch (aahz at netcom.com) wrote:
> : In addition to Paul's excellent comments, if one is writing an
> : interactive process, one cannot afford the unpredictability of
random
> : garbage collection.  Garbage collection could run in a thread, but
then
> : one is restricted to environments where threads are well-supported.
>
> You appear to be making an argument like "when a GC occurs the system
> pauses, and this is bad in an interactive environment". If that's the
> case then this is an old chestnut. Modern GC implementations do not
> have such pauses. Look up papers on real-time garbage collection for
> example.

Hmm... How do you do multi-thread real-time (i.e. incremental) GC being
a) portable (must run on PalmPilot, WinCE, Unix, Mac,...) b) without
relying on assumptions [that will fail in some programs] (such as "very
few writes in shared space"). The supplementary condition would be "c)
efficient" which would be useful for languages such as C/C++ or compiled
Java, but I grant you that this is not necessary for the current Python,
and that we are comparing here to reference counting anyway.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From arcege at shore.net  Wed Jun 23 07:23:44 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 23 Jun 1999 11:23:44 GMT
Subject: Using C to execute Python
References: <7kqcds$4nu$1@kopp.stud.ntnu.no>
Message-ID: <4t3c3.1044$7X1.266606@news.shore.net>

In comp.lang.python Thomas R.berg <thomar at ifi.ntnu.no> wrote:
: I have a C program and I want to call certain Python methods / functions
: passing quite a few parameters.
: How can I do this, in general fashion?
: I would have to include python.h, I reckon - where can I find this, I've
: been searching all over for it (is it just me???)
: I'm no expert, far from, in Python...

: Thomas
: email: thomar at ifi.ntnu.no

Hi Thomas,

You will need the full source distribution of Python (the default Linux
distribution, the binary distributions come with the header files, etc.).

Compile and install that.  Then read "Extending and Embedding the
Python Interpreter" (http://www.python.org/doc/current/ext/ext.html)
and "Python/C API Reference Manual"
(http://www.python.org/doc/current/api/api.html) to describe how to
embed Python.

Very simply,
  1) Initialize Python (with Py_Initialize())
  2) Run some of the Python API calls (high-level or low-level)
  3) When you are done, call Py_Finalize()

int callpython(python_cmd)
  char *python_cmd;
  { int rc;

    if (!Py_IsInitialized())
      Py_Initialize();
    rc = PyRun_SimpleString(python_cmd);
    return rc;
  }

The problem with this is that exceptions are not captured and are
printed to stderr directly.  The lower-level API calls give finer
control, but luckily very easy to code (compared to other P*
languages).

  -Arcege




From llwo at dbtech.net  Mon Jun 14 00:56:46 1999
From: llwo at dbtech.net (Karl Putland)
Date: Sun, 13 Jun 1999 23:56:46 -0500
Subject: New to threading. Right idea???
Message-ID: <37648cec.0@news.dbtech.net>

I need some advise.  Am I on the right track?

The code below dose not work as I would expect.  If a thread is an
idependant path of execution should not the objects created on those paths
and imported on those paths be separate entities?

I need to load the same dll more that once.  I want to avoid mucking with or
renaming files if it can be avoided.  Gordon McMillan sugested possibly
using threads in a post on 4/20/99.  Am I thinking this though properly?
The same library is being loaded on separate threads.  When a thread is
spawned it gets its own copy of the interpreter state(yes?).  But the
library for objects returned for separate threads is the same.  Should not
each get its own?

One of the functions in the dll is EXPENSIVE.  It's not mine and I doubt
they will rework it so I have to live with it.  the EXPENSIVE function must
be called at the begining and ties all activity in the dll to a specific
directory.  If you want to change directories you must call unexpensive then
EXPENSIVE again with the new directory as an argument.  If the dll is loaded
and EXPENSIVE has been called sucessive calls to EXPENSIVE return FALSE.

I experimented with creating a copy of the dll with a diferent name and
loading it.  This worked.  I could call EXPENSIVE for the dll and the
copy_of_dll and both succeed.

Back to threads.  First tryed using the windll module, but it caches the dll
in a module level variable so I didn't see that working so I tryed the
example below and it still doesn't work.

Any ideas, or a swift kick in the right direction is greatly appreciated.

Karl Putland
kperacles at geocities.com

***Begin Example***

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> class GMdll:
 def __init__(self):
  import calldll
  self.calldll=calldll
  self.handle=self.calldll.load_library('GM4S32.DLL')
  self.h_load = self.calldll.get_proc_address(self.handle, "GMW_LoadBDE")
  self.h_unload=self.calldll.get_proc_address(self.handle, "GMW_UnloadBDE")
 def load(self):
  from windll import cstring
  s,g,c,u,p = map(cstring, ('d:\\goldmine',
'd:\\goldmine\\gmbase','d:\\goldmine\\demo','KARL','KAPPASIG'))
  args=(s,g,c,u,p)
  return self.calldll.call_foreign_function(
          self.h_load,
          'l'*len(args),
          'l',
          args
          )
 def unload(self):
  return self.calldll.call_foreign_function(
   self.h_unload,
   '',
   'l',
   ()
   )
 def die(self):
  self.calldll.free_library(self.handle)


>>> from threading import *
>>> def test():
 gm=GMdll()
 lLock.acquire()
 l.append(gm)
 lLock.release()


>>> l=[]
>>> lLock=Lock()
>>> a=Thread(target=test)
>>> b=Thread(target=test)
>>> def tt():
 a.start()
 b.start()


>>> tt()
>>> l
[<__console__.GMdll instance at 839600>, <__console__.GMdll instance at
842910>]
>>> l[0].handle
22020096
>>> l[1].handle
22020096
>>>





From schorsch at schorsch.com  Sun Jun 13 08:24:33 1999
From: schorsch at schorsch.com (Georg Mischler)
Date: Sun, 13 Jun 1999 12:24:33 GMT
Subject: do anonymous pipes normally work on NT?
References: <7juatc$fef$1@nnrp1.deja.com> <7juq84$df3$1@m2.c2.telstra-mm.net.au>
Message-ID: <7k07tu$veb$1@nnrp1.deja.com>

Mark Hammond wrote:
> Nothing is particularly obvious, although the code is too large
> to really be sure.

Thanks for looking into it.


> * Have you seen win32pipe.popen*?

Yes I did. The trouble is that I need to store the pid
(and tid on NT), since the started process can live
longer than the parent. If I want to regain control
over the child again later (for stopping it etc.) then
I can't use any of the popen*s. Also, in some cases
I need to redirect stdout to a file (which  works fine).


> * I can't see where you keep references to the handles
> for the sub-process.
> Is it possible that the handles are being closed by the
> parent  process, thereby closing the child process handles?

I was afraid of that too. I save all necessary handles as
self.hStd{Input|Output|Error}, self.hProcess and self.hThread
(assuming that I actually save the right ones...). Those
handles are only closed in self.__del__() long after my
problem arises.


> * Personally I would use win32file.CreateFile instead of
> os.open/msvcrt.get_osfhandle.  One advantage is that a
> layer of indirection you are now using is removed.

This has a technical point. Otoh, it's one of the things
that *do* work right now. I'll check seperately, if your
suggestion changes anything.


> Second advantage is that you deal with > "PyHANDLE" objects,
> and explicit closing is not necessary (a-la files)

I seem to remember a post by someone (you?) who stated
that *not* closing Windows handles on could lead to serious
resource leaks. Am I confusing something here or are
not all handles created equal?


> * You may have more luck spotting the problem (or getting
> someone else to) if you provide all this in a single,
> general purpose function.
>
> def CoProc( command_line, infile, outfile ):
>   ...
>
> with all command-line parsing and other app specific stuff
> that doesnt relate to the problem removed.

Yes, I guess I'll have to do this. I stripped it down
allready to a certain degree, but there's still some
unrelated stuff in there (In my application context it's
a tiny piece of code the way it is, but that obviuosly
doesn't help an external observer to understand it...)
I'll be back after some more fiddling.


Thanks a lot

-schorsch

--
Georg Mischler  --  simulation developper  --  schorsch at schorsch.com
+schorsch.com+  --  lighting design tools  --  http://www.schorsch.com/


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From dan at feld.cvut.cz  Wed Jun 16 08:08:38 1999
From: dan at feld.cvut.cz (Dan Ohnesorg)
Date: Wed, 16 Jun 1999 14:08:38 +0200
Subject: GNU gettext and python
Message-ID: <376793C6.9DB513F5@feld.cvut.cz>

I will internazionalize one application written python. My favoite
tool
is gettext. I have found some support for gettext, most recent in
py-gnome package. It works well but  only for simple strings like
"Hello
world". The the application uses multi-line string (it is listserv),
like
'''From: %s
To: %s

llll
kkkk
jjjj
'''

etc. 

But I am not able to bring this strings into getext. The format of
.po
files does not expect string like this and when I use many single
line
strings, the application does not find the localized version of
messages.

Is here someone, who has solved this? Or has some sugestion?

Thanks
dan



From suzuki611 at okisoft.co.jp  Wed Jun  2 23:56:03 1999
From: suzuki611 at okisoft.co.jp (Hisao Suzuki)
Date: 03 Jun 1999 12:56:03 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7iv86a$iud@news.dns.microsoft.com> <87vhd7nsra.fsf@ev.netlab.co.jp>
	<3754F5D7.2B6E1954@pop.vet.uu.nl> <aahzFCpG40.EFs@netcom.com>
	<3755714F.294CCB8D@pop.vet.uu.nl>
Message-ID: <u36756gcoc.fsf@ares.sys1.ngo.okisoft.co.jp>

In article <3755714F.294CCB8D at pop.vet.uu.nl>,
Martijn Faassen <faassen at pop.vet.uu.nl> wrote:
| Aahz Maruch wrote:
[snip]
| > I'm curious: does GC provide any advantages other than its ability to
| > deal with circular references?
| 
| I think certain forms of GC can use less memory; I don't think all forms
| of GC need extra data for all objects, like you need to store the ref
| counter in.

Yes.  And it is also often said that GC can be faster than
reference counting.  But this does not solely determine the
whole speed of every interpreter of every language.

For example, mostly Perl runs faster than Ruby (which is Perl's
successor invented by Mr. Matz --- {'June':'Pearl', 'July':'Ruby'}),
whereas Perl uses reference counting and Ruby uses some
sophisticated GC.

As for the issue of circular references, in my opinion, the
future Python should use non-compacting GC at the end of the
last thread and destroy the remained objects by invoking their
__del__ methods --- it is THE time to destroy all the objects
remained, so this is conceptually clean (thus very Pythonic :-),
though it is the way of memory management used in the present
Perl in fact...

--===-----========------------- Sana esprimo naskas sanan ideon.
SUZUKI Hisao            suzuki611 at okisoft.co.jp, suzuki at acm.org.



From john.grayson at gsc.gte.com  Tue Jun  8 11:56:50 1999
From: john.grayson at gsc.gte.com (john.grayson at gsc.gte.com)
Date: Tue, 8 Jun 1999 15:56:50 GMT
Subject: Humbug (was Re: Indentation and grouping proposal
In-Reply-To: <375D1E23.9367A42C@spirit.gcrc.upenn.edu>
References: <375D1E23.9367A42C@spirit.gcrc.upenn.edu>
Message-ID: <7jjeg2$ht60@eGroups.com>

 <375d1e23.9367a42- at spirit.gcrc.upenn.edu> wrote: 
Original Article: http://www.egroups.com/group/python-list/?start=60528
> The day Python has {'s, is the day I learn Dylan. :)
> 
> Tom Fenn
> 

You mean like:

Oh his face was all shot up and his hand was all blown off                          
And he wore a metal brace around his waist.
He whispered kind of slow, in a voice she did not know,                           
While she couldn't even recognize his face!

?? ):




From mvanier at bbb.caltech.edu  Wed Jun  2 20:42:40 1999
From: mvanier at bbb.caltech.edu (Michael Vanier)
Date: 02 Jun 1999 17:42:40 -0700
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net>
Message-ID: <jlaeuiru67.fsf@arathorn.bbb.caltech.edu>

Paul Prescod <paul at prescod.net> writes:

> Martijn Faassen wrote:
> > 
> > Hi there,
> > 
> > I don't intend to start a flamewar or anything, and I'm sure Larry Wall
> > is a nice guy, and Perl is nice and all, and the postmodern
> > rationalisation of the confusion that Perl is to me is *fun* (but does
> > not work for me), but Larry Wall seems to be infected by the whitespace
> > eating nanovirus meme. 
> 
> The first time he mentioned this I thought he was joking. He attached a
> "Hi Guido!" to indicate that he was. Now it seems he is serious. He thinks
> that whitespace usage somehow goes to the heart of the way Python
> programmers think about their craft -- as if in Python, whitespace is as
> central as object orientation is in Java. A language designer should have
> a better feeling for what is essential about a language and what is
> trivial -- especially since he whines that people judge Perl too quickly
> based on its dollar signs. If we were as fair to Perl as he is to Python
> we would call it "Perl: the curly bracket and dollar sign language."
> 

I couldn't agree with this more.  Larry is confusing syntax with semantics.
He also makes a big deal of the fact that perl was "designed to evolve" while
designating (among other languages) lisp as somehow B&D because it "drives
parentheses into the ground".  This is ridiculous.  Lisp was also "designed to
evolve" and (thanks to macros) can do so more than any other language, perl
included.  Interestingly, it's the simple parenthesized syntax that makes this
possible, or at least makes it much easier.  Lots of languages (lisp, perl,
python, C++ among others) are "multiparadigm", which I consider far more
significant than whitespace, parentheses, curly brackets or what not.

P.S.-I-may-like-lisp-but-I-still-use-python-90%-of-the-time-ly y'rs,

Mike



-------------------------------------------------------------------------
Mike Vanier	mvanier at bbb.caltech.edu
Department of Computation and Neural Systems, Caltech 216-76
Will optimize nonlinear functions with complex parameter spaces for food.



From l.a.apiolaza at massey.ac.nz  Wed Jun 23 23:41:19 1999
From: l.a.apiolaza at massey.ac.nz (l.a.apiolaza at massey.ac.nz)
Date: Thu, 24 Jun 1999 03:41:19 GMT
Subject: function with variable number of arguments
Message-ID: <7ks9cu$qht$1@nnrp1.deja.com>

Hi Pythoneers!

I want to write a function that takes a bunch of  'n' matrices (defined
using array from NumPy) and processes them further. Question: how do I
write a function with a variable number 'n' of arguments and then get
the number of arguments? I think I can put all the matrices together in
a list and then pass the list as argument, but is there another more
efficient/elegant way of doing it?

Thanks,


Luis

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Luis A. Apiolaza
IVABS
Massey University
Palmerston North
New Zealand

l.a.apiolaza at massey.ac.nz
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From faassen at pop.vet.uu.nl  Mon Jun 14 06:45:32 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Mon, 14 Jun 1999 12:45:32 +0200
Subject: OO (was: Why is tcl broken?)
References: <000a01beb5dc$5e88aa80$ae9e2299@tim> <87btejjes8.fsf@home.ivm.de>
Message-ID: <3764DD4C.2EDAAC7@pop.vet.uu.nl>

Klaus Schilling wrote:
> 
> "Tim Peters" <tim_one at email.msn.com> writes:
> > OK!  Now we know the answer Fernando is looking for:  Tcl is fatally broken
> > because its syntax differs from Lisp's <0.9 wink>.
> 
> Infix syntax is junk.

The quality of discussion on comp.lang.python is going up by leaps and
bounds, I see. :)

Whitespace-eating-nanoviruses-are-junk-ly yours,

Martijn



From holger at phoenix-edv.netzservice.de  Fri Jun 25 10:13:57 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Fri, 25 Jun 1999 14:13:57 GMT
Subject: version of python?
References: <3773866C.9D29C071@phoenix-edv.netzservice.de>
Message-ID: <37738EA5.D240D995@phoenix-edv.netzservice.de>

BTW:

In that source I could set arguments, working-directory,
and other things... Do y'a know the way to set the size
of starting window (e.g. minimized, maximized ...)?

Thanx,
Holger



From parkw at better.net  Tue Jun  8 23:46:40 1999
From: parkw at better.net (William Park)
Date: Tue, 8 Jun 1999 23:46:40 -0400
Subject: "Programming Python" still worthwhile?
In-Reply-To: <7jki66$qam$1@news.gstis.net>; from Rick Robino on Wed, Jun 09, 1999 at 02:05:58AM +0000
References: <7jki66$qam$1@news.gstis.net>
Message-ID: <19990608234640.C997@better.net>

On Wed, Jun 09, 1999 at 02:05:58AM +0000, Rick Robino wrote:
> Greetings,
> 
> I've been watching this group for some time and just now getting warm to
> Python. Looking at source examples has been great, but for completeness'
> sake I'd like to get more of the "why" maybe found in a book. Since Python
> is at 1.5.2 and people are talking about 2.0 as if it's around the corner,
> I was wondering if anyone here could give me some advice as to whether or
> not "Programming Python" is useful, if the current edition is still
> current, if I should wait for the "Learning Python", or if I should use
> some other resources.
> 
> I expect to come up to speed fairly quickly and I'm just looking for
> those fundamental insights into OOP and Python that can build up the
> kind of inuition and understanding one needs in order to "get
> comfortable" connecting/creating the dots. Currently I only program
> small - medium programs in C and shell (Korn). Any sage pointers to
> resources?
> 
> TIA,
> 
> --Rick 
>                          ^^^^^^^^^^^^^^
> Rick Robino                                  rrobino at wavedivision.com
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list

I recommend you read the tutorials and other official documentations
at <www.python.org/doc/> first.  The "Programming Python" is very
disorganized; it's okey for first reading, but, later, it's very hard
to go back to find the pages you want.

William



From thstr at serop.abb.se  Mon Jun  7 08:26:10 1999
From: thstr at serop.abb.se (Thomas S. Strinnhed)
Date: Mon, 07 Jun 1999 14:26:10 +0200
Subject: calldll and floats/doubles (newbie q)
References: <7jaih5$gk1$1@nntp8.atl.mindspring.net>
Message-ID: <375BBA62.B59C39BD@serop.abb.se>

Hi

If you've got the module up'n working this should be easy.
(At least I think so, I leave it to the wizzes to correct me if
I'm wrong :-)

The calling from python is straight forward:
>>> import spam
>>> x = spam.func(3.14159, 0.0548)

The C-code for the spammodule containing function func() could be like
this (in say spammodule.c):

/* File spammodule.c
 * Almost the same as example in 
 * "Extending and Embedding the Python Interpreter"
 * which has a lot more detail.
 */
#include "Python.h"

/* Declare functions */
static PyObject * spam_func(PyObject *self, PyObject *args);
// ... more functions

/* A method lookup table */
static PyMethodDef spam_methods[] = 
{
   {"func", spam_func, 1},
   // ... more functions
   {NULL, NULL}
};

/* Init function called on module load-time */
void _declspec(dllexport) initspam(void)
{
   (void) Py_InitModule("spam", spam_methods);
}


/* Define functions */
static PyObject * spam_func(PyObject *self, PyObject *args)
{
   double d_val, d_val2;
   PyObject *ret_val;

	/* Extract argument(s) from args) 
         * d is for double, f for float
         */
   if(!PyArg_ParseTuple(args, "dd", &d_val, &d_val2)
      return NULL;  /* Python raises exception upon NULL-value */

	/* Do something useful */
   d_val = d_val / d_val2;

	/* Convert values into PyObject, reverse of Parse_Tuple */
   ret_val = Py_BuildValue("d", d_val);
   return ret_val;
}


Good Luck 
-- Thomas

Wesley Phoa wrote:
> 
> I'd like to write a small library of option pricing functions in C, and test
> it from Python using calldll or windll. Does anyone know how to invoke
> functions that take float or double arguments and return float or double
> results?
> 
> Wesley Phoa.



From matomira at iname.com  Tue Jun 15 04:15:33 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Tue, 15 Jun 1999 10:15:33 +0200
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <rnbn1y2n5jk.fsf@nortelnetworks.com>
Message-ID: <37660BA4.EA85D219@iname.com>


Chris Ebenezer wrote:

> Fernando Mato Mira <matomira at iname.com> writes:
> : of folks end up using one or another GNU tool. As a side effect, some
> : newcomers to Tk that would have normally used tcl could end up going
> : the STk way (unfortunately, STk is not publicized enough). Things like
> : Kawa also increase the feel of security regarding code reusability.
> : And some of those people will end up wanting to write full-fledged
>
> Why use TK at all ?  Once you are using scheme anyway there are nicer
> widget toolkits to use (from a point of view of more "normal" looking
> toolkits that don't suffer from color allocation problems),  there are
> bindings for guile and gtk and guile and Qt.

I don't now. Some people might just `want to do Tk'.





From tismer at appliedbiometrics.com  Thu Jun 17 11:12:54 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Thu, 17 Jun 1999 15:12:54 GMT
Subject: ref counting, access from python
References: <7kad31$o7f$1@ns.mtu.ru>
		<3768C80B.14B245E6@appliedbiometrics.com> <wq7a3.223$7X1.49549@news.shore.net>
Message-ID: <37691076.4AA47CF7@appliedbiometrics.com>


"Michael P. Reilly" wrote:
> 
> Christian Tismer <tismer at appliedbiometrics.com> wrote:
> 
> : Oleg Orlov wrote:
> :>
> :> Hello.
> :> How can i get objects current ref count from my python code?
> :> Thanks.
> 
> :>>> import sys
> :>>> sys.getrefcount(None)
> : 1583
> :>>> print "ciao - chris"
> : ciao - chris
> :>>>
> 
> Wow, 1583 seems a little high to me. ;)

This was from PythonWin.
If you want to see real weirdness, try IDLE.
Every time you create a window, the number increases.

> Python 1.5.2 (#14, Jun 11 1999, 16:31:09) [C] on aix4
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import sys
> >>> sys.getrefcount(None)
> 140

This is slightly platform dependant.
Under Win98, I get 164 (both with standard and stackless Python).

> Anyone wonder what chris has been doing? ;)

:-) what I do all day - ciao - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     not reilly, oh really



From gregc at cgl.ucsf.edu  Mon Jun 28 21:16:39 1999
From: gregc at cgl.ucsf.edu (Greg Couch %CGL)
Date: 29 Jun 99 01:16:39 GMT
Subject: Extension Class patch for 1.5.2 (and C++)
Message-ID: <gregc.930618999@cgl.ucsf.edu>

Enclosed is a patch file for Jim Fulton's Extension Class package that
makes it work with the new PyTypeObject structure in Python 1.5.2.
I've also included changes to ExtensionClass.h that make it work with
C++.  The source that the patch is based on is the 1.2 release that
is available from http://www.digicool.com/releases/ExtensionClass/.

	Greg Couch
	gregc at cgl.ucsf.edu

===================================================================
RCS file: RCS/ExtensionClass.h,v
retrieving revision 1.12
diff -c -r1.12 ExtensionClass.h
*** ExtensionClass.h	1998/11/17 19:53:27	1.12
--- ExtensionClass.h	1999/06/29 00:38:22
***************
*** 1,6 ****
  /*
  
!   $Id: ExtensionClass.h,v 1.12 1998/11/17 19:53:27 jim Exp $
  
    Extension Class Definitions
  
--- 1,6 ----
  /*
  
!   $Id: ExtensionClass.h,v 1.12 1998/11/17 19:53:27 jim Exp gregc $
  
    Extension Class Definitions
  
***************
*** 144,153 ****
--- 144,164 ----
  	getattrofunc tp_getattro;
  	setattrofunc tp_setattro;
  	/* Space for future expansion */
+ #if PY_VERSION_HEX < 0x01050200
  	long tp_xxx3;
  	long tp_xxx4;
+ #else
+ 	PyBufferProcs *tp_as_buffer;
+ 	long tp_flags;
+ #endif
  
  	char *tp_doc; /* Documentation string */
+ #if PY_VERSION_HEX >= 0x01050200
+ 	long tp_xxx5;
+ 	long tp_xxx6;
+ 	long tp_xxx7;
+ 	long tp_xxx8;
+ #endif
  
  #ifdef COUNT_ALLOCS
  	/* these must be last */
***************
*** 278,299 ****
  
  /* Import the ExtensionClass CAPI */
  #define ExtensionClassImported \
!   (PyExtensionClassCAPI=PyCObject_Import("ExtensionClass","CAPI"))
  
  /* Make sure the C interface has been imported and import it if necessary.
     This can be used in an if.
     */
  #define MakeSureExtensionClassImported \
!   (PyExtensionClassCAPI || \
!    (PyExtensionClassCAPI=PyCObject_Import("ExtensionClass","CAPI")))
! 
  
  /* Export an Extension Base class in a given module dictionary with a
     given name and ExtensionClass structure.
     */
  #define PyExtensionClass_Export(D,N,T) \
!  if(PyExtensionClassCAPI || \
!    (PyExtensionClassCAPI= PyCObject_Import("ExtensionClass","CAPI"))) \
     { PyExtensionClassCAPI->Export(D,N,&T); }
  
  /* Convert a method list to a method chain.  */
--- 289,308 ----
  
  /* Import the ExtensionClass CAPI */
  #define ExtensionClassImported \
!   (PyExtensionClassCAPI= (struct ExtensionClassCAPIstruct *) \
!     PyCObject_Import("ExtensionClass","CAPI"))
  
  /* Make sure the C interface has been imported and import it if necessary.
     This can be used in an if.
     */
  #define MakeSureExtensionClassImported \
!   (PyExtensionClassCAPI || ExtensionClassImported)
  
  /* Export an Extension Base class in a given module dictionary with a
     given name and ExtensionClass structure.
     */
  #define PyExtensionClass_Export(D,N,T) \
!  if(MakeSureExtensionClassImported) \
     { PyExtensionClassCAPI->Export(D,N,&T); }
  
  /* Convert a method list to a method chain.  */
***************
*** 328,338 ****
--- 337,360 ----
  
  /* The following macro can be used to define an extension base class
     that only provides method and that is used as a pure mix-in class. */
+ #if PY_VERSION_HEX < 0x01050200 && !defined(COUNT_ALLOCS)
  #define PURE_MIXIN_CLASS(NAME,DOC,METHODS) \
  static PyExtensionClass NAME ## Type = { PyObject_HEAD_INIT(NULL) \
  	0, # NAME, sizeof(PyPureMixinObject), 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, \
  	0, 0, 0, 0, 0, 0, 0, DOC, {METHODS, NULL}, \
          EXTENSIONCLASS_BASICNEW_FLAG}
+ #elif PY_VERSION_HEX >= 0x01050200 && defined(COUNT_ALLOCS)
+ static PyExtensionClass NAME ## Type = { PyObject_HEAD_INIT(NULL) \
+ 	0, # NAME, sizeof(PyPureMixinObject), 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, \
+ 	0, 0, 0, 0, 0, 0, 0, DOC, 0, 0, 0, 0, 0, 0, 0, 0, {METHODS, NULL}, \
+         EXTENSIONCLASS_BASICNEW_FLAG}
+ #else /* either PY_VERSION_HEX >= 0x01050200 or defined(COUNT_ALLOCS) */
+ #define PURE_MIXIN_CLASS(NAME,DOC,METHODS) \
+ static PyExtensionClass NAME ## Type = { PyObject_HEAD_INIT(NULL) \
+ 	0, # NAME, sizeof(PyPureMixinObject), 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, \
+ 	0, 0, 0, 0, 0, 0, 0, DOC, 0, 0, 0, 0, {METHODS, NULL}, \
+         EXTENSIONCLASS_BASICNEW_FLAG}
+ #endif
  
  /* The following macros provide limited access to extension-class
     method facilities. */
===================================================================
RCS file: RCS/ExtensionClass.c,v
retrieving revision 1.30
diff -c -r1.30 ExtensionClass.c
*** ExtensionClass.c	1998/11/17 19:48:32	1.30
--- ExtensionClass.c	1999/06/29 00:40:47
***************
*** 33,39 ****
    USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
    DAMAGE.
  
!   $Id: ExtensionClass.c,v 1.30 1998/11/17 19:48:32 jim Exp $
  
    If you have questions regarding this software,
    contact:
--- 33,39 ----
    USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
    DAMAGE.
  
!   $Id: ExtensionClass.c,v 1.30 1998/11/17 19:48:32 jim Exp gregc $
  
    If you have questions regarding this software,
    contact:
***************
*** 54,60 ****
  "  - They provide access to unbound methods,\n"
  "  - They can be called to create instances.\n"
  "\n"
! "$Id: ExtensionClass.c,v 1.30 1998/11/17 19:48:32 jim Exp $\n"
  ;
  
  #include <stdio.h>
--- 54,60 ----
  "  - They provide access to unbound methods,\n"
  "  - They can be called to create instances.\n"
  "\n"
! "$Id: ExtensionClass.c,v 1.30 1998/11/17 19:48:32 jim Exp gregc $\n"
  ;
  
  #include <stdio.h>
***************
*** 3325,3330 ****
--- 3325,3336 ----
    /* Space for future expansion */
    0L,0L,
    "C classes", /* Documentation string */
+ #if PY_VERSION_HEX >= 0x01050200
+   0, 0, 0, 0,
+ #endif
+ #ifdef COUNT_ALLOCS
+   0, 0, 0, 0,
+ #endif
    METHOD_CHAIN(ExtensionClass_methods)
  };
  
===================================================================
RCS file: RCS/Acquisition.c,v
retrieving revision 1.24
diff -c -r1.24 Acquisition.c
*** Acquisition.c	1998/11/17 19:49:43	1.24
--- Acquisition.c	1999/06/29 00:43:01
***************
*** 33,39 ****
    USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
    DAMAGE.
  
!   $Id: Acquisition.c,v 1.24 1998/11/17 19:49:43 jim Exp $
  
    If you have questions regarding this software,
    contact:
--- 33,39 ----
    USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
    DAMAGE.
  
!   $Id: Acquisition.c,v 1.24 1998/11/17 19:49:43 jim Exp gregc $
  
    If you have questions regarding this software,
    contact:
***************
*** 807,812 ****
--- 807,818 ----
    /* Space for future expansion */
    0L,0L,
    "Wrapper object for implicit acquisition", /* Documentation string */
+ #if PY_VERSION_HEX >= 0x01050200
+   0, 0, 0, 0,
+ #endif
+ #ifdef COUNT_ALLOCS
+   0, 0, 0, 0,
+ #endif
    METHOD_CHAIN(Wrapper_methods),
    EXTENSIONCLASS_BINDABLE_FLAG,
  };
***************
*** 836,841 ****
--- 842,853 ----
    /* Space for future expansion */
    0L,0L,
    "Wrapper object for explicit acquisition", /* Documentation string */
+ #if PY_VERSION_HEX >= 0x01050200
+   0, 0, 0, 0,
+ #endif
+ #ifdef COUNT_ALLOCS
+   0, 0, 0, 0,
+ #endif
    METHOD_CHAIN(Wrapper_methods),
    EXTENSIONCLASS_BINDABLE_FLAG,
  };
***************
*** 913,919 ****
    /* Create the module and add the functions */
    m = Py_InitModule4("Acquisition", methods,
  	   "Provide base classes for acquiring objects\n\n"
! 	   "$Id: Acquisition.c,v 1.24 1998/11/17 19:49:43 jim Exp $\n",
  		     OBJECT(NULL),PYTHON_API_VERSION);
  
    d = PyModule_GetDict(m);
--- 925,931 ----
    /* Create the module and add the functions */
    m = Py_InitModule4("Acquisition", methods,
  	   "Provide base classes for acquiring objects\n\n"
! 	   "$Id: Acquisition.c,v 1.24 1998/11/17 19:49:43 jim Exp gregc $\n",
  		     OBJECT(NULL),PYTHON_API_VERSION);
  
    d = PyModule_GetDict(m);
===================================================================
RCS file: RCS/Missing.c,v
retrieving revision 1.8
diff -c -r1.8 Missing.c
*** Missing.c	1998/11/17 19:54:33	1.8
--- Missing.c	1999/06/29 00:44:12
***************
*** 33,39 ****
    USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
    DAMAGE.
  
!   $Id: Missing.c,v 1.8 1998/11/17 19:54:33 jim Exp $
  
    If you have questions regarding this software,
    contact:
--- 33,39 ----
    USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
    DAMAGE.
  
!   $Id: Missing.c,v 1.8 1998/11/17 19:54:33 jim Exp gregc $
  
    If you have questions regarding this software,
    contact:
***************
*** 47,53 ****
  
  static char Missing_module_documentation[] = 
  ""
! "\n$Id: Missing.c,v 1.8 1998/11/17 19:54:33 jim Exp $"
  ;
  
  #include "ExtensionClass.h"
--- 47,53 ----
  
  static char Missing_module_documentation[] = 
  ""
! "\n$Id: Missing.c,v 1.8 1998/11/17 19:54:33 jim Exp gregc $"
  ;
  
  #include "ExtensionClass.h"
***************
*** 271,276 ****
--- 271,282 ----
    "Missing values are used to represent numberic quantities that are\n"
    "unknown.  They support all mathematical operations except\n"
    "conversions by returning themselves.\n",
+ #if PY_VERSION_HEX >= 0x01050200
+   0, 0, 0, 0,
+ #endif
+ #ifdef COUNT_ALLOCS
+   0, 0, 0, 0,
+ #endif
    METHOD_CHAIN(NULL)
  };
  
===================================================================
RCS file: RCS/MultiMapping.c,v
retrieving revision 1.7
diff -c -r1.7 MultiMapping.c
*** MultiMapping.c	1998/11/17 20:20:17	1.7
--- MultiMapping.c	1999/06/29 00:45:21
***************
*** 33,39 ****
    USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
    DAMAGE.
  
!   $Id: MultiMapping.c,v 1.7 1998/11/17 20:20:17 jim Exp $
  
    If you have questions regarding this software,
    contact:
--- 33,39 ----
    USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
    DAMAGE.
  
!   $Id: MultiMapping.c,v 1.7 1998/11/17 20:20:17 jim Exp gregc $
  
    If you have questions regarding this software,
    contact:
***************
*** 222,227 ****
--- 222,233 ----
  	/* Space for future expansion */
  	0L,0L,0L,0L,
  	MMtype__doc__, /* Documentation string */
+ #if PY_VERSION_HEX >= 0x01050200
+ 	0, 0, 0, 0,
+ #endif
+ #ifdef COUNT_ALLOCS
+ 	0, 0, 0, 0,
+ #endif
  	METHOD_CHAIN(MM_methods)
  };
  
***************
*** 239,245 ****
        "MultiMapping", MultiMapping_methods,
        "MultiMapping -- Wrap multiple mapping objects for lookup"
        "\n\n"
!       "$Id: MultiMapping.c,v 1.7 1998/11/17 20:20:17 jim Exp $\n",
        (PyObject*)NULL,PYTHON_API_VERSION);
    d = PyModule_GetDict(m);
    PyExtensionClass_Export(d,"MultiMapping",MMtype);
--- 245,251 ----
        "MultiMapping", MultiMapping_methods,
        "MultiMapping -- Wrap multiple mapping objects for lookup"
        "\n\n"
!       "$Id: MultiMapping.c,v 1.7 1998/11/17 20:20:17 jim Exp gregc $\n",
        (PyObject*)NULL,PYTHON_API_VERSION);
    d = PyModule_GetDict(m);
    PyExtensionClass_Export(d,"MultiMapping",MMtype);
===================================================================
RCS file: RCS/ThreadLock.c,v
retrieving revision 1.5
diff -c -r1.5 ThreadLock.c
*** ThreadLock.c	1998/11/17 21:43:50	1.5
--- ThreadLock.c	1999/06/29 00:56:32
***************
*** 72,78 ****
    int count;
    long id;
  #ifdef WITH_THREAD
!   type_lock lock;
  #endif
  } ThreadLockObject;
  
--- 72,78 ----
    int count;
    long id;
  #ifdef WITH_THREAD
!   PyThread_type_lock lock;
  #endif
  } ThreadLockObject;
  
***************
*** 248,254 ****
    UNLESS(self = PyObject_NEW(ThreadLockObject, &ThreadLockType)) return NULL;
    self->count=-1;
  #ifdef WITH_THREAD
!   self->lock = allocate_lock();
    if (self->lock == NULL) {
      PyMem_DEL(self);
      self = NULL;
--- 248,254 ----
    UNLESS(self = PyObject_NEW(ThreadLockObject, &ThreadLockType)) return NULL;
    self->count=-1;
  #ifdef WITH_THREAD
!   self->lock = PyThread_allocate_lock();
    if (self->lock == NULL) {
      PyMem_DEL(self);
      self = NULL;
===================================================================
RCS file: RCS/Makefile.pre.in-1.5,v
retrieving revision 1.1
diff -c -r1.1 Makefile.pre.in-1.5
*** Makefile.pre.in-1.5	1999/06/29 00:48:26	1.1
--- Makefile.pre.in-1.5	1999/06/29 00:48:33
***************
*** 136,141 ****
--- 136,145 ----
  # Install prefix for architecture-dependent files
  exec_prefix=	$(prefix)
  
+ # Uncomment the following two lines for AIX
+ #LINKCC= 	$(LIBPL)/makexp_aix $(LIBPL)/python.exp "" $(LIBRARY); $(PURIFY) $(CC)
+ #LDSHARED=	$(LIBPL)/ld_so_aix $(CC) -bI:$(LIBPL)/python.exp
+ 
  # === Fixed definitions ===
  
  # Shell used by make (some versions default to the login shell, which is bad)
***************
*** 164,170 ****
  MAKEFILE=	$(LIBPL)/Makefile
  CONFIGC=	$(LIBPL)/config.c
  CONFIGCIN=	$(LIBPL)/config.c.in
! SETUP=		$(LIBPL)/Setup
  
  SYSLIBS=	$(LIBM) $(LIBC)
  
--- 168,174 ----
  MAKEFILE=	$(LIBPL)/Makefile
  CONFIGC=	$(LIBPL)/config.c
  CONFIGCIN=	$(LIBPL)/config.c.in
! SETUP=		$(LIBPL)/Setup.thread $(LIBPL)/Setup.local $(LIBPL)/Setup
  
  SYSLIBS=	$(LIBM) $(LIBC)
  
***************
*** 251,256 ****
--- 255,261 ----
  	 -e '/^LINKCC=/s/^LINKCC=[ 	]*\(.*\)/s%@LINKCC[@]%\1%/p' \
  	 -e '/^OPT=/s/^OPT=[ 	]*\(.*\)/s%@OPT[@]%\1%/p' \
  	 -e '/^LDFLAGS=/s/^LDFLAGS=[ 	]*\(.*\)/s%@LDFLAGS[@]%\1%/p' \
+ 	 -e '/^LDLAST=/s/^LDLAST=[      ]*\(.*\)/s%@LDLAST[@]%\1%/p' \
  	 -e '/^DEFS=/s/^DEFS=[ 	]*\(.*\)/s%@DEFS[@]%\1%/p' \
  	 -e '/^LIBS=/s/^LIBS=[ 	]*\(.*\)/s%@LIBS[@]%\1%/p' \
  	 -e '/^LIBM=/s/^LIBM=[ 	]*\(.*\)/s%@LIBM[@]%\1%/p' \
***************
*** 260,265 ****
--- 265,271 ----
  	 -e '/^SO=/s/^SO=[ 	]*\(.*\)/s%@SO[@]%\1%/p' \
  	 -e '/^LDSHARED=/s/^LDSHARED=[ 	]*\(.*\)/s%@LDSHARED[@]%\1%/p' \
  	 -e '/^CCSHARED=/s/^CCSHARED=[ 	]*\(.*\)/s%@CCSHARED[@]%\1%/p' \
+ 	 -e '/^SGI_ABI=/s/^SGI_ABI=[ 	]*\(.*\)/s%@SGI_ABI[@]%\1%/p' \
  	 -e '/^$L=/s/^$L=[ 	]*\(.*\)/s%@$L[@]%\1%/p' \
  	 -e '/^$P=/s/^$P=\(.*\)/s%^$P=.*%$P=\1%/p' \
  	 -e '/^$E=/s/^$E=\(.*\)/s%^$E=.*%$E=\1%/p' \



From thomas at xs4all.nl  Fri Jun  4 01:47:44 1999
From: thomas at xs4all.nl (Thomas Wouters)
Date: Fri, 4 Jun 1999 07:47:44 +0200
Subject: Changing environment variables into python variables
In-Reply-To: <19990604072728.I22133@xs4all.nl>; from Thomas Wouters on Fri, Jun 04, 1999 at 07:27:28AM +0200
References: <7j7f0t$165$1@nnrp1.deja.com> <19990604072728.I22133@xs4all.nl>
Message-ID: <19990604074744.J22133@xs4all.nl>

On Fri, Jun 04, 1999 at 07:27:28AM +0200, Thomas Wouters wrote:

> > What am I doing wrong here?  How can I, at runtime, define a variable
> > and assign it a value?

> You need to use exec(), not eval(). This is probably a FAQ, but i'm not sure
> if it's in the FAQ. From the Python Documentation however:

Oh, i forgot to warn you: You probably want to be very sure you dont have an
environment variable containing, for instance:

'You\'re a loony.", posix.system("rm -rf /"), "'

Dont forget -- you already imported posix or os so you can get to the
environment variables, which means you give them access to the 'remove' and
'system' functions, as well.

$ echo $test
You're a loony.", posix._exit(1), "
$ python
Python 1.5.1 (#1, Sep  3 1998, 22:51:17)  [GCC 2.7.2.3] on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import posix
>>> for var in posix.environ.keys():
...     command = var + '="' + posix.environ[var] + '"'
...     print command
...     exec(command)
... 
MACHTYPE="i386"
MAIL="/var/spool/mail/thomas"
OSTYPE="linux"
WINDOWID="25165838"
WINDOW="3"
VENDOR="intel"
test="You're a loony.", posix._exit(1), ""
$ echo $?
1

being-paranoid-doesn't-mean-the-bastards-aren't-out-to-get-you-ly y'rs, Thomas

-- 
Thomas Wouters <thomas at xs4all.net>

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



From greg.ewing at compaq.com  Tue Jun 22 16:41:54 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Wed, 23 Jun 1999 08:41:54 +1200
Subject: newbie idiom question
References: <m3emj4yezq.fsf@monsoon.swcp.com> <slrn7mvl8l.h16.jepler@craie.inetnebr.com>
Message-ID: <376FF512.21F0AE33@compaq.com>

Jeff Epler wrote:
> 
> Is this a wart on python, that "=" means three different things depending
> if you have a name, a [item] or a [slice:] on the LHS?

Syntactically, '=' doesn't stand on its own - you
have to read it in conjunction with the LHS.

In other words, there isn't an '=' operator, there
are 'x=', 'x.y=', 'x[y]=' and 'x[y:z]=' operators.

It's not a wart, it's a feature...

Greg



From graham at sloth.math.uga.edu  Mon Jun  7 12:31:44 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 7 Jun 1999 16:31:44 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net> <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net> <7j34l3$hbj$1@nnrp1.deja.com> <7j66b7$il5$2@cronkite.cc.uga.edu> <3756C693.2C910B29@usa.net> <7j6jnl$58f$3@cronkite.cc.uga.edu> <3757A820.CE3F195F@usa.net> <7j8obc$55v$2@cronkite.cc.uga.edu> <375BEDF8.DD14C4E6@usa.net>
Message-ID: <7jgs5g$q55$1@cronkite.cc.uga.edu>

Graham Matthews wrote:
> so I don't understand the above comment? As for circularly
> referred data it will be deleted in the order determined by the collector,
> but that's better than current Python which doesn't delete it at all.
Salvador =?iso-8859-1?Q?Fandi=F1o?= (fandino at usa.net) wrote:
: If I can't remove objects in the right order I would have to delete it
: explicitly the same way I have to do it now with ref counting. Lots of
: current __del__ methods will not perform properly and will raise
: exceptions if called after deleting related objects.
: You could use __del__ methods only for trivial task like freeing
: resources. It's an incomplete solution like reference counting is.

I really don't understand this. A lot of languages (eg. Java) have
garbage collection and finalisers. All these languages manage to delete
objects in an order that does not cause any problems (all finalisers etc
work). If a collector can do that in these languages why can't it do in
Python (__del__ methods are Python's finalisers)? 

I am very very confused by your question. You seem to have a misconception
about __del__ methods. They are finalisers, used only to close resources
like files, etc. In a garbage collection scheme the __del__ methods are
not used to implement the actual collection.

graham
-- 
             You wanna get close to me
              The feeling's so clear
            But I need some time to see
              Vision through my tears



From NO_SPAMnpirzkal at hq.eso.org  Sun Jun 27 16:40:50 1999
From: NO_SPAMnpirzkal at hq.eso.org (NO_SPAMnpirzkal at hq.eso.org)
Date: 27 Jun 1999 20:40:50 GMT
Subject: Python script background execution...
Message-ID: <7l628i$f2e$1@snatch.hq.eso.org>

Hi,

	I do not seem to be able to run my python script in the background (test.py 
&..). After a couple of second, the OS (Solaris 2.5.1) reports that the 
process has been suspended. 
	Does anyone know what I am missing?

		Nor




From Johann.Spies at p98.f112.n480.z2.fidonet.org  Tue Jun 29 15:03:55 1999
From: Johann.Spies at p98.f112.n480.z2.fidonet.org (Johann Spies)
Date: Tue, 29 Jun 1999 20:03:55 +0100
Subject: Please help me to understand classes
Message-ID: <0000158c@bossar.com.pl>

From: Johann Spies <jhspies at futurenet.co.za>

I have been playing around with python for some time now, but I am a
relative newbie as it is my hobby and normally I have very little time for
programming.  I am on leave now and want to write a program but I find
python a very difficult language to comprehend despite the fact that I
have a few years of experience in programming a language like Pascal.
Maybe it is because I have little experience in OOP.

Can somebody explain to me why the following program produces no output:
---------------------------------------------
class A:

    def druk():
        print 'Jy is in Class A'

class B:

    def druk():
        print 'Jy is in Class B'
        
a = A()
c = B()
c.druk
a.druk
---------------------------------------------


 --------------------------------------------------------------------------
| Johann Spies                                 Windsorlaan 19              |
| jhspies at futurenet.co.za                3201 Pietermaritzburg             |
| Tel/Faks Nr. +27 331-46-1310		       Suid-Afrika (South Africa)  |
 --------------------------------------------------------------------------

     "One thing have I desired of the LORD, that will I seek
      after; that I may dwell in the house of the LORD all
      the days of my life, to behold the beauty of the LORD,
      and to inquire in his temple."            
                               Psalms 27:4       



From amackay at starvision.com  Mon Jun 21 16:56:02 1999
From: amackay at starvision.com (Angus MacKay)
Date: Mon, 21 Jun 1999 13:56:02 -0700
Subject: while(s=infile.readline()):     # reading input
Message-ID: <376EA6E2.4E5EE527@starvision.com>

why can't I have syntax like this?
python is written in C after all isn't it?

this works fine:
  while infile.readline():
    print "Hi!"

this doe not:
  while s = infile.readline():
    print s

which is okay, but alas this does not work:
  while(s=infile.readline()):
    print s

is there any way to avoid using the ugly:
  while 1:
    s = infile.readline()
    if not s: break
    print s

cheers, Angus.



From ivanlan at callware.com  Fri Jun 18 15:20:23 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Fri, 18 Jun 1999 19:20:23 GMT
Subject: Cameron Laird
Message-ID: <376A9BF7.F7DCB30B@callware.com>

Hi All--
This is for Cameron Laird (Sorry, everyone else):  every time I try to
send you mail, even not on purpose (you were cc'd), after two or three
days, I *always* get this back:

The message that you sent was undeliverable to the following:
        claird at starbase.neosoft.com

Transcript of session follows:
        starbase.neosoft.com
        450 MX lookup failure

Possibly truncated original message follows:

<snippety-do-dah>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From amackay at starvision.com  Wed Jun 30 20:10:49 1999
From: amackay at starvision.com (Angus MacKay)
Date: Wed, 30 Jun 1999 17:10:49 -0700
Subject: readline completion
Message-ID: <377AB209.5A5B9D32@starvision.com>

how is readline completion done for python with the readline module?

I looked in the readlinemodule.c and saw that ESCESC was bound to
completion but it never completes on anything for me.

cheers, Angus.



From jlj at cfdrc.com  Fri Jun 25 08:29:16 1999
From: jlj at cfdrc.com (Lyle Johnson)
Date: Fri, 25 Jun 1999 07:29:16 -0500
Subject: C++ and Python
References: <3772AA7D.A7855B93@anisci.com>
Message-ID: <3773761C.1B822655@cfdrc.com>

Joel,

This is indeed a tricky area. I recommend you start by reading this paper,
several times:

    http://www.python.org/workshops/1995-12/papers/ahlstrom1.html

Then to see this implemented in a real extension, take a look at WPY itself,
wxPython or FXPy (and probably others?).

Good luck,

Lyle

Joel Davis wrote:

> I am trying to wrap a bunch of c++ classes so that virtual methods in
> derived classes can be written in python, and still callable from c++,
>
> SWIG and friends seem to be one-way, they let you subclass a C++ class
> in python via shadow classes or fancy extention types, but the
> subclasses behavior only shows up on the python side. I am trying to add
> the ability to mix python derived classes into a large c++ application.
>
> I guess what I'm asking is, the ORA Python book says (p626), "it's also
> possible to route C++ virtual method calls back down to methods in a
> python subclass, but it may require a stub-class on the C++ side of the
> interface, too (this is left as a Exercise to the Reader)." Are there
> any tools to help make this easier, or any good examples of this?
>
> thanks,
> joel
> --
> ----------------------------------------------------------------
>                                                 joeld at anisci.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/19990625/e23f07f3/attachment.html>

From ivnowa at hvision.nl  Sat Jun 19 02:33:54 1999
From: ivnowa at hvision.nl (Hans Nowak)
Date: Sat, 19 Jun 1999 06:33:54 GMT
Subject: Tkinter & socket question
In-Reply-To: <slrn7mllgk.dh.C.Kotsokalis@ppp02.softlab.ece.ntua.gr>
References: <slrn7mllgk.dh.C.Kotsokalis@ppp02.softlab.ece.ntua.gr>
Message-ID: <199906190632.IAA03215@axil.hvision.nl>

On 18 Jun 99, Constantinos A. Kotsokalis wrote:

> On Fri, 18 Jun 1999 21:54:11 GMT, Def P wrote:
> >2) I need to poll the connection periodically, say, every second. This would
> >not have been a problem in console mode... I don't know how to do this in a
> >Tkinter environment though. So, how can I do something every second (or every
> >n milliseconds, or whatever the time span may be)? Are threads the way to go?
> >I haven't used them before.
> 
> Well you can always check for a recv() that returns a zero-length
> buffer. That means that the other end has closed connection.

That is a nice way to detect disconnection, but I think the question 
was, how to receive messages from the other side? In console mode you 
can simply do a while loop with recv() in it and check what comes 
back. This is not so straightforward in Tk, you are either in 
mainloop or in your own loop, and unless there's some special event 
in Tk which can be called periodically, I don't know how to do this 
either.

--Hans Nowak (ivnowa at hvision.nl)
Homepage: http://fly.to/zephyrfalcon



From gmcm at hypernet.com  Wed Jun  9 00:48:00 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 8 Jun 1999 23:48:00 -0500
Subject: "Programming Python" still worthwhile?
In-Reply-To: <7jki66$qam$1@news.gstis.net>
Message-ID: <1283218560-70098357@hypernet.com>

Rick Robino writes:

> I've been watching this group for some time and just now getting
> warm to Python. Looking at source examples has been great, but for
> completeness' sake I'd like to get more of the "why" maybe found in
> a book. Since Python is at 1.5.2 and people are talking about 2.0 as
> if it's around the corner, I was wondering if anyone here could give
> me some advice as to whether or not "Programming Python" is useful,
> if the current edition is still current, if I should wait for the
> "Learning Python", or if I should use some other resources.

Python 2 is around a really, really big corner, and then down the 
block quite a ways. Python 1.6 sometime in 2000 (which will be 
backwards compatible), so no 2 until 2001 or 2002.

Learning Python is on the shelves. Haven't read it. Nobody's 
flamed it, and that's a _very_ good sign. There's only a few things 
out of date in PP, and you'll find updates on Mark's web site. I 
liked PP very much (and the difficult to find Internet Programming 
with Python, Watters et al, M&T books). When Mark wrote PP, he 
thought it might be the only Python book ever, so he stuffed in 
everything he could think of. So don't feel bad if you skip stuff.

> I expect to come up to speed fairly quickly and I'm just looking for
> those fundamental insights into OOP and Python that can build up the
> kind of inuition and understanding one needs in order to "get
> comfortable" connecting/creating the dots. Currently I only program
> small - medium programs in C and shell (Korn). Any sage pointers to
> resources?
> 
> TIA,
> 
> --Rick 
>                          ^^^^^^^^^^^^^^
> Rick Robino                                 
> rrobino at wavedivision.com
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list

- Gordon



From thstr at serop.abb.se  Wed Jun  9 07:39:03 1999
From: thstr at serop.abb.se (Thomas S. Strinnhed)
Date: Wed, 09 Jun 1999 13:39:03 +0200
Subject: Python/C API and setattr
Message-ID: <375E5257.8579541B@serop.abb.se>

Hi

I'm building an extention type for Python in C using NT, MSVC++.
It's built as a dll to link in at Runtime. 

I'm doing this for practice so the type is quite simple, it's
a CounterType, with an attribute called value (an int)
and the simple operations/methods inc() and dec().

I've looked pretty much at Mark Lutz's example StackType  from
Programming Python and I think I'm doing the same, but I've added
setattrfunc to be able to directly manipulate value as c.value = 23
and when I use it something goes wrong.
(inc() and dec() work fine)

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> from CounterType import *
>>> c = Counter()
>>> c.value = 23
Traceback (innermost last):
  File "<stdin>", line 1, in ?
SystemError: error return without exception set
>>> c.value
23
>>>
The value is set properly but I can't figure out the system error, it
seems to be something wrong with my setattrfunc. 

static void counter_setattr
	(counterobject *self, char *name, PyObject *v)
{
	if(strcmp(name, "value") == 0)
	{
		self->value = (int)PyInt_AsLong(v);
	}
}

But what would that be??
I have a hard time finding examples of how to do setattr so
any pointers would be great. Also if someone understands this error,
please post an answer.

My counterobject is this:
typedef struct
{
	PyObject_HEAD
	int value;
} counterobject;

Thanks 
 -- Thomas S. Strinnhed, thstr at serop.abb.se



From holger at phoenix-edv.netzservice.de  Wed Jun  9 09:35:38 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Wed, 9 Jun 1999 13:35:38 GMT
Subject: os.mkdir()
References: <375E5DCE.7005D71A@phoenix-edv.netzservice.de>
Message-ID: <375E6DAA.DD279E10@phoenix-edv.netzservice.de>

Another problem about this:

[source:]
thePath = "c:\\test"
aMode = "0777"
print "...updating filesystem: creating [%s](%s)" % (thePath, aMode)
os.makeDirs(thePath, aMode)

[result and error message:]
...updating filesystem: creating [c:\test](0777)
Traceback (innermost last):    
os.makeDirs(thePath, aMode)
AttributeError: makeDirs

What am I doing wrong?
(There's no problem at python-command-line-interpreter!?!)


Thanx for any hints!

Ciao,
Holger



From druid at princeton.crosswinds.net  Wed Jun  9 20:03:27 1999
From: druid at princeton.crosswinds.net (The Well of Latis)
Date: Thu, 10 Jun 1999 00:03:27 GMT
Subject: "Programming Python" still worthwhile?
References: <7jki66$qam$1@news.gstis.net> <7jl9r8$i1j$1@news06.btx.dtag.de>
Message-ID: <375effa9.1364893@netnews.worldnet.att.net>

On Wed, 9 Jun 1999 10:46:14 +0200, "Monika G?hmann"
<MGoehmann at t-online.de> wrote:

>"Learning Python" is just great!! I have only started reading but it already
>answered many of my newbie-questions. Some time ago I tried reading

I will agree with this, it seems to be a good book for beginners (both
at programming and Python). It was the first Python book that I
purchased, and would reccomend it for either a beginner or a more
experienced programmer who is new at Python.

>"Programming Python" but I find it quite hard to digest. LP seems much more
>readable for the beginner IMHO.

"Programming Python" seems to be more oriented to an experienced
programmer. I particularly liked the follow a simple application
approach from start to end. The book does jump around a bit however,
and I would only reccomend it for someone who is a somewhat
experienced programmer, who wants to learn Python. I do wish it came
with a more up to date CD however. 



Sparrow (Alex)                             | /|\ brewing on the net
http://www.crosswinds.net/~prin_druid      |      



From tismer at appliedbiometrics.com  Thu Jun  3 13:24:06 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Thu, 3 Jun 1999 17:24:06 GMT
Subject: starship.python.net
References: <000601beadc7$75b2ecb0$8b7125a6@cpda6686.mcit.com>
Message-ID: <3756BA36.E92FE999@appliedbiometrics.com>


jkraai at murl.com wrote:
> 
> Ok, I don't go to the starship root very often, but went there this morning.
> 
> It's _beautiful_.  Who did this excellent work?  Chris T?
> 
> Wow.  Clean, elegant, etc.

<blurb>

Not me alone. Meanwhile there are many helpers, and the
website layout is from Robin Friedrich again, as it was
with the first, now old ship.

I still do most of the administrivia, write scripts
for membership and mailing list management.

And the contents is made by the crew. This is the real
starship, all the rest is mostly kind of outfit.

If I have something to claim then only that it's
my genuine idea, and it turns out to be good.

</blurb>

a-blind-old-drunken-sailor-sometimes-finds-a-gem - ly y'rs - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From thomar at ifi.ntnu.no  Wed Jun 23 06:20:29 1999
From: thomar at ifi.ntnu.no (Thomas Røberg)
Date: Wed, 23 Jun 1999 12:20:29 +0200
Subject: Using C to execute Python
Message-ID: <7kqcds$4nu$1@kopp.stud.ntnu.no>

I have a C program and I want to call certain Python methods / functions
passing quite a few parameters.
How can I do this, in general fashion?
I would have to include python.h, I reckon - where can I find this, I've
been searching all over for it (is it just me???)
I'm no expert, far from, in Python...

Thomas
email: thomar at ifi.ntnu.no







From marcoxa at copernico.parades.rm.cnr.it  Mon Jun 28 09:37:33 1999
From: marcoxa at copernico.parades.rm.cnr.it (Marco Antoniotti)
Date: 28 Jun 1999 15:37:33 +0200
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <37682DD3.43923243@compaq.com> <7ka2be$qja$1@nntp6.u.washington.edu> <37776FFA.2D85@mailserver.hursley.ibm.com>
Message-ID: <lw7looxx7m.fsf@copernico.parades.rm.cnr.it>

Paul Duffin <pduffin at mailserver.hursley.ibm.com> writes:

	...

> Correct. With Tcl you can create new control structures which
> are indistinguishable from the built in ones, you cannot do
> that in Python. Python is much more rigid in its syntax although
> it does have a lot of nice hooks to allow objects to behave
> in different ways.

But you can do that much more easily and elegantly in the L-word
language :)

Cheers

-- 
Marco Antoniotti ===========================================
PARADES, Via San Pantaleo 66, I-00186 Rome, ITALY
tel. +39 - 06 68 10 03 17, fax. +39 - 06 68 80 79 26
http://www.parades.rm.cnr.it/~marcoxa



From tim_one at email.msn.com  Wed Jun  2 00:03:59 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 2 Jun 1999 00:03:59 -0400
Subject: Python 2.0
In-Reply-To: <87n1ykr0z9.fsf@ev.netlab.co.jp>
Message-ID: <001401beacac$f199e660$d89e2299@tim>

[Paul Prescod]
> But I think that the bigger problem with "real GC" is that C programs
> often hang on to references to Python objects and any kind of compacting
> GC would invalidate those references.

[Yukihiro Matsumoto]
> It's the common legend especially here in comp.lang.python.
>
> From my experience with Scheme and Ruby, both with real GC, the real
> GC is mostly very fast.  Fast enough no one notices its "hang" on most
> cases.  The legend was formed up by old GC implementation and slower
> machines of the past, I guess.

Matz, this one is just a misunderstanding.  Paul's "hang" doesn't refer to
time.  It's more like "retain", as in "keep a pointer to".  "compacting" is
the key there:  compacting GC (one that relocates live objects) creates
problems for C extensions that have saved away ("hung on to") the addresses
of objects.  But GC doesn't have to be compacting in order to reclaim
cycles; Paul is likely reacting to the "self-contained universe"
inter-language hostility common among languages with compacting GC (like
Icon or Java).

betting-a-nickel-python2-won't-compact-ly y'rs  - tim





From Robert.Meegan at wcom.com  Wed Jun  9 10:46:04 1999
From: Robert.Meegan at wcom.com (Robert Meegan)
Date: Wed, 9 Jun 1999 14:46:04 GMT
Subject: Tkinter Question
Message-ID: <Pine.A41.4.10.9906090941210.65174-100000@barney.mcit.com>

Greetings ---

I have an alarm display that consists of a  Frame that contains multiple
frames within it, stacked one above the other. When a user clicks on one
of the subframes to acknowledge it, I want the subframe to disappear and
the parent frame to resize itself to remove the gap.

Getting the subframe to remove itself is no problem. I bound the
<Button-1> event to a method and had the frame destroy itself. Life is
good.

On the other hand, I can't figure out how to make the parent frame resize
itself.

Suggestions will be greatly appreciated.

--- Robert

--------------------------------
Robert Meegan
MCIWorldCom - Cedar Rapids, Iowa
319.375.2416




From hniksic at srce.hr  Mon Jun  7 14:21:07 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 07 Jun 1999 20:21:07 +0200
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7iqhkb$d7s@wiscnews.wiscnet.net> <7iu7ma$d33$1@cronkite.cc.uga.edu> <02hfonzsx2.fsf@rtp.ericsson.se> <02909zzou0.fsf@rtp.ericsson.se> <87zp2f4qrz.fsf@pc-hrvoje.srce.hr> <wkd7z858c3.fsf@turangalila.harmonixmusic.com>
Message-ID: <87wvxfzxbg.fsf@pc-hrvoje.srce.hr>

Dan Schmidt <dfan at harmonixmusic.com> writes:

> | Since when does Python provide closures?
> 
> The bytecodehacks module allows the creation of closures as I
> understand them.

I've never seen bytecodehacks.  Where can I get it?



From matomira at iname.com  Thu Jun 10 09:24:00 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Thu, 10 Jun 1999 15:24:00 +0200
Subject: Why is tcl broken?
In-Reply-To: <1283099377-77267484@hypernet.com>
References: <375F82C7.CF0F0477@iname.com>
Message-ID: <4.1.19990610150908.00d09790@exchsrv>

At 08:54 AM 6/10/99 -0500, Gordon McMillan wrote:
>Fernando Mato Mira writes:
>> 
>>   I'm trying to collect a list of all the _current_ issues making
>>   tcl
>> a bad language choice.
>
>Why? Convincing your boss to use Python over TCL for your next 
>project is one thing, but in general this is a zero-sum (or worse) 
>game.
>
>The TCLers I know value its simplicity and accept its limitations. 
>Telling them they're wrong will only piss them off.

A `TCLer'  here is going to give a talk presenting the virtues of
tcl to the rest of the organization. As a hard-core lisper, it is obvious
to me that the core philosophical basis of the `New Jersey school' (i.e. the
whole idea of scripting languages) is broken, and obviously the syntax
is tasteless to me. But I'm looking for irrefutable signs of bad language
design that result in low maintenability. For example, dynamic scoping,
ambiguities, etc. Things like the `string issue' are also interesting if they
can be exposed in a way that the answer is not "So what?".

I'm not putting in question the virtues of Tk, as you can access that
through STk, for example. But if there are bad things there, I'd like
to know, too. (eg: is there a lot of tcl code making a Tk interface run?
If so, does speed ever become an issue? I assume that Tk works well
enough that you'd never have to look at the tcl code, and live happily
inside your reflective Scheme world)

Thanks,

Fernando D. Mato Mira                    
Real-Time SW Eng & Networking            
Advanced Systems Engineering Division
CSEM                             
Jaquet-Droz 1                   email: matomira AT acm DOT org
CH-2007 Neuchatel                 tel:       +41 (32) 720-5157
Switzerland                       FAX:       +41 (32) 720-5720

www.csem.ch www.vrai.com
<http://ligwww.epfl.ch/matomira.html>ligwww.epfl.ch/matomira.html




From mwh21 at cam.ac.uk  Sun Jun 27 14:01:01 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 27 Jun 1999 19:01:01 +0100
Subject: Python for large projects?
References: <3774b5d1.690922@news.force9.net>  <T6bd3.2883$U6.12014@newsr2.twcny.rr.com> <slrn7naup9.jk.thantos@brimstone.mecha> <RGid3.2957$U6.12374@newsr2.twcny.rr.com> <87n1xm85mf.fsf@lisboa.i-did-not-set--mail-host-address--so-shoot-me> <Jhsd3.3090$U6.12410@newsr2.twcny.rr.com>
Message-ID: <m3lnd5cyle.fsf@atrus.jesus.cam.ac.uk>

"Darrell" <news at dorb.com> writes:

> > If it is only one object, one can try to wrap it in a class and
> > overwrite setattr, setitem etc.
> > This would inhibit that one changes the buffer data accidently. And
> > put the data into an immutable sequence type (string, tuple).
> >
> > __Janko
> >
> I'll give that a look, but they can always get to it if they want. Maybe
> that's ok, just worries me.
> 
> --Darrell

If you are really really worried about not letting code you let the
user write mucking up your internals, use RExec. I don't think there's
any other way, and I believe it's pretty configurable, although I
haven't used it myself.

HTH
Michael



From arcege at shore.net  Wed Jun 23 13:06:28 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 23 Jun 1999 17:06:28 GMT
Subject: How to obtain an instance's name at runtime?
References: <7kqpu8$7u1$1@nnrp1.deja.com>
Message-ID: <ou8c3.1077$7X1.277903@news.shore.net>

Dinu C. Gherman <gherman at my-deja.com> wrote:
: This is a nasty question, perhaps, so I appologize in
: advance if its stupidity level is far above the accepted
: average in this newsgroup...

: I would like to do the following:

:>>> class C:
: ...    pass
: ...
:>>>
:>>> c = C()
:>>> C.__class__.__name__ # ok
: 'C'
:>>> c.__name__           # not ok
: Traceback (innermost last):
:   File "<interactive input>", line 0, in ?
: AttributeError: __name__
:>>>

: Question: How to obtain an instance's name at runtime?
: That is I'd like to see this happen (one was or the other):

:>>> c.__name__
: 'c'

: I've checked the reference manual, but it's silent about
: this topic or I'm too blind...

: Fearing-there's-no-such-solution'ly,

: Dinu

There is not reliable solution because instances are not named except as
part of a possible class behavior (__init__ setting self.__name__).
(It is the behavior of classes themselves to have names.)

But you can at least find some of the names bound to this object (within
the current scope, or beyond).

  def whatname(obj):
    import sys
    try:
      raise SystemError
    except:
      tb = sys.exc_info()[-1]
    f = tb.tb_next.tb_frame
    locals = f.f_locals
    globals = f.f_globals
    del tb, f
    for key, item in locals.items() + globals.items():
      if item is obj:
        break
    else:
      raise NameError, 'object not found'  # kind of a reverse namelookup ;)
    return key

This is inaccurate because the object could be passed down from another
module, thru numerous function calls.  And the same could be done for
classes:
  >>> class C:
  ...   pass
  >>> A = C
  >>> c = A()
  >>> c.__class__.__name__
  'C'
  >>>

To make the behavior, then set the "__name__" attribute in the
constructor.
  class C:
    def __init__(self, name='c'):
      self.__name__ = name
  b = C('b')

It may not be what you want, but then Python doesn't quite work like
you might suspect. :)

  -Arcege




From matz at netlab.co.jp  Wed Jun  2 05:57:31 1999
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 02 Jun 1999 18:57:31 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
 <87pv3iq6ss.fsf@ev.netlab.co.jp> <7iv86a$iud@news.dns.microsoft.com>
 <u3r9nwtd9q.fsf@ares.sys1.ngo.okisoft.co.jp>
 <87so8bnsd6.fsf@ev.netlab.co.jp>
 <u37lpndko6.fsf@ares.sys1.ngo.okisoft.co.jp>
Message-ID: <87pv3eorg4.fsf@ev.netlab.co.jp>

Hisao Suzuki <suzuki611 at okisoft.co.jp> writes:

|This sounds a sort of advocacy of your Ruby...

Well, maybe..

|Any way, the idea of destructors with "real" GC is a conceptual
|contradiction.  Such GC emulates infinite memories and objects
|are never destroyed *virtually*.  Thus the destructors should
|never be invoked with real GC.

I thought GC is the way to reclaim the unreferenced objects.  I think
it is not directly related to infinite memory space, even though you
can choose it as the conceptial model for GC.

|Certainly you may make your GC routine retrieve not only
|memories but also various resources (such as file descriptors),
|but it is highly error prone to make the routine invoke
|ARBITRARY destructors specified by arbitrary users.  Such
|invocation may introduce serious indeterminism or
|unpredictability into the behavior of user's program.

Do you mean finalizers with real GC (as in Java) is the source of
indeterminism?  Maybe you're right.

BTW, did I say Ruby can invoke arbitrary destructors specified by
arbitrary users?  It can't.

|# Well, in this respect, Ruby might be more indeterministic
|# than Python.

I don't think it is.  But, well, I have to confess Ruby is unstable
than Python yet.

|Even if Python has "real" GC someday, IMHO, the GC should be
|used only to imitate infinite resources.  For the sake of
|conceptual simplicity, the future Python should still use
|reference counting both to invoke __del__ methods and to free
|storages; the GC routine should be used as the last resort.

I agree with it.  I wouldn't claim to remove ref counting from Python.
Combination of ref counting and real GC may be the best for Python,
accomplishes both compatibility and reliability.

						matz.



From jo at delorges.in-berlin.de  Wed Jun 30 20:49:32 1999
From: jo at delorges.in-berlin.de (Jo Meder)
Date: 1 Jul 1999 00:49:32 GMT
Subject: Weird commandline behaviour with SunOS 5.5.1
Message-ID: <slrn7nleos.gf5.jo@delorges.in-berlin.de>


Dear Python community!

On a SunOS 5.5.1 virtual machine I observed, that commandline 
parameters to scripts seem to get mixed up or misinterpreted: 

Entering "python  script.py -r" at the shell prompt does call
python correctly but the  interpreter will barf because there
is no valid option named "-r" for Python itself.  Python will
then stop with a helpful message about allowed options to the 
invocation  of the  our beloved interpreter. According to the 
documentation, this should have  never happened since options 
after the scriptname should be given to the script instead of
the interpreter.

Changing the invocatione to "python script.py %r" (e.g.) does
help, but I'm a little reluctant to use kludges like changing
the option indicator from something widely used to a nonstan-
dard character.

This interpreter was compiled  using "configure" without any 
additional options except a special path for installation.

I observed this originally with Python 1.4 but the problem is
still there in 1.52. I originally  attributed it to stupidity
on my part, but now I'm unsure whom to blame. Is this a shell
misbehaving or misconfigured or is it a pythonic problem?

Thanks in advance for  pointers to documentation I missed or 
for any helpful hints.

Best regards,

                Jo.
		



From johnga at primenet.com  Sat Jun 19 21:35:53 1999
From: johnga at primenet.com (John E. Garrott)
Date: Sat, 19 Jun 1999 18:35:53 -0700
Subject: Newbie question
References: <376BBB7E.A82DA789@primenet.com> <199906191928.VAA20205@axil.hvision.nl>
Message-ID: <376C4579.C7B38E25@primenet.com>

Hans Nowak wrote:
> 
> On 19 Jun 99, John E. Garrott wrote:
> 
> > I am have a problem with the program below.
> > The intent is, eventually, to get a list and display it,
> > in a list box, then print the selected row
> > from the listbox, but I can't get the callback
> > to work.
> >
> > The function dbs.databases() works, putting,
> > in this case, the databases owned by myself
> > into the listbox.  When I click on the listbox
> > items they highlight as expected, but, apparently,
> > list_cb() is never called.
> >
> > I've read all the available literature on
> > binding and signals, tried numerous variations
> > on what is here.  No results.  This program
> > shows no errors, so is syntactically(sp?)
> > correct.  Just doesn't work.
> >
> > What am I doing wrong?
> >
> > Thanks in advance,
> >
> > John
> > --------------------------------------------------------
> > import dbs
> > import string
> > from Tkinter import *
> >
> > class SelectionBox:
> >
> >       def __init__(self, master=None):
> >               self.frame = Frame(master, name="frame")
> >               self.frame.pack(expand=1, fill=BOTH)
> >
> >               self.frame.listbox = Listbox(self.frame, name='listbox')
> >               self.frame.listbox.pack(expand=1, fill=BOTH)
> >               self.frame.listbox.bind('Button-1', self.list_cb)
> >
> >               for item in dbs.databases():
> >                       self.frame.listbox.insert(END, item)
> >
> >       def list_cb(event):
> >               print "Here we are!"
> 
> I don't know for sure, but shouldn't this be
> 
>     def list_cb(self, event):
> 
> ?
> 
> I don't know if that's the solution for your problem, though... It's
> not like *I* know much about Tkinter. :o)
> 
> Veel liefs,
> 
> --Hans Nowak (ivnowa at hvision.nl)
> Homepage: http://fly.to/zephyrfalcon

I believe you are right, but that doesn't work either.  No errors, but
also
no action.

Thanks for your reply,

John



From faassen at pop.vet.uu.nl  Wed Jun  2 08:45:44 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Wed, 02 Jun 1999 14:45:44 +0200
Subject: "?:", "a and b or c" or "iif"
References: <374EC384.446B@mailserver.hursley.ibm.com> <Pine.A41.4.10.9905281330180.64654-100000@barney.mcit.com>
Message-ID: <37552778.BAB4F4DA@pop.vet.uu.nl>

Robert Meegan wrote:
 
> On Fri, 28 May 1999, Paul Duffin wrote:
> > You mean like a proper *** US *** English speaker :)
 
> Do you mean to say that there's any other kind? <wicked grin>

Sure, Dutch English speakers. :)

Regards,

Martijn



From bwarsaw at cnri.reston.va.us  Thu Jun 17 17:17:29 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Thu, 17 Jun 1999 17:17:29 -0400 (EDT)
Subject: [ANNOUNCE] JPython 1.1 beta 2
Message-ID: <14185.26089.959983.601576@anthem.cnri.reston.va.us>

I'm happy to announce the release of the second beta in the JPython
1.1 series.  JPython is the 100% Pure Java implementation of the
Python language, which provides seamless integration with the Java
platform.  JPython 1.1beta2 is compatible with most Java 1.1 and 1.2
JDKs.

JPython 1.1beta2 is essentially a bug fix release to 1.1beta1, fixing
reported problems in the jpythonc tool, other reported bugs and
improving compatibility with CPython.  See below for a summary.

You can download JPython 1.1beta2 from the JPython home page.  Please
visit

    http://www.jpython.org/

Here's a summary of what's changed since 1.1beta1:

    - Updates and bug fixes to jpythonc tool.  New registry entries
      python.jpythonc.compiler, python.jpythonc.classpath, and
      python.jpythonc.compileropts.  getopt style long/short option
      names (old options are deprecated).  New -J/--compileropts
      option.
    - syncing string object methods with experimental CPython 1.6
      string methods.  This includes new optional arguments on
      startswith() and endswith() and the moving of join() to a method 
      of the separator string.
    - Many other bug fixes and CPython syncs

Enjoy,
-Barry

Barry A. Warsaw                    Corp. for Nat'l Research Initiatives (CNRI)
bwarsaw at cnri.reston.va.us          1895 Preston White Drive, Suite 100
bwarsaw at python.org                 Reston, VA USA 20191-5434
http://www.python.org/~bwarsaw     703.620.8990 / (fax) 703.620.0913




From emile at fenx.com  Thu Jun 24 20:28:17 1999
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 24 Jun 1999 17:28:17 -0700
Subject: getstate, setstate and pickle
References: <X%vc3.1830$VL2.283108@news.direcpc.com>
Message-ID: <43Ac3.1839$VL2.284144@news.direcpc.com>

I've identified my problem.  pickle.py uses map(id,..) as part of its
test routine.  I tried to use the same to test.  map(id,...) (which
likes sequences) doesn't work on my class instances.

The script below illustrates that dumps(r) ==dumps(loads(dumps(r))) but
that r <> loads(dumps(r)).

My question now moves to: How can I test/verify my pickled data?  or is
it that is the dumps match I'm ok?  (sound good to me, but are there
traps?)

from pickle import *
class R:
 def __init__(self):
  self.a = {}
  self.b = {}

print r == loads(dumps(r))
print dumps(r) == dumps(loads(dumps(r)))

regards,

--

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






From john at mediamanager.com.sg  Mon Jun  7 01:28:32 1999
From: john at mediamanager.com.sg (John Francis)
Date: Mon, 07 Jun 1999 13:28:32 +0800
Subject: extracting contents from a text file
Message-ID: <375B5880.992B0C6C@mediamanager.com.sg>

Hi,

I would like to extract the text contents residing in another server and
display it as html format using python. Can anyone help me I am new to
python.

Thanks and regards
John




From tlhall at blackhole.nyx.net  Tue Jun 22 16:44:46 1999
From: tlhall at blackhole.nyx.net (Tom Hall)
Date: Tue, 22 Jun 1999 20:44:46 GMT
Subject: W95 binary output from CGI script
Message-ID: <930084286.377082@iris.nyx.net>

I'm working on a CGI script for Windows 95, and trying to send
binary data out stdout. Every time I try to send a \x0a it spits
out \x0d\x0a. This really messes up a JPEG, for example.

How can I get python to print just \x0a ?
I've tried
   sys.stdout.write("\x0a")
and
   rawout=os.fdopen(sys.stdout.fileno(),"wb")
   rawout.write("\x0a")
and even
   rawout=os.fdopen(sys.stdout.fileno(),"wb")
   os.write(rawout.fileno(),"\x0a")
but I still get that damn \x0d.

How do I send only the binary data I want out to stdout ?




From tim_one at email.msn.com  Tue Jun  8 22:27:33 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 8 Jun 1999 22:27:33 -0400
Subject: Python 2.0
In-Reply-To: <7jjrgv$h16$1@cronkite.cc.uga.edu>
Message-ID: <000b01beb21f$a17ff340$2fa22299@tim>

>[Salvador]
> I have been using Java since it was invented and I haven't programed a
> finaliser ever. They should run under so unknown and inconsistent
> environment that they are not usable at all (except for freeing external
> resources).

[Graham Matthews]
> I am not sure I understand this. Can you explain to me how a finaliser
> can run in an unknown environment. Surely everything the finaliser needs
> is referenced by the finaliser (either directly or indirectly) and hence
> is still there.

The caveats are explained well in the Java Language Spec, section 12.6.  In
short, you can't know which thread a finalizer will run in; you can't be
sure that an object is unreachable at the time its finalizer is invoked; you
can't know *when* a finalizer will be invoked; even if it is unreachable at
the time its finalizer is invoked, you can't be sure that it won't become
reachable again *after* the finalizer returns; and in the case of cyclic
object structures nothing is specified about the order in which the objects'
finalizers will be called (indeed, they may all run simultaneously).

All Java defines is enough so that the Java runtime itself never chokes on
e.g. an internal null pointer, i.e., your "still there" guarantee.  Say you
have a mail server holding a ref to a mail client, which in turn holds a ref
back to the mail server.  Can the client object tell the server it's going
away at finalization time?  Not if the server object shut down the
communications link beforehand because the server's finalizer happened to
get called first.  Etc.  Yes, the references are "still there", but without
a lot of pain there's no guarantee they're still in a *useful* state.

"Therefore, we recommend that the design of finalize methods be kept simple
and that they be programmed defensively, so that they will work in all
cases."

good-advice-in-any-language-ly y'rs  - tim





From alrice at swcp.com  Thu Jun  3 16:21:45 1999
From: alrice at swcp.com (Alex Rice)
Date: Thu, 03 Jun 1999 14:21:45 -0600
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl>
Message-ID: <3756E3D9.E44D2B39@swcp.com>


Martijn Faassen wrote:
...

> Okay, sorry for all this, it's out of line, I just needed to blow off
> steam. :)

As the composer John Cage would have told you, silence is music as much
as tones and harmonies are music. silence == whitespace ?

Alex Rice



From wtanksle at dolphin.openprojects.net  Wed Jun 30 16:35:49 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Wed, 30 Jun 1999 20:35:49 GMT
Subject: Does the first argument of a method *has to be* self?
References: <19990630222121.A1121@optiplex.palga.uucp>
Message-ID: <slrn7nkvt4.5ee.wtanksle@dolphin.openprojects.net>

On Wed, 30 Jun 1999 22:21:21 +0200, Gerrit Holl wrote:
>Hello,

>I never saw code where the first argument of a method wasn't self. Is it
>in theory possible?

It's always self, although you're free to call it anything you want.

>Gerrit.

-- 
-William "Billy" Tanksley



From mwh21 at cam.ac.uk  Tue Jun 22 16:02:12 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 22 Jun 1999 21:02:12 +0100
Subject: YAS to the "Reading line-by-line" Problem
References: <Pine.SUN.3.95-heb-2.07.990622183356.3454A-100000@sunset.ma.huji.ac.il> <376FDF70.4756AD38@mindspring.com>
Message-ID: <m3d7yo2ed7.fsf@atrus.jesus.cam.ac.uk>

"Jesse D. Sightler" <jsight at mindspring.com> writes:

> Ack is raising an exception for a normally occuring event really a
> _good_ idea?  Generally exceptions are only used for things that
> actually are error conditions, and it would seem to me that using them
> as nothing more than a fancy flow-control construct is a sure path to
> nasty spaghetti code.

Well, every for loop terminates with an IndexError exception...  But
yes, it does make a mockery of the name.

The performance of exceptions in Python isn't really a concern, if
you're coming from C++ where exceptions can take thousands of times
longer than normal function returns.

> I really do not like this idea.  :)
> 

No, me neither. Oh well, each to their own.
M



From gutier at intergate.bc.ca  Tue Jun 22 00:54:25 1999
From: gutier at intergate.bc.ca (Gerald Gutierrez)
Date: 21 Jun 1999 21:54:25 -0700
Subject: TCL/Tk
References: <B3wb3.67$Rj1.29977@typhoon-sf.snfc21.pbi.net>
Message-ID: <m2zp1sdeda.fsf@intergate.bc.ca>

You can always compile it yourself, can't you? Make your own distribution.

"Zigron" <zigron at jps.net> writes:

>     I personally don't like that TCL/Tk is now included in the main
> distribution,
> even if Tk is the defacto GUI library for it. And, even if you can't use
> IDLE
> without it.
>     But, lots of people do, I'd expect.. so I've a sugguestion :)
> 
>     Have a seperate distribution that /doesn't/ have TCL/Tk in it? :-)
> 
> --Stephen



From ivanlan at callware.com  Wed Jun  2 17:15:48 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Wed, 2 Jun 1999 21:15:48 GMT
Subject: [Off topic] Weird...
References: <l03102809b37b3f90b2a0@[193.78.237.145]>
Message-ID: <37559F04.2062C65D@callware.com>

Hi All--

Just van Rossum wrote:
> 
> The mpeg movie is a bit old, but otherwise this is quite a weird page:
>         http://www.java.net/
> (yes there is Python content)
> 
> Just

Well, I tried to view the movie, and it insisted I needed a plugin.  So
I got Apple Quicktime 4, which was recommended.  I installed it.  It
doesn't support the format of the movie; I followed the link on the web
page, which led me to the download page for M$ Internet Explorer 4.0,
which M$ insisted would let me play the movie, instead of NetScape
Communicator.

Sorry, but it's not worth it to me.  ;-)  Maybe you could give us a
blow-by-blow description, Just?

<we-don't-need-no-stinking-ie-4>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From arcege at shore.net  Tue Jun 22 17:01:25 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 22 Jun 1999 21:01:25 GMT
Subject: creating dictionnaries
References: <376F82A6.82497D2B@der.edf.fr>
Message-ID: <FQSb3.973$7X1.249189@news.shore.net>

Christian Caremoli <Christian.Caremoli at der.edf.fr> wrote:
: Hi,
: Dictionnaries can be created like that :
: d={'a':1,'b':2}

: By calling a function you create a dictionnary like that :
: def f(**d):
:    return d

: d=f(a=1,b=2)

: I would like to be able to create dictionnaries with some similar syntax
: like keyed tuples :
: d=(a=1,b=2)

: Is there a way to do that ?

In a word, No.  The structure of the language would perform a namespace
lookup on "a" and "b" instead of taking them as strings.  Also a tuple
is still an expression, and Python does not allow assignments inside
expressions.  This means that you would get a SyntaxError exception
during bytecode compilation.  And even if you didn't, it would be
likely that you would get a NameError exception.

  -Arcege

Note: I haven't looked at bytecodehacks, but that might let you do what
you wish.  Since it would be extremely non-portable, I wouldn't suggest
it for published or long-term code.




From tcpeter at my-deja.com  Tue Jun  8 14:37:26 1999
From: tcpeter at my-deja.com (tcpeter at my-deja.com)
Date: Tue, 08 Jun 1999 18:37:26 GMT
Subject: In what order does Python read dictionary keys?
Message-ID: <7jjnt5$snk$1@nnrp1.deja.com>

Hello, all.  I've checked the docs and queried DejaNews for this but I
couldn't come up with an answer.  I wrote a script to convert rgb
values to hex for when I'm doing colors in HTML pages and don't have
Photoshop handy.  The code appears below.  My question is, how does
Python decide in which order to read the keys?  Notice that I have the
dictionary set up in this order: red, blue, green.  (I mistyped it while
writing the code).  But when I ran the code, the options were presented
in the desired order. (A trifle creepy, if you ask me.  Turing would be
delighted.)  Anyway, how does Python decide the order. Thanks.


#! /usr/bin/python

# The first thing to do is create a dictionary containing the
# questions to be asked where the color name is the dictionary key.

entries = {'red':'Type the red value, please: ',
	'blue':'Type the blue value, please: ',
	'green':'Type the green value, please: '}

for key in entries.keys():		# Now get a loop going
					# through the dictionaries

	foo = input(entries[key])	# Create a variable to hold
					# the return from the prompt
					# for each key

	if (foo) < 16:			# Since hex numbers don't
					# have leading zeroes below
					# F, I added them.

		print 'The hex equivalent for',foo,'is 0%X' % foo
	else:
		print 'The hex equivalent for',foo,'is %X' % foo


Tim Peter
(Not him, a different guy)


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From movits at lockstar.com  Thu Jun 17 16:27:33 1999
From: movits at lockstar.com (Mordy Ovits)
Date: 17 Jun 1999 13:27:33 PDT
Subject: Parsing HTML
Message-ID: <376984E2.5055E13D@lockstar.com>

What is the best way to parse HTML into a Python data structure, allow me change it,
and output it as HTML?
-- 
o Mordy Ovits
o Cryptographic Engineer
o LockStar Inc.
---------------------------------------------------------------------------
#!/usr/local/bin/python 
from sys import*;from string import*;a=argv;[s,p,q]=filter(lambda x:x[:1]!=
'-',a);d='-d'in a;e,n=atol(p,16),atol(q,16);l=(len(q)+1)/2;o,inb=l-d,l-1+d
while s:s=stdin.read(inb);s and map(stdout.write,map(lambda i,b=pow(reduce(
lambda x,y:(x<<8L)+y,map(ord,s)),e,n):chr(b>>8*i&255),range(o-1,-1,-1)))



From news at dorb.com  Fri Jun  4 12:05:00 1999
From: news at dorb.com (Darrell)
Date: Fri, 4 Jun 1999 12:05:00 -0400
Subject: pyscript help
Message-ID: <BRS53.233$h14.1175@newsr2.twcny.rr.com>

Has anyone seen this ?
D:\python\win32comext\axscript\client>pyscript.py --debug
Traceback (innermost last):
  File "D:\python\win32comext\axscript\client\pyscript.py", line 400, in ?
    Register()
  File "D:\python\win32comext\axscript\client\pyscript.py", line 388, in
Registe
r
    ret = win32com.server.register.UseCommandLine(PyScript)
  File "d:\python\win32com\server\register.py", line 401, in UseCommandLine
    apply(RegisterClasses, classes, flags)
  File "d:\python\win32com\server\register.py", line 358, in RegisterClasses
    addPyComCat, dispatcherSpec, clsctx, addnPath)
  File "d:\python\win32com\server\register.py", line 227, in RegisterServer
    regCat = _cat_registrar()
  File "d:\python\win32com\server\register.py", line 85, in _cat_registrar
    pythoncom.IID_ICatRegister
pywintypes.com_error: (-2147467262, 'No such interface supported', None,
None)



I was able to control word from a test script.
Could this have something to do with being a server ?
And I this works also:
>>> import pythoncom
>>> import pywintypes
>>>

I'm running NT4 SP4, IE5 and Option pack 4.


Thanks--
--Darrell





From greg.ewing at compaq.com  Wed Jun 16 20:14:14 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Thu, 17 Jun 1999 12:14:14 +1200
Subject: Solution to finalisation problem [Re: fork()]
References: <000301beb6c9$a01d2960$b2a22299@tim> <7k4gec$6vg$1@cronkite.cc.uga.edu>
Message-ID: <37683DD6.4C42DC3E@compaq.com>

Graham Matthews wrote:
> 
> All
> objects will be in a state in which they can be finalised in pass 1 since
> nothing is deleted in pass 1. I don't see the problem here at all.

I think what we need at this point is some concrete examples.
What sorts of things do people actually *do* with __del__
methods at the moment?

As far as I can see, the only thing you really need a
finalizer for is freeing an external resource. (The only
non-external resource is memory, and the GC takes care
of that.)

If each external resource is wrapped up in an object which
encapsulates everything needed to identify that resource,
then that object can have a finaliser which does its job
without caring about the state of any other object.

The only way I can see that the 2-pass collection scheme
can fail is if the finaliser of object A destroys something
needed by the finaliser of object B. But if each object
having a finaliser looks after just one external resource,
how can that happen?

Anyone have a real-world example?

Greg



From greg.ewing at compaq.com  Sun Jun 27 20:07:38 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Mon, 28 Jun 1999 12:07:38 +1200
Subject: How do can you make a variable static?
References: <3772AE33.AA764DFA@alliedsignal.com>
Message-ID: <3776BCCA.11FD9915@compaq.com>

The reason you're getting a NameError is that
the name SymHandDict is in a different module
from the one where you're referring to it.
The "global" statement you put in doesn't make
any difference to that.

For what you're doing here, you don't need or
want a "global" statement at all. You only need
one if you want to assign to a module-level variable
from within a function defined in the same module.
You're not doing that here. You're changing the
contents of the object to which SymHandDict refers,
but you're not assigning to the variable SymHandDict 
itself -- it still refers to the same object afterwards.

In general, the way you refer to a name in a *different*
module is by prefixing it with the name of the module
(which you must also import using an "import" statement),
e.g. in module foo:

import test

  def foo():
      handlelist=test.SymHanDict.keys() 
      ...
      test.SymHanDict[handle]=0

***BUT*** I wouldn't recommend you do it exactly that
way in this case, because then your two modules would
be importing from each other, a situation which can
lead to difficulties and is best avoided if possible.

If there will only ever be one instance of SymHanDict,
I'd suggest moving it either into module foo or some
other module which foo imports. 

If you put it in module foo, you will be able to refer
to it directly without any prefix:

  # in module foo:

  SymHanDict = {0:0}

  def foo():
      handlelist = SymHanDict.keys()
      ...
      SymHanDict[handle] = 0

If module test needs to decide how to initialise
SymHanDict, it can do this:

  # in module test:
  import foo
  foo.SymHanDict = {0:0}

> I don't understand the explanation of
> globals in python.

Keep in mind that there are only two namespaces you
can reference without using any prefix: (1) the
"local" scope (parameters and local variables of the
current function) and (2) the so-called "global"
scope, which is not completely global, but is
only the top level of the current module (i.e.
the one containing the piece of code being
executed).

To refer to anything in another module, you have
to use the module.name syntax.

Finally, the "global" statement is only needed in
one special circumstance, which is quite rare:
when you want to assign to a name in the current
module from within a function, without prefixing it 
with a module name. For example,

  x = 42

  def blarg():
    global x
    x = 17

  blarg()
  print x

prints 17. If the "global" statement weren't there,
the assignment would assign to a *local* variable
called x, and the global x would be unchanged.

However, note that you *don't* need a global
statement to do this:

   x = 42

   def blarg():
     print x

nor do you need one to do this:

   import banana

   def blarg():
     banana.x = 17

   blarg()
   print banana.x

Hope I've helped more than I've confused,
Greg

   




> 
> Can anyone clarify? What's the workaround? For a variety of reasons I
> won't go into right now, I really can't pass the dictionary as an
> argument.
> 
> Thanks
> 
> Pete
> 
> **** the following contained in main module test.py
> *****************************
> from foo import foo
> 
> SymHanDict={0:0}
> 
> foo()                            #each call to foo() should add items to
> the dictionary
> foo()
> foo()
> 
> *** the following contained in module foo.py ************************
> def foo():
>     global SymHanDict
> 
>     handlelist=SymHanDict.keys()          # get the list of existing
> handles
>     handlelist.sort()                     # sort the list, then
> 
>     handle=handlelist[len(handlelist)-1]        # find largest handle
>     handle=handle+1                         # generate new handle, and
>     SymHanDict[handle]=0                  # add it to dictionary along
> with
>                                           # a default value of zero.
> 
>     print SymHanDict
> 
> --
> 
> -----------------------------------------------
> | The opinions expressed here are my own,     |
> | and not necessarily those of my employer.   |
> -----------------------------------------------



From bwarsaw at cnri.reston.va.us  Mon Jun 14 11:01:50 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Mon, 14 Jun 1999 11:01:50 -0400 (EDT)
Subject: Variations on a theme by "pound bang"
References: <3763e75f.3265255@news.iol.ie>
	<3763F77E.6094AEE9@appliedbiometrics.com>
	<3764185B.7BCBC488@Lugoj.Com>
Message-ID: <14181.6494.968443.827788@anthem.cnri.reston.va.us>

>>>>> "JL" == James Logajan <JamesL at Lugoj.Com> writes:

    JL> Alas, this is a myth. Not all Unixen put env in /usr/bin, some
    JL> put it in /bin. I can't recall if it was Solaris (probably
    JL> not) or LynxOS (probably) that put env in /bin.

Solaris puts env in both /bin and /usr/bin.  You're right though
/usr/bin/env isn't 100% portable but it's by far the best shot you've
got.

FWIW, in Mailman we hardcode the paths to python based on the one we
find in your $PATH at config/install time.  Very important for CGI
based apps as many people have pointed out, and it's a workable, if
involved, solution if you've got an install step.  For most Python
apps I think it's not worth it, and /usr/bin/env is your best bet.

-Barry



From piers at cs.su.oz.au  Wed Jun  2 02:38:44 1999
From: piers at cs.su.oz.au (Piers Lauder)
Date: Wed, 02 Jun 1999 16:38:44 +1000
Subject: Problem with Cookie module and cPickle
Message-ID: <129.78.10.43.126.21332.928305966.756.11436>

I wonder if anyone has a fix for the following problem?

We've been using the Cookie.py module (by Timothy O'Malley)
successfully for some time, but ever since the conversion
to Python-1.5.2, it tends to dump core. Here's an example:

	% python
	Python 1.5.2 (#10, May 11 1999, 15:32:03)  [GCC 2.8.1] on sunos5
	Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
	>>> import Cookie; C = Cookie.Cookie()
	>>> C.load("mary=ht;")
	Segmentation Fault (core dumped)

My system is:
	% uname -a
	SunOS staff 5.6 Generic_105181-05 sun4u sparc SUNW,Ultra-2

(It also dumps core on Linux, so it's not just a sparc problem.)

It seems to me that this is probably a cPickle problem, as changing
the use of cPickle to Pickle, the core dump goes away.

I'm using version: cPickle.c,v 1.63 1999/02/05 01:40:06 jim Exp







From rdr_nl at my-deja.com  Fri Jun 11 21:53:32 1999
From: rdr_nl at my-deja.com (Henk Jansen)
Date: Sat, 12 Jun 1999 01:53:32 GMT
Subject: embedding question (unresolved name)
Message-ID: <7jseiq$vbv$1@nnrp1.deja.com>

I just want to import Aaron Watters' kjbuckets module in an embedded
Python interpreter. Loading it from within a normal (non-embedded)
Python session goes fine. However, when I (for instance) add the
following lines in the Python1.5.2/Demo/embed/demo.c example:

        PyRun_SimpleString("import kjbuckets\n");
        PyRun_SimpleString("a=kjbuckets.kjDict()\n");
        PyRun_SimpleString("print a\n");

the PyType_Type variable in the kjbuckets module is unsresolved and  the
following error is raised:

Traceback (innermost last):
  File "<string>", line 1, in ?
  File "/mnt/opt/lib/python1.5/site-packages/kjbuckets/__init__.py",
line 2, in ?
    import sys, kjbuckets
ImportError: /mnt/opt/lib/python1.5/site-packages/kjbucketsmodule.so:
undefined symbol: PyType_Type
Traceback (innermost last):
  File "<string>", line 1, in ?
NameError: kjbuckets

Can anyone help me to find the cause?

Thanks,

Henk Jansen
--
 =====================================================================
 Henk Jansen -- Delft University of Technology -- Dept. of Mathematics
 hjansen at math.tudelft.nl
 http://ta.math.tudelft.nl/users/hjansen/H.Jansen.html


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From news at dorb.com  Sat Jun 19 21:17:58 1999
From: news at dorb.com (Darrell)
Date: Sat, 19 Jun 1999 21:17:58 -0400
Subject: Proxying HTTP - end of message?
References: <7kgkbk$uso$1@nnrp1.deja.com>
Message-ID: <GlXa3.1011$U6.3896@newsr2.twcny.rr.com>

http://www.cyberclip.com/webdebug/

--
--Darrell
Andrew Cooke <andrew at andrewcooke.free-online.co.uk> wrote in message
news:7kgkbk$uso$1 at nnrp1.deja.com...
>
>
> Hi,
>
> I'm writing a small http proxy (I know a million already exist -
> my program isn't important, understanding the bug is).  I have a
> couple of threads, spinning away, one copying from client to server
> and one from server to client.  When there is no more data to read
> from the server the blocking read returns None and I close both
> sockets (one socket is proxy-client, the other proxy-server).
>
> But my browser continues to wait.  If I press "stop" I have all the
> page, but the browser thinks the download has been interrupted.
>
> What am I missing?  It appears as though the browser isn't detecting
> that the connection has been closed - why not?
>
> For what it's worth, this is on NT, I'm extending SocketServer, and
> using 1.5.2.  But I suspect this is a problem with my logic rather
> than any piece of software....
>
> Any sugegstions appreciated (I can post/forward code - it's my own
> work, not my employer's).
>
> Thanks,
> Andrew
>
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.





From rhww at erols.com  Mon Jun  7 09:36:17 1999
From: rhww at erols.com (R Wentworth)
Date: Mon, 07 Jun 1999 09:36:17 -0400
Subject: Indentation and grouping proposal
References: <000001beb0a1$bc540a00$5b9e2299@tim>
Message-ID: <375BCAD1.BE6E391B@erols.com>


Tim Peters wrote:
> 
> [R Wentworth, suggests "_(" and ")_" as alternative grouping constructs]
> 
> Note that this won't fly, since "_(" already has a meaning in the grammar
> (call the function whose name is _).

Actually, the suggested tokens were _{ and }_, with braces rather than
parentheses.  Admittedly braces and parenthesis can be hard to distinguish
on a computer screen.  I'm well aware that "_(" is syntactically legal and 
meaningful, and therefore not a candidate.

> > [careful argument & analysis deleted]
> > ...
> > I haven't looked at the Python parser, but naively, this sort
> > of "enhancement" would seem to be relatively easy.  If there were
> > official blessing for the concept, I'd be willing to look into
> > doing the necessary work.
> 
> Guido's last public word on the subject is in Tools/Scripts/pindent.py.

That's a nice tool, and one I wasn't aware of.  Could be applied to some
of the examples I cited, albeit awkwardly.  (And it doesn't allow a
1-line program, which is what I would want in some situations.)

> Go way back (pre-DejaNews; in python.org's archives, though), and you'll
> find some words, and even some patches to implement alternatives; e.g., see
> Donald Beaudry's
> 
>     http://www.python.org/search/hypermail/python-1994q2/0763.html

Interesting.  (Again, it isn't quite as general as what I was proposing,
and does not allow for 1-line programs.)

> The parser hasn't changed all that much since then, although it's likely
> changed a lot more than Guido's opinion <wink>.

So, I guess having this "blessed" is unlikely. :(

(I do suspect that the "embedding" issue is more important now than it was
five years ago, what with the rise of HTML templating systems and the like.
To my mind that makes the issue somewhat less frivolous.  Oh, well.)

> IIRC,

"IIRC" ?

> Steven Majewski christened the alternatives "frowning Guidos", in
> honor of everyone's favorite some 5 years ago:
> 
>     if whatever :(
>     print "no whitespace!" ):

Plausible, if inauspicious.

> or-was-that-"(:"-and-":)"?-ly y'rs  - tim

Bob



From adi at i-data.com  Thu Jun 10 06:35:48 1999
From: adi at i-data.com (Adi)
Date: Thu, 10 Jun 1999 10:35:48 +0000
Subject: Asserting object class
References: <375F8986.73BA3C5@i-data.com> <m3wvxc9xis.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <375F9504.CA3FE6CA@i-data.com>

Thank you, but:

Michael Hudson wrote:

> Adi <adi at i-data.com> writes:
>
> > In a method, I'd like to assert that a parameter is an object which is
> > of a specific class or has this class as it's base class.
>
> I think that this will do what you want:
>
> def method(self,p):
>     assert isinstance(p,self.__class__)
>

Actually it does'nt. When this method is called, it is called on a subclass
of the superclass I wan't to assert the parameter is of.

So it fails. What I do now is:

class X:
    def m(self, p):
        assert isinstance(p, X)
        ...

>
> the following also does the same:
>
> def method(self,p):
>     assert issubclass(p.__class__,self.__class__)
>

I have'nt tested, but I suspect it will do the same.

Anders

--
i-data is a leading vendor of printing connectivity solutions,
e-forms software, and networking products.
More information is available at www.i-data.com

Disclaimer: I speak for myself, not my employer.




From akuchlin at mems-exchange.org  Thu Jun  3 16:51:07 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: Thu, 3 Jun 1999 16:51:07 -0400 (EDT)
Subject: threads
In-Reply-To: <8790a1njeu.fsf@pc-hrvoje.srce.hr>
References: <3756C078.5D998D3B@icrf.icnet.uk>
	<14166.51558.850416.153999@bitdiddle.cnri.reston.va.us>
	<8790a1njeu.fsf@pc-hrvoje.srce.hr>
Message-ID: <14166.60091.353043.829598@amarok.cnri.reston.va.us>

Hrvoje Niksic writes:
>Jeremy Hylton <jeremy at cnri.reston.va.us> writes:
>> The short answer is no.  The Python interpreter has a lock that
>> prevents allows only a single thread to execute at a time.
>
>Eek.  That's depressing.  Isn't supporting threads and not allowing
>simultaneous execution a bit weird?

    I don't think so, because the common case is probably not the one
where lots of work is being done in Python code.  Threads are probably 
being used for one of 2 purposes:

      1) I/O-heavy programs, where most of the threads are waiting for
some external event such as new data arriving on a socket.  Few
threads will run simultaneously at the same time, so it's not much of
a problem in this case.

      2) Computation-heavy programs, where you want multiple threads
to be scheduled on several CPUs.  In this case, your computational
heavy lifting is almost certainly not being done in Python, but in
wrapped calls to some C or Fortran library that does matrix math,
database access, or whatever.  If the person who wrapped the library
put in the proper Py_BEGIN_ALLOW_THREADS/END_ALLOW_THREADS
invocations, then other threads can run while the C code crunches, so
most of the threads spend most of their time not holding the
interpreter lock as they crank away.

      The single interpreter lock matters if you have several threads
all executing pure Python code at the same time; my suggestion is that
this is not the common case.  One exception might be Zope 2; I'm not
sure what the design of the threaded ZODB looks like, or if it avoids
this problem.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
If you go looking for the Ladies... well, I don't know that that's such a good
idea. You might find them.
    -- Euryale, in SANDMAN #61: "The Kindly Ones:5"





From alex at mindlube.com  Thu Jun 24 22:09:29 1999
From: alex at mindlube.com (Alex Rice)
Date: 24 Jun 1999 20:09:29 -0600
Subject: Emacs? (was Re: total idiot question: +=, .=, etc...)
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0o3q.toe.bernhard@alpha1.csd.uwm.edu> <7kr7ii$dss$1@nnrp1.deja.com> <87pv2lk5fl.fsf_-_@pbl.ddns.org>
Message-ID: <m3pv2lhvza.fsf@monsoon.swcp.com>

Preston Landers <planders at mail.utexas.edu> writes:

> I'm thinking of doing a similar thing in (X)Emacs.  It's my editor of
> choice though I've never actually *programmed* something into emacs
> itself.  I'm wondering if anyone's already done a similar thing so I
> can spare the trouble (and the learning exercise, bah.)  
> 
> I'm thinking the best way might be to hook in to abbrev mode or
> something... I guess I have some homework to do in this area.  I was
> just wondering if someone's already done something like this. 

Not that I know of -- and that doesn't mean much --, but since I'm
always willing to learn more about Emacs (wink-wink, nudge-nudge),
here are some macros that kind of approximate it...

In Emacs, creating a macro is easy: C-x( [type stuff here] C-x)
Binding the macros to sensible keystrokes seems to be the hard part in
this case. I challenge you to find a way to integrate these into
emacs's python-mode. :-)

Put this in your .emacs

8< cut

;; Macros for Python  
;; Alex Rice <alex at mindlube.com>

(defalias '++
  (read-kbd-macro "C-SPC ESC b ESC w ESC f = C-y +1 C-e"))
(defalias '--
  (read-kbd-macro "C-SPC ESC b ESC w ESC f = C-y -1 C-e"))
(defalias '+=
  (read-kbd-macro "C-SPC ESC b ESC w ESC f = C-y + C-e"))
(defalias '-=
  (read-kbd-macro "C-SPC ESC b ESC w ESC f = C-y - C-e"))
(defalias '*=
  (read-kbd-macro "C-SPC ESC b ESC w ESC f = C-y * C-e"))
(defalias '/=
  (read-kbd-macro "C-SPC ESC b ESC w ESC f = C-y / C-e"))

8< cut

Now for example typing

spam M-x++

yields

spam=spam+1 

BTW, _Learning GNU Emacs_ by O'Reilly & Assoc is a good gentle intro
to Emacs if you are like me and sometimes can't figure it out just
from the Info browser.

Cheers,

Alex Rice



From tim_one at email.msn.com  Fri Jun  4 01:15:13 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 4 Jun 1999 01:15:13 -0400
Subject: "The Python Way"
In-Reply-To: <9906031407150D.00552@quadra.teleo.net>
Message-ID: <001101beae49$39af7bc0$949e2299@tim>

[Fredrik Lundh]
> sure looks like the "community" thinks that changing the
> language is more important than using it...

[Patrick Phalen]
> Yeah; now that you mention it, the "let's fix Python" discourse
> has been approaching my pain threshold, as it does once in awhile.

That's the spirit <wink>!

> (Have you noticed that, during such periods, both Guido and Tim
> seem to become singularly quiet?)

It's just old.  python.org has archives of the pre-c.l.py days, and you'll
find many of the old-timers arguing vigorously about all the "current
issues" 8 years ago already.  There have been few notable advances in the
art of GC-- let alone whitespace --since then, and repetition is tiring.
That doesn't mean people to whom it's all fresh shouldn't have fun clawing
each others' eyes out, though.  OTOH, doesn't mean they should, either.

> Not that Python shouldn't be open to critique,

It's far too late to complain about 1.5.2 <0.9 wink>.

> but the more I use and learn about the language, the more I find myself
> appreciating the nice balance and heft Guido gave to it. Yet there doesn't
> seem to be a single document that sums up that "aesthetic," but rather it
> tends to appear piecemeal, over time, mostly in the Wisdom of Chairman
> Tim.

Eh?  I make shit up as I go along -- & Guido often appears to think that my
fabrications would be good positions for him to have pretended he had all
along too.   OTOH, after all these years, I *still* have to lecture him
about his true vision of what Python should be ... although he learns pretty
fast, for a Dutch guy <wink>.

> I'd like to suggest something as a sort of balm for those of us who
> come here to try to learn to work with, rather than against, the grain
> of Python:
>
> Would both Guido and TIm Peters be willing to collaborate on a short
> paper -- call it "The Python Way" for lack of a better title -- which
> sets out the 10-20 prescriptives they might offer to those who come to
> Python from other languages and immediately want to find a way to bend
> it into uncomfortable positions -- (implement closures, etc.).

And forego all those juicy Guido Channeling fees?!  Not me.

> What I have in mind is sort of a very brief Strunk-&-White-like
> "Elements of Style" for Python, which suggests fundamental idiomatic
> recommendations for operating within the spirit of the language. A
> distillation of Python Zen is what I'm talking about -- something to go
> off and contemplate when the "fix Python now" decibels become a bit
> much.
>
> Tim? Guido?

Clearly a job for Guido alone -- although I doubt it's one he'll take on
(fwiw, I wish he would too!).  Here's the outline he would start from,
though <wink>:

    Beautiful is better than ugly.
    Explicit is better than implicit.
    Simple is better than complex.
    Complex is better than complicated.
    Flat is better than nested.
    Sparse is better than dense.
    Readability counts.
    Special cases aren't special enough to break the rules.
    Although practicality beats purity.
    Errors should never pass silently.
    Unless explicitly silenced.
    In the face of ambiguity, refuse the temptation to guess.
    There should be one-- and preferably only one --obvious way to do it.
    Although that way may not be obvious at first unless you're Dutch.
    Now is better than never.
    Although never is often better than *right* now.
    If the implementation is hard to explain, it's a bad idea.
    If the implementation is easy to explain, it may be a good idea.
    Namespaces are one honking great idea -- let's do more of those!

There you go:  20 Pythonic Fec^H^H^HTheses on the nose, counting the one I'm
leaving for Guido to fill in.  If the answer to *any* Python design issue
isn't obvious after reading those -- well, I just give up <wink>.

let's-hear-it-for-lambda-in-curly-assignment-stmts-ly y'rs  - tim





From bwarsaw at cnri.reston.va.us  Fri Jun  4 11:36:30 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Fri, 4 Jun 1999 11:36:30 -0400 (EDT)
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7imr6b$1s0$1@nnrp1.deja.com>
	<7j8o26$55v$1@cronkite.cc.uga.edu>
Message-ID: <14167.62078.226416.757006@anthem.cnri.reston.va.us>

What we need is CODE.  Somebody's gotta step up to the plate and
generate some patches that people can play with.  The best idea I've
heard yet is to graft some GC on top of the already existing
refcounting scheme.  Sounds good to me, but I'm no GC expert so I have
no idea how to do this.  I'm just sitting here waiting for someone to
pick a scheme, snag an implementation, and post some hacks.  Then
we'll have a starting point from which the real fun can begin!

garbage-in-garbage-out-who's-gonna-garbage-collect-this-thread?-ly y'rs,
-Barry



From tismer at appliedbiometrics.com  Sat Jun  5 16:21:33 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Sat, 5 Jun 1999 20:21:33 GMT
Subject: Memory leak under Idle?
Message-ID: <375986CD.5DD8BE79@appliedbiometrics.com>

Howdy,

I was testing my stackless Python which now does also
most builtins without involving the C stack, and I thought
to use Idle as a testing horse.
By chance, I discovered a very high reference count
of None, after Idle was run.

Then I tested it with standard python 1.5.2, and discovered
the same problem. (Which makes me happy about my version,
but a little concerned about TK related stuff).

Here my DOS session and the refcounts of None:


D:\python\PNSpeed>python
Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import sys
>>> sys.getrefcount(None)
164
>>> import os
>>> os.chdir("d:/python/tools/idle")
>>> execfile("idle.py")
>>> sys.getrefcount(None)
2232
>>> execfile("idle.py")
>>> sys.getrefcount(None)
2326
>>>

What's going on here?
Just cyclic references piling up?

ciao - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From Klaus.Alexander.Seistrup at p98.f112.n480.z2.fidonet.org  Tue Jun 29 20:36:39 1999
From: Klaus.Alexander.Seistrup at p98.f112.n480.z2.fidonet.org (Klaus Alexander Seistrup)
Date: Wed, 30 Jun 1999 01:36:39 +0100
Subject: [Python|Crypto] on handhelds (was Re: Thanks for Python...)
Message-ID: <00001595@bossar.com.pl>

From: Klaus Alexander Seistrup <kas at maps.magnetic-ink.dk>

Mordy Ovits wrote:

> #!/usr/local/bin/python
> from sys import*;from string import*;a=argv;[s,p,q]=filter(lambda x:x[:1]!=
> '-',a);d='-d'in a;e,n=atol(p,16),atol(q,16);l=(len(q)+1)/2;o,inb=l-d,l-1+d
> while s:s=stdin.read(inb);s and map(stdout.write,map(lambda i,b=pow(reduce(
> lambda x,y:(x<<8L)+y,map(ord,s)),e,n):chr(b>>8*i&255),range(o-1,-1,-1)))

'-',a);d='-d'in a;e,n=atol(p,16),atol(q,16);l=(len(q)+1)/2;o,inb=l-d,l-1+d
ValueError: unpack list of wrong size

??


  //Klaus

-- 
???[ Magnetic Ink ]??????????????????????????????????????????????????????????
???[ http://www.magnetic-ink.dk/ ][ maps == mail abuse prevention system ]???



From algei at mailexcite.com  Tue Jun 15 08:51:07 1999
From: algei at mailexcite.com (algei at mailexcite.com)
Date: 15 Jun 1999 12:51:07 GMT
Subject: We offer all kinds of products from China at very competitive price
Message-ID: <7k5i7r$da213784@rain>

Grand Treasure Industrial Limited

Contact Information

We are a manufacturer and exporter in Hong Kong for all kinds of plastic
products,
We export to worldwide markets. Recently , we join-ventured with a bag
factory in China produce all kinds of shopping , lady's , traveller's
bags.... visit our page and send us your enquiry by email now.
Contact Address :
Rm. 1905, Asian Trade Centre , 79 Lei Muk Rd, Tsuen Wan , Hong Kong.
Telephone : ( 852 ) 2408 9382

Fax : ( 852 ) 2408 9620

Emailto:grandtreas at mailexcite.com

Contact Person  :  Mr. Vicent Li ( Exporting and Marketing Manager)

Home Page: http://business.fortunecity.com/boutique/409





From gerrit at nl.linux.org  Mon Jun  7 11:24:15 1999
From: gerrit at nl.linux.org (Gerrit Holl)
Date: 7 Jun 1999 15:24:15 GMT
Subject: Suggestion: Dutch Python mailinglist
References: <slrn7lli2q.ia.gerrit@gerrit.palga.uucp>
Message-ID: <slrn7lnp5m.kg.gerrit@gerrit.palga.uucp>

Gerrit Holl wrote:
> This e-mail is in two languages. Look below for the Dutch version.
> Deze e-mail is in twee talen. Kijk beneden voor de Nederlandse versie.
> 
> Hello all,
> I'm a 13-years-old Dutch boy, and my English is not good. Now I have the
> chance to set up a Dutch python mailinglist. I know there are many Dutch
> python-users, but most of them know English very well. So will there be
> any users who'll subscribe to such a mailinglist, or will it remain very quiet?
> 

Okay, I heard of you all that my English isn't really bad. Thanks for the
compliments everyone...

If I can't find the anwser on my current question, I'll ask it here.

regards,
Gerrit.

-- 
The Dutch Linuxgames website.		De Nederlandse Linuxgames pagina.
Everything about games on Linux.	Alles over spelletjes onder Linux.
Site address:				http://linuxgames.nl.linux.org
Personal homepage:			http://nl.linux.org/~gerrit/



From gmcm at hypernet.com  Wed Jun  2 11:48:14 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 2 Jun 1999 10:48:14 -0500
Subject: Why won't this work???
In-Reply-To: <7j3e5q$nrq$1@rtpnews.raleigh.ibm.com>
Message-ID: <1283783753-36099768@hypernet.com>

Jr. King writes:

> Uhhhh, excuse me but, I did point out the problem.  The
> PyEval_CallObject call is failing everytime.  I have it commented in
> the code, I guess that wasn't good enuff.  Excuse me, but that is
> the way I have done and seen it done on c.l.c and the questions are
> usually answered minus the attitude.

Generally, one is expected trim the problem to the minimal code that 
exposes the problem. You are also expected to include all the error 
messages and as much context as possible. As I recall, you post said 
"What's wrong with this?" as was followed by a whole lot of code. I 
scrolled to the bottom, didn't see any error messages or output and 
went on to the next post.

> And I will be doggoned if I
> give out my darn email address on newsgroups again, I don't need all
> that trash about porno.

Actually, MLM spam outnumbers porno spam about 5 to 1 lately. It's a 
minor inconvenience for the extra help you'll receive. And unlike 
telemarketing, spam mail won't interrupt your delicious dinner of 
Spam, Spam, Spam, Eggs and Spam.

- Gordon



From pduffin at mailserver.hursley.ibm.com  Mon Jun 28 08:50:05 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Mon, 28 Jun 1999 13:50:05 +0100
Subject: Python as a shared library on Unix
References: <376102C9.5648@mailserver.hursley.ibm.com> <m3vhcuam2b.fsf@atrus.jesus.cam.ac.uk> <emacs-smtp-532-14177-8119-350099@export.andrew.cmu.edu>
Message-ID: <37776F7D.4A7B@mailserver.hursley.ibm.com>

Brian E Gallew wrote:
> 
> Then <mwh21 at cam.ac.uk> spoke up and said:
> > Paul Duffin <pduffin at mailserver.hursley.ibm.com> writes:
> >
> > > Is it possible to build Python itself as a small executable and a
> > > shared library, rather than an archive library which of
> > > course results in a monolithic executable.
> 
> Sure it is.  I've done it.  The problem is, doing this is even more
> platform dependent than just building shared libraries.  When I did
> it, it was a "by-hand" process, where I manually modified some files,
> forced all object files to be relocatable code, manually converted
> libpython.a to libpython.so.1.5.2, and so on.  What makes this
> difficult is that some Modules want to reference things in libpython,
> and sometimes it means linking the modules against -lpython, and

How does this work when Python is a monolithic executable ?

What do Modules link to then ?

They obviously cannot link to the python library as that will result
in duplicate stuff.

I thought that when Python loaded modules it had platform specific code
to manage the binding of the symbols to those exported from Python
itself (except on Windows which does not support this at all).

It sounds to me as though you need something like the Stubs mechanism
that Tcl and its extensions are starting to use.

> sometimes not.  It's ugly, and not for the faint of heart.  Further,
> the utility of such a hack quickly became questionable for me,
> considering that I'm never embedding python itself.
> 


-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From tlng at phileo.com.my  Thu Jun 24 05:43:12 1999
From: tlng at phileo.com.my (tlng at phileo.com.my)
Date: Thu, 24 Jun 1999 09:43:12 GMT
Subject: tkMessageBox returns 'can't invoke "frame" command:' error
References: <7kktvv$550$1@nnrp1.deja.com> <c75so7lklce.fsf@pc142.cosc.canterbury.ac.nz>
Message-ID: <7ksujf$1jp$1@nnrp1.deja.com>

In article <c75so7lklce.fsf at pc142.cosc.canterbury.ac.nz>,
  Timothy R Evans <tre17 at cosc.canterbury.ac.nz> wrote:
> The actuall error is caused by the Tk mainloop cannot be restarted,
> obviously Tkinter was never designed to support what you're doing
> here.

Oh, you mean I can't reload my script if my script contains Tkinter
calls? How do you test your Tkinter scripts then? What I usually do is,
I write my Python script in a text editor, switch to the Python
interpreter and reload my script to see if my script is working. Since
I can't do this with Tkinter scripts, does that mean everytime I want
to retest my Tkinter scripts, I have to quit Python and restart Python?

> The cause of your problem seems be partly that you actually
> expected this to work, putting lots of code in a module
> initialisation is not good style.

That bit about "putting lots of code in a module initialisation" just
went over my head. What did I do that constitutes "module
initialisation"? I've reposted my sample script below so you can point
it out to me easily.

-----
def askyesno():
	if tkMessageBox.askyesno("Testing Askyesno",
				 "Click on something!"):
		print 'You clicked on Yes!';

rootWindow = Tkinter.Tk()

button1 = Tkinter.Button(rootWindow,
			 text="Askyesno",
			 command=askyesno)
button1.pack()

button2 = Tkinter.Button(rootWindow,
			 text="QUIT",
                         command=rootWindow.quit)
button2.pack()

rootWindow.mainloop()
-----

> If you post a description of what you are trying
> to do I can give you a correct and better style solution.
>
> --
> Tim Evans

What I'm trying to do is pretty simple. I want to display a menu of
several buttons. When Button1 is clicked, it will use Windows Notepad
to open a text file and display a askyesno dialog box. If user is
satisfied with what he sees in the text file, he clicks yes and the
file will be uploaded to a server. If not, program quits. Clicking on
Button2 will open up another menu of several buttons which will run
other scripts. All the other buttons are variations of the above and
all the scripts to be called by the buttons have already by written. I
just need to write the Tkinter menu screen. I chose to use
tkMessageBox's standard dialogs because I'm new to Tkinter and the
standard dialogs pretty much cover what I need.

Thanks for your post, Tim

Raymond Tong Leng Ng


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From mnot at pobox.com  Fri Jun 11 21:13:08 1999
From: mnot at pobox.com (Mark Nottingham)
Date: Sat, 12 Jun 1999 01:13:08 GMT
Subject: Using PIL with CGI scripts
References: <19990611223149.63326.qmail@hotmail.com>
Message-ID: <011601beb470$bbb216a0$0301a8c0@cbd.net.au>

Luis,

what you have to do is call the script twice. For instance, if you decide
that you need to display foo.jpg (which you've got in a pickle), you first
need to output html like this:

...
<IMG SRC="/path/to/this/script/scriptname?show=foo.jpg">
...
replacing /path/to/this/script/scriptname with the URL of the script
generating the request.

The browser will get this, and when it gets to the IMG tag, it will re-call
your script, this time with foo.jpg as an argument. Your script needs to be
able to take that request and return the proper HTTP headers, along with the
image data itself as the body. For instance (assuming that img is your
Image() instance):

tmp = StringIO()
img.save(tmp, 'JPEG')
tmp.seek(0)
image = tmp.read()

print '''\
Content-type: image/jpeg
Content-length: %s

%s''' % (len(image), image)


u
----- Original Message -----
From: Luis Roco Arriagada <luisroco at hotmail.com>
To: <mnot at pobox.com>; <python-list at cwi.nl>
Sent: Saturday, June 12, 1999 8:31
Subject: Re: Using PIL with CGI scripts


> Mark:
>
> Sorry for bothering you again, but I can't figure out how can I do this
> work.
>
> I can`t write the image to the disk because all of the images are inside a
> single file where they are compressed using pickle. What I'm trying to do
is
> to get some images from this file according to a word search perform by
the
> user of the web site. This part is already done, so the images are
extracted
> from the file but they are in a PIL format, ie they are an Image type, and
> for that i cannot display them in a web page using the HTML tag <IMG SRC =
>
> because this tag only accepts a file name as input.
>
> I try to use cStringIO to emulate a file, but I don't know how I must use
> this variable with the HTML tag.
>
> This part of the code is something like this:
> from cStringIO import StringIO
>
> print "Content-type: text/html"
> print
> print "<HEAD><TITLE>WEBIMAGE</TITLE></HEAD>"
> print "<BODY BGCOLOR=\"#FFFFFF\" TEXT=\"#000000\" LINK=\"#0000FF\"
> VLINK=\"#000080\" ALINK=\"#FF0000\">"
> print "<center><H1>WEBIMAGE</h1>"
> print "<p>Version 0.01</p>"
> print "<hr>"
> .
> .
> .
> temp = StringIO()
> im.save(temp,"JPEG") #im is an Image variable
> temp.seek(0)
> print "<p><IMG SRC='%s'>"%temp
>
> print "</center>"
> print "</body>"
>
>
> Hope you can help me and thanks in advance
> Luis Roco
>
> >From: "Mark Nottingham" <mnot at pobox.com>
> >To: <luisroco at hotmail.com>,<python-list at cwi.nl>
> >Subject: Re: Using PIL with CGI scripts
> >Date: Wed, 9 Jun 1999 08:27:23 +1000
> >
> >See the StringIO module (actually, cStringIO).
> >
> >One thing to keep in mind - unless the image is *really* dynamic (i.e.,
it
> >changes EVERY time someone accesses it), you're much better off if you
> >write
> >it to disk.
> >
> >See genpix.py at http://www.mnot.net/python/  for an example of this
(it's
> >a
> >command line utility that HTMLifys a directory of images, but it does use
> >cStringIO in this manner).
> >
> >Cheers,
> >
> >
> >----- Original Message -----
> >From: <luisroco at hotmail.com>
> >To: <python-list at cwi.nl>
> >Sent: Wednesday, June 09, 1999 7:10
> >Subject: Using PIL with CGI scripts
> >
> >
> > > Hi all
> > >
> > > How can a CGI scripts outputs graphics images through Python code
using
> >the Python Imaging Library package?
> > >
> > > I just want to produce some images which has been processed using PIL
> >and
> >show them in a web site
> > >
> > >
> >
>
>
> ______________________________________________________
> Get Your Private, Free Email at http://www.hotmail.com
>




From behrends at cse.msu.edu  Tue Jun 29 01:21:25 1999
From: behrends at cse.msu.edu (Reimer Behrends)
Date: 29 Jun 1999 05:21:25 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net>  <7l4nht$jl7$1@nnrp1.deja.com>
Message-ID: <slrn7nglvv.g9.behrends@allegro.cse.msu.edu>

Andrew Cooke (andrew at andrewcooke.free-online.co.uk) wrote:
> Tk isn't the standard GUI for Python - it's the standard GUI.  It's
> been around for aeons and I, for one, would be a tad annoyed if I
> had to learn a new GUI each time I learnt a new language.
> 
> There seem to be two problems with it:
> 
> 1. Speed.

First of all, I do not even agree that the speed of Tkinter (or other
Tk-based libraries) is a problem at all. In my experience, the speed and
responsiveness has always been more than adequate (as opposed to, say,
AWT and Swing). Unless you are going to program highly interactive
graphical games, speed should be more than sufficient.

> For most applications, is this an issue?  Python is, above
> all, a scripting language.  Speed is not important.  What you can do -
> and do fairly simply with Tk, once you know it, is build a GUI front-
> end to whatever programs you are tying together.  The text widget, for
> example, is amazingly flexible with very little code.

I have to disagree at least partly--in a GUI, speed (and in particular,
responsiveness) is an issue. On the other hand, I _did_ write at least
one program where high responsiveness was an issue (read: interactive
gaming), and Tk never gave me any headaches, even on slower machines.

> 2. Widgets.  Tk certainly doesn't have a modern set of widgets.  There
> are extensions, however.  And with Idle, it looks as though the range
> included in standard Python is increasing (does anyone complain Idle
> is slow?!)

However, Tk[inter] is easily extensible (and you don't even have to
recompile anything for that), with fairly powerful basic widgets. For
instance, when I needed gauge widget, I put it together in just a few
minutes.

Also, I can run Tk-based programs on just about any system I have access
to without changes (as opposed to, for instance, Gtk-based code, which
even with the advent of version 1.2 remained pretty much a moving
target).

[...]

> For others, I suspect the lack of good docs for Tk is a more serious
> problem.  There is a pretty steep learning curve involved if you're
> meeting Tk for the first time (and if you do know it, finding the
> right way to express it in Python can still mean reading the Tkinter
> code).

I always thought that Fredrik Lundh's introduction to Tkinter was
excellent (and easily accessible from the Tkinter page on
www.python.org, too). What problems do you see with it?

[...]

				Reimer Behrends



From radbit at my-deja.com  Thu Jun 10 22:39:46 1999
From: radbit at my-deja.com (radbit at my-deja.com)
Date: Fri, 11 Jun 1999 02:39:46 GMT
Subject: WinNT: Corrupt installation detected
References: <7joq9h$qd2$1@schbbs.mot.com>
Message-ID: <7jpsth$3rk$1@nnrp1.deja.com>

In article <7joq9h$qd2$1 at schbbs.mot.com>,
  "Rick Rankin" <Rick_Rankin-P15254 at email.mot.com> wrote:
> I'm trying to install 1.5.2 on my Windows NT system. I previously had
1.5.2
> beta 2 installed, which I uninstalled before trying to install the
final
> release. When I try to run the installer (py152.exe), I immediately
get a
> dialog box that says "Corrupt installation detected".
>
> I'm sure the downloaded installer is not corrupt. I opened it in
WinZip and
> used the test function. All files in the archive tested fine. Has
anyone
> seen this before?
>
> Thanks,
>
> Rick Rankin
> Motorola SATCOM
>
>
Nope. I know that you tested the archive but just as a matter of
interest: how large (in bytes) is the version you downloaded?
Another question: if you downloaded the file per FTP, I trust that your
FTP client downloaded it in binary mode? Not goading - just haven't
noticed this problem in similar circumstances over various NT 4.0 -
py152.exe installations.

Cheers
--
Florent Heyworth


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From dalke at bioreason.com  Thu Jun 24 16:50:25 1999
From: dalke at bioreason.com (Andrew Dalke)
Date: Thu, 24 Jun 1999 14:50:25 -0600
Subject: Thanks for Python...
References: <376ead64@anonymous.newsfeeds.com> <7ktfn0$5q$1@aurwww.aur.alcatel.com> <14194.18324.582362.94739@amarok.cnri.reston.va.us> <37726458.C8CC9623@zipzap.ch> <aahzFDuH2p.9C5@netcom.com>
Message-ID: <37729A11.C93DA513@bioreason.com>

Aahz Maruch <aahz at netcom.com> said, concerning Boris Borcic  <zorro at zipzap.ch>'s
suggestion that _THHGttG_ (quoted by
Andrew Kuchling) was a rebuttal to the _Matrix_:

> _THHGttG_ was written so long before _Matrix_ that if anything
> it's the other way 'round.

  But if the world is a giant computer run by white mice as
Hitchhiker's says, then wouldn't there be time travel?  From a
rough memory, the Guide lifted text from a ceral company then
went back in time and sued the company for copyright infringement.

  So after The Matrix came out, the mice went back in time and
had (pursuaded, brainwashed, induced, etc.) Douglas Adams to
write the book then got it turned into a movie.

  Oh, you think time goes *that* way -- Mostly Harmless (paraphrased)

						Andrew Dalke
						dalke at acm.org



From Klaus.Alexander.Seistrup at p98.f112.n480.z2.fidonet.org  Wed Jun 30 09:39:04 1999
From: Klaus.Alexander.Seistrup at p98.f112.n480.z2.fidonet.org (Klaus Alexander Seistrup)
Date: Wed, 30 Jun 1999 14:39:04 +0100
Subject: I am v. stupid, please help
Message-ID: <000015ae@bossar.com.pl>

From: Klaus Alexander Seistrup <kas at maps.magnetic-ink.dk>

Newbie wrote:

> name = input('What is your name : ')
> 
> This works for integers but not for strings,

Use raw_input() instead.


  //Klaus

-- 
???[ Magnetic Ink ]??????????????????????????????????????????????????????????
???[ http://www.magnetic-ink.dk/ ][ maps == mail abuse prevention system ]???



From dalke at bioreason.com  Mon Jun 14 22:04:47 1999
From: dalke at bioreason.com (Andrew Dalke)
Date: Mon, 14 Jun 1999 20:04:47 -0600
Subject: module import, but different
References: <000301beb508$6fcca140$ce9e2299@tim>
Message-ID: <3765B4BF.7181221D@bioreason.com>

Tim Peters <tim_one at email.msn.com> said:
> This is much harder than the problem you claimed to have at
> the start

  Well..... yeah.  When I'm not sure what I'm doing, I try
to start off with what I think is the problem.

> Ack, enough already.

  And keep on describing the problem in the hopes that someone
can figure out what I want to do.

> The std imp module packages many functions that will be helpful
> when doing this.

  This is just the pointer I needed.  Once I figured out the right
parameters for imp.load_module, it worked:


val> pwd
/home/u03/dalke/src/MCS/test
val> python
Python 1.5.1 (#21, Nov 23 1998, 15:04:47) [C] on irix6
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import MCS
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ImportError: No module named MCS
>>> import imp
>>> imp.load_module("MCS", None, "..", ("", "r", imp.PKG_DIRECTORY))
>>> import MCS.Daylight
>>> MCS.Daylight.test()
time for standard comparison 0.0091609954834
 MCS size 2
>>> 

  Yet another reason why I like the Python implementation :)


G. David Kuhlman <dkuhlman at netcom.com> chimed in:
> If you decide to follow Tim's suggestion and implement your own
> import, then search DejaNews for "imputil".

  Luckily, this is enough for what I need.

						Andrew Dalke
						dalke at bioreason.com



From bmatt at acm.org  Sat Jun  5 18:20:55 1999
From: bmatt at acm.org (Bruno Mattarollo)
Date: Sat, 5 Jun 1999 22:20:55 GMT
Subject: =?iso-8859-1?Q?Spanish_Python_mailing_list_-_Lista_de_Python_en_Espa=F1ol?=
Message-ID: <000201beafa1$adaaf410$6eba0ac8@kuarajy.infosys.com.ar>

Hello

	We are pleased to announce the creation of a spanish Python mailing list.

	Everyone is invited to join this list, share ideas, opinions, ask for
technical advice and share technical knowledge.

	This list is run on Mailman.

	To join you can send an email to: python-ar-request at lists.arnet.net.ar with
"subscribe" as the subject.

	pdinoto at DONTSPAM-arnet.net.ar or bmatt at DONTSPAM-acm.org can be contacted
for any info regarding problems concerning the list. (please remove the
DONTSPAM- string before sending a mail).

	Cheers,

- (Spanish version - Versi?n espa?ol) ----------------

Hola

	Tenemos el orgullo de anunciar la creaci?n de la lista de correo de Python
en espa?ol.

	Todos est?n invitados a suscribirse a dicha lista para compartir ideas,
opiniones, solicitar ayuda t?cnica y compartir conocimiento.

	Esta lista es operada a trav?s de Mailman.

	Para suscribirse debe enviar un mensaje a:
python-ar-request at lists.arnet.net.ar con la palabra "subscribe" en el campo
"subject".

	pdinoto at DONTSPAM-arnet.net.ar o bmatt at DONTSPAM-acm.org pueden ser
contactados por cualquier problema relacionados con la lista (por favor,
quite la cadena DONTSPAM- de las direcciones para poder env?ar un mensaje).

Saludos

/B

Bruno Mattarollo <bmatt at acm.org>
... proud to be a PSA member <http://www.python.org/psa>




From tim_one at email.msn.com  Sun Jun 13 14:55:22 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 13 Jun 1999 14:55:22 -0400
Subject: The Best Parsing Book on Earth (was RE: Interpolation)
In-Reply-To: <m3d7z0t6n0.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <000101beb5ce$4a3aa3c0$ae9e2299@tim>

[Tim]
> I posted a link earlier to the Best Parsing Book on Earth; read the
> first few chapters of that first.

[Michael Hudson]
> Which one was that? I think I missed it!

Bears repeating, then (I've gotten many email inquiries too).  It's a
wonderful intro, now out of print, but thanks to the authors' generosity
downloadable for free:

    http://www.cs.vu.nl/~dick/PTAPG.html

and-changed-the-subject-line-for-future-generations-ly y'rs  - tim





From larsga at ifi.uio.no  Wed Jun 16 18:20:19 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 17 Jun 1999 00:20:19 +0200
Subject: GC In Python: YAS (Yet Another Summary)
References: <1282571877-108998356@hypernet.com> <zoQ93.198$U6.1646@newsr2.twcny.rr.com> <wku2s86nmb.fsf@ifi.uio.no> <aNT93.216$U6.1531@newsr2.twcny.rr.com>
Message-ID: <wk674n7pp8.fsf@ifi.uio.no>

* news at dorb.com
|
| Tried the lazy plumbo and it found a cycle in a large application I'm
| working on.
| And it was easy to fix, once found ::)

Glad to hear it. :)
 
| It took 5min at least to find the cycle. I'm not complaining because
| it would have taken me a lot longer. But maybe there is a way to
| speed this up.

You certainly seem to have a way of getting your wishes fulfilled.
Fredrik Lundh just sent me patch that should speed it up a little. 

May the rest of your wishes be fulfilled as well! (I'm not going to
do it, though. :)

--Lars M.



From delgado at Mathematik.Uni-Bielefeld.DE  Tue Jun 15 15:07:25 1999
From: delgado at Mathematik.Uni-Bielefeld.DE (Olaf Delgado)
Date: Tue, 15 Jun 1999 21:07:25 +0200
Subject: Newbie: Truth values (three-valued logic)
Message-ID: <Pine.OSF.4.05.9906152037050.23333-100000@pinax.Mathematik.Uni-Bielefeld.DE>

Hi everybody!

I am just starting to use python seriously. Currently, I am porting a
small class library from C++ to python, which is kind of fun. Everything's
much more simple and elegant now.

There's one thing I could do in C++, though, which doesn't seem to work in
python: some methods need to return truth values, which can be 1 for
'true', 0 for 'false' and x for 'frankly I don't know'. Now this x,
when used in an if-clause, should evaluate to false, which is easy:

 class uncertain:
   def __nonzero__(self):
     return 0
   def __repr__(self):
     return 'dunno'

So the following works fine:

>>> maybe = uncertain()
>>> maybe
dunno
>>> if maybe: print 't' # prints nothing
...
>>>

Unfortunately:

>>> if not maybe: print 'f'
...
f
>>>

This behaviour is completely consistent in a two-valued logic, but not
what I actually want, because maybe should neither be true nor false, nor
should its negation. As far as I know, the logical negation operator can
not be overloaded, so I have no chance to change this, right? No big deal,
really, but I'd like to know if I missed something here. So, is there some
magical trickery to achieve the behaviour I want?

certainly-uncertain'ly yours,  Olaf

-- 
   ////         Olaf Delgado Friedrichs, Uni Bielefeld, Mathematik
   Olaf         Postfach 100131                  (room:  V4-109)
   `='          D-33501 Bielefeld, Germany       (phone: +49 521 106-4765)
                           http://www.mathematik.uni-bielefeld.de/~delgado




From jeremy at cnri.reston.va.us  Tue Jun  1 13:47:06 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Tue,  1 Jun 1999 13:47:06 -0400 (EDT)
Subject: Python 2.0 (perhaps off topic)
In-Reply-To: <375372F6.A874554@prescod.net>
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7imr6b$1s0$1@nnrp1.deja.com>
	<87pv3iq6ss.fsf@ev.netlab.co.jp>
	<3752C76D.604CD474@prescod.net>
	<87n1ykr0z9.fsf@ev.netlab.co.jp>
	<375372F6.A874554@prescod.net>
Message-ID: <14164.7179.10346.998488@bitdiddle.cnri.reston.va.us>

>>>>> "PP" == Paul Prescod <paul at prescod.net> writes:

  PP> Yukihiro Matsumoto wrote:
  >> |But note that Python's guarantees are more important in small,
  >> simple |programs and Java's guarantees are better for large,
  >> complex programs with |complicated data structures.
  >> 
  >> Do you really mean Python the object-oriented programming
  >> language rarely treat complicated data structures?

  PP> No, I said that Python is optimized for basic users, not for the
  PP> experts.  The experts can figure out how to work around
  PP> problems. The beginning users cannot. "I'm opening all of the
  PP> files in a directory in a loop and in large directories I get
  PP> out of file handle errors."

I don't think this is a good example.  Even beginning users are going
to have to learn a little about resource management about some point.
People should not rely on files getting deallocated and closed by
Python.  All the beginners who start with JPython are going to get
Java's GC.  And they're going to need to close the files by hand.

Jeremy




From graham at ccs.ucsb.edu  Wed Jun 16 16:30:11 1999
From: graham at ccs.ucsb.edu (Graham Hughes)
Date: 16 Jun 1999 13:30:11 -0700
Subject: OO (was: Why is tcl broken?)
References: <000801beb745$02768220$979e2299@tim>
 <871zfdp73f.fsf@home.ivm.de> <wklndk8ly6.fsf@ifi.uio.no>
 <873dzsma6b.fsf@home.ivm.de> <wk3dzs87mr.fsf@ifi.uio.no>
 <87lndkkmda.fsf@home.ivm.de>
Message-ID: <87k8t3opm4.fsf@oak.treepeople.net>

>>>>> "Klaus" == Klaus Schilling <Klaus.Schilling at home.ivm.de> writes:

    Klaus> Lars Marius Garshol <larsga at ifi.uio.no> writes:
    >> * Klaus Schilling | | No, in the sense `Steele merely
    >> rediscovered scheme, not bring it | into being'
    >> 
    >> I don't understand what you mean by this. The two people who
    >> created Scheme originally were Sussman and Steele. If that isn't
    >> bringing it into being I don't know what would be.

    Klaus> No, they discovered Scheme, they did not create it.

In the interests of ending this, I think I can explain what Klaus means
here.

There's a philosophical debate about whether especially mathematical
ideas are `created' or `discovered'; that is, whether they were cobbled
together by a human or lay there secretly for years only to be stumbled
across by the same human.  Mathematicians typically believe that they
only discovered mathematical theorems etc., rather than create them;
it's a sort of tribute to the spare elegance of group theory, for
example.

Klaus appears to mean that Scheme is so mind bogglingly elegant that
Steele could not have created it out of whole cloth, merely discovered
it as a mathematical truth.

As an aside I really don't think Scheme is anywhere near *that* elegant,
and as a matter of fact it hurts me to use the language for more than
trivial things, *precisely* because the damn thing is so small.  Fer
crissakes, you don't even have the ability to create an aggregate type
in R5RS!  You can fake it; but it sucks, and it doesn't integrate well
with the built in type system...

ObPython: Python is far easier to use IMHO, because it has more stuff
available.  I don't need to reinvent the hash table, or cobble together
OOP with closures, 'cause it's done for me, and I don't need to reinvent
SMTP, FTP, network sockets, any of that because it's all in the standard
libraries.  When it isn't, like XML, it's usually available elsewhere,
or failing everything I can hack one together myself and drop it in.
Large languages permit small programs, or something like that.
-- 
Graham Hughes <graham at ccs.ucsb.edu>
GPG Fingerprint: 4FC5 80F0 63EB 00BE F438  E365 084B 4010 60BF 17D3
((lambda (x) (list x (list 'quote x)))
 '(lambda (x) (list x (list 'quote x))))



From janz at easics.be  Fri Jun  4 09:02:03 1999
From: janz at easics.be (Jan Zegers)
Date: Fri, 04 Jun 1999 15:02:03 +0200
Subject: Tkinter problem in Python 1.5.2 on Solaris 2.5.1
References: <1283620640-45912067@hypernet.com>
Message-ID: <3757CE4B.B8B70983@easics.be>

Gordon McMillan wrote:
> 
> Jan Zegers wrote:
> >
> > I have the following problem with Tkinter in Python 1.5.2.
> /snip/
> >   This probably means that Tcl wasn't installed properly.
> >
> > Note the whitespace between the two last test lines. The
> > installation is as follows:
> >   Tcl/Tk 8.1 with --enable-shared so we have libraries
> >     /usr/local/lib/libtcl8.1.so and /usr/local/lib/libtk8.1.so
> 
> Tcl / Tk 8.0 is still the official version for Python.
> 
> On Jun 1, Dieter Maurer posted patches to use 8.1 ("Tkinter-patch for
> use with Tk8.1...").
> 
> - Gordon
Thank you. I applied the patches and it worked!
Kind regards,
Jan
-- 
===================================================================
Jan Zegers                ===              Easics               ===
General Manager           ===  VHDL-based ASIC design services  ===
Tel: +32-16-395 601          ===================================
Fax: +32-16-395 619      Interleuvenlaan 86, B-3001 Leuven, BELGIUM
mailto:janz at easics.be              http://www.easics.com



From gmcm at hypernet.com  Wed Jun 30 18:20:22 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 30 Jun 1999 17:20:22 -0500
Subject: A Python problem?
In-Reply-To: <377A765F.90367AC5@connection.com>
Message-ID: <1281341433-35198011@hypernet.com>

Colin J. Williams wrote:

> This script below illustrates what appears to be a problem with the
> identification of a module in a package.
> 
> The example is borrowed from the linbot material.
> 
> Colin W.
> ''' A Python problem?   .
> 
>   '''
> def main():
>   import sys
>   print 'Start:',dir()
>   s0= sys.modules
>   import plugins.rptlib
>   s1= sys.modules
>   if s0 == s1:
>     print 'sys.modules is unchanged'

Charles already caught that one.

>   print 'After attempt to import plugins.rptlib:', dir()
>   try:
>     print dir(plugins.rptlib)
>   except AttributeError:
>     import sys
>     print sys.exc_info()

Notice that your "import plugins.rptlib" yields something called
"rptlib". This is something done by the plugins package (probably in 
it's __init__.py), not normal behavior. Normally, dir() would show 
"plugins" and "dir(plugins.rptlib)" would show what you expect. 
Unless there's a good reason for doing so, I'd call this a case of 
"clever" turning out to be "stupid" <wink>.

>   from plugins.rptlib import *
>   print 'rptlib names now available:', dir()
> main()
> 
> '''
> ------------------------ Resulting Display
> ----------------------------
> >>> Start: ['sys']
> sys.modules is unchanged
> After attempt to import plugins.rptlib: ['plugins', 's0', 's1',
> 'sys'] (<class exceptions.AttributeError at 76bbc0>,
> <exceptions.AttributeError instance at daa0e0>, <traceback object at
> dab2b0>) rptlib names now available: ['Link', 'add_problem',
> 'check_and_warn', 'config', 'debugio', 'doBotMain', 'doTopMain',
> 'get_title', 'linbot', 'linkList', 'main_index', 'make_link',
> 'nav_bar', 'open_file', 'opener', 'os', 'plugins', 'problem_db',
> 'proxies', 'read_registry', 's0', 's1', 'sort_by_age',
> 'sort_by_author', 'sort_by_rev_age', 'sort_by_size', 'string',
> 'stylesheet', 'sys', 'urllib', 'version']
> 
>   '''
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list

- Gordon



From holger at phoenix-edv.netzservice.de  Fri Jun 25 05:35:51 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Fri, 25 Jun 1999 09:35:51 GMT
Subject: pkzip and using popen on Windows 98 ???
References: <2H1c3.180$T3.411@newsr2.u-net.net>
Message-ID: <37734D77.1979CBD7@phoenix-edv.netzservice.de>

High Mark!

Mark Butterworth - U-Net schrieb:
> 
> Im having trouble using os.system or popen (both win32pipe & os) under
> windows 98.  It will not return the exit value of the command Im running
> (which is infozips - zip).  Also when attempting to use win32pipe.popen3 -
> my pipe read's hang after one or two reads.
> 
> Also does anyone know of any python modules which can read / write pkzip (or
> infozip) format files as I don't like calling external commands.
> 
> Any help would be appreciated?

Tested module GZIP?

f = gzip.GzipFile(fromPath , 'rb')
var = f.read()
f.close()
writeFile(toPath, var)

..works fine for unzipping file.



From paolo.cantore at zxa.basf-ag.de  Tue Jun 22 08:07:38 1999
From: paolo.cantore at zxa.basf-ag.de (Paolo G. Cantore)
Date: Tue, 22 Jun 1999 14:07:38 +0200
Subject: Check for valid date
References: <7klq51$etp$1@nnrp1.deja.com>
Message-ID: <376F7C8A.A809C0A@zxa.basf-ag.de>

Hi Remi,

I use a two-step procedure:  

>>>def check_date(year, month, day):
...	tup1 = (year, month, day, 0,0,0,0,0,0)
...  	try:
...    		date = time.mktime (tup1)
...    		tup2 = time.localtime (date)
...		if tup1[:2] != tup2[:2]:
...			return "Date not valid."
...		else:
...			return "Date valid."
...	except OverflowError:
...		return "Date not valid."
...
>>> import time
>>> print date_valid(1999,2,29)
Date not valid.
>>> print date_valid(1999,2,28)
Date valid.

Paolo


Remi Delon wrote:
> 
> Hi, I'm trying to use the time module to check whether a date is
> actually a valid day or not.
> Unfortunately, it looks like the strptime function only performs a basic
> range check. (for example, it will accept 02/30/199)
> Here is a little test I did:
> 
> >>> strptime('02/30/1999','%m/%d/%Y)
> (1999, 2, 30, 0, 0, 0, 61, 0)
> >>> mktime((1999, 2, 30, 0, 0, 0, 61, 0))
> 920361600.0
> >>> ctime(920361600.0)
> 'Tue Mar  2 00:00:00 1999'
> 
> Is there a way to check test whether the day actually exists or not ?
> 
> I also have a totally different problem : I'm trying to get python 1.5.2
> to run with Tcl/Tk 8.1. Has anyone been able to do that on AIX ?
> I found 2 patches for tkinter that were supposed to fix the problems,
> but unfortunately, they don't work for AIX.
> 
> Thanks
> 
> Remi.
> 
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.

-- 
/* Paolo G. Cantore             paolo.cantore at basf-ag.de */
/* BASF AG            ZOI/AP Zentralabteilung Informatik */
/* D-67056 Ludwigshafen            Tel. +49-621-60-43987 */
/* Germany                         Fax. +49-621-60-20833 */



From rfreedm1 at nospam.rochester.rr.com  Fri Jun 11 20:25:34 1999
From: rfreedm1 at nospam.rochester.rr.com (Benjamin Schollnick)
Date: Sat, 12 Jun 1999 00:25:34 GMT
Subject: Problem with Lists & Index command
References: <rNHvjEdhm5Pp-pn2-zOTjsVjHebb4@d185d18f8.rochester.rr.com> <37618424.8B8F3116@mindspring.com>
Message-ID: <rNHvjEdhm5Pp-pn2-XcIGYOCMcDxI@d185d18f8.rochester.rr.com>

On Fri, 11 Jun 1999 21:44:47, Doug Hellmann 
<doughellmann at mindspring.com> wrote:

> > It appears that INDEX doesn't work on a multiple
> > dimension list????
> 
> Using index() on tcmlist is only going to search through the values in
> tcmlist.  It won't look inside the tuples that tcmlist contains.  So,
> you either have to construct a tuple to pass to index, or use a
> dictionary instead of a list:

Example cut.....to save space...

I'll try this over the weekend...

But, the most obvious question, is.....

Can dictionaries be made & manipulated from
variables?

> tcmvalues = { 'EFGHI':1,
>               'JKLMN':2,
>               'ABCD':3,
>             }

i.e.
	tcmvalues.append ( entry:entry3)

etc...

Please keep in mind, that I don't know anything about
dictionaries... The book I'm using hasn't hit 'em yet....
(I'm not done reading yet)

		- Benjamin



From tim_one at email.msn.com  Tue Jun  8 22:27:39 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 8 Jun 1999 22:27:39 -0400
Subject: dict vs kjBuckets vs ???
In-Reply-To: <aahzFD15sr.3oD@netcom.com>
Message-ID: <000e01beb21f$a4c5a860$2fa22299@tim>

[Aahz Maruch]
> I'm currently using a dict for storing references to objects (call them
> customer objects, referenced by customer ID (a number)).  I'm using a
> dict (instead of SQL or something similar) because we're willing to
> trade lots of memory to get speed.  However, I've heard that after about
> 10K items in a dict, it starts having problems.

11,523 to be exact.  After that, dicts drink to excess and show up for work
late the morning after.  We don't like to talk about it, though.

> Is this a specious rumor,

It wasn't before you spread it <wink>.  Seriously, haven't heard a peep
about this before, and routinely use dicts with 40K objects -- performance
stays O(1) all the way.  It's certainly possible-- but unlikely --that you
assign "customer IDs" in a pattern that interacts badly with the hash
function, leading to lots of collisions.

> should I switch to kjBuckets, or should I do something else?

The latter:  investigate!  Create a reproducible bad case and we'll fix it,
or retract your slander <wink>.

> Does the answer change if we have 100K objects?

No.

>  A million?

No.

It all depends on how much memory you have and how well the hash function is
working.  Note that dicts work hard to "randomize" the addresses at which
they store keys, so memory locality can be terrible.  But if there's no
pattern to customer ID accesses, locality would be terrible in a 1M array
too.

> (The only operations are object insert, update, delete, and lookup.)

Kinda figured that, else you wouldn't be using a dict <wink>.

dicts-are-great-use-more-of-them-ly y'rs  - tim





From andrew at andrewcooke.free-online.co.uk  Tue Jun 29 04:17:52 1999
From: andrew at andrewcooke.free-online.co.uk (Andrew Cooke)
Date: Tue, 29 Jun 1999 08:17:52 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <7l4nht$jl7$1@nnrp1.deja.com> <slrn7nglvv.g9.behrends@allegro.cse.msu.edu>
Message-ID: <7l9vfb$abi$1@nnrp1.deja.com>

In article <slrn7nglvv.g9.behrends at allegro.cse.msu.edu>,
  behrends at cse.msu.edu (Reimer Behrends) wrote:
> Andrew Cooke (andrew at andrewcooke.free-online.co.uk) wrote:
> > Tk isn't the standard GUI for Python - it's the standard GUI.  It's
> > been around for aeons and I, for one, would be a tad annoyed if I
> > had to learn a new GUI each time I learnt a new language.
> >
> > There seem to be two problems with it:
> >
> > 1. Speed.
>
> First of all, I do not even agree that the speed of Tkinter (or other
> Tk-based libraries) is a problem at all. In my experience, the speed
and
> responsiveness has always been more than adequate (as opposed to, say,
> AWT and Swing). Unless you are going to program highly interactive
> graphical games, speed should be more than sufficient.

I agree - but my only use of Tk has been for building GUIs.  I
presumed that the people complaining about speed must have been
trying to do 3d imaging or something, hence my later comments about
going to C to do faster imaging.

[...]
> > For others, I suspect the lack of good docs for Tk is a more serious
> > problem.  There is a pretty steep learning curve involved if you're
> > meeting Tk for the first time (and if you do know it, finding the
> > right way to express it in Python can still mean reading the Tkinter
> > code).
>
> I always thought that Fredrik Lundh's introduction to Tkinter was
> excellent (and easily accessible from the Tkinter page on
> www.python.org, too). What problems do you see with it?

To pick the last example I had problems with - where does anything
tell you how to change the value of a checkbox menu item (not the
same as a checkbox)?  Or if you interactively resize a window and
then pack_forget it, why do you get a chunk of blank canvas?

That's the kind of thing I mean.  OK, it is easy to put something
together to do the basics, but there are a lot of small
(undocumented) wrinkles when it comes to making Tk do exactly what
you want - and I suspect the people complaining about Tk are the
generation after me raised on sophisticated interfaces...  Before
you can get a good application you have to learn a lot (or use a
very basic interface, I guess).

[The answer to the first puzzle is to change the value of the variable
associated with the item - you have to use the special class for
variables.   I have no fix for the second,m except to fix sizes to
make resizing impossible.]

Andrew


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From srn at fn.com.au  Wed Jun 23 08:12:34 1999
From: srn at fn.com.au (Stephen Norris)
Date: 23 Jun 1999 12:12:34 GMT
Subject: currentThread(): no current thread for ... using threading and Events?
Message-ID: <7kqivi$cli$1@m2.c2.telstra-mm.net.au>

I'm seeing 

currentThread(): no current thread for XXXXX

messages (where XXXXX is some number) from some code I'm working on. They
are coming from the threading module, and the only part of that I'm
using is Events.

I'm also using thread style threads. Is this what's going wrong? Does this
confuse things? The lib ref doesn't say it will, but it seems possible,
especially now that I look at the code. Hmmm....

I don't really like the Thread class, and I'd like to avoid changing to
it (since there's a fair bit of code already there), but if this is the
cause I'll have to.

Does anyone have any wisdom to impart?

	Stephen
-- 
Stephen Norris  srn at fn.com.au
PGP key available via finger srn at flibble.fn.com.au.
Farrow Norris Pty. Ltd.     http://www.fn.com.au/



From Sven.Drescher at dlr.de  Wed Jun 23 02:57:36 1999
From: Sven.Drescher at dlr.de (Sven Drescher)
Date: Wed, 23 Jun 1999 08:57:36 +0200
Subject: os.kill()
Message-ID: <37708560.BF23BA36@dlr.de>

Hi!
I have any problems to kill a process started with execv().
The process is right on and the pid is ok. But I tested :
        os.kill(pid, 0)
or    os.kill(pid, 15)
or    os.kill(pid, signal.SIGTERM)
an nothing happens.
I looked for my problem in the newgroups and on the python homepage, but
I haven't found anything.
I killed my process on console with " kill -9 pid".
Is there a stronger version of os.kill() like under UNIX?

Thanks for help and hints!

--
Sven






From perry at ac-copy.net  Mon Jun  7 16:27:45 1999
From: perry at ac-copy.net (Joachim Schmitz)
Date: Mon, 7 Jun 1999 20:27:45 GMT
Subject: mxODBC and MySQL
Message-ID: <Pine.LNX.4.05.9906072222280.5855-100000@ns.ac-copy.net>

Hi,

I compiled and installed  mxODBC for MySQL on SuSE 6.1 without problems.
I get this:
Python 1.5.2b2 (#6, Feb 28 1999, 21:41:19)  [GCC egcs-2.91.60 19981201
(egcs-1.1
.1  on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import ODBC.MySQL
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/local/lib/python1.5/ODBC/MySQL/__init__.py", line 7, in ?
    from mxODBC import *
ImportError: /usr/local/lib/python1.5/ODBC/MySQL/mxODBC.so: undefined
symbol: PyBuffer_Type
>>

regards 

Joachim Schmitz

------------------------------------------------------------------------------
 P R O F I . N E T  A G    *    Joachim Schmitz   * Senior Network Consultant
------------------------------------------------------------------------------
  profi.net                |    Aindlinger Str. 7   | Voice  +49 821 27062-0
  Hassler & Mair AG        |    86167 Augsburg      | Fax    +49 821 27062-99
                           |    Germany             | Email  js at profi.net  




From ddainese34x at x22dsi.unive.it  Fri Jun 11 07:30:01 1999
From: ddainese34x at x22dsi.unive.it (Diego Dainese)
Date: 11 Jun 1999 11:30:01 GMT
Subject: waste of resources ?
References: <375C0F91.1C73E4E3@icrf.icnet.uk> <375C16A9.5ADC4990@icrf.icnet.uk> <375C2003.B01BEC3@arlut.utexas.edu> <87yahvzqz3.fsf@pc-hrvoje.srce.hr> <ytt1zfnen4h.fsf@gilgamesh.cse.ucsc.edu> <87pv37zpex.fsf@pc-hrvoje.srce.hr> <Pine.SUN.3.95-heb-2.07.990608005904.9179A-100000@sunset.ma.huji.ac.il>
Message-ID: <slrn7m1spp.1tq.diego@blanka2.blankanet.it>

On Tue, 8 Jun 1999 01:03:03 +0300, Moshe Zadka wrote:
> On 7 Jun 1999, Hrvoje Niksic wrote:
> <about syscalls throwing EINTR and aborting>
> > I don't know about Perl, but a good fix would be to have the
> > interpreter do the hard work for us.  If a system call fails and errno 
> > is EINTR, then the signal handler should be called, *and* the syscall
> > should be restarted.
> 

I'm agree with this also.

> Ummmmm......
> 
> You can do the hard-work yourself, but once and for all:
> 

Yes, but this is boring ;)

> def myread(file):
> 	while 1:
> 		try:
> 			return file.read()
> 		except (IOError, os.error), detail:
> 			if detail.args[0] != errno.EINTR: raise
> 
> And just call myread(foo) instead of  foo.read(). 
> Where's the catch-22?

IMHO a better approach is a function like this:

  def temp_failure_retry(func, *args):
      while 1:
          try:
              return apply(func, args)
          except (IOError, os.error), detail:
              if detail.args[0] != errno.EINTR: raise  

(the strange function name come from the similar C macro used in the
GNU libc ;)

An example of the use of this function with open:

    fd = temp_failure_retry(open, "pippo", "w");

I don't tryed it, but AFAIK it should work.

Bye,

-- 
d i e g o
--
To reply remove the numbers and the `x' from my address
--
Sorry for my bad English!



From gobry at blanc.idiap.ch  Fri Jun 18 08:04:51 1999
From: gobry at blanc.idiap.ch (Frederic Gobry)
Date: 18 Jun 1999 13:04:51 +0100
Subject: BiBTeX module?
References: <joe-160619991358444033@chinacat.salk.edu> <7k950a$hsb@cs.vu.nl> <slrn7miqcg.tpi.bernhard@alpha1.csd.uwm.edu>
Message-ID: <slrn7mk9ui.dic.gobry@blanc.idiap.ch>

>Having a look at the current programs available for editing and entering
>BiBTeX files someone really ahs to write a good python/tk or
>python/wxpython program for this task, too.

I've also started such a project as free software. The parser and the python
interface are working correctly now. My idea was to write something generic
for bibliographic databases, BibTeX being only the first module. A GNOME
interface is also in progress (does some display but no real edition yet)

So I can make a snapshot for you or you can use an old version : the new
parser tries to interpret things like lists of authors, first/last names,
and to deal with capitalization, whereas the old one just isolates the
BibTeX fields.

Fr?d?ric

-- 
    Fr?d?ric GOBRY                  IDIAP
                               Rue du Simplon 4,
  Research Assistant          CH 1920 - Martigny
Machine Learning Group      Tel: +41 27 721 77 31



From jkraai at murl.com  Wed Jun  9 01:26:25 1999
From: jkraai at murl.com (jkraai)
Date: Wed, 9 Jun 1999 05:26:25 GMT
Subject: SWIG'd HTML Tidy?
References: <1283218560-70098357@hypernet.com>
Message-ID: <375DFB01.937D8B17@murl.com>

Has anyone SWIG'd HTML Tidy?

Thanks!

--jim



From fredrik at pythonware.com  Wed Jun  9 06:03:24 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 9 Jun 1999 10:03:24 GMT
Subject: "Programming Python" still worthwhile?
References: <7jki66$qam$1@news.gstis.net>
Message-ID: <00c301beb260$09e94680$f29b12c2@pythonware.com>

Rick Robino <see-sig at wavedivision.com> wrote:
> I've been watching this group for some time and just now getting warm to
> Python. Looking at source examples has been great, but for completeness'
> sake I'd like to get more of the "why" maybe found in a book.

Programming Python was written to given you the "why";
it's as much about the process of programming (and how
python affects that), as about the python language itself.
and most of it will work just fine with 1.6.  you'll find the
necessary updates (up until 1.5.2) at:

http://home.rmi.net/~lutz/

also note that Programming Python is *not* a tutorial
(though you'll find one in an appendix), and it's *not*
a reference manual.  many PP critics seems to have
missed that.

but you may find Learning Python a better choice,
though.  see below.

> Since Python is at 1.5.2 and people are talking about 2.0
> as if it's around the corner.

it's not.  forget everything you've ever heard about it.
it only exists as a few ideas in Guido's head, and he's
about to embark on a "programming for the masses"
project, so those ideas are likely to change...

next stop is 1.6.

probably early next year, or so. expect unicode, a
more modular interpreter, string methods, and pro-
bably lots of changes under the hood.  hardly any
major changes on the surface.

if I'd were you, I'd stop worrying about the future,
and learn to love what Python can offer you today!

(btw, there are LOTS of books coming out now, so
changing Python in a way that renders them useless
would be very, very stupid...  and if Guido suddenly
loses his mind, I can guarantee that others will step
in and maintain the 1.X series for another 30 years
or so...)

> I was wondering if anyone here could give me some advice as to whether or
> not "Programming Python" is useful, if the current edition is still
> current, if I should wait for the "Learning Python", or if I should use
> some other resources.

Learning Python is available, since long.  if you're a new-
comer to python, it's by far your best choice.

the five-star rating at amazon isn't there by accident...

btw, If you're going to buy it via amazon, consider
supporting the Python Software Association:

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

</F>

(and this concludes my weekly broadcast to c.l.py ;-)




From ovidiu at cup.hp.com  Thu Jun 24 19:40:34 1999
From: ovidiu at cup.hp.com (Ovidiu Predescu)
Date: Thu, 24 Jun 1999 23:40:34 +0000
Subject: getstate, setstate and pickle
References: <X%vc3.1830$VL2.283108@news.direcpc.com>
Message-ID: <3772C1F2.B919C0F2@cup.hp.com>

Emile van Sebille wrote:
> 
> I've written a class, instantiated, then pickled the instance.  At
> first, pickling wouldn't work at all, but through some mix of
> eliminating a class variable 'id' and adding __setstate__ and
> __getstate__ methods it does dump (too soon to call it a pickle).
> 
> What I'm having trouble with is getting the load to match the dump.  I
> lifted parts of the pickle module test code, and the dump == load part
> fails, then my code falls apart, and then nothing works... (I'm in
> California, so picking grapes is an option ;-)
> 
> I'm probably not using get/set state properly.  My instance should be
> pickleable or I'd get an error there.
> 
> Any help or pointers to examples/docs appreciated.

You don't need the __setstate__/__getstate__ methods unless you want to
exclude some instance variables from being pickled. On a normal class
the pickling should work without any problem. I also never had problems
with the name of the instance variables.

If you want to compare your own defined classes, don't forget to define
the __cmp__ method. This method defines how to compare instances of the
same class.

Here is a short example on how to pickle some objects:

from cPickle import Pickler, Unpickler
from StringIO import StringIO

class MyClass:
    def __init__ (self, value):
        self.value = value

    def __repr__ (self):
        return '[MyClass: %s]' % (self.value,)

    def __cmp__ (self, other):
        return self.value == other.value

object = MyClass (MyClass (123))
strIO = StringIO ()
pickler = Pickler (strIO)
pickler.dump (object)
strIO.seek (0, 0)
unpickler = Unpickler (strIO)
object2 = unpickler.load ()

print 'object: '+`object`
print 'object2: '+`object`

print object == object2


Hope this helps,
--
Ovidiu Predescu <ovidiu at cup.hp.com>
http://www.geocities.com/SiliconValley/Monitor/7464/



From mwh21 at cam.ac.uk  Wed Jun 23 02:49:22 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 23 Jun 1999 07:49:22 +0100
Subject: creating dictionnaries
References: <376F82A6.82497D2B@der.edf.fr> <FQSb3.973$7X1.249189@news.shore.net>
Message-ID: <m3bte72yz1.fsf@atrus.jesus.cam.ac.uk>

"Michael P. Reilly" <arcege at shore.net> writes:

> Christian Caremoli <Christian.Caremoli at der.edf.fr> wrote:
> : Hi,
> : Dictionnaries can be created like that :
> : d={'a':1,'b':2}
> 
> : By calling a function you create a dictionnary like that :
> : def f(**d):
> :    return d
> 
> : d=f(a=1,b=2)
> 
> : I would like to be able to create dictionnaries with some similar syntax
> : like keyed tuples :
> : d=(a=1,b=2)
> 
> : Is there a way to do that ?
> 
> In a word, No.  The structure of the language would perform a namespace
> lookup on "a" and "b" instead of taking them as strings.  Also a tuple
> is still an expression, and Python does not allow assignments inside
> expressions.  This means that you would get a SyntaxError exception
> during bytecode compilation.  And even if you didn't, it would be
> likely that you would get a NameError exception.
> 
>   -Arcege
> 
> Note: I haven't looked at bytecodehacks, but that might let you do what
> you wish.  Since it would be extremely non-portable, I wouldn't suggest
> it for published or long-term code.

What? byecodehacks can't affect syntax! I'm quite proud of that bit of
code, but lets not blow it up into things it just isn't...

yours,
Michael



From a.mueller at icrf.icnet.uk  Fri Jun  4 10:08:41 1999
From: a.mueller at icrf.icnet.uk (Arne Mueller)
Date: Fri, 04 Jun 1999 15:08:41 +0100
Subject: threads
References: <001001beae49$389b4ca0$949e2299@tim>
Message-ID: <3757DDE9.F10CC150@icrf.icnet.uk>

Thanks for all the information about threads.

I think the thread behaviour is a heavy disadvantage of python. However
it's not important in a case where you'd like to launch something in
background from a running application like printing or getting data from
a socket or whatever (remember those days when printing or copying files
locked your running application?). 

I think I'll try to solve my problem by switching from multiple threads
to different processes using 'fork'. The only problem I have is that all
these processes must have one object in common that collects all the
result data. Maybe a pipe or shared memory can help. Implementing my
computition intensive function in C is not a solution at the time - I'm
in a hurry ;-)

	cheers,

	Arne

-- 
Arne Mueller
Biomolecular Modelling Laboratory
Imperial Cancer Research Fund
44 Lincoln's Inn Fields
London WC2A 3PX, U.K.
phone : +44-(0)171 2693405      | Fax : +44-(0)171 269 3258
email : a.mueller at icrf.icnet.uk | http://www.icnet.uk/bmm/



From bruises at my-deja.com  Mon Jun 14 23:20:05 1999
From: bruises at my-deja.com (bruises at my-deja.com)
Date: Tue, 15 Jun 1999 03:20:05 GMT
Subject: Installing ODBC on PythonWin
References: <3760D550.2071295E@lemburg.com>
Message-ID: <7k4got$avh$1@nnrp1.deja.com>

Before I go wack and take it out on my computer
like this guy
(http://www.web42.com/badday/badday_small.mpg)
did, I'd like to ask for some help.

I'm trying to install an ODBC module into
PythonWin, but it still *won't* work!

It is written that the ODBC module is included in
the pythonwin distribution but >>>import ODBC
says otherwise. I've d/e'd (downloaded/executed)
Win32, but it died and said that it was looking
for version 1.4!

I was able to partially install mxODBC by copying
all the files into the right place but I'm having
trouble compiling the modules in vc++.

I'm going to pull my hair out!!! Could someone
show me an easier way of installing ODBC? (I
don't want to mess with DAO since I'm expecting
to migrate data from a database.)

Hhelp!!

SL


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From tim_one at email.msn.com  Sun Jun 13 18:56:04 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 13 Jun 1999 18:56:04 -0400
Subject: The Best Parsing Book on Earth (was RE: Interpolation)
In-Reply-To: <m3d7z0t6n0.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <000001beb5ef$eaec04a0$f09e2299@tim>

[Tim]
> I posted a link earlier to the Best Parsing Book on Earth; read the
> first few chapters of that first.

[Michael Hudson]
> Which one was that? I think I missed it!

Bears repeating, then (I've gotten many email inquiries too).  It's a
wonderful intro, now out of print, but thanks to the authors' generosity
downloadable for free:

    http://www.cs.vu.nl/~dick/PTAPG.html

and-changed-the-subject-line-for-future-generations-ly y'rs  - tim





From ronERASEjeffries at ERASEacm.org  Fri Jun 18 21:24:32 1999
From: ronERASEjeffries at ERASEacm.org (Ronald E Jeffries)
Date: Sat, 19 Jun 1999 01:24:32 GMT
Subject: [JPYTHON] Installation Problems
Message-ID: <0D1FF10FB191D45F.A19ABABF1E01F5C3.C3A5123AAD9506A3@lp.airnews.net>

Thought I'd try XP in Python, so I just downloaded JPython1.1beta2 and
when I installed it with JView, got this:

D:\Downloading\jpython>jview JPython11beta2
InstallShield JavaTM Edition
Extracting installation code................................done
Failed to execute MakeScriptAction

The files all appear to be in the directory I chose. There's a
"jpython", and a "jpython.jar".

However, when I type jpython at the DOS prompt, I get the dreaded

D:\Downloading\jpython>jpython
Bad command or file name

I'm supposing this is because the MakeScriptAction didn't run.

Suggestions welcome!

Also ... is there an IDE of any kind available for Python yet ... or
at least something that will get me out of the DOS box?

Thanks!

Ron Jeffries
Extreme Programmer
http://www.armaties.com



From wware-nospam at world.std.com  Wed Jun 23 09:33:46 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Wed, 23 Jun 1999 13:33:46 GMT
Subject: Thanks for Python...
References: <376ead64@anonymous.newsfeeds.com> <VrDQRddcbbs9-pn2-QdftOXYk6Hdd@Fortress>
Message-ID: <FDs8CB.CFF@world.std.com>

Jeff Rush (jrush at summit-research.com) wrote:
: Python is also just plain fun...  Now if I could just get time to port
: it to the PalmPilot, I'd never be away from it.

Do a web search for "uclinux". It's a Linux kernel being worked on
at Ryerson University somewhere in Canada. I don't know how stable
or developed it is, but it might run Python.

I'd like to see somebody take the old HP100LX palmtop design, pop out
the ROM, and put in a Python interpreter (and PIM stuff written in
Python) in its place. That would be cool.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com



From MHammond at skippinet.com.au  Wed Jun  9 18:34:08 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Thu, 10 Jun 1999 08:34:08 +1000
Subject: Python ActiveX Scripting interface
References: <375e87ae.11001540@news.origin-it.com> <6mw73.1227$h14.5409@newsr2.twcny.rr.com> <375ecca3.3972892@news.demon.co.uk>
Message-ID: <7jmq1s$fpq$1@m2.c2.telstra-mm.net.au>

Paul Moore wrote in message <375ecca3.3972892 at news.demon.co.uk>...

>Collecting Python Trace Output...
> ScriptEngine QI - unknown IID {4954E0D0-FBC7-11D1-8410-006008C3FBFC}
>Traceback (innermost last):
>  File "E:\Applications\Python\win32com\client\dynamic.py", line 247,
>in _make_method_
>    fn = self._builtMethods_[name] = tempNameSpace[name]
>KeyError: print

Hmm.  This is a bug.  There was a bug that IE5 has a "print" method, and
this being a reserved word upset Python.  All reserved words are translated,
but it appears this one is not.

I will try and recreate this soon.  But I am short of time.  Would you like
to help?  If so, mail me.

>Interesting - the IID it can't find is IActiveScriptProperty. What's
>that? It's not mentioned in MSDN (not even the online version on the
>MSDN web site, which I assume is fairly up to date...)

That is almost certainly not the problem.  There are a number of interfaces
QI'd for that we dont implement.

Mark.






From sydney96y at aol.com  Tue Jun 29 02:14:05 1999
From: sydney96y at aol.com (sydney96y at aol.com)
Date: Tuesday, 29 Jun 1999 00:14:05 -0600
Subject: 18+  100981
Message-ID: <29069900.1405@aol.com>

Pirate Password Links Below (check 'em out before the page dies):

http://freedailyvirgins.the14u.com





nGK)LUcQ.>



From JamesL at Lugoj.Com  Sun Jun 13 16:45:15 1999
From: JamesL at Lugoj.Com (James Logajan)
Date: Sun, 13 Jun 1999 13:45:15 -0700
Subject: Variations on a theme by "pound bang"
References: <3763e75f.3265255@news.iol.ie> <3763F77E.6094AEE9@appliedbiometrics.com>
Message-ID: <3764185B.7BCBC488@Lugoj.Com>

Christian Tismer wrote:
> 
> Sean Mc Grath wrote:
> >
> > Can anyone shed some light on the benefits of:-
> >
> > # !/usr/bin/env python
> 
> Uses the env program which is allways there, to figure out
> from the environment settings which python to use.

Alas, this is a myth. Not all Unixen put env in /usr/bin, some put it in
/bin. I can't recall if it was Solaris (probably not) or LynxOS (probably)
that put env in /bin. Anyway, I got bit by it.



From jody at ldgo.columbia.edu  Sun Jun 27 09:36:51 1999
From: jody at ldgo.columbia.edu (Jody Winston)
Date: 27 Jun 1999 09:36:51 -0400
Subject: Python for large projects?
References: <3774b5d1.690922@news.force9.net>
Message-ID: <tx3ogi1ahos.fsf@hog.ldgo.columbia.edu>

I've used Python for three large distributed applications (each > 100k
lines of Python).  The only other recommendation that I would add is
to define interfaces to the code.  If this is done with a product like
ILU or FNORB, you can then easily call the Python code from other
languages.  In addition, you can replace critical sections with other
languages if needed.

-- 
Jody Winston
Lamont-Doherty Earth Observatory
RT 9W, Palisades, NY 10964
jody at ldeo.columbia.edu, 914 365 8526, Fax 914 359 1631 

Under US Code Title 47, Sec.227(b)(1)(C), Sec.227(a)(2)(B) This email
address may not be added to any commercial mail list with out my
permission.  Violation of my privacy with advertising or SPAM will
result in a suit for a MINIMUM of $500 damages/incident, $1500 for
repeats.





From wware-nospam at world.std.com  Sun Jun  6 21:55:23 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Mon, 7 Jun 1999 01:55:23 GMT
Subject: "The Python Way"
References: <045501beadf8$610e62e0$f29b12c2@pythonware.com> <9906031407150D.00552@quadra.teleo.net>
Message-ID: <FCxpCB.4rn@world.std.com>

Patrick Phalen (pphalen at teleo.net) wrote:
: Would both Guido and TIm Peters be willing to collaborate on a short
: paper ...
: What I have in mind is sort of a very brief Strunk-&-White-like
: "Elements of Style" for Python, which suggests fundamental idiomatic
: recommendations for operating within the spirit of the language.

In the turmoil following this proposal, its actual usefulness may have
been overlooked. It sounds to me like a great idea. I'm sure I could
learn much stylistically from people who've been using Python for a
long time and have been involved in its design. Just as I might hope
to make myself a better C/Unix programmer by reading any thoughts that
Kernighan/Ritchie/Pike/etc might feel like passing along. Think of it
not as "damage control to keep out the barbarians" but rather as the
well-informed recommendations of some of the world's most skillful
Python programmers (assuming they agree to do it).
-- 
 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Will Ware		email: wware[at]world[dot]std[dot]com
PGP fp (new key 07/15/97) 67683AE2 173FE781 A0D99636 0EAE6117



From radbit at my-deja.com  Wed Jun 16 16:23:21 1999
From: radbit at my-deja.com (Florent Heyworth)
Date: Wed, 16 Jun 1999 22:23:21 +0200
Subject: Anonymous Pipes (Was do anonymous pipes normally work on NT?)
Message-ID: <376807B9.39B84592@radbit.com>

Hi

looking at the test scripts Georg Mischler provided it seems the code is
failing
to pass on the standard handles to the created process. The following
method works:

file runproc.py
---------------

'''runproc.py

start a process with three inherited pipes.
Try to write to and read from those.
'''

import win32api
import win32pipe
import win32file
import win32process
import win32security
import win32event
import os

#Constants for stdandard handles
STD_ERR_HANDLE=-12
STD_OUTPUT_HANDLE=-11
STD_INPUT_HANDLE=-10

class Process:
    def run(self, cmdline):

        # security attributes for pipes
        sAttrs = win32security.SECURITY_ATTRIBUTES()
        sAttrs.bInheritHandle = 1

        # create pipes
        hStdin_r,self.hStdin_w=win32pipe.CreatePipe(sAttrs, 0)
        self.hStdout_r,hStdout_w=win32pipe.CreatePipe(sAttrs,0)
        self.hStderr_r,hStderr_w=win32pipe.CreatePipe(sAttrs, 0)

        #associate Standard output to previously open handle
        win32api.SetStdHandle(STD_OUTPUT_HANDLE,hStdout_w)

        # set the info structure for the new process.
        StartupInfo = win32process.STARTUPINFO()

        # start the process.
        hProcess, hThread, dwPid, dwTid=win32process.CreateProcess(
                None, #program
                cmdline,# command line
                sAttrs,  # process security attributes
                sAttrs,  # thread attributes
                1,      # inherit handles,
                win32process.NORMAL_PRIORITY_CLASS,
                None,  # no new environment
                None,  # current directory (stay where we are)
                StartupInfo)
        # normally, we would save the pid etc. here...


        res, str = win32file.ReadFile(self.hStdout_r,100)
        if res == 0: print 'read:', str
        else: 'read nothing.'


if __name__ == '__main__':
    p = Process()
    p.run('..\\python.exe testpipe.py')
-----------------
end of runproc.py

file testpipe.py - explicit assignement of standard handles
----------------
import win32api
import win32file

STD_ERR_HANDLE=-12
STD_OUTPUT_HANDLE=-11
STD_INPUT_HANDLE=-10

std_in=win32api.GetStdHandle(STD_INPUT_HANDLE)
std_err=win32api.GetStdHandle(STD_ERR_HANDLE)
std_out=win32api.GetStdHandle(STD_OUTPUT_HANDLE)

win32file.WriteFile(std_out, 'this method works....')

end of testpipe.py
------------------

file testpipe1.py implicit standard handles
------------------
print "this method works as well..."

end of testpipe1.py
-----------------

Cheers

Florent Heyworth






From mlh at idt.ntnu.no  Fri Jun  4 08:44:13 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 04 Jun 1999 14:44:13 +0200
Subject: Competing algorithms...
Message-ID: <y0jwvxkuode.fsf@vier.idi.ntnu.no>

Yay!

I have managed to get the course on algorithms here at the university
to use Python as its language!

Now, I have to find some cool uses for it :) I am planning on setting
up some programming competitions as a part of the course; i.e.
students send their algorithms (in Python) to a (rexec-) server which
benchmarks them and ranks them. So... Have you got any good ideas for
fair benchmarking?

The evaluation should preferrably be done automatically, so the only
thing I can think of is straight benchmarking baset on clocking and
test cases... But is this a good and fair measure of how good the
algorithm is? (Of course, a O(n) algorithm would probably out-perform
an O(n^2) algorithm...)

It might also be possible to do manual evaluation of the best ones, or
something - or to have a separate prize for beautiful programs...
Anything is possible...

eagerly-awaiting-several-exciting-suggestions-ly y'rs

 -M

--

  Magnus
  Lie
  Hetland        http://arcadia.laiv.org <arcadia at laiv.org>



From billtut at microsoft.com  Tue Jun 15 01:22:31 1999
From: billtut at microsoft.com (Bill Tutt)
Date: Tue, 15 Jun 1999 05:22:31 GMT
Subject: Installing ODBC on PythonWin
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8100DB90DC6@RED-MSG-50>

After installing the Python interpreter, and the Win32 extensions. 
(They're available at: http://www.python.org/download/download_windows.html)

Check out:
http://www.python.org/windows/OdbcHints.html

For hints on using ODBC with Python on Windows.

The answer to your question is that inorder to import the odbc module, you
need to import the dbi module first.
e.g.:
import dbi
import odbc

conn = odbc.odbc('DSNname')
cursor = conn.cursor()
cursor.execute('select * from EndUser')
....


Bill



From suzuki611 at okisoft.co.jp  Sun Jun 20 19:17:18 1999
From: suzuki611 at okisoft.co.jp (Hisao Suzuki)
Date: 21 Jun 1999 08:17:18 +0900
Subject: fork()
References: <000101beb886$628a1ea0$099e2299@tim>
Message-ID: <u3u2s2mphd.fsf@ares.sys1.ngo.okisoft.co.jp>

In the view of some (or many?) Python programmers, to remove all
the references to an object is the obvious and deterministic way
to destroy the object.  The predictability of destructor
invocation favored by some (or many?) Pythoneers is based on
this rule.

She/he consciously writes codes to destroy the object based on
the rule.  Just the same, any C++ programmer consciously writes
`delete's to destroy the `new'ed object.  Both intents are very
similar to each other.  From _this_ point of view, what
"Tim Peters" <tim_one at email.msn.com> wrote
in article <000101beb886$628a1ea0$099e2299 at tim>:
> To me, that merely describes the conditions under which the respective
> languages treat an object as immortal today.  It makes good sense for
> Stroustrup to say that in C++ a garbage collector shouldn't invoke the
> finalizer for its flavor of immortal objects, but he's not arguing that
> position *because* they're immortal today, but instead because explictly
> new'ed objects never have their finalizer invoked in C++ unless explicitly
> delete'd, and GC is not explicit delete'tion.  

sounds too particular about superficial differences.  However,

> Python has no such rules, so
> the core of his argument doesn't apply to Python without more strain than I
> can swallow <0.9 wink>.

is certainly consistent and valid in the _official_ semantics
described in the Python Reference Manual.

At the same time, in the current Pytnon (or rather CPython), the
idea that to remove all references corresponds to the delete
operation in C++ (and thus being part of cycles in Python
corresponds to being never deleted in C++) is _also_ consistent.
And moreover, it is a valid working rule practically useful to
predict the object destruction.

I described it against what
Moshe Zadka <moshez at math.huji.ac.il> wrote in
article <Pine.SUN.3.95-heb-2.07.990611075346.4942G-100000 at sunset.ma.huji.ac.il>
< On Fri, 11 Jun 1999, Guido van Rossum wrote:
< <snip/>
< > ...this almost seems acceptable: in the formal semantics, objects
< > that are part of cycles would live forever, and as an invisible
< > optimization we recycle their memory anyway if we know they are
< > unreachable.  (I've got the feeling that I've seen this rule before
< > somewhere.)  
< <snip/>
< No, that can't be. If we did, it were in Scheme, and after you put it in,
< you'd be morally obliged to add full-fledged lambda, tail-recursion and 
< user visible continuations. 

There I said that Guido's idea is quite consistent and orthodox
in a sense which has almost nothing to do with Scheme.  If one
adopts the idea such as I described above, one can write rightly
working programs, can predict their behavior more precisely, and
might think well of Guido's idea (even if not definitively; it
is not the sole consequence even for C++, you know).

At the same time, if one adopts the _official_ weaker semantics
(which is also written by Guido), one can still (needless to
say!) write rightly working programs, and the above idea of
Guido would lose some power of persuasion (even if not
rejected).  To remove all the references _no_ longer means the
immediate destruction of the object.

All in all, each idea is a valid model for the current Python.
The definitive difference lies in the semantics of language it
relies on.  Regarding the favor of destruction predictability
such as seen in this newsgroup, the current actual semantics is
perhaps as important as the official one.

Nevertheless, please do not think that I stick to one idea.
Actually I have also a third idea, though it might be useless
for the purpose of storage reclamation (see article
<u36756gcoc.fsf at ares.sys1.ngo.okisoft.co.jp> on 03 Jun 1999
12:56:03 +0900).  I am not yet sure which one is better than
others in total ...


> Whose point <wink>?  Java has always guaranteed to "destroy objects in an
> order that ensures that the destructor for one object doesn't refer to an
> object that has been previously destroyed", and indeed "without help from
> the programmer".  So by "Java solved that one" I meant it satisfactorily
> addressed the points made in the quote.  That doesn't mean there aren't
> other points to be made, but since the quote you gave didn't make any other
> points I don't feel bad about commenting on the points it did make <wink>.

Even if the storage of such objects are not reclaimed yet, their
finalizer()s may have been already invoked.  This may be a real
problem for aggregation objects (if you once rely on finalizers
in Java).  Note that Stroustrup's "destroy" implies "call the
destructor".

The JLS says:
  "12.6.2 Finalizer Invocations are Not Ordered

   Java imposes no ordering on finalize method calls.  Finalizers
   may be called in any order, or even concurrently.

   As an example, if a circularly linked group of unfinalized
   objects becomes unreachable (or finalizer-reachable), then all
   the objects may become finalizable together.  Eventually, the
   finalizers for these objects may be invoked, in any order, or
   even concurrently using multiple threads.  If the automatic
   storage manager later finds that the objects are unreachable,
   then their storage can be reclaimed."

Therefore, the authors of the JLS continue:

  "It is straightforward to implement a Java class that will cause
   a set of finalizer-like methods to be invoked in a specified
   order for a set of objects when all the objects become
   unreachable.  Defining such a class is left as an exercise for
   the reader."

You see, to implement such a class needs _help_ from the
programmer.  It would be like the following.  It might be very
boring to write codes like this every time from the scratch, and
the component classes might lose flexibility and re-usability.

  public class HeadOfAggregation {
      int count;
      int expected_count;  ...
      protected void finalize() { trigger(); }
      public void trigger() {
          boolean r = false;
          synchronized (this) {
              count++;
              r = (count == expected_count);
          }
          if (r) finalize_aggregation_totally();
      }  ....
  }

  public class ComponentOfAggregation_1 { ....
      protected void finalize() { our_head.trigger(); }
  }

  public class ComponentOfAggregation_2 { ....
      protected void finalize() { our_head.trigger(); }
  }

--===-----========------------- Sana esprimo naskas sanan ideon.
SUZUKI Hisao            suzuki611 at okisoft.co.jp, suzuki at acm.org.



From tim_one at email.msn.com  Fri Jun  4 21:42:27 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 4 Jun 1999 21:42:27 -0400
Subject: "The Python Way"
In-Reply-To: <7j8pra$55v$4@cronkite.cc.uga.edu>
Message-ID: <000d01beaef4$ab3b5c40$f19e2299@tim>

[Patrick Phalen, misattributed to Fredrik Lundh]
> Would both Guido and TIm Peters be willing to collaborate on a short
> paper -- call it "The Python Way" for lack of a better title -- which
> sets out the 10-20 prescriptives they might offer to those who come to
> Python from other languages and immediately want to find a way to bend
> it into uncomfortable positions -- (implement closures, etc.).

[Graham Matthews]
> Oh thank you Fredrik for posting the above.

See above; wrong fellow.

> Does anyone else consider that the above sounds slightly paranoid?
> ...

Less so than your response <0.6 wink>.  I saw no hint that Patrick was
referring to you or your current agenda.  In context, he was clearly asking
for the list of "Pythonic Principles" from which Guido supposedly deduces--
with exacting rigor <choke> --what goes in and what stays out.  That's all.

BTW, refcount + gc is a great idea!  Don't hold your breath anyway.

lack-of-time-is-a-near-universal-ly y'rs  - tim


PS:  Rafael Lins has done some very interesting work on adding GC to
refcounted systems, and that was looked into deeply a couple years ago.  The
killer for that approach in the end was that you can reach almost the entire
system from a suprisingly high number of Python objects (e.g., any function
object points to its module namespace, which often imports sys, from which
sys.modules can be reached, from which "the entire system").  Would have
been real slick had it been practical.  Nobody has had time to make a
similar effort with a dumber approach, else 1.5.2 would likely have it
already.





From gmcm at hypernet.com  Fri Jun 18 10:57:47 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 18 Jun 1999 09:57:47 -0500
Subject: Newbie: Truth values (three-valued logic)
In-Reply-To: <Pine.OSF.4.05.9906181317100.413-100000@pinax.Mathematik.Uni-Bielefeld.DE>
References: <7kd3ke$d1n$1@nnrp1.deja.com>
Message-ID: <1282404372-119074879@hypernet.com>

Olaf Delgado writes:

[three valued logic: one interpretation of "not maybe"]

> Okay, but this kind of intuitive semantics is not what I need here.
> The negation of "Maybe, I'll go to the cinema tomorrow." should be
> "Maybe, I wan't go to the cinema tomorrow.". By the way, this is IMO
> another, stronger argument against hardwiring multi-valued logics
> into a programming language. There may be different, equally
> justified views on what the natural behaviour would be.

Note that, if you want to retain DeMorgan's Laws, you don't have 
much choice. ANSI SQL defines a three valued logic which meets these 
conditions, and comes up with 
  NOT maybe == maybe
(well, NOT unknown == unknown), which makes sense as long as you 
rmember that NOT is an operator, and the result of a non-trivial 
operation on an unknown operand is probably unknown.

Which is why most DBA's won't allow NULL values in any column that 
might be used in a WHERE clause - three valued logic is often 
non-obvious.

Then there's fuzzy logic, where MAYBE is a whole range of values, and 
the rules are different.

- Gordon



From htrd90 at zepler.org  Thu Jun 10 16:07:40 1999
From: htrd90 at zepler.org (Toby Dickenson)
Date: Thu, 10 Jun 1999 20:07:40 GMT
Subject: COM Question - QueryInterface?
References: <375C7D56.C9744E39@connix.com> <375fe0b1.1117810@news.freeserve.net> <375EFF29.2C1D0FB7@connix.com>
Message-ID: <3760ffc3.2905460@news.freeserve.net>

(posted and cced to Chris)

Chris Tavares <tavares at connix.com> wrote:

>Thanks. The reason that the interface in question isn't a dual is because my
>object design depends quite heavily on having multiple interfaces. If you have
>multiple duals, you end up with lots of problems with IDispatch, since you can't
>safely implement more than one version of the same interface on a single object
>(and lots of people have tried, REALLY HARD).

This is a fallacy about IDispatch that I've heard from many sources, but it
certainly is not true.

One true statement is that you can only have one interface known as
IID_IDispatch. This causes a problem for scripting languages that can not
QueryInterface (or, worse, which QueryInterface behind your back).[

If you need to support this kind of client (VBScript, JScript, what else?
definitely not Python) then you need to implement IID_IDispatch with the union
of the methods from all your other interfaces. The page Chris mentions (below)
has a great summary of some solutions to this.

>So what I'm doing is having multiple oleautomation compatible interfaces, and then
>using a typelib-driven IDispatch implementation that I got from
>http://www.sellsbrothers.com/tools/multidisp/index.html. This automagically merges
>all my separate interfaces into a pseudo-object model. Really neat stuff, and
>saves a TON of work.

Follow that advice if you need to support those lesser scripting clients, or
don't if you don't.

In any case, there is still no reason not to make all your other interfaces dual
too. This should be an easy step. I am assuming you already have some IDL for
these interfaces? Change the oleautomation line to dual, and hook up the same
typelib-driven IDispatch implementation you are already using for IID_Dispatch.
If you are using ATL for this implementation then it all comes for free. The
advantages? You get easier access from Python, and in VB you get compile time
type safety.

In your specific example, the type library would say your collection object's
Item method returns an ITavaresDrive (or whatever) and python would use an
appropriate makepy-generated wrapper.

>Having tried that, I find that I now have access to IDispatch::Invoke. Well, what
>I really wanted was win32com.client.dynamic.Dispatch. How do I get to the wrapped
>IDispatch rather than the raw one?

drive = drives.Item("c:")._oleobj_.QueryInterface(pythoncom.IID_IDispatch)
win32com.client.Dispatch(drive)

(from memory)

I hope this helps,


Toby Dickenson



From mwh21 at cam.ac.uk  Thu Jun  3 02:27:00 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 03 Jun 1999 07:27:00 +0100
Subject: Bizarre error while ./configure-ing PIL
References: <m3g14dxduq.fsf@ime.net> <m3n1yii8w9.fsf@atrus.jesus.cam.ac.uk> <m3k8tmgmpc.fsf@delorges.in-berlin.de>
Message-ID: <m3k8tliytn.fsf@atrus.jesus.cam.ac.uk>

Jo Meder <jo at delorges.in-berlin.de> writes:

> Michael Hudson <mwh21 at cam.ac.uk> writes:
> 
> > Matt Gushee <mgushee at ime.net> writes:
> 
> [reasonable permissions for ./configure]
> > > bash: ./configure: Permission denied
> 
> > Random guess, but check the first line of ./configure. It should
> > probably read
> > 
> > #!/bin/sh
> > 
> > My guess is either 1) /bin/sh is not executable (very unlikely, all
> > sorts of things would barf at that) 2) the line's gotten mangled
> > somehows - maybe a misplaced space?
> 
> Actually the line reads
> 
> #! /bin/sh
>   ^note the space
> 
> but why this should lead to "Permission denied", I don't know. 
> 
> I tried it with tcsh, zsh, bash, ksh, ash and not one of them barfed
> at the line.

Well, the shell's not getting to see the line (if a shell sees the
line, it sees the # and thinks it's a comment.

It's the kernel that's getting confused. I don't know what the space
between the ! and the / does, but if it reads

#!/bin/ sh

the kernel tries to execute /bin/, finds it's a directory and
complains. Having a space between ! and / works on my system (RedHat
6.0 + kernel 2.2.9).

Also if the executable isn't there you get very bizarre messages:

[mwh21 at atrus mwh21]$ cat > shellscript.sh
#!/bin/frumble
frub
[mwh21 at atrus mwh21]$ chmod u+x shellscript.sh
[mwh21 at atrus mwh21]$ ls shellscript.sh
shellscript.sh
[mwh21 at atrus mwh21]$ ./s
shellscript.sh  src
[mwh21 at atrus mwh21]$ ./shellscript.sh
bash2: ./shellscript.sh: No such file or directory
[mwh21 at atrus mwh21]$ 

Leading to the "what the f.? the file's there you stupid machine"
response.

Michael




From nathan at islanddata.com  Wed Jun 30 13:38:16 1999
From: nathan at islanddata.com (Nathan Clegg)
Date: Wed, 30 Jun 1999 10:38:16 -0700 (PDT)
Subject: socket timeouts
Message-ID: <XFMail.990630103816.nathan@islanddata.com>

Is there any way to set the timeout for a TCP/IP connection? 
Specifically, I'm writing some apps meant to run on the same machine, so
if they can't connect *immediately* (i.e. the server app is not running), I
need to go to plan B.  Any answers or alternate solutions?  I'm also
interested in a similar issue across the internet, but I'm more generous
about timeouts in this circumstance.  Thanks.


----------------------------------
Nathan Clegg
 nathan at islanddata.com





From mdcb at ing.iac.es  Mon Jun 14 06:56:48 1999
From: mdcb at ing.iac.es (Mathieu Bec)
Date: Mon, 14 Jun 1999 11:56:48 +0100
Subject: Extending with POSIX thread
Message-ID: <3764DFEF.1A8889DD@ing.iac.es>

Hi,

Is there any restriction on extending Python with pthread?

Basically, I want to create a background process that blocks on a select
call.
My prototype 'seems' to be working fine using  pthread,
but I'd like to be sure ... or if there's any more viable alternative
(fork ?)

cheers,

--
Matthieu Bec                       Isaac Newton Group
tel  : +34 922 425427              Apartado de Correos 321
fax  : +34 922 425401              38780 Santa Cruz de La Palma
url  : www.ing.iac.es/~mdcb        Canary Islands, Spain






From Drew.Whitehouse at anu.edu.au  Wed Jun 30 06:23:08 1999
From: Drew.Whitehouse at anu.edu.au (Drew Whitehouse)
Date: 30 Jun 1999 20:23:08 +1000
Subject: Python in the public eye !
Message-ID: <wl3dza2dir.fsf@anu.edu.au>


Anyone who visits the Powerhouse Museum (Sydney, Australia) might be
interested to know that the "Wedge" virtual environment exhibit in the
new "Universal Machine" permanent exhibition which opened last night
is Python powered.  The whole program for the exhibit was put together
over the last fortnight on top of our in house "pSpace" VE system
which use python as it's embedded language. I got to watch primary
school kids using it for the first time today and it was a real
buzzzz. Thanks Guido and Co !!!

	http://www.phm.gov.au/universal/
	http://anusf.anu.edu.au/pSpace/


-Drew

-- 
;; mailto:Drew.Whitehouse at anu.edu.au   http://anusf.anu.edu.au/~drw900/
;; Viz programmer, Australian National University Supercomputer Facility




From jtravs at debian.org  Fri Jun 25 18:45:29 1999
From: jtravs at debian.org (John Travers)
Date: Fri, 25 Jun 1999 23:45:29 +0100
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <59EAFD6B662F0706.86BFF3E906CBF8DE.784D2A219F85BC21@lp.airnews.net>
Message-ID: <99062523473500.00769@Vegus>

Has anybody tried the PyKDE/Qt package, (you don't have to have KDE), it is
absoloutly fantastic, quick, easy, powerful, far far better than Tkinter.
I love it!

 --
John Travers

"Sometimes I think the surest sign that intelligent life exists elsewhere
in the universe is that none of it has tried to contact us!"



From m.r.poppenk at sightline.nl  Mon Jun 14 06:34:30 1999
From: m.r.poppenk at sightline.nl (Marc R. Poppenk)
Date: 14 Jun 1999 10:34:30 GMT
Subject: Python and Netware
Message-ID: <3764D7EE.4C87@sightline.nl>

Hello,

I'm looking for some one who can give me the answers to these questions:
- Is a Python version for Novell Netware available?
- Can it be downloaded somewhere?

Or
- Do I have to port Python for Netware myself?
- Does any one have experience with this?
  (Known problems, sollutions, instructions)

Thanks a lot,
Marc Poppenk


-- 
Marc Poppenk
SightLine Informationtechnology
Hengelosestraat 705
Postbus 1379
7500 BJ Enschede

T +31 53 4836680
F +31 53 4836582
E m.r.poppenk at sightline.nl



From sekter at mail.matav.hu  Mon Jun 28 04:54:37 1999
From: sekter at mail.matav.hu (Arpad Kiss)
Date: Mon, 28 Jun 1999 10:54:37 +0200
Subject: ANN: my python widgets
Message-ID: <003901bec143$dae9fd50$499f38c3@geometria.hu>

Hi!
I have completly redesign my widgets(TreeBox, ComboBox, Tabs...) and you can
download them from my starship page(http://starship.python.net/crew/arpadk).
I have tested them on Windows NT, Linux and Win95. It needs Python 1.5.2 and
tcl/tk 8.0.5.
Arpad Kiss





From fbo at dde.dk  Tue Jun 22 06:38:15 1999
From: fbo at dde.dk (Finn Bock)
Date: Tue, 22 Jun 1999 12:38:15 +0200
Subject: How to call a Java superclass from jpython ??
References: <376E9415.8E002ACA@t-online.de>
Message-ID: <376F6797.B4C602EE@dde.dk>

"Markus A. Greiner" wrote:

> I?m trying to port some java applications into jpython.
> Now I got a problem. I?m not able to call a java superclass from
> jpython.
> Did anybody know how to call them? Here the code I?m trying to convert:
> 
> It?s from PropertiesMetalTheme.java of the JDK Metalworks example.
> In method initcolors:
> 
>         ...
>         primary1 = super.getPrimary1()
> 
> Everything I tried in jpython failed. In some other cases replacing the
> super call into a construct like:
> 
>         ...
>         ParentClass.method(self)
> 
> works fine, but in this case
> 
>         ...
>         DefaultMetalTheme.getPrimary1()

I'm sure you mean
          DefaultMetalTheme.getPrimary1(self)
But either way, it does not work in JPython1.1beta2

> and all other constructs failed. I?m using JPython 1.1 beta 2 so
> calls like self.super__method are obsolte and didn?t work anymore.
> (I think :-)) or better I tried without success)

The super__ methods should no longer be necessary, but they still work,
so as a temporary workaround you could perhaps use something like this:

from javax.swing.plaf import metal

class PropertiesMetalTheme(metal.DefaultMetalTheme):
    def initColors(self):
        self._primary1 = self.super__getPrimary1()
    def getPrimary1(self):
        return self._primary1

p = PropertiesMetalTheme()
p.initColors()
print p.primary1

BTW, the chance of an answer to jpython questions are much higher in
jpython-interest mail list.

-- 
Regards
Finn Bock



From tim_one at email.msn.com  Wed Jun  9 22:22:03 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 9 Jun 1999 22:22:03 -0400
Subject: fork()
In-Reply-To: <7jmbe1$r22$2@cronkite.cc.uga.edu>
Message-ID: <000901beb2e8$071f4840$af9e2299@tim>

{Graham]
> Which "hoary old theoretical arguments" are you talking about Tim.

The historical literature is rampant with them.  If you feel I was accusing
*you* of making them in this particular leaf of the thread, sorry, that's
just a misunderstanding.

> I said "copying memory *can* speed applications up". That is an
> observable reality.

I agreed.  Also said it can slow them down, and gave a pointer to the most
helpful accessible reference I know of, for the benefit of anyone
interested.

> Where are the hoary theoretical arguments?

See the Jones & Lins book for a lucid presentation-- and debunking --of
them.

> Or is this just more of the same old bluff and bluster arguing.

Nope -- still too busy beating my wife, I guess <wink>.

oh-damn-now-i-have-to-get-married-ly y'rs  - tim





From aa8vb at yahoo.com  Fri Jun 25 13:14:25 1999
From: aa8vb at yahoo.com (Randall Hopper)
Date: Fri, 25 Jun 1999 13:14:25 -0400
Subject: PyOpenGl problem
In-Reply-To: <99062220260100.01217@Vegus>; from John Travers on Tue, Jun 22, 1999 at 08:21:17PM +0100
References: <99062220260100.01217@Vegus>
Message-ID: <19990625131425.A762231@epa.gov>

John Travers:
 |I have compiled PyOpenGl and installed it but whenever I run a demo I get:
 |
 |File "./trees.py", line 1 in ?
 |	from oglpm import *
 |File "./oglpm.py", line 4, in ?
 |            from OpenGL.GL import *
 |File "/usr/lib/python1.5/site-packages/OpenGL/GL/__init.py", line 4, in ?
 |	from opengl_num import *
 |ImportError: /usr/lib/libMesaGL.so.3: undefined symbol: XFreePixmap.
 |
 |Is this the wrong version of mesa or something?

This dynamic symbol resolution error likely indicates that libMesaGL.so.3
was not correctly linked at compile time (e.g. "ld -o libMesaGL.so ... -lX11")

Randall



From zooko at xs3.xs4all.nl  Tue Jun  8 03:31:11 1999
From: zooko at xs3.xs4all.nl (Zane Lewkowicz)
Date: 8 Jun 1999 09:31:11 +0200
Subject: I'll pay $40 in e-gold if you'll fix this bug.
Message-ID: <7jigrv$ppo$1@xs3.xs4all.nl>

Greetings to Pythonistas, Pythoniandos and Python Hackers 
everywhere!


I have written some Python scripts for my own amusement, and 
I'm stymied by a "bug" involving traversing symlinked 
directories in a Linux ext2 filesystem.


I'm offering 40.00 USD for a patch to my code that makes my 
"scoot()" function do what I want it to, like this:


------- begin imaginary bash transcript
bash$ cd /tmp
bash$ mkdir dirOne
bash$ mkdir dirTwo
bash$ mkdir dirTwo/dirThree
bash$ ln -s dirTwo/dirThree .
bash$ cd dirThree
bash$ echo "hi there" > file.txt
bash$ scoot.py file.txt ../dirOne 
bash$ cd /tmp
bash$ cd dirOne
bash$ ls -al
total 3
drwx------   2 zooko    zooko        1024 Jun  8 01:10 .
drwx------   4 zooko    zooko        1024 Jun  8 01:07 ..
-rw-------   1 zooko    zooko           9 Jun  8 01:09 file.txt
bash$ cd /tmp/dirThree
bash$ ls -al
total 2
drwx------   2 zooko    zooko        1024 Jun  8 01:10 .
drwx------   3 zooko    zooko        1024 Jun  8 01:07 ..
lrwxrwxrwx   1 zooko    zooko          21 Jun  8 01:10 file.txt -> ../../dirOne/file.txt
bash$ cat file.txt
hi there
bash$
------- end imaginary bash transcript


Currently, it does this:


------- begin actual bash transcript
bash$ cd /tmp
bash$ mkdir dirOne
bash$ mkdir dirTwo
bash$ mkdir dirTwo/dirThree
bash$ ln -s dirTwo/dirThree .
bash$ cd dirThree
bash$ echo "hi there" > file.txt
bash$ scoot.py file.txt ../dirOne 
/bin/mv: cannot create regular file `../dirOne/file.txt': No such file or directory
bash$ cd /tmp
bash$ cd dirOne
bash$ ls -al
total 5
drwx------   2 zooko    zooko        1024 Jun  8 01:12 .
drwxrwxrwt  12 root     root         4096 Jun  8 01:12 ..
bash$ cd /tmp/dirThree
bash$ ls -al
total 3
drwx------   2 zooko    zooko        1024 Jun  8 01:12 .
drwx------   3 zooko    zooko        1024 Jun  8 01:12 ..
-rw-------   1 zooko    zooko           9 Jun  8 01:12 file.txt
bash$ cat file.txt
hi there
bash$
------- end actual bash transcript


The code for "scoot.py" and the related functions in "zutil.py"
is available at 
"http://wildgoose.tandu.com/~zooko/PythonHacking/".



I have $40.00 worth of e-gold ("http://e-gold.com/") which 
I'll give to the author of the best, most elegant, solution 
(IMO, of course).


If you create an e-gold account (takes less than 5 minutes) 
then I can transfer the money to you in instantaneous digital 
form, or if you prefer I'll have e-gold.com debit my account 
and cut you a snail mail check.


I'll keep the newsgroup posted of what kind of response I get 
to this offer.


:-)


Regards,

Zooko

-------
"Must it be assumed that because we are engineers beauty is not
 our concern, and that while we make our constructions robust 
 and durable we do not also strive to make them elegant?"
                                              -- Gustave Eiffel



From jhauser at ifm.uni-kiel.de  Sun Jun 27 03:28:08 1999
From: jhauser at ifm.uni-kiel.de (Janko Hauser)
Date: 27 Jun 1999 09:28:08 +0200
Subject: Python for large projects?
References: <3774b5d1.690922@news.force9.net>  <T6bd3.2883$U6.12014@newsr2.twcny.rr.com> <slrn7naup9.jk.thantos@brimstone.mecha> <RGid3.2957$U6.12374@newsr2.twcny.rr.com>
Message-ID: <87n1xm85mf.fsf@lisboa.i-did-not-set--mail-host-address--so-shoot-me>

"Darrell" <news at dorb.com> writes:

> Good or bad, this is one C++ guys aren't comfortable with. The project I'm
> on now will allow people with minimal programming experience to write a
> little python to control the larger application. One of my primary concerns
> is performance and an optimization relies on them not changing a large
> buffer, only look at it. So the people I don't trust aren't on my team. What
> was it Murphy said ? "If it can go wrong"
> I'm tempted to do this object in C++.
> Agreed using __ to hide a name doesn't seem worth while.
> 

If it is only one object, one can try to wrap it in a class and
overwrite setattr, setitem etc. 
This would inhibit that one changes the buffer data accidently. And
put the data into an immutable sequence type (string, tuple).

__Janko

-- 
  Institut fuer Meereskunde             phone: 49-431-597 3989
  Dept. Theoretical Oceanography        fax  : 49-431-565876
  Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
  24105 Kiel, Germany



From avv at quasar.ipa.nw.ru  Fri Jun 25 20:48:54 1999
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Fri, 25 Jun 1999 17:48:54 -0700
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <59EAFD6B662F0706.86BFF3E906CBF8DE.784D2A219F85BC21@lp.airnews.net> <99062523473500.00769@Vegus> <7l16e5$q4j$1@news.tamu.edu>
Message-ID: <37742375.13C68229@quasar.ipa.nw.ru>

Corran Webster wrote:

> This is missing the point - although PyKDE/Qt might be a very nice GUI, it's
> not cross-platform in the same way as Tcl/Tk.  The same goes for Fltk,
> gtk or what have you.  It sounds like wxWindows might be getting near
> the same level of cross-platform compatibility - and it certainly seems
> to be one of the groups' objectives - it's not there yet.
>
> The point of having a standard GUI for python is that it's _standard_.
> ie. that it's likely to work no matter what your platform.  At present,
> Tk seems to be the only mature package with that sort of ubiquity.

Well, let's write a sort of wxTk, and have that everywhere, where no wxNative
runs yet.

Alexander





From kaeber at gmd.de  Thu Jun 24 02:51:13 1999
From: kaeber at gmd.de (Joachim Kaeber)
Date: Thu, 24 Jun 1999 08:51:13 +0200
Subject: function with variable number of arguments
References: <7ks9cu$qht$1@nnrp1.deja.com>
Message-ID: <3771D561.51359D2E@gmd.de>

Hi,

l.a.apiolaza at massey.ac.nz wrote:

> I want to write a function that takes a bunch of  'n' matrices (defined
> using array from NumPy) and processes them further. Question: how do I
> write a function with a variable number 'n' of arguments and then get
> the number of arguments? I think I can put all the matrices together in
> a list and then pass the list as argument, but is there another more
> efficient/elegant way of doing it?

What about:

>>> def f(*args):
...     for l in args:
...             print l
...
>>> f([1,2,3], [4,5,6])
[1, 2, 3]
[4, 5, 6]

-- 
Joachim Kaeber (kaeber at gmd.de) (http://mats.gmd.de/~kaeber)
GMD - Forschungszentrum Informationstechnik GmbH
    - German National Research Center for Information Technology
Schloss Birlinghoven                    email: kaeber at gmd.de
D-53757 Sankt Augustin 1, Germany       phone: (+49 2241) 14-2546



From nathan at islanddata.com  Tue Jun 15 12:27:52 1999
From: nathan at islanddata.com (Nathan Clegg)
Date: Tue, 15 Jun 1999 09:27:52 -0700 (PDT)
Subject: idle tasks
Message-ID: <XFMail.990615092752.nathan@islanddata.com>

I'm writing a multi-threaded application with the threading module.  I
would like one of the threads to run as the lowest priority, effectively
an idle function.  I typically only see idle functions in GUI toolkits,
however, and this needs to be a console app.  Is there a way to lower the
priority of a thread or create an idle function without using tkinter or
pygtk??


----------------------------------
Nathan Clegg
 nathan at islanddata.com





From tre17 at cosc.canterbury.ac.nz  Sun Jun 13 21:25:54 1999
From: tre17 at cosc.canterbury.ac.nz (Timothy R Evans)
Date: 14 Jun 1999 13:25:54 +1200
Subject: Created a Python module for ID3 MP3 tag manipulation
References: <yttaeu7z4o8.fsf@gilgamesh.cse.ucsc.edu>
Message-ID: <c753dzv8tel.fsf@pc142.cosc.canterbury.ac.nz>

Ben Gertzfield <che at debian.org> writes:

> Hey, Pythoners! Just a quick announcement that I've created a Python
> module for easy manipulation of ID3 tags on MP3 files.  These tags
> generally involve the title, artist, and other information about the
> file. I've mailed webmaster at python.org and submitted it to
> Freshmeat, so I hope it can be of use to you all.
> 
> The home-page for the ID3.py module is at:
> 
> http://csl.cse.ucsc.edu/~ben/python/
> 

A quick pendantic note here:

Just out of interest this code apperas to do the same thing as the C
program `id3' by Robert Woodcock, so his warning also applies here.

=== from the `id3' manpage
BUGS
       The ID3 format has some shortcomings.  There  is  a  1  in
       16777216  chance of your mp3 having a certain bit-pattern,
       causing id3 to write over the last 128 bytes of  your  mp3
       upon  tag  creation. Also because the ID3 format is space-
       padded, not null-padded, trailing spaces will  be  ignored
       by most players. Any bugs found in id3 should be forwarded
       to the author, Robert Woodcock <rcw at debian.org>.
===

This is more of a problem with the id3 `standard' than with anything
else.  There is a small chnace that your mp3 file will contain the
data "TAG" at just the wrong place.

--
Tim Evans



From just at letterror.com  Wed Jun  2 16:17:16 1999
From: just at letterror.com (Just van Rossum)
Date: Wed, 2 Jun 1999 20:17:16 GMT
Subject: [Off topic] Weird...
Message-ID: <l03102809b37b3f90b2a0@[193.78.237.145]>

The mpeg movie is a bit old, but otherwise this is quite a weird page:
	http://www.java.net/
(yes there is Python content)

Just





From sams at quinta.com  Fri Jun 18 16:26:11 1999
From: sams at quinta.com (Sam Schulenburg)
Date: Fri, 18 Jun 1999 20:26:11 GMT
Subject: Green or Red for stderr under IDLE?
Message-ID: <7kea0m$rig$1@nnrp1.deja.com>

I notice that stderr for the IDLE environmnet is defaulted to green for
it's color. Is this a standard anywhere or should it be defaulted to
red?

I find that on all the releases that I have used, I have to reset its
value to red as I use this feature to print out error messages on the
status of the scripts that I run.

Sam Schulenburg


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From holger at phoenix-edv.netzservice.de  Tue Jun 22 06:30:42 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Tue, 22 Jun 1999 10:30:42 GMT
Subject: tarlib.py
Message-ID: <376F65D2.DC2E2B36@phoenix-edv.netzservice.de>

High,

I'd like to untar differnt files with the module from
Clemens Hintze (hi, Clemens, are you watching, uh??).

But there occurs an error:

>>> print os.getcwd()
c:\tmp
>>> os.listdir(os.getcwd())
['filecach', 'copycach', 'unbenannt1', 'Tb1_rpt.sdw', 'netzwerk2.sda', 'netzwerk1.sda',
'nls45.tar']
>>> fp=open('nls45.tar')
>>> print fp
<open file 'nls45.tar', mode 'r' at 7f7520>
>>> tar=tarlib.TAR(fp,cached=0)
>>> for entry in tar:
...     print entry.name
...
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "C:\Programme\Python\tarlib.py", line 291, in __getitem__
    return self.__getindex(i)
  File "C:\Programme\Python\tarlib.py", line 639, in __getindex
    entry = self._next()
  File "C:\Programme\Python\tarlib.py", line 487, in _next
    recordcount = (size / RECORDSIZE) + ((size % RECORDSIZE) != 0)
TypeError: bad operand type(s) for /
>>>

[workin' under WinNT, with Python1.52]

What's wrong?

Sorry for asking newbie-questions again:
How could I start extern Gnu-tar inside of python,
or any other program outside of python?
I think 'execfile' is not the thing...

thanx,
Holger



From cjw at connection.com  Thu Jun 10 07:41:28 1999
From: cjw at connection.com (Colin J. Williams)
Date: Thu, 10 Jun 1999 07:41:28 -0400
Subject: Pythonwin - advise sought
Message-ID: <375FA468.98976DDF@connection.com>

I am looking at Pythonwin as an alternative to Tk and would appreciate advice on
the following three questions.

The Pythonwin documentation points to MFC documents as a source of information.
The MSDN site has:

http://msdn.microsoft.com/library/devprods/vs6/vc++/vcmfc/_mfc_about_the_microsoft_foundation_classes.htm'''

Q1:  Are there better sources which can be recommended?

The code fragment below:

       code1= "win32ui.CreateFileDialog(1)"
       fd= eval(code)
       print(repr(fd))

     produces the result:

     object 'PyCFileDialog' - assoc is 007CC240, vf=False,
     notify=0,ch/u=0/0, mh=0,kh=0

Q2.  What do these various symbols and their values mean?

    type(fd) returns the type of the class.

Q3.  Is there some way that one can test that fd is an instance of some class,
any class?

Thanks for any help.

Colin W.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/19990610/527326e6/attachment.html>

From froydnj at cs.rose-hulman.edu  Wed Jun 16 16:41:14 1999
From: froydnj at cs.rose-hulman.edu (Nathan J Froyd)
Date: 16 Jun 1999 15:41:14 -0500
Subject: Why don't my global variables work?
Message-ID: <ttbd7yvho9h.fsf@talon.i-did-not-set--mail-host-address--so-shoot-me>

Can somebody explain this point to me? I have the following program
(posted below to avoid disturbing the continuity):

It's a program for turning a netscape bookmarks file into something
resembling XML. Anyway, you might be puzzled about the evilcount
parameter in topic_read(). If I remove the line `count = evilcount',
python gives up, complaining about a NameError. I know darn well count
already exists, so what is the problem? Especially when a similar
program works as well. (similar program follows)

#!/usr/local/bin/python

count = 0

def evilfunc(num):
	print " " * count, "iamevil"

for x in xrange(0, 100):
	evilfunc(count)
	count = x

Could somebody look at this, perhaps running the program and explain
to me what fundamental fact I am missing here? Because it looks like
my program should run without errors. Thanks!

Nathan

#!/usr/local/bin/python

import sys
import re
import string

titleregex = "<TITLE>(.*)</TITLE>"

topicheadregex = "<DT><H3 [^>]*>([^<]*)</H3>$"
bookmarkregex = "<DT><A HREF=\"([^\"]*)\" .*>([^<]*)</A>$"
topicendregex = "</DL>.*"

# I HAVE DEFINED `count' HERE, PYTHON!
count = 0

ctopic = re.compile(topicheadregex)
cbookmark = re.compile(bookmarkregex)
cend = re.compile(topicendregex)

bmarkfile = open(sys.argv[1], 'r')
print count

def topic_read(regmatch, evilcount):
  # if you comment this line, the program will not work
  count = evilcount
  print "\t" * count, "<topic name=\"%s\">" % regmatch.group(1)
  count = count + 1

  while 1:
    tline = bmarkfile.readline()
    tline = string.strip(tline)

    tmatch = ctopic.match(tline) # first try seeing if we have new topics

    if tmatch is not None:
      topic_read(tmatch, count)

    tmatch = cend.match(tline) # now try checking for </topic>

    if tmatch is not None:
      count = count - 1
      print "\t" * count, "</topic>"
      return # we finished our topic call

    tmatch = cbookmark.match(tline)

    if tmatch is not None:
      print "\t" * count, "<bookmark link=\"%s\">%s</bookmark>" % (tmatch.group(1), tmatch.group(2))

while 1:
  count = 0
  line = bmarkfile.readline()

  if line is None:
    break

  line = string.strip(line)
  match = cbookmark.match(line)

  if match is not None:
    # this is always as the top level, so we don't worry about topiccount
    print "<bookmark link=\"%s\">%s</bookmark>" % (match.group(1), match.group(2))
    continue

  match = ctopic.match(line)

  if match is not None:
    topic_read(match, count)

sys.exit(0)




From arnold at dstc.edu.au  Wed Jun 30 01:46:00 1999
From: arnold at dstc.edu.au (David Arnold)
Date: Wed, 30 Jun 1999 05:46:00 GMT
Subject: A Certifiable Very Cool Trick 
In-Reply-To: Your message of "Tue, 29 Jun 1999 20:52:46 -0400."
             <14201.27231.1003.466027@anthem.cnri.reston.va.us> 
References: <14201.27231.1003.466027@anthem.cnri.reston.va.us>
Message-ID: <199906300546.PAA31064@piglet.dstc.edu.au>

-->"BAW" == Barry A Warsaw <bwarsaw at cnri.reston.va.us> writes:

  BAW> My model was gdb's attach feature where you can attach to, stop
  BAW> and trace through a running external C program.  I would love
  BAW> to be able to do the same thing at the Python level.  We
  BAW> started talking about remote pdb's with threads listening on
  BAW> sockets, etc. (still cool ideas but more work).

on a (vaguely) related note, i was talking to Mark Hammond the other
day (hi Mark) about MS's ActiveDebugging(TM) (sp?) stuff.

as i understood it, each interpreter presents a set of COM APIs which
provide an abstract view of stack frames, etc, suitable for use by a
symbolic script debugger.  you can thus debug across scripting
languages using a single tool, watching calls between interpreters,
etc etc.

all very cool, and so i started thinking about doing the same thing
for Unix via gdb (for C) and the Python interpreter (what other
languages are there?  oh -- ELisp, yeah -- we can do that!)

of course, attaching to a running process should still be possible,
and thus the round-about relationship to the thread.

ahhhhh.  now, where's that spare time again?




d



From Steve at HOSYS.com  Mon Jun 14 11:44:48 1999
From: Steve at HOSYS.com (Steve Hunter)
Date: Mon, 14 Jun 1999 11:44:48 -0400
Subject: Alg Problem (new to Python)
Message-ID: <7k38gf$khr$1@holly.prod.itd.earthlink.net>

Hello,
 The out put of a program is below....the problem is the tasks are being
added
to every project vs. to the specific project....it sees like an alg program
but
I can't see it....brand new to Python...have mercy.

project key: z125 name: Project C
task key: t1241
task key: t1231
project key: y124 name: Project B
task key: t1241
task key: t1231
project key: x123 name: Project A
task key: t1241
task key: t1231

the program below is in on .py file


import sys

projectDict = {} #MAPS PROJECT id string to PROJECT  object
taskDict = {}  #MAPS TASK id string to TASK object

class Project:
 _id = ''
 _name = ''
 _descr = ''
 _tasks = {}    #list of Tasks I contain

 def __init__(self, projectID):
  self._id = projectID

 #setters
 def set_id(self, anID):
  self._id = anID

 def set_name(self, aName):
  self._name = aName

 def set_descr(self, aDescr):
  self._descr = aDescr

 def addTask(self, aTask):
  #require aTask not null
  self._tasks[aTask.id()] = aTask

 #getters
 def id(self):
  return self._id

 def name(self):
  return self._name

#end class Project



class Task:
 _id = ''
 _name = ''
 _descr = ''
 _owner = '' #Person responsible for the task
 _project = ''   #the project that contains me

 def __init__(self,ID):
  self._id = ID

 def test(self):
  #dump internals
  print('in Task test')

 #setters
 def set_id(self,anID):
  self._id = anID

 def set_name(self, aName):
  self._name = aName

 def set_descr(self, aDescr):
  self._descr = aDescr

 def set_project(self, aProject):
  #aProject: PROJECT
  self._project = aProject

 #getters
 def id(self):
  return self._id

 def name(self):
  return self._name

 def descr(self):
  return self._descr


#end class Task


#---------------Call back functions for GUI------------------------

def addTaskToProject(aTaskID, aProjectID):
 #require valid id's and exist in the dictionaries
 prj = projectDict[aProjectID]
 task = taskDict[aTaskID]
 prj.addTask(task)

#end addTaskToProject

def createNewProject(anID,aName):
 p = Project(anID)
 p.set_name(aName)
 projectDict[p.id()] = p

def createNewTask(anID,aName):
 t = Task(anID)
 t.set_name(aName)
 taskDict[t.id()]= t

def dumpProjectDictContents():
 for prjKey in projectDict.keys():
  print 'project key: ' + prjKey + ' name: ' + projectDict[prjKey].name()


def dumpTaskDictContents():
 for key in taskDict.keys():
  print 'task key: ' + key + ' name: ' + taskDict[key].name()

def projectReport():
 for prjKey in projectDict.keys():
  print 'project key: ' + prjKey + ' name: ' + projectDict[prjKey].name()
  for taskKey in projectDict[prjKey]._tasks.keys():
   print 'task key: ' + taskKey
  taskKey = ''


#end createNewProject

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

def testRun():
 createNewProject('x123','Project A')
 createNewProject('y124','Project B')
 createNewProject('z125','Project C')
# dumpProjectDictContents()

 createNewTask('t1231','Task A')
 createNewTask('t1232','Task B')
 createNewTask('t1233','Task C')
 createNewTask('t1241','Task D')
 createNewTask('t1242','Task E')
 createNewTask('t1251','Task F')
 createNewTask('t1252','Task G')
# dumpTaskDictContents()

 addTaskToProject('t1231','x123') #some how all the tasks are added to all
the projects
 addTaskToProject('t1241','y124')

 projectReport()


#end testRun

if __name__ == '__main__':
 testRun()









From Cepl at fpm.cz  Fri Jun 18 03:33:26 1999
From: Cepl at fpm.cz (Matej Cepl)
Date: Fri, 18 Jun 1999 08:33:26 +0100
Subject: HTML bettering?
Message-ID: <1318D78C9072D11195C9006094EA98A72C2164@ocesrv>

> From: KAnton at gmx.de (Konrad Anton)
> Date: Thu, 17 Jun 1999 18:53:36 +0159
> 
> Does it have to be a selfmade Python solution? If you just want to
> number your headlines (and if you speak Perl), have a look at WML[1]
> which combines embedded Perl, m4, MetaHTML, an HTML optimizer and five
> other passes your input script is fed into. It's under GPL and fairly
> well documented.
> 
	[MCepl]   Well, it is not the biggest support of Python's
goodwill, is it? But, to make Pythonians more happy, the story is as
follows: I am writing some stuff in Amaya, which has beautiful heading
numberings, but its priting abilities are rather questionable (it fails
three times from four attempts -- on WinNT) so I would rather use
html2ps which has much better output, but it numbers headings in
different way (still unresolved issue, what it the top level -- H2 or
H1?). So I needed some simple tool to make numbering in the text (BTW,
it would be usefull for latter presentation in, say, widely spread
browsers).

	I sent a notice to amaya list and I have got small script in
Perl in about quarter of hour. It works, it is beautiful, but it is
totally incomprehensible for me, how it works (my knowledge of Perl is
fairly limited -- what is push and pop?). Therefore, I ask again, if
there is some simple way, how to show power of Python's libraries system
(htmllib.py, I guess).

					Have a nice day and sorry for
slightly off-topic


Matthew



From stidolph at origin.ea.com  Tue Jun  1 11:31:16 1999
From: stidolph at origin.ea.com (Stidolph, David)
Date: Tue, 1 Jun 1999 15:31:16 GMT
Subject: Python 2.0
Message-ID: <11A17AA2B9EAD111BCEA00A0C9B4179301E10ED9@forest.origin.ea.com>

Py_TRACE_REFS seems to be defined if Py_DEBUG is defined.

I cannot find a "getobjects" or any other similar function in the source
code or docs.  I am using the Win32 version 1.5.2.

Any help would be appreciated,

David Stidolph.


-----Original Message-----
From: Michael Hudson [mailto:mwh21 at cam.ac.uk]
Sent: Tuesday, June 01, 1999 4:55 AM
To: python-list at cwi.nl
Subject: Re: Python 2.0


"Stidolph, David" <stidolph at origin.ea.com> writes:

> >OK, I'll try to propose something constructive. Maybe we need an object
> >protocol, that would enumerate
> >all references held by an object ?  Writing a portable GC would be then
> much
> >easier.
> 
> Sounds good to me for debugging.  A call that could return a list of
> everything that holds a reference to an object - that would be cool!
> 
> list = GetReferences(object)
> print 'List of referencest to',object
> for item in list:
>   print 'item:',item
> 
> Anybody know of a current way to do this?

If you recompile Python with Py_TRACE_REFS defined, then the sys
module sprouts a "getobjects" function that returns a list of all
objects in existence. This could probably be used to implement
something like this. It would be veeeeery slow, I suspect.

I've never resorted to this approach, tending always to make cyclic
references go away be staring at them very, very hard.

I'm not sure this is at all what is being asked for, but it seems kind
of relavent.

Yours,
Michael



From rhww at erols.com  Sun Jun  6 22:57:41 1999
From: rhww at erols.com (R Wentworth)
Date: Sun, 06 Jun 1999 22:57:41 -0400
Subject: Indentation and grouping proposal
Message-ID: <375B3525.F67D424F@erols.com>

I am reasonably fond of Python's grouping-via-indentation 
paradigm.  Even so, it occurs to me that there may be
legitimate reasons to consider adding a feature to allow
non-indentation-based grouping.

I might envision this being done via the introduction
of two new tokens, e.g., "_{" and "}_".  These would
have the following significance:

    _{  if it is preceded by non-whitespace:
            ends the current logical line and
            begins a new one
        acts like an INDENT token, but the entry it
            pushes onto the indentation stack is
            not a positive integer
        disables interpretation of indentation until
            matching }_ is reached

    }_  ends the current logical line
        acts like a DEDENT token, and pops the top entry
            (which must not be a positive integer)
            off the indentation stack
    
    As a slight extension of the feature, it should be legal to
    use _{ }_ to enclose the outermost set of statements in
    an interpretation unit.  This is in addition to using _{ }_
    to enclose statements that constitute a "suite" that is part
    of an "if" or "def" or "class" statement.

The intent behind adding this feature would be as follows:

1)  Simplifies the task of dealing with Python code 
    in non-standard contexts, e.g., when embedded in a
    free-form language like HTML.

2)  Could help make Python more popular by providing
    an alternative for those many (arguably misguided)
    souls whose main pet peeve about Python is the
    grouping-via-indentation paradigm.

3)  May make code generation marginally simpler.

############################################################

Here are some usage examples.

Legal but pointless, with no new functionality:

    if good: _{
        print "Spam!"
        print "Spam!"
    }_
    print "Yum."

Legal but with pointlessly irregular indentation:

    if good:
    _{
        print "Spam!"
            print "Spam!"
    }_
    print "Yum."

Previously impossible, a one-line version:

    if good: _{ print "Spam!"; print "Spam!" }_  print "Yum."
    
Prettier looking Python embedded in Python (indentation within string
would previously have been illegal):

    statements = '''if good: _{
                print "Spam!"
                print "Spam!"
            }_
            print "Yum." '''
    exec(statements)
            
The last two examples are relevant to embedding in HTML as well
(using a fictitious inclusion syntax):

    Tasty? <!--#pycode if good: _{ print "Spam!"; print "Spam!" }_  print "Yum." -->

    Tasty? <!--#pycode if good: _{ 
                           print "Spam!"
                           print "Spam!"
                       }_
                       print "Yum." -->

Or, using an even more speculative syntax, something which would be
horrendously awkward without this new feature:

    Tasty? 
    <!--#pycode if good: _{ --> 
    Spam! Spam!
    <!--#pycode }_ -->
    Yum.

Up until now, I have not used the subfeature that would
allow _{ }_ to surround the outermost level of code. Taking this 
into account, formatting improvements cited above become possible 
even if there is no "if" or the like involved, e.g.:

    statements = ''' _{ print "Spam!"
                        print "Spam!" }_ '''
    exec(statements)

Here is a code generation example.  Consider the following
rather goofy code:

    n = 0
    print "Computing ceiling(log_2(x))"
    if x <= 1:
        print "Got it!"
        print `n`
    else:
        n = n + 1
        x = x/2
        if x <= 1:
            print "Got it!"
            print `n`
        else:
            n = n + 1
            x = x/2
            if x <= 1:
                print "Got it!"
                print `n`
            else:
                n = n + 1
                x = x/2
                if x <= 1:
                    print "Got it!"
                    print `n`
                else:
                    n = n + 1
                    x = x/2
                    print "I give up."

This code contains four repeated blocks which are identical
except for indentation level.  If we were to try to generate
this code verbatim, we would need to carefully adjust the
indentation level each time we inserted the block.  But with
the _{ }_ feature we could do the following:

    start = '''_{ n = 0
                  print "Computing ceiling(log_2(x))"
               '''
    atom = '''if x <= 1: _{
                  print "Got it!"
                  print `n` }_
              else: _{
                  n = n + 1
                  x = x/2
                  '''
    end = '''print "I give up." }_ }_ }_ }_ }_'''
    code = start + atom + atom + atom + end
    exec(code)

In this version, the indentation in the generated code is
sloppy and meaningless -- but it does the right thing
without as much effort as would have been required previously.

############################################################

A few comments on the choice of symbols used to implement
this new functionality...

Selection criteria would sensibly include:

1)  Syntactically unambiguous; need to choose something that
    currently constitutes illegal syntax.

2)  Should suggest grouping to folks who are used to C, Perl, etc.

3)  Should be reasonably aesthetic.

4)  Not excessively difficult to type.

Here are a few options, and associated comments:

     {  }    Not available -- in use for dictionaries.

    \{ \}    Gets ugly if code needs to be enclosed in a string
             (in Python or another language)

    (@ @)    Could work, but arguably graceless and unfamiliar

    -{ }-    Likely to be confusing even if technically unambiguous.

    {_ _}    Syntactically confusing, as "_" is a legal identifier name,
	     and an identifier could be the first or last thing in a
             dictionary expression.

    .{ }.    Visually as close as possible to { } and perhaps easier
             to type than some alternatives? Perhaps viable?
             (Could be confusing to name these tokens in error messages
             or textual discussion?)

    _{ }_    Probably viable?

############################################################

I haven't looked at the Python parser, but naively, this sort
of "enhancement" would seem to be relatively easy.  If there were
official blessing for the concept, I'd be willing to look into
doing the necessary work.

Robert Wentworth
rhww at erols.com



From larsga at ifi.uio.no  Sun Jun 20 17:25:33 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 20 Jun 1999 23:25:33 +0200
Subject: nerdnews.py
References: <376D52EA.41C6@student.liu.se>
Message-ID: <wkr9n6tvhu.fsf@ifi.uio.no>

* Fredrik Henbj?rk
| 
| Creates a htmlpage with todays Dilbert strip and 
| headlines from Linux Today and Slashdot.

Maybe you'll find this interesting:
<URL: http://www.python.org/pipermail/xml-sig/1999-May/001220.html>

--Lars M.



From arcege at shore.net  Thu Jun 17 10:15:56 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Thu, 17 Jun 1999 14:15:56 GMT
Subject: ref counting, access from python
References: <7kad31$o7f$1@ns.mtu.ru> <3768C80B.14B245E6@appliedbiometrics.com>
Message-ID: <wq7a3.223$7X1.49549@news.shore.net>

Christian Tismer <tismer at appliedbiometrics.com> wrote:

: Oleg Orlov wrote:
:> 
:> Hello.
:> How can i get objects current ref count from my python code?
:> Thanks.

:>>> import sys
:>>> sys.getrefcount(None)
: 1583
:>>> print "ciao - chris"
: ciao - chris
:>>> 

Wow, 1583 seems a little high to me. ;)

Python 1.5.2 (#14, Jun 11 1999, 16:31:09) [C] on aix4
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import sys
>>> sys.getrefcount(None)
140
>>>

Anyone wonder what chris has been doing? ;)

  -Arcege




From cgw at fnal.gov  Wed Jun 30 16:42:25 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Wed, 30 Jun 1999 20:42:25 GMT
Subject: A Python problem?
In-Reply-To: <377A765F.90367AC5@connection.com>
References: <377A765F.90367AC5@connection.com>
Message-ID: <14202.33073.919694.671354@buffalo.fnal.gov>

Colin J. Williams writes:
 > def main():
 >   import sys
 >   print 'Start:',dir()
 >   s0= sys.modules
 >   import plugins.rptlib
 >   s1= sys.modules
 >   if s0 == s1:
 >     print 'sys.modules is unchanged'

This will *always* print "sys.modules is unchanged";  sys.modules is a 
dictionary, which is a mutable type.

If you want to see if sys.modules changed, try this:

s0 = sys.modules.copy()
import stuff
if sys.modules == s0:
   print "it's unchanged"



From amackay at starvision.com  Tue Jun 29 16:00:47 1999
From: amackay at starvision.com (Angus MacKay)
Date: Tue, 29 Jun 1999 13:00:47 -0700
Subject: printing with print
Message-ID: <377925EF.1CBB640F@starvision.com>

is there any other way to print that with print?
perhaps with a real function.

I want to print two things without a space being added between them:
foo = "there"
print "hi" + foo

works fine but: 
foo = 2
print "hi" + foo

does not. I like the printing of any object ability of "print" so I
do not want to use "" % style printing, I just want to be able to
govern the whitespace that gets output.

cheers, Angus.



From faassen at pop.vet.uu.nl  Mon Jun  7 04:34:55 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Mon, 07 Jun 1999 10:34:55 +0200
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net> <aahzFCpHw8.Gxt@netcom.com> <7j4e2u$8rh$1@m2.c2.telstra-mm.net.au> <37564132.2D01B7C3@pop.vet.uu.nl> <3756BAFC.2066A636@home.com> <3756F071.E717D0FA@pop.vet.uu.nl> <37573C16.D0233810@home.com>
Message-ID: <375B842F.BD53DA63@pop.vet.uu.nl>

Jim Meier wrote:
> David Schere's python wrapper, available from the same place as mine. 

Where o where was this again. :)

> > I'm rambling about Python these days. Yesterday I read a paper from
> > Guido about how he was thinking about bringing Python (or some successor
> > language) to *billions* of users. :)
> 
> Just think of all the nasty code _that_ would generate... I can't 
> beleive some of the English that some people speak, and am horrified to think 
> that when I'm 30, I might end up teaching "Python Literacy" to a group of bored 
> and whiny (and by that point, negative-attention-spanned) high school students 
> and trying to explain the design philosophy of "The-Right-Way"... !!

You have to do your Duty for the Cause.. :) Yeah, Guido also talks about
the nasty code that would be produced. Apparently they have some ideas
for a development environment to make this less of a problem, though not
much appears to be known about that development environment yet.

Regards,

Martijn



From faassen at pop.vet.uu.nl  Thu Jun 10 05:36:42 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Thu, 10 Jun 1999 11:36:42 +0200
Subject: Grimoire [was: How do I defect?]
References: <Pine.WNT.4.05.9906092053040.120-100000@david.ski.org>
Message-ID: <375F872A.C8C53B1A@pop.vet.uu.nl>

David Ascher wrote:
> Andrew Kuchling is too modest to advertise it, but his
> Python Grimoire is a good document to get a feel for common tasks in
> Python. It's at:
> 
>     http://starship.python.net/crew/amk/grimoire/
> 
> It's a good thing to read just after the tutorial.

I second that! This looks like a really useful document indeed. Thanks
Andrew!

Me-too-ly,

Martijn



From sragsdale at my-deja.com  Mon Jun  7 22:25:17 1999
From: sragsdale at my-deja.com (sragsdale at my-deja.com)
Date: Tue, 08 Jun 1999 02:25:17 GMT
Subject: where does 'del' come from?
Message-ID: <7jhuu5$8d4$1@nnrp1.deja.com>

A fairly academic question, but I'm curious nonetheless.  Where is 'del'
defined?  Functions like 'join' are obviously in the string package,
'range' is a builtin, but I can't seem to find 'del' anywhere.

Python 1.5.2 (#12, Jun  2 1999, 19:23:30) [C] on irix646
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> dir(__builtins__)
['ArithmeticError', 'AssertionError', 'AttributeError', 'EOFError',
'Ellipsis', 'EnvironmentError', 'Exception', 'FloatingPointError',
'IOError', 'ImportError', 'IndexError', 'KeyError', 'KeyboardInterrupt',
'LookupError', 'MemoryError', 'NameError', 'None',
'NotImplementedError', 'OSError', 'OverflowError', 'RuntimeError',
'StandardError', 'SyntaxError', 'SystemError', 'SystemExit',
'TypeError', 'ValueError', 'ZeroDivisionError', '__debug__', '__doc__',
'__import__', '__name__', 'abs', 'apply', 'buffer', 'callable', 'chr',
'cmp', 'coerce', 'compile', 'complex', 'delattr', 'dir', 'divmod',
'eval', 'execfile', 'exit', 'filter', 'float', 'getattr', 'globals',
'hasattr', 'hash', 'hex', 'id', 'input', 'int', 'intern', 'isinstance',
'issubclass', 'len', 'list', 'locals', 'long', 'map', 'max', 'min',
'oct', 'open', 'ord', 'pow', 'quit', 'range', 'raw_input', 'reduce',
'reload', 'repr', 'round', 'setattr', 'slice', 'str', 'tuple', 'type',
'vars', 'xrange']
>>> range
<built-in function range>
>>> del
  File "<stdin>", line 1
    del
       ^
SyntaxError: invalid syntax


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From Paul.Prescod at p98.f112.n480.z2.fidonet.org  Tue Jun 29 12:20:25 1999
From: Paul.Prescod at p98.f112.n480.z2.fidonet.org (Paul Prescod)
Date: Tue, 29 Jun 1999 17:20:25 +0100
Subject: PySequence_Check
Message-ID: <00001582@bossar.com.pl>

From: Paul Prescod <paul at prescod.net>

Reuben Sumner wrote:
> 
> How can I do the equivalent of a PySequence_Check in Python?
> GvR told me that the question was illdefined (if I explained it correctly)
> and to ask here.  
>...
> What am I missing?  I have a nested structure of lists and tuple and I
> just want to know when I have bottomed out.  I can check for tuple or
> list indivually but that is ugly and inflexible.  I could just try
> indexing and see if there was an exception but that is not much better.

Hi, Reuben.

This latter is probably our best bet. Python in general has no way of
declaring that an object acts as a sequence or mapping. Therefore there
can be no way of querying whether an object acts as a sequence or
mapping.  In general, Python has no way of declaring that an object "acts
as" anything at all (other examples would include a set, a widget, a web
browser). There is no concept of interface or protocol except in the C
API. I think that the goal is to define the concept in general before
applying it specifically to sequences, mappings, file objects and other
"built-in" protocols.

My discussion with Guido on this issue began here:

<Pine.SUN.3.91.980824165825.9662F-100000 at cito.uwaterloo.ca>

or you can do a deja search with "IsMappingType"

There are some tricky issues about what this interface stuff all means in
a dynamically typed language.

The constant re-occurence of discussions like this have lead to the
development of a special interest group:

http://www.python.org/sigs/types-sig/

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for only himself
 http://itrc.uwaterloo.ca/~papresco

"The new revolutionaries believe the time has come for an aggressive 
move against our oppressors. We have established a solid beachhead 
on Friday. We now intend to fight vigorously for 'casual Thursdays.'
  -- who says America's revolutionary spirit is dead?



From mwh21 at cam.ac.uk  Wed Jun  2 17:02:32 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 02 Jun 1999 22:02:32 +0100
Subject: pickle vs .pyc
References: <7j43pn$55l$1@mlv.mit.edu> <m3pv3eian1.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <m3ogiyiadz.fsf@atrus.jesus.cam.ac.uk>

Michael Hudson <mwh21 at cam.ac.uk> writes:

> Michael Vezie <mlv at pobox.com> writes:
> 
> > I need to be able to read a couple very complex (dictionary of arrays 
> > of dictionaries, and array of dictionaries of array of dictionaries) 
> > data structures into python.  To generate it by hand takes too long, 
> > so I want to generate it once, and read it each time (the data doesn't 
> > change).
> > 
> > The obvious choice is, of course pickle, or some flavor thereof.
> > But can someone tell me why this wouldn't be faster:
> > 
> > In the code that does the "pickling", simply do:
> > f = open("cache.py", "w")
> > f.write("# cache file for fast,slow\n")
> > f.write("fast = "+`fast`+'\n')
> > f.write("slow = "+`slow'+'\n')
> > f.close()
> > import cache
> > 
> > Then, later, when I want the data, I just do:
> > 
> > from cache import fast,slow
> > 
> > and it's right there.  It's compiled, and seems really fast (loading a 
> > 50k file in .12 seconds).  I just tried the same data using cPickle, and 
> > it took 1.4 seconds.  It's also not as portable.  There is a space savings 
> > with pickle, but it's only 5% (well, 56% if you count both the .py and 
> > .pyc files), but that doesn't really matter to me.
> > 
> > Am I missing something here?  This sounds like an obvious, and fast, 
> > way to do things.  True, the caching part may take longer.  But I 
> > really don't care about that, since it's done only once, and in the 
> > background.  
> > 
> > Michael
> 
> Hmm, you're relying on all the data you're storing having faithful
> __repr__ methods. This certainly isn't universally true. I'd regard
> this method as too fragile.
> 
> If you're only storing simple data (by which I mean simple types of
> data, not that the data is simple) (and I think you must be for the
> approach you're using to work) give the marshal module a whirl.
> 
> I think it will be substantially faster than your repr-based method
> (cryptic hint: if it wasn't, the marshal module probably wouldn't
> exist).
> 
> Eg:
> 
> import marshal
> 
> complex_data_structure = {'key1':['nested list'],9:"mixed types"}
> 
> marshal.dump(complex_data_structure,open('/tmp/foo','w'))
> 
> print marshal.load(open('/tmp/foo'))
> 
> HTH
> Michael

Duh! Of course, once you've imported cache.py once, it's compiled to a
.pyc file and all the literals within it will be marshalled
anyway. Still using marshal directly is certainly more robust and
probably faster...

one-day-I'll-learn-to-read-thesubject-as-part-of-the-message-ly y'rs
Michael



From neelk at brick.cswv.com  Mon Jun 14 22:56:36 1999
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 14 Jun 1999 21:56:36 -0500
Subject: ANN: Stackless Python 0.2
References: <37628EAA.C682F16C@appliedbiometrics.com>
Message-ID: <7k4fd4$c04$1@brick.cswv.com>

In article <37628EAA.C682F16C at appliedbiometrics.com>,
Christian Tismer  <tismer at appliedbiometrics.com> wrote:
>ANNOUNCING:
>
>               Stackless Python 0.2
>
>The possibilities are for instance:
>
>Coroutines, Continuations, Generators
>
>Coroutines will be able to run at the speed of
>a single C function call, which makes them a considerable
>alternative in certain algorithms.

This is very neat, and you are completely deranged. I know just 
enough to know that I should cheer you on, mind, but I'll try to 
cheer loudly. Feel free to make an announce when you add coroutine
support. Please? :)


Neel



From bwarsaw at cnri.reston.va.us  Thu Jun  3 14:45:56 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Thu, 3 Jun 1999 14:45:56 -0400 (EDT)
Subject: mailing list to news gateway
References: <7j4asp$v4a$1@nnrp1.deja.com>
Message-ID: <14166.52580.600388.729849@anthem.cnri.reston.va.us>

>>>>> "jeremy" ==   <jeremy at cnri.reston.va.us> writes:

    jeremy> Since about last Thursday, messages I have sent to
    jeremy> python-list at python.org haven't been showing up at Deja
    jeremy> News. I would assume they also aren't showing up in the
    jeremy> newsgroup, but I'm not sure.

For some miraculous reason, it appears as if CNRI's internal upstream
feed has -- just today -- become unwedged.  It looks like we may have
lost[1] gatewayed messages posted to the mailing list between 5/27 and
5/31, but that anything posted 6/1 until today seems to have gotten
out.

Our sysadmins, and Jeremy and I looked at our dnews installation and
none of us know why it broke, or why it suddenly got unjammed (tho' I
suspect DNS problems might have caused the original breakage).

Those of you who post to the list via python-list at python.org, please
keep an eye out on Deja.com to make sure messages are making it off
our news server.  One good rule of thumb that Jeremy had: if you don't
see a posting on Usenet from Tim for a couple of days, there might be
a problem :)

If WCTW, I'll point the gateway directly at our upstream UUnet news
server, but I'd rather not do that since it'll probably cause
duplicates in one direction or the other.

-Barry

[1] lost means the messages didn't get propagated to Usenet.  They did 
make it to the Pipermail archive though:

    http://www.python.org/pipermail/python-list



From Andreas.Jung at p98.f112.n480.z2.fidonet.org  Wed Jun 30 11:32:49 1999
From: Andreas.Jung at p98.f112.n480.z2.fidonet.org (Andreas Jung)
Date: Wed, 30 Jun 1999 16:32:49 +0100
Subject: looking for cookie.py
Message-ID: <000015bc@bossar.com.pl>

From: Andreas Jung <ajung at sz-sb.de>

On Wed, Jun 30, 1999 at 01:15:19PM +0200, Mathieu Lecarme wrote:
> the link on python web site doesn't work!
> where can i find it?
> 
ftp://ftp.bbn.com/pub/timo/python/Cookie.py


-- 
                                _\\|//_
                               (' O-O ')
------------------------------ooO-(_)-Ooo--------------------------------------
   Andreas Jung, Saarbr?cker Zeitung Verlag und Druckerei GmbH
   Saarbr?cker Daten-Innovations-Center
   Gutenbergstr. 11-23, D-66103 Saarbr?cken, Germany
   Phone: +49-(0)681-502-1563, Fax: +49-(0)681-502-1509
   Email: ajung at sz-sb.de (PGP key available)
-------------------------------------------------------------------------------



From Barry.A..Warsaw at p98.f112.n480.z2.fidonet.org  Wed Jun 30 03:12:56 1999
From: Barry.A..Warsaw at p98.f112.n480.z2.fidonet.org (Barry A. Warsaw)
Date: Wed, 30 Jun 1999 08:12:56 +0100
Subject: A Certifiable Very Cool Trick
Message-ID: <000015b8@bossar.com.pl>

From: "Barry A. Warsaw" <bwarsaw at cnri.reston.va.us>


>>>>> "DA" == David Arnold <arnold at dstc.edu.au> writes:

    DA> as i understood it, each interpreter presents a set of COM
    DA> APIs which provide an abstract view of stack frames, etc,
    DA> suitable for use by a symbolic script debugger.  you can thus
    DA> debug across scripting languages using a single tool, watching
    DA> calls between interpreters, etc etc.

Another thing that would be really cool, would be a debugging tool
that could cross the implementation language barrier.  E.g. stepping
through Python, I could then drop down into the C or Java at that
point.  Having the ability to move between even those two abstraction
levels would be pretty useful.  Really cool would be the ability to
also drop into the bytecode level (either under the Java source in the
case of JPython, or between the Python and C code in the case of
CPython).

    DA> all very cool, and so i started thinking about doing the same
    DA> thing for Unix via gdb (for C) and the Python interpreter
    DA> (what other languages are there?  oh -- ELisp, yeah -- we can
    DA> do that!)

Ah, so our dream of scripting XEmacs with Python isn't so far away...

    DA> ahhhhh.  now, where's that spare time again?

...or is it? :)  Fun to think about at least.

-Barry




From skip at mojam.com  Wed Jun 30 22:44:37 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 01 Jul 1999 02:44:37 GMT
Subject: coverage tool for Python?
References: <377ABE31.249B1E51@cup.hp.com>
Message-ID: <377AD6AE.56797BD2@mojam.com>

Ovidiu Predescu wrote:

> Are there any coverage tools for Python?

Not terrific, but there's a knock-off of the profiler I wrote a few
years ago available at:

    http://www.musi-cal.com/~skip/python/

Search for "coverage".

Cheers,

-- 
Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
518-372-5583



From mark.butterworth at cis.co.uk  Wed Jun 23 08:35:14 1999
From: mark.butterworth at cis.co.uk (Mark Butterworth - U-Net)
Date: Wed, 23 Jun 1999 13:35:14 +0100
Subject: Freeze?
References: <7kqcj4$3bd$1@nnrp1.deja.com>
Message-ID: <Pk5c3.193$T3.425@newsr2.u-net.net>

Curtis Yanko wrote in message <7kqcj4$3bd$1 at nnrp1.deja.com>...
>I have loaded 1.5.2 and the Win32All-125 but dont see the freeze tool
>anywhere. Searching the FTP site I only see Freeze for older versions.


I think the author's site is at www.nightmare.com





From gmcm at hypernet.com  Tue Jun 15 10:17:10 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 15 Jun 1999 09:17:10 -0500
Subject: Bug in `imp' module?
In-Reply-To: <3765C3F8.D3F8A4CD@bioreason.com>
Message-ID: <1282666010-103335885@hypernet.com>

Andrew Dalke writes:

> Summary: imp.find_file returns the .py file before the .pyc file
> when I think it should be the other way around.  Solution should be
> to swap two lines in Python/importdl.c 

Not so simple. The .pyc file is trash if the .py file is newer.

I suggest you follow the earlier lead to Greg's imputil.py. You will 
still have to write code, but you'll be fitting your code into a 
"sane" architecture.

Greg's is at: http://www.lyra.org/greg/small/

If he hasn't yet patched it to properly pass on reload requests, you 
can get a patched version from: 

www.mcmillan-inc.com/dnld/archive.tar.gz


- Gordon



From matomira at iname.com  Fri Jun 11 06:13:04 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Fri, 11 Jun 1999 12:13:04 +0200
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <wk909syqr4.fsf@ifi.uio.no> <375FF969.C879B4CD@iname.com> <37601668.D9105225@mayo.edu>
Message-ID: <3760E130.7456447@iname.com>


Bob Techentin wrote:

> These flame wars are more than five years old.  Tcl has had namespaces
> and a byte-code compiler for years.

I know that. But I see some serious things there (the `call by name' issue,
for example) that give me the impression must be still around unless
the semantics have gone through a major change.

I would cut this thread right here before someone comes up
with some myth. There's enough work for debunkers there.

Thanks,





From tgm at math.uni-bonn.de  Sun Jun  6 09:46:27 1999
From: tgm at math.uni-bonn.de (Thomas Ackermann)
Date: Sun, 6 Jun 1999 15:46:27 +0200
Subject: Designing Large Systems with Python
References: <m37lqz0yoa.fsf@solo.david-steuber.com> <372599D6.C156C996@pop.vet.uu.nl> <7g4a3l$atk$1@news.worldonline.nl> <m3vhehxjhr.fsf@solo.david-steuber.com> <7g99pj$b1$1@news.worldonline.nl> <slrn7igcdg.is.davecook@sputnik.escnd1.sdca.home.com> <7g9eln$4eq$1@news.worldonline.nl>
Message-ID: <j3udj7.5a.ln@final.math.uni-bonn.de>

Ilja Heitlager <news at helen.demon.nl> wrote:
> I would love too, but Windows doesn't leave any space on my HD ;-(

Reorganize your disk (scandisk or ?) then use "fips" to make this
partition small and create a new, free partion from the saved space, then
install LINUX ...	;-)

	Byebye,
-- 
  Thomas Ackermann | Tel. +49-(0)228/631369|73-7773 | <tgm at math.uni-bonn.de>
             finger tgm at rhein.math.uni-bonn.de for public key
               GNU LINUX Python gtk pygtk MySQL FUDGE GURPS



From wlfraed at ix.netcom.com  Wed Jun  2 00:14:28 1999
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Wed, 02 Jun 1999 04:14:28 GMT
Subject: Python 2.0
References: <11A17AA2B9EAD111BCEA00A0C9B4179301E10ED1@forest.origin.ea.com>  <374F1F6D.F3338F56@prescod.net> <374F5001.1C850DC@Lugoj.Com> <374F4591.3E66FB0@prescod.net> <zTS33.1171$nn.335894@news.shore.net> <374FF31A.6E990179@prescod.net> <37512edf.3024969@news.demon.co.uk> <3752ddac.3812702@nntp.ix.netcom.com> <3753EC5B.4C70860@prescod.net>
Message-ID: <3754ade0.2590819@nntp.ix.netcom.com>

On Tue, 1 Jun 1999 14:21:15 GMT, Paul Prescod <paul at prescod.net>
declaimed the following in comp.lang.python:


> That's all cool, but on Windows you get to choose your scripting language
> which, for Python users, is even cooler. Still, I take your point that
> AREXX was years ahead of Windows. And AREXX was ripped off of mainframe
> REXX which was, of course, even older.

	I've (unfortunately in one way) had to migrate to a W95 machine,
but the last Amiga Python port I saw did incorporate an AREXX interface
mechanism.

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <



From Aahz.Maruch at p98.f112.n480.z2.fidonet.org  Wed Jun 30 12:39:32 1999
From: Aahz.Maruch at p98.f112.n480.z2.fidonet.org (Aahz Maruch)
Date: Wed, 30 Jun 1999 17:39:32 +0100
Subject: Python Editor
Message-ID: <000015bf@bossar.com.pl>

From: aahz at netcom.com (Aahz Maruch)

In article <AJqe3.1318$24.76 at news01.t-net.net.ve>,
Jose A. Antich <jantich at cantv.net> wrote:
>
>   Does anyone know if there's any Python Editor for the MAc?

IDLE, a debugger/editor written in Python.  You'll need Python 1.5.2 and
Tcl/Tk 8.0 (*not* 8.1).
--
                      --- Aahz (@netcom.com)

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



From arcege at shore.net  Mon Jun 28 14:09:48 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Mon, 28 Jun 1999 18:09:48 GMT
Subject: Python GUIs: Summary and Conclusion
References: <99FA4A1A1280BEBD.0A1ABA62CC989479.4C2C6022F695BEAC@lp.airnews.net>
Message-ID: <MTOd3.139$6M6.44310@news.shore.net>

Mikael Lyngvig <mikael at pobox.com> wrote:
: Hi,

: I think there is a sufficient number of replies to my original "Python
: GUIs: Abondining TkInter and welcoming wxPython" post that a
: conclusion can be made.  Thanks to all who replied; the replies did
: exactly what I wanted - presented a lot of arguments for and against a
: switch from TkInter to wxPython.

[Summary of arguments snipped]

: Conclusion (my conclusion, at least...):

: TkInter should continue to be the de-facto standard Python GUI.
: Primarily because of its matureness, some say better design
: philosophy, its portability, and its popularity.  

: wxPython continues to be a good alternative, as does others, which may
: be better in specific situations and to some developers.

: The TkInter module(s) should be rewritten to employ the native Tk API,
: instead of Tcl, which would probably eliminate all sluggishness in
: client applications.

: More widgets should be added to Tk(Inter).

: Anyone disagree?  

: Any volunteers to start work on making a native Tk C API version of
: TkInter, 100 percent backwards compatible of course?

I've already rewritten _tkinter to use a new pytcl module, but the
functionality is still the same, ie. executing Tk calls from within an
interpreter.  But I'd like to see it using the TK/C API (so much for
my hard word on rewriting _tkinter *wink*).

The main issue would be to make it extensible for adding the new
widgets and other extensions (BTL, PIL, etc.).  Also, the interface
between Python and C would not be as "simple" as the existing module.
It's doable tho.

Note that using the "native" Tk API would still require linking with
the Tcl library.

I could scratch my _tkinter rewrite and work on a native solution.
(I could still publish my pytcl module.)  Any thoughts?

We already know that Python 1.6 won't be until the new year, so any
changes to Tkinter would be as a patch.

  -Arcege




From behrends at cse.msu.edu  Sat Jun 26 01:57:38 1999
From: behrends at cse.msu.edu (Reimer Behrends)
Date: 26 Jun 1999 05:57:38 GMT
Subject: GC In Python: YAS (Yet Another Summary)
References: <Pine.SUN.3.95-heb-2.07.990616155228.15761A-100000@sunset.ma.huji.ac.il> <p53dzrfcea.fsf@bidra241.bbn.hp.com> <930267199.139037@clint.waikato.ac.nz> <37730BAA.28A7C738@bioreason.com> <930340397.465868@clint.waikato.ac.nz>  <3773EFF7.27D29899@bioreason.com>
Message-ID: <slrn7n8r03.kl.behrends@allegro.cse.msu.edu>

Andrew Dalke (dalke at bioreason.com) wrote:
[...]
> Yes, the library has a pointer table to the object.  The problem
> I see is, how does the Boehm collector (or other GC) know what to
> collect from this vendor library?  I can't see how, given that
> my Python code stores it as an integer.

The Boehm collector is conservative; it will treat any word in
memory as a potential pointer until proven otherwise. It uses a
number of sophisticated methods to avoid misidentifying integers
as pointers. Surprisingly enough, this approach works pretty well
in practice. I've used it more than once to reign in a misbehaving
C program (which leaked memory or freed storage prematurely).

> That's why I don't see how you say "It fully supports C and C++"
> when this is a package which doesn't do "pointer XORing or other
> crimes" but cannot be usable with a GC w/o modifications.

The few practices that are unsafe should be avoided by programmers,
anyway. Like xoring pointers together, or doing similar ugly things.
Most programs are pretty well-behaved.

[...]

				Reimer Behrends



From 8499 at greenhead.ac.uk  Wed Jun 30 05:31:49 1999
From: 8499 at greenhead.ac.uk (Newbie)
Date: 30 Jun 1999 04:31:49 -0500
Subject: I am v. stupid, please help
Message-ID: <3779d5f5@discussions>

I have just downloaded Python 1.5 and I have 
had a brief look at the helpfiles and some 
source code and my stupid question is this :
How do you define a string and then use it 
for input, e.g,

name = input('What is your name : ')

This works for integers but not for strings,

Help Me Please!!

--Posted from EarthWeb Discussions. http://discussions.earthweb.com




From MHammond at skippinet.com.au  Thu Jun 10 02:24:51 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Thu, 10 Jun 1999 16:24:51 +1000
Subject: Windows Printing Question
References: <375EF682.CCAC69D1@west.boeing.com>
Message-ID: <7jnlkj$r1v$1@m2.c2.telstra-mm.net.au>

If you are using IE4, you should investigate the COM interfaces.  The
win32com\test directory has a COM sample that loads a page in explorer.
Making it print this should be quite simple.

Mark.

Jeff P. wrote in message <375EF682.CCAC69D1 at west.boeing.com>...
>I have a set of Internet shortcuts on a Windows 95 machine that I'd like
>to print out. I know Windows can do this, after all, I can right-click
>on a shortcut and select "Print" and it'll do it. The question is: How
>to do this programaticaly?
...






From ray at commsecure.com.au  Wed Jun  9 18:43:56 1999
From: ray at commsecure.com.au (Ray Loyzaga)
Date: Thu, 10 Jun 1999 08:43:56 +1000
Subject: [Thread-SIG] RE: tstate invalid crash with threads
References: <000101beb237$88b71d80$999e2299@tim> <003201beb296$b9be7d60$d5a245ab@cisco.com>
Message-ID: <375EEE2C.B7E7EF68@commsecure.com.au>

Scott Cothrell wrote:
> 
> I can confirm that we were seeing similar behavior under a highly threaded,
> highly stressed situation.  We backed off on the number of threads and
> slowed things down as a workaround also.  We couldn't trace it down to
> Python, as we have a Java, C, Python mix that has too many suspects.
> 

The locking code added to the create/destroy side of the threads
has yielded good results. On my stripped down server, which was crashing
after between 40,000 and 1.4M transactions managed 14.75M last night.
I will kill this and attempt to run my real task for a day or so,
but it seems to confirm a small race exists ....



From bob at horvath.com  Thu Jun 10 21:44:45 1999
From: bob at horvath.com (Bob Horvath)
Date: Thu, 10 Jun 1999 20:44:45 -0500
Subject: Scripting Telnet
References: <a6U73.1265$VL2.195835@news.direcpc.com>
Message-ID: <37606A0D.6332A9C2@horvath.com>

Emile van Sebille wrote:

> Hello all,
>
> I've accepted a project and I think I can use Python for part of it.
>
> The project is moving data collected in a warehouse management system
> into a proprietary, legacy CHUI application.  The information is to be
> processed hourly by a cron job under HPUX.  Most of the legacy app
> source code is available, but there are some black boxes in there I
> can't get at.  The app doesn't like standard input redirection, but will
> run in a telnet session.
>
> My choices appear to be 1) write a utility in the native app's
> environment the mimics the result set of the current app (and I _really_
> don't want to wade through that crap to figure it out); or 2) script the
> input requirements and feed the application while it runs (I've done
> this for smaller projects using VB sendkeys).
>
> I feel that python ought to be good for this, and I'm hoping you'll
> point me to the appropriate modules to look into or techniques to use.
> I did a little browsing through dejanews and saw a reference to perl's
> expect that looks like it does this kind of work,so maybe there's a
> python tool too?
>
> Trying-to-get-paid-and-learn-more-python-at-it-ly y'rs,
>

Well, the tcl based expect is probably the most appropriate tool for such a
thing, but there is a pythonized version(s) of it as well.  The one I can
think of is ExpectPy, (http://www1.shore.net/~arcege/python/ExpectPy/), but
I seem to remember seeing at least one other one.




From m01ymu00 at cwcom.net  Fri Jun  4 11:36:44 1999
From: m01ymu00 at cwcom.net (Daniel Faulkner)
Date: Fri, 04 Jun 1999 16:36:44 +0100
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl>
Message-ID: <3757F28B.BF6411FE@cwcom.net>

How many messages are there about this white space stuff? Looks like millions whats
it all about?

Martijn Faassen wrote:

> Hi there,
>
> I don't intend to start a flamewar or anything, and I'm sure Larry Wall
> is a nice guy, and Perl is nice and all, and the postmodern
> rationalisation of the confusion that Perl is to me is *fun* (but does
> not work for me), but Larry Wall seems to be infected by the whitespace
> eating nanovirus meme. From an interview at:
>
> http://www.linuxjournal.com/issue61/3394.html
>
> Interviewer:  In what way is Perl better than other scripting languages
> such as Python and Eiffel?
>
> [Eiffel a 'scripting language'? hm]
>
> [Larry Wall:]
> [Perl is postmodern]
> > That is, if something's worth doing, it's worth driving into the ground to the
> > exclusion of all other approaches. Look at the use of parentheses in Lisp or
> > the use of white space as syntax in Python. Or the mandatory use of objects in
> > many languages, including Java. All of these are ways of taking freedom away
> > from the end user ``for their own good''. They're just versions of Orwell's
> > Newspeak, in which it's impossible to think bad thoughts. We escaped from the
> > fashion police in the 1970s, but many programmers are still slaves of the cyber
> > police.
> [Perl is postmodern and therefore allows freedom]
>
> One could say the same about Perl and it requiring it curly braces
> around blocks? In fact, I heard the claim that its requirement to use
> curly braces around single line blocks (unlike C which allows you to
> remove them) makes Perl less prone to errors. And Perl excludes the use
> of meaningful whitespace indentation! (python does allow #{ and #}
> blocks ;)
>
> Okay, I *will* say the same about Perl and its curly braces requirement:
>
> } That is, if something's worth doing, it's worth driving into the
> ground to the
> } exclusion of all other approaches. Look at the use of curly braces in
> } Perl.
>
> *any* language makes it impossible to code in some ways, and encourages
> other coding styles, Perl included. Python, for instance, allows
> procedural and object oriented coding styles; and people keep trying for
> functional styles as well. We've even started using 'assembler style' by
> manipulating Python bytecodes! Silly, silly, whitespace meme.
>
> People think bad thoughts in Python all the time (see the 'evil hacks'
> threads showing up regularly). It's just that we have the sense to
> recognize the bad thoughts for what they are. I suppose that's why
> 'advisory locking' of member data is good enough for Python. :)
>
> Okay, sorry for all this, it's out of line, I just needed to blow off
> steam. :)
>
> Regards,
>
> Martijn






From duncan at rcp.co.uk  Mon Jun 28 04:44:31 1999
From: duncan at rcp.co.uk (Duncan Booth)
Date: 28 Jun 1999 08:44:31 GMT
Subject: [ANN] DejaGrabber 0.1 -- search Dejanews with Python
References: <7kuu73$1ki$1@brick.cswv.com>
Message-ID: <8DF363D15duncanrcpcouk@news.rmplc.co.uk>

Neel Krishnaswami <neelk at alum.mit.edu> wrote in 
<7kuu73$1ki$1 at brick.cswv.com>:

>
>If you'd like a convenient way of searching the DejaNews archive using
>Python, I'm happy to present the DejaGrabber module. This is a module
>to make searching Dejanews easier. There are two classes, DejaGrabber
>and Article. I've tried to add a reasonable number of useful
>docstrings, too.
>
>Here's an example:
>>>> from DejaGrabber import *
>>>> d = DejaGrabber(group='comp.lang.python', author='Tim Peters')
>>>> l = d.get_messages(4)
>>>> print l
>[99/06/23  30 comp.lang.python          RE: Python Exes!      Tim Peters
>, 99/06/18  28 comp.lang.python RE: Newbie: Truth values (th      Tim 
Peters
>, 99/06/18  28 comp.lang.python       RE: DATE ARITHMETIC      Tim Peters
>, 99/06/18  28 comp.lang.python     RE: NameError Problem      Tim Peters
>, 99/05/18  25 comp.lang.python     RE: while (a=b()) ...      Tim Peters
>]
>

Nice bit of work. Two comments though:
Why does get_messages(4) return 5 messages? The doc string implies it 
should return only as many as its argument.

It would be useful if a test example was included in a "if 
__name__=='__main__'" block at the end of the file. At the very least the 
example you give above, but perhaps more usefully a full command line 
driven grabber.

-- 
Duncan Booth                                    duncan at dales.rmplc.co.uk
int month(char *p){return(124864/((p[0]+p[1]-p[2]&0x1f)+1)%12)["\5\x8\3"
"\6\7\xb\1\x9\xa\2\0\4"];} // Who said my code was obscure?
http://dales.rmplc.co.uk/Duncan



From blipf at yahoo.com  Tue Jun  8 22:34:38 1999
From: blipf at yahoo.com (Flbill Blipf)
Date: Tue, 08 Jun 1999 19:34:38 -0700
Subject: makepy not war
Message-ID: <375DD2BE.D918CCCE@yahoo.com>

I hope this is not too general of a question.

After running makepy.py on a type library, how do I use
the resulting file?

I know how to use win32com.client.Dispatch, and I kinda
sort of know how to use CoCreateInstance, QueryInterface,
etc. with C++.



From paul at prescod.net  Wed Jun  2 01:02:54 1999
From: paul at prescod.net (Paul Prescod)
Date: Wed, 02 Jun 1999 00:02:54 -0500
Subject: Python 2.0 (perhaps off topic)
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
		<7imr6b$1s0$1@nnrp1.deja.com>
		<87pv3iq6ss.fsf@ev.netlab.co.jp>
		<3752C76D.604CD474@prescod.net>
		<87n1ykr0z9.fsf@ev.netlab.co.jp>
		<375372F6.A874554@prescod.net>
		<14164.7179.10346.998488@bitdiddle.cnri.reston.va.us>
		<375431E6.6BC64E80@prescod.net> <14164.43542.417731.361360@bitdiddle.cnri.reston.va.us>
Message-ID: <3754BAFE.192D7648@prescod.net>

Jeremy Hylton wrote:
> 
> Perhaps this is what I get for jumping in in the middle of a
> conversation without learning how it got started.  The example I was
> responding to depended on file object being immediately finalized to
> avoid leaking file descriptors.

This is a documented feature of Python (documented in the FAQ at least).
New users do not currently need to learn otherwise unless they use JPython
which is incompatible in this regard. In other words, this IS a
demonstration of a flaw in GC, or at least in Java GC because it required
a change in the Python language semantics.

I think that that is all anyone has been saying. Many people like the
current semantics and don't trust that a full GC would deterministically
invoke finalizers for arbitrary resources.

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for only himself
 http://itrc.uwaterloo.ca/~papresco

"Silence," wrote Melville, "is the only Voice of God." The assertion,
like its subject, cuts both ways, negating and affirming, implying both
absence and presence, offering us a choice; it's a line that the Society
of American Atheists could put on its letterhead and the Society of
Friends could silently endorse while waiting to be moved by the spirit
to speak. - Listening for Silence by Mark Slouka, Apr. 1999, Harper's





From rdev at my-deja.com  Wed Jun 30 13:59:45 1999
From: rdev at my-deja.com (rdev at my-deja.com)
Date: Wed, 30 Jun 1999 17:59:45 GMT
Subject: Reflection in Python?
References: <7ldglr$ita$1@nnrp1.deja.com>
Message-ID: <7ldlu8$lb7$1@nnrp1.deja.com>

In article <7ldglr$ita$1 at nnrp1.deja.com>,
  anders777 at my-deja.com wrote:
>  ...How do I do the
> equivalent of what Java calls reflection? I want
> to do the following:
>
> test = MathTest("testAdd")
>
> where test.run() will run the "testAdd" method
on
> the MathTest object?  I also want to be able to
> tell the user the name of the object (MathTest)
> and the method in case the method fails.  Any
> thoughts?
>
If I understand what you are trying to do, it's
easy in Python.  Assuming MathTest is an
object instance:
test = MathTest.testAdd
will return you a bound method object which you
can then call as:
test()
Getting the name of the object class and method
is also easy.  There are several ways to do this:
1) Define a doc string within the method by
   putting a quoted string as the first line
   under the def line.  You can then refer to it
   as "test.__doc__".  This could say, for
   example, "Method bar of class foo".
2) "test.__name__" will return the name of the
   method.  "test.im_class.__name__" will
   return the name of the class that the method
   is defined within.

Example:
class MathTest:
  def testAdd(self, parm1, parm2):
      "Method testAdd of Class MathTest"
      result = parm1 + parm2
      return result
<<< mt = MathTest()
<<< test = mt.testAdd
<<< test.run(2,2)
4
<<< print test.__doc__
Method testAdd of Class MathTest
<<< print test.__name__
testAdd
<<< print test.im_class.__name__
MathTest


I hope that helps.

- Roger




> Thanks,
> Anders Schneiderman
> National Journal Daily Briefings Group
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
>



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From uzs59g at uni-bonn.de  Wed Jun 30 02:43:06 1999
From: uzs59g at uni-bonn.de (Axel Schmitz-Tewes)
Date: Wed, 30 Jun 1999 08:43:06 +0200
Subject: win32com.client -- beginner question
Message-ID: <3779BC79.720C1281@uni-bonn.de>

Hello,

I ' ll try to use the comclient ability of python. In the Doku I have
found
the example:

o = win32com.client.Dispatch("Excel.Application")
o.Visible = 1
o.Workbooks.Add() # for office 97 ? 95 a bit different!
o.Cells(1,1).Value = "Hello"

Because I have installed Excel95 I have given it a try and that's the
result:

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import win32com.client
>>> o= win32com.client.Dispatch("Excel.Application")
>>> o.Workbooks.Add()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "E:\Programme\Python\win32com\client\dynamic.py", line 400, in
__getattr_
_
    raise AttributeError, "%s.%s" % (self._username_, attr)
AttributeError: Excel.Application.Workbooks

But Workbooks.Add is a available method as I can test with VBScript.
Is my installation corrupt or do I do some silly mistake or perhaps some
more serious?

Thanks for any answer.
Regards,
Axel





From tim_one at email.msn.com  Thu Jun 10 21:10:22 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 10 Jun 1999 21:10:22 -0400
Subject: Broken 1.5.2 installation? Missing cw3215mt.dll
In-Reply-To: <ZbV73.13408$WL.177860@news2.rdc1.on.home.com>
Message-ID: <000101beb3a7$2ea18440$329e2299@tim>

[Mike C. Fletcher]
> Just downloaded the latest 1.5.2 for installation.  Did the
> installation and got an error message telling me that the
> above-mentioned file is not on the path.  Indeed, it does not
> seem to exist anywhere on the machine.  I'm pretty sure this is
> part of TCL (the cygnus TCL distribution has a file named
> cw3215.dll).  However, I have no idea where to get it.

Very strange!  I don't have a cw3215mt.dll, but do have a cw3215.dll.
Presumably the former is a multi-threaded ("mt") version of the latter.  The
strings in the latter identify it as verion 1.5 of the runtime for Borland
C++ 4.50.  So bug Borland, or search the web for this name.  Several
packages out there appear to distribue *some* dll with this name.

BTW, *when* did this error pop up?  It's unclear.  That is, did the
installation complete (in which case what does the error have to do with the
installation?), or did it occur during the install (in which case exactly
when during the install)?  There's no mention of cw*.dll in either the
Tcl/Tk or Python install logs left behind by my 1.5.2 installation.  Makes
me wonder whether the installer itself needs this DLL.

just-another-day-in-dll-hell-ly y'rs  - tim





From Barry.A..Warsaw at p98.f112.n480.z2.fidonet.org  Tue Jun 29 12:52:46 1999
From: Barry.A..Warsaw at p98.f112.n480.z2.fidonet.org (Barry A. Warsaw)
Date: Tue, 29 Jun 1999 17:52:46 +0100
Subject: A Certifiable Very Cool Trick
Message-ID: <000015a3@bossar.com.pl>

From: "Barry A. Warsaw" <bwarsaw at cnri.reston.va.us>


Just a quick note because this is way too cool not to share.  When
python.org came back up today, it almost immediately got cpu slammed
(and has been for at least the last 3 hours).  I know what's going on
-- it has to do with performance problems in Mailman that in this case 
was triggered by a flood of incoming messages to all the mailing
lists.  This Mailman problem has been seen by a few other users, but
it's dang hard to track down.  Guido, Jeremy, and I got to talking
about what tools we'd like to have to help debug this problem.

My model was gdb's attach feature where you can attach to, stop and
trace through a running external C program.  I would love to be able
to do the same thing at the Python level.  We started talking about
remote pdb's with threads listening on sockets, etc. (still cool ideas
but more work).  I mentioned that I'd love to be able to just get a
traceback of where the program is currently stopped.  Even if it hoses
the interpreter afterwards, it would still be incredibly useful.
Guido and I spent some fun after hours hacking time and came up with
the following trick.

[This is described in terms of gdb, but you might be able to do
something similar with dbx or other Unix and non-Unix debuggers, I
dunno]

First, get the pid of the running Python process.  Start up gdb on the
python executable and at the gdb prompt type "attach XXX" where XXX is
the pid.  Now, at the gdb prompt, type the following:

(gdb) call PyRun_SimpleString("import sys, traceback;
sys.stderr=open('/tmp/tb','w',0); traceback.print_stack()")

Sitting in /tmp/tb will be the stack trace of where the Python program
was when you stopped it!  There's reason to believe this will not
always work, and it could seriously confuse your interpreter depending
on where it stops.  But every time I've tried it it /does/ work, and
you can even detach the program and let it continue on!

Chris Tismer, somehow I think your stackless Python might either make
this trick harder or much easier :)

makes-this-geek-quiver-with-joy-but-my-mom-would-just-scratch-her-head-ly y'rs,
-Barry




From dalke at bioreason.com  Sat Jun 12 02:29:13 1999
From: dalke at bioreason.com (Andrew Dalke)
Date: Sat, 12 Jun 1999 00:29:13 -0600
Subject: sneaky Python path
Message-ID: <3761FE39.BF1C3461@bioreason.com>

I have a problem understanding how Python knows which files to
include in the default path.  I get different results if the
binary is run in its build directory as compared to outside of
it:

val> unsetenv PYTHONPATH
val> pwd
/var/tmp
val> ~/Python-1.5.1/python
'import site' failed; use -v for traceback
Python 1.5.1 (#23, Dec  8 1998, 05:12:14) [C] on irix6
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> 
val> cp ~/Python-1.5.1/python .
val> ./python
Python 1.5.1 (#23, Dec  8 1998, 05:12:14) [C] on irix6
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> 


When I do "~/Python-1.5.1/python -v" it says:
'import site' failed; traceback:
Traceback (innermost last):
  File "/home/u03/dalke/Python-1.5.1/Lib/site.py", line 60, in ?
    import sys, os
  File "/home/u03/dalke/Python-1.5.1/Lib/os.py", line 35, in ?
    import posixpath
SystemError: bad argument to internal function


  I think the problem arises from getpath.c in search_for_prefix
where it appends "Lib" to the path if the code is in the build
directory (as determined by looking for Modules/Setup).  That's
the only reason place I can find where "Lib" can be added to
the path.  To double check, if I move Setup to Setup.orig, then
everything is okay

val> pwd
/home/u03/dalke/Python-1.5.1
val> cd Modules/
val> mv Setup Setup.orig
val> pushd /usr/tmp
/usr/tmp ~/Python-1.5.1/Modules 
val> ~/Python-1.5.1/python 
Python 1.5.1 (#26, Jun 11 1999, 23:54:47) [C] on irix6
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> 

  I found this problem when I tried running a "make test" on 
the distribution, and it failed saying:

% make test
   [...some lines omitted...]
        rm -f ./Lib/test/*.py[co]
        PYTHONPATH= ./python ./Lib/test/regrtest.py 
'import site' failed; use -v for traceback
Traceback (innermost last):
  File "./Lib/test/regrtest.py", line 193, in ?
    sys.exit(main())
  File "./Lib/test/regrtest.py", line 60, in main
    tests = args or findtests()
  File "./Lib/test/regrtest.py", line 104, in findtests
    testdir = findtestdir()
  File "./Lib/test/regrtest.py", line 160, in findtestdir
    testdir = os.path.dirname(file) or os.curdir
AttributeError: path


  I'm not sure what changed between then and now.  All I can
think of is that my home account got moved after a recent server
crash, but I can't see any place in the code where that would
cause a problem, and my home directory isn't even stored in
the .o files anywhere.

  Also, I've found a workaround by copying a valid posixpath.pyc
into the Lib directory.  Watch this:

val> rm Lib/*.py
val> ./python 
'import site' failed; use -v for traceback
Python 1.5.1 (#26, Jun 11 1999, 23:54:47) [C] on irix6
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> 
val> cp /usr/local/lib/python1.5/posixpath.pyc Lib
val> ./python 
Python 1.5.1 (#26, Jun 11 1999, 23:54:47) [C] on irix6
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam

 So I'm stumped, stymied and just plain old confused.  Python is
doing something sneaky and I can't figure it out :(

						Andrew
						dalke at bioreason.com



From larsga at ifi.uio.no  Mon Jun 21 02:59:58 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 21 Jun 1999 08:59:58 +0200
Subject: XML-toolkit for Windows - binaries available?
References: <376701a3.72288277@newssrv> <wkzp208udt.fsf@ifi.uio.no> <3767fc51.1547455@news.iol.ie> <wkd7yv7t5p.fsf@ifi.uio.no> <37693395.731772@news.iol.ie> <376da59b.2034245@news>
Message-ID: <wkiu8it4wh.fsf@ifi.uio.no>

* Lee Borkman
| 
| Now I have to edit __init__.py to include sgmlop and pyexpat in the
| __all__ list, is that right?  Now which __init__.py would that be?
| I assume it's C:\Program Files\Python\Lib\xml-0.5.1\__init__.py

I haven't done this, and I don't think you need to do it unless you
care what the result of 

  import xml
  dir(xml)

is.

| But none of the demos will work.  I keep getting:
| ImportError: No module named xml.sax
| or:
| ImportError: No module named pyversioncheck
| 
| 
| I guess I still don't understand where Python (and PythonWin) look for
| their modules, and I guess I don't understand quite a bit of other
| stuff.

It sounds like you haven't set your PYTHONPATH. This is an environment
variable with a list of ';'-separated directories, but you can also
set it in the registry under HKEY_LOCAL_MACHINE\Software\Python
 
Make sure you include C:\Program Files\Python\Lib\xml-0.5.1\ in that
list. 

--Lars M.



From rgolden at x25.net  Thu Jun 24 18:56:13 1999
From: rgolden at x25.net (Ryan Golden)
Date: Thu, 24 Jun 1999 22:56:13 GMT
Subject: python-mySQL module documentation other that python.org?
Message-ID: <3772b3ee.27015456@news2.x25.net>

Can anyone tell a beginner where one might find some example scripts
for using the mySQL.py interface?  I found the documentation on
python.org to be confusing, and couldn't find any links to any example
scripts.





From pduffin at mailserver.hursley.ibm.com  Mon Jun 28 08:52:50 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Mon, 28 Jun 1999 13:52:50 +0100
Subject: Linking an Interpreter dynamically
References: <slrn7mrpes.94.tuna@delirious.fsnnet>
Message-ID: <37777022.19A1@mailserver.hursley.ibm.com>

Lloyd Weehuizen wrote:
> 
> Hi!
> 
> I was just wondering if it is possible to create a dynamically linkable
> python interpreter? I've looked in the docs, but can't seem to find
> anything, and "make sharedinstall" doesn't seem to give me any linkable
> libraries??
> 

On Windows Python is built as a DLL but on Unix Python is built as an
archive library. Apparently Guido decided that he did not want to build
Python as a shared library on Unix (or even support it in the standard
installation) for some reason but I do not know why. The only reason
why it is built as a DLL on Windows is that otherwise Python could not
load dynamic extensions due to linking problems in Windows.


-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From mychoko at my-deja.com  Thu Jun 10 21:30:40 1999
From: mychoko at my-deja.com (my)
Date: Fri, 11 Jun 1999 01:30:40 GMT
Subject: swig' example don't swig !
Message-ID: <7jpos1$2ki$1@nnrp1.deja.com>

hi,

i was trying to "swig" one of the examples available in the
distribution under  Examples/python/manual/pde1
running (almost) exactly the swig command given in the documentation
and i get these errors:

> swig1.2 -python -shadow pde.i

Generating wrappers for Python
pde.h : Line 17. Syntax error in input.
pde.h : Line 21. Syntax error in input.
pde.h : Line 23. Syntax error in input.
pde.h : Line 29. Syntax error in input.

Can anyone help me on that ?

Thanks a lot,


--
my.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From guido at CNRI.Reston.VA.US  Mon Jun 21 16:26:34 1999
From: guido at CNRI.Reston.VA.US (Guido van Rossum)
Date: Mon, 21 Jun 1999 16:26:34 -0400
Subject: mailman outage
Message-ID: <199906212026.QAA15112@eric.cnri.reston.va.us>

Two of our mailing lists, python-list at python.org and
mailman at python.org, may have suffered a brief outage today (Jun 21st)
between 3 and 4:30 pm EST.  A disk volume filled up (ironically with
mail syslog messages) and this caused some mailman processes to fail.
I believe we've fixed the problem now, but if you sent something to
either list during or shortly before the time of the outage, and did
not see it come through, you might consider resending.

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



From kiki at pixar.com  Fri Jun  4 02:06:26 1999
From: kiki at pixar.com (kiki)
Date: Thu, 03 Jun 1999 23:06:26 -0700
Subject: More Tkinter menu problems
References: <3755C714.526B7994@pixar.com> <%Zm53.899$S15.46310@dfiatx1-snr1.gtei.net> <c75g14863sh.fsf@pc142.cosc.canterbury.ac.nz>
Message-ID: <37576CE1.133C4514@pixar.com>

> Another possible way of implemented getting the same result, without
> resorting to unreadable lambda expressions is to create a class
> something like...
>
> class Command:
>     def __init__(self, func, *args, **kw):
>         self.func = func
>         self.args = args
>         self.kw = kw
>
>     def __call__(self, *args, **kw):
>         args = self.args + args
>         kw.update(self.kw)
>         apply(self.func, args, kw)
>
> then...
>
> command = Command(self.select_color, new_color)
>
> This costs one extra function call when it is used, but I think the
> decrease in complexity of the expression make it worth it.
>
> --
> Tim Evans

Great!  Thanx.  That worked well.

Kiki




From paul at prescod.net  Mon Jun 28 15:13:00 1999
From: paul at prescod.net (Paul Prescod)
Date: Mon, 28 Jun 1999 19:13:00 GMT
Subject: Pythonwin and Office 2000
References: <377648A7.9B9BE59F@prescod.net>
		<7l672r$5k0$1@m2.c2.telstra-mm.net.au>
		<3776D1E6.45E1AAF7@prescod.net> <7l7bbt$42u$1@m2.c2.telstra-mm.net.au>
Message-ID: <3777C93C.F0B5E236@prescod.net>

Mark Hammond wrote:
> 
> I agree.  Worse, as you found out, it breaks when upgrading the COM object.
> However, the only way to do it textually would be to search the registry for
> the name.  

Would that be the same speed as .Dispatch("..."), or ar you saying it
would be slower? Anyhow, could we store the mapping from string names to
UUIDs in the gencache directory so that the second time is fast?

> However, it is worth noting that speed is no longer a good justification for
> using makepy.  Recent changes to dynamic dispatch mean that we are now
> comparable speed wise to VB without makepy support.  The reason you often
> now need makepy support is to get decent parameter support, but the problem
> is worse than that.  In many non-trivial cases, the code you write is simply
> not compatible between makepy supported and dynamic dispatch.  So, if you
> want to ensure your code will run when installed, you must _insist_ on
> makepy support.  (ie, silent failure of makepy is usually not a good option,
> as you are simply delaying the inevitable error, but in a less obvious way).

Okay, I don't understand what stuff works with makepy and what stuff
doesn't. Let's put it this way: if we did everything dynamically the same
stuff VBScript does dynamically would we slow down so much that makepy
would be significantly faster again?

Part of the reason I was using makepy was because I needed access to
constants. How does JScript/VBscript handle this? Do you just have to use
numbers?

Even if makepy is 10 times faster than dynamic dispatch, if code at least
behaved the same with and without it, then we could just try to do the
makepy generation when possible and fallback to dynamic dispatch
otherwise. Savvy programmers could manually force makepy generation if
they really care about speed.

> There is plenty of scope to rationalize some of this.  As Python COM support
> has grown from its humble beginnings, some of the existing code is no longer
> appropriate.  It is also big and scary enough that fiddling will not have
> the desired effect - it needs a good, hard think and a revamp.
> Compatibility shouldnt be an issue - there are only about 4 truly public
> functions in the whole package - 99% of users will only care that
> "win32com.client.Dispatch" does what they expect!

You're suggesting a volunteer take your big, scary code and improve upon
it? :)

Is this just Python code that needs fixing or a mix of Python and C++?
(i.e. are the C++ primatives you provide enough to build the cool new
Python system on top?)

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for only himself
 http://itrc.uwaterloo.ca/~papresco

Those who profess to favor freedom and yet depreciate agitation, are men 
who want crops without plowing up the ground. They want rain without 
thunder and lightning. They want the ocean without the roar of its many 
waters. - Fredrick Douglass
 http://www.informamerica.com/Articles/Quotes.htm



From gdd0 at gte.com  Tue Jun 29 15:39:54 1999
From: gdd0 at gte.com (Gary D. Duzan)
Date: Tue, 29 Jun 1999 19:39:54 GMT
Subject: ANN: Stackless Python 0.2 
Message-ID: <199906291939.PAA24426@lion.gte.com>

In Message <37791B4D.43ADC401 at appliedbiometrics.com> ,
   Christian Tismer <tismer at appliedbiometrics.com> wrote:

=>Gary Duzan wrote:
=>> 
=>> Toby J Sargeant wrote:
=>> >
=>...
=>>    This appears to be the conventional wisdom. The optimization class I
=>> took used an ML-like language, and the first thing we did was to move
=>> heap frames, which was the default, to the stack (i.e. closure
=>> optimization.)
=>
=>Loosing much of flexibility by doing that, btw.

   True. For example, (if I remember correctly; it has been a while)
tail call optimization is defeated. In this particular case, however,
we didn't break the language semantics (which didn't guarantee constant
memory tail recursion) because we only did stack conversion after
checking that it wouldn't break anything. Otherwise, the closure is
left on the heap. I see that I wasn't clear about that in my original
post; sorry 'bout that.

=>> It might be less of an issue in an interpreted language if it already has
=>> heap
=>> allocation (or other sorts of) overhead to deal with when making calls.
=>> There may be other allocation tricks which can be used, as well.


=>> > It seems that this should affect Stackless Python equally as much. Does a
nyone
=>> > have anything to add on the subject? I would imagine that frames could be
=>> > allocated and managed in chunks to alleviate a lot of the memory manageme
nt
=>> > load...
=>> 
=>>    I would think that as well (at least at first.) Maybe pools of different
=>> sizes of frame to limit the space overhead.
=>
=>Maybe you should really look into the sources.
=>You are discussing things which are not Python
=>related, since this simply does not apply.

   Freely conceeded.

=>The Python "stacks" are so tiny that you can forget about them.
=>Allocations always never occour, unless your program runs
=>astray in recursion.
=>
=>The cost of a call comes from the slightly complicated
=>function setup which is done all the time. Frames belong
=>to the fastest available Python objects already, 
=>like, say dictionaries.

   I thought that might be the case, which is the point I was making in
my earlier paragraph. I was just briefly pondering possibilities for
reducing frame memory management time, if it were necessary. Stackless
Python certainly seems to me to be a Good Thing.

					Gary Duzan
					GTE Laboratories





From air at apex.net  Thu Jun  3 14:51:16 1999
From: air at apex.net (Todd Neal)
Date: Thu, 3 Jun 1999 13:51:16 -0500
Subject: Re example
Message-ID: <k9A53.62$G2.553414@news.apex.net>

Looking for some examples on how to use re, had a script made with sed ,awk,
and lynx to download and parse webpages for viewing on a palmpilot, I would
like to convert it to Python but I cant find any good examples of how to use
the re modules.
                                                                Thanks, Todd





From jepler at inetnebr.com  Fri Jun  4 12:33:30 1999
From: jepler at inetnebr.com (Jeff Epler)
Date: Fri, 04 Jun 1999 16:33:30 GMT
Subject: Alternative dictionary output format?
References: <7j8pd8$f01$1@nnrp1.deja.com>
Message-ID: <slrn7lg030.i5l.jepler@craie.inetnebr.com>

On Fri, 04 Jun 1999 14:55:42 GMT, rael at my-deja.com
 <rael at my-deja.com> wrote:
>s = "%(href)s/foo/bar?x=%m&y=%n"

(he wants to only substitute the %(href) and not the rest)

s = "%(href)s/foo/bar?x=%%m&y=%%n"

i.e., double the percents you want to have not interpolated at this stage
of the game.

A clever regular expression might be able to do this for you, but the way
to write it 100% bulletproof escapes me.

Jeff
-- 
\/ http://www.slashdot.org/                      Jeff Epler jepler at inetnebr.com
I just forgot my whole philosophy of life!!!



From def-p at usa.net  Sun Jun 20 09:29:57 1999
From: def-p at usa.net (Def P)
Date: Sun, 20 Jun 1999 13:29:57 GMT
Subject: [Tkinter] Placing a ScrolledText on a Canvas?
Message-ID: <19990620132958.27148.qmail@nwcst292.netaddress.usa.net>

The subject says it all, really... I would like to have a window with some
fancy background, yet full functionality. So I thought, I use a Canvas, put an
image on it, then put additional components on it. So far it works
reasonably... I put an Entry on it (== the Canvas with PhotoImage on it), with
no problems at all.

However, I cannot seem to manage to put a ScrolledText on the same Canvas. The
following code demonstrates this:

from Tkinter import *
from ScrolledText import ScrolledText

root = Tk()
root.canvas = Canvas(root, width=400, height=580,
background='gray',relief=SUNKEN)
root.canvas.pack()

photo = PhotoImage(file="c:/multimedia/pix/SmallerDragonBanner.gif")
item = root.canvas.create_image(1, 1, anchor=NW, image=photo)

entry = Entry(root, width=38)
entry1 = root.canvas.create_window(60, 490, anchor=NW, window=entry)

textBox = ScrolledText(root, width=15, height=30)
textBox1 = root.canvas.create_window(50, 100, anchor=NW, window=textBox)

root.mainloop()

I cannot copy the exact error message but it says that it "can't use
.8201376.text in a window item of this canvas"...

Am I doing something wrong, or is a ScrolledText "just" an item that cannot be
put on a canvas?

Thanks,
Def P


____________________________________________________________________
Get free e-mail and a permanent address at http://www.netaddress.com/?N=1



From ljz at asfast.com  Wed Jun  2 00:38:36 1999
From: ljz at asfast.com (Lloyd Zusman)
Date: 02 Jun 1999 00:38:36 -0400
Subject: Is pcgi usable without Zope?
Message-ID: <ltd7zf6wtv.fsf@asfast.com>

I have to admit that I'm a bit confused about pcgi these days.  The
most recent documentation for it was last updated in August, 1998, and
it still refers to Bobo and Principia.

Now that we have Zope, is pcgi now intimately bundled with it, never
to emerge on its own again?  Or is there a way to run pcgi without
Zope?

Don't get me wrong ... I like Zope.  I'm just wondering if there's any
way to make use of pcgi without Zope during those occasional times
when Zope's full-blown functionality isn't necessary.

Thanks in advance for any light that you folks could shed on this
issue.

-- 
 Lloyd Zusman
 ljz at asfast.com



From arcege at shore.net  Thu Jun 10 14:56:15 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Thu, 10 Jun 1999 18:56:15 GMT
Subject: Date formats
References: <375FEA53.95E3D3BC@ausinfo.com>
Message-ID: <jTT73.2684$nn.808830@news.shore.net>

mike milliger <milliger at ausinfo.com> wrote:
: I read the documentation in the Python Reference Library (6.3) on time
: access and conversions and chapter 8 in programming python, yet I can't
: figure out if i need ( or how to use) a tuple to format the return from
: localtime().  i get a string now (wed jun 10 1999 xx:xx:xx) but i need
: to format it to yyyymmdd_hhmmss.  Is there any switch ( %) or function
: that can be useful?  Any help is appreciated.

"Here strftime comes to save the day!"  (Okay, I won't do my bad
impressions of Andy Kaufman anymore.)

The time module has a handy function called strftime, which works like
format strings, but on the time tuple.

  my_date = time.strftime("%Y%m%d_%H%M%S", time.localtime(time.time()))
  print my_date
19990610_145144

Or you can look at the less-standard, more versatile mxDateTime package
(http://starship.python.net/~lemburg/mxExtensions.html).

  -Arcege




From druid at princeton.crosswinds.net  Fri Jun  4 20:26:15 1999
From: druid at princeton.crosswinds.net (The Well of Latis)
Date: Sat, 05 Jun 1999 00:26:15 GMT
Subject: newbie trouble with raw_input and idle
References: <37581eb4.2704370@netnews.worldnet.att.net> <7j75f5$u49$1@nnrp1.deja.com>
Message-ID: <37586cce.3298769@netnews.worldnet.att.net>

On Fri, 04 Jun 1999 00:09:15 GMT, Sam Schulenburg <sams at quinta.com>
wrote:

>In article <37581eb4.2704370 at netnews.worldnet.att.net>,
>  druid at princeton.crosswinds.net wrote:
>> If I enter the following code (or any other using raw_input for that
>> matter) into idle's shell it works, when trying to run it from a file
>> (new window, edit it, save it then run it, nada)
>>
>> def f():
>>     x=raw_input("ladida:")
>>     print x
>>
>> f()
>>
>> systems involved are a win98 and an nt4 machines, tcl version 8.0.5,
>> Python 1.5.2. The same code works under the plain Python interactive
>> shell, and pythonwin, but not idle. Any ideas?
>>

>I just tried your sample under windows NT with IDLE 0.4 and it worked OK
>Sam Schulenburg
>
>

If typed or directly imported into idle it does work. When doing the
following it did not work (specifically the print statement was never
executed)

Ctrl-N (New Window)
type in the code
Ctrl-S (Save)
F5    (Run module)

It does not work. I found a similar post back on deja about a few
months back, but could not find a resolution for it posted. 
Sparrow (Alex)                             | /|\ brewing on the net
http://www.crosswinds.net/~prin_druid      |      



From arcege at shore.net  Wed Jun  2 10:47:41 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 02 Jun 1999 14:47:41 GMT
Subject: python extension problem (newbie)
References: <37548FA5.526A8E48@crystal.uwa.edu.au>
Message-ID: <hub53.1615$nn.483094@news.shore.net>

Douglas du Boulay <ddb at crystal.uwa.edu.au> wrote:
: Hi ,

: I've been trying to write a python extension in C to interface to a
: Fortran program
: and a class wrapper for the extension using the following code loosely
: adapted from Mark Lutz's Programming Python  stackmodule example.

: Unfortunately whereas his example works fine, mine goes into a recursive
: loop whenever an
: instance is instantiated.  It seems that the __getattr__  method of the
: wrapper class
: cannot find the getattr method of the extension library  and defaults to
: trying to call itself, infinitely
: recursively.

: Having spent about 8 hours trying to figure out why, I still have no
: idea.
: Help!
: Thanks

: Doug

:>>file oograsp.py

: import p2fgraspint

: class Grasp:
:     def __init__(self, start=None):
:         self._base = start or p2fgraspint.Grasp()
:         print "doing done"                  # which is never executed
:     def __getattr__(self, name):
:         print "called getattr"              # executed  endlessly
:         return getattr( self._base , name)
:     def __setattr__(self, name, value):
:         return setattr(self._base, name, value)
                         ^^^^^^^^^^
This is the problem: inside your __setattr__ method, you are performing
a __getattr__ on an attribute that does not exist yet.  So your __init__
function is attempting to use a value that isn't set yet.  (cf. section
"3.3.2 Customizing attribute access" in the Python Reference Manual,
http://www.python.org/doc/current/ref/attribute-access.html.)

You either want to:

a) assign to "self.__dict__['_base']" in the __init__ method, or
  def __init__(self, start):
    self.__dict__['_base'] = start or p2fgraspint.Grasp()
    print "doing done"

b) check for the attribute being set in __setattr__ and perform the
operation based on that:
  def __setattr__(self, name, value):
    if name == '_base':
      self.__dict__[name] = value
    else:
      setattr(self._base, name, value)

I suggest (b)  it is better for when you change or add localized
attribute names later, and is better for subclasses (IMO).

(Also, setattr returns None, so you do not need the "return".)

  -Arcege




From andy at robanal.demon.co.uk  Mon Jun 21 14:03:59 1999
From: andy at robanal.demon.co.uk (Andy Robinson)
Date: Mon, 21 Jun 1999 18:03:59 GMT
Subject: Newbie Q: How to call unmodified VB DLL from Python
References: <3769bcd1.54492213@news.earthlink.net>
Message-ID: <376f7e8c.817023@news.demon.co.uk>

lull at acm.org (John Lull) wrote:

>As a first Python project, I'd like to use Python (or JPython) to
>automate some hardware testing under Win95/98/NT.  The hardware in
>question is run by a DLL written in VB, and to which I do not have
>source.
>
>"Extending & Embedding the Python interpreter", ch. 3, has
>instructions for building a DLL which will work, but assumes you're
>working in VC++ and have the source.
>
>If anyone could point me toward instructions on how to make this work,
>I'd be grateful.
You need to use Sam Rushing's calldll/windll package at
www.nightmare.com.  This lets you load and call functions in DLLs
dynamically (i.e. without needing a C compiler).  It also provides
python types to emulate C strings and arbitrary memory buffers, so can
call anything.

Needless to say, it is also possible to crash stuff if you get the
arguments wrong, but once set up it works beautifully.  I have been
using it in production round the clock for months now...

Regards,

Andy



From matomira at iname.com  Thu Jun 10 11:23:00 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Thu, 10 Jun 1999 17:23:00 +0200
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <4.1.19990610150908.00d09790@exchsrv> <7jog8f$ljv$1@Starbase.NeoSoft.COM>
Message-ID: <375FD854.2314D886@iname.com>


Cameron Laird wrote:

> In article <4.1.19990610150908.00d09790 at exchsrv>,
> Fernando Mato Mira  <matomira at iname.com> wrote:
>
> So I take it you're going to end up arguing
> for Eiffel in favor of Scheme?

Oh, no. The crossposting logic is that people
in any comp.lang.whatever-is-unpopular-and-clean tend to know
other things besides their favorite, because they are many times
forced to touch other stuff as well.
I'm not going to argue during the presentation. I just made
sure to pass the message that people should look around
before jumping eyes closed in some bandwagon (but
obviously, I took care to mention things like Java and STk
(Disclaimer: I'm not a Java pundit. And if I needed to to
JVM-compatible stuff, I'd try to use Kawa first (duh)))

My first objective in this research is to try to be honest. I only
have fuzzy memories of the discussions about tcl of years ago,
and as I promised myself never to read a book on it.. (and even
if I did, just for this, then I would be commiting the popular gaffe
of speaking about something in which you have little experience).

> I'll ask a better question:  is maintainability
> to you a philosophically-accessible dimension?

Yes. A sound software engineering philosophy
requires that you think about maintainability
for tools used to build things that will be more than
prototypes. Of course, what different groups
in `sound' differs, so we are indeed getting into
philo-land here.

> If you seriously want to debate syntax and
> semantics in academic abstraction ... well, at
> least we need to know that.

Syntax ans semantics in the sense that
affect a software engineering practice.

<NOT IMPORTANT>

> >If so, does speed ever become an issue? I assume that Tk works well
> Rarely.  Much less often than the uninitiated
> suspect.

That's the impression I have. But then, what are those `less often' cases?
(Besides trying to do silly things, like writing a fast-paced  video game in Tk)

> >enough that you'd never have to look at the tcl code, and live happily
> >inside your reflective Scheme world)
> Some TkSchemers do.  Some don't.  It depends.

Examples?

</NOT IMPORTANT>




From arcege at shore.net  Thu Jun  3 08:21:30 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Thu, 03 Jun 1999 12:21:30 GMT
Subject: Why won't this work???
References: <7j0rkr$1eeu$1@rtpnews.raleigh.ibm.com>
Message-ID: <eru53.1740$nn.518460@news.shore.net>

Jr. King <n at n.com> wrote:
: I know you are getting tired of me, but I have tried repeatedly to solve
: this on my own and I can't get it to work.
: I am trying to do a python callback to call from a c/c++ program.
: Here is the c/c++ code

: // learnPyth.cpp : Defines the entry point for the console application.
: //

: #include "stdafx.h"
: #include "python.h"
: #include "stdio.h"
: #include <iostream.h>
: #include <windows.h>

: static PyObject *my_callback = NULL;

: static PyObject* Test(PyObject *self, PyObject* args){
:  cout<<"Test"<<endl;
:  int command;
:  int aoe;
:     if (!PyArg_ParseTuple(args, "ii", &command,&aoe))
:         return NULL;

:  int xs(5);
:  PyObject* xarg = Py_BuildValue("(i)",xs);

:  PyEval_CallObject(my_callback, xarg);//python returns the PYNone thingie
: here
:  Py_DECREF(xarg);

:  return args;
: }

: static PyObject* setCallback(PyObject *self, PyObject* args){
:     PyObject *result = NULL;
:     PyObject *temp;

:     if (PyArg_ParseTuple(args, "O:set_callback", &temp)) {
:         if (!PyCallable_Check(temp)) {
:             PyErr_SetString(PyExc_TypeError, "parameter must be callable");
:             return NULL;
:         }
:         Py_XINCREF(temp);         /* Add a reference to new callback */
:         Py_XDECREF(my_callback);  /* Dispose of previous callback */
:         my_callback = temp;       /* Remember new callback */
:         /* Boilerplate to return "None" */
:         Py_INCREF(Py_None);
:         result = Py_None;
:     }
:     return result;
: }

: static PyMethodDef meths[]={
:  {"testmeth", Test, METH_VARARGS},
:  {"setCallback", setCallback, METH_VARARGS},
:  {NULL,  NULL}  /* sentinel */
: };

: int main(int argc, char* argv[])
: {
:  Py_Initialize();
:  char input[200];
:  int ret;
:  FILE *fp;
:  cout<<Py_GetPath()<<endl;
:  fp = fopen("beer.py","r+");
:  Py_InitModule("testmeth", meths);
:  Py_InitModule("setCallback",meths);
:  if(fp == NULL)
:   cout<<"Didn't open file";
:  else{
:   ret = PyRun_AnyFile(fp,"beer.py");
:   cout<<ret<<endl;
:   fclose(fp);
:   Py_Exit(0);
:  }
:  return 0;
: }

: Here is the python code

: #! /usr/bin/env python
: # By GvR, demystified after a version by Fredrik Lundh.
: #import sys
: import testmeth
: import setCallback
: n = 20
: #if sys.argv[1:]: n = int(sys.argv[1])

: def setN(x):
:  x = n
:  print n

: def bottle(n):
:     if n == 0: return "no more bottles of beer"
:     if n == 1: return "one bottle of beer"
:     return str(n) + " bottles of beer"

: setCallback.setCallback(setN)
: testmeth.testmeth(13,9)
: for i in range(n):
:     #print bottle(n-i), "on the wall,"
:     #print bottle(n-i) + "."
:     #print "Take one down, pass it around,"
:     print bottle(n-i-1), "on the wall."

: print n

: thanks again.

To be honest, what you think the "error" is here is quite confusing
from your output.  But your comment in the Test() function mentions
that PyEval_CallObject() returns Py_None from setN.  This would be correct,
since a function with no "return" statement would return None (Py_None)
(cf. section 5.3.4 "Calls" in the Python Reference Manual).

But what you do want to do is capture the return value from
PyEval_CallObject to at the very least decrement the reference count
(with Py_DECREF) or to see if there was an exception (NULL).

In addition, you return the argument list passed to the function, this
argument list is owned by the calling C routine and will its reference
count will be decremented (most likely to zero, thereby destroying the
object) when Test() returns.  If you are going to return the argument
list, then you will want to increment the reference count.

  static PyObject* Test(PyObject *self, PyObject* args) {
    cout<<"Test"<<endl;
    int command;
    int aoe;
++  PyObject *res;
    if (!PyArg_ParseTuple(args, "ii", &command,&aoe))
      return NULL;

    int xs(5);
    PyObject* xarg = Py_BuildValue("(i)",xs);

    //python returns the PYNone thingie here
++  res = PyEval_CallObject(my_callback, xarg);
    Py_DECREF(xarg);
++  if (res == NULL) /* propagate the exception raised */
++    return NULL;
++  Py_DECREF(res);

++  Py_INCREF(args);
    return args;
  }

Hope this helps,
  -Arcege




From mark_removethis_ at _removethis_btweng.krakow.pl  Fri Jun 11 04:59:46 1999
From: mark_removethis_ at _removethis_btweng.krakow.pl (MK)
Date: Fri, 11 Jun 1999 08:59:46 GMT
Subject: dict vs kjBuckets vs ???
References: <000801beb3b1$5c5af240$329e2299@tim>
Message-ID: <3760c9d2.1106146@news.cyf-kr.edu.pl>

On Thu, 10 Jun 1999 22:23:15 -0400, "Tim Peters"
<tim_one at email.msn.com> wrote:
>[MK]
>> In some book on algorithms I've read that after inserting limited
>> number of items performance of operating on hash tables
>> drops dramatically.

>Depends on the details.  What you read is true of some kinds of hash tables.
>Python's dicts dynamically expand to keep the "load factor" under 2/3, so
>what you read isn't applicable to Python in normal use.

Great.

>> I plan to write a program that would store lots (in range of 10M or even
>> more) of relatively small objects (a few hundred bytes at most), so what
>> do you think I should use?

>Let's do a little math <wink>:  10M * 100 = ?, a lower bound on what you're
>contemplating.  Do you have gigabytes of RAM?

I'm opening a boutique.

>> I thought about dictionaries, kjBuckets, or maybe even library called
>> Metakit for Python (http://www.equi4.com/metakit/info/README-Python.html).

>> what-do-you-think-ly y'rs

>You don't really want to know <wink>.  Memory-based data structures aren't
>going to work for the size of thing you have in mind.  If you can make it
>fly it all, you'll likely require a powerful database, so of those choices
>Metakit is the only approach that's not dead on arrival.

A few additional informations: items stored would be natural language
text fragments (several sentences at most, several words typically)
+ binary descriptions, primary operation would be lots of searching. 
Is there anything else that would be better for this kind of program? 
Object database?

>better-still-write-it-in-perl<wink>-ly y'rs  - tim

it's-a-stiff-ly yours




--------------------------------------------------
Reality is something that does not disappear after
you cease believing in it - VALIS, Philip K. Dick
--------------------------------------------------

Delete _removethis_ from address to email me



From peterh at sapros.com  Mon Jun  7 02:57:35 1999
From: peterh at sapros.com (Peter Haight)
Date: Sun, 06 Jun 1999 23:57:35 -0700
Subject: extracting contents from a text file 
Message-ID: <199906070657.XAA00519@wartch.sapros.com>

>I would like to extract the text contents residing in another server and
>display it as html format using python. Can anyone help me I am new to
>python.

Well, here's an example. I'm assuming that the 'text contents' you are
talking about is on an HTTP (web) server. I'll grab the Python home page as
an example at 'http://www.python.org/' but you could easily modify this
for other kinds of servers and documents. It writes the html to stdout. You
won't need the 'string.replace' lines if your 'text contents' are not
formatted as html.

--------------------
import string
import sys
import urllib

file = urllib.urlopen('http://www.python.org/')

header = \
"""
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN">
<html>
<head>
<title>Example</title>
</head>
<body>
<pre>
"""

footer = \
"""
</pre>
</body>
</html>
"""

out = sys.stdout

out.write(header)
for line in file.readlines():
	#
	# This keeps a browser from parsing an html file.
	#
	line = string.replace(line, '<', '&lt;')
	line = string.replace(line, '>', '&gt;')
	out.write(line)
out.write(footer)




From janssen at parc.xerox.com  Thu Jun 24 13:21:52 1999
From: janssen at parc.xerox.com (Bill Janssen)
Date: Thu, 24 Jun 1999 10:21:52 PDT
Subject: make clean bug in 1.5.2
Message-ID: <99Jun24.102155pdt."3693"@watson.parc.xerox.com>

Running "make clean" seems to leave a couple of problematic files,
SRC/libpython1.5.a, and getbuildinfo.o, which then cause problems
when I try to "make" again.

Bill



From ovidiu at cup.hp.com  Wed Jun 30 21:02:41 1999
From: ovidiu at cup.hp.com (Ovidiu Predescu)
Date: Thu, 01 Jul 1999 01:02:41 +0000
Subject: coverage tool for Python?
Message-ID: <377ABE31.249B1E51@cup.hp.com>

Hi,

Are there any coverage tools for Python?

Thanks,
-- 
Ovidiu Predescu <ovidiu at cup.hp.com>
http://www.geocities.com/SiliconValley/Monitor/7464/



From dalke at bioreason.com  Mon Jun 14 13:33:29 1999
From: dalke at bioreason.com (Andrew Dalke)
Date: Mon, 14 Jun 1999 11:33:29 -0600
Subject: Checking for Python modules in autoconf
References: <199906031338.JAA28564@amarok.cnri.reston.va.us> <m3emjgt6on.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <37653CE9.ADF884E1@bioreason.com>

Michael Hudson <mwh21 at cam.ac.uk>
> James Henstridge has written some extensions to automake/autoconf as
> part of his Python GTK+ bindings.

I'm going to finish up my work on Python support in automake, which
will be included in the current development version of automake.  I
haven't seen any contributions in the automake tree from James
Henstridge.

Regardless, I will add a way to test for Python modules using
automake.  Given the way GNU works, please don't send me any code
to be added unless you've sent the copyright disclaimer form to
the FSF.

						Andrew Dalke
						dalke at bioreason.com



From fatjim at home.com  Thu Jun  3 13:25:17 1999
From: fatjim at home.com (Jim Meier)
Date: Thu, 03 Jun 1999 17:25:17 GMT
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net> <aahzFCpHw8.Gxt@netcom.com> <7j4e2u$8rh$1@m2.c2.telstra-mm.net.au> <37564132.2D01B7C3@pop.vet.uu.nl>
Message-ID: <3756BAFC.2066A636@home.com>


Martijn Faassen wrote:

> Mark Hammond wrote:
>
> [Aahz Maruch does not like block indenting a lot]
>
> > Of course, this is your HO.  Many other people (including myself) found that
> > a few weeks in, they thought "this is great", and many years later, can't
> > believe that other language still hold on to the "old ways".  When I take a
> > quick look at some of the new little languages referenced on this newsgroup,
> > believe it or not, my _first_ reaction is "God those block closers are
> > stupid and ugly" :-)
>
> I never was much afraid of this indentation means block structure thing,
> probably not having been spoiled by other languages that are horrid and
> practice it, so it was mostly a neat discovery. Now, I am less than a
> year in and I already have the same reaction to other languages. "Neat
> language..but ick, what are those block closers for?" Just the other day
> I was looking at some pseudocode in a book and my eyes kept stumbling
> over the block closers.

Ditto about not being afraid of the indentation. I've been rabidly happy about
using indentation properly ever since I moved up from a commodore vic-20: in the
vic's BASIC, you typed a line number and then just typed until you had trouble
reading the line - it would just wrap around the 20-character wide screen. Moving
up to a DOS machine with MS QBasic, I was very releived to be allowed to indent my
programs properly. But QBasic has crazy block open/closures: each type of
statement is different!

C:
if(cond) {
    foo();
} else {
    bar();
}

while(1) {
    foo()
}

Python:
if cond:
    foo()
else:
    bar()

while 1:
    foo()

BASIC:
if cond then foo else bar
-or-
if cond then
    foo()
else
    barr()
endif

while(1)
    foo()
wend


> I must say that Emacs Python mode helps tremendously in this respect,
> though. Without it or something similar block indentation wouldn't be as
> easy (though I've found I can code in Windows Notepad too if I really
> want to :).

Oh, ditto. I just *love* (x)emacs now, though I would really like it to take far
less memory and be extendable with python.


Wow, talking about old H/W makes me sound old - I'm only 19!
-Jim




From eugene.leitl at lrz.uni-muenchen.de  Mon Jun 14 23:24:06 1999
From: eugene.leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Mon, 14 Jun 1999 20:24:06 -0700 (PDT)
Subject: NEWBIE: opaque error message
Message-ID: <14181.50740.800169.537860@lrz.de>

I've got this piece of code here, kindly donated as a point of
departure. 

from MMTK import *
from MMTK.Lattice import BravaisLattice

a = 4.516*Units.Ang
b = 4.516*Units.Ang
c = 7.354*Units.Ang
alpha = 90*Units.deg
beta = 90*Units.deg
gamma = 120*Units.deg

ex = Vector(1., 0., 0.)
ey = Vector(0., 1., 0.)
ez = Vector(0., 0., 1.)

# Caution: this works only for beta=90*Units.deg
a1 = a*ex
a2 = b*Rotation(ez, alpha)(ex)
a3 = c*Rotation(ey, gamma)(ex)

water = Molecule('water')
water.O.setPosition(0.3333*a1 +  0.6667*a2 + 0.0629*a3)
water.H1.setPosition(0.3333*a1 + 0.6667*a2 + 0.1989*a3)
water.H2.setPosition(0.4551*a1 + 0.9102*a2 + 0.0182*a3)

universe = InfiniteUniverse()
for point in BravaisLattice((a1, a2, a3), (3, 3, 3)):
    molecule = copy(water)
    molecule.translateTo(point)
    universe.addObject(molecule)


which produces the following error message:

Traceback (innermost last):
  File "ice.py", line 27, in ?
    molecule = copy(water)
  File "/usr/lib/python1.5/copy.py", line 76, in copy
    y = copierfunction(x)
  File "/usr/lib/python1.5/copy.py", line 110, in _copy_inst
    return x.__copy__()
  File "/usr/lib/python1.5/site-packages/MMTK/ChemicalObjects.py", line 116, in __copy__
    return copy.deepcopy(self, {id(self.parent): None})
  File "/usr/lib/python1.5/copy.py", line 151, in deepcopy
    y = copierfunction(x, memo)
  File "/usr/lib/python1.5/copy.py", line 236, in _deepcopy_inst
    state = deepcopy(state, memo)
  File "/usr/lib/python1.5/copy.py", line 151, in deepcopy
    y = copierfunction(x, memo)
  File "/usr/lib/python1.5/copy.py", line 199, in _deepcopy_dict
    y[deepcopy(key, memo)] = deepcopy(x[key], memo)
  File "/usr/lib/python1.5/copy.py", line 151, in deepcopy
    y = copierfunction(x, memo)
  File "/usr/lib/python1.5/copy.py", line 172, in _deepcopy_list
    y.append(deepcopy(a, memo))
  File "/usr/lib/python1.5/copy.py", line 151, in deepcopy
    y = copierfunction(x, memo)
  File "/usr/lib/python1.5/copy.py", line 236, in _deepcopy_inst
    state = deepcopy(state, memo)
  File "/usr/lib/python1.5/copy.py", line 151, in deepcopy
    y = copierfunction(x, memo)
  File "/usr/lib/python1.5/copy.py", line 199, in _deepcopy_dict
    y[deepcopy(key, memo)] = deepcopy(x[key], memo)
  File "/usr/lib/python1.5/copy.py", line 151, in deepcopy
    y = copierfunction(x, memo)
  File "/usr/lib/python1.5/copy.py", line 221, in _deepcopy_inst
    return x.__deepcopy__(memo)
TypeError: too many arguments; expected 1, got 2

<duh>Huh? Can somebody comment what it means?</duh>



From bwarsaw at cnri.reston.va.us  Fri Jun  4 19:07:01 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Fri, 4 Jun 1999 19:07:01 -0400 (EDT)
Subject: [ANNOUNCE] JPython 1.1 beta 1 released
Message-ID: <14168.23573.4173.886534@anthem.cnri.reston.va.us>

I'm happy to announce the release of JPython 1.1beta1.  JPython is
the 100% Pure Java implementation of the Python language, which
provides seamless integration with the Java platform.  JPython
1.1beta1 is compatible with most Java 1.1 and 1.2 JDKs.

JPython 1.1beta1 is much improved over the previous public release of
version 1.0.3.  Many bugs have been fixed, performance has been
improved, and most new CPython 1.5.2 features have been implemented.
See below for highlights of the changes between 1.1beta1 and 1.0.3.

You can download JPython 1.1beta1 from the JPython home page.  Please
visit

    http://www.jpython.org/

Much thanks go to Jim Hugunin -- JPython's inventor, Finn Bock -- for
his contribution of modules and test driving, Guido, and the rest of
the folks at CNRI.

Here's a summary of what's changed since 1.0.3:

    - New and improved jpythonc tool by Jim Hugunin
    - New modules contributed by Finn Bock: cStringIO, cPickle,
      struct, binascii
    - New module synchronize, equivalent to Java's synchronized
      keyword
    - Operator module now complete
    - sys.add_package() no longer necessary, "import java.lang" just works
    - Class-based standard exceptions (w/ support for -X option)
    - super__ methods are deprecated, use regular Python style
    - time.strftime() added
    - Java 1.2 style collections support
    - Multiple interpreter states
    - Most CPython 1.5.2 features have been implemented
    - String objects now have methods -- expected CPython 1.6 feature
    - Improved performance
    - Java inner class support
    - keywords are allowed in some locations (e.g. as attribute names)
    - New license
    - bug fixes

Enjoy,
-Barry

Barry A. Warsaw                    Corp. for Nat'l Research Initiatives (CNRI)
bwarsaw at cnri.reston.va.us          1895 Preston White Drive, Suite 100
bwarsaw at python.org                 Reston, VA USA 20191-5434
http://www.python.org/~bwarsaw     703.620.8990 / (fax) 703.620.0913

P.S. contrary to Python tradition, I will /not/ be going on vacation
tomorrow :)



From amackay at starvision.com  Wed Jun 30 20:06:38 1999
From: amackay at starvision.com (Angus MacKay)
Date: Wed, 30 Jun 1999 17:06:38 -0700
Subject: ioctl in python
References: <377A71D7.470AF9DA@starvision.com> <377A7374.C6919FD3@starvision.com> <377AA64A.B4C63805@bigpond.NO.SPAM.com>
Message-ID: <377AB10E.B0B85931@starvision.com>

hun?

why is this expecting a 4 byte buffer? my C version uses an int*
and it works no problem.

as for 80 being a input I think you have it backwards (for linux-i386 at least):
0x80046601  EXT2_IOC_GETFLAGS      int *
0x40046602  EXT2_IOC_SETFLAGS      const int *
0x80047601  EXT2_IOC_GETVERSION         int *
0x40047602  EXT2_IOC_SETVERSION         const int *

cheers, Angus.

Perry Faulkner wrote:
> 
> Hi Angus,
> 
> The ioctl cmd code (0x80047601) is expecting a 4 byte buffer to be
> passed to it. The
> simplest thing is something like :
> 
>     rec = pack('BBBB', 0,0,0,0)
> or
>     rec = pack('L', 0)
> 
> depending on what's expected, then
> 
>     ioctl(handle, 0x80047601, rec)
> 
> By the way, the 80 part of the command also says this is an input
> function, so you
> are writing the data, rec in this case! For an output function you would
> follow the
> ioctl with an unpack of the data buffer, as well. But you still need to
> supply the
> packed input buffer in any case.
> 
> Hope this helps!
> 
> Regards,
>     Perry
> 
> Angus MacKay wrote:
> 
> > well it seems I have sort of solved my problem. if I use a
> > string for the 3rd arg then it works properly:
> > >>> fo=open('/tmp')
> > >>> fcntl.ioctl(fo.fileno(), 0x80047601, "hi")
> > '\001\000'
> > >>> fo=open('/')
> > >>> fcntl.ioctl(fo.fileno(), 0x80047601, "hi")
> > '\000\000'
> > >>> fo=open('/proc')
> > >>> fcntl.ioctl(fo.fileno(), 0x80047601, "hi")
> > Traceback (innermost last):
> >   File "<stdin>", line 1, in ?
> > IOError: (25, 'Inappropriate ioctl for device')
> >
> > the question is why? it will still fail with only 2 args or an int:
> > >>> fcntl.ioctl(fo.fileno(), 0x80047601)
> > Traceback (innermost last):
> >   File "<stdin>", line 1, in ?
> > IOError: (14, 'Bad address')
> > >>> fcntl.ioctl(fo.fileno(), 0x80047601, 1)
> > Traceback (innermost last):
> >   File "<stdin>", line 1, in ?
> > IOError: (14, 'Bad address')
> >
> > cheers, Angus.
> >
> > Angus MacKay wrote:
> > >
> > > is there some magic to the fcntl.ioctl() call?
> > >
> > > I can do these ioctl calls in C no problem (with a small wrapper):
> > > (amackay at phat)~/tmp$ ./ioctltest /tmp 0x80047601
> > > 0x80047601: 1
> > > (amackay at phat)~/tmp$ ./ioctltest / 0x80047601
> > > 0x80047601: 0
> > > (amackay at phat)~/tmp$ ./ioctltest /proc 0x80047601
> > > /proc: Inappropriate ioctl for device
> > >
> > > that was the Linux ioctl for EXT2 version.
> > >
> > > but in python:
> > > >>> a = 0
> > > >>> req = 0x80047601
> > > >>> import fcntl
> > > >>> fo=open('/tmp')
> > > >>> fcntl.ioctl(fo.fileno(), req, a)
> > > Traceback (innermost last):
> > >   File "<stdin>", line 1, in ?
> > > IOError: (14, 'Bad address')
> > > >>>



From pduffin at mailserver.hursley.ibm.com  Tue Jun 29 05:02:49 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Tue, 29 Jun 1999 10:02:49 +0100
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <37682DD3.43923243@compaq.com> <7ka2be$qja$1@nntp6.u.washington.edu> <37776FFA.2D85@mailserver.hursley.ibm.com> <lw7looxx7m.fsf@copernico.parades.rm.cnr.it> <876748mcxg.fsf@home.ivm.de>
Message-ID: <37788BB9.ABD@mailserver.hursley.ibm.com>

Klaus Schilling wrote:
> 
> Marco Antoniotti <marcoxa at copernico.parades.rm.cnr.it> writes:
> 
> > Paul Duffin <pduffin at mailserver.hursley.ibm.com> writes:
> >
> >       ...
> >
> > > Correct. With Tcl you can create new control structures which
> > > are indistinguishable from the built in ones, you cannot do
> > > that in Python. Python is much more rigid in its syntax although
> > > it does have a lot of nice hooks to allow objects to behave
> > > in different ways.
> >
> > But you can do that much more easily and elegantly in the L-word
> > language :)
> 
> It can be done best in the Scheme, by means of the almighty
> call-with-current-continuation, the best of all control structures,
> and define-syntax on top of it.
> 

Both Tcl and Scheme can do it, "best" is subjective.

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From ivanlan at callware.com  Thu Jun  3 13:47:34 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Thu, 3 Jun 1999 17:47:34 GMT
Subject: Python 2.0
References: <1283872945-30734792@hypernet.com>
Message-ID: <3756BFB6.90A1B9D1@callware.com>

Hi All--

Gordon McMillan wrote:
> 
> Michael Hudson writes:
> 
> [re Py_TRACE_REFS]
> > I've never resorted to this approach, tending always to make cyclic
> > references go away be staring at them very, very hard.
> 
> Ay. This is my approach. WIth the added advantage that now I find I
> can hypnotize a brick, simply by starting at it.
> 
> that-one's-already-asleep-ly y'rs
> 
> - Gordon

Yes, but the question we really want to know the answer to is ``Who is
the brick?  The starer or the staree?''

<dialectic>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From alex at mindlube.com  Tue Jun 22 01:35:21 1999
From: alex at mindlube.com (Alex Rice)
Date: 21 Jun 1999 23:35:21 -0600
Subject: newbie idiom question
Message-ID: <m3emj4yezq.fsf@monsoon.swcp.com>

Something I keep getting tripped up about is that objects being
iterated in "for" statements cannot be modified.

I keep trying to do this:

      >>> eggs = [1,2,3]
      >>> for spam in eggs:
      ...	spam = 'cooked'
      ...
      >>> eggs
      [1, 2, 3]
      >>> 

The tutorial says this:

      If you need to modify the list you are iterating over, e.g., duplicate
      selected items, you must iterate over a copy. The slice notation makes
      this particularly convenient:

      >>> for x in a[:]: # make a slice copy of the entire list
      ...    if len(x) > 6: a.insert(0, x)
      ... 
      >>> a
      ['defenestrate', 'cat', 'window', 'defenestrate']

Understood, but what if you want to modify each element in a list?
What's the best way to do this in terms of speed and elegance? I guess
I just haven't seen a good example of this yet in Python

What I'm unfortunately used to is this, in Perl:

      @eggs = (1,2,3);
      foreach $spam (@eggs) {
	$spam = 'cooked'; 
      }
      print "@eggs";
      >>> cooked cooked cooked

Thanks,
Alex Rice



From thornley at visi.com  Wed Jun 16 17:27:52 1999
From: thornley at visi.com (David Thornley)
Date: Wed, 16 Jun 1999 21:27:52 GMT
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <rnbn1y2n5jk.fsf@nortelnetworks.com>
Message-ID: <sFU93.1567$kS.233449@ptah.visi.com>

In article <rnbn1y2n5jk.fsf at nortelnetworks.com>,
Chris Ebenezer  <chriseb at nortelnetworks.com> wrote:
>
>Fernando Mato Mira <matomira at iname.com> writes:
>: of folks end up using one or another GNU tool. As a side effect, some
>: newcomers to Tk that would have normally used tcl could end up going
>: the STk way (unfortunately, STk is not publicized enough). Things like
>: Kawa also increase the feel of security regarding code reusability.
>: And some of those people will end up wanting to write full-fledged
>
>Why use TK at all ?  Once you are using scheme anyway there are nicer
>widget toolkits to use (from a point of view of more "normal" looking
>toolkits that don't suffer from color allocation problems),  there are
>bindings for guile and gtk and guile and Qt. 
>
The attraction, to me, would be that Tk is a free, open-source
graphics environment that will run on Unix with X-Windows, the
Macintosh, and Microsoft Windows.  CLIM is commercial, Garnet last
I looked didn't run on Microsoft Windows, and it looks to me like
I should be able to get Macintosh Common Lisp communicating with
Tk through Tcl without *that* much work.

I know little of this Qt animal.  Is it free?  Is it open source?
Does it run on the Macintosh?  Is it easy to set up an interface with?

--
David H. Thornley                        | If you want my opinion, ask.
david at thornley.net                       | If you don't, flee.
http://www.thornley.net/~thornley/david/ | O-



From jam at newimage.com  Sun Jun 13 20:14:29 1999
From: jam at newimage.com (jam)
Date: Sun, 13 Jun 1999 20:14:29 -0400
Subject: Python as script language?
In-Reply-To: <37644DC7.DE422A3B@hotmail.com>; from Denis Lamarche on Sun, Jun 13, 1999 at 09:38:30PM +0000
References: <37644DC7.DE422A3B@hotmail.com>
Message-ID: <19990613201428.K723@toast.internal>

On Sun, Jun 13, 1999 at 09:38:30PM +0000, Denis Lamarche wrote:
> 
> How can I set up Python so that I can call the scripts from within an
> application (that I wrote) so that it will run and interface with the
> application (that I wrote)?  Is this possible?
> 
> Denis Lamarche
> 

yes, it is possible, and it's relatively simple. check the python.org
website (specifically, read through the extending and embedding manual) to
get a feel for what is required. if you are familiar with C and object
oriented programming concepts, you should have no trouble.. and if you do,
feel free to post questons to the mailing list.

regards,
J
-- 
|| visit gfd <http://quark.newimage.com/> 
|| psa member #293 <http://www.python.org/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>



From S.I.Reynolds at cs.bham.ac.uk  Tue Jun  1 17:42:38 1999
From: S.I.Reynolds at cs.bham.ac.uk (Stuart I Reynolds)
Date: Tue, 01 Jun 1999 22:42:38 +0100
Subject: Cannot store Sequences in Numeric arrays of Objects
Message-ID: <375453CE.3192@cs.bham.ac.uk>

Hi,

I don't seem to be able to store a whole range of different type in
Numeric's PyObject arrays:

>>> from Numeric import *
>>> a = array( [None,]*9 , PyObject)
>>> b = array( [None,]*9 , PyObject)
>>> a.shape = 3,3
>>> a
array([[None , None , None ],
       [None , None , None ],
       [None , None , None ]],'O')

>>> #Storing ints works fine
>>> a[0,0] = 123
>>> a
array([[123 , None , None ],
       [None , None , None ],
       [None , None , None ]],'O')

>>>#Cannot store an array
>>> a[0,0] = b
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ValueError: array too large for destination

In fact, trying to store any sequence causes the error:

  a[0,0] = []
  a[0,0] = (None, None)
  a[0,0] = UserList()

... whereas storing these works fine:

  a[0,0] = Pickler(file)
  a[0,0] = 'hello'       #Also a sequence?

What's going on?


Cheers,

Stuart



From too at pk.highway.ne.jp  Tue Jun 22 08:55:39 1999
From: too at pk.highway.ne.jp (Thooney Millennier)
Date: Tue, 22 Jun 1999 21:55:39 +0900
Subject: signal handler
Message-ID: <376F87CB.DC3802E6@pk.highway.ne.jp>

Hi, I want to use "signal()" in my program.

  import signal;
  def func(arg1,arg2):
      DO_SOMETHING_GOOD;
      print "OK!";
  signal.signal(SOME_SIGNAL,func);

It seems "func()" is not called 
when the program gets signal "SOME_SIGNAL".
Why?
Any help welcome.

(I have installed Python 1.5.2 on Linux 2.0.34 and
running.)

Thanks for reading.



From mvanier at bbb.caltech.edu  Thu Jun  3 17:51:58 1999
From: mvanier at bbb.caltech.edu (Michael Vanier)
Date: 03 Jun 1999 14:51:58 -0700
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <3756E3D9.E44D2B39@swcp.com> <3756E476.ACB50DF1@swcp.com>
Message-ID: <jl7lplrlz5.fsf@arathorn.bbb.caltech.edu>

Alex Rice <alrice at swcp.com> writes:

> Therefore python is more post-modern than perl.
> 
> :-)


Post-post-modern?  Ultra-post-modern?  Magna-cum-post-modern?

Attack-of-the-killer-buzzwords-ly y'rs,

Mike


-------------------------------------------------------------------------
Mike Vanier	mvanier at bbb.caltech.edu
Department of Computation and Neural Systems, Caltech 216-76
Will optimize nonlinear functions with complex parameter spaces for food.



From arcege at shore.net  Wed Jun 30 14:49:18 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 30 Jun 1999 18:49:18 GMT
Subject: socket timeouts
References: <XFMail.990630103816.nathan@islanddata.com>
Message-ID: <OEte3.383$6M6.125233@news.shore.net>

Nathan Clegg <nathan at islanddata.com> wrote:
: Is there any way to set the timeout for a TCP/IP connection? 
: Specifically, I'm writing some apps meant to run on the same machine, so
: if they can't connect *immediately* (i.e. the server app is not running), I
: need to go to plan B.  Any answers or alternate solutions?  I'm also
: interested in a similar issue across the internet, but I'm more generous
: about timeouts in this circumstance.  Thanks.

Since you are using a TCP/IP connection (ie. a socket), using select()
should be cross-platform enough.

  import select
  class TimeoutError(Exception):
    pass
  def wait_for_socket(sock, timeo):
    timeo = float(timeo)
    rdrs, wtrs, excs = select.select([sock], [], [], timeo)
    if not rdrs:
      raise TimeoutError, "no responce"

If the exception is raised, then the connection did not complete.

  -Arcege




From mark at _spamspamlovelyspam_btweng.krakow.pl  Sat Jun 19 05:37:57 1999
From: mark at _spamspamlovelyspam_btweng.krakow.pl (MK)
Date: Sat, 19 Jun 1999 09:37:57 GMT
Subject: Crawling Python
References: <000601beb91d$e1321ae0$839e2299@tim> <376a1ac1.10261180@news.tpnet.pl> <7ke6va$qa9$1@nnrp1.deja.com>
Message-ID: <376b5b32.6312672@news.tpnet.pl>

On Fri, 18 Jun 1999 19:34:03 GMT, Sam Schulenburg <sams at quinta.com> wrote:

>> >>> len(s)
>> 1178515

>I just tried both environments on a WindowsNT system and obtained
>130000 for len(s) under IDLE and Console mode python.

Win98. I tried the same several more times in DOS box and got different results
every time:

>>> for i in range(10000):
...     s=s+`range(i,i+20)`
...
>>>
>>>
>>> len(s)
1645360
>>>
>>>
>>> s=''
>>>
>>> for i in range(10000):
...     s=s+`range(i,i+20)`
...
>>> len(s)
1178515
>>>
>>> for i in range(10000):
...     s=s+`range(i,i+20)`
...
>>> len(s)
2357030
>>>

Now _that_ is weird.





MK


--------------------------------------------------
Reality is something that does not disappear after 
you cease believing in it - VALIS, Philip K. Dick
--------------------------------------------------

Delete _spamspamlovelyspam_ from address to email me

postmaster at 127.0.0.1
root at 127.0.0.1
webmaster at 127.0.0.1






From kiftdn at ffff.com  Sat Jun 12 02:51:56 1999
From: kiftdn at ffff.com (kiftdn at ffff.com)
Date: 12 Jun 1999 06:51:56 GMT
Subject: Free Sex Links  7457
Message-ID: <7jt02c$2pe$4640@nclient1-gui.server.virgin.net>

For the nastiest sex pictures visit:

http://freespace.virgin.net/efrwer.werewrd/




From Klaus.Schilling at home.ivm.de  Fri Jun 18 02:52:15 1999
From: Klaus.Schilling at home.ivm.de (Klaus Schilling)
Date: 18 Jun 1999 08:52:15 +0200
Subject: Problem with Curses
References: <ppgmj7.s7r.ln@solling.seismic.de> <jpeck7.1661.ln@solling.seismic.de>
Message-ID: <87aety0zmo.fsf@home.ivm.de>

kjwolf at online.de (Klaus-Juergen Wolf) writes:

> Klaus-Juergen Wolf <kjwolf at online.de> wrote:
>  > (Python 1.5.2 on FreeBSD 3.2-RELEASE (i386))
> 
>  > I found that (after having compiled Python with cursesmodule.c -

Andrich's curses support is more complete, particularly suited for
x/openpg compliant curses. It turns to be very useful to replace it.

>  > testall.py succeeds) the following program gives a segmentation
>  > fault:
> 
> As it appears,
> 
> -*-
> import curses
> 
> curses.initscr()
> curses.endwin()
> -*-
> 
> alone does exactly the same. I've tested it in debugging environment,
> where it doesn't happen, and I've checked the cursesmodule code, and I
> still have no idea why this happens. (Inconsistency? But where?)
> 
> Stack trace says that it happens when the first actual write call to
> libtermcap takes place.

So curses on that platform still needs to be linked to libtermcap?
> 
> The problem occurs under Solaris in the same way, while Linux doesn't
> seem to know anything of it.

Using libtermcap on those systems? Linux usually uses terminfo which is
already burried in the ncurses stuff. Solaris might use sun curses.

Btw why is there no direct termcap support for python? 
Perl has Term::Cap , which doesn't even require the termcap library,
just the termcap data file (e.g. /etc/termcap). 
Would it be very hard to provide? 

Klaus Schilling



From thomas.heller at ion-tof.com  Wed Jun 23 10:45:04 1999
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Wed, 23 Jun 1999 16:45:04 +0200
Subject: How to obtain an instance's name at runtime?
References: <7kqpu8$7u1$1@nnrp1.deja.com>
Message-ID: <003f01bebd86$fb5631f0$4500a8c0@thomasnotebook>

> I would like to do the following:
> 
> >>> class C:
> ...    pass
> ...
> >>>
> >>> c = C()
> >>> C.__class__.__name__ # ok
> 'C'
> >>> c.__name__           # not ok
> Traceback (innermost last):
>   File "<interactive input>", line 0, in ?
> AttributeError: __name__
> >>>

Huh?

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> class C:
...   pass
...
>>> c = C()
>>> c.__class__.__name__
'C'
>>> C.__name__
'C'
>>> c.__name__
Traceback (innermost last):
  File "<stdin>", line 1, in ?
AttributeError: __name__
>>> C.__name__
'C'
>>>
Use c.__class__.__name__!

Thomas Heller
ION-TOF GmbH




From billtut at microsoft.com  Fri Jun 18 11:03:22 1999
From: billtut at microsoft.com (Bill Tutt)
Date: Fri, 18 Jun 1999 15:03:22 GMT
Subject: do anonymous pipes normally work on NT?
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8100DB90DEF@RED-MSG-50>

> From: Georg Mischler [mailto:schorsch at schorsch.com]

> Do I need to swig this stuff myself? or...

Nope, they already come pre-swigged.
Download, load into VC++, tweak away so that it can find all of the
necessary .h/.libs it needs and build away.

Bill



From kjwolf at online.de  Sun Jun 20 20:54:23 1999
From: kjwolf at online.de (Klaus-Juergen Wolf)
Date: Mon, 21 Jun 1999 02:54:23 +0200
Subject: Problem with Curses
References: <ppgmj7.s7r.ln@solling.seismic.de> <jpeck7.1661.ln@solling.seismic.de> <87aety0zmo.fsf@home.ivm.de>
Message-ID: <vf2kk7.gj.ln@solling.seismic.de>

Klaus Schilling <Klaus.Schilling at home.ivm.de> wrote:
 >> Klaus-Juergen Wolf <kjwolf at online.de> wrote:
 >>  > (Python 1.5.2 on FreeBSD 3.2-RELEASE (i386))
 >> 
 >>  > I found that (after having compiled Python with cursesmodule.c -

 > Andrich's curses support is more complete, particularly suited for
 > x/openpg compliant curses. It turns to be very useful to replace it.

There are more Curses' and NCurses' in the world than I have fingers
and toes. You can easily write a Curses module that makes _you_ happy.
And the rest of the world... Why, do you think, do Oliver Andrich's
approaches have a 'b' in the version number? It's simple: they won't
work everywhere, here, for instance.

cu
 k.j.

-- 
 Support Julia Hill - save the Redwood trees: http://www.lunatree.org/



From hniksic at srce.hr  Mon Jun  7 17:11:50 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 07 Jun 1999 23:11:50 +0200
Subject: waste of resources ?
References: <375C0F91.1C73E4E3@icrf.icnet.uk> <375C16A9.5ADC4990@icrf.icnet.uk> <375C2003.B01BEC3@arlut.utexas.edu> <87yahvzqz3.fsf@pc-hrvoje.srce.hr> <ytt1zfnen4h.fsf@gilgamesh.cse.ucsc.edu>
Message-ID: <87pv37zpex.fsf@pc-hrvoje.srce.hr>

Ben Gertzfield <che at debian.org> writes:

> If I recall correctly, Perl has just about the exact same problem.
> Is there a good way to fix it?

I don't know about Perl, but a good fix would be to have the
interpreter do the hard work for us.  If a system call fails and errno 
is EINTR, then the signal handler should be called, *and* the syscall
should be restarted.



From dkuhlman at netcom.com  Fri Jun  4 18:59:06 1999
From: dkuhlman at netcom.com (G. David Kuhlman)
Date: Fri, 4 Jun 1999 22:59:06 GMT
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net> <aahzFCpHw8.Gxt@netcom.com> <7j4e2u$8rh$1@m2.c2.telstra-mm.net.au> <37564132.2D01B7C3@pop.vet.uu.nl> <3756BAFC.2066A636@home.com> <3756F071.E717D0FA@pop.vet.uu.nl> <37573C16.D0233810@home.com> <etdr9nrx39r.fsf@m2-225-11.mit.edu>
Message-ID: <dkuhlmanFCtruI.Ms9@netcom.com>

Look at:

    http://www.python.org/doc/essays/everybody.html

  - Dave


Alex (alex at somewhere.round.here) wrote:

> > I'm rambling about Python these days. Yesterday I read a paper from
> > Guido about how he was thinking about bringing Python (or some
> > successor language) to *billions* of users. :)

> Could someone point me at this paper?

> Alex.



From wlfraed at ix.netcom.com  Wed Jun  9 22:43:13 1999
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Thu, 10 Jun 1999 02:43:13 GMT
Subject: "Programming Python" still worthwhile?
References: <1283218560-70098357@hypernet.com> <375E9943.F9FEA7BA@gssec.bt.co.uk>
Message-ID: <375f23fe.1133449@nntp.ix.netcom.com>

On Wed, 09 Jun 1999 17:41:39 +0100, Alan Gauld
<alan.gauld at gssec.bt.co.uk> declaimed the following in comp.lang.python:


> At work we have Python 1.3 (don't ask!) and at home I use 

	Some system admin bought the book, and installed Python from the
CD, then forgot about it? <G>

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <



From hank at prosysplus.com  Mon Jun 28 23:34:14 1999
From: hank at prosysplus.com (Hank Fay)
Date: Mon, 28 Jun 1999 23:34:14 -0400
Subject: ActiveX, IDE
References: <7cKd3.3449$lr3.133866@news2.jacksonville.net> <7l96c1$5hn$1@m2.c2.telstra-mm.net.au>
Message-ID: <f8Xd3.3539$lr3.140731@news2.jacksonville.net>

Mark,

actually, I want a command window just like pythonwin 1.52 or whatever,
which I already have installed; but also want to be able to visually design
forms, including drag-n-drop, visually size, etc., activex controls, also as
I currently do.  FWIW: I think the VB IDE is a straightjacket.

thanks,

Hank

--
http://www.prosysplus.com
Mark Hammond <MHammond at skippinet.com.au> wrote in message
news:7l96c1$5hn$1 at m2.c2.telstra-mm.net.au...
> Hank Fay wrote in message
<7cKd3.3449$lr3.133866 at news2.jacksonville.net>...
> >I'd like to make Wintel apps using ActiveX controls under Python: is
there
> >an IDE for Python that will support this?
>
> Pythonwin supports hosting ActiveX controls, but if you are expecting an
> ActiveX IDE similar to VB, you will be dissapointed.
>
> http://starship.python.net/crew/mhammond
>
> Mark.
>
>
>





From anders777 at my-deja.com  Wed Jun 30 12:30:04 1999
From: anders777 at my-deja.com (anders777 at my-deja.com)
Date: Wed, 30 Jun 1999 17:30:04 +0100
Subject: Reflection in Python?
Message-ID: <000015c0@bossar.com.pl>

From: anders777 at my-deja.com

I've just started programming in Python, and I
could really use some help figuring out something.
 I'd like to implement Kent Beck's unit tester in
Python (someone posted a very nice alternative a
few weeks ago, but Junit's approach works better
for my project's needs).  How do I do the
equivalent of what Java calls reflection? I want
to do the following:

test = MathTest("testAdd")

where test.run() will run the "testAdd" method on
the MathTest object?  I also want to be able to
tell the user the name of the object (MathTest)
and the method in case the method fails.  Any
thoughts?

Thanks,
Anders Schneiderman
National Journal Daily Briefings Group




Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From zorro at zipzap.ch  Thu Jun 24 13:01:12 1999
From: zorro at zipzap.ch (Boris Borcic)
Date: Thu, 24 Jun 1999 19:01:12 +0200
Subject: Thanks for Python...
References: <376ead64@anonymous.newsfeeds.com>
		<VrDQRddcbbs9-pn2-QdftOXYk6Hdd@Fortress>
		<37716901.43CB604@lockstar.com>
		<slrn7n2gkr.i3k.wtanksle@dolphin.openprojects.net>
		<7ktfn0$5q$1@aurwww.aur.alcatel.com> <14194.18324.582362.94739@amarok.cnri.reston.va.us>
Message-ID: <37726458.C8CC9623@zipzap.ch>

Andrew M. Kuchling wrote:

> Once you accept that the world is a giant computer run by white mice, all
> other movies fade into insignificance.
>     -- Mutsumi Takahashi, on _The Hitch-hiker's Guide to the Galaxy_

Should this be taken as a rebuttal of "Matrix" ?

BB



From larsga at ifi.uio.no  Fri Jun 18 12:56:31 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 18 Jun 1999 18:56:31 +0200
Subject: Why don't my global variables work?
References: <ttbd7yvho9h.fsf@talon.i-did-not-set--mail-host-address--so-shoot-me>
Message-ID: <wkyahhv45c.fsf@ifi.uio.no>

* Nathan J. Froyd
| 
| It's a program for turning a netscape bookmarks file into something
| resembling XML.

Hmmm. Maybe you'd like to have a look at this:

<URL: http://www.python.org/topics/xml/xbel/>

:-)

--Lars M.



From hniksic at srce.hr  Fri Jun  4 16:24:04 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 04 Jun 1999 22:24:04 +0200
Subject: threads
References: <1283593620-47537497@hypernet.com>
Message-ID: <87emjr67ff.fsf@pc-hrvoje.srce.hr>

"Gordon McMillan" <gmcm at hypernet.com> writes:

> Hrvoje Niksic writes:
> 
> > "Barry A. Warsaw" <bwarsaw at cnri.reston.va.us> writes:
> > 
> > > You might be interesting in digging around for Greg Stein's
> > > free-threading patches.  Greg did a lot of work, I think in the
> > > Python 1.4 days, to remove the central interpreter lock and make all
> > > the data structures and such thread safe.
> > 
> > Why would I want to do that? 
> 
> See that other thread.

Huh?  What other thread?  Maybe the question wasn't clear enough.

What I mean is: at this time, my need for simultaneous thread
execution in Python is not so urgent that to warrant forking off (no
pun intended).  Once I do need it, I'd prefer to be able to choose
Python than to dismiss Python on the grounds of not properly
supporting threading.

> > If Greg's patches haven't been applied to the base, surely mine
> > would not be either.
> 
> As I recall, Greg's patches didn't make it because (1) they were late 
> and (2) they were slow.

Interesting.  What was the source of slowness?  (If you _remove_
locks, I'd expect things to go faster, even in a single-threaded
world.)  Maybe the patches simply introduced a different set of
problems?



From janssen at parc.xerox.com  Wed Jun  2 10:04:59 1999
From: janssen at parc.xerox.com (Bill Janssen)
Date: Wed, 2 Jun 1999 14:04:59 GMT
Subject: Inter-process communication
In-Reply-To: <928118414snz@vision25.demon.co.uk>
References: <928118414snz@vision25.demon.co.uk>
Message-ID: <ArJHc=4B0KGW9RtUJm@holmes.parc.xerox.com>

A typical way to do it is to give one or both of the programs ILU
interfaces, and use ILU to do the IPC.  Information on ILU is available
at ftp://ftp.parc.xerox.com/pub/ilu/ilu.html; a Python tutorial for ILU
is at ftp://ftp.parc.xerox.com/pub/ilu/misc/tutpython.html.

Bill



From rdr_nl at my-deja.com  Tue Jun 15 06:23:03 1999
From: rdr_nl at my-deja.com (Henk Jansen)
Date: Tue, 15 Jun 1999 10:23:03 GMT
Subject: extension called from embedded code
Message-ID: <7k59i2$ijd$1@nnrp1.deja.com>

Trying to get the most out off the C++/Python team I run into problems
when I try to load extension modules (written in C, like kjbuckets or
ranlib) from Python code that is embedded in a C++ application. The API
I use is 'PyRun_SimpleString("import kjbuckets\n");' and causes
unresolved symbol error messages from the Python interpreter (e.g.
PyType_Type for kjbucketsmodule.so and PyExc_RuntimeError for
ranlib.so). Loading these modules when Python is in control goes fine.
Has anyone any idea how to do this?

Thanks,

Henk
--
 =====================================================================
 Henk Jansen -- Delft University of Technology -- Dept. of Mathematics
 hjansen at math.tudelft.nl
 http://ta.twi.tudelft.nl/users/hjansen/H.Jansen.html


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From behrends at cse.msu.edu  Wed Jun 23 01:22:49 1999
From: behrends at cse.msu.edu (Reimer Behrends)
Date: 23 Jun 1999 05:22:49 GMT
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org>
Message-ID: <slrn7n0rr2.2fi.behrends@allegro.cse.msu.edu>

Preston Landers (planders at mail.utexas.edu) wrote:
> Can someone *please* give me a good reason why Python doesn't yet have 
> +=, -=, and .= and other such operators?  I know it's just syntactic
> sugar, but damn, it is such sweet sugar!

Few people, I think, have ever tried to objectively count how often such
a construct is used. The Python 1.5.2 source, for instance, has
approximately one of ++, --, +=, -=, *=, /= every 40 lines. If you do
not count the occurrences if ++ and -- in for loops, for which there's a
construct in Python anyway, you can roughly double that number. Now, C
is not Python, but it's a good first-order approximation of how often
you need it--the Python source is one of the cleanest C programs I know
of.

Personally, I can easily live without a construct that I can use on
average once every three screens of code or so. On the other hand, I
_am_ getting pretty sick of having to type self.whatever countless
times, where pretty much any other language allows me to discard the
"self." part. Not only is it annoying to type, it also reduces the
readability of OO code a lot, much like the gratuitious use of
punctuation characters as variable prefixes in Perl does.

Among all the suggestions for Python 2.0, removing the necessity of
prefixing methods and instance variables with "self." (and not having
to list it as an argument, either) would be my #1 wish.

[...]

				Reimer Behrends



From fredrik at pythonware.com  Tue Jun 22 09:39:27 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 22 Jun 1999 13:39:27 GMT
Subject: htmllib: CR in CDATA
References: <00d201bebc5a$adb129a0$0301a8c0@mnot.net> <012601bebc91$f3ded760$0301a8c0@mnot.net> <014401bebc9a$1963a9e0$0301a8c0@mnot.net> <000b01bebca0$8a022c20$f29b12c2@pythonware.com> <017501bebcad$830a3180$0301a8c0@mnot.net>
Message-ID: <003701bebcb4$a6d33fb0$f29b12c2@pythonware.com>

> Now, htmllib certainly:
> * is a component
> * part of a distributed system (i.e., the Web)
> * presents an interface (programmatic)
> * processes requests on behalf of a user

> I'm curious... if it's not a user agent in the quoted context, what is it?

I'm pretty sure you know what I meant, but
alright...

htmllib is a parser, just like the documentation says.
you have to add an application to get an HTML user
agent (see section 1.2.3 of the 2.0 spec for more
info on user agents).

imho, it's pretty reasonable for an SGML parser to
behave like an XML parser: split the document up
into pieces, but pass them all to the application as
untouched as possible. if you wish to implement
additional behaviour, do that on the application
level. otherwise, you'll end up in a situation where
some users cannot use the standard library...

(like I did only a few hours ago, trying to use sgmllib
to parse SGML data with case-sensitive tags. sigh...)

</F>




From greg.ewing at compaq.com  Thu Jun 17 19:00:13 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 18 Jun 1999 11:00:13 +1200
Subject: Newbie: Truth values (three-valued logic)
References: <000f01beb7b8$2722a540$2ca22299@tim> <Pine.OSF.4.05.9906161106470.295-100000@pinax.Mathematik.Uni-Bielefeld.DE>
Message-ID: <37697DFD.91C2B8F6@compaq.com>

Olaf Delgado wrote:
> 
> Next time I'll want to redefine 'and' and 'or'. 

You could use '*' and '+', if you always know
when you're dealing with a 3-valued boolean
(troolean? trillian?).

By the way, I think Python 2.0 should let you define
new infix operators -- and design new characters to
represent them! We all have bitmapped screens now --
there's no reason why we should be restricted to a
fixed character set any more...

Greg



From xcvx at sdofi.com  Mon Jun 21 19:06:58 1999
From: xcvx at sdofi.com (ATS Public Labs)
Date: Mon, 21 Jun 1999 19:06:58 -0400
Subject: More Problems With Curses
Message-ID: <376EC592.A97D117D@sdofi.com>

Hello,

I'm trying to write an IRC client in Python using the curses and
threading modules.  Basically one thread will read user input while the
other threads print the messages received from the servers the user is
connected to.   This works fine if I use print() and input() instead of
curses, but it's annoying because the other threads will print stuff to
the screen in the middle of what you are typing.  However, when I try to
use curses instead, only the user-input thread seems to have control of
the screen; the other threads will hang and not print anything until you
press return.

I wrote the following to try to figure out what's going on:


import time
import threading
import curses

def repd():
  x=0
  while 1:
    time.sleep(1)
    print 'x=%d' % x
    x=x+1

thr=threading.Thread(target=repd)
thr.start()
while 1:
  time.sleep(1)
  str=raw_input()
  print 'got %s' % str


This prints out 'x=0', 'x=1' etc. at one second intervals, and in
addition prints whatever you type as soon as you press return.  However,
if I add the following two lines:

stdscr=curses.initscr()
curses.endwin()

right before I start the thread, then what happens is that 'x=0' gets
printed, and then nothing will get printed again until you type
something and press return, at which point all the 'x=n' messages which
you should have seen get printed all at once.  So, just the act of
initializing curses causes this annoying blocking behavior which I don't
want.

So, what am I doing wrong?  How can I get curses and threading to play
nice with each other?

now-I-know-why-it's-called-"curses"-ly yrs, Cliff





From pduffin at mailserver.hursley.ibm.com  Thu Jun 10 11:41:27 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Thu, 10 Jun 1999 16:41:27 +0100
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com>
Message-ID: <375FDCA7.62FF@mailserver.hursley.ibm.com>

John Mitchell wrote:
> 
> On Thu, 10 Jun 1999, Gordon McMillan wrote:
> 
> > Fernando Mato Mira writes:
> > > I'm trying to collect a list of all the _current_ issues making tcl
> > > a bad language choice.
> >
> > Why? Convincing your boss to use Python over TCL for your next
> > project is one thing, but in general this is a zero-sum (or worse)
> > game.
> >
> > The TCLers I know value its simplicity and accept its limitations.
> > Telling them they're wrong will only piss them off.
> >
> > - Gordon
> 
> For what it's worth, I've tried convincing a friend at AOL to start using
> Python.  They mostly use TCL with AOLServer[1] (webserver with TCL built
> in at a very low level, quite fast); with a bit of Perl.
> 

I was going to say "Sensible folk" until I read the Perl bit ;-).

> He's not a trained programmer, and really likes TCL.   Since they've more
> or less standardized on *one* language across many developers, they have a
> large library of code to steal from, which makes nearly all new projects
> trivial to whip up.  Switching to a "real" (OO, scripting) language really
> wouldnt help them out.
> 

"real", "REAL" !!!!

> Ironically, TCL *does* have some object-oriented features.  When I tried
> it out many years ago, there was a [itcl] extension to basically add
> C++-style object suppport.  But, all the TCL types I've talked to since
> have never heard of it.  Odd.
> 

Tcl has many OO features, you can have C++ like, Java like, SmallTalk
like ... it is up to you. A lot of them are implemented in pure Tcl.

> Has anyone messed with Guile, GNU's extension language?  One specific
> thing they want to do is be a "universal" back end, that is you can write
> translators from language A to Guile, and their page[2] mentions Python
> several times...
> 

Nice idea but I doubt they will even manage one language completely,
never mind 2, or 3, or 4 ...

> 
> [1] http://www.aolserver.com/
> 
> [2] http://www.gnu.org/software/guile/guile.html
> 

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From cfelling at iae.nl  Thu Jun 24 15:41:25 1999
From: cfelling at iae.nl (Carel Fellinger)
Date: 24 Jun 1999 21:41:25 +0200
Subject: tkMessageBox returns 'can't invoke "frame" command:' error
References: <7kktvv$550$1@nnrp1.deja.com>
Message-ID: <7ku1l5$13q$1@vvs.superst.iae.nl>

Hai Raymond,

though I'm new to python and Tkinter myself, I think I know what the
problem is. I tried it out at my site and it seems to work. And if
there is any truth in what I am going to tell (please you higher
gods, correct me if i'm wrong) I have finally fathomed the difference
between quit and destroy whilst figuring out your bug:).

The point is that you repeatedly try to initialise Tkinter without
properly finishing up prior uses. You correctly call widget.quit(), but
this doesn't destroy the widget, it merely exits the widget.mainloop().
This can easily be seen: after finishing the first import of your code
in the interactive interpretor by pressing the 'quit' button, you will
notice that the window doesn't disappear. Though this window is kind of
dead in not reacting to clicking it is still there. And so is your
invocation of Tkinter. If at this point you reload your script you might
notice that the default window label now reads "Tk #2" to remind you of
the fact that you now have a second invocation of Tkinter running at the
same time.

The way out is simple, you have to explicitly destroy your widgets.
a simple "rootWindow.destroy()" after the call to mainloop will do.


As for your error message complaining about the frame command not working
because the application has been destroyed, that's a little trickier.
Just where should I start...

First, Tkinter keeps track of the current Tk invocation using the var
Tkinter._default_root, initially this equals None, and upon invocing
Tk() it gets initialised. Unfortunately Tkinter doesn't reset this
var to None upon termination of Tk through an appropriate destroy()!
(maybe this should be considered a bug? who should know about it?)

Second, many (all?) widget commands have None as default for the master
keyword-arg. And if you don't specify an explicit master for the widget
the before mentioned Tkinter._default_root will be used. But elas,
the second time around (reloading your script) this still refers to the
now obsolete Tk instantiation. Hence the Tcl error: that Tk doesn't
exist anymore.

There are two ways to remedy this:
1) add something like "master=rootWindow" to constructor calls of MessageBox
   widgets.
2) do what Tkinter should have done, i.e. "Tkinter._default_root==None"
   but only do this after you have destroyed the root window with something
   like "rootWindow.destroy()", the latter is done automatically upon
   pressing the root-window-close-button, but has to be programmed otherwise.

Hope this helps, and sure hope it's correct:}

-- 
groetjes, carel



From nikolai.kirsebom at siemens.no  Tue Jun 15 09:36:49 1999
From: nikolai.kirsebom at siemens.no (Nikolai Kirsebom)
Date: Tue, 15 Jun 1999 13:36:49 GMT
Subject: TAPI and Pythonwin
Message-ID: <376656be.101659969@news.mch.sni.de>

Does anyone know if there is a module for the TAPI (telephone API) on
NT ?

Nikolai




From tim_one at email.msn.com  Tue Jun 29 04:12:07 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 29 Jun 1999 04:12:07 -0400
Subject: ANN: Stackless Python 0.2
In-Reply-To: <7l8fut$es6@dfw-ixnews6.ix.netcom.com>
Message-ID: <000801bec207$142b6360$719e2299@tim>

[G. David Kuhlman]
> I'll bite. How *does* a thread that calls out to C release the global
> lock?

This is covered in detail in the Python/C API manual (look in your Python
doc distribution), chapter 8.  You typically just use a pair of
Python-supplied bracket macros:

... here you own the global lock

Py_BEGIN_ALLOW_THREADS

... now you do not:  other Python threads can run, and you can do
... C stuff here as long as you like, in parallel with them

Py_END_ALLOW_THREADS

... now you own the global lock again, and can return to Python

Many examples of these macros can be found in the distribution's C source
code too.

> Would it have to do this thingy that is at the beginning of the
> interpreter loop in python/ceval.c?
>
>     if (PyThreadState_Swap(NULL) != tstate)
>         Py_FatalError("ceval: tstate mix-up");
>     PyThread_release_lock(interpreter_lock);
>
> Would this work? or would it screw up the interpreter?

It won't even compile.  interpreter_lock is not an extern symbol:  you can't
get at it directly.  That's a feature, of course, to discourage chuckleheads
from trying to subvert the published API <wink>.

> Is there a way to do this in a Python script *before* it calls out
> to C.

No:  Python-level code must never run unless the thread running it holds the
global lock.  Therefore no Python-level code can ever release the global
lock (the instant it did so, it would be illegal code -- so no possibility
for screwing up this way is provided).

This is no real restriction, though.  To run external C code at all you have
to put it in a module known to Python (so that Python can resolve the name
*from* Python), and in the typical case where you're calling some
pre-existing thread-safe C function ThreadedSpam, you're not going to stick
ThreadedSpam in a Python module anyway.  Instead you'd write your Python C
module with a trivial wrapper function and tell Python about *that*; its
guts will consist of

Py_BEGIN_ALLOW_THREADS
ThreadedSpam();
Py_END_ALLOW_THREADS

it's-only-confusing-if-you-think-about-it-too-much<wink>-ly y'rs  - tim





From gmcm at hypernet.com  Wed Jun 16 12:25:57 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 16 Jun 1999 11:25:57 -0500
Subject: GNU python?
In-Reply-To: <022c01beb7ec$f9617070$f29b12c2@pythonware.com>
Message-ID: <1282571880-108998174@hypernet.com>

/F wrote:

> Skip Montanaro <skip at mojam.com> wrote:
> > > Looking to try out python... is there a GNU python for Win9x?  and a
> > > tutorial maybe? 
> > 
> > Not sure just how a GNU Python would differ from the real thing.
> 
> a bit more like GNUscript, I suppose:
> 
> http://art.net/Studios/Hackers/Hopkins/Don/text/gnuscript.html

There goes pythonware. Eaten by the TGPLV.

- Gordon



From hniksic at srce.hr  Thu Jun  3 12:09:47 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 03 Jun 1999 18:09:47 +0200
Subject: Use of StringIO vs cStringIO in standard modules
In-Reply-To: Guido van Rossum's message of "Thu, 03 Jun 1999 10:25:03 -0400"
References: <8790a1620b.fsf@pc-hrvoje.srce.hr> <199906031425.KAA06935@eric.cnri.reston.va.us>
Message-ID: <87aeuh5kqc.fsf@pc-hrvoje.srce.hr>

Guido van Rossum <guido at CNRI.Reston.VA.US> writes:

> Hrvoje Niksic <hniksic at srce.hr>:
> 
> > I noticed that many standard modules use StringIO and not
> > cStringIO, although they don't need subclassing.  Is this
> > intentional?
> > 
> > For example, base64.py uses StringIO to implement encodestring()
> > and decodestring().  Since both functions write to output line by
> > line, I imagine the performance hit of StringIO vs cStringIO might
> > be non-negligible.
> 
> Have you noticed any speed difference?

Yes, quite a bit.  Trivially replacing StringIO with cStringIO in
base64.py makes encoding 2.3 times and decoding 3.6 times faster.
That's on my system (Ultra 2 under Solaris 2.6), measured repeatedly
with time.clock() and an approx. 1M sample string.  I can post the
script if there is interest.

Maybe the correct solution for base64.py would be to use something
like this at top-level:

try:
    from cStringIO import StringIO
except:
    from StringIO import StringIO

> cPickle, because calling it from C is much faster than calling
> StringIO from C; however I believe that for calls from Python,
> StringIO isn't that much slower.

I've looked at the code, and to me it seems that the slowness comes
from creating new strings on each write, where cStringIO just resizes
its internal buffer and creates the string only at the end.

> > Furthermore, is there a particular reason for maintaining two
> > parallel StringIO implementations?  If subclassing is the reason,
> > I assume it would be trivial to rewrite StringIO to encapsulate
> > cStringIO the same way that UserDict encapsulates dictionary
> > objects.
> 
> That's not the reason; it's got more to do with not requiring a C
> extension where plain Python code will do.  Also to have a reference
> implementation.

But then you have to maintain both, *and* you get much slower code.
Is it worth it?



From mwh21 at cam.ac.uk  Thu Jun  3 02:21:00 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 03 Jun 1999 07:21:00 +0100
Subject: pickle vs .pyc
References: <7j43pn$55l$1@mlv.mit.edu> <m3pv3eian1.fsf@atrus.jesus.cam.ac.uk> <87vhd69tdm.fsf@pc-hrvoje.srce.hr>
Message-ID: <m3lne1iz3n.fsf@atrus.jesus.cam.ac.uk>

Hrvoje Niksic <hniksic at srce.hr> writes:

> Michael Hudson <mwh21 at cam.ac.uk> writes:
> 
> > Random aside: something fishy's going on when I try to try to marshal
> > *arrays* (as opposed to mere lists):
> 
> The documentation for marshal says:
> 
>     Not all Python object types are supported; in general, only
>     objects whose value is independent from a particular invocation of
>     Python can be written and read by this module. The following types
>     are supported: `None', integers, long integers, floating point
>     numbers, strings, tuples, lists, dictionaries, and code objects
>     (...)

Yes, but usually you get an exception to show that somethings gone
awry:

>>> import md5
>>> marshal.loads(marshal.dumps(md5.md5()))
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ValueError: unmarshallable object
>>> class F:
...  pass
...
>>> marshal.loads(marshal.dumps(F()))
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ValueError: unmarshallable object

It's not that important I suppose, it's just odd.

Michael



From htrd90 at zepler.org  Wed Jun  2 16:23:54 1999
From: htrd90 at zepler.org (Toby Dickenson)
Date: Wed, 02 Jun 1999 20:23:54 GMT
Subject: need help checking for stack overflow in Win32
References: <5lk8to3su3.fsf@eric.cnri.reston.va.us> <7j1pcc$256$1@m2.c2.telstra-mm.net.au> <kSso3CA5VPV3EwYB@jessikat.demon.co.uk>
Message-ID: <375992b6.8610970@news.freeserve.net>

Robin Becker <robin at jessikat.demon.co.uk> wrote:

>In article <7j1pcc$256$1 at m2.c2.telstra-mm.net.au>, Mark Hammond
><MHammond at skippinet.com.au> writes
>>Actually, I was about to post this yesterday, and to prove my point I wrote:
>>
>>def die():
>>    die()
>>
>>Expecting to see the crash.  I was then going to go into a rant about how
>>win32api.Apply would actually let you catch this (and any other Win32)
>>exception as a temporary solution.
>>
>>But it worked correctly - ie, Python raised a RuntimeError: Max Recursion.
>>
>...
>I just tried this with my bog standard 1.5.2 under win95 osr2 and it
>faults to the system. Strange :)

It could be that Mark did actually exceed the 1MB stack, but windows was able to
extend it since the relevent bit of his process's address space had not been
reserved.

Mark also wrote:

>A better mechanism would be to use Win32 exception handlers.  Then you dont
>need artificial recursion checks at all - leave it to the OS.  However, it
>isnt clear to me where this exception handler should be placed, nor indeed
>if setting up a Win32 exception handler inside the VM would have a
>significant performance hit.  A plus would be we could also catch Access
>Violations (Im not sure what we would do with them - but a traceback would
>be nicer than Dr Watson when dealing with buggy extensions!)

Getting this to work could be tricky. The win32 exception could be generated
when the VM or an extension was in an inconsistant state, and it may not be
possible to continue processing after swallowing the RuntimeError. If we don't
need to do that, then what is wrong with Dr Watson?

>Thomas' solution is simple, and requires either a simple source file change,
>or using the MSVC GUI to adjust the stack size.

I don't think adjusting the stack size is a viable option. Python is used in
some places where it is not in control of this parameter; in-process com servers
being the obvious example.



Toby Dickenson



From jeffrey at Digicool.com  Fri Jun  4 19:06:43 1999
From: jeffrey at Digicool.com (Jeffrey P Shell)
Date: Fri, 04 Jun 1999 19:06:43 -0400
Subject: python active web pages
References: <jonathan-0406991640390001@borg.onegoodidea.net>
	 <7j98g1$atj$1@autumn.news.rcn.net>
	 <jonathan-0406992311300001@borg.onegoodidea.net>
Message-ID: <7j9m2n$6kr$1@autumn.news.rcn.net>

> does anyone else use active web pages? surely all you python/web hackers
> don't just use CGI scripts?

No, that's why Gods created Bobo ;-)

Actually, here's another alternative that's only gone non-updated since
late november:: (Uses JPython and Servlets)

http://www.dstc.edu.au/aspy/

A _long_ time ago (about three years methinks), I did something like all
this based on the only web application server/platform (besides Zope) that
I really respect, WebObjects, called Pummel.

It had a syntax for putting code on the page, and then addressing it later.
In that phase it was _fairly_ simple.  But then it started trying to be a
little bit like Bobo (mainly:: i wanted more powerful components that I
could share).  And once I actually figured out Bobo and DTML, I put the
project away.  I don't even know where the source is anymore.

Basically, it looked like:

<HTML><HEAD>....
<APYML>
import foobar
import hti

hello=hti.Input(type='text', name='uname', value=uname)
</APYML>
</HEAD>
<BODY>
<FORM ACTION...>
<PYML NAME="hello"></PYML>
<INPUT TYPE="submit">
</FORM>
</BODY></HTML>

(hti was a very very very very simple HTMLGen-ish module without the smarts
that HTMLGen has.  Remember this was many many many moons ago).

I'm about to unplug and go into a sooperfire weekend, but one quick
question:  is there any reason why Zope isn't an option for what you're
doing?  It seems like all these pseudo-ASP-ish projects get to about a 0.8
release functionality and then die off (my own PYML project included).
HTMLGen and Bobo seem to be the most widely supported and up-to-date
options from what I've seen over the past few years.

There used to be something called "Boilerplate" (Broilerplate?) listed on
the Python web site that did ASP-esque stuff too.  Again, I don't know how
current the development is on that one either.

.jPS | jeffrey at digicool.com
zope, through the web. www.zope.org



From avv at quasar.ipa.nw.ru  Thu Jun  3 13:21:05 1999
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Thu, 03 Jun 1999 10:21:05 -0700
Subject: pyGTK on Windows
References: <3757D453.A302E757@interar.com.ar>
Message-ID: <3756B981.AACF7F1@quasar.ipa.nw.ru>

Nahuel wrote:

> Hi, im about to do a cross-plataform GUI application in Linux and
> Windoze... i must
> choose the GUI toolkit.. my question is .. can i use pyGTK under
> Windows???
>
> (I need a gui that can display a table with about 4000 lines without
> crashing (not like Tk :))

wxPython is mapped onto GTK under Linux, native API under Windows and
has a 'grid' widget. If it crashes (suppose this) on 4000 lines, you
will have a patch within a day (as I understand).

Alexander





From hniksic at srce.hr  Mon Jun  7 12:06:43 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 07 Jun 1999 18:06:43 +0200
Subject: fork()
References: <375BE64A.EB997F24@icrf.icnet.uk>
Message-ID: <87so840zcc.fsf@pc-hrvoje.srce.hr>

Arne Mueller <a.mueller at icrf.icnet.uk> writes:

> All that woks fine now (after I spend a long time fiddling with the
> pipe and select stuff to synchronize I/O). However the forking is a
> waste of time and memory because all the children get their own
> large dictionary!

If you don't change the dictionary, the memory will not be copied.
Most modern Unixes support COW (copy on write) which means that the
actual fork()ing does surprisingly little.  Only when the memory is
written to, a new copy is spawned off.

If the children are modifying the dictionary, that's well something
different...



From faassen at pop.vet.uu.nl  Wed Jun  2 05:14:00 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Wed, 02 Jun 1999 11:14:00 +0200
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
	 <87pv3iq6ss.fsf@ev.netlab.co.jp> <7iv86a$iud@news.dns.microsoft.com> <87vhd7nsra.fsf@ev.netlab.co.jp>
Message-ID: <3754F5D7.2B6E1954@pop.vet.uu.nl>

Yukihiro Matsumoto wrote:
 
> Some languages use both.  Ref counting for usual basis, real GC for
> full clean up.  How about adding real GC to Python 2.0, co-working
> with ref counting?

This sounds like an interesting idea; I believe I saw references to
garbage collecting schemes that help clean up circular references caused
by referencing counting. I forget where, though. :)

I agree with the general principle that for Python 2, garbage collecting
should be taken a good look at. I suppose the two main issues are:

* The predictability of reference counting is nice in the simple case.
In the complex case, you get into a mess of circular references too
easily, though.

* In the complex case, garbage collecting is nice as it cleans up
circular references. There are unwanted side-effects of its
unpredictability, though; a possible leaking of resources (files which
aren't closed), and some difficulties interfacing with C.

In that light combining the two schemes might be very interesting.

Repeatingly yours,

Martijn



From fleck at informatik.uni-bonn.de  Tue Jun  8 21:01:56 1999
From: fleck at informatik.uni-bonn.de (Markus Fleck)
Date: 9 Jun 1999 01:01:56 GMT
Subject: GC SIG
References: <Pine.A41.4.10.9906081526430.65174-100000@barney.mcit.com>
Message-ID: <7jkee4$qd8@news.rhrz.uni-bonn.de>

Robert Meegan <Robert.Meegan at wcom.com> wrote:
> Are there enough potential Sanitation Engineers to create a Garbage
> Collection SIG?
> ...
> 2) The GC discussions would be moved out of the general mailing
> list/newsgroup, freeing them of the constant debate. <huzzah again!>

Hmmm... exactly which kind of "garbage collection" are you referring to? :-)

Yours,
Markus.

-- 
If it ain't broken, better don't tempt to affix it. <0.5 broken wink>



From matz at netlab.co.jp  Tue Jun  1 23:24:13 1999
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 02 Jun 1999 12:24:13 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
 <87n1ykr0z9.fsf@ev.netlab.co.jp> <7ivs9q$dc5@news.dns.microsoft.com>
 <87k8toqv70.fsf@ev.netlab.co.jp> <aahzFCnx28.1Jp@netcom.com>
Message-ID: <871zfvp9nm.fsf@ev.netlab.co.jp>

aahz at netcom.com (Aahz Maruch) writes:

|In addition to Paul's excellent comments, if one is writing an
|interactive process, one cannot afford the unpredictability of random
|garbage collection.  Garbage collection could run in a thread, but then
|one is restricted to environments where threads are well-supported.

Well, in <87n1ykr0z9.fsf at ev.netlab.co.jp> I said:

|From my experience with Scheme and Ruby, both with real GC, the real
|GC is mostly very fast.  Fast enough no one notices its "hang" on most
|cases.  The legend was formed up by old GC implementation and slower
|machines of the past, I guess.

You can see STk (Scheme Tk, of cource with real GC) works just fine
for interactive process.
						matz.



From just at letterror.com  Mon Jun 14 14:00:58 1999
From: just at letterror.com (Just van Rossum)
Date: Mon, 14 Jun 1999 18:00:58 GMT
Subject: Embedding Questions on the Macintosh
In-Reply-To: <DH993.59$uF.6339@news.uswest.net>
References: <DH993.59$uF.6339@news.uswest.net>
Message-ID: <l03102805b38af25cece9@[193.78.237.136]>

At 9:01 AM -0700 6/14/99, Sean Hummel wrote:
>Hi, I am writing a game which uses Python as the control language for a
>bunch of sprites.  I don't have any problems doing the embedding part.
>
>What I want to know is if it is easy to package the entire Python Library,
>much like the "macfreeze" module does?    I want to distributes as few
>modifiable files as possible, and would prefer to distribute only two files:
>the application, and the data file.

That's indeed possible, you need to pack all modules as 'PYC ' resources
into the app. Check out PackLibDir.py.

Finally you have to add
$(APPLICATION)
to your app's path prefs. One way is to create a new resource file, drop it
onto EditPythonPrefs and add it to your C project.

Just





From suzuki611 at okisoft.co.jp  Wed Jun  2 21:08:42 1999
From: suzuki611 at okisoft.co.jp (Hisao Suzuki)
Date: 03 Jun 1999 10:08:42 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
	<87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>
	<87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net>
	<87hfosquc0.fsf@ev.netlab.co.jp> <3753F121.E4A9BE8B@prescod.net>
	<87yai3ntdt.fsf@ev.netlab.co.jp>
	<7j3iif$134e$1@nntp6.u.washington.edu>
	<7j45iv$908$4@cronkite.cc.uga.edu>
Message-ID: <u390a2ccpx.fsf@ares.sys1.ngo.okisoft.co.jp>

In article <7j45iv$908$4 at cronkite.cc.uga.edu>,
graham at sloth.math.uga.edu (Graham Matthews) wrote:
> You also seem to believe that with GC finalisation will disappear. Why
> do you believe that (a lot of GC schemes have finalisation). 

You might misunderstand Donn's saying.  As far as I understand,
he says that explicit finalization (or close call, in your
wording) disappears if reference counting is used; finalization
will be done implicitly with destructor such as __del__(self)
all at the expected time then.

> Finally I
> believe you are mis-using finalisation if you use it to close up files,
> sockets, etc. That's what close calls are for.

You are all right in some other languages.  However, if we adopt
such a manner as Pythonic style, we would have to write the code
like this:

	a = SomeClass(....)
	try:
	    ...do something with a...
	finally:
	    a.close()

instead of providing the __del__ method for SomeClass once and
for all.

--===-----========------------- Sana esprimo naskas sanan ideon.
SUZUKI Hisao            suzuki611 at okisoft.co.jp, suzuki at acm.org.



From phd at emerald.netskate.ru  Sat Jun 12 12:12:33 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Sat, 12 Jun 1999 20:12:33 +0400 (MSD)
Subject: regular expression question: last occurence
In-Reply-To: <19990612180242.A6075@optiplex.palga.uucp>
Message-ID: <Pine.LNX.4.04.9906122010160.28879-100000@emerald.netskate.ru>

On Sat, 12 Jun 1999, Gerrit Holl wrote:
> currentdir = os.getcwd()
> mo = re.search('/linuxgames', currentdir)
> eind = mo.end()
> subdir = currentdir[eind:] + '/'
> 
> But that doesn't solve my problem.

print os.path.basename(currentdir)

> kind regards,
> Gerrit.
> 
> -- 
> The Dutch Linuxgames website.		De Nederlandse Linuxgames pagina.
> Everything about games on Linux.	Alles over spelletjes onder Linux.
> Site address:				http://linuxgames.nl.linux.org
> Personal homepage:			http://nl.linux.org/~gerrit/

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From aahz at netcom.com  Tue Jun  1 15:06:08 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Tue, 1 Jun 1999 19:06:08 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <87n1ykr0z9.fsf@ev.netlab.co.jp> <7ivs9q$dc5@news.dns.microsoft.com> <87k8toqv70.fsf@ev.netlab.co.jp>
Message-ID: <aahzFCnx28.1Jp@netcom.com>

In article <87k8toqv70.fsf at ev.netlab.co.jp>,
Yukihiro Matsumoto  <matz at netlab.co.jp> wrote:
>
>Well, OK.  I understand now.  Now only thing I can say is, why do you
>emphasize so much on reclaiming object at the time you exepcted?
>Because of your C++ background?

In addition to Paul's excellent comments, if one is writing an
interactive process, one cannot afford the unpredictability of random
garbage collection.  Garbage collection could run in a thread, but then
one is restricted to environments where threads are well-supported.
-- 
                      --- Aahz (@netcom.com)

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

"NRE is a lot like cocaine. Feels great while you're on it, makes you
feel very special and talented, and you become highly annoying to those
who aren't on it."  -- Stef



From fatjim at home.com  Wed Jun 16 05:26:10 1999
From: fatjim at home.com (Jim Meier)
Date: Wed, 16 Jun 1999 09:26:10 GMT
Subject: Cut down on writing repetitive code
References: <3766f6a8.19474000@news2.mainnet.net> <3767128E.E049636C@mojam.com>
Message-ID: <37676DCA.3594C959@home.com>


Skip Montanaro wrote:

> Tim Gallati wrote:
>
> > If you do a lot of repetitive typing, check out Perfect Keyboard Lite.
> > I use it everyday and consider it my most useful program.
>
> Gee, I thought the usual OO way to cut down on repetitive typing was to
> abstract common code into base classes.  I guess the original OO folks
> (Simula? Smalltalk?) got it all wrong 20- to 30-odd years ago.  They
> should have invested in Mavis whats-her-name instead!

:) ROTFLMAO - again, and again, and again, .. oh, wait..
class again:
    def __init__(self, num, arg):
        tmp=arg[:]
        tmp=tmp+", and "
        self.val=tmp*(num-1)
        self.val=self.val+arg
    def __str(self):
        return self.val

print again(900, "again")

:)

>
> No disclaimers in the original post.  I guess it was an off-charter
> advertisement.  Another company hits my bit bucket.

I wouldn't be _THAT_ hard on them. True, they didn't mention that they were
actually the company in question, and for that they certainly deserve a
smack upside the head. But they certainly posted a relevant pointer to
something that may help a lot of people associated with this newsgroup. They
may not be perfect, but they are certainly on my "good usenet advertisers"
list for not randomly spamming, for spelling everything properly, for being
short and direct, and for being reasonably on-topic.

Better this than the recent porn postings!!

>
> --
> Skip Montanaro  | http://www.mojam.com/
> skip at mojam.com  | http://www.musi-cal.com/~skip/
> 518-372-5583

-Jim.




From smjoshi at bellsouth.net  Wed Jun 23 23:01:18 1999
From: smjoshi at bellsouth.net (SunitJoshi)
Date: Wed, 23 Jun 1999 22:01:18 -0500
Subject: Python Exes!
Message-ID: <%fhc3.2901$Lk4.35205@news4.mia>

Could someone please point me to the URL where Gordon Mcmillians, installer
is located. I had it but can't seem to locate it

thanks
Sunit





From kernr at mail.ncifcrf.gov  Wed Jun  9 17:33:17 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Wed, 09 Jun 1999 21:33:17 GMT
Subject: Unicode handling Wstring Win95/98 Binary
References: <000801beb250$11cb0c40$b8e1fea9@computername>
Message-ID: <375edb49.8685256@news.erols.com>

On Wed, 9 Jun 1999 16:14:13 +0800, "Sin Hang Kin"
<kentsin at poboxes.com> wrote:

[snip MIME]

>Hello,
>
>Does any one have wstring package in winxx binary format? I do not have =
>the necessary compiler to do it.

I do.  I can get wstrop.pyd to compile and work (I'm guessing on that
last claim; test.py doesn't report any errors.  I just downloaded the
package, so I don't know how it is supposed to work).  I still can't
get intl.pyd to compile (I don't have libintl.h and whatever else is
necessary).  I can e-mail wstrop.pyd if you want.  If there is more
interest, I'll place the binary in my forthcoming Starship cabin.

>Rgs,
>
>Kent Sin
>kentsin at poboxes.com

[snip MIME]




From clarence at silcom.com  Thu Jun 24 10:29:23 1999
From: clarence at silcom.com (Clarence Gardner)
Date: Thu, 24 Jun 1999 14:29:23 GMT
Subject: It's just magic, and I hate it ...
References: <slrn7mss5c.f90.mgm@unpkhswm04.bscc.bls.com> <14190.35530.772547.166720@buffalo.fnal.gov> <slrn7mvlhr.4is.mgm@unpkhswm04.bscc.bls.com> <37703621.B29B72CB@bioreason.com> <slrn7n1r6p.keg.mgm@unpkhswm04.bscc.bls.com> <slrn7n2aj0.n0c.mgm@unpkhswm04.bscc.bls.com> <930181790.156.59@news.remarQ.com>
Message-ID: <7hrc3.659$dR.56613@newsfeed.avtel.net>

Mitchell Morris (mmorris at mindspring.com) wrote:
: In article <slrn7n2aj0.n0c.mgm at unpkhswm04.bscc.bls.com>, 
: mmorris at mindspring.com wrote:
: [snip]
: 
: In a very polite private e-mail, Tim Peters pointed out that this is a FAQ, 
: number 4.87 in fact. That will teach me to not check *ALL* the docs before 
: opening my trap.
: 
: Thanks to everyone ... I appreciate your time and efforts on my behalf,
: +Mitchell

I wonder if it will teach me to read the entire thread before following
up to a message :)

-- 
-=-=-=-=-=-=-=-=
Clarence Gardner
AvTel Communications
Software Products and Services Division
clarence at avtel.com



From tim_one at email.msn.com  Thu Jun 10 21:22:37 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 10 Jun 1999 21:22:37 -0400
Subject: None inconsistent
In-Reply-To: <7jp51f$rci$1@nnrp1.deja.com>
Message-ID: <000201beb3a8$e3e0f100$329e2299@tim>

[mjhand at concentric.net]
> Pardon the repetition (if this is repetitive).  I have posted
> this message three times but have yet to see it appear on the
> newsgroup.

First time I've seen it <wink>.

> In 1.5.1, "None" was "less" than all numbers:
> >>> 0 < None
> 0
> >>> sys.maxint > None
> 1

Yes, "incomparable" objects were compared first by the *string* name of
their type in 1.5.1, and as strings "None" < "int".  An accident, really.

> In 1.5.2, "None" became "greater" than all numbers:
> >>> 0 < None
> 1
> >>> sys.maxint > None
> 0
>
> I have three questions about this:
> (1) Was this change made intentionally?  (I hope so.)

What happened to None was pure accident, but the rules were deliberately
changed to treat all builtin numeric types (int, long, float, complex) as
having string name "", i.e. an empty string.  Else e.g. because, as strings

    "int" < "list" < "long"

we also had

    42 < [42] < 42L

return true while

    42 < 42L

returned false.  Python is happy to let you create your own non-transitive
comparisons, but didn't want to set a bad example <0.9 wink>.

In 1.5.2 they both return false.

> (2) Can we count on it staying this way forever?  (I hope so.)

The language doesn't define what happens when doing silly <wink>
comparisons, beyond promising consistency involving comparisons of objects
of builtin types.  So, no, in theory you can't count on it.  In practice,
the change in 1.5.2 achieved the consistency it was aiming for, so as a
practical matter it's unlikely to change again.

> (3) When will the starship be updated to use 1.5.2 by default?
>     (/usr/bin/python == 1.5.1, /usr/local/bin/python == 1.5.2b2)

Beats me -- bring it up on the starship crew mailing list.

> I absolutely LOVE that "None" is accepted as a valid argument in
> min(), max(), and cmp(), but I would like it to behave consistently.
> The 1.5.2 way is best for my code.  Please tell me it will stay
> that way forever.

I won't lie without better financial inducement than that.

besides-only-guido-is-eternal-ly y'rs  - tim





From kernr at mail.ncifcrf.gov  Tue Jun 22 19:13:53 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Tue, 22 Jun 1999 23:13:53 GMT
Subject: interactive-logging
References: <376FF09A.8170CB@icrf.icnet.uk>
Message-ID: <3770183d.84268602@news.erols.com>

On Tue, 22 Jun 1999 21:22:50 +0100, Arne Mueller
<a.mueller at icrf.icnet.uk> wrote:

>Hi All,
>
>I'd like to add some logging to the interpreter's interactive modus so
>that everything I type in and everything what's printed to the screen
>(errors and 'normal' output) is copied to a log-file.

[snip]

>Is there already something like this written? I've no idea how how the
>interactive modus works. For aprogram I've written that copies a
>file-object, say everything that goes to 'sys.stderr' is also written to
>a log-file, how to do that interactive modus?

I'm not sure how many of your desired features are in it, but try out
IPP (the Interactive Python Prompt) by Janko Hauser at
(http://starship.python.net/crew/jhauser/IPP.tgz).

>	thanks very much,
>
>	Arne

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From claird at Starbase.NeoSoft.COM  Tue Jun 29 11:30:08 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 29 Jun 1999 10:30:08 -0500
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <876748mcxg.fsf@home.ivm.de> <37788BB9.ABD@mailserver.hursley.ibm.com> <lw67471fyi.fsf@copernico.parades.rm.cnr.it>
Message-ID: <7laoq0$bvh$1@Starbase.NeoSoft.COM>

In article <lw67471fyi.fsf at copernico.parades.rm.cnr.it>,
Marco Antoniotti  <marcoxa at copernico.parades.rm.cnr.it> wrote:
			.
			.
			.
>spin. :) Finally, AFAIU, a Tcl "macro" must run as an interpreter of
>the spec. A Common Lisp (Scheme) macro is compiled into regular code
>by read-time expansion.
			.
			.
			.
Are you making a semantic point, or arguing
on the basis of implementation-specific per-
formance?  If the latter, please be aware
that the Tcl community is actively investi-
gating dramatic speedups of [eval].
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From l.a.apiolaza at massey.ac.nz  Tue Jun 29 01:38:35 1999
From: l.a.apiolaza at massey.ac.nz (l.a.apiolaza at massey.ac.nz)
Date: Tue, 29 Jun 1999 05:38:35 GMT
Subject: reloading modules in idle
Message-ID: <7l9m4q$6r1$1@nnrp1.deja.com>

Hi Pythoneers,

I'm writing my Python code using IDLE. When working with multiple files
and making corrections to some of them IDLE doesn't use the latest
version of the files, but the one previously interpreted. Is there any
way (from inside IDLE) to reload the latest versions and run the code
with all the corrections?

Thanks,


Luis
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Luis A. Apiolaza
IVABS
Massey University
Palmerston North
New Zealand

l.a.apiolaza at massey.ac.nz
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From petrilli at trump.amber.org  Mon Jun 28 12:17:31 1999
From: petrilli at trump.amber.org (Christopher Petrilli)
Date: 28 Jun 1999 09:17:31 PDT
Subject: ConfigParser.py?
References: <rNHvjEdhm5Pp-pn2-ob9OW6dCFSRN@d185d18f8.rochester.rr.com>
Message-ID: <7l876r$nrs@chronicle.concentric.net>

Benjamin Schollnick <rfreedm1 at nospam.rochester.rr.com> wrote:
> Folks,

> 	I'm trying to use ConfigParser... And I believe I'm doing everything
> write, except I don't see any documentation (including in the 
> configparser.py file)
> telling my what function(s) to use to write the file out to disk???

That's simple... it doesn't exist :-)

> 	Can someone give me a example or syntax on that?

Would love ot, but you'll need to write the code to do this.  This was
designed to be "read only", although it's been talked about making it
bi-directional.

Chris
-- 
| Christopher Petrilli                      ``Television is bubble-gum for
| petrilli at amber.org                          the mind.''-Frank Lloyd Wright



From toby at puckish.demon.co.uk  Thu Jun 10 18:59:30 1999
From: toby at puckish.demon.co.uk (Toby Kelsey)
Date: Thu, 10 Jun 1999 23:59:30 +0100
Subject: Newbie class questions
Message-ID: <tN13wIASNEY3EwAl@puckish.demon.co.uk>

I have just started to learn Python (v1.5.2), and have some questions
about classes.  Here is something I prepared earlier...

--- cut ---
def y(x): return 'a global function y(x)'
def g0(): return 'some arbitrary function (no args)'
def g1(x): return 'another arbitrary function (one arg)'
i = 0
class MyClass:
  """A simple class demonstrating scope."""
  i = 12345
  def __init__(self):
    print '\n\tI output, therefore I am.\n'
  def y(x):
    return 'default class method y'
  def f(x):
    # i = 7 # hides global
    print 'Global variable:', i
    print 'Class variable:', MyClass.i
    print 'Instance variable:', x.i
    print 'Global function:', y(x)
    print 'Class method:', MyClass.y(x) # explicit instance arg.
    print 'Instance method:', x.y() # implicit instance arg.
    return
  def q():
    """This method is inaccessible - should be a parse-time error"""
    return

t = MyClass()
print "Doc:", t.__doc__
t.f()
print "\nModified:"
i = 21
t.i = 169
#t.y = g1
t.y = g0
t.f()
--- cut ---

BTW I think an example class like this in the tutorial would help
explain scope, but there are also a few things I don't understand here
(probably my C++ background).

(1) Is there a reason why I cannot call MyClass.q()?  Why require a
meaningless instance arg to a class-static method?  Alternatively
shouldn't it be an error at parse time?

(2) Why doesn't t.y = g1 work?  How am I meant to access instance
attributes in g0 without an instance argument?

(3) Some mention was made in the tutorial of the scope rules being in
flux.  Are any impending changes likely to affect the output of this
class?

Cheers,
Toby
-- 
Toby Kelsey



From gmcm at hypernet.com  Mon Jun 14 10:02:48 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 14 Jun 1999 09:02:48 -0500
Subject: Problem with embedded python
In-Reply-To: <7jk0cm$4hk@wapping.ecs.soton.ac.uk>
Message-ID: <1282753272-98086574@hypernet.com>

Ben writes:
>     I have embedded python into a win32 c++ application using
> PyRun_SimpleFile. This executes a python script at regular (25 times
> a second-ish) intervals. Also included in this C++ application is a
> procedure which handles the window messages, which I have also tried
> to script. Unfortunately, running any script in this function
> crashes windows with a nasty protection fault. Do I need to run my
> scripts on seperate threads... or in different interpreters. or is
> this genuinely a bug..

It seems likely that the procedure handling windows messages is being 
executed by a different thread than the one running PyRun_SimpleFile. 
In which case you will need to establish 2 thread states when you 
start up Python, and use one for each purpose. If the work they are 
doing is completely disjoint, you could use 2 different interpreter 
states. But Python needs to know that it is operating in a threaded 
environment, and has to build objects which shadow the OS threads. 

The hard part is getting it started; after that it should just work.
And it's not actually hard to get it started, just confusing. Check
out the pysvr demo. You should probably direct further questions to
the Thread SIG. 

- Gordon



From joe at strout.net  Thu Jun 17 16:49:16 1999
From: joe at strout.net (Joe Strout)
Date: Thu, 17 Jun 1999 13:49:16 -0700
Subject: [ANN] Graphite 0.1
Message-ID: <joe-170619991349164707@chinacat.salk.edu>

[[ This message was both posted and mailed: see
   the "To," "Cc," and "Newsgroups" headers for details. ]]


The Graphite development team is pleased to announce the first public
release of Graphite, a 100% Python, open-source, cross-platform
plotting/graphing library.

      http://www.strout.net/python/graphite/

Graphite will enable the easy creation of scientific, engineering, and
business graphs. Some graph types to be included are line graphs, bar
charts, function plots, polar plots, parametric plots, and mesh plots.
Graphite will be able to produce output for formats such as postcript,
pdf, gif, Windows, QuickDraw, Tk by using the PIDDLE drawing system. It
will also have interactive capability (e.g., positioning labels and
other overlays via the mouse or arrow keys), and should run on any
platform that supports Python. 

This is the very first public release -- note the version number, 0.1. 
Many of the features mentioned above are not yet implemented.  However,
it does run, and the intrepid Python hacker should be able to produce
some plots in short order.  Documentation is sparse, so please refer to
the web site for frequent updates.  You may also wish to join the
mailing list, at <http://www.egroups.com/group/graphite-list> (click on
"group info" to get to the subscription form).

This is an open-source project; we are actively seeking input and
contributions from the Python community.  That's why we're releasing it
now, in its current embryonic stage.  Our hope is that with your input,
Graphite will grow rapidly into a powerful, professional-quality yet
easy-to-use plotting package.

Cheers,
-- Joe

-- 
,------------------------------------------------------------------.
|    Joseph J. Strout           Biocomputing -- The Salk Institute |
|    joe at strout.net             http://www.strout.net              |
`------------------------------------------------------------------'



From mlh at idt.ntnu.no  Wed Jun 23 09:44:52 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 23 Jun 1999 15:44:52 +0200
Subject: interactive-logging
References: <376FF09A.8170CB@icrf.icnet.uk>
Message-ID: <y0jso7jghez.fsf@vier.idi.ntnu.no>

Arne Mueller <a.mueller at icrf.icnet.uk> writes:

> Hi All,
> 
> I'd like to add some logging to the interpreter's interactive modus so
> that everything I type in and everything what's printed to the screen
> (errors and 'normal' output) is copied to a log-file.

Hm. Isn't there a UN*X command that does that? (That logs everything
going on in a terminal...)

I'm damned if I can remember what it was, though...

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From jonathan at onegoodidea.com  Fri Jun  4 11:40:39 1999
From: jonathan at onegoodidea.com (Jonathan Hogg)
Date: Fri, 04 Jun 1999 16:40:39 +0100
Subject: python active web pages
Message-ID: <jonathan-0406991640390001@borg.onegoodidea.net>

hi there,

i'm looking to embed python into web pages in a similar way to ASP and
PHP3 and so forth.

i found a thing called Active Server Python,

   <http://starship.skyport.net/crew/ejohn/asp/>,

but i note that it doesn't seem to have had any attention in the last year.

i also looked at ZTemplates (a component of Zope),

   <http://www.zope.org/Download/ZTemplates/>,

and while it doesn't actually embed python, it has a limited embedding
language of its own, DTML, which gets munged into python. i must say that
DTML looks a bit hideous (if anyone from digicool is reading, sorry guys).

anyone have any experience with active web pages using either of these or
something else?

thanks in advance,

:-j

-- 
jonathan hogg, one good idea limited, 25 clouston street, glasgow g20 8qr, uk
jonathan at onegoodidea.com www.onegoodidea.com t:(0)976-614338 f:(0)7970-537451



From wware-nospam at world.std.com  Fri Jun 25 17:40:41 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Fri, 25 Jun 1999 21:40:41 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net>
Message-ID: <FDwK7t.BM7@world.std.com>

Mikael Lyngvig (mikael at pobox.com) wrote:
: I'd like some discussion on whether or not it is a good idea to
: abandon TkInter, before it gets too widely used, and instead adopt
: wxPython as the primary GUI framework for Python.

The last time I tried playing with wxWindows on a Linux box, it
demanded some library I didn't have. I don't remember if it was
Mesa, KDE, Gnome, or what, but a couple days' fooling around with
it got me no progress. For all its performance problems, Tcl/Tk does
work pretty much everwhere.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com



From bernhard at alpha1.csd.uwm.edu  Sat Jun 19 19:33:48 1999
From: bernhard at alpha1.csd.uwm.edu (Bernhard Reiter)
Date: 19 Jun 1999 23:33:48 GMT
Subject: BiBTeX module?
References: <joe-160619991358444033@chinacat.salk.edu> <7k950a$hsb@cs.vu.nl> <slrn7miqcg.tpi.bernhard@alpha1.csd.uwm.edu> <slrn7mk9ui.dic.gobry@blanc.idiap.ch>
Message-ID: <slrn7moa6s.4l5.bernhard@alpha1.csd.uwm.edu>

On 18 Jun 1999 13:04:51 +0100, Frederic Gobry <gobry at blanc.idiap.ch> wrote:
>>Having a look at the current programs available for editing and entering
>>BiBTeX files someone really ahs to write a good python/tk or
>>python/wxpython program for this task, too.
>
>I've also started such a project as free software. The parser and the python
>interface are working correctly now. My idea was to write something generic
>for bibliographic databases, BibTeX being only the first module. A GNOME
>interface is also in progress (does some display but no real edition yet)

Great!

>So I can make a snapshot for you or you can use an old version 
Please make snapshot and more infos available.
I might be able to playtest a bit.
	Bernhard

-- 
Research Assistant, Geog Dept UM-Milwaukee, USA.         (www.uwm.edu/~bernhard)
Funding/Promoting Association for a Free Informational Infrastructure (ffii.org)



From suzuki611 at okisoft.co.jp  Wed Jun  2 19:27:00 1999
From: suzuki611 at okisoft.co.jp (Hisao Suzuki)
Date: 03 Jun 1999 08:27:00 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
	<87pv3iq6ss.fsf@ev.netlab.co.jp> <7iv86a$iud@news.dns.microsoft.com>
	<u3r9nwtd9q.fsf@ares.sys1.ngo.okisoft.co.jp>
	<87so8bnsd6.fsf@ev.netlab.co.jp>
	<u37lpndko6.fsf@ares.sys1.ngo.okisoft.co.jp>
	<87pv3eorg4.fsf@ev.netlab.co.jp>
Message-ID: <u3aeuichff.fsf@ares.sys1.ngo.okisoft.co.jp>

In article <87pv3eorg4.fsf at ev.netlab.co.jp>,
Yukihiro Matsumoto <matz at netlab.co.jp> wrote:
> BTW, did I say Ruby can invoke arbitrary destructors specified by
> arbitrary users?  It can't.

Is this your mental premise in your opinion about garbage
collection?  Then, the unpredictability might not be matter at
all...

I see that is the way of your own excellent Ruby, but it is not
the manner of the present Python.

--===-----========------------- Sana esprimo naskas sanan ideon.
SUZUKI Hisao            suzuki611 at okisoft.co.jp, suzuki at acm.org.



From tismer at appliedbiometrics.com  Thu Jun  3 07:09:45 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Thu, 3 Jun 1999 11:09:45 GMT
Subject: need help checking for stack overflow in Win32
References: <5lk8to3su3.fsf@eric.cnri.reston.va.us>
		<7j1pcc$256$1@m2.c2.telstra-mm.net.au>
		<375502C7.39C4FD0E@appliedbiometrics.com> <7j38bs$ipe$1@m2.c2.telstra-mm.net.au>
Message-ID: <37566279.15E62594@appliedbiometrics.com>


Mark Hammond wrote:
> 
> Christian Tismer wrote in message
> <375502C7.39C4FD0E at appliedbiometrics.com>...
> 
> >Nope. The trap is:
> >You compiled with VC++ 6.0 which has better register spilling.
> >Therefore, recusion costs only 100 bytes, not 104.
> >Not calculate 104 * 10000 and you see why 1 MB stack
> >lets it crash.
> 
> Cool.  Thanks.
> 
> But doesnt the fact remain that counting the recursion depth is dangerous?
> Im thinking of embedded environments, where a) the stack size of the hosting
> .exe can't be known and b) the stack used by the app before Python is even
> called can't be determined.  Add to this callbacks in GUI and COM type
> applications, and we seem to be applying more band-aids.
> 
> The idea of counting space used seems better, but still assumes a known
> stack size.

Absolutely right.

The shorthand solution is to either modify the stack size
of the program, or to modify the recursion depth in
python15.dll.

The better solution is stackless Python which I'm
working on. But I'm no sourcerer, it takes time.

> 
> >I submitted a patch program for this kind of
> >problems, so I think there is no need to change
> >Python 1.5.2 at all.
> 
> Huh?  Are you suggesting people continue to build from sources, then run
> your script to get an "official" version?  Wouldnt it be better to get Guido
> to apply these changes to the build process?

No. People who build from sources migtht simply change
their stack size or the recursion depth parameter.
There is no reason to submit such a patch to Guido,
since he can adjust his parameters alone.

The reason why I provided a binary patch is, that this
problem is not worth messing around with another python release.
If someone cares, he just adjusts his max_recursion_depth
with my little program, and he's done.

This is even more powerful than changing the source,
since this parameter can be patched whenever needed,
without recompilation.
Of course, such a parameter could be made visible and
changeable through a variable like sys.max_recursion,
but this is a change which Guido would not want to
include into 1.5.2 .
So well, my patch does it. Many people don't like
binary patches. I don't care, it works and it's there.

> You have provided an excellent interim fix, but IMO the problem remains.

The interim fix is the final fix, since the problem is
about to vanish completely.

ciao - chris.stackless

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From ivnowa at hvision.nl  Tue Jun 22 02:47:24 1999
From: ivnowa at hvision.nl (ivnowa at hvision.nl)
Date: Tue, 22 Jun 1999 06:47:24 GMT
Subject: newbie idiom question
In-Reply-To: <m3emj4yezq.fsf@monsoon.swcp.com>
References: <m3emj4yezq.fsf@monsoon.swcp.com>
Message-ID: <199906220645.IAA15728@axil.hvision.nl>

On 21 Jun 99, Alex Rice wrote:

> Something I keep getting tripped up about is that objects being
> iterated in "for" statements cannot be modified.
> 
> I keep trying to do this:
> 
>       >>> eggs = [1,2,3]
>       >>> for spam in eggs:
>       ...	spam = 'cooked'
>       ...
>       >>> eggs
>       [1, 2, 3]
>       >>> 
> 
> The tutorial says this:
> 
>       If you need to modify the list you are iterating over, e.g., duplicate
>       selected items, you must iterate over a copy. The slice notation makes
>       this particularly convenient:

> Understood, but what if you want to modify each element in a list?
> What's the best way to do this in terms of speed and elegance? I guess
> I just haven't seen a good example of this yet in Python

> What I'm unfortunately used to is this, in Perl:
> 
>       @eggs = (1,2,3);
>       foreach $spam (@eggs) {
> 	$spam = 'cooked'; 
>       }
>       print "@eggs";
>       >>> cooked cooked cooked

An unwelcome side effect, methinx. You can use range:

for i in range(len(eggs)):
    eggs[i] = 'cooked'

Since the list you're iterating over is the same as the list you're 
modifying, it may be wise to use a slice (not in this particular 
case, but in general):

for i in range(len(mylist[:])):
    if mylist[i] == 'ham':
        mylist[i:i] = ['eggs']  # insert something

You can also use map:

eggs = map(lambda s: 'cooked', eggs)

But this can get difficult for complex lambdas, and it doesn't change 
the list in place.


Hans Nowak (ivnowa at hvision.nl)
Homepage: http://fly.to/zephyrfalcon



From faassen at pop.vet.uu.nl  Thu Jun  3 17:18:13 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Thu, 03 Jun 1999 23:18:13 +0200
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <3756E3D9.E44D2B39@swcp.com>
Message-ID: <3756F115.CBBAC9D8@pop.vet.uu.nl>

Alex Rice wrote:
> 
> Martijn Faassen wrote:
> ...
> 
> > Okay, sorry for all this, it's out of line, I just needed to blow off
> > steam. :)
> 
> As the composer John Cage would have told you, silence is music as much
> as tones and harmonies are music. silence == whitespace ?

What did John Cage say about curly braces and dollar signs? :)

Regards,

Martijn



From htrd90 at zepler.org  Fri Jun 11 21:45:09 1999
From: htrd90 at zepler.org (Toby Dickenson)
Date: Sat, 12 Jun 1999 01:45:09 GMT
Subject: fork()
References: <000901beb424$b2878200$a49e2299@tim>
Message-ID: <3761b44e.1186639@news.freeserve.net>

"Tim Peters" <tim_one at email.msn.com> wrote:

>Hard to weigh that against the probable increases in memory and runtime;
>and, philosophically, you gotta admit it's a mess.  OTOH, the combo of
>finalizers, cycles and resurrection seems to be a philosophical mess even in
>languages that take it all <ahem> seriously.
>
>Don't know, Guido!  I could live happily with it provided it didn't slow
>things down much, and especially provided I don't think about the semantics
>at all <wink>.  But then I'm not a __del__-lover to begin with.  Anyone
>following this who is?

All of my recent __del__ methods have been similar to a file object, where
__del__ calls self.close and the close method must expect to be called more than
once.

I would be happy for a m&s to just call __close__, multiple times if necessary,
trusting the user to manually break any cycles.

>resurrection-should-be-reserved-to-gods-ly y'rs  - tim


Toby Dickenson



From skott1870 at my-deja.com  Tue Jun  1 14:53:54 1999
From: skott1870 at my-deja.com (skott1870 at my-deja.com)
Date: Tue, 01 Jun 1999 18:53:54 GMT
Subject: Capturing stderr from a call to os.popen
Message-ID: <7j1a88$1vq$1@nnrp2.deja.com>


Is it possible to redirect stderr from a call to os.popen? If so,
will someone post an example.

Thanks,
TST


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From aahz at netcom.com  Sat Jun  5 16:32:47 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Sat, 5 Jun 1999 20:32:47 GMT
Subject: Deleting characters from a string
References: <87n1ye3g4v.fsf@pc-hrvoje.srce.hr> <y0jyahye89i.fsf@vier.idi.ntnu.no>
Message-ID: <aahzFCvFqo.7wz@netcom.com>

In article <y0jyahye89i.fsf at vier.idi.ntnu.no>,
Magnus L. Hetland <mlh at idt.ntnu.no> wrote:
>
>result = ""
>for char in host:
>    if char not in ' "':
>        result = result+char
>
>host = result

That's a bad way to do it.  At the very least, that code should be
something like:

# There may be a couple of syntax errors below
result = []
for char in host:
    if char not in ' "':
        result.append(char)
host = string.join ( result, "" )

This has *much* better performance if host is at all long.
-- 
                      --- Aahz (@netcom.com)

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

"Why do you like my boyfriend to tie you up and beat me.?"
    -- ELIZA generates a poly moment



From skip at mojam.com  Tue Jun 15 11:17:01 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 15 Jun 1999 15:17:01 GMT
Subject: Variations on a theme by "pound bang"
References: <3763e75f.3265255@news.iol.ie> <3763F77E.6094AEE9@appliedbiometrics.com> <3764185B.7BCBC488@Lugoj.Com>
Message-ID: <37666EFD.B31EBCAC@mojam.com>

James Logajan wrote:
> Christian Tismer wrote:
> > Uses the env program which is allways there, to figure out
> > from the environment settings which python to use.
> Alas, this is a myth. Not all Unixen put env in /usr/bin, some put it in
> /bin. I can't recall if it was Solaris (probably not) or LynxOS (probably)
> that put env in /bin. Anyway, I got bit by it.

I think it's a case of the 80/20 rule.  *Most* Unix variants put env it
in /usr/bin these days.  (I think there was a thread about this when the
topic came up in the past couple of years.) In fact, it appears that
/bin is falling out of favor for most tools, at least on Linux:

    % ls /bin | wc -l
         86		(18 are symlinks)
    % ls /usr/bin | wc -l
        969		(97 are symlinks)

I don't seem to recall that being the case in the "good old days".

-- 
Skip Montanaro	| Mojam: "Uniting the World of Music"
http://www.mojam.com/
skip at mojam.com  | Musi-Cal: http://www.musi-cal.com/
518-372-5583



From thornley at visi.com  Sat Jun 19 13:20:22 1999
From: thornley at visi.com (David Thornley)
Date: Sat, 19 Jun 1999 17:20:22 GMT
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <rnbn1y2n5jk.fsf@nortelnetworks.com> <sFU93.1567$kS.233449@ptah.visi.com> <rnbpv2v5ejh.fsf@nortelnetworks.com>
Message-ID: <qjQa3.2280$kS.310807@ptah.visi.com>

In article <rnbpv2v5ejh.fsf at nortelnetworks.com>,
Chris Ebenezer  <chriseb at nortelnetworks.com> wrote:
>
>thornley at visi.com (David Thornley) writes:
>: >Why use TK at all ?  Once you are using scheme anyway there are nicer
>: >widget toolkits to use (from a point of view of more "normal" looking
>: >toolkits that don't suffer from color allocation problems),  there are
>: >bindings for guile and gtk and guile and Qt. 
>: >
>: The attraction, to me, would be that Tk is a free, open-source
>: graphics environment that will run on Unix with X-Windows, the
>: Macintosh, and Microsoft Windows.  CLIM is commercial, Garnet last
>
I haven't gotten any responses that indicate that there's anything
besides Tk that meets the requirements I've specified:  free (in some
sense or another), open source (in some sense or another) and running
on MacOS, Microsoft Windows and X Windows.  Apparently Qt is free in
a very restricted sense, and doesn't run on MacOS.

>Hi David,
>
>If you want something that is completely free (in the GNU sense) go to
>http://www.gtk.org/, scroll down until you reach the bit on the sidebar
>that reads "Language Bindings".  You should be able to get guile-gtk
>from there.
> 
I'm willing to consider free in the GNU sense, but GTK claims to be
for X under Unix.  I didn't see anything about GTK for the Macintosh
(yes, I have heard RMS discuss Apple Corporation) or Microsoft Windows,
aside from a note from somebody porting it in his spare time.  Given
that I have limited spare time, I'm interested in a toolkit that I can
use on the Mac without doing my own port first.  (Besides, the impression
I get from the discussion on the MS Windows port is that GTK was
designed for X Windows and Unix, and is unnecessarily hard to port
compared to a toolkit designed with a wider focus to begin with.)

It still looks like Tk is the best available graphics package for my
purposes.

--
David H. Thornley                        | If you want my opinion, ask.
david at thornley.net                       | If you don't, flee.
http://www.thornley.net/~thornley/david/ | O-



From scarblac-spamtrap at pino.selwerd.cx  Sun Jun 27 13:51:47 1999
From: scarblac-spamtrap at pino.selwerd.cx (Remco Gerlich)
Date: 27 Jun 1999 17:51:47 GMT
Subject: Questions; freeze, tkinter, internet connection...
Message-ID: <slrn7ncopu.84g.scarblac-spamtrap@flits104-37.flits.rug.nl>

I need to make a few programs that will be given to our clients,
with which they can periodically enter some information that will
be mailed to us. I'd like to do it in Python, since I recently
learned it and I'm in love with its readability, mainly. I also
expect development to be very quick. Before I can decide to use
it I need a few answers though, hopefully someone here can help :)

I understand I can use Freeze and the compiled Python/Tk program
will work on Win32 without needing anything else installed?

The FAQ says that Freeze "maybe works with Visual C++ 6.0".
Does it? (can't test, I don't own any version yet)

Can I use the extensions to Tk like BLT or Tix? I'd like to use
tab bars...

I've seen the examples of how to send mail and so on, but how do
I know if there is a connection to the Net at the moment? How do I
start a dial-in if there isn't any yet? 


Now if I can get one thing to work well with Python, I'll convince
my boss we should use it a lot more...
-- 
Remco Gerlich,  scarblac at pino.selwerd.cx



From tim_one at email.msn.com  Wed Jun 23 01:14:20 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 23 Jun 1999 01:14:20 -0400
Subject: It's just magic, and I hate it ...
In-Reply-To: <slrn7mvlhr.4is.mgm@unpkhswm04.bscc.bls.com>
Message-ID: <000b01bebd37$3ff19340$cb9e2299@tim>

[Charles G Waldman]
> Try running Python in unbuffered mode - with a "-u" commandline flag,
> or by setting the PYTHONUNUFFERED environment variable to a non-empty
> string.

[Mitchell Morris]
> Are you saying that this worked for you? Or was this a guess?
>
> I ask because it makes no difference when I try it.

Something like this popped up a while back, which made me <wink> write FAQ

4.87. Why doesn't closing sys.stdout (stdin, stderr) really close it?

This solved the problem for the person who had it then.  So read the FAQ and
try what it says.  If that doesn't work, I'll try to find the original
thread in DejaNews.

charles'-suggestion-was-worth-a-try-but-you're-getting-buffered-
    somewhere-else-ly y'rs  - tim





From news at dorb.com  Wed Jun 30 17:40:17 1999
From: news at dorb.com (Darrell)
Date: Wed, 30 Jun 1999 17:40:17 -0400
Subject: problem with Fnorb and module new
References: <377A8D17.9B4D17C8@sr.hp.com>
Message-ID: <kbwe3.4027$U6.20290@newsr2.twcny.rr.com>

I don't think that "new" is part of the default build.
It shows up in the Windows build.

--
--Darrell
Jim Kerr <jbkerr at sr.hp.com> wrote in message
news:377A8D17.9B4D17C8 at sr.hp.com...
>   I just tried installing Fnorb on HP-UX 10.20... the install went
> smoothly, but when I tried stepping through the Hello World example,
> python complained about not being able to find module new. The
> complaint
> is certainly valid -- I can't find new.py anywhere either. Anyone know
> what's going on here?
>
> $ fnidl HelloWorld.idl
> Traceback (innermost last):
>   File
> "/users/jbkerr/python/Python-1.5.1/site-packages/Fnorb/script/fnidl",
> line 2, in ?
>     import sys; from Fnorb.script import fnidl;
> sys.exit(fnidl.main(sys.argv))
>   File
> "/users/jbkerr/python/Python-1.5.1/site-packages/Fnorb/script/fnidl.py",
> line 41, in ?
>     from Fnorb.orb      import CORBA, Util
>   File
> "/users/jbkerr/python/Python-1.5.1/site-packages/Fnorb/orb/CORBA.py",
> line 38, in ?
>     import fnorb_thread, TypeManager, Util
>   File
> "/users/jbkerr/python/Python-1.5.1/site-packages/Fnorb/orb/Util.py",
> line 38, in ?
>     import keyword, new, string, UserList
> ImportError: No module named new
>
> --
> Jim Kerr
> Hewlett-Packard
> Santa Rosa Systems Division
> 1400 Fountaingrove Pkwy, MS 3USZ
> Santa Rosa, CA 95403
> Phone: 707-577-5575
> Telnet: 577-5575





From graham at sloth.math.uga.edu  Fri Jun  4 10:32:38 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 4 Jun 1999 14:32:38 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
Message-ID: <7j8o26$55v$1@cronkite.cc.uga.edu>

Hisao Suzuki (suzuki611 at okisoft.co.jp) wrote:
: Me too.  And I am afraid that some evangelists would do nothing
: but _only_ point out Python's some so-called fault in order to
: drive us into convert to their favorite or invented language...

I am sorry Hisao but this is pure fantasy. This kind of "anyone
suggesting improvement to Python is a critic" mentality pervades
this thread and this newsgroup. Can you give even one example of
where someone has tried to convert anyone to another language in
this thread? The only other language mentioned in this thread is
Ruby, and the author of the article about Ruby was very clear to
point out that he was reporting his experience with GC in Ruby
and not evangelising it. If you cannot give an example then what
you write above is fantasy. 

Hisao Suzuki (suzuki611 at okisoft.co.jp) wrote:
: Anyway, as you say, I'd think that we need an actual experience.

But people do have experience with GC. Sure that experience is not
in Python, but why do you believe that GC in Python will be radically
different to GC in any other language?

graham
-- 
                  A cat named Easter
              He say "will you ever learn"
              Its just an empty cage girl
                 If you kill the bird



From bwarsaw at cnri.reston.va.us  Tue Jun  1 10:34:38 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Tue, 1 Jun 1999 10:34:38 -0400 (EDT)
Subject: File locking 
References: <199905301615.MAA00996@mira.erols.com>
Message-ID: <14163.61310.936743.849211@anthem.cnri.reston.va.us>

>>>>> "AMK" == A M Kuchling <amk1 at erols.com> writes:

    AMK> What is the correct solution for doing file locking in Python
    AMK> programs?

You might want to look at Mailman's flock.py module.  The doc string
says:

    Portable (?) file locking with timeouts.  
    This code should work with all versions of NFS.
    The algorithm was suggested by the GNU/Linux open() man page.  Make
    sure no malicious people have access to link() to the lock file.

It's also been banged on quite a bit, although I'll make no guarantees 
that it's bug free.  It's probably pretty close tho'.

-Barry



From zigron at jps.net  Tue Jun  8 00:19:21 1999
From: zigron at jps.net (Zigron)
Date: Mon, 7 Jun 1999 21:19:21 -0700
Subject: where does 'del' come from?
References: <7jhuu5$8d4$1@nnrp1.deja.com>
Message-ID: <yS073.6192$F7.635459@typhoon-sf.snfc21.pbi.net>

    I *believe* that 'del' is a keyword, thus, isn't exactly /defined/
anywhere
like functions and the likes are. Its like 'if', and 'while', and 'print'..

--Stephen

<sragsdale at my-deja.com> wrote in message news:7jhuu5$8d4$1 at nnrp1.deja.com...
> A fairly academic question, but I'm curious nonetheless.  Where is 'del'
> defined?  Functions like 'join' are obviously in the string package,
> 'range' is a builtin, but I can't seem to find 'del' anywhere.
>
> Python 1.5.2 (#12, Jun  2 1999, 19:23:30) [C] on irix646
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> dir(__builtins__)
> ['ArithmeticError', 'AssertionError', 'AttributeError', 'EOFError',
> 'Ellipsis', 'EnvironmentError', 'Exception', 'FloatingPointError',
> 'IOError', 'ImportError', 'IndexError', 'KeyError', 'KeyboardInterrupt',
> 'LookupError', 'MemoryError', 'NameError', 'None',
> 'NotImplementedError', 'OSError', 'OverflowError', 'RuntimeError',
> 'StandardError', 'SyntaxError', 'SystemError', 'SystemExit',
> 'TypeError', 'ValueError', 'ZeroDivisionError', '__debug__', '__doc__',
> '__import__', '__name__', 'abs', 'apply', 'buffer', 'callable', 'chr',
> 'cmp', 'coerce', 'compile', 'complex', 'delattr', 'dir', 'divmod',
> 'eval', 'execfile', 'exit', 'filter', 'float', 'getattr', 'globals',
> 'hasattr', 'hash', 'hex', 'id', 'input', 'int', 'intern', 'isinstance',
> 'issubclass', 'len', 'list', 'locals', 'long', 'map', 'max', 'min',
> 'oct', 'open', 'ord', 'pow', 'quit', 'range', 'raw_input', 'reduce',
> 'reload', 'repr', 'round', 'setattr', 'slice', 'str', 'tuple', 'type',
> 'vars', 'xrange']
> >>> range
> <built-in function range>
> >>> del
>   File "<stdin>", line 1
>     del
>        ^
> SyntaxError: invalid syntax
>
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.





From jam at newimage.com  Fri Jun  4 10:56:31 1999
From: jam at newimage.com (jam)
Date: Fri, 4 Jun 1999 10:56:31 -0400
Subject: Competing algorithms...
In-Reply-To: <y0jwvxkuode.fsf@vier.idi.ntnu.no>; from Magnus L. Hetland on Fri, Jun 04, 1999 at 02:44:13PM +0200
References: <y0jwvxkuode.fsf@vier.idi.ntnu.no>
Message-ID: <19990604105631.F24062@toast.internal>

On Fri, Jun 04, 1999 at 02:44:13PM +0200, Magnus L. Hetland wrote:
> 
> Yay!
> 
> I have managed to get the course on algorithms here at the university
> to use Python as its language!
> 

woowoo! ;)

> Now, I have to find some cool uses for it :) I am planning on setting
> up some programming competitions as a part of the course; i.e.
> students send their algorithms (in Python) to a (rexec-) server which
> benchmarks them and ranks them. So... Have you got any good ideas for
> fair benchmarking?
> 

hmm.. well, not really. I didn't find computing big-oh an exciting part of
school. ;)

> The evaluation should preferrably be done automatically, so the only
> thing I can think of is straight benchmarking baset on clocking and
> test cases... But is this a good and fair measure of how good the
> algorithm is? (Of course, a O(n) algorithm would probably out-perform
> an O(n^2) algorithm...)
> 
> It might also be possible to do manual evaluation of the best ones, or
> something - or to have a separate prize for beautiful programs...
> Anything is possible...
> 
> eagerly-awaiting-several-exciting-suggestions-ly y'rs
> 
>  -M

my idea would be to make a way to let the *students* rank themselves.. get
them to come up with a way to evaluate their peers.. this way the ones that
latch onto it quickly will end up teaching the ones that aren't as far along
yet.. and they will learn a lot more by seeing sometimes completely
different ways to solve the same problem. put the survey system up on the
web.. you can make a form that will allow them to post their code.. make
something that lets them authenticate to the system some how (simple
password will do), then after the assignments have been turned in they rank
the code of their peers based on some criteria (and have the students come
up with that, too).. say, a ranking system. you (as the organizer) get to
have the final call on what the grade is, so you can catch problems before
they become a problem.. the goal is to stir discussion in class about *why*
a certain algorithm is 'better'.. maybe one is way faster, but far less
beautiful, but no less correct than someones elses
beautiful-but-not-yet-fast-enough version-- they both deserve a passing
grade, but for different reasons.

setting-the-wheels-in-motion-ly y'rs,
J ;)
-- 
|| visit gfd <http://quark.newimage.com/> 
|| psa member #293 <http://www.python.org/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 210 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/19990604/ec63b42e/attachment.sig>

From dank at deming.com  Fri Jun 25 21:58:26 1999
From: dank at deming.com (dank at deming.com)
Date: Sat, 26 Jun 1999 01:58:26 GMT
Subject: getting memory information under NT
References: <002201beab54$66218d30$f29b12c2@pythonware.com> <003301beab56$88cb3af0$f29b12c2@pythonware.com> <375E88A2.68C1F118@opticominc.com>
Message-ID: <7l1c3v$l3o$1@nnrp1.deja.com>

In article <375E88A2.68C1F118 at opticominc.com>,
  Eric Renouf <erenouf at opticominc.com> wrote:
> Does anyone know of an easy way to get the the total physical memory
> and/or available memory on an NT system using Python 1.5.2?  Is there
a
> module that has functions for things like this already?
>
> Thanks,
> Eric Renouf
>
See function CeGlobalMemoryStatus() in module wincerapi, part of the
Python Win32 Extensions by Mark Hammond.  It is in the win32all download
available from www.python.org.  There is an example of usage in
demos\cerapi.py.

The win32 extensions also provide an interface to get performance
monitor data, which has several additional measures of memory usage.

Good luck,
Dan Knierim


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From gherron at aw.sgi.com  Tue Jun 15 12:46:23 1999
From: gherron at aw.sgi.com (Gary Herron)
Date: Tue, 15 Jun 1999 16:46:23 GMT
Subject: Tk Scrollbar
References: <37663C34.2957622F@dlr.de>
Message-ID: <3766835F.9C3DDB9C@aw.sgi.com>

Sven Drescher wrote:
> 
> Hi!
> I have only an easy question. I think it is possible to bind a scrollbar
> and an other widget e.g. a frame with buttons. But I have not found any
> literature or programming examples. I want to bind the scrollbar to an
> frame and I don't know how. Please help me.
> Thanks for links and Python-code or other help.
> 
> Sven

The communication between a widget and a scrollbar is carried out by two
procedures:
  
  1.  The scrollbar calls the widget's "yview" (or "xview") method to
cause it to scroll to a new position.

  2. The widget calls the scrollbar's "set" method to indicate
to invoke a change in the sliders position or size.

The only widgets which implement the "yview" method are "Canvas",
"Listbox", and "Text".  (And to be very clear here, this implementation
and the connection between the two widgets is at the tcl/tk level.)

So...
To add scrolling capabilities to an arbitrary widget, you would need to
implement the "yview" command (in its several versions) on the widget. 
I don't know if this would need to be implemented in tcl, or if you
could rig things so that the scrollbar's call to "yview" would come all
the way back into python so that you could implement it in python.

-- 
Dr. Gary Herron <gherron at aw.sgi.com>
206-287-5616
Alias | Wavefront
1218 3rd Ave, Suite 800, Seattle WA 98101



From xhyuan at oe.fau.edu  Fri Jun  4 09:29:12 1999
From: xhyuan at oe.fau.edu (Xiaohong Yuan)
Date: Fri, 4 Jun 1999 06:29:12 -0700 (PDT)
Subject: ask help for problem with static linking (fwd)
Message-ID: <Pine.SGI.3.92.990604062558.15744A-100000@sea30>

I am trying to link my c extension module staticly with python
interpreter. But in my c extension module, I have external variable
declaration, which makes it fail to be linked with Python interpreter.
There must be some way to solve this, could anybody help me with this?

Thank you very much.

Xiaohong Yuan


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




From clarence at silcom.com  Wed Jun 16 01:24:58 1999
From: clarence at silcom.com (Clarence Gardner)
Date: Wed, 16 Jun 1999 05:24:58 GMT
Subject: Compiled main program
Message-ID: <KyG93.386$dR.22261@newsfeed.avtel.net>

[Clarence, posting a request for help managed to find the solution
before committing himself -- but he's posting it anyway (with his
solution) just to see what people think...]


One of the applications that I've written in Python (in fact, the
first big one, a billing system for ISPs in operation for about
two years now) consists of about 120 main programs, all running
via CGI.  There is one that is particularly larger than the
rest (the main program is about 900 lines; all the rest are less
than 400) and, unlike the others, is used in a very 'interactive' way;
that is, you'll have hundreds or maybe thousands of http transactions
between your browser and it over the course of a half hour or an hour.  

Today I finally got around to something that's bugged me for a long
time, namely, the fact that Python is compiling that program every
one of those hundreds of times.  So I tried the obvious:

$ cat nph-pmtbatch
#!/usr/local/bin/python
import PmtBatchGuts
$

This worked fine (and with a noticeable speed increase) until the
execution path happened to intersect with one of various spots in my
various modules that reference a name in __main__ (it's an object that
gives access to the context of a login session).  Hoping against hope, I
made sure nobody was looking and put this at the top of PmtBatchGuts.py:

# Because this is really a main, but pre-compiled, program
__name__ = '__main__'

Didn't think it would work, and it didn't.  I figured that what I really
wanted to do was mess with sys.modules somehow, but I don't know how
to get a reference to the current module.  I assume that
	sys.modules['__main__'] = <this module>
would do it.  As distasteful as that is (do I hear retching in the back
of the room?), I like it much more than changing
	S = Session.Session()
to
	__main__.S = Session.Session()



Remembering that people have on occasion mentioned attributes that
seem to be invisible unless you know they're there, I've just been
trying random things and came up with this:

$ cat PmtBatchGuts.py
import sys

class MakeMeMain:
	def __init__(self):
        sys.modules['__main__'] = sys.modules[self.__module__]
x = MakeMeMain()
del x

import Session
S = Session.Session()
...


This works, but I'm interested in the what the community thinks of it.
TIA

-- 
-=-=-=-=-=-=-=-=
Clarence Gardner
AvTel Communications
Software Products and Services Division
clarence at avtel.com



From nathan at islanddata.com  Tue Jun 22 13:31:49 1999
From: nathan at islanddata.com (Nathan Clegg)
Date: Tue, 22 Jun 1999 10:31:49 -0700 (PDT)
Subject: renegotiating socket connections
Message-ID: <XFMail.990622103149.nathan@islanddata.com>

I'm writing a server application in python.  I understand that most
servers fork off a process for every incoming connection and then
renegotiate on a different port to free up the well-known server port.  I
would rather do a multithread model than a forked process model because of
shared memory needs.  I've heard that forking the process "automatically"
renegotiates the port somehow, but that doesn't make much sense to me.

More importantly, since I don't want to fork at all, how does one switch
to a different port?  Is this a manual process?  It seems it would be slow
to have so much round-trip communication before even establishing a
connection, and none of my network manuals say anything more than this
sort of thing "happens."  Any tips?


----------------------------------
Nathan Clegg
 nathan at islanddata.com





From csdayton+usenet at cs.uchicago.edu  Thu Jun 17 18:39:43 1999
From: csdayton+usenet at cs.uchicago.edu (Soren Dayton)
Date: Thu, 17 Jun 1999 22:39:43 GMT
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <37682DD3.43923243@compaq.com>
Message-ID: <xcdg13qo3io.fsf@sloohah.cs.uchicago.edu>

Greg Ewing <greg.ewing at compaq.com> writes:

> Fernando Mato Mira wrote:
> > 
> > I wanted to find out whether the disgust for tcl I've seen many people
> > express had some basis on some generally accepted principles.
> 
> I think I can give some reasonably objective reasons why
> I find that I enjoy using Python in a way that I don't
> enjoy using Tcl. (I have used both, and in one case have
> written versions of the same program in both languages.)
> 
> 1. Syntax
> 
> Tcl's syntax is based on the principle that an unadorned
> sequence of alphanumeric characters represents a literal.
> Anything else (such as referring to a variable) requires
> extra syntax.
> 
> Ousterhout's justification is that this is optimal
> for entering commands interactively, which is probably true.
> However, this seems to me the wrong thing to optimise for,
> given that tcl is meant to be a *programming* language.
> The case being optimised for -- typing in a command which
> is to be used once and thrown away -- simply doesn't occur.
> Programs, even tiny ones, get used more than once!

but it occurs to me that this is a GOOD argument for the best uses of
tcl, that is as an interactive style shell or scriptable configuration
file language. 

tcl _IS_ too annoying to do real programming in, but it is not clear
that it is not _by far_ the best tool for this kind of niche use.  (Yes, 
I've heard about guile, but frankly, tcl looks like my ipfilter
configuration files and we are _NOT_ after expressibility here but a
sort of clarity)

Soren



From paul at prescod.net  Wed Jun  2 09:50:58 1999
From: paul at prescod.net (Paul Prescod)
Date: Wed, 2 Jun 1999 13:50:58 GMT
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl>
Message-ID: <375536C2.DD917F99@prescod.net>

Martijn Faassen wrote:
> 
> Hi there,
> 
> I don't intend to start a flamewar or anything, and I'm sure Larry Wall
> is a nice guy, and Perl is nice and all, and the postmodern
> rationalisation of the confusion that Perl is to me is *fun* (but does
> not work for me), but Larry Wall seems to be infected by the whitespace
> eating nanovirus meme. 

The first time he mentioned this I thought he was joking. He attached a
"Hi Guido!" to indicate that he was. Now it seems he is serious. He thinks
that whitespace usage somehow goes to the heart of the way Python
programmers think about their craft -- as if in Python, whitespace is as
central as object orientation is in Java. A language designer should have
a better feeling for what is essential about a language and what is
trivial -- especially since he whines that people judge Perl too quickly
based on its dollar signs. If we were as fair to Perl as he is to Python
we would call it "Perl: the curly bracket and dollar sign language."

I consider it a compliment that he can't find anything real to point out
as Python's contribution to language fascism but I also find it
distressing that he feels the need to sow the seeds of fear, uncertainty
and doubt in the minds of would-be Python users. The Python feature that
most of us (in retrospect, unnecessarily) worry about before trying Python
is the whitespace handling. Larry can only turn people off of Python by
suggesting that that is Python's central feature or philosophy.

This whole modern, postmodern distinction is bogus anyhow. Almost every
language designer has a bone to pick with history and they are all pushing
some meme. They are all "modern" in this sense. Perl's meme is that
lexical flexibility overrides all other concerns. 

It is "classically modern" for the promoter of an ideology (in this case,
TMTOWTDI) to disclaim adherence to an ideology and claim inspiration in
the natural world: "Capitalism is just how things work. Perl works just
like human languages!" It is also charmingly modern to ignore the
unintended consequences of ideology: unreadable code, expensive
maintenance, wasted money and wasted time. Just as with communism, the
difficulty of truly measuring productivity will hide the problems for a
while but eventually the house will topple.

Now compare a Wall interview to one with Guido. Does Guido point to some
overriding ideology that makes Python great? Object Orientation? TMTOWTDI?
Whitespace? He promotes Python as a nice language that tries to
incorporate many neat features from other languages, to remain easy for
beginners and to scale nicely to complex problems. There is no single
dominant philosophy similar to TMTOWTDI that drives Python's growth. Or
one might say that Python's dominant philosophy is Aristotle's Golden
Mean. Or perhaps Python's philosophy is that code is communication and
readable code is clear communication. But even so, Python has many
features that exchange readability for flexibility...

I truly think that Perl is a unique and innovative language. Larry can
promote it without playing on people's fears of whitespace and fascism.

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for only himself
 http://itrc.uwaterloo.ca/~papresco

"Silence," wrote Melville, "is the only Voice of God." The assertion,
like its subject, cuts both ways, negating and affirming, implying both
absence and presence, offering us a choice; it's a line that the Society
of American Atheists could put on its letterhead and the Society of
Friends could silently endorse while waiting to be moved by the spirit
to speak. - Listening for Silence by Mark Slouka, Apr. 1999, Harper's



From mwh21 at cam.ac.uk  Mon Jun 21 19:17:54 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 22 Jun 1999 00:17:54 +0100
Subject: choosing random numbers with weights/probability?
References: <929987267.6227@www.remarq.com>
Message-ID: <m3g13l2lel.fsf@atrus.jesus.cam.ac.uk>

kevinsl <kevinsl at yahoo.com> writes:

> I've been using the whrandom.choice routine and it's very
> useful. But is there any way to add weights or
> probabilities of being chosen to the items in the list?
> 
> example:
> 
> list=['one','two','three']
> item=whrandom.choice(list)
> 
> Is there any way to say that 'one' and 'two' have a 25%
> chance of being chosen, and 'three' has a 50% chance?

One very obvious, not very extensible, way:

list=['one','two','three','three']
item=whrandom.choice(list)

(NB: list is a builtin; using list as a variable name can lead to
confusion)

A not very efficient but more general method:

def weighted_choice(choices):
    tot = 0
    for w,v in choices:
        tot = tot + w
    d = random.random()*tot
    tot = 0
    for w,v in choices:
        tot = tot + w
        if tot > d:
            return v

list=[(0.25,'one'),(0.25,'two'),(0.5,'three')]
weighted_choice(list)

seems to work. Haven't tested it much, so please don't rely on it...

HTH
Michael


> I'm hoping there's already a module to do this... or else
> I'll be writing my own..
> 
> thanks,
> 
> Kevin



From da at ski.org  Tue Jun 29 14:50:35 1999
From: da at ski.org (David Ascher)
Date: Tue, 29 Jun 1999 11:50:35 -0700 (Pacific Daylight Time)
Subject: BOFs (Birds-Of-a-Feather) sessions at the Monterey Python conference
Message-ID: <Pine.WNT.4.04.9906291133130.243-100000@rigoletto.ski.org>

As you may know, O'Reilly & Associates is organizing a Python Conference
as part of its Open Source Convention in Monterey in August.  (For more
information, see http://conferences.oreilly.com).  Early word from
O'Reilly is that the Python conference is getting quite a few
early registrants eager to get the discount & free t-shirts!

In addition to the tutorials, talks and panels, there will be informal
"Birds-Of-a-Feather" sessions in the tradition of Usenix BOFs.  These meet
at "off" times:  Saturday and Sunday from 7pm to 10pm, and Monday from 8pm
to 10pm -- meeting spaces will be made available by ORA.  The format for
these is deliberately left up to the birds -- however, one or more
organizers are expected to show up and help lead discussions.

I'd like to solicit discussion on what BOFs would be good to organize, and
find some organizers.  Topics that come to mind include field-specific
BOFs such as a Zope BOF, an image-processing BOF, a database BOF, etc.,
tool-BOFs (Tkinter, wxPython, ...) or "market segment" BOFs such as
"Python in Education", "Selling Python programs", etc.  Other ideas are
welcome.  Anyone can register a BOF through ORA's website at
http://conferences.oreilly.com/bof/, but it might make sense to talk about
it on the mailing list first to minimize overlaps and work out scheduling.

If a BOF is scheduled by July 15, it will be included in the conference
program given to attendees.  BOFs can be scheduled later, and will be
available on the conference website (as well as onsite, I imagine).

Looking forward to seeing many of you there,

--David Ascher & the rest of the Python Conference Organizing Committee

PS: If you have any questions about the Python conference, you can contact
    the conference staff at oscon at oreilly.com or you can contact me at
    da at python.net.  In addition, the folks at oscon at oreilly.com can mail
    you a nifty glossy printed brochure showing the entire schedule of the
    convention, all the talk descriptions, etc.





From kjwolf at online.de  Sat Jun 19 15:32:30 1999
From: kjwolf at online.de (Klaus-Juergen Wolf)
Date: Sat, 19 Jun 1999 21:32:30 +0200
Subject: Curses for MSDOS
References: <376BD33B.5A6F54BB@erols.com>
Message-ID: <e8rgk7.hj81.ln@solling.seismic.de>

Swartz <swartz at erols.com> wrote:
 > Is there anything like curses for MS-DOS?  Thanx.

I know of PC-Curses. Written in the early 80s. Maybe someone has
extended it, but in the state I had seen it had no NCurses extensions
and wouldn't work with Python.

Probably very few people used it, because that screen access mechanism
was relatively slow on MS-DOS.

cu
 k.j.



From def-p at usa.net  Fri Jun 18 17:54:11 1999
From: def-p at usa.net (Def P)
Date: Fri, 18 Jun 1999 21:54:11 GMT
Subject: Tkinter & socket question
Message-ID: <19990618215411.23898.qmail@nwcst277.netaddress.usa.net>

Hello,

I am writing a little chat program in Python and Tkinter. I can't figure out
two things though:

1) When there's no connection, I don't want people to use the Entry on the
window (which is used to enter text and send it after pressing Enter). So, I
want to disable it. I used entry.state = DISABLED, but I still can enter text.
Is this normal? Is there a way to make it "unusable"?

2) I need to poll the connection periodically, say, every second. This would
not have been a problem in console mode... I don't know how to do this in a
Tkinter environment though. So, how can I do something every second (or every
n milliseconds, or whatever the time span may be)? Are threads the way to go?
I haven't used them before.

Thanks,
Def P


____________________________________________________________________
Get free e-mail and a permanent address at http://www.netaddress.com/?N=1



From rozzin at geekspace.com  Fri Jun 25 19:27:19 1999
From: rozzin at geekspace.com (Joshua Rosen)
Date: Fri, 25 Jun 1999 19:27:19 -0400
Subject: Could someone please give an example?
References: <7kqpu8$7u1$1@nnrp1.deja.com> <3771127C.E40F9BF3@home.com> <7kvd8b$ueb$1@nnrp1.deja.com> <w1Lc3.1850$VL2.286812@news.direcpc.com>
Message-ID: <37741057.83741A41@geekspace.com>

Emile van Sebille wrote:
> 
> I'm new at this, but it seems I've read that the tracestack is
> available.  Might it be possible to read in the calling line number,
> then parse the source code for your 'name'?

Sure, but what'd be the point?
Let's say that you define a function to tell you the name:

def tn(inst):
    ....

Now, if you can call that function, with an instance as an argument, then you
already know what `name' you are calling it by.

If you've got an instanced produced and `stored' by something like...

class C:
  ...

i = C()

... then you need to call `tn(i)'--there you go, "i" is the name of the class,
eh?

If you want the program to decide which variable to look at, then you need to
use something like `eval(XXX)', where `XXX' is an expression that returns a
string, so, if you want to know the `name' of the variable, there, then you can
just turn to whatever `XXX' returned.

I've yet to see a reason why this `finding the name of an instance' is needed,
or what it accomplishes. Just by virtue of `the only name that an instance has
is what you give it', there really /shouldn't/ be a problem--if you give it a
name, you should be able to `remember' what the name is, or at least where you
stored it.

		Would someone please give an example of the use of this, assuming that we
have a function that does it (say..., `tn(xxx)')?

			-Rozzin.




From quinn at auto.ugcs.caltech.edu  Wed Jun 23 19:53:28 1999
From: quinn at auto.ugcs.caltech.edu (Quinn Dunkan)
Date: 23 Jun 1999 23:53:28 GMT
Subject: file system quotas
References: <Pine.LNX.4.10.9906211845450.684-100000@madre.servicom2000.com>
Message-ID: <slrn7n2srr.b6l.quinn@auto.ugcs.caltech.edu>

On Mon, 21 Jun 1999 18:53:45 +0200, vmas <vmas at servicom2000.com> wrote:
>Hi,
>I'm using Python for writing system administration scripts with a Unix
>system and I'd like to set quota limits for a given user. Do you know how
>to manipulate the file system quotas using Python? Or, is there a
>Quota module in Python?
>Thanks.

Various unices all seem to have their own way to do quotas.  Under Irix I
use os.system() and os.popen() to run the apropriate commands.  Generally,
if you're doing shellish sorts of things, these two functions are your friends.



From hniksic at srce.hr  Sat Jun  5 14:30:40 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 05 Jun 1999 20:30:40 +0200
Subject: Deleting characters from a string
References: <000101beaf74$646413e0$8c9e2299@tim>
Message-ID: <87bteu1ovj.fsf@pc-hrvoje.srce.hr>

"Tim Peters" <tim_one at email.msn.com> writes:

> It's (yours, not mine ...) a decent idea; in the meantime how about
> writing a little wrapper
> 
> def deletechars(s, todelete, id=string.maketrans('', '')):
>     return string.translate(s, id, todelete)

I think I'd rather write a patch that allows None as the second
argument to string.translate(), if that's OK with the developers.

Thanks for the tip, though.



From graham at sloth.math.uga.edu  Thu Jun  3 18:07:00 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 3 Jun 1999 22:07:00 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7j66b7$il5$2@cronkite.cc.uga.edu> <3756C693.2C910B29@usa.net> <7j6jnl$58f$3@cronkite.cc.uga.edu> <aahzFCrnB0.DIE@netcom.com>
Message-ID: <7j6ua4$fp6$1@cronkite.cc.uga.edu>

Graham Matthews <graham at sloth.math.uga.edu> wrote:
>Why is it so hard for people to see that adding a mark sweep collector
>*ON TOP OF* the existing ref count scheme essentially changes nothing.
>All the current ref count behaviour that people love will stay. The
>only additional behaviour will be that circular refs are mopped up
>correctly.
Aahz Maruch (aahz at netcom.com) wrote:
: I don't disagree with this; I only claim that GC should only happen by
: explicit call from the programmer.  I also claim that there should be a
: debug mode for GC that merely informs the programmer about cyclic
: references so that zie can locate their source (and that this should be
: the primary mode for GC).

I don't understand why you want GC only by explict call and in "inform"
mode only?

graham
-- 
          The fool escaped from paradise
        will look over his shoulder and cry
          Sit and chew on daffodils and
             struggle to answer why?



From graham at sloth.math.uga.edu  Fri Jun  4 10:45:04 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 4 Jun 1999 14:45:04 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net> <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net> <7j34l3$hbj$1@nnrp1.deja.com> <7j66b7$il5$2@cronkite.cc.uga.edu> <3756C693.2C910B29@usa.net> <7j6jnl$58f$3@cronkite.cc.uga.edu> <3757C181.477E0BD5@inrialpes.fr>
Message-ID: <7j8opg$55v$3@cronkite.cc.uga.edu>

Graham Matthews wrote:
> Why is it so hard for people to see that adding a mark sweep collector
> *ON TOP OF* the existing ref count scheme essentially changes nothing.
> All the current ref count behaviour that people love will stay. The
> only additional behaviour will be that circular refs are mopped up
> correctly.
Vladimir Marangozov (Vladimir.Marangozov at inrialpes.fr) wrote:
: Why don't you try implementing it?
: There's nothing more convincing than good working code.
: 
: If you try, I suspect that not we, but you'll realize that it's not so
: immediate and that there would be other additional effects, either
: inacceptable or undesirable.

How does one respond to the above? The old "I suspect" line. Well I don't
suspect because:

a) many many languages have garbage collection outlined exactly as I 
have proposed and they all work just fine. 

b) I work on a product which pretty much does exaclty what I outlined
for a collection scheme, and it works beautifully. Moreover it contains
many similarites with Python (the need to interface to C, complex
objects, etc). 

c) I have implemented 3 collectors in my "coding life", so I have some
experience.

You on the other hand just "suspect that I will realise that it's not so 
immediate"? Do you have any basis for these suspicions? For example have
you ever implemented a reference counting mark sweep collector, maybe
with colouring. 

I am sorry to sound sooo sarcastic, but what can I reply to "I suspect ...".
I put up facts, and you put up "I suspect ..."
: 
graham
-- 
                  A cat named Easter
              He say "will you ever learn"
              Its just an empty cage girl
                 If you kill the bird



From tim_one at email.msn.com  Wed Jun 23 23:43:37 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 23 Jun 1999 23:43:37 -0400
Subject: Python Exes!
In-Reply-To: <%fhc3.2901$Lk4.35205@news4.mia>
Message-ID: <000401bebdf3$be7e3f80$549e2299@tim>

[SunitJoshi]
> Could someone please point me to the URL where Gordon Mcmillians, 
> installer is located.

Start at

    http://www.mcmillan-inc.com/python.html

would-be-more-specific-but-the-server-appears-to-be-resting-ly y'rs  - tim





From chriseb at nortelnetworks.com  Mon Jun 14 11:52:15 1999
From: chriseb at nortelnetworks.com (Chris Ebenezer)
Date: 14 Jun 1999 16:52:15 +0100
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com>
Message-ID: <rnbn1y2n5jk.fsf@nortelnetworks.com>

Fernando Mato Mira <matomira at iname.com> writes:
: of folks end up using one or another GNU tool. As a side effect, some
: newcomers to Tk that would have normally used tcl could end up going
: the STk way (unfortunately, STk is not publicized enough). Things like
: Kawa also increase the feel of security regarding code reusability.
: And some of those people will end up wanting to write full-fledged

Why use TK at all ?  Once you are using scheme anyway there are nicer
widget toolkits to use (from a point of view of more "normal" looking
toolkits that don't suffer from color allocation problems),  there are
bindings for guile and gtk and guile and Qt. 

-- 
chris



From robin at alldunn.com  Tue Jun 29 03:07:59 1999
From: robin at alldunn.com (Robin Dunn)
Date: Tue, 29 Jun 1999 00:07:59 -0700
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <Pine.GSO.3.96.990625170056.28322B-100000@saga3.Stanford.EDU>
Message-ID: <Bh_d3.28$oY.1310857@news.randori.com>

Jeffrey Chang wrote in message ...
>On Fri, 25 Jun 1999, Mikael Lyngvig wrote:
>> I'd like some discussion on whether or not it is a good idea to
>> abandon TkInter, before it gets too widely used, and instead adopt
>> wxPython as the primary GUI framework for Python.
>
>Abandoning Tkinter seems a little bit extreme, since there are some
>applications for which its event model is well-suited.
>
>A few weeks ago I needed to build a viewer that displays binary trees
>whose branches have variable lengths (actually, phylogenetic trees for
>biological data).  Since the trees are large, I need to be able to slide
>it around the window.  In addition, I needed each node to be clickable so
>that the user can query for more data.
>
>Tkinter is very good for that type of thing.  All I needed to do was
>create a Canvas and then for each node, create an Oval with a callback to
>a function that handles node clicks.  Tk keeps track of the coordinates of
>each Item so that I don't have to.  Also, whenever I wanted to move the
>tree, all I had to do was:
>for widget in self._canvas.find_all():
>    self._canvas.move(widget, xoff, yoff)
>
>Of course, Tk incurrs an overhead by making everything you draw on the
>canvas its own object, but for this case, it made thigns a lot easier.
>
>Although I haven't used wxWindows before, but the documentation suggests
>that it would be a lot more difficult for this kind of application.  In
>wxWindows, I would need to draw each of my nodes into a window using a
>device context.  Then, I'd have to write a function that would handle
>mouse clicks in the window.  That function would need to get the
>coordinates and then figure out whether it intersects with any of my
>nodes, which Tk would've handled for me.  In addition, whenever I wanted
>to move the tree, I'd have to erase the window and redraw each node.  If I
>wanted that optimized, I'd need to write routines that would only redraw
>the uncovered parts of the tree, which means figuring out which nodes are
>newly revealed.
>
>From what other people have said about wxPython, it seems like it may be
>appropriate for other types of applications.  However, I'd hate to see
>Tkinter dumped because it's not appropriate for everything!
>
>Jeff
>





From lull at acm.org  Thu Jun 17 23:34:29 1999
From: lull at acm.org (John Lull)
Date: Fri, 18 Jun 1999 03:34:29 GMT
Subject: Newbie Q: How to call unmodified VB DLL from Python
Message-ID: <3769bcd1.54492213@news.earthlink.net>

As a first Python project, I'd like to use Python (or JPython) to
automate some hardware testing under Win95/98/NT.  The hardware in
question is run by a DLL written in VB, and to which I do not have
source.

"Extending & Embedding the Python interpreter", ch. 3, has
instructions for building a DLL which will work, but assumes you're
working in VC++ and have the source.

If anyone could point me toward instructions on how to make this work,
I'd be grateful.

Thanks.

Regards,
John




From nospam at bitbucket.com  Wed Jun 23 19:07:28 1999
From: nospam at bitbucket.com (Phil Mayes)
Date: Wed, 23 Jun 1999 16:07:28 -0700
Subject: Freeze?
References: <7kqcj4$3bd$1@nnrp1.deja.com>
Message-ID: <1Ndc3.648$dR.53682@newsfeed.avtel.net>

Curtis Yanko wrote in message <7kqcj4$3bd$1 at nnrp1.deja.com>...
>I have loaded 1.5.2 and the Win32All-125 but dont see the freeze tool
>anywhere. Searching the FTP site I only see Freeze for older versions.
>Hopefully I'm just confused but if somebody could give me a nudge I'd
>appreciate it.

It's not in the Python binaries - you need to get the Python source
(you won't need to build it, though).
--
Phil Mayes    pmayes AT olivebr DOT com








From bwarsaw at cnri.reston.va.us  Tue Jun  8 11:01:48 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Tue, 8 Jun 1999 11:01:48 -0400 (EDT)
Subject: waste of resources ?
References: <375C0F91.1C73E4E3@icrf.icnet.uk>
	<375C16A9.5ADC4990@icrf.icnet.uk>
	<375C2003.B01BEC3@arlut.utexas.edu>
	<87yahvzqz3.fsf@pc-hrvoje.srce.hr>
	<14172.34081.978798.911000@anthem.cnri.reston.va.us>
	<87n1ybxj3c.fsf@pc-hrvoje.srce.hr>
Message-ID: <14173.12380.479032.202036@anthem.cnri.reston.va.us>

>>>>> "Hrvoje" == Hrvoje Niksic <hniksic at srce.hr> writes:

    Hrvoje> It probably didn't exist in Python 1.5.1, under which I
    Hrvoje> was writing the code in question.

Oops, yes, EnvironmentError showed up in 1.5.2a1

-Barry



From lutz at rmi.net  Fri Jun 11 14:07:42 1999
From: lutz at rmi.net (Mark Lutz)
Date: Fri, 11 Jun 1999 18:07:42 GMT
Subject: "Programming Python" still worthwhile?
Message-ID: <199906111807.MAA13401@shell.rmi.net>

Curtis Yanko <cmyanko at bigfoot.com> writes:
>
> Isn't PP due for 2nd edition soon?

Frank Willison, O'Reilly's editor-in-chief, posted a 
few words on this subject at:

    http://www.oreilly.com/frank/python.html

An update will happen, but we're not yet sure about
the timing.  Given the way publishing works, even if an
update was started today, it probably wouldn't show up 
until early 2000 at best.  Depending on 2.0's release, 
that could make the update out of date in a year or less.

Given that, I doubt a 2nd edition will appear very soon.
OTOH, I may move up the schedule if 2.0 turns out to be 
far away or mostly compatible with 1.6.  I may also move
it up if I get bored with between-books slacking, but that
seems much less likely ;-).

For the time being, I maintain a list of recent Python
changes--which are mostly minor, in terms of the core
language--at my web site:

    http://rmi.net/~lutz/errata-python-changes.html

Consider it a new appendix.

--Mark Lutz  (http://rmi.net/~lutz)



From gmcm at hypernet.com  Wed Jun 16 12:25:57 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 16 Jun 1999 11:25:57 -0500
Subject: GC In Python: YAS (Yet Another Summary)
In-Reply-To: <FtO93.191$U6.1770@newsr2.twcny.rr.com>
Message-ID: <1282571877-108998356@hypernet.com>

Darrell writes:

> Guido's idea sounds least intrusive.
> 
> Is there any merit to building a tool to tell me if I have cycles
> and where they are ? Then I apply one of N strategies to kill them.
> 
> How often does the normal Joe have or care that he has cycles ? I
> find them hard to track down.

Check out Lars's plumbo.py:

<URL:http://www.stud.ifi.uio.no/~larsga/download/python/plumbo.py>

- Gordon



From falk.hueffner at student.uni-tuebingen.de  Tue Jun  8 07:38:16 1999
From: falk.hueffner at student.uni-tuebingen.de (Falk Hueffner)
Date: 08 Jun 1999 13:38:16 +0200
Subject: Higher level FTP lib?
Message-ID: <st2yahuorbr.fsf@mirjam.informatik.uni-tuebingen.de>

Hi,

I wonder if there is a FTP module for python that allows for higher
level operations than ftplib. With ftplib, I have to parse LIST lines
myself (footnote: is there a module for *that* like perl's
File::Listine?) if I want to know file size or date.

Due to the OO nature of Python I was pretty sure the FTP library would 
make FTP sites look like local directories. Would be really nice.

	Falk




From c.kotsokalis at noc.ntua.gr  Sun Jun  6 13:33:50 1999
From: c.kotsokalis at noc.ntua.gr (Constantinos A. Kotsokalis)
Date: 6 Jun 1999 17:33:50 GMT
Subject: oh, python gurus, shall thy help a beginner?
Message-ID: <slrn7llc6v.fc7.c.kotsokalis@ajax.noc.ntua.gr>

Heya,
  I've started with python and I've written the following script.
  It is supposed to use (a subclassed) ThreadedTCPServer and act
  like a little proxy server. It is working perfectly on linux
  where I am mainly developing it, but it denies to work properly
  under Solaris (at least 2.7 which I tried it on). I even recompiled
  Python, using gcc one time, SUN cc the other, trying with statically
  linked/shared modules, anything I could think of. It simply does
  not work. In both cases, it's threaded python 1.5.2.
  What happens under Solaris, is that the server starts all right,
  seems to be waiting for connections, but when I try to connect
  to that port it fails (connection refused). The 2 or 3 times that
  it _did_ connnect, the process was taking up to 95% of the CPU
  usage time right after (and then started not to accept connections
  again). I really can't find out anything, so if any of you might
  know what's wrong (or even speculate about it), please reply.

    Thanks!
	  Costas

PS: Any other suggestions are also welcomed, as well as code
contributions :-) This software acts like a proxy server that will
leave advertisment banners out of your sight, when browsing. If
anyone is interested in using it, the full distribution is at
http://www.softlab.ece.ntua.gr/~ckotso/CTC/

-----------------------START OF SCRIPT--------------------------
#!/usr/bin/env python

import os, time, re, SocketServer, urlparse, string, os, sys
try:
	import thread
except:
	print "You do not have a thread-enabled Python version."
	print "Please download and (re)compile the latest Python"
	print "version -- 1.5.2 at the moment -- with thread support."
	print "You will need to use the --with-thread option when you"
	print "./configure it.\n"
	sys.exit(0)
from socket import *

def SocketErrorMsg(msg, rem_srv, rem_port):
	err_htm_file = open(ERROR_HTML_FILE, "r")
	err_msg = err_htm_file.read(ERROR_HTML_SIZE)
	err_htm_file.close()
	err_msg = string.replace(err_msg, '%SERVER', rem_srv)
	err_msg = string.replace(err_msg, '%PORT', `rem_port`)
	err_msg = string.replace(err_msg, '%ERROR', `msg`)
	err_msg_size = len(err_msg)
	return_buff = "HTTP/1.1 403 ERROR\nServer: Cut The Crap\nContent-Length: "+`err_msg_size`+"\nContent-Type: text/html\nProxy-Connection: close\n\n\n"
	return_buff = return_buff+err_msg
	return return_buff

class MyTCPServer(SocketServer.ThreadingTCPServer):
	def server_bind(self):
		self.socket.setsockopt(SOL_SOCKET, SO_REUSEADDR, 1)
		self.socket.bind(self.server_address)

class handle_con:
	def __init__(req, sock, remaddr, srv):
		rhost, rport = remaddr
		req.proc_req(sock)

	def w_configure(self):
		None  # This will be used for web-based configuration

	def proc_req(self, sock):
		buff = sock.recv(BUFF_SIZE)
		if not len(buff):
			sys.exit(0)
		if string.find(buff, "POST ") == 0:
			time.sleep(1) # Sleep one sec to make sure all POST info has come
			sock.setblocking(0)
			try:
				while (1):
					buff = buff + sock.recv(BUFF_SIZE)
			except:
				None
			sock.setblocking(1)
		tokens = string.split(buff)
		if cre.search(tokens[1]):
			sock.send(ADV_RET_BUFF)
			sock.send(retgif)
			sys.exit(0)

		if ((tokens[1] == "http://ctc/") or (tokens[1] == "http://CTC/")):
			self.w_configure() # Not used yet
			sys.exit(0)

		server = socket(AF_INET, SOCK_STREAM, 0)
		if not len(PARENT_PROXY_HOST):
			url_tuple = urlparse.urlparse(tokens[1])
			host_port = url_tuple[1]
			if (string.find(host_port, ':') != -1):
				rem_srv, rem_port = string.split(host_port, ':')
				rem_port = string.atoi(rem_port)
			else:
				rem_srv, rem_port = host_port, 80
			if (len(url_tuple[4])):
				buff = string.replace(buff, tokens[1], url_tuple[2]+'?'+url_tuple[4])
			else:
				buff = string.replace(buff, tokens[1], url_tuple[2])
		else:
			rem_srv = PARENT_PROXY_HOST
			rem_port = PARENT_PROXY_PORT
		
		try:
			server.connect(rem_srv, rem_port)
			server.send(buff)
			retval = server.recv(BUFF_SIZE)
			while (retval):
				sock.send(retval)
				retval = server.recv(BUFF_SIZE)
			server.close()
		except error, msg:
			sock.send(SocketErrorMsg(msg[0], rem_srv, rem_port))

		sock.close()
		sys.exit(0)


VERSION = '0.2.4'

BIND_PORT = 5050
PARENT_PROXY_HOST = ''
PARENT_PROXY_PORT = 8080
WISH_FORK = 0

RETURN_GIF = "/usr/local/etc/white.gif"
ERROR_HTML_FILE = "/usr/local/etc/ctc_error.html"

cre_str = "\/ad\/|\/ads\/|\/ads2\/|\/ad\.|\/ads\.|\/ads2\.|\.ad\.|\.ads\.|\/ads1\.|\.ads1\.|\/adv\/|\/cadv\/|\/adserver\.|\/adlog\.|\.adbannercenter\.|\/nsads\.|adclick\.|\/banner\.|\/banner\/|\/adverts\/|\.linkexchange\.|\/acc_clickthru|\/adfinity|event\.ng|\/surfad|\/redirect\.|\/gen_addframe|\/m\=|\/ad\=|\/adclick\.|\/adfu\.|\?adnum\=|\.adzerver\.|\?advert\=|\/advert\/|\/SHOP\/|\/advertiser\/"

if (len(sys.argv) > 1):
	try:
		cf = open(sys.argv[1], "r")
	except:
		sys.stderr.write("Cannot open configuration file "+conf+"\n")
		sys.exit(0)
	lines = cf.readlines()
	cf.close()
	for line in lines:
		line = string.strip(line)
		if (len(line) == 0) or re.search("\s*#", line):
			continue
		if string.find(line, '#'):
			ltmp = string.split(line, '#')
			line = ltmp[0]
		tokens = re.split('\s+', line)
		if (tokens[0] == 'BIND_PORT'):
			BIND_PORT = string.atoi(tokens[1])
		elif (tokens[0] == 'PARENT_PROXY_HOST'):
			PARENT_PROXY_HOST = tokens[1]
		elif (tokens[0] == 'PARENT_PROXY_PORT'):
			PARENT_PROXY_PORT = string.atoi(tokens[1])
		elif (tokens[0] == 'WISH_FORK'):
			WISH_FORK = string.atoi(tokens[1])
		elif (tokens[0] == 'RETURN_GIF'):
			RETURN_GIF = tokens[1]
		elif (tokens[0] == 'ERROR_HTML_FILE'):
			ERROR_HTML_FILE = tokens[1]
		elif (tokens[0] == 'BLOCK'):
			re_add = re.escape(tokens[1])
			if (string.find(line, re_add) >= 0):
				continue
			cre_str = cre_str + '|' + re_add
		else:
			sys.stderr.write("Unknown option: "+tokens[0]+"\n")

BUFF_SIZE = 8192
MAX_COMMAND_BUFF = 1024

tmpval = os.stat(RETURN_GIF)
RETURN_GIF_SIZE = tmpval[6]

tmpval = os.stat(ERROR_HTML_FILE)
ERROR_HTML_SIZE = tmpval[6]

ADV_RET_BUFF = "HTTP/1.1 200 OK\nServer: Cut The Crap\nContent-Length: "+`RETURN_GIF_SIZE`+"\nProxy-Connection: close\nContent-Type: image/gif\n\n"

locf = open(RETURN_GIF, 'r')
retgif = locf.read(RETURN_GIF_SIZE)
locf.close()

addr = '127.0.0.1', BIND_PORT
srv = MyTCPServer(addr, handle_con)
cre = re.compile(cre_str, re.I)

if (WISH_FORK == 1):
	try:
		fv = os.fork()
	except:
		sys.stderr.write("Fork failed, exiting\n")
		sys.exit(0)
	
	if (fv != 0):  #Father process
		sys.exit(0)
	else: 
		None

print "Cut The Crap -- Version "+VERSION+" \"Goin' down\" launched..."
srv.serve_forever()
----------------------------END OF SCRIPT-------------------------

-- 
Constantinos A. Kotsokalis || c.kotsokalis at noc.ntua.gr
National Technical University of Athens
Network Management Center
Tel. No: +30 1 7721861



From sjturner at ix.netcom.com  Mon Jun  7 17:24:33 1999
From: sjturner at ix.netcom.com (Stephen J. Turner)
Date: Mon, 07 Jun 1999 17:24:33 -0400
Subject: Question about envvars
References: <7jgl9n$p13$1@nnrp1.deja.com> <375BFE38.3FB3E83C@ix.netcom.com> <7jh7pc$as$1@nnrp1.deja.com>
Message-ID: <375C3891.1D11E1A4@ix.netcom.com>

Hi Rob,

> Boy I was excited to get a reasonable explaination so quickly, but alas,
> it still does not work.  Here is my test script:
> 
> #!/usr/local/bin/python
> 
> import MDIreg
> 
> reg = MDIreg.FileRegistry(1,1)
> reg.exportEnvVars()
> 
> import os
> print "child has it : "
> os.system("env | grep MDI_REG_VERSION")
> print "but do I have it? :"
> print os.environ['MDI_REG_VERSION']
> 
> ---
> 
> I am interested in this "MDI_REG_VERSION" envvar which gets exported by
> my library.  Here is the output:
> 
> child has it :
> MDI_REG_VERSION=10.0
> but do I have it? :
> Traceback (innermost last):
>   File "regtest.py", line 12, in ?
>     print os.environ['MDI_REG_VERSION']
>   File "/usr/local/lib/python1.5/UserDict.py", line 12, in __getitem__
>     def __getitem__(self, key): return self.data[key]
> KeyError: MDI_REG_VERSION

Oops!  I just realized that the site module, which is imported by
default when the Python interpreter starts, imports the os module -- so
by the time your script begins to execute, it's already too late.  Just
for grins, you could try disabling the site module by adding the '-S'
option to the Python command line, as in:

#!/usr/local/bin/python -S

I'm not advocating this approach, since it's generally a good idea to
import the site module.  I'd just like to know if it works for you.

> Any other ideas?  Any way to tell the os/posix module to "refresh"?

None that I can see from looking in Modules/posixmodule.c.  Sorry.

> By the way, thanks for the library implementation suggestion, but I must
> keep it the way it is, because other c, and c++ (at the class level)
> programs access this library.

OK, how 'bout this?  Suppose you added a virtual "putenv" method to your
FileRegistry C++ class whose default implementation was to call the
global putenv function.  Then in your Python C extension, you could
subclass FileRegistry and override the putenv method with one that
modifies os.environ as described in the prior post.

Regards,
Stephen

--
Stephen J. Turner <sjturner at ix.netcom.com>



From claird at Starbase.NeoSoft.COM  Thu Jun 10 08:07:41 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 10 Jun 1999 07:07:41 -0500
Subject: Using perl in python
References: <7jjoqh$t56$1@nnrp1.deja.com> <6cs73.2432$nn.739920@news.shore.net>
Message-ID: <7jo9qd$hgi$1@Starbase.NeoSoft.COM>

In article <6cs73.2432$nn.739920 at news.shore.net>,
Michael P. Reilly <arcege at shore.net> wrote:
>Brenda <bbailey08 at my-deja.com> wrote:
>: I have a bunch of perl libraries that I would like to call in my python
>: scripts.  Is this possible?
>
>It is possible;  Tkinter utilizes an embedded Tk interpreter.  You
>would have to write a Perl interpreter.  There may be some people
>around who might help you with that.  I'm in the middle of embedding
>another interpreter myself.
>
>I've found that it is more efficient (and more fun) to convert the
>libraries to Python.  The code is invariably clearer.
			.
			.
			.
There's an even grosser approach that I've semi-automated.
I build Perl scripts as strings from within Python, and
send them out to an external (Perl) process for execution.

The automation part is that I have conventions for expressing
variable references so I can get data to flow conveniently
back and forth.  It's only "semi-", 'cause I tinker with
exception-handling and such most times I use it.

When do I use it?  Typically, when CPAN has something I
want in Python.  I do a cognate translation of the CPAN
interface into Python, and steal the implementation from our
Perl colleagues.  When it becomes important enough, or I
make time for fun, I rewrite the implementation in Python.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From MGoehmann at t-online.de  Wed Jun  9 04:46:14 1999
From: MGoehmann at t-online.de (Monika Göhmann)
Date: Wed, 9 Jun 1999 10:46:14 +0200
Subject: "Programming Python" still worthwhile?
References: <7jki66$qam$1@news.gstis.net>
Message-ID: <7jl9r8$i1j$1@news06.btx.dtag.de>

"Learning Python" is just great!! I have only started reading but it already
answered many of my newbie-questions. Some time ago I tried reading
"Programming Python" but I find it quite hard to digest. LP seems much more
readable for the beginner IMHO.

Monika

Rick Robino <see-sig at wavedivision.com> schrieb in im Newsbeitrag:
7jki66$qam$1 at news.gstis.net...
> Greetings,
>
> I've been watching this group for some time and just now getting warm to
> Python. Looking at source examples has been great, but for completeness'
> sake I'd like to get more of the "why" maybe found in a book. Since Python
> is at 1.5.2 and people are talking about 2.0 as if it's around the corner,
> I was wondering if anyone here could give me some advice as to whether or
> not "Programming Python" is useful, if the current edition is still
> current, if I should wait for the "Learning Python", or if I should use
> some other resources.
>
> I expect to come up to speed fairly quickly and I'm just looking for
> those fundamental insights into OOP and Python that can build up the
> kind of inuition and understanding one needs in order to "get
> comfortable" connecting/creating the dots. Currently I only program
> small - medium programs in C and shell (Korn). Any sage pointers to
> resources?
>
> TIA,
>
> --Rick
>                          ^^^^^^^^^^^^^^
> Rick Robino                                  rrobino at wavedivision.com





From matomira at iname.com  Fri Jun 11 08:33:33 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Fri, 11 Jun 1999 14:33:33 +0200
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <wk909syqr4.fsf@ifi.uio.no> <375FF969.C879B4CD@iname.com> <37601668.D9105225@mayo.edu> <3760E130.7456447@iname.com> <3760FA1F.7BCD@mailserver.hursley.ibm.com>
Message-ID: <3761021C.8C029142@iname.com>


Paul Duffin wrote:

>     In Tcl if you want to pass a "reference" you pass a variable name

I was refering to this, hence the quotes.






From pereira at research.att.com  Tue Jun 15 08:51:17 1999
From: pereira at research.att.com (Fernando Pereira)
Date: Tue, 15 Jun 1999 12:51:17 GMT
Subject: <insert favourite flame-bait here> broken
References: <1282792849-95706075@hypernet.com> <3764D7C3.718E125E@iname.com> <002f01beb64d$58e7b5f0$f29b12c2@pythonware.com> <37651951.ED54894E@acm.org> <005501beb674$c5575890$f29b12c2@pythonware.com> <376621CF.88C95B5E@iname.com>
Message-ID: <150619990851173299%pereira@research.att.com>

In article <376621CF.88C95B5E at iname.com>, Fernando Mato Mira
<matomira at iname.com> wrote:
> Fredrik Lundh wrote:
> 
> > btw, while we're on this topic, here's an interesting
> > article by kent pitman:
> >
> > http://world.std.com/~pitman/pfaq/cross-posting.html
> Well, I have no time for reading that. Obviously crossposts
> are useful. I'll tell you how it works:
> <snip: long off-topic ramblings>
No time? You do seem to have plenty of time to flood several language
newsgroups with off-topic postings like the present one. Revealing
priorities.

-- 
-- F



From Christian.Caremoli at der.edf.fr  Wed Jun 23 02:07:27 1999
From: Christian.Caremoli at der.edf.fr (Christian Caremoli)
Date: Wed, 23 Jun 1999 08:07:27 +0200
Subject: creating dictionnaries
References: <376F82A6.82497D2B@der.edf.fr> <199906222048.WAA08720@axil.hvision.nl>
Message-ID: <3770799F.BDAC8FC@der.edf.fr>

Thanks for all the answers
I know that it s invalid and what I use now is a function named _F to
avoid name clashes
def _F(**args):
   return args
What I want is to have an uniform syntax to write :
a=mycommand(x=1,y=2,z=(d=1,e=2),s=1)
or alike
instead of
a=mycommand(x=1,y=2,z=_F(d=1,e=2),s=1)
I wonder if it exists any syntactic sugar or any trick that could help me


Hans Nowak wrote:

> On 22 Jun 99, Christian Caremoli wrote:
>
> > Hi,
> > Dictionnaries can be created like that :
> > d={'a':1,'b':2}
> >
> > By calling a function you create a dictionnary like that :
> > def f(**d):
> >    return d
> >
> > d=f(a=1,b=2)
> >
> > I would like to be able to create dictionnaries with some similar syntax
> > like keyed tuples :
> > d=(a=1,b=2)
> >
> > Is there a way to do that ?
>
> Nope. :-) I wonder why you would want this? As far as I'm concerned,
> there's not screamingly much difference between
>
> d=f(a=1,b=2)    # valid
>
> and
>
> d=(a=1,b=2)     # invalid
>
> --Hans Nowak (ivnowa at hvision.nl)
> Homepage: http://fly.to/zephyrfalcon






From news at dorb.com  Sun Jun  6 02:12:15 1999
From: news at dorb.com (Darrell)
Date: Sun, 6 Jun 1999 02:12:15 -0400
Subject: pyscript help
References: <BRS53.233$h14.1175@newsr2.twcny.rr.com>
Message-ID: <Tlo63.546$h14.2355@newsr2.twcny.rr.com>

Life is good.
After an evening of install reboot.....
I got Aaron Watters Calculator Demo to work.
D:\python\win32comext\axscript\demos\client\ie\demo.htm
<SCRIPT LANGUAGE="Python">
Removed the "<!--"


and "-->"
</SCRIPT>

I just wanted to see an ASP something or other work.

Replaced comcat.dll with an older one.
NT4 SP5
IE5
And I'm afraid to install option pack 4.

--Darrell





From wtanksle at dolphin.openprojects.net  Tue Jun 22 21:26:01 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Wed, 23 Jun 1999 01:26:01 GMT
Subject: Help with String Manipulation
References: <5650A1190E4FD111BC7E0000F8034D26A0EE67@HUINA>
Message-ID: <slrn7n0dt9.gh6.wtanksle@dolphin.openprojects.net>

On Wed, 23 Jun 1999 01:06:25 GMT, Doug Stanfield wrote:
>If you're new to Python it might seem like some magic in this:

>>>> testring = "This is an example"
>>>> testlist = map (ord, testring)
>>>> testlist.reverse()
>>>> import array
>>>> finalstring = array.array('b', testlist).tostring()

Whew!  Is there any reason to not use:

>>> x = list("This is an example.")
>>> x.reverse()
>>> import string
>>> string.join(x,"")

??

In other words, since a string is immutable, just convert the string to a
list explicitly.

If you MUST be obscure, try this:

>>> x = map(None, "This is an example.")
>>> x.reverse()

This way you don't have to play with ordinals.

>-Doug-

-- 
-William "Billy" Tanksley
Utinam logica falsa tuam philosophiam totam suffodiant!
   :-: May faulty logic undermine your entire philosophy!



From tismer at appliedbiometrics.com  Tue Jun 29 05:41:56 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Tue, 29 Jun 1999 09:41:56 GMT
Subject: ANN: Stackless Python 0.2
References: <000301bec1ff$22ac6fe0$719e2299@tim>
Message-ID: <377894E4.4897D2E8@appliedbiometrics.com>


Tim Peters wrote:
> 
> [Tim sez Python stores space for each code object's "Python stack" in a
>  runtime frame object]
> 
> [Toby J Sargeant wondering about frame allocs]
...
> > It seems that this should affect Stackless Python equally as
> > much.
> 
> No more so than Stackful Python:  Python has always worked this way;
> Stackless Python doesn't change this aspect.

And this was one of the temptations to start this at all.
We pay already some amount of runtime for the frame thing.
Then, we should make full use of it. Doing not so would
be a waste and not optimum.

> 
> > Does anyone have anything to add on the subject? I would imagine that
> > frames could be allocated and managed in chunks to alleviate a lot of
> > the memory management load...
> 
> The code is in Objects/frameobject.c, and easy to follow.  frameobjects are
> recycled  via their own free list.  Typically the total number of
> frame-associated mallocs is proportional to the maximum depth of the call
> stack, not to the number of calls made.
> 
> Setting up a Python frame remains expensive, but for other reasons.

The last word on this isn't spoken yet, since I didn't try
to optimize frames. But I will see what to do (later).

There are two things, which at the moment are still too
intermingled. I will try to break things apart more in the
next stackless verison. Currently, a frame is constructed
from a code object, by examining the code object's properties.
This is done in frameobject.c. The engine then modifies
the local registers in the frame object, according to
parameters of the given function call.

This is by far not optimum. If a function is called repeatedly
in the same context, the frame could stay initialized and be
reused. I believe there are very many contexts where just
one or two arguments of a function call change, and the
frame would only need to update these two fast locals!

Maybe caching this a little, by code object and caller
location would pay off. Will think more about it.

tim-this-is-continuation-prediction - ly y'rs  - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From jblake at eskimo.com  Wed Jun  9 19:35:22 1999
From: jblake at eskimo.com (Jonathon)
Date: 9 Jun 1999 23:35:22 GMT
Subject: Python Paterns
Message-ID: <7jmtnq$fc7$1@eskinews.eskimo.com>

Searching Python.org, I didn't come across 
anything _recent_ about 
patterns in Python.  Did that SIG die, get swallowed up
or did I miss their home?

The only documentation I could find was 
http://starship.python.net/crew/gandalf/DNET/classes/index.html
which was for Python 1.4 on an NT machine, last updated in
1997 and a paper given at the sixth conference on Python.


	xan

	jonathon


-- 
	I'm still looking for a good book on
		3:	The Recent Unpleasantness
		1:	The War Of Northern Aggression.
		2:	The War of Southern Rebellion.



From mfletch at vrtelecom.com  Tue Jun 29 03:43:46 1999
From: mfletch at vrtelecom.com (Mike C. Fletcher)
Date: Tue, 29 Jun 1999 07:43:46 GMT
Subject: PythonWin VSS Integration
Message-ID: <002a01bec203$1fc90230$a9807018@yec1.on.wave.home.com>

I've just finished setting up visual source safe here, and thought that I
would take the opportunity to see what PythonWin's integration was like.
However, not knowing what it is supposed to do, I don't know what I am
supposed to do, or whether it is working.  I was assuming that the
integration would be something similar to what visual studio does, i.e.,
when I attempt to edit a file in PythonWin, visual source safe would prompt
me to checkout the file.  That doesn't seem to be the case.

So, any hints?  What does visual source safe integration mean?

Thanks,
Mike




From scott at chronis.pobox.com  Wed Jun  2 13:34:43 1999
From: scott at chronis.pobox.com (scott cotton)
Date: Wed, 2 Jun 1999 13:34:43 -0400
Subject: metaclasses and setting __class__
Message-ID: <19990602133443.A3846@chronis.pobox.com>

Last night i was playing with meta classes, and found out
that we are not able to assign an instance's __class__
attribute to a metaclass instance.  was just wondering if
this is probably always going to be the case, or if there
are any plans to make .__class__ assignable to such beasts.

scott



From catlee at my-deja.com  Tue Jun 29 15:15:36 1999
From: catlee at my-deja.com (catlee at my-deja.com)
Date: Tue, 29 Jun 1999 20:15:36 +0100
Subject: Arbitrary precision calculations
Message-ID: <0000158b@bossar.com.pl>

From: catlee at my-deja.com

Hi,

I'm looking for a package that will let me do arbitrary precision
arithmatic in python.  I've come across the real.py module written by
Jurjen N.E. Bos, and it seems like a really nice module except for the
fact that you can't compare two numbers for equality,inequality,etc.

Cheers,
Chris

P.S. I found the real.py package by doing a search on www.deja.com for
arbirary precision in comp.land.python


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From mikael at pobox.com  Wed Jun 30 20:48:11 1999
From: mikael at pobox.com (Mikael Lyngvig)
Date: Thu, 01 Jul 1999 00:48:11 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773c3a0.102825445@news.oh.verio.com> <xPPc3.535$or1.22866@typhoon-sf.snfc21.pbi.net> <xJRc3.292$Of3.67094@news.shore.net> <7laovo$c4s$1@Starbase.NeoSoft.COM> <377A24B7.AD6FAA1A@cfdrc.com>
Message-ID: <3A174E56AD89F7F6.47941772B9D545B9.852EB0CCFCA68614@lp.airnews.net>

On Wed, 30 Jun 1999 09:07:51 -0500, Lyle Johnson <jlj at cfdrc.com>
wrote:

>Just to add fuel to the fire, we now have a Python interface to FOX, a
>cross-platform GUI toolkit:
>
>    http://www.cfdrc.com/pub/FOX/fox.html
>
>The latest version of the Python wrappers can be picked up at:
>
>    ftp://ftp.cfdrc.com/pub/FOX/FXPy-0.99.42.tar.gz
>
>I don't know how the speed of FXPy would compare with some of the others
>mentioned (Tk, wxWindows, Qt, GTK, ...) but it would be worthwhile for
>some independent opinions.

I'm downloading it now.  The http URL is wrong, it should be
http://www.cfdrc.com/FOX/fox.html.

The presentation seems promising, but I'd like to suggest a binary
distribution for PC/Intel computers.  I know this is not common for
Unix distributions, but it's the norm for PCs.


-- Mikael




From eugene.leitl at lrz.uni-muenchen.de  Mon Jun 28 14:32:14 1999
From: eugene.leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Mon, 28 Jun 1999 11:32:14 -0700 (PDT)
Subject: SWIGging VolPack (voxel rendering library)
Message-ID: <14199.48948.94150.900645@lrz.de>

Has anybody SWIGed VolPack here yet?
http://www-graphics.stanford.edu/software/volpack/ 

'cause if nobody did, I'd like to give it a spin.



From MHammond at skippinet.com.au  Wed Jun  2 03:25:14 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Wed, 2 Jun 1999 17:25:14 +1000
Subject: IDLE Extension: CallTips
Message-ID: <7j2m5m$kuf$1@m2.c2.telstra-mm.net.au>

Humming the mantra "Release early and release often", I humbly submit my
first attempt at an IDLE extension.

Some of you may be wondering if I have a) seen the light, or b) lost my
mind, depending on your viewpoint <wink> But fear not - Guido and I have
been working on allowing IDLE extensions to exist in Pythonwin, so this
extension will indeed be used by the next version of Pythonwin (in fact, was
ported from the Pythonwin version)

CallTips is an extension that uses a pop-up window similar to "Tool Tips" or
"Balloon Help", and displays the parameter description for functions as you
open a paren.  The text for the description is derived from the function
object itself if possible, and the function's docstring.

It has the following problems:

* It is not incredibly smart about determining the function to use to
display the tips.  Simple things like "string.split" and "os.path.split"
work correctly, but many other function calls, such as "self.foo" will not
invoke any help at all.

* The tips may hang around longer than you expect.  Specifically,
back-spacing after the window is open, or moving to a new location with the
mouse will not close the window.  Almost all other cursor movement keys
will, so it should not hang around too long.  Any Tkinter experts with
advice on how to fix this would be appreciated!

* Only been tested on WinNT, but has absolutely no Windows specific code -
should work fine on any IDLE platform.  The CallTip window code is based on
Guido's ToolTips.py, which is based on PySol - so it should work anywhere!

* Doesn't handle nested calls correctly - eg, closing a nested function call
will not re-display the help for the outer call.  This is slowly being
addressed.

* No way to disable the CallTips or delay their display (other than to
remove the extension, of course)

For all this, I still find it incredibly useful.  Plus the code is small
enough to attach :-)

The instructions are very simple:
* Copy the attached files to your IDLE directory.
* Open extend.py, and add "CallTips" to the list of extensions.  See
extend.txt for details
* Restart IDLE.  From the PyShell window type "string.split(" and see it do
it's stuff!

All comments and enhancements appreciated!

Mark.


begin 666 CallTipWindow.py
M(R!!($-A;&Q4:7 @=VEN9&]W(&-L87-S(&9O<B!4:VEN=&5R+TE$3$4N#0HC
M($%F=&5R(%1O;VQ4:7 N<'DL('=H:6-H('5S97,@:61E87, at 9VQE86YE9"!F
M<F]M(%!Y4V]L#0H-"B, at 57-E9"!B>2!T:&4 at 0V%L;%1I<', at 241,12!E>'1E
M;G-I;VXN#0II;7!O<G0@;W,-"F9R;VT at 5&MI;G1E<B!I;7!O<G0@*@T*#0IC
M;&%S<R!#86QL5&EP. at T*#0H@(" @9&5F(%]?:6YI=%]?*'-E;&8L('=I9&=E
M="DZ#0H@(" @(" @('-E;&8N=VED9V5T(#T@=VED9V5T#0H@(" @(" @('-E
M;&8N=&EP=VEN9&]W(#T at 3F]N90T*(" @(" @("!S96QF+FED(#T at 3F]N90T*
M(" @(" @("!S96QF+G@@/2!S96QF+GD@/2 P#0H-"B @("!D968@<VAO=W1I
M<"AS96QF+"!T97AT*3H-"B @(" @(" @<V5L9BYT97AT(#T@=&5X= T*(" @
M(" @("!I9B!S96QF+G1I<'=I;F1O=R!O<B!N;W0@<V5L9BYT97AT. at T*(" @
M(" @(" @(" @<F5T=7)N#0H@(" @(" @('-E;&8N=VED9V5T+G-E92 at B:6YS
M97)T(BD-"B @(" @(" @>"P@>2P at 8W@L(&-Y(#T@<V5L9BYW:61G970N8F)O
M>"@B:6YS97)T(BD-"B @(" @(" @>" ]('@@*R!S96QF+G=I9&=E="YW:6YF
M;U]R;V]T>"@I("L@, at T*(" @(" @("!Y(#T@>2 K(&-Y("L@<V5L9BYW:61G
M970N=VEN9F]?<F]O='DH*0T*(" @(" @("!S96QF+G1I<'=I;F1O=R ]('1W
M(#T at 5&]P;&5V96PH<V5L9BYW:61G970I#0H@(" @(" @('1W+G=M7V]V97)R
M:61E<F5D:7)E8W0H,2D-"B @(" @(" @='<N=VU?9V5O;65T<GDH(BLE9"LE
M9"(@)2 H>"P@>2DI#0H@(" @(" @(&QA8F5L(#T at 3&%B96PH='<L('1E>'0]
M<V5L9BYT97AT+"!J=7-T:69Y/4Q%1E0L#0H@(" @(" @(" @(" @(" @(" @
M(" @8F%C:V=R;W5N9#TB(V9F9F9E,"(L(')E;&EE9CU33TQ)1"P at 8F]R9&5R
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M0V]D90T*(PT*8VQA<W, at 8V]N=&%I;F5R.B C($-O;F-E<'1U86QL>2!A;B!E
M9&ET;W)?=VEN9&]W#0H@(" @9&5F(%]?:6YI=%]?*'-E;&8I. at T*(" @(" @
M("!R;V]T(#T at 5&LH*0T*(" @(" @("!T97AT(#T@<V5L9BYT97AT(#T at 5&5X
M="AR;V]T*0T*(" @(" @("!T97AT+G!A8VLH<VED93U,1494+"!F:6QL/4)/
M5$@L(&5X<&%N9#TQ*0T*(" @(" @("!T97AT+FEN<V5R="@B:6YS97)T(BP@
M(G-T<FEN9RYS<&QI="(I#0H@(" @(" @(')O;W0N=7!D871E*"D-"B @(" @
M(" @<V5L9BYC86QL=&EP(#T at 0V%L;%1I<"AT97AT*0T*#0H@(" @(" @('1E
M>'0N979E;G1?861D*"(\/&-A;&QT:7 M<VAO=SX^(BP@(B at B*0T*(" @(" @
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M="!I;B!I9&QE#0H-"B @("!D968 at 8V%L;'1I<%]S:&]W*'-E;&8L(&5V96YT
M*3H-"B @(" @(" @<V5L9BYC86QL=&EP+G-H;W=T:7 H(DAE;&QO('=O<FQD
M(BD-"@T*(" @(&1E9B!C86QL=&EP7VAI9&4H<V5L9BP at 979E;G0I. at T*(" @
M(" @("!S96QF+F-A;&QT:7 N:&ED971I<"@I#0H-"F1E9B!M86EN*"DZ#0H@
M(" @(R!497-T(&-O9&4-"B @("!C/6-O;G1A:6YE<B at I#0H-"FEF(%]?;F%M
>95]?/3TG7U]M86EN7U\G. at T*(" @(&UA:6XH*0T*
`
end

begin 666 CallTips.py
M:6UP;W)T('-T<FEN9PT*:6UP;W)T('-Y<PT*:6UP;W)T('1Y<&5S#0H-"B,@
M5&AE<V4@:V5Y<R!C:&5C:R!I9B!C86YC96QL871I;VX@:7,@;F5S97-S87)Y
M+ at T*7V-H96-K7V-A;F-E;%]K97ES(#T at 6R),969T(BP@(E)I9VAT(ET-"@T*
M(R!4:&5S92!K97ES("AW:71H('1H92!M;V1I9FEE<G,I(&-A;&P at 8V%N8V5L
M('5N8V]N9&ET:6]N86QL>2X-"E]C86YC96Q?:V5Y<U]B87-E(#T at 6R)5<"(L
M(")$;W=N(BP@(DYE>'0B+" B4')I;W(B+" B2&]M92)=#0I?8V%N8V5L7VME
M>7-?;6]D:69I97)S(#T at 6R=+97DM)RP@)T-O;G1R;VPM)RP@)U-H:69T+2<L
M("=#;VYT<F]L+5-H:69T+2==#0H-"E]C86YC96Q?:V5Y<R ](%M=#0IF;W(@
M:V5Y(&EN(%]C86YC96Q?:V5Y<U]B87-E. at T*(" @(&9O<B!M;V0@:6X at 7V-A
M;F-E;%]K97ES7VUO9&EF:65R<SH-"B @(" @(" @7V-A;F-E;%]K97ES+F%P
M<&5N9"AM;V0K:V5Y*0T*#0IC;&%S<R!#86QL5&EP<SH-"@T*(" @(&UE;G5D
M969S(#T at 6PT*(" @(%T-"@T*(" @(&ME>61E9G,@/2![#0H@(" @(" @("<\
M/'!A<F5N+6]P96X^/B<Z(%LG/$ME>2TH/B==+ T*(" @(" @(" G/#QP87)E
M;BUC;&]S93X^)SH at 6R<\2V5Y+2D^)UTL#0H@(" @(" @("<\/&-H96-K+6-A
M;&QT:7 M8V%N8V5L/CXG.B!M87 H;&%M8F1A(&ME>3H@(CQ+97DM)7,^(B E
M(&ME>2P at 7V-H96-K7V-A;F-E;%]K97ES*2P-"B @(" @(" @)SP\8V%L;'1I
M<"UC86YC96P^/B<Z(&UA<"AL86UB9&$@:V5Y.B B/"5S/B(@)2!K97DL(%]C
M86YC96Q?:V5Y<RDL#0H@(" @?0T*#0H@(" @=VEN9&]W<U]K97ED969S(#T@
M>PT*(" @('T-"@T*(" @('5N:7A?:V5Y9&5F<R ]('L-"B @("!]#0H-"B @
M("!D968 at 7U]I;FET7U\H<V5L9BP at 961I='=I;BDZ#0H@(" @(" @('-E;&8N
M961I='=I;B ](&5D:71W:6X-"B @(" @(" @<V5L9BYT97AT(#T at 961I='=I
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M;%1I<"AS96QF+G1E>'0I#0H-"B @("!D968 at 7W)E;6]V95]C86QL=&EP7W=I
M;F1O=RAS96QF*3H-"B @(" @(" @:68@<V5L9BYC86QL=&EP. at T*(" @(" @
M(" @(" @<V5L9BYC86QL=&EP+FAI9&5T:7 H*0T*(" @(" @(" @(" @<V5L
M9BYC86QL=&EP(#T at 3F]N90T*(" @(" @(" -"B @("!D968@<&%R96Y?;W!E
M;E]E=F5N="AS96QF+"!E=F5N="DZ#0H@(" @(" @('-E;&8N7W)E;6]V95]C
M86QL=&EP7W=I;F1O=R at I#0H@(" @(" @(&%R9U]T97AT(#T at 9V5T7V%R9U]T
M97AT*'-E;&8N9V5T7V]B:F5C=%]A=%]C=7)S;W(H*2D-"B @(" @(" @:68@
M87)G7W1E>'0Z#0H@(" @(" @(" @("!S96QF+F-A;&QT:7!?<W1A<G0@/2!S
M96QF+G1E>'0N:6YD97 at H(FEN<V5R="(I#0H@(" @(" @(" @("!S96QF+F-A
M;&QT:7 @/2!S96QF+E]M86ME7V-A;&QT:7!?=VEN9&]W*"D-"B @(" @(" @
M(" @('-E;&8N8V%L;'1I<"YS:&]W=&EP*&%R9U]T97AT*0T*(" @(" @("!R
M971U<FX@(B(@(W-O('1H92!E=F5N="!I<R!H86YD;&5D(&YO<FUA;&QY+ at T*
M#0H@(" @9&5F('!A<F5N7V-L;W-E7V5V96YT*'-E;&8L(&5V96YT*3H-"B @
M(" @(" @(R!.;W<@:G5S="!H:61E<RP at 8G5T(&QA=&5R('=E('-H;W5L9"!C
M:&5C:R!I9B!O=&AE<@T*(" @(" @(" C('!A<F5N)V0 at 97AP<F5S<VEO;G,@
M<F5M86EN(&]P96XN#0H@(" @(" @('-E;&8N7W)E;6]V95]C86QL=&EP7W=I
M;F1O=R at I#0H@(" @(" @(')E='5R;B B(B C<V\@=&AE(&5V96YT(&ES(&AA
M;F1L960@;F]R;6%L;'DN#0H-"B @("!D968 at 8VAE8VM?8V%L;'1I<%]C86YC
M96Q?979E;G0H<V5L9BP at 979E;G0I. at T*(" @(" @(" C(%1H:7, at 9&]E<VYT
M('%U:71E('=O<FL at 8V]R<F5C=&QY(&%S(&ET(&ES('!R;V-E<W-E9 T*(" @
M(" @(" C(%]B969O<F5?('1H92!K97D@:7,@:&%N9&QE9"X@(%1H=7,L('=H
M96X@=&AE(")5<"(@:V5Y#0H@(" @(" @(",@:7,@<')E<W-E9"P@=&AI<R!T
M97-T(&AA<'!E;G, at 8F5F;W)E('1H92!C=7)S;W(@:7,@;6]V960N#0H@(" @
M(" @(", at 5&AI<R!W:6QL(&1O(&9O<B!N;W<N#0H@(" @(" @(&EF('-E;&8N
M8V%L;'1I<#H-"B @(" @(" @(" @(", at 268@=V4@:&%V92!M;W9E9"!B969O
M<F4@=&AE('-T87)T(&]F('1H92!C86QL=&EP+ T*(" @(" @(" @(" @(R!O
M<B!O9F8@=&AE(&-A;&QT:7 @;&EN92P@=&AE;B!C86YC96P@=&AE('1I<"X-
M"B @(" @(" @(" @(",@*$QA=&5R(&YE960@=&\@8F4@<VUA<G1E<B!A8F]U
M="!M=6QT:2UL:6YE+"!E=&,I#0H@(" @(" @(" @("!I9B!S96QF+G1E>'0N
M8V]M<&%R92 at B:6YS97)T(BP@(CP](BP@<V5L9BYC86QL=&EP7W-T87)T*2!O
M<B!<#0H@(" @(" @(" @(" @("!S96QF+G1E>'0N8V]M<&%R92 at B:6YS97)T
M(BP@(CXB+"!S96QF+F-A;&QT:7!?<W1A<G0@*R B(&QI;F5E;F0B*3H-"B @
M(" @(" @(" @(" @("!S96QF+E]R96UO=F5?8V%L;'1I<%]W:6YD;W<H*0T*
M(" @(" @("!R971U<FX@(B(@(W-O('1H92!E=F5N="!I<R!H86YD;&5D(&YO
M<FUA;&QY+ at T*#0H@(" @9&5F(&-A;&QT:7!?8V%N8V5L7V5V96YT*'-E;&8L
M(&5V96YT*3H-"B @(" @(" @<V5L9BY?<F5M;W9E7V-A;&QT:7!?=VEN9&]W
M*"D-"B @(" @(" @<F5T=7)N("(B("-S;R!T:&4 at 979E;G0@:7,@:&%N9&QE
M9"!N;W)M86QL>2X-"@T*(" @(&1E9B!G971?;V)J96-T7V%T7V-U<G-O<BAS
M96QF+ T*(" @(" @(" @(" @(" @(" @(" @(" @(" @("!W;W)D8VAA<G,]
M(BY?(B K('-T<FEN9RYU<'!E<F-A<V4@*R!S=')I;F<N;&]W97)C87-E("L@
M<W1R:6YG+F1I9VET<RDZ#0H@(" @(" @(", at 6%A8("T at 5&AI<R!N965D('1O
M(&)E(&UO=F5D('1O(&$@8F5T=&5R('!L86-E#0H@(" @(" @(", at 87,@=&AE
M("(N(B!A='1R:6)U=&4@;&]O:W5P(&-O9&4 at 8V%N(&%L<V\@=7-E(&ET+ at T*
M(" @(" @("!T97AT(#T@<V5L9BYT97AT#0H@(" @(" @(&-H87)S(#T@=&5X
M="YG970H(FEN<V5R="!L:6YE<W1A<G0B+" B:6YS97)T(BD-"B @(" @(" @
M:2 ](&QE;BAC:&%R<RD-"B @(" @(" @=VAI;&4@:2!A;F0 at 8VAA<G-;:2TQ
M72!I;B!W;W)D8VAA<G,Z#0H@(" @(" @(" @("!I(#T@:2TQ#0H@(" @(" @
M('=O<F0@/2!C:&%R<UMI.ET-"B @(" @(" @:68@=V]R9#H-"B @(" @(" @
M(" @(", at 2&]W(&ES('1H:7, at 9F]R(&$@:&%C:R$-"B @(" @(" @(" @(&EM
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M<PT*(" @(" @("!R971U<FX at 3F]N92 C($-A;B=T(&9I;F0 at 86X@;V)J96-T
M+ at T*#0ID968 at 9V5T7V%R9U]T97AT*&]B*3H-"B @(" C($=E="!A('-T<FEN
M9R!D97-C<FEB:6YG('1H92!A<F=U;65N=', at 9F]R('1H92!G:79E;B!O8FIE
M8W0N#0H@(" @87)G5&5X=" ]("(B#0H@(" @:68@;V(@:7,@;F]T($YO;F4Z
M#0H@(" @(" @(&%R9T]F9G-E=" ](# -"B @(" @(" @(R!B:70@;V8 at 82!H
M86-K(&9O<B!M971H;V1S("T@='5R;B!I="!I;G1O(&$@9G5N8W1I;VX-"B @
M(" @(" @(R!B=70@=V4 at 9')O<"!T:&4@(G-E;&8B('!A<F%M+ at T*(" @(" @
M("!I9B!T>7!E*&]B*3T]='EP97,N365T:&]D5'EP93H-"B @(" @(" @(" @
M(&]B(#T@;V(N:6U?9G5N8PT*(" @(" @(" @(" @87)G3V9F<V5T(#T@,0T*
M(" @(" @(" C(%1R>2!A;F0 at 8G5I;&0@;VYE(&9O<B!0>71H;VX at 9&5F:6YE
M9"!F=6YC=&EO;G,-"B @(" @(" @:68@='EP92AO8BD@:6X at 6W1Y<&5S+D9U
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M5&5X=" ]('-T<FEN9RYJ;VEN*"!M87 H;&%M8F1A(&%R9RP at 9&9L=#H at 87)G
M*V1F;'0L(')E86Q!<F=S+"!D969A=6QT<RDL("(L("(I#0H@(" @(" @(" @
M(" @(" @:68@;&5N*')E86Q!<F=S*2MA<F=/9F9S970@/" H;&5N*&]B+F9U
M;F-?8V]D92YC;U]V87)N86UE<RD at +2!L96XH;V(N9G5N8U]C;V1E+F-O7VYA
M;65S*2 I. at T*(" @(" @(" @(" @(" @(" @("!I9B!A<F=497AT.B!A<F=4
M97AT(#T at 87)G5&5X=" K("(L("(-"B @(" @(" @(" @(" @(" @(" @87)G
M5&5X=" ](&%R9U1E>'0@*R B+BXN(@T*(" @(" @(" @(" @(" @(&%R9U1E
M>'0@/2 B*"5S*2(@)2!A<F=497AT#0H@(" @(" @(" @("!E>&-E<'0Z#0H@
M(" @(" @(" @(" @(" @<&%S<PT*(" @(" @(" C($-A;B=T(&)U:6QD(&%N
M(&%R9W5M96YT(&QI<W0 at +2!S964@:68@=V4 at 8V%N('5S92!A(&1O8W-T<FEN
M9RX-"B @(" @(" @:68@:&%S871T<BAO8BP@(E]?9&]C7U\B*2!A;F0@;V(N
M7U]D;V-?7SH-"B @(" @(" @(" @('!O<R ]('-T<FEN9RYF:6YD*&]B+E]?
M9&]C7U\L(")<;B(I#0H@(" @(" @(" @("!I9B!P;W,\,"!O<B!P;W,^-S Z
M('!O<STW, T*(" @(" @(" @(" @:68 at 87)G5&5X=#H at 87)G5&5X=" ](&%R
M9U1E>'0@*R B7&XB#0H@(" @(" @(" @("!A<F=497AT(#T at 87)G5&5X=" K
M(&]B+E]?9&]C7U];.G!O<UT-"@T*(" @(')E='5R;B!A<F=497AT#0H-"B,C
M(R,C(R,C(R,C(R,C(R,C(R,C(R,C(R,C(R,C(R,C(R,C(R,C(R,C(R,C(R,C
M(R,-"B,-"B, at 5&5S="!C;V1E#0HC#0II9B!?7VYA;65?7ST])U]?;6%I;E]?
M)SH-"@T*(" @(&1E9B!T,2 at I.B B*"DB#0H@(" @9&5F('0R*&$L(&(]3F]N
M92DZ("(H82P at 8CU.;VYE*2(-"B @("!D968@=#,H82P@*F%R9W,I.B B*&$L
M("XN+BDB#0H@(" @9&5F('0T*"IA<F=S*3H@(B at N+BXI(@T*(" @(&1E9B!T
M-2AA+" J87)G<RDZ("(H82P at +BXN*2(-"B @("!D968@=#8H82P at 8CU.;VYE
M+" J87)G<RP@*BIK=RDZ("(H82P at 8CU.;VYE+" N+BXI(@T*#0H@(" @8VQA
M<W, at 5$,Z#0H@(" @(" @(&1E9B!T,2AS96QF*3H@(B at I(@T*(" @(" @("!D
M968@=#(H<V5L9BP at 82P@8CU.;VYE*3H@(BAA+"!B/4YO;F4I(@T*(" @(" @
M("!D968@=#,H<V5L9BP at 82P@*F%R9W,I.B B*&$L("XN+BDB#0H@(" @(" @
M(&1E9B!T-"AS96QF+" J87)G<RDZ("(H+BXN*2(-"B @(" @(" @9&5F('0U
M*'-E;&8L(&$L("IA<F=S*3H@(BAA+" N+BXI(@T*(" @(" @("!D968@=#8H
M<V5L9BP at 82P@8CU.;VYE+" J87)G<RP@*BIK=RDZ("(H82P at 8CU.;VYE+" N
M+BXI(@T*#0H@(" @9&5F('1E<W0H('1E<W1S("DZ#0H@(" @(" @(&9A:6QE
M9#U;70T*(" @(" @("!F;W(@="!I;B!T97-T<SH-"B @(" @(" @(" @(&EF
M(&=E=%]A<F=?=&5X="AT*2 A/2!T+E]?9&]C7U\@*R B7&XB("L@="Y?7V1O
M8U]?. at T*(" @(" @(" @(" @(" @(&9A:6QE9"YA<'!E;F0H="D-"B @(" @
M(" @(" @(" @("!P<FEN=" B)7, at +2!E>'!E8W1E9" E<RP at 8G5T(&=O=" E
M<R(@)2 H="P at 8'0N7U]D;V-?7V L(&!G971?87)G7W1E>'0H="E@*0T*(" @
M(" @("!P<FEN=" B)60@;V8@)60@=&5S=', at 9F%I;&5D(B E("AL96XH9F%I
M;&5D*2P@;&5N*'1E<W1S*2D-"@T*(" @('1C(#T at 5$,H*0T*(" @('1E<W1S
M(#T@=#$L('0R+"!T,RP@=#0L('0U+"!T-BP at 7 T*(" @(" @(" @(" @=&,N
M=#$L('1C+G0R+"!T8RYT,RP@=&,N=#0L('1C+G0U+"!T8RYT- at T*#0H@(" @
-=&5S="AT97-T<RD-"@``
`
end




From billtut at microsoft.com  Sat Jun 19 16:27:59 1999
From: billtut at microsoft.com (Bill Tutt)
Date: Sat, 19 Jun 1999 20:27:59 GMT
Subject: NT
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8100DB90DF7@RED-MSG-50>

No, you've got it right. When you use the file association for some unknown
reason you can't use redirection.
If you just do: "python blah.py < blah > blah2" everything works perfectly.

My current theory is that there is a design point/bug in the file
association code that we just don't understand at the momemnt.

Bill

> -----Original Message-----
> From: Holger Jannsen [mailto:holger at phoenix-edv.netzservice.de]
> Sent: Friday, June 18, 1999 9:02 AM
> To: python-list at cwi.nl
> Subject: NT
> 
> 
> Hi,
> 
> another problem under NT: I'd like to message the user a 
> longer description about the
> programm with sys.stdout.write(usagetxt).
> The user should have possibility to pipe the output like he
> wants, e.g. with 'less' or 'more'.
> But I get an error:
> 
> C:\myPython>installation15.py /? |more
> Traceback (innermost last):
>   File "C:\myPython\installation15.py", line 642, in ?
>     switcher()
>   File "C:\myPython\installation15.py", line 629, in switcher
>     __usage()
>   File "C:\myPython\installation15.py", line 189, in __usage
>     sys.stdout.write(usagetxt)
> IOError: [Errno 9] Bad file descriptor
> 
> If I start with 'C:\myPython>python installation15.py /? |more'
> there's no problem and it works allright.
> 
> I've read the messages about NT and piping in this group, but I think
> it's something differnt, ouh?!
> 
> hand (='have a nice day'!),
> Holger
> 



From da at ski.org  Mon Jun 14 01:25:26 1999
From: da at ski.org (David Ascher)
Date: Sun, 13 Jun 1999 22:25:26 -0700 (Pacific Daylight Time)
Subject: New to threading. Right idea???
In-Reply-To: <37648cec.0@news.dbtech.net>
Message-ID: <Pine.WNT.4.05.9906132220070.103-100000@david.ski.org>

On Sun, 13 Jun 1999, Karl Putland wrote:

> I need some advise.  Am I on the right track?

Not quite.  As you say, threads are independent paths of execution.  The
key word there is "path of execution" =).  In other words, everything but
the path of execution is shared, especially globals, and in Python, the
state of the interpreter.  The two threads share the same list of imported
modules, etc.

What you're thinking of is processes, not threads.  On unix, that's done
with a fork().  There is no simple equivalent on Windows, alas.  Not that
it would help (see below).

> One of the functions in the dll is EXPENSIVE.  It's not mine and I doubt
> they will rework it so I have to live with it.  the EXPENSIVE function must
> be called at the begining and ties all activity in the dll to a specific
> directory.  If you want to change directories you must call unexpensive then
> EXPENSIVE again with the new directory as an argument.  If the dll is loaded
> and EXPENSIVE has been called sucessive calls to EXPENSIVE return FALSE.

If you must call EXPENSIVE for each directory, then you must -- I don't
think there's any way around that.  Loading the same DLL multiple times
won't help if you have to call EXPENSIVE for each (especially if
unexpensive is, well, unexpensive =).  If EXPENSIVE is VERY_EXPENSIVE,
where VERY_EXPENSIVE >> cost of loading the DLL, then even a fork()
wouldn't help much, even if it were available on your OS.

--david ascher





From fredrik at pythonware.com  Fri Jun 18 05:12:01 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 18 Jun 1999 09:12:01 GMT
Subject: tkwait
References: <3769FE25.6ECF09D4@dlr.de>
Message-ID: <000701beb96a$a15d70c0$f29b12c2@pythonware.com>

Sven Drescher <Sven.Drescher at dlr.de> wrote:
> I know, that is a very simple question, but I didn't find a solution.
> Please help me with the syntax of Tk's tkwait window. I found the Tk
> Manual page and Deja-news entries, but no solution for my problem.

http://www.pythonware.com/library/tkinter/introduction/basic-widget-methods.htm
(look under event processing).

btw, a PDF version of the current tkinter introduction
draft is now available from:

    http://www.pythonware.com/library.htm

</F>




From bwinton at tor.dhs.org  Wed Jun  9 23:42:26 1999
From: bwinton at tor.dhs.org (Blake Winton)
Date: Thu, 10 Jun 1999 03:42:26 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7j90va$6lk$1@cronkite.cc.uga.edu> <u3g1443iht.fsf@ares.sys1.ngo.okisoft.co.jp> <87pv384koy.fsf@ev.netlab.co.jp> <u3909slwbo.fsf@ares.sys1.ngo.okisoft.co.jp>
Message-ID: <slrn7lud8t.g2c.bwinton@tor.dhs.org>

On 10 Jun 1999 09:43:07 +0900, Hisao Suzuki <suzuki611 at okisoft.co.jp> wrote:
>In article <87pv384koy.fsf at ev.netlab.co.jp>,
>Yukihiro Matsumoto <matz at netlab.co.jp> wrote:
>| I claimed in summary:
>|   * I don't like mere ref counting; it cannot reclaim cyclic data, and
>|     easy to forget INCREF/DECREF.
>|   * Many among Python users misunderstood about real GC.
>|     (slow, hangs, non portable, etc.)
>|   * ref counting with GC may be good for Python 2.0.
>| Did I do any advocacy on Ruby?  I was trying constructive discussion.
>I'm sorry to say, but your argument is too trivial (or rather
>naive) to consider as a constructive discussion.  Having almost
>nothing to do with real problems on Python, your referrings to
>your Ruby (surely in triumphant tones) sound evangelic.  I'd be
>glad if you would present a non-trivial viewpoint or propose a
>practical solution _relevant_ to Python actually.

I think this is a little harsh.  I've been reading this entire thread
(No, I don't know why), and Matz has seemed to me to be remarkably
restrained in his evangelical attempts.  Whenever he mentions Ruby, he
makes the disclaimer that it's not particularly portable, and only seems
to use it as a proof-of-concept, or demonstration that he has some skill
in writing GCs.

Ref counting with GC sounds to me like a good idea for Python 2.0, as it
will eliminate a class of memory leaks that Python can currently fall
prey to.  As long as you don't move stuff around in memory, and only
collect things where all the refcounts are accounted for, and can't be
reached, you should be safe.  Granted, this means that the people
writing the extensions will have to be careful to INCREF/DECREF where
they're supposed to, but I don't see that as a bad thing since they
should already be careful about that sort of stuff.

Later,
Blake.

-- 
One Will.  One Dream.  One Truth.  One Destiny.  One Love.



From claird at Starbase.NeoSoft.COM  Thu Jun 10 12:00:40 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 10 Jun 1999 11:00:40 -0500
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <4.1.19990610150908.00d09790@exchsrv> <7jog8f$ljv$1@Starbase.NeoSoft.COM> <375FD854.2314D886@iname.com>
Message-ID: <7jonf8$6n6$1@Starbase.NeoSoft.COM>

In article <375FD854.2314D886 at iname.com>,
Fernando Mato Mira  <matomira at iname.com> wrote:
			.
			.
			.
>> >If so, does speed ever become an issue? I assume that Tk works well
>> Rarely.  Much less often than the uninitiated
>> suspect.
>
>That's the impression I have. But then, what are those `less often' cases?
>(Besides trying to do silly things, like writing a fast-paced  video game in Tk)
			.
			.
			.
Tk's bad at pixel-oriented graphics; its [canvas]
is designed for higher-order manipulations (and
can be breathtakingly effective for many inter-
esting tasks).  People coming from other paradigms
can usually find ways to misuse Tk into sluggish-
ness.  The Unicode stuff in 8.1 is still being
shaken down; severe performance issues have turned
up there, but plenty of fixes are available and
are gradually being worked into the core.

If you're doing things that fit well in [canvas]es
and [text]s, you'll go far with Tk.  Several
widgets that moderns expect are *not* available
in the standard core, and some of the extensions
that provide them are not high in performance.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From news at dorb.com  Sun Jun 27 12:24:52 1999
From: news at dorb.com (Darrell)
Date: Sun, 27 Jun 1999 12:24:52 -0400
Subject: Python for large projects?
References: <3774b5d1.690922@news.force9.net>  <T6bd3.2883$U6.12014@newsr2.twcny.rr.com> <slrn7naup9.jk.thantos@brimstone.mecha> <RGid3.2957$U6.12374@newsr2.twcny.rr.com> <87n1xm85mf.fsf@lisboa.i-did-not-set--mail-host-address--so-shoot-me>
Message-ID: <Jhsd3.3090$U6.12410@newsr2.twcny.rr.com>

> If it is only one object, one can try to wrap it in a class and
> overwrite setattr, setitem etc.
> This would inhibit that one changes the buffer data accidently. And
> put the data into an immutable sequence type (string, tuple).
>
> __Janko
>
I'll give that a look, but they can always get to it if they want. Maybe
that's ok, just worries me.

--Darrell





From tcropley at nospam.prodigy.net  Fri Jun 11 20:55:10 1999
From: tcropley at nospam.prodigy.net (Phaedrus42)
Date: Fri, 11 Jun 1999 20:55:10 -0400
Subject: Newbie question: DOS console commands?
Message-ID: <7jsb5s$k7q$1@newssvr01-int.news.prodigy.com>

I'm new to Python this week and was wondering if there are any console
commands to use on a DOS screen (e.g. ClearScreen, GotoXY, Foreground and
Background colors, etc.)

I've got PythonWin running on Windows 95 and am having fun learning the
language and Tkinter, but I have a need to write a simple DOS interface
program.  I could pull out my copy of TurboPascal, but I'd like to write it
in Python to further my education.

Any help?





From scott at chronis.pobox.com  Tue Jun 22 23:09:51 1999
From: scott at chronis.pobox.com (scott cotton)
Date: Tue, 22 Jun 1999 23:09:51 -0400
Subject: help?  strptime ....
In-Reply-To: <rNHvjEdhm5Pp-pn2-dCsXnyPPztJy@d185d18f8.rochester.rr.com>; from Benjamin Schollnick on Wed, Jun 23, 1999 at 02:33:56AM +0000
References: <rNHvjEdhm5Pp-pn2-dCsXnyPPztJy@d185d18f8.rochester.rr.com>
Message-ID: <19990622230951.A11935@chronis.pobox.com>

>>> import time
>>> print time.strptime.__doc__
strptime(string, format) -> tuple
Parse a string to a time tuple according to a format specification.
See the library reference manual for formatting codes (same as strftime()).
>>>
>>>
>>> time.strptime("19990822", "%Y%m%d")
(1999, 8, 22, 0, 0, 0, 6, 1, 0)
>>> time.mktime(time.strptime("19990822", "%Y%m%d"))
935294400.0

scott



On Wed, Jun 23, 1999 at 02:33:56AM +0000, Benjamin Schollnick wrote:
| Folks,
| 
| 	I'm trying to use strptime (from the time module) and can't....
| 
| Could someone please post a example, so that I can try to figure out 
| what
| I'm doing wrong?
| 
| 			- Benjamin
| 
| -- 
| http://www.python.org/mailman/listinfo/python-list



From c.evans at clear.net.nz  Sun Jun  6 01:52:00 1999
From: c.evans at clear.net.nz (Carey Evans)
Date: 06 Jun 1999 17:52:00 +1200
Subject: Sockets and lemon curry
References: <qDf63.375$os.20566@typ42b.nn.bcandid.com>
Message-ID: <87d7z97u67.fsf@psyche.evansnet>

"Ian King" <iking at killthewabbit.org> writes:

> I'm trying to implement an rshd server in Python.  The problem seems to be
> with establishing the secondary socket for stderr.

[snip]

> s2 = socket(AF_INET, SOCK_STREAM)
> s2.connect(addr[0], port2)    #addr tuple is host, port for conn
> 
> ....and that's where the client says, "Pining for the fjords???" and gives
> up.

I think you've missed something.  The OS will assign the local port
for this socket from 1024 up.  However, rsh requires that the remote
port that connects to it for stderr is less than 1024, to prove that
it's not just another user on the box pretending to be rshd, but a
process running as root.

You'll have to have the Python server bind() to a reserved port before 
trying to connect back to rsh.  Something like the following code may
be appropriate, and acts like the code the real rshd uses.

------------------------------------------------------------
import exceptions
import socket
import errno
import types

class RResvPortError(exceptions.Exception):
    def __str__(self):
        return 'No reserved ports available'

def rresvport():
    port = 1023
    s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
    while port >= 1:
        try:
            s.bind('', port)
        except socket.error, detail:
            if type(detail) is not types.TupleType \
               or detail[0] != errno.EADDRINUSE:
                raise
        else:
            return s
        port = port - 1

    raise RResvPortError
------------------------------------------------------------

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

	     "I'm not a god.  I've just been misquoted."



From holger at phoenix-edv.netzservice.de  Tue Jun 15 10:02:48 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Tue, 15 Jun 1999 14:02:48 GMT
Subject: Newbie: easy filecopy?
References: <376531D3.7E42AC42@phoenix-edv.netzservice.de>
Message-ID: <37665D08.267A697C@phoenix-edv.netzservice.de>

Another ascpect to my question of yesterday:

The problem with 
f=open(aFile)
x=f.read()
f.close
f.open(toFile)
f.write(x)
f.close
is, that some special ASCII-codes are not copied as usual 
under Windows-systems, so e.g. CarriageReturn (tested with normal
Asccii-text).

Hints?

Ciao,
Holger

Holger Jannsen schrieb:
> 
> Sorry,
> it's me again.;-)
> 
> I couldn't believe it. I think there must be an easy copy-function
> in python (I only want to copy files under ext2- or FAT-FileSystems).
> Call me "4 blind mice"-Joey, but I couldn't find it.
> For sure, I could write a thing like open-read-write-close, but I
> think, someone else has done already, hasn't he?
> 
> Thanx again for patience with me...[snuff]
> 
> Ciao,
> Holger



From mwh21 at cam.ac.uk  Thu Jun 10 06:10:19 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 10 Jun 1999 11:10:19 +0100
Subject: Asserting object class
References: <375F8986.73BA3C5@i-data.com>
Message-ID: <m3wvxc9xis.fsf@atrus.jesus.cam.ac.uk>

Adi <adi at i-data.com> writes:

> In a method, I'd like to assert that a parameter is an object which is
> of a specific class or has this class as it's base class.
> 
> I can test on the __class__ and its __bases__ attributes, but its clumsy
> and I'm wondering if there is an easier notation, like:
> 
> def method(self, p)
>     assert class(p) == class(self)
> 
> This is comparable to - in a strongly typed language - specifying a
> required base class as the type of the parameter.

I think that this will do what you want:

def method(self,p):
    assert isinstance(p,self.__class__)

the following also does the same:

def method(self,p):
    assert issubclass(p.__class__,self.__class__)

HTH
Michael



From aahz at netcom.com  Wed Jun 30 16:39:30 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 30 Jun 1999 20:39:30 GMT
Subject: Does the first argument of a method *has to be* self?
References: <19990630222121.A1121@optiplex.palga.uucp>
Message-ID: <7ldva2$mhq@dfw-ixnews21.ix.netcom.com>

In article <19990630222121.A1121 at optiplex.palga.uucp>,
Gerrit Holl  <gerrit.holl at pobox.com> wrote:
>
>I never saw code where the first argument of a method wasn't self. Is it
>in theory possible?

Yes and no.  The first argument always *is* self, but it can be named
something else.  For example:

class foo :
	def bar(s):
		s.bar()  # recursive call
--
                      --- Aahz (@netcom.com)

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



From chriseb at nortelnetworks.com  Thu Jun 17 06:04:18 1999
From: chriseb at nortelnetworks.com (Chris Ebenezer)
Date: 17 Jun 1999 11:04:18 +0100
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <rnbn1y2n5jk.fsf@nortelnetworks.com> <sFU93.1567$kS.233449@ptah.visi.com>
Message-ID: <rnbpv2v5ejh.fsf@nortelnetworks.com>

thornley at visi.com (David Thornley) writes:
: >Why use TK at all ?  Once you are using scheme anyway there are nicer
: >widget toolkits to use (from a point of view of more "normal" looking
: >toolkits that don't suffer from color allocation problems),  there are
: >bindings for guile and gtk and guile and Qt. 
: >
: The attraction, to me, would be that Tk is a free, open-source
: graphics environment that will run on Unix with X-Windows, the
: Macintosh, and Microsoft Windows.  CLIM is commercial, Garnet last

Hi David,

If you want something that is completely free (in the GNU sense) go to
http://www.gtk.org/, scroll down until you reach the bit on the sidebar
that reads "Language Bindings".  You should be able to get guile-gtk
from there.
 
-- 
chris



From matomira at iname.com  Tue Jun 15 10:37:46 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Tue, 15 Jun 1999 16:37:46 +0200
Subject: <insert favourite flame-bait here> broken
References: <1282792849-95706075@hypernet.com> <3764D7C3.718E125E@iname.com> <002f01beb64d$58e7b5f0$f29b12c2@pythonware.com> <37651951.ED54894E@acm.org> <005501beb674$c5575890$f29b12c2@pythonware.com> <376621CF.88C95B5E@iname.com> <150619990851173299%pereira@research.att.com>
Message-ID: <3766653A.D9F6E99B@iname.com>


Fernando Pereira wrote:

> In article <376621CF.88C95B5E at iname.com>, Fernando Mato Mira
>
> No time? You do seem to have plenty of time to flood several language

No time to read a philosophy of `closed mindness' that I don't share.
What is off-topic to you is not to everyone. For comp.lang, I rarely
look outside of comp.lang.lisp. Even comp.object, which would be
a natural environment for me was going unread. And if a comp.lang.misc
exists, who said the `people that know' looks there?

If all you do is code in your own favorite language, this kind of
discussion
will obviously not be intersting to you.

And nobody still gave me a reason why I should look beyond
the www.python.org front page (I'd already been there).

I'm out of here. If one of the previous flamers wants to send me email
regarding
that question, you're welcome to do so. But I won't waste any more time
defending myself and showing other points of view regarding crossposting
to dogmatic people that think that when an authority puts up a web page,
it must obviously be `the truth'.
Actually, you should read that better. The first sentence says
`generally'.

Obviously, comp.lang.python falls into the generality. No one from
any other group complained, even when I explicitly pointed out something
that "they" (i.e. I'm part of them) might not like. What does that say
about you?

[The cool people around (including _every_single_one_ who stood up for
tcl)
  please excuse me (specially Guido, since I have a hunch he must be a
really cool guy).
  I'm not talking about you. But really, this infuriates me so much,
  if someone asks me what I think about Python I'll probably just answer
"Don't go there"]

You can cheer now.





From dickey at shell.clark.net  Mon Jun 21 05:26:14 1999
From: dickey at shell.clark.net (T.E.Dickey)
Date: Mon, 21 Jun 1999 09:26:14 GMT
Subject: Problem with Curses
References: <ppgmj7.s7r.ln@solling.seismic.de> <jpeck7.1661.ln@solling.seismic.de> <ve1dk7.1e02.ln@solling.seismic.de> <87ogidzs37.fsf@home.ivm.de> <6Zeb3.526$4e1.32195@iad-read.news.verio.net> <5c5kk7.fn.ln@solling.seismic.de>
Message-ID: <Wynb3.608$4e1.36336@iad-read.news.verio.net>

Klaus-Juergen Wolf <kjwolf at online.de> wrote:
> T.E.Dickey <dickey at shell.clark.net> wrote:
>  >>> It depends on when the executable is linked to libmytinfo, which only
>  >>> takes place on platforms that don't know the terminfo database, like
>  >>> BSD and Solaris; there are two competing declarations of tputs().

>  >> Why doesn't Freebsd know the terminfo database? Can't the one that is 
>  >> integrated into ncurses be used on FreeBSD? 

>  > I've been told ncurses 4.2 is incorporated in the FreeBSD 3.1 - but that
>  > may still just be a port...

> Yes, there is such a port. Do you think, one could release code that
> works on several machines, while some of them must install a special
> program first?

I don't understand this comment (you seem to be saying that because there
is some reluctance to install a working version of ncurses, that there's
no reason to use packages which use it).

> Curses now is a very old UI, very old, it was old when I first learned
> it in the beginning of the 80's. Most people have forgotten that there

not very old at that point (iirc, it dates to the late 70's, and was still
considered "new" when I encountered it in 1983)

> was a Curses before NCurses came up. cursesmodule now, it bases on
> Curses with NCurses extensions. It is not acceptable to release a

as I understand it, most of the "NCurses extensions" you refer to are
standard in SVr4 curses (color, function keys).

> cursesmodule that wouldn't work on all those platforms it formerly
> worked on. If you intend that, you'll have to give it a new name.
> Furthermore, you can't say, to make cursesmodule work, you have got to
> download another one, it's beta, but maybe, it works on your desolated
> site. But that's no solution. That's a cramp.

> Anyway, that wouldn't have helped in my case, I think, since the name
> conflict would have still existed - libtermcap would have been linked
> before, since libreadline demands it, and, I'm afraid, since ncurses
> originally defines tputs(), it would also been mis-linked. I don't
> really know, I guess. I'm not so familiar with the dynamic linking
> rules under ELF/BSD.

that's a different problem (libreadline doesn't mix with curses on any
system because it makes incompatible assumptions about terminal initalization).

> cu
>  k.j.


> -- 
>  Support Julia Hill - save the Redwood trees: http://www.lunatree.org/

-- 
Thomas E. Dickey
dickey at clark.net
http://www.clark.net/pub/dickey



From tim_one at email.msn.com  Wed Jun 23 01:14:31 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 23 Jun 1999 01:14:31 -0400
Subject: W95 binary output from CGI script
In-Reply-To: <930084286.377082@iris.nyx.net>
Message-ID: <000c01bebd37$4639da00$cb9e2299@tim>

[Tom Hall]
> I'm working on a CGI script for Windows 95, and trying to send
> binary data out stdout. Every time I try to send a \x0a it spits
> out \x0d\x0a.

Python opens the std file streams in text mode, and under all flavors of
Windows that does horrible stuff to binary data.  Quoting from a different
thread,

[Charles G Waldman]
> Try running Python in unbuffered mode - with a "-u" commandline flag,
> or by setting the PYTHONUNUFFERED environment variable to a non-empty
> string.

That disables buffering, but under Windows it also causes Python to open the
std streams in binary mode.

blame-niklaus-wirth-cuz-bill-gates-is-untouchable-ly y'rs  - tim





From joe at strout.net  Thu Jun 17 14:20:28 1999
From: joe at strout.net (Joe Strout)
Date: Thu, 17 Jun 1999 11:20:28 -0700
Subject: [ANN] PIDDLE v1.0a1
Message-ID: <joe-170619991120282505@chinacat.salk.edu>

The PIDDLE development team is pleased to announce the first alpha
release of PIDDLE -- Plug In Drawing, Does Little Else.

PIDDLE is an easy-to-use drawing API with backends that do the actual
drawing to a variety of formats, including PDF, Postscript, QuickDraw,
Tk, PIL, and eventually OpenGL and Windows.

This first alpha release is a little rough around the edges, and not
for the faint of heart.  But if you're up for some adventure, you may
find it quite interesting.  For more information and to get the
distribution, go to:

      http://www.strout.net/python/piddle/

If you have any trouble, you can write to me directly, or post to the
PIDDLE mailing list (http://www.egroups.com/group/pythonpiddle).

Cheers,
-- Joe

-- 
,------------------------------------------------------------------.
|    Joseph J. Strout           Biocomputing -- The Salk Institute |
|    joe at strout.net             http://www.strout.net              |
`------------------------------------------------------------------'



From nikolai at micon.no  Mon Jun 28 16:07:57 1999
From: nikolai at micon.no (Nikolai Kirsebom)
Date: Mon, 28 Jun 1999 22:07:57 +0200
Subject: ASP and namespace
Message-ID: <2BQd3.511$vv1.4827@news1.online.no>

I'm trying to use Python as scripting language on the web-server, however -
have problems to see how the namespace scope is handled.

FILE - GenDb.py
=============
class GenDb:
   def __init__(self):
      self.ConnStr = ""

   def SetDbAndTable(self, db, table):
      self.Db = db
      self.Table = table
      self.ConnStr = "Driver={Microsoft Access Driver (*.mdb)};
DBQ=f:\\Databases\\%s" % db

   def Attr(self, attrName):
      if attrName == "abc":
         return(self.ConnStr)
      else:
         return("-non-")

FILE - KSdebatt.py
===============
from GenDb import *

class KSdebatt(GenDb):
   def __init__(self):
      self.SetDbAndTable("KSdebatt.mdb", "BillMrk")

FILE - SCRIPT1.ASP
=================
<%@ LANGUAGE = Python %>
<%
import sys
sys.path.append("f:\\PythonScripts")
from KSdebatt import *
o=KSdebatt()
%>
<HTML>
<INPUT TYPE="text" NAME="abc" VALUE="<%= o.Attr("abc") %>" SIZE=20>
</HTML>


when I run the statements in the script1.asp file in an interactive window,
I can create the KSdebatt object, however when I run the script on the
server - I  get the following message:

Python ActiveX Scripting Engine error '80020009'
Traceback (innermost last): File "<Script Block >", line 4, in ?
o=KSdebatt() NameError: KSdebatt
/ksdebatt/pages/Script1.asp, line 6
o=KSdebatt()

If it is possible I would like to 'hide' the object (o) inside the KSdebatt
module - that is to have the instanciation inside the module itself - but
still be able to access it from the ASP-code in the script (o.Attr....).  Is
this possible ?


I'm running on NT 4.0 SP 3 and python 1.5.2b2


Thanks for any help

Nikolai Kirsebom






From rdev at my-deja.com  Wed Jun 30 14:18:57 1999
From: rdev at my-deja.com (rdev at my-deja.com)
Date: Wed, 30 Jun 1999 18:18:57 GMT
Subject: Problem with bsddb on NT
Message-ID: <7ldn23$lst$1@nnrp1.deja.com>

I've been using bsddb on both Unix
and NT and I found a very serious bug
on the NT side:

Using database opened with btopen:

If I write random length records using
a six byte key (arbitrary), the bsddb
will crash with a protection exception
after writing, typically, between 50 and
5000 records.

Using database openend with hashopen:

Same test as above, but rather than
crashing with a protection exception,
it generates a python exception as follows:
Traceback (innermost last):
  File "Prog\bsddb_test.py", line 52, in ?
    test ()
  File "Prog\bsddb_test.py", line 49, in test
    adb[key] = rec
bsddb.error: (0, 'No error')

It's easy to recreate using the test program below.
It works just fine on Unix but dies quickly on NT.

If anyone has any ideas at all, please help.  This
is a critical problem and I can't seem to find any
workaround.

Test Program:

import random
import bsddb
import os

# Test overview:
#
# Write a random length record between 1 and 1000 bytes,
# using a 6 byte random valued
# key.  Repeat until the database crashes (typically between
# 50 and 5000 writes.
#
# Crashes in different ways depending on whether bsddb was
# opened as btree or hash database.


def test ():
        count = 0
        # Remove any old database in order to start clean
        try:
                os.remove('test.db')
        except:
                pass
        # Either open bsddb in hash mode or btree mode.
        # Each fails in its own way.
        adb = bsddb.btopen ('test.db', 'c')
        #adb = bsddb.hashopen ('test.db', 'c')
        # Generate a long constant string (1001 chars)
        longstring1 = 'a' * 1001
        while 1:
                # Generate a large random string
                num = random.random()
                longstring = str(int(num * 1000000)) * 200
                # Pick a record length from 1 to 1000
                reclen = int(random.random() * 1000) + 1
                # Use a keylen of 6 (arbitrary)
                keylen = 6
                #if keylen + reclen > 214:
                #       continue

                # Generate a key by taking the first keylen chars
                # from the long string
                key = longstring[:keylen]
                # Generate a record by taking the first reclen chars
                # from the constant string (contents don't matter here)
                rec = longstring1[:reclen]
                count = count + 1
                print 'count = ', count
                print 'lengths = ', keylen, reclen, keylen + reclen
                print 'Key = ', key
                # Add it to the database
                adb[key] = rec


test ()


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From tim_one at email.msn.com  Thu Jun 10 22:23:15 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 10 Jun 1999 22:23:15 -0400
Subject: dict vs kjBuckets vs ???
In-Reply-To: <375ffb1d.26189164@news.cyf-kr.edu.pl>
Message-ID: <000801beb3b1$5c5af240$329e2299@tim>

[MK]
> In some book on algorithms I've read that after inserting limited
> number of items performance of operating on hash tables
> drops dramatically.

Depends on the details.  What you read is true of some kinds of hash tables.
Python's dicts dynamically expand to keep the "load factor" under 2/3, so
what you read isn't applicable to Python in normal use.

> I plan to write a program that would store lots (in range of 10M or even
> more) of relatively small objects (a few hundred bytes at most), so what
> do you think I should use?

Let's do a little math <wink>:  10M * 100 = ?, a lower bound on what you're
contemplating.  Do you have gigabytes of RAM?

> I thought about dictionaries, kjBuckets, or maybe even library called
> Metakit for Python (http://www.equi4.com/metakit/info/README-Python.html).
>
> what-do-you-think-ly y'rs

You don't really want to know <wink>.  Memory-based data structures aren't
going to work for the size of thing you have in mind.  If you can make it
fly it all, you'll likely require a powerful database, so of those choices
Metakit is the only approach that's not dead on arrival.

better-still-write-it-in-perl<wink>-ly y'rs  - tim





From graham at sloth.math.uga.edu  Mon Jun  7 10:14:43 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 7 Jun 1999 14:14:43 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
Message-ID: <7jgk4j$s8p$1@cronkite.cc.uga.edu>

Hisao Suzuki (suzuki611 at okisoft.co.jp) wrote:
: Just now I tried to make the recent stable version of Ruby on a
: box of the same type.  Regrettably it crushed as the similar way
: as before.  Certainly Ruby _is_ suffering from its insufficient
: portability.

And what do we conclude from this?

graham
-- 
             You wanna get close to me
              The feeling's so clear
            But I need some time to see
              Vision through my tears



From greg.ewing at compaq.com  Thu Jun 10 21:24:53 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 11 Jun 1999 13:24:53 +1200
Subject: waste of resources ?
References: <375C0F91.1C73E4E3@icrf.icnet.uk> <375C16A9.5ADC4990@icrf.icnet.uk> <375C1FD1.A85565C@aw.sgi.com> <c75so8370by.fsf@pc142.cosc.canterbury.ac.nz> <375D216B.61787B66@icrf.icnet.uk> <87r9nl398o.fsf@psyche.evansnet> <375E86C1.73B2CF80@icrf.icnet.uk>
Message-ID: <37606565.5A93BCE9@compaq.com>

Arne Mueller wrote:
> 
>                     chfds.remove(i)
>                     os.waitpid(chpids[i], os.WNOHANG)
>                     del chpids[i]

If you're explicitly waiting for a particular child,
*don't* use WNOHANG, because you *want* to block until
that child has exited -- which it may not have done
yet, even though it has written its data.

> I mean maybe ther child exits when the parent
> hasn't reached the os.waitpid?

Other way around -- the parent can reach the waitpid
before the child has exited.

Greg



From MHammond at skippinet.com.au  Mon Jun 28 20:25:54 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Tue, 29 Jun 1999 10:25:54 +1000
Subject: Pythonwin and Office 2000
References: <377648A7.9B9BE59F@prescod.net> <7l672r$5k0$1@m2.c2.telstra-mm.net.au> <3776D1E6.45E1AAF7@prescod.net> <7l7bbt$42u$1@m2.c2.telstra-mm.net.au> <377cdaaa.6963555@news.freeserve.net>
Message-ID: <7l93mr$1in$1@m2.c2.telstra-mm.net.au>

Toby Dickenson wrote in message <377cdaaa.6963555 at news.freeserve.net>...

>If you need to ensure that the makepy wrapper is always available,
independent
>of type library, then I have found the most convenient approach is to copy
the
>appropriate module out of the genpy directory and import it directly. No
UUIDs
>in sight, and it works with freeze. This is effectively what Mark describes
>below.

And back to how makepy worked originally.  Damn it.

>> (VB puts the
>>generated code in the same .exe, hiding this extra distribution - in
Python
>>terms, this would be integrating with freeze etc).  To go this route would
>>really require that makepy no longer generate source code, but instead
>>generate a tight binary format (which may be desirable anyway).
>
>Why single out the makepy wrapper for this compacting effort? Several other
>modules lead to embarrasingle big .pyc and frozen .exe files, but I have
never
>found this a practical problem.
>
>VB and C++ (and maybe Delphi) gain an advantage from the compile-time
approach
>since, in combination with static typing, they can optimise into vtbl
method
>calls. Today, Python is firmly tied to Invoke. We could gain this same

Sure.  I wasnt thinking about going that far, simply optimizing the makepy
approach.  Eg, rather than generating Python code, it may be better
generating Pythonish datastructures that can be used at runtime.  A good
example is the "propMap???" entries for each class.  Generating these maps
as Python code is a little pointless.

Im not sure moving to vtables is appropriate for win32com.  I think we
should be looking at the COM+ space, and making sure any significant work
like this will tie in with the future.

>>  The reason you often
>>now need makepy support is to get decent parameter support, but the
problem
>>is worse than that.
>
>Is the parameter support still a continuing problem? I had the impression
that
>in general only older servers had this problem, with newer servers being
much
>better behaved?

It is a problem mainly with byrefs, and only for those byrefs used on a
dynamic dispatch object that has a type library available, but does not
provide type info at runtime.  However, for most well-behaved objects this
should work fine, either for makepy or dynamic dispatch objects.

[Then we have servers like Notes, that provide 1/2 the information necessary
as a type library, and dont provide _any_ type info at runtime, leaving poor
win32com with a bit of a mess on its hands.]

Mark.





From zeitlin at seth.lpthe.jussieu.fr  Sat Jun 26 17:37:12 1999
From: zeitlin at seth.lpthe.jussieu.fr (Vadim Zeitlin)
Date: 26 Jun 1999 21:37:12 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net>
Message-ID: <slrn7nai1j.8ll.zeitlin@seth.lpthe.jussieu.fr>

 Hello all,

 I've been reading with a lot of interest the messages in this thread and I'd
like to contribute to it from a different point of view - not as a Python
programmer, but as a C++ programmer working on wxWindows development.

 While I don't think wxPython can replace TkInter as a de-facto GUI standard
for Python I still think that it has several advantages compared to TkInter.
Instead of saying how much better, prettier, faster (this one has really made
me smile, BTW: I don't think wxWin is currently very fast, if it has been
cited as its advantage it must really mean that TkInter is extremely slow. But
we're working on performance problems too...) - instead, I'd just like to say
that unlike probably any other GUI toolkit, wxWindows developers are
interested in making their toolkit useful to the Python community. I don't
speak only about Robin Dunn, of course, who developed wxPython, but about
"core" wxWindows developers as well.

 This won't go as far as making wxWin less MFC like just to please folks who
hate MFC (but then, except for the idea of event handlers which I believe is
quite elegant I don't think wxWindows resembles much to MFC nowadays), but
adding new features which lack in wxWin compared to TkInter, for example, or
improving the performance is certainly in the realm of possible.

 TkInter has, for the moment, several advantages compared to wxWindows (it
would be nice if someone did the comparison of 2 toolkits - I'm afraid I don't
know TkInter well enough to do it), but we're working hard to change this and
I sincerely believe that wxWin will have all the features we still miss and
which are present in TkInter really soon (an advanced text widget and
integrated HTML viewer are in alpha and beta stage; image support in wxWin
already includes GIF/JPEG/PNG/BMP/XPM; multimedia classes are waiting to
be integrated in the main distribution; the grid control is being actively
worked on by several people right now).

 Best regards,
VZ



From billtut at microsoft.com  Sun Jun 27 20:55:55 1999
From: billtut at microsoft.com (Bill Tutt)
Date: Mon, 28 Jun 1999 00:55:55 GMT
Subject: Pythonwin and Office 2000
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8100DB90E6D@RED-MSG-50>

> From: Mark Hammond [mailto:MHammond at skippinet.com.au]
> 
> 
> Paul Prescod wrote in message <377648A7.9B9BE59F at prescod.net>...
> >I believe that Pythonwin is not compatible with Office 2000. 
> Here are some
> >symptoms:
> 

Heh, now there's a vague statement...
My Python COM code is talking to Excel 2k quite happily, although I don't
generate the .py wrapper, those things are just too big, and take too long
to initially create on other peoples machines.

> I havent tried this, although I believe it!
> 
> >  File "testMSOffice.py", line 46, in TestWord
> >    if 
> gencache.EnsureModule("{00020905-0000-0000-C000-000000000046}",
> >1033, 8,
> >0) is None:
> 
> This is expected (unfortunately).  Word will have a new 
> typelibrary, and
> hence a new IID for the library.  The "fix" will be to work 
> out what the
> library ID of the new one is (by using "makepy -i") and 
> changing TestWord.py
> to gracefully allow either version.
> 

Actually, the GUID hasn't changed. Its just has a version of 8.1 instead of
8.0. 

> However, this is a PITA in general, and a less-clunky 
> solution would be
> great!
> 

You're the one who required the version number. :)

> >>>> a=win32com.client.Dispatch("MSWord.Application")
> ...
> >com_error: (-2147221005, 'Invalid class string', None, None)
> 
> That is a strange one :-(

Err. I think thats the "can't find ProgID" error message.
The progID is "Word.Application" unless something else wierd is going on. :)

Bill



From holger at phoenix-edv.netzservice.de  Mon Jun 14 05:40:47 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Mon, 14 Jun 1999 09:40:47 GMT
Subject: global vars?
References: <1283099378-77267406@hypernet.com>
Message-ID: <3764CE1F.DA7A6E8F@phoenix-edv.netzservice.de>

Hi,


Gordon McMillan wrote:
> 
>...
> Without some code to demonstrate your confusion, it's difficult to
> guess exactly why you're feeling confused.

Sorry, but I think, you already answered my question.

> In both cases, changing the binding of a global variable requires a
> "global" statement within the function / method. Without that
> statement, Python takes an assignment to the variable (a binding) to
> be local in scope (that is, it's a new variable that exists only
^^^^^^^^^^^^^ ^^^^^^
That's it.

>...
> 
> - Gordon

Thanx,
Holger



From fredrik at pythonware.com  Wed Jun 30 12:33:34 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 30 Jun 1999 16:33:34 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773c3a0.102825445@news.oh.verio.com> <xPPc3.535$or1.22866@typhoon-sf.snfc21.pbi.net> <xJRc3.292$Of3.67094@news.shore.net> <00f701bec212$f6a83050$f29b12c2@secret.pythonware.com> <qmqe3.337$6M6.119017@news.shore.net>
Message-ID: <002e01bec316$4c8bf4d0$f29b12c2@secret.pythonware.com>

Michael P. Reilly wrote:
> Hmm... something about that argument makes it circularly invalid.
> Non-idle event a comes in, but must be handled after idle event d.
> Event d can only execute when event a is taken care of.  Sounds like
> deadlock to me. ;)

hmm.  I thought I'd meant that idle tasks kept piling up
when the windowing system generates a lot of events
at once (like when you're redrawing an entire system).
instead of redrawing everything at once, Tkinter waits
until the window event queue is empty...

</F>




From AInsingaExciseThis at InfoMation.com  Wed Jun  2 11:10:03 1999
From: AInsingaExciseThis at InfoMation.com (Aron K. Insinga)
Date: Wed, 02 Jun 1999 11:10:03 -0400
Subject: Package for Solaris
Message-ID: <3755494B.61C863B2@InfoMation.com>

http://www.python.org/ftp/python/INDEX lists python 1.4 but not 1.5.
By extension of the 1.4 path I found the 1.5 directory,
but there doesn't seem to be a solaris kit there.
(I sent mail to ftpmaster at python.org about 1.5 missing from INDEX.)

I built 1.5.2 from the sources, but is there a Solaris "package" for it?

- Aron Insinga (ainsinga(at)infomation.com)



From dickey at shell.clark.net  Mon Jun 21 05:20:58 1999
From: dickey at shell.clark.net (T.E.Dickey)
Date: Mon, 21 Jun 1999 09:20:58 GMT
Subject: Problem with Curses
References: <ppgmj7.s7r.ln@solling.seismic.de> <jpeck7.1661.ln@solling.seismic.de> <ve1dk7.1e02.ln@solling.seismic.de> <87ogidzs37.fsf@home.ivm.de> <b52kk7.gj.ln@solling.seismic.de>
Message-ID: <_tnb3.607$4e1.36336@iad-read.news.verio.net>

Klaus-Juergen Wolf <kjwolf at online.de> wrote:
>  > Why doesn't Freebsd know the terminfo database? Can't the one that is 
>  > integrated into ncurses be used on FreeBSD? 

> Some versions in the past still knew it, but, it seems, nowadays,
> termcap is the only valid standard for declaring terminal interactions.
> In fact, NCurses _is_ integrated into FreeBSD, the bridge between the
> two systems is libmytinfo, it seems. mytinfo now declares tputs(),
> which termcap also also declares, therefore, there is a small conflict,
> if you don't link them libraries in the correct order.

no.  mytinfo is a rather old/nonstandard interface for terminfo (my copy of
FreeBSD comes with no documentation for it ;-), which is used to support an
old (read: obsolete & buggy) version of ncurses.

-- 
Thomas E. Dickey
dickey at clark.net
http://www.clark.net/pub/dickey



From arcege at shore.net  Wed Jun  9 16:24:37 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 09 Jun 1999 20:24:37 GMT
Subject: fork()
References: <000a01beb21f$a051d380$2fa22299@tim> <7jmbe1$r22$2@cronkite.cc.uga.edu> <7jmg18$1h4$1@aurwww.aur.alcatel.com>
Message-ID: <94A73.2519$nn.762484@news.shore.net>

John Klassa <klassa at aur.alcatel.com> wrote:
: On 9 Jun 1999 18:22:57 GMT, Graham Matthews <graham at sloth.math.uga.edu> wrote:
:   
:   > observable reality. Where are the hoary theoretical arguments? Or 
:   > is this just more of the same old bluff and bluster arguing.
:  
: Uh oh...  This is starting to look like comp.lang.perl.misc. :-)

Yes, some of us have noticed.  That may be a main reason why the person
to actually make the final decisions (Guido) often does not follow
these threads.

  -Arcege




From greg.ewing at compaq.com  Tue Jun 29 00:13:34 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Tue, 29 Jun 1999 16:13:34 +1200
Subject: total idiot question: +=, .=, etc...
References: <000d01bec11f$8f966e20$e19e2299@tim>
Message-ID: <377847EE.D5B97A23@compaq.com>

Tim Peters wrote:
> 
>     class A(B):
>         def __init__(self, x, y, z):
>             B.__init__(self, x, y, z)
> 
> can be grating too

I could live with having to explicitly name the
superclass, provided something checked that the
class I named was actually a direct base class
of the one where the method is defined. That
way things would be less likely to break
mysteriously when I rearrange the class
hierarchy.

I'm still thinking about how to implement
this...

Greg



From tjreedy at udel.edu  Tue Jun 22 02:42:01 1999
From: tjreedy at udel.edu (Terry Reedy)
Date: 22 Jun 1999 06:42:01 GMT
Subject: Thanks for Python...
References: <376ead64@anonymous.newsfeeds.com>
Message-ID: <7knb7p$kas$2@news.udel.edu>

In article <376ead64 at anonymous.newsfeeds.com>,

John Brainard posting anonymously as nobody at newsfeeds.com says...

>write a mach-server that emulated the real server. I was successful at

>mach server that I had originally written in Perl and it works much 

and >I hope that my boss is impressed.

To impress your boss more, use 'mock' (imitation) instead of 'mach' 
(speed number) when you write your report.

TJR (With a real email address, I would have sent this privately.)




From gmcm at hypernet.com  Sun Jun 13 23:03:12 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sun, 13 Jun 1999 22:03:12 -0500
Subject: <language> broken
In-Reply-To: <37645B8D.1A267771@acm.org>
Message-ID: <1282792849-95706075@hypernet.com>

So far, you have not made one post to comp.lang.python that comes 
anywhere close to the posting guidelines.

Would you mind taking this to comp.lang.snobbery, or wherever it 
belongs?

> Fernando Mato Mira wrote:
> 
> > [For example, I can think the Eiffel syntax is `philosophically' broken, but
> > as `infix' syntaxes go, it is not. And it's a good language (I programmed
> 
> Correction. There's one thing broken. The Subject Oriented syntax,
> but as Eiffel was started in '85, I guess that is understandable.
> 
> Ada 95 got this one right. Unfortunately, the semantics qualify as
> `multiply broken' regarding this issue:
> 
> 1. They thought about multiple dispatch, and they broke it on
> purpose. 2. If you specialize on more than one parameter, it should
> be the same type specifier
>      for all of them, i.e. only applicable to predicates, copying,
>      arithmetic and
> things like that,
>      not useful most of the times one wants several specializers. 3.
> Worse of all, instead of just adopting a left-to-right precedence
> rule, one that looks very
>      ugly was added, increasing the complexity of a document that is
>      already big
> enough.
> 
> In '95, at VRAI we refocused our application from AI-oriented to
> (soft) real-time. I grabbed the 9X specs, and the initial excitement
> turned to dissapointment (the prototype was in CLOS). Moving would
> have not only meant losing a lot in flexibility, but, worse of all,
> we would move only from being third-class to second-class citizens
> (SGI platform). And I would have rather dedicate the extra effort to
> CL. So after a little successful experience, and given the simpler
> requirements, we went with C++. Needless to say, the project became
> a death march. You don't go with C++ unless you can afford paying
> 5-10 times more for a lot less functionality.
> 
> --
> Fernando D. Mato Mira
> Real-Time SW Eng & Networking
> Advanced Systems Engineering Division
> CSEM
> Jaquet-Droz 1                   email: matomira AT acm DOT org
> CH-2007 Neuchatel                 tel:       +41 (32) 720-5157
> Switzerland                       FAX:       +41 (32) 720-5720
> 
> www.csem.ch      www.vrai.com     ligwww.epfl.ch/matomira.html
> 
> 
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list

- Gordon



From mjhand at concentric.net  Thu Jun 10 15:52:20 1999
From: mjhand at concentric.net (mjhand at concentric.net)
Date: Thu, 10 Jun 1999 19:52:20 GMT
Subject: None inconsistent
Message-ID: <7jp51f$rci$1@nnrp1.deja.com>

Pardon the repetition (if this is repetitive).  I have posted
this message three times but have yet to see it appear on the
newsgroup.

In 1.5.1, "None" was "less" than all numbers:
>>> 0 < None
0
>>> sys.maxint > None
1

In 1.5.2, "None" became "greater" than all numbers:
>>> 0 < None
1
>>> sys.maxint > None
0

I have three questions about this:
(1) Was this change made intentionally?  (I hope so.)
(2) Can we count on it staying this way forever?  (I hope so.)
(3) When will the starship be updated to use 1.5.2 by default?
    (/usr/bin/python == 1.5.1, /usr/local/bin/python == 1.5.2b2)

I absolutely LOVE that "None" is accepted as a valid argument in
min(), max(), and cmp(), but I would like it to behave consistently.
The 1.5.2 way is best for my code.  Please tell me it will stay
that way forever.

Thanks,
Manus


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From claird at Starbase.NeoSoft.COM  Thu Jun 10 07:58:05 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 10 Jun 1999 06:58:05 -0500
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <375f92b4.618467559@news.u-psud.fr>
Message-ID: <7jo98d$h5i$1@Starbase.NeoSoft.COM>

In article <375f92b4.618467559 at news.u-psud.fr>,
Laurent POINTAL <pointal at lure.u-psud.fr> wrote:
>[note Follow-up to: set to comp.lang.python]
>
>On Thu, 10 Jun 1999 11:17:59 +0200, Fernando Mato Mira
><matomira at iname.com> wrote:
>
>for Python vs Tcl
>>:   If you have references to articles of the sort, that will be useful,
>>: too.
>
>See http://www.cwi.nl/~sjoerd/PythonVsTcl.html
>
>I remeber having seen an article of a guy explaining why he switched
>from Tcl to Python for a large project... but I loose the URL.
			.
			.
			.
I collect material of this sort at <URL:http://
starbase.neosoft.com/~claird/comp.lang.misc/language_comparisons.html>.
For the Tcl -> Python switch, I suspect you're looking for Frank
Stajano's paper, which you'll find in both the previous URL and at
<URL:http://starbase.neosoft.com/~claird/comp.lang.python/7IPC.html>.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From radbit at my-deja.com  Sun Jun 20 19:55:35 1999
From: radbit at my-deja.com (radbit at my-deja.com)
Date: Sun, 20 Jun 1999 23:55:35 GMT
Subject: do anonymous pipes normally work on NT?
References: <4D0A23B3F74DD111ACCD00805F31D8100DB90DFB@RED-MSG-50> <7kjgt1$noj$1@nnrp1.deja.com>
Message-ID: <7kjv1k$ro6$1@nnrp1.deja.com>

In article <7kjgt1$noj$1 at nnrp1.deja.com>,
  Georg Mischler <schorsch at schorsch.com> wrote:
> Bill Tutt wrote:
> > After digging through this stuff, I went back and looked at the
code I
> had
> > written for win32popen.cpp. I realized that I had forgotten why I
had
>
> Uh oh, Bill...
>
> If you should ever happen to visit the Munich Oktoberfest,
> please remind me that I owe you a few beer!
>
>
> Thanks a lot for all the tips and bug hunting!
>

Well......

(I) never dug up the source - but I still spent 3 hours of my time
trying to get your code to work...... do I get a beer at the Fest 8-)?

Cheers
--
Florent Heyworth


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From tismer at appliedbiometrics.com  Wed Jun 16 07:58:09 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Wed, 16 Jun 1999 11:58:09 GMT
Subject: ANN: Stackless Python 0.2
References: <37628EAA.C682F16C@appliedbiometrics.com>
		<7k4fd4$c04$1@brick.cswv.com>
		<376636EA.84240129@appliedbiometrics.com>
		<7k6ttr$i5g$1@brick.cswv.com>
		<m3909lszww.fsf@atrus.jesus.cam.ac.uk> <m37lp4tonr.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <37679151.6C6B29DF@appliedbiometrics.com>


Michael Hudson wrote:

> Got this one; PyEval_Frame_Dispatch is defined static in
> Python/ceval.c. Take the keyowrd static out, and it builds and runs
> fine (in my limited testing so far).

Ah, thanks. Yes, this function was a local eval_frame_dispatch,
before I realized that it needs to be exposed as an API
function. Ahem, forgot to change the decl.

Someone else told be that in frameobject.h, my forward
type declaration gives problems with some compilers.
What is the standard way to do this?

thanks - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From kentsin at poboxes.com  Wed Jun  9 04:14:13 1999
From: kentsin at poboxes.com (Sin Hang Kin)
Date: Wed, 9 Jun 1999 16:14:13 +0800
Subject: Unicode handling Wstring Win95/98 Binary 
Message-ID: <000801beb250$11cb0c40$b8e1fea9@computername>

Hello,

Does any one have wstring package in winxx binary format? I do not have the necessary compiler to do it.

Rgs,

Kent Sin
kentsin at poboxes.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/19990609/376f377e/attachment.html>

From Cepl at fpm.cz  Thu Jun 17 04:35:22 1999
From: Cepl at fpm.cz (Matej Cepl)
Date: Thu, 17 Jun 1999 09:35:22 +0100
Subject: Heading numbering again
Message-ID: <1318D78C9072D11195C9006094EA98A72C2157@ocesrv>

What's wrong with my use of htmllib.py, that when I define in my class
(derived from htmllib.HTMLParser) following method, nothing happens on
output?

	Thanks
			Matthew
-------------------------------------------------------------

    def start_h2(self,attrib):
        "set level on 2 and increment last_num"
        self.out.write("<%s" % tag)
        for pair in attributes:
            self.out.write(" %s = \"%s\"" % pair)
        self.out.write(">")
        self.level = 2
        self.last_num = self.last_num + 1
        self.out.write("%s " % self.last_num)



From guido at cnri.reston.va.us  Tue Jun 29 07:27:05 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 29 Jun 1999 07:27:05 -0400
Subject: total idiot question: +=, .=, etc...
References: <000d01bec11f$8f966e20$e19e2299@tim> <377847EE.D5B97A23@compaq.com>
Message-ID: <5l67476yd2.fsf@eric.cnri.reston.va.us>

Greg Ewing <greg.ewing at compaq.com> writes:

> >     class A(B):
> >         def __init__(self, x, y, z):
> >             B.__init__(self, x, y, z)
> 
> I could live with having to explicitly name the
> superclass, provided something checked that the
> class I named was actually a direct base class
> of the one where the method is defined. That
> way things would be less likely to break
> mysteriously when I rearrange the class
> hierarchy.

Oh, but that check *is* being made!  (In fact this is the same check
that prevents "class methods" from working ;-)

>>> class A:
	def __init__(self):
		self.a = 1


>>> class B:
	def __init__(self):
		self.b = 1


>>> class C(B):
	def __init__(self):
		A.__init__(self)


>>> C()
Traceback (innermost last):
  File "<pyshell#12>", line 1, in ?
    C()
  File "<pyshell#11>", line 3, in __init__
    A.__init__(self)
TypeError: unbound method must be called with class instance 1st argument
>>> 

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



From paulb at infercor.no  Fri Jun  4 05:47:15 1999
From: paulb at infercor.no (Paul Boddie)
Date: Fri, 04 Jun 1999 11:47:15 +0200
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>  <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>  <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net>  <87hfosquc0.fsf@ev.netlab.co.jp> <3753F121.E4A9BE8B@prescod.net> <87yai3ntdt.fsf@ev.netlab.co.jp> <7j3iif$134e$1@nntp6.u.washington.edu> <7j45iv$908$4@cronkite.cc.uga.edu> <7j4hhj$118u$1@nntp6.u.washington.edu> <7j4jbe$dhh$1@cronkite.cc.uga.edu> <37564A02.F115B9C5@infercor.no> <045501beadf8$610e62e0$f29b12c2@pythonware.com>
Message-ID: <3757A0A3.180B21C0@infercor.no>

Fredrik Lundh wrote:
> 
> sure looks like the "community" thinks that changing the
> language is more important that using it...

Improvements should always be considered if Python is to continue to get better
- not that it isn't very good already. However, there is a difference between
those suggestions which genuinely attempt to improve Python "under the hood",
and those which are primarily concerned with changing the syntax. The latter
often seem to stem from the desire to import "cool tricks" from other languages
rather than genuinely improve the language.

It would arguably be better if people concentrated their attention on
applications of Python through the development and usage of modules rather than
getting bogged down with issues which aren't necessarily going to increase the
adoption of Python (of concern to those who feel confident in a larger
community), or make it appropriate or more convenient for a larger number of
tasks.

Just as applications sell operating systems, so modules "sell" languages. Well,
why else would Perl be as successful as it is? (Ruling out for a moment that it
all wasn't a big accident resulting from fortuitous timing.)

Paul



From robin at jessikat.demon.co.uk  Mon Jun  7 07:50:12 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 7 Jun 1999 12:50:12 +0100
Subject: bug in os.path.samefile (Python 1.5.1)?
References: <1318D78C9072D11195C9006094EA98A72C20EF@ocesrv>
 <Pine.LNX.4.10.9906070536020.2300-100000@tbryanpc.arlut.utexas.edu>
Message-ID: <22Ms+CA0H7W3EwBA@jessikat.demon.co.uk>

In article <Pine.LNX.4.10.9906070536020.2300-100000 at tbryanpc.arlut.utexa
s.edu>, Thomas Bryan <tbryan at tbryanpc.arlut.utexas.edu> writes
>
>On Mon, 7 Jun 1999, Matej Cepl wrote:
>
>
...
>
>Aren't filenames case-sensitive on NT?  Look at your traceback.  Notice
last time I used NT4.0 in November paths were case insensitive
...
>-------------------------------------------
>Tom Bryan
>Applied Research Laboratories
>University of Texas at Austin
>
>
>
>
>
>

-- 
Robin Becker



From aa8vb at yahoo.com  Wed Jun  2 13:23:31 1999
From: aa8vb at yahoo.com (Randall Hopper)
Date: Wed, 2 Jun 1999 13:23:31 -0400
Subject: Alien whitespace eating nanovirus strikes again!
In-Reply-To: <aahzFCpHw8.Gxt@netcom.com>; from Aahz Maruch on Wed, Jun 02, 1999 at 03:33:44PM +0000
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net> <aahzFCpHw8.Gxt@netcom.com>
Message-ID: <19990602132331.A281155@epa.gov>

Aahz Maruch:
 |Paul Prescod  <paul at prescod.net> wrote:
 |>
 |>...whitespace handling....
 |
 |...this is one point that still annoys me.
 |
 |Oh, I've gotten used to it, no question about that, and I even
 |appreciate the readability when it comes to working over someone else's
 |code (which is what I've mostly been doing).  But the whitespace problem
 |is a real PITA when it comes time to make a change to the structure of a
 |program -- it's an investment I don't mind making when I'm doing "real"
 |programming, but when I'm making a bunch of quick, fiddly changes to
 |debug something it continually gets in my way.

I actually have like Python's whitespace-as-syntax, but I can see your
perspective.  

If I was still locked onto vi, it'd be a pain, and I can see that being an
issue with some editors.  Though with emacs, there's rectangular kill
(^Xrk) and block indent/dedent (^X^I) -- key sequences I use without even
thinking anymore.

Randall




From larsga at ifi.uio.no  Mon Jun 21 16:21:47 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 21 Jun 1999 22:21:47 +0200
Subject: It's just magic, and I hate it ...
References: <slrn7mss5c.f90.mgm@unpkhswm04.bscc.bls.com>
Message-ID: <wkr9n5s3s4.fsf@ifi.uio.no>

* Mitchell Morris
|
| When run from the browser, however, the redirect doesn't happen
| until the child finishes. To recap, this is exactly the situation I
| need to avoid.

Try doing a sys.stdout.close() at the point where you want the
redirect to happen.

--Lars M.



From tim_one at email.msn.com  Thu Jun 17 01:58:26 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 17 Jun 1999 01:58:26 -0400
Subject: Solution to finalisation problem [Re: fork()]
In-Reply-To: <37683DD6.4C42DC3E@compaq.com>
Message-ID: <000401beb886$6a7f2b00$099e2299@tim>

[Greg Ewing]
> ...
> The only way I can see that the 2-pass collection scheme
> can fail is if the finaliser of object A destroys something
> needed by the finaliser of object B. But if each object
> having a finaliser looks after just one external resource,
> how can that happen?

Greg, this is kind of like asking "if each loop terminates, how can we have
an infinite loop?".  That is, sure, if you write code that works, it works
<wink>.

> Anyone have a real-world example?

Not in *my* Python code <0.5 wink>.  It's all too tricky for me.  Doesn't
mean bigger minds than mine should be dissuaded, though.

Note that the Boehm-Demers-Weiser collector has most often been suggested
for Python (heck, it's a FAQ), and you may want to read about why *they*
guarantee that if A points to B, and both are registered for finalization,
A's finalizer gets called first; e.g., at

    http://reality.sgi.com/boehm_mti/finalization.html

This is an expensive and inconvenient decision for them to have made, so you
can be sure they didn't take it lightly.

BTW, a little-known characteristic of BDW is that given a cycle containing
finalizers, they let the cycle leak rather than risk running finalizers in a
wrong order.

when-the-reactor-is-going-critical-you-don't-want-to-lock-the-doors-
    before-the-people-get-out<wink>-ly y'rs  - tim





From tim_one at email.msn.com  Tue Jun  8 22:27:31 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 8 Jun 1999 22:27:31 -0400
Subject: fork()
In-Reply-To: <7jjghd$9il$1@cronkite.cc.uga.edu>
Message-ID: <000a01beb21f$a051d380$2fa22299@tim>

[Tim]
> OTOH, in a straightfoward mark-and-sweep GC scheme there will
> at least be a "mark bit" in each object header.  So under that,
> or under any form of compacting GC, the *entire live object space*
> will get copied, regardless of whether *you* reference anything in
> it or not.  So if RC truly sucks, what does that make GC <wink>?

{Graham]
> But the reality is that copying memory can in fact speed applications
> up! Think about locality of reference after you have copied all live
> blocks to the start of memory.

Context:  this offshoot of the thread was talking about memory use after a
fork() call in a copy-on-write implementation of fork.  So long as
bookkeeping info is stored in object headers, "real GC" is at a real
disadvantage, in that respect, compared to RC.  Simply because RC won't
touch something unless *you* do.

LOR is an unrelated issue, and harder to analyze.  Yes, copying can speed
applications.  It can also slow them.  "Although copying garbage collectors
have predominated in the past, recent studies suggest that the choice
between mark-sweep and copying garbage collection may well depend as much on
the behavior of the client program as on the inherent properties of the
garbage collection algorithm" [Jones & Lins, "Garbage Collection", Wiley,
1996].  The hoary old theoretical arguments really aren't worth crap in this
field <0.5 wink>.

modern-hardware-is-too-hard-to-predict-ly y'rs  - tim





From andrew at andrewcooke.free-online.co.uk  Sat Jun 19 13:34:44 1999
From: andrew at andrewcooke.free-online.co.uk (Andrew Cooke)
Date: Sat, 19 Jun 1999 17:34:44 GMT
Subject: Proxying HTTP - end of message?
Message-ID: <7kgkbk$uso$1@nnrp1.deja.com>


Hi,

I'm writing a small http proxy (I know a million already exist -
my program isn't important, understanding the bug is).  I have a
couple of threads, spinning away, one copying from client to server
and one from server to client.  When there is no more data to read
from the server the blocking read returns None and I close both
sockets (one socket is proxy-client, the other proxy-server).

But my browser continues to wait.  If I press "stop" I have all the
page, but the browser thinks the download has been interrupted.

What am I missing?  It appears as though the browser isn't detecting
that the connection has been closed - why not?

For what it's worth, this is on NT, I'm extending SocketServer, and
using 1.5.2.  But I suspect this is a problem with my logic rather
than any piece of software....

Any sugegstions appreciated (I can post/forward code - it's my own
work, not my employer's).

Thanks,
Andrew


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From larsga at ifi.uio.no  Wed Jun 16 17:05:38 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 16 Jun 1999 23:05:38 +0200
Subject: XML-toolkit for Windows - binaries available?
References: <376701a3.72288277@newssrv> <wkzp208udt.fsf@ifi.uio.no> <3767fc51.1547455@news.iol.ie>
Message-ID: <wkd7yv7t5p.fsf@ifi.uio.no>

* Sean Mc Grath
|
| The binaries are there but a bit of fiddling is required to get
| things working. (This is from memory but here goes)
| 
| The pyexpat and sgmlop dlls have to be copied into the xml/parsers
| subdirectory. You also need to edit the __init__.py file to include
| 'pyexpat' and 'sgmlop' in the __all__ string.

Heh. I didn't know you could do this! :)

I just put them in the python/dlls directory, and that worked for me.

--Lars M.



From jam at newimage.com  Tue Jun 22 17:48:47 1999
From: jam at newimage.com (jam)
Date: Tue, 22 Jun 1999 17:48:47 -0400
Subject: Unbound method
In-Reply-To: <001701bebcdd$73d3b080$656211c8@miro.lami.pucpr.br>; from faria on Tue, Jun 22, 1999 at 03:31:31PM -0300
References: <001701bebcdd$73d3b080$656211c8@miro.lami.pucpr.br>
Message-ID: <19990622174847.N5115@toast.internal>

On Tue, Jun 22, 1999 at 03:31:31PM -0300, faria wrote:
> 
> Hi everyone!!!
> 
>     I'm new in Python and in Brasil we don't wave any book of Python (and
> it's very expensive to import!), so I have to ask you all this boring
> questions!
> 
>     Why it doens't work ?
> 
> --- test.py
> |
> |class A:
> |    a = 'Greetings from Brasil!!!'
> |    def pa(self):
> |        print self.a
> 
> 
> Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> from test import *
> >>> x = A
> >>> x.a
> 'Greetings from Brasil!!!'
> >>> x.pa
> <unbound method A.pa>
> >>>
> 
> 
> 

hi.

I'm going to take a stab at explaining this, since I ran into similar
confusion in some code I'm working on for Image.

what you did above was to lookup and use *unbound* methods and attributes of
the class called 'A'.

let's simplify this, and rename the class to 'spam':

class spam:
	a = "greetings from brasil!"
	def pa(self):
		print self.a

now, instantiate (or 'create') an *instance* of type class 'spam':

[~] [5:44pm] [jam at toast-pts/4] % python
Python 1.5.2 (#1, Apr 18 1999, 16:03:16)  [GCC pgcc-2.91.60 19981201 (egcs-1.1.1  on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> class spam:
...     a = "greetings from brasil!"
...     def pa(self):
...             print self.a
... 
>>> egg = spam()
>>> egg.a
'greetings from brasil!'
>>> egg.pa
<method spam.pa of spam instance at 80e1250>
>>> egg.pa()
greetings from brasil!
>>> type(egg)
<type 'instance'>
>>> type(spam)
<type 'class'>

does that help to explain what is going on a little better? I'm sorry I
don't have the precise terms used to describe these things.. what you used
in your example was a perfectly valid expression in python (it's *ok* to
reference unbound methods of objects.. I'm sure there are uses for them),
but it's probably not what you expected.

regards,
J
-- 
|| visit gfd <http://quark.newimage.com/> 
|| psa member #293 <http://www.python.org/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>



From frankn=news at cs.vu.nl  Wed Jun 16 17:29:46 1999
From: frankn=news at cs.vu.nl (Frank Niessink)
Date: Wed, 16 Jun 1999 21:29:46 GMT
Subject: BiBTeX module?
References: <joe-160619991358444033@chinacat.salk.edu>
Message-ID: <7k950a$hsb@cs.vu.nl>

Joe Strout <joe at strout.net> wrote:
> Is there a BiBTeX module for Python?  I did a search at python.org, but
> apart from a Perl module (ick!), I didn't find anything.

A collegue and myself hacked up a package (called Bridge) which enables
you to generate BibTeX from SGML/XML (so it includes both a BibTeX SGML dtd
and a BibTeX XML dtd) and vice versa and also has a web interface through 
which you can search your bibfiles (using glimpse). See 
http://www.cs.vu.nl/~frankn/bridge for an example.

The package itself is in a state of flux right now, but if you're
interested I guess we can make it available nevertheless.

Cheers, Frank

-- 
The Hitch Hiker's Guide to the Galaxy has this to say on the subject of 
flying. There is an art, it says, or rather a knack to flying. The knack lies
in learning how to throw yourself at the ground and miss. Pick a nice day, it
suggests, and try it.
	-- Douglas Adams, 'Life, the Universe, and Everything'



From matz at netlab.co.jp  Mon Jun  7 02:10:49 1999
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 07 Jun 1999 15:10:49 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
 <7j450j$908$2@cronkite.cc.uga.edu> <7j5sr7$eb4$1@nnrp1.deja.com>
 <7j6j8d$58f$2@cronkite.cc.uga.edu> <7j902g$hob$1@nnrp1.deja.com>
Message-ID: <87n1yc4k2e.fsf@ev.netlab.co.jp>

adjih at technologist.com writes:

|  It is true that it shouldn't be hard to make a mark-sweep collector
|that is working in practice on many platforms.
|
|  Pedantically I doubt that ANSI-C tells you the current stack top
|address, or guarantees that the whole address space between the current
|stack pointer and this top stack address is stack space. But I agree
|that actual C compilers wouldn't do something too stupid, we
|shouldn't care about this :-)

You don't have to know stack address to implement mark-sweep register,
unless you want to make it conservative.  The mark-sweep GC can be
written in ANSI-C portable.
						matz.



From mikael at pobox.com  Mon Jun 28 13:08:40 1999
From: mikael at pobox.com (Mikael Lyngvig)
Date: Mon, 28 Jun 1999 17:08:40 GMT
Subject: Python GUIs: Summary and Conclusion
Message-ID: <99FA4A1A1280BEBD.0A1ABA62CC989479.4C2C6022F695BEAC@lp.airnews.net>

Hi,

I think there is a sufficient number of replies to my original "Python
GUIs: Abondining TkInter and welcoming wxPython" post that a
conclusion can be made.  Thanks to all who replied; the replies did
exactly what I wanted - presented a lot of arguments for and against a
switch from TkInter to wxPython.

Arguments for keeping TkInter:

1. It is the most widely portable GUI.
2. It is the most flexible GUI (at least compared to wxWindows).
3. It is the most widely used GUI, documented in several books.

Arguments for promoting wxWindows:

1. It is the fastest GUI (compared to TkInter).

Two other GUIs, The FastLight Toolkit and PyKDE/Qt, were mentioned,
but they both fall outside the domain of this discussion because
they're not portable enough (one of the primary tasks of a standard
GUI must be to be available on any, or most, Python hosts). 

One other "GUI", JPI, was mentioned and dismissed because it requires
Java (so the GUI developer must know two scripting languages to make
GUIs in one of them).  However, this may be useful to some since it
gives JPython and CPython developers one GUI tool.

Really useful clues to why TkInter does not perform as well as it
should:

(Michael P. Reilley)
But.. the argument over (these two) GUIs is not exactly a valid one.
Tkinter was written using the Tcl interface not the Tk/C API.  If
Tkinter was rewritten to use the C API, it probably wouldn't be slower
than wxPython.

Ideas to make wxPython/wxWindows more portable, which the wxPython
community may consider to implement:

(Alexander V. Voinov)
Well, let's write a sort of wxTk, and have that everywhere, where no
wxNative runs yet.

URLs for the various GUIs (excluding TkInter):

wxPython - http://alldunn.com/wxPython/
PyKDE/Qt - http://www.river-bank.demon.co.uk/software/
JPI - http://www.ndim.edrc.cmu.edu/dougc/jpi/Home.html


Conclusion (my conclusion, at least...):

TkInter should continue to be the de-facto standard Python GUI.
Primarily because of its matureness, some say better design
philosophy, its portability, and its popularity.  

wxPython continues to be a good alternative, as does others, which may
be better in specific situations and to some developers.

The TkInter module(s) should be rewritten to employ the native Tk API,
instead of Tcl, which would probably eliminate all sluggishness in
client applications.

More widgets should be added to Tk(Inter).

Anyone disagree?  

Any volunteers to start work on making a native Tk C API version of
TkInter, 100 percent backwards compatible of course?


-- Mikael




From phd at emerald.netskate.ru  Wed Jun 16 03:55:17 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Wed, 16 Jun 1999 11:55:17 +0400 (MSD)
Subject: pygtk - clist title button
In-Reply-To: <376755A5.15621A4@onera.fr>
Message-ID: <Pine.LNX.4.04.9906161153290.8375-100000@emerald.netskate.ru>

Hi!

On Wed, 16 Jun 1999, Marc POINOT wrote:
> What about using Gtk instead of Tkinter in the Python "core" built ?

   Portability problem - GTK (and especially PyGTK) don't work on some
platforms where Tk and Tkinter work. Either we should destroy these
platforms or make GTK/PyGTK more portable.

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From mfletch at vrtelecom.com  Mon Jun 28 09:26:19 1999
From: mfletch at vrtelecom.com (Mike C. Fletcher)
Date: Mon, 28 Jun 1999 13:26:19 GMT
Subject: Extracting Text file contents using Python
References: <37782f9a.0@news.smartnet.com.sg>
Message-ID: <%JKd3.12371$o6.228862@news2.rdc1.on.home.com>

Note that this algo closes the html table as soon as it opens, not sure if
that's the problem, but you probably want an elif in here...
>         if begin_table:
>                 out.write('<table border = "1">')
>                 begin_table = 0
>         elif line[:5] == "=====":
>                 out.write('</table>')
>                 out.write('<br>')
>                 begin_table = 1
>                 continue

Here's a function that seems to do pretty-much what you want, not really
very clearly written, but that's mostly cause I prefer maps and filters and
such, a per-line-looping function with counters and such would, of course,
work...

import string
DEFRECSEP = '================================='
DEFFIELDSEP = '\n'
DEFKEYSEP = ':'
def loadfile( filename,
        recordsep = DEFRECSEP,
        fieldsep= DEFFIELDSEP,
        keysep=DEFKEYSEP
    ):
    # read in whole file
    tempdata = open( filename).read()
    # break into records
    tempdata = string.split( tempdata, recordsep )
    # Get rid of extra whitespace
    # and null records...
    tempdata = filter( None, map( string.strip, tempdata ))
    # for each record, get the field values
    for i in range( len( tempdata)):
        # split into the lines
        fields = string.split( tempdata[i], fieldsep )
        # strip trailing/leading whitespace
        fields = map( string.strip, fields )
        # get the key-value pairs
        fields = map( string.split, fields, [ keysep ]*len(fields),
[1]*len(fields) )
        # should probably do a strip here too...
        tempdata[i] = fields
    return tempdata

So, you get a structure that's like this
[
    [
        [key, value],
        [key,value],
        ...
    ]
    [
        [key, value],
        [key,value],
        ...
    ]
]

for record in dataset:
    for key, value in record:
        dosomething_with_key_and_value(key, value)

Incidentally, you'd probably find this kind of thing easier with a simple
database module.  Wouldn't be plain-text, but much easier (and faster) to
use.  If you really wanted convenience, you could even use shelve and just
dump instances and/or data structures directly to disk so you don't need to
do any parsing at all.

Hope this helps,
Mike




john <john at mediamanager.com.sg> wrote in message
news:37782f9a.0 at news.smartnet.com.sg...
> Hi,
>
> I am trying to extract the following text file using python and the
program
> & text file are as shown below.
>
> But unfortunately I am unable to get every separate data displayed except
> for the first data and after the separation line which is
> "====================" the next data is not getting picked up by the
> program.
> Could anyone kindly let me know what could be wrong with the program.
>
> Example Text File - But will be similar
> ----------------------------------------------------
>
> Title:Meet Me
> Http Host:www.www.www
> Remote Address:10.2.0.1
> Remote Host:10.2.0.1
> Rating:8
> Date:06/25/99
> Time:17.15.28
> =================================
> Title:Meet Me
> Http Host:www.www.www
> Remote Address:10.2.0.1
> Remote Host:10.2.0.1
> Rating:8
> Date:06/25/99
> Time:17.15.28
> =================================
> Title:Meet Me
> Http Host:www.www.www
> Remote Address:10.2.0.1
> Remote Host:10.2.0.1
> Rating:8
> Date:06/25/99
> Time:17.15.28
> =================================
> Title:Meet Me
> Http Host:www.www.www
> Remote Address:10.2.0.1
> Remote Host:10.2.0.1
> Rating:8
> Date:06/25/99
> Time:17.15.28
> =================================
> Title:Meet Me
> Http Host:www.www.www
> Remote Address:10.2.0.1
> Remote Host:10.2.0.1
> Rating:8
> Date:06/25/99
> Time:17.15.28
> =================================
>
> The python program is as follows:
>
> #!c:/progra~1/python
>
> import string
> import sys
> import cgi
> #import urllib
>
> data_file = open('c:/xitami/cgi-bin/textfiles/FilmFestReview.txt', 'r')
> form = cgi.FieldStorage()
> header = \
> """
> <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN">
> <html>
> <head>
> <title>Extract & Write</title>
> </head>
> <body>
> """
>
> footer = \
> """
> </body>
> </html>
> """
>
> out = sys.stdout
>
> begin_table = 1
> out.write(header)
> for line in data_file.readlines():
>         if begin_table:
>                 out.write('<table border = "1">')
>                 begin_table = 0
>         if line[:5] == "=====":
>                 out.write('</table>')
>                 out.write('<br>')
>                 begin_table = 1
>                 continue
>         field_name, field_value = string.split(line, ':')
>         out.write('\t<tr>\n')
>         out.write('\t\t<td>\n')
>         out.write('\t\t\t%s\n' % field_name)
>         out.write('\t\t</td>\n')
>         out.write('\t\t<td>\n')
>         out.write('\t\t\t%s\n' % field_value)
>         out.write('\t\t</td>\n')
>         out.write('\t</tr>\n')
>
>
>
> out.write(footer)
>
>
>
>





From zessin at my-deja.com  Mon Jun 21 11:40:10 1999
From: zessin at my-deja.com (Uwe Zessin)
Date: Mon, 21 Jun 1999 15:40:10 GMT
Subject: Creating a new type with an extension module
References: <Ctp*vl52n@news.chiark.greenend.org.uk>
Message-ID: <7klmcn$dd0$1@nnrp1.deja.com>

In article <Ctp*vl52n at news.chiark.greenend.org.uk>,
  Sion Arrowsmith <siona at chiark.greenend.org.uk> wrote:
> Is there any documentation anywhere on how to go about writing an
> extension which implements a new type? The extending/embedding
> tutorial give me, very quickly and painlessly, a module with
> functions, but I could really do with them being object methods.
> http://www.python.org/doc/current/api/newTypes.html is unhelpful
> in the extreme -- do I have to resort to reading source?

I am not sure if this is what you are after, but I have found XXMODULE.C
+ XXOBJECT.C very helpful as small examples. I was able to easily turn
them into what I wanted.

--
Uwe Zessin


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From fdrake at cnri.reston.va.us  Tue Jun 15 13:20:39 1999
From: fdrake at cnri.reston.va.us (Fred L. Drake)
Date: Tue, 15 Jun 1999 13:20:39 -0400 (EDT)
Subject: easy filecopy? - JPython?
In-Reply-To: <1282653264-104102639@hypernet.com>
References: <PBt93.843$B04.2678@newsfeed.slurp.net>
	<1282653264-104102639@hypernet.com>
Message-ID: <14182.35687.906156.126592@weyr.cnri.reston.va.us>

Gordon McMillan writes:
 > Last I recall (a while ago), os.chmod() doesn't work on Win9* and 
 > acts as a noop on WinNT under CPython. This might explain why the 
 > functionality isn't there in Java?

Gordon,
  This certainly would make sense.  The question becomes "If os.chmod
doesn't exist, should shutil.copy() fail?"  I'm not at all sure what
the answer is, but I can see it going either way.  shutil.copymode()
and shutil.copystat() should certainly fail, but it's not at all clear 
that shutil.copy() should.
  Someone more aware of the Java/JPython issues than myself should
make the call; I seem only to start up JPython when a thread like this 
pops up.  ;-)


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From esuzm at primrose.csv.warwick.ac.uk  Wed Jun 23 15:37:31 1999
From: esuzm at primrose.csv.warwick.ac.uk (Andrew Clover)
Date: 23 Jun 1999 19:37:31 GMT
Subject: file.tell problems (1.5.2/Win32)
Message-ID: <7krd1r$2oa$1@holly.csv.warwick.ac.uk>

I've been meeting some surprises using the seek and tell on file objects
recently. I don't know if these are bugs in Python or stdio, in Win32 or
in general, or are simply broken as designed, but:

 - If a file is opened in 'a'ppend mode, but nothing has yet been written to
   it, tell() returns zero.

 - Values returned by tell() - possibly only on a append file, I don't know -
   seem to be randomly a bit wrong unless flush() is used before reading.

 Can anyone confirm this?

-- 
This posting was brought to you by And Clover.
(Sorry.)



From nikolai at micon.no  Tue Jun 29 14:15:42 1999
From: nikolai at micon.no (Nikolai Kirsebom)
Date: Tue, 29 Jun 1999 20:15:42 +0200
Subject: Python, ASP - Namespace (again)
Message-ID: <O18e3.449$U%.3201@news1.online.no>

Trying again - since I'm rather stuck, and I would like to stick to Python.


I'm trying to use Python as scripting language on the web-server, however -
have problems to see how the namespace scope is handled.

FILE - GenDb.py
=============
class GenDb:
   def __init__(self):
      self.ConnStr = ""

   def SetDbAndTable(self, db, table):
      self.Db = db
      self.Table = table
      self.ConnStr = "Driver={Microsoft Access Driver (*.mdb)};
DBQ=f:\\Databases\\%s" % db

   def Attr(self, attrName):
      if attrName == "abc":
         return(self.ConnStr)
      else:
         return("-non-")

FILE - KSdebatt.py
===============
from GenDb import *

class KSdebatt(GenDb):
   def __init__(self):
      self.SetDbAndTable("KSdebatt.mdb", "BillMrk")

FILE - SCRIPT1.ASP
=================
<%@ LANGUAGE = Python %>
<%
import sys
sys.path.append("f:\\PythonScripts")
from KSdebatt import *
o=KSdebatt()
%>
<HTML>
<INPUT TYPE="text" NAME="abc" VALUE="<%= o.Attr("abc") %>" SIZE=20>
</HTML>


when I run the statements in the script1.asp file in an interactive window,
I can create the KSdebatt object, however when I run the script on the
server - I  get the following message:

Python ActiveX Scripting Engine error '80020009'
Traceback (innermost last): File "<Script Block >", line 4, in ?
o=KSdebatt() NameError: KSdebatt
/ksdebatt/pages/Script1.asp, line 6
o=KSdebatt()

If it is possible I would like to 'hide' the object (o) inside the KSdebatt
module - that is to have the instanciation inside the module itself - but
still be able to access it from the ASP-code in the script (o.Attr....).  Is
this possible ?


I'm running on NT 4.0 SP 3 and python 1.5.2b2


Thanks for any help

Nikolai Kirsebom







From faassen at pop.vet.uu.nl  Mon Jun 14 05:26:40 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Mon, 14 Jun 1999 11:26:40 +0200
Subject: Any Production ZOPE sites we can visit?
References: <37614A32.79FE554E@metricom.com> <Xyd83.25$KR2.61466@news.corridex.com>
Message-ID: <3764CAD0.B800CE7B@pop.vet.uu.nl>

Evan Simpson wrote:

> 
> Someone on Zope-list recently opened a
> super-duper-all-singing-all-dancing-framed-JScripted site running on Zope,
> at:
> http://www.master.cit.be/c8/default.html

As far as I know this isn't running on Zope, unfortunately. It *is*
super-duper-all-singing-all-dancing-framed-JScripted, however, and looks
very nice. *and* it is _about_ Zope, and well worth a read.

Regards,

Martijn



From greg.ewing at compaq.com  Sun Jun 27 23:34:59 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Mon, 28 Jun 1999 15:34:59 +1200
Subject: file.tell problems (1.5.2/Win32)
References: <7krd1r$2oa$1@holly.csv.warwick.ac.uk> <7l3aao$65c$1@nnrp1.deja.com>
Message-ID: <3776ED63.35D87F89@compaq.com>

Bill Wilkinson wrote:
> 
> hmmm.... not too sure what your asking, here is a session on my NT box.
> 
> >>> g = open("c:\\temp\\trick.ini","a")
> >>> g.tell()
> 0

The problem is, tell() returns 0 at this point
even if the file being appended to is NOT EMPTY!
Witness:

>>> f = open("blarg.txt")
>>> print f.read()
echo Pan galactic gargle blaster


>>> f.close()
>>> f = open("blarg.txt", "a")
>>> f.tell()
0
>>> f.write("furbie")
>>> f.tell()
42

Only after something has been written does tell()
return the correct file position.

Greg



From bwarsaw at cnri.reston.va.us  Thu Jun  3 15:48:48 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Thu, 3 Jun 1999 15:48:48 -0400 (EDT)
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7imr6b$1s0$1@nnrp1.deja.com>
	<87pv3iq6ss.fsf@ev.netlab.co.jp>
	<3752C76D.604CD474@prescod.net>
	<87n1ykr0z9.fsf@ev.netlab.co.jp>
	<375372F6.A874554@prescod.net>
	<87hfosquc0.fsf@ev.netlab.co.jp>
	<3753F121.E4A9BE8B@prescod.net>
	<87yai3ntdt.fsf@ev.netlab.co.jp>
	<7j3iif$134e$1@nntp6.u.washington.edu>
	<7j45iv$908$4@cronkite.cc.uga.edu>
	<7j4hhj$118u$1@nntp6.u.washington.edu>
	<14166.46756.866947.64007@anthem.cnri.reston.va.us>
	<87g14942co.fsf@pc-hrvoje.srce.hr>
	<14166.53171.289824.873109@anthem.cnri.reston.va.us>
	<87yai1nkjj.fsf@pc-hrvoje.srce.hr>
Message-ID: <14166.56352.769611.537949@anthem.cnri.reston.va.us>

>>>>> "Hrvoje" == Hrvoje Niksic <hniksic at srce.hr> writes:

    Hrvoje> Would this right approach ever work in Python?  Given
    Hrvoje> Python's dynamism, it's perfectly legal to write:

    Hrvoje>     x.__del__ = FN

    Hrvoje> where x is an instance of a class without a __del__()
    Hrvoje> method.  But then again, maybe it would be OK to silently
    Hrvoje> ignore such __del__ methods...

Yep, that's exactly what would be broken with my approach.

-Barry



From suzuki611 at okisoft.co.jp  Mon Jun  7 00:54:45 1999
From: suzuki611 at okisoft.co.jp (Hisao Suzuki)
Date: 07 Jun 1999 13:54:45 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7j90va$6lk$1@cronkite.cc.uga.edu>
	<u3g1443iht.fsf@ares.sys1.ngo.okisoft.co.jp>
Message-ID: <u3btesy5ii.fsf@ares.sys1.ngo.okisoft.co.jp>

In article <u3g1443iht.fsf at ares.sys1.ngo.okisoft.co.jp>,
I wrote:
| Regrettably, I'd have to say that Ruby is (was?) suffering from
| its insufficient portability.  Once I have used a certain i860
| SVR4 box.  On that box, Ruby did not run but dumped immediately
| with its twiddling the machine stack, while Python ran perfectly
| with no difficulty.  As far as I know, the collector of Ruby has
| not changed so much since then.

Just now I tried to make the recent stable version of Ruby on a
box of the same type.  Regrettably it crushed as the similar way
as before.  Certainly Ruby _is_ suffering from its insufficient
portability.

--===-----========------------- Sana esprimo naskas sanan ideon.
SUZUKI Hisao            suzuki611 at okisoft.co.jp, suzuki at acm.org.



From larsga at ifi.uio.no  Wed Jun 23 10:50:46 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 23 Jun 1999 16:50:46 +0200
Subject: How to obtain an instance's name at runtime?
References: <7kqpu8$7u1$1@nnrp1.deja.com>
Message-ID: <wkzp1rotrt.fsf@ifi.uio.no>

* Dinu C. Gherman
|
| 
| Question: How to obtain an instance's name at runtime?
| That is I'd like to see this happen (one was or the other):
| 
| >>> c.__name__
| 'c'

How do you want this to behave if you go on from the situation you
have to type:

>>> d=c
>>> d.__name__
???
>>> c.__name__
???

--Lars M.



From evan at tokenexchange.com  Tue Jun 29 16:00:14 1999
From: evan at tokenexchange.com (Evan Simpson)
Date: Tue, 29 Jun 1999 15:00:14 -0500
Subject: Please help me to understand classes
References: <Pine.LNX.3.96.990629205752.3228A-100000@Johann>
Message-ID: <3B9e3.2$P4.4712@news-dal.corridex.com>

Your Pascal experience is showing <wink>.  While you can call a
parameterless function or procedure without parentheses in Pascal, Python
always requires them, since functions are first-class objects.

So when you write "a.druk", you are fetching the method "druk" from "a", but
then doing nothing with it.  You should write "a.druk()", which will call
the method with no parameters.

One reason for the required parentheses is the ability to do this:

f = a.druk  # get the method and make f refer to it.
f()  # call the method.

Johann Spies wrote in message ...
I have been playing around with python for some time now, but I am a
relative newbie as it is my hobby and normally I have very little time for
programming.  I am on leave now and want to write a program but I find
python a very difficult language to comprehend despite the fact that I
have a few years of experience in programming a language like Pascal.
Maybe it is because I have little experience in OOP.

Can somebody explain to me why the following program produces no output:
---------------------------------------------
class A:

    def druk():
        print 'Jy is in Class A'

class B:

    def druk():
        print 'Jy is in Class B'

a = A()
c = B()
c.druk
a.druk






From tim_one at email.msn.com  Tue Jun  1 21:20:16 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 1 Jun 1999 21:20:16 -0400
Subject: When I run a module written in C, core dumps at Python exit.
In-Reply-To: <37529504.669DE3D8@erols.com>
Message-ID: <000b01beac96$1262f6a0$d89e2299@tim>

[Edward C. Jones]
> I wrote an extension module written in C. It seems to work OK.
> But core dumps when exiting Python. When I run the code from the
> interpreter, Python works fine until I try to exit with ^D. Then
> there is a core dump. What are the most common causes of such
> problems?

Buggy extension modules written in C.

suspecting-that's-obvious-ly y'rs  - tim





From glazer at scicomp.com  Thu Jun 24 13:04:14 1999
From: glazer at scicomp.com (Ben Glazer)
Date: Thu, 24 Jun 1999 17:04:14 GMT
Subject: Default printing behavior for classes
Message-ID: <7ktoe5$bbl$1@nnrp1.deja.com>

I'm looking for a way to set default printing behavior for a class I've
defined.  Lists are printed in a certain format by default.  I'd like to
be able to define this for any type of class.  It'd be really nice if
there were something I could include in my class definition, e.g.

	class Board:
	    def __init__(self, width=2, height=4, length=20):
		self.width = width
		self.height = height
		self.length = length

	    def __print__:
		print "[%s,%s,%s]" % (width,height,length)

such that

	b = Board(8,8,2)
	print b

would print "[8,8,2]".

Is there anything in Python that encapsulates this functionality, or
should I resign this desire to a feature req for Python 2.0?


Thanks,
Ben


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From markusk at bidra241.bbn.hp.com  Thu Jun 17 04:40:29 1999
From: markusk at bidra241.bbn.hp.com (Markus Kohler)
Date: 17 Jun 1999 10:40:29 +0200
Subject: GC In Python: YAS (Yet Another Summary)
References: <Pine.SUN.3.95-heb-2.07.990616155228.15761A-100000@sunset.ma.huji.ac.il>
Message-ID: <p53dzrfcea.fsf@bidra241.bbn.hp.com>

Moshe Zadka <moshez at math.huji.ac.il> writes:

> On Wed, 16 Jun 1999, Robert Meegan wrote:
> 
> > On Wed, 16 Jun 1999, Moshe Zadka wrote:
> > 
> > > Regarding the recend discussion here about garbage collection, I wrote 
> > > a small article.
> > 
> > Moshe ---
> > 
> > You didn't provide the document or a URL.
> 
> You're absolutely right! Shame on me!
> http://www.geocities.com/TheTropics/Island/2932/gcpy.html
> 

First I think it would be interesting to see how much code is affected
by the problem you describe in your document. Ssince JPython
would have a problem with this kind of code one needs to fix the
standard libraries anyway. 

Your example could also be solved by using a GC with a finalization
mechanism that is aware of file objects. It would keep track how many
files are open and eventually do a collection.   

All the other solutions seems to make the implementation more
complicated and probably would decrease Pythons already not so good
performance. 

The main problem with introducing a GC in Python is that a lot of C-code
might directly point to Python objects which could be moved at runtime 
by a state of the art GC  (generational mark and sweep + copying GC). 

Markus
-- 
Markus Kohler  mailto:markus_kohler at hp.com



From aahz at netcom.com  Thu Jun  3 18:33:18 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Thu, 3 Jun 1999 22:33:18 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7j6jnl$58f$3@cronkite.cc.uga.edu> <aahzFCrnB0.DIE@netcom.com> <7j6ua4$fp6$1@cronkite.cc.uga.edu>
Message-ID: <aahzFCrvzJ.FuI@netcom.com>

In article <7j6ua4$fp6$1 at cronkite.cc.uga.edu>,
Graham Matthews <graham at sloth.math.uga.edu> wrote:
>
>I don't understand why you want GC only by explict call and in "inform"
>mode only?

I don't think inform mode should be the only mode, but it should be the
default and primary mode.  As for explicit GC, just call me a Luddite.
(IOW, I think of GC as primarily a debugging tool and I don't trust
autoGC to not impact real-time programs.)
-- 
                      --- Aahz (@netcom.com)

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

"NRE is a lot like cocaine. Feels great while you're on it, makes you
feel very special and talented, and you become highly annoying to those
who aren't on it."  -- Stef



From jmrober1 at ingr.com  Thu Jun 10 09:39:11 1999
From: jmrober1 at ingr.com (Joseph Robertson)
Date: Thu, 10 Jun 1999 08:39:11 -0500
Subject: "Programming Python" still worthwhile?
References: <7jki66$qam$1@news.gstis.net>
Message-ID: <375FBFFF.3B37D5DE@ingr.com>

If you feel you have to buy a book on Python, then get "Learning Python".  As
an experienced LumberJack, er Pythoner, I must say that Mark and his cohorts
writing style is much improved over "Programming Python".  I've been at Python
for several years, written apps, com objects, lots of stuff.  LP is a very
good, insightful and useful read.

Joe


Rick Robino wrote:

> Greetings,
>
> I've been watching this group for some time and just now getting warm to
> Python. Looking at source examples has been great, but for completeness'
> sake I'd like to get more of the "why" maybe found in a book. Since Python
> is at 1.5.2 and people are talking about 2.0 as if it's around the corner,
> I was wondering if anyone here could give me some advice as to whether or
> not "Programming Python" is useful, if the current edition is still
> current, if I should wait for the "Learning Python", or if I should use
> some other resources.
>
> I expect to come up to speed fairly quickly and I'm just looking for
> those fundamental insights into OOP and Python that can build up the
> kind of inuition and understanding one needs in order to "get
> comfortable" connecting/creating the dots. Currently I only program
> small - medium programs in C and shell (Korn). Any sage pointers to
> resources?
>
> TIA,
>
> --Rick
>                          ^^^^^^^^^^^^^^
> Rick Robino                                  rrobino at wavedivision.com






From arcege at shore.net  Tue Jun  1 15:39:08 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 01 Jun 1999 19:39:08 GMT
Subject: Whats missing?
References: <37541767.AB2B250B@cwcom.net>
Message-ID: <wFW43.1497$nn.453396@news.shore.net>

Daniel Faulkner <m01ymu00 at cwcom.net> wrote:
: I was just messing about and found out that if I made some thing like
: this it would not work but if I did it by typing at the python prompt

: from urllib import urlopen
: Data = urlopen("http://www.python.org/").read()
: print Data

: It would work but why is this?

: from Tkinter import *
: from urllib import urlopen

: class HTMLBrowser(Frame):
:     def Create_Widgets(self):
:         # frames
:         butFrame = Frame()
:         butFrame.pack(side='top')
:         # normal widgets
:         self.QuitB = Button(butFrame, text='Quit', fg='red',
: command=self.quit)
:         self.UrlEnt = Entry(self, relief=SUNKEN)
:         self.source_box = Text(self)
:         self.startB = Button(butFrame, text='Start',
: command=self.UrlGeting)
:         # Pack it
:         self.QuitB.pack(side='left')
:         self.startB.pack(side='left')
:         self.UrlEnt.pack(side='top', fill='x')
:         self.source_box.pack(side='bottom')
:     def lets_see_it(self):
:         self.source_box.insert('0.0', self.Data)
:     def UrlGeting(self):
:         UrlCurr = self.UrlEnt.get()
:         Data = urlopen('"' + UrlCurr + '"').read()

Off-hand, I would say it is the quoting here.

This statement is adding double quotes to the URL.  UrlCurr is already
a string, you do not need to enclose it in quotes.

  urlopen("http://www.python.org/spam").read()
is the same as:
  url = "http://www.python.org/spam"
  urlopen(url).read()

The statement above is attempting to call:
  urlopen("\"http://www.python.org/spam\"").read()

:         self.source_box.insert('0.0', Data)
:     def __init__(self, master=None):
:         Frame.__init__(self, master)
:         self.pack()
:         self.Create_Widgets()

: # Test
: Dat = urlopen("http://www.python.org/")
: # print Dat

: Browser = HTMLBrowser()
: Browser.mainloop()




From holger at phoenix-edv.netzservice.de  Wed Jun  9 06:34:26 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Wed, 9 Jun 1999 10:34:26 GMT
Subject: global vars?
Message-ID: <375E4332.645E6BBF@phoenix-edv.netzservice.de>

Hi there,

I'm a newbie at python and there's something I don't understand:

I'd like to have some global variables e.g. PREDEFINED_PROGDIR.
It should have a predefined value e.g. "C:\\programs" in an
init-procedure. Then, there's a loop in another procedure, 
which reinitializes that variable every time calling that 
init-procedure. 
Also the loop creates every time a new instance of an class where 
is another procedure that could give another value to that variable 
e.g. "C:\\programs\\myprogram", so I want to work with that 
new value further until a new item of the loop occurs.
[Check it?]

Why do I have to declare that variable again as global in that
class-procedre (with :	global PREDEFINED_PROGDIR)?
If I don't do that the value of that variable outside of that 
procedure is the old one.
It seems that python takes a global variable inside a procedure
in a class but not outside.

What's about my "narf" in my brain?

Ciao,
Holger



From siona at chiark.greenend.org.uk  Wed Jun 23 08:02:56 1999
From: siona at chiark.greenend.org.uk (Sion Arrowsmith)
Date: 23 Jun 1999 13:02:56 +0100 (BST)
Subject: Creating a new type with an extension module
References: <Ctp*vl52n@news.chiark.greenend.org.uk> <rvy*CQ92n@news.chiark.greenend.org.uk> <QpMb3.908$7X1.231730@news.shore.net> <faSb3.968$7X1.247240@news.shore.net>
Message-ID: <ugu*8md3n@news.chiark.greenend.org.uk>

In article <faSb3.968$7X1.247240 at news.shore.net>,
Michael P. Reilly <arcege at shore.net> wrote:
>  http://starship.python.net/crew/arcege/extwriting/pyext.html
>
>Let me know how well it reads (I doubt it reads well since it was
>started this morning).

It's a lot more readable than just source code, and a lot more
informative with respect to the type and method structures. Any
chance of you borrowing Guido's time machine so I can read it
48 hours ago?

-- 
\S -- siona at chiark.greenend.org.uk -- http://www.chaos.org.uk/~sion/
  ___  |  "Frankly I have no feelings towards penguins one way or the other"
  \X/  |    -- Arthur C. Clarke
       |  If music be the food of love, bring me a doggie bag.



From delgado at Mathematik.Uni-Bielefeld.DE  Tue Jun 29 14:25:16 1999
From: delgado at Mathematik.Uni-Bielefeld.DE (Olaf Delgado)
Date: Tue, 29 Jun 1999 20:25:16 +0200
Subject: Binary pickle portability
In-Reply-To: <Pine.OSF.4.05.9906291852100.1376-100000@pinax.Mathematik.Uni-Bielefeld.DE>
References: <Pine.OSF.4.05.9906291852100.1376-100000@pinax.Mathematik.Uni-Bielefeld.DE>
Message-ID: <Pine.OSF.4.05.9906292018360.28230-100000@pinax.Mathematik.Uni-Bielefeld.DE>

On Tue, 29 Jun 1999, Olaf Delgado wrote:

> Lately I observed strange things when trying to unpickle a binary pickle
> (including floating point numbers) produced on a different architecture.

Sorry, that was too quick.

After looking a bit closer, I found that the problem is probably related
to the Numeric module (i.e., pickling matrices) and doesn't depend on
whether binary or ordinary mode is used. So please don't bother now!
Still, I am interested in whether people use binary pickles.

Olaf





From kumar*xspam* at *xspam*rtp.ericsson.se  Fri Jun  4 16:35:19 1999
From: kumar*xspam* at *xspam*rtp.ericsson.se (Kumar Balachandran)
Date: 04 Jun 1999 16:35:19 -0400
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7iqhkb$d7s@wiscnews.wiscnet.net> <7iu7ma$d33$1@cronkite.cc.uga.edu> <02hfonzsx2.fsf@rtp.ericsson.se>
Message-ID: <02909zzou0.fsf@rtp.ericsson.se>

What you said is quite wrong. Since Python provides closures, I can
change my programming style to be completely different from the
standard C or Fortran style and use a functional interface. Or, I
could just as easily implement a message-passing interface. So, your
comment about beauty in the language due to discouragement of
individuality is not grounded in reality. The issues that a good
programming language should address include a structure that makes it
easy to use, write in and debug as well. The first and the last are
somewhat marred by the subject of my complaint.

It is quite clear that the authors of the language treat the design of 
the syntax very religiously, and my intent is not to denigrate their
efforts, but I must point out that valuable criticism rarely comes
from people that have accepted the philosophy being espoused, but
rather from persons that are able to rise to a metalevel corresponding 
to the system (in this case Python).

I like to quote an analogy that occured to me sometime ago. When I
started learning C many years ago, I wondered how life without
malloc had been possible. Then I learnt C++ and was equally enamoured by the
new operator and the concept of destructors. Finally I learnt Scheme
and Lisp, and now wonder why people use languages without closures and 
garbage collection. I still have to master continuations, something I
feel ashamed about constantly. The problem of course is that Lisp and its
dialects are like Latin -- quite difficult to use with too many tools
for any given task. The notation tends to throw people off as
well. So, in any collaborative work we have to settle for a lesser
language such as Matlab (blech!)  or Python. What do we get out of it
-- we get closures. We also get a nice interactive environment with
easy prototyping ability. But wait a minute, it's like FORTRAN. It uses
indentation and blank lines as part of the syntax. At least it does
not require a label in a particular column of the line. In conclusion, 
lets face the fact that that design decision was flawed, and people
have to conjure up excuses to save face. 

I did suggest "wend" instead of "elihw", and in any case I was not
presenting a specification, just a suggestion. 

>>>>> "Charles" == Charles G Waldman <cgw at fnal.gov> writes:

    Charles> I vociferously disagree that adding junk like "elihw"
    Charles> adds anything to readability.

    Charles> And one of the beautiful things about Python, especially
    Charles> in projects with more than one programmer, is that it
    Charles> explicitly *discourages* individual programmers from
    Charles> developing their own coding styles, since in general
    Charles> there are *not* multiple ways to do the same thing.
    Charles> "Optional" syntax is very much against this spirit.







From mlh at idt.ntnu.no  Sun Jun  6 18:47:10 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 07 Jun 1999 00:47:10 +0200
Subject: Deleting characters from a string
References: <87n1ye3g4v.fsf@pc-hrvoje.srce.hr> <y0jyahye89i.fsf@vier.idi.ntnu.no> <87vhd2zaju.fsf@pc-hrvoje.srce.hr>
Message-ID: <y0jvhd1dk0h.fsf@vier.idi.ntnu.no>

Hrvoje Niksic <hniksic at srce.hr> writes:

> mlh at idt.ntnu.no (Magnus L. Hetland) writes:
> 
> > result = ""
> > for char in host:
> >     if char not in ' "':
> >         result = result+char
> > 
> > host = result
> > 
> > What!? To me that's more elegant... <wink>
> 
> To me it would be if Python actually had chars.  <handshake>

What do you mean? The example above works...

> 
> :-)

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From tim_one at email.msn.com  Wed Jun  9 22:22:10 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 9 Jun 1999 22:22:10 -0400
Subject: Python 2.0
In-Reply-To: <u37lpdssbq.fsf@ares.sys1.ngo.okisoft.co.jp>
Message-ID: <000c01beb2e8$0b825440$af9e2299@tim>

[Hisao Suzuki]
> ...
> It would be an issue of distributed garbage collection.  And it
> is a difficulty for automatic GC with finalization.

Yes, there are real problems here, and Python certainly isn't the first to
face them.

> In addition, perhaps we should also take a (close) look at
> ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
>
> It summarizes other (or rather more general) issues to consider
> in the design of a finalization mechanism (and presents one
> solution for Scheme---actually it is adapted by gnu guile).  I'd
> think anyone who is planning improvements on Python's storage
> management has to _notice_ what are summarized there first.

Excellent suggestion; excellent paper, too <wink>.

> (The solution is another story.  It is not so obvious for
> Python.)
>
> # Certainly the current reference counting system of Python
> # reduces the issues to fairly (though _not_ fully) tolerable
> # ones...

It automatically deals with the order-of-finalization problem in a correct
way, by following a topological sort of the "references" graph.  Which says
almost nothing, since that's *natural* for RC.  In return it can't deal with
cycles at all, and indeed it's still easy to write __del__ methods that
trigger mysterious exceptions at system shutdown time, when Python *does*
clean up *some* kinds of cyclic trash (namely the cycles the interpreter
itself creates for its own nefarious purposes).

All in all, though, the lack of "real GC" has never stopped me from doing
anything in Python, nor the lack of "real RC" from doing anything in any
GC'd language.  The lack of "any C" in C <wink> has stopped me often in that
language, though.

the-ocean-holds-many-mysteries-but-we-live-longer-than-fish-ly y'rs  - tim





From jblake at eskimo.com  Fri Jun 18 19:48:20 1999
From: jblake at eskimo.com (Jonathon)
Date: 18 Jun 1999 23:48:20 GMT
Subject: NameError Problem
Message-ID: <7kels4$9tv$1@eskinews.eskimo.com>

	I'm writing a long << 25K line >> script to verify data from a
 	database.
		
	The following bit of script is giving me a NameError, and
	I can't figure out why.

	datafile = open ( the_data_file_is, "r" )
	new_datafile = open ( the_temp_data_file_is, "w" )
	line_kounter = 0
	while line_kounter < 50000
	    line_kounter = line_kounter + 1
	    templine = datafile.readline(line_kounter)
            validate_line = templine
	    if ( the_file_is == "john_www.db" ) :
	         temp_record_id = int(validate_line[2:18])
		 temp_country  = strip(validate_line[20:50])
            elif ( the_file_is == "jan_www.db " ):
                 temp_record_id = int(validate_line[2:10])
                 temp_country = strip(validate_line[12:20])
            print temp_record_id
		
	At that point I get a NameError message.
	All I'm trying to do is have it print to screen
	the record number it is working on.  
	
	I've double checked the white spaces, and they are
	all lined up at 0, 4, 8, 12, 16 etc.
	
	What am I missing here?  << And how do I correct it. >>
	
	<< If it matters, there are actually six more
	elif constructs  each containing
	containing thirty or so initializations.  >>

	<< The person who wrote the original database input functions
	didn't believe in error trapping, with the result that it is
	replete with errors -- and now that database is going onto
	the web --- but has to be error free.  >>

	xan

	jonathon
-- 
	I'm still looking for a good book on
		3:	The Recent Unpleasantness
		1:	The War Of Northern Aggression.
		2:	The War of Southern Rebellion.



From pierre at saiph.com  Sat Jun 12 06:40:44 1999
From: pierre at saiph.com (Pierre Imbaud)
Date: Sat, 12 Jun 1999 11:40:44 +0100
Subject: tkinter and signals
Message-ID: <3762392C.AE94C44D@saiph.com>

Hi everyone.
Im quite new to tkinter. I try to make a program using tkinter, that
is able to listen to some signals: read a pipe, perform periodic
actions. I first tried with SIGALRM, with a widget changing fg color
every second. I dont wait() for SIGALRM, just return in mainloop().
the behavior I get is: I need the mouse to hover on the app, to get
the SIGALRM signal. My diagnostic, right or wrong is: in the mainloop,
the idle state is a wait() for a signal (dunno which) related to an X
event. In this state, an unwaited signal just wont wake up the
process.
It just occurs to me threads might be the solution: for every possible
signal source, spawn a thread that will explicitly wait for this
signal? Im not familiar with thread either, so I better wait for a
confirmation from someone who knows....
Oh, btw, Im using Solaris 2.5.1 on X86 (its an Unix V.4)
Thanks for answering!

-- 
Pierre Imbaud <pierre at saiph.com>
12 Rue des Bosquets 91480 Quincy Sous S?nart France
Tel:  01 69 00 94 57 Fax 79 65



From claird at Starbase.NeoSoft.COM  Fri Jun 11 00:41:41 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 10 Jun 1999 23:41:41 -0500
Subject: High performance Python on NT
References: <000301beb3b7$4f973810$6eba0ac8@kuarajy.infosys.com.ar>
Message-ID: <7jq425$cqb$1@Starbase.NeoSoft.COM>

In article <000301beb3b7$4f973810$6eba0ac8 at kuarajy.infosys.com.ar>,
Bruno Mattarollo <bmatt at acm.org> wrote:
>Hi!
>
>	I am about to start a project where I will need to do some high performance
>Web CGI scripts on WinNT ... I have no choice ... I know it's not the best
>combination. It's not my decision ;-) I am looking for some advice. Is it
>better if I use some kind of Active Python Scripting Page (ASP but with
>Python)? Or just doing CGI with threads it will be fine... ?
>
>	I know this is a very vague question... I just want to know if someone had
>to take a similar decision... The app I will be doing should support several
>hundred users connecting to the Web server and doing some serious CGI
>activity ... Almost all hits that the server will receive will be scripts.
>The hardware is no problem in this case... When can scale up to Quad PII
>Xeon with 1GB RAM ... U2W SCSI discs...
>
>	I know exactly what to do in a UNIX env... but I am not sure of the
>decision to take regarding win32...
			.
			.
			.
ASP is likely to perform a lot better than CGI in
the situation you describe.

It depends on details, though.  It's possible that
there will be enough constraints elsewhere in your
system that users won't notice the difference.  If
CGI is much easier for you, start with that.

I don't understand "CGI with threads".  I suspect
there's a misconception, or at least a vocabulary
mismatch, afoot.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From gmcm at hypernet.com  Wed Jun  9 10:57:42 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 9 Jun 1999 09:57:42 -0500
Subject: Windows 95/98: Python uses console, not stdio
In-Reply-To: <87r9nl33de.fsf@parus.itea.ntnu.no>
Message-ID: <1283181980-72298734@hypernet.com>

Jon K Hellan writes:

> I am unable to run Python in a shell window in emacs on Windows
> 95/98. I use the bash shell, but don't think this makes any
> difference.
> 
> If Python reads/writes from/to console, not stdio, this would be an
> explanation. 
 
But it doesn't. On Windows, stdio (and the whole c runtime lib) is a 
compatibility layer. NT achieves the best compatibility, though 
it still requires workarounds and has a few doesn't-work-at-alls. The 
number of the latter increases on Win95/98. This is probably related 
to a known Win95 bug that prevents piping from working. 

- Gordon



From a.mueller at icrf.icnet.uk  Wed Jun 23 14:02:54 1999
From: a.mueller at icrf.icnet.uk (Arne Mueller)
Date: Wed, 23 Jun 1999 19:02:54 +0100
Subject: interactive-logging
References: <376FF09A.8170CB@icrf.icnet.uk> <y0jso7jghez.fsf@vier.idi.ntnu.no> <87yahbc9fe.fsf@banet.net>
Message-ID: <3771214E.83A76264@icrf.icnet.uk>

meow wrote:
> 
> Magnus L Hetland <mlh at idt.ntnu.no> writes:
> 
> > Arne Mueller <a.mueller at icrf.icnet.uk> writes:
> >> I'd like to add some logging to the interpreter's interactive modus so
> >> that everything I type in and everything what's printed to the screen
> >> (errors and 'normal' output) is copied to a log-file.
> 
> > Hm. Isn't there a UN*X command that does that? (That logs everything
> > going on in a terminal...)
> 
> 'python | tee log.file' would be a simple way to get that kind of
> thing.
Right! Thanks very much, simple and very effective solution ... I
completely forgot about good old tee ;-)

my interactive session:

bash% python 2>&1 | tee python.log 
Python 1.5.2 (#14, Jun 23 1999, 04:05:01)  [GCC egcs-2.91.66 19990314
(egcs-1.1.2  on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
python> x = [1,2,3,4,5]
python> x
[1, 2, 3, 4, 5]
python> for i in x:
...     print i
... 
1
2
3
4
5
python> 
python> 
python> 
python> fir in a:
  File "<stdin>", line 1
    fir in a:
            ^
SyntaxError: invalid syntax
python> 
python> 


and now look in python.log:

bash% cat python.log 
Python 1.5.2 (#14, Jun 23 1999, 04:05:01)  [GCC egcs-2.91.66 19990314
(egcs-1.1.2  on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
python> x = [1,2,3,4,5]
python> x
[1, 2, 3, 4, 5]
python> for i in x:
...     print i
... 
1
2
3
4
5
python> 
python> 
python> 
python> fir in a:
  File "<stdin>", line 1
    fir in a:
            ^
SyntaxError: invalid syntax
python> 
python> 


        thans,

        Arne



From tim_one at email.msn.com  Sun Jun  6 13:50:01 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 6 Jun 1999 13:50:01 -0400
Subject: [pickle] Different version of a class
In-Reply-To: <375a9760@nntp.server.uni-frankfurt.de>
Message-ID: <000001beb045$00b75740$3c9e2299@tim>

[mlauer at asmoday-atm.rz.uni-frankfurt.de]
> my application uses pickle to store and load data.
>
> During developing cycle, many variables are added
> to the classes which are pickled. Whenever I load
> a file pickled with a previous version, some
> variables are not existent (naturally, because
> __init__ is not called). Is there an easy way
> to initialize the "new" variables or must I
> call __init__ - which would overwrite some of
> my saved variables.

As the 1.5.2 pickle docs hint,

    If you plan to have long-lived objects that will see many
    versions of a class, it may be worthwhile to put a version
    number in the objects so that suitable conversions can be
    made by the class's __setstate__() method.

For example, say you start with this class:

class Point:
    version = 1

    def __init__(self, x, y):
        print "in init w/ x, y", x, y
        self.x, self.y = x, y

    def __getstate__(self):
        return (Point.version, self.__dict__)

    def __setstate__(self, (version, dict)):
        if version == 1:
            self.__dict__ = dict
        else:
            raise ValueError("I'm only version %d of Point -- "
                             "can't set state from version %d" %
                             (Point.version, version))

When an instance is pickled, __getstate__ tells it to save away a 2-tuple,
consisting of the class version number and the instance dict.  On
unpickling, this 2-tuple is passed to __setstate__, which can do whatever it
wants.  In this case it's trivial because there is only one version of the
class, but at least version 1 of the class can detect it doesn't know how to
restore data from later versions!

Version 2 of the class may be:

class Point:
    version = 2

    def __init__(self, x, y):
        print "in init w/ x, y", x, y
        self.x, self.y = x, y
        import math
        self.dist = math.sqrt(x**2 + y**2)

    def __getstate__(self):
        return (Point.version, self.__dict__)

    def __setstate__(self, (version, dict)):
        if version == 1:
            self.__dict__ = dict
            import math
            self.dist = math.sqrt(self.x**2 + self.y**2)
        elif version == 2:
            self.__dict__ = dict
        else:
            raise ValueError("I'm only version %d of Point -- "
                             "can't set state from version %d" %
                             (Point.version, version))

That is, __getstate__ doesn't change, but Point.version gets bumped and
__setstate__  looks at the version number in effect at the time the pickle
was made.  Unpickling a version 1 instance now sprouts a dist attr by magic.

Note:  despite the simple-minded <wink> appeal of version numbers, in Python
I find it's almost always better to use hasattr or try/except; that is, to
probe the actual *behavior* rather than rely on always-poorly-documented
"version numbers".  So I'd be more likely to write these toy classes as:

class Point:
    version = 1

    def __init__(self, x, y):
        print "in init w/ x, y", x, y
        self.x, self.y = x, y

and later:

class Point:
    version = 2

    def __init__(self, x, y):
        print "in init w/ x, y", x, y
        self.x, self.y = x, y
        import math
        self.dist = math.sqrt(x**2 + y**2)

    def __setstate__(self, dict):
        self.__dict__ = dict
        if not hasattr(self, 'dist'):
            import math
            self.dist = math.sqrt(self.x**2 + self.y**2)

convolute-to-taste-ly y'rs  - tim





From sams at quinta.com  Wed Jun 23 19:55:31 1999
From: sams at quinta.com (Sam Schulenburg)
Date: Wed, 23 Jun 1999 23:55:31 GMT
Subject: newbie IDLE question
References: <37714A8C.C6E581@aol.com>
Message-ID: <7krs5e$m8a$1@nnrp1.deja.com>

In article <37714A8C.C6E581 at aol.com>,
  Mike Callahan <mikey1cal at aol.com> wrote:
> Where can I find documentation on IDLE? Also, is version 0.5 ready to
be
> downloaded? What about any IDLE vs PTUI comparisons.
>
> Mike Callahan
>

Version 0.5 is part of the beta code at python.org's public CVS mirror.
You need to have CVS installed and running to download this code.
The documentation for IDLE is in the tools\idle directory with all the
source code.
(Check out the information on the following URL to find out how to
obtain CVS and download the code.)
http://www.python.org/download/cvs.html

Samuel W. Schulenburg


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From graham at sloth.math.uga.edu  Wed Jun  9 14:22:57 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 9 Jun 1999 18:22:57 GMT
Subject: fork()
References: <000a01beb21f$a051d380$2fa22299@tim>
Message-ID: <7jmbe1$r22$2@cronkite.cc.uga.edu>

{Graham]
> But the reality is that copying memory can in fact speed applications
> up! Think about locality of reference after you have copied all live
> blocks to the start of memory.
Tim Peters (tim_one at email.msn.com) wrote:
: LOR is an unrelated issue, and harder to analyze.  Yes, copying can speed
: applications.  It can also slow them.  "Although copying garbage collectors
: have predominated in the past, recent studies suggest that the choice
: between mark-sweep and copying garbage collection may well depend as much on
: the behavior of the client program as on the inherent properties of the
: garbage collection algorithm" [Jones & Lins, "Garbage Collection", Wiley,
: 1996].  The hoary old theoretical arguments really aren't worth crap in this
: field <0.5 wink>.

Which "hoary old theoretical arguments" are you talking about Tim.
I said "copying memory *can* speed applications up". That is an
observable reality. Where are the hoary theoretical arguments? Or 
is this just more of the same old bluff and bluster arguing.

graham
-- 
           As you grow up and leave the playground
       where you kissed your prince and found your frog
          Remember the jester that showed you tears
                   the script for tears



From garry at sage.att.com  Fri Jun 11 13:50:07 1999
From: garry at sage.att.com (Garrett G. Hodgson)
Date: Fri, 11 Jun 1999 17:50:07 GMT
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <375FA62E.CCEF5079@cfmu.eurocontrol.be> <19990610153423.A23327@optiplex.palga.uucp>
Message-ID: <37614C4F.AE8D0343@sage.att.com>

Gerrit Holl wrote:

> > Fernando Mato Mira wrote:
> >
> > >   I'm trying to collect a list of all the _current_ issues making tcl
> > > a bad language choice. I'd like concrete examples,
> >
> > (I mean, it's theoretically possible that, in some
> > parallel universe, there's a even problem for which
> > C++ is the ideal choice.)
> >
> 
> Can I conclude that C++ isn't a good language? I'm a starting programmer: why
> isn't C++ good for many problems?

well, no one else bit.
c++ is actually a very useful and powerful language, used 
in a great many problem domains.  however, it is very complex,
and demands a good bit of expertise to use effectively.  in the
hands of someone lacking that expertise, it is easy to develop
programs that fail in mysterious ways.

as a starting programmer, you'd be better off using a more
forgiving language, like, say, python.

-- 
Garry Hodgson			comes a time
garry at sage.att.com		when the blind man
Software Innovation Services	takes your hand, says,
AT&T Labs			"don't you see?"



From rfreedm1 at nospam.rochester.rr.com  Mon Jun 14 20:38:12 1999
From: rfreedm1 at nospam.rochester.rr.com (Benjamin Schollnick)
Date: Tue, 15 Jun 1999 00:38:12 GMT
Subject: Looking for modules?  Config files?  Ini files?  in python....
Message-ID: <rNHvjEdhm5Pp-pn2-WD108jh22jtO@d185d18f8.rochester.rr.com>

Folks,

	I'm trying to find a simple method to deal with *.INI or other styles
of
configuration files....

	I haven't seen anything, and would hate to re-invent the wheel....

	Any suggestions?

				Benjamin



From none at microsoft.com  Thu Jun 10 02:21:25 1999
From: none at microsoft.com (plz_reply_to_group)
Date: Wed, 09 Jun 1999 23:21:25 -0700
Subject: How do I match the last occurence of a character?
References: <375F21B3.D23A5EA9@west.boeing.com>
Message-ID: <375F5965.5923EB3D@microsoft.com>

This is. essentially, what worked. 



import string

file_in = open( 'c:/temp/dd.txt', 'r+' )
for line in file_in.readlines():
	name2 = line[0:string.rfind(line,"/" )]
	print name2



Thanks, everyone!

Jeff P.



"Jeff P." wrote:
> 
> I've been messing with this for a little while now, but I can't get this
> to work. How do I match the last occurence of '\' in a string? I want to
> remove the file name from a url (e.g. news/breakingnews/981001b.html -->
> news/breakingnews/ ).
> 
> Here is the simplest version (that doesn't work). It matches the _first_
> backslash.
> 
> import regex
> 
> file_in = open( 'dir.txt', 'r+' )
> for line in file_in.readlines():
>         print regex.search( '/', line )
> 
> Thanks for any help!
> 
> Jeff P.




From rdev at my-deja.com  Wed Jun 30 13:59:44 1999
From: rdev at my-deja.com (rdev at my-deja.com)
Date: Wed, 30 Jun 1999 18:59:44 +0100
Subject: Reflection in Python?
Message-ID: <000015c7@bossar.com.pl>

From: rdev at my-deja.com

In article <7ldglr$ita$1 at nnrp1.deja.com>,
  anders777 at my-deja.com wrote:
>  ...How do I do the
> equivalent of what Java calls reflection? I want
> to do the following:
>
> test = MathTest("testAdd")
>
> where test.run() will run the "testAdd" method
on
> the MathTest object?  I also want to be able to
> tell the user the name of the object (MathTest)
> and the method in case the method fails.  Any
> thoughts?
>
If I understand what you are trying to do, it's
easy in Python.  Assuming MathTest is an
object instance:
test = MathTest.testAdd
will return you a bound method object which you
can then call as:
test()
Getting the name of the object class and method
is also easy.  There are several ways to do this:
1) Define a doc string within the method by
   putting a quoted string as the first line
   under the def line.  You can then refer to it
   as "test.__doc__".  This could say, for
   example, "Method bar of class foo".
2) "test.__name__" will return the name of the
   method.  "test.im_class.__name__" will
   return the name of the class that the method
   is defined within.

Example:
class MathTest:
  def testAdd(self, parm1, parm2):
      "Method testAdd of Class MathTest"
      result = parm1 + parm2
      return result
<<< mt = MathTest()
<<< test = mt.testAdd
<<< test.run(2,2)
4
<<< print test.__doc__
Method testAdd of Class MathTest
<<< print test.__name__
testAdd
<<< print test.im_class.__name__
MathTest


I hope that helps.

- Roger




> Thanks,
> Anders Schneiderman
> National Journal Daily Briefings Group
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
>



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From jbauer at rubic.com  Thu Jun  3 17:59:43 1999
From: jbauer at rubic.com (Jeff Bauer)
Date: Thu, 3 Jun 1999 21:59:43 GMT
Subject: build of python 1.5.2 with threads on SunOs 5.6 fails
References: <Pine.LNX.4.05.9906032310190.6933-100000@ns.ac-copy.net>
Message-ID: <3756FACF.D62ABE2A@rubic.com>

Joachim Schmitz wrote:
> I compiled Python 1.5.2 on SunOS 5.6/Solaris 2.6 without threads
> successfully. But doing a configure --with-threads, and make results
> in the following error:
> .....

Joachim,

I'd guess this problem is caused by first configuring
Python without threads.  Configure is probably using a
stale config.status -- you can check Modules/Setup.thread.

A 'make clean' might be the easy way to fix your problem.

Best regards,

Jeff Bauer
Rubicon, Inc.



From news at dorb.com  Mon Jun 21 14:49:21 1999
From: news at dorb.com (Darrell)
Date: Mon, 21 Jun 1999 14:49:21 -0400
Subject: choosing random numbers with weights/probability?
References: <929987267.6227@www.remarq.com>
Message-ID: <jRvb3.1444$U6.5827@newsr2.twcny.rr.com>

I'm no expert but....

import whrandom

def popList(l):
 selList=[]
 for i in l:
  for v in range(i[1]):
   selList.append(i[0])
 return selList

list=popList([('one',25),('two',50),('three',25)])
item=whrandom.choice(list)


--
--Darrell





From janssen at parc.xerox.com  Tue Jun 22 22:52:48 1999
From: janssen at parc.xerox.com (Bill Janssen)
Date: Tue, 22 Jun 1999 19:52:48 PDT
Subject: bsddb bloat in shelve?
Message-ID: <crQ4k0UB0KGW0DiuNH@holmes.parc.xerox.com>

Hi!

I've been using shelve (apparently with bsddb, but how does one tell?)
to maintain a few dictionaries of some of the headers in mail messages,
for a mail folder.  I'm finding that the files containing the persistent
data are incredibly bloated.  I'm just keeping a few of the headers for
each message, such as From, Date, and Subject, but the index files for
40MB of total mail are 150MB!  Even full-text-indexing them with Verity
does better:  50 MB of indexing for 40MB of mail.

I wonder if there is some tuning that needs to be done.  For example, my
app never calls sync() on the dictionaries.  Should it?  Will this make
a difference?  What would?

Bill



From a.mueller at icrf.icnet.uk  Tue Jun  8 09:58:03 1999
From: a.mueller at icrf.icnet.uk (Arne Mueller)
Date: Tue, 08 Jun 1999 14:58:03 +0100
Subject: waste of resources ?
References: <375C0F91.1C73E4E3@icrf.icnet.uk> <375C16A9.5ADC4990@icrf.icnet.uk> <375C1FD1.A85565C@aw.sgi.com> <c75so8370by.fsf@pc142.cosc.canterbury.ac.nz>
Message-ID: <375D216B.61787B66@icrf.icnet.uk>

Hi All,


> 
> Another thing to note is that when a child process dies the parent
> will be sent a SIGCHLD signal (I think this also applies if the child
> process was paused with SIGSTOP).  If you just want to ignore the
> return code of the child process, use code like this:
> 
> import signal
> import os
> 
> def sigchild_handler(signum, frame):
>     os.wait()
> 
> signal.signal(signal.SIGCHLD, sigchild_handler)
> 
> This will run os.wait for each child process that exits.  You will
> need to catch the return value of os.wait and inspect it if you care
> what the process returned or whether it was killed by a signal.
> 

Hm, that doesn't work properly  with my program. THe number of zombie
processes is steadily at about 20, which means there are 20 unhandled
died children.  

	Arne



From milliger at ausinfo.com  Mon Jun  7 13:14:55 1999
From: milliger at ausinfo.com (mike milliger)
Date: Mon, 07 Jun 1999 12:14:55 -0500
Subject: Calling .exe's from module
References: <375BEA17.B03B528D@ausinfo.com>
Message-ID: <375BFE0F.600C93F2@ausinfo.com>

thanks for the tips. it works now.
--mike

mike milliger wrote:

> Shouldn't the following module call test.exe? Could someone point me to
> the section in the programming python book which covers this
> functionality?   No objects persist from my executable, so I don't have
> to worry about ref counts and obj's returning to my module. i'm
> basically going to pkzip a list of filenames passed to my module as
> argv's.
>
> #start
> import string,sys,os
>
> # --------------------------------------------------------------------
> # run test
>
> if __name__ == "__main__":
>
>     os.system ("D:\Archiver\test.exe")




From jhauser at ifm.uni-kiel.de  Wed Jun  2 02:57:11 1999
From: jhauser at ifm.uni-kiel.de (Janko Hauser)
Date: 02 Jun 1999 08:57:11 +0200
Subject: complete GNU readline support?
References: <37445622.1EE18CFD@cup.hp.com> <374B8030.A29351CA@_nospam_ken.x13.com> <3754600B.F78CD96A@cup.hp.com>
Message-ID: <87pv3ft7i0.fsf@ifm.uni-kiel.de>

Would you mind to release the new module?

TIA,

__Janko Hauser
-- 
  Institut fuer Meereskunde             phone: 49-431-597 3989
  Dept. Theoretical Oceanography        fax  : 49-431-565876
  Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
  24105 Kiel, Germany



From banderson at boi.hp.com  Fri Jun 18 16:50:36 1999
From: banderson at boi.hp.com (Bill Anderson)
Date: Fri, 18 Jun 1999 14:50:36 -0600
Subject: All this self dot is making me...
References: <Vn653.468$HD.99317@typhoon-sf.snfc21.pbi.net> <m33e0a52x2.fsf@chinon.cnrs-orleans.fr> <37551fa6.90445994@news.oh.verio.com> <8Rb53.1622$nn.483094@news.shore.net> <3757bdfb.262052080@news.oh.verio.com>
Message-ID: <376AB11C.ED2B3040@boi.hp.com>

Kevin Dahlhausen wrote:
> 
> "Michael P. Reilly" <arcege at shore.net> wrote:
> 
> >I think abbreviations would be better:
> >  :abbrev s. self.
> 
> Give it a shot.   The abbreviation has to been seen as a word before it is
> replaced - that is you must have white space after hitting 's.'  So then you
> you have to back up to just after the .
> 
> The mapping runs as soon as you hit the '.' and you can continue typing on the
> fly.


And a similiar problem with mapping .s to self:

"There she was"
becomes:"there she waself"

-- 
Bill Anderson 					Linux Administrator
MCS-Boise (ARC)					banderson at boi.hp.com
My opinions are just that; _my_ opinions.



From jsight at mindspring.com  Mon Jun 21 23:27:06 1999
From: jsight at mindspring.com (Jesse D. Sightler)
Date: Mon, 21 Jun 1999 23:27:06 -0400
Subject: pyGTK on Windows
References: <3757D453.A302E757@interar.com.ar>
Message-ID: <376F028A.E85CD6E9@mindspring.com>

There's always: http://user.sgic.fi/~tml/gimp/win32/

The problem is that I doubt it will support Python right now.  It could
probably be done, though, with some work on your part. :)

Nahuel wrote:
> 
> Hi, im about to do a cross-plataform GUI application in Linux and
> Windoze... i must
> choose the GUI toolkit.. my question is .. can i use pyGTK under
> Windows???
> 
> (I need a gui that can display a table with about 4000 lines without
> crashing (not like Tk :))
> 
> Nahuel Greco.

-- 
---------------
Jesse D. Sightler
http://www3.pair.com/jsight/

"Do not use a hatchet to remove a fly from your friend's forehead." 
      - Chinese Proverb



From nomad at vagabond.com  Thu Jun 10 22:10:09 1999
From: nomad at vagabond.com (nomad at vagabond.com)
Date: Fri, 11 Jun 1999 02:10:09 GMT
Subject: test  8712
Message-ID: <5e_73.12599$mu5.4851@news.rdc1.ct.home.com>

test
ofhzxzqqwulvupmpcjrtlwnnsuuzqzpmygqgxuwhxryyg




From mark at _spamspamlovelyspam_btweng.krakow.pl  Sat Jun 19 05:37:59 1999
From: mark at _spamspamlovelyspam_btweng.krakow.pl (MK)
Date: Sat, 19 Jun 1999 09:37:59 GMT
Subject: Crawling Python
References: <37695707.3168269@news.tpnet.pl> <xj0vhcl5kut.fsf@cyclone.harvard.edu>
Message-ID: <376b63a2.8472828@news.tpnet.pl>

On 18 Jun 1999 16:12:26 -0400, Michael Haggerty <mhagger at blizzard.harvard.edu>
wrote:

>mark at _spamspamlovelyspam_btweng.krakow.pl (MK) writes:
>
>> >>> s=''
>> >>> for i in d.keys():
>> 	s=s+d[i]
>
>Please, before you complain about Python's speed, think twice about
>the algorithm that you are using.  

I don't complain about Python's speed. The speed of Python is fine with me. 
What I rather meant to test was performance of anydbm. Now I do know
that daemon of speed it is not. I only wanted to get a grasp how _actually_
it perfoms.


>The above algorithm is O(n^2),
>which means that if you have n keys the time it takes is proportional
>to n squared.  

Now I am not _that_ lame not to know what O(n^2) means.

>That's because each time you add to the end of the
>string, the whole string has to be copied.  

Oh shit, you're right, only know I realized that s=s+d[i] actually does that
which results in given class of algorithm.

>(Unless there is a
>nontrivial optimization built into Python for this special case, but I
>don't think so.)

>But if instead you do
>
>>>> import string
>>>> s = []
>>>> for i in d.keys():
>>>>   s.append(d[i])
>>>> s = string.join(s, '')

>then you get exactly the same result in O(n) time, because s.append()
>doesn't have to copy each time.  

Thanks. I suspected that .append does not have to copy but was not sure.

>For large n the second algorithm will
>beat the first one every time, by an increasingly huge margin.

Well, good performance of this thing was not important to
me in first place, I actually tried to do some 'crash tests' in
order to find out limits of the system. If I meant performance,
I would write this thing in a different way definitely.

>It may be that idle has problems too, but at least part of this
>problem can be blamed on the algorithm.

Version of IDLE is still 0.4. It works surprisingly good for such an early
version. I would like a few things here and there (for example, the only thing I
sorely miss in IDLE are breakpoints), but overally it's already fine. 





MK


--------------------------------------------------
Reality is something that does not disappear after 
you cease believing in it - VALIS, Philip K. Dick
--------------------------------------------------

Delete _spamspamlovelyspam_ from address to email me

postmaster at 127.0.0.1
root at 127.0.0.1
webmaster at 127.0.0.1






From fredrik at pythonware.com  Thu Jun 24 12:58:33 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 24 Jun 1999 16:58:33 GMT
Subject: htmllib: CR in CDATA
References: <00d201bebc5a$adb129a0$0301a8c0@mnot.net> <012601bebc91$f3ded760$0301a8c0@mnot.net> <014401bebc9a$1963a9e0$0301a8c0@mnot.net> <000b01bebca0$8a022c20$f29b12c2@pythonware.com> <017501bebcad$830a3180$0301a8c0@mnot.net> <003701bebcb4$a6d33fb0$f29b12c2@pythonware.com> <002901bebd00$0a35cbe0$0301a8c0@mnot.net>
Message-ID: <002701bebe64$5e5b6e90$f29b12c2@secret.pythonware.com>

Mark Nottingham wrote:
> > imho, it's pretty reasonable for an SGML parser to
> > behave like an XML parser: split the document up
> > into pieces, but pass them all to the application as
> > untouched as possible. if you wish to implement
> > additional behaviour, do that on the application
> > level. otherwise, you'll end up in a situation where
> > some users cannot use the standard library...
> 
> I see your point. I suppose in a perfect world, the standards would be
> coherent enough that these kinds of choices wouldn't have to be made; this
> shouldn't have to be additional behaviour.

well, I guess this is one of the main reasons why
XML was invented...

</F>




From gerrit.holl at pobox.com  Mon Jun 14 14:47:54 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Mon, 14 Jun 1999 20:47:54 +0200
Subject: WHY are all those sigs ending??
Message-ID: <19990614204754.A26177@optiplex.palga.uucp>

Hello all,

I was very happy when I saw all thos SIGs, but they are all retiring? WHY?
XML-sig, DB-sig, Doc-sig, String-sig: if it's because there isn't interest:
I'm interested!

So why is it?

regards,
Gerrit.

-- 
The Dutch Linuxgames website.		De Nederlandse Linuxgames pagina.
Everything about games on Linux.	Alles over spelletjes onder Linux.
Site address:				http://linuxgames.nl.linux.org
Personal homepage:			http://nl.linux.org/~gerrit/



From marcoxa at copernico.parades.rm.cnr.it  Wed Jun 30 07:21:06 1999
From: marcoxa at copernico.parades.rm.cnr.it (Marco Antoniotti)
Date: 30 Jun 1999 13:21:06 +0200
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <37682DD3.43923243@compaq.com> <7ka2be$qja$1@nntp6.u.washington.edu> <37776FFA.2D85@mailserver.hursley.ibm.com> <lw7looxx7m.fsf@copernico.parades.rm.cnr.it> <876748mcxg.fsf@home.ivm.de> <37788BB9.ABD@mailserver.hursley.ibm.com> <lw67471fyi.fsf@copernico.parades.rm.cnr.it> <3779ED0F.2C67@mailserver.hursley.ibm.com>
Message-ID: <lwoghyexy5.fsf@copernico.parades.rm.cnr.it>

Paul Duffin <pduffin at mailserver.hursley.ibm.com> writes:

> Marco Antoniotti wrote:
> > 
> > Nope. *best* is the Common Lisp Macro system, with Scheme/Dylan syntax
> > stuff a second best.  The notion of 'upvar' in Tcl makes my head
> 
> I am surprised at this because Tcl's [upvar] is simply an explicit form
> of dynamic binding which I seem to remember is what Lisp uses. (That
> is how the 'let' function can be (is) implemented as a lambda
> function).

Ahem!  LET is essentially a macro built on top of LAMBDA application.
Dynamic binding is another beast altogether, which Common Lisp and
Scheme (statically scoped languages) allow in through a back door.
The notion of [upvar] and [uplevel] (with the optional numeric argument which allows
you to inspect N levels of stack is truly hackish.  The utility of
this construct is not in question.  But its use as a
'macro-approximating' device is questionable.

BTW. Is TCL statically or dynamically scoped?

> > spin. :) Finally, AFAIU, a Tcl "macro" must run as an interpreter of
> > the spec. A Common Lisp (Scheme) macro is compiled into regular code
> > by read-time expansion.
> > 
> 
> I find that Lisp macros (while a very powerful and necessary mechanism)
> are sooo confusing. They are in essence another language inside Lisp,
> and as such introduce inconsistencies.

You are very mistaken.  Common Lisp macros are definitively not
another language.  They manipulate S-expressions, which are what Lisp
is made of.  They generate inconsistencies insofar as you are careless
when dealing with possible name capture problems. GENSYM and GENTEMP
are there to help you.

> Tcl on the other hand doesn't
> need a macro language and as such is much more consistent than Lisp.

You severely underestimate the power of Common Lisp (Scheme and Dylan)
macro systems and ascribe to them inconsistencies which are not
there.  It is true that Tcl does not need a macro system. But mostly
because, since you have Common Lisp, you do not need Tcl altogether :)
Apart from some good ole flaming, why doesn't Tcl need a Lisp style
macro system? Just because you have [eval]?  Not a very good
argument. Especially when brought up in front of a Lisp audience.

> 
> If I needed a macro language in Tcl I can just write one.

I just recieved an email from Cameron Laird, citing a 'procm' form in
Tcl, which is supposed to do 'scan time' macros - probably something
in line with the real thing. However, the manual pages for 8.1 at Scriptics
does not mention it.  Yet, I suppose that it is still an experimental
feature that maybe will appear in Tcl in a later edition - a few
lustres after Lisp had them :)

> Lisp and Tcl have a lot more in common than Lispers seem to want to 
> acknowledge.

Ahem!  Do you remember the "revised" Tcl paper by Ousterhout (sorry
for the spelling mistakes :) ), where the essential
addition/correction was the mention of the L-word? :)

Cheers

-- 
Marco Antoniotti ===========================================
PARADES, Via San Pantaleo 66, I-00186 Rome, ITALY
tel. +39 - 06 68 10 03 17, fax. +39 - 06 68 80 79 26
http://www.parades.rm.cnr.it/~marcoxa



From tim_one at email.msn.com  Sat Jun  5 23:25:29 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 5 Jun 1999 23:25:29 -0400
Subject: Looking for for some ideas for argument processing
In-Reply-To: <7jcdhc$htd$1@nnrp1.deja.com>
Message-ID: <000401beafcc$3acf35c0$639e2299@tim>

[Chuck <medcoff at my-deja.com>]
> ...
>         a = BuildArgumentProcessor()
>
>         # process arguments
>         for x in sys.argv[1:]:
>             opt = x[1:2]
>             optArg = x[2:]
>             a.Process(opt, optArg)
>
> This would work if somehow inside the for loop one could "increment
> this list iterator more that once" (sorry I'm coming from a C++
> background).  I don't see anyway of doing this.  Is it possible?  Any
> other ideas?

Some problems are better approached from a Fortran frame of mind <wink>:

    # process arguments
    i, n = 1, len(sys.argv)
    while i < n:
        x = sys.argv[1:]
        opt = x[1:2]
        # etc; each branch needs to bump i appropriately

Or you could build your own class to fake a multi-increment (or even random
access) iterator:  you'd have a lot of fun, but nobody else would understand
it.

as-much-like-c++-as-you-want<wink>-ly y'rs  - tim





From cgw at fnal.gov  Tue Jun 15 22:08:25 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Tue, 15 Jun 1999 21:08:25 -0500 (CDT)
Subject: ANN: Stackless Python 0.2
In-Reply-To: <7k6ttr$i5g$1@brick.cswv.com>
References: <37628EAA.C682F16C@appliedbiometrics.com>
	<7k4fd4$c04$1@brick.cswv.com>
	<376636EA.84240129@appliedbiometrics.com>
	<7k6ttr$i5g$1@brick.cswv.com>
Message-ID: <14183.1817.654330.469062@buffalo.fnal.gov>

Sure sounds like some stray whitespace got between the
backslash and the newline it was intended to protect.  Possibly
due to different conventions about newlines and linefeeds
between Unix and other bogus operating systems?

I've always hated the Backslash-Newline convention! It's one thing to
have *leading* whitespace be significat (a la Python) but when code
doesn't work because of *trailing* whitespace (invisible!) it makes me 
mad!




From guido at cnri.reston.va.us  Thu Jun 10 16:26:14 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 10 Jun 1999 16:26:14 -0400
Subject: fork()
References: <000601beb1c7$682c6220$559e2299@tim> <7jjghd$9il$1@cronkite.cc.uga.edu> <375D5826.23197692@inrialpes.fr> <375D6C60.FF9998DE@appliedbiometrics.com> <5laeu8zmza.fsf@eric.cnri.reston.va.us> <14175.51022.577016.265867@amarok.cnri.reston.va.us>
Message-ID: <5lzp27yf89.fsf@eric.cnri.reston.va.us>

"Andrew M. Kuchling" <akuchlin at mems-exchange.org> writes:

> Guido van Rossum writes:
> >(You can create a cycle using only lists, but I bet that happens only
> >for demonstration purposes, as in ``a = []; a.append(a)''.  So I can
> >live with not freeing such cycles.)
> 
>       I'm doubtful of that argument; I've written classes which track
> all instances that get created, and this can be done with a class
> attribute containing a list, that gets mutated as instances are
> created.  You reconsider this decision later in your post.

You misread me (or I didn't state it very carefully).  I didn't mean
to say that cycles wouldn't contain lists; I meant to say that cycles
typically contain at least one dictionary.

> 	  In this solution, I don't understand how Python would keep
> track of all the dictionaries that get created, in order to determine
> which ones are left unreachable, and hence can be garbage collected.
> From my memory of the Jones-Lins book, this is usually handling by
> doing all the allocation yourself by slicing memory into chunks, and
> then sweeping through it.  It's not obvious (to me, at least) how
> you'd do this with objects created by malloc().  Perhaps lists and
> dictionaries could contain prev/next pointers that are used to
> maintain them in a doubly-linked list, and a linear-time traversal
> would be done to find unmarked objects, but there has to be a better
> way.

A doubly-linked list is exactly what I had in mind.  Sorry I wasn't
more explicit.  I would think that linear time to find unmarked
objects is fine...  Isn't it?  The elegance (I hope) of my scheme is
that it only needs to mark relatively few objects (all dicts and
perhaps all lists) compared to the total number of allocated objects.

Of course finalizers are still screwed, since if an instance is
involved in a cycle, I may have zapped its instance variables before
calling its __del__.  (To break the cycles, the collector would call
the clear() method of all unmarked dicts, which would break the
cycles, and then the ref counting would free the depending objects.)

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



From jonathan at meanwhile.freeserve.co.uk  Fri Jun  4 08:06:42 1999
From: jonathan at meanwhile.freeserve.co.uk (Jonathan)
Date: Fri, 4 Jun 1999 13:06:42 +0100
Subject: Help for beginner
References: <7ipplg$31d7@enews3.newsguy.com> <3751885A.DCF055D6@acm.org>
Message-ID: <7j8fgi$72r$1@news5.svr.pol.co.uk>

Max M. Stalnaker wrote in message <3751885A.DCF055D6 at acm.org>...

>You have to buy all the python books and once you work through those,
(being an
>old-timer) I suggest you work through volume I of Knuth's opus on
programming.
>Other people will have more current suggestions, but no one will complain
about
>a Knuth recommendation.  Embrace the object orientation of python but stay
away
>from GUI implementations for awhile.


Knuth's great, but I *wouldn't* suggest it to someone just starting
programming. It's for people who want to make programming their life's work
and who are happy reading examples in assembly code, happy with mathematical
analyses of algorithms, happy to read thousands of pages of deeply
thoughtful thought, admittedly crystalline but sometimes very dense prose. I
don't think even Knuth would suggest that his books are for beginners.

Get Learning Python if you like the look of it. Then perhaps buy Kernighan
and Pike's Practice Of Programming - it doesn't have any Python in, but
it's an excellent introduction to the issues that matter whatever language
you're using, which is what Max rightly suggests that you need. It gives a
fast intro to data structures and covers program design, debugging,
readability, efficiency. But you can read it in days instead of  the months
and years that Knuth demands to really understand.

Jonathan Coupe, XWare






From Vladimir.Marangozov at inrialpes.fr  Fri Jun  4 08:07:29 1999
From: Vladimir.Marangozov at inrialpes.fr (Vladimir Marangozov)
Date: Fri, 04 Jun 1999 14:07:29 +0200
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net> <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net> <7j34l3$hbj$1@nnrp1.deja.com> <7j66b7$il5$2@cronkite.cc.uga.edu> <3756C693.2C910B29@usa.net> <7j6jnl$58f$3@cronkite.cc.uga.edu>
Message-ID: <3757C181.477E0BD5@inrialpes.fr>

Graham Matthews wrote:
> 
> Why is it so hard for people to see that adding a mark sweep collector
> *ON TOP OF* the existing ref count scheme essentially changes nothing.
> All the current ref count behaviour that people love will stay. The
> only additional behaviour will be that circular refs are mopped up
> correctly.

Why don't you try implementing it?
There's nothing more convincing than good working code.

If you try, I suspect that not we, but you'll realize that it's not so
immediate and that there would be other additional effects, either
inacceptable or undesirable.

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



From zehp at francemel.com  Wed Jun 30 07:15:19 1999
From: zehp at francemel.com (Mathieu Lecarme)
Date: Wed, 30 Jun 1999 13:15:19 +0200
Subject: looking for cookie.py
Message-ID: <7lcu6r$6p$1@minus.oleane.net>

the link on python web site doesn't work!
where can i find it?





From arcege at shore.net  Wed Jun  2 11:02:34 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 02 Jun 1999 15:02:34 GMT
Subject: Posix: Exec format error
References: <3754ED47.54C2F0AF@ilt.fhg.de>
Message-ID: <eIb53.1620$nn.483094@news.shore.net>

Rolf Wester <WESTER at ilt.fhg.de> wrote:
: Hi,

: when doing the following:

: import os
: id=os.fork()
: if id == 0:
:     os.execl("/home/wester/test/test","-d", "indata")
: else:
:     os.wait()

: I get the error message:

: OSError: [Errno 8] Exec format error

: Can aybody tell me how the format should look like?

Usually Errno 8 (ENOEXEC) means that the file is executable but not a
valid, understandable format.  This could be because it is not a valid
binary executable for that machine, the binary is corrupted (eg. copied
over FTP in ASCII mode), or if a shell script does not have a proper
shebang line (#!).

Can you run /home/wester/test/test from the command line?  If it is as
script, is the first line (#!) correct?  Incidently, for OS problems,
it is usually a good idea to say what OS you are using.

  -Arcege




From Fred.L..Drake at p98.f112.n480.z2.fidonet.org  Tue Jun 29 14:02:26 1999
From: Fred.L..Drake at p98.f112.n480.z2.fidonet.org (Fred L. Drake)
Date: Tue, 29 Jun 1999 19:02:26 +0100
Subject: RuntimeError?
Message-ID: <00001583@bossar.com.pl>

From: "Fred L. Drake" <fdrake at cnri.reston.va.us>


Tim Peters writes:
 > Honestly, I had never seen a RuntimeError before!  I found this one by

  Ooh, lucky you!  ;-)  Actually, I've seen very few.

 > digging thru the source code to see when it gets raised.  There's really no
 > pattern to it, and the RuntimeError docstring in Lib/exceptions.py

  When RuntimeError is raised by the interpreter itself, that seems to 
be the case.  When documenting modules, I've seen that a lot of them
do things like "raise RuntimeError" (note: no parameters) whenever
they need to raise an exception and the programmer hadn't already
defined an exception specific to the module.  I consider this a real
problem, since it becomes hard to catch just the right exception.  The 
ones I've seen in modules didn't seem to be so serious, just lazy
about defining new exceptions the might only get used in one place.
  I've not maintained a list of where I've see this, unfortunately:
fixing them would be a worthwhile activity.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From mgm at unpkhswm04.bscc.bls.com  Wed Jun 23 14:41:36 1999
From: mgm at unpkhswm04.bscc.bls.com (Mitchell Morris)
Date: 23 Jun 1999 18:41:36 GMT
Subject: It's just magic, and I hate it ...
References: <slrn7mss5c.f90.mgm@unpkhswm04.bscc.bls.com> <14190.35530.772547.166720@buffalo.fnal.gov> <slrn7mvlhr.4is.mgm@unpkhswm04.bscc.bls.com> <37703621.B29B72CB@bioreason.com> <slrn7n1r6p.keg.mgm@unpkhswm04.bscc.bls.com>
Message-ID: <slrn7n2aj0.n0c.mgm@unpkhswm04.bscc.bls.com>

As much as I hate to admit I'm an idiot, I got caught by the simplest of
things. When the child forks, it inherits all the open file descriptors of
the parent ... which includes sys.stdin and sys.stdout. Since these two (at
least) are attached to the socket from the browser, the browser doesn't
follow the redirect until *ALL* the fds close.

My new question is how do I close all the connections to the browser after
the fork? The obvious:
	sys.stdin.close()
	sys.stdout.close()
	sys.stderr.close()
didn't get them all, since the browser continues to wait. Is there a
mechanism to determine all opened files?




To recap, here is the offending source

#! /usr/bin/env python

import sys
import cgi
import time
import os
import string
import socket

form = cgi.FieldStorage()
if form.has_key("title"):
	hostname = socket.gethostbyaddr(socket.gethostname())[0]
	uniq = hex(os.getpid())[2:] + hex(int(time.time()))[2:]
	addr = "http://%s/~mgm/results/%s.html" % (hostname, uniq)
	fn = "/home/mgm/public_html/results/%s.html" % uniq

	# Generate the results file
	open(fn, "w").write("""<HTML>
<HEAD>
<META HTTP-EQUIV=REFRESH CONTENT=5>
<TITLE>Results for [%s]</TITLE>
</HEAD>
<BODY>
<P>Beginning search at %s:
</BODY>
</HTML>
""" % (form["title"].value, time.asctime(time.localtime(time.time()))))

	# Parent returns a redirect, child pretends to work for a while
	# accumulating results
	if os.fork() == 0:
		sys.stdin.close()
		sys.stdout.close()
		sys.stderr.close()
		for i in (1, 2, 3):
			txt = open(fn, 'r').readlines()
			txt.insert(len(txt)-2, "<P>Still searching\n")
			open(fn, 'w').write(string.join(txt, ''))
			time.sleep(10)
		open(fn, 'w').write("""<HTML>
<HEAD>
<TITLE>Finished</TITLE>
</HEAD>
<BODY>
<H1>Finished</H1>
<P>If you had results, they'd be here
</BODY>
</HTML>
""")
	else:
		print "Status: 302 Moved"
		print "Location:", addr
else:
	print """Content-type: text/html

<HTML>
<HEAD>
<TITLE>Sample Form</TITLE>
</HEAD>
<BODY>
<P>
<H1>Sample Form</H1>
<FORM METHOD="POST">
Title: <INPUT TYPE="TEXT" NAME="title"><P>
<INPUT TYPE="SUBMIT" NAME="submit">
</FORM>
</BODY>
</HTML>"""


-- 
Mitchell Morris

If at first you don't succeed, see if the loser gets anything.



From ovidiu at cup.hp.com  Tue Jun  1 18:34:51 1999
From: ovidiu at cup.hp.com (Ovidiu Predescu)
Date: Tue, 01 Jun 1999 22:34:51 +0000
Subject: complete GNU readline support?
References: <37445622.1EE18CFD@cup.hp.com> <374B8030.A29351CA@_nospam_ken.x13.com>
Message-ID: <3754600B.F78CD96A@cup.hp.com>

Ken Kinder wrote:
> 
> Ovidiu Predescu wrote:
> >
> > Hi,
> >
> > I'm trying to access some advanced GNU readline capabilities from Python
> > 1.5.2 but it doesn't seem that the readline module supports them. Is
> > there an extension to it that has bindings to GNU readline in a similar
> > way the Perl module has? I'm interested in supporting keymaps and
> > asynchronous reads from the terminal.
> 
> Yeah, I think I saw something like that, but I can't find a URL.  You
> also might take a look at the rlcompleter module.  I think together,
> they do give you all the readline stuff.  Make sure and look at all the
> module functions, not just their enhancements to raw_input.
> 
> - Ken

I did take a look at the readline support but could find what I was
interested in. I was interested in having support for all the various
hooks present in readline, for keymaps support, undo and most important
of all, in the alternate interface, that allows my programs to be listen
to multiple sockets and in the same time have a readline interface.

I ended up writing a new, complete, readline module for Python. The old
interface seems to work pretty nice. The rlcompleter module works the
same the only modification I had to do was to import the new module.

Greetings,
-- 
Ovidiu Predescu <ovidiu at cup.hp.com>
http://www.geocities.com/SiliconValley/Monitor/7464/



From greg.ewing at compaq.com  Sun Jun 27 22:33:14 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Mon, 28 Jun 1999 14:33:14 +1200
Subject: Ellipsis?
References: <19990627112841.A31130@optiplex.palga.uucp>
Message-ID: <3776DEEA.7137E92@compaq.com>

Gerrit Holl wrote:
> 
> Can someone please explain that to me? What's so special about '...'?

I don't think that the core of Python uses it for
anything, but the Numeric module uses it as part
of its array slicing syntax. See the documentation
for the Numeric module for details.

Greg



From nathan at islanddata.com  Sat Jun 26 14:08:24 1999
From: nathan at islanddata.com (nathan at islanddata.com)
Date: Sat, 26 Jun 1999 11:08:24 -0700 (PDT)
Subject: if threads != processes:
Message-ID: <19990626180824.20405.qmail@islanddata.com>

I chose to make my current project multi-threaded rather than forking
for two reasons: speed of context-switches and shared memory.  The
memory part is working out great.  The Queue class seems to be suiting
me just fine.  However, I've noticed that every thread that gets
spawned off is assigned its own process id.  Is this truly less
expensive than forking?  And those processes seem to stick around even
after the target function returns...am I doing something wrong?
Thanks for any input.



From Gaetan_Corneau at baan.com  Wed Jun  9 17:19:25 1999
From: Gaetan_Corneau at baan.com (Gaetan Corneau)
Date: Wed, 9 Jun 1999 21:19:25 GMT
Subject: COM dates
Message-ID: <816010E2456BD111A48700805FBBE2EEFDF353@ex-quebec-u1.baan.com>

Hello,

I have a COM object with a field calles "StartDate". When I type:

print Obj.StartDate, I get:

<PyTime:1/1/70 12:00:00 AM>

Two questions:

1) What is that PyTime object? Is it documented somewhere?

2) Suppose I want to set StartDate to January 1, 1999, what do I write in my
script?

Thanks in advance!
______________________________________________________
   Gaetan Corneau
   Software Developer (System integration Team)
   BaaN  Supply Chain Solutions  
   E-mail: Gaetan_Corneau at baan.com        
   Compuserve: Gaetan_Corneau at compuserve.com  
   ICQ Number: 7395494             
   Tel: (418) 654-1454 ext. 252          
______________________________________________________
"Profanity is the one language all programmers know best"




From ovidiu at cup.hp.com  Fri Jun 18 15:58:54 1999
From: ovidiu at cup.hp.com (Ovidiu Predescu)
Date: Fri, 18 Jun 1999 19:58:54 +0000
Subject: Python source code documentation?
References: <7kdv0l$n0g$1@nnrp1.deja.com>
Message-ID: <376AA4FE.8116C36A@cup.hp.com>

Sam Schulenburg wrote:
> 
> I have been using a freeware product called Autoduck to document my
> Python source code.
>
> [...]
> 
> I would prefer to have a Python based solution that would generate HTML
> files. Can anyone point me towards any information that may start me
> down this path, If a solution does not exist, then I may have to write
> my own Python parser to handle this.

Take a look at pythondoc or gendoc:

http://starship.python.net/crew/danilo/pystuff.html

Greetings,
-- 
Ovidiu Predescu <ovidiu at cup.hp.com>
http://www.geocities.com/SiliconValley/Monitor/7464/



From cgw at fnal.gov  Tue Jun 15 15:53:43 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Tue, 15 Jun 1999 14:53:43 -0500 (CDT)
Subject: socket bindings
In-Reply-To: <XFMail.990615114447.nathan@islanddata.com>
References: <XFMail.990615114447.nathan@islanddata.com>
Message-ID: <14182.44871.31356.207716@buffalo.fnal.gov>


Nathan Clegg writes:
 >> ... if I start the application again, it will die, saying
 >> the address is already bound to.

You need to specify the REUSEADDR flag on the socket, like this:

import socket
s=socket.socket(socket.AF_INET,socket.SOCK_STREAM)
s.setsockopt(socket.SOL_SOCKET,socket.SO_REUSEADDR,1)
s.bind(host,port)



See the Unix manual pages or any decent network programming book for
more info on setsockopt.



From gerrit.holl at pobox.com  Thu Jun 10 15:17:44 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Thu, 10 Jun 1999 21:17:44 +0200
Subject: Free Sex Links  7546
In-Reply-To: <375fd775@nemo.idirect.com>; from Snoopy :-\ on Thu, Jun 10, 1999 at 11:13:36AM -0400
References: <3CN73.93097$pl3.57568@newreader.ukcore.bt.net> <375fd775@nemo.idirect.com>
Message-ID: <19990610211744.A7465@optiplex.palga.uucp>

On Thu, Jun 10, 1999 at 11:13:36AM -0400, Snoopy :-\ wrote:
> From: "Snoopy :-\)\)" <genius at acme.ca>
> Newsgroups: comp.lang.python
> Subject: Re: Free Sex Links  7546
> Date: Thu, 10 Jun 1999 11:13:36 -0400
> X-Newsreader: Microsoft Outlook Express 4.72.3155.0
> 
> F#@*^ OFF ;-((
> 

True thing, I totally agree to that.
kvrekp, do whatever you want, but don't bother us please...

regards,
Gerrit.

-- 
The Dutch Linuxgames website.		De Nederlandse Linuxgames pagina.
Everything about games on Linux.	Alles over spelletjes onder Linux.
Site address:				http://linuxgames.nl.linux.org
Personal homepage:			http://nl.linux.org/~gerrit/



From ivnowa at hvision.nl  Fri Jun  4 17:27:13 1999
From: ivnowa at hvision.nl (Hans Nowak)
Date: Fri, 4 Jun 1999 21:27:13 GMT
Subject: Python 2.0
In-Reply-To: <02hfonzsx2.fsf@rtp.ericsson.se>
References: <02hfonzsx2.fsf@rtp.ericsson.se>
Message-ID: <199906042125.XAA12398@axil.hvision.nl>


On 4 Jun 99, Kumar Balachandran wrote:

> Here is some useful evangelizing (methinks). One of the irritating things about
> Python is the use of whitespace in syntax. Agreed, the code is
> readable without parentheses or braces, but why not have optional
> syntactic sugar such as
> 
> if
> ...
> elif
> ...
> else
> ...
> fi
> 
> def
> ...
> ...
> fed
> 
> 
> while
> ...
> elihw or wend
> 
> 
> 
> for
> ...
> rof
> 
> etc. It makes the language more elegant. When I see code using
> indentation or blank likes to achieve blocking of constructs, it
> reminds me of an old language (FORTRAN I think it was called:-).

If you think it's more elegant, why not try things like this:

if blahblah:
    do_this()
    do_that()
#fi

for x in my_sequence:
    print x
    print booze.append(x)
#rof

Or use #endif, #endfor, etc. People have even "implemented" curly 
braces this way. It's as optional as you can get, and it really 
doesn't require a syntax change.

Veel liefs,

--Hans Nowak (ivnowa at hvision.nl)
Homepage: http://fly.to/zephyrfalcon



From rceci at my-deja.com  Mon Jun  7 19:08:31 1999
From: rceci at my-deja.com (rceci at my-deja.com)
Date: Mon, 07 Jun 1999 23:08:31 GMT
Subject: Question about envvars
References: <7jgl9n$p13$1@nnrp1.deja.com> <375BFE38.3FB3E83C@ix.netcom.com> <7jh7pc$as$1@nnrp1.deja.com> <375C3891.1D11E1A4@ix.netcom.com>
Message-ID: <7jhjd8$4g2$1@nnrp1.deja.com>

Stephen,

Yes -S does work!.  Being rather new to python, what exactly do I lose
if I specify that switch ?  What is in the site module?  Can I
compensate for the -S with manual "import <module>" commands in my
script?


In article <375C3891.1D11E1A4 at ix.netcom.com>,
  "Stephen J. Turner" <sjturner at ix.netcom.com> wrote:
> Hi Rob,
>
> > Boy I was excited to get a reasonable explaination so quickly, but
alas,
> > it still does not work.  Here is my test script:
> >
> > #!/usr/local/bin/python
> >
> > import MDIreg
> >
> > reg = MDIreg.FileRegistry(1,1)
> > reg.exportEnvVars()
> >
> > import os
> > print "child has it : "
> > os.system("env | grep MDI_REG_VERSION")
> > print "but do I have it? :"
> > print os.environ['MDI_REG_VERSION']
> >
> > ---
> >
> > I am interested in this "MDI_REG_VERSION" envvar which gets exported
by
> > my library.  Here is the output:
> >
> > child has it :
> > MDI_REG_VERSION=10.0
> > but do I have it? :
> > Traceback (innermost last):
> >   File "regtest.py", line 12, in ?
> >     print os.environ['MDI_REG_VERSION']
> >   File "/usr/local/lib/python1.5/UserDict.py", line 12, in
__getitem__
> >     def __getitem__(self, key): return self.data[key]
> > KeyError: MDI_REG_VERSION
>
> Oops!  I just realized that the site module, which is imported by
> default when the Python interpreter starts, imports the os module --
so
> by the time your script begins to execute, it's already too late.
Just
> for grins, you could try disabling the site module by adding the '-S'
> option to the Python command line, as in:
>
> #!/usr/local/bin/python -S
>
> I'm not advocating this approach, since it's generally a good idea to
> import the site module.  I'd just like to know if it works for you.
>
> > Any other ideas?  Any way to tell the os/posix module to "refresh"?
>
> None that I can see from looking in Modules/posixmodule.c.  Sorry.
>
> > By the way, thanks for the library implementation suggestion, but I
must
> > keep it the way it is, because other c, and c++ (at the class level)
> > programs access this library.
>
> OK, how 'bout this?  Suppose you added a virtual "putenv" method to
your
> FileRegistry C++ class whose default implementation was to call the
> global putenv function.  Then in your Python C extension, you could
> subclass FileRegistry and override the putenv method with one that
> modifies os.environ as described in the prior post.
>
> Regards,
> Stephen
>
> --
> Stephen J. Turner <sjturner at ix.netcom.com>
>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From neelk at brick.cswv.com  Sun Jun 13 12:52:53 1999
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 13 Jun 1999 11:52:53 -0500
Subject: dict vs kjBuckets vs ???
References: <000801beb3b1$5c5af240$329e2299@tim>
Message-ID: <7k0nl5$bgu$1@brick.cswv.com>

In article <000801beb3b1$5c5af240$329e2299 at tim>,
Tim Peters <tim_one at email.msn.com> wrote:
>[MK]
>> I plan to write a program that would store lots (in range of 10M or even
>> more) of relatively small objects (a few hundred bytes at most), so what
>> do you think I should use?
>
>Let's do a little math <wink>:  10M * 100 = ?, a lower bound on what you're
>contemplating.  Do you have gigabytes of RAM?

That might actually be reasonable, if you do a little more math. RAM
is going for about a dollar a meg right now, so 2 gigs of RAM will
cost around $2000. This is price-competitive with tying up a
programmer for a week or two.

I'm always surprised by how effective brute force and ignorance can
be. :)


Neel



From phd at emerald.netskate.ru  Mon Jun  7 05:21:42 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Mon, 7 Jun 1999 13:21:42 +0400 (MSD)
Subject: CGI module misbehaving?
In-Reply-To: <375B8BCF.C55B1F31@home.com>
Message-ID: <Pine.LNX.4.04.9906071320360.5443-100000@emerald.netskate.ru>

Hi!

On Mon, 7 Jun 1999, Jim Meier wrote:
> if form.has_key["name"]:

if form.has_key("name"):

   has_key is a function!

Oleg.
---- 
    Oleg Broytmann  National Research Surgery Centre  http://sun.med.ru/~phd/
           Programmers don't die, they just GOSUB without RETURN.




From news at dorb.com  Tue Jun 29 10:32:31 1999
From: news at dorb.com (Darrell)
Date: Tue, 29 Jun 1999 10:32:31 -0400
Subject: Python for large projects?
References: <3774b5d1.690922@news.force9.net>  <T6bd3.2883$U6.12014@newsr2.twcny.rr.com> <slrn7naup9.jk.thantos@brimstone.mecha> <RGid3.2957$U6.12374@newsr2.twcny.rr.com> <87n1xm85mf.fsf@lisboa.i-did-not-set--mail-host-address--so-shoot-me> <Jhsd3.3090$U6.12410@newsr2.twcny.rr.com> <3776EF65.A07CBEFC@compaq.com> <slrn7ne533.1rmr.marc@oscar.noc.cv.net> <xeLd3.3224$U6.13605@newsr2.twcny.rr.com> <q6Od3.3254$U6.12368@newsr2.twcny.rr.com> <377829FE.EA457FDB@compaq.com>
Message-ID: <lQ4e3.3642$U6.15685@newsr2.twcny.rr.com>

Looks like Marc-Andr? Lemburg already did this.
http://starship.python.net/crew/lemburg/

--
--Darrell
Greg Ewing <greg.ewing at compaq.com> wrote in message
news:377829FE.EA457FDB at compaq.com...
> Darrell wrote:
> >
> > Read up on RExec and don't think it's the solution.
>
> I think the only way to make an exposed data object
> completely tamper-proof in Python is, as was suggested,
> implement it as a C extension.
>
> Maybe it could maybe be a thin C layer that has Python on
> the other side. Would it be feasible to design a
> general-purpose "rinstance" object that does for
> data structures what rexec does for namespaces?
>
> Greg





From arcege at shore.net  Tue Jun  8 11:57:22 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 08 Jun 1999 15:57:22 GMT
Subject: Indentation and grouping proposal
References: <000001beb0a1$bc540a00$5b9e2299@tim> <375BCAD1.BE6E391B@erols.com> <iCQ63.1184$VL2.175421@news.direcpc.com>
Message-ID: <C3b73.2301$nn.709127@news.shore.net>

Emile van Sebille <emile at fenx.com> wrote:
: Re: One Liners:

:>>> def myexec(b):
: ...  for stmnt in string.split(b,'\n'):
: ...   exec(stmnt)
: ...
:>>> b="if a: print 'Spam!';print 'Spam2!';\nprint'Yum!'"
:>>> myexec(b)
: Yum!
:>>> a=1
:>>> myexec(b)
: Spam!
: Spam2!
: Yum!
:>>>

Unfortunately, this doesn't work in other namespaces and frames,  but
the following does.

$ cat fexec.py
def myexec(stmts):
  import string, sys
  try:
    raise SystemError
  except SystemError:
    tb = sys.exc_info()[-1]
  f = tb.tb_frame.f_back
  g, l = f.f_globals, f.f_locals
  del tb, f
  for stmt in string.splitfields(stmts, '\n'):
    exec stmt in g, l

$ cat foo.py
import fexec

def foo(x):
  fexec.myexec('print x')
foo(10)

$ python foo.py
10
$

The trick is to get locals and globals of myexec's caller (which could
be in another module).  But, do we know how safe this is? ;)

  -Arcege




From fredrik at pythonware.com  Thu Jun 24 13:09:47 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 24 Jun 1999 17:09:47 GMT
Subject: tkMessageBox returns 'can't invoke "frame" command:' error
References: <7kktvv$550$1@nnrp1.deja.com> <c75so7lklce.fsf@pc142.cosc.canterbury.ac.nz> <7ksujf$1jp$1@nnrp1.deja.com>
Message-ID: <002801bebe64$5ebec440$f29b12c2@secret.pythonware.com>

tlng at phileo.com.my wrote:
> If I import my script above for the first time, my script runs fine.
> However, after I quit my script and try to reload my script at the
> Python interpreter by typing reload(myscriptname), I get the error
> message: "TclError: Can't invoke "frame" command:". Actually, it is
> quite a long-winded error message, I'm just giving the last line.

you could try passing the root window to askyesno, like
this:

reply = tkMessageBox.askyesno("foo", "bar!", master=rootWindow)

no guarantees, though.

> > The actuall error is caused by the Tk mainloop cannot be restarted,
> > obviously Tkinter was never designed to support what you're doing
> > here.
> 
> Oh, you mean I can't reload my script if my script contains Tkinter
> calls? How do you test your Tkinter scripts then?

well, dejanews has the answer, as usual:

>    "Incidently, in a recent project, we successfully used a combination of
>    random generators, chimps, and 3-year olds to produce the actual code
>    base, while the engineers concentrated their efforts on testing."

(since then, all the engineers have quit. go figure...)

> What I usually do is, I write my Python script in a text editor, switch
> to the Python interpreter and reload my script to see if my script is
> working. Since I can't do this with Tkinter scripts, does that mean
> everytime I want to retest my Tkinter scripts, I have to quit Python
> and restart Python?

I usually type ctrl-C twice to test my code.  maybe you
could train your editor to do something similar for you?

> > The cause of your problem seems be partly that you actually
> > expected this to work, putting lots of code in a module
> > initialisation is not good style.
> 
> That bit about "putting lots of code in a module initialisation" just
> went over my head. What did I do that constitutes "module
> initialisation"? I've reposted my sample script below so you can point
> it out to me easily.

code that's not in a def or a class are executed when you
run your program as a script, *and* if you import your pro-
gram as a module.  it's usually better to put the important
stuff in a function instead, by various reasons (it runs faster,
it's easier to deal with the day you decide to turn your code
into a module, etc, etc).

</F>




From bruises at my-deja.com  Mon Jun 14 23:20:08 1999
From: bruises at my-deja.com (bruises at my-deja.com)
Date: Tue, 15 Jun 1999 03:20:08 GMT
Subject: Installing ODBC on PythonWin
Message-ID: <7k4gp1$avj$1@nnrp1.deja.com>

Before I go wack and take it out on my computer
like this guy
(http://www.web42.com/badday/badday_small.mpg)
did, I'd like to ask for some help.

I'm trying to install an ODBC module into
PythonWin, but it still *won't* work!

It is written that the ODBC module is included in
the pythonwin distribution but >>>import ODBC
says otherwise. I've d/e'd (downloaded/executed)
Win32, but it died and said that it was looking
for version 1.4!

I was able to partially install mxODBC by copying
all the files into the right place but I'm having
trouble compiling the modules in vc++.

I'm going to pull my hair out!!! Could someone
show me an easier way of installing ODBC? (I
don't want to mess with DAO since I'm expecting
to migrate data from a database.)

Hhelp!!

SL


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From kumar*xspam* at *xspam*rtp.ericsson.se  Fri Jun  4 15:07:05 1999
From: kumar*xspam* at *xspam*rtp.ericsson.se (Kumar Balachandran)
Date: 04 Jun 1999 15:07:05 -0400
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7iqhkb$d7s@wiscnews.wiscnet.net> <7iu7ma$d33$1@cronkite.cc.uga.edu>
Message-ID: <02hfonzsx2.fsf@rtp.ericsson.se>


>>>>> "Graham" == Graham Matthews <graham at sloth.math.uga.edu> writes:

    Graham> Graham Matthews wrote in message
    Graham> <7ik6mi$lbk$1 at cronkite.cc.uga.edu>...
    Graham> You are envangelising (a not uncommon response when
    Graham> someone remotely criticises Python). Stop evangelising and
    Graham> start considering the technical issues involve (read (!) 
    Graham> other posts for what those issues are).
 
    Graham> graham
Here is some useful evangelizing (methinks). One of the irritating things about
Python is the use of whitespace in syntax. Agreed, the code is
readable without parentheses or braces, but why not have optional
syntactic sugar such as

if
...
elif
...
else
...
fi

def
...
...
fed


while
...
elihw or wend



for
...
rof

etc. It makes the language more elegant. When I see code using
indentation or blank likes to achieve blocking of constructs, it
reminds me of an old language (FORTRAN I think it was called:-).

The change is simple to achieve if backward compatibility is given to
accomodate people that grew up with FORTRAN.





From da at ski.org  Tue Jun 22 12:59:30 1999
From: da at ski.org (David Ascher)
Date: Tue, 22 Jun 1999 09:59:30 -0700 (Pacific Daylight Time)
Subject: Q: httplib patches for proxy support
Message-ID: <Pine.WNT.4.04.9906220958590.310-100000@rigoletto.ski.org>

Has anyone patched httplib.py to add support for proxies?

--david




From c.evans at clear.net.nz  Wed Jun  9 07:26:47 1999
From: c.evans at clear.net.nz (Carey Evans)
Date: 09 Jun 1999 23:26:47 +1200
Subject: waste of resources ?
References: <375C0F91.1C73E4E3@icrf.icnet.uk> <375C16A9.5ADC4990@icrf.icnet.uk> <375C1FD1.A85565C@aw.sgi.com> <c75so8370by.fsf@pc142.cosc.canterbury.ac.nz> <375D216B.61787B66@icrf.icnet.uk>
Message-ID: <87r9nl398o.fsf@psyche.evansnet>

Arne Mueller <a.mueller at icrf.icnet.uk> writes:


[snip]

> > def sigchild_handler(signum, frame):
> >     os.wait()

[snip]

> Hm, that doesn't work properly  with my program. THe number of zombie
> processes is steadily at about 20, which means there are 20 unhandled
> died children.  

Unix doesn't guarantee you'll get exactly one signal delivered per
event.  To be sure to get all the children, you need to loop over the
processes with waitpid().

I think something like this should do it:

----------
def sigchld_handler(signum, frame):
    print "Checking for processes..."
    try:
        while 1:
            pid, sts = os.waitpid(-1, os.WNOHANG)
            if pid == 0:
                break
            print "Process %d ended with status %04x." % (pid, sts)
    except OSError, detail:
        if detail.args[0] != errno.ECHILD:
            raise
        else:
            print "No more children."
    else:
        print "Some children yet to finish."
----------

I can't actually see any way to keep count well enough and avoid race
conditions, to get rid of the try/except.  Sometimes I wish Python had 
POSIX signals, but not badly enough to actually write it.

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

	     "I'm not a god.  I've just been misquoted."



From mkersey at metricom.com  Fri Jun 11 13:41:06 1999
From: mkersey at metricom.com (Michael Kersey)
Date: Fri, 11 Jun 1999 12:41:06 -0500
Subject: Any Production ZOPE sites we can visit?
Message-ID: <37614A32.79FE554E@metricom.com>

Are there many production Zope sites visible on the www?

Of the links from the Zope home page (Case Studies), only Musi-Cal at
http://www.musi-cal.com/ _appears_ (to the untrained eye!-)) to be a
Zope site.

It's a little tough to get my hands around how Zope works and what it
can do without good examples. 

Is Zope being used primarily for intranets?



From billtut at microsoft.com  Sat Jun 19 18:51:11 1999
From: billtut at microsoft.com (Bill Tutt)
Date: Sat, 19 Jun 1999 22:51:11 GMT
Subject: do anonymous pipes normally work on NT?
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8100DB90DFB@RED-MSG-50>

After digging through this stuff, I went back and looked at the code I had
written for win32popen.cpp. I realized that I had forgotten why I had
written the DuplicateHandle() code, so I went back commented it out from the
KB article that I had originally used, and then modified runproc.py to do
the same thing.

cat.py is now much happier with life with the new and improved runproc.py.
The other important thing is to remember to close the stdin writing handle
only once.
Once you call msvcrt.open_osfhandle, you don't need to call
win32file.CloseHandle on it any more, simply issue a CRT close, and all will
be well.

Notice I also changed the DETACHED_PROCESS to a zero. This is to not do
those wierd things that DETACHED_PROCESS seems to do. Since stdin, etc.. are
pipes, it doesn't matter if the process technically has access to
runproc.py's NT console. (if any)

The other alternative from 0 would be to pass in CREATE_NEW_CONSOLE, and
tweak win32process_win32module.cpp to add:
	{"wShowWindow",	     T_INT,  OFF(m_startupinfo.wShowWindow)},
//@prop integer|wShowWindow|

onto the end of PySTARTUPINFO::memberlist, and then do the following:
        StartupInfo = win32process.STARTUPINFO()
        StartupInfo.hStdInput  = hStdin_r
        StartupInfo.hStdOutput = hStdout_w
        StartupInfo.hStdError  = hStderr_w
        StartupInfo.dwFlags = win32process.STARTF_USESTDHANDLES | 
		win32process.STARTF_USESHOWWINDOW
	  StartupInfo.wShowWindow = win32con.SW_HIDE

Mark: please add that line when you get a chance. :)

Bill

'''runproc.py

start a process with three inherited pipes.
Try to write to and read from those.
'''

import win32api
import win32pipe
import win32file
import win32process
import win32security
import win32con
import msvcrt
import os

class Process:
    def run(self, cmdline):
        # security attributes for pipes
        sAttrs = win32security.SECURITY_ATTRIBUTES()
        sAttrs.bInheritHandle = 1

        # create pipes
        hStdin_r,  self.hStdin_w  = win32pipe.CreatePipe(sAttrs, 0)
        self.hStdout_r, hStdout_w = win32pipe.CreatePipe(sAttrs, 0)
        self.hStderr_r, hStderr_w = win32pipe.CreatePipe(sAttrs, 0)

        # set the info structure for the new process.
        StartupInfo = win32process.STARTUPINFO()
        StartupInfo.hStdInput  = hStdin_r
        StartupInfo.hStdOutput = hStdout_w
        StartupInfo.hStdError  = hStderr_w
        StartupInfo.dwFlags = win32process.STARTF_USESTDHANDLES
        
        # Create new output read handles and the input write handle. Set
        # the inheritance properties to FALSE. Otherwise, the child inherits
        # the these handles; resulting in non-closeable handles to the pipes
        # being created.
        pid = win32api.GetCurrentProcess()

        tmp = win32api.DuplicateHandle(
            pid,
            self.hStdin_w,
            pid,
            0,
            0,     # non-inheritable!!
            win32con.DUPLICATE_SAME_ACCESS)
        # Close the inhertible version of the handle
        win32file.CloseHandle(self.hStdin_w)
        self.hStdin_w = tmp
        tmp = win32api.DuplicateHandle(
            pid,
            self.hStdout_r,
            pid,
            0,
            0,     # non-inheritable!
            win32con.DUPLICATE_SAME_ACCESS)
        # Close the inhertible version of the handle
        win32file.CloseHandle(self.hStdout_r)
        self.hStdout_r = tmp

        # start the process.
        hProcess, hThread, dwPid, dwTid = win32process.CreateProcess(
                None,   # program
                cmdline,# command line
                None,   # process security attributes
                None,   # thread attributes
                1,      # inherit handles, or USESTDHANDLES won't work.
                        # creation flags. Don't access the console.
                0,
                None,   # no new environment
                None,   # current directory (stay where we are)
                StartupInfo)
        # normally, we would save the pid etc. here...

        # Child is launched. Close the parents copy of those pipe handles
        # that only the child should have open.
        # You need to make sure that no handles to the write end of the
        # output pipe are maintained in this process or else the pipe will
        # not close when the child process exits and the ReadFile will hang.
        win32file.CloseHandle(hStderr_w)
        win32file.CloseHandle(hStdout_w)
        win32file.CloseHandle(hStdin_r)

        self.stdin = os.fdopen(msvcrt.open_osfhandle(self.hStdin_w, 0),
"wb")
        self.stdin.write('hmmmmm\n')
        self.stdin.flush()
        self.stdin.close()

        self.stdout = os.fdopen(msvcrt.open_osfhandle(self.hStdout_r, 0),
"rb")
        print "Read on stdout: ", repr(self.stdout.read())

        self.stderr = os.fdopen(msvcrt.open_osfhandle(self.hStderr_r, 0),
"rb")
        print "Read on stderr: ", repr(self.stderr.read())
        

if __name__ == '__main__':
    p = Process()
    p.run(r'd:\pythondevtree\python\dist\src\pcbuild\python_d.exe
d:\cat.py')

# end of runproc.py




From mstenber at cc.Helsinki.FI  Thu Jun  3 00:58:46 1999
From: mstenber at cc.Helsinki.FI (Markus Stenberg)
Date: 03 Jun 1999 07:58:46 +0300
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <199906022028.WAA28874@axil.hvision.nl>
Message-ID: <al87lplg9rt.fsf@kruuna.Helsinki.FI>

"Hans Nowak" <ivnowa at hvision.nl> writes:
> Sometimes I dare read a few messages in the Perl newsgroups, and some 
> people there seem to think of Python as the summum of inflexibility. 
> It doesn't allow you enough freedom to code what you want. Just look 
> at that whitespace! -- Nonsense, of course. The fact that Python 
> heavily uses whitespace doesn't mean you cannot code what you want, 
> and in a way you want it. The same thing can be said for Scheme and 
> Lisp. They use lots of parentheses, and while the languages look 
> peculiar because of that, it doesn't make them any less flexible. On 
> the contrary.

Au contraire; Python isn't whitespace-fasist, really. For basic
GNU-indentation-style CurlyPython it takes only few-line preprocessor to
produce from this, beautiful GNU-style CurlyPython example: (Ok, I should
call it 'slant'-style, however, it _is_ whitespace-neutral)

-------- cut -------- curly_hello.cpy -------- cut -------- 
        class foo
       {
      def x(self): print 'hello'
     #.. or..
    def y(self)
   {
  print 'hello'
 }
}

foo().y()
-------- cut -------- 

Ugly, whitespace-infested result like this:

-------- cut -------- curly_hello.py -------- cut -------- 
class foo:
   def x(self): print 'hello'
   #.. or..
   def y(self):
      print 'hello'


foo().y()
-------- cut -------- 

.. Now to add CurlyPython support to __import__ so it can eat .cpy:s and
preprocess them automatically. Hmm. :) 

(As a joke, simplified CurlyPython converter supplied at end of the
message)

> Whitespace in Python enforces readability, it does not force you to 
> use a certain coding style. I really cannot understand why some 
> people seem to think Python is not flexible. It actually allows very 
> much, just takes out brain-damaged things, like for instance adding 
> strings and integers, which makes as much sense as adding file 
> objects and complex numbers. :o)

True.

> > Okay, I *will* say the same about Perl and its curly braces requirement:
> > 
> > } That is, if something's worth doing, it's worth driving into the
> > ground to the
> > } exclusion of all other approaches. Look at the use of curly braces in
> > } Perl.
> $Yes, (%(and) quite) @a { $few %other $annoying {characters...} }

No shit - but that's being _post-modern_, not fascistic forcing to using
obscure non-whitespace markers for all variables! Really!

> > Okay, sorry for all this, it's out of line, I just needed to blow off
> > steam. :)
> Me too. :)

Join the club.

> Veel liefs,
> 
> --Hans Nowak (ivnowa at hvision.nl)
> Homepage: http://fly.to/zephyrfalcon

-Not-yet-member-of-Larry-Wall-fanclub-ly yours, 
	Markus

(and now the uber-ugly CurlyPython converter as I promised ;>)

-------- cut -------- 
import re, string

INDENT_LEVEL=4 # per curly

def curl(l):
    res = []
    start_curly = re.compile("^\s*{").match
    end_curly = re.compile("^\s*}").match
    # Note: real parser would make this actually robust + pretty.
    # But CurlyPython is just a joke, anyway ;-)
    ind = 0
    comm = 0
    SPACES = ""* INDENT_LEVEL
    for line in l:
	line = string.strip(line)
	if start_curly(line):
	    res[-1] = res[-1] + ":"
	    ind = ind + 1
	    continue
	if end_curly(line):
	    ind = ind - 1
	    continue
	line = "  "* ind + line
	res.append(line)
    return map(lambda x:x+"\n", res )

if __name__ == '__main__':
	# Usage: python <whatever> filename_base (without .cpy)
	assert(len(sys.argv)>1)
	open(sys.argv[1]+".py",'w').writelines(\
		curl(open(sys.argv[1]+".cpy").readlines()))
-------- cut -------- 
    


-- 
	"If you want to travel around the world and be invited to speak at
	a lot of different places, just write a Unix operating system."
		-- Linus Torvalds



From nahuel at interar.com.ar  Fri Jun  4 09:27:48 1999
From: nahuel at interar.com.ar (Nahuel)
Date: Fri, 04 Jun 1999 10:27:48 -0300
Subject: pyGTK on Windows
Message-ID: <3757D453.A302E757@interar.com.ar>

Hi, im about to do a cross-plataform GUI application in Linux and
Windoze... i must
choose the GUI toolkit.. my question is .. can i use pyGTK under
Windows???

(I need a gui that can display a table with about 4000 lines without
crashing (not like Tk :))

Nahuel Greco.





From no at spam.net  Thu Jun  3 00:19:50 1999
From: no at spam.net (Tom Vrankar)
Date: 3 Jun 1999 04:19:50 GMT
Subject: Q: Proper Radix Formatting of Longs?
Message-ID: <7j4vp6$je8$1@bashir.ici.net>

I use Python a lot for computer hardware testing and communications. It 
generally works great.

However, the (seemingly) wacky asymmetry between things like the % operator 
and string.atoi() when I must handle 32-bit fields such as address and data 
busses continues to drive me crazy. I often run into situations where I may 
need to handle a value like 0x8000_0000 and do math on it, while converting 
between strings and numbers. The % operator won't let me format longs, hex() 
has a fixed format that includes the sometimes inappropriate "0x" or "L", 
eval() considers 0x8000_0000 to be negative, and string.atoi() refuses to set 
the MSb of an integer unless it sees a dash at the beginning. So far, I 
special-case things a lot.

Is there a standard pythonic idiom for using 32-bit integers as if they were 
unsigned, so that 0x8000_0000 +0x8000_0000 =0, just like in C or my hardware 
ALU? Is there a standard pythonic way to format longs into arbitrary radices 
(sort of like an inverse string.atol)? Is there a secret handshake required to 
at least get string.atoi() to happily convert "0x8000_0000" symmetrically with 
"%x" %0x8000_0000? Yes, I know I could probably write some elaborate python 
functions to manually convert back and forth, but when handling megabytes, the 
performance would be unacceptable.

Any help is appreciated.

						twv@




From jam at newimage.com  Tue Jun 15 09:23:39 1999
From: jam at newimage.com (jam)
Date: Tue, 15 Jun 1999 09:23:39 -0400
Subject: Time display without colon
In-Reply-To: <3766234B.3B347856@mediamanager.com.sg>; from John Francis on Tue, Jun 15, 1999 at 05:56:27PM +0800
References: <3766234B.3B347856@mediamanager.com.sg>
Message-ID: <19990615092339.N6454@toast.internal>

On Tue, Jun 15, 1999 at 05:56:27PM +0800, John Francis wrote:
> 
> Hi,
> 
> I want to display time without colon separting them
> 
> I am using strftime(format,tuple) function.
> 
> But the display is like this HH:MM:SS
> 
> But I am trying to put it in a table form by extracting from the text
> file
> 
> TIME: HH:MM:SS
> 
> My extraction of each line is based on the colon splitting the field
> name and the value in the field, but my script goes heywire when I have
> the above format (TIME:HH:MM:SS) is there a way to extract this or to
> display TIME in different format.
> 
> Thanking you
> 
> Regards
> John
> 

greetings,

if you are storing the data to be machine readable (in a database or a file
of some sort), why not just use the 'raw' 'time.time()' value, and don't
bother with strftime at all? this way you can display it for the user in any
way they wish, at runtime even, with or without colons, in different
timezones, etc... it's much cleaner and easier to work with, and it (should)
be portable across platforms.

on some systems, time.time() returns a float, so I usually use
'long(time.time())' and use the result in the database or the datafile or
whatever.

hope that helps.

regards,
J
-- 
|| visit gfd <http://quark.newimage.com/> 
|| psa member #293 <http://www.python.org/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>



From Johann.Hibschman at p98.f112.n480.z2.fidonet.org  Wed Jun 30 14:20:00 1999
From: Johann.Hibschman at p98.f112.n480.z2.fidonet.org (Johann Hibschman)
Date: Wed, 30 Jun 1999 19:20:00 +0100
Subject: sound-playing modules?
Message-ID: <000015ff@bossar.com.pl>

From: Johann Hibschman <johann at physics.berkeley.edu>

Hi,

Are there any modules out there for playing back sound files, such as
.au, .wav, .mp3, and other formats?

Are any of these modules portable between Windows and Linux?

-- Johann

-- 
Johann Hibschman                           johann at physics.berkeley.edu



From jraven at psu.edu  Tue Jun 29 23:31:00 1999
From: jraven at psu.edu (Jeff Raven)
Date: Tue, 29 Jun 1999 23:31:00 -0400
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <59EAFD6B662F0706.86BFF3E906CBF8DE.784D2A219F85BC21@lp.airnews.net> <99062523473500.00769@Vegus> <7l16e5$q4j$1@news.tamu.edu> <3778ACCF.ACE29713@hk.super.net>
Message-ID: <37798F74.6C1DE9A@psu.edu>

Scott Deerwester wrote:
> 
> That's probably true, but Qt IS cross-platform. Has anybody ever
> tried running pyKDE with the Windows version of Qt? I've used
> pyKDE a bit and it seems to win in regard to completeness and
> usability. One "problem" is that pyKDE (unlike Qt) seems to have
> tightly coupled KDE (which isn't cross-platform) with Qt (which
> is).
> 

Well, although Qt is cross-platform, it isn't cross platform
_free_software_. Only the X11/Unix version of Qt is freely
available for use in free software. To develop with Qt for
Windows you need a developer's license -- regardless of whether
your program is open source or not (the Windows version of the
library is not, to my knowledge, available for download). At
$1500+ a seat, that's going to cost you just a wee bit more
than using Tk or wxWindows as your cross-platform toolkit.

And God forbid your software not be open source -- then you'll
need to shell out for a license for both Unix and Windows...

Jeff Raven



From peter.stoehr at weihenstephan.org  Fri Jun 11 17:22:06 1999
From: peter.stoehr at weihenstephan.org (Dr. Peter Stoehr)
Date: Fri, 11 Jun 1999 23:22:06 +0200
Subject: Problem with Lists & Index command
References: <rNHvjEdhm5Pp-pn2-zOTjsVjHebb4@d185d18f8.rochester.rr.com>
Message-ID: <37617DFE.9BE0E03F@weihenstephan.org>

Hi Benjamin,

maybe I missed something, but why don't you a dictionary ?

    Greetings form Bavaria
	Peter
-- 
---------------------------------------------------------------------------
   Dr. Peter Stoehr --- Teisenbergweg 6 --- 85435 Erding --- 08122/47232
---------------------------------------------------------------------------
		I'm the terror that flaps through the night



From dieter at handshake.de  Mon Jun  7 16:42:47 1999
From: dieter at handshake.de (Dieter Maurer)
Date: 07 Jun 1999 22:42:47 +0200
Subject: Python an Tcl/Tk 8.1
References: <7jaeof$7ja$1@news.rapidnet.com> <375BD2BA.9C960A06@bigfoot.com>
Message-ID: <x7pv37d9o8.fsf@lindm.dm>

Hello Jared

Jared Hanson wrote:
> I am new to all of this.  I am running Windows 98, and I just downloaded
> Python and the newest version of Tcl/Tk (8.1).  Unfortunately, it appears
> that Python doesn't like this version.  Doing an import Tkinter command
> prints out an error message saying that its looking for 8.0 dlls.  I was
> wondering what files I need to change to point to 8.1 dlls in order to make
> it work (if that would make it work at all)  Or do I just need to get a copy
> of the 8.0 dlls and stick them in the same folder.  Any help you could offer
> me would be appreciated, just reply to my post.
"Tkinter" as distributed in 1.5.2 will not work with Tk8.1 due
to minor changes in the Tk startup requirements.

To use "Tkinter" with Tk8.1, you must apply the two patches
I posted recently to this newsgroup, and then regenerate
"_tkinter.pyd" with the Tcl/Tk 8.1 distribution.
Thus, you need a source distribution of both
Python and Tcl/Tk and a C development system.

- Dieter




From mbcgpd at hotmail.com  Sun Jun  6 10:11:04 1999
From: mbcgpd at hotmail.com (mbcgpd at hotmail.com)
Date: Sun, 06 Jun 1999 14:11:04 GMT
Subject: FREE XXX SEX PICTURES FOR YOU!  1716
Message-ID: <Yjv63.16993$pl3.21476@newreader.ukcore.bt.net>

For literally 1000s of free sex pictures goto:

http://www.bigwig.net/reality/default.htm
sevhdwbrdtfebytejpoxryqgjjizevkjfkzzdjfopbcjzpgpkgenrngwcbr




From gregm at iname.com  Wed Jun  2 22:29:02 1999
From: gregm at iname.com (Greg McFarlane)
Date: Thu, 3 Jun 1999 02:29:02 GMT
Subject: intractable(?) Tkinter scrollbar problem
In-Reply-To: <3753ECEE.58AE682A@mindspring.com>; from Doug Hellmann on 01 Jun 1999 at 02:20:17PM
References: <19990601001054.36247@nms.otc.com.au> <jonathan-3105992257320001@borg.onegoodidea.net> <19990601124302.48567@nms.otc.com.au> <3753ECEE.58AE682A@mindspring.com>
Message-ID: <19990603122902.18684@nms.otc.com.au>

In private mail, someone suggested adding a delay before actually
mapping/unmapping the scrollbar.  This fix stops the continuous
mapping and unmapping of the scrollbars.  The change was to the
AutoScrollbar class, but I include the whole script below if you want
to try it out. 

The problem with this fix is that the length of the delay is
significant and may be different depending on your window system or
network delays.  For example, on my system (Solaris host, with X
server running locally), a delay of 100ms works well.  If I reduce
this to 10ms, the window resizes twice before settling.  If I reduce
the time even more, the window resizes more times before settling,
until I get to 1ms when the resizing is continuous.

It's not a good solution, but the best I've seen so far :(

Has anyone else seen this problem?

======================================================================
# Tkinter-only script which demonstrates continuous flashing
# (mapping/unmapping) of dynamic scrollbars (autoscrollbars).
#
# When this script is run, the two scrollbars will be continuously
# mapped and unmapped and the window will continuously change size.
#
# Script modified from Mark C Favas, using Fredrik Lundh's
# AutoScrollbar.

import Tkinter

class AutoScrollbar(Tkinter.Scrollbar): 
    def set(self, lo, hi):
	# Delay 100ms before deciding to map or unmap.
	self.after(100,lambda s=self,l=lo,h=hi: s._set(l,h))

    def _set(self,lo,hi):
        print lo, hi
        if float(lo) <= 0.0 and float(hi) >= 1.0:
            self.tk.call("grid", "remove", self)
        else:
            self.grid()
        Tkinter.Scrollbar.set(self, lo, hi)


root = Tkinter.Tk()

frame = Tkinter.Frame(root)
frame.grid()
frame.columnconfigure(0, weight=1)
frame.rowconfigure(0, weight=1)

scrollFrame = Tkinter.Frame(frame)
scrollFrame.columnconfigure(1, weight=1)
scrollFrame.rowconfigure(1, weight=1)
scrollFrame.grid(sticky='news')

vscrollbar = AutoScrollbar(scrollFrame)
vscrollbar.grid(row=1, column=2, sticky='ns')
hscrollbar = AutoScrollbar(scrollFrame, orient='horizontal')
hscrollbar.grid(row=2, column=1, sticky='ew')

borderframe = Tkinter.Frame(scrollFrame)
borderframe.rowconfigure(0, weight=1)
borderframe.columnconfigure(0, weight=1)
borderframe.grid(row=1, column=1, sticky='news')

canvas = Tkinter.Canvas(borderframe,
        highlightthickness=0,
        borderwidth=0,
        background = 'red',
        width = 300,
        height = 200,
        scrollregion = (0, 0, 301, 200),
        yscrollcommand=vscrollbar.set,
        xscrollcommand=hscrollbar.set)
canvas.grid(sticky='news')
vscrollbar.config(command=canvas.yview)
hscrollbar.config(command=canvas.xview)

root.mainloop()
======================================================================

-- 
Greg McFarlane:    INMS Telstra Australia (gregm at iname.com)
Today's forecast:  Sunny, with occasional cloudy periods and a chance
		   of rain in some areas.



From arcege at shore.net  Thu Jun 24 14:05:51 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Thu, 24 Jun 1999 18:05:51 GMT
Subject: tkMessageBox returns 'can't invoke "frame" command:' error
References: <7kktvv$550$1@nnrp1.deja.com> <c75so7lklce.fsf@pc142.cosc.canterbury.ac.nz> <7ksujf$1jp$1@nnrp1.deja.com> <002801bebe64$5ebec440$f29b12c2@secret.pythonware.com>
Message-ID: <3suc3.102$Of3.18161@news.shore.net>

Fredrik Lundh <fredrik at pythonware.com> wrote:
: tlng at phileo.com.my wrote:
:> If I import my script above for the first time, my script runs fine.
:> However, after I quit my script and try to reload my script at the
:> Python interpreter by typing reload(myscriptname), I get the error
:> message: "TclError: Can't invoke "frame" command:". Actually, it is
:> quite a long-winded error message, I'm just giving the last line.

: you could try passing the root window to askyesno, like
: this:

: reply = tkMessageBox.askyesno("foo", "bar!", master=rootWindow)

: no guarantees, though.

:> > The actuall error is caused by the Tk mainloop cannot be restarted,
:> > obviously Tkinter was never designed to support what you're doing
:> > here.
:> 
:> Oh, you mean I can't reload my script if my script contains Tkinter
:> calls? How do you test your Tkinter scripts then?

: well, dejanews has the answer, as usual:

:>    "Incidently, in a recent project, we successfully used a combination of
:>    random generators, chimps, and 3-year olds to produce the actual code
:>    base, while the engineers concentrated their efforts on testing."

: (since then, all the engineers have quit. go figure...)

:> What I usually do is, I write my Python script in a text editor, switch
:> to the Python interpreter and reload my script to see if my script is
:> working. Since I can't do this with Tkinter scripts, does that mean
:> everytime I want to retest my Tkinter scripts, I have to quit Python
:> and restart Python?

: I usually type ctrl-C twice to test my code.  maybe you
: could train your editor to do something similar for you?

:> > The cause of your problem seems be partly that you actually
:> > expected this to work, putting lots of code in a module
:> > initialisation is not good style.
:> 
:> That bit about "putting lots of code in a module initialisation" just
:> went over my head. What did I do that constitutes "module
:> initialisation"? I've reposted my sample script below so you can point
:> it out to me easily.

: code that's not in a def or a class are executed when you
: run your program as a script, *and* if you import your pro-
: gram as a module.  it's usually better to put the important
: stuff in a function instead, by various reasons (it runs faster,
: it's easier to deal with the day you decide to turn your code
: into a module, etc, etc).

: </F>

I haven't been following this thread too closely, but in general, I
believe that as long as you call the "quit" method of any Tkinter
widget, then "mainloop()" is restartable.  (Having just redesigned the
underlying Tcl/Tk/Tkinter modules, I've seen how this works.)  If you
instead call the "destroy" method, then it is unlikely that you can
restart the event loop (the Tk object has usually been destroyed, but
not always).

  -Arcege




From florent.heyworth at radbit.com  Wed Jun 16 19:55:10 1999
From: florent.heyworth at radbit.com (Florent Heyworth)
Date: Thu, 17 Jun 1999 01:55:10 +0200
Subject: do anonymous pipes normally work on NT?
References: <4D0A23B3F74DD111ACCD00805F31D8100DB90DE0@RED-MSG-50>
Message-ID: <3768395E.A27EAE33@radbit.com>


Bill Tutt wrote:

> > From: radbit at my-deja.com [mailto:radbit at my-deja.com]
> >
> >
> > Hi
> >
> > maybe I'm missing something about what you are looking for: here are 2
> > test scripts that work as anonymous pipes:
> >
>
> That's not the question... Anonymous pipes on NT work wonderfully. (just not
> the way we're expecting from this particular Python context)
> Calling SetStdHandle() is particularly silly as well, you shouldn't need to
> overwrite the current stdout just to get this to work.

Agreed - it was just the only way I could get it to work.

>
>
> > win32file.WriteFile(std_out, 'this method works....')
> >
>
> Calling GetStdHandle shouldn't be necessary either. The CRT already grabbed
> that information for us when Python started up.
>

True. That was just for the sake of being pedantic. Of course:
print "this method works..."
works just fine.

>
> The question is why does sys.stdout.write() not seem to work without -u when
> the above line works wonderfully.... I'm not worried about a working example
> of anonymous pipes in general, thats the easy part.

Interestingly enough I couldn't get at the outpost of testpipe.py even when
when passing "..\python -u testpipe1.py" as the command line argument in
win32process.CreateProcess().

The only workaround I found was to use SetStdHandle 8-(. Once that is done the
child process can use sys.stdout.write("this works....") without problems. For
safety I should have saved the parent's Stdout handle first.

Just a question: I'm pretty sure that this leads nowhere but: do you think what
we're seeing might have something to do with the fact that Python is built in
with multithreaded support?

Anyway - off to bed soon 8-)

Cheers
Florent Heyworth





From Friedrich.Dominicus at inka.de  Wed Jun 23 07:23:40 1999
From: Friedrich.Dominicus at inka.de (Friedrich Dominicus)
Date: Wed, 23 Jun 1999 13:23:40 +0200
Subject: Using C to execute Python
References: <7kqcds$4nu$1@kopp.stud.ntnu.no>
Message-ID: <3770C3BC.287A6FBB@inka.de>

Thomas R?berg wrote:
> 
> I have a C program and I want to call certain Python methods / functions
> passing quite a few parameters.
> How can I do this, in general fashion?
> I would have to include python.h, I reckon - where can I find this, I've
> been searching all over for it (is it just me???)
> I'm no expert, far from, in Python...

Have you read the Python/C Api? you can found it here:
http://www.python.org/doc/current/api/api.html


if you want to extend python using C have a look at:
http://www.python.org/doc/current/ext/ext.html

if you don't know where to start www.python.org is quite first choice.

Regards
Friedrich



From gmcm at hypernet.com  Fri Jun  4 09:06:45 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 4 Jun 1999 08:06:45 -0500
Subject: Tkinter problem in Python 1.5.2 on Solaris 2.5.1
In-Reply-To: <37578FDD.179459EE@easics.be>
Message-ID: <1283620640-45912067@hypernet.com>

Jan Zegers wrote:
> 
> I have the following problem with Tkinter in Python 1.5.2.
/snip/
>   This probably means that Tcl wasn't installed properly.
> 
> Note the whitespace between the two last test lines. The
> installation is as follows:
>   Tcl/Tk 8.1 with --enable-shared so we have libraries 
>     /usr/local/lib/libtcl8.1.so and /usr/local/lib/libtk8.1.so

Tcl / Tk 8.0 is still the official version for Python.

On Jun 1, Dieter Maurer posted patches to use 8.1 ("Tkinter-patch for 
use with Tk8.1...").

- Gordon



From fdrake at cnri.reston.va.us  Tue Jun 15 11:16:33 1999
From: fdrake at cnri.reston.va.us (Fred L. Drake)
Date: Tue, 15 Jun 1999 15:16:33 GMT
Subject: easy filecopy? - JPython?
In-Reply-To: <PBt93.843$B04.2678@newsfeed.slurp.net>
References: <376531D3.7E42AC42@phoenix-edv.netzservice.de>
	<aahzFDCq4D.Do1@netcom.com>
	<PBt93.843$B04.2678@newsfeed.slurp.net>
Message-ID: <14182.28241.949065.590839@weyr.cnri.reston.va.us>

A.K. Masters, Frumious Bandersnatch writes:
 > Copy, copyfile, copymode in shutil apparently haven't been implemented
 > under jpython.  Any copy ideas for jpython (Unix, of course)?

  shutil.copyfile() works just fine (JPython 1.0.3 on java1.2), but
shutil.copymode() relies on os.chmod(), which isn't there.  I'm sure
Barry Warsaw (bwarsaw at python.org) would love to receive a patch for
JPython to support that!  ;-)


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From cgw at fnal.gov  Wed Jun 23 11:27:35 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Wed, 23 Jun 1999 15:27:35 GMT
Subject: How to obtain an instance's name at runtime?
In-Reply-To: <7kqpu8$7u1$1@nnrp1.deja.com>
References: <7kqpu8$7u1$1@nnrp1.deja.com>
Message-ID: <14192.64743.381841.670099@buffalo.fnal.gov>

Dinu C. Gherman writes:

 > >>> class C:
 > ...    pass
 > ...
 > >>>
 > >>> c = C()
 > >>> C.__class__.__name__ # ok
 > 'C'
 > >>> c.__name__           # not ok
 > Traceback (innermost last):
 >   File "<interactive input>", line 0, in ?
 > AttributeError: __name__
 > >>>
 > 
 > Question: How to obtain an instance's name at runtime?
 > That is I'd like to see this happen (one was or the other):
 > 
 > >>> c.__name__
 > 'c'
 > 
 > Fearing-there's-no-such-solution'ly,

Do not fear!  Simply write

 >>> c.__class__.__name__
 'C'
















From jepler at inetnebr.com  Tue Jun 22 14:21:00 1999
From: jepler at inetnebr.com (Jeff Epler)
Date: Tue, 22 Jun 1999 18:21:00 GMT
Subject: newbie idiom question
References: <m3emj4yezq.fsf@monsoon.swcp.com>
Message-ID: <slrn7mvl8l.h16.jepler@craie.inetnebr.com>

To add to the confusion, consider

>>> eggs = [[1], [2], [3]]
>>> for spam in eggs:
...      spam[:]=['cooked'] # or spam[0] = 'cooked'
...
... eggs
[['cooked'], ['cooked'], ['cooked']]

Here, the item "spam" is a mutable object, and I modify it.

In your case, 
>      >>> for spam in eggs:
>      ...	spam = 'cooked'

spam begins the loop as a reference to something from eggs, but all you do
is make spam a reference to something else. (which doesn't change the
references in eggs)

Remember, in Python the "=" operator means "put a reference to the RHS in
the LHS", and has no effect on the object that the LHS used to be a
reference to (if it had a prior value).  Well, except in the case of
setitem/setslice, of course. (which is what I used in the above example)

Is this a wart on python, that "=" means three different things depending
if you have a name, a [item] or a [slice:] on the LHS?

Jeff



From mstenber at cc.Helsinki.FI  Thu Jun  3 08:36:05 1999
From: mstenber at cc.Helsinki.FI (Markus Stenberg)
Date: 03 Jun 1999 15:36:05 +0300
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <199906022028.WAA28874@axil.hvision.nl> <al87lplg9rt.fsf@kruuna.Helsinki.FI> <37564310.3131B5B4@pop.vet.uu.nl>
Message-ID: <al8zp2hea16.fsf@kruuna.Helsinki.FI>

Martijn Faassen <faassen at pop.vet.uu.nl> writes:
> Markus Stenberg wrote:
> Funny, I was prompted to write just about the same thing yesterday. I
> was calling it WallPython, though, and provided an 'unWall' function to
> remove the curly braces. Perhaps you could rename CurlyPython into that.

Nice. I thought about it too (it's actually much easier than doing the
'wall' part, multiline comments and escaped comments can get fairly
painful). Wonder if someone'd seriously write extension like that for
Python, *grin*

> Unlike you I never got my (slightly more ambitious) code debugged
> properly, though. :)

Slightly more ambitious isn't very hard.. Proper Way To Go would be to
build parser for it, but I felt lazy and that was enoough to demonstrate
the point ;) I wouldn't touch something like that with a long stick, but
demonstrations like that are neat way to foil
'whitespace-is-the-only-way' arguments people occasionally leverage against
my language of choice (read: Python).

> Regards,
> 
> Martijn

-Markus

-- 
"...Unix, MS-DOS, and Windows NT (also known as the Good, the Bad, and
the Ugly)."
	- Matt Welsh



From Lars.Marius.Garshol at p98.f112.n480.z2.fidonet.org  Tue Jun 29 22:55:03 1999
From: Lars.Marius.Garshol at p98.f112.n480.z2.fidonet.org (Lars Marius Garshol)
Date: Wed, 30 Jun 1999 03:55:03 +0100
Subject: Please help me to understand classes
Message-ID: <0000159a@bossar.com.pl>

From: Lars Marius Garshol <larsga at ifi.uio.no>


* Johann Spies
| 
| Can somebody explain to me why the following program produces no
| output:

A good trick to learn in Python is that if you use the interpreter you
can fool around and figure things out much more easily and quicker
than if you use something that resembles the edit-compile-run loop of
Pascal.

| [...skipping class defs]
|         
| a = A()
| c = B()
| c.druk
| a.druk

If you do this in the interpreter you get:

>>> a = A()
>>> c = B()
>>> c.druk
<method B.druk of B instance at 80bdd50>
>>> a.druk
<method A.druk of A instance at 809a4d8>

This is a general rule in Python: if you mention a function or method
by name you just get a reference to it. If you add parentheses it is
called. 

>>> c.druk()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: no arguments expected
>>> a.druk()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: no arguments expected

In your example this failed, since you forgot the self argument. If
you redeclare as:

class A:

    def druk(self):
        print 'Jy is in Class A'

To illustrate:

>>> a=A()
>>> a.druk
<method A.druk of A instance at 80b1e60>
>>> a.druk()
Jy is in Class A
>>> method=a.druk
>>> method
<method A.druk of A instance at 80b1e60>
>>> method()
Jy is in Class A


--Lars M.



From nospam at bitbucket.com  Tue Jun 15 14:37:16 1999
From: nospam at bitbucket.com (Phil Mayes)
Date: Tue, 15 Jun 1999 11:37:16 -0700
Subject: Freezing a PythonWin application
Message-ID: <Q3x93.358$dR.20163@newsfeed.avtel.net>

I'm trying to freeze a PythonWin application.

I can successfully freeze a simple script:
  import win32api
  win32api.MessageBox(0, 'Open-source', 'groovy')

My PythonWin application, however, has an MFC .exe stub based on
pythonwin.exe; when I freeze this, it doesn't even reach
CxxxApp::InitApplication().  I presume that I need to meld this
and the code in frozen.c; If anyone has done this before (or failed
to!) I would appreciate some advice.

TIA, Phil
--
Phil Mayes    pmayes AT olivebr DOT com






From krp02078 at my-deja.com  Wed Jun 30 05:10:51 1999
From: krp02078 at my-deja.com (krp02078 at my-deja.com)
Date: Wed, 30 Jun 1999 10:10:51 +0100
Subject: announcement: DIALOG.py
Message-ID: <000015ad@bossar.com.pl>

From: krp02078 at my-deja.com

Hi everybody!

Just finished this module, and thought it could be useful:
DIALOG.py - an interface to the famous dialog program
get it from http://www.brigadoon.de/peter/DIALOG.py

ciao,

	pete

--
-rw-r--r--   1 kruse    users          66 Nov  2 11:50 .signature


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From herzog at online.de  Wed Jun 16 16:17:14 1999
From: herzog at online.de (Bernhard Herzog)
Date: 16 Jun 1999 22:17:14 +0200
Subject: GNU gettext and python
References: <376793C6.9DB513F5@feld.cvut.cz>
Message-ID: <m34sk7sxx1.fsf@greebo.nodomain.de>

Dan Ohnesorg <dan at feld.cvut.cz> writes:

> I will internazionalize one application written python. My favoite
> tool
> is gettext. I have found some support for gettext, most recent in
> py-gnome package. It works well but  only for simple strings like
> "Hello
> world". The the application uses multi-line string (it is listserv),
> like
> '''From: %s
> To: %s
> 
> llll
> kkkk
> jjjj
> '''
> 
> etc. 
> 
> But I am not able to bring this strings into getext. The format of
> .po
> files does not expect string like this and when I use many single
> line
> strings, the application does not find the localized version of
> messages.
> 
> Is here someone, who has solved this? Or has some sugestion?

You can put multi-line messages into po-files like this (taken from
Sketch):

#: ../Sketch/UI/styledlg.py:175
msgid ""
"The name `%(name)s' is already used.\n"
"Please choose another one."
msgstr ""
"Der Name ?%(name)s? ist bereits vergeben.\n"
"Bitten w?hlen Sie einen Anderen"

It certainly works for me under Linux/glibc2.

HTH


-- 
Bernhard Herzog	  | Sketch, a python based drawing program
herzog at online.de  | http://www.online.de/home/sketch/



From ashishg at pobox.com  Mon Jun 21 07:40:40 1999
From: ashishg at pobox.com (Ashish Goyal)
Date: Mon, 21 Jun 1999 05:40:40 -0600
Subject: Using perl in python
References: <7jjoqh$t56$1@nnrp1.deja.com> <6cs73.2432$nn.739920@news.shore.net>
Message-ID: <376D7F38.D1500244@pobox.com>

hi,
    try and use swig (www.swig.org) - that is an interface generator that i
had come across some time back. Since i don't remember things clearly,  it
might not be able to glue python and perl together (i used it to glue C and
python code).

Good luck
ashish

"Michael P. Reilly" wrote:

> Brenda <bbailey08 at my-deja.com> wrote:
> : I have a bunch of perl libraries that I would like to call in my python
> : scripts.  Is this possible?
>
> It is possible;  Tkinter utilizes an embedded Tk interpreter.  You
> would have to write a Perl interpreter.  There may be some people
> around who might help you with that.  I'm in the middle of embedding
> another interpreter myself.
>
> I've found that it is more efficient (and more fun) to convert the
> libraries to Python.  The code is invariably clearer.
>
>   -Arcege




From gauravm at hotmail.com  Fri Jun 25 13:16:37 1999
From: gauravm at hotmail.com (gauravm at hotmail.com)
Date: Fri, 25 Jun 1999 17:16:37 GMT
Subject: Python in India
Message-ID: <7l0dh9$9r1$1@nnrp1.deja.com>

Hi,

Are there any Python user groups in India? Or, would anyone like to form
such a group out here in India? If so, please let me know.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From arcege at shore.net  Fri Jun 25 16:34:37 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Fri, 25 Jun 1999 20:34:37 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773c3a0.102825445@news.oh.verio.com> <xPPc3.535$or1.22866@typhoon-sf.snfc21.pbi.net>
Message-ID: <xJRc3.292$Of3.67094@news.shore.net>

Zigron <zigron at jps.net> wrote:
: Kevin Dahlhausen <morse at harborcom.net> wrote in message
: news:3773c3a0.102825445 at news.oh.verio.com...
:> >I'd like some discussion on whether or not it is a good idea to
:> >abandon TkInter, before it gets too widely used, and instead adopt
:> >wxPython as the primary GUI framework for Python.

:     I'm thinking this is too late, actually. :) From my understanding, its
: the de-facto standard or whatever, and plenty of people use it already.
: Not that that's a good thing.. I find it rather unpleasant that the de-facto
: GUI for Python requires you to install a /different/ scripting language
: just to run,.. and, does seem to be rather slower then wxPython from the
: uses i've had for both... which doesn't surprise me, since Tkinter /does/
: have to run through a second scripting-interpreter and all.

How is this different from requiring you to install two additional
graphical packages (on UNIX), which aren't as mature (there is little
support for the more mature wxMotif according to the literature), as
well-used or as concise.  And Tkinter only requires that you install
libraries (with embedded interpreter engines), not the interpreters
themselves.

But.. the argument over (these two) GUIs is not exactly a valid one.
Tkinter was written using the Tcl interface not the Tk/C API.  If
Tkinter was rewritten to use the C API, it probably wouldn't be slower
than wxPython.  In fact, since wxPython has to go thru extra layers, it
would probably be slower (Tk/C is written on the native API on all the
platforms).  Tcl/Tk is far more portable than wxWindows (as people in
this thread have attested to), which makes it a more widespread choice;
not necessarily a better one on all platforms, but a more mentioned
anyway.  Even if Tkinter was to use the Tk/C directly, the difference
in what is installed is still the same: Tkinter requires one package,
wxWindows requires one or more.

I wouldn't care if wxPython was standard, I would still use Tkinter,
better design philosophy, IMTO (most anything is better than
MFC-oriented systems, even curses).

There are other graphical packages for Python out there too, it would
be appropriate to contrast them as well.

Python is a good language, with good extension capabilities.  I think
there is enough room for more than one GUI package.

  -Arcege

Note for Guido: interfaces for Stdwin and Tk are already shipped with
Python; is it inappropriate to think about shipping wxPython with
Python?




From kvbalaji at wipsys.soft.net  Thu Jun 24 10:22:29 1999
From: kvbalaji at wipsys.soft.net (Balaji Yogesh K V)
Date: Thu, 24 Jun 1999 19:52:29 +0530
Subject: ODBC connectivity to Access!
Message-ID: <37723F25.31033307@wipsys.soft.net>

Hi ,
  Can any body tell me how to connect and retrieve data stored in a
MS Access database using ODBC in python? 
I am new to this language, Hints, pointers welcome!

regards,
bala
-------------- next part --------------
A non-text attachment was scrubbed...
Name: kvbalaji.vcf
Type: text/x-vcard
Size: 407 bytes
Desc: Card for Balaji Yogesh K.V.
URL: <http://mail.python.org/pipermail/python-list/attachments/19990624/605fee77/attachment.vcf>

From mnot at pobox.com  Tue Jun 15 06:52:37 1999
From: mnot at pobox.com (Mark Nottingham)
Date: Tue, 15 Jun 1999 10:52:37 GMT
Subject: <insert favourite flame-bait here> broken
References: <1282792849-95706075@hypernet.com><3764D7C3.718E125E@iname.com><002f01beb64d$58e7b5f0$f29b12c2@pythonware.com><37651951.ED54894E@acm.org><005501beb674$c5575890$f29b12c2@pythonware.com> <376621CF.88C95B5E@iname.com>
Message-ID: <00e701beb71d$2ecd65a0$0301a8c0@cbd.net.au>

> Well, I have no time for reading that. Obviously crossposts
> are useful. I'll tell you how it works:

Congratulations! Welcome to my kill file!






From r14382 at email.sps.mot.com  Fri Jun 25 17:30:02 1999
From: r14382 at email.sps.mot.com (Matthew Cincera)
Date: Fri, 25 Jun 1999 14:30:02 -0700
Subject: VB/win32com data passing question - be gentle!
Message-ID: <3773F4D9.D355E7D2@email.sps.mot.com>

Hello,

I am not really experienced with win32com so hold your laughter to
a minimum if this is a stupid question.  I have a python socket
based system to which I created a com server where I can execute
methods on the main system.  Basically the com server is a 'wrapper'
which does everything in python, but accepts strings as input from
VB, and all replies are converted to strings and parsed in VB.
We are having performance problems since the returned python objects
are large and the repr() function is taking ~9 seconds on large
transactions.  My questions are:

1. Is there an easy way to convert python datatypes into ones
   VB can use.  The datatypes can be nested, for example:

   reply = [ [1, { this is data } ], [2, { this is more data }]]

   The replys can be rather large.  I do not reply anything
   but python datatypes, so I only need to handle numbers, strings,
   dictionaries, lists and the like.  Currently, I convert the
   reply to a string and return that to VB.  Then my VB program
   re-parses the string into the individual elements.  This is
   obviously not efficient but does work.

2. If I can't do 1), does anyone know of a more efficient way
   to do repr()?

Thanks in advance for any insights.

Regards,

Matt




From tim_one at email.msn.com  Sat Jun  5 23:25:43 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 5 Jun 1999 23:25:43 -0400
Subject: Memory leak under Idle?
In-Reply-To: <375986CD.5DD8BE79@appliedbiometrics.com>
Message-ID: <000901beafcc$424ec400$639e2299@tim>

[Christian Tismer]
> ...
> What's going on here?
> Just cyclic references piling up?

Cyclic references are piling up (at least proportional to the number of
windows you've opened), but haven't looked into it deeper than that (an
EditorWindow holds refs to the extensions it installs, and extensions often
keep a reference to the EditorWindow that installed them).  Played briefly
with trying to clear an EditorWindow's __dict__ at window close time, but Tk
started griping about bogus callbacks.  If you grab the latest development
IDLE off of CVS, though, it's bound to be even worse <wink>.

if-pystone-works-ship-it-ly y'rs  - tim





From delgado at Mathematik.Uni-Bielefeld.DE  Wed Jun 16 06:27:04 1999
From: delgado at Mathematik.Uni-Bielefeld.DE (Olaf Delgado)
Date: Wed, 16 Jun 1999 12:27:04 +0200
Subject: Newbie: Truth values (three-valued logic)
In-Reply-To: <000f01beb7b8$2722a540$2ca22299@tim>
References: <000f01beb7b8$2722a540$2ca22299@tim>
Message-ID: <Pine.OSF.4.05.9906161106470.295-100000@pinax.Mathematik.Uni-Bielefeld.DE>

Hi again!

Thanks to everyone for giving hints.

On Wed, 16 Jun 1999, Tim Peters wrote:

> [Olaf Delgado]
> > I am just starting to use python seriously. Currently, I am porting a
> > small class library from C++ to python, which is kind of fun. Everything's
> > much more simple and elegant now.
> 
> Except for C++ <wink>.

Right! C++ looks even more stupid.

> > As far as I know, the logical negation operator can not be overloaded,
> > so I have no chance to change this, right?
> 
> Right.  I'd try overloading "~" instead (the unary prefix "complement"
> operator, special method name __invert__):

I'll consider this. It came to my mind, though, that redefining negation
alone will not help. Next time I'll want to redefine 'and' and 'or'. So,
unless python 2 implements three-valued logic <hint, hint :)>, I'll have
to come up with a routine for myself to handle these truth values
consistently.

> OTOH, overloading sucks when you have to fight the language -- and 90% of
> the time even when you don't <0.9 wink>.

I agree. On the first hand, again, having to use function calls to
evaluate simple expressions increases the risk that I'll just do it the
usual (infix) way when I think it's save.

lisp-would-save-me-from-infix-temptation-ly your's,  Olaf

-- 
   ////         Olaf Delgado Friedrichs, Uni Bielefeld, Mathematik
   Olaf         Postfach 100131                  (room:  V4-109)
   `='          D-33501 Bielefeld, Germany       (phone: +49 521 106-4765)
                           http://www.mathematik.uni-bielefeld.de/~delgado





From gobry at blanc.idiap.ch  Mon Jun 21 10:01:51 1999
From: gobry at blanc.idiap.ch (Frederic Gobry)
Date: 21 Jun 1999 15:01:51 +0100
Subject: BiBTeX module?
References: <joe-160619991358444033@chinacat.salk.edu> <7k950a$hsb@cs.vu.nl> <slrn7miqcg.tpi.bernhard@alpha1.csd.uwm.edu> <slrn7mk9ui.dic.gobry@blanc.idiap.ch> <slrn7moa6s.4l5.bernhard@alpha1.csd.uwm.edu>
Message-ID: <slrn7msdtt.at5.gobry@blanc.idiap.ch>

>Please make snapshot and more infos available.
>I might be able to playtest a bit.
>	Bernhard

Have a look at:

	http://www.idiap.ch/~gobry/pybliographer.html
	
In the snapshot version, I try to interpret authors and dates. Authors are
represented as

	Honorific / First / Last / Lineage

fields. It has been inspired from BibTeX format, but I would like to be able
to handle DocBook bibliographic type and eventually XML if there is some
standard DTD for that...

Fr?d?ric



From jam at newimage.com  Sun Jun 13 20:15:48 1999
From: jam at newimage.com (jam)
Date: Sun, 13 Jun 1999 20:15:48 -0400
Subject: Python Newbie
In-Reply-To: <37644D2B.CC7D0BB3@hotmail.com>; from Denis Lamarche on Sun, Jun 13, 1999 at 09:35:54PM +0000
References: <37644D2B.CC7D0BB3@hotmail.com>
Message-ID: <19990613201548.L723@toast.internal>

On Sun, Jun 13, 1999 at 09:35:54PM +0000, Denis Lamarche wrote:
>
> Where can I find "Python.h"?  I am using dos.  I cant find it anywhere.
> 
> Denis Lamarche
> 

if you downloaded a binary distribution of the interpreter for DOS, you will
probably need to download the 'source' or perhaps the 'devel' package (I am
not familiar with the DOS version of python) to get the header and .lib
file.

hope that helps.

regards,
J
-- 
|| visit gfd <http://quark.newimage.com/> 
|| psa member #293 <http://www.python.org/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>



From vmas at servicom2000.com  Mon Jun 21 12:53:45 1999
From: vmas at servicom2000.com (vmas)
Date: Mon, 21 Jun 1999 18:53:45 +0200
Subject: file system quotas
Message-ID: <Pine.LNX.4.10.9906211845450.684-100000@madre.servicom2000.com>

Hi,
I'm using Python for writing system administration scripts with a Unix
system and I'd like to set quota limits for a given user. Do you know how
to manipulate the file system quotas using Python? Or, is there a
Quota module in Python?
Thanks.

-- 
=========================================================================
Vicente Mas                       ||
Dpto. Tecnico, Servicom 2000      ||    Do or do not ... There is no try.
e-mail: vmas at servicom2000.com     ||    Yoda (The Empire Strikes Back)
Tel: +34963618776                 ||    
Fax: +34963605508                 ||
=========================================================================




From should at beinmessage.com  Wed Jun  9 09:42:04 1999
From: should at beinmessage.com (Arinte)
Date: Wed, 9 Jun 1999 09:42:04 -0400
Subject: Python classes in C
Message-ID: <7jlqtp$o8e$1@rtpnews.raleigh.ibm.com>

How can I get the member variable's values from a python class from inside
my c .dlls???
I have seen PyObject_GetAttr C function, but I am unsure on how to use it.
Could someone post a small example showing both c and python side or point
me to one.  I didn't see on in the docs.





From NoSpam at SeeBelow.net  Thu Jun 10 07:08:06 1999
From: NoSpam at SeeBelow.net (Robert Zimmerle)
Date: Thu, 10 Jun 1999 11:08:06 GMT
Subject: KBHIT in a Windows Environment
Message-ID: <375f9a49.4431852@netnews.worldnet.att.net>

I am using Python 1.52 in a Windows 98 enviroment.  I
recently tried using the kbhit() function in the msvcrt
module.  It appears to work fine in the Python command
line environment but not in the Pythonwin or IDLE
enviroments.  Should kbhit() work in those enviroments?
If not, is their an equivalent function that will?
Thanks in advance.



From MHammond at skippinet.com.au  Thu Jun 10 02:22:58 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Thu, 10 Jun 1999 16:22:58 +1000
Subject: COM dates
References: <816010E2456BD111A48700805FBBE2EEFDF354@ex-quebec-u1.baan.com>
Message-ID: <7jnlh3$qu5$1@m2.c2.telstra-mm.net.au>

Gaetan Corneau wrote in message
<816010E2456BD111A48700805FBBE2EEFDF354 at ex-quebec-u1.baan.com>...
>Thanks Mark!
>
>Another question: why no "strptime" function under WIN32? This function is
>not, as far as I know, difficult to implement.

Because it doesnt exist in the raw MS CRTL, and no one has offered a
replacement.

Mark.






From kas at maps.magnetic-ink.dk  Tue Jun 29 16:36:39 1999
From: kas at maps.magnetic-ink.dk (Klaus Alexander Seistrup)
Date: Tue, 29 Jun 1999 22:36:39 +0200
Subject: [Python|Crypto] on handhelds (was Re: Thanks for Python...)
References: <376ead64@anonymous.newsfeeds.com> <VrDQRddcbbs9-pn2-QdftOXYk6Hdd@Fortress> <37716901.43CB604@lockstar.com> <slrn7n2gkr.i3k.wtanksle@dolphin.openprojects.net> <37727194.D9CD6B7@lockstar.com> <slrn7n5j99.k00.wtanksle@dolphin.openprojects.net> <37790460.47FAD8AF@lockstar.com>
Message-ID: <37792E57.8E5865A8@maps.magnetic-ink.dk>

Mordy Ovits wrote:

> #!/usr/local/bin/python
> from sys import*;from string import*;a=argv;[s,p,q]=filter(lambda x:x[:1]!=
> '-',a);d='-d'in a;e,n=atol(p,16),atol(q,16);l=(len(q)+1)/2;o,inb=l-d,l-1+d
> while s:s=stdin.read(inb);s and map(stdout.write,map(lambda i,b=pow(reduce(
> lambda x,y:(x<<8L)+y,map(ord,s)),e,n):chr(b>>8*i&255),range(o-1,-1,-1)))

'-',a);d='-d'in a;e,n=atol(p,16),atol(q,16);l=(len(q)+1)/2;o,inb=l-d,l-1+d
ValueError: unpack list of wrong size

??


  //Klaus

-- 
???[ Magnetic Ink ]??????????????????????????????????????????????????????????
???[ http://www.magnetic-ink.dk/ ][ maps == mail abuse prevention system ]???



From arcege at shore.net  Thu Jun 10 15:36:50 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Thu, 10 Jun 1999 19:36:50 GMT
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com>
Message-ID: <mtU73.2731$nn.808830@news.shore.net>

In comp.lang.python Fernando Mato Mira <matomira at iname.com> wrote:
: Hello,

:   I'm trying to collect a list of all the _current_ issues making tcl
: a bad language choice. I'd like concrete examples, and not just
: vague assertions that cannot be put porward honestly. I think
: that as is usual in this newsgroups, there's going to be people that
: actually know what they are talking about (and even better, experienced)

: when referring to a language that is not their favorite.
:   If you have references to articles of the sort, that will be useful,
: too.
:   Note that I've not listed comp.lang.tcl for increased productivity,
: and to avoid starting a flame war. Hopefully some misconceptions
: will be filtered down here, and _then_, the summary can be presented
: in that newsgroup for their defense.

I won't go into too many details, it was on an older version of Tcl.
But some years ago, I played around with making C library wrappers
(before SWIG was around) so the developers could write Tcl test scripts
for their libraries.

The problem was that null bytes ('\000') we frequently used in the
code, and Tcl (at the time) could not handle null bytes.  I abandoned
the proof-of-concept project for that reason.  I did have a little
following of developers who were looking forward to that functionality
tho.

I believe that current releases of Tcl can handle null bytes, but I
have not verified that.

In general, the lack of complex data structures is what I find to be
most troublesome in the Tcl language itself.  Beyond lists and
associative arrays (hashes), you have numbers and strings.  The
language makes it convenient to pack these (split, join, etc.), but not
always to utilize them (array names, etc.).  And evaluation is
sometimes difficult to deal with as part of the language.

But for its problem set, I find Tcl to be flexible, quick and easily
extensible (in C or Tcl).

  -Arcege




From phXineXly at haXwaii.edXu  Sat Jun 12 20:37:59 1999
From: phXineXly at haXwaii.edXu (Peter)
Date: Sat, 12 Jun 1999 14:37:59 -1000
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <3762988E.C94AA252@iname.com>
Message-ID: <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu>

In article <3762988E.C94AA252 at iname.com>, Fernando Mato Mira
<matomira at iname.com> wrote:

> Smalltalk taught me OO.
> Occam taught me about concurrency.
> Eiffel taught me SE.
> CLOS taught me what OO really means.

If you are a fan of CLOS, you might want to check out the Dylan
programming language.

It is basically a thoroughly modernized descendant of Lisp/CLOS with an
infix instead of prefix syntax.  In Dylan, almost everything is an object
including the primitive data types (such as integers).  Methods and
classes are also first class objects.

--------------------------------------------------------------------
p h i n e l y @hawaii.edu

Dylan... the high-performance dynamic language
 * open-source Unix version:  http://www.gwydiondylan.org/
 * free Win32 version: http://www.harlequin.com/products/ads/dylan/
---------------------------------------------------------------------



From kjwolf at online.de  Fri Jun 18 04:53:51 1999
From: kjwolf at online.de (Klaus-Juergen Wolf)
Date: Fri, 18 Jun 1999 10:53:51 +0200
Subject: Problem with Curses
References: <ppgmj7.s7r.ln@solling.seismic.de> <jpeck7.1661.ln@solling.seismic.de>
Message-ID: <ve1dk7.1e02.ln@solling.seismic.de>

Klaus-Juergen Wolf <kjwolf at online.de> wrote:
 > -*-
 > import curses

 > curses.initscr()
 > curses.endwin()
 > -*-

 > alone does exactly the same. I've tested it in debugging environment,
 > where it doesn't happen, and I've checked the cursesmodule code, and I
 > still have no idea why this happens. (Inconsistency? But where?)

I now have solved the problem.
It depends on when the executable is linked to libmytinfo, which only
takes place on platforms that don't know the terminfo database, like
BSD and Solaris; there are two competing declarations of tputs().

libmytinfo has to be linked before libtermcap.

(Hours of debugging went for such a laugh... oh no!)

cu
 k.j.



From Sven.Drescher at dlr.de  Thu Jun 24 01:49:46 1999
From: Sven.Drescher at dlr.de (Sven Drescher)
Date: Thu, 24 Jun 1999 07:49:46 +0200
Subject: os.kill()
References: <37708560.BF23BA36@dlr.de> <m37lou3jd8.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <3771C6FA.193BF870@dlr.de>

Michael Hudson wrote:

> Sven Drescher <Sven.Drescher at dlr.de> writes:
>
> > Hi!
> > I have any problems to kill a process started with execv().
> > The process is right on and the pid is ok. But I tested :
> >         os.kill(pid, 0)
> > or    os.kill(pid, 15)
> > or    os.kill(pid, signal.SIGTERM)
> > an nothing happens.
> > I looked for my problem in the newgroups and on the python homepage, but
> > I haven't found anything.
> > I killed my process on console with " kill -9 pid".
> > Is there a stronger version of os.kill() like under UNIX?
>
> os.kill(pid,9)?

Thanks!The parameter '9' does the job! I know, signals are my problem.

Sven




From sams at quinta.com  Thu Jun 10 13:21:13 1999
From: sams at quinta.com (Sam Schulenburg)
Date: Thu, 10 Jun 1999 17:21:13 GMT
Subject: KBHIT in a Windows Environment
References: <375f9a49.4431852@netnews.worldnet.att.net>
Message-ID: <7jos5s$nm5$1@nnrp1.deja.com>

In article <375f9a49.4431852 at netnews.worldnet.att.net>,
  NoSpam at SeeBelow.net (Robert Zimmerle) wrote:
> I am using Python 1.52 in a Windows 98 enviroment.  I
> recently tried using the kbhit() function in the msvcrt
> module.  It appears to work fine in the Python command
> line environment but not in the Pythonwin or IDLE
> enviroments.  Should kbhit() work in those enviroments?
> If not, is their an equivalent function that will?
> Thanks in advance.
>
I have had the same problem i ended up using the following format:
  try:
    main body of code here
  except KeyboardInterrupt:
    if cntrl C is hit this code gets executed

This works fine as long as you are working within a Python module. I
have code that calls a DLL, and if I am in the DLL when the Cntrl C is
hit Python will not execute the excepthon, although I get a message
that it saw the excepthon. ( This is a known problem)

Sam Schulenburg


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From paul at prescod.net  Tue Jun  1 15:17:58 1999
From: paul at prescod.net (Paul Prescod)
Date: Tue, 01 Jun 1999 14:17:58 -0500
Subject: Python 2.0 (perhaps off topic)
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
		<7imr6b$1s0$1@nnrp1.deja.com>
		<87pv3iq6ss.fsf@ev.netlab.co.jp>
		<3752C76D.604CD474@prescod.net>
		<87n1ykr0z9.fsf@ev.netlab.co.jp>
		<375372F6.A874554@prescod.net> <14164.7179.10346.998488@bitdiddle.cnri.reston.va.us>
Message-ID: <375431E6.6BC64E80@prescod.net>

Jeremy Hylton wrote:
> 
> I don't think this is a good example.  Even beginning users are going
> to have to learn a little about resource management about some point.
> People should not rely on files getting deallocated and closed by
> Python.  All the beginners who start with JPython are going to get
> Java's GC.  And they're going to need to close the files by hand.

It looks like this conversation has come all of the way around. We are
talking about whether the JVM should be the default, standard Python
distribution platform. It isn't fair, then, to use the flaws in the JVM
version of Python to argue FOR that change.

"Oh well, we'll lose compatibility with PIL but JPython users have
alreadly lost that so I guess it isn't a big deal."

Personally, I wouldn't mind if Python ran on the JVM or had real GC. I
just think that both sides need to be told. Python is the way it is not by
accident but through design decisions intended to keep it simple to use
and embed.

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for only himself
 http://itrc.uwaterloo.ca/~papresco

"Silence," wrote Melville, "is the only Voice of God." The assertion,
like its subject, cuts both ways, negating and affirming, implying both
absence and presence, offering us a choice; it's a line that the Society
of American Atheists could put on its letterhead and the Society of
Friends could silently endorse while waiting to be moved by the spirit
to speak. - Listening for Silence by Mark Slouka, Apr. 1999, Harper's



From bboard at parc.xerox.com  Mon Jun 21 13:29:52 1999
From: bboard at parc.xerox.com (Bill Janssen - Owner - Intended for posting PARC-wide mail archive system bboard)
Date: Mon, 21 Jun 1999 10:29:52 PDT
Subject: Python fn to invent mail Message-ID? or an UUID?
Message-ID: <99Jun21.102957pdt."3671"@watson.parc.xerox.com>

Is there a Python function which will invent and return a good UUID string?
Suitable for use as an RFC822 Message-ID?

Bill



From fabien.coutant at steria.fr  Mon Jun 28 08:08:07 1999
From: fabien.coutant at steria.fr (Fabien COUTANT)
Date: Mon, 28 Jun 1999 12:08:07 GMT
Subject: strange behaviour at termination time
Message-ID: <377765F0.38230773@steria.fr>

Basically there is a base class name that I found bound to None,
which prevents me to call base methods.

I tested it with 1.5.1 on Linux and 1.5.2 on Windows. Was working
fine with 1.4 in Debian/Linux 1.3.1 till I upgraded to Debian 2.1...

Bug or feature ? Quite annoying if feature...

More that it, I discovered during testing that sometimes it is correct
if I change the base class name !

Here is sample code to show it:

----------------------------------
class A :
    def __del__ (self) :
        print "A.__del__", self

class B (A) :
    def __del__ (self) :
        print "B.__del__", self, A
        # can't do A.__del__ (self) for example

x = A ()
y = B ()
----------------------------------

In this configuration it should show ok.
If you change the name of the first class from A to X everywhere
then X is bound to None in B.__del__ !!!

Please can anyone explain me ?!



From doughellmann at mindspring.com  Wed Jun  2 16:56:45 1999
From: doughellmann at mindspring.com (Doug Hellmann)
Date: Wed, 02 Jun 1999 20:56:45 GMT
Subject: A tkinter kick-off
References: <37544a32.0@news1.mcmail.com> <37546939.E9CE0264@club-internet.fr>
Message-ID: <37559B70.69ACC7DF@mindspring.com>

Olivier Berger wrote:
> 
> My suggestion is then to have a finite state machine (simple to code if
> you use disctionnaries, lists and eval() ...) to interprete these
> inputs, then rendering them as you like in the window...
> 
> I'm myself writing such a tool, so this is only my actual experiments
> results... Maybe you can find allmost complete application frmaework
> elsewhere for such a "simple" tool...
> 

You might be intereseted in text widget I subclassed from
Pmw.ScrolledText.  You can map regular expressions to type styles when
you create the widget.  Then, when you want to run a shell command, you
give it the command (or sequence of commands).  The output is displayed
in the widget using the mappings to render each line appropriately
(right now it works on a line by line basis), based on the regular
expression that matches the string.  When the command (or sequence)
finishes, the widget will even call a callback to let you know.

Check out http://www.mindspring.com/~doughellmann/PmwContribD/ to
download the archive.

The widget in question is in PipeOutputWindow.py.

Doug



From matomira at iname.com  Sat Jun 12 13:26:23 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Sat, 12 Jun 1999 19:26:23 +0200
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com>
Message-ID: <3762983F.D1E09DD5@iname.com>

Fernando Mato Mira wrote:

> [For example, I can think the Eiffel syntax is `philosophically' broken, but
> as `infix' syntaxes go, it is not. And it's a good language (I programmed

Actually, it's a _great_ language to learn SE (for OO pretty much too,  but of
course only in the restricted
single-dispatch way most people know about. I like to call that "Subject Oriented").

Smalltalk taught me OO.
Occam taught me about concurrency.
Eiffel taught me SE.
CLOS taught me what OO really means.

--

"Object Oriented is to not think only about your `subject' and forget about your
`arguments' - the syntactic objects of your sentence"

Fernando D. Mato Mira
Real-Time SW Eng & Networking
Advanced Systems Engineering Division
CSEM
Jaquet-Droz 1                   email: matomira AT acm DOT org
CH-2007 Neuchatel                 tel:       +41 (32) 720-5157
Switzerland                       FAX:       +41 (32) 720-5720

www.csem.ch      www.vrai.com     ligwww.epfl.ch/matomira.html







From eeghead84 at yahoo.com  Mon Jun 14 23:21:32 1999
From: eeghead84 at yahoo.com (eeghead84 at yahoo.com)
Date: Tue, 15 Jun 1999 03:21:32 GMT
Subject: Installing ODBC on PythonWin
Message-ID: <7k4grk$b19$1@nnrp1.deja.com>

Before I go wack and take it out on my computer like this guy
(http://www.web42.com/badday/badday_small.mpg) did, I'd like to ask for
some help.

I'm trying to install an ODBC module into PythonWin, but it still
*won't* work!

It is written that the ODBC module is included in the pythonwin
distribution but >>>import ODBC says otherwise. I've d/e'd
(downloaded/executed) Win32, but it died and said that it was looking
for version 1.4!

I was able to partially install mxODBC by copying all the files into
the right place but I'm having trouble compiling the modules in vc++.

I'm going to pull my hair out!!! Could someone show me an easier way of
installing ODBC? (I don't want to mess with DAO since I'm expecting to
migrate data from a database.)

Hhelp!!

SL


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From arcege at shore.net  Thu Jun 24 17:05:06 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Thu, 24 Jun 1999 21:05:06 GMT
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0rr2.2fi.behrends@allegro.cse.msu.edu>  <8L_b3.1213$4s.136166@typhoon-sf.snfc21.pbi.net> <slrn7n3gi0.44d.behrends@allegro.cse.msu.edu> <377287C5.4E096380@geekspace.com>
Message-ID: <64xc3.130$Of3.30385@news.shore.net>

Joshua Rosen <rozzin at geekspace.com> wrote:
: Reimer Behrends wrote:
:> To remind you, pretty much every single OO language other
:> than Python does not require you to add a self; yet at the same time
:> they have per-method and per-object data. It's a matter of how you
:> declare it.

: Of course, Python doesn't have pre-assignment declarations of variables....

Not quite true.. global variables must declared as such within a local
scope, even before any assignment.

  $ cat foo.py
  def foo():
    global bar
    bar = "Joe's"
  foo()
  print bar
  $ python foo.py
  Joe's
  $

Formal function parameters (which are placed in the local namespace)
must be declared before envocation.

So there are five ways to..

No, two sir!

But it IS true that within functions (and methods and class definitions
and...) that unqualified local variable names do not have to be
declared.

It is also true that a lot of speghetti code can more easily come out
of "OO" languages with intermeshed object namespaces (this said as a
software configuration manager of many years, organizations and
projects writing in different OO languages).  Having to qualify the
names leads to fewer scoping problems down the road (when someone else
is picking up the code and porting it).

  -Arcege




From matomira at iname.com  Thu Jun 10 08:44:35 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Thu, 10 Jun 1999 14:44:35 +0200
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <375FA62E.CCEF5079@cfmu.eurocontrol.be>
Message-ID: <375FB333.5054A4AC@iname.com>

Ian Wild wrote:

> Fernando Mato Mira wrote:
>
> >   I'm trying to collect a list of all the _current_ issues making tcl
> > a bad language choice. I'd like concrete examples,
>
> Without knowing what you're choosing it FOR, how
> can anyone legitimately say it's a bad choice?

Choosing for nothing in particular. It's mainly
a language design question (syntax/semantics).
Although if there're other kinds of problems when
trying to use it in the realm it was designed for, that
would also be interesting.

> >   Note that I've not listed comp.lang.tcl for increased productivity,
> > and to avoid starting a flame war.
>
> Yeah - right.  And no Tcl users ever read here?

Obviously not. The idea was avoid the usual `Why does XYZ suck' flame
bait on comp.lang.XYZ as far as possible, and not irritate and waste XYZ
users' time on issues that some ABC user will point out as false.

There's some tcl lawyer close to me. That's why I did not ask about
the advantages. But nobody knows everything, so in order to make answers
easier to catalogue, I suggest starting another one, "Why is tcl better"
for that.

Thank you for your help,

--
Fernando D. Mato Mira
Real-Time SW Eng & Networking
Advanced Systems Engineering Division
CSEM
Jaquet-Droz 1                   email: matomira AT acm DOT org
CH-2007 Neuchatel                 tel:       +41 (32) 720-5157
Switzerland                       FAX:       +41 (32) 720-5720

www.csem.ch      www.vrai.com     ligwww.epfl.ch/matomira.html






From bizzaro at bc.edu  Sat Jun  5 00:46:33 1999
From: bizzaro at bc.edu (J.W. Bizzaro)
Date: Sat, 05 Jun 1999 04:46:33 +0000
Subject: pyGTK on Windows
References: <3757D453.A302E757@interar.com.ar> <slrn7leo4r.fh.davecook@sputnik.escnd1.sdca.home.com>
Message-ID: <3758ABA9.D77619D0@bc.edu>

"David M. Cook" wrote:
> 
> As someone has mentioned, even if it is made to work under windows, pygtk
> would not have a windows look and feel.

Unless you use a Windows theme with GTK :-)


Jeff
-- 
J.W. Bizzaro                  mailto:bizzaro at bc.edu
Boston College Chemistry      http://www.uml.edu/Dept/Chem/Bizzaro/
--



From gmcm at hypernet.com  Mon Jun 14 21:53:17 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 14 Jun 1999 20:53:17 -0500
Subject: Looking for modules?  Config files?  Ini files?  in python..
In-Reply-To: <rNHvjEdhm5Pp-pn2-WD108jh22jtO@d185d18f8.rochester.rr.com>
Message-ID: <1282710643-100650854@hypernet.com>

Benjamin Schollnick writes:
> 
>  I'm trying to find a simple method to deal with *.INI or other
>  styles
> of
> configuration files....
> 
>  I haven't seen anything, and would hate to re-invent the wheel....
> 
>  Any suggestions?

How 'bout ConfigParser.py? It's in the standard distribution.

- Gordon



From gerrit.holl at pobox.com  Thu Jun 24 10:40:30 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Thu, 24 Jun 1999 14:40:30 GMT
Subject: Question - Returning exit codes
In-Reply-To: <7ktbmu$5os$1@nnrp1.deja.com>; from sjoshi@ingr.com on Thu, Jun 24, 1999 at 01:27:07PM +0000
References: <7ktbmu$5os$1@nnrp1.deja.com>
Message-ID: <19990624164030.A28977@optiplex.palga.uucp>

On Thu, Jun 24, 1999 at 01:27:07PM +0000, sjoshi at ingr.com wrote:
> Hello All
>    I'm calling my python program from a perl script using the system
> command. I was wondering what would be the best way to return exit
> codes from my python script so that they can be intercepted in the perl
> program.
> 

#!/usr/bin/python
"""example on using exit codes"""

import sys

def exit_with_code():
    number = input("What exit code do you want? ")
    sys.exit(number)

if __name__ == '__main__':
    exit_with_code()



or easier:

sys.exit(1)

groeten,
Gerrit.

-- 
The Dutch Linuxgames homepage:	http://linuxgames.nl.linux.org
Personal homepage:		http://www.nl.linux.org/~gerrit/

Discoverb is a python program in Dutch and English which tests the words you
learned by asking it. Homepage: http://www.nl.linux.org/~gerrit/discoverb/



From hniksic at srce.hr  Mon Jun  7 07:00:56 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 07 Jun 1999 13:00:56 +0200
Subject: Deleting characters from a string
References: <87n1ye3g4v.fsf@pc-hrvoje.srce.hr> <y0jyahye89i.fsf@vier.idi.ntnu.no> <87vhd2zaju.fsf@pc-hrvoje.srce.hr> <y0jvhd1dk0h.fsf@vier.idi.ntnu.no>
Message-ID: <87g14446mv.fsf@pc-hrvoje.srce.hr>

mlh at idt.ntnu.no (Magnus L. Hetland) writes:

> > > What!? To me that's more elegant... <wink>
> > 
> > To me it would be if Python actually had chars.  <handshake>
> 
> What do you mean? The example above works...

I mean that to me it's not elegant at all.  Breaking up strings into
more strings is anything but elegant.

Just forget it, I guess.  It was an attempt at humor.



From ivanlan at callware.com  Fri Jun 11 10:29:06 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Fri, 11 Jun 1999 14:29:06 GMT
Subject: Broken 1.5.2 installation? Missing cw3215mt.dll
References: <000901beb3b2$baa6d980$329e2299@tim>
Message-ID: <37611D32.47BED973@callware.com>

Hi All--

Tim Peters wrote:
> 

[snip]

> Aha!  Stop your grinnin' and drop your linen, Mike.  Having more than one
> version of tk*.dll or tcl*.dll anywhere in your entire *city* is like
> putting a male brain-eater in your left ear and a female brain-eater in your
> right ear at the peak of brain-eater mating season, then betting you can
> keep them apart by force of moral suasion.  You should reformat your drives
> and reinstall NT to be *sure* <wink>, but before that try uninstalling every
> Tcl/Tk you have, "dir/s" your drives from the roots to make sure nothing
> matching tk*.dll or tcl*.dll survived, uninstall Python, and reinstall
> Python with the One True Tcl/Tk that comes with it.
> 

[snort]

I've just been installing Python/Tcl/PyWin/Win32All/blah blah on several
Win machines (95/98/NTW/NTS) over the past several days.  It's a long
story (soon to be a major motion picture starring Mel Gibson), but
suffice it to say that in *every* case IDLE refused to run out of the
box.  All that was required, though, was to update PATH to include the
Python directory, the Tcl directory and the Tcl/bin directory.  I've
also installed Python/Tcl on my three Linux machines--RH5.2 and
RH6.0-SMP--but I didn't have to fiddle with anything at all in order to
get IDLE to run just fine.  Of course, Python and Tcl etc. have been in
my path permanently for a long time on Linux;-)

> followups-to-cameron-laird<wink>-ly y'rs  - tim

I'll gladly follow Cameron around if he's responsible for the install;-)

<follow-the-one-true-path-to-enlightenment>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From graham at sloth.math.uga.edu  Wed Jun 16 11:55:53 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 16 Jun 1999 15:55:53 GMT
Subject: fork()
References: <000901beb6ef$f73c0f60$069e2299@tim>
Message-ID: <7k8he9$609$1@cronkite.cc.uga.edu>

[Hisao Suzuki]
>     Note that it is
>     always important to destroy objects in an order that ensures
>     that the destructor for one object doesn't refer to an
>     object that has been previously destroyed.  Such ordering
>     isn't easily achieved by a garbage collector without help
>     from the programmer."

I think people are confusing implementation with semantics here.
The Java implementation of GC might have the problem of destroying objects
too early, and hence having to deal with objects that refer to
previously destroyed objects. But this is an implementation issue
and not a semantic issue. A two pass collector avoids this problem
completely. In pass 1 all finalisers are called, but no object
deleteion is done. In pass 2 object deletion occurs. The point
is to define what you want for the semantics of finalisation and
then implement it. The Java approach you refer to above is to define
and implementation and make that the semantics. Indeed this is
largely the Python approach (the language semantics is really defined
by the C implementation). For my money semantics by implementation
leads to serious problems.

graham

-- 
           As you grow up and leave the playground
       where you kissed your prince and found your frog
          Remember the jester that showed you tears
                   the script for tears



From tools at tools.com  Wed Jun 30 12:45:19 1999
From: tools at tools.com (TOOLS Conferences)
Date: Wed, 30 Jun 1999 09:45:19 -0700
Subject: TOOLS USA '99 - Conference Program & Registration Information
Message-ID: <7ldhhq$nnp$1@news.rain.org>

**************************************************************

   C O N F E R E N C E   A N N O U N C E M E N T

                    TOOLS USA '99
   Technology of Object-Oriented Languages and Systems
          30 Int'l Conference & Exhibition
    "DELIVERING QUALITY SOFTWARE - THE WAY AHEAD"

   Santa Barbara, Calif., August 1-5, 1999
          Fess Parker's Double Tree Resort

              HELD IN COOPERATION WITH
  Regional Small Business Innovation Research (SBIR)
              Workshop - August 4, 1999
                        &
National Defense Planning And Analysis Society (DPAAS)
          Annual Symposium - August 5, 1999

               http://www.tools.com/usa

**************************************************************

                CONFERENCE PROGRAM
               ~~~~~~~~~~~~~~~~~~~~

* Keynote presentations by world-class visionaries:
     David Parnas, McMaster University, Canada
     Bertrand Meyer, ISE, USA
     Don Box, DevelopMentor, USA
     Tucker Taft, AverStar, USA

* Hands-on tutorials by recognized experts:
     30 tutorials, in six tracks: Quality, Languages,
     Technology, Methodology, Experience, and Management

TUTORIAL PROGRAM
Systematic Techniques for Inspecting Critical Software
   David L. Parnas, McMaster University, USA
The Development and Verification of Safety Critical Software
   George Romanski, Aonix, USA
Managing By Contract
   Bertrand Meyer, Interactive Software Engineering, USA
Verification and Validation of Object-Oriented Software Systems
   Kasi Periyasamy, University of Manitoba, Canada
Mastering JFC 'Look and Feel' and UI Delegation
   Mitchell Goldstein, Modis Inc., USA
Quality Component Development: Making the Most of JavaBeans and
Enterprise JavaBeans Features
   Gilda Pour, San Jose State University, USA
Ada and the Objects
   Benjamin Brosgol, Aonix, USA
Programming with C++ Exceptions
   Angelika Langer, Germany
Behind the Beans
   Michael Stal, Siemens AG, Germany
Java, Eiffel and C++: The Language Comparison
   Ian Joyner, Microsoft Research Institute, Australia
Mastering the Windows Eiffel Library
   Glenn Maughan, Westernport Group Pty Ltd, Australia & Raphael Simon, ISE,
USA
Distributed Objects from a Patterns Perspective
   Michael Stal, Siemens AG, Germany
Building Scalable ODBMS Applications
   Matt BenDaniel, Object Design, USA
Real-Time Object-Oriented Distributed Systems: RT CORBA and RMA
   Antonios Broumas, Tri-Pacific Software, USA
System Design: Architectures and Archetypes
   Stephen J. Mellor, Project Technology, USA
A Short Use Case Writing Workshop
   Alistair Cockburn, Humans & Technology, USA
Modularization Revisited: Aspects in the Design and Evolution of Software
Systems
   Cristina Videira Lopes, Xerox Research Center, USA
Advanced Visual Modelling: Beyond UML
   Stuart Kent, University of Kent, UK; Joseph Gil, Technion, Israel;
   and John Howse, University of Brighton, UK
Advanced OO Modelling: Metamodels and Notations for the Next Millenium
   Brian Henderson-Sellers, University of Technology, Sydney, Australia
Analysis by Contract: An Introduction to Precise OO Modeling with UML and
OCL
   Richard Mitchell, University of Brighton, UK
Design of Commercial Object-Oriented Applications - The Patterns
   Frieder Monninger, Object Tools GmbH, Germany
Octopus/UML: An Object-Oriented Method Used in Industry
   Rallis Farfarakis, Nokia Research Center, Finland
Analysis, Design and Implementation of Distributed Java Business
Frameworks using Domain Patterns
   Ali Arsanjani, IBM Enterprise Java Services, USA
Requirement Patterns
   Christopher Creel, Technical Resource Connection, TRC
Introduction to the OPEN Method with UML
   Brian Henderson-Sellers, University of Technology, Sydney, Australia
Effective Strategies and Techniques for Rapid Object-Oriented Application
Development
   Michael P. Anton, Financial Technology Solutions, USA
Software Development "in the Zone"
   Todd Lauinger, Reliastar Inc., USA
Creating Social Agents with Reusable Components: A Practitioner's Guide
   Dana Moore, AT&T Labs., USA
Features of UML Tools (live demo)
   Tony Wasserman, Software Methods and Tools, USA

* Interactive workshops on the most innovative issues

* Lively panel discussions:
     including the famous "Great Language Debate"

* 37 Technical paper presentations:
     selected by the international program committee

* Eiffel Summit '99
     annual Eiffel users forum

* Symposium on Teaching Object Technology (STOT)
     "Using Information Technology to Teach Object Technology"

* Special pre-conference session:
     "Object Technology: An Executive Overview"

* Regional Small Business Innovation Research (SBIR)
  Workshop
      Sponsored by Frontier Technology, Inc.

* National Defense Planning And Analysis Society (DPAAS)
  Annual Symposium
      Sponsored by Frontier Technology, Inc.

* Exhibition of the latest products, services, and research
      only a few booths left...

* Complementary events:
     user meetings, working group discussions, BOFs, etc.

* An attractive social program in the superb Santa Barbara area.


********************************************************************

FULL CONFERENCE PROGRAM & REGISTRATION INFORMATION
           at http://www.tools.com/usa
         (also see registration from below)

If you wish to obtain your own copy of the conference program,
please send your mailing information to tools at tools.com

TOOLS?the major international events entirely devoted
to the applications of object-oriented technology

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
FOR MORE INFORMATION ON TOOLS USA '99 OR ANY OTHER
EVENTS IN THE TOOLS CONFERENCE SERIES,
VISIT OUR WEB SITE AT http://www.tools.com
OR CONTACT US AT tools at tools.com
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

         T O O L S   U S A   ' 9 9
     R E G I S T R A T I O N   F O R M
       email form to tools at tools.com
       print and fax to 805-685-6869
        phone us at 805-685-1006
           or register online at
         http://www.tools.com/usa

1. PERSONAL INFORMATION
First Name:
Last Name:
Job Title:
Company:
Address:
City:
State:
Zip:
Country:
Phone:
Fax:
Email:

2. REGISTRATION PACKAGES
|_| 1 day ($495): August __
|_| 2 consecutive days ($795): August __-__
|_| 3 consecutive days ($1045): August __-__
|_| 4 consecutive days ($1295): August 2-5

|_| SBIR ($195): Wednesday, August 4
|_| DPAAS ($195): Thursday, August 5

|_| Eiffel Summit '99 ($495): Tuesday, August 3

|_| Introduction to Object Technology : Sunday, August 1
($395 with conference registration*; $495 on its own)

|_| Symposium on Teaching Object Technology (STOT): Sunday, August 1
($300 with conference registration*; $395 on its own)

|_| Free Exhibit-only Pass (free)

*if registered for at least two TOOLS conference days

NOTE:
- Rates shown above are for regular registrations
- Academic and Group Discounts available
- Early-Bird prices shown above are valid until July 2
See details online at http://www.tools.com/usa/reg.html


I PLAN TO ATTEND:
(list TOOLS sessions by code; e.g. LANG1, METH2)


3. MEAL TICKETS AND ACTIVITIES
Full registrations include lunches, BBQ dinner, and conference dinner.
Academic Participants must purchase meal tickets separately

|_| __ Conference lunches ($20 each)
|_| __ Beach Barbecue ($30): Monday, August 2
|_| __ Conference Dinner ($60): Wednesday, August 4
       |_| I am a vegetarian.
|_| __ Golf Tournament ($95)


4. PAYMENT

Registration Total $________

I wish to pay by:
|_| Check or International Money Order #___________ enclosed
|_| Inter-bank wire transfer ($15 additional charge)
|_| Credit Card: __ Visa  __ Mastercard  __ American Express
Credit card #:
Expiration date:
Cardholder Name:


5. HOW DID YOU FIND OUT ABOUT TOOLS?








From travel at mailcity.com  Sat Jun  5 12:52:26 1999
From: travel at mailcity.com (Travel Now!)
Date: Sat, 5 Jun 1999 16:52:26 GMT
Subject: Complimentary hotel stay!
Message-ID: <419.436316.53684965travel@mailcity.com>

This email list was received from Maxnet promo.com

Free Vacation!
***********************************************************
 Just for stopping by our site and reviewing some
material  We will send you and your companion for
a 2 night stay with luxury hotel accomodations-
absolutely no strings attached!  ($175 value)  You
will need to fill out the form completely with honest, 
accurate information. It is very short, only ten questions.  
You will then receive your travel voucher via email 
after the form is submitted and information verified.
They will ask for a $14.95 shipping and handling
fee for the vacation vouchers.
http://www.inficad.com/~dreams2000

The business material will let you look at how to 
take advantage of a Mult-Billion dollar travel industry-
working from your own home. Featuring a huge 
compensation plan, it is not uncommon to make 
six figures!  This is not MLM or a travel industry 
franchise.You will want to review this information.
Visit & take advantage of this free mini vacation
for a limited time only!
http://www.inficad.com/~dreams2000
*************************************************************




From faria at ppgia.pucpr.br  Tue Jun 22 15:16:10 1999
From: faria at ppgia.pucpr.br (faria)
Date: Tue, 22 Jun 1999 16:16:10 -0300
Subject: Tkinter canvas configuration
Message-ID: <000b01bebce3$b1f87200$656211c8@miro.lami.pucpr.br>

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> from Tkinter import *
>>> root = Tk()
>>> canvas = Canvas(root)
>>> canvas.pack()
>>>

    Now I have a canvas, but I still don't know how to configure it (e.g.
make it bigger, change the color...)

    Tanx for the attention.


Henrique Faria
Desenvolvedor Multim?dia
LAMI - Laborat?rio de M?dias Interativas Puc-PR
faria at ppgia.pucpr.br

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Henrique Faria.vcf
Type: text/x-vcard
Size: 436 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/19990622/d642d41c/attachment.vcf>

From MHammond at skippinet.com.au  Fri Jun 18 20:28:27 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Sat, 19 Jun 1999 10:28:27 +1000
Subject: Examples of usage for win32pdh.ExpandCounterPath()?
References: <376ac08a.13773775@news.worldtalk.com>
Message-ID: <7keo4r$2q9$1@m2.c2.telstra-mm.net.au>

calouro wrote in message <376ac08a.13773775 at news.worldtalk.com>...

>The win32 extensions work well once I figure out how to call them
>correctly -- which is much easier if I have some examples to study and
>experiment with.

The help files and looking at win32pdhutil contain the hints you need.  The
biggest problem with win32pdh is knowing what counters to use, and their
exact format.

OK - lets browse for the counters.  In the dialog that is presented, we
select a counter, and press "Add":

>>> import win32pdhutil
>>> win32pdhutil.browse()
Browsing for counters...
Value of '\\BOBCAT\Process(_Total)\% Processor Time' is 95.5669019211
Added '% Processor Time' on object 'Process

OK - now we know the format of at least one counter.  The documentation to
ExpandCounterPath implies you can add wildcards.  So, lets try it:

>>> import win32pdh
>>> from pprint import pprint
>>> pprint(win32pdh.ExpandCounterPath(r"\\bobcat\Process(_Total)\*"))
['\\\\bobcat\\Process/_Total)\\% Processor Time',
 '\\\\bobcat\\Process/_Total)\\% User Time',
 '\\\\bobcat\\Process/_Total)\\% Privileged Time',
 '\\\\bobcat\\Process/_Total)\\Virtual Bytes Peak',
 '\\\\bobcat\\Process/_Total)\\Virtual Bytes',
etc!
]

Mark.





From news at dorb.com  Sun Jun 27 01:29:01 1999
From: news at dorb.com (Darrell)
Date: Sun, 27 Jun 1999 01:29:01 -0400
Subject: Python for large projects?
References: <3774b5d1.690922@news.force9.net>  <T6bd3.2883$U6.12014@newsr2.twcny.rr.com> <slrn7naup9.jk.thantos@brimstone.mecha>
Message-ID: <RGid3.2957$U6.12374@newsr2.twcny.rr.com>

>
>   buf = open(file).read()
>   doTheBigJob(buf)
>   del buf
>
> No need to introduce new passing schema or wrap the buffer object in
> another object at all; this will clean it up for you.
>
My point was this isn't cleaned up until you return from doTheBigJob.

>
> If the only thing you're doing with the content of file is passing it
> to doTheBigJob, then its even /better/ to simply call it as:
>
>   doTheBigJob(open(file).read())
>
No doubt this is a clean way to open the file. But it doesn't work if you
wanted to act on the buffer before passing it on. I'm not attacking python
or reference counting here and agree that for most programs these aren't
even concerns. But when your shinny new application comes together and eats
200meg, you start to wonder. Tracking though everyone's code trying to
understand who has all this memory is real fun. My point was to think about
this stuff from the start, keep an eye on memory usage. Reference counting
can give a false sense of security.


> >6. No const or private.
> >    You have to trust everyone.
>
> This always struck me as a really /odd/ concern.  If you don't trust
> the people on your programming team, why are they on your team?  If
> you're afraid the hooks into your inner-code are visible, then rethink
> why you're concerned.  Mathods/slots on Python objects can be prefixed
> with _ and __ to make them harder to casually examine, but I've always
> equated 'reference hiding' with 'security through obscurity,' it
> doesn't protect you and it lends a false sense of hope.
>
Good or bad, this is one C++ guys aren't comfortable with. The project I'm
on now will allow people with minimal programming experience to write a
little python to control the larger application. One of my primary concerns
is performance and an optimization relies on them not changing a large
buffer, only look at it. So the people I don't trust aren't on my team. What
was it Murphy said ? "If it can go wrong"
I'm tempted to do this object in C++.
Agreed using __ to hide a name doesn't seem worth while.

I appreciate your reply because I'm still seeking the truth of these
matters.
--Darrell






From kiki at pixar.com  Wed Jun  2 20:06:44 1999
From: kiki at pixar.com (kiki)
Date: Wed, 02 Jun 1999 17:06:44 -0700
Subject: More Tkinter menu problems
Message-ID: <3755C714.526B7994@pixar.com>

Here's a new problem which is much like my old problem with a new
twist...

I'm making a color menu class now.  In the menu is mainly "Add Color"
and the current list of colors.  When you choose "Add Color", the user
makes a color, then names it.  From here, I want to append the color to
the menu, in much the same way I appended my mondo_menu call -- again, I
have a group of menu items which are all the same action on slightly
different data.

I tried the suggested lambda idea:

  self.menu.add_command(label = new_name,
   command = lambda temp_color = new_color:
   self.select_color(temp_color))

but I get a Name Error on "self".  So now "self" is out of scope when
the menu call happens.

Is there *really* no better way to have a single function handle a group
of similar menu items?  Not only would that be useful, but in many cases
like these it's *necessary*.  I still think it would sure be nice to get
an event record back from *all* events, including menu events... :(

Thanx!

Kiki

PS Greg McFarlane: you are absolutely right in scolding me for my
over-use of eval. I've never used it nor had a need for such a thing
before, but now I'm getting interpreter-giddy. :) Learning those "safer"
and more lovely methods is what I'm-a trying to do. :) Speaking of, is
there a way to take the *string* "Kikis_modules.Fractal" and turn it
into "Fractal" gracefully without doing eval(class_name).__name__?  I
tried "import Perl" but that didn't work. ;) hee-hee




From lull at acm.org  Tue Jun 22 00:42:31 1999
From: lull at acm.org (John Lull)
Date: Tue, 22 Jun 1999 04:42:31 GMT
Subject: Newbie Q: How to call unmodified VB DLL from Python
References: <3769bcd1.54492213@news.earthlink.net> <7kjvsl$rtu$1@nnrp1.deja.com>
Message-ID: <377112bf.57865747@news.earthlink.net>

In the waning years of the 20th century, radbit at my-deja.com wrote
(with possible deletions):

> since VB has only been able to do in-process OLE (Com) Dlls since VB4.0
> I take it that this DLL exposes its methods via COM. In which case it
> should be a snip to talk to it via Marc Hammond's Pythonwin COM
> support. Find out what it's programmatic id is (progid) from
> Pythonwin's COM browser or MS VC++'s COM Viewer. Once you know what
> progid it has you can do:
> 
> >>> from win32com.client import Dispatch
> >>> vbObj=Dispatch('progid')

This should get me started.

Thanks.

Regards,
John




From graham at sloth.math.uga.edu  Wed Jun  2 16:52:47 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 2 Jun 1999 20:52:47 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>  <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>  <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net>  <87hfosquc0.fsf@ev.netlab.co.jp> <3753F121.E4A9BE8B@prescod.net> <87yai3ntdt.fsf@ev.netlab.co.jp> <7j3iif$134e$1@nntp6.u.washington.edu>
Message-ID: <7j45iv$908$4@cronkite.cc.uga.edu>

Donn Cave (donn at u.washington.edu) wrote:
: You're in good company here, I think the leading lights of the Python
: world all prefer explicit termination in these cases.  Ref counting sure
: offers a great convenience here, though.  Files close and flush their
: buffers, dialogues disappear, all in accordance with their programmatic
: lifetime.  I personally will miss this finalization more than any resource
: management if reference counting is taken away.

You seem to believe that garbage collection implies no reference counts.
Why do you believe that? (a lot of GC schemes use reference counts).
You also seem to believe that with GC finalisation will disappear. Why
do you believe that (a lot of GC schemes have finalisation). Finally I
believe you are mis-using finalisation if you use it to close up files,
sockets, etc. That's what close calls are for.

graham
-- 
             Like a bird on a wire
        Like a drunk in a midnight choir
            I have tried in my way
                  To be free



From hgkang at etri.re.kr  Wed Jun 30 20:43:28 1999
From: hgkang at etri.re.kr (hgkang)
Date: Thu, 1 Jul 1999 09:43:28 +0900
Subject: [Q] Static type checker for Python
Message-ID: <7lee1g$sod$1@news.etri.re.kr>

hello.
Probabilly this is dummy question 'cause I'm a new user of Python.

Can I make static type checker for Python?
or Is there any previous work about this?

if impossible, why?
what language feature should be changed to be possible?

thanks in advance.





From morse at harborcom.net  Tue Jun 29 08:00:37 1999
From: morse at harborcom.net (Kevin Dahlhausen)
Date: Tue, 29 Jun 1999 12:00:37 GMT
Subject: Python GUIs: Summary and Conclusion
References: <99FA4A1A1280BEBD.0A1ABA62CC989479.4C2C6022F695BEAC@lp.airnews.net>
Message-ID: <3778ae18.87993968@news.oh.verio.com>

>Two other GUIs, The FastLight Toolkit and PyKDE/Qt, were mentioned,
>but they both fall outside the domain of this discussion because
>they're not portable enough (one of the primary tasks of a standard
>GUI must be to be available on any, or most, Python hosts). 

Just for accuracy - I'm advocating it as another option for  users, but not as
an  official replacement in the distributions-  the FastLight Toolkit builds on
Unix, OS2, and MS-Windows.







From tim_one at email.msn.com  Sat Jun 12 15:19:04 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 12 Jun 1999 15:19:04 -0400
Subject: module import, but different
In-Reply-To: <3762087C.98EA86DA@bioreason.com>
Message-ID: <000301beb508$6fcca140$ce9e2299@tim>

[Andrew Dalke]
>   I would like to add a test suite to our software so at the top
> level of the distribution I can do a "make check" and run the tests.
> (We are planning to do something based to Python's regrtest code.)
> I'm running into a problem because I don't know how to deal with
> packages; and we do almost everything as packages.  It comes down
> to needing a way to import a package when you can't include the
> package's parent directory in the PYTHONPATH.
>
>   I'll describe the problem with some detail first.
>
>   Here's an example package layout:
>
> MCS/
>   Makefile
>   __init__.py
>   Graph.py
>   ...
>   doc/
>
> to which I plan to add
>   test/
>     br_regrtest.py     ("br_" for "Bioreason's version of")
>     test_graph.py
>     ...
>
> And in the Makefile add something like
>
> check:
>   cd test && $(PYTHON) br_regrtest.py
>
> Inside the test directory will be programs like "test_graph.py".
> They will have code like "import MCS.Graph".  This will fail
> because MCS isn't a module or package on the PYTHONPATH.  (Setting
> PYTHONPATH to ".." will fail for modules which have non-trivial
> __init__.py files; and we have a couple of that form.)
>
> [and much more]

I don't understand this stuff, so let me ask a dumb question:  why can't the
"check" target simply be

check:
    cd .. && $(PYTHON) MCS/test/br_regrtest.py

The directory from which Python is invoked is always the first thing on
sys.path, so MCS will be found.  IOW, make test into a normal MCS package
too.

> One solution would be to include "../.." but that wouldn't be
> appropriate because that may include files which get imported
> by accident.  For example, I keep all of my development code
> under src/, so src/MCS, src/daylight, etc.  But when I'm doing
> unit testing I want to import the globally installed "daylight"
> and not the ../../daylight imported by adding ../.. to the path.

Ah.  Trying to get something done with directory trees *is* a lot like
trying to get something done with regexps <0.7 wink>.  This is much harder
than the problem you claimed to have at the start:  the problem isn't that
you don't have the package's parent on PYTHONPATH, the problem is that your
intended package structure differs from your actual directory structure.
That means your only hope is to write a custom importer that knows about
your secret desires -- all Python can grok on its own is the directory
structure that actually exists.

> ...
> Besides, we distribute the packages with version numbers, as in
> MCS-1.0.  Including ../.. means we would have to convert the name
> somehow.

OK, your intended package names have *nothing* to do with your actual
directory names!

> The Python distribution solves the problem, I think, because in
> getpath.c it adds `Lib' to the PYTHONPATH if python was was run
> in the build tree.  Seems sneaky to me!

Yes, but not half as sneaky as what you're trying to do <wink>.

> Another possibility is to make a symbolic link in the test/
> directory, as in
>   ln -s ../ MCS
>
> This will work under unix, but isn't very portable, and I expect
> we'll have to support Windows someday.  (Or do the other OSes have
> something equivalent to this?)

Windows doesn't have links, symbolic or hard.  This was a very successful
strategy to increase sales of hard drives and OS upgrades <wink>.

> A third solution ...

Ack, enough already.  If your packages don't match your directories, the
builtin importer simply doesn't do what you want.  Look in the Library
manual's Built-in Functions section for __import__:  that's the builtin
function an "import" stmt invokes.  You can replace it with whatever you
like.  The std imp module packages many functions that will be helpful when
doing this.  As I said at the start, I don't understand this stuff (never
had reason to use it!), but Greg Stein is widely reputed to have developed a
much more flexible import mechanism; maybe DejaNews will turn up something
useful.

the-distutils-sig-may-also-be-your-friend-ly y'rs  - tim





From John.Michelsen at p98.f112.n480.z2.fidonet.org  Tue Jun 29 17:23:40 1999
From: John.Michelsen at p98.f112.n480.z2.fidonet.org (John Michelsen)
Date: Tue, 29 Jun 1999 22:23:40 +0100
Subject: Python GUIs: Summary and Conclusion
Message-ID: <00001598@bossar.com.pl>

From: "John Michelsen" <john.michelsen at gte.net>

Philippe, they look very nice. How did you use them with Python
and Tkinter? Any Python code that you would like to share
demonstrating BWidgets?  I would be especially interested
to see use of the tree widget.

Thanks,

John


Philippe Chassany wrote in message <377805C8.C6BBD4C1 at unifix-online.com>...
>> ....
>> More widgets should be added to Tk(Inter).
>> ....
>
>Our company has made a 100% Tcl/Tk Toolkit named "BWidget".
>BWidgets are "well known" in the Tcl community because
>they are free, flexible, powerful and completly interpreted (no C).
>More than 25 widgets are available.
>
>We have tested BWidget with Tkinter the result is quite fine :
>Quick and good looking. It's a good alternative to GTk or Qt
>compiled programming to have professional application GUIs.
>
>Try it to make your own idea (free download, screenshots,
>full HTML manuals)
>
>http://www.unifix-online.com/BWidget
>
>--
>Philippe Chassany
>UNIFIX - http://www.unifix-online.com
>--
>
>




From thstr at serop.abb.se  Wed Jun  9 09:48:48 1999
From: thstr at serop.abb.se (Thomas S. Strinnhed)
Date: Wed, 09 Jun 1999 15:48:48 +0200
Subject: Python/C API and setattr
References: <1283183791-72189810@hypernet.com>
Message-ID: <375E70C0.7D19FB6D@serop.abb.se>

Hi,

Thanks, but unfortunately it doesn't seem to be the problem.
Please take a look at the response posted to myself.

Thank you

-- Thomas S. Strinnhed, thstr at serop.abb.se

Gordon McMillan wrote:
> 
> Thomas S. Strinnhed writes:
> >
> > I'm building an extention type for Python in C using NT, MSVC++.
> 
> > I've looked pretty much at Mark Lutz's example StackType  from
> > Programming Python and I think I'm doing the same, but I've added
> > setattrfunc to be able to directly manipulate value as c.value = 23
> > and when I use it something goes wrong. (inc() and dec() work fine)
> >
> > Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on
> > win32 Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> > >>> from CounterType import *
> > >>> c = Counter()
> > >>> c.value = 23
> > Traceback (innermost last):
> >   File "<stdin>", line 1, in ?
> > SystemError: error return without exception set
> 
> You returned NULL (or 0, or the compiler generated a return of NULL
> because you didn't state anything) from your C code. That signifies
> an error to Python, which expects you to have set an error string.
> 
> Return (an increfed) None and Python will be happy.
> 
> > >>> c.value
> > 23
> > >>>
> > The value is set properly but I can't figure out the system error,
> > it seems to be something wrong with my setattrfunc.
> >
> > static void counter_setattr
> >  (counterobject *self, char *name, PyObject *v)
> > {
> >  if(strcmp(name, "value") == 0)
> >  {
> >   self->value = (int)PyInt_AsLong(v);
> >  }
> > }
> >
> > But what would that be??
> > I have a hard time finding examples of how to do setattr so
> > any pointers would be great. Also if someone understands this error,
> > please post an answer.
> >
> > My counterobject is this:
> > typedef struct
> > {
> >  PyObject_HEAD
> >  int value;
> > } counterobject;
> >
> > Thanks
> >  -- Thomas S. Strinnhed, thstr at serop.abb.se
> >
> > --
> > http://www.python.org/mailman/listinfo/python-list
> 
> - Gordon



From sjoshi at ingr.com  Tue Jun 22 13:53:19 1999
From: sjoshi at ingr.com (sjoshi at ingr.com)
Date: Tue, 22 Jun 1999 17:53:19 GMT
Subject: Win32 Question
Message-ID: <7koiib$fdu$1@nnrp1.deja.com>

 I want to find out if a certaing application is releasing memor or not
when it terminates and was wondering how to go about doing that using
win32api.
  I would appreciate if someone could point me in the right direction

thanks
Sunit


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From greg.ewing at compaq.com  Tue Jun 29 00:04:45 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Tue, 29 Jun 1999 16:04:45 +1200
Subject: Python GUIs: Summary and Conclusion
References: <99FA4A1A1280BEBD.0A1ABA62CC989479.4C2C6022F695BEAC@lp.airnews.net> <MTOd3.139$6M6.44310@news.shore.net>
Message-ID: <377845DD.23AD193A@compaq.com>

"Michael P. Reilly" wrote:
> 
> I could scratch my _tkinter rewrite and work on a native solution.
> (I could still publish my pytcl module.)  Any thoughts?

I'd certainly be in favour of a native
Tk interface, especially if it became the 
"standard" one included in the distribution.
Even if it wasn't any faster, I'd feel
better while using it!

Wasn't there a Tcl-less Tk interface once
before, that wasn't kept up to date?
Called something beginning with R that
I can't remember.

Maybe you could use that as a starting
point if the Tk C interface hasn't
changed too radically.

Greg



From Skip.Montanaro at p98.f112.n480.z2.fidonet.org  Wed Jun 30 23:02:22 1999
From: Skip.Montanaro at p98.f112.n480.z2.fidonet.org (Skip Montanaro)
Date: Thu, 01 Jul 1999 04:02:22 +0100
Subject: [Q] Static type checker for Python
Message-ID: <000015f3@bossar.com.pl>

From: Skip Montanaro <skip at mojam.com>

> Can I make static type checker for Python?
> or Is there any previous work about this?

In theory, you could do some type inference and generate some warning
messages, but since you don't declare variables to be of a particular
type in Python, it's unlikely it would be very useful.  The same
variable can point to many different types at runtime.

For example, the following code may be just what the author intended:

    tmp = 1.7 * math.sin(a)
    tmp = `tmp`
    print tmp
    tmp = len(tmp)

After the first statement you might *presume* that tmp holds a floating
point number, but there's no guarantee, since you don't know that
math.sin returns a float.  You can infer that tmp holds a string after
the second statement because its builtin to the language that ``
converts what it encloses to a string (though the timbot will probably
chime in with a case where I'm wrong about that <wink>).  After the last
statement you no longer know what tmp holds because you can't be sure
that len() returns an int.

The topic of static type declarations has consumed more than its fair
share of heat and light on the newsgroup and at Python conferences.  For
one recent example, check out:

   
http://www.foretec.com/python/workshops/1998-11/proceedings/papers/masse/masse.h
tml
    http://www.foretec.com/python/workshops/1998-11/dd-rmasse-sum.html
    http://www.python.org/~rmasse/presentations/python-types/

You can scan the comp.lang.python archives for "static types" using the
form at

    http://www.python.org/search/

to browse around the previous discussions about this topic on the
newsgroup.

Also, it's worth taking a look at Guido's most recent "Python Futures"
talk to get an idea where he things Python might be headed:

   
http://www.foretec.com/python/workshops/1998-11/proceedings/guido/index.html

Note, however, that the above talk was given last November.  Things have
probably changed a bit since then.

<hungarian-notation-anyone-l'yrs?>

-- 
Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
518-372-5583



From wjones at usachoice.net  Thu Jun  3 18:11:09 1999
From: wjones at usachoice.net (wjones at usachoice.net)
Date: Thu, 3 Jun 1999 22:11:09 GMT
Subject: Python 1.5 help
Message-ID: <7j6uht$q1tk@eGroups.com>

I am new to the python language and just downloaded Python 1.5. I want to know where I can find more info on the web (or anywhere) about programing in this language? www.python.com is good but complacated! Please help! Thanks.




From jmrober1 at ingr.com  Thu Jun  3 14:14:42 1999
From: jmrober1 at ingr.com (Robertson, Joseph M)
Date: Thu, 3 Jun 1999 18:14:42 GMT
Subject: [tkinter] two questions regarding Canvas
Message-ID: <FE1ABBF2573FD2119417080009C15B1A2A3274@HQ3>

Hi Greg,

Thats really cool!  You've got me going now.  What we need is a way to have
'create_text' inherit from Entry.  We need someone who nows Tk well enough
to build one that have a 'create_text_entry' object.

Thanks for sharing the ideas.
Joe Robertson
256-730-8216
jmrober1 at ingr.com
jmrobert at ro.com
http://ro.com/~jmrobert/





From matz at netlab.co.jp  Wed Jun  2 00:23:01 1999
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 02 Jun 1999 13:23:01 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
 <87pv3iq6ss.fsf@ev.netlab.co.jp> <7iv86a$iud@news.dns.microsoft.com>
 <u3r9nwtd9q.fsf@ares.sys1.ngo.okisoft.co.jp>
Message-ID: <87so8bnsd6.fsf@ev.netlab.co.jp>

Hisao Suzuki <suzuki611 at okisoft.co.jp> writes:

|In fact, there is no guarantee that all garbages will be reused,
|up to their last one byte, within the life time of the java
|process, and there is also no guarantee that ALL finalizers will
|be eventually invoked.  It is an issue of the quality of JVM
|implementation, you know.

In Java, right?

|Perhaps it is safe to say that destructors with real GC are more
|error prone.  (The present Python is safer in this respect!)

Safer than Java, probably.

Ruby garantees (well, trying to garantee) that ALL finalizers will be
called before process termination, so that Ruby is as safe as Python,
at least theoretically.

|BTW, Mr. Matz has invented and implemented his excellent
|language, Ruby, which is a sort of Perl's successor with some
|intrinsic object-oriented features.  IMHO, we must have respect
|for his opinion as such or, at the very least, as the evangelist
|of his own language.  See http://www.netlab.co.jp/ruby/

Thank you for mentioning it.  But my first intention is not
evangelism.  I wanted to resolve the misunderstanding abount real GC
in this newsgroup, and make Python better.  Believe it or not.

But, of cource, new comers are always welcome. :-)

						matz.



From pjc_2 at geocities.com  Sat Jun  5 15:56:55 1999
From: pjc_2 at geocities.com (Philip C.)
Date: Sat, 5 Jun 1999 14:56:55 -0500
Subject: smtp module 
Message-ID: <0gf63.28$Vv4.277@news.megsinet.net>

I have pretty much figured out how to use the module but I can't seem
set the subject or the from of the bcc etc....

please tell me how to do it?





From William.Tanksley at p98.f112.n480.z2.fidonet.org  Wed Jun 30 16:35:49 1999
From: William.Tanksley at p98.f112.n480.z2.fidonet.org (William Tanksley)
Date: Wed, 30 Jun 1999 21:35:49 +0100
Subject: Does the first argument of a method *has to be* self?
Message-ID: <000015d6@bossar.com.pl>

From: wtanksle at dolphin.openprojects.net (William Tanksley)

On Wed, 30 Jun 1999 22:21:21 +0200, Gerrit Holl wrote:
>Hello,

>I never saw code where the first argument of a method wasn't self. Is it
>in theory possible?

It's always self, although you're free to call it anything you want.

>Gerrit.

-- 
-William "Billy" Tanksley



From paul at prescod.net  Tue Jun  1 01:43:18 1999
From: paul at prescod.net (Paul Prescod)
Date: Tue, 1 Jun 1999 05:43:18 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
		<7imr6b$1s0$1@nnrp1.deja.com>
	 <87pv3iq6ss.fsf@ev.netlab.co.jp>
		<3752C76D.604CD474@prescod.net> <87n1ykr0z9.fsf@ev.netlab.co.jp>
Message-ID: <375372F6.A874554@prescod.net>

Yukihiro Matsumoto wrote:
> 
> >As many people have noted before, you would not be able to use
> >destructors to free up resources like memory or file descriptors.
> 
> which, as I understand, means destructors can not be used with real
> GC.  That's just not true.

Is there a language with real GC where destructors are called as
predictably as they are in Python?

> |But I think that the bigger problem with "real GC" is that C programs
> |often hang on to references to Python objects and any kind of compacting
> |GC would invalidate those references.
> 
> It's the common legend especially here in comp.lang.python.
> 
> >From my experience with Scheme and Ruby, both with real GC, the real
> GC is mostly very fast.  

Speed wasn't my concern. I asked about references in C code. Do we use
PyObject **s instead of PyObject *s now?

> |But note that Python's guarantees are more important in small, simple
> |programs and Java's guarantees are better for large, complex programs with
> |complicated data structures.
> 
> Do you really mean Python the object-oriented programming language
> rarely treat complicated data structures?  

No, I said that Python is optimized for basic users, not for the experts.
The experts can figure out how to work around problems. The beginning
users cannot. "I'm opening all of the files in a directory in a loop and
in large directories I get out of file handle errors."

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for only himself
 http://itrc.uwaterloo.ca/~papresco

"Silence," wrote Melville, "is the only Voice of God." The assertion,
like its subject, cuts both ways, negating and affirming, implying both
absence and presence, offering us a choice; it's a line that the Society
of American Atheists could put on its letterhead and the Society of
Friends could silently endorse while waiting to be moved by the spirit
to speak. - Listening for Silence by Mark Slouka, Apr. 1999, Harper's



From gmcm at hypernet.com  Wed Jun  2 12:21:22 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 2 Jun 1999 11:21:22 -0500
Subject: Alien whitespace eating nanovirus strikes again!
In-Reply-To: <375536C2.DD917F99@prescod.net>
Message-ID: <1283781766-36219323@hypernet.com>

Paul Prescod  wrote:

> ... There is no single dominant philosophy similar to
> TMTOWTDI that drives Python's growth. Or one might say that Python's
> dominant philosophy is Aristotle's Golden Mean. Or perhaps Python's
> philosophy is that code is communication and readable code is clear
> communication. 

Python has 3 core philosophies...

Um, make that 4 core philosopies...

> I truly think that Perl is a unique and innovative language. Larry
> can promote it without playing on people's fears of whitespace and
> fascism.

How else do you explain all the veiled references to Domicans? 
Evidently Larry doesn't know about the secret signs and signals <wink 
wink nudge nudge>.

conspiratorily-y'rs


- Gordon



From arcege at shore.net  Fri Jun 25 08:44:46 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Fri, 25 Jun 1999 12:44:46 GMT
Subject: Embedding Question
References: <slrn7n4699.g36.tuna@delirious.fsnnet> <Frrc3.63$Of3.12274@news.shore.net> <slrn7n5eds.5ns.tuna@delirious.fsnnet>
Message-ID: <2RKc3.211$Of3.49674@news.shore.net>

Lloyd Weehuizen <tuna at delirious.fsnnet> wrote:
: On Thu, 24 Jun 1999 14:40:37 GMT, Michael P. Reilly <arcege at shore.net> wrote:
:>Lloyd Weehuizen <tuna at delirious.fsnnet> wrote:
:>: Hi
:>
:>: If I use the PyRun_String method, with a string that defines a method called

:>You can use any number of the "del" based operations.  But from the
:>high-level (using the PyRun_* functions), you would want:

: Ok, that would work, but I don't really know what functions are being
: defined, as the user can edit them. Is there no way to flush everything and
: start with a clean slate except doing a Finalize() and Initialize()?

: Hmm I spose I could search through the strings looking for all the "def"
: keyworks ;)

: Also, with the PyRun_String method, what is the difference between the
: locals and globals definition? I've set them to be the same, like the
: PyRun_SimpleString method does, could this be causing the problem? Are
: globals remembered the entire time the interpreter is running and locals
: only for that PyRun_String method? Into what namespace do "def" statements
: get loaded? Globals or locals? 

You can do a number of things.  But I would suggest creating a
new module and running the user code there, this is essentially the
purpose of the globals and locals values.

  PyObject *
  New_Python_Module()
    { static int modulenumber = 0;
      char name[50];

      sprintf(name, "run%d", modulenumber++);
      return PyModule_New(name);
    }
  /* if module is NULL, then create a new module to run the commands in
   * this allows multiple calls within the same module
   */
  PyObject *
  call_Python_string(pycmd, module)
    char *pycmd;
    PyObject **module;
    { PyObject *result = NULL;
      PyObject *globals;

      if (*module == NULL)
        *module = New_Python_Module();
      globals = PyModule_GetDict(*module);
      result = PyRun_String(pycmd, 0, globals, globals);
      return result;
    }

Notice that I did indeed keep the locals and globals the same.  Having
the locals is for evaluation within a function or class (or other such
"inner" namespace, cf. "Code blocks, execution frames, and namespaces"
in the Python Reference Manual).

Now to clean up, you dereference the module created:

  /* run two files in the same module, return 1 if successful, 0 otherwise
   * the module (and data) is destroyed.
   */
  int runfiles(file1, file2)
    char *file1, *file2;
    { int rc = 0;
      char *file_contents;
      PyObject *module = NULL;
      PyObject *result;

      file_contents = read_file(file1);
      result = call_Python_string(file_contents, &module);
      /* module should now not be NULL, unless the module could not
         be created */
      if (module == NULL || result == NULL) {
        Py_XDECREF(module);  /* destroy the module */
        Py_XDECREF(result);  /* destroy the object */
        return 0;
      }
      Py_DECREF(result);  /* python returns an object, so destroy it now */
      /* now run the next file, but in the same module */
      result = call_Python_string(file_contents, &module);
      Py_XDECREF(module); /* destroy the module */
      rc = (result != NULL);
      Py_XDECREF(result);
      return rc;
    }

Unless the user code modifies other modules (which it might), then this
is pretty good.  If you want to prevent even this security hole, then
look at the rexec module.

Alternatively, you can do the same that is done in some Python code:
  exec command in global_dicts

  int run_command(char *pycmd)
    { PyObject *result, *globals;
      PyObject *builtins;

      globals = PyDict_New();
      builtins = PyImport_ImportModule("__builtin__");
      PyDict_SetItemString(globals, "__builtins__", builtins);

      result = PyRun_String(pycmd, 0, globals, globals);
      Py_DECREF(globals);
      if (result == NULL)
        return 0;
      Py_DECREF(result);
      return 1;
    }

This is simpler, but sometimes you just need a module. :)

  -Arcege




From htrd90 at zepler.org  Wed Jun  9 18:00:07 1999
From: htrd90 at zepler.org (Toby Dickenson)
Date: Wed, 09 Jun 1999 22:00:07 GMT
Subject: COM Question - QueryInterface?
References: <375C7D56.C9744E39@connix.com>
Message-ID: <375fe0b1.1117810@news.freeserve.net>

Chris Tavares <tavares at connix.com> wrote:

>I've got a (hopefully easy) question about python's COM support.
>
>I'm developing a COM object in C++ which implements the COM collection
>idiom - an Item method that returns another object. The code in Python
>I'm using is something like this:
>
>from win32com.client import Dispatch
>drives = Dispatch("AmAccess.Drives")
>drive = drives.Item("c:")
>dir = drive.Item("some-area")
>
>On this last line, I get an AttributeError for Item.
>
>I think I know the reason why - the interface returned from Drives.Item
>isn't a dual, it's an oleautomation compatible vtable interface.
>IDispatch on that object is implemented separately. So, PythonCOM sees
>the VT_UNKNOWN return type and wraps a PYIUnknown object around it. I
>need to be able to do a QueryInterface for IDispatch.

Your analysis sounds about right, this should give you the IDispatch interface
returned by querying for IID_IDispatch:

drive = drives.Item("c:")._oleobj_.QueryInterface(pythoncom.IID_IDispatch)

A better approach might be to make the interface dual? It sounds like you have
already tackled the painful bits of making your interface dual; implementing
IDispatch (under IID_IDispatch) and sticking to the oleautomation compatible
restrictions. Making it dual should be an easy step, and will avoid this and
several different non-python problems too.




Toby Dickenson



From mlh at idt.ntnu.no  Sat Jun 12 12:15:27 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 12 Jun 1999 18:15:27 +0200
Subject: Prolog module
References: <99061120061800.01147@pruneta.titinet.enst-bretagne.fr>
Message-ID: <y0jaeu52y5c.fsf@vier.idi.ntnu.no>

Jordi Inglada <jordi.inglada at enst-bretagne.fr> writes:

> Hi,
> 
> I am new to Python Programming (just did the tutorial and played
> a little with it). It is really a great language!
> 
> I have seen that Python can interact with other languages. My question
> is: can Python access a Prolog kernel?

Sure. (Actually, I have been thinking of implementing a simple Prolog
kernel *in* Python... But that might be far into the future... :) You
could either wrap the Prolog kernel with swig (a tool for interfacing
between c-programs/modules and scripting languages) or you could check
out pyexpect which lets you interact with command-line programs (like
a Prolog interpreter) programmatically.

> Or even, is there a module that
> implements in Python a Prolog kernel? (in the same way one can use
> Java from Python).
> 
> Thanks.

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From ajung at sz-sb.de  Mon Jun 21 00:57:30 1999
From: ajung at sz-sb.de (Andreas Jung)
Date: Mon, 21 Jun 1999 04:57:30 GMT
Subject: Module for Solaris /proc ?
Message-ID: <19990621065729.A17907@sz-sb.de>

Reply-To: 
Is there a module for accessing the /proc filesystem
under Solaris 2.X available ?

Thanks in advance,
Andreas




From tim_one at email.msn.com  Thu Jun 17 20:02:39 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 17 Jun 1999 20:02:39 -0400
Subject: Crawling Python
In-Reply-To: <37695707.3168269@news.tpnet.pl>
Message-ID: <000601beb91d$e1321ae0$839e2299@tim>

[MK]
> While playing with anydbm module using IDLE 0.4, I have made
> Python to do the following:
>
> Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> [... a bunch of code building a long string ...]
> >>> len(s)
> 49035
> >>>
>
> Then I tried was typing s only to see the string. Shell printed
> it after some 20 seconds, and then came to a crawl.

I've noticed that too.  Don't know whether it's the IDLE code or the Tk Text
widget that's unhappy.  Probably both; e.g., IDLE spends a lot of time
syntax-coloring the junk, and the Tk Text docs warn that long lines can be
expensive.

Everything speeds up back to normal if you simply delete the giant string
from the window.

> ...
> PS. I have tried the same thing in DOS executable -- above does
> not happen.

Sure, and in return none of IDLE's history, undo/redo, searching, replacing,
scrolling or editing cmds work in a DOS box either <wink>.

adaptively y'rs  - tim





From jim at interet.com  Wed Jun 23 12:03:16 1999
From: jim at interet.com (James C. Ahlstrom)
Date: Wed, 23 Jun 1999 12:03:16 -0400
Subject: Build the Python library into the binary
References: <376F9518.D3252CEE@interet.com> <7kolnv$gp6$1@nnrp1.deja.com>
Message-ID: <37710544.F47F6B2C@interet.com>

radbit at my-deja.com wrote:

> did you have a look at mxCGIPython at
> http://starship.skyport.net/~lemburg/mxCGIPython.html ?
> This is a single executable not shared library with the
> standard Python library embedded.

Thanks.  I the problem is I want only part of the Python library,
and I still need to be able to use freeze.

Jim Ahlstrom



From bwinton at tor.dhs.org  Wed Jun  9 23:23:06 1999
From: bwinton at tor.dhs.org (Blake Winton)
Date: Thu, 10 Jun 1999 03:23:06 GMT
Subject: How do I match the last occurence of a character?
References: <375F21B3.D23A5EA9@west.boeing.com>
Message-ID: <slrn7luc4m.g2c.bwinton@tor.dhs.org>

On Thu, 10 Jun 1999, Jeff P. <jeffrey.polaski at west.boeing.com> wrote:
>How do I match the last occurence of '\' in a string?
>import regex

This is your first problem.  Why use a tool more powerful than you
really want?  Check out the string module.

>>> import string
>>> name = "test\\a\\b\\c.html"
>>> string.rfind(name,"\\")
8
>>> name2 = name[0:string.rfind(name,"\\" )]
>>> name2
'test\\a\\b'

Note that I have to use "\\" to match a single "\".
Oh, and to get output similar to the one in your example, you would have
to use something like

baseurl = url[ 0 : string.rfind( url, "\\" )+1 ]

Hope this helps,
Blake.

-- 
One Will.  One Dream.  One Truth.  One Destiny.  One Love.



From graham at sloth.math.uga.edu  Fri Jun  4 13:04:42 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 4 Jun 1999 17:04:42 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
Message-ID: <7j90va$6lk$1@cronkite.cc.uga.edu>

Hisao Suzuki (suzuki611 at okisoft.co.jp) wrote: : 
> Anyway, as you
> say, I'd think that we need an actual experience.
Graham Matthews <graham at sloth.math.uga.edu> writes:
> But people do have experience with GC. Sure that experience is
> not in Python, but why do you believe that GC in Python will be
> radically different to GC in any other language?
Jeremy Hylton (jeremy at cnri.reston.va.us) wrote:
: I think the problem is that the people who have experience
: implementing garbage collection are not the people who advocating
: adding it to Python.  It's one thing to have used languages with
: garbage collectors or to have read books and papers about them; it's
: quite another thing to have done the implementation.  The gap between
: theory and practice is greater in practice than in theory.

I am completely confused by this statement. The two people that have
been advocating GC the hardest are myself and the author of Ruby. Since
Ruby has GC the author has clearly implemented a collector. As for myself
I have implemented 3 collectors, one of which is used in a system that
is used by quite a few people worldwide. So I am not sure where your
comment is coming from?

Jeremy Hylton (jeremy at cnri.reston.va.us) wrote:
: So go for it.  Many of us would love to look at some rough, but
: running code.  Just make sure it is portable enough to run on all the
: platforms that Python runs on, and make sure it has minimal impact on
: C extension modules.  I doubt that it's impossible, but I expect that
: it's hard.

I have no plans to implement a GC for Python. I have already implemented
enough standard collectors for one lifetime. Besides which someone far
more intimate with the implementation details of Python would take less
time. That's not a cop out. I just don't have the time.

graham
-- 
                    One day in a nuclear age 
                  They may understand our rage
              We build machines that we can't control
              And bury the waste in a great big hole



From vkolosov at unitedmedia.com  Fri Jun 11 10:06:32 1999
From: vkolosov at unitedmedia.com (Kolosov, Victor)
Date: Fri, 11 Jun 1999 10:06:32 -0400
Subject: Help: MIME Attachments
Message-ID: <461182F0B9B7D111B4000008C7282ADC22B080@nynt04.umny.scripps.com>

Here is the right sequence(first write header then body):

		subwriter = w.nextpart()
		subwriter.addheader("Content-Transfer-Encoding", "base64")
		subwriter.addheader("Content-Disposition", 'attachment;
filename="DiaryFile.notes"')
		subwriter.flushheaders()
		f = subwriter.startbody('application/octet-stream;
name="DiaryFile.notes"')

		base64.encode(open('./DiaryFile.notes', 'r'), f)
	w.lastpart()

I could not quite figure where do you write the encoded file to your message
and I whould not use 'base64' directry but 'encode' function from mimetools.
If you want here is the my function that does it for me:

def encode_me(file):
	from StringIO import StringIO
	from mimetools import encode
	file_to_send=open(file,'rb')
	encoded_object = StringIO()
	encode(file_to_send,encoded_object,'base64')
	file_to_send.close()
	file = encoded_object.getvalue()
	return file

So if you have this function - you just do:

		subwriter = w.nextpart()
		subwriter.addheader("Content-Transfer-Encoding", "base64")
		subwriter.addheader("Content-Disposition", 'attachment;
filename="DiaryFile.notes"')
		subwriter.flushheaders()
		f = subwriter.startbody('application/octet-stream;
name="DiaryFile.notes"')
		f.write(encode_me('DiaryFile.notes'))
		w.lastpart()

--
Hope it helps


> -----Original Message-----
> From:	k_mcdermott at my-deja.com [SMTP:k_mcdermott at my-deja.com]
> Sent:	Friday, June 11, 1999 5:37 AM
> To:	python-list at python.org
> Subject:	Help: MIME Attachments
> 
> Hi,
> 
> I am trying to send a plain text file, as an attachment through e-mail.
> 
> This is a hacked version of my test code below, basically the
> program interprets a dump from a diary system, and then writes
> Lotus Notes importable Calendar entries.  I want to then mail
> them to users with an accompanying instruction header to allow them
> to import them into their calendars...
> 
> The only problem is the mailing them to users bit, I either get
> the plain text in the file (which I don't want, I just want them
> to detach the text file and then import) or I get the Base64 encoded
> text, either way there is no attachment...
> 
> My Problem is that the MIME headers indicating Base64 encoding
> appears a line after the content type, and I believe that this blank
> line terminates the headers...
> 
> So my question is, at what point should I be introducing these
> headers...?
> 
> TIA
> 
> Kevin
> 
> ps Python has brought a project that was originally quoted at "tens of
> thousands of pounds and hundreds of man-hours" down to around 4 hours
> so far, and if I can crack this, ZERO outlay :-)
> 
> import base64
> import sys
> import StringIO
> import MimeWriter
> from smtplib import SMTP
> 
> # Mails the Diary Import file to the user in MIME format
> # Based upon code by GVR
> def mailFileToUser(_userName, _fileName):
> 	outputfp = StringIO.StringIO()
> 	w = MimeWriter.MimeWriter(outputfp)
> 	w.addheader("subject", "Diary Entries from Mainframe")
> 	w.flushheaders()
> 	w.startmultipartbody("mixed")
> 	subwriter = w.nextpart()
> 	f = subwriter.startbody('application/octet-stream;
> name="DiaryFile.notes"')
> 
> >>> This is the puzzling bit...
> 
> 	subwriter.addheader("Content-Transfer-Encoding", "base64")
> 	subwriter.addheader("Content-Disposition", 'attachment;
> filename="DiaryFile.notes"')
> >>>
> 
> 	subwriter.flushheaders()
> 	base64.encode(open('./DiaryFile.notes', 'r'), f)
> 	w.lastpart()
> 	# s = SMTP("localhost")
> 	# s.sendmail("diary at system.com", ["my at email.address.com"],
> outputfp.getvalue())
> 	# s.close()
> 	print outputfp.getvalue()
> if __name__=='__main__':
> 	mailFileToUser('Kevin McDermott', 'TestFile.out')
> 
> 
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.



From mcfletch at home.com  Thu Jun 10 16:26:33 1999
From: mcfletch at home.com (Mike C. Fletcher)
Date: Thu, 10 Jun 1999 20:26:33 GMT
Subject: Broken 1.5.2 installation? Missing cw3215mt.dll
Message-ID: <ZbV73.13408$WL.177860@news2.rdc1.on.home.com>

Just downloaded the latest 1.5.2 for installation.  Did the installation and
got an error message telling me that the above-mentioned file is not on the
path.  Indeed, it does not seem to exist anywhere on the machine.  I'm
pretty sure this is part of TCL (the cygnus TCL distribution has a file
named cw3215.dll).  However, I have no idea where to get it.

Help appreciated,
Mike

--
______________________________________
   M i k e   C .  F l e t c h e r

       mcfletch at vrtelecom.com

        Designer, Researcher






From nathan at islanddata.com  Tue Jun 15 14:44:47 1999
From: nathan at islanddata.com (Nathan Clegg)
Date: Tue, 15 Jun 1999 11:44:47 -0700 (PDT)
Subject: socket bindings
Message-ID: <XFMail.990615114447.nathan@islanddata.com>

I have a server application that binds itself to a specified port and
waits for connections.  It handles all connections fine and quits cleanly
after closing all of the connected sockets as well as the listening
socket.  However, if I start the application again, it will die, saying
the address is already bound to.  If I wait 2-3 minutes, the port will
become available again.  I can "solve" this by implementing a loop that
continues until the socket binds successfully, but this seems a little
klunky.  Can anyone tell me why the socket might not be correctly
releasing itself?  Relevant code follows:

    server = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
    connected = None
    
    while not connected:
        try:
            server.bind('', PORT)
            connected = 1
        except socket.error:
            time.sleep(.1)
    server.listen(128)  # let the OS determine the limit
    print 'Listening...'

# code making connections deleted

    server.close()



----------------------------------
Nathan Clegg
 nathan at islanddata.com





From pduffin at mailserver.hursley.ibm.com  Fri Jun  4 06:39:15 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Fri, 04 Jun 1999 11:39:15 +0100
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <3756E3D9.E44D2B39@swcp.com> <3756F115.CBBAC9D8@pop.vet.uu.nl>
Message-ID: <3757ACD3.3F54@mailserver.hursley.ibm.com>

Martijn Faassen wrote:
> 
> Alex Rice wrote:
> >
> > Martijn Faassen wrote:
> > ...
> >
> > > Okay, sorry for all this, it's out of line, I just needed to blow off
> > > steam. :)
> >
> > As the composer John Cage would have told you, silence is music as much
> > as tones and harmonies are music. silence == whitespace ?
> 
> What did John Cage say about curly braces and dollar signs? :)
> 

He probably likes them because they remind him of quavers and 
semi-quavers and clefs etc.

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From faassen at pop.vet.uu.nl  Thu Jun  3 04:55:44 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Thu, 03 Jun 1999 10:55:44 +0200
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <199906022028.WAA28874@axil.hvision.nl> <al87lplg9rt.fsf@kruuna.Helsinki.FI>
Message-ID: <37564310.3131B5B4@pop.vet.uu.nl>

Markus Stenberg wrote:
[curly python]

Yay! :)

Funny, I was prompted to write just about the same thing yesterday. I
was calling it WallPython, though, and provided an 'unWall' function to
remove the curly braces. Perhaps you could rename CurlyPython into that.

Unlike you I never got my (slightly more ambitious) code debugged
properly, though. :)

Regards,

Martijn



From daniels at mindspring.com  Wed Jun 30 23:55:35 1999
From: daniels at mindspring.com (Alan Daniels)
Date: 1 Jul 1999 03:55:35 GMT
Subject: Python suitable for a game engine?
Message-ID: <slrn7nlpuk.eu.daniels@localhost.localdomain>

I'm working on a Quake-style game engine, where my goal is to give it
a fully integrated editor, similar to the excellent ones in Starcraft
and Unreal. I'm planning on doing this by giving the game and the
editor as much back-end code in common as possible.

What I want to do is be able to design the level in the editor, and
then sprinkle objects and monsters around the level. All of these will
have default low-level behaviors, so that level designers don't have
to do monster AI from scratch. However, all the objects and monsters
will have simple properties manipulated through a property list,
similar to that in VB. This way, they can set properties such as, is
the monster aggressive or passive, etc.

Here's the thing. I want to have a script as its main property, so that
the monsters can have elaborate behaviors, such as being mournful when
their leader is killed, etc. The events would be things such as
"create", "monster sees player", "clock tick", etc. All the functions
in the script would be callbacks written in C++, to manipulate the
monster to do things such as "patrol area", "play an animation",
"sleep for X seconds", etc. In other words, the scripts would be glue,
not the main thread of execution.

I'm been using Python for a couple of years now, and it is my favorite
scripting language. But, I don't have much experience with embedding
it in a another program, and I don't know enough about the speed
issues. So, my questions are...

- Will Python be fast enough? Even with possibly 100 or 200 active
  items on a level?
- Especially when all of the scripts will be live, and usually in the
  middle of a callback?
- Should I use the interpreter in the library, or should I take an
  approach of crunching the byte-codes myself?

I've seen plenty of cases where Python is used as a "VBA" style
embedded language (load up the script, run it, and be done with it),
but I haven't seen anything like this mentioned before. My other two
candidate languages are Scheme and Perl, and I hate the syntax of
both. So I really REALLY want to use Python if its viable. :=) If
anybody can shed some light on this, I'd appreciate it. Thanks.

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



From nelson at crynwr.com  Tue Jun 22 10:44:27 1999
From: nelson at crynwr.com (Russell Nelson)
Date: 22 Jun 1999 10:44:27 -0400
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <z%Q93.4399$5E4.36952@news2.randori.com>
Message-ID: <m2n1xsmh10.fsf@desk.crynwr.com>

>   I'm trying to collect a list of all the _current_ issues making tcl
> a bad language choice...

Because it necessarily parses strings.  Been there, done that, have
the scars to prove it.

-- 
-russ nelson <rn-sig at crynwr.com>  http://crynwr.com/~nelson
Crynwr supports Open Source(tm) Software| PGPok | Government schools are so
521 Pleasant Valley Rd. | +1 315 268 1925 voice | bad that any rank amateur
Potsdam, NY 13676-3213  | +1 315 268 9201 FAX   | can outdo them. Homeschool!



From jeremy at cnri.reston.va.us  Fri Jun  4 11:20:39 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Fri,  4 Jun 1999 11:20:39 -0400 (EDT)
Subject: Python 2.0
In-Reply-To: <7j8o26$55v$1@cronkite.cc.uga.edu>
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7imr6b$1s0$1@nnrp1.deja.com>
	<7j8o26$55v$1@cronkite.cc.uga.edu>
Message-ID: <14167.60178.820265.375878@bitdiddle.cnri.reston.va.us>

>>>>> "GM" == Graham Matthews <graham at sloth.math.uga.edu> writes:

  GM> Hisao Suzuki (suzuki611 at okisoft.co.jp) wrote: : Me too.  And I
  GM> am afraid that some evangelists would do nothing : but _only_
  GM> point out Python's some so-called fault in order to : drive us
  GM> into convert to their favorite or invented language...

  GM> I am sorry Hisao but this is pure fantasy. This kind of "anyone
  GM> suggesting improvement to Python is a critic" mentality pervades
  GM> this thread and this newsgroup.

Strictly speaking, I believe a critic is "a person who judges,
evaluates, or criticizes."  (http://infoplease.lycos.com/ipd/A0393235.html)
So, yeah, people suggesting improvements are critics.  I think we all
benefit from having good critics.  All sorts of changes have come as
the result of constructive critism -- packages, the threading module,
[].pop, {}.get, etc.  Probably lots more that I'm not thinking of.

  GM> Hisao Suzuki (suzuki611 at okisoft.co.jp) wrote: : Anyway, as you
  GM> say, I'd think that we need an actual experience.

  GM> But people do have experience with GC. Sure that experience is
  GM> not in Python, but why do you believe that GC in Python will be
  GM> radically different to GC in any other language?

I think the problem is that the people who have experience
implementing garbage collection are not the people who advocating
adding it to Python.  It's one thing to have used languages with
garbage collectors or to have read books and papers about them; it's
quite another thing to have done the implementation.  The gap between
theory and practice is greater in practice than in theory.

So go for it.  Many of us would love to look at some rough, but
running code.  Just make sure it is portable enough to run on all the
platforms that Python runs on, and make sure it has minimal impact on
C extension modules.  I doubt that it's impossible, but I expect that
it's hard.

Jeremy



From guido at CNRI.Reston.VA.US  Fri Jun 11 00:43:32 1999
From: guido at CNRI.Reston.VA.US (Guido van Rossum)
Date: Fri, 11 Jun 1999 00:43:32 -0400
Subject: fork()
In-Reply-To: Your message of "Fri, 11 Jun 1999 00:19:46 EDT."
             <000c01beb3c1$a3f428a0$329e2299@tim> 
References: <000c01beb3c1$a3f428a0$329e2299@tim> 
Message-ID: <199906110443.AAA02408@eric.cnri.reston.va.us>

> From: "Tim Peters" <tim_one at email.msn.com>

> In this kind of approach, I doubt there's a substantially better way.  Guido
> will also want a doubly-linked list (where the links don't count against the
> refcount!) so that when a dict goes out of existence via the normal
> refcount-hits-0 route, the dict can efficiently unlink itself from the list.

Probably, although everytime this comes up I try to figure out a
clever scheme to use a singly-linked list, where the zero-refcnt ones
are recycled at the next mark-and-sweep.

> So, ignoring that, you make one pass over the trash list just to run
> finalizers.

Bzzt!  We only have a trash list of dictionaries.  The finalizers are
typically on class instances (or extension type objects) that are
somehow referenced by those dictionaries!  But not every object
reachable from a trash dictionary is trash, of course.  Nasty... :-(

Could we get away with not calling (user) finalizers on objects in
trash cycles at all?  Since the finalization order is problematic at
best, this almost seems acceptable: in the formal semantics, objects
that are part of cycles would live forever, and as an invisible
optimization we recycle their memory anyway if we know they are
unreachable.  (I've got the feeling that I've seen this rule before
somewhere.)  We'd still get complaints "my __del__ doesn't get called"
from some users, and the answer would still be "it's in a cycle -- use
an explicit close"; but those who know what they are doing can be
guaranteed that their memory cycles get recycled.  In other words it
would be no worse than today and for some people it would be better.

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



From rozzin at geekspace.com  Fri Jun 25 11:02:08 1999
From: rozzin at geekspace.com (Joshua Rosen)
Date: Fri, 25 Jun 1999 11:02:08 -0400
Subject: Crawling Python
References: <37695707.3168269@news.tpnet.pl> <xj0vhcl5kut.fsf@cyclone.harvard.edu>
Message-ID: <377399F0.A51634DC@geekspace.com>

Michael Haggerty wrote:
> >>> import string
> >>> s = []
> >>> for i in d.keys():
> >>>   s.append(d[i])
> >>> s = string.join(s, '')
> 
> then you get exactly the same result in O(n) time, because s.append()
> doesn't have to copy each time.  For large n the second algorithm will
> beat the first one every time, by an increasingly huge margin.

How about...

from array import array
s = array('c')
for e in d.values():
s.fromstring(e)
s = s.tostring()

...?




From amitp at Xenon.Stanford.EDU  Tue Jun  1 15:52:59 1999
From: amitp at Xenon.Stanford.EDU (Amit Patel)
Date: 1 Jun 1999 19:52:59 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <87n1ykr0z9.fsf@ev.netlab.co.jp> <7ivs9q$dc5@news.dns.microsoft.com> <87k8toqv70.fsf@ev.netlab.co.jp>
Message-ID: <7j1dmr$dr7$1@nntp.Stanford.EDU>

 Yukihiro Matsumoto  <matz at netlab.co.jp> wrote:
| Hi.
| 
| "Vadim Chugunov" <chega_ at yahoo.com> writes:
| 
| |> Remember the original statement is like this:
| |>
| |> >As many people have noted before, you would not be able to use
| |> >destructors to free up resources like memory or file descriptors.
| |>
| |>which, as I understand, means destructors can not be used with real
| |>GC.  That's just not true.
| |
| |I'm sorry for being unclear, but the key words here were: "to free up
| |resources".
| |And I meant "to free up resources the moment I expect them to", not "some time
| |later".
| 
| Well, OK.  I understand now.  Now only thing I can say is, why do you
| emphasize so much on reclaiming object at the time you exepcted?
| Because of your C++ background?

In my case it's because in Python and C++ I tend to deal with
resources a LOT.  Sockets especially, but also files.  When I write
Scheme and ML code, I rarely ever deal with resources other than
memory, so it doesn't bother me.

It's not the language but how I tend to use it that colors my opinion
on destructors.

	- Amit



From tim_one at email.msn.com  Wed Jun 23 23:58:53 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 23 Jun 1999 23:58:53 -0400
Subject: List <-> news hosed again?
In-Reply-To: <199906231508.LAA23048@eric.cnri.reston.va.us>
Message-ID: <000601bebdf5$e0572fc0$549e2299@tim>

[Guido]
> We're looking into this; uunet changed their configuration and we
> neglected to make a matching change.  It should be fixed now
> (presuming this message shows up on the uunet side of things).

Alas, don't think it worked -- DejaNews doesn't have your msg yet, nor my
ISP's (admittedly flaky) c.l.py feed.  Appears to have stopped working very
early AM (EDT) on 18-June-99.

in-the-last-hideous-stage-of-ego-surfing-withdrawal<wink>-ly y'rs  - tim





From jbauer at rubic.com  Tue Jun 15 11:57:35 1999
From: jbauer at rubic.com (Jeff Bauer)
Date: Tue, 15 Jun 1999 15:57:35 GMT
Subject: Newbie: easy filecopy?
References: <376531D3.7E42AC42@phoenix-edv.netzservice.de> <14182.27386.991483.999566@weyr.cnri.reston.va.us>
Message-ID: <376677EF.B48E0736@rubic.com>

"Fred L. Drake" wrote:
> As someone has pointed out, the shutil module contains the
> beast you're looking for.

I believe shutil.py is an obscure name for this
particular module.  I only discovered it (years ago) 
by accident when looking for an example of os.path.walk.

It's probably too late, but it would be great if we
could change the name to filecopy.py, retaining
shutil.py (which could import filecopy's code).

-Jeff



From guido at cnri.reston.va.us  Wed Jun  2 13:44:13 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 02 Jun 1999 13:44:13 -0400
Subject: Windows Stack Size
References: <3752C225.65C98B52@cloud9.net> <3752F154.6C444928@ricochet.net> <375413D8.C3B02D2@cloud9.net> <7j1q41$3fu$1@m2.c2.telstra-mm.net.au> <3755259C.3513D254@cloud9.net>
Message-ID: <5liu96zedu.fsf@eric.cnri.reston.va.us>

Michael Muller <proteus at cloud9.net> writes:

> Using the raw Tkinter 'Text' control instead of my ScrollingText
> class.  It doesn't seem to have this problem.  Unfortunately, it
> doesn't have scrollbars either.

Try the standard (though under-publicized) module ScrolledText.py,
which does what you are trying to do and (as far as I know) doesn't
trap.

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



From da at ski.org  Mon Jun 28 15:41:09 1999
From: da at ski.org (David Ascher)
Date: Mon, 28 Jun 1999 12:41:09 -0700 (Pacific Daylight Time)
Subject: high-level "read" function?
In-Reply-To: <3777D967.484F0B6D@lbl.gov>
Message-ID: <Pine.WNT.4.04.9906281238220.155-100000@rigoletto.ski.org>

On Mon, 28 Jun 1999, Christine Celata wrote:

> I am just starting to use Python.  It is foreign to me-- I'm a physicist
> used to Fortran, not C.  I have succeeded in getting Python to read
> program user input, but as far as I can see, in order to get it to read
> numbers separated by commas from the terminal and store them as floating
> point variables,

input() can do most of the job for you:

  >>> userinput = input("Enter a bunch of number: ")
  Enter a bunch of number: 1.24, 400, 15.5
  >>> print userinput
  (1.24, 400, 15.5)
  >>> type(userinput)
  <type 'tuple'>
  >>> type(userinput[0])
  <type 'float'>
  >>> type(userinput[1])
  <type 'int'>

If you really need them all to be floating point variables, you can do
that with:

  >>> floats = map(float, userinput)
  >>> print floats
  [1.24, 400.0, 15.5]  

--david ascher




From tgm at math.uni-bonn.de  Mon Jun 28 04:48:09 1999
From: tgm at math.uni-bonn.de (Thomas Ackermann)
Date: Mon, 28 Jun 1999 10:48:09 +0200
Subject: Python socket performance testing
References: <7l1ptf$ov9$1@nnrp1.deja.com>
Message-ID: <9sc7l7.9d.ln@final.math.uni-bonn.de>

Bruce Fletcher <befletch at my-deja.com> wrote:
> Something else interesting; time.clock() doesn't seem to work on Linux
> (RedHat 6.0) in either 1.5.1 or 1.5.2.  The time is always < 1, and it
> doesn't change.

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

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


  Byebye,
-- 
  Thomas Ackermann | Tel. +49-(0)228/631369|73-7773 | <tgm at math.uni-bonn.de>
             finger tgm at rhein.math.uni-bonn.de for public key
               GNU LINUX Python gtk pygtk MySQL FUDGE GURPS



From news at dorb.com  Tue Jun 22 13:29:58 1999
From: news at dorb.com (Darrell)
Date: Tue, 22 Jun 1999 13:29:58 -0400
Subject: Build the Python library into the binary
References: <376F9518.D3252CEE@interet.com>
Message-ID: <TMPb3.1886$U6.6789@newsr2.twcny.rr.com>

I thought these guys were working on these problems.
http://www.python.org/sigs/distutils-sig/

This also seems close to what you want.
http://x40.deja.com/getdoc.xp?AN=487778745&CONTEXT=930072342.1479737351&hitn
um=0
--Darrell





From akuchlin at mems-exchange.org  Thu Jun 10 08:54:44 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: Thu, 10 Jun 1999 08:54:44 -0400 (EDT)
Subject: How do I match the last occurence of a character?
In-Reply-To: <375F21B3.D23A5EA9@west.boeing.com>
References: <375F21B3.D23A5EA9@west.boeing.com>
Message-ID: <14175.46484.543974.29825@amarok.cnri.reston.va.us>

Jeff P. writes:
>I've been messing with this for a little while now, but I can't get this
>to work. How do I match the last occurence of '\' in a string? I want to
>remove the file name from a url (e.g. news/breakingnews/981001b.html -->
>news/breakingnews/ ).

    Rather than using all the machinery of regular expressions, use
the string module: string.rfind(S, Ssub) returns the last occurrence
of Ssub in S:

>>> import string
>>> s='a/b/c/d'
>>> string.rfind(s, '/')
5
>>> s[5:]
'/d'

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
The world is full of people whose notion of a satisfactory future is, in fact,
a return to an idealised past.
    -- Robertson Davies, _A Voice from the Attic_




From matomira at iname.com  Thu Jun 10 05:17:59 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Thu, 10 Jun 1999 11:17:59 +0200
Subject: Why is tcl broken?
Message-ID: <375F82C7.CF0F0477@iname.com>

Hello,

  I'm trying to collect a list of all the _current_ issues making tcl
a bad language choice. I'd like concrete examples, and not just
vague assertions that cannot be put porward honestly. I think
that as is usual in this newsgroups, there's going to be people that
actually know what they are talking about (and even better, experienced)

when referring to a language that is not their favorite.
  If you have references to articles of the sort, that will be useful,
too.
  Note that I've not listed comp.lang.tcl for increased productivity,
and to avoid starting a flame war. Hopefully some misconceptions
will be filtered down here, and _then_, the summary can be presented
in that newsgroup for their defense.

Thanks in advance,






From eugene.leitl at lrz.uni-muenchen.de  Wed Jun 23 16:47:47 1999
From: eugene.leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Wed, 23 Jun 1999 13:47:47 -0700 (PDT)
Subject: generating valid random SMILES, water ice modelling
Message-ID: <14193.17579.54805.701994@lrz.de>

Does anybody have (preferably Python, but I'd be thankful for
anything) code to generate random but valid SMILES? Thanks a bunch.

Also, I'd be thankful for any comments whether it is feasible
(accuracy and accessiblity of the time domain -- the planned system
will have 8 nodes of about ~500 MHz Alpha equivalent) to model water 
ice formation and melting using the Amber or CHARMm force fields. 

Pointers to papers on large-scale water/water ice modelling are 
extremely welcome.

TIA,

Eugene Leitl



From faria at ppgia.pucpr.br  Mon Jun 14 16:02:02 1999
From: faria at ppgia.pucpr.br (faria)
Date: Mon, 14 Jun 1999 17:02:02 -0300
Subject: Class Library
Message-ID: <000d01beb6a0$c703f320$656211c8@miro.lami.pucpr.br>

Hi there!

    I'm trying to build a class library called "figuras.py" but when I try
to import it ( from figuras import * ) in the python shell the methods to
draw a line for example don't work. Any clue?

Tanx all.


Henrique Faria
Desenvolvedor Multim?dia
LAMI - Laborat?rio de M?dias Interativas Puc-PR
faria at ppgia.pucpr.br

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Henrique Faria.vcf
Type: text/x-vcard
Size: 436 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/19990614/d90e25a9/attachment.vcf>

From python-list at teleo.net  Wed Jun 16 10:13:59 1999
From: python-list at teleo.net (Patrick Phalen)
Date: Wed, 16 Jun 1999 07:13:59 -0700
Subject: GC In Python: YAS (Yet Another Summary)
References: <Pine.SUN.3.95-heb-2.07.990616133606.13037B-100000@sunset.ma.huji.ac.il>
Message-ID: <9906160718471G.00557@quadra.teleo.net>

On Wed, 16 Jun 1999, Moshe Zadka wrote:
> Regarding the recend discussion here about garbage collection, I wrote 
> a small article.

Very excellent summary Moshe; but why didn't you write it two weeks ago,
so that the entire thread would have been unnecessary? ;)




From seanh at unforgettable.com  Mon Jun 14 12:01:38 1999
From: seanh at unforgettable.com (Sean Hummel)
Date: Mon, 14 Jun 1999 09:01:38 -0700
Subject: Embedding Questions on the Macintosh
Message-ID: <DH993.59$uF.6339@news.uswest.net>

Hi, I am writing a game which uses Python as the control language for a 
bunch of sprites.  I don't have any problems doing the embedding part.

What I want to know is if it is easy to package the entire Python Library,
much like the "macfreeze" module does?    I want to distributes as few
modifiable files as possible, and would prefer to distribute only two files:
the application, and the data file.

Thanks!



From graham at sloth.math.uga.edu  Sun Jun  6 20:07:52 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 7 Jun 1999 00:07:52 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7iqhkb$d7s@wiscnews.wiscnet.net> <7iu7ma$d33$1@cronkite.cc.uga.edu> <02hfonzsx2.fsf@rtp.ericsson.se>
Message-ID: <7jf2go$4t4$1@cronkite.cc.uga.edu>

Graham Matthews <graham at sloth.math.uga.edu> writes:
:> You are envangelising (a not uncommon response when
:> someone remotely criticises Python). Stop evangelising and
:> start considering the technical issues involve (read (!) 
:> other posts for what those issues are).
Kumar Balachandran (kumar*xspam*@*xspam*rtp.ericsson.se) wrote:
: Here is some useful evangelizing (methinks). One of the irritating things 
: about Python is the use of whitespace in syntax. Agreed, the code is
: readable without parentheses or braces, but why not have optional
: syntactic sugar such as

I am not sure what the point here is? I don't like whitespace in syntax
either but I also don't think it's worth trying to change (unlike something
really major like garbage collection). So I must have missed the point.

graham
-- 
                    One day in a nuclear age 
                  They may understand our rage
              We build machines that we can't control
              And bury the waste in a great big hole



From n at n.com  Thu Jun  3 16:05:16 1999
From: n at n.com (Jr. King)
Date: Thu, 3 Jun 1999 16:05:16 -0400
Subject: Where to find samples???
Message-ID: <7j6n4g$1jqq$1@rtpnews.raleigh.ibm.com>

Where can I find some good c/c++ integrated with python, besides the 2 or 3
that comes with the download of Python for windows???





From mgutierrez at dss.es  Fri Jun 18 05:30:53 1999
From: mgutierrez at dss.es (mgutierrez at dss.es)
Date: Fri, 18 Jun 1999 09:30:53 GMT
Subject: Newbie: Truth values (three-valued logic)
References: <Pine.OSF.4.05.9906161106470.295-100000@pinax.Mathematik.Uni-Bielefeld.DE> <000201beb946$b9c99e00$969e2299@tim>
Message-ID: <7kd3ke$d1n$1@nnrp1.deja.com>

In article <000201beb946$b9c99e00$969e2299 at tim>,
  "Tim Peters" <tim_one at email.msn.com> wrote:
> [Olaf Delgado]
> > ...
> > It came to my mind, though, that redefining negation alone will not
> > help. Next time I'll want to redefine 'and' and 'or'. So, unless
> > python 2 implements three-valued logic <hint, hint :)>, I'll have
> > to come up with a routine for myself to handle these truth values
> > consistently.
>
> Well, "and" and "or" do short-ciruit evaluation (i.e., they don't
evaluate
> their right-hand operand at all if the value of their left-hand
operand
> suffices to determine the result).  For that reason, they're more
properly
> viewed as control structures than operators, and so unlikely to get
"opened
> up" to user intervention.
>
> But in the spirit of overloading "~" instead of "not", consider
overloading
> "&" and "|" instead of "and" and "or".  Then you can do whatever you
want.
> If I were you I'd consider not overloading anything, and using ~ & |
> directly with a naming trick, like so:
>
> f, t, m = 0, -1, 42
> name = {0: 'false', -1: 'true', 42: 'maybe', ~42: 'maybe'}

"Not maybe" equals to "not".
Imagine this case:
Maybe, I'll go to the cinema tomorrow.
If you say, not "maybe, I'll go to the cinema tomorrow", then you are
saying not to go the cinema. Imagine:
Johan: Maybe, we could go to the cinema tomorrow.
Anthony: No
Johan : ok, let's go to the zoo, to see that python.

I'd propose another solution, if we currently have:

class a:
  def __init__():
    pass
  def __add__(c,d):
    pass

e=a(); f=a()
h = e + f

why not :
class a:
  ....
  def __or__(c,d):
    pass
  def __and__(e,f):
    pass

you could even include three logic behaviour in the operator,
so if c = true d= true then __or__ returns true if it has certain
probability.
if c = true d= false returns true if a random value is greater than
0.34 ,...
...

Regards
Manolo


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From CMCelata at lbl.gov  Mon Jun 28 16:22:06 1999
From: CMCelata at lbl.gov (Christine Celata)
Date: Mon, 28 Jun 1999 12:22:06 -0800
Subject: high-level "read" function?
Message-ID: <3777D967.484F0B6D@lbl.gov>

I am just starting to use Python.  It is foreign to me-- I'm a physicist
used to Fortran, not C.  I have succeeded in getting Python to read
program user input, but as far as I can see, in order to get it to read
numbers separated by commas from the terminal and store them as floating
point variables, I have to use a sequence of low-level call to:  read
the input as one character string, then strip off the carriage return,
then split the string into several strings using the commas as markers
for the end of the string, then convert the separated strings into
floating point numbers.  In the case where I am reading in strings
separated by commas the last step is omitted, but I have to add a step
to eliminate leading whitespace.  I am about to try to write a function
for myself that does all this, and a function to read integers, and a
function to read lists of strings, but--  Since lots of people must be
inputting numbers, isn't this stuff already written and lying around
somewhere?  I can't find it in the documentation for modules.   Thanks.




From graham at sloth.math.uga.edu  Mon Jun 14 15:58:40 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 14 Jun 1999 19:58:40 GMT
Subject: Solution to finalisation problem [Re: fork()]
References: <000001beb60e$1fd4f500$559e2299@tim>
Message-ID: <7k3mtg$65t$1@cronkite.cc.uga.edu>

[Greg Ewing]
> Indeed, which leads me to believe that the idea of giving objects
> a __del__ method is fundamentally flawed in the first place.
Tim Peters (tim_one at email.msn.com) wrote:
: Except it works so nice in "the usual" case (no cycle, no resurrection):
: everyone agrees on what should be done then, and it's remarkably
: surprise-free!

The problem with this argument Tim is that is is very difficult for the
programmer to recognise when they are in "the usual case", since circular
references are easy to create. Hence it is very difficult for the 
programmer to verify the correctness of their code.

Tim Peters (tim_one at email.msn.com) wrote:
: It's cycles that break its back, in Python and Java:

No cycles are not the problem. The problem is that finalisers are run in
"incomplete contexts". These incomplete contexts arise because implementations
of GC delete or partially delete collected objects before calling finalisers. 
That is a mistake. If you run *all* finalisers of *all* collected objects 
before you actually delete or partially delete any of those objects then 
all finalisers run in well defined contexts and there are no problems
with finalisers. That is the point of two pass collection.

graham
-- 
           As you grow up and leave the playground
       where you kissed your prince and found your frog
          Remember the jester that showed you tears
                   the script for tears



From hongkongmail at iname.com  Sun Jun 13 07:20:35 1999
From: hongkongmail at iname.com (HANG CHEONG)
Date: 13 Jun 1999 11:20:35 GMT
Subject: A new successful marketing:
Message-ID: <7k0463$8lb$11125@hfc.pacific.net.hk>

              HANG CHEONG INTERNATIONAL

A new successful marketing tactic to increase your sales :

    Do you always have problems to find channels to increase your sales?
The advertise in newspapers, magazines and doing some Direct Mailing, but
still can not achieve you goal, so why still dumping your money to a media
that could not help your sales and not using the most efficient and modern
way of selling stragies E-MAIL.Now many big companies are using this,
because it is economical, fast and efficient, and the other way is selling
by fax through the internets, and the result are remarkable.

  According to report pointed out that the ratio of internet selling is
1,000:1.5 that is 20,000,000 names x 0.0015 = 30,000.  These 30,000 clients
will buy from you and if your net profit is $50 that is to say your total
profit is:

           HK$50 x 30,000 = HK$1,500,000.  1.5 Million Dollars !!!

How to make your first 1 million in your life, now it is quite possible that
you can make your first 1 million within one month.  But the thing is ??Are 
you selling the right products, through the right media.??

***************************************************************************
A full set of sales tool with 20 million of client??s names will make you a SUPER SALES.  

A professional mail software (no need to go through any ISP Mail Server,making your computer
as the professional Mail Server, its super speed.  Easy to use.

* 500,000 E-mail addresses in Hong Kong
* 1,000,000 E-mail address in Taiwan
* 20,000,000 E-mail address in the World
 (latest revised on 1-6-1999.)

* A free revision one year (around 100,000 every month)

* Free of charge of 200,000 names of fax addresses in Hong Kong for 1999.

* Video Teaching  * Technical Support  * Software Update  * Data Update.

* Using CD-R dish and CD-AutoRun Menu.

(Only HK$680 for one full set) for unlimited use.

HK$680 is very small amount, you may not need it now but you will sure to use it when you 
have every thing ready.

***************************************************************************
If you need any E-mail address, call us any time!!
***************************************************************************
Don??t miss this chance, you may make lots of money with this new marketing technique.

*********************************************************************************************************
PLEASE TRANSFER THE MONEY $680(HK DOLLAR) TO: 

BANK NAME    : (HONG KONG) HANG SENG BANK. 
ACCOUNT NO   :  269-226981-882. 
ACCOUNT NAME :  WONG CHANG SANG. 

AFTER I GET THE MONEY WILL SEND THE E-Mail Marketing CD TO
YOU AT SOON(IN 7 DAY) 
**********************************************************************************************************

Booking Hotline : (852) 23488365 Mr Wong

2ND FL.FRONT BLOCK HING YIP HOUSE.24,SAI YEE ST,HK



From tim_one at email.msn.com  Mon Jun  7 21:24:51 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 7 Jun 1999 21:24:51 -0400
Subject: Indentation and grouping proposal
In-Reply-To: <375BCAD1.BE6E391B@erols.com>
Message-ID: <000b01beb14d$b502a480$aaa02299@tim>

[Tim, lying thru his teeth again, claims ...]
>> [R Wentworth, suggests "_(" and ")_" as alternative grouping constructs]
>> Note that this won't fly, since "_(" already has a meaning in
>> the grammar (call the function whose name is _).

[R Wentworth, the offended party, retaliates]
> Actually, the suggested tokens were _{ and }_, with braces rather than
> parentheses.  Admittedly braces and parenthesis can be hard to distinguish
> on a computer screen.  I'm well aware that "_(" is syntactically
> legal and meaningful, and therefore not a candidate.

Blame it on old eyes and a terrible email font -- both on my end.  OK, it's
"_{" and "}_" that will never make it in <wink>.

[notes that pindent.py & one of Donald's old patches ...]

> [do] not allow for 1-line programs.

Let's skip to:

> So, I guess having this "blessed" is unlikely. :(

Exceedingly, yes.  That was the point in trotting out (just a few of the)
previous attempts -- not that their details are in any way superior to your
proposal, but that this kind of suggestion is as old as Python and has never
made progress.

> (I do suspect that the "embedding" issue is more important now than it
> was five years ago, what with the rise of HTML templating systems and
> the like.  To my mind that makes the issue somewhat less frivolous.  Oh,
> well.)

Christian (Tismer) once invented a cute scheme, sticking a number of periods
at the start of each stmt equal to the stmt's intended indent level.  This
preserved his Python programs on a server that stripped leading whitespace,
at the cost of a trivial preprocessing step.  pyindent can also be used in a
similar fashion.  IIRC,

[
>> IIRC,

> "IIRC" ?

If I Remember/Recall Correctly
]

that flavor of server later stopped destroying whitespace, rendering the
scheme moot.  The Perlish urge to write a bunch of stuff on one line anyway
can be resisted, with practice <wink>.  And if codegen is properly
abstracted at the start (via open_block() & close_block() calls, etc),
generating appropriate whitespace is really no harder than generating
visible bracketing glyphs.

Granted that some people don't-- and never will --*like* whitespace, the
notion that there's a killer *technical* argument (on either side) is pretty
much hogwash.  Python's decision was inherited from ABC, where usability
studies with newbies concluded that grouping via indentation was easier to
learn and less likely to confuse.  That's where the colon after
block-opening stmts comes from too, despite that the grammar almost never
really *needs* it.

>> Steven Majewski christened the alternatives "frowning Guidos", in
>> honor of everyone's favorite some 5 years ago:
>>
>>     if whatever :(
>>     print "no whitespace!" ):

> Plausible, if inauspicious.

More than plausible, I believe it got fully implemented.  Wasn't able to
find the patches, though.

although-did-not-suffer-a-coronary-occlusion-from-trying-ly y'rs  - tim





From MHammond at skippinet.com.au  Fri Jun 18 19:44:12 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Sat, 19 Jun 1999 09:44:12 +1000
Subject: do anonymous pipes normally work on NT?
References: <4D0A23B3F74DD111ACCD00805F31D8100DB90DEF@RED-MSG-50>
Message-ID: <7kelhn$sal$1@m2.c2.telstra-mm.net.au>

Bill Tutt wrote in message
<4D0A23B3F74DD111ACCD00805F31D8100DB90DEF at RED-MSG-50>...
>> From: Georg Mischler [mailto:schorsch at schorsch.com]
>
>> Do I need to swig this stuff myself? or...
>
>Nope, they already come pre-swigged.
>Download, load into VC++, tweak away so that it can find all of the
>necessary .h/.libs it needs and build away.

FWIW, I am now recommending that you place the Python directories on your
MSVC global include and lib paths (using View/Options/Directories).  This
makes it easier to change versions or locations of Python - simply change
the global settings once, and all the project still work.

This is the way I ship my project files - with no embedded Python etc
paths - as I can't possibly predict your system.  So setting these paths
globally also means you can update the sources without re-tweaking the
projects.

If there are wrinkles in this that I can iron out, let me know...  [although
longer term it may make sense to move to DA's compile.py tool, then I can
just ship a setup.in and the projects files are built locally.]

Mark.

Mark.





From faria at ppgia.pucpr.br  Wed Jun 23 18:54:35 1999
From: faria at ppgia.pucpr.br (faria)
Date: Wed, 23 Jun 1999 19:54:35 -0300
Subject: [ANN]: Article about Python in Hungary
Message-ID: <001401bebdcb$5d9c7ce0$656211c8@miro.lami.pucpr.br>

Greetings from Brasil !

It's a very nice idea to tell the world about Python, maybe I will follow
the exemple and write a paper in Portuguese... I just don't know where to
publish it yet.

Tanx all


-----Mensagem original-----
De: Zsolt Hever <never at matavnet.hu>
Para: python-list at python.org <python-list at python.org>
Data: Quarta-feira, 23 de Junho de 1999 18:57
Assunto: [ANN]: Article about Python in Hungary


>Hi there,
>
>I saw  a discussion about "Is Python dying?" on the mailing list some
>months ago. That time I decided to write an article about Python in a
>paper.
>Now the article was published in the CHIP Magazine in Hungary yesterday.
>The title is The friendly snake.
>I made a HTML version of that <
>http://www.cab.u-szeged.hu/~h532851/cikk/index.htm >.
>It was written on the base of the FAQ, interviews, articles about Python
>and other languages (Perl, Rebol) and my experience.
>
>The content:
>========
>Shortly about Python
>What language is that?
>Installing of Python
>Documentation and examples
>Learning the language
>Writing and running of Python programs
>Life of Python
>Some interesting developments
>Summary
>
>Hungary is not a big country (10 million inhabitants), and not everybody is
>a programmer.
>I put a counter on the main page <
>http://www.cab.u-szeged.hu/~h532851/python.htm > to see the interest.
>
>I hope it helps Python to be known better.
>
>Zsolt
>
>
>
>
>
>--
>http://www.python.org/mailman/listinfo/python-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Henrique Faria.vcf
Type: text/x-vcard
Size: 436 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/19990623/a1d3f6a7/attachment.vcf>

From janssen at parc.xerox.com  Fri Jun 25 15:37:14 1999
From: janssen at parc.xerox.com (Bill Janssen)
Date: Fri, 25 Jun 1999 12:37:14 PDT
Subject: loading C++ into Python; static instance initialization?
Message-ID: <crQxdegB0KGW13wM0Y@holmes.parc.xerox.com>

Hello!

I'd like to develop a "C" extension module for Python in C++.  When I
try this, it almost seems to work.  The problem I'm seeing is that the
initializers for static instances in my C++ code are not being run. 
Does my module initialization code have to do something funky to make
this happen?  Or do I have to build the shared library in some special
way?  Any hints?

I'm using Sun's SparcWorks SunPro C++ 4.2, with Python 1.5, on Solaris
2.6, but could also use the same info for g++, if anyone knows it.

Thanks for any help!

Bill



From gmcm at hypernet.com  Wed Jun  2 00:01:41 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 1 Jun 1999 23:01:41 -0500
Subject: python extension problem (newbie)
In-Reply-To: <37548FA5.526A8E48@crystal.uwa.edu.au>
Message-ID: <1283826147-33549759@hypernet.com>

Douglas du Boulay writes:
> 
> I've been trying to write a python extension in C to interface to a
> Fortran program and a class wrapper for the extension using the
> following code loosely adapted from Mark Lutz's Programming Python 
> stackmodule example.
> 
> Unfortunately whereas his example works fine, mine goes into a
> recursive loop whenever an instance is instantiated.  It seems that
> the __getattr__  method of the wrapper class cannot find the getattr
> method of the extension library  and defaults to trying to call
> itself, infinitely recursively.

You didn't need an extension module to do this. It's all in the 
wrapper. While __getattr__ is only called if an appropriate attribute 
is not found, __setattr__ is *always* called when trying to set an 
attribute.

So the assignment to self._base in __init__ invokes __setattr__ 
recursively. A print statement (the only one you're missing, of 
course) will prove this (and make you go boom in slow motion).

If you use __setattr__, you need to do all your assignments (in 
methods of the class like this:
    self.__dict__['_base'] = start or p2fgraspint.Grasp()

(I do this about half the time I use __setattr__. Now you'll know 
what to look for, too).

> >>file oograsp.py
> 
> import p2fgraspint
> 
> class Grasp:
>     def __init__(self, start=None):
>         self._base = start or p2fgraspint.Grasp()
>         print "doing done"                  # which is never
>         executed
>     def __getattr__(self, name):
>         print "called getattr"              # executed  endlessly
>         return getattr( self._base , name)
>     def __setattr__(self, name, value):
>         return setattr(self._base, name, value)
> 
> >> file  http://www.crystal.uwa.edu.au/~ddb/pyextend/p2fgraspint.c
> 
> >> file  graspsub.py
> 
> #!/usr/bin/env python
> import oograsp
> x = oograsp.Grasp()
> 
> 
> 

- Gordon



From fdrake at cnri.reston.va.us  Mon Jun 21 10:45:42 1999
From: fdrake at cnri.reston.va.us (Fred L. Drake)
Date: Mon, 21 Jun 1999 14:45:42 GMT
Subject: grail browser failure
In-Reply-To: <376D4A9E.B9B90C3D@netzone.com>
References: <376D4A9E.B9B90C3D@netzone.com>
Message-ID: <14190.20502.944277.743491@weyr.cnri.reston.va.us>

Farrell Kenimer writes:
 > fails with something like isabs not available
 > I can import grail and after a try or two I can also import Browser
 > How would one try to run grail after it's imported?

  There's a patch for this on the Grail Web site.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From m01ymu00 at cwcom.net  Fri Jun  4 12:04:28 1999
From: m01ymu00 at cwcom.net (Daniel Faulkner)
Date: Fri, 04 Jun 1999 17:04:28 +0100
Subject: Problems with Install of LLNLDistribution11 (beginner)
References: <02k8ttz0ep.fsf@rtp.ericsson.se> <37553B49.24624DA7@arlut.utexas.edu>
Message-ID: <3757F90C.A19B6646@cwcom.net>

Or if it is to do with the directory try:
import sys
sys.path.insert(0, "filename")
import Numeric

Thomas Bryan wrote:

> Kumar Balachandran wrote:
> >
> > I am trying to install the Numeric and Graphics packages and I have
> > the follwing problems
> > 1. The numeric package was installed as
> >         $PREFIX/lib/python1.5/site-packages/numeric
> >    However, I am not able to import numeric using
> >         from numeric import *
>
> The package is called Numeric.
>
> Python 1.5.1 (#1, Mar 21 1999, 22:49:36)  [GCC egcs-2.91.66
> 19990314/Li on linux-i386
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> from Numeric import *
> >>> array
> <built-in function array>
> >>>
>
> -------------------------------------------
> Tom Bryan
> Applied Research Laboratories
> University of Texas at Austin


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/19990604/8c0523cb/attachment.html>

From guido at cnri.reston.va.us  Thu Jun 10 00:41:13 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 10 Jun 1999 00:41:13 -0400
Subject: fork()
References: <000601beb1c7$682c6220$559e2299@tim> <7jjghd$9il$1@cronkite.cc.uga.edu> <375D5826.23197692@inrialpes.fr> <375D6C60.FF9998DE@appliedbiometrics.com>
Message-ID: <5laeu8zmza.fsf@eric.cnri.reston.va.us>

Christian Tismer <tismer at appliedbiometrics.com> writes:

> If we have knowlegde of the internals of every Python structure,
> and if there were no hidden internal references to Python
> objects, then we could quite easily build a non-pessimistic
> GC, without the need to be hardware specific, scan the stack
> and so on.
> You know that I have an (early) Python with no stack.
> Would this help?

I don't see how: the roots you are looking for are all in the frame
objects anyway.  (There are a few variables in the C stack of
eval_code() but they point to objects that are also in the frame's list
of local variables or on the eval stack in the frame.)

[Christian, please don't call eval_code() "the interpreter" -- there
are already two or three other concepts that bear that name: the
interpreter state objects defined in pystate.[ch], the interactive
read-eval-print loop, and the entire Python executable.  We don't need
a fourth one!]

But here's another dumb idea based on the same premise (since this
thread won't die anyway, I might as well join in :-).

The idea is to add mark-and-sweep on top of refcounting, and only
invoke it on relatively rare occasions.  Perl does a mark-and-sweep at
certain moments to destroy cycles; I believe when a thread dies or
when an interpreter state object dies.  Something like this (plus a
way to force it) might be enough for Python, too.

Some hunches:

- Almost all cycles involve a dictionary, in practice (probably also
an instance, but those cycles involving an instance always also
involve a dictionary, since all references from instances go through
dictionaries).  Why does this matter?  Because now I only need to put
a mark bit in dioctionary objects, and I can maintain a list of all
dictionaries instead of a list of all objects, and apply
mark-and-sweep to it, without occurring overhead for all objects.

(You can create a cycle using only lists, but I bet that happens only
for demonstration purposes, as in ``a = []; a.append(a)''.  So I can
live with not freeing such cycles.)

- As roots should suffice the objects referenced by the threadstates
of all the interpreterstates.  (From there you can find all stack
frames, in particular.)  And perhaps a small number of global
variables at the C level.  Maybe we have to mark "freshly created"
dictionaries as live, to allow for C code that creates a dictionary,
then calls other parts of Python that will eventually fill the
dictionary and which might trigger a mark-and-sweep.  But other than
that I doubt that the C stack is needed to find roots.

- We can easily add a "mark" function to the list of function pointers
in the type object.  The mark() for a dictionary returns immediately
if the dictionary is already marked; otherwise, it marks it and then
calls the mark() function for all values in the dictionary.  The
mark() function for other container functions (tuples, lists,
instances, classes, frames, tracebacks, modules, function objects)
simply calls mark() for all objects contained in it.  Non-containers
(e.g. numbers, strings, most extension types) have a null mark()
function.  (I think code objects don't need a mark() function either,
they never contribute cycles -- unless one uses the bytecodehacks! :-)

(Hm, heavily shared containers, especially lists, might be traversed
many time by this algorithm, and that would be expensive if were
large.  Maybe it would pay to also have a mark bit in all lists.)

- The only tricky part is extension modules that might have additional
roots.  For example, _tkinter keeps references to callback functions
alive that might in some circumstances be methods of class instances
that have no other references to keep them alive.  We could invent a
convention whereby an extension (or any!) module can define a
__mark__() function which is called to provide additional roots.

(Extension types that participate in cycles should simply define a
mark() function in their type object.)


Why this compromise instead of "true GC" like Java?  Because one of
Python's main reasons of existence is its good integration with
external C code -- in a glue language, you can't set the rules for
management of all memory!

Then why not Boehm-Weiser?  It's been tried, and with disappointing
results.

May this thread spurn some code,

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



From jimbag at kw.igs.net  Tue Jun  1 15:00:56 1999
From: jimbag at kw.igs.net (JB)
Date: Tue, 01 Jun 1999 19:00:56 GMT
Subject: SWIG'in python question
Message-ID: <37542E8B.ABEAA531@kw.igs.net>

I've got a .so that I built with swig for python. Everything builds fine
but when I import i get this message from python "_mymod.so: undefined
symbol: __eh_pc". I wondered if anyone had seen this before and could
give me a clue. I think I'm missing a library somewhere. I'm building on
RedHat 5.2.

thanks
jb



From fredrik at pythonware.com  Thu Jun  3 14:37:20 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 3 Jun 1999 18:37:20 GMT
Subject: unable to access Tk from Python152/WinNT
References: <7j67lu$ibe$1@nnrp1.deja.com>
Message-ID: <02fa01beadf0$b3251220$f29b12c2@pythonware.com>

riftweaver at hotmail.com wrote:
> I am trying to run the tk interface from the Python 1.5.2 release on
> WindowsNT. I have installed Python with both a pre-existing Tcl/TK
> installation, and with the included Tcl/Tk installation.
> 
> When I couldn't start idle, I followed the troubleshooting selection
> from python.org. First Pyhon couldn't find a lib/Tcl8.0 directory; so I
> copied that from the Tcl install to Python/lib. Then it couldn't find a
> lib/Tk8.0 directory; again I copied from Tcl install to Python/lib.
> 
> Then I got a stack trace with an error about binding to a <MouseWheel>.
> 
> Any help or suggestions are appreciated. Thanks.

http://www.pythonware.com/people/fredrik/fyi/fyi02.htm

</F>




From tim_one at email.msn.com  Tue Jun  1 21:20:14 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 1 Jun 1999 21:20:14 -0400
Subject: Gzip.py
In-Reply-To: <14164.7764.546060.19420@bitdiddle.cnri.reston.va.us>
Message-ID: <000a01beac96$114ec780$d89e2299@tim>

> >>>>> "MC" == Matej Cepl <Cepl at fpm.cz> writes:
>
>   MC> Hi, I tried to play with gzip.py and it doesn't seem to
>   MC> work. Enclosed module produced output which Windows Commander as
>   MC> well as gzip refused to accept as correct gzip archive.

[Jeremy Hylton]
> I created my own file, fre-en_2.htm, and ran your script.  It worked
> correctly; gzip uncompressed it.  What exactly was your problem? Can
> you send me a copy of the .gz file produced?

The script opened the gzip file for output in text mode, and Matej was
running on Windows.  Replacing 'w' with 'wb' fixed it.

if-re.search('[Ww](in)?9[58]|indow)|NT|DOS')-and-
    re.search('(read|write)\(')-and-
    re.search("""['"][rw]['"]"""):-
    print-"try binary mode"-ly y'rs  -
    timbot's first and still most useful pattern





From DOUGS at oceanic.com  Tue Jun 22 21:06:25 1999
From: DOUGS at oceanic.com (Doug Stanfield)
Date: Wed, 23 Jun 1999 01:06:25 GMT
Subject: Help with String Manipulation
Message-ID: <5650A1190E4FD111BC7E0000F8034D26A0EE67@HUINA>

If you're new to Python it might seem like some magic in this:

Python 1.5.1 (#1, Sep  3 1998, 22:51:17)  [GCC 2.7.2.3] on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> testring = "This is an example"
>>> testlist = map (ord, testring)
>>> testlist
[84, 104, 105, 115, 32, 105, 115, 32, 97, 110, 32, 101, 120, 97, 109, 112,
108, 101]
>>> testlist.reverse()
>>> testlist
[101, 108, 112, 109, 97, 120, 101, 32, 110, 97, 32, 115, 105, 32, 115, 105,
104, 84]
>>> import array
>>> finalstring = array.array('b', testlist).tostring()
>>> finalstring
'elpmaxe na si sihT'
>>>

A string is immutable thus the conversion to a list.  The reverse() function
does so in place and the array magic (ref
http://www.python.org/doc/essays/list2str.html for other ways to do it)
gives back a string.  Function creation is left as an exercise...

-Doug-

> -----Original Message-----
> From: Samuel G. Williams [mailto:samw at masu.wwa.com]
> Sent: Tuesday, June 22, 1999 1:22 PM
> To: python-list at cwi.nl
> Subject: Help with String Manipulation
> 
> 
> I know this is a very stupid question and one that has been addressed
> within available documentation, however I need a quick hand. 
> I am in need
> of a routine to reverse the order of a string. I have tried 
> several things
> that have resulted in failures. It is really stupid, but I am just
> learning python. I can do this readily in C, perl, tcl, but 
> am at a loss.
> I would appreciate any help I can get.
> 
> Thanks in advance.
> 
> -- 
> 
> 
> Sam Williams  wb5yni					samurai at acm.org
> Senior IS Staff Specialist			        samw at wwa.com
> --------------------------------------------------------------
> -------------
> 
> 	Thinking of running your critical apps on NT?
> 
> 		Isn't there enough world suffering?
> 



From vlachid at otenet.gr  Sat Jun 26 01:56:58 1999
From: vlachid at otenet.gr (Vlachidis Costas)
Date: Sat, 26 Jun 1999 08:56:58 +0300
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <Pine.GSO.3.96.990625170056.28322B-100000@saga3.Stanford.EDU>
Message-ID: <37746BAA.DB304AB@otenet.gr>

Hi
Yes , me too I thinc that wxwindows are very good in programming very simple
GUIs and no some complex ones.,but they are not made to handle complex canvas
aporations in a easy way.
 This hapens because wxWindows abd wxPython are developed from few people like
Julian Smart, and some others, and then thinks stoped there.In contrast
,Tcl/Tk community is  yet developing their software.In the other hand
wxWindows is compact, fast and very effective for simple applications.
Costas


Jeffrey Chang wrote:

> On Fri, 25 Jun 1999, Mikael Lyngvig wrote:
> > I'd like some discussion on whether or not it is a good idea to
> > abandon TkInter, before it gets too widely used, and instead adopt
> > wxPython as the primary GUI framework for Python.
>
> Abandoning Tkinter seems a little bit extreme, since there are some
> applications for which its event model is well-suited.
>
> A few weeks ago I needed to build a viewer that displays binary trees
> whose branches have variable lengths (actually, phylogenetic trees for
> biological data).  Since the trees are large, I need to be able to slide
> it around the window.  In addition, I needed each node to be clickable so
> that the user can query for more data.
>
> Tkinter is very good for that type of thing.  All I needed to do was
> create a Canvas and then for each node, create an Oval with a callback to
> a function that handles node clicks.  Tk keeps track of the coordinates of
> each Item so that I don't have to.  Also, whenever I wanted to move the
> tree, all I had to do was:
> for widget in self._canvas.find_all():
>     self._canvas.move(widget, xoff, yoff)
>
> Of course, Tk incurrs an overhead by making everything you draw on the
> canvas its own object, but for this case, it made thigns a lot easier.
>
> Although I haven't used wxWindows before, but the documentation suggests
> that it would be a lot more difficult for this kind of application.  In
> wxWindows, I would need to draw each of my nodes into a window using a
> device context.  Then, I'd have to write a function that would handle
> mouse clicks in the window.  That function would need to get the
> coordinates and then figure out whether it intersects with any of my
> nodes, which Tk would've handled for me.  In addition, whenever I wanted
> to move the tree, I'd have to erase the window and redraw each node.  If I
> wanted that optimized, I'd need to write routines that would only redraw
> the uncovered parts of the tree, which means figuring out which nodes are
> newly revealed.
>
> From what other people have said about wxPython, it seems like it may be
> appropriate for other types of applications.  However, I'd hate to see
> Tkinter dumped because it's not appropriate for everything!
>
> Jeff
-------------- next part --------------
A non-text attachment was scrubbed...
Name: vlachid.vcf
Type: text/x-vcard
Size: 366 bytes
Desc: Card for Vlachidis Costas
URL: <http://mail.python.org/pipermail/python-list/attachments/19990626/c3ccf06b/attachment.vcf>

From soundhack at geocities.com  Mon Jun  7 22:08:41 1999
From: soundhack at geocities.com (Robert)
Date: Mon, 7 Jun 1999 22:08:41 -0400
Subject: vbCr in Pythoncom?
Message-ID: <7jhtml$hlu@senator-bedfellow.MIT.EDU>

Hi,
 I'm playing around with pythoncom and MS Word. There's a useful book by
O'Reilly, 'Learning Word Programming' that I'm using. In a section on how
to enter text, it mentions using the constant 'vbCr' to insert Word
paragraph
marks. Can someone tell me how i can get pythoncom/makepy to generate
this and other vb* constants? the makepy generated interface for
Word.Application.8
does not seem to have any.

Thanks!
Robert






From c.evans at clear.net.nz  Sat Jun  5 20:33:46 1999
From: c.evans at clear.net.nz (Carey Evans)
Date: 06 Jun 1999 12:33:46 +1200
Subject: smtp module
References: <0gf63.28$Vv4.277@news.megsinet.net> <Idi63.2035$nn.613205@news.shore.net>
Message-ID: <87g1466uc5.fsf@psyche.evansnet>

"Michael P. Reilly" <arcege at shore.net> writes:

> Philip C. <pjc_2 at geocities.com> wrote:
> : I have pretty much figured out how to use the module but I can't seem
> : set the subject or the from of the bcc etc....

[snip]

> The protocol for SMTP requires that the headers are included.

Actually, Bcc: shouldn't be, even if some mail servers seem to work as
you expect.  If you want to Bcc: to someone, add their address to the
envelope recipients, but not the message:

>>> fromaddr = 'c.evans at clear.net.nz'
>>> toaddr = 'eggs at example.com'
>>> msg = '''From: Carey Evans <c.evans at clear.net.nz>
To: Spam, spam, spam, eggs and spam <eggs at example.com>
Subject: Something completely different

Blah, blah, blah...'''
>>> s = smtplib.SMTP('localhost')
>>> s.sendmail(fromaddr, [fromaddr, toaddr], msg)

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

	     "I'm not a god.  I've just been misquoted."



From r14382 at email.sps.mot.com  Fri Jun 25 17:19:58 1999
From: r14382 at email.sps.mot.com (Matthew Cincera)
Date: Fri, 25 Jun 1999 14:19:58 -0700
Subject: Tkinter & socket question
References: <19990618215411.23898.qmail@nwcst277.netaddress.usa.net> <sogic333t.fsf@foreman.ac.rwth-aachen.de> <cOVa3.642$7X1.149581@news.shore.net>
Message-ID: <3773F27E.60AD622F@email.sps.mot.com>

Hello,

I have had a lot of experience with socket/Tkinter over the years
and have one simple recommendation.  If you are going to use a
lot of event based updates of UI's based on socket interactions,
use the tcl sockets instantiated in a tcl interpreter inside
of your GUI program.  The createfilehandler is not supported
on Windows any longer (starting with tcl/tk 8.0).  Using after()
and polling are alternatives, but using the native tcl sockets
will put everything in the tcl/tk event loop (this is my humble
opinion, so take it for what it is worth).  I have used native
python sockets but the limitation is really on the tcl/tk side,
not the python side.  Another bonus is tcl handles more of
the socket housecleaning stuff.

Here's some code snips to do some simple socket serving inside
of Python using the tcl sockets layer.  This example is not fully
functional but hopefully this gives you ideas.  It is based on
prior postings in this newsgroup.

Regards,

Matt
--------------------------------------------------------------
import Tkinter

class _Callback:
 """ A foundation class used for representing callbacks.
  Methods added by this class:

  __call__()  - Apply function for the callback

 """

     # =========================================================================

 def __init__(self, callback):
  self.callback = callback
  self.name     = "cb%d" % id(self)

     # =========================================================================

 def __call__(self, *args):
  return apply(self.callback, args)

class Server:

  def __init__(self):

    self.acceptCb = _Callback(self.acceptRequest)
    self.connection.tk.createcommand(self.acceptCb.name, self.acceptCb)
    self.handleCb = _Callback(self.handleRequest)
    self.connection.tk.createcommand(self.handleCb.name, self.handleCb)

    self.connection = Tkinter.Tk()
    self.connection.withdraw()


  #=================================================================================
  def acceptRequest(self, sock, addr, port):
    # Accept the connection request
    self.connection.tk.eval('fconfigure ' + sock + ' -buffering none')
    self.connection.tk.eval('fconfigure ' + sock + ' -translation binary')
    self.connection.tk.eval('fileevent ' + sock + ' readable [list ' + self.handleCb.name + ' '
+ sock + ' ' + addr + ' ' + port + ']')

  #=================================================================================
  def handleRequest(self, sock, addr, port):
    try:
      self.processMessage( sock, addr )
    except _ClientShutdown:
      self.handleClientShutdown( sock, addr )
    except:
      print 'error'

....
  def serveForever(self):
    self.connection.tk.mainloop()


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

Michael P. Reilly wrote:

> Gregory A. Landrum <landrum at foreman.ac.rwth-aachen.de> wrote:
>
> :>2) I need to poll the connection periodically, say, every second. This would
> :>not have been a problem in console mode... I don't know how to do this in a
> :>Tkinter environment though. So, how can I do something every second (or every
> :>n milliseconds, or whatever the time span may be)? Are threads the way to go?
> :>I haven't used them before.
>
> : Wow, a question I can answer.... amazing.
> : Obligatory-Mandatory-Necessary-Essential Disclaimer:  I bear no
> : resemblance whatsoever to a python wizard, I just happen to have done
> : something like what you want to do.  There's almost definitely a
> : better way of doing this, but my method seems to work.
>
> : I assume that you have already opened a nonblocking socket connection,
> : and that that is what you are polling.  Your main Tkinter window has a
> : method called "after" which can be used to do the polling.
> : Documentation can be found at:
> : http://www.pythonware.com/library/tkinter/introduction/x8550-alarm-handlers-and-other.htm
>
> : An example, in my stupid little networked python game, I have the
> : following polling function:
> :     def ListenForMessages(self):
> :         msg = None
> :         theConn = None
> :
> :  # Check to see if we are the client or the server
> :         if self.serverActive == 1:
> :             if self.clientConn is None:
> :                 print "No Client!"
> :                 return
> :             theConn = self.clientConn
> :         elif self.clientActive == 1:
> :             theConn = self.socket
>
> :         if theConn is None: return
>
> :  # see if there is anything there to read
> :         try:
> :             msg = theConn.recv(1024)
> :      # Got it!  Dispatch the message.
> :             self.Dispatch(self.message.Type(msg),msg)
> :         except error:
> :      # nope... no one loves us.
> :             pass
>
> :  # re-register the after callback, with a 200msec delay
> :         self.listenID = self.Game.rootWin.after(200,self.ListenForMessages)
>
> : The last line is important, because the after callback is removed
> : after being called.  So it needs to be re-registered with Tk.
>
> : The callback is first registered immediately upon establishing the
> : socket connection.
>
> You can also deal directly with Tcl/Tk's filehandler:
>
>   from Tkinter import tkinter
>   self.root.tk.createfilehandler(socket, tkinter.READABLE,
>     self.handle_msg
>   )
>   ...
>   self.root.tk.deletefilehandler(socket)
>
> The mainloop() method handles the rest for you.
>
>   -Arcege




From greg.ewing at compaq.com  Sun Jun 27 19:17:01 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Mon, 28 Jun 1999 11:17:01 +1200
Subject: Why exceptions shouldn't be used for flow control [Re: YAS to the   
 "Reading line-by-line" Problem]
References: <Pine.SUN.3.95-heb-2.07.990622183356.3454A-100000@sunset.ma.huji.ac.il> <376FDF70.4756AD38@mindspring.com> <slrn7n0chl.gh6.wtanksle@dolphin.openprojects.net> <377173F7.36BCFF6B@compaq.com> <slrn7n31k4.idj.wtanksle@dolphin.openprojects.net> <3772AED9.D9E1110@compaq.com> <slrn7na4hq.3vt.wtanksle@dolphin.openprojects.net>
Message-ID: <3776B0ED.C1ABBEC1@compaq.com>

William Tanksley wrote:
> 
> while not f.eof():
>   line = f.readline()
>   # ...

I'd love to be able to write reading loops that way.
Unfortunately, it's a fact of life that some of the
things read() has to deal with are unable to detect
eof without trying to read something. I think the
existing definition of read() and friends is the best 
that can be done in those circumstances.

> It's been a while since I've used read, so I
> don't recall what it actually returns

It returns an empty string if it can't read more
than 0 characters, same as readline().

> in a function designed to get n!=0
> more characters, an EOF really is an error.  Unfortunately for me, that's
> not the definition of 'read', and never will be.

Perhaps there should be a 2-parameter version of read:

   read(min, max)

which would raise an exception if it couldn't read at
least min characters. Setting min=0 would give the
current behaviour, and setting min=max would allow
reading a fixed-length record without having to check
for errors. Then everyone would be happy!

Greg



From phd at emerald.netskate.ru  Tue Jun 15 06:37:48 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Tue, 15 Jun 1999 14:37:48 +0400 (MSD)
Subject: Time display without colon
In-Reply-To: <3766234B.3B347856@mediamanager.com.sg>
Message-ID: <Pine.LNX.4.04.9906151435360.30587-100000@emerald.netskate.ru>

Hi!

On Tue, 15 Jun 1999, John Francis wrote:
> I want to display time without colon separting them
> I am using strftime(format,tuple) function.
> 
> But the display is like this HH:MM:SS

strftime("%H-%M-%S", tuple)

> But I am trying to put it in a table form by extracting from the text
> file
> 
> TIME: HH:MM:SS
> 
> My extraction of each line is based on the colon splitting the field
> name and the value in the field, but my script goes heywire when I have
> the above format (TIME:HH:MM:SS) is there a way to extract this or to
> display TIME in different format.

string.split("TIME: 12:01:47", ":", 1)

> Thanking you
> 
> Regards
> John

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From garry at sage.att.com  Wed Jun  9 14:04:22 1999
From: garry at sage.att.com (Garrett G. Hodgson)
Date: Wed, 9 Jun 1999 18:04:22 GMT
Subject: Designing Large Systems with Python
References: <m37lqz0yoa.fsf@solo.david-steuber.com> <372599D6.C156C996@pop.vet.uu.nl> <7g4a3l$atk$1@news.worldonline.nl> <m3vhehxjhr.fsf@solo.david-steuber.com> <7g99pj$b1$1@news.worldonline.nl> <slrn7igcdg.is.davecook@sputnik.escnd1.sdca.home.com> <7g9eln$4eq$1@news.worldonline.nl> <j3udj7.5a.ln@final.math.uni-bonn.de>
Message-ID: <375EACA6.CC631742@sage.att.com>

Thomas Ackermann wrote:
> 
> Ilja Heitlager <news at helen.demon.nl> wrote:
> > I would love too, but Windows doesn't leave any space on my HD ;-(
> 
> Reorganize your disk (scandisk or ?) then use "fips" to make this
> partition small and create a new, free partion from the saved space, then
> install LINUX ...       ;-)

i actually found that i saved even more disk space by just reformatting
the disk and installing linux.

-- 
Garry Hodgson			comes a time
garry at sage.att.com		when the blind man
Software Innovation Services	takes your hand, says,
AT&T Labs			"don't you see?"



From bernhard at alpha1.csd.uwm.edu  Mon Jun 28 13:41:22 1999
From: bernhard at alpha1.csd.uwm.edu (Bernhard Reiter)
Date: 28 Jun 1999 17:41:22 GMT
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0o3q.toe.bernhard@alpha1.csd.uwm.edu> <37715134.399AFB2D@compaq.com>
Message-ID: <slrn7nfcu2.6su.bernhard@alpha1.csd.uwm.edu>

On Thu, 24 Jun 1999 09:27:16 +1200, Greg Ewing <greg.ewing at compaq.com> wrote:
>Bernhard Reiter wrote:
>> += just saves some space and typing time.
>It's not quite as simple as that, because
>
>  a[i] = a[i] + expr
>and
>  a[i] += expr
>
>would be semantically different if the former
>did the index lookup only once. 
True.
I guess that using multiple evaluations and their side effects
are a not a good thing anyway. :)
And the python interpreter might optimise the former... >;-)
	Bernhard


-- 
Research Assistant, Geog Dept UM-Milwaukee, USA.         (www.uwm.edu/~bernhard)
Funding/Promoting Association for a Free Informational Infrastructure (ffii.org)



From tim_one at email.msn.com  Tue Jun  8 11:56:01 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 8 Jun 1999 11:56:01 -0400
Subject: fork()
In-Reply-To: <m3aeuac0en.fsf@pcrm.win.tue.nl>
Message-ID: <000601beb1c7$682c6220$559e2299@tim>

[Bernhard Herzog]
> I fear that in reality the memory is copied after all. If they use
> the dictionary in any way they are modifying refcounts of at least
> some of the objects in the dictionary. From the OS's point of view
> this counts as modifying the memory.

[Hrvoje Niksic]
> Eek.  You're right.  Refcounting truly sucks.  :-(

OTOH, in a straightfoward mark-and-sweep GC scheme there will at least be a
"mark bit" in each object header.  So under that, or under any form of
compacting GC, the *entire live object space* will get copied, regardless of
whether *you* reference anything in it or not.  So if RC truly sucks, what
does that make GC <wink>?

easy-answers-don't-always-work-ly y'rs  - tim





From rhww at erols.com  Fri Jun 11 10:17:23 1999
From: rhww at erols.com (R Wentworth)
Date: Fri, 11 Jun 1999 10:17:23 -0400
Subject: High performance Python on NT
References: <000301beb3b7$4f973810$6eba0ac8@kuarajy.infosys.com.ar> <37609429.65A7E802@palladion.com>
Message-ID: <37611A73.348C6C5C@erols.com>

Tres Seaver wrote:
> Run, don't walk, to http://www.zope.org.  You'll think you've died and gone to
> heaven.

I looked at Zope a little while ago, and it did seem rather exciting.
Depends a little bit on your constraints, though, whether it's the
right solution.  I had a project that needed to be hosted via a commercial
hosting provider, and nobody was willing to run Zope unless I rented
a dedicated server.  (Actually, not really nobody, but with web hosting
providers it can be devastatingly bad to become dependent on a single
hosting service unless you've got a lot of prior positive experience
with them.)

I chose instead to go with a solution using the Java "servlet" interface
(instead of CGI), and writing JPython (http://www.jpython.org/) servlets 
instead of Java servlets.  Don't know if this could be considered a
"high-performance" solution; it avoids the process creation and
DB connection overhead of CGI solutions, but JPython is inherently
slower than Python.  So, depends what sort of performance matters.
I expect JPython to be fine as long as I'm not too compute-intensive;
we'll see.  In this model, if I run into performance bottlenecks
I can improve things by selectively rewriting code into Java.

(I did have a few other concerns about Zope: I wanted to do some
non-hierarchical design, and found that the limited documentation
made attempting such a thing difficult.  Still, I hope to have
an excuse to learn/use Zope someday in the future.)

Robert Wentworth



From wtanksle at dolphin.openprojects.net  Thu Jun  3 23:02:48 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Fri, 04 Jun 1999 03:02:48 GMT
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net> <aahzFCpHw8.Gxt@netcom.com> <7j4e2u$8rh$1@m2.c2.telstra-mm.net.au> <aahzFCq5CC.KFK@netcom.com>
Message-ID: <slrn7legem.932.wtanksle@dolphin.openprojects.net>

On Thu, 3 Jun 1999 00:00:11 GMT, Aahz Maruch wrote:
>Mark Hammond <MHammond at skippinet.com.au> wrote:

>>Not to me.  If I make structural changes to C or other programs, I always
>>re-indent the reorganised code correctly, and I would bet that you do too.
>>In almost all languages block indenting is _very_ important to the human
>>reader.

>I think you misunderstood what I meant by "structural" (and I wasn't
>particularly clear -- my bad): if I need to add a quick "if" statement
>during debugging, I don't like having to make sure I clean up the block
>structure formatting.  With just about any other language, I don't have
>to.  I do, as I think I did make clear, clean up the formatting after
>the debugging.

I see what you're saying -- it's the difference between adding two braces
and adjusting spacing for a mess of things.  Yes, that is harder.

All I can do is appeal to two things: first, I don't debug messes, so I
don't allow my debug code to become a mess (in other words, I always fix
the indentation whether the language needs it or not); and second, I use
editors which handle indentation for me (Jed and IDLE).

>                      --- Aahz (@netcom.com)

-- 
-William "Billy" Tanksley
Utinam logica falsa tuam philosophiam totam suffodiant!
   :-: May faulty logic undermine your entire philosophy!



From gerrit.holl at pobox.com  Thu Jun 17 09:46:43 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Thu, 17 Jun 1999 15:46:43 +0200
Subject: Starting a project: any hints?
Message-ID: <19990617154643.A1680@optiplex.palga.uucp>

Hello python-aholics,

I'm a 13-years-old-boy and I've about half a year programming experience. And
now I'm going to start a serious project, called "discoverb" (name can still
change). Below is the discription. I just want to know if anyone has hints for
a half-newbie.

Discoverb is an interactive program meant to test learned words. It reads the
words from a file. Discoverb does also supports comments (#) in the file, one
word with more words in the foreign language (all words needed), one word with
more possible words in the foreign language (one of the words needed). If there
are more words, you'll be asked. All the wrong words can be written to a file,
which can be tested (and hand-edited) again.

Design:
discoverb will exist of 3 classes: Parser, Writer and Tester. Parser reads the
words from a file and make a dictionairy, Writer writes the wrong words (see
above) to a file, and Tester asks the user for words and tests them, and will
also do the status update.

Why discoverb?
When I was learning latin words, I came on this idea. Discere is the latin word
for "learn", and disco means "I learn". Verb does not only means verb, but also
word. So it means: I learn word. Beside that, it can also be interpreted as
discover-b. I still have to get a nice story for that b... :-)

Is there anyone with about half a year first-programming experience who've had
a project but encounterd problems?

The project webpage will be at nl.linux.org/~gerrit/discoverb/ (today, I hope).

regards,
Gerrit.

--
The Dutch Linuxgames website.		De Nederlandse Linuxgames pagina.
Everything about games on Linux.	Alles over spelletjes onder Linux.
Site address:				http://linuxgames.nl.linux.org
Personal homepage:			http://nl.linux.org/~gerrit/



From wtanksle at dolphin.openprojects.net  Tue Jun 29 19:27:42 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Tue, 29 Jun 1999 23:27:42 GMT
Subject: [Python|Crypto] on handhelds (was Re: Thanks for Python...)
References: <376ead64@anonymous.newsfeeds.com> <VrDQRddcbbs9-pn2-QdftOXYk6Hdd@Fortress> <37716901.43CB604@lockstar.com> <slrn7n2gkr.i3k.wtanksle@dolphin.openprojects.net> <37727194.D9CD6B7@lockstar.com> <slrn7n5j99.k00.wtanksle@dolphin.openprojects.net> <37790460.47FAD8AF@lockstar.com>
Message-ID: <slrn7niljd.3rr.wtanksle@dolphin.openprojects.net>

On 29 Jun 1999 07:35:42 PDT, Mordy Ovits wrote:
>William Tanksley wrote:

>To each his own. :-)  Don't you wish all flame were required to end like that
>(i.e. before they started?)

Yes, it is kinda nice.  Can we please argue about something now?  This IS
Usenet, after all -- I can't pass the Turing test on other topics.

>> >> I used Quartus on my Crypto exam to solve a modular equation after I
>> >> forgot how to solve modular equations (I had it use brute force).

>> >Wow!  I wish I had had my E100 when I was in college.  I just realized how
>> >awesome that would have been. :-)

>> Actually, I sometimes miss my old OmniGo, which was a pocket computer, in
>> spite of being slower than the Pilot for some tasks.  I really needed the
>> glorified organizer (and had been fooled into thinking that the OmniGo was
>> one), but the OmniGo had some abilties that the Pilot doesn't.

>I'm not familiar with the OmniGo.  Got a link for me?

Sure.  It's discontinued, because HP got touched by Microsoft, but it was
a decent little tool, just a little slow to handle things that an
organizer needs to do quickly (in spite of the fact that it's faster than
a Pilot in actual speed).

http://www.geoworks.com/devices/omnigo/ seems to be the best thing I can
find -- there was a really good user site, but I lost the link and Google
can't find it (there are too many Omnigo mailing lists).

>> >The SECOND time I saw that sig was in ESR's mail.  The first was mine.  I doubt
>> >ESR copied mine, but who knows? :-)

>> You mean you wrote it?  If so, there seems to have crept into it a bug --
>> one of the implicit unpackings don't work.

>No, I didn't write it.  It was written by the all-talented AMK.  I did write a
>nifty version of Bruuce Schneier's Solitaire algorithm, though.  check out:
>http://www.counterpane.com/solitaire.html

I've read that -- I always read Counterpane's site :-).

>Mordy

-- 
-William "Billy" Tanksley



From ddb at crystal.uwa.edu.au  Tue Jun  1 21:57:57 1999
From: ddb at crystal.uwa.edu.au (Douglas du Boulay)
Date: Wed, 02 Jun 1999 09:57:57 +0800
Subject: python extension problem (newbie)
Message-ID: <37548FA5.526A8E48@crystal.uwa.edu.au>

Hi ,

I've been trying to write a python extension in C to interface to a
Fortran program
and a class wrapper for the extension using the following code loosely
adapted from Mark Lutz's Programming Python  stackmodule example.

Unfortunately whereas his example works fine, mine goes into a recursive
loop whenever an
instance is instantiated.  It seems that the __getattr__  method of the
wrapper class
cannot find the getattr method of the extension library  and defaults to
trying to call itself, infinitely
recursively.

Having spent about 8 hours trying to figure out why, I still have no
idea.
Help!
Thanks

Doug

>>file oograsp.py

import p2fgraspint

class Grasp:
    def __init__(self, start=None):
        self._base = start or p2fgraspint.Grasp()
        print "doing done"                  # which is never executed
    def __getattr__(self, name):
        print "called getattr"              # executed  endlessly
        return getattr( self._base , name)
    def __setattr__(self, name, value):
        return setattr(self._base, name, value)

>> file  http://www.crystal.uwa.edu.au/~ddb/pyextend/p2fgraspint.c

>> file  graspsub.py

#!/usr/bin/env python
import oograsp
x = oograsp.Grasp()


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/19990602/36571375/attachment.html>

From tbryan at arlut.utexas.edu  Sat Jun  5 18:28:17 1999
From: tbryan at arlut.utexas.edu (Thomas A. Bryan)
Date: Sat, 05 Jun 1999 18:28:17 -0400
Subject: Looking for for some ideas for argument processing
References: <7jcdhc$htd$1@nnrp1.deja.com>
Message-ID: <3759A481.135D19CF@arlut.utexas.edu>

cmedcoff at my-deja.com wrote:
> 
> I'm writing a script with is going to make heavy use of script
> arguments.  I'd like to hear some ideas on some ways to support this.

I usually just use Python's getopt or DPyGetOpt.

The first is in the standard library.

The second is modeled after Perl's GetOpt::Long, which is very  
flexible and supports many options.  DPyGetOpt is available at
python.org:
http://www.python.org/ftp/python/contrib/All/DPyGetOpt.py
http://www.python.org/ftp/python/contrib/All/DPyGetOpt.README



From guido at cnri.reston.va.us  Thu Jun 10 16:31:24 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 10 Jun 1999 16:31:24 -0400
Subject: A small module question
References: <375ECD2A.71AA9751@lysator.liu.se>
Message-ID: <5lwvxbyezn.fsf@eric.cnri.reston.va.us>

redhog at lysator.liu.se writes:

> I am having a problem, which is reducible to the following, which I
> would be happy if someone could explain to me:
> 
> [redhog at k35 Python]$ ls -R test/
> test:
> __init__.py  a.py  b.py
> [redhog at k35 Python]$ cat test/__init__.py
> __all__ = ["a", "b"]
> [redhog at k35 Python]$ cat test/a.py
> from test import b
> [redhog at k35 Python]$ cat test/b.py
> from test import a
> [redhog at k35 Python]$ python
> Python 1.5.1 (#1, Mar 21 1999, 22:49:36)  [GCC egcs-2.91.66 19990314/Li
> on linux-i386
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> from test import a
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "test/a.py", line 1, in ?
>     from test import b
>   File "test/b.py", line 1, in ?
>     from test import a
> ImportError: cannot import name a
> >>>

The "from test import a" statement ends with referencing test.a while
a isn't completely imported.  The key "test.a" exists in sys.modules,
and "import test.a" would work; however the attribute a isn't added to
the module test until after the import of a is completed.

So the solution is to use "import a" and "import b" instead -- inside
the package test, this will refer to modules test.a and test.b anyway, 
if they exist in the package.

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



From matomira at acm.org  Mon Jun 14 12:19:21 1999
From: matomira at acm.org (Fernando D. Mato Mira)
Date: Mon, 14 Jun 1999 18:19:21 +0200
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com>
	 <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com>
	 <375FDCA7.62FF@mailserver.hursley.ibm.com>
	 <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com>
	 <3762988E.C94AA252@iname.com>
	 <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu>
	 <3763C746.5CA8FE78@iname.com> <m3u2sby4gf.fsf@soma.andreas.org>
	 <87pv2zjhhl.fsf@home.ivm.de> <3764F3FE.F7CBE6BA@acm.org> <0Sgs3BA5FPZ3Ew6E@jessikat.demon.co.uk>
Message-ID: <37652B89.6ED1E8CA@acm.org>

Robin Becker wrote:

> postfix, prefix, infix etc etc are isosemantic

Yup. But the deal is to be able to write a parser in an hour (say, in
1950something), and
be setup until the next comet hits the Earth.

Not to mention how invaluable it is not to waste the mandatory compiler
course with the unnecessary half of the `Dragon book'.


And when infix goes wild, a LOT of dollars go down the drain (the one
with a bug-free C++ frontend please raise your hand!).

--
Fernando D. Mato Mira
Real-Time SW Eng & Networking
Advanced Systems Engineering Division
CSEM
Jaquet-Droz 1                   email: matomira AT acm DOT org
CH-2007 Neuchatel                 tel:       +41 (32) 720-5157
Switzerland                       FAX:       +41 (32) 720-5720

www.csem.ch      www.vrai.com     ligwww.epfl.ch/matomira.html






From dsheets at wam.umd.edu  Sat Jun 26 21:33:17 1999
From: dsheets at wam.umd.edu (David Sheets)
Date: Sat, 26 Jun 1999 21:33:17 -0400
Subject: I am having difficulties in embedding python in a win32 application...
Message-ID: <7l3utr$h9s$1@autumn.news.rcn.net>

    I have been trying to create a program that does some math and graphics
using win32 code, and then uses Python to do a lot of GUI stuff. I'd like to
be able (eventually) to send a  message to the main windows code whenever
certain actions take place in the Python GUI. I'd also like to be able to
have Python create a window, and pass a handle to this window back to the
main windows code to have windows draw into this window. But, really, all of
these problems are secondary...
    The main problem I am having right now is just getting the dang Python
code to run in the first place! :)  I must admit, that my Python skills are
poor at best. But, I have created a python program, that I can run from the
command line (ie. python.exe test.py) which creates a little window, lets me
press some buttons, and then close it. I've been trying to have my windows
code call this python code, but for some reason it is not working. The
module loads fine, the python function loads fine, but then when I try to
call the function, nothing happens. A snipet of code follows:

void Python_thread(PVOID pvoid)
{
    volatile P_THREAD_PARAMS pparams; // pointer to the parameters passed
into the thread
    PyObject *p_module,    // python objects used to start python
        *p_function,
        *p_result;
    int status, result;
    char *c_str;      // result returned when python exits

    pparams = (P_THREAD_PARAMS) pvoid;

    Py_Initialize();

    p_module = PyImport_ImportModule("P_Menu");
    if(p_module == NULL)
        _RPT0(_CRT_ERROR, "\nPython module P_Menu was not loaded.\n");

    p_dictionary = PyModule_GetDict(p_module);
    if(p_dictionary == NULL)
        _RPT0(_CRT_ERROR, "\nPython dictionary was not loaded.\n");

    p_function = PyObject_GetAttrString(p_module, "start_menu");
    if(p_function == NULL)
        _RPT0(_CRT_ERROR, "\nPython function start_menu was not loaded.\n");

    // this calls the function start_menu in the P_Menu module
    //_RPT0(_CRT_ERROR, "\n Python start_menu is now being called.\n\n");
    p_result = PyEval_CallObject(p_function, NULL);
    // p_result = PyRun_String("P_Menu.start_menu()", eval_input,
p_dictionary, p_dictionary);

    // etc....
    _endthread();
}

    What am I doing wrong here? I don't get any errors loading the module or
the function, but then when I call PyEval_CallObject, nothing happens. The
python function I am calling does not take any arguments, it's just creating
a simple window... Here is the Python code:

def kill_python():
 print 'kill python'

def kill_draw():
 print 'kill draw'

def start_menu():
 from Tkinter import *
 win = Frame(None)
 win.pack()
 Label(win, text='Python menu').pack(side=TOP)
 Button(win, text='Kill python', command=kill_python).pack(side=LEFT)
 Button(win, text='Kill draw', command=kill_draw).pack(side=RIGHT)
 win.mainloop()

    Another strange problem, which I am not quite sure if it is related, is
that if I move the "from Tkinter import *" to before all the def's then the
module will not load at all from the win32 code. ie. when I call
PyImport_ImportModule("P_Menu"), it returns null. But I can still load the
module perfectly find from the Python command line by doing an "import
P_Menu" and then "start_menu()".
    I'm really not sure if I am doing something wrong in my C code, or my
Python code. Any help would be greatly appreciated. Because before I start
spending my time creating my Python GUI, I'd like to make sure I'll be able
to use it from my C code! :)
    Also, is there some way to create a window using Python, and then have
win32 code draw into that window? (and/or X/Windows code, since I will
eventually be tryng to create a program that will run under both
environments). Thanks to anyone who might be able to help me out...

    --David Sheets





From zigron at jps.net  Mon Jun 21 18:10:11 1999
From: zigron at jps.net (Zigron)
Date: Mon, 21 Jun 1999 15:10:11 -0700
Subject: TCL/Tk
Message-ID: <B3wb3.67$Rj1.29977@typhoon-sf.snfc21.pbi.net>



From gvinther at atex.com  Mon Jun  7 16:35:51 1999
From: gvinther at atex.com (Gordon Vinther)
Date: Mon, 7 Jun 1999 16:35:51 -0400
Subject: problems using Tkinter
Message-ID: <928787726.241.78@news.remarQ.com>

I get the following error whenever I attempt to use Tk

Os AIX4.2.1
Python 1.5.2
Tk/TCL 8.0.1

from Tkinter import  *  .............this works
root=Tk() ........... this errors with the following message

File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 886, in __init__
self.tk = _tkinter.create(screenName, baseName, className)

TclError: invalid command name "tcl_findLibrary"


Doesn't this look like a Tcl problem?  All modules Tcl/Tk/Python compile and
test ok.





From c.evans at clear.net.nz  Wed Jun  9 07:26:47 1999
From: c.evans at clear.net.nz (Carey Evans)
Date: 09 Jun 1999 23:26:47 +1200
Subject: waste of resources ?
Message-ID: <mailman.1.1572195710.21486.python-list@python.org>

Unix doesn't guarantee you'll get exactly one signal delivered per
event.  To be sure to get all the children, you need to loop over the
processes with waitpid().

I think something like this should do it:

----------
def sigchld_handler(signum, frame):
    print "Checking for processes..."
    try:
        while 1:
            pid, sts = os.waitpid(-1, os.WNOHANG)
            if pid == 0:
                break
            print "Process %d ended with status %04x." % (pid, sts)
    except OSError, detail:
        if detail.args[0] != errno.ECHILD:
            raise
        else:
            print "No more children."
    else:
        print "Some children yet to finish."
----------

I can't actually see any way to keep count well enough and avoid race
conditions, to get rid of the try/except.  Sometimes I wish Python had 
POSIX signals, but not badly enough to actually write it.

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

	     "I'm not a god.  I've just been misquoted."

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




From ron at library.ucsf.edu  Mon Jun  7 20:10:31 1999
From: ron at library.ucsf.edu (Ron Klatchko)
Date: Mon, 07 Jun 1999 17:10:31 -0700
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
	 <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>
	 <87n1ykr0z9.fsf@ev.netlab.co.jp> <7ivs9q$dc5@news.dns.microsoft.com> <87k8toqv70.fsf@ev.netlab.co.jp>
Message-ID: <375C5F77.DAA900A@library.ucsf.edu>

Yukihiro Matsumoto wrote:
> Well, OK.  I understand now.  Now only thing I can say is, why do you
> emphasize so much on reclaiming object at the time you exepcted?

Because it never hurts to release a resource as soon as you're done with
it and it can hurt to hold onto it longer then necessaary.  Think about
an object the implements a mutex, do you want to hold it until GC is
done?

moo
----------------------------------------------------------------------
          Ron Klatchko - Manager, Advanced Technology Group           
           UCSF Library and Center for Knowledge Management           
                        ron at library.ucsf.edu



From tbryan at tbryanpc.arlut.utexas.edu  Mon Jun  7 06:47:57 1999
From: tbryan at tbryanpc.arlut.utexas.edu (Thomas Bryan)
Date: Mon, 7 Jun 1999 05:47:57 -0500 (CDT)
Subject: bug in os.path.samefile (Python 1.5.1)?
In-Reply-To: <1318D78C9072D11195C9006094EA98A72C20EF@ocesrv>
Message-ID: <Pine.LNX.4.10.9906070536020.2300-100000@tbryanpc.arlut.utexas.edu>

On Mon, 7 Jun 1999, Matej Cepl wrote:

> trying to finish my port of sitemap.py (for original Sitemap.pl from ESR
> see http://www.tuxedo.org/%7Eesr/sitemap-1.9.tar.gz for it), I am not
> able to debug the attached procedure.

[to comp.lang.python]
Matej and I have been working on this little project of and on for a
little while.  I plan to make the code available at
http://starship.python.net/~tbryan/ as soon as I've degugged some stuff
that I was adding this weekend.

> Do you have any idea, what's up (ntpath.samefile seems to be OK) and
> what should I do (please, do not recommend me 1.5.2)?

I don't use NT, but I have one suggestion.

Aren't filenames case-sensitive on NT?  Look at your traceback.  Notice
that the error is actually coming from a call to os.stat() inside of
os.path.samefile().  The error says that there is "no such file."  I think
that's because you're calling string.lower() on the filenames before
calling os.path.samefile().  If you have a file called MatejsFile,
string.lower() turns it into matejsfile...which may or may not exist.
When it does exist, os.stat() is actually looking at the wrong file.  
When it doesn't exist, os.stat() returns a "no such file" error.  Try
removing the string.lower() or moving the calls to string.lower() lower
in the code, after the call to os.path.samefile().

For example, 
Python 1.5.1 (#1, Sep  3 1998, 22:51:17)  [GCC 2.7.2.3] on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import os
>>> import string 
>>> s = '/home/crew/tbryan/public_html/FAQ/'
>>> os.stat(s)
(16893, 1091592, 777, 4, 1253, 1254, 1024, 928747153, 928638646,
928638646)
>>> s_lower = string.lower(s)
>>> s_lower
'/home/crew/tbryan/public_html/faq/'
>>> os.stat(s_lower)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
os.error: (2, 'No such file or directory')



-------------------------------------------
Tom Bryan
Applied Research Laboratories
University of Texas at Austin







From gem at infomap.gr  Tue Jun 22 09:05:17 1999
From: gem at infomap.gr (Giorgos Marentakis)
Date: Tue, 22 Jun 1999 16:05:17 +0300
Subject: Python and SQL server (Greek text)
Message-ID: <376F8A0C.3D87461A@infomap.gr>

Hallo all ,

I'm using Python to write my cgi-scripts in order to access a Microsoft
SQL server through an applet , using the odbc module included in the
win-32 extensions. I'm experiencing a problem . When accesing text data
type  ( english characters ) everything works fine , but when acesing
text ( greek chracters e.g) python returns numbers instead of characters
. I suppose this happens because Python tries to convert the data types
returned by the SQL server to its own objects and when dealing with
greek characters cannot do the appropriate mapping . Is there anyway to
workaround this problem?

Any help will be greatly appreciated
Giorgos




From larsga at ifi.uio.no  Wed Jun 16 12:09:34 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 16 Jun 1999 18:09:34 +0200
Subject: Python-list digest, Vol 1 #455 - 20 msgs
References: <1318D78C9072D11195C9006094EA98A72C2151@ocesrv>
Message-ID: <wk1zfc86v5.fsf@ifi.uio.no>

* Matej Cepl
|
| That was exactly my question: how to make filter with htmllib.py
| doing nothing and just copying HTML input on output?

Here's a simple implementation. Note that you'll need to escape all
'<'s and '&'s in element content and in attribute values. And in
attribute values you'll also need to escape '"'s.

import sys,htmllib

class IDTransform(htmllib.HTMLParser):

    def __init__(self,out=sys.stdout):
        htmllib.HTMLParser.__init__(self,None)
        self.out=out
    
    def handle_pi(self,data):
        self.out.write("<?%s>\n" % data)

    def handle_starttag(self,tag,method,attributes):
        self.out.write("<%s " % tag)
        for pair in attributes:
            self.out.write("%s=\"%s\"" % pair)
        self.out.write(">\n")

    def handle_endtag(self,tag,method):
        self.out.write("</%s>\n" % tag)
        
    def unknown_entityref(self,name):
        self.out.write("&%s;" % name)        

    def unknown_charref(self,no):
        self.out.write("&#%s;" % no)

    def handle_data(self,data):
        self.out.write(data)


--Lars M.



From gmcm at hypernet.com  Thu Jun 17 18:44:36 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 17 Jun 1999 17:44:36 -0500
Subject: sleep()ing
In-Reply-To: <3769698C.FD19B200@starvision.com>
Message-ID: <1282462763-115562301@hypernet.com>

Angus MacKay writes:

> is there anyway to sleep in a python program asside from:
> import os
> os.system("sleep 1")
> 
> or
> 
> import threading
> threading._sleep(1)
> 
> I mean it is part of unistd.h so you would think that it would
> appear in os or posix or somewhere.

Would time.sleep do? (There is a low level index to the library docs, 
you know).

- Gordon



From hniksic at srce.hr  Thu Jun  3 15:58:00 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 03 Jun 1999 21:58:00 +0200
Subject: Use of StringIO vs cStringIO in standard modules
References: <8790a1620b.fsf@pc-hrvoje.srce.hr> <199906031425.KAA06935@eric.cnri.reston.va.us> <87aeuh5kqc.fsf@pc-hrvoje.srce.hr> <yBz53.1777$nn.529906@news.shore.net>
Message-ID: <87btexnjjr.fsf@pc-hrvoje.srce.hr>

"Michael P. Reilly" <arcege at shore.net> writes:

> There are some interface differences between calls to
> cStringIO.StringIO and StringIO.StringIO which I think makes
> StringIO more "usable" in some cases.

In the case of base64.py, it simply doesn't use those differences.  In
the case of creating a "UsercStringIO" wrapper, the additional
functionality could be cheaply implemented by the wrapper.  Or, the
additional stuff could simply be added to cStringIO.

> I would want to see cStringIO mimic StringIO before the two are
> merged.

Absolutely.  I would not propose breaking backward compatibility.



From emile at fenx.com  Thu Jun 24 15:51:56 1999
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 24 Jun 1999 12:51:56 -0700
Subject: getstate, setstate and pickle
Message-ID: <X%vc3.1830$VL2.283108@news.direcpc.com>

I've written a class, instantiated, then pickled the instance.  At
first, pickling wouldn't work at all, but through some mix of
eliminating a class variable 'id' and adding __setstate__ and
__getstate__ methods it does dump (too soon to call it a pickle).

What I'm having trouble with is getting the load to match the dump.  I
lifted parts of the pickle module test code, and the dump == load part
fails, then my code falls apart, and then nothing works... (I'm in
California, so picking grapes is an option ;-)

I'm probably not using get/set state properly.  My instance should be
pickleable or I'd get an error there.

Any help or pointers to examples/docs appreciated.

--

Emile van Sebille
emile at fenx.com

NB:  Perhaps builtins could be honorably mentioned along with reserved
words in the documentation.  I'm now typing type(target) to get the
error as a result of my id problem above.  In time, I'll know them.  But
for now, it wasn't a pleasant bug to chase.
-------------------







From gmcm at hypernet.com  Mon Jun  7 09:22:49 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 7 Jun 1999 08:22:49 -0500
Subject: Programming Python - examples from the book work, no?
In-Reply-To: <375B78AC.5DB962E4@serop.abb.se>
Message-ID: <1283360448-61563154@hypernet.com>

Thomas S. Strinnhed writes:
> 
> Doing everything in MSVC++ on NT I try to run the Example 14-2
> (p534-540 ) from "Programming Python" by Mark Lutz. I run it just
> like it comes on the CD included with the book
> (CD-DRIVE:/Examples/DOS/Extend/Stacktyp.c) trying to extend Python
> with this Stack-type. BUT: I get an I-don't-know-what-this-is-about
> kind of compiler error, complaing about this call:
> PyObject_HEAD_INIT(&PyType_Type)  with message "initializer is not a
> constant".

While every other C compiler on earth allows this, MS decided on a 
very literal reading of the C language spec. The easiest way around 
it is to change the extension on your file to .cpp. The "proper" way 
around it is to define the type object with a NULL in this slot, and 
then assign &PyType_Type to it after it has been created.

sand-in-the-vaseline-goes-with-embrace-and-extend-ly y'rs

- Gordon



From news at dorb.com  Wed Jun  9 12:07:47 1999
From: news at dorb.com (Darrell)
Date: Wed, 9 Jun 1999 12:07:47 -0400
Subject: Python ActiveX Scripting interface
References: <375e87ae.11001540@news.origin-it.com>
Message-ID: <6mw73.1227$h14.5409@newsr2.twcny.rr.com>

Check out an earlier thread called "pyscript help" on 6/4/99
I believe that pyscript.py registers python as a scripting engine.
I had two problems one was a microsoft dll version problem with comcat.dll
comcat.dll is used to register com interfaces and I think it gets screwed up
in SP4.
The other was a change to the calculator demo.
--
--Darrell





From cgw at fnal.gov  Fri Jun 18 17:58:41 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Fri, 18 Jun 1999 21:58:41 GMT
Subject: All this self dot is making me...
In-Reply-To: <376AB11C.ED2B3040@boi.hp.com>
References: <Vn653.468$HD.99317@typhoon-sf.snfc21.pbi.net>
	<m33e0a52x2.fsf@chinon.cnrs-orleans.fr>
	<37551fa6.90445994@news.oh.verio.com>
	<8Rb53.1622$nn.483094@news.shore.net>
	<3757bdfb.262052080@news.oh.verio.com>
	<376AB11C.ED2B3040@boi.hp.com>
Message-ID: <14186.49425.740795.91149@buffalo.fnal.gov>

I'd suggest the potent combination of (X)Emacs "dabbrev" mode and
Python-mode.

Dabbrev has probably saved me from carpal tunnel syndrome...



From wlfraed at ix.netcom.com  Fri Jun  4 23:14:41 1999
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Sat, 05 Jun 1999 03:14:41 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7iqhkb$d7s@wiscnews.wiscnet.net> <7iu7ma$d33$1@cronkite.cc.uga.edu> <02hfonzsx2.fsf@rtp.ericsson.se>
Message-ID: <37588ff9.962233@nntp.ix.netcom.com>

On 04 Jun 1999 15:07:05 -0400, Kumar Balachandran
<kumar*xspam*@*xspam*rtp.ericsson.se> declaimed the following in
comp.lang.python:

> if
> ...
> elif
> ...
> else
> ...
> fi
> 
> def
> ...
> ...
> fed
> 
> 
> while
> ...
> elihw or wend
>
	Ugh... I have never liked languages that use a reversed spelling
to signify the end of a block... "end if", "end while" all read easier
to me than "fi", "elihw"...
 
	At that level, I'll take Ada's more generic "end loop", as "for"
and "while" loops are both built up as conditions on a "loop ... end
loop" block.

> etc. It makes the language more elegant. When I see code using
> indentation or blank likes to achieve blocking of constructs, it
> reminds me of an old language (FORTRAN I think it was called:-).
>
	No FORTRAN that I know of ever relied on indentation, etc. to
block constructs. Heck, a true FORTRAN totally ignored white space.

	do 10 i = 1.20
	do10i=1.20
assignments, identical.

	do10i=1,20
	do 10 i = 1, 20 
loop, identical.

> The change is simple to achieve if backward compatibility is given to
> accomodate people that grew up with FORTRAN.
>
	FORTRAN 77 added a nice blocked IF

	IF (cond) THEN
	ELSE IF (cond) THEN
	ELSE
	END IF
but again, no required indentation. Indentation is solely up to the
programmer for understanding of code -- Python's usage I find nice, as
it ensures visible blocking without having to match start and end
markers.

	Fortran 90 (for some reason it has now been declared that
"Fortran" is not a contraction of "FORmula TRANslator" but a proper
name, and to be treated with lowercase <G>) even permits the use of <, >
in conditionals !
 

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <



From greg.ewing at compaq.com  Thu Jun 24 16:43:28 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 25 Jun 1999 08:43:28 +1200
Subject: How to obtain an instance's name at runtime?
References: <7kqpu8$7u1$1@nnrp1.deja.com> <37718424.C70BB9DE@geekspace.com> <7ksgad$nvp$1@news.udel.edu> <377277B7.AC477B79@geekspace.com>
Message-ID: <37729870.397A59CD@compaq.com>

Joshua Rosen wrote:
> 
>         Instances aren't `definers'--they don't act as the root of the definition for
> `what something is'.

I think it's more a matter of the fact that the syntaxes for
creating classes and functions happen to incorporate a name, 
which Python squirrels away just in case someone is interested
later (which they often are).

When an instance is created, on the other hand, there is
no obvious name available. So, if you want one, you have to
arrange for it yourself.

Greg



From phxken at netzone.com  Thu Jun 17 17:09:22 1999
From: phxken at netzone.com (Farrell Kenimer)
Date: Thu, 17 Jun 1999 21:09:22 GMT
Subject: tk/tcl in python?
Message-ID: <376962AF.59EB5B1B@netzone.com>

My 1st post on this group:
Have COL 2.2 and have installed all the devel rpms for tk/tcl and
python.
Edited the /config/Setup file to show actual location of tk.h and tcl.h
Did the make -f Makefile.pre.in boot
and got Errors 2
Did the make Makefile.pre VERSION=1.5
and got Errors 2
I also edited the Setup file for actual location of tk/tcl libraries or
at least where the *.so files are.
The .tk and .tcl files are one directory lower.




From faassen at pop.vet.uu.nl  Thu Jun  3 04:47:46 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Thu, 03 Jun 1999 10:47:46 +0200
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net> <aahzFCpHw8.Gxt@netcom.com> <7j4e2u$8rh$1@m2.c2.telstra-mm.net.au>
Message-ID: <37564132.2D01B7C3@pop.vet.uu.nl>

Mark Hammond wrote:

[Aahz Maruch does not like block indenting a lot]
 
> Of course, this is your HO.  Many other people (including myself) found that
> a few weeks in, they thought "this is great", and many years later, can't
> believe that other language still hold on to the "old ways".  When I take a
> quick look at some of the new little languages referenced on this newsgroup,
> believe it or not, my _first_ reaction is "God those block closers are
> stupid and ugly" :-)

I never was much afraid of this indentation means block structure thing,
probably not having been spoiled by other languages that are horrid and
practice it, so it was mostly a neat discovery. Now, I am less than a
year in and I already have the same reaction to other languages. "Neat
language..but ick, what are those block closers for?" Just the other day
I was looking at some pseudocode in a book and my eyes kept stumbling
over the block closers.

I must say that Emacs Python mode helps tremendously in this respect,
though. Without it or something similar block indentation wouldn't be as
easy (though I've found I can code in Windows Notepad too if I really
want to :).

Regards,

Martijn



From arcege at shore.net  Mon Jun  7 11:19:22 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Mon, 07 Jun 1999 15:19:22 GMT
Subject: Programming Python - examples from the book work, no?
References: <1283360448-61563154@hypernet.com>
Message-ID: <_pR63.2126$nn.660380@news.shore.net>

Gordon McMillan <gmcm at hypernet.com> wrote:
: Thomas S. Strinnhed writes:
:> 
:> Doing everything in MSVC++ on NT I try to run the Example 14-2
:> (p534-540 ) from "Programming Python" by Mark Lutz. I run it just
:> like it comes on the CD included with the book
:> (CD-DRIVE:/Examples/DOS/Extend/Stacktyp.c) trying to extend Python
:> with this Stack-type. BUT: I get an I-don't-know-what-this-is-about
:> kind of compiler error, complaing about this call:
:> PyObject_HEAD_INIT(&PyType_Type)  with message "initializer is not a
:> constant".

: While every other C compiler on earth allows this, MS decided on a 
: very literal reading of the C language spec. The easiest way around 
: it is to change the extension on your file to .cpp. The "proper" way 
: around it is to define the type object with a NULL in this slot, and 
: then assign &PyType_Type to it after it has been created.

Ah.. thank you, Gordon.. I was wondering why that was there in _tkinter.c
when I was making mods. ;)

  -Arcege




From gvinther at atex.com  Wed Jun  2 15:09:29 1999
From: gvinther at atex.com (Gordon Vinther)
Date: Wed, 2 Jun 1999 15:09:29 -0400
Subject: problems using Tkinter
Message-ID: <928350546.726.80@news.remarQ.com>

I get the following error whenever I attempt to use Tk

Os AIX4.2.1
Python 1.5.2
Tk/TCL 8.0.1

from Tkinter import  *  .............this works
root=Tk() ........... this errors with the following message

File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 886, in __init__
self.tk = _tkinter.create(screenName, baseName, className)

TclError: invalid command name "tcl_findLibrary"






From dalke at bioreason.com  Tue Jun 15 11:59:14 1999
From: dalke at bioreason.com (Andrew Dalke)
Date: Tue, 15 Jun 1999 09:59:14 -0600
Subject: Bug in `imp' module?
References: <1282666010-103335885@hypernet.com>
Message-ID: <37667852.5A692CE7@bioreason.com>

Gordon McMillan <gmcm at hypernet.com> said:
> Not so simple. The .pyc file is trash if the .py file is newer.

Yeah, that makes perfect sense.  I guess I figured somehow it
would just *know* what to do.

Still, I don't need that ability for what we're doing so I'll
have to put off develing farther into that Python depth until
I have some free time; I figure that will be about 4 years ater
I'm buried :)

						Andrew
						dalke at bioreason.com



From tim_one at email.msn.com  Wed Jun  9 22:22:08 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 9 Jun 1999 22:22:08 -0400
Subject: fork()
In-Reply-To: <877lpd528t.fsf@ev.netlab.co.jp>
Message-ID: <000b01beb2e8$09deb660$af9e2299@tim>

[Tim]
> Context:  this offshoot of the thread was talking about memory
> use after a fork() call in a copy-on-write implementation of fork.
> So long as bookkeeping info is stored in object headers, "real GC"
> is at a real disadvantage, in that respect, compared to RC.  Simply
> because RC won't touch something unless *you* do.

[Matz]
> Indeed.  Though some "real GC" keeps mark bits in separated pages to
> avoid copying bunch of pages in such situation.

And, as at least Barry pointed out earlier, some refcounting systems
separate the counts from the objects they're counting, and largely for the
same reason.

> BTW, ref counting modifies objects for mere referencing, it causes
> copying too.

That's actually the observation that *started* "this offshoot of the
thread" -- but as if it were unique to RC, which it isn't.

>  Also, it would be performance pain if Python's global thread lock is
> removed someday, since ref counting requires mutual lock to protect
> ref count update.  I'm not sure this problem can be solved or not.

Back at Python 1.4, a set of patches was developed (by Greg Stein) that
*did* remove the global lock.  This meant fine-grained locking of other
stuff as needed, refcounts included.  The interpreter has changed quite a
bit since then, so old numbers aren't quantitatively relevant anymore; but
it was indeed a significant slowdown for single-threaded Python programs.

It's particularly acute for Python because *everything* "is boxed"; nothing
is exempt, and refcounting is never delayed.

OTOH, people who gripe about needing to keep refcounts straight in
extensions really haven't lived until they've wrestled with slamming in
exactly the right macros in exactly the right places to stop threaded
non-conservative GC from finding things in an inconsistent state <0.1 wink>.

what-end-users-don't-see-would-kill-them<wink>-ly y'rs  - tim





From arcege at shore.net  Thu Jun 24 10:55:22 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Thu, 24 Jun 1999 14:55:22 GMT
Subject: Question - Returning exit codes
References: <7ktbmu$5os$1@nnrp1.deja.com>
Message-ID: <uFrc3.66$Of3.12274@news.shore.net>

sjoshi at ingr.com wrote:
: Hello All
:    I'm calling my python program from a perl script using the system
: command. I was wondering what would be the best way to return exit
: codes from my python script so that they can be intercepted in the perl
: program.

: thanks
: Sunit

Either call sys.exit(rc) or raise the SystemExit exception with the
numeric return code.

If the exception data is not an integer, the string representation is
printed to stderr and the return code is 1.  Otherwise, nothing is
printed (no traceback) and integer is the return code.

Note: the sys.exit() function raises the SystemExit exception.

  -Arcege




From graham at sloth.math.uga.edu  Thu Jun 10 11:08:40 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 10 Jun 1999 15:08:40 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7j90va$6lk$1@cronkite.cc.uga.edu> <u3g1443iht.fsf@ares.sys1.ngo.okisoft.co.jp> <87pv384koy.fsf@ev.netlab.co.jp> <u3909slwbo.fsf@ares.sys1.ngo.okisoft.co.jp> <slrn7lud8t.g2c.bwinton@tor.dhs.org>
Message-ID: <7jokdo$hih$3@cronkite.cc.uga.edu>

Blake Winton (bwinton at tor.dhs.org) wrote:
: Ref counting with GC sounds to me like a good idea for Python 2.0, as it
: will eliminate a class of memory leaks that Python can currently fall
: prey to.  As long as you don't move stuff around in memory, and only
: collect things where all the refcounts are accounted for, and can't be
: reached, you should be safe.  Granted, this means that the people
: writing the extensions will have to be careful to INCREF/DECREF where
: they're supposed to, but I don't see that as a bad thing since they
: should already be careful about that sort of stuff.

Actually Blake I have a feeling that I have read a paper somewhere that
says that a mark sweep collector with refcounts can in fact patch up
refcount errors, so it won't be so critical that extension writers get
their INCREF/DECREFs just right. I am not 100% sure of this but at least
on the surface it seems plausible (even if it can only give a warning
about such errors). Anyone know more about this?
graham

-- 
           As you grow up and leave the playground
       where you kissed your prince and found your frog
          Remember the jester that showed you tears
                   the script for tears



From jeff at norden.math.tntech.edu  Fri Jun 11 11:12:03 1999
From: jeff at norden.math.tntech.edu (Jeff Norden)
Date: 11 Jun 1999 10:12:03 -0500
Subject: newbie questions
Message-ID: <x7emjizs8s.fsf@norden.math.tntech.edu>

I've been tinkering with python for a few days.  I have to admit that my
first impression was that this language would turn me into a newt, but I got
better :-)

A few questions:

1) Is there no equivalent to perl's  "use strict" in which all variables
have to be declared?  How about perl's "-w" flag, which at least gives a
warning when a variable is only used once.  These features are nice for
catching common mistakes, i.e., if you try to increment spam with

   span = spam + 1

python will just create a new span variable, and no error message.
Or, consider the even more insidious error:   vikingl = viking1 + spam

2) Is the above the correct way to increment a variable?  I can't find an
equivalent for "++" or "+="

3) These are rather minor, but is there no way to "forward declare" function
definitions?  I'd rather put them at the end of the file instead of the
beginning.  Python does parse the whole file before it starts execution,
doesn't it?  Also, is there a way to just check for syntax errors without
execution, like perl's "-c" switch?

Thanks in advance,
  -Jeff Norden, jeff at math.tntech.edu
   Dept of Math, Tenn Tech Univ, Cookeville TN



From fredrik at pythonware.com  Wed Jun 16 17:32:26 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 16 Jun 1999 21:32:26 GMT
Subject: need exit from 'exec' execution
References: <3768057F.D7473931@online.no>
Message-ID: <002301beb83f$bbd40f80$f29b12c2@pythonware.com>

(quick, before the clock strikes midnight, and I'm
switched back to single-process mode!)

Anders Moe <andermoe at online.no> wrote:
> My python-embedded app frequently makes use of  'exec <program-string>'.
> I'd like to create a Tk Button to halt this execution at some point, with the
> option of aborting it. The only command that comes close is the exit state-
> ment, but its to powerful - it exits from the whole app, whereas I only
> wanted a local exit from the exec's program-string. Anyone know how to
> make this kind of local exit ?

are we talking sys.exit here?  if so,

import sys

try:
    exec "sys.exit()"
except SystemExit:
    print "got you"

does what you want.

see the library reference for more info:
http://www.python.org/doc/current/lib/module-sys.html

<F href="http://www.pythonware.com/people/fredrik" />




From zorro at zipzap.ch  Thu Jun 24 09:41:18 1999
From: zorro at zipzap.ch (Boris Borcic)
Date: Thu, 24 Jun 1999 15:41:18 +0200
Subject: My thoughts on Python 1.618
Message-ID: <3772357E.F4876C79@zipzap.ch>

Greetings to the people and God(s) of Pythonland,

A few quasi-random thoughts on Python NG...

Bootstrap thought : First of all, the long speculative threads
on Python 2, I found disturbing, because in any case, the
design of Python is Guido's property, and the artist's
prerogative is to remain free to dream up his designs,
without interference.

Second thought : "Python 2 is a contradiction in terms".
Obviously, the Python version numbering scheme is destined
to converge on the value of the golden mean 1.618, therefore,
Python 2 is out of bounds.

Third thought : Still, I can dream up ideas for Python, but
purely as an exercize, in the hope that one of these ideas
will draw a positive smile from Guido's face.

The most recent of these ideas, takes as a pretext of
expression : to have the recurrent debate on Python's braces
by newbies, C-ers, Perl-ers and Tcl-ers, further itself one bit.

I know that one key factor (by no means the only one) of
the beauty of Python to my eyes, is precisely the "effect"
of having braces "replaced" in Python, by the INDENT/DEDENT
scheme.

Reflecting on it, I discovered that this component of my
aesthetic pleasure with Python does not reduce to the idea
that making indentation machine-significant was the right
thing to do, but may well also have to be explained in terms
of a theory of the taste for braces.

It came to me in a flash, that BOTH the general taste for
braces in programming languages AND my distaste with the
way they are used TRACE BACK to an education in mathematics,
that introduced me, as well as the others, to braces AS DELIMITERS
OF THE WRITTEN REPRESENTATION OF SETS.

Now, SETS are not the last word, even in mathematics, and
one does well IMHO to think that what is today known as
"Category Theory" is actually "Sets done right", that is,
Sets done the way they should have been (and were not)
to fit the promises of the "new maths" that conquered the
educative system(s) during the sixties.

This is to say, that although SETS aren't the last word,
even IMHO, I believe today that, besides my love of using
indentation significatively, hides behind my taste for Python
syntax, an aesthetical DISpleasure with HOW braces are used
in "those OTHER languages", namely as block delimiters,
BECAUSE that use doesn't "chord" with set syntax.

And indeed, Python DOES use braces, to delimit dictionary
constructors, which are not sets, but... but are actually
a way to use braces that SMOOTHLY EXTENDS to classical set
syntax. { 1, 2, 3 } is presently illegal syntax in Python.
It could be made natural legal shorthand for { 1:1 , 2:2, 3:3}.

To make it interesting, a corresponding semantics should be
given to the currently illegal :

for item in { 1, 2, 3 } : do_your_thing(item)

I suggest the spec : "for I in M, M mapping", guarantees to loop
varying I, and the set of values I takes is comprised between

filter((lambda x,M=Mapping : M[x] is x),Mapping.keys())

And

M.keys()+filter((lambda x,K=M.keys() : x not in K),M.values())

Which boil down to the same in the case M is a set.

Now I remember that Guido comes from Dijkstra's country,
who, together with Wirth, made "syntactic sugar" a bad name.
And the above sounds ridiculously LIKE a case for a
syntactic sugar addition (MnotsoHO, is that it would be
more precise to speak of semantic sugar, ah, well).

Well, if it is a matter of sugar, why not start by going
full length, and ask for a ridiculously syntactical further
extension. While SETS, in the school where I was thaught,
were delimited by braces, their elements where SEPARATED
BY ";" rather than commas. And this early training burns
me regularly when I am writing dictionnary constructors
in Python. So I would wish Python 1.618 to accept ';' as
item or key-value pair separator in that context, just as
it already accepts ',' in the same role.

But that's not all. As I said in the beginning of this
little diatribe, it's banner is to further the debate
on Python's "bracelessness" quite a bit, in a way that
will incite the evercomplaining C-ers etc... to silence
their "critics"... and the method is to show better the
starring role that braces already have in Python.

This is a motive for my further proposition : ascribe,
as a matter of principle and standard implementation,
to *mappings* the CALL SEMANTICS of an ANONYMOUS CLASS.

So that :

	doubler = { '__mul__' : lambda s,y : 2*y }()

does the obvious.

...

To fullfill my promise at the beginning, I shall
come back with completely unrelated, and perhaps
even contradictory, propositions for Python NG aka
1.618 aka 2.001.

As I said, I view myself here, in the role of the
King's fool. Guido is the King. The design contract
of my proposals is no more than trying to bring
a fleeting smile on the King's face.

Boris Borcic
--
"c++ is a contradiction in terms" -- Einstein



From tim_one at email.msn.com  Wed Jun  2 19:31:01 1999
From: tim_one at email.msn.com (tim_one at email.msn.com)
Date: Wed, 02 Jun 1999 23:31:01 GMT
Subject: mailing list to news gateway
References: <7j4asp$v4a$1@nnrp1.deja.com>
Message-ID: <7j4erl$i9$1@nnrp1.deja.com>

In article <7j4asp$v4a$1 at nnrp1.deja.com>,
  jeremy at cnri.reston.va.us wrote:
> I have noticed what appear to be problems with the gateway between
> the mailing list (python-list at python.org) and the newsgroup -- and
> more specifically with the view of the newsgroup at dejanews.
>
> Since about last Thursday, messages I have sent to
> python-list at python.org haven't been showing up at Deja News. I
> would assume they also aren't showing up in the newsgroup, but
> I'm not sure.
>
> So two questions: First, has anyone else noticed problems with
> messages sent to the mailing list not appearing in the newsgroup?

I hadn't checked, but now that you mention it I see that DejaNews
doesn't having anything from me after 27-May-99 (last Thursday -- I
think -- let's see -- yes -- I think) either.

> Trying to establish the scope of the problem.  Perhaps it's just
> weirdness related to my messages.

Nope!

> Second, has anyone seen posts from me on the newsgroup lately --
> say in the Python 2.0 thread?

Not from the newsfeed at work, but it's an exceptionally poor one.  Did
see them on the mailing list.

> (I'm sure everybody keeps track of all the witty and insightful
> comments I make <1.0 wink>.)

indeed-i-hear-they're-building-a-shrine-in-your-honor<wink>-ly y'rs  -
    tim


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From Michael.P..Reilly at p98.f112.n480.z2.fidonet.org  Wed Jun 30 10:07:14 1999
From: Michael.P..Reilly at p98.f112.n480.z2.fidonet.org (Michael P. Reilly)
Date: Wed, 30 Jun 1999 15:07:14 +0100
Subject: PySequence_Check
Message-ID: <000015b5@bossar.com.pl>

From: "Michael P. Reilly" <arcege at shore.net>

Tim Peters <tim_one at email.msn.com> wrote:
: [Reuben Sumner]
:> How can I do the equivalent of a PySequence_Check in Python?
:> GvR told me that the question was illdefined (if I explained it correctly)
:> and to ask here.  It seems that some of the functions in abstract.h
:> are available from python (len(), callable()) and others are not
:> (PySequence_Check, PyNumber_Check).
:>
:> What am I missing?

: For starters, unquestioning faith in Guido's pronouncements <wink>.

:> I have a nested structure of lists and tuple and I just want to know
:> when I have bottomed out.

: Then why not check for that directly and be done with it?

:> I can check for tuple or list indivually but that is ugly and
:> inflexible.

: Not to mention simple, obvious and unsurprising <wink>.

:> I could just try indexing and see if there was an exception but that
:> is not much better.

: OK, you tell me:  what *is* a sequence?  Answer that, and you'll know how to
: check for it <0.5 wink>.  Before answering too fast, note that mapping types
: can be indexed without raising an exception, and so can strings.  Do you
: want your nest of "lists and tuples" *not* to "bottom out" at a string?
: Probably not -- but if so, you're not really interested in testing whether
: it's a sequence!  Note too that user-defined classes can support the
: __getitem__ protocol without supporting the __getslice__ protocol, or vice
: versa.  Are either of those sequences, or do they need to support both?
: Whichever way you answer, half the world will disagree.

: I think that, in most cases, when people say "sequence":  (A) it's
: ill-defined <wink>; and, (B) after five rounds of tedious questioning they
: end up deciding they mean "responds to slicing but isn't a string".  So
: there you go:

: def issequence(x):
:     try:
:         x[0:0]
:     except:
:         return 0
:     return type(x) is not type("")

My basis (in my own programming) has been:
  import types, UserList
  def issequence(x):
    return (isinstance(x, types.ListType) or
            isinstance(x, types.TupleType) or
            isinstance(x, UserList.UserList))

It is very restrictive in some cases, but if I want a user defined
sequence, I usually _try_ to make one from UserList.

  -Arcege



From Gary.Duzan at p98.f112.n480.z2.fidonet.org  Tue Jun 29 06:26:41 1999
From: Gary.Duzan at p98.f112.n480.z2.fidonet.org (Gary Duzan)
Date: Tue, 29 Jun 1999 11:26:41 +0100
Subject: ANN: Stackless Python 0.2
Message-ID: <00001586@bossar.com.pl>

From: Gary Duzan <gduzan at gte.com>

Toby J Sargeant wrote:
> 
> On Mon, Jun 28, 1999 at 12:34:48AM -0400, Tim Peters wrote:
> > Nope!  It's a beauty of the implementation already that each code object
> > knows exactly how much "Python stack space" it needs, and (just) that much
> > is allocated directly into the code object's runtime frame object.  IOW,
> > there isn't "a Python stack" as such, so there's nothing to change here --
> > the stack is implicit in the way frames link up to each other.
> 
> This might be completely irrelevant, but during the course of my masters, I
> considered doing this kind of thing to java in order to allow asynchronous
> threads to share stack frames (don't ask...). My supervisor complained
bitterly
> on the grounds that function invocations where orders of magnitude more
> common than object creation, and hence associating memory
> allocation/deallocation with every call was considered horrendously
> inefficient.

   This appears to be the conventional wisdom. The optimization class I
took used an ML-like language, and the first thing we did was to move
heap frames, which was the default, to the stack (i.e. closure
optimization.)
It might be less of an issue in an interpreted language if it already has
heap
allocation (or other sorts of) overhead to deal with when making calls.
There may be other allocation tricks which can be used, as well.

> It seems that this should affect Stackless Python equally as much. Does
anyone
> have anything to add on the subject? I would imagine that frames could be
> allocated and managed in chunks to alleviate a lot of the memory management
> load...

   I would think that as well (at least at first.) Maybe pools of different
sizes of frame to limit the space overhead.

> Toby.

					Gary Duzan
					GTE Laboratories



From gerrit.holl at pobox.com  Tue Jun 29 03:27:57 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Tue, 29 Jun 1999 09:27:57 +0200
Subject: Python GUIs: Summary and Conclusion
In-Reply-To: <99FA4A1A1280BEBD.0A1ABA62CC989479.4C2C6022F695BEAC@lp.airnews.net>; from Mikael Lyngvig on Mon, Jun 28, 1999 at 05:08:40PM +0000
References: <99FA4A1A1280BEBD.0A1ABA62CC989479.4C2C6022F695BEAC@lp.airnews.net>
Message-ID: <19990629092757.A4998@optiplex.palga.uucp>

On Mon, Jun 28, 1999 at 05:08:40PM +0000, Mikael Lyngvig wrote:
> From: mikael at pobox.com (Mikael Lyngvig)
> Subject: Python GUIs: Summary and Conclusion
> Date: 	Mon, 28 Jun 1999 17:08:40 GMT
> To: python-list at python.org
> 
> Hi,
> 
> Anyone disagree?  
> 

I think one could also make an application "multiple-interface". The user can
choose: svgalib (any support in python, by the way?), curses, commandline,
tk, qt/kde or some of the other stuff. The developer can put the interface
in a seperate class or even in a seperate module (I'll do with my discoverb
project (see signature)), so the user can exactly download what he want. I
believe KDE users would like KDE applications...

groeten,
Gerrit.

-- 
The Dutch Linuxgames homepage:	http://linuxgames.nl.linux.org
Personal homepage:		http://www.nl.linux.org/~gerrit/

Discoverb is a python program (in several languages) which tests the words you
learned by asking it. Homepage: http://www.nl.linux.org/~gerrit/discoverb/



From a.mueller at icrf.icnet.uk  Tue Jun 22 16:22:50 1999
From: a.mueller at icrf.icnet.uk (Arne Mueller)
Date: Tue, 22 Jun 1999 21:22:50 +0100
Subject: interactive-logging
Message-ID: <376FF09A.8170CB@icrf.icnet.uk>

Hi All,

I'd like to add some logging to the interpreter's interactive modus so
that everything I type in and everything what's printed to the screen
(errors and 'normal' output) is copied to a log-file.

Usage could be as follows:

Python 1.5.2b2 (#11, Jun  3 1999, 17:27:58) [C] on irix646
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
python> import ilogger
python> ilogger('lastputhon.log')
python> x = [1,2,3,4,5]
python> for i in x: 
...     print i
... 
1
2
3
4
5
python> for i x:
  File "<stdin>", line 1
    for i x:
          ^
SyntaxError: invalid syntax
python> 

This interactive session should than be written to 'lastputhon.log' e.g:
# python session from mueller at jumbo
# Tue Jun 22 21:13:47 BST 1999
x = [1,2,3,4,5]
for i in x:
   print i
out>
out> 1
out> 2
out> 3
out> 4
out> 5
error> for i x:
error>   File "<stdin>", line 1
error>     for i x:
error>           ^
error> SyntaxError: invalid syntax

It's easy to parse the log-file for normal output indicated by 'out>'
and for erros. The normal python-prompt and the '...' in the loop-body
is removed. 

Is there already something like this written? I've no idea how how the
interactive modus works. For aprogram I've written that copies a
file-object, say everything that goes to 'sys.stderr' is also written to
a log-file, how to do that interactive modus?

	thanks very much,

	Arne



From entree at my-deja.com  Wed Jun 23 06:20:24 1999
From: entree at my-deja.com (entree at my-deja.com)
Date: Wed, 23 Jun 1999 10:20:24 GMT
Subject: PIL and cgi link problem
Message-ID: <7kqcd3$3at$1@nnrp1.deja.com>

Hi,

We are experiencing a strange problem with the
PIL library.
We have compiled it and installed it succesfully
into our Python environment (linux Redhat 5.2).
The strange thing is that it works fine when
imported in Python from a shell.

We want to use the library from a python cgi
script. Here we encounter a strange problem with
the Image module. The cgi script tries to import
it in response to a web-query. but it failes with
an ImportError Exception: _imaging C module is
not installed. further investigation of the
exception showed the following (link) error:
undefined symbol PyExc_IOError

Does anybody has an idea?.

We are using Python 1.5.1 and LibImaging 1.0b1

Thanx in advance,

Robbert
Entree





Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From jsight at mindspring.com  Tue Jun 22 15:09:36 1999
From: jsight at mindspring.com (Jesse D. Sightler)
Date: Tue, 22 Jun 1999 15:09:36 -0400
Subject: YAS to the "Reading line-by-line" Problem
References: <Pine.SUN.3.95-heb-2.07.990622183356.3454A-100000@sunset.ma.huji.ac.il>
Message-ID: <376FDF70.4756AD38@mindspring.com>

Ack is raising an exception for a normally occuring event really a
_good_ idea?  Generally exceptions are only used for things that
actually are error conditions, and it would seem to me that using them
as nothing more than a fancy flow-control construct is a sure path to
nasty spaghetti code.

I really do not like this idea.  :)

Moshe Zadka wrote:
> 
> (YAS == Yet Another Solution)
> 
> This solution introduces file-like objects which do the Right Thing(TM) at
> EOF: raise EOFError (instead of returning an empty string)
> 
> ----------------- cut here ------------------
> class File:
>         '''\
>         File-like objects which throw EOFError on End-of-File.
> 
>         Initialize with a file-like object.
>         '''
>         def __init__(self, file):
>                 self.__file=file
> 
>         def __getattr__(self, name):
>                 if name[:4]!='read':
>                         return getattr(self.__file, name)
>                 return _Wrap(getattr(self.__file, name))
> 
> class _Wrap:
>         def __init__(self, function):
>                 self.function=function
> 
>         def __call__(self, *args, **kw):
>                 ret=apply(self.function, args, kw)
>                 if ret=='': raise EOFError, 'no more data in file'
>                 return ret
> 
> def open(file, mode='r'):
>         '''\
>         return a file like object open to /file/ with mode /mode/,
>         which throws an EOFError on EOF
>         '''
>         import __builtin__
>         return File(__builtin__.open(file, mode))
> 
> def _test():
>         import sys
>         file=open('/etc/passwd')
>         try:
>                 while 1:
>                         sys.stdout.write(file.readline())
>         except EOFError: pass
> 
> if __name__=='__main__':
>         _test()
> ---------------------------- cut here ---------------------------
> 
> --
> Moshe Zadka <mzadka at geocities.com>.
> #!/usr/bin/tail -1
> Just another tail hacker.

-- 
---------------
Jesse D. Sightler
http://www3.pair.com/jsight/

"Do not use a hatchet to remove a fly from your friend's forehead." 
      - Chinese Proverb



From matz at netlab.co.jp  Fri Jun  4 04:18:47 1999
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 04 Jun 1999 17:18:47 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
 <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>
 <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net>
 <87hfosquc0.fsf@ev.netlab.co.jp> <3753F121.E4A9BE8B@prescod.net>
 <87yai3ntdt.fsf@ev.netlab.co.jp> <7j3iif$134e$1@nntp6.u.washington.edu>
 <7j45iv$908$4@cronkite.cc.uga.edu> <7j4hhj$118u$1@nntp6.u.washington.edu>
 <7j4jbe$dhh$1@cronkite.cc.uga.edu> <37564A02.F115B9C5@infercor.no>
 <045501beadf8$610e62e0$f29b12c2@pythonware.com>
 <14166.61071.201719.405639@bitdiddle.cnri.reston.va.us>
 <u31zfszath.fsf@ares.sys1.ngo.okisoft.co.jp>
Message-ID: <871zfsbcpk.fsf@ev.netlab.co.jp>

Hisao Suzuki <suzuki611 at okisoft.co.jp> writes:

|Me too.  And I am afraid that some evangelists would do nothing
|but _only_ point out Python's some so-called fault in order to
|drive us into convert to their favorite or invented language...

Sorry for bothering the newsgroup.  But I couldn't resist pointing out
misunderstanding about "real GC" among some part of Python users.

	http://www.iecc.com/gclist/GC-faq.html

It's totally OK for Python to use ref counting.



From suzuki611 at okisoft.co.jp  Mon Jun  7 19:26:31 1999
From: suzuki611 at okisoft.co.jp (Hisao Suzuki)
Date: 08 Jun 1999 08:26:31 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7jgk4j$s8p$1@cronkite.cc.uga.edu>
Message-ID: <u3vhcztwwo.fsf@ares.sys1.ngo.okisoft.co.jp>

In article <7jgk4j$s8p$1 at cronkite.cc.uga.edu>,
graham at sloth.math.uga.edu (Graham Matthews) wrote:
> Hisao Suzuki (suzuki611 at okisoft.co.jp) wrote:
> : Just now I tried to make the recent stable version of Ruby on a
> : box of the same type.  Regrettably it crushed as the similar way
> : as before.  Certainly Ruby _is_ suffering from its insufficient
> : portability.
> 
> And what do we conclude from this?

In article <u3g1443iht.fsf at ares.sys1.ngo.okisoft.co.jp>,
I also wrote:
| ... On that box, Ruby did not run but dumped immediately
| with its twiddling the machine stack, while Python ran perfectly
| with no difficulty.  ...

We have been enjoying Python's portability for many years.

--===-----========------------- Sana esprimo naskas sanan ideon.
SUZUKI Hisao            suzuki611 at okisoft.co.jp, suzuki at acm.org.



From faassen at pop.vet.uu.nl  Tue Jun 15 06:26:32 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Tue, 15 Jun 1999 12:26:32 +0200
Subject: GNU python?
References: <3765B8B4.18DE0257@spots.ab.ca>
Message-ID: <37662A58.C8930BC4@pop.vet.uu.nl>

Maxwell Sayles wrote:
> 
> Looking to try out python... is there a GNU python for Win9x?  and a
> tutorial maybe?  thx

Python is open source/free software, but not GNU. So you won't find a
'GNU' python for Win9X. The good news it that you can download Python
for windows just fine from www.python.org. You can also find a tutorial
on that site.

Good luck!

Martijn



From heiland at ncsa.uiuc.edu  Fri Jun 11 14:06:56 1999
From: heiland at ncsa.uiuc.edu (Randy Heiland)
Date: Fri, 11 Jun 1999 13:06:56 -0500
Subject: Embedding question
References: <37613D8A.BD4E07FA@ncsa.uiuc.edu>
Message-ID: <37615040.AF246F5C@ncsa.uiuc.edu>

>

Well, I was wrong.  I can indeed do this - in a simple C++ pgm.  My earlier
situation had it buried inside a much larger ball of goo... it seems the
goo may have been causing problems.  Will explore further.





From paul.moore at uk.origin-it.com  Wed Jun  9 11:37:02 1999
From: paul.moore at uk.origin-it.com (Paul Moore)
Date: Wed, 9 Jun 1999 17:37:02 +0200 
Subject: Python ActiveX Scripting interface
Message-ID: <375e87ae.11001540@news.origin-it.com>

I've just tried to set up Python on my Windows NT box (NT4 SP4). I
installed Python (1.5.2) and Mark Hammond's Python Win32 extensions
(build 125) with no problems. I ran the ActiveX scripting registration
script, to install the scripting interfaces, but I still have no joy
in trying to use Python as an ActiveX scripting language.

In WSH 2.0, if I specify LANGUAGE="Python", I get a message about that
not being a known language. The Internet Explorer examples fail to
work (with script errors) - I'm using IE5.

I know this is so vague a report as to be pretty much useless - I
can't clarify much at the moment as I'm away from the machine in
question. But I was wondering two things:

1. Is there any reason why the Python script interface would fail to
work with WSH 2.0 or IE5? It seems unlikely, but I suppose it's
possible...

2. Can anybody tell me exactly what I should look for in the registry,
to be sure that Python is registered properly as a scripting host? (I
don't relish trying to understand the Python script that does the
registration...!)

Thanks for any help. When I'm next at the machine in question, I'll
try to isolate the problem a bit better.

Paul Moore.




From graham at sloth.math.uga.edu  Wed Jun  2 16:59:16 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 2 Jun 1999 20:59:16 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <7j0urf$2dm$3@cronkite.cc.uga.edu> <3754FBE2.4BFD7319@infercor.no>
Message-ID: <7j45v4$908$5@cronkite.cc.uga.edu>

Paul Boddie (paulb at infercor.no) wrote:
:Well, my statement was actually an invitation for more enlightened contributors
:to give examples of such schemes, and this has evidently had the intended
:effect. 

cf the posts on Ruby.

graham

-- 
             Like a bird on a wire
        Like a drunk in a midnight choir
            I have tried in my way
                  To be free



From smithkf at OMIT_THIS.us.ibm.com  Mon Jun 21 19:38:23 1999
From: smithkf at OMIT_THIS.us.ibm.com (Kevin F. Smith)
Date: Mon, 21 Jun 1999 16:38:23 -0700
Subject: sleep()ing
References: <3769698C.FD19B200@starvision.com>
Message-ID: <376ECCEF.B738DFBF@OMIT_THIS.us.ibm.com>

It's in time.

import time
time.sleep(1)



Angus MacKay wrote:

> is there anyway to sleep in a python program asside from:
> import os
> os.system("sleep 1")
>
> or
>
> import threading
> threading._sleep(1)
>
> I mean it is part of unistd.h so you would think that it would
> appear in os or posix or somewhere.
>
> cheers, Angus.




From greg.ewing at compaq.com  Wed Jun 16 19:05:55 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Thu, 17 Jun 1999 11:05:55 +1200
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com>
Message-ID: <37682DD3.43923243@compaq.com>

Fernando Mato Mira wrote:
> 
> I wanted to find out whether the disgust for tcl I've seen many people
> express had some basis on some generally accepted principles.

I think I can give some reasonably objective reasons why
I find that I enjoy using Python in a way that I don't
enjoy using Tcl. (I have used both, and in one case have
written versions of the same program in both languages.)

1. Syntax

Tcl's syntax is based on the principle that an unadorned
sequence of alphanumeric characters represents a literal.
Anything else (such as referring to a variable) requires
extra syntax.

Ousterhout's justification is that this is optimal
for entering commands interactively, which is probably true.
However, this seems to me the wrong thing to optimise for,
given that tcl is meant to be a *programming* language.
The case being optimised for -- typing in a command which
is to be used once and thrown away -- simply doesn't occur.
Programs, even tiny ones, get used more than once!

In my opinion, the small amount of extra syntax needed
to quote literals and express function calls in a language
like Python is well worth the benefits you get further
down the track. An example of these is that, where in
Tcl you find yourself writing

   set a [expr $b + $c]

in Python you get to write

   a = b + c

Since the vast majority of words in my programs refer
to variables, and string literals are relatively rare,
I much prefer the Python way of writing things to
the Tcl way.

2. Data structures

Briefly, Python has them, Tcl doesn't. Tcl has character
strings, which are sometimes interpreted according to a
certain set of rules as lists of other character strings.
The rules are pedantic and can trip you up sometimes.

Tcl also has associative arrays, which is a powerful
feature compared to what some languages give you, but
they're not first-class objects and all you can put
in them are character strings.

Python has a rich set of first-class data types, including
lists, dictionaries (associative arrays) and classes/instances,
from which you can build real data structures and manipulate
them in easy and intuitive ways.

Tcl proponents will tell you that you can emulate all of
that using tcl lists and arrays, which is true, but misses
the point. Why bother with emulation when you can have the
real thing at no extra cost?

Summary

In my experience, I find that I can do with Python everything
that Tcl was designed for, do it more easily, do a lot
more besides, and have more fun in the process. I believe
the reason for this is rooted in some fundamental design
features of these languages, which I have sketched above.
I think that's about as close as one can get to providing
an objective argument as to whether one language is better
than another for a given purpose.

Greg



From siona at chiark.greenend.org.uk  Mon Jun 21 10:25:24 1999
From: siona at chiark.greenend.org.uk (Sion Arrowsmith)
Date: 21 Jun 1999 15:25:24 +0100 (BST)
Subject: Creating a new type with an extension module
Message-ID: <Ctp*vl52n@news.chiark.greenend.org.uk>

Is there any documentation anywhere on how to go about writing an
extension which implements a new type? The extending/embedding
tutorial give me, very quickly and painlessly, a module with
functions, but I could really do with them being object methods.
http://www.python.org/doc/current/api/newTypes.html is unhelpful
in the extreme -- do I have to resort to reading source?

-- 
\S -- siona at chiark.greenend.org.uk -- http://www.chaos.org.uk/~sion/
  ___  |  "Frankly I have no feelings towards penguins one way or the other"
  \X/  |    -- Arthur C. Clarke
       |  If music be the food of love, bring me a doggie bag.



From jam at newimage.com  Wed Jun 23 09:00:41 1999
From: jam at newimage.com (jam)
Date: Wed, 23 Jun 1999 09:00:41 -0400
Subject: [FYI] Future and Status of the RPM based Linux distribution
In-Reply-To: <7kqcpl$u4l$1@news.rhein-zeitung.DE>; from olli@rhein-zeitung.de on Wed, Jun 23, 1999 at 10:27:01AM +0000
References: <7kqcpl$u4l$1@news.rhein-zeitung.DE>
Message-ID: <19990623090041.A7683@toast.internal>

On Wed, Jun 23, 1999 at 10:27:01AM +0000, olli at rhein-zeitung.de wrote:
> 
> Hi,
> 
> as I am receiving more and more emails when I will release a RedHat 6
> compatible ready to use version of my distribution and when which module is
> available for the distribution, I like to comment on this in the newsgroup,
> cause all of the senders seem to read this group. And so I can reach some
> other people too.
>
[..snipped..]
> 
> But I also have some plans concerning the distribution and the non x86
> versions. I like to setup some group writeable FTP directory on Starship
> Python so that I can give the contributors a password to upload their packages
> in a certain directory. That is just an idea, and I have to talk to the
> Starship admins about that. Otherwise I will offer this access on my own
> server and rsync the Starship dir from here.
> 
> Best regards,
> 
>     Oliver Andrich
> 

greetings,

I'm a big fan of your RPM packages, and I'm currently running python 1.5.2
on a redhat 6.0 based machine. ;)

as a general point, why don't we make up a web page that will allow the
uploading of (small) modules, to be approved by a moderator, via a webpage?
this would seem to simplify the problem of a 'group writable' directory. I
have some code I've been working on that might help push this along based on
medusa, and being able to have a file section is something I've wanted to
kick in the right direction for a while.

how about this.. if you can try to help me get 'getdatemodule' out to the
masses with your spec file skills (mine are a bit weak), I'll try to help
you with my web application skills (which I am stronger in). does that sound
cool?

regards,
J
-- 
|| visit gfd <http://quark.newimage.com/> 
|| psa member #293 <http://www.python.org/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>



From jeremy at cnri.reston.va.us  Sat Jun 12 13:24:23 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Sat, 12 Jun 1999 13:24:23 -0400 (EDT)
Subject: urllib: an alternative URLOpener
In-Reply-To: <7jt7uc$sh4@news.rhrz.uni-bonn.de>
References: <Pine.SUN.3.95-heb-2.07.990611155013.10151A-100000@sunset.ma.huji.ac.il>
	<7jt7uc$sh4@news.rhrz.uni-bonn.de>
Message-ID: <14178.38855.323759.698523@bitdiddle.cnri.reston.va.us>

I've been working on-and-off on a replacement for urllib that is more
easily customizable; they basic philosophy has been to avoid
inheritance as the customization mechanism.  Instead, a single
director class coordinates the activities of many small handlers that
implement one feature/protocol.

The resulting architecture makes it much easier to create a UrlOpener
that handles redirects (which is probably the crucial feature of
FancyUrlOpener) without getting all of the other cruft (like prompting 
the user to enter a password on stdin).

I've been delayed a bit because I wanted to add better support for
proxies, e.g. using a proxy only for URLs that match some pattern.
Since people are asking about this, I'll expedite my release plans
:-).  I'll try to make a useful release available early next week.

Jeremy




From Ovidiu.Predescu at p98.f112.n480.z2.fidonet.org  Wed Jun 30 21:02:41 1999
From: Ovidiu.Predescu at p98.f112.n480.z2.fidonet.org (Ovidiu Predescu)
Date: Thu, 01 Jul 1999 02:02:41 +0100
Subject: coverage tool for Python?
Message-ID: <000015ef@bossar.com.pl>

From: Ovidiu Predescu <ovidiu at cup.hp.com>

Hi,

Are there any coverage tools for Python?

Thanks,
-- 
Ovidiu Predescu <ovidiu at cup.hp.com>
http://www.geocities.com/SiliconValley/Monitor/7464/



From Oleg.Broytmann at p98.f112.n480.z2.fidonet.org  Wed Jun 30 19:40:19 1999
From: Oleg.Broytmann at p98.f112.n480.z2.fidonet.org (Oleg Broytmann)
Date: Thu, 01 Jul 1999 00:40:19 +0100
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
Message-ID: <000015bd@bossar.com.pl>

From: Oleg Broytmann <phd at emerald.netskate.ru>

Hi!

On Wed, 30 Jun 1999, Paul Prescod wrote:
> I find it fascinating that there are so many GUI toolkits. The GUI is more
> than twenty years right and we haven't agreed on the basic programming
> model or API yet??

   I think for many years that there is something wrong with the windowing
systems. Don't know what. May be the very abstraction of window?

> -- 
>  Paul Prescod  - ISOGEN Consulting Engineer speaking for only himself
>  http://itrc.uwaterloo.ca/~papresco
> 
> "The new revolutionaries believe the time has come for an aggressive 
> move against our oppressors. We have established a solid beachhead 
> on Friday. We now intend to fight vigorously for 'casual Thursdays.'
>   -- who says America's revolutionary spirit is dead?

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.





From gerrit.holl at pobox.com  Wed Jun 30 16:21:21 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Wed, 30 Jun 1999 22:21:21 +0200
Subject: Does the first argument of a method *has to be* self?
Message-ID: <19990630222121.A1121@optiplex.palga.uucp>

Hello,

I never saw code where the first argument of a method wasn't self. Is it
in theory possible?

groeten,
Gerrit.

-- 
The Dutch Linuxgames homepage:	http://linuxgames.nl.linux.org
Personal homepage:		http://www.nl.linux.org/~gerrit/

Discoverb is a python program (in several languages) which tests the words you
learned by asking it. Homepage: http://www.nl.linux.org/~gerrit/discoverb/



From Newbie at p98.f112.n480.z2.fidonet.org  Tue Jun 29 19:31:49 1999
From: Newbie at p98.f112.n480.z2.fidonet.org (Newbie)
Date: Wed, 30 Jun 1999 00:31:49 +0100
Subject: I am v. stupid, please help
Message-ID: <000015aa@bossar.com.pl>

From: "Newbie" <8499 at greenhead.ac.uk>


I have just downloaded Python 1.5 and I have 
had a brief look at the helpfiles and some 
source code and my stupid question is this :
How do you define a string and then use it 
for input, e.g,

name = input('What is your name : ')

This works for integers but not for strings,

Help Me Please!!

--Posted from EarthWeb Discussions. http://discussions.earthweb.com



From phd at emerald.netskate.ru  Mon Jun  7 12:10:52 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Mon, 7 Jun 1999 20:10:52 +0400 (MSD)
Subject: Calling .exe's from module
In-Reply-To: <375BEA17.B03B528D@ausinfo.com>
Message-ID: <Pine.LNX.4.04.9906072009480.7098-100000@emerald.netskate.ru>

On Mon, 7 Jun 1999, mike milliger wrote:

> Shouldn't the following module call test.exe? Could someone point me to
> the section in the programming python book which covers this
> functionality?   No objects persist from my executable, so I don't have
> to worry about ref counts and obj's returning to my module. i'm
> basically going to pkzip a list of filenames passed to my module as
> argv's.
> 
> #start
> import string,sys,os
> 
> # --------------------------------------------------------------------
> # run test
> 
> if __name__ == "__main__":
> 
>     os.system ("D:\Archiver\test.exe")

\A is just A, so you run
     os.system ("D:Archivertest.exe")

   Double backslashes:
>     os.system ("D:\\Archiver\\test.exe")

Oleg.
---- 
    Oleg Broytmann  National Research Surgery Centre  http://sun.med.ru/~phd/
           Programmers don't die, they just GOSUB without RETURN.




From Michael.P..Reilly at p98.f112.n480.z2.fidonet.org  Wed Jun 30 08:05:45 1999
From: Michael.P..Reilly at p98.f112.n480.z2.fidonet.org (Michael P. Reilly)
Date: Wed, 30 Jun 1999 13:05:45 +0100
Subject: I am v. stupid, please help
Message-ID: <000015b2@bossar.com.pl>

From: "Michael P. Reilly" <arcege at shore.net>

Klaus Alexander Seistrup <kas at maps.magnetic-ink.dk> wrote:
: Newbie wrote:

:> name = input('What is your name : ')
:> 
:> This works for integers but not for strings,

: Use raw_input() instead.

The input() function takes Python expressions, so strings must be
surrounded in quotes, tuples with parentheses, etc.  You would would get
a NameError attempting to enter an unquoted string, because the string
would be taken as a variable name (referencing variables that do not
exist raise a NameError exception).

The raw_input() function returns the string as entered (as per UNIX
cooked mode).

  -Arcege



From arcege at shore.net  Wed Jun 30 08:05:45 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 30 Jun 1999 12:05:45 GMT
Subject: I am v. stupid, please help
References: <3779d5f5@discussions> <3779E5B8.3CDB36CB@maps.magnetic-ink.dk>
Message-ID: <tKne3.314$6M6.111762@news.shore.net>

Klaus Alexander Seistrup <kas at maps.magnetic-ink.dk> wrote:
: Newbie wrote:

:> name = input('What is your name : ')
:> 
:> This works for integers but not for strings,

: Use raw_input() instead.

The input() function takes Python expressions, so strings must be
surrounded in quotes, tuples with parentheses, etc.  You would would get
a NameError attempting to enter an unquoted string, because the string
would be taken as a variable name (referencing variables that do not
exist raise a NameError exception).

The raw_input() function returns the string as entered (as per UNIX
cooked mode).

  -Arcege




From gerrit.holl at pobox.com  Sat Jun 12 06:12:21 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sat, 12 Jun 1999 12:12:21 +0200
Subject: Problem with Lists & Index command
In-Reply-To: <rNHvjEdhm5Pp-pn2-qhHq6lGpn6DZ@d185d18f8.rochester.rr.com>; from Benjamin Schollnick on Sat, Jun 12, 1999 at 12:22:19AM +0000
References: <rNHvjEdhm5Pp-pn2-zOTjsVjHebb4@d185d18f8.rochester.rr.com> <37617DFE.9BE0E03F@weihenstephan.org> <rNHvjEdhm5Pp-pn2-qhHq6lGpn6DZ@d185d18f8.rochester.rr.com>
Message-ID: <19990612121221.A18518@optiplex.palga.uucp>

On Sat, Jun 12, 1999 at 12:22:19AM +0000, Benjamin Schollnick wrote:
> From: rfreedm1 at nospam.rochester.rr.com (Benjamin Schollnick)
> Subject: Re: Problem with Lists & Index command
> To: python-list at python.org
> 
> On Fri, 11 Jun 1999 21:22:06, "Dr. Peter Stoehr" 
> <peter.stoehr at weihenstephan.org> wrote:
> 
> > Hi Benjamin,
> > 
> > maybe I missed something, but why don't you a dictionary ?
> 
> Because I'm still learning and haven't really read up up on
> dictionary's yet....
> 
> I'm only about 1/3 (somewhere around pg 150) in the 
> Python book I have >g<  (Out of somewhere around
> page 500 or so).  And dictionaries haven't been covered
> yet....
> 

Which book is that??
I think dictionairies should be in the tutorial!

regards,
Gerrit.

-- 
The Dutch Linuxgames website.		De Nederlandse Linuxgames pagina.
Everything about games on Linux.	Alles over spelletjes onder Linux.
Site address:				http://linuxgames.nl.linux.org
Personal homepage:			http://nl.linux.org/~gerrit/



From phd at emerald.netskate.ru  Wed Jun 30 11:40:19 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Wed, 30 Jun 1999 19:40:19 +0400 (MSD)
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
In-Reply-To: <377A204C.D843868D@prescod.net>
Message-ID: <Pine.LNX.4.04.9906301938120.15623-100000@emerald.netskate.ru>

Hi!

On Wed, 30 Jun 1999, Paul Prescod wrote:
> I find it fascinating that there are so many GUI toolkits. The GUI is more
> than twenty years right and we haven't agreed on the basic programming
> model or API yet??

   I think for many years that there is something wrong with the windowing
systems. Don't know what. May be the very abstraction of window?

> -- 
>  Paul Prescod  - ISOGEN Consulting Engineer speaking for only himself
>  http://itrc.uwaterloo.ca/~papresco
> 
> "The new revolutionaries believe the time has come for an aggressive 
> move against our oppressors. We have established a solid beachhead 
> on Friday. We now intend to fight vigorously for 'casual Thursdays.'
>   -- who says America's revolutionary spirit is dead?

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From rfreedm1 at nospam.rochester.rr.com  Mon Jun 28 17:37:15 1999
From: rfreedm1 at nospam.rochester.rr.com (Benjamin Schollnick)
Date: Mon, 28 Jun 1999 21:37:15 GMT
Subject: ConfigParser.py?
References: <rNHvjEdhm5Pp-pn2-ob9OW6dCFSRN@d185d18f8.rochester.rr.com> <7l876r$nrs@chronicle.concentric.net>
Message-ID: <rNHvjEdhm5Pp-pn2-obELMofjPdsy@d185d18f8.rochester.rr.com>

On Mon, 28 Jun 1999 09:17:31, Christopher Petrilli 
<petrilli at trump.amber.org> wrote:

> > 	I'm trying to use ConfigParser... And I believe I'm doing everything
> > write, except I don't see any documentation (including in the 
> > configparser.py file)
> > telling my what function(s) to use to write the file out to disk???
> 
> That's simple... it doesn't exist :-)

What's the rationale behind this???  It seems to be silly to allow 
read, and not some
sort of output???

		- Benjamin



From Fred.L..Drake at p98.f112.n480.z2.fidonet.org  Tue Jun 29 14:31:50 1999
From: Fred.L..Drake at p98.f112.n480.z2.fidonet.org (Fred L. Drake)
Date: Tue, 29 Jun 1999 19:31:50 +0100
Subject: Tkinter slide question
Message-ID: <00001587@bossar.com.pl>

From: "Fred L. Drake" <fdrake at cnri.reston.va.us>


David Stokes writes:
 > Code snipit:
 > Scale(label="Rescan",variable=self.yspeed,orient=HORIZONTAL,).grid(column=0)
 > 
 > currently returns PY_VAR12 instead of the slider value.

David,
  Appearantly, you're refering to self.yspeed to get the value of the
variable, although you seem to have initialized self.yspeed with an
instance of Tkinter.StringVar or one of it's friends.
  To get the value, use self.yspeed.get(); the object itself is simply 
a reference to a Tcl variable.  The get() method will return the value 
as the data type indicated by the name of the class you instantiated
to create self.yspeed.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From vvetwq at thankyou.com  Wed Jun 23 18:55:07 1999
From: vvetwq at thankyou.com (vvetwq at thankyou.com)
Date: 23 Jun 1999 22:55:07 GMT
Subject: Disney vacation  1597
Message-ID: <7krokb$4dt$5624@news.gdi.net>

Would you like to find out where can you get, the best priced ever, Disney vacation packages?

Go To: http://www.altavista.com/cgi-bin/query?pg=q&kl=XX&q=Disney+vacation+packages

and you will find.

Thank you.





 
cywvkkjbyjydmuviqcuzdngwdtbcb




From a.mueller at icrf.icnet.uk  Mon Jun  7 16:08:57 1999
From: a.mueller at icrf.icnet.uk (Arne Mueller)
Date: Mon, 07 Jun 1999 21:08:57 +0100
Subject: fork()
References: <375BE64A.EB997F24@icrf.icnet.uk> <87so840zcc.fsf@pc-hrvoje.srce.hr> <375BF721.4B1864F2@icrf.icnet.uk> <m31zfn99td.fsf@greebo.nodomain.de> <87hfojzudh.fsf@pc-hrvoje.srce.hr>
Message-ID: <375C26D9.327BA92@icrf.icnet.uk>

Hrvoje Niksic wrote:
> 
> Bernhard Herzog <herzog at online.de> writes:
> 
> > I fear that in reality the memory is copied after all. If they use
> > the dictionary in any way they are modifying refcounts of at least
> > some of the objects in the dictionary. From the OS's point of view
> > this counts as modifying the memory.
> 
> Eek.  You're right.  Refcounting truly sucks.  :-(

Hm, that means as long as I realy don't look at the data of the parent
the data is shared between parent and child. Thats funny - it's a
useless feature. As soon as a child reference that dictionary  it gets
it's own copy. 

What about shared memory, is it possible to put objects into a shared
memory segment, most advanced language can do that?

	greetings,

	Arne



From kjwolf at online.de  Thu Jun 17 23:35:15 1999
From: kjwolf at online.de (Klaus-Juergen Wolf)
Date: Fri, 18 Jun 1999 05:35:15 +0200
Subject: Problem with Curses
References: <ppgmj7.s7r.ln@solling.seismic.de>
Message-ID: <jpeck7.1661.ln@solling.seismic.de>

Klaus-Juergen Wolf <kjwolf at online.de> wrote:
 > (Python 1.5.2 on FreeBSD 3.2-RELEASE (i386))

 > I found that (after having compiled Python with cursesmodule.c -
 > testall.py succeeds) the following program gives a segmentation
 > fault:

As it appears,

-*-
import curses

curses.initscr()
curses.endwin()
-*-

alone does exactly the same. I've tested it in debugging environment,
where it doesn't happen, and I've checked the cursesmodule code, and I
still have no idea why this happens. (Inconsistency? But where?)

Stack trace says that it happens when the first actual write call to
libtermcap takes place.

The problem occurs under Solaris in the same way, while Linux doesn't
seem to know anything of it.

Anyone got an idea?

cu
 k.j.



From aahz at netcom.com  Wed Jun 16 10:26:18 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Wed, 16 Jun 1999 14:26:18 GMT
Subject: urllib: an alternative URLOpener
References: <Pine.SUN.3.95-heb-2.07.990611155013.10151A-100000@sunset.ma.huji.ac.il> <14178.38855.323759.698523@bitdiddle.cnri.reston.va.us> <14181.8918.956963.762261@bitdiddle.cnri.reston.va.us> <slrn7mef1g.1b0.davecook@sputnik.escnd1.sdca.home.com>
Message-ID: <aahzFDFC3u.IqH@netcom.com>

In article <slrn7mef1g.1b0.davecook at sputnik.escnd1.sdca.home.com>,
David M. Cook <davecook at home.com> wrote:
>
>Please forgive the naive question, but how does one get the "realm" info
>from a site?

IIRC, the challenge contains that info.  urllib should be handling this
automatically; if you want to see what it is directly, I don't know if
there's a Python mechanism, but you can do something like this:

% telnet www.foo.com 80
GET / HTTP/1.0\n
\n
-- 
                      --- Aahz (@netcom.com)

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

"That doesn't stop me from wanting to work all three of them over with
the clue stick for while, with no safewords allowed."  --abostick



From tim_one at email.msn.com  Wed Jun  9 01:47:54 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 9 Jun 1999 01:47:54 -0400
Subject: fork()
In-Reply-To: <7jjghd$9il$1@cronkite.cc.uga.edu>
Message-ID: <000001beb23b$9f17bc20$a69e2299@tim>

[Tim]
> OTOH, in a straightfoward mark-and-sweep GC scheme there will
> at least be a "mark bit" in each object header.  So under that,
> or under any form of compacting GC, the *entire live object space*
> will get copied, regardless of whether *you* reference anything in
> it or not.  So if RC truly sucks, what does that make GC <wink>?

{Graham]
> But the reality is that copying memory can in fact speed applications
> up! Think about locality of reference after you have copied all live
> blocks to the start of memory.

Context:  this offshoot of the thread was talking about memory use after a
fork() call in a copy-on-write implementation of fork.  So long as
bookkeeping info is stored in object headers, "real GC" is at a real
disadvantage, in that respect, compared to RC.  Simply because RC won't
touch something unless *you* do.

LOR is an unrelated issue, and harder to analyze.  Yes, copying can speed
applications.  It can also slow them.  "Although copying garbage collectors
have predominated in the past, recent studies suggest that the choice
between mark-sweep and copying garbage collection may well depend as much on
the behavior of the client program as on the inherent properties of the
garbage collection algorithm" [Jones & Lins, "Garbage Collection", Wiley,
1996].  The hoary old theoretical arguments really aren't worth crap in this
field <0.5 wink>.

modern-hardware-is-too-hard-to-predict-ly y'rs  - tim





From hochberg at psn.net  Tue Jun  1 21:00:25 1999
From: hochberg at psn.net (Tim Hochberg)
Date: Tue, 1 Jun 1999 19:00:25 -0600
Subject: Cannot store Sequences in Numeric arrays of Objects
References: <375453CE.3192@cs.bham.ac.uk>
Message-ID: <Pd053.429$gg5.24830@typ32b.nn.bcandid.com>

Stuart I Reynolds wrote in message <375453CE.3192 at cs.bham.ac.uk>...


[Stuart demonstrates that Numeric doesn't support PyObjects very well.
 In particular you can't store a sequence into an array of PyObjects]

The problem that you're running into is that Numeric is trying to be too
helpful! When you assign a sequence to an array it tries to unpack it into
the array. This works great under "normal" circumstances when you are using
Numeric arrays for numeric types. For example:

>>> a = zeros((3,3), Int)
>>> a[0] = (1,5,9)
>>> a
array([[1, 5, 9],
       [0, 0, 0],
       [0, 0, 0]])

Unfortunately when you try to store a sequence of PyObjects into a slot of
an array, it trys to unpack the sequence, decides that sequence is too long
for the slot and fails like so:

>>> b = zeros((3,3), PyObject)
>>> b[0,0] = (1,2,3)
Traceback (innermost last):
  File "<pyshell#6>", line 1, in ?
    b[0,0] = (1,2,3)
ValueError: array too large for destination

Numeric could probably be smarter about how it treats PyObjects, but as far
as I know, the PyObject feature is used very little.

-tim

PS I could swear I used to know a work around for this, but I sure can't
remember it now.





From dfan at harmonixmusic.com  Mon Jun  7 16:34:57 1999
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 07 Jun 1999 16:34:57 -0400
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7iqhkb$d7s@wiscnews.wiscnet.net> <7iu7ma$d33$1@cronkite.cc.uga.edu> <02hfonzsx2.fsf@rtp.ericsson.se> <02909zzou0.fsf@rtp.ericsson.se> <87zp2f4qrz.fsf@pc-hrvoje.srce.hr> <wkd7z858c3.fsf@turangalila.harmonixmusic.com> <87wvxfzxbg.fsf@pc-hrvoje.srce.hr>
Message-ID: <wk674z3g26.fsf@turangalila.harmonixmusic.com>

Hrvoje Niksic <hniksic at srce.hr> writes:

| Dan Schmidt <dfan at harmonixmusic.com> writes:
| 
| > | Since when does Python provide closures?
| > 
| > The bytecodehacks module allows the creation of closures as I
| > understand them.
| 
| I've never seen bytecodehacks.  Where can I get it?

ftp://starship.python.net/pub/crew/mwh/

Thanks to Jeremy Hylton for his post pointing out that closure
capability is still somewhat limited, even with bytecodehacks.  But
my original, carefully phrased, statement is still correct :)

-- 
                 Dan Schmidt -> dfan at harmonixmusic.com, dfan at alum.mit.edu
Honest Bob & the                http://www2.thecia.net/users/dfan/
Factory-to-Dealer Incentives -> http://www2.thecia.net/users/dfan/hbob/
          Gamelan Galak Tika -> http://web.mit.edu/galak-tika/www/



From aahz at netcom.com  Thu Jun  3 15:25:48 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Thu, 3 Jun 1999 19:25:48 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7j66b7$il5$2@cronkite.cc.uga.edu> <3756C693.2C910B29@usa.net> <7j6jnl$58f$3@cronkite.cc.uga.edu>
Message-ID: <aahzFCrnB0.DIE@netcom.com>

In article <7j6jnl$58f$3 at cronkite.cc.uga.edu>,
Graham Matthews <graham at sloth.math.uga.edu> wrote:
>
>Why is it so hard for people to see that adding a mark sweep collector
>*ON TOP OF* the existing ref count scheme essentially changes nothing.
>All the current ref count behaviour that people love will stay. The
>only additional behaviour will be that circular refs are mopped up
>correctly.

I don't disagree with this; I only claim that GC should only happen by
explicit call from the programmer.  I also claim that there should be a
debug mode for GC that merely informs the programmer about cyclic
references so that zie can locate their source (and that this should be
the primary mode for GC).
-- 
                      --- Aahz (@netcom.com)

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

"NRE is a lot like cocaine. Feels great while you're on it, makes you
feel very special and talented, and you become highly annoying to those
who aren't on it."  -- Stef



From tismer at appliedbiometrics.com  Wed Jun 16 07:27:39 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Wed, 16 Jun 1999 11:27:39 GMT
Subject: ANN: Stackless Python 0.2
References: <37628EAA.C682F16C@appliedbiometrics.com>
		<7k4fd4$c04$1@brick.cswv.com>
		<376636EA.84240129@appliedbiometrics.com>
		<7k6ttr$i5g$1@brick.cswv.com> <m3909lszww.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <37678A2B.6C85F4F6@appliedbiometrics.com>


Michael Hudson wrote:

> Ah, got this one; Naughty Chris's archive contains files with CRLF
> line endings. This doesn't bother gcc in most places, except where a
> backslash is used to continue a line, when it gets the screaming
> heebie-jeebies. Eliminate the nasty ^M's and it all gets better. Also
> diff stops reporting that every single line in the file as changed,
> which was confusing me.

Sorry about that. I developed under Windows, which silently
insertes the CR/LFs for (against) me.

The next release (coming soon) will run through a filter
before I post it.

ciao - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From matz at netlab.co.jp  Thu Jun  3 21:34:15 1999
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 04 Jun 1999 10:34:15 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
 <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>
 <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net>
 <7j34l3$hbj$1@nnrp1.deja.com> <7j66b7$il5$2@cronkite.cc.uga.edu>
 <3756C693.2C910B29@usa.net> <7j6jnl$58f$3@cronkite.cc.uga.edu>
Message-ID: <877lpkbvfs.fsf@ev.netlab.co.jp>

graham at sloth.math.uga.edu (Graham Matthews) writes:

|Why is it so hard for people to see that adding a mark sweep collector
|*ON TOP OF* the existing ref count scheme essentially changes nothing.
|All the current ref count behaviour that people love will stay. The
|only additional behaviour will be that circular refs are mopped up
|correctly.

They may be worrying about resurrection of the dead objects by __del__
finalizers, which is solvable by Java approach (maybe inefficient though).

						matz.



From gmcm at hypernet.com  Tue Jun  1 11:01:43 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 1 Jun 1999 10:01:43 -0500
Subject: Python 2.0
In-Reply-To: <m3r9nwi6uf.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <1283872945-30734792@hypernet.com>

Michael Hudson writes:

[re Py_TRACE_REFS]
> I've never resorted to this approach, tending always to make cyclic
> references go away be staring at them very, very hard.

Ay. This is my approach. WIth the added advantage that now I find I 
can hypnotize a brick, simply by starting at it.

that-one's-already-asleep-ly y'rs

- Gordon



From ovidiu at cup.hp.com  Wed Jun 16 15:44:49 1999
From: ovidiu at cup.hp.com (Ovidiu Predescu)
Date: Wed, 16 Jun 1999 19:44:49 +0000
Subject: Emacs package to the Python debugger?
Message-ID: <3767FEB1.87D18512@cup.hp.com>

Hi,

Is there any package that allows debugging Python programs from within
Emacs? Something that is integrated with Emacs' GUD would be very
convenient.

Thanks,
-- 
Ovidiu Predescu <ovidiu at cup.hp.com>
http://www.geocities.com/SiliconValley/Monitor/7464/



From Gordon.McMillan at p98.f112.n480.z2.fidonet.org  Wed Jun 30 14:14:42 1999
From: Gordon.McMillan at p98.f112.n480.z2.fidonet.org (Gordon McMillan)
Date: Wed, 30 Jun 1999 19:14:42 +0100
Subject: Mailbox, tell and read
Message-ID: <000015f5@bossar.com.pl>

From: "Gordon McMillan" <gmcm at hypernet.com>

Paul Prescod writes:

> The mailbox message objects simulate a file interface by seeking
> through a mailbox. A read() returns the entire message. It does this
> by seeking to the start, reading to the message end, and returning
> the data. The problem is that message offsets are derived by
> file.tell() which I presume is in terms of bytes. But then it does a
> read() based on the offsets and read() is in terms of characters. 
> 
> These don't match up on Windows because of CR/LF pairs. I think that
> the right fix is to never use tell() and instead keep track of the
> location by counting characters. Does that make sense? -- 

How 'bout opening in binary, and munging line endings yourself just 
before returning the message? If you can't use tell(), you can't 
trust seek(), either.

- Gordon




From thomas at xs4all.nl  Fri Jun  4 12:13:14 1999
From: thomas at xs4all.nl (Thomas Wouters)
Date: Fri, 4 Jun 1999 18:13:14 +0200
Subject: Alien whitespace eating nanovirus strikes again!
In-Reply-To: <3757F28B.BF6411FE@cwcom.net>; from Daniel Faulkner on Fri, Jun 04, 1999 at 04:36:44PM +0100
References: <375500C9.742DC539@pop.vet.uu.nl> <3757F28B.BF6411FE@cwcom.net>
Message-ID: <19990604181314.K22614@xs4all.nl>

On Fri, Jun 04, 1999 at 04:36:44PM +0100, Daniel Faulkner wrote:

> How many messages are there about this white space stuff? Looks like
> millions whats it all about?

Whitespace, oddly enough. Read the articles if you want more info.

Yes-i'm-the-one-Tim-said-'faq-you'-to-not-so-long-ago-ly y'rs Thomas.

-- 
Thomas Wouters <thomas at xs4all.net>

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



From esuzm at primrose.csv.warwick.ac.uk  Fri Jun 18 16:19:38 1999
From: esuzm at primrose.csv.warwick.ac.uk (Andrew Clover)
Date: 18 Jun 1999 20:19:38 GMT
Subject: Installing ODBC on PythonWin
References: <7k4grk$b19$1@nnrp1.deja.com>
Message-ID: <7ke9kq$3c2$1@holly.csv.warwick.ac.uk>

eeghead84 at yahoo.com wrote:

> It is written that the ODBC module is included in the pythonwin
> distribution but >>>import ODBC says otherwise. I've d/e'd
> (downloaded/executed) Win32, but it died and said that it was looking
> for version 1.4!

 If you download the latest Python and Win32 from
http://www.python.org/download/download_windows.html it should be fine.
You'll want to write "import dbi, odbc" though.

 Now for my problems.

 I've got an Access database accessed by Python scripts under CGI. However,
sometimes, calls to odbc.odbc(dsn) hang up and never return. (Eventually the
web server times out and kills the process.) This seems to happen randomly,
but increasingly. Does anyone have any idea why? A fault with the odbc
module, or Access, or what?

-- 
This posting was brought to you by And Clover.
(Sorry.)



From graham at sloth.math.uga.edu  Tue Jun  1 11:27:12 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 1 Jun 1999 15:27:12 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
Message-ID: <7j0u4g$2dm$1@cronkite.cc.uga.edu>

Paul Prescod (paul at prescod.net) wrote:
: But I think that the bigger problem with "real GC" is that C programs
: often hang on to references to Python objects and any kind of compacting
: GC would invalidate those references.

You have two classes of references, compactable and non-compactable.
That solves that little problem.

graham
-- 
      This mountain is not visible from any inhabitable place. 
       It has no name, and since it's discovery in 1856 by a 
      British survey team, it has mysteriously resisted every
    attempt to name it. The mountain is known simply and starkly
     by it's original designation on the surveyor's map ... K2.



From phxken at netzone.com  Sun Jun 20 16:15:56 1999
From: phxken at netzone.com (Farrell Kenimer)
Date: Sun, 20 Jun 1999 20:15:56 GMT
Subject: grail browser failure
Message-ID: <376D4A9E.B9B90C3D@netzone.com>

fails with something like isabs not available
I can import grail and after a try or two I can also import Browser
How would one try to run grail after it's imported?


--
Farrell Kenimer of Phoenix, Arizona  USA


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/19990620/57cfa0e4/attachment.html>

From news at dorb.com  Sat Jun 19 13:20:34 1999
From: news at dorb.com (Darrell)
Date: Sat, 19 Jun 1999 13:20:34 -0400
Subject: GC play
References: <mp8a3.324$U6.2508@newsr2.twcny.rr.com> <3769A840.A5AE6778@compaq.com> <p6qlndgrwtk.fsf@pandora.inst-inf-1.hu-berlin.de>
Message-ID: <8mQa3.948$U6.4000@newsr2.twcny.rr.com>

When searching for leaks I can live with less preformance.
It doesn't seem that tracing works for some cases.
The "All Done" comes out with any mention of l.append(l1)
So I don't know how to record the traceback.
Would I be wasting my time to think about a new or modified sys.trace ?

def t1(f,e,a):
 print 'T1:',e
 return t1

def t2(f,e,a):
 print 'T2:',e
 return t2

sys.setprofile(t1)
sys.settrace(t2)

def myFunc(x):
 print x

l=[]
l1=[]
l.append(l1)
print 'All Done:'
myFunc('PassedValue')

******** Output *********
All Done:
T2: call
T1: call
T2: line
T2: line
PassedValue
T2: return
T1: return
T1: return

--
--Darrell
Martin von Loewis <loewis at informatik.hu-berlin.de> wrote in message
news:p6qlndgrwtk.fsf at pandora.inst-inf-1.hu-berlin.de...
> Greg Ewing <greg.ewing at compaq.com> writes:
>
> > I don't think there's anything you can find out about the source
> > of any other kind of object.
>
> Although I wonder: Would it be possible to record a traceback for each
> object allocated, to support purify-like leak analysis? It would
> surely be very inefficient, and you'd also need to release all the
> locals in the traceback - but would it work?
>
> Regards,
> Martin





From too at pk.highway.ne.jp  Mon Jun 28 10:30:41 1999
From: too at pk.highway.ne.jp (Thooney Millennier)
Date: Mon, 28 Jun 1999 23:30:41 +0900
Subject: signal handler
References: <376F87CB.DC3802E6@pk.highway.ne.jp> <3772CD4B.38677556@cup.hp.com>
Message-ID: <37778711.A2752414@pk.highway.ne.jp>

I am sorry for my late reply,
and thank for your suggestion.
I did't know functions such as
"getpid()","kill()".

Well, I tried and realized...
If you use "kill(getpid(),SIGINT)",
it works. (ie. callback ok!)
Instead,if you send "Control-C" by keyboard input,
it does not. (nothing happens,signal ignored)

It is confusing.

Thooney.
 
Ovidiu Predescu wrote:
>> Hi, I want to use "signal()" in my program.
>> 
>>   import signal;
>>   def func(arg1,arg2):
>>       DO_SOMETHING_GOOD;
>>       print "OK!";
>>   signal.signal(SOME_SIGNAL,func);
>> 
>> It seems "func()" is not called
>> when the program gets signal "SOME_SIGNAL".
>> Why?
>
> It looks like you forgot to send the signal. One way you could 
> implement this code is:
>
> from signal import signal, SIGINT
> from os import getpid, kill
>
> def signalHandler (signum, frame):
>    print 'signal handler called for signal %d' % (signum,)
>
> signal (SIGINT, signalHandler)
> kill (getpid (), 
>



From eugene.leitl at lrz.uni-muenchen.de  Wed Jun 23 16:23:55 1999
From: eugene.leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Wed, 23 Jun 1999 13:23:55 -0700 (PDT)
Subject: Thanks for Python...
In-Reply-To: <37716901.43CB604@lockstar.com>
References: <376ead64@anonymous.newsfeeds.com>
	<VrDQRddcbbs9-pn2-QdftOXYk6Hdd@Fortress>
	<37716901.43CB604@lockstar.com>
Message-ID: <14193.16609.554364.194800@lrz.de>

Mordy Ovits writes:

 > It sure is nifty to be coding during my commute :-)
 
More than a bit offtopic, but you can program in Forth if you use
free Quartus here http://www.interlog.com/~nbridges/quartus.html
After I downloaded a debug ROM image from 3com I was able to run it
on http://xcopilot.cuspy.com/ under Linux very nicely.

*Is* there a Python for the PalmPilot out there? 'cause I would be
buying one if there was. I've got a Garmin 25 OEM GPS board I'd like
to put to some use...

-- Eugene

P.S. Our headers seem to be screwed. python-list at python.org doesn't
appear in Reply-to.



From Voynow  Fri Jun 25 15:50:46 1999
From: Voynow (Voynow)
Date: Fri, 25 Jun 1999 11:50:46 -0800
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773c3a0.102825445@news.oh.verio.com>
Message-ID: <930340248.9300@www.remarq.com>

Has anyone SWIG'd or otherwise wrapped FLTK?

SV



**** Posted from RemarQ - http://www.remarq.com - Discussions Start Here (tm) ****



From mnot at pobox.com  Sun Jun 20 02:47:44 1999
From: mnot at pobox.com (Mark Nottingham)
Date: Sun, 20 Jun 1999 06:47:44 GMT
Subject: Proxying HTTP - end of message?
References: <7kgkbk$uso$1@nnrp1.deja.com>
Message-ID: <009501bebae8$cd9314e0$0301a8c0@cbd.net.au>

Are you using Connection: close and friends? Your browser may be setting up
a persistant connection...


----- Original Message -----
From: Andrew Cooke <andrew at andrewcooke.free-online.co.uk>
Newsgroups: comp.lang.python
To: <python-list at cwi.nl>
Sent: Sunday, June 20, 1999 3:34
Subject: Proxying HTTP - end of message?


>
>
> Hi,
>
> I'm writing a small http proxy (I know a million already exist -
> my program isn't important, understanding the bug is).  I have a
> couple of threads, spinning away, one copying from client to server
> and one from server to client.  When there is no more data to read
> from the server the blocking read returns None and I close both
> sockets (one socket is proxy-client, the other proxy-server).
>
> But my browser continues to wait.  If I press "stop" I have all the
> page, but the browser thinks the download has been interrupted.
>
> What am I missing?  It appears as though the browser isn't detecting
> that the connection has been closed - why not?
>
> For what it's worth, this is on NT, I'm extending SocketServer, and
> using 1.5.2.  But I suspect this is a problem with my logic rather
> than any piece of software....
>
> Any sugegstions appreciated (I can post/forward code - it's my own
> work, not my employer's).
>
> Thanks,
> Andrew
>
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
>
>




From dalke at bioreason.com  Thu Jun 17 00:33:18 1999
From: dalke at bioreason.com (Andrew Dalke)
Date: Wed, 16 Jun 1999 22:33:18 -0600
Subject: Solution to finalisation problem [Re: fork()]
References: <000301beb6c9$a01d2960$b2a22299@tim> <7k4gec$6vg$1@cronkite.cc.uga.edu> <37683DD6.4C42DC3E@compaq.com>
Message-ID: <37687A8E.7F9DB5CF@bioreason.com>

Greg Ewing <greg.ewing at compaq.com> said:
> I think what we need at this point is some concrete examples.
> What sorts of things do people actually *do* with __del__
> methods at the moment?

I use them for a few things.  One is to free memory allocated
from a C library, or to remove a temporary file.  This is
effectively your statement:

> As far as I can see, the only thing you really need a
> finalizer for is freeing an external resource.

Most of the uses of __del__ in the std. Python lib. is of this form.



I use __del__ to implement the "resource acquisition is initialization"
idea, as in:

class PreserveMod:
    """Internal class which returns the molecule to its original "mod"
state"""
    def __init__(self, mol):
        self.mol = mol
        self.modify = mol.mod
    def __del__(self):
        self.mol.mod = self.modify

Looking at the std. lib., audiodev also uses this style.


Finally, I use __del__ to free cycles which might otherwise arise, as
in this (untested) example:

class Molecule:
  def __init__(self):
    self.atoms = []
    self.bonds = []
  def __del__(self):
    for atom in self.atoms:
      del atom.bonds

class Atom:
  def __init__(self, name):
    self.name = name
    self.bonds = []
  def add_bond(self, bond):
    self.bonds.append(bond)

class Bond:
  def __init__(self, atoms):
    assert len(atoms) == 2
    self.atoms = atoms

mol = Molecule()
a1 = Atom("C")
a2 = Atom("O")
mol.atoms.append(a1)
mol.atoms.append(a2)
b = Bond([a1, a2])
a1.add_bond(b)
a2.add_bond(b)


This has cycles between the atoms and the bonds, but as neither of
them point back to the molecule, I can use the deletion of the
molecule (ref count goes to 0) to trigger a cleanup of the cycles.

						Andrew
						dalke at bioreason.com



From sjmachin at lexicon.net  Mon Jun 21 08:55:28 1999
From: sjmachin at lexicon.net (John Machin)
Date: Mon, 21 Jun 1999 12:55:28 GMT
Subject: NameError Problem
In-Reply-To: <7kh6kn$6cr$1@eskinews.eskimo.com>
References: <7kh6kn$6cr$1@eskinews.eskimo.com>
Message-ID: <19990621125609298.AAA130.185@pc6558.cmutual.com.au>

A non-text attachment was scrubbed...
Name: not available
Type: text/enriched
Size: 8378 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/19990621/6de29a2f/attachment.bin>

From gmcm at hypernet.com  Wed Jun 16 15:58:44 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 16 Jun 1999 14:58:44 -0500
Subject: Newbie again: computing attributes on the fly
In-Reply-To: <Pine.OSF.4.05.9906161916040.10750-100000@pinax.Mathematik.Uni-Bielefeld.DE>
References: <1282571882-108998091@hypernet.com>
Message-ID: <1282559115-109766087@hypernet.com>

[Olaf on my very simple getattr hack]:

> Yes, I definitely want to generalize that. The problem is that that
> method may be defined in a superclass, so "self.__dict__" won't find
> it. AFAIK (which may be not be very far), I have to use "hasattr"
> and/or "getattr", which in turn will call "__getattr__".

Oh, getattr isn't bad. Wait 'till you try setattr! Anyway, this 
should demonstrate how to control the recursion:

class X:
  def __getattr__(self, nm):
    print "Looking for", nm
    mthdnm = "calc_%s" %nm
    print "Looking in instance..."
    mthd = self.__dict__.get(mthdnm, None)
    if mthd is None:
      print "Looking in class hierarchy..."
      try:
        mthd = getattr(self.__class__, mthdnm)
      except AttributeError:
        pass
    if mthd:
      val = mthd(self)
      self.__dict__[nm] = val
      return val
    raise AttributeError("no attribute %s" % nm)

x = X()
try:
  print x.foo
except AttributeError:
  print "foo not found in x"

class Y(X):
  def calc_foo(self):
    return 33.0 * 17

y = Y()
print y.foo
print "second time..."
print y.foo

Which should yield:

Looking for foo
Looking in instance...
Looking in class hierarchy...
foo not found in x
Looking for foo
Looking in instance...
Looking in class hierarchy...
561.0
second time...
561.0

- Gordon



From kullstam at ne.mediaone.net  Tue Jun 15 18:27:24 1999
From: kullstam at ne.mediaone.net (Johan Kullstam)
Date: 15 Jun 1999 18:27:24 -0400
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <3762988E.C94AA252@iname.com> <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu> <3763C746.5CA8FE78@iname.com> <m3u2sby4gf.fsf@soma.andreas.org> <87pv2zjhhl.fsf@home.ivm.de> <ekvn1y2efz0.fsf@gnoll.pvv.ntnu.no> <slrn7mb2oq.lpd.wtanksle@dolphin.openprojects.net> <37660FEC.D4F3371B@iname.com> <37661C02.9CA6E258@cfmu.eurocontrol.be> <376650FA.1AED1606@iname.com>
Message-ID: <m2d7yxrtf7.fsf@sophia.axel.nom>

Fernando Mato Mira <matomira at iname.com> writes:

> I seem to have forgotten C, too..

you lucky bastard!

-- 
J o h a n  K u l l s t a m
[kullstam at ne.mediaone.net]
Don't Fear the Penguin!



From andrew at andrewcooke.free-online.co.uk  Mon Jun 14 03:42:11 1999
From: andrew at andrewcooke.free-online.co.uk (Andrew Cooke)
Date: Mon, 14 Jun 1999 07:42:11 GMT
Subject: Python 2.0
References: <000b01beb21f$a17ff340$2fa22299@tim> <u37lpdssbq.fsf@ares.sys1.ngo.okisoft.co.jp>
Message-ID: <7k2boi$j9m$1@nnrp1.deja.com>

In article <u37lpdssbq.fsf at ares.sys1.ngo.okisoft.co.jp>,
  Hisao Suzuki <suzuki611 at okisoft.co.jp> wrote:
> | >[Salvador]
> | > I have been using Java since it was invented and I haven't
programed a
> | > finaliser ever. They should run under so unknown and inconsistent
> | > environment that they are not usable at all (except for freeing
external
> | > resources).
>
> I have not ever programmed any finalizer, too, while I have
> programmed a lot of codes in Java.

I know this is getting off-topic, but just for general info, one example
of where a finalizer is useful in Java is when using JNI (the Java/C
interface) - it is often useful for a Java object to mirror a C
structure, and when the Java object is GCed you need to free() the C
stuff (which is an example of "freeing external resources", I guess).

Andrew


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From billtut at microsoft.com  Thu Jun 17 03:40:04 1999
From: billtut at microsoft.com (Bill Tutt)
Date: Thu, 17 Jun 1999 07:40:04 GMT
Subject: do anonymous pipes normally work on NT?
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8100DB90DE4@RED-MSG-50>

> From: Georg Mischler [mailto:schorsch at schorsch.com]

> The Documentaion provided with VC++ about this stuff
> it not very helpful, so you have to try and guess.
> Here's what I figured by now: The CreateProcess()
> function has several ways to pass handles to child
> processes. The way I tried it (passing through
> StartupInfo.hStdXxx) only seems to work with disk files,
> and not with pipes.
> 

It may seem that way, but thats far from the truth.
Because win32popen.cpp uses StartupInfo.hStdxxxx and it works quite well.
The real cause of this problem is that there is a bug in win32process's
startupinfo support. (see info at bottom of email) 

> The other way is to assign the handles to the stdxxx
> of the current process, and inherit them to the
> child. This does work with pipes, and is fundamentally
> incompatible with the first approach. In fact, as soon
> as you just tell it you want to pass any handles in the
> StartupInfo, by assigning STARTF_USESTDHANDLES to it's
> dwFlags attribute, inheritance is dead. Still, if
> you *want* to pass (file) handles through the
> StartupInfo object, you nonetheless *need* to set the
> bInheritHandles parameter of CreateProcess() to TRUE.
> Go figure...
> 

Well, inheritance of stdin, stdout, stderr is dead, yes. (All other
inhertible handles will be valid in the child process)
They sure like to make life confusing don't they? :)

> The least obvious thing though, which I only found
> by comparing the working code with mine very carefully,
> is that for inheritance to work, the dwCreationFlags
> must *not* contain win32process.DETACHED_PROCESS.
> The VisualStudio docs (and consequently the win32proc
> help file) tell us that this would deny acess to the
> *console* of the parent process. In fact, it appears
> to cut all connections between the two processes.
> 

In theory, this should only disallow the new process from acessing the
current program's NT Console (if any). This should have nothing to do with
hStdXxx being anonymous pipe handles. Although after some experimentation it
certainly seems to be doing just a little bit more than revoking access to
the NT console. i.e. passing DETACHED_PROCESS in the example below works
fine, but it completly failed if I tried to fire up Mark's cool debugger.
(It asserted at objcore.cpp:45 in MFC land...)

> In my latest experiments this *does* work if the parent
> process manages to get the handles across. CreateProcess()
> has so many mutually exclusive combinations of parameters
> that it's just hard to find the right ones for a particular
> purpose.
> 

Hrm, now when I go to retest what I did last night, I wasn't paying close
enough attention.  If i do "cat.py < xxx > yyy" it complains because
something in the file extension association code is screwing things up. :(

python cat.py < xxxx > yyy works just fine.

> I don't mind setting and resetting the std handles in the
> parent, as long as I get the result I want. But then, I'm
> also trying to get along with as little knowledge as possible
> about Windows programming. I assume you look at it from a
> slightly different angle.
> 

Heh, if you know more about it you can do more useful things when life
becomes complicated. :) I'm just digging through the docs like you, I don't
work on the NT team and can't go look up the source for CreateProcess(). 

Attached below is the fix information, along with a Georg's example slightly
tweaked to be useful.

Bill

Patch info:

Alter the PyHANDLE_Check() branch of win32processmodule_win32.cpp:sethandle
to:
		*pobHandle = v;
		if (PyWinObject_AsHANDLE(v, ph))
		{
			Py_INCREF(v);
		}
		else
		{
			rc = -1;
		}
The cast that was there before was setting *ph equal to the Python reference
count on v.

Working sample:

'''cat.py
a "enhanced" version of the good old cat unix filter.
'''

import sys
try:
    # write status to a file, in order not to lose it.
    f = open('log.txt', 'a')
    f.write('before write\n')
    sys.stdout.write('huh?')
    f.write('after write, before read\n')
    data = sys.stdin.read()
    f.write('data: ' + data)
    f.write('after read\n')
    f.write('in loop, before write\n')
    sys.stdout.write(data)
except:
    import traceback, string
    tb_strs = traceback.format_exception(
            sys.exc_type,
            sys.exc_value,
            sys.exc_traceback)
    f.write(string.join(tb_strs))

# end of cat.py

'''runproc.py

start a process with three inherited pipes.
Try to write to and read from those.
'''

import win32pipe
import win32file
import win32process
import win32security
import win32con

class Process:
    def run(self, cmdline):
        # security attributes for pipes
        sAttrs = win32security.SECURITY_ATTRIBUTES()
        sAttrs.bInheritHandle = 1

        # create pipes
        hStdin_r,  self.hStdin_w  = win32pipe.CreatePipe(sAttrs, 0)
        self.hStdout_r, hStdout_w = win32pipe.CreatePipe(sAttrs, 0)
        self.hStderr_r, hStderr_w = win32pipe.CreatePipe(sAttrs, 0)

        # set the info structure for the new process.
        StartupInfo = win32process.STARTUPINFO()
        StartupInfo.hStdInput  = hStdin_r
        StartupInfo.hStdOutput = hStdout_w
        StartupInfo.hStdError  = hStderr_w
        StartupInfo.dwFlags = win32process.STARTF_USESTDHANDLES 

        # start the process.
        hProcess, hThread, dwPid, dwTid = win32process.CreateProcess(
                None,   # program
                cmdline,# command line
                None,   # process security attributes
                None,   # thread attributes
                1,      # inherit handles, or USESTDHANDLES won't work.
                        # creation flags. Don't access the console.
                win32process.DETACHED_PROCESS,
                None,   # no new environment
                None,   # current directory (stay where we are)
                StartupInfo)
        # normally, we would save the pid etc. here...

        err, bytes = win32file.WriteFile(self.hStdin_w, 'hmmm\n')
        print 'wrote:', bytes, 'bytes'

        res, str = win32file.ReadFile(self.hStdout_r, 7)
        if res == 0: print 'read:', str
        else: print 'read nothing on stdin.'

if __name__ == '__main__':
    p = Process()
    p.run(r'd:\pythondevtree\python\dist\src\pcbuild\python_d.exe
d:\cat.py')

# end of runproc.py



From rasumner at bach.wisdom.weizmann.ac.il  Tue Jun 29 09:54:33 1999
From: rasumner at bach.wisdom.weizmann.ac.il (Reuben Sumner)
Date: 29 Jun 1999 13:54:33 GMT
Subject: PySequence_Check
Message-ID: <slrn7nhjv9.lt6.rasumner@bach.wisdom.weizmann.ac.il>

How can I do the equivalent of a PySequence_Check in Python?
GvR told me that the question was illdefined (if I explained it correctly)
and to ask here.  It seems that some of the functions in abstract.h
are available from python (len(), callable()) and others are not
(PySequence_Check, PyNumber_Check).

What am I missing?  I have a nested structure of lists and tuple and I
just want to know when I have bottomed out.  I can check for tuple or
list indivually but that is ugly and inflexible.  I could just try
indexing and see if there was an exception but that is not much better.

Reuben



From Paul.Prescod at p98.f112.n480.z2.fidonet.org  Wed Jun 30 19:53:38 1999
From: Paul.Prescod at p98.f112.n480.z2.fidonet.org (Paul Prescod)
Date: Thu, 01 Jul 1999 00:53:38 +0100
Subject: Mailbox, tell and read
Message-ID: <000015f1@bossar.com.pl>

From: Paul Prescod <paul at prescod.net>

The mailbox message objects simulate a file interface by seeking through a
mailbox. A read() returns the entire message. It does this by seeking to
the start, reading to the message end, and returning the data. The problem
is that message offsets are derived by file.tell() which I presume is in
terms of bytes. But then it does a read() based on the offsets and read()
is in terms of characters. 

These don't match up on Windows because of CR/LF pairs. I think that the
right fix is to never use tell() and instead keep track of the location by
counting characters. Does that make sense?
-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for only himself
 http://itrc.uwaterloo.ca/~papresco

"I don't want you to describe to me -- not ever -- what you were doing
to that poor boy to make him sound like that; but if you ever do it
again, please cover his mouth with your hand," Grandmother said.
        -- John Irving, "A Prayer for Owen Meany"



From jmagave at amazon.com.br  Mon Jun 28 00:58:06 1999
From: jmagave at amazon.com.br (Jayr Magave)
Date: Mon, 28 Jun 1999 01:58:06 -0300
Subject: Python Comm Lib
Message-ID: <67Dd3.271$Xw4.18913@typ32b.nn.bcandid.com>

I'm trying to write a TAP wraper module (pager protocol www.pcia.com) in
Python as my 1st project. I am currently running Windows 32 bits and I am
using Serial I/O I found at Win95/NT area in www.python.org.

The project is:

1. Setup a web homepage FORM where users may  fill a TEXTEDIT with message
as (PHONE, PIN, MSG)
2. The server should dial-up the PAGER SERVICE, probably throug CGI Python
module, say:

    def TAP_send('222-9999', '123','Hello World')


3. When connected SEND MSG, then disconnects.

Any hints?

Thanx

Jair





From olipt at mayo.edu  Wed Jun 16 18:23:39 1999
From: olipt at mayo.edu (Travis Oliphant)
Date: Wed, 16 Jun 1999 17:23:39 -0500
Subject: BiBTeX module?
In-Reply-To: <joe-160619991358444033@chinacat.salk.edu>
References: <joe-160619991358444033@chinacat.salk.edu>
Message-ID: <Pine.LNX.4.10.9906161722220.9477-100000@us2.mayo.edu>

I don't have a BiBTeX module but I do have a python program that converts
medline or OVID references to the BiBTeX format.  It does not parse BiBTeX
files --- it just writes them.

Have you used bibtool?

Travis





From tim_one at email.msn.com  Wed Jun  9 22:22:05 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 9 Jun 1999 22:22:05 -0400
Subject: Python 2.0
In-Reply-To: <871zfsbcpk.fsf@ev.netlab.co.jp>
Message-ID: <000a01beb2e8$087fc2a0$af9e2299@tim>

[Hisao Suzuki]
> Me too.  And I am afraid that some evangelists would do nothing
> but _only_ point out Python's some so-called fault in order to
> drive us into convert to their favorite or invented language...

[Yukihiro Matsumoto]
> Sorry for bothering the newsgroup.

Hmm.  The level of paranoia in this thread is quite remarkable!  I don't
think Suzuki was talking about you (or Ruby) here, and your contributions to
the thread have been civil and helpful.  Always said you should hang out
here more often <wink>.

> But I couldn't resist pointing out misunderstanding about "real GC"
> among some part of Python users.
>
> 	http://www.iecc.com/gclist/GC-faq.html

Nice link!  Thank you.

> It's totally OK for Python to use ref counting.

Ya, but it's not OK for Ruby to use Scheme gc, so stop doing that <wink>.

like-any-of-this-is-going-to-change-anything-ly y'rs  - tim





From paulb at infercor.no  Wed Jun  2 05:39:46 1999
From: paulb at infercor.no (Paul Boddie)
Date: Wed, 02 Jun 1999 11:39:46 +0200
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <7j0urf$2dm$3@cronkite.cc.uga.edu>
Message-ID: <3754FBE2.4BFD7319@infercor.no>

Graham Matthews wrote:
> 
> Paul Boddie (paulb at infercor.no) wrote:
> : As has been discussed before on this newsgroup/list, such resources get
> : overlooked in many a garbage collection scheme. I'm sure a perusal of
>                 ----
> : various news/list archives would provide a substantial amount of relevant
> : literature.
> 
> That word "many" is your key Paul. Just because many a garbage collection
> scheme works that way doesn't mean all do, or all have to. It's entirely
> an implementation dependent thing.

Absolutely. I'm not ruling anything out here. ;-)

> You are creating a strawman argument that GC cannot properly free resources
> like files. But some GC do just that so what is your point?

Well, my statement was actually an invitation for more enlightened contributors
to give examples of such schemes, and this has evidently had the intended
effect. Since you have probably played as important a role as anyone in the
previous discussions I mentioned, Graham, perhaps you could indeed enlighten us.
I really don't have any objections to "full" GC, provided it is done in a way
which avoids the pitfalls described in previous posts.

Paul



From Fredrik.Lundh at p98.f112.n480.z2.fidonet.org  Wed Jun 30 12:33:34 1999
From: Fredrik.Lundh at p98.f112.n480.z2.fidonet.org (Fredrik Lundh)
Date: Wed, 30 Jun 1999 17:33:34 +0100
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
Message-ID: <000015be@bossar.com.pl>

From: "Fredrik Lundh" <fredrik at pythonware.com>

Michael P. Reilly wrote:
> Hmm... something about that argument makes it circularly invalid.
> Non-idle event a comes in, but must be handled after idle event d.
> Event d can only execute when event a is taken care of.  Sounds like
> deadlock to me. ;)

hmm.  I thought I'd meant that idle tasks kept piling up
when the windowing system generates a lot of events
at once (like when you're redrawing an entire system).
instead of redrawing everything at once, Tkinter waits
until the window event queue is empty...

</F>



From Axel.Schmitz-Tewes at p98.f112.n480.z2.fidonet.org  Wed Jun 30 06:43:06 1999
From: Axel.Schmitz-Tewes at p98.f112.n480.z2.fidonet.org (Axel Schmitz-Tewes)
Date: Wed, 30 Jun 1999 11:43:06 +0100
Subject: win32com.client -- beginner question
Message-ID: <000015a6@bossar.com.pl>

From: Axel Schmitz-Tewes <uzs59g at uni-bonn.de>

Hello,

I ' ll try to use the comclient ability of python. In the Doku I have
found
the example:

o = win32com.client.Dispatch("Excel.Application")
o.Visible = 1
o.Workbooks.Add() # for office 97 ? 95 a bit different!
o.Cells(1,1).Value = "Hello"

Because I have installed Excel95 I have given it a try and that's the
result:

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import win32com.client
>>> o= win32com.client.Dispatch("Excel.Application")
>>> o.Workbooks.Add()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "E:\Programme\Python\win32com\client\dynamic.py", line 400, in
__getattr_
_
    raise AttributeError, "%s.%s" % (self._username_, attr)
AttributeError: Excel.Application.Workbooks

But Workbooks.Add is a available method as I can test with VBScript.
Is my installation corrupt or do I do some silly mistake or perhaps some
more serious?

Thanks for any answer.
Regards,
Axel




From aahz at netcom.com  Wed Jun  2 10:55:12 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Wed, 2 Jun 1999 14:55:12 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7iv86a$iud@news.dns.microsoft.com> <87vhd7nsra.fsf@ev.netlab.co.jp> <3754F5D7.2B6E1954@pop.vet.uu.nl>
Message-ID: <aahzFCpG40.EFs@netcom.com>

In article <3754F5D7.2B6E1954 at pop.vet.uu.nl>,
Martijn Faassen  <M.Faassen at vet.uu.nl> wrote:
>
>* The predictability of reference counting is nice in the simple case.
>In the complex case, you get into a mess of circular references too
>easily, though.
>
>* In the complex case, garbage collecting is nice as it cleans up
>circular references. There are unwanted side-effects of its
>unpredictability, though; a possible leaking of resources (files which
>aren't closed), and some difficulties interfacing with C.

It might make more sense to simply add tools to locate and stomp
circular references, perhaps augmented by explicit calls to delete
specific instances of circular references.

I'm curious: does GC provide any advantages other than its ability to
deal with circular references?
-- 
                      --- Aahz (@netcom.com)

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

"NRE is a lot like cocaine. Feels great while you're on it, makes you
feel very special and talented, and you become highly annoying to those
who aren't on it."  -- Stef



From gherman at my-deja.com  Wed Jun 23 13:39:14 1999
From: gherman at my-deja.com (Dinu C. Gherman)
Date: Wed, 23 Jun 1999 17:39:14 GMT
Subject: How to obtain an instance's name at runtime?
References: <7kqpu8$7u1$1@nnrp1.deja.com> <7kqrc5$9fs@cs.vu.nl>
Message-ID: <7kr640$d8j$1@nnrp1.deja.com>

In article <7kqrc5$9fs at cs.vu.nl>,
  Frank Niessink <frankn=news at cs.vu.nl> wrote:
> Dinu C. Gherman <gherman at my-deja.com> wrote:
>
> > Question: How to obtain an instance's name at runtime?
>
> That instance doesn't have a name. Take for example:
>
> >>> a = b = c = C()
>
> That creates one instance of C, with three references to it.
> Now what is the `real' name of the instance?

Very true, very true - I new there was something good about
Python, here: object references, making something else rather
impossible! Still, I wish it was possible to access the "name
administrator" somewhere deep inside the Python interpreter to
have that. I'll give more reasons for that only in a couple
of days, I hope... Sorry for not being more precize, now!

Dinu



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From philc at unifix-online.com  Mon Jun 28 19:31:20 1999
From: philc at unifix-online.com (Philippe Chassany)
Date: Tue, 29 Jun 1999 01:31:20 +0200
Subject: Python GUIs: Summary and Conclusion
References: <99FA4A1A1280BEBD.0A1ABA62CC989479.4C2C6022F695BEAC@lp.airnews.net>
Message-ID: <377805C8.C6BBD4C1@unifix-online.com>

> ....
> More widgets should be added to Tk(Inter).
> ....

Our company has made a 100% Tcl/Tk Toolkit named "BWidget".
BWidgets are "well known" in the Tcl community because
they are free, flexible, powerful and completly interpreted (no C).
More than 25 widgets are available.

We have tested BWidget with Tkinter the result is quite fine :
Quick and good looking. It's a good alternative to GTk or Qt
compiled programming to have professional application GUIs.

Try it to make your own idea (free download, screenshots,
full HTML manuals)

http://www.unifix-online.com/BWidget

--
Philippe Chassany
UNIFIX - http://www.unifix-online.com
--





From Klaus.Schilling at home.ivm.de  Sat Jun 12 05:25:18 1999
From: Klaus.Schilling at home.ivm.de (Klaus Schilling)
Date: 12 Jun 1999 11:25:18 +0200
Subject: Newbie question: DOS console commands?
References: <7jsb5s$k7q$1@newssvr01-int.news.prodigy.com>
Message-ID: <87vhctbwjl.fsf@home.ivm.de>

"Phaedrus42" <tcropley at nospam.prodigy.net> writes:

> I'm new to Python this week and was wondering if there are any console
> commands to use on a DOS screen (e.g. ClearScreen, GotoXY, Foreground and
> Background colors, etc.)
> 
> I've got PythonWin running on Windows 95 and am having fun learning the
> language and Tkinter, but I have a need to write a simple DOS interface
> program.  I could pull out my copy of TurboPascal, but I'd like to write it
> in Python to further my education.

Maybe try to use Andrich's Curses Module with M. Hessling's PDcurses?
One prolly would neet to cut out some stuff that isn't supported
by PDcurses.

Klaus Schilling



From doughellmann at mindspring.com  Mon Jun 14 10:00:59 1999
From: doughellmann at mindspring.com (Doug Hellmann)
Date: Mon, 14 Jun 1999 14:00:59 GMT
Subject: Problem with Lists & Index command
References: <rNHvjEdhm5Pp-pn2-zOTjsVjHebb4@d185d18f8.rochester.rr.com> <37618424.8B8F3116@mindspring.com> <rNHvjEdhm5Pp-pn2-XcIGYOCMcDxI@d185d18f8.rochester.rr.com>
Message-ID: <37650BEB.564B2B20@mindspring.com>

Benjamin Schollnick wrote:
> 
> But, the most obvious question, is.....
> 
> Can dictionaries be made & manipulated from
> variables?

Of course!

Think of dictionaries as arrays with indexes that can be values other
than numbers.  Indeed, dictionaries are sometimes called "associative
arrays" because thinking about them this way makes them easier to
understand.

So, if you have a variable 'entry' with the value "EFGHI" and you want
to use a dictionary to associate the value of another variable entryval
(which is 1) with that value, you could do something like:

entry = 'EFGHI'
entryval = 1
tcmvalues = {}   # create an empty dictionary (similar to creating an
emtpy list with [])
tcmvalues[ entry ] = entryval

This creates an association between the "value" in entryval with the
"key" in entry.

Now, suppose you are writing a program to count the number of times a
word occurs in a document.  In this case, you would want to increment a
counter stored at the appropriate location.  A first try might look
something like this:

wordcounts = {}
for word in document.words():
  wordcounts[word] = wordcounts[word] + 1

But what happens when you see a word you've never seen before?  On the
right side of the = sign we are retrieving the current value of
wordcounts[word] to add one to it.  But if we've never seen the value of
word before, we'll have a problem.  Some languages solve this by
returning a null value of some type.  Python solves the problem by
raising an exception to let you know you have a situation you need to
deal with.  

> Please keep in mind, that I don't know anything about
> dictionaries... The book I'm using hasn't hit 'em yet....
> (I'm not done reading yet)

If you haven't gotten far enough in the book to find a discussion of
dictionaries, you probably haven't seen anything about exception
handling either.  Throwing and catching exceptions is (usually) a way to
simplify program logic.  It is sort of like saying, "try doing X except
when you encounter a problem.  Handle problem type Y like this, handle
problem type Z like this..."  The standard library functions and built
in types usually raise exceptions to indicate problems or special
cases.  The exception raised by a dictionary when you ask for a value
associated with a key that is not present will be an instance of the
class "KeyError."  

The first occurance of a word is a special case, but is not an error. 
That means we can just handle the exception and keep going through the
list of words.

wordcounts = {}
for word in document.words():
  try:
    wordcounts[word] = wordcounts[word] + 1
  except KeyError:
    wordcounts[word] = 1


Good luck,
Doug



From srn at fn.com.au  Fri Jun 25 03:12:49 1999
From: srn at fn.com.au (Stephen Norris)
Date: 25 Jun 1999 07:12:49 GMT
Subject: "Pre-forking" SocketServer...
Message-ID: <7kva5h$9t1$1@m2.c2.telstra-mm.net.au>

I thought someone might be interested - I've written a
pre-thread-generating socket server, which maintains a set of threads to
handle incoming requests. Considerably faster than the forking server,
but a drop-in replacement for a TCPServer.

It's at: http://flibble.fn.com.au/~srn/Python/

The trace module is needed for debugging output - it's easy to remove.

Comments encouraged.

	Stephen
-- 
Stephen Norris  srn at fn.com.au
PGP key available via finger srn at flibble.fn.com.au.
Farrow Norris Pty. Ltd.     http://www.fn.com.au/



From dkuhlman at netcom.com  Sun Jun 13 20:59:12 1999
From: dkuhlman at netcom.com (G. David Kuhlman)
Date: Mon, 14 Jun 1999 00:59:12 GMT
Subject: Python as script language?
References: <37644DC7.DE422A3B@hotmail.com> <19990613201428.K723@toast.internal>
Message-ID: <dkuhlmanFDALEo.B6@netcom.com>

jam (jam at newimage.com) wrote:
> On Sun, Jun 13, 1999 at 09:38:30PM +0000, Denis Lamarche wrote:
> > 
> > How can I set up Python so that I can call the scripts from within an
> > application (that I wrote) so that it will run and interface with the
> > application (that I wrote)?  Is this possible?
> > 
> > Denis Lamarche
> > 
> 
[ good suggestions snipped ]

Also note that the Python source code distribution comes with a demo
called "embed".  A good first step is to modify and extend this
demo.

And you should consider going to www.python.org and searching the
newsgroup for "embed".

Embedding Python in an application is not only possible, there is
support in Python intended for this purpose.

  - Dave



From tbryan at arlut.utexas.edu  Fri Jun  4 09:03:51 1999
From: tbryan at arlut.utexas.edu (Thomas Bryan)
Date: Fri, 04 Jun 1999 08:03:51 -0500
Subject: build of python 1.5.2 with threads on SunOs 5.6 fails
References: <3756FACF.D62ABE2A@rubic.com> <Pine.LNX.4.05.9906040828480.12456-100000@ns.ac-copy.net>
Message-ID: <3757CEB7.916962BA@arlut.utexas.edu>

Joachim Schmitz wrote:
> 
> > A 'make clean' might be the easy way to fix your problem.
> >
> 
> That obviously didn?t remove libpython1.5.a. After removing it and doing a
> make again, it works now.

I blew a couple of hours on this problem once.  
I should have looked at the Makefile a lot sooner.

"make clean" simply removes object files and such.

"make clobber" does a "make clean" and removes a previous run's 
libpython, python executable, and configuration information.  If 
you want to build Python a second time with a new configuration, 
run "make clobber" before reconfiguring and running make.

-------------------------------------------
Tom Bryan
Applied Research Laboratories
University of Texas at Austin



From tim_one at email.msn.com  Sun Jun 27 22:27:04 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 27 Jun 1999 22:27:04 -0400
Subject: RuntimeError?
In-Reply-To: <19990627112432.A29097@optiplex.palga.uucp>
Message-ID: <000401bec10d$b5bc3a60$e19e2299@tim>

[Gerrit Holl]
> can someone please give me an example when a RuntimeError is raised?
>
> I don't understand: if none of the others are raised.

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import sys
>>> del sys.stdout
>>> print "hi"
Traceback (innermost last):
  File "<stdin>", line 1, in ?
RuntimeError: lost sys.stdout
>>>

Honestly, I had never seen a RuntimeError before!  I found this one by
digging thru the source code to see when it gets raised.  There's really no
pattern to it, and the RuntimeError docstring in Lib/exceptions.py
("Unspecified run-time error.") is as accurate as anything that says nothing
can be <wink>.  It generally seems to mean something bad and unexpected
happened, but not bad enough to shut down the interpreter (things *that* bad
raise SystemError).

more-art-than-science-here-ly y'rs  - tim





From 4mschulz at informatik.uni-hamburg.de  Wed Jun  9 13:58:11 1999
From: 4mschulz at informatik.uni-hamburg.de (Marko Schulz)
Date: 9 Jun 1999 17:58:11 GMT
Subject: "Programming Python" still worthwhile?
References: <7jki66$qam$1@news.gstis.net> <00c301beb260$09e94680$f29b12c2@pythonware.com> <7jlqfq$jgs$1@nnrp1.deja.com>
Message-ID: <7jm9vj$5ft$1@rzsun02.rrz.uni-hamburg.de>

Curtis Yanko <cmyanko at bigfoot.com> wrote:
> Isn't PP due for 2nd edition soon?

On www.ora.com in the 'Frankly speaking'-column they say, they will
publish PP after Python 2.0 comes out.

I wouldn't call this 'soon'. 

-- 
marko schulz
              "Die Welt ist gut, die Welt ist schlecht.
               Ich seh' mehr als ich begreifen kann. Ich sehe in 3-D."
                                                         '3-D', Extrabreit



From tbryan at arlut.utexas.edu  Thu Jun 24 20:24:51 1999
From: tbryan at arlut.utexas.edu (Thomas Bryan)
Date: Thu, 24 Jun 1999 19:24:51 -0500
Subject: make clean bug in 1.5.2
References: <99Jun24.102155pdt."3693"@watson.parc.xerox.com>
Message-ID: <3772CC53.274BFBEB@arlut.utexas.edu>

Bill Janssen wrote:
> 
> Running "make clean" seems to leave a couple of problematic files,
> SRC/libpython1.5.a, and getbuildinfo.o, which then cause problems
> when I try to "make" again.

After over an hour of utter confusion, I noticed the same thing
when I built Python 1.5.2.  Then I checked the Makefile.  

Solution: 
make clobber

My guess:
Since Python needs the build directory for tools like Freeze, you
don't want to take up space on your hard drive with all of the
object files in the build directory.  Thus, you can run make clean
to recover some disk space after a successful Python install.  
Since the clean target was taken, the clobber target was given 
the task of cleaning up between builds.

Maybe I'll add this to the FAQ since it's the third time I've seen it...

-------------------------------------------
Tom Bryan
Applied Research Laboratories
University of Texas at Austin



From behrends at cse.msu.edu  Wed Jun 30 09:14:09 1999
From: behrends at cse.msu.edu (Reimer Behrends)
Date: 30 Jun 1999 13:14:09 GMT
Subject: total idiot question: +=, .=, etc...
References: <000001bebffc$12060200$da9e2299@tim> <slrn7ngj6b.g9.behrends@allegro.cse.msu.edu>  <m3so7buvr6.fsf@greebo.nodomain.de>
Message-ID: <slrn7nk628.18i.behrends@allegro.cse.msu.edu>

Bernhard Herzog (herzog at online.de) wrote:
[...]
> I think a better solution that better fits the python syntax is to
> simply omit the self but not the dot:
> 
> 	.last = .lines[-1]
>  	.lines = .lines[:-1]

Such a syntax was proposed a few years ago for private variables,
IIRC, but was discarded because it had various problems.

And personally, I'd rather live with the inconvenience of the explicit
self rather than having Python's design obfuscated by some strange
syntax; syntax highlighting can go a long way towards making the current
state of affairs more manageable.

			Reimer Behrends



From Sim.&.Golda.Zacks at p98.f112.n480.z2.fidonet.org  Wed Jun 30 12:14:15 1999
From: Sim.&.Golda.Zacks at p98.f112.n480.z2.fidonet.org (Sim & Golda Zacks)
Date: Wed, 30 Jun 1999 17:14:15 +0100
Subject: MySQL interface
Message-ID: <000015e9@bossar.com.pl>

From: Sim & Golda Zacks <simngolda at voyager.net>

My ISP has a MySQL server and I want to use it through Python 1.4. The
version of MySQL they have is:  mysql  Ver 9.16 Distrib 3.21.33, for
pc-linux-gnu (i586).
I looked at the Python database site and they have a module for MySQL,
but it is a C file and the instructions are to build it with the
distribution. I am not building anything, all I want to do is use the
database server. Can someone help me figure out what I need to use, and
how to get it. I was going to use Gadfly, but I'm also going to need to
use a database with JAVA JDBC so I want to use the same product for all
my stuff.
The ISP only has Python 1.4.

Thanks

Sim



From cgw at fnal.gov  Thu Jun 24 14:03:26 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Thu, 24 Jun 1999 18:03:26 GMT
Subject: Default printing behavior for classes
In-Reply-To: <7ktoe5$bbl$1@nnrp1.deja.com>
References: <7ktoe5$bbl$1@nnrp1.deja.com>
Message-ID: <14194.29422.173193.823343@buffalo.fnal.gov>

You're looking for the __repr__ and/or __str__ methods:

>From the docs:

 __repr__ (self) 
     Called by the repr() built-in function and by string conversions
 (reverse quotes) to compute the ``official'' string representation of
 an object. This should normally look like a valid Python expression
 that can be used to recreate an object with the same value. By
 convention, objects which cannot be trivially converted to strings
 which can be used to create a similar object produce a string of the
 form "<...some useful description...>".

 __str__ (self) 
      Called by the str() built-in function and by the print statement
 to compute the ``informal'' string representation of an object. This
 differs from __repr__() in that it does not have to be a valid Python
 expression: a more convenient or concise representation may be used
 instead.

Note that these methods should *return* the string value, not print
it.

If c is a class instance, and the class has __repr__ or __str__
methods,  these will be invoked when you say "print c".

If you have both a __str__ method and a __repr__ method, the __str__
will be used when you say "print c".  But if you only have a 
__repr__ that will be used.






From graham at sloth.math.uga.edu  Tue Jun  8 12:31:41 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 8 Jun 1999 16:31:41 GMT
Subject: fork()
References: <000601beb1c7$682c6220$559e2299@tim>
Message-ID: <7jjghd$9il$1@cronkite.cc.uga.edu>

Tim Peters (tim_one at email.msn.com) wrote:
: OTOH, in a straightfoward mark-and-sweep GC scheme there will at least be a
: "mark bit" in each object header.  So under that, or under any form of
: compacting GC, the *entire live object space* will get copied, regardless of
: whether *you* reference anything in it or not.  So if RC truly sucks, what
: does that make GC <wink>?

But the reality is that copying memory can in fact speed applications
up! Think about locality of reference after you have copied all live
blocks to the start of memory.

graham

-- 
                      Je suis pour le communisme
                      Je suis pour le socialisme
                      Je suis pour le capitalisme
                     Parce que je suis opportuniste



From jonathan at onegoodidea.com  Fri Jun  4 18:18:34 1999
From: jonathan at onegoodidea.com (Jonathan Hogg)
Date: Fri, 04 Jun 1999 23:18:34 +0100
Subject: python active web pages
References: <jonathan-0406991640390001@borg.onegoodidea.net> <37582D9D.3D7A4B1D@hp.com.NOSPAM>
Message-ID: <jonathan-0406992318350001@borg.onegoodidea.net>

In article <37582D9D.3D7A4B1D at hp.com.NOSPAM>, Bruce Wheeler
<NOSPAMbruce_wheeler at hp.com.NOSPAM> wrote:

>Jonathan Hogg wrote:
>> 
>> hi there,
>> 
>> i'm looking to embed python into web pages in a similar way to ASP and
>> PHP3 and so forth.
>> 
>take a look at <http://starship.python.net/crew/mhammond/>

sorry, i should perhaps have made myself more clear. i was wondering as i
posted whether mention of ASP would confuse the matter.

i'm not implementing on Windows or using "Active Server Pages" as defined
by Microsoft. i'm interested in dynamic web pages on unix/apache.

thanks anyway,

:-j


(and yes, i know about PyApache - but that's just a
convenience/optimisation for CGI, not an embedding of Python in web pages)

-- 
jonathan hogg, one good idea limited, 25 clouston street, glasgow g20 8qr, uk
jonathan at onegoodidea.com www.onegoodidea.com t:(0)976-614338 f:(0)7970-537451



From arcege at shore.net  Fri Jun 11 11:39:58 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Fri, 11 Jun 1999 15:39:58 GMT
Subject: Serial programming
References: <7jp1jt$8hg$1@imsp009a.netvigator.com>
Message-ID: <i5a83.2876$nn.847925@news.shore.net>

DORAEMON.HK <abc at def.ghi.jkl> wrote:
: I am looking for some example in python about send and receive data from
: serial port.  Like send "ATZ" and receive "OK".  Any help?? Thanks a lot!!

You can open the device file directly and handle the information
yourself.  UNIX devices are handle just like files.  You can change
device attributes (speed, parity, etc.) with the termios module or
fcntl.ioctl function.

  modem = open('/dev/cua0', 'r+')
  modem.write('AT\r')
  f = modem.read()
  if f[:2] != 'OK':
    raise ValueError, "modem not okay"
  ...

There are some Expect modules/extensions around that could help you as
well;  my ExpectPy extention is at
http://starship.python.net/~arcege/ExpectPy/.  You can find others
through the Python website.

  import ExpectPy
  modem = ExpectPy.spawnfile(open('/dev/cua0', 'r+'))
  modem.send("AT\r")
  try:
    f = modem.expect(
      (ExpectPy.EXACT, "OK\r\n", "OK"),
      (ExpectPy.GLOB, "*\r\n", "Not OK")
    )
    if f == "Not OK":
      raise ValueError, "modem not okay"
  except (ValueError, ExpectPy.TimeoutError), value:
    raise ValueError, value

I hope this helps.

  -Arcege




From avv at quasar.ipa.nw.ru  Wed Jun 30 18:17:58 1999
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Wed, 30 Jun 1999 15:17:58 -0700
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773c3a0.102825445@news.oh.verio.com> <xPPc3.535$or1.22866@typhoon-sf.snfc21.pbi.net> <xJRc3.292$Of3.67094@news.shore.net> <7laovo$c4s$1@Starbase.NeoSoft.COM> <377A24B7.AD6FAA1A@cfdrc.com>
Message-ID: <377A9796.176DA97E@quasar.ipa.nw.ru>

Lyle Johnson wrote:

> Just to add fuel to the fire, we now have a Python interface to FOX, a
> cross-platform GUI toolkit:
>
>     http://www.cfdrc.com/pub/FOX/fox.html
>
> The latest version of the Python wrappers can be picked up at:
>
>     ftp://ftp.cfdrc.com/pub/FOX/FXPy-0.99.42.tar.gz
>
> I don't know how the speed of FXPy would compare with some of the others
> mentioned (Tk, wxWindows, Qt, GTK, ...) but it would be worthwhile for
> some independent opinions.

I installed it on Red Hat 6.0. I failed to recompile it without Mesa, but
even with it, it complains as follows:

[avoinov at dhcp101 tests]$ python window.py
Traceback (innermost last):
  File "window.py", line 1, in ?
    from FXPy import *
  File "/usr/lib/python1.5/site-packages/FXPy/__init__.py", line 1, in ?
    from fox import *
  File "/usr/lib/python1.5/site-packages/FXPy/fox.py", line 2, in ?
    import foxc
ImportError: /usr/lib/python1.5/site-packages/FXPy/foxcmodule.so: undefined
symbol: glXQueryExtension

It happens both with Mesa-3.0 (Nov 98) and some Mesa-3.1beta, found in
RedHat contrib.

Alexander





From loredo at spacenet.tn.cornell.edu  Fri Jun 25 15:15:39 1999
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Fri, 25 Jun 1999 15:15:39 -0400
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773CA67.7057F797@quasar.ipa.nw.ru>
Message-ID: <3773D55B.8C966174@spacenet.tn.cornell.edu>

Well, one reason not to rush to wxWindows is that there is
no "real" Mac version.  I realize that a Mac version is "underway",
but the wxWindows site has said that for a *long* time now.
A beta version was posted about a month ago, but I've not
heard much about it, and most of the wx web site doesn't
seem to know it exists (e.g., the FAQ still says the beta
is forthcoming, and the dowload section appears to still
have the old preview alpha release).  I develop on both Unix
and Mac, and although the Unix part of me finds wx appealing,
the Mac side of me feels like a 2nd class citizen in the wx world.

-Tom Loredo



From gmcm at hypernet.com  Wed Jun 16 12:25:58 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 16 Jun 1999 11:25:58 -0500
Subject: Newbie again: computing attributes on the fly
In-Reply-To: <Pine.OSF.4.05.9906161547450.4875-100000@pinax.Mathematik.Uni-Bielefeld.DE>
Message-ID: <1282571882-108998091@hypernet.com>

Olaf Delgado writes:
> 
> here's another couple of questions to the enlightened ones. I'd like
> to compute certain properties of a class instance only once and
> store it for future use. I'd like to do those computations on demand
> only, because some might be really, really expensive. My first try
> looks like this:
> 
> class x:
>     def __getattr__(self, attr):
>         import string
>         if string.find(attr, "get_") != 0:
>             val = ( lambda x = getattr(self, "get_"+attr)(): x )
>             self.__dict__[attr] = val return val
>         else:
>             raise AttributeError("no such attribute")
> 
>     def get_lost(self):
>  return 0 # really, really expensive :)
> 
>     def get_a_life(self):
>  return 1
> 
> >>> y = x()
> >>> y.lost()
> 0
> >>> y.a_life()
> 1
> 
> Here, the lambda is pointless somehow, but hints to a potential
> desire to add parameter support in the future (which I think I don't
> need, but who knows?).
> 
> Question(s):
> 1) Is this 'good', or at least 'okay' style?
> 2) Any immediate suggestions for improvement?
> 3) Has anyone ever bothered to implement a more elegant or more
> complete
>    approach, which might, for example, include such things as
>    parameters, adding memorization to existing classes, you name it?

You're making things too complicated. __getattr__ is only called if 
the attribute is not found. So something like this will suffice:

>>> class X:
...   def __getattr__(self, nm):
...     if nm == 'foo':
...       self.foo = 17.0 * 33.1 + .01
...       return self.foo
...     raise AttributeError("no attribute %s" % nm)
...
>>> x = X()
>>> dir(x)
[]			#no instance attributes
>>> x.foo
562.71
>>> dir(x)
['foo']		#now foo is an instance attribute

If you want to generalize that, I'd see if we have an attribute 
"calc_%s" % nm and use that method (vs the inline computation).

- Gordon



From fatjim at home.com  Tue Jun  8 02:56:15 1999
From: fatjim at home.com (Jim Meier)
Date: Tue, 08 Jun 1999 06:56:15 GMT
Subject: where does 'del' come from?
References: <7jhuu5$8d4$1@nnrp1.deja.com>
Message-ID: <375CBEF3.B1D2D288@home.com>


sragsdale at my-deja.com wrote:

> A fairly academic question, but I'm curious nonetheless.  Where is 'del'
> defined?  Functions like 'join' are obviously in the string package,
> 'range' is a builtin, but I can't seem to find 'del' anywhere.
>
> Python 1.5.2 (#12, Jun  2 1999, 19:23:30) [C] on irix646
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> dir(__builtins__)
>

[snipped list of python builtins, which include the range function and don't
include "del"]

> >>> range
> <built-in function range>
> >>> del
>   File "<stdin>", line 1
>     del
>        ^
> SyntaxError: invalid syntax

The SyntaxError is the clue, as should del's invocation (parentheses-less).
del is not a function, but a part of the Python grammar, the same way if,
while, and, is, etc.. are. If you'd like to see where it comes from, have a
look in your Python source tree:

(python source home)/Grammar/Grammar

I think that should answer the question. Until you start wanting to know how
Python turns the grammar file into an working Python lexer/parser.. that
one's for the CS guys. And I'm not yet a CS guy!

-Jim





From Klaus.Schilling at home.ivm.de  Mon Jun 14 05:46:14 1999
From: Klaus.Schilling at home.ivm.de (Klaus Schilling)
Date: 14 Jun 1999 11:46:14 +0200
Subject: OO (was: Why is tcl broken?)
References: <000a01beb5dc$5e88aa80$ae9e2299@tim>
Message-ID: <87btejjes8.fsf@home.ivm.de>

"Tim Peters" <tim_one at email.msn.com> writes:

> [Fernando Mato Mira]
> > Although Dylan never really made it into "Fernando's elegibility list"
> > because it was not a serious contender against Scheme and CL yet, it
> > definitely got banished the day the Lisp syntax was dropped. I'm still
> > waiting for the day when the Dylan fans will reinstate the alternative.
> 
> OK!  Now we know the answer Fernando is looking for:  Tcl is fatally broken
> because its syntax differs from Lisp's <0.9 wink>.

Infix syntax is junk.

Klaus Schilling



From Reimer.Behrends at p98.f112.n480.z2.fidonet.org  Wed Jun 30 08:30:49 1999
From: Reimer.Behrends at p98.f112.n480.z2.fidonet.org (Reimer Behrends)
Date: Wed, 30 Jun 1999 13:30:49 +0100
Subject: Typing [Re: total idiot question: +=, .=, etc...]
Message-ID: <000015b3@bossar.com.pl>

From: behrends at cse.msu.edu (Reimer Behrends)

Joshua Rosen (rozzin at geekspace.com) wrote:
> Reimer Behrends wrote:
> >         Require variables to be prefixed with "$" to denote scalar
> >         values--strings and integers, "@" for lists, and "%" for
> >         dictionaries.
> > 
> > Given the precedent of having to use self to access instance variables,
> > this is entirely consistent with the current language, and is going to
> > eliminate all those nasty problems where you can't tell from the context
> > what the type of a variable or function is. :)

> 	+There is no difficulty in getting the program to know what type
> 	some piece of data is, which lessens the need for the
> 	type-declaring prefixes--use type(), and then check
> 	obj.__class__ if obj is a class-instance, and you need to -know-
> 	what class it is (if you want to -force- it to be a certain
> 	type, keep reading).

Ahem. I happen to know Python fairly intimately after I don't know how
many years of constant use. What I wrote above was mostly a tongue in
cheek comment (heck, I even said so). The point I was making was that
the very same argument that justifies a self prefix can also be used to
justify type prefixes--in fact, I pretty much paraphrased it from an old
posting by Tom Christiansen on 10/26/98 (the message id is
<710kg1$mr0$1 at csnews.cs.colorado.edu>)

> 	+Requiring variables to bear a prefix denoting type sortof makes
> 	Python into a statically-typed language, where one of the big
> 	reasons for using it is that it's a dynamically-typed language.

I never said otherwise. But the same argument can be used mutatis
mutandum to support a "self"-less notation. :)

> 	+There is no such thing as a `scalare', in Python--strings and
> 	numbers are different data-types, numeric data can actually be
> 	one of many different classes.

I know. It was a facetious argument.

> 	+Forcing data to be of a specific type is already possible--see
> 	documentation on int(), str(), list(), tuple(), float(), long(),
> 	complex(), string.atoi(), et al. This functionality also
> 	lessens, if not completely eliminates the need for typing
> 	prefixes; the program knows what type the data is being coered
> 	into, and so does the programmer. In fact, all of the
> 	functionality that you're proposing (and then some) is already
> 	there--it just looks different.

And I just happen to be aware of that, too.

> 	+If you really want a feature that -really- isn't offered by a
> 	language, the best decision may be to just use another language
> 	that -does- offer it. Try Pike, for one (I suggest it because it
> 	has nifty typing which you may like: you can declare a variable
> 	to be one or -more- types, or use a mixed type). Choose your
> 	languages based upon what you need them to do.

And I know Pike, too. In fact, I'm pretty familiar with it.

By the way, would you mind using shorter lines next time?

				Reimer Behrends



From mikael at pobox.com  Wed Jun 30 19:51:17 1999
From: mikael at pobox.com (Mikael Lyngvig)
Date: Wed, 30 Jun 1999 23:51:17 GMT
Subject: Python GUIs: Summary and Conclusion
References: <99FA4A1A1280BEBD.0A1ABA62CC989479.4C2C6022F695BEAC@lp.airnews.net> <19990629092757.A4998@optiplex.palga.uucp>
Message-ID: <46F9FB3A1847A014.2CBD270176138613.20F84A39E6ED95DA@lp.airnews.net>

On Tue, 29 Jun 1999 09:27:57 +0200, Gerrit Holl
<gerrit.holl at pobox.com> wrote:

>I think one could also make an application "multiple-interface". The user can
>choose: svgalib (any support in python, by the way?), curses, commandline,
>tk, qt/kde or some of the other stuff. The developer can put the interface
>in a seperate class or even in a seperate module (I'll do with my discoverb
>project (see signature)), so the user can exactly download what he want. I
>believe KDE users would like KDE applications...

I think that would be a good idea, but I don't think it is directly
related to the issue at hand.  Such an API would certainly be slower
than the actual implentation and the problem is, IMHO, that the actual
implementation (of Tk) is too slow to be really interesting.

As a matter of fact, I've always wished that I could make my apps so
they took an option, say -ui:name, what would allow the user to
specify which kind of user interface it had:

   bui - batch user interface; Unix style command-line interface.
   cui - console user interface; Unix curses interface.
   gui - graphical user interface; X/Windows style interface.

That way the user could use the application in any way he/she like,
without having to resort to Win32 COM (automation server) APIs etc.


-- Mikael




From edlsoft at laker.net  Thu Jun 17 18:05:25 1999
From: edlsoft at laker.net (Burt Leavenworth)
Date: Thu, 17 Jun 1999 22:05:25 GMT
Subject: Tk8.0.5 for Linux not compiling!
Message-ID: <37697070.381356@news.laker.net>

I compiled Python 1.5.2 under Caldera OpenLinux 2.2 on an IBM Aptiva
and everything works fine. Also tcl8.0.5/unix compiled successfully.

But ./configure under tk8.0.5/unix produced:
...
checking for X11 header files...couldn't find any!
checking for X11 libraries...checking for -1Xwindow...no
couldn't find any! Using -1X11

and make produced a whole bunch of parse errors in ... /generic/tk.h

Any help appreciated.

Burt Leavenworth




From zhaoway at mud.jlonline.com  Fri Jun  4 09:22:26 1999
From: zhaoway at mud.jlonline.com (zhaoway)
Date: Fri, 4 Jun 1999 21:22:26 +0800
Subject: Help for beginner
Message-ID: <003b01beae8e$2b1d92a0$714166ca@zhaoway>

Spam detection software, running on the system "albatross.python.org", has
identified this incoming email as possible spam.  The original message
has been attached to this so you can view it (if it isn't spam) or label
similar future email.  If you have any questions, see
the administrator of that system for details.

Content preview:  Max M. Stalnaker wrote in message <3751885A.DCF055D6 at acm.org>...
   >You have to buy all the python books and once you work through those, >(being
   an old-timer) I suggest you work through volume I of Knuth's opus >on programming.
   Other people will have more current suggestions, but no >one will complain
   about a Knuth recommendation. Embrace the object >orientation of python but
   stay away from GUI implementations for awhile. [...] 

Content analysis details:   (5.1 points, 5.0 required)

 pts rule name              description
---- ---------------------- --------------------------------------------------
 3.0 FH_HELO_ENDS_DOT       Helo ends with a dot.
 2.0 FH_DATE_IS_19XX        The date is not 19xx.
 0.1 RDNS_NONE              Delivered to trusted network by a host with no rDNS


-------------- next part --------------
An embedded message was scrubbed...
From: "zhaoway" <zhaoway at mud.jlonline.com>
Subject: Re: Help for beginner
Date: Fri, 4 Jun 1999 21:22:26 +0800
Size: 2268
URL: <http://mail.python.org/pipermail/python-list/attachments/19990604/8fe7c017/attachment.mht>

From arcege at shore.net  Thu Jun 10 14:45:37 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Thu, 10 Jun 1999 18:45:37 GMT
Subject: Embedded Tcl for Python
Message-ID: <lJT73.2677$nn.808830@news.shore.net>

A few weeks ago, someone asked if there was a Tcl module for Python.
People responded (including myself) that Tkinter uses a Tcl/Tk
interpreter, but you couldn't get rid of the annoying window.

I've since written an extension ("pytcl") that creates Tcl interpreter
objects.  As part of that, I have also rewritten _tkinter to use these
new objects.

I've had a short discussion with Guido and he's interested in the
idea.  Now I need some alpha testers; I only have access to UNIX boxes,
and very few types at that.

The _tkinter module has the same functionality as before, including the
same threading issues.  I've include Deiter's Tcl/Tk8.1 patch but have
not had the opportunity to test that yet (right not it is just do-no-
damage).

Contact me at arcege at shore.net if you are interested in helping me test
these modules.

  -Arcege




From zorro at zipzap.ch  Tue Jun  1 11:36:20 1999
From: zorro at zipzap.ch (Boris Borcic)
Date: Tue, 01 Jun 1999 17:36:20 +0200
Subject: Looking for a lib fun
Message-ID: <3753FDF4.5A13F8B8@zipzap.ch>

There is a function I can't seem to find in the
standard library. One that will turn XML/HTML
offending characters to corresponding entity- or
character references.

Is it indeed in the lib ? Where ? If not, could a kind
soul hand me one ? Writing it really feels like
reinventing the wheel for the 10^nth time.

TIA



From da at ski.org  Wed Jun 16 14:10:08 1999
From: da at ski.org (David Ascher)
Date: Wed, 16 Jun 1999 11:10:08 -0700 (Pacific Daylight Time)
Subject: Newbie again: computing attributes on the fly
In-Reply-To: <Pine.OSF.4.05.9906161916040.10750-100000@pinax.Mathematik.Uni-Bielefeld.DE>
Message-ID: <Pine.WNT.4.04.9906161108430.310-100000@rigoletto.ski.org>

On Wed, 16 Jun 1999, Olaf Delgado wrote:

> Yes, I definitely want to generalize that. The problem is that that method
> may be defined in a superclass, so "self.__dict__" won't find it. AFAIK
> (which may be not be very far), I have to use "hasattr" and/or "getattr",
> which in turn will call "__getattr__".

You may find that using ExtensionClass could help.  See:

    http://www.digicool.com/releases/ExtensionClass/

(note especially the section on computed attributes in the documentation).

--david





From tim_one at email.msn.com  Tue Jun 29 11:56:56 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 29 Jun 1999 11:56:56 -0400
Subject: total idiot question: +=, .=, etc...
In-Reply-To: <5l67476yd2.fsf@eric.cnri.reston.va.us>
Message-ID: <000001bec248$040ac0c0$6d9e2299@tim>

>>>     class A(B):
>>>         def __init__(self, x, y, z):
>>>             B.__init__(self, x, y, z)

[Greg Ewing]
> I could live with having to explicitly name the
> superclass, provided something checked that the
> class I named was actually a direct base class
> of the one where the method is defined. That
> way things would be less likely to break
> mysteriously when I rearrange the class
> hierarchy.

[Guido]
> Oh, but that check *is* being made!  (In fact this is the same check
> that prevents "class methods" from working ;-)

Nope, that's a different check.  If C derives from B derives from A, this
check allows either C.__init__ or B.__init__ to be called from A.__init__
with an instance of A.  Greg wants  B.__init__ to be OK here, but not
C.__init__ (C is not a direct base class of the one (A) where the method
(A.__init__) is defined).  IOW, he wants a way to restrict __init__ calls to
immediate bases of the class from which the __init__ is invoked, to e.g.
prevent level-skipping under class rearrangement.

I posted some ugly code to address that, but think it's really a job for a
text editor and a pair of eyeballs <wink>.

although-something-to-be-said-for-automating-the-eyeballs-ly y'rs  - tim





From behrends at cse.msu.edu  Fri Jun 25 02:02:29 1999
From: behrends at cse.msu.edu (Reimer Behrends)
Date: 25 Jun 1999 06:02:29 GMT
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0rr2.2fi.behrends@allegro.cse.msu.edu>  <37722979.2FCD9A5F@gssec.bt.co.uk>
Message-ID: <slrn7n66t8.rh7.behrends@allegro.cse.msu.edu>

Alan Gauld (alan.gauld at gssec.bt.co.uk) wrote:
["self" prefix vs. readability.]
> I was amazed to discover how difficult it was to understand 
> the code because I couldn't identify which variables were 
> local to the function, which were instance variables and 
> which were global.

While I'd be the first to agree that C++ is not exactly the epitome of
language design, I would think that at least the local variables should
be easy to identify. The only real problem is to distinguish global
variables from instance variables. And this is made worse in C++ by
spreading all the information related to a class over more than one
file. It should be noted that other languages do not necessarily have
the same problem and that careful design can easily avoid it.

But I do not think that the most basic operations of an OO language
should be complicated by cumbersome syntax (this includes instance
variable access and local method calls). The language should encourage
use of these elements, not get in your way.

[...]

				Reimer Behrends



From fandino at usa.net  Thu Jun  3 14:16:51 1999
From: fandino at usa.net (Salvador =?iso-8859-1?Q?Fandi=F1o?=)
Date: Thu, 03 Jun 1999 20:16:51 +0200
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net> <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net> <7j34l3$hbj$1@nnrp1.deja.com> <7j66b7$il5$2@cronkite.cc.uga.edu>
Message-ID: <3756C693.2C910B29@usa.net>

Graham Matthews wrote:

> You don't need a new method call -- all you need is a mark and sweep
> collector with refcounts. I really am amazed that this has generated so
> much traffic. It's very simple. You have ref counts just like you do
> now in Python. When an objects refcount goes to 0 its __del__ method
> is called. On top of this you add a mark and sweep collector to handle
> circular references. During a sweep if it finds dead objects it calls
> the __del__ method for that object.

GC usually deletes objects in random order which is not a very good
behavior. With a new magic method like __gc__ you could determine
adequate entry points to ciclid data structures deletion.

Suppose you have an object DB that encapsulates a database and a
persistent object O that automatically stores itself in DB when changed:

class DB_class:
	def __del__(s): s.db_close()
	def store(s,o):
		...
	...

class O_class:
	def store(s,DB): DB.store(s)
	def __del__(s):
		if s.changed_p:
			self.DB.store(s)
	
So if you delete DB before O (that could happen if the GC behavior was
to call __del__), O will never become stored.

Adding the new method __gc__ to O_class and changing the GC behavior to
just call __gc__ solves the problem:

	def __gc__(s):
		s.__del__()
		s.DB=None


- Salva.



From rdev at my-deja.com  Wed Jun 30 13:59:44 1999
From: rdev at my-deja.com (rdev at my-deja.com)
Date: Wed, 30 Jun 1999 17:59:44 GMT
Subject: Reflection in Python?
References: <7ldglr$ita$1@nnrp1.deja.com>
Message-ID: <7ldlu6$lb5$1@nnrp1.deja.com>

In article <7ldglr$ita$1 at nnrp1.deja.com>,
  anders777 at my-deja.com wrote:
>  ...How do I do the
> equivalent of what Java calls reflection? I want
> to do the following:
>
> test = MathTest("testAdd")
>
> where test.run() will run the "testAdd" method
on
> the MathTest object?  I also want to be able to
> tell the user the name of the object (MathTest)
> and the method in case the method fails.  Any
> thoughts?
>
If I understand what you are trying to do, it's
easy in Python.  Assuming MathTest is an
object instance:
test = MathTest.testAdd
will return you a bound method object which you
can then call as:
test()
Getting the name of the object class and method
is also easy.  There are several ways to do this:
1) Define a doc string within the method by
   putting a quoted string as the first line
   under the def line.  You can then refer to it
   as "test.__doc__".  This could say, for
   example, "Method bar of class foo".
2) "test.__name__" will return the name of the
   method.  "test.im_class.__name__" will
   return the name of the class that the method
   is defined within.

Example:
class MathTest:
  def testAdd(self, parm1, parm2):
      "Method testAdd of Class MathTest"
      result = parm1 + parm2
      return result
<<< mt = MathTest()
<<< test = mt.testAdd
<<< test.run(2,2)
4
<<< print test.__doc__
Method testAdd of Class MathTest
<<< print test.__name__
testAdd
<<< print test.im_class.__name__
MathTest


I hope that helps.

- Roger




> Thanks,
> Anders Schneiderman
> National Journal Daily Briefings Group
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
>



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From clarence at silcom.com  Wed Jun 16 11:19:05 1999
From: clarence at silcom.com (Clarence Gardner)
Date: Wed, 16 Jun 1999 15:19:05 GMT
Subject: Newbie again: computing attributes on the fly
References: <Pine.OSF.4.05.9906161547450.4875-100000@pinax.Mathematik.Uni-Bielefeld.DE>
Message-ID: <JfP93.395$dR.23133@newsfeed.avtel.net>

Olaf Delgado (delgado at Mathematik.Uni-Bielefeld.DE) wrote:
: Hi everybody,
: 
: here's another couple of questions to the enlightened ones. I'd like to
: compute certain properties of a class instance only once and store it for
: future use. I'd like to do those computations on demand only, because some
: might be really, really expensive. My first try looks like this:
: 
: class x:
:     def __getattr__(self, attr):
:         import string
:         if string.find(attr, "get_") != 0:
:             val = ( lambda x = getattr(self, "get_"+attr)(): x )
:             self.__dict__[attr] = val
:             return val
:         else:
:             raise AttributeError("no such attribute")
: 
:     def get_lost(self):
: 	return 0 # really, really expensive :)
: 
:     def get_a_life(self):
: 	return 1

On occasion I've done things like this (not necessarily for cost reasons,
but I just might want to do something different the first time).  What
I've done is:

class x:
	def dosomething(self, arg):
		initial blah, blah, blah
		self.dosomething = self._dosomething2
	def _dosomething2(self, arg):
		second_and_subsequent blah, blah, blah
		
-- 
-=-=-=-=-=-=-=-=
Clarence Gardner
AvTel Communications
Software Products and Services Division
clarence at avtel.com



From proteus at cloud9.net  Thu Jun 24 10:46:20 1999
From: proteus at cloud9.net (Michael Muller)
Date: Thu, 24 Jun 1999 10:46:20 -0400
Subject: Question - Returning exit codes
References: <7ktbmu$5os$1@nnrp1.deja.com>
Message-ID: <377244BC.6707D3C9@cloud9.net>

sjoshi at ingr.com wrote:
> 
> Hello All
>    I'm calling my python program from a perl script using the system
> command. I was wondering what would be the best way to return exit
> codes from my python script so that they can be intercepted in the perl
> program.
> 
> thanks
> Sunit
> 
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.

   import sys
   sys.exit(1) # replace 1 with your exit code of choice

=============================================================================
michaelMuller = proteus at cloud9.net | http://www.cloud9.net/~proteus
-----------------------------------------------------------------------------
Government, like dress, is the badge of lost innocence; the palaces of
kings
are built on the ruins of the bowers of paradise. - Thomas Paine
=============================================================================



From mstenber at cc.Helsinki.FI  Mon Jun  7 00:47:02 1999
From: mstenber at cc.Helsinki.FI (Markus Stenberg)
Date: 07 Jun 1999 07:47:02 +0300
Subject: File locking
References: <199905301615.MAA00996@mira.erols.com> <14163.61310.936743.849211@anthem.cnri.reston.va.us>
Message-ID: <al8ogisbos9.fsf@kruuna.Helsinki.FI>

"Barry A. Warsaw" <bwarsaw at cnri.reston.va.us> writes:
> You might want to look at Mailman's flock.py module.  The doc string
> says:
>
>     Portable (?) file locking with timeouts.  
>     This code should work with all versions of NFS.
>     The algorithm was suggested by the GNU/Linux open() man page.  Make
>     sure no malicious people have access to link() to the lock file.
> 
> It's also been banged on quite a bit, although I'll make no guarantees 
> that it's bug free.  It's probably pretty close tho'.

At least on our HP-UX devel box (HP-UX 11.0) it failed miserably, didn't
have the time to examine why, though. 

> 
> -Barry

-Markus



From brainwav at hotmail.com  Tue Jun 15 03:15:31 1999
From: brainwav at hotmail.com (Denis Lamarche)
Date: Tue, 15 Jun 1999 07:15:31 GMT
Subject: Newbie for Embeding
Message-ID: <3766267D.AF89C56B@hotmail.com>

I was wondering something.  If you can embed python into c++, can you
then call your c++ fuctions from python?  Can you also modify variable
in your c++ program using python?

Denis Lamarche



From loredo at spacenet.tn.cornell.edu  Wed Jun 23 18:33:52 1999
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Wed, 23 Jun 1999 18:33:52 -0400
Subject: PIL/Tkinter problem on Solaris
Message-ID: <377160CF.B2F1EC29@spacenet.tn.cornell.edu>

Hi folks-

I've been using Python 1.5.2 on Solaris, and everything has been
going fine until I tried to use piddleTK and ImageTk.  After installing
PIL and Piddle, running the piddleTK "sample1.py" file (with the 
required .gif file on hand) gives this traceback:

PIL imported!
Traceback (innermost last):
  File "pd_test.py", line 62, in ?
    test()
  File "pd_test.py", line 57, in test
    sample1(canvas)
  File "pd_test.py", line 52, in sample1
    canvas.drawImage( i, 5, 220)
  File "/home/laplace/lib/python1.5/site-packages/piddle/piddleTK.py", line 320,
in drawImage
    itk = ImageTk.PhotoImage(img)
  File "/home/laplace/lib/python1.5/site-packages/PIL/ImageTk.py", line 83, in
__init__
    self.paste(image)
  File "/home/laplace/lib/python1.5/site-packages/PIL/ImageTk.py", line 113, in
paste
    self.__photo.tk.call("PyImagingPhoto", self.__photo, block.id)
TclError: invalid command name "PyImagingPhoto"


I understand that this is a reported bug for PIL, but I've followed
the instructions in the PIL README and 1.5.2 Setup files to make
sure the Tcl command "PyImagingPhoto" is registered, yet it doesn't
seem to have happened.  By the way, this was done after a "make clean".

Any clues as to what else I should change?

Thanks in advance,
Tom Loredo

PS:  I tried recompiling python both with the included tkappinit.c file,
and with one slightly modified.  The one provided with 1.5.2 has this
line in the WITH_PIL section:

		extern void TkImaging_Init(Tcl_Interp *);

but the PIL README says it should look like this:

	    extern void TkImaging_Init(Tcl_Interp* interp);

Both versions give the same error.



From matz at netlab.co.jp  Thu Jun 10 05:08:02 1999
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 10 Jun 1999 18:08:02 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
 <7j450j$908$2@cronkite.cc.uga.edu> <7j5sr7$eb4$1@nnrp1.deja.com>
 <7j6j8d$58f$2@cronkite.cc.uga.edu> <7j902g$hob$1@nnrp1.deja.com>
 <87n1yc4k2e.fsf@ev.netlab.co.jp> <7jmoo5$f6$1@nnrp1.deja.com>
Message-ID: <87wvxc2zkd.fsf@ev.netlab.co.jp>

adjih at technologist.com writes:

|  Yukihiro Matsumoto <matz at netlab.co.jp> wrote:
|> You don't have to know stack address to implement mark-sweep register,
|> unless you want to make it conservative.  The mark-sweep GC can be
|> written in ANSI-C portable.
|
|But then this can break Python C modules, because references are held
|from the C stack.

In case you abandon ref counting, yes.  But with the method like which
described by Guido in <5laeu8zmza.fsf at eric.cnri.reston.va.us>, you
still don't have to peep in C stack, I think.

						matz.



From xhyuan at oe.fau.edu  Sat Jun  5 02:13:29 1999
From: xhyuan at oe.fau.edu (Xiaohong Yuan)
Date: Fri, 4 Jun 1999 23:13:29 -0700 (PDT)
Subject: ask help for problem with static linking (fwd)
In-Reply-To: <1283593623-47537420@hypernet.com>
Message-ID: <Pine.SGI.3.92.990604230514.11561A-100000@sea30>


On Fri, 4 Jun 1999, Gordon McMillan wrote:

> Xiaohong Yuan wrote:
>
> > I am trying to link my c extension module staticly with python
> > interpreter. But in my c extension module, I have external variable
> > declaration, which makes it fail to be linked with Python
> > interpreter. There must be some way to solve this, could anybody
> > help me with this?
>
> Please forgive the lack of response to your query, but I think people
> are confused about what you are asking.
>
> A variable declared as "extern" will have to be found in some object
> when you link. Is this variable something you expected the Python
> interpreter to expose? If so, what is it? What version of Python?
> What platform? What compiler?
>
> - Gordon
>

I am sorry I didn't make it clear. In my C extension module, I declare
some extern variables. Those extern variables are for my own application.
They can not be found in Python interpreter source code. So when I do
static linking, my c extension module fails to be linked with the python
interpreter. I don't want to add into Python source code the definition of
my extern variables. There might be some way to wrap my c extensions such
that extern variables would not cause a problem. My platform is QNX, my
compiler is watcom compiler. I don't think platform and compiler matter,
the problem might be how should I deal with extern variable when writing
python c extensions.

--Xiaohong Yuan




From gustav at morpheus.demon.co.uk  Wed Jun  9 16:43:00 1999
From: gustav at morpheus.demon.co.uk (Paul Moore)
Date: Wed, 09 Jun 1999 20:43:00 GMT
Subject: Python ActiveX Scripting interface
References: <375e87ae.11001540@news.origin-it.com> <6mw73.1227$h14.5409@newsr2.twcny.rr.com>
Message-ID: <375ecca3.3972892@news.demon.co.uk>

On Wed, 9 Jun 1999 12:07:47 -0400, "Darrell" <news at dorb.com> wrote:

>Check out an earlier thread called "pyscript help" on 6/4/99
>I believe that pyscript.py registers python as a scripting engine.
>I had two problems one was a microsoft dll version problem with comcat.dll
>comcat.dll is used to register com interfaces and I think it gets screwed up
>in SP4.
>The other was a change to the calculator demo.

Hmm, thanks for the pointer. However, I seem to have a matched set of
comcat.dll and ole32.dll (the older versions) so that's not my problem
:-(

For a bit more detail, when I run the following script file (test.ws)
in WSH 2.0 beta, I get the error

D:\winnt\Profiles\Administrator\Desktop\test.ws(0, 1) Windows
Scripting Host: Cannot create script engine : Python

[Script]

<job ID="Test">
<script language="Python">
WScript.Echo("Hello, world!")
</script>
</job>

If I run the supplied "foo2.htm" demo, I just get "Done - errors on
page" displayed, with no indication of *what* error. None of the
python scripted buttons work, but the VBScript one does. The initial
bit of script does execute, though - I can see the results of the
"document.write" calls.

All in all, the whole thing looks very odd...

Actually, I just noticed that the web page directs trace output to the
"win32trace" destination. On starting the client I find that the trace
output says

Collecting Python Trace Output...
 ScriptEngine QI - unknown IID {4954E0D0-FBC7-11D1-8410-006008C3FBFC}
Traceback (innermost last):
  File "E:\Applications\Python\win32com\client\dynamic.py", line 247,
in _make_method_
    fn = self._builtMethods_[name] = tempNameSpace[name]
KeyError: print
Redirecting output to win32trace remote collector
 Hello
Location is
file:///E:/Applications/Python/win32comext/axscript/demos/client/ie/
foo2.htm
Hello from in the form
X is 13
a is Hi there
MyForm is X is 13
a is Hi there
MyForm is

Interesting - the IID it can't find is IActiveScriptProperty. What's
that? It's not mentioned in MSDN (not even the online version on the
MSDN web site, which I assume is fairly up to date...)

Curiouser and curiouser...

Paul.



From tim_one at email.msn.com  Mon Jun  7 00:53:50 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 7 Jun 1999 00:53:50 -0400
Subject: Indentation and grouping proposal
In-Reply-To: <375B3525.F67D424F@erols.com>
Message-ID: <000001beb0a1$bc540a00$5b9e2299@tim>

[R Wentworth, suggests "_(" and ")_" as alternative grouping constructs]

Note that this won't fly, since "_(" already has a meaning in the grammar
(call the function whose name is _).

> [careful argument & analysis deleted]
> ...
> I haven't looked at the Python parser, but naively, this sort
> of "enhancement" would seem to be relatively easy.  If there were
> official blessing for the concept, I'd be willing to look into
> doing the necessary work.

Guido's last public word on the subject is in Tools/Scripts/pindent.py.

Go way back (pre-DejaNews; in python.org's archives, though), and you'll
find some words, and even some patches to implement alternatives; e.g., see
Donald Beaudry's

    http://www.python.org/search/hypermail/python-1994q2/0763.html

The parser hasn't changed all that much since then, although it's likely
changed a lot more than Guido's opinion <wink>.

IIRC, Steven Majewski christened the alternatives "frowning Guidos", in
honor of everyone's favorite some 5 years ago:

    if whatever :(
    print "no whitespace!" ):

or-was-that-"(:"-and-":)"?-ly y'rs  - tim





From ray at commsecure.com.au  Tue Jun  8 11:17:56 1999
From: ray at commsecure.com.au (Ray Loyzaga)
Date: Wed, 09 Jun 1999 01:17:56 +1000
Subject: tstate invalid crash with threads
References: <7jik35$esd$1@nntp6.atl.mindspring.net>
Message-ID: <375D3424.6EF8E85F@commsecure.com.au>

I have been playing with a multithreaded tcp based server for a while,
and during some serious stress testing continually hit a problem
involving the interpreter crashing with "Fatal Python error:
PyThreadState_Delete: invalid tstate".

I have stripped the server down to a pretty basic SocketServer
based process which replies to a client request. It still shows the
bug, and also crashes with a Segmentation fault sometimes.
I have tested the same code on Linux 2.0.36/Python1.5.1, Linux 2.0.36/
Python1.5.2 and Solaris 2.6/python 1.5.2, all crash the same way.
Unfortunately to produce the crash I need to run 12 clients on various
other machines via 100Mbit Ethernet, the server usually dies between
150k
and 1.4M transactions later (it could take an hour).
Sometimes it crashes within 15 minutes.

It appears to be a subtle race in PyThreadState_Delete ....
interestingly, if I uncomment the small sleep in "handle" in the server,
ie.
make the server slower, it seems to work for ever ... 4m transactions
before I gave up. I think the problem only comes if you are creating and
destroying
threads quickly in parallel.

Any comments would be appreciated, note however that the following is
seriously stripped down, it isn't supposed to do anything useful.
The queue is used to make sure that I'm not keeling over due
to thread exhaustion, it isn't required, I just wanted to know if
the process goes thread crazy prior to crashing ... it doesn't.
All this was done on an otherwise idle system with 384Mb memory,
so resource exhaustion seems unlikely. I've seen this bug before
but that particular application didn't really require threads
so I converted to a synchronous server by using an ordinary
TCPServer.

The server is:
from SocketServer import ThreadingTCPServer, StreamRequestHandler
from socket import *
import Queue
import cPickle
import time
from threading import *

server_address = ('', 8000)

cnt = 0L

threadq = Queue.Queue(200)
for i in range(200):
        threadq.put(i)

class BankHandler(StreamRequestHandler):
        def handle(self):
                global cnt

                val = self.connection.recv(1024)
                if (cnt % 10000) == 0:
                        print time.ctime(time.time()), cnt
                cnt = cnt + 1
                #time.sleep(0.01)
                self.connection.send('OK')

class MyServer(ThreadingTCPServer):
        def process_request(self, request, client_address):
                """Start a new thread to process the request."""
                import thread

                if threadq.empty():
                        self.handle_error(request, client_address)
                else:
                        thread.start_new_thread(self.finish_request,
                                        (request, client_address))

        def finish_request(self, request, client_address):
                """Finish one request by instantiating
RequestHandlerClass."""
                t = threadq.get()
                self.RequestHandlerClass(request, client_address, self)
                threadq.put(t)


        def server_bind(self):
                """Called by constructor to bind the socket.
                May be overridden.
                """
                self.socket.setsockopt(SOL_SOCKET, SO_REUSEADDR, 1)
                self.socket.bind(self.server_address)

s = MyServer(server_address, BankHandler)

s.serve_forever()

*********************************************************************************************
And the client is: (client_speed has a single line with .01 in it)

from socket import *
import string
import cPickle
import sys
import time
from select import select

TIMEOUT=30
trans = 0L
while 1:
        try:
                f = open('client_speed')
                delay = string.atof(f.readline())
                f.close()
                sok = socket(AF_INET, SOCK_STREAM)
                sok.connect('bozo', 8000)
                sok.send('Hello')
                x = select([sok.fileno()], [], [], TIMEOUT)
                if x != ([],[],[]):
                        data = sok.recv(2048)
                        if (trans % 10240) == 0:
                                print time.ctime(time.time()), trans,
data
                else:
                        print 'Recv timeout'
                sok.close()
        except:
                print 'failure %s'%(sys.exc_info()[1])
        trans = trans + 1
        time.sleep(0.01)



From Sven.Drescher at dlr.de  Fri Jun 18 04:07:01 1999
From: Sven.Drescher at dlr.de (Sven Drescher)
Date: Fri, 18 Jun 1999 10:07:01 +0200
Subject: tkwait
Message-ID: <3769FE25.6ECF09D4@dlr.de>

Hi!

I know, that is a very simple question, but I didn't find a solution.
Please help me with the syntax of Tk's tkwait window. I found the Tk
Manual page and Deja-news entries, but no solution for my problem.

Thanks
Sven





From larsga at ifi.uio.no  Tue Jun 29 18:55:03 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 30 Jun 1999 00:55:03 +0200
Subject: Please help me to understand classes
References: <Pine.LNX.3.96.990629205752.3228A-100000@Johann>
Message-ID: <wkiu8662ig.fsf@ifi.uio.no>

* Johann Spies
| 
| Can somebody explain to me why the following program produces no
| output:

A good trick to learn in Python is that if you use the interpreter you
can fool around and figure things out much more easily and quicker
than if you use something that resembles the edit-compile-run loop of
Pascal.

| [...skipping class defs]
|         
| a = A()
| c = B()
| c.druk
| a.druk

If you do this in the interpreter you get:

>>> a = A()
>>> c = B()
>>> c.druk
<method B.druk of B instance at 80bdd50>
>>> a.druk
<method A.druk of A instance at 809a4d8>

This is a general rule in Python: if you mention a function or method
by name you just get a reference to it. If you add parentheses it is
called. 

>>> c.druk()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: no arguments expected
>>> a.druk()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: no arguments expected

In your example this failed, since you forgot the self argument. If
you redeclare as:

class A:

    def druk(self):
        print 'Jy is in Class A'

To illustrate:

>>> a=A()
>>> a.druk
<method A.druk of A instance at 80b1e60>
>>> a.druk()
Jy is in Class A
>>> method=a.druk
>>> method
<method A.druk of A instance at 80b1e60>
>>> method()
Jy is in Class A


--Lars M.



From tom at spirit.gcrc.upenn.edu  Tue Jun  8 09:44:03 1999
From: tom at spirit.gcrc.upenn.edu (Tom Fenn)
Date: Tue, 08 Jun 1999 09:44:03 -0400
Subject: Humbug (was Re: Indentation and grouping proposal)
References: <000b01beb14d$b502a480$aaa02299@tim>
Message-ID: <375D1E23.9367A42C@spirit.gcrc.upenn.edu>

The day Python has {'s, is the day I learn Dylan. :)

Tom Fenn



From c.evans at clear.net.nz  Wed Jun  9 07:26:47 1999
From: c.evans at clear.net.nz (Carey Evans)
Date: 09 Jun 1999 23:26:47 +1200
Subject: waste of resources ?
Message-ID: <mailman.2.1572195733.21486.python-list@python.org>

Unix doesn't guarantee you'll get exactly one signal delivered per
event.  To be sure to get all the children, you need to loop over the
processes with waitpid().

I think something like this should do it:

----------
def sigchld_handler(signum, frame):
    print "Checking for processes..."
    try:
        while 1:
            pid, sts = os.waitpid(-1, os.WNOHANG)
            if pid == 0:
                break
            print "Process %d ended with status %04x." % (pid, sts)
    except OSError, detail:
        if detail.args[0] != errno.ECHILD:
            raise
        else:
            print "No more children."
    else:
        print "Some children yet to finish."
----------

I can't actually see any way to keep count well enough and avoid race
conditions, to get rid of the try/except.  Sometimes I wish Python had 
POSIX signals, but not badly enough to actually write it.

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

	     "I'm not a god.  I've just been misquoted."

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




From phd at emerald.netskate.ru  Sun Jun 20 05:07:49 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Sun, 20 Jun 1999 13:07:49 +0400 (MSD)
Subject: DATE ARITHMETIC
In-Reply-To: <7kejrv$e57$1@violet.singnet.com.sg>
Message-ID: <Pine.LNX.4.04.9906201306180.25630-100000@emerald.netskate.ru>

On Sat, 19 Jun 1999, Ajith Prasad wrote:
> I need to do some simple date arithmetic - finding the number of days
> between any 2 days, the age of a person on a day given his birthday. The
> mxDateTime module downloadable from the Python website is more complex than
> what I need. Any other suggestions?

http://members.xoom.com/phd2/Software/Python/
http://www.fortunecity.com/skyscraper/unix/797/Software/Python/

   Download m_lib.tgz and look into opdate.py. Very simple, but powerful
enough for many tasks.

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From avv at quasar.ipa.nw.ru  Wed Jun 30 16:36:13 1999
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Wed, 30 Jun 1999 13:36:13 -0700
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773c3a0.102825445@news.oh.verio.com> <xPPc3.535$or1.22866@typhoon-sf.snfc21.pbi.net> <xJRc3.292$Of3.67094@news.shore.net> <7laovo$c4s$1@Starbase.NeoSoft.COM> <377A24B7.AD6FAA1A@cfdrc.com>
Message-ID: <377A7FBD.DBC69FEB@quasar.ipa.nw.ru>

Lyle Johnson wrote:

> Just to add fuel to the fire, we now have a Python interface to FOX, a
> cross-platform GUI toolkit:
>
>     http://www.cfdrc.com/pub/FOX/fox.html
>
> The latest version of the Python wrappers can be picked up at:
>
>     ftp://ftp.cfdrc.com/pub/FOX/FXPy-0.99.42.tar.gz
>
> I don't know how the speed of FXPy would compare with some of the others
> mentioned (Tk, wxWindows, Qt, GTK, ...) but it would be worthwhile for
> some independent opinions.

It looks great. It seems "intuitively obvious" (for me, of course :-) both
in programming and in using (I mean keyboard handling, at least under
Windows). Yes, the principal question, I ask always when see a new GUI is:

IS IT POSSIBLE TO DO EVERYTHING WITHOUT MOUSE?

Gtk still fails to answer this positively. Qt-2.0 is very close to the
ideal, as I can see.

The only thing surprised me a bit: I failed to switch to a non-US keyboard
layout while in text entry field (window.py example). I run Windows NT 4.0,
SP4. I.e. the toolbar indicator said I'm in the required mode, but the
letters I entered were just those ABC-xyz... I can proceed without this in
getting started and evaluation, but it's necessary for any application I
would deliver to users.

Alexander





From wtanksle at dolphin.openprojects.net  Wed Jun 23 21:14:44 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Thu, 24 Jun 1999 01:14:44 GMT
Subject: Why exceptions shouldn't be used for flow control [Re: YAS to the  "Reading line-by-line" Problem]
References: <Pine.SUN.3.95-heb-2.07.990622183356.3454A-100000@sunset.ma.huji.ac.il> <376FDF70.4756AD38@mindspring.com> <slrn7n0chl.gh6.wtanksle@dolphin.openprojects.net> <377173F7.36BCFF6B@compaq.com>
Message-ID: <slrn7n31k4.idj.wtanksle@dolphin.openprojects.net>

On Thu, 24 Jun 1999 11:55:35 +1200, Greg Ewing wrote:
>William Tanksley wrote:

>> Both should be handled by exception, not by returning a special
>> value.

>I don't agree with that. Reaching the end of a file while
>reading it is not exceptional enough to warrant requiring
>the use of an exception handler to catch it.

"Not exceptional enough".  What does that mean?  (I ain't never had too
much exception! ;-)

It's an exception to the behavior of a function defined to return the next
entity in a file.  It's not an exception to a function which returns the
next entity, or some reserved value otherwise.  There's no such thing as a
partial exception; it's a quality, not a quantity.

>What bothers me about using exceptions for flow control

Presumably you're talking about command flow.  Okay, but exceptions do
nothing BUT affect flow control.  What else can they do?

>in situations like this is that the effect of an exception
>handler is *non-local*, whereas what you're trying to
>catch is really only a local concern.

Is it?  Are they?  The effects of an exception toss are as local as the
user of the function want them to be.

If EOF was truly local, then file.read would always be able to decide what
to do with it -- print a message, abort(), pop up a window, or whatever
else it takes.

>When you write something like

>  while 1:
>    try:
>      line = read_me_a_line()
>    except EOFError:
>      break

>what you really mean by the try-except is "if THIS
>PARTICULAR reading operation reaches the end of
>the file". But that's not what it does - it catches
>any EOFError raised by any operation invoked by
>read_me_a_line() that wasn't caught by something
>else first.

I see your problem, but it's pretty hard to sympathize with it, at least
in this case (and it seems to me in any case).  Consider that 'for' loops
have been terminated by IndexError exceptions for a long time now, with no
problems reported.

If someone else's uncaught EOFError hits this function, something nasty is
almost always going to happen, because of how far this is away from its
correct target.  This is robust by a definition of 'robust' with which
some are not perfectly acquainted: if it fails, it does so loudly so that
the tester notices and reports the problem.

In addition, the complexity of the failure required to produce this is
pretty high.  How many files do you need to read to get the next line from
this file?  What would have happened if you'd ignored the EOF if it had
been returned otherwise?

>In my experience, that sort of behaviour tends to
>mask bugs. I prefer results which are not errors to
>be returned normally, so that I can be sure where
>they came from.

But reading data after EOF _is_ an error condition!  A normal return is
precisely what you don't want.  A normal return offers permission for the
program to keep going as usual.

What you want is not a normal return, it's an _ab_normal return. Something
like returning (int)-1 when the function's only expected to return char,
or (in Python) returning None.  The problem?  These things are SO easy to
ignore!!!!  C is famous for noiselessly casting the -1 to a char (look at
the old Bash security bug), and Python is even worse -- any return you can
possibly make will almost certainly fit in noiselessly with anything you
do with it.  This is Python's ease of use.

The only solution which works in each case is exceptions.  This doesn't
mean that using exceptions guerantees bug freedom; nothing can do that.
It simply means that it catches the majority of the bugs the majority of
the time.

You ought to feel lucky.  Some respected programming gurus recommend
calling abort() on any odd results, to make misuse of their API as
blatantly obvious as possible.

>Greg

-- 
-William "Billy" Tanksley
Utinam logica falsa tuam philosophiam totam suffodiant!
   :-: May faulty logic undermine your entire philosophy!



From delgado at Mathematik.Uni-Bielefeld.DE  Fri Jun 18 11:55:39 1999
From: delgado at Mathematik.Uni-Bielefeld.DE (Olaf Delgado)
Date: Fri, 18 Jun 1999 17:55:39 +0200
Subject: Newbie: Truth values (three-valued logic)
In-Reply-To: <1282404372-119074879@hypernet.com>
References: <7kd3ke$d1n$1@nnrp1.deja.com> <1282404372-119074879@hypernet.com>
Message-ID: <Pine.OSF.4.05.9906181751100.23895-100000@pinax.Mathematik.Uni-Bielefeld.DE>

On Fri, 18 Jun 1999, Gordon McMillan wrote:

> Note that, if you want to retain DeMorgan's Laws, you don't have 
> much choice. ANSI SQL defines a three valued logic which meets these 
> conditions, and comes up with 
>   NOT maybe == maybe
> (well, NOT unknown == unknown), which makes sense as long as you 
> rmember that NOT is an operator, and the result of a non-trivial 
> operation on an unknown operand is probably unknown.
> 
> Which is why most DBA's won't allow NULL values in any column that 
> might be used in a WHERE clause - three valued logic is often 
> non-obvious.

Well, NOT unknown == unknown seem perfectly obvious to me, but then I'm a
Mathematician and everyone knows we're strange guys.

Obviously yours,
Olaf

-- 
   ////         Olaf Delgado Friedrichs, Uni Bielefeld, Mathematik
   Olaf         Postfach 100131                  (room:  V4-109)
   `='          D-33501 Bielefeld, Germany       (phone: +49 521 106-4765)
                           http://www.mathematik.uni-bielefeld.de/~delgado




From " HPeter.Friedrichs" at alliedsignal.com  Wed Jun 30 19:49:58 1999
From: " HPeter.Friedrichs" at alliedsignal.com (H. P. Friedrichs)
Date: Wed, 30 Jun 1999 16:49:58 -0700
Subject: Help: Can you explain this error? (contains code snip)
Message-ID: <377AAD25.96947EBA@alliedsignal.com>

Hello:

This is week 2 for me in python...What fun!

I have stumbled across an interesting problem. When the code below is
run under Win95 (python 1.5.2),it works without problems. If the same
code is run under Linux (python 1.4) I get the error:

Traceback (innermost last):
    File "socksrv3.py", line 100, in ?
        han=int(handle)    #extract integer representation of handle
TypeError: int() argument can't be converted to int

The code section below is part of a larger program that receives ascii
"link messages" through a TCPIP socket. The format for the message is as
follows: an ascii "handle" followed by a tilde, followed by an ascii
"verb", followed by a tilde, and concludes with an ascii "value" An
example looks like this: "123~opensym~0"

Anyhow, the code below uses splitfields to break the message into it's
constituent parts. Later, at the line signified by the marker "====>"
the
program blows up when I attempt to convert the string "handle" into it's

integer equivalent.

What gives? Is the 1.4 version of python buggy, or am I overlooking
something? Thanks for your help in advance.

Pete




        #*** receive link  msg ****
        cmd=conn.recv(1024)             #try to get a message
        if not cmd:
            print"closing connection..."
            conn.close()                #good housekeeping...close
cnnctn
            break


        #*** extract components of link message ***********
        fieldlist=splitfields(cmd,"~")      #break message to
constituents
        handle=fieldlist[0]                 #extract handle
        verb=fieldlist[1]                   #extract verb
        value=fieldlist[2]                  #extract value



        #****** if command requested write to existing symbol, do this:
        if verb == "writesym":
====>       han=int(handle)        #extract integer representation of
handle
            masterdict[han]=float(value)

            response=handle+"~"+verb+"~"+value #build response string
            print "tx'd response:",response
            print "dict status:  ",masterdict
            conn.send(response)         #send response
            continue


--

-----------------------------------------------
| The opinions expressed here are my own,     |
| and not necessarily those of my employer.   |
-----------------------------------------------





From techentin.robert at mayo.edu  Thu Jun 10 15:47:52 1999
From: techentin.robert at mayo.edu (Bob Techentin)
Date: Thu, 10 Jun 1999 14:47:52 -0500
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <wk909syqr4.fsf@ifi.uio.no> <375FF969.C879B4CD@iname.com>
Message-ID: <37601668.D9105225@mayo.edu>

These flame wars are more than five years old.  Tcl has had namespaces
and a byte-code compiler for years.

If you want "current" issues, you might just read comp.lang.tcl, or the
"frequently made mistakes" available from the FAQs at
http://www.tclfaq.wservice.com/tcl-faq/  You'll find complaints about
commenting out blocks of code and accidental conversions to octal.

Fernando Mato Mira wrote:
> 
> Lars Marius Garshol wrote:
> 
> > * Fernando Mato Mira
> > |
> > | I'm trying to collect a list of all the _current_ issues making tcl
> > | a bad language choice.
> >
> > You can find three interesting articles on this at
> >
> > <URL: http://itrc.uwaterloo.ca/~papresco/tcl/>
> 
  [snip]
> 
> http://www.vanderburg.org/~glv/Tcl/war/
> http://www.vanderburg.org/~glv/Tcl/war2/

-- 
Bob Techentin                                  techentin.robert at mayo.edu
Mayo Foundation                                           (507) 284-2702
Rochester MN, 55905 USA   http://www.mayo.edu/sppdg/sppdg_home_page.html



From dave_kirby at eur.3com.com  Thu Jun 17 12:32:09 1999
From: dave_kirby at eur.3com.com (Dave Kirby)
Date: Thu, 17 Jun 1999 16:32:09 GMT
Subject: wierdness with pythonwin dialogs, threads & refcounts
References: <1282492928-113747642@hypernet.com>
Message-ID: <37690c3e.27197698@nnrp.uk.insnet.net>

On Thu, 17 Jun 1999 09:21:51 -0500, "Gordon McMillan"
<gmcm at hypernet.com> wrote:

>Dave Kirby writes:
> 
>> I have found strange behaviour (strange to me at least) when
>> creating a pythonwin dialog on a separate thread created using the
>> standard python thread module.  The dialog object is never deleted
>> since the thread is keeping references to it. 
>> [example snipped]
>
>First, there's no incompatibility with the thread module in 
>Pythonwin. 
>
>Second, you should use threading.py, not the low level thread module 
>(which is very, very primitive). Or you can use 
>win32process.beginthreadex which is a wrapper around MFC threads. 
>This lets you manipulate priorities. (As of a couple years ago, there 
>was a small memory leak involved with using MFC threads, however).

I have just tried both the threading.py and beginthreadex with the
same results - vis:

[carrying on from the previous example]
>>> import win32process
>>> test()
Edit text changed!
refs: 2
>>> x = win32process.beginthreadex( None, 10000, test, (), 0 )
>>> refs: 6
>>> import threading
>>> thread = threading.Thread(target = test)
>>> thread.start()
>>> refs: 6

Again the refcount is 6 when test is run from a separate thread, but 2
when run in the main thread. 


>
>As for the source of your questions - without doing things in a debug 
>build of python / Pythonwin, it's almost impossible to tell what's 
>going on, or even if there really is a problem.
>
>Using sys.getrefcount increments the refcount. Printing something in 
>the interactive interp binds the thing to a var named "_" (to 
>optimize looking at it again). Printing to stdout in Pythonwin 
>involves all kinds of hidden manipulations.
>
>If you want to know whether you're leaking TestDialogs you could code 
>a loop that creates threading.Thread's that show a TestDialog and 
>then joins the Thread. Run it for as many iterations as you can 
>stand, and watch the process in TaskManager.

I am pretty sure there is a real problem here. Whatever the hidden
manipulations are it shouldnt make any difference which thread it is
running in. I discovered this problem because the __del__ method was
not being called for the dialog class in the program I am writing.
When I moved it to the main thread it was destroyed ok. 

I will play around with this some more and try to narrow it down
further, but I am not intimate with either the internals of pythonwin
or Windows threads, so don't hold your breath.  One thought that did
occur to me is that the python thread/threading modules and
win32process.beginthreadex are all creating worker threads so there is
no message processing being done to clean up the dialog when it is
closed. I will test this hypothesis as soon as I have figured out how
to create GUI threads in pythonwin.

Regards,

	Dave K




From graham at sloth.math.uga.edu  Wed Jun  9 14:17:43 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 9 Jun 1999 18:17:43 GMT
Subject: Python 2.0
References: <000b01beb21f$a17ff340$2fa22299@tim>
Message-ID: <7jmb47$r22$1@cronkite.cc.uga.edu>

Tim Peters (tim_one at email.msn.com) wrote:
: The caveats are explained well in the Java Language Spec, section 12.6.  In
: short, you can't know which thread a finalizer will run in; you can't be
: sure that an object is unreachable at the time its finalizer is invoked; you
: can't know *when* a finalizer will be invoked; even if it is unreachable at
: the time its finalizer is invoked, you can't be sure that it won't become
: reachable again *after* the finalizer returns; and in the case of cyclic
: object structures nothing is specified about the order in which the objects'
: finalizers will be called (indeed, they may all run simultaneously).

There are two obvious answers to almost all the above problems in Python

a) don't call finalisers for circularly referenced collected objects.
	 Instead just collect their memory. This is not perfect but better than 
	 current Python which doesn't run finalisers on circularly referenced
	 objects nor collect their memory.

b) better still use a two pass collector. If you don't know what that is
	 just think about what you need to do (or not do) in pass 1 to ensure
	 that all the objects needed for a __del__ method to execute in a known
	 environment are there. <wink> 

graham
-- 
           As you grow up and leave the playground
       where you kissed your prince and found your frog
          Remember the jester that showed you tears
                   the script for tears



From mlh at idt.ntnu.no  Mon Jun 21 08:29:24 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 21 Jun 1999 14:29:24 +0200
Subject: simple text file 'parsing' question
References: <376B1AAC.19FE8BCE@mysolution.com> <m3ogiaq31k.fsf@havenrock.com>
Message-ID: <y0jyahdhh3v.fsf@vier.idi.ntnu.no>

Matt Gushee <mgushee at havenrock.com> writes:

> KP <terocr at mysolution.com> writes:
> 
> > Here's my dilema: a directory filled (200+) with small emails. My goal
> > is to strip all the headers and combine them into one file. I can read
> > all the files just fine and write them all to one file, but I cannot
> > discern how to strip the headers.

Just remove everything up to and including the first blank line. That
represents the end of the header. For instance:

  file = open(resultFileName,'w')

  # ...

  lines = open(name).readlines()
  lines = lines[lines.index("")+1:]

  file.writelines(lines)

  # ...

  file.close()



--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From mukundd at my-deja.com  Wed Jun 30 15:17:28 1999
From: mukundd at my-deja.com (mukundd at my-deja.com)
Date: Wed, 30 Jun 1999 19:17:28 GMT
Subject: Problem embedding C++ code
Message-ID: <7ldqfs$nbj$1@nnrp1.deja.com>

Hi,

I am trying to write a wrapper around a library written in C++ and I am
running into lots of problems.

I am running Python 1.5.2 on SunOS 5.5.1 (compiled with SUN's cc and
linked with Sun's CC)

The wrapper is written in a file called link-wrap.c and currently it
contains only one very simple function.

CC -c -KPIC link-wrap.c -I/home/mdeshpan/SW/include/python1.5
"link-wrap.c", line 13: Warning: String literal converted to char* in
formal argument 1 in call to Py_BuildValue(char*, ...).
1 Warning(s) detected.
CC -G link-wrap.o -lapic -lpthread -o link.so

libapic.so is the library which I am trying to write the wrapper for,
it was compiled using Sun's CC compiler

when I try to import link module in python I get the following error..

daffy:~/Swig$python
Python 1.5.2 (#3, Jun 29 1999, 11:17:23) [C] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import link
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ImportError: ld.so.1: python: fatal: relocation error: file
./libapic.so: symbol _ex_rethrow_q: referenced symbol not found

any help would be greatly appreciated..

Thanks in advance

Mukund


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From tim_one at email.msn.com  Fri Jun  4 21:42:17 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 4 Jun 1999 21:42:17 -0400
Subject: threads
In-Reply-To: <14168.18308.709890.544779@bitdiddle.cnri.reston.va.us>
Message-ID: <000a01beaef4$a53dda20$f19e2299@tim>

[Jeremy Hylton, on why free-threading may slow things down]
> If I recall correctly, the problem is that taking away one big lock
> usually requires introducing lots of little locks.  (There's some kind
> of conservation of concurrency going on.)  Right now, there is no need
> to have locks on access to builtin types like dictionaries, because
> only a single thread can hold the interpreter lock anyway.  When that
> goes away, you need to add lots of fine-grained locking on the builtin
> types.

Refcount incref/decref are also unsafe without a lock, and one of those
occurs even more often than it's suggested "real GC" would eliminate the
need for them <wink>.

the-guarantee-of-exclusion-simplifies-much-of-the-implementation-ly y'rs  -
tim





From tismer at appliedbiometrics.com  Wed Jun  2 06:09:11 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Wed, 2 Jun 1999 10:09:11 GMT
Subject: need help checking for stack overflow in Win32
References: <5lk8to3su3.fsf@eric.cnri.reston.va.us> <7j1pcc$256$1@m2.c2.telstra-mm.net.au>
Message-ID: <375502C7.39C4FD0E@appliedbiometrics.com>


Mark Hammond wrote:
> 
> Actually, I was about to post this yesterday, and to prove my point I wrote:
> 
> def die():
>     die()
> 
> Expecting to see the crash.  I was then going to go into a rant about how
> win32api.Apply would actually let you catch this (and any other Win32)
> exception as a temporary solution.
> 
> But it worked correctly - ie, Python raised a RuntimeError: Max Recursion.
> 
> So I guessed that 1.5.2 had fixed this, presumably by reducing the recursion
> count for Win32.  Apparently not!

Nope. The trap is:
You compiled with VC++ 6.0 which has better register spilling.
Therefore, recusion costs only 100 bytes, not 104.
Not calculate 104 * 10000 and you see why 1 MB stack
lets it crash.

I submitted a patch program for this kind of
problems, so I think there is no need to change
Python 1.5.2 at all.

cheers - chris


-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From ivnowa at hvision.nl  Fri Jun  4 12:42:15 1999
From: ivnowa at hvision.nl (Hans Nowak)
Date: Fri, 4 Jun 1999 16:42:15 GMT
Subject: Alien whitespace eating nanovirus strikes again!
In-Reply-To: <3757F28B.BF6411FE@cwcom.net>
References: <3757F28B.BF6411FE@cwcom.net>
Message-ID: <199906041640.SAA05103@axil.hvision.nl>

On 4 Jun 99, Daniel Faulkner wrote:

> How many messages are there about this white space stuff? Looks like millions whats
> it all about?

Oh, this thread is only trivial compared to the one about the Python 
version that only exists in Guido's head so far. :o)

You are new to Python? The short story is:

in languages like C, you define blocks using block delimiters:

void p() 
{
    int x;
    blahblah();
}

or Pascal,

procedure P;
var x;
begin
    blahblah;
end;

(OK, so these examples don't make much sense. :-) Anyway, you'll get 
the point. Python does not have such delimiters; rather than that, 
indentation (thus, use of whitespace) is used:

def p():
    x = 0
    blahblah()

Same thing for if-constructs, while, etc.

Not everybody likes to use whitespace for indentation. This seems to 
be especially true for people coming from a C or Perl background. If 
you search DejaNews for old messages, you will find tons of messages 
dealing with this subject (usually people asking for block delimiters 
vs people defending the Python style).

I dare say that many people like this Python feature a lot, since it 
improves readability greatly. But as said, not everyone does, and 
this fuels discussions greatly. Those are really a waste of bandwidth 
though, since (IMO) Guido is not going to change it. :-)

Veel liefs,


--Hans Nowak (ivnowa at hvision.nl)
Homepage: http://fly.to/zephyrfalcon



From tim_one at email.msn.com  Wed Jun 16 01:21:57 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 16 Jun 1999 01:21:57 -0400
Subject: Newbie: Truth values (three-valued logic)
In-Reply-To: <Pine.OSF.4.05.9906152037050.23333-100000@pinax.Mathematik.Uni-Bielefeld.DE>
Message-ID: <000f01beb7b8$2722a540$2ca22299@tim>

[Olaf Delgado]
> I am just starting to use python seriously. Currently, I am porting a
> small class library from C++ to python, which is kind of fun. Everything's
> much more simple and elegant now.

Except for C++ <wink>.

> There's one thing I could do in C++, though, which doesn't seem to work in
> python: some methods need to return truth values, which can be 1 for
> 'true', 0 for 'false' and x for 'frankly I don't know'.
> ... [but 'not' only delivers true or false] ...
> As far as I know, the logical negation operator can not be overloaded,
> so I have no chance to change this, right?

Right.  I'd try overloading "~" instead (the unary prefix "complement"
operator, special method name __invert__):

>>> class uncertain:
        def __nonzero__(self):
            return 0
        def __repr__(self):
            return 'dunno'
        def __invert__(self):
            return self

>>> maybe = uncertain()
>>> if maybe or ~maybe:
    print "maybe or not maybe!"
else:
    print "we don't need no stinking excluded middle!"

we don't need no stinking excluded middle!
>>>

Under this hack, may work best if you use 0 to represent false and -1 to
represent true:

>>> ~0
-1
>>> ~-1
0
>>>

OTOH, overloading sucks when you have to fight the language -- and 90% of
the time even when you don't <0.9 wink>.

functions-spelled-as-functions-rarely-confuse-ly y'rs  - tim





From tim_one at email.msn.com  Sun Jun  6 16:11:39 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 6 Jun 1999 16:11:39 -0400
Subject: Suggestion: Dutch Python mailinglist
In-Reply-To: <slrn7lli2q.ia.gerrit@gerrit.palga.uucp>
Message-ID: <000101beb058$c992bc00$e39e2299@tim>

[Gerrit Holl]
> This e-mail is in two languages. Look below for the Dutch version.
> Deze e-mail is in twee talen. Kijk beneden voor de Nederlandse versie.
>
> Hello all,
> I'm a 13-years-old Dutch boy, and my English is not good. Now I have the
> chance to set up a Dutch python mailinglist. I know there are many Dutch
> python-users, but most of them know English very well. So will there be
> any users who'll subscribe to such a mailinglist, or will it
> remain very quiet?
>
> regards,
> Gerrit.

Gerrit, rest assured that your English is better than that of most
13-year-old Americans!  Don't know about young Englishmen, but they spell
funny anyway <wink>.

one-day-the-whole-world-will-speak-python-ly y'rs  - tim





From akuchlin at mems-exchange.org  Tue Jun 22 17:24:26 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: Tue, 22 Jun 1999 17:24:26 -0400 (EDT)
Subject: Unbound method
In-Reply-To: <001701bebcdd$73d3b080$656211c8@miro.lami.pucpr.br>
References: <001701bebcdd$73d3b080$656211c8@miro.lami.pucpr.br>
Message-ID: <14191.65290.126498.62145@amarok.cnri.reston.va.us>

faria writes:
>Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
>Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>>> from test import *
>>>> x = A
>>>> x.a
>'Greetings from Brasil!!!'
>>>> x.pa
><unbound method A.pa>

x is a class, not an instance.  This can be made clearer with
different variable names:

>>> klass = A
>>> instance = A()
>>> klass
<class __main__.A at e21d8>
>>> instance
<__main__.A instance at f22c8>
>>> klass.a
'Greetings from Brasil!!!'
>>> instance.a
'Greetings from Brasil!!!'
>>> klass.pa
<unbound method A.pa>
>>> instance.pa
<method A.pa of A instance at f22c8>

If you're expecting x = A to create an instance of the class A, that
expection is incorrect; you have to do x = A() to create an instance.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
In the design of fission reactors man was not an innovator but an unwitting
imitator of nature.
    -- George A. Cowan, "A Natural Fission Reactor"




From rozzin at hotmail.com  Sat Jun 12 01:43:02 1999
From: rozzin at hotmail.com (Joshua Rosen)
Date: Sat, 12 Jun 1999 01:43:02 -0400
Subject: import posixpath
References: <199906100438.AAA16802@tor.dhs.org>
Message-ID: <3761F366.5AA93D00@hotmail.com>

    Can't one import posixpath, regardless of what OS one is running on, and
expect posixpath.split to give the proper result, here?






From mwh21 at cam.ac.uk  Wed Jun 16 17:16:58 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 16 Jun 1999 22:16:58 +0100
Subject: ANN: Stackless Python 0.2
References: <37628EAA.C682F16C@appliedbiometrics.com> 		<7k4fd4$c04$1@brick.cswv.com> 		<376636EA.84240129@appliedbiometrics.com> 		<7k6ttr$i5g$1@brick.cswv.com> 		<m3909lszww.fsf@atrus.jesus.cam.ac.uk> <m37lp4tonr.fsf@atrus.jesus.cam.ac.uk> <37679151.6C6B29DF@appliedbiometrics.com>
Message-ID: <m3674nu9px.fsf@atrus.jesus.cam.ac.uk>

Christian Tismer <tismer at appliedbiometrics.com> writes:

> Michael Hudson wrote:
> 
> > Got this one; PyEval_Frame_Dispatch is defined static in
> > Python/ceval.c. Take the keyowrd static out, and it builds and runs
> > fine (in my limited testing so far).
> 
> Ah, thanks. Yes, this function was a local eval_frame_dispatch,
> before I realized that it needs to be exposed as an API
> function. Ahem, forgot to change the decl.
> 
> Someone else told be that in frameobject.h, my forward
> type declaration gives problems with some compilers.
> What is the standard way to do this?

The problem I had with this was that you had code like

typedef struct _frame PyFrameObject;

... stuff ...
 
typedef struct _frame {
    ... stuff ...
} PyFrameObject;

which egcs didn't like; I changed it to 

typedef struct _frame PyFrameObject;

... stuff ...
 
struct _frame {
    ... stuff ...
};

and everything went smoothly.

> thanks - chris

HTH
Michael



From zigron at jps.net  Fri Jun 25 22:57:48 1999
From: zigron at jps.net (Zigron)
Date: Fri, 25 Jun 1999 19:57:48 -0700
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773c3a0.102825445@news.oh.verio.com> <xPPc3.535$or1.22866@typhoon-sf.snfc21.pbi.net> <xJRc3.292$Of3.67094@news.shore.net>
Message-ID: <AlXc3.1578$or1.81548@typhoon-sf.snfc21.pbi.net>

Michael P. Reilly <arcege at shore.net> wrote in message
news:xJRc3.292$Of3.67094 at news.shore.net...
> Zigron <zigron at jps.net> wrote:
> : Kevin Dahlhausen <morse at harborcom.net> wrote in message
(snip)
> I wouldn't care if wxPython was standard, I would still use Tkinter,
> better design philosophy, IMTO (most anything is better than
> MFC-oriented systems, even curses).
>
> There are other graphical packages for Python out there too, it would
> be appropriate to contrast them as well.
    Exactly. There are plenty of GUI packages out there, why have a
standard one? For me, for everything I want to do, wxPython is a very
obvious and clear choice. For you, its Tkinter.
    What annoys me is that Tkinter *is* the de-facto standard, and is
actually now shipped with python itself. If TCL wasn't involved, if they
DID use the TK/C api, that wouldn't be such a problem.. but they do,
so, the whole 'if they did.....' argument is moot.

> Python is a good language, with good extension capabilities.  I think
> there is enough room for more than one GUI package.

    I think there will always be handfulls, regardless. No one will ever
agree that one is the best.

--S





From ovidiu at cup.hp.com  Thu Jun 10 22:25:19 1999
From: ovidiu at cup.hp.com (Ovidiu Predescu)
Date: 10 Jun 1999 19:25:19 -0700
Subject: complete GNU readline support?
References: <37445622.1EE18CFD@cup.hp.com> <374B8030.A29351CA@_nospam_ken.x13.com> <3754600B.F78CD96A@cup.hp.com> <87pv3ft7i0.fsf@ifm.uni-kiel.de>
Message-ID: <yosd7z3v5gw.fsf@hpcll563.cup.hp.com>

I need to figure out some copyright issues with my company, as it was
written for a project here. I don't expect to have major problems with
the last item but it may take a while (hopefully not too much).

-- 
Ovidiu Predescu <ovidiu at cup.hp.com>
http://www.geocities.com/SiliconValley/Monitor/7464/



From sragsdale at my-deja.com  Thu Jun  3 22:52:16 1999
From: sragsdale at my-deja.com (sragsdale at my-deja.com)
Date: Fri, 04 Jun 1999 02:52:16 GMT
Subject: Changing environment variables into python variables
Message-ID: <7j7f0t$165$1@nnrp1.deja.com>

I'm writing a script that needs to import environment variables into
python as standard variables.  I.E. if the environment variable USER is
set to 'soren', after my program starts up and does its thing I want
there to be a variable named USER with a value of 'soren'.

I can write this program with little effort in perl:

foreach my $var (keys(%ENV)) {
  my $cmd = "\$".$var." = '".$ENV{$var}."'";
  print $cmd,"\n"; # just to see what I'm doing
  eval($cmd);
}

When I try to write this in python, I have this error:

-------------------------------
import posix
for var in posix.environ.keys():
    command = var+" = \""+posix.environ[var]+"\""
    print command
    eval(command)
-------------------------------
> ./importer.py
Traceback (innermost last):
  File "./importer.py", line 12, in ?
    eval(command)
  File "<string>", line 1
    WF_VC_PLUG_INS = "/usr/local/aw/userconfig/composer/plug-ins"
                   ^
SyntaxError: invalid syntax
-------------------------------

What am I doing wrong here?  How can I, at runtime, define a variable
and assign it a value?


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From never at matavnet.hu  Wed Jun 23 17:51:20 1999
From: never at matavnet.hu (Zsolt Hever)
Date: Wed, 23 Jun 1999 23:51:20 +0200
Subject: [ANN]: Article about Python in Hungary
Message-ID: <199906232156.RAA20257@python.org>

Hi there,

I saw  a discussion about "Is Python dying?" on the mailing list some
months ago. That time I decided to write an article about Python in a
paper.
Now the article was published in the CHIP Magazine in Hungary yesterday.
The title is The friendly snake. 
I made a HTML version of that <
http://www.cab.u-szeged.hu/~h532851/cikk/index.htm >.  
It was written on the base of the FAQ, interviews, articles about Python
and other languages (Perl, Rebol) and my experience.

The content:
========
Shortly about Python
What language is that?
Installing of Python 
Documentation and examples
Learning the language        
Writing and running of Python programs
Life of Python
Some interesting developments         
Summary

Hungary is not a big country (10 million inhabitants), and not everybody is
a programmer.
I put a counter on the main page <
http://www.cab.u-szeged.hu/~h532851/python.htm > to see the interest.

I hope it helps Python to be known better. 

Zsolt
   

          




From gduzan at gte.com  Tue Jun 29 14:26:41 1999
From: gduzan at gte.com (Gary Duzan)
Date: Tue, 29 Jun 1999 14:26:41 -0400
Subject: ANN: Stackless Python 0.2
References: <000c01bec11f$8dfbf800$e19e2299@tim> <19990629121441.A19011@longford.cs.monash.edu.au>
Message-ID: <37790FE1.4F778B4C@gte.com>

Toby J Sargeant wrote:
> 
> On Mon, Jun 28, 1999 at 12:34:48AM -0400, Tim Peters wrote:
> > Nope!  It's a beauty of the implementation already that each code object
> > knows exactly how much "Python stack space" it needs, and (just) that much
> > is allocated directly into the code object's runtime frame object.  IOW,
> > there isn't "a Python stack" as such, so there's nothing to change here --
> > the stack is implicit in the way frames link up to each other.
> 
> This might be completely irrelevant, but during the course of my masters, I
> considered doing this kind of thing to java in order to allow asynchronous
> threads to share stack frames (don't ask...). My supervisor complained bitterly
> on the grounds that function invocations where orders of magnitude more
> common than object creation, and hence associating memory
> allocation/deallocation with every call was considered horrendously
> inefficient.

   This appears to be the conventional wisdom. The optimization class I
took used an ML-like language, and the first thing we did was to move
heap frames, which was the default, to the stack (i.e. closure
optimization.)
It might be less of an issue in an interpreted language if it already has
heap
allocation (or other sorts of) overhead to deal with when making calls.
There may be other allocation tricks which can be used, as well.

> It seems that this should affect Stackless Python equally as much. Does anyone
> have anything to add on the subject? I would imagine that frames could be
> allocated and managed in chunks to alleviate a lot of the memory management
> load...

   I would think that as well (at least at first.) Maybe pools of different
sizes of frame to limit the space overhead.

> Toby.

					Gary Duzan
					GTE Laboratories



From graham at sloth.math.uga.edu  Thu Jun  3 11:22:57 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 3 Jun 1999 15:22:57 GMT
Subject: Python 2.0 (perhaps off topic)
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
Message-ID: <7j66kh$il5$5@cronkite.cc.uga.edu>

Paul Prescod (paul at prescod.net) wrote:
: I think that that is all anyone has been saying. Many people like the
: current semantics and don't trust that a full GC would deterministically
: invoke finalizers for arbitrary resources.

Why do you think this? See my post about adding a mark and sweep
collector *on top of* the existing ref count semantics.

graham
-- 
                          So I turned around
                    And 40,000 headmen bit the dirt
                      Firing twenty shotguns each
                        And man it really hurt



From news at dorb.com  Tue Jun 22 20:25:04 1999
From: news at dorb.com (Darrell)
Date: Tue, 22 Jun 1999 20:25:04 -0400
Subject: Help with String Manipulation
References: <0UUb3.1337$el4.44932@ord-read.news.verio.net>
Message-ID: <0SVb3.1951$U6.7016@newsr2.twcny.rr.com>

>>> s='now is the time'
>>> l=list(s)
>>> l.reverse()
>>> l
['e', 'm', 'i', 't', ' ', 'e', 'h', 't', ' ', 's', 'i', ' ', 'w', 'o', 'n']
>>> import string
>>> string.join(l,'')
'emit eht si won'

--
--Darrell
Samuel G. Williams <samw at masu.wwa.com> wrote in message
news:0UUb3.1337$el4.44932 at ord-read.news.verio.net...
> I know this is a very stupid question and one that has been addressed
> within available documentation, however I need a quick hand. I am in need
> of a routine to reverse the order of a string. I have tried several things
> that have resulted in failures. It is really stupid, but I am just
> learning python. I can do this readily in C, perl, tcl, but am at a loss.
> I would appreciate any help I can get.
>
> Thanks in advance.
>
> --
>
>
> Sam Williams  wb5yni samurai at acm.org
> Senior IS Staff Specialist         samw at wwa.com
> --------------------------------------------------------------------------
-
>
> Thinking of running your critical apps on NT?
>
> Isn't there enough world suffering?





From tjreedy at udel.edu  Tue Jun 22 02:52:22 1999
From: tjreedy at udel.edu (Terry Reedy)
Date: 22 Jun 1999 06:52:22 GMT
Subject: newbie idiom question
References: <m3emj4yezq.fsf@monsoon.swcp.com>
Message-ID: <7knbr6$kas$3@news.udel.edu>

In article <m3emj4yezq.fsf at monsoon.swcp.com>, alex at mindlube.com says...
>
>Something I keep getting tripped up about is that objects being
>iterated in "for" statements cannot be modified.

It depends on whether the objects are mutable or not.

>I keep trying to do this:
>
>      >>> eggs = [1,2,3]
>      >>> for spam in eggs:
>      ...       spam = 'cooked'
>      ...
>      >>> eggs
>      [1, 2, 3]
>      >>> 

Modifying the list by replacement is relatively easy.

eggs = [1,2,3]
for i in range(len(eggs)): eggs[i] = 'cooked'

>The tutorial says this:
>
>      If you need to modify the list you are iterating over, e.g., duplicate
>      selected items, you must iterate over a copy. The slice notation makes
>      this particularly convenient:
>
>      >>> for x in a[:]: # make a slice copy of the entire list
>      ...    if len(x) > 6: a.insert(0, x)
>      ... 
>      >>> a
>      ['defenestrate', 'cat', 'window', 'defenestrate']

This applies to insertions and deletions.  Without making a copy, the result 
in difficult to predict.

>Understood, but what if you want to modify each element in a list?

Do you actually want to change the elements (leaving the list the same) or 
change the list?

Terry J. Reedy




From avv at quasar.ipa.nw.ru  Fri Jun 25 14:28:55 1999
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Fri, 25 Jun 1999 11:28:55 -0700
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net>
Message-ID: <3773CA67.7057F797@quasar.ipa.nw.ru>

Mikael Lyngvig wrote:

> I'd like some discussion on whether or not it is a good idea to
> abandon TkInter,

Yes, it's a good idea, I vote.

> before it gets too widely used, and instead adopt
> wxPython as the primary GUI framework for Python.

I trace the mailing lists for the wxWindows itself and wxPython. As I
understand, Robin Dunn *alone* tries to keep in sync with the wxWindows,
which is *very* intensively developed by a dozen or more of active
contributors. It seems that if it were more possible to extend wxPython
and let it do more collaboratively it would be best we can gain for a
native Python GUI. (Probably I'm not absolutely correct in all of my
observations, unfortunately I do not have enough spare time to participate
in wxPython deeper).

I'm also glad to observe that wxWindows community is *very* friendly to
wxPython. It even extends its manual to describe in detail small
deviations which wxPython imposes on the underlying toolkit to let it look
and feel more Pythonically.

Alexander





From mlv at pobox.com  Wed Jun  2 23:06:13 1999
From: mlv at pobox.com (Michael Vezie)
Date: 2 Jun 1999 23:06:13 -0400
Subject: pickle vs .pyc
References: <7j43pn$55l$1@mlv.mit.edu> <m3pv3eian1.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <7j4rf5$5h0$1@mlv.mit.edu>

In article <m3pv3eian1.fsf at atrus.jesus.cam.ac.uk>,
Michael Hudson  <mwh21 at cam.ac.uk> wrote:
>
>Hmm, you're relying on all the data you're storing having faithful
>__repr__ methods. This certainly isn't universally true. I'd regard
>this method as too fragile.

Granted.  It wouldn't do for the file to contain:

fast = <__main__.fast instance at 8137590>
slow = <__main__.slow instance at 8137590>

But the data I'm dealing with is all dictionaries of arrays of dictionaries, 
all containing strings.  Pretty boring; no class at all.


>If you're only storing simple data (by which I mean simple types of
>data, not that the data is simple) (and I think you must be for the
>approach you're using to work) give the marshal module a whirl.

This is the first I've heard of marshal.  I take it from various 
hints here and there that the .pyc files are handled internally 
by marshal, yes?

Thanks for your (and everyone else's) help.

One final asside, I've seen disagreements both online and in the 
web docs about the spelling of marshal.  I did a search and found 
12 hits for "marshall" (now granted, some are probably unrelated, I 
didn't trace them all).  But there were enough to make me think 
that 'marshall' was the right spelling.

But the fact that 'import marshal' worked while 'import marshall' 
failed convinced me otherwise.  :)

marshaling-up-the-courage-to-convert-working-code-to-marshal'ly yrs

... (hoping-this-'ly yrs-tag-isn't-trademarked-or-only-available-
to-those-who've-achieved-some-level-of-pythondom'ly yrs)

Michael



From 8499 at greenhead.ac.uk  Wed Jun 30 06:55:49 1999
From: 8499 at greenhead.ac.uk (Newbie)
Date: 30 Jun 1999 05:55:49 -0500
Subject: Decent Tutorial
References: <377925EF.1CBB640F@starvision.com>  <011a01bec2d3$c215a900$f29b12c2@secret.pythonware.com> 
Message-ID: <3779e9a5@discussions>

Where can I find a decent tutorial which is 
for beginners to Python, preferably including 
some documented source code and avaliable for 
download.

Thanks.



From wware-nospam at world.std.com  Wed Jun  9 23:14:35 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Thu, 10 Jun 1999 03:14:35 GMT
Subject: "Programming Python" still worthwhile?
References: <1283218560-70098357@hypernet.com> <375E9943.F9FEA7BA@gssec.bt.co.uk> <375f23fe.1133449@nntp.ix.netcom.com>
Message-ID: <FD3D0C.G8v@world.std.com>

: > At work we have Python 1.3 (don't ask!) and at home I use 

Where I work, the publicly available version is 1.2. I maintain a
private copy of 1.5.2 for my own use. As far as I'm aware, nobody
else in the company is using Python at all.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Will Ware		email: wware[at]world[dot]std[dot]com
PGP fp (new key 07/15/97) 67683AE2 173FE781 A0D99636 0EAE6117



From rdev at my-deja.com  Wed Jun 30 13:59:45 1999
From: rdev at my-deja.com (rdev at my-deja.com)
Date: Wed, 30 Jun 1999 18:59:45 +0100
Subject: Reflection in Python?
Message-ID: <000015c6@bossar.com.pl>

From: rdev at my-deja.com

In article <7ldglr$ita$1 at nnrp1.deja.com>,
  anders777 at my-deja.com wrote:
>  ...How do I do the
> equivalent of what Java calls reflection? I want
> to do the following:
>
> test = MathTest("testAdd")
>
> where test.run() will run the "testAdd" method
on
> the MathTest object?  I also want to be able to
> tell the user the name of the object (MathTest)
> and the method in case the method fails.  Any
> thoughts?
>
If I understand what you are trying to do, it's
easy in Python.  Assuming MathTest is an
object instance:
test = MathTest.testAdd
will return you a bound method object which you
can then call as:
test()
Getting the name of the object class and method
is also easy.  There are several ways to do this:
1) Define a doc string within the method by
   putting a quoted string as the first line
   under the def line.  You can then refer to it
   as "test.__doc__".  This could say, for
   example, "Method bar of class foo".
2) "test.__name__" will return the name of the
   method.  "test.im_class.__name__" will
   return the name of the class that the method
   is defined within.

Example:
class MathTest:
  def testAdd(self, parm1, parm2):
      "Method testAdd of Class MathTest"
      result = parm1 + parm2
      return result
<<< mt = MathTest()
<<< test = mt.testAdd
<<< test.run(2,2)
4
<<< print test.__doc__
Method testAdd of Class MathTest
<<< print test.__name__
testAdd
<<< print test.im_class.__name__
MathTest


I hope that helps.

- Roger




> Thanks,
> Anders Schneiderman
> National Journal Daily Briefings Group
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
>



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From greg.ewing at compaq.com  Thu Jun 24 18:19:05 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 25 Jun 1999 10:19:05 +1200
Subject: Why exceptions shouldn't be used for flow control [Re: YAS to the  
 "Reading line-by-line" Problem]
References: <Pine.SUN.3.95-heb-2.07.990622183356.3454A-100000@sunset.ma.huji.ac.il> <376FDF70.4756AD38@mindspring.com> <slrn7n0chl.gh6.wtanksle@dolphin.openprojects.net> <377173F7.36BCFF6B@compaq.com> <slrn7n31k4.idj.wtanksle@dolphin.openprojects.net>
Message-ID: <3772AED9.D9E1110@compaq.com>

William Tanksley wrote:
> 
> "Not exceptional enough".  What does that mean?  (I ain't never had too
> much exception! ;-)

Sorry, I was using "exceptional" in a non-technical sense there.
I meant "not abnormal enough".

I don't agree that reaching EOF is necessarily "abnormal".
If I want to read all the data in a file, and I don't know
in advance how much there is, I write a loop that keeps reading
until EOF. In that case, reaching EOF is not abnormal at all -
it's bound to happen eventually!

On the other hand, if I have reason to believe that there
should be a certain amount of data left, and I hit EOF while
trying to read it, then that is certainly an error, and
raising an exception is a reasonable thing to do.

The question, then, is what a general-purpose library routine
should do on EOF, given that the "right thing" depends on the
circumstances. You would have it always throw an exception,
and require the programmer to catch it if it is an "expected"
exception.

Personally, I prefer it the way it is. If I want an exception
thrown, I can always write a wrapper which does so. Doing it
the other way around - wrapping exception-throwing code into
something which doesn't - is trickier to get right; as I've
shown, it's hard to make sure that it catches only what you
want to catch.

> In addition, the complexity of the failure required to produce this is
> pretty high.  How many files do you need to read to get the next line from
> this file?

It's perhaps not the best example of this kind of problem.
A better one, which I actually have experienced, is from the
days when it was common to catch KeyError as a way of telling
when a key wasn't in a dictionary:

   try:
     v = d[the_key()]
   except KeyError:
     v = some_default_value

The subtle problem with that piece of code is that if there
is a bug which causes the_key() to raise a KeyError (which I
think you'll agree is *not* an unlikely event) it gets
incorrectly caught instead of triggering a traceback.
To guard against that, you have to be very careful what
you put inside the try:

   i = the_key()
   try:
     v = d[i]
   except KeyError:
     v = some_default_value

This is one of the reasons that the get() method was
added to dictionaries. Using it, you can write

   v = d.get(the_key(), some_default_value)

which is not only less error-prone but shorter and
clearer as well.

> If someone else's uncaught EOFError hits this function, something nasty is
> almost always going to happen... if it fails, it does so loudly so that
> the tester notices and reports the problem.

But that's exactly what *doesn't* happen! The misdirected exception
silently terminates a loop that it wasn't meant to terminate, and
later on the program fails with some set of symptoms that give little
or no clue as to what the original cause was. I would much rather get 
a traceback pinpointing exactly what was thrown and where it was thrown 
from.

> C is famous for noiselessly casting the -1 to a char (look at
> the old Bash security bug), and Python is even worse -- any return you can
> possibly make will almost certainly fit in noiselessly with anything you
> do with it.

The design of read() is perhaps not the best here -- it might be
better to return None, which would cause most things expecting a
string to blow up rather more quickly.

> You ought to feel lucky.  Some respected programming gurus recommend
> calling abort() on any odd results

Raising an exception is Python's equivalent of calling abort(). But
it only works as such if you can be sure that your exception isn't
going to be caught by something that makes unwarranted assumptions
about its cause. The more you use exceptions for "normal" things,
such as EOF or KeyError or IndexError when they aren't really errors, 
the more likely that is to happen. In my experience, anyway.

Greg



From scott at chronis.pobox.com  Mon Jun 21 17:22:37 1999
From: scott at chronis.pobox.com (scott cotton)
Date: Mon, 21 Jun 1999 17:22:37 -0400
Subject: Check for valid date
In-Reply-To: <7klq51$etp$1@nnrp1.deja.com>; from Remi Delon on Mon, Jun 21, 1999 at 04:44:20PM +0000
References: <7klq51$etp$1@nnrp1.deja.com>
Message-ID: <19990621172237.A4612@chronis.pobox.com>

you could try making sure that 

strftime(your_format, strptime(date2check, your_format)) == date2check.

and checking for OverflowError & ValueError in the process.

scott



On Mon, Jun 21, 1999 at 04:44:20PM +0000, Remi Delon wrote:
| Hi, I'm trying to use the time module to check whether a date is
| actually a valid day or not.
| Unfortunately, it looks like the strptime function only performs a basic
| range check. (for example, it will accept 02/30/199)
| Here is a little test I did:
| 
| >>> strptime('02/30/1999','%m/%d/%Y)
| (1999, 2, 30, 0, 0, 0, 61, 0)
| >>> mktime((1999, 2, 30, 0, 0, 0, 61, 0))
| 920361600.0
| >>> ctime(920361600.0)
| 'Tue Mar  2 00:00:00 1999'
| 
| Is there a way to check test whether the day actually exists or not ?
| 
| 
| I also have a totally different problem : I'm trying to get python 1.5.2
| to run with Tcl/Tk 8.1. Has anyone been able to do that on AIX ?
| I found 2 patches for tkinter that were supposed to fix the problems,
| but unfortunately, they don't work for AIX.
| 
| Thanks
| 
| Remi.
| 
| 
| Sent via Deja.com http://www.deja.com/
| Share what you know. Learn what you don't.
| 
| -- 
| http://www.python.org/mailman/listinfo/python-list



From mmorris at mindspring.com  Wed Jun 23 19:52:11 1999
From: mmorris at mindspring.com (Mitchell Morris)
Date: Wed, 23 Jun 1999 23:52:11 GMT
Subject: It's just magic, and I hate it ...
References: <slrn7mss5c.f90.mgm@unpkhswm04.bscc.bls.com> <14190.35530.772547.166720@buffalo.fnal.gov> <slrn7mvlhr.4is.mgm@unpkhswm04.bscc.bls.com> <37703621.B29B72CB@bioreason.com> <slrn7n1r6p.keg.mgm@unpkhswm04.bscc.bls.com> <slrn7n2aj0.n0c.mgm@unpkhswm04.bscc.bls.com>
Message-ID: <930181790.156.59@news.remarQ.com>

In article <slrn7n2aj0.n0c.mgm at unpkhswm04.bscc.bls.com>, 
mmorris at mindspring.com wrote:
[snip]

In a very polite private e-mail, Tim Peters pointed out that this is a FAQ, 
number 4.87 in fact. That will teach me to not check *ALL* the docs before 
opening my trap.

Thanks to everyone ... I appreciate your time and efforts on my behalf,
+Mitchell



From srn at fn.com.au  Wed Jun  9 02:23:45 1999
From: srn at fn.com.au (Stephen Robert Norris)
Date: 9 Jun 1999 06:23:45 GMT
Subject: tstate invalid crash with threads
References: <000101beb237$88b71d80$999e2299@tim>
Message-ID: <7jl19h$ofs$1@crux.cs.usyd.edu.au>

In article <000101beb237$88b71d80$999e2299 at tim>,
	"Tim Peters" <tim_one at email.msn.com> intoned:
> Followups to the Thread-SIG, please.
> 
> [Ray Loyzaga]
>> I have been playing with a multithreaded tcp based server for a while,
>> and during some serious stress testing continually hit a problem
>> involving the interpreter crashing with "Fatal Python error:
>> PyThreadState_Delete: invalid tstate".
>> ...
>> It appears to be a subtle race in PyThreadState_Delete ....
>> interestingly, if I uncomment the small sleep in "handle" in the server,
>> ie. make the server slower, it seems to work for ever ... 4m transactions
>> before I gave up. I think the problem only comes if you are creating and
>> destroying threads quickly in parallel.
> 
> PyThreadState_Delete is called from very few places, and one of them strikes
> me as suspicious:  at the end of threadmodule.c's t_bootstrap, we have:
> 
> 	PyThreadState_Clear(tstate);
> 	PyEval_ReleaseThread(tstate);
> 	PyThreadState_Delete(tstate);
> 	PyThread_exit_thread();
> 
> The suspicious thing here is that PyEval_ReleaseThread releases the global
> interpreter lock, so nothing is serializing calls to PyThreadState_Delete
> made from the following line.  PyThreadState_Delete in turn does no locking
> of its own either, but mutates a shared list.
> 
> If this isn't plain wrong, it's certainly not plain right <wink>.  Matches
> your symptoms, too (very rare blowups during high rates of thread death).
> 
> Guido?  I haven't been able to provoke Ray's problem under Win95, but the
> above just doesn't smell right.
> 
> win95-didn't-crash-but-the-TAB-and-ESC-keys-did-swap-their-
>     meanings!-ly y'rs  - tim

I've put a lock in pystate.c around anything that touches tstate,
and it seems to have fixed the problem (although reproducing it is
time-consuming).

This seems to be the same solution that Tim suggests, but at one remove.

	Stephen
-- 
Stephen Norris  srn at fn.com.au
PGP key available via finger srn at flibble.fn.com.au.
Farrow Norris Pty. Ltd.     http://www.fn.com.au/



From thantos at brimstone.mecha  Wed Jun 23 23:25:28 1999
From: thantos at brimstone.mecha (Alexander Williams)
Date: Thu, 24 Jun 1999 03:25:28 GMT
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0rr2.2fi.behrends@allegro.cse.msu.edu> <m3aetr2yk9.fsf@atrus.jesus.cam.ac.uk>  <7ks38p$l4m$1@brick.cswv.com>
Message-ID: <slrn7f68br.2bp.thantos@brimstone.mecha>

On 23 Jun 1999 20:56:41 -0500, Neel Krishnaswami <neelk at brick.cswv.com> wrote:
>It's just that writing self.__class__ is so *ugly*! Everything else
>in Python is beautiful, too, which makes this stand out even more. 

Er, that could be because you're doing it wrong?  Try:

class Foo:
    def __init__(me):
        me.x = 9

    def setX(me, val):
        me.x = val

>>> y = Foo()
>>> y.x
9
>>> y.setX(12)
>>> y.x
12

If you /really/ want to go frobbing around with the value that all
instances of the class inherit from, instance.__class__.slot is likely
a /great/ way of pointing out to the reader you're doing something
funky (in that classes are usually thought of as immutable, despite
the reality).  The 'unsightly' nature of the construct makes it clear
you're doing something unasthetic.

-- 
Alexander Williams (thantos at gw.total-web.net)
"In the end ... Oblivion Always Wins."



From pduffin at mailserver.hursley.ibm.com  Mon Jun 28 08:51:03 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Mon, 28 Jun 1999 13:51:03 +0100
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <375f92b4.618467559@news.u-psud.fr> <376100E9.855C11C0@mojam.com>
Message-ID: <37776FB7.31D2@mailserver.hursley.ibm.com>

Skip Montanaro wrote:
> 
> Laurent POINTAL wrote:
> > Fernando Mato Mira wrote:
> > for Python vs Tcl
> > >:   If you have references to articles of the sort, that will be useful,
> > >: too.
> >
> > See http://www.cwi.nl/~sjoerd/PythonVsTcl.html
> 
> While I'm sure that what Sjoerd wrote is accurate, it described Python
> and Tcl as they existed around late 1995 or early 1996.  Both languages
> have changed more than a little since then.
> 

His descriptions of the languages were reasonably accurate when it was
written but some of his comments about them and his comparisons between
the two show a definite bias towards Python and show up his limited
use of Tcl.

> > I remeber having seen an article of a guy explaining why he switched
> > from Tcl to Python for a large project... but I loose the URL.
> 
> That would be Frank Stajano. He presented a paper at last November's
> Python conference entitled "Implementing the SMS server, or why I
> switched from Tcl to Python".
> 
> 
> http://www.foretec.com/python/workshops/1998-11/proceedings/papers/stajano/stajano.html
> 
> Frank's perspective is an interesting one, because unlike most of us, he
> used Tcl and [incr Tcl] happily for several years before encountering
> Python, so he brings a depth of understanding to the discussion most of

His article is very good and the main reason why he uses Python and not
Tcl is the large number of libraries shipped with the standard Python
distribution and the object oriented nature of Python.

> us don't have.  (I, on the other hand, got an immediate bad taste from
> Tcl and spit it right out. ;-)
> 

My son did that when he first tasted red cabbage but since then his
taste buds have matured and he know loves it ;-).


-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From arcege at shore.net  Wed Jun 23 16:48:03 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 23 Jun 1999 20:48:03 GMT
Subject: TkInter callback - how to ?
References: <376F4E57.6A23A66F@cellware.de>
Message-ID: <7Kbc3.1113$7X1.284549@news.shore.net>

bernard <bh at cellware.de> wrote:

: I have an array of buttons (a keypad) all using the same callback
: function - is there any way of finding out which button invoked
: the callback ? I suppose I am looking for a tcl/tk solution
: something like
:   -command "myCallback $thisButton"
: but tkInter only seems to provide
:   -command myCallback

: Any help welcome

for index in range(5):
  button = Button(None,
    text='%d' % index,
    command=lambda i=index: myCallback(i)
  )
  button.pack()

Generally it is easier to give it a bound method or other callable object
with the needed information.

  -Arcege




From arcege at shore.net  Wed Jun 30 13:22:07 1999
From: arcege at shore.net (arcege at shore.net)
Date: Wed, 30 Jun 1999 17:22:07 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773c3a0.102825445@news.oh.verio.com> <xPPc3.535$or1.22866@typhoon-sf.snfc21.pbi.net> <xJRc3.292$Of3.67094@news.shore.net> <00f701bec212$f6a83050$f29b12c2@secret.pythonware.com> <qmqe3.337$6M6.119017@news.shore.net> <002e01bec316$4c8bf4d0$f29b12c2@secret.pythonware.com>
Message-ID: <3nse3.362$6M6.123714@news.shore.net>

Fredrik Lundh <fredrik at pythonware.com> wrote:
: Michael P. Reilly wrote:
:> Hmm... something about that argument makes it circularly invalid.
:> Non-idle event a comes in, but must be handled after idle event d.
:> Event d can only execute when event a is taken care of.  Sounds like
:> deadlock to me. ;)

: hmm.  I thought I'd meant that idle tasks kept piling up
: when the windowing system generates a lot of events
: at once (like when you're redrawing an entire system).
: instead of redrawing everything at once, Tkinter waits
: until the window event queue is empty...

Yes, indefinate postponement is a problem with the current definition
of Tcl_DoOneEvent.  But idle tasks are for when the event loop is idle,
and indefinate postponement is the standard, accepted trade-off for
that.  Timer handlers exist for handling more time critical actions.

You had made the comment(s) earlier in the thread:
> The interface is only part of the story; many complaints
> seem to come from the fact that Tk uses idle tasks to
> redraw things.  When an "expose" event arrives from
> the windowing system, Tk calls "after_idle", rather
> than redrawing the widget at once.  And idle tasks
> are only executed when there are no other events
> to take care of...

This lead me to think that you meant windowing events (such as expose)
were handled as idle tasks and I don't see this (MotionNotify events,
yes, not all others are handled by Tcl_QueueEvent).  Hence my comment
about the deadlock.

  -Arcege




From fatjim at home.com  Tue Jun 22 17:08:19 1999
From: fatjim at home.com (Jim Meier)
Date: Tue, 22 Jun 1999 21:08:19 GMT
Subject: Win32 Question
References: <7koiib$fdu$1@nnrp1.deja.com> <376FF68A.5698AF68@home.com>
Message-ID: <376FFAB1.4BAD1D1E@home.com>

Jim Meier wrote:

> does file on my system. I'd like to see more elegant solutions, though!

er, does "FINE" on my system..


-Jim




From amaranda at nospam.com  Thu Jun  3 12:58:26 1999
From: amaranda at nospam.com (Alex Maranda)
Date: Thu, 03 Jun 1999 17:58:26 +0100
Subject: pyGTK on Windows
References: <3757D453.A302E757@interar.com.ar>
Message-ID: <3756B432.3858@nospam.com>

Nahuel wrote:
> 
> Hi, im about to do a cross-plataform GUI application in Linux and
> Windoze... i must
> choose the GUI toolkit.. my question is .. can i use pyGTK under
> Windows???
It's probably doable (as somebody's doing a port of GTK to Windoze -
check out www.gtk.org for the URL), but not worthy in my oppinion (the
port does not use Common Controls and the like).
A much better choice is wxPython http://alldunn.com/wxPython, which is
Python on top of wxWindows, a cross-platform GUI framework. You get
Windoze, Unix (Gtk and Motif), and Mac.
> 
> (I need a gui that can display a table with about 4000 lines without
> crashing (not like Tk :))
> 
there is a wxGrid widget in wxWindows.

Cheers,
Alex
-- 
Alex Maranda         mailto: amaranda at spider dot com
Spider Software Ltd. Tel: +44 (0)131 4757036
Edinburgh, UK        http://members.xoom.com/Alex_Maranda
STREAMS based communications protocols for embedded systems



From zeitlin at seth.lpthe.jussieu.fr  Sat Jun 26 17:50:24 1999
From: zeitlin at seth.lpthe.jussieu.fr (Vadim Zeitlin)
Date: 26 Jun 1999 21:50:24 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773CA67.7057F797@quasar.ipa.nw.ru> <3773D55B.8C966174@spacenet.tn.cornell.edu>
Message-ID: <slrn7naiqb.8ll.zeitlin@seth.lpthe.jussieu.fr>

On Fri, 25 Jun 1999 15:15:39 -0400, Tom Loredo <loredo at spacenet.tn.cornell.edu> wrote:
>Well, one reason not to rush to wxWindows is that there is
>no "real" Mac version.  I realize that a Mac version is "underway",
>but the wxWindows site has said that for a *long* time now.

 This is true, but someone (Stefan Csomor) has really worked on it hard during
the last few months only. We all hope that we will have a "real" wxMac version
soon now.

>A beta version was posted about a month ago, but I've not
>heard much about it, and most of the wx web site doesn't
>seem to know it exists (e.g., the FAQ still says the beta
>is forthcoming, and the dowload section appears to still
>have the old preview alpha release).

 Oops, it's just an oversight. I will ask Julian to update it a.s.a.p.

>  I develop on both Unix
>and Mac, and although the Unix part of me finds wx appealing,
>the Mac side of me feels like a 2nd class citizen in the wx world.

 The sad truth is that macs are not very common nowadays. I would like to help
develop wxMac, but I don't have one (nor at home nor anywhere near me in the
office), so I really can't. If there were more people working on wxMac it
would surely advance more quickly...

 Regards,
VZ



From alex at somewhere.round.here  Sat Jun  5 15:24:04 1999
From: alex at somewhere.round.here (Alex)
Date: 05 Jun 1999 15:24:04 -0400
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net> <aahzFCpHw8.Gxt@netcom.com> <7j4e2u$8rh$1@m2.c2.telstra-mm.net.au> <37564132.2D01B7C3@pop.vet.uu.nl> <3756BAFC.2066A636@home.com> <3756F071.E717D0FA@pop.vet.uu.nl> <37573C16.D0233810@home.com> <etdr9nrx39r.fsf@m2-225-11.mit.edu> <dkuhlmanFCtruI.Ms9@netcom.com>
Message-ID: <etdpv3awiwb.fsf@m2-225-11.mit.edu>

Thanks.

Alex.

> Look at:
> 
>     http://www.python.org/doc/essays/everybody.html
> 
>   - Dave



From matomira at iname.com  Mon Jun 14 06:21:55 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Mon, 14 Jun 1999 12:21:55 +0200
Subject: <language> broken
References: <1282792849-95706075@hypernet.com>
Message-ID: <3764D7C3.718E125E@iname.com>

Gordon McMillan wrote:
> 
> So far, you have not made one post to comp.lang.python that comes
> anywhere close to the posting guidelines.
> 
> Would you mind taking this to comp.lang.snobbery, or wherever it
> belongs?

Don't go into a flame war. I don't know ANY language that 
is not `broken' somehow, even my favorite ones or the ones
that are not my favorite but that are also `good'.



From iking at killthewabbit.org  Sun Jun  6 02:09:00 1999
From: iking at killthewabbit.org (Ian King)
Date: Sat, 5 Jun 1999 23:09:00 -0700
Subject: Sockets and lemon curry
References: <1283483353-54170129@hypernet.com>
Message-ID: <001401beafe3$12d50560$450010ac@killthewabbit.org>

From: Gordon McMillan <gmcm at hypernet.com>
To: Ian King <iking at killthewabbit.org>; <python-list at python.org>
Sent: Saturday, June 05, 1999 8:14 PM
Subject: Re: Sockets and lemon curry


> Ian King writes:
>
> > I'm trying to implement an rshd server in Python.  The problem seems
> > to be with establishing the secondary socket for stderr.  I'm using
> > a Linux 2.0 box as client, and a Windows machine as server
[snip]
>
> Yes. On *nix, it takes root authority to create sockets on ports
> below 1024. No such limit on NT. You didn't say what 'port2' was, but
> it sure fits the symptoms...
>
> - Gordon
>
This sounded right on, but as I played with it, I realized: the secondary
port (which is, yes, a reserved port) is created on the Linux box; the NT
machine is simply doing a connect() (and there are no problems as Python
under NT creates *its* socket).  Nonetheless, I played with .rhosts and
such, to create as benign (and dangerous!) an environment as possible on the
client, to no avail.  Besides, if I couldn't connect() to reserved ports,
that would pretty much preclude most Internet services!  :-)








From mlauer at asmoday-atm.rz.uni-frankfurt.de  Sun Jun  6 12:44:32 1999
From: mlauer at asmoday-atm.rz.uni-frankfurt.de (mlauer at asmoday-atm.rz.uni-frankfurt.de)
Date: 6 Jun 1999 17:44:32 +0100
Subject: [pickle] Different version of a class
Message-ID: <375a9760@nntp.server.uni-frankfurt.de>

Hi,

my application uses pickle to store and load data.

During developing cycle, many variables are added
to the classes which are pickled. Whenever I load
a file pickled with a previous version, some
variables are not existent (naturally, because
__init__ is not called). Is there an easy way
to initialize the "new" variables or must I
call __init__ - which would overwrite some of
my saved variables.

Any idea ?

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





From moshez at math.huji.ac.il  Tue Jun 22 11:37:06 1999
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 22 Jun 1999 18:37:06 +0300
Subject: YAS to the "Reading line-by-line" Problem
Message-ID: <Pine.SUN.3.95-heb-2.07.990622183356.3454A-100000@sunset.ma.huji.ac.il>

(YAS == Yet Another Solution)

This solution introduces file-like objects which do the Right Thing(TM) at
EOF: raise EOFError (instead of returning an empty string)

----------------- cut here ------------------
class File:
        '''\
        File-like objects which throw EOFError on End-of-File.

        Initialize with a file-like object.
        '''
        def __init__(self, file):
                self.__file=file

        def __getattr__(self, name):
                if name[:4]!='read':
                        return getattr(self.__file, name)
                return _Wrap(getattr(self.__file, name))

class _Wrap:
        def __init__(self, function):
                self.function=function

        def __call__(self, *args, **kw):
                ret=apply(self.function, args, kw)
                if ret=='': raise EOFError, 'no more data in file'
                return ret

def open(file, mode='r'):
        '''\
        return a file like object open to /file/ with mode /mode/,
        which throws an EOFError on EOF
        '''
        import __builtin__
        return File(__builtin__.open(file, mode))

def _test():
        import sys
        file=open('/etc/passwd')
        try:
                while 1:
                        sys.stdout.write(file.readline())
        except EOFError: pass

if __name__=='__main__':
        _test()
---------------------------- cut here ---------------------------

--
Moshe Zadka <mzadka at geocities.com>. 
#!/usr/bin/tail -1
Just another tail hacker.




From claird at Starbase.NeoSoft.COM  Sun Jun 13 09:59:16 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 13 Jun 1999 08:59:16 -0500
Subject: Broken 1.5.2 installation? Missing cw3215mt.dll
References: <000901beb3b2$baa6d980$329e2299@tim> <37611D32.47BED973@callware.com>
Message-ID: <7k0dfk$811$1@Starbase.NeoSoft.COM>

In article <37611D32.47BED973 at callware.com>,
Ivan Van Laningham  <ivanlan at callware.com> wrote:
			.
			.
			.
>[snort]
>
>I've just been installing Python/Tcl/PyWin/Win32All/blah blah on several
>Win machines (95/98/NTW/NTS) over the past several days.  It's a long
>story (soon to be a major motion picture starring Mel Gibson), but
>suffice it to say that in *every* case IDLE refused to run out of the
>box.  All that was required, though, was to update PATH to include the
>Python directory, the Tcl directory and the Tcl/bin directory.  I've
>also installed Python/Tcl on my three Linux machines--RH5.2 and
>RH6.0-SMP--but I didn't have to fiddle with anything at all in order to
>get IDLE to run just fine.  Of course, Python and Tcl etc. have been in
>my path permanently for a long time on Linux;-)
>
>> followups-to-cameron-laird<wink>-ly y'rs  - tim
>
>I'll gladly follow Cameron around if he's responsible for the install;-)
			.
			.
			.
I regard installation as comparable to sanitation services
(garbage removal) in the real world:  an enormous enabler
of comfort, theoretically inchoate, culturally underappre-
ciated--and not my specialty.  My ignorance about
installation is even more massive than the one I have about
art, for truly all I know about installation is what works
for me.

My personal belief, in fact, is that there are *no* experts
in installation.  It's turtles all the way, for everyone
I've asked seems to have a mentor whom he trusts beyond his
own experience.  Elementary graph theory questions how this
embeds in a finite population, but all I can do is report
my own observations.

I don't know off-hand who's responsible for Tk installations.
If you want to track down the guilty parties, I'll be happy
to help.  In regard to Python, of course, The Guido knows all.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From paul at prescod.net  Wed Jun 30 09:49:00 1999
From: paul at prescod.net (Paul Prescod)
Date: Wed, 30 Jun 1999 13:49:00 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net>
		<3773c3a0.102825445@news.oh.verio.com>
		<xPPc3.535$or1.22866@typhoon-sf.snfc21.pbi.net>
		<xJRc3.292$Of3.67094@news.shore.net>
		<7laovo$c4s$1@Starbase.NeoSoft.COM> <377A24B7.AD6FAA1A@cfdrc.com>
Message-ID: <377A204C.D843868D@prescod.net>

Lyle Johnson wrote:
> 
> Just to add fuel to the fire, we now have a Python interface to FOX, a
> cross-platform GUI toolkit:

I find it fascinating that there are so many GUI toolkits. The GUI is more
than twenty years right and we haven't agreed on the basic programming
model or API yet??

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for only himself
 http://itrc.uwaterloo.ca/~papresco

"The new revolutionaries believe the time has come for an aggressive 
move against our oppressors. We have established a solid beachhead 
on Friday. We now intend to fight vigorously for 'casual Thursdays.'
  -- who says America's revolutionary spirit is dead?



From faassen at pop.vet.uu.nl  Thu Jun  3 17:15:29 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Thu, 03 Jun 1999 23:15:29 +0200
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net> <aahzFCpHw8.Gxt@netcom.com> <7j4e2u$8rh$1@m2.c2.telstra-mm.net.au> <37564132.2D01B7C3@pop.vet.uu.nl> <3756BAFC.2066A636@home.com>
Message-ID: <3756F071.E717D0FA@pop.vet.uu.nl>

Jim Meier wrote:
[snip]

You're the guy who did the GGI wrapper, right? I've been playing with it
a little, now I need more docs... Perhaps I'll go write some.

> > I must say that Emacs Python mode helps tremendously in this respect,
> > though. Without it or something similar block indentation wouldn't be as
> > easy (though I've found I can code in Windows Notepad too if I really
> > want to :).
> 
> Oh, ditto. I just *love* (x)emacs now, though I would really like it to take far
> less memory and be extendable with python.
> 
> Wow, talking about old H/W makes me sound old - I'm only 19!

Well, computer tech moves quickly (I'd say 'in internet time' but it
already did before the net was popular).

I remember the days before the great internet explosion, 1991 and such,
where I could talk about the marvels of the net and get looks 'Martijn
is talking about computer stuff again'. These days I sit in the bus and
hear people talk about 'getting email'. And I'm only 25. :) I can go on
and on about the MSX-2, its VDP chip and Z80 assembler. :)

This made me come up with the Quantum-Martijn-Solipsist theory of
Reality Alteration: 'If Martijn rambles about something exciting to
people that people don't get, reality will change within a few years so
that suddenly everybody is using it.'

I'm rambling about Python these days. Yesterday I read a paper from
Guido about how he was thinking about bringing Python (or some successor
language) to *billions* of users. :)

Regards,

Martijn



From aahz at netcom.com  Fri Jun 11 13:56:25 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Fri, 11 Jun 1999 17:56:25 GMT
Subject: waste of resources ?
References: <375C0F91.1C73E4E3@icrf.icnet.uk> <87r9nl398o.fsf@psyche.evansnet> <375E86C1.73B2CF80@icrf.icnet.uk> <37606565.5A93BCE9@compaq.com>
Message-ID: <aahzFD6CI1.1qD@netcom.com>

In article <37606565.5A93BCE9 at compaq.com>,
Greg Ewing  <greg.ewing at compaq.com> wrote:
>Arne Mueller wrote:
>> 
>>                     chfds.remove(i)
>>                     os.waitpid(chpids[i], os.WNOHANG)
>>                     del chpids[i]
>
>If you're explicitly waiting for a particular child, *don't* use
>WNOHANG, because you *want* to block until that child has exited --
>which it may not have done yet, even though it has written its data.

I can't speak to Arne's situation, but I'm about to rewrite a chunk of
code to fork off a specific single child process precisely because I
*don't* want the parent to block while this code executes, but I still
want to reap the child later (and perform some final cleanup).
-- 
                      --- Aahz (@netcom.com)

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

"That doesn't stop me from wanting to work all three of them over with
the clue stick for while, with no safewords allowed."  --abostick



From delgado at Mathematik.Uni-Bielefeld.DE  Wed Jun 16 16:39:56 1999
From: delgado at Mathematik.Uni-Bielefeld.DE (Olaf Delgado)
Date: Wed, 16 Jun 1999 22:39:56 +0200
Subject: Newbie again: computing attributes on the fly
In-Reply-To: <Pine.WNT.4.04.9906161108430.310-100000@rigoletto.ski.org>
References: <Pine.WNT.4.04.9906161108430.310-100000@rigoletto.ski.org>
Message-ID: <Pine.OSF.4.05.9906162236210.19856-100000@pinax.Mathematik.Uni-Bielefeld.DE>

On Wed, 16 Jun 1999, David Ascher wrote:

> You may find that using ExtensionClass could help.  See:
> 
>     http://www.digicool.com/releases/ExtensionClass/
> 
> (note especially the section on computed attributes in the documentation).

Oops! That's the kind of advice I was fearing. Last time I tried to
understand that stuff, I really got dizzy.

Anyway, thanks! I'll have a look at it once more.

Yours,
Olaf

-- 
   ////         Olaf Delgado Friedrichs, Uni Bielefeld, Mathematik
   Olaf         Postfach 100131                  (room:  V4-109)
   `='          D-33501 Bielefeld, Germany       (phone: +49 521 106-4765)
                           http://www.mathematik.uni-bielefeld.de/~delgado




From paul at prescod.net  Tue Jun 29 12:20:25 1999
From: paul at prescod.net (Paul Prescod)
Date: Tue, 29 Jun 1999 16:20:25 GMT
Subject: PySequence_Check
References: <slrn7nhjv9.lt6.rasumner@bach.wisdom.weizmann.ac.il>
Message-ID: <3778F249.878FCCC@prescod.net>

Reuben Sumner wrote:
> 
> How can I do the equivalent of a PySequence_Check in Python?
> GvR told me that the question was illdefined (if I explained it correctly)
> and to ask here.  
>...
> What am I missing?  I have a nested structure of lists and tuple and I
> just want to know when I have bottomed out.  I can check for tuple or
> list indivually but that is ugly and inflexible.  I could just try
> indexing and see if there was an exception but that is not much better.

Hi, Reuben.

This latter is probably our best bet. Python in general has no way of
declaring that an object acts as a sequence or mapping. Therefore there
can be no way of querying whether an object acts as a sequence or
mapping.  In general, Python has no way of declaring that an object "acts
as" anything at all (other examples would include a set, a widget, a web
browser). There is no concept of interface or protocol except in the C
API. I think that the goal is to define the concept in general before
applying it specifically to sequences, mappings, file objects and other
"built-in" protocols.

My discussion with Guido on this issue began here:

<Pine.SUN.3.91.980824165825.9662F-100000 at cito.uwaterloo.ca>

or you can do a deja search with "IsMappingType"

There are some tricky issues about what this interface stuff all means in
a dynamically typed language.

The constant re-occurence of discussions like this have lead to the
development of a special interest group:

http://www.python.org/sigs/types-sig/

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for only himself
 http://itrc.uwaterloo.ca/~papresco

"The new revolutionaries believe the time has come for an aggressive 
move against our oppressors. We have established a solid beachhead 
on Friday. We now intend to fight vigorously for 'casual Thursdays.'
  -- who says America's revolutionary spirit is dead?



From tseaver at palladion.com  Wed Jun 23 23:18:00 1999
From: tseaver at palladion.com (Tres Seaver)
Date: Wed, 23 Jun 1999 22:18:00 -0500
Subject: Why exceptions shouldn't be used for flow control [Re: YAS to the  "Reading line-by-line" Problem]
References: <Pine.SUN.3.95-heb-2.07.990622183356.3454A-100000@sunset.ma.huji.ac.il> <376FDF70.4756AD38@mindspring.com> <slrn7n0chl.gh6.wtanksle@dolphin.openprojects.net> <377173F7.36BCFF6B@compaq.com> <slrn7n31k4.idj.wtanksle@dolphin.openprojects.net>
Message-ID: <3771A368.1F491DD0@palladion.com>

William Tanksley wrote:
> 
> On Thu, 24 Jun 1999 11:55:35 +1200, Greg Ewing wrote:
> >William Tanksley wrote:
> 
> >> Both should be handled by exception, not by returning a special
> >> value.
> 
> >I don't agree with that. Reaching the end of a file while
> >reading it is not exceptional enough to warrant requiring
> >the use of an exception handler to catch it.
> 
> "Not exceptional enough".  What does that mean?  (I ain't never had too
> much exception! ;-)
> 
> It's an exception to the behavior of a function defined to return the next
> entity in a file.  It's not an exception to a function which returns the
> next entity, or some reserved value otherwise.  There's no such thing as a
> partial exception; it's a quality, not a quantity.

In Design-by-Contract terms, a service may throw exceptions for two reasons:

 *  The caller violated the precondition of the contract (passing an
    invalid parameter, for instance).  Note that ALL such exceptions
    are supposed to be avoidable, given enough diligence on the caller's
    part.

 *  The service was unable to fulfill the contract due to the moral 
    equivalent of force majeure (out of disk space, etc.).  No action the
    caller could possibly take beforehand would obviate the need to catch 
    / propagate these exceptions.

The "EOF" problem in Python acts like one of the second group, because the
file-object-protocol does not supply a test for EOF (callers who can't test for
EOF, can't be blamed for reading off the end).  The tradeoff here is that the
FOP is a "bigger tent" than if it provided EOF():  many FOP-obeying objects
literally can't determine EOF without trying to do the read anyway (sockets,
FIFOs, etc).


Using exceptions for other purposes requires a more definite protocol for them
between caller and service, and thus requires more design justification.   Two
cases I see often:

Because they unwind the stack cleanly, exceptions provide an expedient way to
return "success" from deeply-recursive routines, such as graph searches.  The
justification is along the following lines:  "non-exception implemtations will
spend as much effort and code to handle returning the OOB data in line as they
do performing the search".

In Soft-n-GUI ("fewmet") programming:  given a suitable framework which catches
the exceptions at the outermost layer and displays them to the user, exceptions
drastically simplify event-handling code:  the event handlers check for nasty
bits at the top, and throw if found;  the rest of the code is unencumbered with
the kinds of ugly control flow which would otherwise be required.

> 
> >What bothers me about using exceptions for flow control
> 
> Presumably you're talking about command flow.  Okay, but exceptions do
> nothing BUT affect flow control.  What else can they do?
> 
> >in situations like this is that the effect of an exception
> >handler is *non-local*, whereas what you're trying to
> >catch is really only a local concern.
> 
> Is it?  Are they?  The effects of an exception toss are as local as the
> user of the function want them to be.
> 
> If EOF was truly local, then file.read would always be able to decide what
> to do with it -- print a message, abort(), pop up a window, or whatever
> else it takes.

Here's the rub:  if read() throws on EOF, it promulgates a policy that EOF is a
"back out" condition:  everyone must wrap almost any call to read() in a try
block, even if only to ignore it, or break out of a local loop.  Because read()
doesn't throw (and thereby doesn't enforce policy at the mechanism level),
applications which don't need the policy don't pay for it;  yet it is trivial
(as the original "YAS" poster shows) to layer the policy on top of the
mechanism, if so desired.

Use-exceptions-daily-in-GUI-code-_except_-when-threads-are-involved'ly

Tres.
-- 
=========================================================
Tres Seaver         tseaver at palladion.com    713-523-6582
Palladion Software  http://www.palladion.com



From arcege at shore.net  Tue Jun 22 16:16:11 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 22 Jun 1999 20:16:11 GMT
Subject: Creating a new type with an extension module
References: <Ctp*vl52n@news.chiark.greenend.org.uk> <rvy*CQ92n@news.chiark.greenend.org.uk> <QpMb3.908$7X1.231730@news.shore.net>
Message-ID: <faSb3.968$7X1.247240@news.shore.net>

Michael P. Reilly <arcege at shore.net> wrote:

: But the documentation on extending Python is pretty well known
: deficiency.  Luckily it is far easier than most languages, both to
: accomplish and to understand (yet more kudos to Guido).  Actually, in
: responce to your earlier posting, I started writing a _SHORT_ webpage
: explaining what is needed, which I hope to have available later today.
: I plan (and hope) to make it into a How-To later.

I've put a copy of my short, quick tutorial on how to write an
extension.  It is very terse, but has all the basics, including a full
example, hints and some minor tricks.

  http://starship.python.net/crew/arcege/extwriting/pyext.html

Let me know how well it reads (I doubt it reads well since it was
started this morning).

  -Arcege




From redsand at dsuper.net  Sat Jun 26 22:14:01 1999
From: redsand at dsuper.net (Dave)
Date: Sat, 26 Jun 1999 22:14:01 -0400
Subject: Advice for Newbie
Message-ID: <377588E9.82E8F567@dsuper.net>

I read the Python homepage. I sounds promising. What books do you
commend for a beginner?




From evan at tokenexchange.com  Mon Jun 21 15:15:51 1999
From: evan at tokenexchange.com (Evan Simpson)
Date: Mon, 21 Jun 1999 14:15:51 -0500
Subject: choosing random numbers with weights/probability?
References: <929987267.6227@www.remarq.com>
Message-ID: <lbwb3.3$V4.11640@news-dal.corridex.com>

In this simple case, you could use:

weighted = [0, 1, 2, 2]
item = list[whrandom.choice(weighted)]

If you really need arbitrary weights, try:

list = [('one', 0.25), ('two', 0.25), ('three', 0.5)]
def weighted_choice(list):
  from whrandom import uniform
  n = uniform(0, 1)
  for item, weight in list:
    if n < weight:
      break
    n = n - weight
  return item

kevinsl wrote in message <929987267.6227 at www.remarq.com>...
I've been using the whrandom.choice routine and it's very
useful. But is there any way to add weights or
probabilities of being chosen to the items in the list?

example:

list=['one','two','three']
item=whrandom.choice(list)

Is there any way to say that 'one' and 'two' have a 25%
chance of being chosen, and 'three' has a 50% chance?






From graham at sloth.math.uga.edu  Thu Jun  3 14:58:21 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 3 Jun 1999 18:58:21 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <7j450j$908$2@cronkite.cc.uga.edu> <7j5sr7$eb4$1@nnrp1.deja.com>
Message-ID: <7j6j8d$58f$2@cronkite.cc.uga.edu>

adjih at technologist.com wrote:
:   But there will be cases where full garbage collection will be
: definitly a nuisance.
:   First, you must scan all the address space of the C code, which can be
: a problem if you're just using the Python interpreter in a library (i.e.
: don't have the ability to control or modifying the main program) ; or if
: your address space is too big (I use python for instance in a program
: that does an mmap of a 350 MB file).

a) you can turn gc off

b) I routinely use garbage collection on 1 Gig memory spaces with no
	 speed problems.

adjih at technologist.com wrote:
:   Second you kill the portability: you could use a subset of Python with
: ANSI-C alone ; but you can't make a incremental GC in ANSI-C.

Then don't use an incremental collector! Use a mark sweep collector.
I use such a thing almost every day -- written entirely in ANSI-C.

:   You'd get problems when embedding Python in unusual environments ; for
: instance if you embed Python in the Linux kernel, then the GC should
: also trace all the pages in the swap, or be rewritten specially for that
: case.

Why do you need to do any of these things?

adjih at technologist.com wrote:
:   The advantages of GC must be traded with its drawbacks:
: - advantages: reclaims cyclic structures.
: - drawbacks: non-portable, requires a special handling when linking with
: C, high complexity for decent GC (multi-thread incremental with
: finalization and support for C/C++ structures referencing Python
: structures).

The drawbacks you list are all misleading. Mark sweep collection (with
refcounts) is highly portable. Non compacting (or even partially
compacting) collectors require no special code to link with C.
Complexity of a mark sweep collector is not very high at all (registering
roots and traversing foreign objects being the main areas of
complexity). I am sure that for *some* collection schemes some of
the drawbacks you list are true, but not for all.

graham

-- 
          The fool escaped from paradise
        will look over his shoulder and cry
          Sit and chew on daffodils and
             struggle to answer why?



From proteus at cloud9.net  Tue Jun 29 11:38:31 1999
From: proteus at cloud9.net (Michael Muller)
Date: Tue, 29 Jun 1999 11:38:31 -0400
Subject: ssh with telnetlib
References: <37778244.C4528C5B@hrz.uni-paderborn.de>
Message-ID: <3778E876.6E4B86A5@cloud9.net>

m liss wrote:
> 
> Hello,
> 
> has someone already made telnetlib.py support ssh ?
> I want to use Python for remote administration.
> 
> regards,
> 
> Mirko
> 
> --
> M Li?, <n89553 at hrz.uni-paderborn.de>

I wouldn't expect to see this any time soon: telnet and ssh are very
two different protocols.

Your best options are:

1) use telnetlib to tunnel through an established ssh connection
2) write a module that wraps the ssh client, communicating to it through
   popen2 or popen3

Option 1 will require a telnet daemon to be available on the remote
host.

If you're interested, and can wait, I'm working on a python based 
distributed object system which will (eventually) support ssh tunneling.

=============================================================================
michaelMuller = proteus at cloud9.net | http://www.cloud9.net/~proteus
-----------------------------------------------------------------------------
Mantra for the 60's: Tune in, turn on, drop out
Mantra for the 90's: Turn on, jack in, jerk off
=============================================================================



From news at dorb.com  Tue Jun 29 10:11:07 1999
From: news at dorb.com (Darrell)
Date: Tue, 29 Jun 1999 10:11:07 -0400
Subject: Uploading files
References: <377874C6.6941CCFF@boxer.no.--->
Message-ID: <hw4e3.3640$U6.15775@newsr2.twcny.rr.com>

Jostein S?iland <jso at boxer.no.---> wrote in message
news:377874C6.6941CCFF at boxer.no.---...
> Hi, I'm writing a script that needs to upload files from the user's
> machine to a server, and it's used in combination with a form and
> HTML-file that lies on the server side. I wrote a piece of code that I
> thought would work, but it turned out that when i specify c:\foo.exe, it
> is the file on the server side that is uploaded and not the foo.exe on
> the client side. Evidently since the script is run on the server side.
>
> So my question is: How do I specify that it is the client's c:\foo.exe
> and not the server file that gets uploaded?
>
>
> Thanks
>
> Jostein S?iland
>

html>
<form name="Attachments" ENCTYPE="multipart/form-data" method="post"
action="http://www.xxx/cgibin/run.cgi/upload" >
<input type=file name=photo size=20 accept="image/*">
<input type=submit>
</form>
</html>


I'm using bobo to read the response to the post that contains the file.
This is a cut up snippet of how I did that, you just need something to
accept the data and write it to disk.
My skill level in these matters is about a 1, so I'm sure you could do
better.

def upload(REQUEST, RESPONSE):
        """
        Get a new file from a user
        """
        name= REQUEST.form[fieldName].filename
        if name=='':
                return
        names= string.split(name,'\\')
        names = string.split(names[-1],'/')
        name=names[-1]
        name = string.split(name)
        name = string.join(name,'')

        global baseDir
        fp = open(baseDir+name,'wb')
        buf = REQUEST.form[fieldName].read()
        fp.write(buf)
        fp.close()


--
--Darrell





From jim at interet.com  Fri Jun 25 10:57:50 1999
From: jim at interet.com (James C. Ahlstrom)
Date: Fri, 25 Jun 1999 10:57:50 -0400
Subject: Environment inheritance under Windows NT 4, Attempt 2
References: <Pine.SGI.3.95q.990623204139.149495A-100000@masox202>
Message-ID: <377398EE.C224497B@interet.com>

"Milton L. Hankins" wrote:
> 
> Since I've not yet gotten a response to my first post, here's an
> elaboration.
> 
> Windows NT Workstation 4.00.1381, SP5.
> C:\>echo %WINDIR%
> C:\WINNT
> 
> C:\>python -c "import os; print os.environ['windir']"
> C:\WINNT
> 
> C:\>set WINDIR=foo
> 
> C:\>echo %WINDIR%
> foo
> 
> C:\>python -c "import os; print os.environ['windir']"
> C:\WINNT
> 
> -----------------------------------------------------

I don't get this behavior on NT 4.00.1381.  This was copied from
cmd.exe:

N:\python\Python-1.5.2\PCbuild>ver

Windows NT Version 4.0

N:\python\Python-1.5.2\PCbuild>./python -c "import os; print
os.environ['WINDIR']"
C:\WINNT35

N:\python\Python-1.5.2\PCbuild>set WINDIR=foo

N:\python\Python-1.5.2\PCbuild>echo %WINDIR%
foo

N:\python\Python-1.5.2\PCbuild>./python -c "import os; print
os.environ['WINDIR']"
foo

N:\python\Python-1.5.2\PCbuild>

I think I am at service pack 4.  I am using Python 1.5.2.  What is your
Python version (not that it should matter).  Very peculiar.

The code seems to be getting the environment from char **environ.  The
VC 6.0 docs say to use _environ, and I don't see a reference to
_environ.
The docs also say getenv() uses _environ.

Although I can't duplicate the behavior you describe, it seems to
me Python should be using _environ instead of environ.  Does anyone
know what environ is on NT?  I see "#define environ _environ" in
stdlib.h, but it is inside a bunch of #ifdef's.

Jim Ahlstrom



From Peter.Haight at p98.f112.n480.z2.fidonet.org  Wed Jun 30 22:54:27 1999
From: Peter.Haight at p98.f112.n480.z2.fidonet.org (Peter Haight)
Date: Thu, 01 Jul 1999 03:54:27 +0100
Subject: Why is no exception raised for file.read() when file is only writeable?
Message-ID: <00001623@bossar.com.pl>

From: Peter Haight <peterh at sapros.com>


I've been helping someone learn the language and he was having trouble
understanding why this didn't work:

test=open('testfile','w')
testread=test.readline()

I explained to him the problem, but I was wondering why Python doesn't raise
an exception when you do this. He probably wouldn't have had to ask me
anything if Python printed something like this:

IOError: [Errno X] You can't read from a file only open for writing.






From bernhard at alpha1.csd.uwm.edu  Thu Jun 17 17:33:04 1999
From: bernhard at alpha1.csd.uwm.edu (Bernhard Reiter)
Date: 17 Jun 1999 21:33:04 GMT
Subject: BiBTeX module?
References: <joe-160619991358444033@chinacat.salk.edu> <7k950a$hsb@cs.vu.nl>
Message-ID: <slrn7miqcg.tpi.bernhard@alpha1.csd.uwm.edu>

On Wed, 16 Jun 1999 21:29:46 GMT, Frank Niessink <frankn=news at cs.vu.nl> wrote:
>Joe Strout <joe at strout.net> wrote:
>> Is there a BiBTeX module for Python?  

Having a look at the current programs available for editing and entering
BiBTeX files someone really ahs to write a good python/tk or
python/wxpython program for this task, too.

(sidebar: tkbibtex is not really good, and bibview-2.2, the old X11
program still seems to be the best, but hey old fashoined.)


>A collegue and myself hacked up a package (called Bridge) which enables
>you to generate BibTeX from SGML/XML (so it includes both a BibTeX SGML dtd
>and a BibTeX XML dtd) and vice versa 


>The package itself is in a state of flux right now, but if you're
>interested I guess we can make it available nevertheless.
Make it available as free software, it sounds interesintg.

thanks, 
	Bernhard

-- 
Research Assistant, Geog Dept UM-Milwaukee, USA.         (www.uwm.edu/~bernhard)
Funding/Promoting Association for a Free Informational Infrastructure (ffii.org)



From tismer at appliedbiometrics.com  Thu Jun  3 14:01:26 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Thu, 3 Jun 1999 18:01:26 GMT
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl>
		<375536C2.DD917F99@prescod.net>
		<aahzFCpHw8.Gxt@netcom.com>
		<7j4e2u$8rh$1@m2.c2.telstra-mm.net.au>
		<37564132.2D01B7C3@pop.vet.uu.nl> <3756BAFC.2066A636@home.com>
Message-ID: <3756C2F6.9D7E6EFD@appliedbiometrics.com>

...

> Ditto about not being afraid of the indentation. I've been rabidly happy about
> using indentation properly ever since I moved up from a commodore vic-20: in the
> vic's BASIC, you typed a line number and then just typed until you had trouble
> reading the line - it would just wrap around the 20-character wide screen. Moving
> up to a DOS machine with MS QBasic, I was very releived to be allowed to indent my
> programs properly. But QBasic has crazy block open/closures: each type of
> statement is different!
...
> Wow, talking about old H/W makes me sound old - I'm only 19!

Are you sure? How did you get your hands at a VIC-20, then? :-))

ciao - chris     (do you want my AIM65 board with an 8 kb Forth?)

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From Guido.van.Rossum at p98.f112.n480.z2.fidonet.org  Tue Jun 29 04:08:06 1999
From: Guido.van.Rossum at p98.f112.n480.z2.fidonet.org (Guido van Rossum)
Date: Tue, 29 Jun 1999 09:08:06 +0100
Subject: total idiot question: +=, .=, etc...
Message-ID: <000015a4@bossar.com.pl>

From: Guido van Rossum <guido at cnri.reston.va.us>

> >>>     class A(B):
> >>>         def __init__(self, x, y, z):
> >>>             B.__init__(self, x, y, z)
> 
> [Greg Ewing]
> > I could live with having to explicitly name the
> > superclass, provided something checked that the
> > class I named was actually a direct base class
> > of the one where the method is defined. That
> > way things would be less likely to break
> > mysteriously when I rearrange the class
> > hierarchy.
> 
> [Guido]
> > Oh, but that check *is* being made!  (In fact this is the same check
> > that prevents "class methods" from working ;-)
> 
> Nope, that's a different check.  If C derives from B derives from A, this
> check allows either C.__init__ or B.__init__ to be called from A.__init__
> with an instance of A.  Greg wants  B.__init__ to be OK here, but not
> C.__init__ (C is not a direct base class of the one (A) where the method
> (A.__init__) is defined).  IOW, he wants a way to restrict __init__ calls to
> immediate bases of the class from which the __init__ is invoked, to e.g.
> prevent level-skipping under class rearrangement.

Sigh.  Indeed.

> I posted some ugly code to address that, but think it's really a job for a
> text editor and a pair of eyeballs <wink>.

Hm -- he may be right!  I was reading some IDLE code the other day and
noticed that PyShell's close() method was calling
PyShellEditorWindow.close(self) while it didn't inherit from that
class at all...  Turns out the actual method was defined in a common
base class -- a silent (and in this case harmless) bug, caused by,
indeed, a class restructuring.  So there is a point to his wish.  On
the other hand Python explicitly allows you to ignore base class
methods in overriding ones, and even constructors...  (And then again,
there is something to say for requiring constructors to play be
slightly different rules -- certainly they don't have to duplicate the
signature of the base class constructor like regular methods need to
do.)

Enough,

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





From kjwolf at online.de  Sun Jun 20 21:01:09 1999
From: kjwolf at online.de (Klaus-Juergen Wolf)
Date: Mon, 21 Jun 1999 03:01:09 +0200
Subject: Problem with Curses
References: <ppgmj7.s7r.ln@solling.seismic.de> <jpeck7.1661.ln@solling.seismic.de> <ve1dk7.1e02.ln@solling.seismic.de> <87ogidzs37.fsf@home.ivm.de> <6Zeb3.526$4e1.32195@iad-read.news.verio.net>
Message-ID: <ls2kk7.sl.ln@solling.seismic.de>

T.E.Dickey <dickey at shell.clark.net> wrote:
 >>> It depends on when the executable is linked to libmytinfo, which only
 >>> takes place on platforms that don't know the terminfo database, like
 >>> BSD and Solaris; there are two competing declarations of tputs().

 >> Why doesn't Freebsd know the terminfo database? Can't the one that is 
 >> integrated into ncurses be used on FreeBSD? 

 > I've been told ncurses 4.2 is incorporated in the FreeBSD 3.1 - but that
 > may still just be a port...

Yes, there is such a port. Do you think, one could release code that
works on several machines, while some of them must install a special
program first?

cu
 k.j.

PS: It wouldn't have worked anyway.

-- 
 Support Julia Hill - save the Redwood trees: http://www.lunatree.org/



From ivnowa at hvision.nl  Wed Jun  2 16:29:48 1999
From: ivnowa at hvision.nl (Hans Nowak)
Date: Wed, 2 Jun 1999 20:29:48 GMT
Subject: Alien whitespace eating nanovirus strikes again!
In-Reply-To: <375500C9.742DC539@pop.vet.uu.nl>
References: <375500C9.742DC539@pop.vet.uu.nl>
Message-ID: <199906022028.WAA28874@axil.hvision.nl>

On 2 Jun 99, Martijn Faassen wrote:

> [Larry Wall:]
> [Perl is postmodern]
> > That is, if something's worth doing, it's worth driving into the ground to the
> > exclusion of all other approaches. Look at the use of parentheses in Lisp or 
> > the use of white space as syntax in Python. Or the mandatory use of objects in
> > many languages, including Java. All of these are ways of taking freedom away
> > from the end user ``for their own good''. They're just versions of Orwell's
> > Newspeak, in which it's impossible to think bad thoughts. We escaped from the
> > fashion police in the 1970s, but many programmers are still slaves of the cyber
> > police.
> [Perl is postmodern and therefore allows freedom]
> 
> One could say the same about Perl and it requiring it curly braces
> around blocks? In fact, I heard the claim that its requirement to use
> curly braces around single line blocks (unlike C which allows you to
> remove them) makes Perl less prone to errors. And Perl excludes the use
> of meaningful whitespace indentation! (python does allow #{ and #}
> blocks ;)

When I read his articles about postmodern Perl, I initially thought 
he was joking... this shtick about 'languages with one big idea... 
Java has objects, Lisp has parentheses, Python has whitespace', etc. 
couldn't be serious, right? But it seems he is serious after all...

Sometimes I dare read a few messages in the Perl newsgroups, and some 
people there seem to think of Python as the summum of inflexibility. 
It doesn't allow you enough freedom to code what you want. Just look 
at that whitespace! -- Nonsense, of course. The fact that Python 
heavily uses whitespace doesn't mean you cannot code what you want, 
and in a way you want it. The same thing can be said for Scheme and 
Lisp. They use lots of parentheses, and while the languages look 
peculiar because of that, it doesn't make them any less flexible. On 
the contrary.

Whitespace in Python enforces readability, it does not force you to 
use a certain coding style. I really cannot understand why some 
people seem to think Python is not flexible. It actually allows very 
much, just takes out brain-damaged things, like for instance adding 
strings and integers, which makes as much sense as adding file 
objects and complex numbers. :o)

> Okay, I *will* say the same about Perl and its curly braces requirement:
> 
> } That is, if something's worth doing, it's worth driving into the
> ground to the
> } exclusion of all other approaches. Look at the use of curly braces in
> } Perl.

$Yes, (%(and) quite) @a { $few %other $annoying {characters...} }

> Okay, sorry for all this, it's out of line, I just needed to blow off
> steam. :)

Me too. :)

Veel liefs,

--Hans Nowak (ivnowa at hvision.nl)
Homepage: http://fly.to/zephyrfalcon



From movits at lockstar.com  Tue Jun 29 10:35:42 1999
From: movits at lockstar.com (Mordy Ovits)
Date: 29 Jun 1999 07:35:42 PDT
Subject: [Python|Crypto] on handhelds (was Re: Thanks for Python...)
References: <376ead64@anonymous.newsfeeds.com> <VrDQRddcbbs9-pn2-QdftOXYk6Hdd@Fortress> <37716901.43CB604@lockstar.com> <slrn7n2gkr.i3k.wtanksle@dolphin.openprojects.net> <37727194.D9CD6B7@lockstar.com> <slrn7n5j99.k00.wtanksle@dolphin.openprojects.net>
Message-ID: <37790460.47FAD8AF@lockstar.com>

William Tanksley wrote:
> I have a certain amount of leeriness about posting advocacy on unrelated
> NGs, so I'll limit myself to saying that the difference is years, not
> miles.  The two machines work differently: one is a glorified organizer,
> the other is a pocket computer.  Both of them can fool you into thinking
> they're something else, because they're both powerful.
> 
> I need a personal organizer.

To each his own. :-)  Don't you wish all flame were required to end like that
(i.e. before they started?)
 
> >> I used Quartus on my Crypto exam to solve a modular equation after I
> >> forgot how to solve modular equations (I had it use brute force).
> 
> >Wow!  I wish I had had my E100 when I was in college.  I just realized how
> >awesome that would have been. :-)
> 
> Oh yeah.
> 
> Actually, I sometimes miss my old OmniGo, which was a pocket computer, in
> spite of being slower than the Pilot for some tasks.  I really needed the
> glorified organizer (and had been fooled into thinking that the OmniGo was
> one), but the OmniGo had some abilties that the Pilot doesn't.

I'm not familiar with the OmniGo.  Got a link for me?
 
> >The SECOND time I saw that sig was in ESR's mail.  The first was mine.  I doubt
> >ESR copied mine, but who knows? :-)
> 
> You mean you wrote it?  If so, there seems to have crept into it a bug --
> one of the implicit unpackings don't work.

No, I didn't write it.  It was written by the all-talented AMK.  I did write a
nifty version of Bruuce Schneier's Solitaire algorithm, though.  check out:

http://www.counterpane.com/solitaire.html

Mordy

-- 
o Mordy Ovits
o Cryptographic Engineer
o LockStar Inc.
---------------------------------------------------------------------------
#!/usr/local/bin/python 
from sys import*;from string import*;a=argv;[s,p,q]=filter(lambda x:x[:1]!=
'-',a);d='-d'in a;e,n=atol(p,16),atol(q,16);l=(len(q)+1)/2;o,inb=l-d,l-1+d
while s:s=stdin.read(inb);s and map(stdout.write,map(lambda i,b=pow(reduce(
lambda x,y:(x<<8L)+y,map(ord,s)),e,n):chr(b>>8*i&255),range(o-1,-1,-1)))



From tgm at math.uni-bonn.de  Mon Jun 28 04:58:14 1999
From: tgm at math.uni-bonn.de (Thomas Ackermann)
Date: Mon, 28 Jun 1999 10:58:14 +0200
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <37766E1A.49FC9062@emsoft.de>
Message-ID: <6fd7l7.9d.ln@final.math.uni-bonn.de>

Could you possible resent this as an ordinary (text) mail, please?

Michael S. Schliephake <mss at emsoft.de> wrote:
> <!doctype html public "-//w3c//dtd html 4.0 transitional//en">
> <html>
> Hallo -
> <blockquote TYPE=CITE>As a matter of fact, it appears to be very easy to
> make commercial grade GUIs using wxPython.</blockquote>
> Commercial grade GUI's ?
> <p>Think about some points:
> <ol>
> <li>
> Tcl/Tk is freeware but has a strong commercial support. Scriptics and a
> growing number of companies support with commercial tools or provide software
> based on Tcl/Tk.<br>
> This could make sure a long-time existence of this language, improvements
> will be made possibly quickly. Even very large companies use it for in-house
> developments,</li>

> <li>
> several books exists about programming,</li>

> <li>
> Performance for Tcl/Tk under Python to improve seems not to bee too complicated,</li>

> <li>
> it is very easily to program, just in the sense of a scripting language.
> With some self-written utilities and classes one can realize the GUI even
> much more faster,</li>

> <li>
> the actual Tkinter-interface allows it to incorporate Tcl/Tk-extensions
> for test purposes without writing new extension code (just new wrapper
> classes or so, implemented in Python, are necessary). It's o.k to write
> extension code later, if the Tcl extension will be used permanently or
> never if there is enough performance.</li>

> <li>
> There exist several GUI-Builders.</li>
> </ol>
> More pointes necessary? It's a world of freeware. So I do not see any reason
> to define the one and only GUI. One should use the best fitting tool. But
> in my point of view, Tcl/Tk seems to be - at least actually - the most
> wise decision.
> <p>Michael Schliephake.</html>


-- 
  Thomas Ackermann | Tel. +49-(0)228/631369|73-7773 | <tgm at math.uni-bonn.de>
             finger tgm at rhein.math.uni-bonn.de for public key
               GNU LINUX Python gtk pygtk MySQL FUDGE GURPS



From fatjim at home.com  Wed Jun 16 13:24:12 1999
From: fatjim at home.com (Jim Meier)
Date: Wed, 16 Jun 1999 17:24:12 GMT
Subject: Fnorb Q..
Message-ID: <3767DD6F.595A04A5@home.com>

I'd post this to the Fnorb list, but don't seem to have been added yet..
I appologise if this offends anyone's sense of relevance.

In installing Fnorb on a Debian 2.1 linux system, I follow the
instructions very carefully. I'm actually not using debian's Python
package, but a self-installed/compile 1.5.2 Python, so I can fiddle with
the source without needing to learn about packages.

Fnorb seems to install fine, but when I try running fnidl on anything,
it seems to chew for a while (disk activity, slight increase on load avg
meter), but then waits an ridiculous amount of time (2-3 minutes), then
spits out the error message at the bottom of this page.

Am I doing something wrong? The command line I'm using is simple "fnidl
(idl file)", which works fine under windows.

Any help would be appreciated.. I really want to play with CORBA!! :)

-- Traceback (innermost last):
  File "/home/fatjim/projects/balderdash/Fnorb/script/fnidl", line 2, in
?
    import sys; from Fnorb.script import fnidl;
sys.exit(fnidl.main(sys.argv))
  File "/home/fatjim/projects/balderdash/Fnorb/script/fnidl.py", line
102, in ma
in
    result = main_batch(context, cpp_flags, idl_files)
  File "/home/fatjim/projects/balderdash/Fnorb/script/fnidl.py", line
147, in ma
in_batch
    ifr = IntRepImpl.RepositoryImpl()
  File
"/home/fatjim/projects/balderdash/Fnorb/cos/interface_repository/IntRepIm
pl.py", line 581, in __init__
    self.get_primitive(CORBA.pk_boolean),
  File
"/home/fatjim/projects/balderdash/Fnorb/cos/interface_repository/IntRepIm
pl.py", line 654, in get_primitive
    self._fnorb_register_impl(key, PrimitiveDefImpl._FNORB_ID,
definition)
  File
"/home/fatjim/projects/balderdash/Fnorb/cos/interface_repository/IntRepIm
pl.py", line 77, in _fnorb_register_impl
    boa = BOA.BOA_init()
  File "/home/fatjim/projects/balderdash/Fnorb/orb/BOA.py", line 59, in
BOA_init
    boa = BOA(argv, boa_id)
  File "/home/fatjim/projects/balderdash/Fnorb/orb/BOA.py", line 149, in
__init_
_
    protocol.enable(host, port)
  File "/home/fatjim/projects/balderdash/Fnorb/orb/IIOPProtocol.py",
line 61, in
 enable
    self.__nudger = Nudger.Nudger()
  File "/home/fatjim/projects/balderdash/Fnorb/orb/Nudger.py", line 76,
in __ini
t__
    self.__client.connect(listen_host, listen_port)
KeyboardInterrupt
Exception exceptions.AttributeError: '_Nudger__reactor' in <method
Nudger.__del_
_ of Nudger instance at 8236c10> ignored



(ps: 'ignored'???)

...
You know it's time to rethink your design when you can't debug your
#defines.
-Jim.



From Rick_Rankin-P15254 at email.mot.com  Thu Jun 10 12:50:21 1999
From: Rick_Rankin-P15254 at email.mot.com (Rick Rankin)
Date: Thu, 10 Jun 1999 09:50:21 -0700
Subject: WinNT: Corrupt installation detected
Message-ID: <7joq9h$qd2$1@schbbs.mot.com>

I'm trying to install 1.5.2 on my Windows NT system. I previously had 1.5.2
beta 2 installed, which I uninstalled before trying to install the final
release. When I try to run the installer (py152.exe), I immediately get a
dialog box that says "Corrupt installation detected".

I'm sure the downloaded installer is not corrupt. I opened it in WinZip and
used the test function. All files in the archive tested fine. Has anyone
seen this before?

Thanks,

Rick Rankin
Motorola SATCOM






From nospam at bitbucket.com  Sun Jun 27 01:21:43 1999
From: nospam at bitbucket.com (Phil Mayes)
Date: Sat, 26 Jun 1999 22:21:43 -0700
Subject: Memory leaks
References: <7l35rf$dl3$1@news3.infoave.net>
Message-ID: <Txid3.782$dR.68184@newsfeed.avtel.net>

Steven L. Kille wrote in message <7l35rf$dl3$1 at news3.infoave.net>...
>I'm embedding python in an MFC application.  When I close the application I
>get a bunch memory leak messages (16).  Im sure this is just memory
>allocated on startup and does not accumulate during the execution of my
>program so it's not big deal.  It's just kind of annoying, especially if
>your looking for memory leaks in your own program.  Does anyone know of any
>other function I need to call other than Py_Finalize() when closing down my
>app?  If I just place the following two lines in my InitInstance() function
>I get the memory leak messages:
>
>BOOL CFormulasApp::InitInstance()
>{
>
>  Py_Initialize();
>  Py_Finalize();
>return FALSE;
>}


I ran into this a while back, then discovered it was only in the debug
build, so I ignored it.  See if that's the case for you, too.  I forget
how I measured non-debug leaks - BoundsChecker, maybe.
--
Phil Mayes    pmayes AT olivebr DOT com
Check out our PalmPilot news page at http://www.olivebr.com/pilot/news.htm






From hniksic at srce.hr  Wed Jun  2 16:46:49 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 02 Jun 1999 22:46:49 +0200
Subject: pickle vs .pyc
References: <7j43pn$55l$1@mlv.mit.edu>
Message-ID: <87wvxmba9y.fsf@pc-hrvoje.srce.hr>

Michael Vezie <mlv at pobox.com> writes:

> from cache import fast,slow
> 
> and it's right there.  It's compiled, and seems really fast (loading
> a 50k file in .12 seconds).

I understand that .pyc files use marshal.  Maybe this is simply a
matter of marshal being faster than cPickle?  Have you tried using
marshal directly?

(It makes sense for pickling to be slower than marshaling because it
does more; for example, it takes care about recursive relationships
and such.)



From boud at rempt.xs4all.nl  Sat Jun 26 16:00:58 1999
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 26 Jun 1999 20:00:58 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <59EAFD6B662F0706.86BFF3E906CBF8DE.784D2A219F85BC21@lp.airnews.net> <99062523473500.00769@Vegus>
Message-ID: <7l3bhq$1se$1@essle.valdyas.org>

John Travers <jtravs at debian.org> wrote:

i Has anybody tried the PyKDE/Qt package, (you don't have to have KDE), it is
> absoloutly fantastic, quick, easy, powerful, far far better than Tkinter.
> I love it!
>

So do I - I realize that it's currently a Unix only solution, but
that's maybe because no-one has forked out for the Windows license
of Qt! A few months ago I went through all the options myself,
and PyKDE was the most satisfactory, especially in terms
of pleasure in working with. It's now in version 0.8, and
pretty good:

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

is the place to get it.

-- 

Boudewijn Rempt  | www.xs4all.nl/~bsarempt




From gutier at intergate.bc.ca  Thu Jun  3 02:53:01 1999
From: gutier at intergate.bc.ca (Gerald Gutierrez)
Date: 02 Jun 1999 23:53:01 -0700
Subject: Threaded Tkinter
Message-ID: <m3btex4vxu.fsf@tiana.kli.prv>

Hi all.

I have two questions and I'd really appreciate it if someone could
help me sort it out.

My first one -- I'm starting out with Tkinter. I wrote the code below,
which will supposedly create a button with an image, that image
changing every second. If I set the button to call swapIcons() when
pressed, the image changes fine, but using the SpinThread, the icons
don't change. Is this because Tkinter is not thread-safe and I cannot
safely change the button's image property from within a thread other
than the main one?

import Tkinter

class App(Tkinter.Frame):

    def __init__(self):
	Tkinter.Frame.__init__(self)

	self.im1 = Tkinter.Image("photo", "tux", {"file":"oldtux.gif"})
	self.im2 = Tkinter.Image("photo", "smiley", {"file":"smiley.gif"})
	self.curImage = "tux"
	self.b = Tkinter.Button(self)
	self.b["image"] = self.curImage
	self.b["command"] = self.swapIcons
	self.b.pack({"side":"left"})
	self.pack()

    def swapIcons(self):
	if (self.curImage == "tux"):
	    self.curImage = "smiley"
	elif (self.curImage == "smiley"):
	    self.curImage = "tux"

	self.b["image"] = self.curImage

import threading
import time
class SpinThread(threading.Thread):

    def __init__(self, app):
	threading.Thread.__init__(self)
	self.app = app
	self.setDaemon(1)

    def run(self):
	while (1):
	    self.app.swapIcons()
	    time.sleep(1)
	    
app = App()
st = SpinThread(app)
st.start()
app.mainloop()


My second question -- I'd like to use Tkinter an Fnorb (CORBA)
together as the UI and the network agent, respectively. Incoming CORBA
calls will change appliation state and cause a change in the UI.

Now, since both Tkinter and Fnorb have their own event loops, I
probably have to run them each in a separate thread. This is fine with
me. The problem is that, if my first problem above is caused by the
fact that I cannot use Tkinter in mutiple threads, I will need to send
UI change requests for processing from within the main thread. 

One way I know to do this is to have some sort of queue to which I can
send messages. Tkinter's event loop will have to process the
queue. How do I go about accomplishing this?

If this is not the correct way, what is?

Thanks for any help.

Please forward replies to gutier at intergate.bc.ca.





From jtravs at debian.org  Sat Jun 26 19:12:04 1999
From: jtravs at debian.org (John Travers)
Date: Sun, 27 Jun 1999 00:12:04 +0100
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <37741264.941B5B49@bioreason.com>
Message-ID: <99062700165200.00448@Vegus>

> John,
> 
> Can you point me to any docs to help me get started with PyKDE?

Have you tried the tutorial at:

http://www.xs4all.nl/~bsarempt/python/tutorial.html

Thats what got me started.

Also you can use the Qt documentation almost literally, any deviations are
described in the PyKDE docs. 

Hope this (very little) helps, I'm faily new to this too, just impressed by the
ease of it all compared to GTK in C!!

--
John Travers

"Sometimes I think the surest sign that intelligent life exists elsewhere
in the universe is that none of it has tried to contact us!"



From arcege at shore.net  Tue Jun  1 12:54:41 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 01 Jun 1999 16:54:41 GMT
Subject: Looking for a lib fun
References: <3753FDF4.5A13F8B8@zipzap.ch>
Message-ID: <lfU43.1452$nn.444697@news.shore.net>

Boris Borcic <zorro at zipzap.ch> wrote:
: There is a function I can't seem to find in the
: standard library. One that will turn XML/HTML
: offending characters to corresponding entity- or
: character references.

: Is it indeed in the lib ? Where ? If not, could a kind
: soul hand me one ? Writing it really feels like
: reinventing the wheel for the 10^nth time.

: TIA

Python 1.5 has the htmlentitydef module which you can use:

import htmlentitydefs
charmap = {}
for key, val in htmlentitydefs.entitydefs.items():
  charmap[val] = key

At this point, charmap will contain character to entity references.
  charmap['>'] == 'gt'
  charmap[chr(160)] == 'nbsp'
  etc.

I don't know if anyone has a more "complete" set tho.

  -Arcege




From aahz at netcom.com  Wed Jun 30 12:39:32 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 30 Jun 1999 16:39:32 GMT
Subject: Python Editor
References: <AJqe3.1318$24.76@news01.t-net.net.ve>
Message-ID: <7ldh84$rle@dfw-ixnews12.ix.netcom.com>

In article <AJqe3.1318$24.76 at news01.t-net.net.ve>,
Jose A. Antich <jantich at cantv.net> wrote:
>
>   Does anyone know if there's any Python Editor for the MAc?

IDLE, a debugger/editor written in Python.  You'll need Python 1.5.2 and
Tcl/Tk 8.0 (*not* 8.1).
--
                      --- Aahz (@netcom.com)

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



From amaranda at nospam.com  Tue Jun 15 09:32:05 1999
From: amaranda at nospam.com (Alex Maranda)
Date: Tue, 15 Jun 1999 14:32:05 +0100
Subject: (wxPython) Double left clicking and right clicking with ListBoxes
References: <7k5ec1$pl3$1@zonnetje.NL.net>
Message-ID: <376655D5.692A@nospam.com>

Vivienne wrote:
> 
> Hi there,
[snip]

Vivienne,
The right place to ask about wxPython is on the m-list:
http://starship.python.net/mailman/listinfo/wxpython-users

This way you'll be sure that TOWK (The One Who Knows :) will see your
messages (i.e. Robin Dunn).

Cheers,
-- 
Alex Maranda         mailto: amaranda at spider dot com
Spider Software Ltd. Tel: +44 (0)131 4757036
Edinburgh, UK        http://members.xoom.com/Alex_Maranda
STREAMS based communications protocols for embedded systems



From fredrik at pythonware.com  Tue Jun 22 07:15:28 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 22 Jun 1999 11:15:28 GMT
Subject: htmllib: CR in CDATA
References: <00d201bebc5a$adb129a0$0301a8c0@mnot.net> <012601bebc91$f3ded760$0301a8c0@mnot.net> <014401bebc9a$1963a9e0$0301a8c0@mnot.net>
Message-ID: <000b01bebca0$8a022c20$f29b12c2@pythonware.com>

Mark Nottingham wrote:
> It appears that htmllib doesn't ignore returns in CDATA fields, as HTML 4.0
> says it should.

well, htmllib doesn't claim to be HTML 4.0 compliant...

> OK, I'm starting to have a really nice conversation with myself now ;-)
> 
> htmllib DOESN'T change the newline to a single space - it leaves it in.
> 
> CDATA is a sequence of characters from the document character set and may
> include character entities. User agents should interpret attribute values as
> follows:
>           Replace character entities with characters,
>           Ignore line feeds,
>           Replace each carriage return or tab with a single space.
>      User agents may ignore leading and trailing white space in CDATA
> attribute values (e.g., "   myval   " may be interpreted as "myval").
> Authors should not declare attribute values with leading or trailing white
> space.

...and it doesn't claim to be a "user agent", either...

</F>




From mss at emsoft.de  Sun Jun 27 14:31:54 1999
From: mss at emsoft.de (Michael S. Schliephake)
Date: Sun, 27 Jun 1999 20:31:54 +0200
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net>
Message-ID: <37766E1A.49FC9062@emsoft.de>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/19990627/94fa5ac6/attachment.html>

From rdev at my-deja.com  Wed Jun 30 14:18:57 1999
From: rdev at my-deja.com (rdev at my-deja.com)
Date: Wed, 30 Jun 1999 19:18:57 +0100
Subject: Problem with bsddb on NT
Message-ID: <000015ca@bossar.com.pl>

From: rdev at my-deja.com

I've been using bsddb on both Unix
and NT and I found a very serious bug
on the NT side:

Using database opened with btopen:

If I write random length records using
a six byte key (arbitrary), the bsddb
will crash with a protection exception
after writing, typically, between 50 and
5000 records.

Using database openend with hashopen:

Same test as above, but rather than
crashing with a protection exception,
it generates a python exception as follows:
Traceback (innermost last):
  File "Prog\bsddb_test.py", line 52, in ?
    test ()
  File "Prog\bsddb_test.py", line 49, in test
    adb[key] = rec
bsddb.error: (0, 'No error')

It's easy to recreate using the test program below.
It works just fine on Unix but dies quickly on NT.

If anyone has any ideas at all, please help.  This
is a critical problem and I can't seem to find any
workaround.

Test Program:

import random
import bsddb
import os

# Test overview:
#
# Write a random length record between 1 and 1000 bytes,
# using a 6 byte random valued
# key.  Repeat until the database crashes (typically between
# 50 and 5000 writes.
#
# Crashes in different ways depending on whether bsddb was
# opened as btree or hash database.


def test ():
        count = 0
        # Remove any old database in order to start clean
        try:
                os.remove('test.db')
        except:
                pass
        # Either open bsddb in hash mode or btree mode.
        # Each fails in its own way.
        adb = bsddb.btopen ('test.db', 'c')
        #adb = bsddb.hashopen ('test.db', 'c')
        # Generate a long constant string (1001 chars)
        longstring1 = 'a' * 1001
        while 1:
                # Generate a large random string
                num = random.random()
                longstring = str(int(num * 1000000)) * 200
                # Pick a record length from 1 to 1000
                reclen = int(random.random() * 1000) + 1
                # Use a keylen of 6 (arbitrary)
                keylen = 6
                #if keylen + reclen > 214:
                #       continue

                # Generate a key by taking the first keylen chars
                # from the long string
                key = longstring[:keylen]
                # Generate a record by taking the first reclen chars
                # from the constant string (contents don't matter here)
                rec = longstring1[:reclen]
                count = count + 1
                print 'count = ', count
                print 'lengths = ', keylen, reclen, keylen + reclen
                print 'Key = ', key
                # Add it to the database
                adb[key] = rec


test ()


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From achrist at easystreet.com  Mon Jun 21 20:08:48 1999
From: achrist at easystreet.com (Al Christians)
Date: Mon, 21 Jun 1999 17:08:48 -0700
Subject: choosing random numbers with weights/probability?
References: <929987267.6227@www.remarq.com>
Message-ID: <376ED410.6D8B1562@easystreet.com>

The standard way to do this (with non-integer weights) is to have
a weight associated with each choice, normalized (scaled) so that the
maximum weight is equal to 1.0.  Then, after randomly selecting an item
from the collection, generate a uniform random number  between 0.0 and
1.0 and accept the choice only if the second random number is less 
than the weight of the randomly selected item.  If it is not, the item
is not used and you have to repeat the entire process with new
random numbers as many times as needed to get an item for which the
weight exeeds the second random number.

Al
 

kevinsl wrote:
> 
> I've been using the whrandom.choice routine and it's very
> useful. But is there any way to add weights or
> probabilities of being chosen to the items in the list?
> 
> example:
> 
> list=['one','two','three']
> item=whrandom.choice(list)
> 
> Is there any way to say that 'one' and 'two' have a 25%
> chance of being chosen, and 'three' has a 50% chance?
> 
> I'm hoping there's already a module to do this... or else
> I'll be writing my own..
> 
> thanks,
> 
> Kevin
> 
> **** Posted from RemarQ - http://www.remarq.com - Discussions Start Here (tm) ****



From hniksic at srce.hr  Wed Jun  2 09:49:41 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 02 Jun 1999 15:49:41 +0200
Subject: A problem with re.split()
Message-ID: <87pv3en24q.fsf@pc-hrvoje.srce.hr>

I understand that the idea with "re" module is for it to behave as
closely to Perl's regular expressions as possible (which is why the
order of arguments in string.split and re.split is different,
string.split being the logical one).

The problem is with re.split() in this case:

$ python
Python 1.5.2 (#3, May 23 1999, 19:57:40)  [GCC 2.8.1] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import re
>>> re.split('', 'foo')
['foo']

Perl splits it to ['f', 'o', 'o']:

$ perl -e 'print join(":", split("", "foo"))'
f:o:o



From skip at mojam.com  Fri Jun 11 08:26:03 1999
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 11 Jun 1999 12:26:03 GMT
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <375f92b4.618467559@news.u-psud.fr>
Message-ID: <376100E9.855C11C0@mojam.com>

Laurent POINTAL wrote:
> Fernando Mato Mira wrote:
> for Python vs Tcl
> >:   If you have references to articles of the sort, that will be useful,
> >: too.
> 
> See http://www.cwi.nl/~sjoerd/PythonVsTcl.html

While I'm sure that what Sjoerd wrote is accurate, it described Python
and Tcl as they existed around late 1995 or early 1996.  Both languages
have changed more than a little since then. 
 
> I remeber having seen an article of a guy explaining why he switched
> from Tcl to Python for a large project... but I loose the URL.

That would be Frank Stajano. He presented a paper at last November's
Python conference entitled "Implementing the SMS server, or why I
switched from Tcl to Python".

   
http://www.foretec.com/python/workshops/1998-11/proceedings/papers/stajano/stajano.html

Frank's perspective is an interesting one, because unlike most of us, he
used Tcl and [incr Tcl] happily for several years before encountering
Python, so he brings a depth of understanding to the discussion most of
us don't have.  (I, on the other hand, got an immediate bad taste from
Tcl and spit it right out. ;-)

-- 
Skip Montanaro	| Mojam: "Uniting the World of Music"
http://www.mojam.com/
skip at mojam.com  | Musi-Cal: http://www.musi-cal.com/
518-372-5583



From piers at cs.su.oz.au  Wed Jun  2 21:29:36 1999
From: piers at cs.su.oz.au (Piers Lauder)
Date: Thu, 03 Jun 1999 11:29:36 +1000
Subject: cPickle dumps core
Message-ID: <129.78.10.43.126.1589.928373856.156.11041>

I have a problem with Python-1.5.2 and cPickle 1.63.

Her's an example:

	# s Python-1.5.2 ; ./python
	Python 1.5.2 (#10, May 11 1999, 15:32:03)  [GCC 2.8.1] on sunos5
	Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
	>>> import cPickle
	>>> cPickle.loads("ht")
	Traceback (innermost last):
	  File "<stdin>", line 1, in ?
	cPickle.BadPickleGetBus error - core dumped

(cPickle behaves similarly on a Linux/386 box.)

I'd be very grateful if anyone can suggest a fix!







From S.I.Reynolds at cs.bham.ac.uk  Thu Jun 17 09:01:43 1999
From: S.I.Reynolds at cs.bham.ac.uk (Stuart Reynolds)
Date: Thu, 17 Jun 1999 14:01:43 +0100
Subject: BiBTeX module?
References: <joe-160619991358444033@chinacat.salk.edu>
Message-ID: <3768F1B7.644A@cs.bham.ac.uk>

Joe Strout wrote:
> 
> Is there a BiBTeX module for Python?  I did a search at python.org, but
> apart from a Perl module (ick!), I didn't find anything.
> 

I'm in the middle of writing a program to extract BibTeX tags from bib
files at the moment. We'll be using it to produce an online
reinforcement learning research papers database. If you can wait a
couple of weeks it'll probably be finished and on the web. If you want,
I'll send you the source when its done.

Stuart



From behrends at cse.msu.edu  Sat Jun 26 01:50:38 1999
From: behrends at cse.msu.edu (Reimer Behrends)
Date: 26 Jun 1999 05:50:38 GMT
Subject: total idiot question: +=, .=, etc...
References: <000d01bebeda$1b7430a0$d29e2299@tim>
Message-ID: <slrn7n8qiv.kl.behrends@allegro.cse.msu.edu>

Tim Peters (tim_one at email.msn.com) wrote:
> [Reimer Behrends]
> > While I'd be the first to agree that C++ is not exactly the epitome of
> > language design, I would think that at least the local variables should
> > be easy to identify. The only real problem is to distinguish global
> > variables from instance variables. And this is made worse in C++ by
> > spreading all the information related to a class over more than one
> > file. It should be noted that other languages do not necessarily have
> > the same problem and that careful design can easily avoid it.
> 
> But even with declarations up the wazoo, distinguishing local from instance
> vars is a major problem in real-life C++:  any OO language with declarations
> has to allow local vars to shadow instance vars,

To the contrary; Eiffel, under no circumstances allows shadowing of
instance variables by formal parameters or by local variables. Yes,
changing stuff in a superclass can thus break code in a subclass; but
this can always happen (like if you add a new method that also exists in
a subclass). Doing a major internal change to a superclass requires at
least rechecking subclasses, anyway. (And I prefer compilers that are
too picky rather than too lenient.)

> else method implementations
> can be broken by trivial changes in far-removed superclasses (e.g., Java
> flatly prohibits shadowing in most cases -- but specifically allows this
> one).

Can you point me to the place in the Java language specification where
this is said? I'm asking, because I've heard that, too, but haven't ever
been able to find it and because the JDK happily allows to me shadow
instance variables left and right.

Of course, there are languages that don't have this problem. For
instance, in either Smalltalk or Eiffel, there are no modules outside
the class concept, and a variable is either local (which you can easily
see in the method declaration, or an attribute.

But I digress ...

[...]
> > But I do not think that the most basic operations of an OO language
> > should be complicated by cumbersome syntax (this includes instance
> > variable access and local method calls). The language should encourage
> > use of these elements, not get in your way.
> 
> Readability is more important to me over the long haul.

Huh? Where have I said otherwise? I just happen to think that these two
goals are not mutually exclusive. And there should be a better solution
than the language ordaining a strange kind of Hungarian notation.

> Say what you like,
> but which vars belong to the instance is screamingly obvious the instant you
> see
> 
>     self.x = self.y + k
> 
> while in
> 
>     x = y + k
> 
> it's at best a minor puzzle and at worst a long & tedious grep project.

I tell you what--I have this great idea for an addition to Python. You
know that you often can't tell what type a variable is? Well, we could
introduce the following convention:

	Require variables to be prefixed with "$" to denote scalar
	values--strings and integers, "@" for lists, and "%" for
	dictionaries.

Given the precedent of having to use self to access instance variables,
this is entirely consistent with the current language, and is going to
eliminate all those nasty problems where you can't tell from the context
what the type of a variable or function is. :)

	Tongue-firmly-in-cheek-ly yours, Reimer




From MHammond at skippinet.com.au  Mon Jun 28 21:11:21 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Tue, 29 Jun 1999 11:11:21 +1000
Subject: ActiveX, IDE
References: <7cKd3.3449$lr3.133866@news2.jacksonville.net>
Message-ID: <7l96c1$5hn$1@m2.c2.telstra-mm.net.au>

Hank Fay wrote in message <7cKd3.3449$lr3.133866 at news2.jacksonville.net>...
>I'd like to make Wintel apps using ActiveX controls under Python: is there
>an IDE for Python that will support this?

Pythonwin supports hosting ActiveX controls, but if you are expecting an
ActiveX IDE similar to VB, you will be dissapointed.

http://starship.python.net/crew/mhammond

Mark.






From Sven.Drescher at dlr.de  Fri Jun 18 04:59:47 1999
From: Sven.Drescher at dlr.de (Sven Drescher)
Date: Fri, 18 Jun 1999 10:59:47 +0200
Subject: Dumb key-binding question
References: <376434de@nntp.server.uni-frankfurt.de>
Message-ID: <376A0A83.D9D11DF2@dlr.de>

Hallo!
Ist mal schoen jemanden auf Deutsch zu fragen, nicht das ich was gegen
Fremdsprachen habe, aber ist halt einfacher.
Ich las, dass Du es geschafft hast Key-Bindings zu implementieren.
Ich versuche das schon seit laengerem, aber ohne Erfolg.
Mit:
    name.bind("<keypress-a>", command)
klappt das naemlich nicht so richtig. Und das mit den Keysyms hab ich
nicht so richtig verstanden. Waere schoen, wenn Du mir weiter helfen
koenntest. Ueber ein kleines Beispiel wuerde ich mich freuen.

Danke
Sven

mlauer at amalia-atm.rz.uni-frankfurt.de wrote:

> Pardon, but I can not get binding
> <Control-KeyPress> to work correctly.
> <Control-KeyRelease> works fine, but
> <Control-KeyPress> only functions while
> the holding the key. This is not exactly
> what I was expecting... Hmm ?
>
> --
> --
>     Regards & Gruesse from Mickey @ http://www.Vanille.de
>   ---------------------------------------------------------
>   How could anyone know me - when I don't even know myself ?



--
_______________________________________
German Aerospace Research Establishment
Institute of Flight Guidance
Air Traffic Management Systems
Tel.: (0531 -) 295 2515






From news at dorb.com  Mon Jun 21 21:51:52 1999
From: news at dorb.com (Darrell)
Date: Mon, 21 Jun 1999 21:51:52 -0400
Subject: choosing random numbers with weights/probability?
References: <929987267.6227@www.remarq.com> <m3g13l2lel.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <q1Cb3.1523$U6.5922@newsr2.twcny.rr.com>

Did a run of 100000. Looks like it works pretty well.

 Time: 5.33 sec
{'one': 24781, 'three': 50148, 'two': 25071}

--Darrell





From MHammond at skippinet.com.au  Tue Jun  1 19:30:27 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Wed, 2 Jun 1999 09:30:27 +1000
Subject: Why won't this work???
References: <7j0rkr$1eeu$1@rtpnews.raleigh.ibm.com>
Message-ID: <7j1qc5$3rj$1@m2.c2.telstra-mm.net.au>

Jr. King wrote in message <7j0rkr$1eeu$1 at rtpnews.raleigh.ibm.com>...
>I know you are getting tired of me, but I have tried repeatedly to solve

Well, I am now a little tired of you:

* You post around 100 lines of C++/Python code with nothing more than "it
doesnt work".  Not even a hint of a symptom.

* I try to mail you to let you know this to avoid further clogging of the
newsgroup, and find your address spam protected!

If you must ask simple questions about simple bugs in your own code, at
least have the decency to put a valid email address so the rest of the
newsgroup doesnt need to hear that you forgot to initialize a variable or
some other trivial error.

Mark.






From mike at proteus.dialup.cloud9.net  Fri Jun  4 10:55:31 1999
From: mike at proteus.dialup.cloud9.net (Michael Muller)
Date: Fri, 4 Jun 1999 10:55:31 -0400
Subject: Threaded Tkinter
In-Reply-To: <14167.21340.954229.2711@tiana.kli.prv>
References: <m3btex4vxu.fsf@tiana.kli.prv>
	<375690B9.37C1B8B6@cloud9.net> <14167.21340.954229.2711@tiana.kli.prv>
Message-ID: <Pine.LNX.3.91.990604104432.242E-100000@proteus.dialup.cloud9.net>

On Thu, 3 Jun 1999 gutier at intergate.bc.ca wrote:

> 
> Hi Michael.
> 
> Thanks for the suggestion.
> 
> I had been mulling over the idea of somehow getting the Tk event loop
> to periodically call a method of mine which would read a queue and do
> what's needed. Would the lock you describe interfere with this in any
> way? If not, how do I go about injecting messages into the Tk event
> loop?
> 
> Also, as you can guess I know little about Tk. Where is the event loop 
> in the code you provided below? Does the tk.after() call look at Tk's
> event loop and perform necessary actions?
[snip]

I omitted the event loop call in this example.  Add the line:

   tk.mainloop()

This is the standard event loop, and it will continue to process events
until the application is terminated.

tk.after() creates a "timer"; the event handler subsequently reacts to the
timer and calls the function at the appropriate time.

=============================================================================
michaelMuller = proteus at cloud9.net | http://www.cloud9.net/~proteus
-----------------------------------------------------------------------------
Mantra for the 60's: Tune in, turn on, drop out
Mantra for the 90's: Turn on, jack in, jerk off
=============================================================================




From matomira at iname.com  Wed Jun 30 11:10:59 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Wed, 30 Jun 1999 17:10:59 +0200
Subject: Language extensibility (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <37682DD3.43923243@compaq.com> <7ka2be$qja$1@nntp6.u.washington.edu> <37776FFA.2D85@mailserver.hursley.ibm.com> <lw7looxx7m.fsf@copernico.parades.rm.cnr.it> <876748mcxg.fsf@home.ivm.de> <37788BB9.ABD@mailserver.hursley.ibm.com> <lw67471fyi.fsf@copernico.parades.rm.cnr.it> <3779ED0F.2C67@mailserver.hursley.ibm.com> <nkjzp1hizvb.fsf@tfeb.org>
Message-ID: <377A3383.D453468@iname.com>


Tim Bradshaw wrote:

> But the language of Lisp macros is Lisp, that's really the whole
> point!  Without knowing TCL, I find it hard to see how you can
> introduce new constructs to the language *without* a macro language,
> even if that language is TCL.

Err.. What about Forth keywords?





From greg.ewing at compaq.com  Tue Jun  1 00:11:11 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Tue, 01 Jun 1999 16:11:11 +1200
Subject: Circular references (was: Defining VCL-like framework for Py
References: <7hvnah$ern$1@cronkite.cc.uga.edu> <1284891865-6060292@hypernet.com>
Message-ID: <37535D5F.10A86B10@compaq.com>

Gordon McMillan wrote:
> 
> Ce'Nedra, er, Hans wrote:
> 
> Another way is to put one of them (the parent, normally) in some
> kind of collection, (say, a dict keyed by a name, like id(parent)).
> Tell the child the parent's name and the collection. When you cut the
> parent loose from the dict, it goes away (the child does not directly
> reference it). Then the child goes away (nobody is referencing it).

I don't see how this helps, because then you have to do something
special to make the parent go away (the dict is holding a reference
to it). If you know when the parent is supposed to go away, you
might as well call parent.free().

Greg



From robin at jessikat.demon.co.uk  Wed Jun  2 08:33:32 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Wed, 2 Jun 1999 13:33:32 +0100
Subject: Patch: Win32 stack overflow
References: <375503B1.9893E25B@appliedbiometrics.com>
Message-ID: <CmfBuDAcSSV3Ewuc@jessikat.demon.co.uk>

In article <375503B1.9893E25B at appliedbiometrics.com>, Christian Tismer
<tismer at appliedbiometrics.com> writes
>Patch for python15.dll
>
>stack152.py
>
works great
-- 
Robin Becker



From graham at sloth.math.uga.edu  Fri Jun  4 11:03:06 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 4 Jun 1999 15:03:06 GMT
Subject: "The Python Way"
References: <001101beae49$39af7bc0$949e2299@tim>
Message-ID: <7j8pra$55v$4@cronkite.cc.uga.edu>

[Fredrik Lundh]
> Would both Guido and TIm Peters be willing to collaborate on a short
> paper -- call it "The Python Way" for lack of a better title -- which
> sets out the 10-20 prescriptives they might offer to those who come to
> Python from other languages and immediately want to find a way to bend
> it into uncomfortable positions -- (implement closures, etc.).

Oh thank you Fredrik for posting the above. Does anyone else consider
that the above sounds slightly paranoid? Can I make a suggestion. When
someone, for example myself, makes a suggestion for adding something
to Python or changing something, or dare I say it "improving" Python,
perhaps you could consider the faint possibility that they are not
trying to "find a way to bend Python into uncomfortable positions", 
but rather they are earnest Python users who happen to have some
expertise/experience in the new idea being suggested, and think that
it could not only greatly benefit Python, but also fit in rather well
with the mystical "Python way". 

A short case in point. Recently I suggested adding mark-sweep-refcounting 
garbage collection to Python.  I was immediately met with a wall of posts 
(some from you), saying basically what you say above, namely that "you 
want to bend Python into uncomfortable positions, ... positions that I 
suspect won't work or will be hard to implement". The reality is far from 
this. I don't want to bend Python at all. I quite *deliberately* proposed 
mark-sweep *with refcounts* (as opposed to any of the more high tech
techniques) so that no existing Python code need change at all, so that the
existing refcount mechanism stays the same, so that C extensions stay
the same, etc. Moreover I quite *deliberately* proposed mark-sweep with
refcounts because I know from experience with implementing such a scheme
that it won't be hard to implement.

In short when someone posts some suggested "improvement" to Python don't
be so defensive. Open your mind to the possibility that the poster may
have a point, that they are not trying to bend Python into uncomfortable
positions, that they may know something about what they are positing 
about, and may even be positing something that would be good for Python. 

graham
-- 
                  A cat named Easter
              He say "will you ever learn"
              Its just an empty cage girl
                 If you kill the bird



From asang at yahoo.com  Wed Jun 30 19:58:16 1999
From: asang at yahoo.com (Asang Dani)
Date: Wed, 30 Jun 1999 23:58:16 GMT
Subject: Problem using freeze with pywintypes
Message-ID: <7leauj$u2p$1@nnrp1.deja.com>

I have discovered minor problem with using freeze with win32all
modules by Mark Hammond. This occurs mainly because the name
"PyWinTypes15.dll" given for the "pywintypes" module dll. (I am posting
this because I don't want someone else to waste time on understanding
the same problem).

The problem and workaround are described as follows.

1. Create a small script test.py

------------------------------------------------------------
test.py
------------------------------------------------------------
import pywintypes
import win32file

ovstruct = pywintypes.OVERLAPPED()
print 'hello World\n'.

Unless "PYTHONEX" enviroment variable is set to the top level
directory where sources for "win32all" modules reside, none of the
win32* (and also pywintypes) module will get built into test.exe.

Case A: PYTHONEX not set
-------------------------

    2. freeze.py -p %pythonsrc% -o Build will try to build test.exe to
       be built in subdirectory "Build".

    3. Copy test.exe, PyWinTypes15.dll,?python15.dll and win32file.pyd
       to a machine on which python is not installed. move
       PyWinTypes15.dll and python15.dll %systemroot%\system32
       directory.

    4. set PYTHONPATH=%cd%

    5. type test.exe.

       Following error appears:
            "ImportError: No module named pywintypes".

    This occurs because "import pywintypes" statement causes core
python's
    "find_module" function to be called,?which looks for

    1. pywintypes.dll
    2. pywintypes.pyd
    3. pywintypes.pyc
    4. pywintypes.py

    in %PYTHONPATH% and cannot find it there. Python on win32 also
    examines registry key:

        HKLM\SOFTWARE\Python\PythonCore\1.5\Modules\<module-name>

    In case such a key exists, it's value is taken as the
    "full-pathname" of the dll from which module can be loaded. When
    win32all-xxx.exe is installed, it creates such key for pywintypes
    and some other modules. For pywintypes, the value is of REG_SZ
    type and points to "C:\WINNT\system32\PyWinTypes15.dll". On the
    machine where it is not installed, "import fails".

    Renaming PyWinTypes15.dll PyWinTypes.dll will work if the script
    used *only* pywintypes module. If it uses any other module like
    win32file, it rename creates problem because win32file.pyd
    "depends" on PyWinTypes15.dll.

    WORKAROUND:
    ----------

    1. Create a text file follwing contents and name it as "pyreg.ini"

\registry\machine\software\python
    PythonCore
        1.5
            Modules
                =
                pythoncom
                  = REG_EXPAND_SZ %systemroot%\System32\pythoncom15.dll
                pywintypes
                  = REG_EXPAND_SZ %systemroot%\System32\PyWinTypes15.dll


    2. copy pyreg.ini to the target machine on which exe is used, and
    type "regini pyreg.ini". (This can be automated through simple
    windows cmd script very easliy. The script can launch test.exe
    automatically after creating registry keys).


Case B: PYTHONEX is set to correct value
----------------------------------------
    freeze.py -p %pythonsrc% -o Build will try to build test.exe to
    be built in subdirectory "Build".

    This fails with a linker error,
        "pywintypes.lib" not found.

    because the Makefile created by freeze.py does not give full path
    for pywintypes.lib. This can be patched by changing
    win32makemakefile.py to treat pywintypes module specially. (All
    the win32****.pyd modules, depend on pywintypes.lib).

Hope this helps.

Thanks
asang..


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From news at dorb.com  Wed Jun  2 17:03:38 1999
From: news at dorb.com (Darrell)
Date: Wed, 2 Jun 1999 17:03:38 -0400
Subject: Python 2.0
References: <11A17AA2B9EAD111BCEA00A0C9B4179301E10ED1@forest.origin.ea.com> <374F1F6D.F3338F56@prescod.net> <374F5001.1C850DC@Lugoj.Com> <374F4591.3E66FB0@prescod.net> <zTS33.1171$nn.335894@news.shore.net> <374FF31A.6E990179@prescod.net> <37512edf.3024969@news.demon.co.uk> <3751573C.CAB4BFCB@hex.net> <w0g43.1119$H5.5564@newsr1.twcny.rr.com> <al8pv3i6wbg.fsf@kruuna.Helsinki.FI>
Message-ID: <G1h53.1795$j7.26459@newsr2.twcny.rr.com>

IONA has such a tool
http://www.iona.com/products/orbix/comet1.html
Also saw a book on this product but can't recall the name.
I go buy it.

--
--Darrell





From tim_one at email.msn.com  Tue Jun 15 01:29:06 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 15 Jun 1999 01:29:06 -0400
Subject: fork()
In-Reply-To: <Ri993.1$V4.6985@news-dal.corridex.com>
Message-ID: <000b01beb6ef$fcaba960$069e2299@tim>

[Tim]
> What you suggest is weaker, effectively lumping the SCCs into two gross
> classes (singletons and non-singletons), and nuking the latter first.

[Evan Simpson]
> Sorry if I was unclear, but that's not what I meant by "dangly bits".
> Rather, take the set of nodes from which you cannot reach a cycle.  Their
> subgraph is a union of DAGs, which can be released normally.  In your
> example, handing "c" a reference to "d<->e" converts it from a
> dangly bit to a cycle-bridge.

Ah, pulling the old DB-to-CB trick <wink>.  I can't say for sure whether you
were unclear before, but it's clear to me now.  Thanks!

> On the other hand, the SCC approach does look stronger than this, since it
> can handle cycle-bridges as well.  I'll happily take your word that
> SCC-counting is likely to be too much trouble, but how about just doing an
> SCC-analysis once M&S has discovered a cycle?  If this is all happening in
> the background, with no deadlines, and with rare, naughty data structures,
> then it doesn't have to be terribly efficient.

Well, we can *never* "win".  The instant the "why aren't my cycles
collected?" complaints go away, the "why does collecting my cycles take so
bloody long?" complaints arrive.  Note that the strongest proponents of GC
over the years here have not, like the Perl camp, been worried about
accidental cycles slowly piling up in long-running servers -- they're keen
to create cycles all over the place.

So slow won't fly forever.  That said, identifying SCCs can actually be done
quite efficiently; making an exact accounting of inter-SCC pointers is
harder but still not a killer.  I'm more bothered by the (lack of) *sense*
in it all:  ignoring a __del__ method on an object that has one is darned
hard to justify, yet Python can't (any more than Java can!) do anything
sensible with __del__ in cycles on its own.

So my plan for the hour <wink> is:  if M&S finds a cycle, it will clean it
all up iff no involved object has a __del__.  If an object in a trash cycle
does have a __del__, tough, the cycle won't be reclaimed.  Instead we'll
supply a function you can call if you care, that will return those objects
(one at a time?  in a list?  whatever), and you do with them what you will;
if you don't clean them up, you'll keep getting them back every time you
call that function until you resurrect them or break the cycle by hand in
the way that makes most sense to your algorithm.

In the face of ambiguity, refuse the temptation to guess <0.5 wink>.

dear-lord-now-he's-quoting-himself-ly y'rs  - tim





From MHammond at skippinet.com.au  Sun Jun 20 19:06:37 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Mon, 21 Jun 1999 09:06:37 +1000
Subject: Python source code documentation?
References: <7kdv0l$n0g$1@nnrp1.deja.com> <7kenc7$1od$1@m2.c2.telstra-mm.net.au> <376D5BA2.70D845C4@compaq.com>
Message-ID: <7kjs2b$k5n$1@m2.c2.telstra-mm.net.au>

Greg Ewing wrote in message <376D5BA2.70D845C4 at compaq.com>...
>Mark Hammond wrote:
>>
>> Pythondoc
>> really doesnt want to introduce ugly mark-ups in the docstring
>
>What about non-ugly markups? Surely it's possible to
>design a form of markup which doesn't look ugly.

Well, "ugly" is in the eye of the beholder :-)

The issue is a phrase such as:
"This method is very similar to <om OtherClass.ItsMethod>, but different"

Clearly, when reading the docstring as a docstring in the source-code, this
would qualify as ugly.  It distracts from the user reading the text.
Similarly, <B> </B> is also ugly, hence Pythondoc/structured text can
convert *bold* into bold (for example).

However, as far as marked-up text goes, I really dont consider my example
ugly.  Once you know that "om==object method", it really isnt that bad.

I suppose what I was trying to say was "Pythondoc doesnt want to introduce
mark-ups in the docstring as they are ugly" :-)

Indeed, Pythondoes goes some way to move all links to the bottom of the
docstring.  I cant remember the exact format, but the text and the just
target are kept seperate from each other.  The analogy would be that in my
world, I would consider:
Please see <jump The Python homepage:www.python.org>

quite acceptable, and indeed quite an improvement (wrt maintenance) if you
consider the generated documentation the "primary documentation" - ie, the
format people will most often use to read the docstrings.

In fact, I believe this is the biggest difference between the pythondoc
focus and the autoduck type focus.  Pythondoc strives to keep completely
readable docstrings and generate good documentation.  Autoduck type
formatting is happy to sacrifice some of the docstring readability in order
to create better documentation.

[Important to recall autoduck started as autodoc, which can parse C code and
marked up comments.  This is quite a different environment to Python
docstrings available at runtime.]

Mark.








From gherron at aw.sgi.com  Tue Jun 29 13:28:03 1999
From: gherron at aw.sgi.com (Gary Herron)
Date: Tue, 29 Jun 1999 17:28:03 GMT
Subject: Tkinter slide question
References: <3778EC8B.833DEBC4@hti.com>
Message-ID: <37790223.F469BC4D@aw.sgi.com>

David Stokes wrote:
> 
> I'm trying to implement a slide to control how often a function restarts
> but it appears that I'm not getting the result back from the 'variable'
> variable.  Could some kind soul out there on the proper way to read the
> slide 'position'?
> 
> Code snipit:
> Scale(label="Rescan",variable=self.yspeed,orient=HORIZONTAL,).grid(column=0)
> 
> currently returns PY_VAR12 instead of the slider value.

Try
  self.yspeed.get()
to get the value of the variable.

-- 
Dr. Gary Herron <gherron at aw.sgi.com>
206-287-5616
Alias | Wavefront
1218 3rd Ave, Suite 800, Seattle WA 98101



From jeffrey at Digicool.com  Fri Jun  4 15:20:26 1999
From: jeffrey at Digicool.com (Jeffrey P Shell)
Date: Fri, 04 Jun 1999 15:20:26 -0400
Subject: python active web pages
References: <jonathan-0406991640390001@borg.onegoodidea.net>
Message-ID: <7j98qf$c96$1@autumn.news.rcn.net>

> i also looked at ZTemplates (a component of Zope),
> 
>    <http://www.zope.org/Download/ZTemplates/>,
>
> and while it doesn't actually embed python, it has a limited embedding
> language of its own, DTML, which gets munged into python. i must say that
> DTML looks a bit hideous (if anyone from digicool is reading, sorry guys).

ZTemplates, aka DTML Syntax is this::

 <!--#var foo-->

DTML Scripting is this::

 <!--#var "foo(bar=something + _['sequence-item']"-->

It's not munged into python, it IS python.  It's just expressions, and it's
run through a Safe Evaluator (that (in Zope) works with security machinery,
and tries to prevent things like infinite recursion and some other
potentially unsafe things).

Just wanted to clear that up.

In Zope where everything can potentially be managed through the web and
control can be turned over to other people, the safety of expressions and
hook-ins to the Zope security machinery is a very powerful notion.

.jPS | jeffrey at digicool.com
zope, through the web. www.zope.org



From xhyuan at oe.fau.edu  Wed Jun  2 20:16:42 1999
From: xhyuan at oe.fau.edu (Xiaohong Yuan)
Date: Wed, 2 Jun 1999 17:16:42 -0700 (PDT)
Subject: problem with static linking
Message-ID: <Pine.SGI.3.92.990602171041.19399A-100000@sea30>

I am working on Python c extensions on QNX. Since QNX doesn't support
dynamic linking. I have to link my c extension module staticly with python
interpreter. But in my c extension module, I have external variable
declaration, which makes it fail to be linked with Python interpreter.
There must be some way to solve this, could anybody help me with this?

Thank you very much.

Xiaohong Yuan




From jason at harlequin.com  Thu Jun 10 06:58:23 1999
From: jason at harlequin.com (Jason Trenouth)
Date: Thu, 10 Jun 1999 10:58:23 GMT
Subject: Asserting object class
References: <375F8986.73BA3C5@i-data.com>
Message-ID: <379899bf.2325681250@newshost>

On Thu, 10 Jun 1999 09:46:46 +0000, Adi <adi at i-data.com> wrote:

> In a method, I'd like to assert that a parameter is an object which is
> of a specific class or has this class as it's base class.
> 
> I can test on the __class__ and its __bases__ attributes, but its clumsy
> and I'm wondering if there is an easier notation, like:
> 
> def method(self, p)
>     assert class(p) == class(self)
> 
> This is comparable to - in a strongly typed language - specifying a
> required base class as the type of the parameter.

FWIW, in Dylan you'd write it as:

	define method m (self, p)
	  assert(object-class(p) == object-class(self));
	  ..
	end;

__Jason



From jdrudnicki at yahoo.com  Wed Jun  2 04:59:35 1999
From: jdrudnicki at yahoo.com (Jim Rudnicki)
Date: Wed, 2 Jun 1999 01:59:35 -0700
Subject: All this self dot is making me...
Message-ID: <Vn653.468$HD.99317@typhoon-sf.snfc21.pbi.net>

To the beginner, having to type all these self. 's in class methods seems a
bit painful.  So all three of these methods below do the same thing.  That
leaves one to wonder "why have classes at all?"  A little thought and I
remember that this same thing is happening in C++: the compiler is adding a
this pointer to the passed parameters.  So Func and fooFunc are the same
thing.  The only difference between a method and a class method is
scope/access rules and tacking on this pointers, which in C++ is done by the
compiler. Correct?

So what has me perplexed is why all the self. 's?  Is this a namespace issue
or maybe just to expensive to do at runtime?

For those of us who have less cartilage left in their hands, would it be
acceptable to use "my." or better "me."  Or is self. strongly ingrained?
The two less characters add up and the keys are on stronger fingers.

nite,
Jim

def Func( this, a, b ):
  this.x = a + b

class Foo:
  def __init__( self ):
    self.x = 0
    self.f = Func
  def fooFunc( me, a, b ):
    me.x = a + b
  #if it was C++
  # fooFunc( a, b ) {
  # x = a + b; }

# who needs class methods?
z1 = Foo()
Func( z1, 5, 7 )
print z1.x
# it doesn't really matter if the method is in a class
z2 = Foo()
z2.f( z2, 5, 7 )
print z2.x
# this is the same as z1 approach
z3 = Foo()
z3.fooFunc( 5, 7 )
print z3.x





From mlv at pobox.com  Wed Jun  2 16:22:15 1999
From: mlv at pobox.com (Michael Vezie)
Date: 2 Jun 1999 16:22:15 -0400
Subject: pickle vs .pyc
Message-ID: <7j43pn$55l$1@mlv.mit.edu>

I need to be able to read a couple very complex (dictionary of arrays 
of dictionaries, and array of dictionaries of array of dictionaries) 
data structures into python.  To generate it by hand takes too long, 
so I want to generate it once, and read it each time (the data doesn't 
change).

The obvious choice is, of course pickle, or some flavor thereof.
But can someone tell me why this wouldn't be faster:

In the code that does the "pickling", simply do:
f = open("cache.py", "w")
f.write("# cache file for fast,slow\n")
f.write("fast = "+`fast`+'\n')
f.write("slow = "+`slow'+'\n')
f.close()
import cache

Then, later, when I want the data, I just do:

from cache import fast,slow

and it's right there.  It's compiled, and seems really fast (loading a 
50k file in .12 seconds).  I just tried the same data using cPickle, and 
it took 1.4 seconds.  It's also not as portable.  There is a space savings 
with pickle, but it's only 5% (well, 56% if you count both the .py and 
.pyc files), but that doesn't really matter to me.

Am I missing something here?  This sounds like an obvious, and fast, 
way to do things.  True, the caching part may take longer.  But I 
really don't care about that, since it's done only once, and in the 
background.  

Michael



From guido at cnri.reston.va.us  Thu Jun 10 00:50:10 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 10 Jun 1999 00:50:10 -0400
Subject: GC SIG
References: <Pine.A41.4.10.9906081526430.65174-100000@barney.mcit.com>
Message-ID: <5l7lpczmkd.fsf@eric.cnri.reston.va.us>

Robert Meegan <Robert.Meegan at wcom.com> writes:

> Are there enough potential Sanitation Engineers to create a Garbage
> Collection SIG?

Rather than create a hell (mailing list) where the discussion can go
on forever without ever concluding, why not create a bible (FAQ) that
summarizes all arguments pro and con once and for all?  Then when the
issue is raised, we can simply point to the GC FAQ.

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



From tim_one at email.msn.com  Wed Jun  9 01:18:39 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 9 Jun 1999 01:18:39 -0400
Subject: tstate invalid crash with threads
In-Reply-To: <375D3424.6EF8E85F@commsecure.com.au>
Message-ID: <000101beb237$88b71d80$999e2299@tim>

Followups to the Thread-SIG, please.

[Ray Loyzaga]
> I have been playing with a multithreaded tcp based server for a while,
> and during some serious stress testing continually hit a problem
> involving the interpreter crashing with "Fatal Python error:
> PyThreadState_Delete: invalid tstate".
> ...
> It appears to be a subtle race in PyThreadState_Delete ....
> interestingly, if I uncomment the small sleep in "handle" in the server,
> ie. make the server slower, it seems to work for ever ... 4m transactions
> before I gave up. I think the problem only comes if you are creating and
> destroying threads quickly in parallel.

PyThreadState_Delete is called from very few places, and one of them strikes
me as suspicious:  at the end of threadmodule.c's t_bootstrap, we have:

	PyThreadState_Clear(tstate);
	PyEval_ReleaseThread(tstate);
	PyThreadState_Delete(tstate);
	PyThread_exit_thread();

The suspicious thing here is that PyEval_ReleaseThread releases the global
interpreter lock, so nothing is serializing calls to PyThreadState_Delete
made from the following line.  PyThreadState_Delete in turn does no locking
of its own either, but mutates a shared list.

If this isn't plain wrong, it's certainly not plain right <wink>.  Matches
your symptoms, too (very rare blowups during high rates of thread death).

Guido?  I haven't been able to provoke Ray's problem under Win95, but the
above just doesn't smell right.

win95-didn't-crash-but-the-TAB-and-ESC-keys-did-swap-their-
    meanings!-ly y'rs  - tim





From tismer at appliedbiometrics.com  Tue Jun 15 07:20:10 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Tue, 15 Jun 1999 11:20:10 GMT
Subject: ANN: Stackless Python 0.2
References: <37628EAA.C682F16C@appliedbiometrics.com> <7k4fd4$c04$1@brick.cswv.com>
Message-ID: <376636EA.84240129@appliedbiometrics.com>


Neel Krishnaswami wrote:

[Stackless Python 0.2]

> This is very neat, and you are completely deranged. I know just
> enough to know that I should cheer you on, mind, but I'll try to
> cheer loudly. Feel free to make an announce when you add coroutine
> support. Please? :)

Thanks :-)

Sure I will. I'd like to know wether my binary works ok for you.
You might also try a recursive function, it should raise an
exception after 29999 recursions (just to have *a* limit).

I'm just undecided on the design. There are a couple of
different coroutine interfaces. You find them for instance in
Scheme, in Modula II, some Oberons, Icon, and in Tim Peter's
example which I added to my stackless archive.
Do you have a proposal?

For Stackless V. 0.3 I'm planning for some more nonrecursive
builtins, and also for a proper "invalid opcode" handling,
which makes it possible to extend the standard interpreter
without having to build Python or rewrite eval_code completely.
This will give a new playing ground for Michael Hudson.

Then the Stackless implementation will only undergo changes
if something needs internal support. Everything else will
go into extension modules which can be loaded on demand.

One extension I'm planning is an API engine, which knows all
C API calls and all internal structures. Together with a
small fast engine, this will allow to build new fast
functions without a compler. Maybe this is not for this year.

But first I have to get out of this asylum :-?

lunaticly y'rs - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From moshez at math.huji.ac.il  Mon Jun  7 18:03:03 1999
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 8 Jun 1999 01:03:03 +0300
Subject: waste of resources ?
In-Reply-To: <87pv37zpex.fsf@pc-hrvoje.srce.hr>
References: <375C0F91.1C73E4E3@icrf.icnet.uk> <375C16A9.5ADC4990@icrf.icnet.uk> <375C2003.B01BEC3@arlut.utexas.edu> <87yahvzqz3.fsf@pc-hrvoje.srce.hr> <ytt1zfnen4h.fsf@gilgamesh.cse.ucsc.edu> <87pv37zpex.fsf@pc-hrvoje.srce.hr>
Message-ID: <Pine.SUN.3.95-heb-2.07.990608005904.9179A-100000@sunset.ma.huji.ac.il>

On 7 Jun 1999, Hrvoje Niksic wrote:

<about syscalls throwing EINTR and aborting>
> I don't know about Perl, but a good fix would be to have the
> interpreter do the hard work for us.  If a system call fails and errno 
> is EINTR, then the signal handler should be called, *and* the syscall
> should be restarted.

Ummmmm......

You can do the hard-work yourself, but once and for all:

def myread(file):
	while 1:
		try:
			return file.read()
		except (IOError, os.error), detail:
			if detail.args[0] != errno.EINTR: raise

And just call myread(foo) instead of  foo.read(). 
Where's the catch-22?
--
Moshe Zadka <mzadka at geocities.com>. 
QOTD: My own exit is more likely to be horizontal then perpendicular.






From tim_one at email.msn.com  Thu Jun 17 01:58:13 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 17 Jun 1999 01:58:13 -0400
Subject: fork()
In-Reply-To: <u3k8t3ohh8.fsf@ares.sys1.ngo.okisoft.co.jp>
Message-ID: <000101beb886$628a1ea0$099e2299@tim>

[Hisao Suzuki and Tim, jockeying over new/delete in C++ vs __del__ in
Python,
 and their relation to cycles and some Stroustrup quotes]

[Hisao]
> ...
> You miss the point.

Could be.  Expect it's really that I disagree with it, though.  I agree that
(barring pathologies) the only a way an object is immortal in C++ is if it's
explicitly new'ed but never explicitly delete'd, and that the only way an
object is immortal in Python (make that CPython, & make that today) is if
it's in a cycle.
You want to infer something from that, while I think the languages are so
different that trying to read anything into it is like trying to see animal
figures in the passing clouds <wink -- but, yes, your "It sounds like a
theoretical nonsense.">.

> ...
> If an object is part of a cycle, the object will never lose the
> reference(s) to it.  So the phrase `(be) never deleted' for C++
> can be safely translated as `(be) part of cycles' for Python.
> You see?

To me, that merely describes the conditions under which the respective
languages treat an object as immortal today.  It makes good sense for
Stroustrup to say that in C++ a garbage collector shouldn't invoke the
finalizer for its flavor of immortal objects, but he's not arguing that
position *because* they're immortal today, but instead because explictly
new'ed objects never have their finalizer invoked in C++ unless explicitly
delete'd, and GC is not explicit delete'tion.  Python has no such rules, so
the core of his argument doesn't apply to Python without more strain than I
can swallow <0.9 wink>.

> | Further the C++ 3rd Ed. says:
> |    "It is possible to design a garbage collector to invoke the
> |     destructors for objects that have been specifically
> |     registered' with the collector.  However, there is no
> |     standard way of `registering' objects.  Note that it is
> |     always important to destroy objects in an order that ensures
> |     that the destructor for one object doesn't refer to an
> |     object that has been previously destroyed.  Such ordering
> |     isn't easily achieved by a garbage collector without help
> |     from the programmer."

[Tim]
>> Java solved that one -- for all the good it did <wink>.

[Hisao]
> Are you referring to java.lang.ref.PhantomReference of Java2?

No.

> Otherwise you wholly miss the point.

Whose point <wink>?  Java has always guaranteed to "destroy objects in an
order that ensures that the destructor for one object doesn't refer to an
object that has been previously destroyed", and indeed "without help from
the programmer".  So by "Java solved that one" I meant it satisfactorily
addressed the points made in the quote.  That doesn't mean there aren't
other points to be made, but since the quote you gave didn't make any other
points I don't feel bad about commenting on the points it did make <wink>.

> The traditional mechanism of finalization in Java serves as only a
> safety net at _best_.

Sure.  It can't be faulted for any of the points *Stroustrup* made, though.

> ...
> The idea of guardians is not so particular to Scheme.  It might
> be safe to say that Scheme is a language that happened to be
> used.  Needless to say, it is important to keep _Python_ simple,
> handy, and consistent.

Yes, except I'd rather Python be consistent with its own object semantics
than with isolated portions of C++'s; and despite their frequent happy
coexistence, "simple" for the user is not always "simple" for the
implementation.

> ...
> However, the programmer often knows the expected refcount of the
> object or the aggregation object of several objects when it
> becomes unreachable from any other (aggregation of) objects.
> Usually (though not always) he/she can pass the expected
> refcount to the guardian.

Afraid not!  refcounts in Python are often unintuitive:

D:\Python>python
Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import sys
>>> sys.getrefcount(3)
8
>>> sys.getrefcount("Suzuki")
4
>>> sys.getrefcount("Hisao Suzuki")
2
>>> sys.getrefcount(sys)
5
>>>

People use getrefcount at their own risk, and the specific values returned
are not bound to remain the same from one release to the next, or even one
run to the next (toss in a thread or two, and they can bounce around
seemingly "at random" even within a single run).  The refcounts are there
for the convenience of the implementation, not the user; so this isn't a
foundation on which to build anything at the user level (although some
people do, and some of those even live long enough to regret it <0.9 wink>).

> ...
> And it is our usual practice to make prototypes in Python first,
> if possible, to experiment on some ideas, isn't it? :-)

Definitely.  In this case, though, it's probably easier to modify the Python
source code <0.5 wink>.

curmudgeonly y'rs  - tim





From Karthik.Ganesan at p98.f112.n480.z2.fidonet.org  Wed Jun 30 13:32:03 1999
From: Karthik.Ganesan at p98.f112.n480.z2.fidonet.org (Karthik Ganesan)
Date: Wed, 30 Jun 1999 18:32:03 +0100
Subject: idle- setting breakpoints
Message-ID: <000015fd@bossar.com.pl>

From: Karthik Ganesan <k_ganesan at yahoo.com>

Hi,

To set a breakpoint under idle:
-have a "Python Shell" (the one which opens up when you fire up idle) open
-Select Debug->Debugger, a window should pop open having the GUI for the
debugger
Now you can set breakpoint in your code.

If you dont have this debug window open, you get a bell from the function
PyShellEditorWindow.set_breakpoint_here(self, event=none) as the attribute
self.flist.pyshell.interp.debugger is <None>.

It seems that the function PyShell.toggle_debugger() (called when you select
the above
menu option) gives this attribute a valid value.

-karthik




From guido at CNRI.Reston.VA.US  Wed Jun 23 11:08:32 1999
From: guido at CNRI.Reston.VA.US (Guido van Rossum)
Date: Wed, 23 Jun 1999 11:08:32 -0400
Subject: List <-> news hosed again?
Message-ID: <199906231508.LAA23048@eric.cnri.reston.va.us>

Tim Peters wrote:

> Seems to me the c.l.py mailing list -> news gateway stopped cold
> again late last week.  Less sure of the other direction, but I do
> believe I've seen some older recent articles on DejaNews that
> haven't come thru on the list.

> If you're reading this and are not on the mailing list, share the
> secret <wink>.

We're looking into this; uunet changed their configuration and we
neglected to make a matching change.  It should be fixed now
(presuming this message shows up on the uunet side of things).

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



From fw at cygnus.stuttgart.netsurf.de  Sun Jun  6 15:14:49 1999
From: fw at cygnus.stuttgart.netsurf.de (Florian Weimer)
Date: 06 Jun 1999 21:14:49 +0200
Subject: Using CVS under Win95
References: <yoy63.1157$VL2.171084@news.direcpc.com>
Message-ID: <m3u2slcf9y.fsf@deneb.cygnus.stuttgart.netsurf.de>

"Emile van Sebille" <emile at fenx.com> writes:

> D>cvs -d :pserver:pythoncvs at cvs.python.org:/projects/cvsroot login
> (Logging in to pythoncvs at cvs.python.org)
> CVS.EXE [login aborted]: could not find out home directory

Did you set your HOME environment variable?

>From the CVS documentation:

| `$HOME'
| `$HOMEPATH'
| `$HOMEDRIVE'
|      Used to locate the directory where the `.cvsrc' file, and other
|      such files, are searched.  On Unix, CVS just checks for HOME.  On
|      Windows NT, the system will set HOMEDRIVE, for example to `d:' and
|      HOMEPATH, for example to `\joe'.  On Windows 95, you'll probably
|      need to set HOMEDRIVE and HOMEPATH yourself.

(I don't use CVS on Windows, so I don't know whether this really works.)



From graham at sloth.math.uga.edu  Sun Jun  6 20:46:14 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 7 Jun 1999 00:46:14 GMT
Subject: "The Python Way"
References: <000d01beaef4$ab3b5c40$f19e2299@tim>
Message-ID: <7jf4om$bbo$1@cronkite.cc.uga.edu>

[Patrick Phalen, misattributed to Fredrik Lundh]
> Would both Guido and TIm Peters be willing to collaborate on a short
> paper -- call it "The Python Way" for lack of a better title -- which
> sets out the 10-20 prescriptives they might offer to those who come to
> Python from other languages and immediately want to find a way to bend
> it into uncomfortable positions -- (implement closures, etc.).
[Graham Matthews]
> Does anyone else consider that the above sounds slightly paranoid?
Tim Peters (tim_one at email.msn.com) wrote:
: Less so than your response <0.6 wink>.  

I don't think so Tim. I think my post was pretty fair actually. I think
a fair evaluation of this group is that critiques of Python (and that
includes suggestions for improvements) are met pretty hostily (if there
is such a word). I think Python misses out on a lot of good ideas as
a result, and that's a shame. The whole notion of "the Python way" is
pretty much anathema to the notion of open discussion. 

Tim Peters (tim_one at email.msn.com) wrote:
: I saw no hint that Patrick was
: referring to you or your current agenda.  In context, he was clearly asking
: for the list of "Pythonic Principles" from which Guido supposedly deduces--
: with exacting rigor <choke> --what goes in and what stays out.  That's all.

I think you are showing your spots here Tim. My post clearly cited used
my contributions to the GC thread *as an example* of the hostility that
news (or old, as in this GC case) are treated. It clearly did *not*
say that I thought Patrick's comments (badly attributed to Fredrik)
were directed at me.

graham
-- 
                    One day in a nuclear age 
                  They may understand our rage
              We build machines that we can't control
              And bury the waste in a great big hole



From hniksic at srce.hr  Mon Jun  7 14:48:15 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 07 Jun 1999 20:48:15 +0200
Subject: fork()
References: <375BE64A.EB997F24@icrf.icnet.uk> <87so840zcc.fsf@pc-hrvoje.srce.hr> <375BF721.4B1864F2@icrf.icnet.uk>
Message-ID: <87u2sjzw28.fsf@pc-hrvoje.srce.hr>

Arne Mueller <a.mueller at icrf.icnet.uk> writes:

> > If you don't change the dictionary, the memory will not be copied.
[...]
> Huh, that's realy good news! But I don't understand how that works,

It works by MMU (memory management unit) magic.  The OS doesn't really 
"watch" the system activity -- it's more that when the child writes to 
one of the pages in question, a trap is invoked which copies the page
and remaps the child's memory address to point to the copy.

> However my children don't change anything in the dictionary ;-)

Then the memory should be kept.  You can test this by running a script 
like this:

    import os, time

    largestring = "x" * 10000000

    for i in range(50):
      if os.fork() != 0:
        time.sleep(60)
        os._exit(0)

    time.sleep(60)

This creates fifty processes, each of which sleep a minute and exit,
and their parent does the same.  Each process can access a 10M string.
Needless to say, I don't have 500M virtual memory on my Linux box, so
copy-on-write definitely works here.  Also, `free' reports:

{pc-hrvoje}[~]$ free
             total       used       free     shared    buffers     cached
Mem:         63272      62572        700     534736         88       9844
-/+ buffers/cache:      52640      10632
Swap:       120924      22112      98812

Of 534M memory shared between the processes (under the "shared"
column), about 500 comes from the 50 python processes.



From greg.ewing at compaq.com  Mon Jun 28 23:16:51 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Tue, 29 Jun 1999 15:16:51 +1200
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <37682DD3.43923243@compaq.com> <7ka2be$qja$1@nntp6.u.washington.edu> <37776FFA.2D85@mailserver.hursley.ibm.com>
Message-ID: <37783AA3.CC63664C@compaq.com>

Paul Duffin wrote:
> 
> Actually you find that you can do everything that YOU want to
> in Python and YOU find it easier than doing it in Tcl. I do not
> believe that you can do everything in Python that you can do
> in Tcl (at least as regards extending the language itself).

I meant that Python addresses all the major areas
of application that Tcl's designer claims Tcl was
designed for.

I don't regard extending Tcl as an "application"
of Tcl -- it's just a means to an end. Python
achieves many of the same ends as Tcl using 
different means.

Greg



From MHammond at skippinet.com.au  Tue Jun 22 10:07:24 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Wed, 23 Jun 1999 00:07:24 +1000
Subject: IDLE 0.5 Function parameter display?
References: <7kmpnu$red$1@nnrp1.deja.com>
Message-ID: <7ko57c$cfk$1@m2.c2.telstra-mm.net.au>

The best way at the moment is to add a docstring to the function in the C
code.  If you look at the help provided for (eg) string.split, it is quite
effective even though it is only from the docstring.

Mark.

Sam Schulenburg wrote in message <7kmpnu$red$1 at nnrp1.deja.com>...
>I love the ability of IDLE 0.5 to display the function parameter
>requirements as defined in the original function definition. My problem
>is when I have a function defined in a windows dll written in 'C', then






From jan at q-bus.de  Tue Jun  8 06:59:35 1999
From: jan at q-bus.de (Jan Walter)
Date: Tue, 8 Jun 1999 12:59:35 +0200 
Subject: I'll pay $40 in e-gold if you'll fix this bug.
References: <7jigrv$ppo$1@xs3.xs4all.nl>
Message-ID: <19990608120140.11DF2310440@bireme.oratrix.nl>

Hi,

Maybe you get confused a bit by the bash!!!

Try following:
bash$ tcsh
tcsh: cd /tmp/
tcsh: mkdir dirOne
tcsh: mkdir dirTwo
tcsh: mkdir dirTwo/dirThree
tcsh: ln -s dirTwo/dirThree/ .
tcsh: cd dirThree/
tcsh: pwd
/tmp/dirTwo/dirThree

Do the same with the bash (removing the directories of course):
bash$ cd /tmp/
bash$ mkdir dirOne
bash$ mkdir dirTwo
bash$ mkdir dirTwo/dirThree
bash$ ln -s dirTwo/dirThree/ .
bash$ cd dirThree/
bash$ pwd
/tmp/dirThree

The tcsh gives the right answer. You are in the /tmp/dirTwo/dirThree directory. If you try
to move a file with "mv file.txt ../dirOne/file.txt" you can?t do that because there is no
directory dirOne !!!

Python works like tcsh. If you try the following you get the right answer:
bash$ cd /tmp/dirThree/
bash$ python
Python 1.5.1 (#1, May  6 1998, 01:48:27)  [GCC 2.7.2.3] on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import os
>>> os.getcwd()
'/tmp/dirTwo/dirThree'

My suggestion:

You can test if there are symbolic links in your path (e.g. os.path.islink("/tmp/dirThree")
within Python) or you can prove that /tmp/dirThree is in reality /tmp/dirTwo/dirThree and
life with that fact ...

Cheers,

Jan






From bwarsaw at cnri.reston.va.us  Thu Jun  3 13:08:52 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Thu, 3 Jun 1999 13:08:52 -0400 (EDT)
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7imr6b$1s0$1@nnrp1.deja.com>
	<87pv3iq6ss.fsf@ev.netlab.co.jp>
	<3752C76D.604CD474@prescod.net>
	<87n1ykr0z9.fsf@ev.netlab.co.jp>
	<375372F6.A874554@prescod.net>
	<87hfosquc0.fsf@ev.netlab.co.jp>
	<3753F121.E4A9BE8B@prescod.net>
	<87yai3ntdt.fsf@ev.netlab.co.jp>
	<7j3iif$134e$1@nntp6.u.washington.edu>
	<7j45iv$908$4@cronkite.cc.uga.edu>
	<7j4hhj$118u$1@nntp6.u.washington.edu>
Message-ID: <14166.46756.866947.64007@anthem.cnri.reston.va.us>

>>>>> "DC" == Donn Cave <donn at u.washington.edu> writes:

    DC> Does JPython use reference counts, and invoke delete methods
    DC> as soon as reference count drops to 0?

JPython used to call __del__() when the Java object (PyInstance) was
finalized.  See my previous message on the subject for why this has
been commented out.  I don't believe __del__() is ever invoked now in
JPython.

-Barry



From peterb at arch.sel.sony.com  Tue Jun 22 05:10:42 1999
From: peterb at arch.sel.sony.com (Peter)
Date: Tue, 22 Jun 1999 10:10:42 +0100
Subject: Drag and dropable scripts on NT
Message-ID: <376F5312.C404B9E0@arch.sel.sony.com>

I'm interested in packaging up some Python code such that I can drop a
file
onto the script icon and have it do it's thing...

I have the script working fine. Don't know where to turn to figure out
the
next step.

Thanks -





From python-list at teleo.net  Wed Jun 23 18:52:37 1999
From: python-list at teleo.net (Patrick Phalen)
Date: Wed, 23 Jun 1999 15:52:37 -0700
Subject: Thanks for Python...
References: <slrn7n2gkr.i3k.wtanksle@dolphin.openprojects.net>
Message-ID: <99062316154001.00558@quadra.teleo.net>

Jeff Rush wrote:

> ...  Now if I could just get time to port
>  it to the PalmPilot, I'd never be away from it.

Sounds like a job for Itsy, if Compaq ever decides to release it as a
consumer product.

http://www.research.digital.com/wrl/itsy/

Runs under Linux with GNU tools on the StrongARM SA-1100 with 16MB of
DRAM (expandable), onboard USB, etc.

They're already fooling around Java and Squeak versions.

In fact, it appears generally rather Pythonic in design. :)



From rozzin at geekspace.com  Tue Jun 29 08:57:43 1999
From: rozzin at geekspace.com (Joshua Rosen)
Date: Tue, 29 Jun 1999 08:57:43 -0400
Subject: Typing [Re: total idiot question: +=, .=, etc...]
References: <000d01bebeda$1b7430a0$d29e2299@tim> <slrn7n8qiv.kl.behrends@allegro.cse.msu.edu>
Message-ID: <3778C2C7.3B101FD0@geekspace.com>

Reimer Behrends wrote:
>         Require variables to be prefixed with "$" to denote scalar
>         values--strings and integers, "@" for lists, and "%" for
>         dictionaries.
> 
> Given the precedent of having to use self to access instance variables,
> this is entirely consistent with the current language, and is going to
> eliminate all those nasty problems where you can't tell from the context
> what the type of a variable or function is. :)

	+There is no difficulty in getting the program to know what type some piece of
data is, which lessens the need for the type-declaring prefixes--use type(),
and then check obj.__class__ if obj is a class-instance, and you need to -know-
what class it is (if you want to -force- it to be a certain type, keep
reading).
	+Requiring variables to bear a prefix denoting type sortof makes Python into a
statically-typed language, where one of the big reasons for using it is that
it's a dynamically-typed language.
	+There is no such thing as a `scalare', in Python--strings and numbers are
different data-types, numeric data can actually be one of many different
classes.
	+Forcing data to be of a specific type is already possible--see documentation
on int(), str(), list(), tuple(), float(), long(), complex(), string.atoi(), et
al. This functionality also lessens, if not completely eliminates the need for
typing prefixes; the program knows what type the data is being coered into, and
so does the programmer. In fact, all of the functionality that you're proposing
(and then some) is already there--it just looks different.
	+If you really want a feature that -really- isn't offered by a language, the
best decision may be to just use another language that -does- offer it. Try
Pike, for one (I suggest it because it has nifty typing which you may like: you
can declare a variable to be one or -more- types, or use a mixed type). Choose
your languages based upon what you need them to do.


		-Rozzin.




From none at microsoft.com  Thu Jun 10 00:03:01 1999
From: none at microsoft.com (plz_reply_to_group)
Date: Wed, 09 Jun 1999 21:03:01 -0700
Subject: How do I match . . . Thank you both very much!
References: <375F21B3.D23A5EA9@west.boeing.com>
Message-ID: <375F38F5.F610E07C@microsoft.com>

Dang, those are both good ideas! Now that I have taken a little break
from what i was doing, I can see that I was not thinking too clearly
about what I was trying to do. I was thinking in terms of "regular
expressions", not strings! I was almost going to resort to reversing the
string, then matching the '/', clipping off the file name, then
reversing it again, or using a while loop to match each occurance. Both
seemed inelegant. 
 
Thanks!

Jeff Polaski



"Jeff P." wrote:
> 
> I've been messing with this for a little while now, but I can't get this
> to work. How do I match the last occurence of '\' in a string? I want to
> remove the file name from a url (e.g. news/breakingnews/981001b.html -->
> news/breakingnews/ ).
> 
> Here is the simplest version (that doesn't work). It matches the _first_
> backslash.
> 
> import regex
> 
> file_in = open( 'dir.txt', 'r+' )
> for line in file_in.readlines():
>         print regex.search( '/', line )
> 
> Thanks for any help!
> 
> Jeff P.




From mikael at pobox.com  Fri Jun 25 18:21:06 1999
From: mikael at pobox.com (Mikael Lyngvig)
Date: Fri, 25 Jun 1999 22:21:06 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773c3a0.102825445@news.oh.verio.com>
Message-ID: <59EAFD6B662F0706.86BFF3E906CBF8DE.784D2A219F85BC21@lp.airnews.net>

On Fri, 25 Jun 1999 18:05:13 GMT, morse at harborcom.net (Kevin
Dahlhausen) wrote:

>I agree with your observations on TK.  I looked at wxPython and some other
>GUIs.  I liked the Fast Light Toolkit API the best.  At least Python allows us
>to use the best GUI, with 'best' being defined by people according to their own
>preferences.

Could you please post an URL?  I think this discussion is moving
toward "Which GUI is the best for Python?", since at least one reader
has .  

If everybody throws their favorite Python GUI on the table, we can
(eventually) make a feature/performance comparison chart and,
hopefully, agree about which one should be The Python GUI.

I don't have any strong preferences, except the wxPython GUI seems to
perform well.  But, I haven't evaluated others, so perhaps the right
one is hiding out there somewhere :)


-- Mikael




From iking at killthewabbit.org  Sat Jun  5 16:23:20 1999
From: iking at killthewabbit.org (Ian King)
Date: Sat, 5 Jun 1999 13:23:20 -0700
Subject: Sockets and lemon curry
Message-ID: <qDf63.375$os.20566@typ42b.nn.bcandid.com>

No, no, I don't know where the lemon curry fits in... with a melon?

I'm trying to implement an rshd server in Python.  The problem seems to be
with establishing the secondary socket for stderr.  I'm using a Linux 2.0
box as client, and a Windows machine as server (ultimately, this is to
support rmt access to the Windows box's tape drive).  In experimenting, I've
gotten things to work with utilities that don't require the second socket
(e.g. rcp).  But the moment I try to establish the secondary socket, the
client says, "protocol violation" and gives up.  Ideas?

In more detail, here's what I'm doing (without error handling, etc.):

s = socket(AF_INET, SOCK_STREAM)
s.bind('', 514)  # RSH well-known port
s.listen(1)
conn, addr = s.listen()    # blocks for client request
response = conn.recv(1024)    # this response has the secondary port number
...do some string stuff to get the port number into integer type as 'port2'
s2 = socket(AF_INET, SOCK_STREAM)
s2.connect(addr[0], port2)    #addr tuple is host, port for conn

...and that's where the client says, "Pining for the fjords???" and gives
up.  I've done network tracing and can't find anything odd going on.  This
seems like a simple handshake, and if I use (for instance) the Windows
machine as a client against the Linux server (and do a network trace),
everything works and looks great.

Any ideas?  TIA - Ian

Ian King
iking at killthewabbit.org





From bernhard at alpha1.csd.uwm.edu  Wed Jun 23 00:20:10 1999
From: bernhard at alpha1.csd.uwm.edu (Bernhard Reiter)
Date: 23 Jun 1999 04:20:10 GMT
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org>
Message-ID: <slrn7n0o3q.toe.bernhard@alpha1.csd.uwm.edu>

On 22 Jun 1999 21:54:42 -0500, Preston Landers <planders at mail.utexas.edu> wrote:
>Can someone *please* give me a good reason why Python doesn't yet have 
>+=, -=, and .= and other such operators?  I know it's just syntactic
>sugar, but damn, it is such sweet sugar!

>From what I understand, the python creators consider it unnecessary
syntactic sugar. In fact they want one good way of writing expressions.
The lack of these operators makes the source code more readable.

+= just saves some space and typing time.
Your editor (given it is powerful enough can easily) immitate
that piece of sugar for your and make it personal syntactic honey!

Here is my posted vim example:
|>> For vim, the following does the trick (without the special word matching
|>> caps new vim versions provide. the ":noh is for people using hlsearch in
|>> vim 5.x version.)
|
|:map! += ^[:s/\([a-zA-Z_][a-zA-Z0-9_]*\)$/\1=\1+/^M:noh^MA
|
|Where   ":" means go into vim Commandmode
|        "^[" means Esc, you have to enter it pressing Ctrl-v and then Esc
|        "^M" means "Return', you have to enter pressing Ctrl-v and then Return


>Also, if this is in 1.5.2+, please forgive me, as I'm not up to speed
>on the new versions yet.  
It's not in there, I guess.

	Bernhard

-- 
Research Assistant, Geog Dept UM-Milwaukee, USA.         (www.uwm.edu/~bernhard)
Funding/Promoting Association for a Free Informational Infrastructure (ffii.org)



From paul at prescod.net  Tue Jun  1 10:21:15 1999
From: paul at prescod.net (Paul Prescod)
Date: Tue, 1 Jun 1999 14:21:15 GMT
Subject: Python 2.0
References: <11A17AA2B9EAD111BCEA00A0C9B4179301E10ED1@forest.origin.ea.com>  <374F1F6D.F3338F56@prescod.net>
			<374F5001.1C850DC@Lugoj.Com>
			<374F4591.3E66FB0@prescod.net>
			<zTS33.1171$nn.335894@news.shore.net>
			<374FF31A.6E990179@prescod.net>
			<37512edf.3024969@news.demon.co.uk> <3752ddac.3812702@nntp.ix.netcom.com>
Message-ID: <3753EC5B.4C70860@prescod.net>

Dennis Lee Bieber wrote:
> 
>         Ever seen the Amiga variant of REXX? AREXX had a very simple
> interface to compatible applications (compatible meaning that the
> application creates an "AREXX message port"). In (A)REXX, a statement
> line that is not recognized as part of the language itself is sent to
> the current "command interpreter" -- default is the command shell, but
> it can just as easily be an application (ADDRESS
> <application-port-name>).

That's all cool, but on Windows you get to choose your scripting language
which, for Python users, is even cooler. Still, I take your point that
AREXX was years ahead of Windows. And AREXX was ripped off of mainframe
REXX which was, of course, even older.

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for only himself
 http://itrc.uwaterloo.ca/~papresco

"Silence," wrote Melville, "is the only Voice of God." The assertion,
like its subject, cuts both ways, negating and affirming, implying both
absence and presence, offering us a choice; it's a line that the Society
of American Atheists could put on its letterhead and the Society of
Friends could silently endorse while waiting to be moved by the spirit
to speak. - Listening for Silence by Mark Slouka, Apr. 1999, Harper's



From avv at quasar.ipa.nw.ru  Sun Jun 20 01:58:19 1999
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Sat, 19 Jun 1999 22:58:19 -0700
Subject: Curses for MSDOS
References: <376BD33B.5A6F54BB@erols.com> <e8rgk7.hj81.ln@solling.seismic.de>
Message-ID: <376C82FB.4B7F4350@quasar.ipa.nw.ru>

Hi,

Klaus-Juergen Wolf wrote:

> Swartz <swartz at erols.com> wrote:
>  > Is there anything like curses for MS-DOS?  Thanx.
>
> I know of PC-Curses. Written in the early 80s. Maybe someone has
> extended it, but in the state I had seen it had no NCurses extensions
> and wouldn't work with Python.
>
> Probably very few people used it, because that screen access mechanism
> was relatively slow on MS-DOS.

I learned that the emulation of curses in SLang
(ftp://space.mit.edu/pub/davis/slang/) is sometimes even better than the
original curses, ncurses, etc. I may be valid also for MS-DOS. If you can
build the SLang library for MS-DOS and link your curses-demanding
application with it you may have solved your problems.

You may want to take a look at the Python wrapper for SLang:
ftp://starship.python.net/pub/crew/avv/pyslang/pyslang-0.2.0.tgz

I haven't yet tried it for MS-DOS (there were no demand, especially from
myself :-), but it may compile out of box.

For MS-DOS with djgpp everybody is welcom to try my very very alpha
wrapper around TurboVision, please take a look at the binary
"distribution" (just try and let me know if that makes sense for you)
ftp://starship.python.net/pub/crew/avv/tvpython

Regards

Alexander





From MHammond at skippinet.com.au  Sun Jun 27 18:04:22 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Mon, 28 Jun 1999 08:04:22 +1000
Subject: Pythonwin and Office 2000
References: <377648A7.9B9BE59F@prescod.net>
Message-ID: <7l672r$5k0$1@m2.c2.telstra-mm.net.au>

Paul Prescod wrote in message <377648A7.9B9BE59F at prescod.net>...
>I believe that Pythonwin is not compatible with Office 2000. Here are some
>symptoms:

I havent tried this, although I believe it!

>  File "testMSOffice.py", line 46, in TestWord
>    if gencache.EnsureModule("{00020905-0000-0000-C000-000000000046}",
>1033, 8,
>0) is None:

This is expected (unfortunately).  Word will have a new typelibrary, and
hence a new IID for the library.  The "fix" will be to work out what the
library ID of the new one is (by using "makepy -i") and changing TestWord.py
to gracefully allow either version.

However, this is a PITA in general, and a less-clunky solution would be
great!

>>>> a=win32com.client.Dispatch("MSWord.Application")
...
>com_error: (-2147221005, 'Invalid class string', None, None)

That is a strange one :-(

>    def AddOLEObject(self, Left=0.0, Top=0.0, Width=-1.0, Height=-1.0,
>ClassName
>=L'', FileName=L'', DisplayAsIcon=0, IconFileName=L'', IconIndex=0,
>IconLabel=L'
>', Link=0):
>
>Is this "L" syntax a future syntax for Unicode that Python 1.5.2 doesn't
>support? Perhaps PythonWin should test for support for that syntax before
>it generates the Python code?

Eeeek.  I have no idea where that is coming from.  I havent added special
Unicode features anywhere.  This _is_ a strange one!

I hope to install Office2000 and IE5 soon, but until then Im not sure I can
do much :-(

Mark.





From MHammond at skippinet.com.au  Sun Jun 13 18:10:31 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Mon, 14 Jun 1999 08:10:31 +1000
Subject: do anonymous pipes normally work on NT?
References: <7juatc$fef$1@nnrp1.deja.com> <7juq84$df3$1@m2.c2.telstra-mm.net.au> <7k07tu$veb$1@nnrp1.deja.com>
Message-ID: <7k1a5f$krj$1@m2.c2.telstra-mm.net.au>

Georg Mischler wrote in message <7k07tu$veb$1 at nnrp1.deja.com>...

>> Second advantage is that you deal with > "PyHANDLE" objects,
>> and explicit closing is not necessary (a-la files)
>
>I seem to remember a post by someone (you?) who stated
>that *not* closing Windows handles on could lead to serious
>resource leaks. Am I confusing something here or are
>not all handles created equal?

Well, not all handles are created equal anyway, but the point I was making
that PyHANDLE object automatically close the handle when it dies - like a
file object.  So the handles will be closed even if you do not explicitely
close them.

My comment wasnt intended to throw light on the specific problem - just
point out an advantage that PyHANDLE objects have.  OTOH, I am wildly
guessing that using CreateFile could solve the problem - not due to the
"auto-close" behaviour, but simply because they remove the indirection.  I
have no basis for believing it will solve the problem, just a guess!

Mark.






From wtanksle at dolphin.openprojects.net  Wed Jun 23 16:24:59 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Wed, 23 Jun 1999 20:24:59 GMT
Subject: Thanks for Python...
References: <376ead64@anonymous.newsfeeds.com> <VrDQRddcbbs9-pn2-QdftOXYk6Hdd@Fortress> <37716901.43CB604@lockstar.com>
Message-ID: <slrn7n2gkr.i3k.wtanksle@dolphin.openprojects.net>

On 23 Jun 1999 13:06:49 PDT, Mordy Ovits wrote:
>Jeff Rush wrote:

>> Welcome to the Python community, John.  Yes, Python is a truly
>> wonderful language for cranking out quick prototypes or test jigs.
>> I prototyped a complete web-based general ledger system in Python,
>> figuring I'd downcode it to C for speed later, but so far I haven't needed
>> to.  Python is also just plain fun...  Now if I could just get time to port
>> it to the PalmPilot, I'd never be away from it.

>Get a Windows CE Palm-sized PC if you want to take Python with you whereever you
>go.  Python runs stunningly well on my Casio E-100, and has nifty GUI
>libraries.  It's not polished by any sense of the word but it works, and the
>source is a barely changed version of the standard windows Python.  This makes
>for easy improvement.

Sure, but we've got Pilots, so that's not an open option.  Fortunately,
the Pilot has Quartus Forth available for it -- it's not Python, but
honestly, Forth fits the limitations of the device better.

I used Quartus on my Crypto exam to solve a modular equation after I
forgot how to solve modular equations (I had it use brute force).

I'd still like to have Python, but dang, Quartus is good at what it does.
Python and Forth -- for two different reasons, pretty much my favorite
languages.

>o Mordy Ovits
>o Cryptographic Engineer
>o LockStar Inc.

>#!/usr/local/bin/python 
>from sys import*;from string import*;a=argv;[s,p,q]=filter(lambda x:x[:1]!=
>'-',a);d='-d'in a;e,n=atol(p,16),atol(q,16);l=(len(q)+1)/2;o,inb=l-d,l-1+d
>while s:s=stdin.read(inb);s and map(stdout.write,map(lambda i,b=pow(reduce(
>lambda x,y:(x<<8L)+y,map(ord,s)),e,n):chr(b>>8*i&255),range(o-1,-1,-1)))

You know, the first time I saw this it was in ESR's .sig, and I tried to
remember how long the Perl version had been.  A few minutes of searching
revealed that it was three lines long -- but also revealed that it was
using bc rather than Perl to do most of the work!  I spent about half an
hour searching for the real Perl version, only to find that it was five
lines long.

The Python for this .sig is shorter than the Perl!  Cool, eh?

-- 
-William "Billy" Tanksley
Utinam logica falsa tuam philosophiam totam suffodiant!
   :-: May faulty logic undermine your entire philosophy!



From gvinther at atex.com  Mon Jun  7 16:44:37 1999
From: gvinther at atex.com (Gordon Vinther)
Date: Mon, 7 Jun 1999 16:44:37 -0400
Subject: problems using Tkinter
References: <928350546.726.80@news.remarQ.com> <02fc01beadf0$b3c3ea30$f29b12c2@pythonware.com>
Message-ID: <928788254.441.84@news.remarQ.com>

Fredrik Lundh <fredrik at pythonware.com> wrote in message
news:02fc01beadf0$b3c3ea30$f29b12c2 at pythonware.com...
> Gordon Vinther <gvinther at atex.com> wrote:
> > I get the following error whenever I attempt to use Tk
> >
> > Os AIX4.2.1
> > Python 1.5.2
> > Tk/TCL 8.0.1
> >
> > from Tkinter import  *  .............this works
> > root=Tk() ........... this errors with the following message
> >
> > File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 886, in __init__
> > self.tk = _tkinter.create(screenName, baseName, className)
> >
> > TclError: invalid command name "tcl_findLibrary"
>
> looks like tkinter is picking up the wrong Tcl/Tk
> library files.   try setting the TCL_LIBRARY and
> TK_LIBRARY environment variables.
>
> </F>
>
I tried the environment variables with no help.  Doesn't this look like a
Tcl/tk problem?  Has anyone run Python with the new version of  Tcl/Tk8.nn??
This is completely new code.  Mayby Tkinter doesn't run with it.
It seems Tcl is complaining about a command it doesn't understand....!!!!!!
and not that it is not being communicated with.








From jeremy at cnri.reston.va.us  Wed Jun  9 11:58:37 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Wed,  9 Jun 1999 11:58:37 -0400 (EDT)
Subject: "Programming Python" still worthwhile?
In-Reply-To: <1283218560-70098357@hypernet.com>
References: <7jki66$qam$1@news.gstis.net>
	<1283218560-70098357@hypernet.com>
Message-ID: <14174.36177.272502.237392@bitdiddle.cnri.reston.va.us>

>>>>> "RR" == Rick Robino <see-sig at wavedivision.com> writes:

  RR> wondering if anyone here could give me some advice as to whether
  RR> or not "Programming Python" is useful, if the current edition is
  RR> still current, if I should wait for the "Learning Python", or if
  RR> I should use some other resources.

>>>>> "GMcM" == Gordon McMillan <gmcm at hypernet.com> writes:

  GMcM> Learning Python is on the shelves. Haven't read it. Nobody's
  GMcM> flamed it, and that's a _very_ good sign. 

I bought a copy of Learning Python in a local technical bookstore that 
doesn't have a very good selection, so it ought to be easy to find.

I haven't read the book carefully, but I did read through the chapter
on classes (Chap. 6).  I thought it was an excellent introduction to
OOP and Python.  It is much more thorough than the Python tutorial and 
other free documentation.  So I don't think you'll go wrong with
Learning Python.

Jeremy



From akuchlin at mems-exchange.org  Tue Jun 22 13:47:55 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: Tue, 22 Jun 1999 13:47:55 -0400 (EDT)
Subject: renegotiating socket connections
In-Reply-To: <XFMail.990622103149.nathan@islanddata.com>
References: <XFMail.990622103149.nathan@islanddata.com>
Message-ID: <14191.52299.153087.487734@amarok.cnri.reston.va.us>

Nathan Clegg writes:
>I'm writing a server application in python.  I understand that most
>servers fork off a process for every incoming connection and then
>renegotiate on a different port to free up the well-known server port.  I

Where does this bit of incorrect conventional wisdom come from?
netstat on my machine is currently reporting:

amarok.80            amarok.40531         32768      0 32768      0 ESTABLISHED
amarok.80            amarok.40532         32768      0 32768      0 TIME_WAIT
amarok.80            amarok.40534         32768      0 32768      0 ESTABLISHED
amarok.80            amarok.40535         32768      0 32768      0 ESTABLISHED

Several connections to port 80, and everything's fine.  Now, you might 
have an architecture where a single gatekeeper process hands things
off to another process, and therefore open a new socket to this
second process, but for a simple forking server, I don't see why 
this port renegotiation would be required.  (Even in this case, you
could use file descriptor passing to avoid the second open, though I
don't know if you can pass file descriptors from Python code easily.)

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
My nose shall never be touched while heaven give me strength.
    -- Sterne, _Tristram Shandy_




From arcege at shore.net  Wed Jun  2 08:15:43 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 02 Jun 1999 12:15:43 GMT
Subject: A tkinter kick-off
References: <37544a32.0@news1.mcmail.com>
Message-ID: <Pf953.1593$nn.478377@news.shore.net>

Mark Butterworth <mrkbutty at mcmail.com> wrote:
: I've been using python for a few months now (I've actually managed to
: persuade my company that freeware is a viable tool) and I've just started
: looking at tkinter.  However,  my first little project doesn't  require an
: interactive window or dialog box.  What I do require is a window which
: displays the result's of my background processing.  The processing will
: simply be the compression and network transfer of some files.  I'm not
: looking for a full solution however, I just need a help starting off.  I can
: create a window containing a listbox alright, I just need to know how to
: process in the background and display results within the listbox (maybe a
: cancel button would also be useful).

Most windowing applications over the last ten to fifteen years have been
based on the event loop, and Tkinter is not exception.  But (of course
there is a but), Tkinter also has the nice feature of including various
other control mechanisms.

In this case I think you would want to be using the update_idletasks()
or update() methods.

  root = Tk()
  done = 0
  # function to turn off computations
  def canceler():
    global done
    done = 1
  # set up your widgets using root
  cancel = Button(root, text="Cancel", command=canceler)
  cancel.pack(side=BOTTOM)
  root.update_idletasks()
  while not done:
    # perform your computations, including updating the widgets
    root.update_idletasks()

You might also want to look into the after_idle() method.  The choice
to use after_idle or update_idletasks should be based on which is going
to take more time (and which is going to block for longer), the
computations/networking or the windowing.  You might also want to look
into threads (just made sure that Tkinter is controlled in the main
thread).

FYI, Curses uses this "idletasks" method (called refresh).

  -Arcege




From klassa at aur.alcatel.com  Wed Jun  9 15:41:28 1999
From: klassa at aur.alcatel.com (John Klassa)
Date: 9 Jun 1999 19:41:28 GMT
Subject: fork()
References: <000a01beb21f$a051d380$2fa22299@tim> <7jmbe1$r22$2@cronkite.cc.uga.edu>
Message-ID: <7jmg18$1h4$1@aurwww.aur.alcatel.com>

On 9 Jun 1999 18:22:57 GMT, Graham Matthews <graham at sloth.math.uga.edu> wrote:
  
  > observable reality. Where are the hoary theoretical arguments? Or 
  > is this just more of the same old bluff and bluster arguing.
 
Uh oh...  This is starting to look like comp.lang.perl.misc. :-)

-- 
John Klassa / Alcatel USA / Raleigh, NC, USA



From jam at newimage.com  Wed Jun  2 06:17:12 1999
From: jam at newimage.com (jam)
Date: Wed, 2 Jun 1999 06:17:12 -0400
Subject: Python & IRC?
In-Reply-To: <rNHvjEdhm5Pp-pn2-tyCQ80XVscVt@d185d18f8.rochester.rr.com>; from Benjamin Schollnick on Mon, May 31, 1999 at 08:21:17PM +0000
References: <rNHvjEdhm5Pp-pn2-tyCQ80XVscVt@d185d18f8.rochester.rr.com>
Message-ID: <19990602061712.F17948@toast.internal>

On Mon, May 31, 1999 at 08:21:17PM +0000, Benjamin Schollnick wrote:
> 
> Folks,
> 
> 	Has anyone done anything in regards to making a IRC
> daemon/server in Python?
> 
> 	I'm thinking about sitting down and doing some socket
> work with Python, and I wouldn't want to reproduce the wheel again...
> 
> 			Benjamin

yes.

I have such code, and I'll be willing to send it to you presuming youare
willing to honor an LGPL license for it.. I will also be willing to answer
any questions or concerns you may have regarding how IRC works.. I've
already been through the process, and it can get a bit hairy.. actually, if
you look through dejanews, you may be able to find other posts along these
lines.. what would be *really* cool is if we pooled resouces and wrote an
irc daemon in python to compliment the irc client(s).

let me know how you'd like to proceed.

regards,
J
-- 
|| visit gfd <http://quark.newimage.com/> 
|| psa member #293 <http://www.python.org/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>



From andy at robanal.demon.co.uk  Mon Jun 21 14:03:53 1999
From: andy at robanal.demon.co.uk (Andy Robinson)
Date: Mon, 21 Jun 1999 18:03:53 GMT
Subject: IPwP books available, 99cents
References: <376a479b.131911984@news.texoma.net>
Message-ID: <376e7e32.727764@news.demon.co.uk>

jhouchin at texoma.net (Jimmie Houchin) wrote:

>Hello,
>
>I was visiting a bookstore in Plano, TX yesterday and saw two new
>copies of Internet Programming with Python on their shelves for $.99,
>yes 99 cents.
>
Well, here in the U.K. they are appearing back on the shelves in the
good bookstores after a long absence, and still at full price.  I
really hope your bookstore has misread Python's direction and not
mine!

- Andy Robinson




From akuchlin at mems-exchange.org  Thu Jun  3 17:03:50 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: Thu, 3 Jun 1999 17:03:50 -0400 (EDT)
Subject: need help with decompress
In-Reply-To: <66A53.34$a43.449@news.megsinet.net>
References: <66A53.34$a43.449@news.megsinet.net>
Message-ID: <14166.60854.252981.324857@amarok.cnri.reston.va.us>

Philip C. writes:
>using the zlib module o found out how to compress springs but I having
>trouble with decompressing them.

Decompressing is just as simple as compressing:

amarok akuchlin>python
Python 1.5.2 (#80, May 25 1999, 18:06:07)  [GCC 2.8.1] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
i>>> import zlib
>>> c = zlib.compress("""So, if this were indeed my Final Hour, ... """)
>>> c   
'x\234MQKn\3350\014\334\347\024\263\353\306\3601\212\006...'
>>> print zlib.decompress(c)
So, if this were indeed my Final Hour, these would be my words to you.
   I would not claim to pass on any secret of life, for there is none, or
   any wisdom except the passionate plea of caring ... Try to feel, in
   your heart's core, the reality of others. This is the most painful
   thing in the world, probably, and the most necessary. In times of
   personal adversity, know that you are not alone. Know that although in
   the eternal scheme of things you are small, you are also unique and
   irreplaceable, as are all of your fellow humans everywhere in the 
   world. Know that your commitment is above all to life itself.

>>> 

>is there any way to scramble string sort of encrypt them

    Python includes the "rotor" module, which does some simple but not
necessarily secure encryption.  mxCrypto
(http://starship.python.net/crew/lemburg/mxCrypto.html) contains more
industrial-strength encryption for Python.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/




From robin at alldunn.com  Tue Jun 29 03:41:34 1999
From: robin at alldunn.com (Robin Dunn)
Date: Tue, 29 Jun 1999 00:41:34 -0700
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <Pine.GSO.3.96.990625170056.28322B-100000@saga3.Stanford.EDU>
Message-ID: <1N_d3.30$oY.3604744@news.randori.com>

Jeffrey Chang wrote in message ...
>
>Tkinter is very good for that type of thing.  All I needed to do was
>create a Canvas and then for each node, create an Oval with a callback to
>a function that handles node clicks.  Tk keeps track of the coordinates of
>each Item so that I don't have to.  Also, whenever I wanted to move the
>tree, all I had to do was:
>for widget in self._canvas.find_all():
>    self._canvas.move(widget, xoff, yoff)
>

This will shortly be very easy to do with wxPython as well.  If all goes
well the next release will include the wxOGL library (Object Graphics
Library) which handles drawing various shapes and such on a canvas,
including event notification for those shapes, dragging, etc.

--
Robin Dunn
robin at AllDunn.com
http://AllDunn.com/robin/
http://AllDunn.com/wxPython/  Check it out!
Try http://AllDunn.com/laughworks/ for a good laugh.






From cjw at connection.com  Wed Jun 30 15:56:16 1999
From: cjw at connection.com (Colin J. Williams)
Date: Wed, 30 Jun 1999 15:56:16 -0400
Subject: A Python problem?
Message-ID: <377A765F.90367AC5@connection.com>

This script below illustrates what appears to be a problem with the
identification of a module in a package.

The example is borrowed from the linbot material.

Colin W.
''' A Python problem?   .

  '''
def main():
  import sys
  print 'Start:',dir()
  s0= sys.modules
  import plugins.rptlib
  s1= sys.modules
  if s0 == s1:
    print 'sys.modules is unchanged'
  print 'After attempt to import plugins.rptlib:', dir()
  try:
    print dir(plugins.rptlib)
  except AttributeError:
    import sys
    print sys.exc_info()
  from plugins.rptlib import *
  print 'rptlib names now available:', dir()
main()

'''
------------------------ Resulting Display ----------------------------
>>> Start: ['sys']
sys.modules is unchanged
After attempt to import plugins.rptlib: ['plugins', 's0', 's1', 'sys']
(<class exceptions.AttributeError at 76bbc0>, <exceptions.AttributeError
instance at daa0e0>, <traceback object at dab2b0>)
rptlib names now available: ['Link', 'add_problem', 'check_and_warn', 'config',
'debugio', 'doBotMain', 'doTopMain', 'get_title', 'linbot', 'linkList',
'main_index', 'make_link', 'nav_bar', 'open_file', 'opener', 'os', 'plugins',
'problem_db', 'proxies', 'read_registry', 's0', 's1', 'sort_by_age',
'sort_by_author', 'sort_by_rev_age', 'sort_by_size', 'string', 'stylesheet',
'sys', 'urllib', 'version']

  '''




From tim_one at email.msn.com  Tue Jun 15 11:37:43 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 15 Jun 1999 11:37:43 -0400
Subject: OO (was: Why is tcl broken?)
In-Reply-To: <37660FEC.D4F3371B@iname.com>
Message-ID: <000801beb745$02768220$979e2299@tim>

[William Tanksley]
> (I like all the languages I know, except Fortran.  I even like C++,
> although I wish they could break backwards compatibility.)

[Fernando Mato Mira]
> Quiz: What do Scheme, Common Lisp, HPF and Java have in common?

Guy Steele's loving ministrations -- but even he gave up on Scheme <wink>.





From jam at newimage.com  Mon Jun 28 08:46:36 1999
From: jam at newimage.com (jam)
Date: Mon, 28 Jun 1999 08:46:36 -0400
Subject: Extracting Text file contents using Python
In-Reply-To: <37782f9a.0@news.smartnet.com.sg>; from john on Mon, Jun 28, 1999 at 07:37:50PM +0800
References: <37782f9a.0@news.smartnet.com.sg>
Message-ID: <19990628084636.B16195@toast.internal>

On Mon, Jun 28, 1999 at 07:37:50PM +0800, john wrote:
> 
> Hi,
> 
> I am trying to extract the following text file using python and the program
> & text file are as shown below.
> 
> But unfortunately I am unable to get every separate data displayed except
> for the first data and after the separation line which is
> "====================" the next data is not getting picked up by the
> program.
> Could anyone kindly let me know what could be wrong with the program.
> 
> Example Text File - But will be similar
> ----------------------------------------------------
[..snipped..]
> 
> The python program is as follows:
> 
[..snipped..]

I didn't have any trouble with this code on my linux machine. 

the one thing that might be confusing the issue is that all of your records
are identical.. one thing I did while testing your code was to change the
'rating' field to a unique number (only 8 records, so I changed each to have
a number in that sequence).. then you can see the data changing, and it's
much clearer whether or not it's working.

give that a shot, and see what happens.

regards,
J

-- 
|| visit gfd <http://quark.newimage.com/> 
|| psa member #293 <http://www.python.org/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>



From William.Tanksley at p98.f112.n480.z2.fidonet.org  Tue Jun 29 19:15:57 1999
From: William.Tanksley at p98.f112.n480.z2.fidonet.org (William Tanksley)
Date: Wed, 30 Jun 1999 00:15:57 +0100
Subject: Why exceptions shouldn't be used for flow control [Re: YAS to the
Message-ID: <0000159d@bossar.com.pl>

From: wtanksle at dolphin.openprojects.net (William Tanksley)

On Mon, 28 Jun 1999 11:17:01 +1200, Greg Ewing wrote:
>William Tanksley wrote:

>> while not f.eof():
>>   line = f.readline()
>>   # ...

>I'd love to be able to write reading loops that way.
>Unfortunately, it's a fact of life that some of the
>things read() has to deal with are unable to detect
>eof without trying to read something. I think the
>existing definition of read() and friends is the best 
>that can be done in those circumstances.

I'd guess that you're talking about things like text file parsing and
such.  I would use readline there, but that's fine.  Yes, unstructured
data is more difficult to handle.

>> It's been a while since I've used read, so I
>> don't recall what it actually returns

>It returns an empty string if it can't read more
>than 0 characters, same as readline().

Ah.  There's a justification for that, as you observed.  I don't like the
fact that readline is stuck doing the same thing (record delimiters are
not really part of the record).

>> in a function designed to get n!=0
>> more characters, an EOF really is an error.  Unfortunately for me, that's
>> not the definition of 'read', and never will be.

>Perhaps there should be a 2-parameter version of read:

>   read(min, max)

>which would raise an exception if it couldn't read at
>least min characters. Setting min=0 would give the
>current behaviour, and setting min=max would allow
>reading a fixed-length record without having to check
>for errors. Then everyone would be happy!

A facinating solution.  Well, since I don't use read, I probably wouldn't
use it, and I'm the only person whining.  Therefore, it's probably not
worth implementing (even though it is a cool solution).

>Greg

-- 
-William "Billy" Tanksley



From shallib at netscape.com  Thu Jun 10 00:22:32 1999
From: shallib at netscape.com (Steve Halliburton)
Date: Wed, 09 Jun 1999 21:22:32 -0700
Subject: How do I match the last occurence of a character?
Message-ID: <375F3D88.A3D1B408@netscape.com>

Spam detection software, running on the system "albatross.python.org", has
identified this incoming email as possible spam.  The original message
has been attached to this so you can view it (if it isn't spam) or label
similar future email.  If you have any questions, see
the administrator of that system for details.

Content preview:  I haven't used it on NT, but I know on unix it handles everything
   gracefully: >>> import os >>> t1 = 'http://tor.dhs.org/~bwinton/index.html'
   >>> t2 = 'home/bwinton/' >>> os.path.split(t1) ('http://tor.dhs.org/~bwinton',
   'index.html') >>> os.path.split(t2) ('home/bwinton', '') [...] 

Content analysis details:   (5.9 points, 5.0 required)

 pts rule name              description
---- ---------------------- --------------------------------------------------
 2.0 FH_DATE_IS_19XX        The date is not 19xx.
 0.5 RCVD_IN_PBL            RBL: Received via a relay in Spamhaus PBL
                            [205.217.237.46 listed in zen.spamhaus.org]
 1.1 RCVD_IN_SORBS_WEB      RBL: SORBS: sender is a abuseable web server
                            [205.217.237.54 listed in dnsbl.sorbs.net]
 0.1 RDNS_DYNAMIC           Delivered to trusted network by host with
                            dynamic-looking rDNS
 2.2 FAKE_REPLY_C           FAKE_REPLY_C


-------------- next part --------------
An embedded message was scrubbed...
From: Steve Halliburton <shallib at netscape.com>
Subject: Re: How do I match the last occurence of a character?
Date: Wed, 09 Jun 1999 21:22:32 -0700
Size: 2573
URL: <http://mail.python.org/pipermail/python-list/attachments/19990609/a3191618/attachment.mht>

From steffen at biomedcomp.com  Fri Jun 11 17:39:10 1999
From: steffen at biomedcomp.com (David Steffen)
Date: Fri, 11 Jun 1999 16:39:10 -0500
Subject: Can't make python1.5.2 on rhLinux5.1
Message-ID: <37618200.91CC7370@biomedcomp.com>

Howdy!

  I am trying to install Python 1.5.2 on a Red Hat Linux 5.1 / 386
(actually pentium) system and have run into a problem I can't find
mention of in the README or the faq.  I am a regular reader of this
group, but don't read every single message, so if I have missed this
one, my apologies.

  The problem occurs during make:

cd Modules ; make OPT="-g -O2" VERSION="1.5" \
                prefix="/usr/local" exec_prefix="/usr/local" all
make[1]: Entering directory `/root/Python-1.5.2/Modules'
gcc  -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H -c ./regexmodule.c
         <10 modules compiled OK>
gcc  -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H -c ./timemodule.c
./timemodule.c: In function `time_strptime':
./timemodule.c:429: warning: assignment makes pointer from integer
without a cast
         <5 modules compile OK>
gcc  -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H -c ./socketmodule.c
./socketmodule.c: In function `initsocket':
./socketmodule.c:2212: `uint32_t' undeclared (first use this function)
./socketmodule.c:2212: (Each undeclared identifier is reported only once
./socketmodule.c:2212: for each function it appears in.)
./socketmodule.c:2212: parse error before `0xe0000000U'
./socketmodule.c:2217: parse error before `0xe0000001U'
./socketmodule.c:2222: parse error before `0xe00000ffU'
make[1]: *** [socketmodule.o] Error 1
make[1]: Leaving directory `/root/Python-1.5.2/Modules'
make: *** [Modules] Error 2

  My Linux is mostly out of the box, though we have installed this and
that on it; Perl5.005, Gimp, etc.

  The Python install is plain vanilla - I didn't hand edit anything and ran
./configure without any command line switches.

  In terms of the issues I have found in the README:

1) % rpm -qa | grep ^linuxthread    <nothing returned>
2) glibc I have installed is version 2.0.7
3) crypt module commented out as per default
4) I tried:
   make clean
   ./configure
   make
   ...with the same results

Any suggestions?

THANKS!

-David Steffen-



From landrum at foreman.ac.rwth-aachen.de  Sat Jun 19 12:21:02 1999
From: landrum at foreman.ac.rwth-aachen.de (Gregory A. Landrum)
Date: 19 Jun 1999 18:21:02 +0200
Subject: A Newbie Learns (part II)
Message-ID: <sk8t02mc1.fsf@foreman.ac.rwth-aachen.de>

Continuing revelations of a learner... part II
This time with a question for the experts

---------------------------------
Matrix Multiplies

The Matrix class (part of the NumPy distribution) provides a
convenient way of dealing with matrices, so you can do stuff like:
    from Matrix import *
    M1 = Matrix([[1.,2.,3.,4.],[1.,2.,3.,4.],[1.,2.,3.,4.],[1.,2.,3.,4.]])
    M2 = Matrix([[.1,.2,.3,.4],[.1,.2,.3,.4],[.1,.2,.3,.4],[.1,.2,.3,.4]])
    M3 = M1*M2
This is considerably more convenient than using the array type
defined in the Numeric module:
    from Numeric import *
    M1 = array([[1.,2.,3.,4.],[1.,2.,3.,4.],[1.,2.,3.,4.],[1.,2.,3.,4.]])
    M2 = array([[.1,.2,.3,.4],[.1,.2,.3,.4],[.1,.2,.3,.4],[.1,.2,.3,.4]])
    M3 = matrixmultiply(M1,M2)
Unfortunately, it's also a ton slower.  This is true because the
Matrix class actually calls matrixmultiply itself.  So the program
ends up doing all of the work of matrixmultiply plus a bunch of
overhead to convert the matrices to arrays.
Some quick profiling indicates that using arrays instead of matrices
takes approximately half as long when doing 10000 4x4 matrix
multiplies.

So the lesson I learned is: Skip the convenience of using the Matrix
class if speed is at all important.
-------------------

After these explorations and poking around in the Numeric source, I am
left with a question.  In Numeric.py, the following appears:

#This is obsolete, don't use in new code
matrixmultiply = dot

If matrixmultiply is obsolete, what exactly are we supposed to use for
doing matrix multiplies with 2D arrays?  Sure, I can just call dot
myself, but that leaves the code less readable.

-greg

---------------------
Dr. Greg Landrum  (landrumSPAM at foreman.ac.rwth-aachen.de)
Institute of Inorganic Chemistry
Aachen University of Technology




From greg.ewing at compaq.com  Sun Jun 27 18:48:28 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Mon, 28 Jun 1999 10:48:28 +1200
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0rr2.2fi.behrends@allegro.cse.msu.edu> <m3aetr2yk9.fsf@atrus.jesus.cam.ac.uk> <7ks38p$l4m$1@brick.cswv.com> <m3674c3ghb.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <3776AA3C.218936D8@compaq.com>

Michael Hudson wrote:
> 
> class Derived(base.Base):
>     def __init__(self,x,y,z,**kw):
>         apply(base.Base.__init__,(self,x,y),kw)
>         print z
> 
> which I think is one of the ugliest things I've ever seen in Python
> that I couldn't think of a prettier way round. Ideas, anyone?

Yes! With the calling syntax extension that I
have proposed before, that would be written

   Base.__init__(self, x, y, **kw)

Still requires the self argument, but I think
you'll agree it's a lot better than having to
use apply!

I'd also like a way of getting rid of the self
argument in that situation as well, but this seems
to be quite a difficult problem to solve within
Python's existing execution model. I haven't
thought of a good solution yet.

Greg



From cwebster at math.tamu.edu  Fri Jun 25 20:21:25 1999
From: cwebster at math.tamu.edu (Corran Webster)
Date: 26 Jun 1999 00:21:25 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <59EAFD6B662F0706.86BFF3E906CBF8DE.784D2A219F85BC21@lp.airnews.net> <99062523473500.00769@Vegus>
Message-ID: <7l16e5$q4j$1@news.tamu.edu>

In article <99062523473500.00769 at Vegus>,
John Travers  <jtravs at debian.org> wrote:
>Has anybody tried the PyKDE/Qt package, (you don't have to have KDE), it is
>absoloutly fantastic, quick, easy, powerful, far far better than Tkinter.
>I love it!

This is missing the point - although PyKDE/Qt might be a very nice GUI, it's
not cross-platform in the same way as Tcl/Tk.  The same goes for Fltk,
gtk or what have you.  It sounds like wxWindows might be getting near
the same level of cross-platform compatibility - and it certainly seems
to be one of the groups' objectives - it's not there yet.

The point of having a standard GUI for python is that it's _standard_.
ie. that it's likely to work no matter what your platform.  At present,
Tk seems to be the only mature package with that sort of ubiquity.  And
for all it's shortcomings, Tkinter does at least get you your basic
widgets up on the screen.

The other possible approach, of course, is to have a standard GUI API
for python which can have your favourite GUI slipped in under the
covers.  I believe that people have looked into that sort of approach,
but I haven't looked recently at how things have developed.  However
because this has to be a lowest common denominator approach, you are
sure to lose your favourite bells and whistles from your favourite
GUI.

In the meantime, by all means, develop using your favourite GUI, but
the time is not right, it seems, for replacing Tk as the standard.

Corran




From fdrake at cnri.reston.va.us  Tue Jun 15 11:02:18 1999
From: fdrake at cnri.reston.va.us (Fred L. Drake)
Date: Tue, 15 Jun 1999 15:02:18 GMT
Subject: Newbie: easy filecopy?
In-Reply-To: <376531D3.7E42AC42@phoenix-edv.netzservice.de>
References: <376531D3.7E42AC42@phoenix-edv.netzservice.de>
Message-ID: <14182.27386.991483.999566@weyr.cnri.reston.va.us>

Holger Jannsen writes:
 > I couldn't believe it. I think there must be an easy copy-function
 > in python (I only want to copy files under ext2- or FAT-FileSystems).
 > Call me "4 blind mice"-Joey, but I couldn't find it. 

  Yes.  As someone has pointed out, the shutil module contains the
beast you're looking for.
  The Index of the Library Reference manual is often pretty helpful
about things like this (not the Module Index, the general Index).
Just click on the "i" icon at the top of any page of the library
reference if you're using the HTML version.  The index may be long,
but I'm told it isn't too shabby.
  (If anyone thinks additional entries are needed, send a note telling
what they are and where they should link to in the manual to
python-docs at python.org; this kind of suggestion is usually well
received by the damn fool^H^H^H^H^H^H^H^H^H^H poor bloke who receives
mail at that address.)


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From gerrit.holl at pobox.com  Mon Jun 28 04:59:21 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Mon, 28 Jun 1999 08:59:21 GMT
Subject: RuntimeError?
In-Reply-To: <000401bec10d$b5bc3a60$e19e2299@tim>; from Tim Peters on Sun, Jun 27, 1999 at 10:27:04PM -0400
References: <19990627112432.A29097@optiplex.palga.uucp> <000401bec10d$b5bc3a60$e19e2299@tim>
Message-ID: <19990628105921.A26266@optiplex.palga.uucp>

On Sun, Jun 27, 1999 at 10:27:04PM -0400, Tim Peters wrote:
> From: "Tim Peters" <tim_one at email.msn.com>
> To: "Gerrit Holl" <gerrit.holl at pobox.com>, <python-list at python.org>
> Subject: RE: RuntimeError?
> Date: 	Sun, 27 Jun 1999 22:27:04 -0400
> 
> [Gerrit Holl]
> > can someone please give me an example when a RuntimeError is raised?
> >
> > I don't understand: if none of the others are raised.
> 
> Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import sys
> >>> del sys.stdout
> >>> print "hi"
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> RuntimeError: lost sys.stdout
> >>>
> 
> Honestly, I had never seen a RuntimeError before!  I found this one by
> digging thru the source code to see when it gets raised.  There's really no
> pattern to it, and the RuntimeError docstring in Lib/exceptions.py
> ("Unspecified run-time error.") is as accurate as anything that says nothing
> can be <wink>.  It generally seems to mean something bad and unexpected
> happened, but not bad enough to shut down the interpreter (things *that* bad
> raise SystemError).
> 

Ah, I understand. But what kind of things are bad enough to raise SystemError?
core dump?

groeten,
Gerrit.

-- 
The Dutch Linuxgames homepage:	http://linuxgames.nl.linux.org
Personal homepage:		http://www.nl.linux.org/~gerrit/

Discoverb is a python program (in several languages) which tests the words you
learned by asking it. Homepage: http://www.nl.linux.org/~gerrit/discoverb/



From fandino at usa.net  Mon Jun  7 12:06:16 1999
From: fandino at usa.net (Salvador =?iso-8859-1?Q?Fandi=F1o?=)
Date: Mon, 07 Jun 1999 18:06:16 +0200
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net> <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net> <7j34l3$hbj$1@nnrp1.deja.com> <7j66b7$il5$2@cronkite.cc.uga.edu> <3756C693.2C910B29@usa.net> <7j6jnl$58f$3@cronkite.cc.uga.edu> <3757A820.CE3F195F@usa.net> <7j8obc$55v$2@cronkite.cc.uga.edu>
Message-ID: <375BEDF8.DD14C4E6@usa.net>

Graham Matthews wrote:

> The proposal I gave for adding mark sweep collection on top of reference
> counts will delete non circularly referred objects in precisely the order
> they are now deleted in current Python. That order doesn't appear to
> random for you.
You are right.

> so I don't understand the above comment? As for circularly
> referred data it will be deleted in the order determined by the collector,
> but that's better than current Python which doesn't delete it at all.
If I can't remove objects in the right order I would have to delete it
explicitly the same way I have to do it now with ref counting. Lots of
current __del__ methods will not perform properly and will raise
exceptions if called after deleting related objects.
You could use __del__ methods only for trivial task like freeing
resources. It's an incomplete solution like reference counting is.

 - Salva.



From peter.sommerfeld at gmx.de  Tue Jun 15 16:29:21 1999
From: peter.sommerfeld at gmx.de (Peter Sommerfeld)
Date: Tue, 15 Jun 1999 22:29:21 +0200
Subject: [Python-de] Python-Tutorium-1.5.2-de quasi fertig
In-Reply-To: <3766745C.B9A88BB8@darwin.in-berlin.de>
References: <37665567.6FD97DC0@darwin.in-berlin.de>
Message-ID: <v04003a00b38c63c3f842@[195.180.62.144]>

Dino Gherman wrote:

>Voila (uhmm, weiss jemand was "Odds and Ends" heissen soll,
>"Vor- und Nachteile" vielleicht?

Tips und Trix (Bin mir aber auch nicht ganz sicher). Sinngem?ss
hei?t das glaube ich Ecken und Kanten, also dass was ?brigbleibt.
Ich w?rde da einfach wortsch?pferisch t?tig werden ;)

-- Peter

PS: Wenn's um so etwas wie HTML Layout o.? geht kann ich gerne
was tun. Kann allerdings kein TeX lesen (Mac).







From thstr at serop.abb.se  Thu Jun 10 01:37:56 1999
From: thstr at serop.abb.se (Thomas S. Strinnhed)
Date: Thu, 10 Jun 1999 07:37:56 +0200
Subject: Python/C API and setattr
References: <375E5257.8579541B@serop.abb.se> <50v73.2457$nn.746290@news.shore.net>
Message-ID: <375F4F34.7EF9D260@serop.abb.se>

Hi

Michael P. Reilly wrote:
> [quoted my message] 
> Thomas, the prototype for functions in the setattr slot of a Python type
> is:
>   typedef int (*setattrfunc)(PyObject *, char *, PyObject *);
> 
> Your counter_setattr function should return an int, true (0) for
> success, false (-1) for error.  I believe the source code and Mark's
> book (somewhere in Ch 15? but I don't have the book with me) are about
> the only places this is documented, and it is very obscure.
> 
> "This looks like a job for Documentation Man!"
> 
>   -Arcege

Thanks everyone, I finally managed to get it all to work.
Embarrassing enough, the answer was in the Programming Python-book all
the time (in ch.14 at p.561), but as said below: "it is very obscure".

Actually we were all almost right, the prototype in the book is
 static int mytype_setattr(mytype *x, char *name, PyObject *value);
so the problem was me trying to return nothing at all, None, the
int value assigned and so on.

I guess this answers a question from an earlier thread:
 "Programming Python" still worthwhile?
with a big: "Yes, very much so!" (as goes for c.l.py :-)

If anyone is interrested I've pasted the code to my function
as it turned out below. 

Many Thanks
 -- Thomas S. Strinnhed. thstr at serop.abb.se

static int counter_setattr(counterobject *self, char *name, PyObject *v)
{
   if(strcmp(name, "value") == 0)
   {
      self->value = (int)PyInt_AsLong(v);
      return 0;	
   }
   return -1;
}



From carlos.velasquez at ssmb.com  Fri Jun 25 16:41:59 1999
From: carlos.velasquez at ssmb.com (Velasquez, Carlos)
Date: Fri, 25 Jun 1999 20:41:59 GMT
Subject: Please add me to the list
Message-ID: <BFC0587634C3D111AACB00805FA766FA02C1C5DB@exchny12.ny.smb.com>

Please add carlos.velasquez at smb.com to the list

______________________________________________________________________
C.F. Velasquez
Asst. Vice President
Global Database and Technology Division
Salomon Smith Barney, Inc.




From Gary.D..Duzan at p98.f112.n480.z2.fidonet.org  Tue Jun 29 15:39:54 1999
From: Gary.D..Duzan at p98.f112.n480.z2.fidonet.org (Gary D. Duzan)
Date: Tue, 29 Jun 1999 20:39:54 +0100
Subject: ANN: Stackless Python 0.2
Message-ID: <0000158d@bossar.com.pl>

From: "Gary D. Duzan" <gdd0 at gte.com>

In Message <37791B4D.43ADC401 at appliedbiometrics.com> ,
   Christian Tismer <tismer at appliedbiometrics.com> wrote:

=>Gary Duzan wrote:
=>> 
=>> Toby J Sargeant wrote:
=>> >
=>...
=>>    This appears to be the conventional wisdom. The optimization class I
=>> took used an ML-like language, and the first thing we did was to move
=>> heap frames, which was the default, to the stack (i.e. closure
=>> optimization.)
=>
=>Loosing much of flexibility by doing that, btw.

   True. For example, (if I remember correctly; it has been a while)
tail call optimization is defeated. In this particular case, however,
we didn't break the language semantics (which didn't guarantee constant
memory tail recursion) because we only did stack conversion after
checking that it wouldn't break anything. Otherwise, the closure is
left on the heap. I see that I wasn't clear about that in my original
post; sorry 'bout that.

=>> It might be less of an issue in an interpreted language if it already has
=>> heap
=>> allocation (or other sorts of) overhead to deal with when making calls.
=>> There may be other allocation tricks which can be used, as well.


=>> > It seems that this should affect Stackless Python equally as much. Does a
nyone
=>> > have anything to add on the subject? I would imagine that frames could be
=>> > allocated and managed in chunks to alleviate a lot of the memory manageme
nt
=>> > load...
=>> 
=>>    I would think that as well (at least at first.) Maybe pools of different
=>> sizes of frame to limit the space overhead.
=>
=>Maybe you should really look into the sources.
=>You are discussing things which are not Python
=>related, since this simply does not apply.

   Freely conceeded.

=>The Python "stacks" are so tiny that you can forget about them.
=>Allocations always never occour, unless your program runs
=>astray in recursion.
=>
=>The cost of a call comes from the slightly complicated
=>function setup which is done all the time. Frames belong
=>to the fastest available Python objects already, 
=>like, say dictionaries.

   I thought that might be the case, which is the point I was making in
my earlier paragraph. I was just briefly pondering possibilities for
reducing frame memory management time, if it were necessary. Stackless
Python certainly seems to me to be a Good Thing.

					Gary Duzan
					GTE Laboratories




From bwarsaw at cnri.reston.va.us  Mon Jun  7 22:51:13 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Mon, 7 Jun 1999 22:51:13 -0400 (EDT)
Subject: waste of resources ?
References: <375C0F91.1C73E4E3@icrf.icnet.uk>
	<375C16A9.5ADC4990@icrf.icnet.uk>
	<375C2003.B01BEC3@arlut.utexas.edu>
	<87yahvzqz3.fsf@pc-hrvoje.srce.hr>
Message-ID: <14172.34081.978798.911000@anthem.cnri.reston.va.us>

>>>>> "HN" == Hrvoje Niksic <hniksic at srce.hr> writes:

    |     except (IOError, os.error), detail:

Or you could just catch EnvironmentError. :)

-Barry



From jmrober1 at ingr.com  Wed Jun  2 13:12:34 1999
From: jmrober1 at ingr.com (Joseph Robertson)
Date: Wed, 02 Jun 1999 12:12:34 -0500
Subject: [tkinter] two questions regarding Canvas
References: <3752a555@nntp.server.uni-frankfurt.de>
Message-ID: <37556602.415AFA46@ingr.com>

Sure does,  heres some relevant clips.  Sorry I couldn't send the whole
thing but its too big.
I have a class that creates the canvas, then I set the binding:

def __init__(self)
...
self.canvas.bind('<Button-1>', self.CellClick)
...

Heres the working method:
def CellClick(self, event):
        self.canvas.delete('rect')
        cellbbox = self.canvas.bbox('current')
        self.canvas.create_rectangle(cellbbox, outline='blue',
tags='rect')


Now I don't have it editable yet, I am using this,
    item = self.canvas.create_text(xpos, ypos,  text=field, font=self.f)

in a loop to put data onto the canvas.

I'm sure that creating a hidden Entry and the moving it and making it
visible would work fine, or don't move it and make a 'status line' type
of data editing, like Lotus did.

Good Luck,
Joe



mlauer at amalia-atm.rz.uni-frankfurt.de wrote:

> Howdy, though I'm gaining knowledge through
> browsing through all the available documentation
> (that is, the tkinter life-preserver, tkinter.py
> and the tk/tcl documentation) some things
> remain misterious...
>
> For instance:
>
> a) Does the tk Canvas support "inplace" editing of
> text items or what are the functions Canvas.icursor
> and Canvas.focus for ? If so, how ? For instance,
> binding the <Any-Enter> to a text item function,
> which sets icursor and focus doesn't work for me
> (yes, the canvas has the current focus)
>
> b) Does the Canvas support a visible selection
> of items (such as a boundaring rectangle or so)
> on its own or have I to do this via binding <Any-Enter>
> and <Any-Leave> ?
>
> Have a nice day!
>
> --
>     Regards & Gruesse from Mickey @ http://www.Vanille.de
>   ---------------------------------------------------------
>   How could anyone know me - when I don't even know myself ?






From ian at cfmu.eurocontrol.be  Thu Jun 10 07:49:02 1999
From: ian at cfmu.eurocontrol.be (Ian Wild)
Date: Thu, 10 Jun 1999 13:49:02 +0200
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com>
Message-ID: <375FA62E.CCEF5079@cfmu.eurocontrol.be>

Fernando Mato Mira wrote:
 
>   I'm trying to collect a list of all the _current_ issues making tcl
> a bad language choice. I'd like concrete examples,

Without knowing what you're choosing it FOR, how
can anyone legitimately say it's a bad choice?
For some problem domains it might well be ideal.
(I mean, it's theoretically possible that, in some
parallel universe, there's a even problem for which
C++ is the ideal choice.)

>   Note that I've not listed comp.lang.tcl for increased productivity,
> and to avoid starting a flame war.

Yeah - right.  And no Tcl users ever read here?



From claird at Starbase.NeoSoft.COM  Thu Jun 10 09:57:35 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 10 Jun 1999 08:57:35 -0500
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <4.1.19990610150908.00d09790@exchsrv>
Message-ID: <7jog8f$ljv$1@Starbase.NeoSoft.COM>

In article <4.1.19990610150908.00d09790 at exchsrv>,
Fernando Mato Mira  <matomira at iname.com> wrote:
			.
			.
			.
>A `TCLer'  here is going to give a talk presenting the virtues of
>tcl to the rest of the organization. As a hard-core lisper, it is obvious
>to me that the core philosophical basis of the `New Jersey school' (i.e. the
>whole idea of scripting languages) is broken, and obviously the syntax
>is tasteless to me. But I'm looking for irrefutable signs of bad language
Oh, so *that*'s your point.  Well, RMS has
already said it all--in fact, more than all,
'cause some of his claims are inaccurate.
Myself, I see Tcl and LISP as more alike
than different, but I gather that's not what
you're after so much as <URL:
http://www.vanderburg.org/~glv/Tcl/war/>.
>design that result in low maintenability. For example, dynamic scoping,
>ambiguities, etc. Things like the `string issue' are also interesting if they
>can be exposed in a way that the answer is not "So what?".
So I take it you're going to end up arguing
for Eiffel in favor of Scheme?

I'll ask a better question:  is maintainability
to you a philosophically-accessible dimension?
Are you involved in an academic or "practical"
inquiry?  I respect both, but I find your ap-
proach ambiguous.  As a practical engineering
matter, I find the distinctions you highlight
are lost in comparison to larger organizational
and architectural issues.  Concretely, Tcl's
characteristics are not absolutely bad, only
(potentially so) in relation to the work
practices of your organization.

If you seriously want to debate syntax and
semantics in academic abstraction ... well, at
least we need to know that.
>
>I'm not putting in question the virtues of Tk, as you can access that
>through STk, for example. But if there are bad things there, I'd like
>to know, too. (eg: is there a lot of tcl code making a Tk interface run?
Approximately 0.46 of a lot, yes.
>If so, does speed ever become an issue? I assume that Tk works well
Rarely.  Much less often than the uninitiated
suspect.
>enough that you'd never have to look at the tcl code, and live happily
>inside your reflective Scheme world)
Some TkSchemers do.  Some don't.  It depends.
			.
			.
			.
Be careful with your audience to make clear
*which* "whole idea of scripting languages"
you disfavor.  The hazard is that they'll
turn on Scheme.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From mlv at pobox.com  Thu Jun 24 12:42:40 1999
From: mlv at pobox.com (Michael Vezie)
Date: 24 Jun 1999 12:42:40 -0400
Subject: tkSimpleDialog.py & tvtwm
Message-ID: <7ktn60$pun$1@mlv.mit.edu>

I think I've found a bug in tkSimpleDialog (Python 1.5.1, apologies 
if it's been fixed in 1.5.2).

In the Dialog class, in __init__(), it sets it's geometry by:

    self.geometry("+%d+%d" % (parent.winfo_rootx()+50,
			      parent.winfo_rooty()+50))

Problem is, in tvtwm, if you're not in the first window, the rootx 
and rooty are that of the full virtual window (e.g., if you're in 
one window over, at X position 250 (relative to the current window), 
and your screen is 1000 pixels wide, rootx returns 1250.  When the 
geometry is set, it then sets it to be 1250 relative to the current 
window, which, if your virtual space is only two screens wide, is off 
the screen, never to be seen again.

A solution that I believe will work with both tvtwm and fvwm2 (only 
two I've checked, but probably others) is to add winfo_vrootx() to 
the mix.  In fvwm2, vrootx is always 0 (I think), while in tvtwm, it 
is (in the above example), -1000.  Adding it would give a current 
position of 250, which will work for future windows.

I hope those who know much more about window managers than I do will 
correct my theories as they see fit.  For myself, I'm fixing the problem 
locally by adding winfo_vrootx/y to that line.  Let me know if there's a 
better way that is more portable (or preferably, whatever way will be 
used in future versions of python).

Michael



From tre17 at cosc.canterbury.ac.nz  Thu Jun  3 23:30:22 1999
From: tre17 at cosc.canterbury.ac.nz (Timothy R Evans)
Date: 04 Jun 1999 15:30:22 +1200
Subject: More Tkinter menu problems
References: <3755C714.526B7994@pixar.com> <%Zm53.899$S15.46310@dfiatx1-snr1.gtei.net>
Message-ID: <c75g14863sh.fsf@pc142.cosc.canterbury.ac.nz>

"John Michelsen" <john.michelsen at gte.net> writes:

> >  self.menu.add_command(label = new_name,
> >   command = lambda temp_color = new_color:
> >   self.select_color(temp_color))
> >
> >but I get a Name Error on "self".  So now "self" is out of scope when
> >the menu call happens.
> 
> 
> try:
> self.menu.add_command(label = new_name,
>     command = lambda c=new_color, f=self.select_color: f(c))
> 
> John

Another possible way of implemented getting the same result, without
resorting to unreadable lambda expressions is to create a class
something like...

class Command:
    def __init__(self, func, *args, **kw):
        self.func = func
        self.args = args
        self.kw = kw

    def __call__(self, *args, **kw):
        args = self.args + args
        kw.update(self.kw)
        apply(self.func, args, kw)

then...

command = Command(self.select_color, new_color)

This costs one extra function call when it is used, but I think the
decrease in complexity of the expression make it worth it.

--
Tim Evans






From hniksic at srce.hr  Sat Jun  5 09:56:32 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 05 Jun 1999 15:56:32 +0200
Subject: Deleting characters from a string
Message-ID: <87n1ye3g4v.fsf@pc-hrvoje.srce.hr>

string.translate can delete characters from a string, as well as
translate them.  But what if I want *only* to delete them?  Currently
I do this:

dummytable = string.maketrans('', '')

...
  host = string.translate(host, dummytable, ' "')
...

What I would really like to be able to do is write:

  host = string.translate(host, None, ' "')

Am I missing a more elegant solution?



From milliger at ausinfo.com  Thu Jun 10 12:39:47 1999
From: milliger at ausinfo.com (mike milliger)
Date: Thu, 10 Jun 1999 11:39:47 -0500
Subject: Date formats
Message-ID: <375FEA53.95E3D3BC@ausinfo.com>

I read the documentation in the Python Reference Library (6.3) on time
access and conversions and chapter 8 in programming python, yet I can't
figure out if i need ( or how to use) a tuple to format the return from
localtime().  i get a string now (wed jun 10 1999 xx:xx:xx) but i need
to format it to yyyymmdd_hhmmss.  Is there any switch ( %) or function
that can be useful?  Any help is appreciated.
--mike




From ragica at news.vex.net  Wed Jun  2 14:47:14 1999
From: ragica at news.vex.net (Ragica)
Date: 2 Jun 1999 18:47:14 GMT
Subject: Idle comments
References: <7ipm6c$ln0$1@nnrp1.deja.com> <7iucf0$drp@vern.bga.com>
Message-ID: <7j3u7h$45j$2@hub.org>

On 31 May 1999 11:13:20 -0500, David Friedman <dhfx at realtime.net> wrote:
>   I think I've posted this before but I'll mention it again. I have my
>Win95 color scheme set for white text on a dark background. The default

I too have posted before asking about how to change the blinding
(to me) IDLE background colour... and have seen your previous post
also (and watched avidly for a response to it, as my message got no 
responses... but neither did yours).

So hopefully someone anwsers this mystery this time! Just to let
you know that there are others with the same problem who want to
use IDLE but can't figure out how to change the background colour
to something tolerable.... 

And now i see a new message about a nifty new extension to 
IDLE... callTips... can't someone make an extension to simply 
change the background colour?! Heh.




From tre17 at cosc.canterbury.ac.nz  Wed Jun 23 19:39:05 1999
From: tre17 at cosc.canterbury.ac.nz (Timothy R Evans)
Date: 24 Jun 1999 11:39:05 +1200
Subject: TkInter callback - how to ?
References: <376F4E57.6A23A66F@cellware.de>
Message-ID: <c75iu8eqyg6.fsf@pc142.cosc.canterbury.ac.nz>

bernard <bh at cellware.de> writes:

> I have an array of buttons (a keypad) all using the same callback
> function - is there any way of finding out which button invoked
> the callback ? I suppose I am looking for a tcl/tk solution
> something like
>   -command "myCallback $thisButton"
> but tkInter only seems to provide
>   -command myCallback
> 
> Any help welcome
> 
> bernard

I posted a `Command' class that solves this a while ago, here it is
again:

class Command:
    def __init__(self, func, *args, **kw):
        self.func = func
        self.args = args
        self.kw = kw

    def __call__(self, *args, **kw):
        args = self.args + args
        kw.update(self.kw)
        apply(self.func, args, kw)

then do this:

command=Command(myCallback, thisbutton)

The class given is rather general, and I'm not quite sure if my
precedence for the keyword arguments makes sense.  As you won't use
the keyword arguments this won't matter.

--
Tim Evans




From sdm7g at Virginia.EDU  Sat Jun  5 00:31:28 1999
From: sdm7g at Virginia.EDU (Steven D. Majewski)
Date: Sat, 5 Jun 1999 00:31:28 -0400 (EDT)
Subject: threads
In-Reply-To: <000a01beaef4$a53dda20$f19e2299@tim>
References: <000a01beaef4$a53dda20$f19e2299@tim>
Message-ID: <Pine.A32.3.90.990605002137.16533A-100000@elvis.med.Virginia.EDU>

On Fri, 4 Jun 1999, Tim Peters wrote:

 
> the-guarantee-of-exclusion-simplifies-much-of-the-implementation-ly y'rs  -
> tim

Ahh -- getting nostalgic for old arguments again: 

 I recall this was one of the reasons I favored coroutines over threads --
the explicit yield means that you don't need locks. Sort of like the
non-preemptive multitasking on the Mac. The Mac, of course, is now 
moving towards a fully preemptive multitasking model -- which makes
better sense for a OS because you can't control all of the code in all
of the apps and ensure that they all behave like good citizens. However,
in a single app, that's not a problem, and cooperative multitasking 
with coroutines is a simpler model. 

BTW: Does anyone have anything to add about how threads and locking 
work in JPython? Is it Bug/Feature-compatible with CPython wrt threads,
or does it manage to use any of the Java thread feature to advantage here?


---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---

 "IA-64 looks just about like what you would expect from a PA-RISC
  and IA-32 train wreck with a little VLIW thrown in for spice."
 * Thomas J. Merritt  <tjm at spam.codegen.com> in <news:comp.arch> *




From mal at lemburg.com  Fri Jun 11 05:22:24 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 11 Jun 1999 11:22:24 +0200
Subject: ANN: mxODBC Package - Version 1.1.1
Message-ID: <3760D550.2071295E@lemburg.com>

ANNOUNCING:

                      mxODBC Version 1.1.1

          A Python Extension Package providing a generic
                Interface to ODBC 2.x API compliant                    
                    Database Drivers or Managers


WHAT IT IS:

mxODBC is an extension package that provides a Python Database
API compliant interface to ODBC 2.x capable database drivers
and managers. In addition to the capabilities provided through
the standard API it also provides a rich set of catalog methods
that allow you to scan the database for tables, procedures, etc.
Furthermore, it uses the mxDateTime package for date/time value
interfacing eliminating most of the problems these types normally
introduce.

mxODBC is known to work with Adabas, MySQL, iODBC Unix ODBC Manager,
Solid, Sybase, OpenLink and Intersolv/Merant ODBC drivers, Oracle/NT,
EasySofts ODBC-ODBC bridge and the Windows ODBC Manager.
This covers pretty much the whole range of well known relational
database engines and gives you all the connectivity you'll need for
doing great database work in Python.

WHAT'S NEW ?

The 1.1.1 version fixes a few small bugs with Solid and Access
and includes an improved test script that allows checking the
database connection as well as the database's capabilities.

The subpackages for EasySoft, MySQL and iODBC were updated
to the latest available versions of the drivers/managers.

Some helper functions for pretty printing result sets have
also made their way into the package.

WHERE CAN I GET IT ?

The full documentation and instructions for downloading and
installing can be found at:

        http://starship.skyport.net/~lemburg/mxODBC.html

The mxDateTime package needed for mxODBC can be found at:

        http://starship.skyport.net/~lemburg/mxDateTime.html

WHAT DOES IT COST ?

mxODBC comes with a slightly modified Python-style license:
Usage is free for non-commercial and commercial internal use.
Redistribution of the package in commercial products requires a
separate license and will only be free if the product itself is
free.

Detailed license information is available at:

        http://starship.skyport.net/~lemburg/mxODBC-License.html

WHERE CAN I GET SUPPORT ?

I am offering commercial support for this package through
Python Professional Services Inc. (http://www.pythonpros.com).
Look on their support pages for details or contact me directly.

REFERENCE:

<P><A HREF="http://starship.skyport.net/~lemburg/mxODBC.html>
mxODBC 1.1.1</A> - ODBC Interface. (11-Jun-1999)

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   203 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From Niel.M..Bornstein at p98.f112.n480.z2.fidonet.org  Tue Jun 29 11:54:27 1999
From: Niel.M..Bornstein at p98.f112.n480.z2.fidonet.org (Niel M. Bornstein)
Date: Tue, 29 Jun 1999 16:54:27 +0100
Subject: FTPlib - how to optimize transfer?
Message-ID: <000015a1@bossar.com.pl>

From: nbornstein at plr.com (Niel M. Bornstein)

I am using FTPlib on the Mac, and I am noticing binary uploads seem to go
very slowly.  Is there any way to optimize the block size or something to
speed it up?  Should the block size match that of the open() call used to
open the file?  What's a good block size to use, anyway?

Email replies preferred but I'll check back here too.

Thanks,
Niel



From scott at chronis.pobox.com  Mon Jun 28 05:29:11 1999
From: scott at chronis.pobox.com (scott cotton)
Date: Mon, 28 Jun 1999 05:29:11 -0400
Subject: ConfigParser & STRPTIME ????
In-Reply-To: <rNHvjEdhm5Pp-pn2-OUBSngdMEQZP@d185d18f8.rochester.rr.com>; from Benjamin Schollnick on Mon, Jun 28, 1999 at 04:05:38AM +0000
References: <rNHvjEdhm5Pp-pn2-OUBSngdMEQZP@d185d18f8.rochester.rr.com>
Message-ID: <19990628052910.A17490@chronis.pobox.com>

On Mon, Jun 28, 1999 at 04:05:38AM +0000, Benjamin Schollnick wrote:
| Folks,
| 
| 	Sorry to repost this, but I haven't had any real answers in
| regards to these questions....
| 
| 	1) Can someone show me a example for STRPTIME?
| 		I'm following the language reference, and getting
| 		errors with STRPTIME.  I suspect that either the manual
| 		is in error, or that it doesn't work.


I thought I posted an example:

0628 05:28 bort:~/mailcrypt-3.5.3% python
Python 1.5.2 (#8, May 28 1999, 21:09:50)  [GCC 2.7.2.3] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum,
Amsterdam
>>> import time
>>> time.strptime("19990403", "%Y%m%d")
(1999, 4, 3, 0, 0, 0, 6, 1, 0)
>>>

| 
| 	2) Can you use CONFIGPARSER to write a *.INI file that
| 		you have made, or read into config parser, back
| 		out to disk? From reading Dejanews, I believe that
| 		it can't be?!?!?!?!?!? If so, why?  It seems silly to support
| 		read but not write.
| 
| 			- Benjamin

this one I can't help with, i prefer LParser for
configuration parsing :)

scott



From alan.gauld at gssec.bt.co.uk  Fri Jun 25 07:34:48 1999
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Fri, 25 Jun 1999 12:34:48 +0100
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0rr2.2fi.behrends@allegro.cse.msu.edu>  <37722979.2FCD9A5F@gssec.bt.co.uk> <slrn7n66t8.rh7.behrends@allegro.cse.msu.edu>
Message-ID: <37736958.5BED2666@gssec.bt.co.uk>

Reimer Behrends wrote:
> While I'd be the first to agree that C++ is not exactly the epitome of
> language design, I would think that at least the local variables should
> be easy to identify. The only real problem is to distinguish global

Not so. If the method is long(as these were - over 200 lines 
in some cases, mostly initialising huge DCE Data structures) 
you can't see the local definitions. C++ compounds this with 
its ability to declare local variables anywhere in the code
(so a split screen editor doesn't help).

Alan G.

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



From olipt at mayo.edu  Thu Jun 10 12:45:06 1999
From: olipt at mayo.edu (Travis Oliphant)
Date: Thu, 10 Jun 1999 11:45:06 -0500
Subject: Advice on an how to handle errors in extension module.
Message-ID: <Pine.LNX.4.10.9906101133010.8120-100000@us2.mayo.edu>

I'm writing some Python-C interfaces to some mathematical libraries in
FORTRAN.  In particular, I'm working on interfacing QUADPACK with Python.
The code is basically complete and an old version can be found in 
Multipack at http://oliphant.netpedia.net

The problem is that the FORTRAN code requires that there be an (external)
function defined which computes the integrand.  This function must be of a
certain type.  The function I use is a just wrapper that handles calling
the user defined Python function (stored in a static global variable).

The question I have is that if a Python error occurs while evaluation of
the user-defined function, I'm not sure what to do inside this wrapper
function which can only return a double.  Currently, I'm printing the
error (if it hasn't been printed before) and returning a zero.  This won't
stop the integrator from callling the function again but it will have the
effect of treating the function as if it were zero.

This actually works really well for 1-D integration.  Lately, though, I've
just made the C-interface so that it can be re-entrant (so the Python
function itself could call an integration routine --- allows for easy
multiple integration).  But, this exposes the inelegance of my solution
for handling a Python error.

I'm just fishing for ideas from more knowledgeable people on this list.

Travis Oliphant




From mlh at swl.msd.ray.com  Mon Jun 21 17:49:15 1999
From: mlh at swl.msd.ray.com (Milton L. Hankins)
Date: Mon, 21 Jun 1999 17:49:15 -0400
Subject: Environment inheritance under Windows NT
Message-ID: <Pine.SGI.3.95q.990621173009.142968O-100000@masox202>

Python doesn't seem to pick up the environment from its parent process
under Windows NT.

If I set an EV (environment variable) in the Windows NT System control
panel applet, then run a command prompt and use the SET command to change
the variable's value, then run Python at that same command prompt,
os.environ reflects the old (control panel) value of the EV. 

setlocal and endlocal do not appear to help.

Am I simply expecting Windows to behave too much like UNIX here?


(Please Cc your response to me.)

--
              Milton L. Hankins              \\  ><> Ezekiel 18:21 ><>
Software Engineer, Raytheon Systems Company  //  <mlh at swl.msd.ray.com>
      http://amasts.msd.ray.com/~mlh         \\  RayComNet  7-225-4728




From tismer at appliedbiometrics.com  Sun Jun 13 14:25:02 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Sun, 13 Jun 1999 18:25:02 GMT
Subject: Variations on a theme by "pound bang"
References: <3763e75f.3265255@news.iol.ie>
Message-ID: <3763F77E.6094AEE9@appliedbiometrics.com>


Sean Mc Grath wrote:
> 
> Can anyone shed some light on the benefits of:-
> 
> # !/usr/bin/env python

Uses the env program which is allways there, to figure out
from the environment settings which python to use.
This can be useful when you are running multiple Python
versions. Use this in a secure setup, on a local
machine, and NEVER AS ROOT.

This feature is dangerous in an CGI environment.
CGI applications should depend on environment
settings to the bare minimum. Whatever can be avoided,
should be avoided.

> as opposed to
> 
> # !/usr/bin/python

The way to go for CGI. Simple, explicit, not dependant
from someone messing up with server settings.

ciao - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From gmcm at hypernet.com  Fri Jun  4 16:37:02 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 4 Jun 1999 15:37:02 -0500
Subject: threads
In-Reply-To: <87aeug7yn7.fsf@pc-hrvoje.srce.hr>
Message-ID: <1283593620-47537497@hypernet.com>

Hrvoje Niksic writes:

> "Barry A. Warsaw" <bwarsaw at cnri.reston.va.us> writes:
> 
> > You might be interesting in digging around for Greg Stein's
> > free-threading patches.  Greg did a lot of work, I think in the
> > Python 1.4 days, to remove the central interpreter lock and make all
> > the data structures and such thread safe.
> 
> Why would I want to do that? 

See that other thread.

> If Greg's patches haven't been applied
> to the base, surely mine would not be either.  

As I recall, Greg's patches didn't make it because (1) they were late 
and (2) they were slow. Guido has never issued an "over my dead body" 
message on the subject, which considering how often it has come up, 
could be considered expressing interest in the idea.

OTOH, if you don't feel strongly enough, or don't have the time, it 
can wait for someone else.

- Gordon



From amackay at starvision.com  Wed Jun 30 15:36:55 1999
From: amackay at starvision.com (Angus MacKay)
Date: Wed, 30 Jun 1999 12:36:55 -0700
Subject: ioctl in python
Message-ID: <377A71D7.470AF9DA@starvision.com>

is there some magic to the fcntl.ioctl() call?

I can do these ioctl calls in C no problem (with a small wrapper):
(amackay at phat)~/tmp$ ./ioctltest /tmp 0x80047601
0x80047601: 1
(amackay at phat)~/tmp$ ./ioctltest / 0x80047601
0x80047601: 0
(amackay at phat)~/tmp$ ./ioctltest /proc 0x80047601
/proc: Inappropriate ioctl for device

that was the Linux ioctl for EXT2 version.

but in python:
>>> a = 0
>>> req = 0x80047601
>>> import fcntl
>>> fo=open('/tmp')
>>> fcntl.ioctl(fo.fileno(), req, a)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
IOError: (14, 'Bad address')
>>> 

thanks in advance.
cheers, Angus.



From joechan25 at hotmail.com  Sat Jun 26 07:47:45 1999
From: joechan25 at hotmail.com (Joe Chan)
Date: Sat, 26 Jun 1999 11:47:45 GMT
Subject: Python for large projects?
Message-ID: <3774b5d1.690922@news.force9.net>

Hi, I've been doing mostly desktop and Client/Server development work
but I'm considering how best to develop web-based and distributed
projects. 

I read a paper on the Python.org site comparing Java and Python and
was impressed. I'm interested in other peoples views and experiences
of using Python for: large projects; team developments; distributed
objects; intranet development etc.

TIA



From matomira at iname.com  Sun Jun 13 10:59:18 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Sun, 13 Jun 1999 16:59:18 +0200
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <3762988E.C94AA252@iname.com> <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu>
Message-ID: <3763C746.5CA8FE78@iname.com>


Peter wrote:

> If you are a fan of CLOS, you might want to check out the Dylan
> programming language.
>
> It is basically a thoroughly modernized descendant of Lisp/CLOS with an
> infix instead of prefix syntax.  In Dylan, almost everything is an object

Although Dylan never really made it into "Fernando's elegibility list"
because
it was not a serious contender against Scheme and CL yet, it definitely got
banished the day the Lisp syntax was dropped. I'm still waiting for the day
when the Dylan fans will reinstate the alternative.

But my dream is a `parenthesized Cecil'
http://www.cs.washington.edu/research/projects/cecil/

or "all you wanted in CLOS (and more) since you got `corrupted' by Eiffel"
;-)

--
Fernando D. Mato Mira
Real-Time SW Eng & Networking
Advanced Systems Engineering Division
CSEM
Jaquet-Droz 1                   email: matomira AT acm DOT org
CH-2007 Neuchatel                 tel:       +41 (32) 720-5157
Switzerland                       FAX:       +41 (32) 720-5720

www.csem.ch      www.vrai.com     ligwww.epfl.ch/matomira.html







From claird at Starbase.NeoSoft.COM  Tue Jun 29 11:45:16 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 29 Jun 1999 10:45:16 -0500
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <7l4nht$jl7$1@nnrp1.deja.com> <9DDBCC52A7B26002.649D772F7232D418.CE2709D4B08E7401@lp.airnews.net> <ui6747qhvz.fsf@bacon.ethz.ch>
Message-ID: <7lapmc$d2g$1@Starbase.NeoSoft.COM>

In article <ui6747qhvz.fsf at bacon.ethz.ch>,
Ernst-Udo Wallenborn  <wallenborn at phys.chem.ethz.ch> wrote:
>
>
>To ask a completely different question: What about
>threads and GUI? With Tkinter you have to rewirte the
>mainloop like this
>
>    def mainloop(self):
>        while 1:
>            Frame.update(self)
>            time.sleep(0.02)
>            if self.quitting == 1:
>                sys.exit(0)
>
>
>to convince it to run with thread or threading.  I remember having
>read a comment from Guido van Rossum, citing Jan Osterhout that Tk
>will never be thread-safe and that therefore this problem will not go
>away.
>
>Would this be any different with a different GUI kit?
			.
			.
			.
As Mr. Zeitlin asserts in his follow-up, multi-threading
the GUI is generally a futile hazard.

Dr. Ousterhout always, as far as I know, goes by his
Americanized given name John.  I make the point because
one of the people working most actively with Tk capabil-
ities is Jan Nijtmans.

Yes, John has said he expects Tk never to be thread-safe.
That has changed, though, just in the last few months.
You're quite right that many toolkits have issues in re-
gard to threading.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From samschul at pacbell.net  Tue Jun 22 11:33:54 1999
From: samschul at pacbell.net (Sam Schulenburg)
Date: Tue, 22 Jun 1999 08:33:54 -0700
Subject: IDLE 0.5 Function parameter display?
References: <7kmpnu$red$1@nnrp1.deja.com> <7ko57c$cfk$1@m2.c2.telstra-mm.net.au>
Message-ID: <T0Ob3.5$4s.799@typhoon-sf.snfc21.pbi.net>

Thanks Mark;
I founnd that the following will also work. I do not know if it is an
acceptable solution.

from MyDll import my_func   # first I import the function from my dll

my_func.__doc__ = "This is my documentation string"





Mark Hammond wrote in message <7ko57c$cfk$1 at m2.c2.telstra-mm.net.au>...
>The best way at the moment is to add a docstring to the function in the C
>code.  If you look at the help provided for (eg) string.split, it is quite
>effective even though it is only from the docstring.
>
>Mark.
>
>Sam Schulenburg wrote in message <7kmpnu$red$1 at nnrp1.deja.com>...
>>I love the ability of IDLE 0.5 to display the function parameter
>>requirements as defined in the original function definition. My problem
>>is when I have a function defined in a windows dll written in 'C', then
>
>
>





From gmcm at hypernet.com  Sun Jun  6 22:11:41 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sun, 6 Jun 1999 21:11:41 -0500
Subject: "The Python Way"
In-Reply-To: <7jf39e$4t4$2@cronkite.cc.uga.edu>
Message-ID: <1283400745-59139041@hypernet.com>

Graham Matthews wrote:

> The key word is "discuss". To have a "discussion" one must have
> participants who don't simply want to defend "the python way".
> That's where that word "parachial" comes in!

Please, Graham. Your posts about GC to collect cyclic references got 
quite a few favorable replies. No one surfaced with both the time and 
enthusiasm to implement it, but that should not be a surprise. Look 
at the state of Greg's free-threading patches, or the total lack of 
activity on the Types SIG. Remember also that most responses will be 
from people who feel strongly one way or the other, which will often
have a negative correlation to the thoughtfulness of the replies. 
That's  "Usenet".

You did, however, start by proposing using Java as the implementation 
language for Python 2, which is a very different matter.

petrolium-distillates-on-troubled-waters-ly y'rs

- Gordon



From kjwolf at online.de  Wed Jun  9 15:54:33 1999
From: kjwolf at online.de (Klaus-Juergen Wolf)
Date: Wed, 9 Jun 1999 21:54:33 +0200
Subject: Problem with Curses
Message-ID: <ppgmj7.s7r.ln@solling.seismic.de>

(Python 1.5.2 on FreeBSD 3.2-RELEASE (i386))

I found that (after having compiled Python with cursesmodule.c -
testall.py succeeds) the following program gives a segmentation fault:

-*-
import curses

mainscr=curses.initscr()
testwin=mainscr.subwin(2,2,2,2)
testwin.erase()
del testwin
curses.endwin()
-*-

I don't understand this - the C equivalent doesn't show any problems.
(You can drop "del testwin", it has no effect anyway.)

In no way, Python should end up in a segmentation fault - in this case,
it appears that the first screen-writing call to libtermcap causes the
problem.

Can anyone help me?

thanx & cu
k.j.



From jam at newimage.com  Sun Jun 20 16:12:53 1999
From: jam at newimage.com (jam)
Date: Sun, 20 Jun 1999 16:12:53 -0400
Subject: simple text file 'parsing' question
In-Reply-To: <m3ogiaq31k.fsf@havenrock.com>; from Matt Gushee on Mon, Jun 21, 1999 at 12:56:07AM +0900
References: <376B1AAC.19FE8BCE@mysolution.com> <m3ogiaq31k.fsf@havenrock.com>
Message-ID: <19990620161253.A814@toast.internal>

On Mon, Jun 21, 1999 at 12:56:07AM +0900, Matt Gushee wrote:
> KP <terocr at mysolution.com> writes:
> 
> > Here's my dilema: a directory filled (200+) with small emails. My goal
> > is to strip all the headers and combine them into one file. I can read
> > all the files just fine and write them all to one file, but I cannot
> > discern how to strip the headers.
> 
> I have no expertise in this area, but I've been reading the "Internet
> Data Handling" section of the Library Reference (Ch. 12 of the 1.5.2
> edition), and it seems like there are several modules that might help
> you. In particular, check out 'rfc822.'
> 
> Hope this helps.
> 
> Matt Gushee
> Portland, Maine, USA
> mgushee at havenrock.com
> 

I wrote a small piece of code that does *exactly* what you are describing.
it doesn't exactly strip the headers, but it parses the message using rfc822
and deals with it. you'll find it attached to this message. if for some
reason it doesn't come through, let me know, and I'll resend it.

regards,
Jeff
-- 
|| visit gfd <http://quark.newimage.com/> 
|| psa member #293 <http://www.python.org/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>
-------------- next part --------------
#!/usr/bin/env python

import os
import dircache
import mimetools

import colacanister

import getdate

from rfc822 import Message

_COLAROOT="/home/jam/projects/cola/cola.archive"
_COLABASEHREF="http://www.cs.helsinki.fi/%7Emjrauhal/linux/cola.archive/"

if __name__ == "__main__":
	l = dircache.listdir(_COLAROOT)
	print len(l)
	for item in l:
		p = os.path.join(_COLAROOT, item)
		if os.path.isdir(p):
			articles = dircache.listdir(p)
			for a in articles:
				if a[:5] != "cola." and a[:4] != "mjr.":
					continue

				fp = open(os.path.join(p, a), "r")
				m = Message(fp, seekable=0)
				fp.close()

				if not m.has_key("subject"):
					print "** message does not have subject line. skipped."
					continue

				url = os.path.join(item, a)
				print "processing '%s'" % (url),
				if colacanister.get_cola_by_archiveurl(url) is None:
					c = colacanister.colacanister()
					c["cola_from"] = m["from"]

					if m.has_key("date"):
						c["cola_dateposted"] = getdate.getdate(m["date"])

					c["cola_subject"] = m["subject"]
					c["cola_archiveurl"] = url
					c.insert()
					print "added."
				else:
					print "already archived."


From just at letterror.com  Sat Jun 12 05:07:04 1999
From: just at letterror.com (Just van Rossum)
Date: Sat, 12 Jun 1999 09:07:04 GMT
Subject: gadfly install problem
In-Reply-To: <7jrn7g$n2f$0@205.138.136.179>
References: <7jrn7g$n2f$0@205.138.136.179>
Message-ID: <l03102800b387d2619d65@[193.78.237.153]>

At 7:14 PM +0000 6/11/99, Douglas Nichols wrote:
>I am trying to install gadfly for use on my mac with python 1.5.1
>
>The installation seems to have worked.  When I try to test gadfly by
>creating a new database, I get the following error message:
>
>Python 1.5.1 (#37, Apr 27 1998, 13:36:04)  [CW PPC w/GUSI w/MSL]

Ah, you're on a Mac.

>Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>>> import gadfly
>>>> connection = gadfly.gadfly()
>>>> connection.startup("testbase", "testfolder")
>>>> cursor = connection.cursor()
>>>> cursor.execute("create table ph (nm varchar, ph varchar)")
>shift: no mem in addchild
>Traceback (innermost last):
>  File "<stdin>", line 1, in ?
>  File "Macintosh HD:Development:Python 1.5.1:Lib:lib-gadfly:gadfly.py",
>line 320, in execute
>    from sqlsem import Parse_Context
>MemoryError
>>>>
>
>Any suggestions?  I am very new to python and am still working my way
>through the Learning Python book.

You'll need to increase Python's memory partition. Python's parser uses a
*lot* of memory if a source file is big.

Just





From ivnowa at hvision.nl  Sun Jun  6 16:27:20 1999
From: ivnowa at hvision.nl (Hans Nowak)
Date: Sun, 6 Jun 1999 20:27:20 GMT
Subject: Newbie question
In-Reply-To: <7je3sk$46$1@nnrp1.deja.com>
References: <7je3sk$46$1@nnrp1.deja.com>
Message-ID: <199906062025.WAA15357@axil.hvision.nl>

On 6 Jun 99, mesak1234 at my-deja.com wrote:

> Hi!
> Well, i get problems with getting date from OS....(win32 at this
> moment)... All what i need that to find out date attribute of a file.
> But i even cannot find out what the function i can use to get the real
> time (anything like Date date = new java.util.Date(); ???? :( )
> All functions of 'time library' seem fine but don't help so much...
> Any suggestions?
> Thanks,
>   Daniel

To get the date and time attributes of a file, you can use os.stat. 
E.g. 

os.stat("myfile.txt")

will return a tuple with file info... its size, attributes, etc. 
Element -2 (!) is the time used by the file system. You will see a 
weird long number like, for example, 928696080; this is because date 
and time are packed into one number. Unpack it with the 
time.localtime function. This returns another tuple whose elements 
are pretty straightforward (year, month, etc).

The system time, on the other hand, can be retrieved with 
time.time(). Once again, you'll get one number, which needs to be 
unpacked with time.localtime().

For further information, I suggest you study the time module and 
os.stat. :)

Veel liefs,

--Hans Nowak (ivnowa at hvision.nl)
Homepage: http://fly.to/zephyrfalcon



From pduffin at mailserver.hursley.ibm.com  Fri Jun 11 07:18:58 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Fri, 11 Jun 1999 12:18:58 +0100
Subject: Embedded Tcl for Python
References: <lJT73.2677$nn.808830@news.shore.net>
Message-ID: <3760F0A1.1DDC@mailserver.hursley.ibm.com>

Michael P. Reilly wrote:
> 
> A few weeks ago, someone asked if there was a Tcl module for Python.
> People responded (including myself) that Tkinter uses a Tcl/Tk
> interpreter, but you couldn't get rid of the annoying window.
> 
> I've since written an extension ("pytcl") that creates Tcl interpreter
> objects.  As part of that, I have also rewritten _tkinter to use these
> new objects.
> 
> I've had a short discussion with Guido and he's interested in the
> idea.  Now I need some alpha testers; I only have access to UNIX boxes,
> and very few types at that.
> 
> The _tkinter module has the same functionality as before, including the
> same threading issues.  I've include Deiter's Tcl/Tk8.1 patch but have
> not had the opportunity to test that yet (right not it is just do-no-
> damage).
> 
> Contact me at arcege at shore.net if you are interested in helping me test
> these modules.
> 

I was interested in doing something similar in Tcl (e.g. creating a
python interpreter object) and I was wondering if someone could give
me some pointers as to how to go about doing it. 

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From behrends at cse.msu.edu  Thu Jun 24 01:28:40 1999
From: behrends at cse.msu.edu (Reimer Behrends)
Date: 24 Jun 1999 05:28:40 GMT
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0rr2.2fi.behrends@allegro.cse.msu.edu>  <8L_b3.1213$4s.136166@typhoon-sf.snfc21.pbi.net>
Message-ID: <slrn7n3gi0.44d.behrends@allegro.cse.msu.edu>

Zigron (zigron at jps.net) wrote:
> Reimer Behrends <behrends at cse.msu.edu> wrote in message
> news:slrn7n0rr2.2fi.behrends at allegro.cse.msu.edu...
[...]
> > Among all the suggestions for Python 2.0, removing the necessity of
> > prefixing methods and instance variables with "self." (and not having
> > to list it as an argument, either) would be my #1 wish.
> 
>     Whyfor? I personally like the seperate scopes, beween the class
> and the method. It makes it so I don't have to clean up as much. I can
> use as many temporary variables as I want in a method, and as soon
> as that method is gone...... its gone.

I said nothing like that. I am simply saying that the syntax is a
major pain. To remind you, pretty much every single OO language other
than Python does not require you to add a self; yet at the same time
they have per-method and per-object data. It's a matter of how you
declare it.

>     The self.attribute thing makes sense. Each object has its own scope,
> er, namespace, whatever. Methods are considered objects, as are
> classes. Removing the 'self' requirement would, to me, change that.
> Methods wouldn't be objects anymore. :) At least it seems so to me,

I'm sorry, but I cannot make much sense of what you try to say here.
Could you perhaps elaborate?

[...]

				Reimer Behrends



From fredrik at pythonware.com  Fri Jun  4 12:35:15 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 4 Jun 1999 16:35:15 GMT
Subject: "The Python Way"
References: <001101beae49$39af7bc0$949e2299@tim> <7j8pra$55v$4@cronkite.cc.uga.edu>
Message-ID: <00f301beaea8$39f419b0$f29b12c2@pythonware.com>

Man looking for an argument (M) wrote:

M> [Fredrik Lundh]
M> > Would both Guido and TIm Peters be willing to collaborate on a short
M> > paper
M> 
M> Oh thank you Fredrik for posting the above. Does anyone else consider
M> that the above sounds slightly paranoid?

I didn't write that.  Sure sounds like you're slightly
paranoid, Graham.

(If you want me to go on arguing, you'll have to pay for
another five minutes).

</F>





From neelk at brick.cswv.com  Thu Jun 24 23:48:51 1999
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 24 Jun 1999 22:48:51 -0500
Subject: [ANN] DejaGrabber 0.1 -- search Dejanews with Python
Message-ID: <7kuu73$1ki$1@brick.cswv.com>

If you'd like a convenient way of searching the DejaNews archive using
Python, I'm happy to present the DejaGrabber module. This is a module
to make searching Dejanews easier. There are two classes, DejaGrabber
and Article. I've tried to add a reasonable number of useful
docstrings, too.

The interface to DejaGrabber could probably use a bit of work. If I
ever use it enough to find out what the flaws are, I will fix
it. Patches are welcome. :)

Here's an example:
>>> from DejaGrabber import *
>>> d = DejaGrabber(group='comp.lang.python', author='Tim Peters')
>>> l = d.get_messages(4)
>>> print l
[99/06/23  30 comp.lang.python          RE: Python Exes!      Tim Peters
, 99/06/18  28 comp.lang.python RE: Newbie: Truth values (th      Tim Peters
, 99/06/18  28 comp.lang.python       RE: DATE ARITHMETIC      Tim Peters
, 99/06/18  28 comp.lang.python     RE: NameError Problem      Tim Peters
, 99/05/18  25 comp.lang.python     RE: while (a=b()) ...      Tim Peters
]

You can grab a copy at 

http://www.sff.net/people/neelk/free-software/DejaGrabber.py


Neel



From gmcm at hypernet.com  Tue Jun 15 10:17:10 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 15 Jun 1999 09:17:10 -0500
Subject: extension called from embedded code
In-Reply-To: <7k59i2$ijd$1@nnrp1.deja.com>
Message-ID: <1282666009-103335976@hypernet.com>

Henk Jansen writes:

> Trying to get the most out off the C++/Python team I run into
> problems when I try to load extension modules (written in C, like
> kjbuckets or ranlib) from Python code that is embedded in a C++
> application. The API I use is 'PyRun_SimpleString("import
> kjbuckets\n");' and causes unresolved symbol error messages from the
> Python interpreter (e.g. PyType_Type for kjbucketsmodule.so and
> PyExc_RuntimeError for ranlib.so). Loading these modules when Python
> is in control goes fine. Has anyone any idea how to do this?

This has something (platform / compiler / linker specific) 
to do with the way you've built your app. For some reason the symbols 
in your embedded Python are not getting exposed to the dynamic 
linker. Perhaps if you post the relevant snippets from your Makefile, 
someone with experience on that platform (compiler...) can help.

- Gordon



From sams at quinta.com  Mon Jun 21 21:43:28 1999
From: sams at quinta.com (Sam Schulenburg)
Date: Tue, 22 Jun 1999 01:43:28 GMT
Subject: IDLE 0.5 Function parameter display?
Message-ID: <7kmpnu$red$1@nnrp1.deja.com>

I love the ability of IDLE 0.5 to display the function parameter
requirements as defined in the original function definition. My problem
is when I have a function defined in a windows dll written in 'C', then
I do not have access to the function prototype information. Is their a
way that I can re-prototype these functions without forgoing access to
the original functions?

Sam Schulenburg


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From gherron at aw.sgi.com  Wed Jun  9 12:12:54 1999
From: gherron at aw.sgi.com (Gary Herron)
Date: Wed, 9 Jun 1999 16:12:54 GMT
Subject: Tkinter Question
References: <Pine.A41.4.10.9906090941210.65174-100000@barney.mcit.com>
Message-ID: <375E9286.E39EF45A@aw.sgi.com>

Robert Meegan wrote:
> 
> Greetings ---
> 
> I have an alarm display that consists of a  Frame that contains multiple
> frames within it, stacked one above the other. When a user clicks on one
> of the subframes to acknowledge it, I want the subframe to disappear and
> the parent frame to resize itself to remove the gap.
> 
> Getting the subframe to remove itself is no problem. I bound the
> <Button-1> event to a method and had the frame destroy itself. Life is
> good.
> 
> On the other hand, I can't figure out how to make the parent frame resize
> itself.
> 
> Suggestions will be greatly appreciated.
> 
> --- Robert
> 
> --------------------------------
> Robert Meegan
> MCIWorldCom - Cedar Rapids, Iowa
> 319.375.2416

You probably used `pack' to put each subframe into the main frame, so
you should use its inverse operation `forget' to take it out.  Then the
parent frame resizes automatically.

Here is a very small example:  poke any of the buttons and watch it
disappear and the frame resize around the remaining buttons.

-- 
Dr. Gary Herron <gherron at aw.sgi.com>
206-287-5616
Alias | Wavefront
1218 3rd Ave, Suite 800, Seattle WA 98101
-------------- next part --------------
#!/usr/bin/env python

from Tkinter import *

root = Tk()

for i in range(10):
	f = Frame(root)
	b = Button(f, text='%s'%i,
			   command=lambda f=f: f.forget())
	b.pack()
	f.pack()

root.mainloop()

From larsga at ifi.uio.no  Tue Jun  8 05:04:38 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 08 Jun 1999 11:04:38 +0200
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <7j0urf$2dm$3@cronkite.cc.uga.edu> <19990601122134.A30248@chronis.pobox.com>
Message-ID: <wkn1ybdpw9.fsf@ifi.uio.no>

* scott cotton
|
| kaffe, the gnu java implementation, has some interesting GC
| features which might be explored for python 2.0:

So does Allegro Common Lisp. I recommend anyone who's interested to go
to <URL: http://www.franz.com/> and download an evaluation copy of ACL
and read the doc/cl/gc.htm file.

Personally, I don't even hesitate to say that Python 2.0 should have
garbage collection, and preferably one that has both compactable and
non-compactable references, as well as weak references. Ref-counting
is slow and error-prone, as far as I'm aware, and should be avoided.

--Lars M.



From bwinton at tor.dhs.org  Fri Jun 18 23:39:49 1999
From: bwinton at tor.dhs.org (Blake Winton)
Date: Sat, 19 Jun 1999 03:39:49 GMT
Subject: Newbie: Truth values (three-valued logic)
References: <000201beb946$b9c99e00$969e2299@tim> <Pine.OSF.4.05.9906181025150.6087-100000@pinax.Mathematik.Uni-Bielefeld.DE>
Message-ID: <slrn7mm0qu.2if.bwinton@tor.dhs.org>

On Fri, 18 Jun 1999, Olaf Delgado wrote:
>On Fri, 18 Jun 1999, Tim Peters wrote:
>> [Olaf Delgado]
>I understand that. A new meaning for "and" and "or" would have to be
>built into the language, but I don't think the merits would justify the
>efforts.  The simplest way I see would be to introduce a new type with
>a single value "Maybe", similar to "None", which would have to be
>handled specially by "if", "not", "and" and "or". It should be
>relatively simple to implement this, as far as I can see, but of course
>it would slow down things by having to consider all those special
>cases.

[snip]

>Or I could set __and__(x,y)=x*y and __or__(x,y)=x+y-x*y. Then I only
>would have to invent a way to partially execute a block of code. :)

Wasn't it Intercal which first had something like this?

if Maybe:
    # do this some of the time
else:
    # do this other times
    # or is that do part of this, and part of that?

I think the first is fairly easy to hack up, but the second would
probably require some ByteCodeHacks (tm).  Now I'm starting to be
interested in one-value logic.  I'm just not sure whether I want the
value to be "True", "False", or "Maybe".

Later,
Blake.

-- 
One Will.  One Dream.  One Truth.  One Destiny.  One Love.



From gmcm at hypernet.com  Mon Jun 14 22:44:36 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 14 Jun 1999 21:44:36 -0500
Subject: Basic python OOPs question
In-Reply-To: <7jot0r$p32$1@nntpd.lkg.dec.com>
Message-ID: <1282707564-100836116@hypernet.com>

Daniel Mendyke asks:
> 
> Do I understand correctly that objects are removed
> from the name space when they go out of scope?

They are reference counted. Going out of scope is one way to decrease 
the ref count. Rebinding the reference is another ("obj = None").

> And that they go out of scope based on indentation?

Not exactly. An "if" statement, for example, does not have a new 
namespace. Functions and methods do.
 
> If so is there a default function called on 
> objects that go out of scope?  The __init__
> function is called when an object is created.
> Is there a function that is called when the
> object is destroyed?

Yes. It's __del__. It is called when the refcount drops to 0. It's 
also called at clean-up time, which can be troublesome, because 
things you refer to in __del__ may well have already been cleaned up. 
There's a FAQ entry on the subject.

- Gordon



From mal at lemburg.com  Mon Jun  7 17:12:00 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Mon, 7 Jun 1999 21:12:00 GMT
Subject: mxODBC and MySQL
References: <Pine.LNX.4.05.9906072222280.5855-100000@ns.ac-copy.net>
Message-ID: <375C35A0.7754C3BD@lemburg.com>

Joachim Schmitz wrote:
> 
> Hi,
> 
> I compiled and installed  mxODBC for MySQL on SuSE 6.1 without problems.
> I get this:
> Python 1.5.2b2 (#6, Feb 28 1999, 21:41:19)  [GCC egcs-2.91.60 19981201
> (egcs-1.1
> .1  on linux2
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import ODBC.MySQL
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "/usr/local/lib/python1.5/ODBC/MySQL/__init__.py", line 7, in ?
>     from mxODBC import *
> ImportError: /usr/local/lib/python1.5/ODBC/MySQL/mxODBC.so: undefined
> symbol: PyBuffer_Type
> >>

You can either upgrade to 1.5.2 final or turn off the buffer
type code in mxODBC.c manually (enclosed in HAVE_PYTHON_BUFFEROBJECTS
#defines). I guess you stumbled into test code that I forgot to remove
for the distribution; sorry.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   207 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From schorsch at schorsch.com  Mon Jun 14 19:40:05 1999
From: schorsch at schorsch.com (Georg Mischler)
Date: Mon, 14 Jun 1999 23:40:05 GMT
Subject: do anonymous pipes normally work on NT?
References: <7juatc$fef$1@nnrp1.deja.com>
Message-ID: <7k43sk$6vd$1@nnrp1.deja.com>

Ok, here we go again.

I simplified my class to the absolute minimum I could.
It now does nothing but create three pipes, save the
parent ends to self.xxx, and create a process with
the child ends of the pipes. After that it tries to
write to the stdin pipe of the child, which goes
without error, and tries to read from stdout, which
blocks.

To have a look at the other end, I wrote a little
script that mimicks the unix "cat" command (in a
very verbose logging incarnation...). This script
first tries to write something to stdout as well,
which goes without error, and then enters a loop
reading from stdin and writing to stdout.

This script is called from my class as a child
process. The interesting thing is now, that in this
case the child gets a "Bad file descriptor" when
reading from stdin!

There must be something going fundamentally wrong
with the pipes, but without any Windows programming
background, I have no chance left figuring out what.
This happens on NT4sp3 with Python 1.5.2 (the
binary from python.org) and the win32 extensions
build 125, if that makes any difference.

If this should be caused by a mistake I made
(which may be likely), could anyone recommend a
book where I can learn how to do this stuff
right? And in any case, what is going on here?

To try this out, copy both files together into
a directory, change the path of python in the
last code line of numproc.py to something that
exists on your system, and call "python runproc.py".
You'll have to kill the process with the task
manager (at least I have to do this). The new
file "log.txt" will contain the results of the
child process.


Thanks a lot for any help!

-schorsch

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

'''cat.py

a "enhanced" version of the good old cat unix filter.
'''

import sys

try:
    # write status to a file, in order not to lose it.
    f = open('log.txt', 'a')
    f.write('before write\n')
    sys.stdout.write('huh?')
    f.write('after write, before read\n')
    data = sys.stdin.read()
    f.write('after read\n')
    while data:
        f.write('in loop, before write\n')
        sys.stdout.write(data)
        f.write('in loop, after write, before read\n')
        data = sys.stdin.read()
        f.write('in loop, after read\n')
except:
    import traceback, string
    tb_strs = traceback.format_exception(
            sys.exc_type,
            sys.exc_value,
            sys.exc_traceback)
    f.write(string.join(tb_strs))

# end of cat.py

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

'''cat.py

an "enhanced" version of the good old cat unix filter.
'''

import sys

try:
    # write status to a file, in order not to lose it.
    f = open('log.txt', 'a')
    f.write('before write\n')
    sys.stdout.write('huh?')
    f.write('after write, before read\n')
    data = sys.stdin.read()
    f.write('after read\n')
    while data:
        f.write('in loop, before write\n')
        sys.stdout.write(data)
        f.write('in loop, after write, before read\n')
        data = sys.stdin.read()
        f.write('in loop, after read\n')
except:
    import traceback, string
    tb_strs = traceback.format_exception(
            sys.exc_type,
            sys.exc_value,
            sys.exc_traceback)
    f.write(string.join(tb_strs))

# end of cat.py

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

'''runproc.py

start a process with three inherited pipes.
Try to write to and read from those.
'''

import win32pipe
import win32file
import win32process
import win32security

class Process:
    def run(self, cmdline):

        # security attributes for pipes
        sAttrs = win32security.SECURITY_ATTRIBUTES()
        sAttrs.bInheritHandle = 1

        # create pipes
        hStdin_r,  self.hStdin_w  = win32pipe.CreatePipe(sAttrs, 0)
        self.hStdout_r, hStdout_w = win32pipe.CreatePipe(sAttrs, 0)
        self.hStderr_r, hStderr_w = win32pipe.CreatePipe(sAttrs, 0)

        # set the info structure for the new process.
        StartupInfo = win32process.STARTUPINFO()
        StartupInfo.hStdInput  = hStdin_r
        StartupInfo.hStdOutput = hStdout_w
        StartupInfo.hStdError  = hStderr_w
        StartupInfo.dwFlags = win32process.STARTF_USESTDHANDLES

        # start the process.
        hProcess, hThread, dwPid, dwTid = win32process.CreateProcess(
                None,   # program
                cmdline,# command line
                None,   # process security attributes
                None,   # thread attributes
                1,      # inherit handles, or USESTDHANDLES won't work.
                        # creation flags. Don't access the console.
                win32process.DETACHED_PROCESS,
                None,   # no new environment
                None,   # current directory (stay where we are)
                StartupInfo)
        # normally, we would save the pid etc. here...

        err, bytes = win32file.WriteFile(self.hStdin_w, 'hmmm')
        print 'wrote:', bytes, 'bytes'

        res, str = win32file.ReadFile(self.hStdout_r, 1)
        if res == 0: print 'read:', str
        else: 'read nothing.'

        res, str = win32file.ReadFile(self.hStderr_r, 1)
        if res == 0: print 'read:', str
        else: 'read nothing.'


if __name__ == '__main__':
    p = Process()
    p.run('c:\programme\python\python.exe cat.py')

# end of runproc.py

--
Georg Mischler  --  simulation developper  --  schorsch at schorsch.com
+schorsch.com+  --  lighting design tools  --  http://www.schorsch.com/


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From claird at Starbase.NeoSoft.COM  Tue Jun 29 11:33:11 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 29 Jun 1999 10:33:11 -0500
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773c3a0.102825445@news.oh.verio.com> <xPPc3.535$or1.22866@typhoon-sf.snfc21.pbi.net> <xJRc3.292$Of3.67094@news.shore.net>
Message-ID: <7laovo$c4s$1@Starbase.NeoSoft.COM>

In article <xJRc3.292$Of3.67094 at news.shore.net>,
Michael P. Reilly <arcege at shore.net> wrote:
			.
			.
			.
>There are other graphical packages for Python out there too, it would
>be appropriate to contrast them as well.
			.
			.
			.
I occasionally work on such a comparison, currently available at <URL:
http://starbase.neosoft.com/~claird/comp.lang.python/python_GUI.html>.
I welcome updates from readers.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From fredrik at pythonware.com  Wed Jun 30 18:23:22 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 30 Jun 1999 22:23:22 GMT
Subject: problem with Fnorb and module new
References: <377A8D17.9B4D17C8@sr.hp.com>
Message-ID: <00ba01bec347$2aa95f70$f29b12c2@secret.pythonware.com>

>   I just tried installing Fnorb on HP-UX 10.20... the install went
> smoothly, but when I tried stepping through the Hello World example,
> python complained about not being able to find module new. The
> complaint is certainly valid -- I can't find new.py anywhere either.
> Anyone know what's going on here?

it's a C module, not a Python module:
http://www.python.org/doc/FAQ.html#4.70

and it's not built by default in 1.5.1.  you have
to rebuild Python with this module enabled (look
for newmodule.c in the Modules/Setup file).

or you can upgrade to 1.5.2...

</F>




From news at dorb.com  Wed Jun 30 13:19:24 1999
From: news at dorb.com (Darrell)
Date: Wed, 30 Jun 1999 13:19:24 -0400
Subject: Reflection in Python?
References: <7ldglr$ita$1@nnrp1.deja.com>
Message-ID: <sose3.3972$U6.19223@newsr2.twcny.rr.com>

    # Get the testAdd method or attribute of MathTest
test=getattr(MathTest,'testAdd')

>>> class A:
...     def myX(self):
...             print 'xxx'
...
>>> a=A()
>>> af=getattr(a,'myX')
>>> apply(af)
xxx
>>> af()
xxx
        # Get the name of 'a'
>>> print a
<__main__.A instance at 7f79d0>
>>> print `a`
<__main__.A instance at 7f79d0>
>>> print str(a)
<__main__.A instance at 7f79d0>
>>> print af
<method A.myX of A instance at 7f79d0>
>>>
--
--Darrell
<anders777 at my-deja.com> wrote in message news:7ldglr$ita$1 at nnrp1.deja.com...
> test = MathTest("testAdd")
>
> where test.run() will run the "testAdd" method on
> the MathTest object?  I also want to be able to
> tell the user the name of the object (MathTest)
> and the method in case the method fails.  Any
> thoughts?
>
> Thanks,
> Anders Schneiderman
> National Journal Daily Briefings Group
>
>
>
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.





From daniel at virtualvoices.com  Mon Jun 14 23:09:50 1999
From: daniel at virtualvoices.com (Daniel)
Date: 14 Jun 1999 22:09:50 -0500
Subject: Support for mSQL?
Message-ID: <3765b5ee@news.mdc.net>

>From the 'python.org' web site the mSQL module is
no longer being supported.  Or has someone else picked 
up it's support?

-Daniel





From jeremy at cnri.reston.va.us  Tue Jun  1 23:50:46 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Tue, 1 Jun 1999 23:50:46 -0400 (EDT)
Subject: Python 2.0 (perhaps off topic)
In-Reply-To: <375431E6.6BC64E80@prescod.net>
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7imr6b$1s0$1@nnrp1.deja.com>
	<87pv3iq6ss.fsf@ev.netlab.co.jp>
	<3752C76D.604CD474@prescod.net>
	<87n1ykr0z9.fsf@ev.netlab.co.jp>
	<375372F6.A874554@prescod.net>
	<14164.7179.10346.998488@bitdiddle.cnri.reston.va.us>
	<375431E6.6BC64E80@prescod.net>
Message-ID: <14164.43542.417731.361360@bitdiddle.cnri.reston.va.us>

>>>>> "PP" == Paul Prescod <paul at prescod.net> writes:

  PP> Jeremy Hylton wrote:
  >>  I don't think this is a good example.  Even beginning users are
  >> going to have to learn a little about resource management about
  >> some point.  People should not rely on files getting deallocated
  >> and closed by Python.  All the beginners who start with JPython
  >> are going to get Java's GC.  And they're going to need to close
  >> the files by hand.

  PP> It looks like this conversation has come all of the way
  PP> around. We are talking about whether the JVM should be the
  PP> default, standard Python distribution platform. It isn't fair,
  PP> then, to use the flaws in the JVM version of Python to argue FOR
  PP> that change.

Perhaps this is what I get for jumping in in the middle of a
conversation without learning how it got started.  The example I was
responding to depended on file object being immediately finalized to
avoid leaking file descriptors.

It isn't a flaw in JPython that performs differently.  It's just a
difference.  Your skate on thin ice when you write code that depends
on implementation issues rather than the language spec.  It may be
fine if you know what you're doing (love those bytecodehacks!), but
it's hardly something you want to encourage a beginner to do.  

Of course, the implementation was sometimes posing as the language
spec until JPython came along.

  PP> "Oh well, we'll lose compatibility with PIL but JPython users
  PP> have alreadly lost that so I guess it isn't a big deal."

Tilt away <wink>.  I took issue with your non-explicit close of a
file, and didn't say anything about the merits of JPython
vs. CPython.  See my earlier post in this thread.  I think I was
unequivocal in my belief that the next version of Python needed to be
in C or C++.  

Jeremy



From da at ski.org  Wed Jun  9 23:56:31 1999
From: da at ski.org (David Ascher)
Date: Wed, 9 Jun 1999 20:56:31 -0700 (Pacific Daylight Time)
Subject: Grimoire [was: How do I defect?]
In-Reply-To: <matt-0906992046290001@scotch.mondoinfo.com>
Message-ID: <Pine.WNT.4.05.9906092053040.120-100000@david.ski.org>

On Wed, 9 Jun 1999, Matthew Dixon Cowles wrote:

> > So how do I defect from Perl to Python?  I 
> > have the Programming Python book on order.  
> > What can I read to get started ASAP?

The tutorial is what I recommend you get started with. However, I'd like
to also put a little plug for a new document I discovered by accident
(well, surfing =). Andrew Kuchling is too modest to advertise it, but his
Python Grimoire is a good document to get a feel for common tasks in
Python. It's at:

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

It's a good thing to read just after the tutorial.

Cheers,

--david ascher




From gubitz at netcologne.de  Tue Jun 22 09:39:07 1999
From: gubitz at netcologne.de (Hans Gubitz)
Date: 22 Jun 1999 15:39:07 +0200
Subject: PhotoImage
Message-ID: <lzyahctkw4.fsf@netcologne.de>

Python 1.5.1 (#1, Apr  5 1999, 00:08:45)  [GCC egcs-2.91.66 19990314 (e on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import Tkinter
>>> tk=Tkinter.Tk()
>>> image1=Tkinter.PhotoImage(width=2,height=2)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/lib/python1.5/lib-tk/Tkinter.py", line 1768, in __init__
    apply(Image.__init__, (self, 'photo', name, cnf, master), kw)
  File "/usr/lib/python1.5/lib-tk/Tkinter.py", line 1736, in __init__
    apply(self.tk.call,
TclError: unknown floating-point error, errno = 84
>>> 

What's wrong?

--
Hans Gubitz
email: gubitz at netcologne.de



From luisroco at hotmail.com  Tue Jun  8 17:10:15 1999
From: luisroco at hotmail.com (luisroco at hotmail.com)
Date: Tue, 8 Jun 1999 21:10:15 GMT
Subject: Using PIL with CGI scripts
Message-ID: <7jk0rm$o9ar@eGroups.com>

Hi all

How can a CGI scripts outputs graphics images through Python code using the Python Imaging Library package?

I just want to produce some images which has been processed using PIL and show them in a web site




From robin at alldunn.com  Tue Jun 22 01:13:38 1999
From: robin at alldunn.com (Robin Dunn)
Date: Mon, 21 Jun 1999 22:13:38 -0700
Subject: wxPython under linux
References: <7ke110$o4r$1@nntpd.lkg.dec.com>
Message-ID: <MQEb3.11$Yk2.1114283@news.randori.com>

>I've installed the wxPython rpm under RH 6.0
>and I'm having the following problems.
>
>First, python can't locate the main module.
>wx.py is located in /usr/lib/python1.5/site-packages/wxPython.
>I've tried adding this path to PYTHONPATH but it
>still has no effect.


This should work if you are using wxPython as a package, e.g from
"wxPython.wx import *".  In other words, the wxPython directory does not
need to be on the PYTHONPATH, but the directory which contains it.  (And the
site-packages directory is supposed to be there automagically.)

>
>Second, I can copy a simply script to the wxPython
>directory and run it from there, however, if I'm
>under KDE it fails with X-errors.  If I switch to
>a generic window manager the script runs fine.
>

I havn't heard of this one.  What kind of X-errors?  GTK and KDE are
supposed to play nice together and I havn't had any troubles in this area.
(Though I don't run KDE that often...)

--
Robin Dunn
robin at AllDunn.com
http://AllDunn.com/robin/
http://AllDunn.com/wxPython/  Check it out!
Try http://AllDunn.com/laughworks/ for a good laugh.






From esuzm at primrose.csv.warwick.ac.uk  Sat Jun 19 00:59:58 1999
From: esuzm at primrose.csv.warwick.ac.uk (Andrew Clover)
Date: 19 Jun 1999 04:59:58 GMT
Subject: Installing ODBC on PythonWin
Message-ID: <7kf84e$fog$1@holly.csv.warwick.ac.uk>

(References lost due to having to post through a slow newsfeed whilst
 reading replies in deja.com...)

Gordon McMillan <gmcm at hypernet.com> soothed my Access pain with:

> Repeat after me: "Access is NOT a server, Access is NOT a server".

 Erm... Access is not a server. I guess. :-)

 Access is barely competent as a client...

> If accessed through ODBC it is most definitely absolutely NOT safe. It's
> not even safe for mutiple connections from one process.
 
> Access is dandy on the desktop as an overgrown spreadsheet. It is the LAST
> thing you should use for CGI.
 
 It would seem so. I knew we'd have to move to a non-cruddy dbms at some
point, but I had imagined Access would be able to cope with a single user
(viz me) clicking Reload on a CGI page repeatedly. I guess it was naive of
me to expect anything much of MS's flagship Office suite...

 Oh well. I downloaded and installed MySQL and though the process wasn't
terribly pleasant, at least the damned thing *works* now. Anyone want to
criticise this choice of server/suggest something else?

 ObPython - actually Ob<eeghead84 at yahoo.com>'s-original posting:

> I was able to partially install mxODBC by copying all the files into the
> right place but I'm having trouble compiling the modules in vc++. 

> I'm going to pull my hair out!!! Could someone show me an easier way of
> installing ODBC?

 Hmm... earlier whilst trying to figure out what was going on, I DLed mxODBC
as a replacement for ODBC, and it worked relatively painlessly - the
distribution at http://starship.skyport.net/crew/lemburg/mxODBC-1.1.1.zip
included precompiled .pyds so there was no need to compile anything, though
it is necessary to include the ODBC directory (and DateTime) in PythonPath
somewhere.

 wwww - wish I weren't working with Windows.

-- 
This posting was brought to you by And Clover.
(Sorry.)



From claird at Starbase.NeoSoft.COM  Tue Jun 29 11:41:00 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 29 Jun 1999 10:41:00 -0500
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <Pine.GSO.3.96.990625170056.28322B-100000@saga3.Stanford.EDU>
Message-ID: <7lapec$cn0$1@Starbase.NeoSoft.COM>

In article <Pine.GSO.3.96.990625170056.28322B-100000 at saga3.Stanford.EDU>,
Jeffrey Chang  <jefftc at leland.Stanford.EDU> wrote:
			.
			.
			.
>Abandoning Tkinter seems a little bit extreme, since there are some
>applications for which its event model is well-suited. 
>
>A few weeks ago I needed to build a viewer that displays binary trees
>whose branches have variable lengths (actually, phylogenetic trees for
>biological data).  Since the trees are large, I need to be able to slide
>it around the window.  In addition, I needed each node to be clickable so
>that the user can query for more data. 
>
>Tkinter is very good for that type of thing.  All I needed to do was
>create a Canvas and then for each node, create an Oval with a callback to
>a function that handles node clicks.  Tk keeps track of the coordinates of
>each Item so that I don't have to.  Also, whenever I wanted to move the
>tree, all I had to do was:
>for widget in self._canvas.find_all():
>    self._canvas.move(widget, xoff, yoff)
>
>Of course, Tk incurrs an overhead by making everything you draw on the
>canvas its own object, but for this case, it made thigns a lot easier.
			.
			.
			.
I don't get the point of the last paragraph.  I think you're
saying that good style in canvas programming is in terms of
objects; if you're doing a lot of pixel-pushing, then you're
in trouble, because Tk wraps even pixels up as objects, with
a predictable performance hit.  Is that correct?
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From doctorwes at mindspring.com  Wed Jun  9 14:07:27 1999
From: doctorwes at mindspring.com (Wesley Phoa)
Date: Wed, 9 Jun 1999 11:07:27 -0700
Subject: Using the Windows clipboard in Python
References: <7jik35$esd$1@nntp6.atl.mindspring.net>
Message-ID: <7jmaou$hdr$1@nntp9.atl.mindspring.net>

Small modification: removed 'import win32con' as all I needed were two
constants. So the Win32 extensions are no longer required. Also, added the
ability to read/write NumPy arrays (via theclipboard.array attribute).

Wesley Phoa wrote in message <7jik35$esd$1 at nntp6.atl.mindspring.net>...
>This is a preliminary version of a module which lets you exchange data via
>the Windows clipboard, either as a string or as a 2D array. For example,
>this lets you copy and paste data between Python programs and Excel.






From Vladimir.Marangozov at inrialpes.fr  Tue Jun  8 13:51:34 1999
From: Vladimir.Marangozov at inrialpes.fr (Vladimir Marangozov)
Date: Tue, 08 Jun 1999 19:51:34 +0200
Subject: fork()
References: <000601beb1c7$682c6220$559e2299@tim> <7jjghd$9il$1@cronkite.cc.uga.edu>
Message-ID: <375D5826.23197692@inrialpes.fr>

Graham Matthews wrote:
> 
> Tim Peters (tim_one at email.msn.com) wrote:
> : OTOH, in a straightfoward mark-and-sweep GC scheme there will
> : at least be a "mark bit" in each object header.  So under that,
> : or under any form of compacting GC, the *entire live object space*
> : will get copied, regardless of whether *you* reference anything in
> : it or not.  So if RC truly sucks, what does that make GC <wink>?
> 
> But the reality is that copying memory can in fact speed applications
> up! Think about locality of reference after you have copied all live
> blocks to the start of memory.
> 

True enough. At least, it sounds true and promising ;-)
However, locality of reference (LOR) is something quite hard to measure
in practice. One has to tweak the kernel to collect some stats on this,
and even if one achieves LOR improvements on a partuclar system, say
Solaris, it could result in a LOR degradation on Linux or Windows.
Not to mention that nowadays nobody wants to play this game...

I also thought that Python suffers from very bad LOR, but this is and
will remain speculation (as well as statements regarding RC, GC or
RC+GC) until someone proves that one strategy is better or worse
than the other, in the average case, on the average system.

Graham, you said that you won't contribute a RC+GC scheme to Python, despite
your positive experience. If you change your mind and consider
devoting some spare time to give it a try, I'll devote some spare
time to help you with Python's internals and we'll see whether we
could come up with something viable, which could compete with the
actual RC scheme alone. Does this sound constructive enough? :-)

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



From tre17 at cosc.canterbury.ac.nz  Tue Jun  8 01:01:05 1999
From: tre17 at cosc.canterbury.ac.nz (Timothy R Evans)
Date: 08 Jun 1999 17:01:05 +1200
Subject: waste of resources ?
References: <375C0F91.1C73E4E3@icrf.icnet.uk> <375C16A9.5ADC4990@icrf.icnet.uk> <375C1FD1.A85565C@aw.sgi.com>
Message-ID: <c75so8370by.fsf@pc142.cosc.canterbury.ac.nz>

Gary Herron <gherron at aw.sgi.com> writes:

> Arne Mueller wrote: 
> > Im running python 1.5.2b2 on SGI power challange irix64 version 6.5.
> > Everytime a child process is created via fork(), it realy does it's job
> > but never dies, instead the number of zombie processes (as reported by
> > the 'top' program) steadily increases, though I call the sys.exit(0)
> > method for the child. Any idea how exit a process without creating a
> > zombie?
> > 
> >         thanks alot,
> > 
> >         Arne
> 
> Yes.  The parent process must execute a os.wait() or os.waitpid() in
> order for the child process's return code to be sent back to the parent
> program.  Only after that does the child process (now a zombie process)
> actually get eliminated.
> 
> Hope this helps.
> 
> -- 
> Dr. Gary Herron <gherron at aw.sgi.com>
> 206-287-5616
> Alias | Wavefront
> 1218 3rd Ave, Suite 800, Seattle WA 98101

Another thing to note is that when a child process dies the parent
will be sent a SIGCHLD signal (I think this also applies if the child
process was paused with SIGSTOP).  If you just want to ignore the
return code of the child process, use code like this:


import signal
import os

def sigchild_handler(signum, frame):
    os.wait()

signal.signal(signal.SIGCHLD, sigchild_handler)


This will run os.wait for each child process that exits.  You will
need to catch the return value of os.wait and inspect it if you care
what the process returned or whether it was killed by a signal.

--
Tim Evans



From larsga at ifi.uio.no  Fri Jun 18 10:58:49 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 18 Jun 1999 16:58:49 +0200
Subject: BiBTeX module?
References: <joe-160619991358444033@chinacat.salk.edu> <7k950a$hsb@cs.vu.nl> <slrn7miqcg.tpi.bernhard@alpha1.csd.uwm.edu> <slrn7mk9ui.dic.gobry@blanc.idiap.ch>
Message-ID: <wk1zf9wo5y.fsf@ifi.uio.no>

* Frederic Gobry
| 
| I've also started such a project as free software. The parser and
| the python interface are working correctly now. My idea was to write
| something generic for bibliographic databases, BibTeX being only the
| first module. A GNOME interface is also in progress (does some
| display but no real edition yet)

BTW: an XML version of BibTeX would be very useful. People keep asking
for it in XML/SGML newsgroups, and it just plain makes sense, since
XML is much better suited for this sort of thing than TeX is.

Just an idea.

--Lars M.



From faria at ppgia.pucpr.br  Tue Jun  1 14:32:31 1999
From: faria at ppgia.pucpr.br (Yen)
Date: Tue, 1 Jun 1999 15:32:31 -0300
Subject: keeping the fraction...?
Message-ID: <000d01beac5d$1c598f40$0301010a@poty>

Hi everybody,

    Is there someway to keep the fraction notation like in CLOS:

5 + 3 = 8 or ( 5/3 + 3/3 = 8/3 )
3    3    3

Tanx,

Henrique Faria
faria at ppgia.pucpr.br




From delgado at Mathematik.Uni-Bielefeld.DE  Wed Jun 16 10:07:28 1999
From: delgado at Mathematik.Uni-Bielefeld.DE (Olaf Delgado)
Date: Wed, 16 Jun 1999 16:07:28 +0200
Subject: Newbie again: computing attributes on the fly
Message-ID: <Pine.OSF.4.05.9906161547450.4875-100000@pinax.Mathematik.Uni-Bielefeld.DE>

Hi everybody,

here's another couple of questions to the enlightened ones. I'd like to
compute certain properties of a class instance only once and store it for
future use. I'd like to do those computations on demand only, because some
might be really, really expensive. My first try looks like this:

class x:
    def __getattr__(self, attr):
        import string
        if string.find(attr, "get_") != 0:
            val = ( lambda x = getattr(self, "get_"+attr)(): x )
            self.__dict__[attr] = val
            return val
        else:
            raise AttributeError("no such attribute")

    def get_lost(self):
	return 0 # really, really expensive :)

    def get_a_life(self):
	return 1

>>> y = x()
>>> y.lost()
0
>>> y.a_life()
1

Here, the lambda is pointless somehow, but hints to a potential desire to
add parameter support in the future (which I think I don't need, but who
knows?).

Question(s):
1) Is this 'good', or at least 'okay' style?
2) Any immediate suggestions for improvement?
3) Has anyone ever bothered to implement a more elegant or more complete
   approach, which might, for example, include such things as parameters,
   adding memorization to existing classes, you name it?

Thanks for reading,
Olaf

-- 
   ////         Olaf Delgado Friedrichs, Uni Bielefeld, Mathematik
   Olaf         Postfach 100131                  (room:  V4-109)
   `='          D-33501 Bielefeld, Germany       (phone: +49 521 106-4765)
                           http://www.mathematik.uni-bielefeld.de/~delgado




From claird at Starbase.NeoSoft.COM  Thu Jun 10 14:35:02 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 10 Jun 1999 13:35:02 -0500
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <wk909syqr4.fsf@ifi.uio.no> <375FF969.C879B4CD@iname.com>
Message-ID: <7jp0gm$fjd$1@Starbase.NeoSoft.COM>

In article <375FF969.C879B4CD at iname.com>,
Fernando Mato Mira  <matomira at iname.com> wrote:
			.
			.
			.
>About the Connoly one, the namespace issue seems to
>be solved by [incr Tcl]. But what about that "interpreted
>multiple times"? Is it just an efficiency issue, or a semantic
>nightmare (maybe both)?
			.
			.
			.
It's a different, but no less defensible, choice for
quoting rules than the one LISP made.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From holger at phoenix-edv.netzservice.de  Fri Jun 25 05:26:54 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Fri, 25 Jun 1999 09:26:54 GMT
Subject: glint?
References: <376F48B9.133D37C6@phoenix-edv.netzservice.de> <shlndavhzw.fsf@tux.gnu.franken.de>
Message-ID: <37734B5E.B01510C6@phoenix-edv.netzservice.de>

Hello Karl,

we just like to have a nice and easy tool written in python to 
install our different products on different platforms like
WinNT, Win9X, Linux. 
I started to write python-code a few weeks ago. I couldn't
imagine what a great publicity that language has. So I began
to write my own tool without looking for ready solutions.
Inside this book of Himstedt/M?tzel I found this notice
about RedHat, but I don't think, it's that what we need.

...
glint -- (Version 2.4.2, 429K)

Glint is a graphical interface to the RPM package management tool. It allows you to browse
packages installed on your system, verify and query those package. It
allows allows you to update packages with new versions and install new packages. 
...

We don't like to manage RPM-packages. We need to untar/unzip many files
from installation medium, create directories, set registry-entries or environments,
links and so on...
If we decide, that the tool is ready for public, we'll open source, I think. Then 
you could get a look at it, and perhaps it may be useful for somebody. But don't 
expect too much, I'm a newbie in python, and there may be many objects that you
may be have done in a better way...;-)

ciao,
Holger Jannsen
www.phoenix-edv.com



Karl Eichwalder schrieb:
> 
> Holger Jannsen <holger at phoenix-edv.netzservice.de> writes:
> 
> |   I'm working on an installation-tool for linux/win without any
> |   graphical interface but file-controlled.
> 
> How do you define "installation-tool"?  There's "xrpm" that's able to
> (de)install RPM packages nicely, and more.  But "xrpm" has a GUI, too.
> 
> It comes with SuSE Linux.
> 
> --
> Karl Eichwalder



From thstr at serop.abb.se  Tue Jun  1 01:32:09 1999
From: thstr at serop.abb.se (Thomas S. Strinnhed)
Date: Tue, 01 Jun 1999 07:32:09 +0200
Subject: More questions on dll's
Message-ID: <37537059.C87856A9@serop.abb.se>

Hi

I'm embedding Python in another app, but I still want to
run the interactive loop. (so I do) But, when I try to
import a module (from the prompt or using PyRun_SimpleString()
the trouble starts: only .py/.pyc files are found!!
My .dll's just won't be imported, no matter what I append
to sys.path or what directory they are in.

Using Python in it's not embedded shape everything is OK,
my .dll's are imported without fuzz an their functions are
ready to use.

I run on WinNT and MSVC++. That's what i used to build Python15 dll's
and libs, my own dll's and the app with Python embedded.

*And* (listen to this) I think this problem was not here from the
beginning, as I recall I could import dll's as late as yesterday!!
So if anyone knows what VC++-checkbox I've accidently touched;
     
              Please help me out on this one!!

Desperately Thanking You
 -- Thomas S. Strinnhed, thstr at serop.abb.se



From evan at tokenexchange.com  Fri Jun 11 10:42:52 1999
From: evan at tokenexchange.com (Evan Simpson)
Date: Fri, 11 Jun 1999 09:42:52 -0500
Subject: fork()
References: <000c01beb3c1$a3f428a0$329e2299@tim>  <199906110443.AAA02408@eric.cnri.reston.va.us>
Message-ID: <df983.24$KR2.60812@news.corridex.com>

Sounds good to me, if the mechanism reports/logs cycles in such a way that
you can figure out how to cleanly break them or avoid making them in the
first place (#10 Errors should never pass silently).  Perhaps I'm being
naive, but I have the impression that when cycles arise it's usually a
simple accident and easy to fix once you know you've done it.  The hard part
is noticing that you've pinched one off, yes?  If a cycle is really hard to
eliminate without contorting your code, you could make an explicit decision
to allow it to be collected and make sure the affected objects don't need
finalization (#11 Unless explicitly silenced).

Would it be possible (and not too expensive) to distinguish actual cycle
members from dangly bits (maximal non-cyclic subgraphs)?  If so, it might be
worth releasing the dangly bits normally through their root object after
flushing the cycle.  That way you could have cycles in a program yet still
have a place to put resource finalizers.

Tim-will-regret-those-20-Theses-yet-ly y'rs
Evan Simpson

Guido van Rossum wrote in message
<199906110443.AAA02408 at eric.cnri.reston.va.us>...
Could we get away with not calling (user) finalizers on objects in
trash cycles at all?  Since the finalization order is problematic at
best, this almost seems acceptable: in the formal semantics, objects
that are part of cycles would live forever, and as an invisible
optimization we recycle their memory anyway if we know they are
unreachable.  (I've got the feeling that I've seen this rule before
somewhere.)  We'd still get complaints "my __del__ doesn't get called"
from some users, and the answer would still be "it's in a cycle -- use
an explicit close"; but those who know what they are doing can be
guaranteed that their memory cycles get recycled.  In other words it
would be no worse than today and for some people it would be better.






From arcege at shore.net  Wed Jun 23 06:50:26 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 23 Jun 1999 10:50:26 GMT
Subject: creating dictionnaries
References: <376F82A6.82497D2B@der.edf.fr> <FQSb3.973$7X1.249189@news.shore.net> <m3bte72yz1.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <SZ2c3.1042$7X1.266606@news.shore.net>

Michael Hudson <mwh21 at cam.ac.uk> wrote:
: "Michael P. Reilly" <arcege at shore.net> writes:

:> Christian Caremoli <Christian.Caremoli at der.edf.fr> wrote:
:> : Hi,
:> : Dictionnaries can be created like that :
:> : d={'a':1,'b':2}
:> 
:> : By calling a function you create a dictionnary like that :
:> : def f(**d):
:> :    return d
:> 
:> : d=f(a=1,b=2)
:> 
:> : I would like to be able to create dictionnaries with some similar syntax
:> : like keyed tuples :
:> : d=(a=1,b=2)
:> 
:> : Is there a way to do that ?
:> 
:> In a word, No.  The structure of the language would perform a namespace
:> lookup on "a" and "b" instead of taking them as strings.  Also a tuple
:> is still an expression, and Python does not allow assignments inside
:> expressions.  This means that you would get a SyntaxError exception
:> during bytecode compilation.  And even if you didn't, it would be
:> likely that you would get a NameError exception.
:> 
:>   -Arcege
:> 
:> Note: I haven't looked at bytecodehacks, but that might let you do what
:> you wish.  Since it would be extremely non-portable, I wouldn't suggest
:> it for published or long-term code.

: What? byecodehacks can't affect syntax! I'm quite proud of that bit of
: code, but lets not blow it up into things it just isn't...

: yours,
: Michael

I only said might.. my thought was to splice a CALL_FUNCTION op in
where the expression was.  I did say that I haven't looked at the
module so I wasn't sure of it's capabilities, or if it worked inside
the parser or in the eval loop, but it seemed to be a reasonably
possible capability.

  -Arcege

PS: From what I've heard of the module, you should be proud of it.




From bwinton at tor.dhs.org  Wed Jun  9 23:50:13 1999
From: bwinton at tor.dhs.org (Blake Winton)
Date: Thu, 10 Jun 1999 03:50:13 GMT
Subject: Python 2.0
References: <000801beb2e8$05e98760$af9e2299@tim>
Message-ID: <slrn7ludnh.g2c.bwinton@tor.dhs.org>

On Wed, 9 Jun 1999, Tim Peters <tim_one at email.msn.com> wrote:
>[Graham Matthews]
>> There are two obvious answers to almost all the above problems in
>> Python
>> a) don't call finalisers for circularly referenced collected objects.
>>    Instead just collect their memory. This is not perfect but better
>>    than current Python which doesn't run finalisers on circularly
>>    referenced objects nor collect their memory.
>Solve the problem by ignoring it?  Guido would find that more
>attractive than I would <wink>.

If so, then he would have given up on this thread a long time ago...

Later,
Blake.

-- 
One Will.  One Dream.  One Truth.  One Destiny.  One Love.



From scothrel at cisco.com  Wed Jun  9 12:40:03 1999
From: scothrel at cisco.com (Scott Cothrell)
Date: Wed, 9 Jun 1999 11:40:03 -0500
Subject: [Thread-SIG] RE: tstate invalid crash with threads
References: <000101beb237$88b71d80$999e2299@tim>
Message-ID: <003201beb296$b9be7d60$d5a245ab@cisco.com>

I can confirm that we were seeing similar behavior under a highly threaded,
highly stressed situation.  We backed off on the number of threads and
slowed things down as a workaround also.  We couldn't trace it down to
Python, as we have a Java, C, Python mix that has too many suspects.

Scott Cothrell
Cisco Systems, Inc.

----- Original Message -----
From: Tim Peters <tim_one at email.msn.com>
To: Ray Loyzaga <ray at commsecure.com.au>; Guido van Rossum
<guido at python.org>; Thread-SIG <thread-sig at python.org>
Cc: <python-list at python.org>
Sent: Wednesday, June 09, 1999 12:18 AM
Subject: [Thread-SIG] RE: tstate invalid crash with threads


> Followups to the Thread-SIG, please.
>
> [Ray Loyzaga]
> > I have been playing with a multithreaded tcp based server for a while,
> > and during some serious stress testing continually hit a problem
> > involving the interpreter crashing with "Fatal Python error:
> > PyThreadState_Delete: invalid tstate".
> > ...
> > It appears to be a subtle race in PyThreadState_Delete ....
> > interestingly, if I uncomment the small sleep in "handle" in the server,
> > ie. make the server slower, it seems to work for ever ... 4m
transactions
> > before I gave up. I think the problem only comes if you are creating and
> > destroying threads quickly in parallel.
>
> PyThreadState_Delete is called from very few places, and one of them
strikes
> me as suspicious:  at the end of threadmodule.c's t_bootstrap, we have:
>
> PyThreadState_Clear(tstate);
> PyEval_ReleaseThread(tstate);
> PyThreadState_Delete(tstate);
> PyThread_exit_thread();
>
> The suspicious thing here is that PyEval_ReleaseThread releases the global
> interpreter lock, so nothing is serializing calls to PyThreadState_Delete
> made from the following line.  PyThreadState_Delete in turn does no
locking
> of its own either, but mutates a shared list.
>
> If this isn't plain wrong, it's certainly not plain right <wink>.  Matches
> your symptoms, too (very rare blowups during high rates of thread death).
>
> Guido?  I haven't been able to provoke Ray's problem under Win95, but the
> above just doesn't smell right.
>
> win95-didn't-crash-but-the-TAB-and-ESC-keys-did-swap-their-
>     meanings!-ly y'rs  - tim
>
>
>
> _______________________________________________
> Thread-SIG maillist  -  Thread-SIG at python.org
> http://www.python.org/mailman/listinfo/thread-sig
>




From ivanlan at callware.com  Wed Jun  9 19:17:57 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Wed, 9 Jun 1999 23:17:57 GMT
Subject: getargs.py
Message-ID: <375EF625.8D265AD1@callware.com>

Hi All--
I did some cutting and pasting, and put together a doc page for
getargs.py.  You can find it at:

	http://www.pauahtun.org/getargs.html

The downloadable file includes getargs.py, testarg.py and getargs.html.

The doc is not great, but it does exist.

The nameserver for www.pauahtun.org is now back online, so anyone who
had problems accessing the site should no longer have those problems. 
And the ftp daemon is running again, although it's not my registered
copy--that's at home, & I have to bring that in.  So anyone who happens
to read the ftp daemon welcome message rest assured that I do not need
to be urged to support shareware.

<existence-is-better-than-nothing-but-nothing-is-better-than-spam-and-eggs>-ly
y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From MHammond at skippinet.com.au  Thu Jun 17 18:28:24 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Fri, 18 Jun 1999 08:28:24 +1000
Subject: wierdness with pythonwin dialogs, threads & refcounts
References: <3768b75e.5469224@NNRP.UK.INSNET.NET>
Message-ID: <7kbsn1$dno$1@m2.c2.telstra-mm.net.au>

There could be a number of leaks in Pythonwin, but I dont think dialogs do.

The ref count is likely to be 6, as the pywin.mfc.Dialog base class hooks a
number of commands and messages.  Each of these hooks is a circular
reference, as a bound method is saved away.  These references are deleted
when a window recieves the WM_DESTROY message.

So in a nutshell, there is nothing obvious as to why the dialog is leaking.
The very first step would be to ensure it does not leak without threads.

Otherwise, it may come down to me tracking this down, as there are some
complicated interactions with MFC going on...

Finally, there is a "threadedGui.py" (or something) that should demonstrate
how to create "native Pythonwin threads".  However, now that we have the
threading module, there really isnt any real advantage to using Pythonwin
threads over threading.

Mark.

Dave Kirby wrote in message <3768b75e.5469224 at NNRP.UK.INSNET.NET>...
>
>I have found strange behaviour (strange to me at least) when creating
>a pythonwin dialog on a separate thread created using the standard






From schorsch at schorsch.com  Fri Jun 18 08:15:09 1999
From: schorsch at schorsch.com (Georg Mischler)
Date: Fri, 18 Jun 1999 12:15:09 GMT
Subject: do anonymous pipes normally work on NT?
References: <4D0A23B3F74DD111ACCD00805F31D8100DB90DE4@RED-MSG-50> <7kbq78$vgm$1@nnrp1.deja.com> <7kckoh$ith$1@m2.c2.telstra-mm.net.au>
Message-ID: <7kdd87$fnt$1@nnrp1.deja.com>

Mark Hammond wrote:
> Inside win32_src.zip you should find win32process.i and
> win32processmodule.cpp - the latter is generated by SWIG from
> the former.
>
> Bill is working from the same archive, so it must be there!

Do I need to swig this stuff myself? or...

> The only thing I can think of is that you didnt grab the
> archives from  my starship page...

Uh, indeed. Are the source packages on www.python.org so
heavily out of date? Someone might want to update them in
that case. The binary package is obviously build 125, but
a quick search didn't reveal the "build" number of the sources.
I'll get them from the starship in any case to make sure
it's the real thing.

Thanks a lot!

-schorsch

--
Georg Mischler  --  simulation developper  --  schorsch at schorsch.com
+schorsch.com+  --  lighting design tools  --  http://www.schorsch.com/


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From vlachid at otenet.gr  Fri Jun 25 04:51:19 1999
From: vlachid at otenet.gr (Vlachidis Costas)
Date: Fri, 25 Jun 1999 11:51:19 +0300
Subject: Newbie questions
Message-ID: <37734307.DD77AA30@otenet.gr>

Hi all
After examining  all python related modules ,I decided to use graphapp
package in C language mixed with python scripts, to a project that I,m
working on .
My problem is that I never have enough time to read all related python
documentation,So I appreciated any  help in the following (in the
graphapp package) .
1)With the askfilesave dialog box the scrollbar button sticks and fill
all scrolling area after pressing once the scrolling buttons. Is there
some tip in the dialog.c distribution file  able to resolve this
problem?
2)I'm not able to make newsubmenu function working properly.Can someone
tel me  how to use it in the midle of a menu, show submenu items and
then show the rest of the parrent menu items?I thing thatit is a object
handling related problem.
3)How to colorise menu items background?
4)Finaly how to make a python script call (in my case some openfile.py
script) from inside a  myprogramm.c file?
Best regards
Costas
-------------- next part --------------
A non-text attachment was scrubbed...
Name: vlachid.vcf
Type: text/x-vcard
Size: 366 bytes
Desc: Card for Vlachidis Costas
URL: <http://mail.python.org/pipermail/python-list/attachments/19990625/8319a61c/attachment.vcf>

From smjoshi at bellsouth.net  Sat Jun 12 11:30:16 1999
From: smjoshi at bellsouth.net (SunitJoshi)
Date: Sat, 12 Jun 1999 10:30:16 -0500
Subject: Raw Strings!
Message-ID: <P3v83.7006$Yb7.1482@news4.mia>

Hi All
        If someone could show me how to use raw-strings with variables, I
would really appreciate it.
        For e.g.: if str_path = "C:\\windows\\temp"

thanks
Sunit





From hniksic at srce.hr  Thu Jun  3 04:53:22 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 03 Jun 1999 10:53:22 +0200
Subject: pickle vs .pyc
References: <7j43pn$55l$1@mlv.mit.edu> <m3pv3eian1.fsf@atrus.jesus.cam.ac.uk> <87vhd69tdm.fsf@pc-hrvoje.srce.hr> <m3lne1iz3n.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <87k8tl7ji5.fsf@pc-hrvoje.srce.hr>

Michael Hudson <mwh21 at cam.ac.uk> writes:

> Yes, but usually you get an exception to show that somethings gone
> awry:
> 
> >>> import md5
> >>> marshal.loads(marshal.dumps(md5.md5()))
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ValueError: unmarshallable object

A good point.  However, that is exactly the kind of behaviour I get
for array objects -- in Python 1.5.1.  In 1.5.2, I can repeat the bug
you describe.  Maybe 1.5.2 is supposed to allow marshalling arrays,
but the support is buggy?

<search search grep grep>

OK, I think I got it.  In 1.5.2, marshal.c contains this code:

	else if ((pb = v->ob_type->tp_as_buffer) != NULL &&
		 pb->bf_getsegcount != NULL &&
		 pb->bf_getreadbuffer != NULL &&
		 (*pb->bf_getsegcount)(v, NULL) == 1)
	{
		/* Write unknown buffer-style objects as a string */
		char *s;
		w_byte(TYPE_STRING, p);
		n = (*pb->bf_getreadbuffer)(v, 0, (void **)&s);
		w_long((long)n, p);
		w_string(s, n, p);
	}

...which means that objects with tp_as_buffer property (whatever that
means) get marshalled as strings.  This wasn't the case in 1.5.1 and I
have no idea why it would be useful to anyone because during
unmarshalling, such simply remain strings instead of getting converted
to the original form.  Not that the conversion would work across
different architectures anyway.

I'm not sure what the new feature is supposed to buy us, but 1.5.1
behaviour looks more correct to me.



From jtravs at debian.org  Tue Jun 22 15:21:17 1999
From: jtravs at debian.org (John Travers)
Date: Tue, 22 Jun 1999 20:21:17 +0100
Subject: PyOpenGl problem
Message-ID: <99062220260100.01217@Vegus>

I have compiled PyOpenGl and installed it but whenever I run a demo I get:

File "./trees.py", line 1 in ?
	from oglpm import *
File "./oglpm.py", line 4, in ?
            from OpenGL.GL import *
File "/usr/lib/python1.5/site-packages/OpenGL/GL/__init.py", line 4, in ?
	from opengl_num import *
ImportError: /usr/lib/libMesaGL.so.3: undefined symbol: XFreePixmap.

Is this the wrong version of mesa or something?

--
John Travers

"Sometimes I think the surest sign that intelligent life exists elsewhere
in the universe is that none of it has tried to contact us!"



From MHammond at skippinet.com.au  Tue Jun  1 19:26:05 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Wed, 2 Jun 1999 09:26:05 +1000
Subject: Windows Stack Size
References: <3752C225.65C98B52@cloud9.net> <3752F154.6C444928@ricochet.net> <375413D8.C3B02D2@cloud9.net>
Message-ID: <7j1q41$3fu$1@m2.c2.telstra-mm.net.au>

>AFAIK, the code is not recursive (sometimes with GUI callbacks, you can
>get a lot deeper into the stack than you might expect :-).  It's been
>running fine under Linux for months, but under Win 95 it traps
>consistantly in some windows, _but not in all of them_, which is why I
>thought there might be a tiny-stack issue. 1 meg should be plenty of
>room.

In this case, you definately would appear to have an infinite recursion bug.
Fixing the stack size will mean it just takes longer to crash :-)


>   def __getattr__(self, attr):
>      if self.text:
>         return getattr(self.text, attr)
>      else:
>         raise AttributeError(attr)
>
>Upon further consideration, I wonder if the getattr could have something
>to do with this.

Certainly a good cause of accidental death by recursion.  For example, had
your code used "self.test" (ie, made a typo on "self.text") you would
recurse infinitely, as you would trigger a __getattr_ for the attribute
name.  However, as you dont have a __setattr__ and you define self.text in
__init__, this *looks* fine to me.  However, a few well-chosen print
statements in __getattr__ would make this immediately obvious.  Also, the
death should take a few seconds to occur - try hitting Ctrl+C in this
period - you may get a traceback telling you exactly where the problem is.

>I'd debug it myself except that I'm happily Windows-free (only creates
>problems when I have to write code for Windows users).  If there is no
>obvious answer, I have a work-around.

And what would the work-around be?

Mark.






From tim_one at email.msn.com  Fri Jun 18 17:13:34 1999
From: tim_one at email.msn.com (tim_one at email.msn.com)
Date: Fri, 18 Jun 1999 21:13:34 GMT
Subject: error? inconsistent dedent
References: <376A6796.814CBA24@phoenix-edv.netzservice.de>
Message-ID: <7kecpk$snf$1@nnrp1.deja.com>

In article <376A6796.814CBA24 at phoenix-edv.netzservice.de>,
  Holger Jannsen <holger at phoenix-edv.netzservice.de> wrote:
> Couldn't understand following error:
>
> C:\myPython>installation16.py -x -d test
> inconsistent dedent
>   File "C:\myPython\installation16.py", line 619
>     opts, args = getopt.getopt(sys.argv[1:], 'x')
>                                         ^
> SyntaxError: invalid token
>
> What does 'inconsistent dedent' mean?
> Why does that error occur?

Because you're running with the -x switch, the line number in the
traceback is off by 1, and it's not showing the correct line either.
This is fixed in the current development version of Python, but didn't
make it in time for the 1.5.2 release.  Go into your source file and
look at the lines *next* to the line it's showing you here.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From adjih at technologist.com  Thu Jun  3 08:35:51 1999
From: adjih at technologist.com (adjih at technologist.com)
Date: Thu, 03 Jun 1999 12:35:51 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <7j450j$908$2@cronkite.cc.uga.edu>
Message-ID: <7j5sr7$eb4$1@nnrp1.deja.com>

In article <7j450j$908$2 at cronkite.cc.uga.edu>,
  graham at sloth.math.uga.edu (Graham Matthews) wrote:
> Martijn Faassen (faassen at pop.vet.uu.nl) wrote:
> : This sounds like an interesting idea; I believe I saw references to
> : garbage collecting schemes that help clean up circular references
caused
> : by referencing counting. I forget where, though. :)
>
> Here. I posted comments in this thread and in a similar thread about
> such systems. They are generally all variations on mark and sweep
> collection. The more efficient ones use colouring.
>
> Martijn Faassen (faassen at pop.vet.uu.nl) wrote:
> : * In the complex case, garbage collecting is nice as it cleans up
> : circular references. There are unwanted side-effects of its
> : unpredictability, though; a possible leaking of resources (files
which
> : aren't closed), and some difficulties interfacing with C.
>
> I don't buy either of these difficulties since I have worked with and
> implemented systems which have full garbage collection and have no
> problems interfacing to C, or dealing with resources like files.

  But there will be cases where full garbage collection will be
definitly a nuisance.
  First, you must scan all the address space of the C code, which can be
a problem if you're just using the Python interpreter in a library (i.e.
don't have the ability to control or modifying the main program) ; or if
your address space is too big (I use python for instance in a program
that does an mmap of a 350 MB file).
  Second you kill the portability: you could use a subset of Python with
ANSI-C alone ; but you can't make a incremental GC in ANSI-C.
  You'd get problems when embedding Python in unusual environments ; for
instance if you embed Python in the Linux kernel, then the GC should
also trace all the pages in the swap, or be rewritten specially for that
case.

  The advantages of GC must be traded with its drawbacks:
- advantages: reclaims cyclic structures.
- drawbacks: non-portable, requires a special handling when linking with
C, high complexity for decent GC (multi-thread incremental with
finalization and support for C/C++ structures referencing Python
structures).


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From fw at cygnus.stuttgart.netsurf.de  Mon Jun  7 14:38:53 1999
From: fw at cygnus.stuttgart.netsurf.de (Florian Weimer)
Date: 07 Jun 1999 20:38:53 +0200
Subject: fork()
References: <375BE64A.EB997F24@icrf.icnet.uk>
Message-ID: <m3pv37onya.fsf@deneb.cygnus.stuttgart.netsurf.de>

Arne Mueller <a.mueller at icrf.icnet.uk> writes:

> All that woks fine now (after I spend a long time fiddling with the pipe
> and select stuff to synchronize I/O). However the forking is a waste of
> time and memory because all the children get their own large dictionary!
> Is there a way to use shared memory in python, I mean how can different
> processes access (for read only) one object without passing the keys of
> the dictionary object through a pipe (the children need _quick_ access
> to that dictionary)? 

If you've got a modern operating system, fork() doesn't copy the
whole address space at once, the memory is shared as long as it is
not modified.  Unfortunately, even if you only fetch an object from a
dictionary, Python is writing to previously shared memory behind the
scene (because the reference counts are updated).  This means that the
memory sharing mechanism of the operating system doesn't work.

I hope others will have better suggestions than I.  First of all, you
should be sure that you cannot solve this problem by throwing more RAM
on it.  RAM is so cheap these days, so it might be too expensive to spend
hours on optimizing your program.  (This obviously depends on the purpose
of your program.)

My second, more serious suggestion: Put the data contained in the
objects stored in the dictionary into a C struct, write a simple C
implementation of a hash table which stores those structs and a lookup
function which creates suitable Python objects on the fly.  This way,
there are no reference counts which are perpetually updated and thus
break he copy-on-write scheme, so the main data structure can be shared
efficiently among the processes.

Another solution is probably a Python version using the Boem garbage
collector which gets rid of the reference counts as well.  There are
some patches floating around, but I don't know whether they are suited
to recent Python releases or production use.



From adjih at technologist.com  Sat Jun  5 09:26:00 1999
From: adjih at technologist.com (adjih at technologist.com)
Date: Sat, 05 Jun 1999 13:26:00 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7iqhkb$d7s@wiscnews.wiscnet.net> <7iu7ma$d33$1@cronkite.cc.uga.edu> <02hfonzsx2.fsf@rtp.ericsson.se>
Message-ID: <7jb8h5$7rb$1@nnrp1.deja.com>

In article <02hfonzsx2.fsf at rtp.ericsson.se>,
  Kumar Balachandran <kumar*xspam*@*xspam*rtp.ericsson.se> wrote:
>
>
> >>>>> "Graham" == Graham Matthews <graham at sloth.math.uga.edu> writes:
>
>     Graham> Graham Matthews wrote in message
>     Graham> <7ik6mi$lbk$1 at cronkite.cc.uga.edu>...
>     Graham> You are envangelising (a not uncommon response when
>     Graham> someone remotely criticises Python). Stop evangelising and
>     Graham> start considering the technical issues involve (read (!)
>     Graham> other posts for what those issues are).
>
>     Graham> graham
> Here is some useful evangelizing (methinks). One of the irritating
things about
> Python is the use of whitespace in syntax. Agreed, the code is
> readable without parentheses or braces, but why not have optional
> syntactic sugar such as
>
> if
> ...
> elif
> ...
> else
> ...
> fi
>
> def
> ...
> ...
> fed
>
> while
> ...
> elihw or wend
>
> for
> ...
> rof
>
> etc. It makes the language more elegant. When I see code using
> indentation or blank likes to achieve blocking of constructs, it
> reminds me of an old language (FORTRAN I think it was called:-).
>
> The change is simple to achieve if backward compatibility is given to
> accomodate people that grew up with FORTRAN.

As I just responded in the "Alien whitespace eating nanovirus strikes
again!" it is possible, although this will be only advisory (Python will
only rely on white space).
All you need to do is to write:

fi,fed,elihw,wend,rof=0,0,0,0,0

and then you could write

for i in range(100):
  if i%5==0:
    print "%d is divisible by 5" %i
  fi
rof


-- Cedric


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From Cepl at fpm.cz  Mon Jun  7 05:32:05 1999
From: Cepl at fpm.cz (Matej Cepl)
Date: Mon, 7 Jun 1999 10:32:05 +0100
Subject: bug in os.path.samefile (Python 1.5.1)?
Message-ID: <1318D78C9072D11195C9006094EA98A72C20EF@ocesrv>

Hi,

trying to finish my port of sitemap.py (for original Sitemap.pl from ESR
see http://www.tuxedo.org/%7Eesr/sitemap-1.9.tar.gz for it), I am not
able to debug the attached procedure.

There is a huge dictionary, where key is abspath of the file and content
is list of some stuff to be pressented in the page. This procedure walks
through this dictionary (recursively) and make the writings into output
HTML.

Now, I need to check whether the processed file is in the "current"
directory or I should make steps to make recursive calling on another
directory. I guess, that the best way how to do it, is by calling
os.path.samefile(os.path.dirname(full_name),base) {see below for
context). However, Python 1.5.1 (I cannot use 1.5.2 not having
administrator's rights on WinNT 4.0 WK) complains by following report:
---------------------------
C:\Program Files\Python\Contrib\SiteMap>python report.py > neco.txt
Traceback (innermost last):
  File "report.py", line 121, in ?
    P = ReportPage(List,"","f:\\website\\",1)
  File "report.py", line 22, in __init__
    self.generate()
  File "html_page.py", line 44, in generate
    self.generate_body ()
  File "report.py", line 67, in generate_body
    if os.path.samefile(name,base):
  File "C:\Program Files\python\lib\ntpath.py", line 197, in samefile
    s2 = os.stat(f2)
os.error: (2, 'No such file or directory')
---------------------------

Do you have any idea, what's up (ntpath.samefile seems to be OK) and
what should I do (please, do not recommend me 1.5.2)?

			Thanks

				Matthew


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

    def generate_body(self,base="",curIndex=0):
        """
        prepare the list of all material in one directory and its
subdirectories
        """
        if base == "":
            base = string.lower(self.total_base)
        LCDir = len(base)
        Index = curIndex
        LenDict = len(self.klice)
        B = 1
        self.f.write("<b>" + base + "</b>\n")
        self.prlog("Base = " + base)
        self.f.write("<dl>\n")
        while ((Index + 1) <= LenDict) and B:
            full_name = string.lower(self.klice[Index])
            self.prlog("Full_name = " + full_name)
            name = os.path.dirname(full_name)
            self.prlog("Shortened name = " + name)
# THIS SEEMS TO BE A PROBLEM!!! ********
            if os.path.samefile(name,base):
                # there is not dirname in name, therefore we can
                # printout current file information
                filetitle = self.WholeDict[full_name][0]
                filedesc = self.WholeDict[full_name][1]
                if filetitle <> "":
                    self.f.write ("<dt><i>")
                    self.f.write ("<a href=\"" + full_name + "\">")
                    self.f.write (filetitle)
                    self.f.write ("</a></i></dt>\n")
                    self.f.write ("<dd>" + filedesc + "<br></dd>\n")
                Index = Index + 1
            else:
                # there is a dirname in a name
                next_base = string.lower(os.path.dirname(full_name))
                self.prlog("Next dir is " + next_base)
                self.f.write("</dl></li>\n")
                self.f.write ("<li>")
                self.generate_body(next_base,Index)
                B = 0



From mgm at unpkhswm04.bscc.bls.com  Wed Jun 23 10:19:04 1999
From: mgm at unpkhswm04.bscc.bls.com (Mitchell Morris)
Date: 23 Jun 1999 14:19:04 GMT
Subject: It's just magic, and I hate it ...
References: <slrn7mss5c.f90.mgm@unpkhswm04.bscc.bls.com> <14190.35530.772547.166720@buffalo.fnal.gov> <slrn7mvlhr.4is.mgm@unpkhswm04.bscc.bls.com> <37703621.B29B72CB@bioreason.com>
Message-ID: <slrn7n1r6p.keg.mgm@unpkhswm04.bscc.bls.com>

In article <37703621.B29B72CB at bioreason.com>, Andrew Dalke wrote:
[snip]
>
>  2) in the web server.  From the Apache FAQ
>http://www.apache.org/docs/misc/FAQ.html#nph-scripts
> that's likely not the case because versions 1.3 and later don't
>do an intermediate buffer, but you didn't say what web server
>you are using nor the version number.  That documentation may
>describe what to do.
>
[snip]
>
>						Andrew
>						dalke at acm.org

Aha! A clue! It's an old Apache, v1.2.6, as it turns out. You may have saved
my few remaining shreds of sanity!

Thanks for the pointer,
+Mitchell


-- 
Mitchell Morris

My wife and I married for better or worse ... I couldn't do better and she
couldn't do worse



From tismer at appliedbiometrics.com  Mon Jun 28 07:09:09 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Mon, 28 Jun 1999 11:09:09 GMT
Subject: ANN: Stackless Python 0.2
References: <37628EAA.C682F16C@appliedbiometrics.com> <7l6gc8$du4$1@news.tamu.edu>
Message-ID: <377757D5.25A2AFF6@appliedbiometrics.com>


Corran Webster wrote:
> 
> In article <37628EAA.C682F16C at appliedbiometrics.com>,
> Christian Tismer  <tismer at appliedbiometrics.com> wrote:
> >ANNOUNCING:
> >
> >               Stackless Python 0.2
> >           A Python Implementation Which
> >             Does Not Ese The C Stack
> >
> >What is it?
> >A plugin-replacement for core Python.
> >It should run any program which runs under Python 1.5.2 .
> >But it does not need space on the C stack.
> [snip]
> 
> I managed to get this to build eventually (thanks to the hints from Michael
> Hudson).  Looking at it, I wonder whether there's the potential here for
> more than coroutines and to write an implementation of threading within
> Python.  Each thread would presumably need its own Python stack, and a
> queueing and locking system would need to be added somehow, but because
> the Python and C stacks are no longer intertangled, switching between
> threads should be easy (as opposed to impossible <wink>).

Please be a little patient. :-)

Yesterday I finished an alpha version
of first class continuations in Python. This allows you to do
everything, and your threads are on my todo-list from the first
time.

> This wouldn't be quite as flexible as the current implimentations of
> threads at the C level - C extensions would be called in a single block
> with no way to swap threads until they return.  On the other hand, this
> would allow every platform to have some sort of threading available,
> which would be a boon.

Not really. C extensions can be written in a stackless
manner. Whenever they need to call into the interpreter,
they can do it in a conformant manner. They do so by just
using their own frame with their own executor.
You need to change your thinking here: There is no return
in that sense any longer. We just have frames in some chains,
and all local state is kept in frames.

> Unfortunately I'm not familiar enough with threads to go out there and
> implement it right away, but I thought I'd at least raise it as a
> possibility and see what people think and what the pros and cons are.

It will be available soon, and it will be nothing more than
a Python module which tames continuations to behave as threads.

ciao - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From kas at maps.magnetic-ink.dk  Sun Jun 20 12:05:38 1999
From: kas at maps.magnetic-ink.dk (Klaus Alexander Seistrup)
Date: Sun, 20 Jun 1999 18:05:38 +0200
Subject: Questions?  Time Module?
References: <rNHvjEdhm5Pp-pn2-5b4dbg1xd7dt@d185d18f8.rochester.rr.com>
Message-ID: <376D1152.DC59E2BC@maps.magnetic-ink.dk>

Benjamin Schollnick wrote:

>         I'm trying to figure out where the "time" functions are
> stored, but GREP isn't finding anything in the \lib directory??!?!?

It's in the binary timemodule.


  //Klaus

-- 
???[ Magnetic Ink ]?????????????????????????????????????????????????????????
???[ http://www.magnetic-ink.dk/ ]??????????????????????????????????????????



From holger at phoenix-edv.netzservice.de  Fri Jun 18 11:36:54 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Fri, 18 Jun 1999 15:36:54 GMT
Subject: error? inconsistent dedent
Message-ID: <376A6796.814CBA24@phoenix-edv.netzservice.de>

Couldn't understand following error:

C:\myPython>installation16.py -x -d test
inconsistent dedent
  File "C:\myPython\installation16.py", line 619
    opts, args = getopt.getopt(sys.argv[1:], 'x')
                                        ^
SyntaxError: invalid token


What does 'inconsistent dedent' mean? 
Why does that error occur?

Thanx,
Holger



From fdrake at cnri.reston.va.us  Tue Jun 15 16:57:17 1999
From: fdrake at cnri.reston.va.us (Fred L. Drake)
Date: Tue, 15 Jun 1999 16:57:17 -0400 (EDT)
Subject: Newbie: Truth values (three-valued logic)
In-Reply-To: <Pine.WNT.4.04.9906151331390.296-100000@rigoletto.ski.org>
References: <Pine.OSF.4.05.9906152037050.23333-100000@pinax.Mathematik.Uni-Bielefeld.DE>
	<Pine.WNT.4.04.9906151331390.296-100000@rigoletto.ski.org>
Message-ID: <14182.48685.82121.105521@weyr.cnri.reston.va.us>

David Ascher writes:
 > Actually, after looking at the documentation for the operator module, I
 > think that there's a documentation problem -- reading it, one would think
 > that there is a __not__ special method, but it is not supported by the
 > core.  I think a special comment should be added to the doc to that

  Done.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From sass at dresearch.de  Tue Jun 22 04:17:46 1999
From: sass at dresearch.de (Daniel =?iso-8859-1?Q?Sa=DF?=)
Date: Tue, 22 Jun 1999 10:17:46 +0200
Subject: How to call a Java superclass from jpython ??
References: <376E9415.8E002ACA@t-online.de>
Message-ID: <376F46AA.67E2ADF2@dresearch.de>

"Markus A. Greiner" wrote:
> 
> Hello,
> 
> I?m trying to port some java applications into jpython.
> Now I got a problem. I?m not able to call a java superclass from
> jpython.
> Did anybody know how to call them? Here the code I?m trying to convert:
> 
> It?s from PropertiesMetalTheme.java of the JDK Metalworks example.
> In method initcolors:
> 
>         ...
>         primary1 = super.getPrimary1()
> 
> Everything I tried in jpython failed. In some other cases replacing the
> super call into a construct like:
> 
>         ...
>         ParentClass.method(self)
> 
> works fine, but in this case
> 
>         ...
>         DefaultMetalTheme.getPrimary1()
> 
> and all other constructs failed. I?m using JPython 1.1 beta 2 so
> calls like self.super__method are obsolte and didn?t work anymore.
> (I think :-)) or better I tried without success)
> 
> Thanks Markus

You try to call a protected method. This problem should be fixed in 
JPython 1.1.


-- 

Daniel Sass



From fdrake at cnri.reston.va.us  Thu Jun 17 12:52:42 1999
From: fdrake at cnri.reston.va.us (Fred L. Drake)
Date: Thu, 17 Jun 1999 16:52:42 GMT
Subject: HTML bettering?
In-Reply-To: <wk7lp48a3a.fsf@ifi.uio.no>
References: <1318D78C9072D11195C9006094EA98A72C214D@ocesrv>
	<wk7lp48a3a.fsf@ifi.uio.no>
Message-ID: <14185.10202.444275.557283@weyr.cnri.reston.va.us>

Lars Marius Garshol writes:
 > You could very well use htmllib and just make an htmllib application
 > that writes the HTML back out again, and modify it slightly to do
 > heading numbering. I don't think HTMLFormatter would do you any good
 > for this sort of thing.

  Or you could add a skip=0 or seqnum=1 attribute to the first H?
element, and then use Grail to display it all... ;-)  (If Grail
detects header sequence information, it turns on auto-numbering.
Kinda neat! ;)


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From faassen at pop.vet.uu.nl  Wed Jun  2 05:27:28 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Wed, 02 Jun 1999 11:27:28 +0200
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7iqhkb$d7s@wiscnews.wiscnet.net> <7iu7ma$d33$1@cronkite.cc.uga.edu> <007901beac05$5c964b00$f29b12c2@pythonware.com> <7j0un1$2dm$2@cronkite.cc.uga.edu>
Message-ID: <3754F900.D742FB67@pop.vet.uu.nl>

[Gang Li is against Python on top of the JVM]
[Graham says you're just evangelizing]
[Fredrik replies Gang Li did say something: Python not on Java, and
offers some statistics on what developers do now]

Graham Matthews wrote:
> The only "reasons" he offers
> for why Python shouldn't be "tied to Java" are all hyperbole and
> evangelism. No technical content whatsoever.

Please read my own reply about strategical arguments, which are in my
opinion important as well, along with technical arguments. You offer
some strategical arguments for building Python on top of Java, after
all. For instance, the argument that building Python on top of the JVM
is good because of continued heavy development in speeding up the JVM is
not a purely technical argument, but also a strategical one.

My post didn't have much technical content but seems well-reasoned and
not overly evangelical, I think. I do agree Gang Li's arguments appeal
to emotions a bit too much, but after all I'm butting in on you being
involved in an emotional interchange yourself right now. :) (i.e. "But
then again I should have known better than to start such a thread.")

So let's try to calm down a little and go back to the technical and
strategical concerns behind our emotions. And emotions aren't that bad
after all, as long as we recognize that they're that, instead of shining
rationality. :)

Emotions-are-useful-ly yours,

Martijn




> Graham Matthews <graham at sloth.math.uga.edu> wrote:
> > Stop evangelising and start considering the technical
> > issues involve (read (!) other posts for what those issues
> > are).
> Fredrik Lundh (fredrik at pythonware.com) wrote:
> : exactly.  and since you seem to imply that you have con-
> : sidered those issues, maybe you should follow up on some
> : of the "non-envangelising" posts instead of miscrediting
> : people who don't happen to agree with you?
> 
> I have followed up Fredrik -- cf. posts on classes of references (some
> compactable, some not), colouring mark and sweep collectors, JNI, etc.
> But then again I should have known better than to start such a thread.
> 
> graham
> 
> --
>       This mountain is not visible from any inhabitable place.
>        It has no name, and since it's discovery in 1856 by a
>       British survey team, it has mysteriously resisted every
>     attempt to name it. The mountain is known simply and starkly
>      by it's original designation on the surveyor's map ... K2.



From graham at sloth.math.uga.edu  Thu Jun 10 11:12:44 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 10 Jun 1999 15:12:44 GMT
Subject: fork()
References: <000b01beb2e8$09deb660$af9e2299@tim>
Message-ID: <7joklc$hih$4@cronkite.cc.uga.edu>

Tim Peters (tim_one at email.msn.com) wrote:
: Back at Python 1.4, a set of patches was developed (by Greg Stein) that
: *did* remove the global lock.  This meant fine-grained locking of other
: stuff as needed, refcounts included.  The interpreter has changed quite a
: bit since then, so old numbers aren't quantitatively relevant anymore; but
: it was indeed a significant slowdown for single-threaded Python programs.
: 
: It's particularly acute for Python because *everything* "is boxed"; nothing
: is exempt, and refcounting is never delayed.
: 
The Sisal language has an interesting approach to this problem. Rather
than try to lock each refcount, it had n areas of memory you could
allocate out of. When changing area i you had to lock it, but all users
of the other areas could proceed without blocking. This was useful
for do reference count fiddling.

grahaml


-- 
           As you grow up and leave the playground
       where you kissed your prince and found your frog
          Remember the jester that showed you tears
                   the script for tears



From perryf at bigpond.NO.SPAM.com  Wed Jun 30 19:20:43 1999
From: perryf at bigpond.NO.SPAM.com (Perry Faulkner)
Date: Thu, 01 Jul 1999 09:20:43 +1000
Subject: ioctl in python
References: <377A71D7.470AF9DA@starvision.com> <377A7374.C6919FD3@starvision.com>
Message-ID: <377AA64A.B4C63805@bigpond.NO.SPAM.com>

Hi Angus,

The ioctl cmd code (0x80047601) is expecting a 4 byte buffer to be
passed to it. The
simplest thing is something like :

    rec = pack('BBBB', 0,0,0,0)
or
    rec = pack('L', 0)

depending on what's expected, then

    ioctl(handle, 0x80047601, rec)

By the way, the 80 part of the command also says this is an input
function, so you
are writing the data, rec in this case! For an output function you would
follow the
ioctl with an unpack of the data buffer, as well. But you still need to
supply the
packed input buffer in any case.

Hope this helps!

Regards,
    Perry

Angus MacKay wrote:

> well it seems I have sort of solved my problem. if I use a
> string for the 3rd arg then it works properly:
> >>> fo=open('/tmp')
> >>> fcntl.ioctl(fo.fileno(), 0x80047601, "hi")
> '\001\000'
> >>> fo=open('/')
> >>> fcntl.ioctl(fo.fileno(), 0x80047601, "hi")
> '\000\000'
> >>> fo=open('/proc')
> >>> fcntl.ioctl(fo.fileno(), 0x80047601, "hi")
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> IOError: (25, 'Inappropriate ioctl for device')
>
> the question is why? it will still fail with only 2 args or an int:
> >>> fcntl.ioctl(fo.fileno(), 0x80047601)
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> IOError: (14, 'Bad address')
> >>> fcntl.ioctl(fo.fileno(), 0x80047601, 1)
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> IOError: (14, 'Bad address')
>
> cheers, Angus.
>
> Angus MacKay wrote:
> >
> > is there some magic to the fcntl.ioctl() call?
> >
> > I can do these ioctl calls in C no problem (with a small wrapper):
> > (amackay at phat)~/tmp$ ./ioctltest /tmp 0x80047601
> > 0x80047601: 1
> > (amackay at phat)~/tmp$ ./ioctltest / 0x80047601
> > 0x80047601: 0
> > (amackay at phat)~/tmp$ ./ioctltest /proc 0x80047601
> > /proc: Inappropriate ioctl for device
> >
> > that was the Linux ioctl for EXT2 version.
> >
> > but in python:
> > >>> a = 0
> > >>> req = 0x80047601
> > >>> import fcntl
> > >>> fo=open('/tmp')
> > >>> fcntl.ioctl(fo.fileno(), req, a)
> > Traceback (innermost last):
> >   File "<stdin>", line 1, in ?
> > IOError: (14, 'Bad address')
> > >>>




From sjturner at ix.netcom.com  Tue Jun  8 09:17:16 1999
From: sjturner at ix.netcom.com (Stephen J. Turner)
Date: Tue, 08 Jun 1999 09:17:16 -0400
Subject: Question about envvars
References: <7jgl9n$p13$1@nnrp1.deja.com> <375BFE38.3FB3E83C@ix.netcom.com> <7jh7pc$as$1@nnrp1.deja.com> <375C3891.1D11E1A4@ix.netcom.com> <7ji1bl$955$1@nnrp1.deja.com>
Message-ID: <375D17DC.9B524EFF@ix.netcom.com>

Rob <rceci at my-deja.com> wrote:
> If I specify a -S, can I compensate by specifically saying at the
> appropriate time, "import site" ?

Yes you can.  In fact, older versions of Python required this; the
implicit "import site" on interpreter startup was added as of 1.5a4. 
For full information on the site module, refer to the documentation at:
  http://www.python.org/doc/current/lib/module-site.html

However, if you plan to go with this approach, consider that you
probably won't want to install your MDIreg extension module in the
$exec_prefix/lib/python1.5/site-packages directory (the usual place to
install site-wide extensions), since that directory is added to sys.path
by the site module.  You'll more likely have to put it in a custom
directory and make sure that directory in named in the PYTHONPATH
environment variable.

Even so, I'd be concerned about the fragility introduced into your
MDIreg module by this approach -- namely, that the
FileRegistry.exportEnvVars method cannot be used once the os module is
imported.  I'd much prefer to see you try the C++ subclassing suggested
below, so that your extension module works regardless of the module
import order.

Regards,
Stephen

> In article <375C3891.1D11E1A4 at ix.netcom.com>,
>   "Stephen J. Turner" <sjturner at ix.netcom.com> wrote:
>> Hi Rob,
>>
>>> Boy I was excited to get a reasonable explaination so quickly, but
>>> alas, it still does not work.  Here is my test script:
>>>
>>> #!/usr/local/bin/python
>>>
>>> import MDIreg
>>>
>>> reg = MDIreg.FileRegistry(1,1)
>>> reg.exportEnvVars()
>>>
>>> import os
>>> print "child has it : "
>>> os.system("env | grep MDI_REG_VERSION")
>>> print "but do I have it? :"
>>> print os.environ['MDI_REG_VERSION']
>>>
>>> ---
>>>
>>> I am interested in this "MDI_REG_VERSION" envvar which gets exported
>>> by my library.  Here is the output:
>>>
>>> child has it :
>>> MDI_REG_VERSION=10.0
>>> but do I have it? :
>>> Traceback (innermost last):
>>>   File "regtest.py", line 12, in ?
>>>     print os.environ['MDI_REG_VERSION']
>>>   File "/usr/local/lib/python1.5/UserDict.py", line 12, in __getitem__
>>>     def __getitem__(self, key): return self.data[key]
>>> KeyError: MDI_REG_VERSION
>>
>> Oops!  I just realized that the site module, which is imported by
>> default when the Python interpreter starts, imports the os module --
>> so by the time your script begins to execute, it's already too late.
>> Just for grins, you could try disabling the site module by adding the
>> '-S' option to the Python command line, as in:
>>
>> #!/usr/local/bin/python -S
>>
>> I'm not advocating this approach, since it's generally a good idea to
>> import the site module.  I'd just like to know if it works for you.
>>
>>> Any other ideas?  Any way to tell the os/posix module to "refresh"?
>>
>> None that I can see from looking in Modules/posixmodule.c.  Sorry.
>>
>>> By the way, thanks for the library implementation suggestion, but I
>>> must keep it the way it is, because other c, and c++ (at the class
>>> level) programs access this library.
>>
>> OK, how 'bout this?  Suppose you added a virtual "putenv" method to
>> your FileRegistry C++ class whose default implementation was to call
>> the global putenv function.  Then in your Python C extension, you
>> could subclass FileRegistry and override the putenv method with one
>> that modifies os.environ as described in the prior post.

--
Stephen J. Turner <sjturner at ix.netcom.com>



From Taylor.Anderson at p98.f112.n480.z2.fidonet.org  Wed Jun 30 06:06:01 1999
From: Taylor.Anderson at p98.f112.n480.z2.fidonet.org (Taylor Anderson)
Date: Wed, 30 Jun 1999 11:06:01 +0100
Subject: theads & global namespaces
Message-ID: <000015c5@bossar.com.pl>

From: Taylor Anderson <andersdt at notes.cba.ufl.edu>

Greetings all,

I am trying to share globals between threads.  Is is possible for me to
share globals between threads that call functions from different
modules?  Thus far I have had little success.  I ask this question
because  I would prefer not to put all my code in one module.

For example, I have 2 .py files: a.py and b.py.  b.py contains the
function ('foo()') to be executed  under the new thread, as well as the
global variable 'x'.  a.py contains the call to
start_new_thread(foo,()).  a.py also tries to use the global variable x.

=========================================
b.py

x = 0

def foo():
   global x
   x  = x + 1

=========================================
a.py

from b.py import *

if __name__ == '__main__':

    start_new_thread(foo, ())
    while 1:
        if x != 0:
            print x

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

So far, my code in a.py does not detect any change in x.  Any
suggestions for a workaround?

Many thanks,

Taylor



From n at n.com  Wed Jun  2 10:13:59 1999
From: n at n.com (Jr. King)
Date: Wed, 2 Jun 1999 10:13:59 -0400
Subject: Why won't this work???
References: <7j0rkr$1eeu$1@rtpnews.raleigh.ibm.com> <7j1qc5$3rj$1@m2.c2.telstra-mm.net.au>
Message-ID: <7j3e5q$nrq$1@rtpnews.raleigh.ibm.com>

Uhhhh, excuse me but, I did point out the problem.  The PyEval_CallObject
call is failing everytime.  I have it commented in the code, I guess that
wasn't good enuff.  Excuse me, but that is the way I have done and seen it
done on c.l.c and the questions are usually answered minus the attitude.
And I will be doggoned if I give out my darn email address on newsgroups
again, I don't need all that trash about porno.





From tim_one at email.msn.com  Mon Jun 21 00:02:55 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 21 Jun 1999 00:02:55 -0400
Subject: DATE ARITHMETIC
In-Reply-To: <7kj3jv$ssu$1@coco.singnet.com.sg>
Message-ID: <000301bebb9a$f0e2b6a0$619e2299@tim>

[Ajith Prasad]
> ...
> The difficulty I had with the mxDateTime module was that results
> seem to be expressed as date time objects with timestamps embedded
> with dates  and, given my limited knowledge of Python (just a few
> chapters into "Learning Python") I have no clue as to how to extract
> just the dates out. But I am sure that mxDateTime has all the
> functionality that justifies further investment in learning it.

An interactive shell (IDLE or PythonWin work great) is your best friend,
along with the "dir" function, __doc__ strings, and just *trying* stuff;
e.g.,

>>> import DateTime
>>> dir(DateTime)  # see what's in the module
['ARPA', 'Date', 'DateTime', ...]
>>> print DateTime.Date.__doc__  # what does Date do?
DateTime(year,month=1,day=1,hour=0,minute=0,second=0.0)

Returns a DateTime-object reflecting the given date
and time. Seconds can be given as float to indicate
fractions.
>>> t = DateTime.Date(1999, 6, 20)
>>> t
<DateTime object for '1999-06-20 00:00:00.00' at 794810>
>>> dir(t)   # see what t can do
['COMDate', 'Format', '__copy__', '__deepcopy__', 'absdate',
 'absdays', 'abstime', 'absvalues', 'day', 'day_of_week',
 'day_of_year', 'days_in_month', 'hour', 'is_leapyear', 'minute',
 'month', 'second', 'strftime', 'ticks', 'tuple', 'tz', 'year',
 'yearoffset']
>>> t.day  # try something
20
>>> t + 20  # try adding
<DateTime object for '1999-07-10 00:00:00.00' at 794100>
>>> # looks like "+ int" adds that many days;
>>> # wonder whether printing is prettier
>>> print t + 20
1999-07-10 00:00:00.00
>>> # yup!

That's how you get amazingly good at Python amazingly fast <wink>.

> In response to my posting, Tim Peters and Jeff Bauer have pointed out
> simpler date-time routines that could be used. Tim's routines (available
> together with the Python source code disribution) use the American-style
> "month-day-year" format while Jeff's use the less ambiguous
> "year-month-date" format which also happens to be the format in which my
> date data are in.

FYI, Tim's code was actually written as a demo of then-new operator
overloading facilities.  That it happens to do anything sensible with dates
for anyone is a reliable accident <wink>.

OTOH, don't let MDY year stop you!  Python excels at gluing things together,
forcing nasty interfaces into forms that *you* like; e.g., it's a one-liner
to write your own little function to accept YMD argument order instead,
passing it on to Date's constructor in the order *it* likes.  Do it once &
never think about it again.

> Jeff's notes at http://starship.python.net/crew/jbauer/normaldate/ make
> some useful points on the need for simple routines which handle the
> more common date arithmetic tasks without the added complications of
> dealing with specific calendars, time zones, etc.

Jeff has a very nice pkg there.  I recall that when I first tried it, I was
baffled by the results until I dug into the code & saw it special-cased the
snot out of the year 1582 <gregorian wink>.  Dates.py takes a more extreme
view of simplicity, pretending today's rules extend infinitely in both
directions.  Of course that's *wrong* -- but then so is ignoring leap
seconds <wink>.

> Both Tim's and Jeff's routines are geared towards this. I will look at
> the further references that Ivan has suggested for a deeper
> understanding of the issues.

No no no, Ivan is a calendar *expert*:  he'll only break your heart.  It's
like asking me what 1.1 + 0.9 should return <wink>.

experts-can't-abide-simple-answers-because-reality-is-a-
    frickin'-mess-ly y'rs  - tim





From MHammond at skippinet.com.au  Thu Jun  3 03:04:28 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Thu, 3 Jun 1999 17:04:28 +1000
Subject: IDLE Extension: CallTips
References: <7j2m5m$kuf$1@m2.c2.telstra-mm.net.au>
Message-ID: <7j59be$ijh$1@m2.c2.telstra-mm.net.au>

Guido suggested a few changes to CallTips.py as posted, making it work very
nicely.  I attach a version with the following changes:

* Cancelling of the Call Tips is now robust, due to knowing the magic events
to hook.

* Turns out my code was Windows specific (surprise :-)  Seems Tkinter on
windows recongises "Key-(" but Unix does not.  Changed to "<Key-parenleft>",
and it works fine on Unix (so Im told <wink>)

* Comments updated, and code cleaned slightly.

This has now been checked into the Python CVS tree - if you are interested,
please follow further updates there.

Enjoy!

Mark.



begin 666 CallTips.py
M(R!#86QL5&EP<RYP>2 M($%N($E$3$4 at 97AT96YS:6]N('1H870@<')O=FED
M97,@(D-A;&P at 5&EP<R(@+2!I92P at 82!F;&]A=&EN9R!W:6YD;W<@=&AA= T*
M(R!D:7-P;&%Y<R!P87)A;65T97(@:6YF;W)M871I;VX at 87,@>6]U(&]P96X@
M<&%R96YS+ at T*#0II;7!O<G0@<W1R:6YG#0II;7!O<G0@<WES#0II;7!O<G0@
M='EP97,-"@T*8VQA<W, at 0V%L;%1I<',Z#0H-"B @("!M96YU9&5F<R ](%L-
M"B @("!=#0H-"B @("!K97ED969S(#T@>PT*(" @(" @(" G/#QP87)E;BUO
M<&5N/CXG.B!;)SQ+97DM<&%R96YL969T/B==+ T*(" @(" @(" G/#QP87)E
M;BUC;&]S93X^)SH at 6R<\2V5Y+7!A<F5N<FEG:'0^)UTL#0H@(" @(" @("<\
M/&-H96-K+6-A;&QT:7 M8V%N8V5L/CXG.B!;)SQ+97E296QE87-E/B==+ T*
M(" @(" @(" G/#QC86QL=&EP+6-A;F-E;#X^)SH at 6R<\0G5T=&]N4')E<W,^
M)RP@)SQ+97DM17-C87!E/B==+ T*(" @('T-"@T*(" @('=I;F1O=W-?:V5Y
M9&5F<R ]('L-"B @("!]#0H-"B @("!U;FEX7VME>61E9G,@/2![#0H@(" @
M?0T*#0H@(" @9&5F(%]?:6YI=%]?*'-E;&8L(&5D:71W:6XI. at T*(" @(" @
M("!S96QF+F5D:71W:6X@/2!E9&ET=VEN#0H@(" @(" @('-E;&8N=&5X=" ]
M(&5D:71W:6XN=&5X= T*(" @(" @("!S96QF+F-A;&QT:7 @/2!.;VYE#0H@
M(" @(" @(&EF(&AA<V%T='(H<V5L9BYT97AT+" B;6%K95]C86QL=&EP7W=I
M;F1O=R(I. at T*(" @(" @(" @(" @<V5L9BY?;6%K95]C86QL=&EP7W=I;F1O
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M:&]W('1H:7,@:7,@=7-E9"X-"B @("!D968 at 7VUA:V5?=&M?8V%L;'1I<%]W
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M979E;G0I. at T*(" @(" @("!S96QF+E]R96UO=F5?8V%L;'1I<%]W:6YD;W<H
M*0T*(" @(" @("!A<F=?=&5X=" ](&=E=%]A<F=?=&5X="AS96QF+F=E=%]O
M8FIE8W1?871?8W5R<V]R*"DI#0H@(" @(" @(&EF(&%R9U]T97AT. at T*(" @
M(" @(" @(" @<V5L9BYC86QL=&EP7W-T87)T(#T@<V5L9BYT97AT+FEN9&5X
M*")I;G-E<G0B*0T*(" @(" @(" @(" @<V5L9BYC86QL=&EP(#T@<V5L9BY?
M;6%K95]C86QL=&EP7W=I;F1O=R at I#0H@(" @(" @(" @("!S96QF+F-A;&QT
M:7 N<VAO=W1I<"AA<F=?=&5X="D-"B @(" @(" @<F5T=7)N("(B("-S;R!T
M:&4 at 979E;G0@:7,@:&%N9&QE9"!N;W)M86QL>2X-"@T*(" @(&1E9B!P87)E
M;E]C;&]S95]E=F5N="AS96QF+"!E=F5N="DZ#0H@(" @(" @(", at 3F]W(&IU
M<W0@:&ED97,L(&)U="!L871E<B!W92!S:&]U;&0 at 8VAE8VL@:68@;W1H97(-
M"B @(" @(" @(R!P87)E;B=D(&5X<')E<W-I;VYS(')E;6%I;B!O<&5N+ at T*
M(" @(" @("!S96QF+E]R96UO=F5?8V%L;'1I<%]W:6YD;W<H*0T*(" @(" @
M("!R971U<FX@(B(@(W-O('1H92!E=F5N="!I<R!H86YD;&5D(&YO<FUA;&QY
M+ at T*#0H@(" @9&5F(&-H96-K7V-A;&QT:7!?8V%N8V5L7V5V96YT*'-E;&8L
M(&5V96YT*3H-"B @(" @(" @:68@<V5L9BYC86QL=&EP. at T*(" @(" @(" @
M(" @(R!)9B!W92!H879E(&UO=F5D(&)E9F]R92!T:&4@<W1A<G0@;V8@=&AE
M(&-A;&QT:7 L#0H@(" @(" @(" @(" C(&]R(&]F9B!T:&4 at 8V%L;'1I<"!L
M:6YE+"!T:&5N(&-A;F-E;"!T:&4@=&EP+ at T*(" @(" @(" @(" @(R H3&%T
M97(@;F5E9"!T;R!B92!S;6%R=&5R(&%B;W5T(&UU;'1I+6QI;F4L(&5T8RD-
M"B @(" @(" @(" @(&EF('-E;&8N=&5X="YC;VUP87)E*")I;G-E<G0B+" B
M/#TB+"!S96QF+F-A;&QT:7!?<W1A<G0I(&]R(%P-"B @(" @(" @(" @(" @
M('-E;&8N=&5X="YC;VUP87)E*")I;G-E<G0B+" B/B(L('-E;&8N8V%L;'1I
M<%]S=&%R=" K("(@;&EN965N9"(I. at T*(" @(" @(" @(" @(" @('-E;&8N
M7W)E;6]V95]C86QL=&EP7W=I;F1O=R at I#0H@(" @(" @(')E='5R;B B(B C
M<V\@=&AE(&5V96YT(&ES(&AA;F1L960@;F]R;6%L;'DN#0H-"B @("!D968@
M8V%L;'1I<%]C86YC96Q?979E;G0H<V5L9BP at 979E;G0I. at T*(" @(" @("!S
M96QF+E]R96UO=F5?8V%L;'1I<%]W:6YD;W<H*0T*(" @(" @("!R971U<FX@
M(B(@(W-O('1H92!E=F5N="!I<R!H86YD;&5D(&YO<FUA;&QY+ at T*#0H@(" @
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M:W5P(&-O9&4 at 8V%N(&%L<V\@=7-E(&ET+ at T*(" @(" @("!T97AT(#T@<V5L
M9BYT97AT#0H@(" @(" @(&-H87)S(#T@=&5X="YG970H(FEN<V5R="!L:6YE
M<W1A<G0B+" B:6YS97)T(BD-"B @(" @(" @:2 ](&QE;BAC:&%R<RD-"B @
M(" @(" @=VAI;&4@:2!A;F0 at 8VAA<G-;:2TQ72!I;B!W;W)D8VAA<G,Z#0H@
M(" @(" @(" @("!I(#T@:2TQ#0H@(" @(" @('=O<F0@/2!C:&%R<UMI.ET-
M"B @(" @(" @:68@=V]R9#H-"B @(" @(" @(" @(", at 2&]W(&ES('1H:7,@
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M3F]N92 C($-A;B=T(&9I;F0 at 86X@;V)J96-T+ at T*#0ID968 at 9V5T7V%R9U]T
M97AT*&]B*3H-"B @(" C($=E="!A('-T<FEN9R!D97-C<FEB:6YG('1H92!A
M<F=U;65N=', at 9F]R('1H92!G:79E;B!O8FIE8W0N#0H@(" @87)G5&5X=" ]
M("(B#0H@(" @:68@;V(@:7,@;F]T($YO;F4Z#0H@(" @(" @(&%R9T]F9G-E
M=" ](# -"B @(" @(" @(R!B:70@;V8 at 82!H86-K(&9O<B!M971H;V1S("T@
M='5R;B!I="!I;G1O(&$@9G5N8W1I;VX-"B @(" @(" @(R!B=70@=V4 at 9')O
M<"!T:&4@(G-E;&8B('!A<F%M+ at T*(" @(" @("!I9B!T>7!E*&]B*3T]='EP
M97,N365T:&]D5'EP93H-"B @(" @(" @(" @(&]B(#T@;V(N:6U?9G5N8PT*
M(" @(" @(" @(" @87)G3V9F<V5T(#T@,0T*(" @(" @(" C(%1R>2!A;F0@
M8G5I;&0@;VYE(&9O<B!0>71H;VX at 9&5F:6YE9"!F=6YC=&EO;G,-"B @(" @
M(" @:68@='EP92AO8BD@:6X at 6W1Y<&5S+D9U;F-T:6]N5'EP92P@='EP97,N
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M9RP at 9&9L=#H at 87)G*V1F;'0L(')E86Q!<F=S+"!D969A=6QT<RD-"B @(" @
M(" @(" @(" @("!I9B!L96XH<F5A;$%R9W,I*V%R9T]F9G-E=" \("AL96XH
M;V(N9G5N8U]C;V1E+F-O7W9A<FYA;65S*2 M(&QE;BAO8BYF=6YC7V-O9&4N
M8V]?;F%M97,I("DZ#0H@(" @(" @(" @(" @(" @(" @(&ET96US+F%P<&5N
M9"@B+BXN(BD-"B @(" @(" @(" @(" @("!A<F=497AT(#T@<W1R:6YG+FIO
M:6XH:71E;7, at +" B+" B*0T*(" @(" @(" @(" @(" @(&%R9U1E>'0@/2 B
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M<B!P;W,^-S Z('!O<STW, T*(" @(" @(" @(" @:68 at 87)G5&5X=#H at 87)G
M5&5X=" ](&%R9U1E>'0@*R B7&XB#0H@(" @(" @(" @("!A<F=497AT(#T@
M87)G5&5X=" K(&]B+E]?9&]C7U];.G!O<UT-"@T*(" @(')E='5R;B!A<F=4
M97AT#0H-"B,C(R,C(R,C(R,C(R,C(R,C(R,C(R,C(R,C(R,C(R,C(R,C(R,C
M(R,C(R,C(R,C(R,-"B,-"B, at 5&5S="!C;V1E#0HC#0II9B!?7VYA;65?7ST]
M)U]?;6%I;E]?)SH-"@T*(" @(&1E9B!T,2 at I.B B*"DB#0H@(" @9&5F('0R
M*&$L(&(]3F]N92DZ("(H82P at 8CU.;VYE*2(-"B @("!D968@=#,H82P@*F%R
M9W,I.B B*&$L("XN+BDB#0H@(" @9&5F('0T*"IA<F=S*3H@(B at N+BXI(@T*
M(" @(&1E9B!T-2AA+" J87)G<RDZ("(H82P at +BXN*2(-"B @("!D968@=#8H
M82P at 8CU.;VYE+" J87)G<RP@*BIK=RDZ("(H82P at 8CU.;VYE+" N+BXI(@T*
M#0H@(" @8VQA<W, at 5$,Z#0H@(" @(" @(&1E9B!T,2AS96QF*3H@(B at I(@T*
M(" @(" @("!D968@=#(H<V5L9BP at 82P@8CU.;VYE*3H@(BAA+"!B/4YO;F4I
M(@T*(" @(" @("!D968@=#,H<V5L9BP at 82P@*F%R9W,I.B B*&$L("XN+BDB
M#0H@(" @(" @(&1E9B!T-"AS96QF+" J87)G<RDZ("(H+BXN*2(-"B @(" @
M(" @9&5F('0U*'-E;&8L(&$L("IA<F=S*3H@(BAA+" N+BXI(@T*(" @(" @
M("!D968@=#8H<V5L9BP at 82P@8CU.;VYE+" J87)G<RP@*BIK=RDZ("(H82P@
M8CU.;VYE+" N+BXI(@T*#0H@(" @9&5F('1E<W0H('1E<W1S("DZ#0H@(" @
M(" @(&9A:6QE9#U;70T*(" @(" @("!F;W(@="!I;B!T97-T<SH-"B @(" @
M(" @(" @(&EF(&=E=%]A<F=?=&5X="AT*2 A/2!T+E]?9&]C7U\@*R B7&XB
M("L@="Y?7V1O8U]?. at T*(" @(" @(" @(" @(" @(&9A:6QE9"YA<'!E;F0H
M="D-"B @(" @(" @(" @(" @("!P<FEN=" B)7, at +2!E>'!E8W1E9" E<RP@
M8G5T(&=O=" E<R(@)2 H="P at 8'0N7U]D;V-?7V L(&!G971?87)G7W1E>'0H
M="E@*0T*(" @(" @("!P<FEN=" B)60@;V8@)60@=&5S=', at 9F%I;&5D(B E
M("AL96XH9F%I;&5D*2P@;&5N*'1E<W1S*2D-"@T*(" @('1C(#T at 5$,H*0T*
M(" @('1E<W1S(#T@=#$L('0R+"!T,RP@=#0L('0U+"!T-BP at 7 T*(" @(" @
M(" @(" @=&,N=#$L('1C+G0R+"!T8RYT,RP@=&,N=#0L('1C+G0U+"!T8RYT
6- at T*#0H@(" @=&5S="AT97-T<RD-"@``
`
end




From Klaus.Schilling at home.ivm.de  Sat Jun 19 15:27:28 1999
From: Klaus.Schilling at home.ivm.de (Klaus Schilling)
Date: 19 Jun 1999 21:27:28 +0200
Subject: Curses for MSDOS
References: <376BD33B.5A6F54BB@erols.com>
Message-ID: <87iu8kdm8v.fsf@home.ivm.de>

Swartz <swartz at erols.com> writes:

> Is there anything like curses for MS-DOS?  Thanx.

pdcurses by M. Hessling , it can be found on some sites that also 
carry djgpp.
dunno whether it can be brought to run with the pytrhon curses module  
whether by ellinghouse or andrich

Klaus Schilling



From mal at lemburg.com  Tue Jun 15 10:47:05 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Tue, 15 Jun 1999 16:47:05 +0200
Subject: Installing ODBC on PythonWin
References: <3760D550.2071295E@lemburg.com> <7k4got$avh$1@nnrp1.deja.com>
Message-ID: <37666769.55555E8A@lemburg.com>

bruises at my-deja.com wrote:
> 
> Before I go wack and take it out on my computer
> like this guy
> (http://www.web42.com/badday/badday_small.mpg)
> did, I'd like to ask for some help.
> 
> I'm trying to install an ODBC module into
> PythonWin, but it still *won't* work!
> 
> It is written that the ODBC module is included in
> the pythonwin distribution but >>>import ODBC
> says otherwise. I've d/e'd (downloaded/executed)
> Win32, but it died and said that it was looking
> for version 1.4!
> 
> I was able to partially install mxODBC by copying
> all the files into the right place but I'm having
> trouble compiling the modules in vc++.

There should be no need to compile the module: the
package comes with a precompiled version. All you have
to do is unzip the archive in e.g. \python\lib and
you're set. If you have the win32 libs installed, be
sure to rename the included odbc.pyd to win32odbc.pyd.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   199 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From balaji at platinum.com  Fri Jun 11 11:47:11 1999
From: balaji at platinum.com (b s)
Date: 11 Jun 1999 11:47:11 -0400
Subject: fork()
References: <000c01beb3c1$a3f428a0$329e2299@tim>
Message-ID: <u4sken3i8.fsf@mailserver.ab.platinum.com>

"Tim Peters" <tim_one at email.msn.com> writes:

When it comes to GC and whether 'ref count' or 'mark and sweep' is better,
here are some opinions from Ken Thompson, on GC and it's implementation in
Limbo.

http://www.computer.org/computer/thompson.htm



From aahz at netcom.com  Sat Jun 26 12:06:32 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 26 Jun 1999 16:06:32 GMT
Subject: How do can you make a variable static?
References: <3772AE33.AA764DFA@alliedsignal.com>
Message-ID: <7l2tq8$s5i@dfw-ixnews7.ix.netcom.com>

In article <3772AE33.AA764DFA at alliedsignal.com>,
H. P. Friedrichs <HPeter.Friedrichs at alliedsignal.com> wrote:
>
>Can anyone clarify? What's the workaround? For a variety of reasons I
>won't go into right now, I really can't pass the dictionary as an
>argument.

Just as a reminder, "pass the dictionary" actually means "pass a
*reference* to the dictionary"; if copying a huge block of memory is an
issue, that's simply not happening.

If you really want to do exactly what you're doing (and Darrell's
suggestion of using a class is far better), you could create something
like this (not tested, but I'm pretty sure it works):

*** module foo.py

def initFoo ( dict ) :
	global SymHanDict
	SymHanDict = dict

Now your Foo() function should work correctly if you call initFoo()
first in the main script.
--
                      --- Aahz (@netcom.com)

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



From proteus at cloud9.net  Thu Jun 17 11:06:15 1999
From: proteus at cloud9.net (Michael Muller)
Date: Thu, 17 Jun 1999 11:06:15 -0400
Subject: GNU python? (GNUbe question)
References: <3765B8B4.18DE0257@spots.ab.ca> <37667309.F9BBAE9A@mojam.com> <022c01beb7ec$f9617070$f29b12c2@pythonware.com> <37689AC2.B8F6579A@serop.abb.se> <87k8t3jh6j.fsf@home.ivm.de>
Message-ID: <37690EE7.4CE24C26@cloud9.net>

Klaus Schilling wrote:
> 
> "Thomas S. Strinnhed" <thstr at serop.abb.se> writes:
> >
> > Sounds like a good thing this GNUscript, it'll do almost anything.
> > Is there a download site ;-)
> 
> ftp://ftp.gnu.org/pub/gnu/guile
> 
> Klaus Schilling

The "do_it" function doesn't appear to be implemented in this release
;-).

=============================================================================
michaelMuller = proteus at cloud9.net | http://www.cloud9.net/~proteus
-----------------------------------------------------------------------------
There is no concept that is more demeaning to the human spirit than the
notion that our freedom must be limited in the interests of our own
protection.
=============================================================================



From loewis at informatik.hu-berlin.de  Sat Jun 19 12:15:19 1999
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 19 Jun 1999 18:15:19 +0200
Subject: GC play
References: <mp8a3.324$U6.2508@newsr2.twcny.rr.com> <3769A840.A5AE6778@compaq.com>
Message-ID: <p6qlndgrwtk.fsf@pandora.inst-inf-1.hu-berlin.de>

Greg Ewing <greg.ewing at compaq.com> writes:

> I don't think there's anything you can find out about the source
> of any other kind of object.

Although I wonder: Would it be possible to record a traceback for each
object allocated, to support purify-like leak analysis? It would
surely be very inefficient, and you'd also need to release all the
locals in the traceback - but would it work?

Regards,
Martin



From wtanksle at dolphin.openprojects.net  Thu Jun 24 20:28:26 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Fri, 25 Jun 1999 00:28:26 GMT
Subject: Thanks for Python...
References: <376ead64@anonymous.newsfeeds.com> <VrDQRddcbbs9-pn2-QdftOXYk6Hdd@Fortress> <37716901.43CB604@lockstar.com> <slrn7n2gkr.i3k.wtanksle@dolphin.openprojects.net> <37727194.D9CD6B7@lockstar.com>
Message-ID: <slrn7n5j99.k00.wtanksle@dolphin.openprojects.net>

On 24 Jun 1999 07:55:38 PDT, Mordy Ovits wrote:
>William Tanksley wrote:
>> Sure, but we've got Pilots, so that's not an open option.  Fortunately,
>> the Pilot has Quartus Forth available for it -- it's not Python, but
>> honestly, Forth fits the limitations of the device better.

>I had a Pilot for a year before I realized that it wasn;t what I needed.  It's a
>glorified organizer, when what I wanted was a pocket computer.  So I shelled out
>for the E-100 and I've never looked back.  The difference is light years, not
>miles.

I have a certain amount of leeriness about posting advocacy on unrelated
NGs, so I'll limit myself to saying that the difference is years, not
miles.  The two machines work differently: one is a glorified organizer,
the other is a pocket computer.  Both of them can fool you into thinking
they're something else, because they're both powerful.

I need a personal organizer.

>> I used Quartus on my Crypto exam to solve a modular equation after I
>> forgot how to solve modular equations (I had it use brute force).

>Wow!  I wish I had had my E100 when I was in college.  I just realized how
>awesome that would have been. :-)

Oh yeah.

Actually, I sometimes miss my old OmniGo, which was a pocket computer, in
spite of being slower than the Pilot for some tasks.  I really needed the
glorified organizer (and had been fooled into thinking that the OmniGo was
one), but the OmniGo had some abilties that the Pilot doesn't.

>The SECOND time I saw that sig was in ESR's mail.  The first was mine.  I doubt
>ESR copied mine, but who knows? :-)

You mean you wrote it?  If so, there seems to have crept into it a bug --
one of the implicit unpackings don't work.

>Actually there IS a 3 line PERL version, but it does use dc for the heavy
>lifting.  I consider that cheating.

That's what I said -- the first one I found was written in Perl, but used
bc.

>Python *is* my favorite language,

I'd generally agree, although I like Forth a lot too.

>Mordy

-- 
-William "Billy" Tanksley
Utinam logica falsa tuam philosophiam totam suffodiant!
   :-: May faulty logic undermine your entire philosophy!



From MHammond at skippinet.com.au  Mon Jun 14 19:08:49 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Tue, 15 Jun 1999 09:08:49 +1000
Subject: Embedding python in win32 (newbie question)
References: <7jh81b$ftc@wapping.ecs.soton.ac.uk>
Message-ID: <7k41va$c82$1@m2.c2.telstra-mm.net.au>

You need to compile your app with "/MD".  This is to ensure that Python and
your app agree on the C runtime library, and thus understand each others
FILE and memory objects etc.

Mark.

Ben wrote in message <7jh81b$ftc at wapping.ecs.soton.ac.uk>...
>it evaluates fine. Unfortunately, when I pass a file pointer e.g.
>
>FILE *fp;
>fp=fopen("test.py","r");
>if(!fp) exit(0);
>Py_Initialize();
>PyRun_SimpleFile(fp,"test.py");
>
>it crashes the computer with a nasty access error.
>






From jstok at bluedog.apana.org.au  Wed Jun 16 21:07:22 1999
From: jstok at bluedog.apana.org.au (Jason Stokes)
Date: Thu, 17 Jun 1999 01:07:22 GMT
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <rnbn1y2n5jk.fsf@nortelnetworks.com> <sFU93.1567$kS.233449@ptah.visi.com>
Message-ID: <eTX93.1201$lc5.5327@ozemail.com.au>

On Wed, 16 Jun 1999 21:27:52 GMT, David Thornley <thornley at visi.com> wrote:


>I know little of this Qt animal.  Is it free?  Is it open source?
>Does it run on the Macintosh?  Is it easy to set up an interface with?

See www.troll.no, but here's the synopsis:

Qt is a cross-platform GUI toolkit for Unix and Windows.  Qt releases a
special "free edition" for Unix only which does match the open source
definition, but with some licensing issues.

The problem is that linking with the Qt free edition libraries is only
allowed if you write open source software yourself.  If you want to write a
commercial application for Qt, you have to buy a license from Troll Tech. 

-- 
Jason Stokes: jstok at bluedog.apana.org.au



From mlauer at amalia-atm.rz.uni-frankfurt.de  Thu Jun  3 17:52:40 1999
From: mlauer at amalia-atm.rz.uni-frankfurt.de (mlauer at amalia-atm.rz.uni-frankfurt.de)
Date: 3 Jun 1999 22:52:40 +0100
Subject: [tkinter] two questions regarding Canvas
References: <3752a555@nntp.server.uni-frankfurt.de> <37556602.415AFA46@ingr.com> <19990604005922.07590@nms.otc.com.au> <02fb01beadf0$b358f360$f29b12c2@pythonware.com>
Message-ID: <3756eb18@nntp.server.uni-frankfurt.de>

Fredrik Lundh (fredrik at pythonware.com) wrote:
> Greg McFarlane <gregm at iname.com> wrote:
> > This is small attempt at adding editing bindings to canvas text items. 
> > This *must* have been done before, but a search of comp.lang.tcl did
> > not come up with anything, except a mention of pg_access in postgresql
> > which I could not find.  Does anyone know of it?

> http://www.pythonware.com/people/fredrik/fyi/fyi52.htm

This is a GREAT resource! I'm very much appreciating all
what will be published there in the future!

Thanks for your work.

A side note: I tried to emulate an editing canvas text item
with a windows item which has a Tkinter.Text field inside.
This is ok - but: While my other canvas items do much work within
<Any-Enter> and <Any-Leave> bindings, the text field seems
to eat up all events - which is ok for editing mode, but
not when the user for instance wants to drag around the
complete text field.

Can I change this behaviour programmatically ?

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





From samw at masu.wwa.com  Tue Jun 22 19:21:32 1999
From: samw at masu.wwa.com (Samuel G. Williams)
Date: Tue, 22 Jun 1999 23:21:32 GMT
Subject: Help with String Manipulation
Message-ID: <0UUb3.1337$el4.44932@ord-read.news.verio.net>

I know this is a very stupid question and one that has been addressed
within available documentation, however I need a quick hand. I am in need
of a routine to reverse the order of a string. I have tried several things
that have resulted in failures. It is really stupid, but I am just
learning python. I can do this readily in C, perl, tcl, but am at a loss.
I would appreciate any help I can get.

Thanks in advance.

-- 


Sam Williams  wb5yni					samurai at acm.org
Senior IS Staff Specialist			        samw at wwa.com
---------------------------------------------------------------------------

	Thinking of running your critical apps on NT?

		Isn't there enough world suffering?



From gvinther at atex.com  Tue Jun  8 14:04:25 1999
From: gvinther at atex.com (Gordon Vinther)
Date: Tue, 8 Jun 1999 14:04:25 -0400
Subject: Tkinter doesn't work with Tcl/Tk 8.0.5 and probably 8.1.
Message-ID: <928865040.722.4@news.remarQ.com>

Tried to get Python 1.5.2 to work with Tcl/Tk 8.0.5, Aix 4.2.1 get the
following error:
from Tkinter import *  ............. this works
root=Tk()    this fails  with the message:

TclError: invalid command name "tcl_findLibrary"

In addition I still get one compile error under AIX :
unable to convert unsignedlong* to int* in socketmodule.c






From tim_one at email.msn.com  Fri Jun 25 02:38:05 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 25 Jun 1999 02:38:05 -0400
Subject: Environment inheritance under Windows NT 4, Attempt 2
In-Reply-To: <Pine.SGI.3.95q.990623204139.149495A-100000@masox202>
Message-ID: <000901bebed5$480437a0$d29e2299@tim>

[Milton L. Hankins, has trouble with envar WINDIR in NT]

    http://support.microsoft.com/support/kb/articles/Q100/8/43.asp

will tell you that WINDIR is "special".  Not exactly clear how, though.
Best I can suggest is that you fight with NT until it delivers a correct
value for WINDIR as soon as you boot; you really shouldn't be messing with
this envar.

search-ms-support-for-"windir"-to-find-out-why-not<wink>-ly y'rs  - tim





From tjreedy at udel.edu  Thu Jun  3 09:58:53 1999
From: tjreedy at udel.edu (Terry Reedy)
Date: 3 Jun 1999 13:58:53 GMT
Subject: All this self dot is making me...
References: <Vn653.468$HD.99317@typhoon-sf.snfc21.pbi.net>
Message-ID: <7j61mt$g7n$1@news.udel.edu>

>For those of us who have less cartilage left in their hands, would it be
>acceptable to use "my." or better "me."

'Acceptable' to whom?  Python is a free language; use it as you wish 8).

For private work, I abbreviate 'self' to 's' and 'other' (in __add__(), etc) 
to 'o'.  This works precisely because the existing conventions make it 
obvious (to me) what is meant.  

TJ Reedy




From schorsch at schorsch.com  Mon Jun 21 08:03:21 1999
From: schorsch at schorsch.com (Georg Mischler)
Date: Mon, 21 Jun 1999 12:03:21 GMT
Subject: do anonymous pipes normally work on NT?
References: <4D0A23B3F74DD111ACCD00805F31D8100DB90DFB@RED-MSG-50> <7kjgt1$noj$1@nnrp1.deja.com> <7kjv1k$ro6$1@nnrp1.deja.com>
Message-ID: <7kl9m4$8dg$1@nnrp1.deja.com>

Florent Heyworth wrote:
> Georg Mischler  wrote:
> > Bill Tutt wrote:
> > > After digging through this stuff, I went back and looked at
> > > the code I had written for win32popen.cpp.
> > > I realized that I had forgotten why I had
> >
> > Uh oh, Bill...
> >
> > If you should ever happen to visit the Munich Oktoberfest,
> > please remind me that I owe you a few beer!
> >
> >
> > Thanks a lot for all the tips and bug hunting!
> >
>
> Well......
>
> (I) never dug up the source - but I still spent 3 hours of my time
> trying to get your code to work...... do I get a beer at the Fest 8-)?


Oops! I knew this usenet thing was going to be expensive...
Just warn me beforehand when you turn up there. And remember,
as the name suggests, the Munich Oktoberfest is in September. ;)


Have fun!

-schorsch

--
Georg Mischler  --  simulation developper  --  schorsch at schorsch.com
+schorsch.com+  --  lighting design tools  --  http://www.schorsch.com/


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From kvrekp at hotmail.com  Thu Jun 10 07:48:15 1999
From: kvrekp at hotmail.com (kvrekp at hotmail.com)
Date: Thu, 10 Jun 1999 11:48:15 GMT
Subject: Free Sex Links  7546
Message-ID: <3CN73.93097$pl3.57568@newreader.ukcore.bt.net>

For Nasty Sex links visit:

http://website.lineone.net/~simonsi/sexlinks.htm
ztjorzgfvo




From fdrake at cnri.reston.va.us  Tue Jun 29 15:10:33 1999
From: fdrake at cnri.reston.va.us (Fred L. Drake)
Date: Tue, 29 Jun 1999 19:10:33 GMT
Subject: Binary pickle portability
In-Reply-To: <Pine.OSF.4.05.9906292018360.28230-100000@pinax.Mathematik.Uni-Bielefeld.DE>
References: <Pine.OSF.4.05.9906291852100.1376-100000@pinax.Mathematik.Uni-Bielefeld.DE>
	<Pine.OSF.4.05.9906292018360.28230-100000@pinax.Mathematik.Uni-Bielefeld.DE>
Message-ID: <14201.6697.811188.198805@weyr.cnri.reston.va.us>

Olaf Delgado writes:
 > After looking a bit closer, I found that the problem is probably related
 > to the Numeric module (i.e., pickling matrices) and doesn't depend on
 > whether binary or ordinary mode is used. So please don't bother now!
 > Still, I am interested in whether people use binary pickles.

  The last release of Grail uses binary pickles to store caches of
parsed bookmark files.  There are no floating point or NumPy objects
in the pickles, so that wouldn't test cases you may be interested in.
My choice to use binary pickles was entirely based on the performance
improvement; I doubt there are any substantial cross-platform issues
for bookmark files, especially when only used as caches!


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From orlov at diasoft.ru  Thu Jun 17 05:02:25 1999
From: orlov at diasoft.ru (Oleg Orlov)
Date: Thu, 17 Jun 1999 13:02:25 +0400
Subject: ref counting, access from python
Message-ID: <7kad31$o7f$1@ns.mtu.ru>

Hello.
How can i get objects current ref count from my python code?
Thanks.
--
====================
Oleg Orlov
orlov at diasoft.ru





From Drew.Whitehouse at anu.edu.au  Mon Jun 21 19:00:27 1999
From: Drew.Whitehouse at anu.edu.au (Drew Whitehouse)
Date: 22 Jun 1999 09:00:27 +1000
Subject: WIN32/python/COM/Multimedia
Message-ID: <wlyahdjh10.fsf@anu.edu.au>

Hi, I was wondering if anyone has any code/examples of using python to
control the playing back of things like audio files and CD's via
python ? I've used makepy to generate the media player COM interface
but have been unsuccessful (except for being able to pop up the
"About" box....) in wrapping my simple UNIX brain around what's going
on. Ideally I'd like to be able to do this without having to use the
GUI, but if necessary I could have another program send commands to
an iconified/hidden GUI app (maybe looking like a COM server ?). BTW I
have seen the winsound module, but I'd like something a more
powerful/flexible.

-Drew

-- 
;; mailto:Drew.Whitehouse at anu.edu.au   http://anusf.anu.edu.au/~drw900/
;; Viz programmer, Australian National University Supercomputer Facility




From suzuki611 at okisoft.co.jp  Tue Jun  1 06:40:17 1999
From: suzuki611 at okisoft.co.jp (Hisao Suzuki)
Date: 01 Jun 1999 19:40:17 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
	<87pv3iq6ss.fsf@ev.netlab.co.jp> <7iv86a$iud@news.dns.microsoft.com>
Message-ID: <u3r9nwtd9q.fsf@ares.sys1.ngo.okisoft.co.jp>

In article <7iv86a$iud at news.dns.microsoft.com>,
"Vadim Chugunov" <chega_ at yahoo.com> wrote:
> Right, but java finalizers will not be called till the moment the memory is
> collected.
> Depending on the GC configuration and the program's memory consumption this
> may be quite a while.

You are right.  The JLS (Java Language Specification) says:
	The Java language does not specify how soon a finalizer
	will be invoked, except to say that it will happen before
	the storage for the object is reused. (sect.12.6)

In fact, there is no guarantee that all garbages will be reused,
up to their last one byte, within the life time of the java
process, and there is also no guarantee that ALL finalizers will
be eventually invoked.  It is an issue of the quality of JVM
implementation, you know.

> Can you explain exactly why is it more error prone?
> If something can be done automatically, without writing explicit
> deallocation code,
> why can't we let it be that way?

Perhaps it is safe to say that destructors with real GC are more
error prone.  (The present Python is safer in this respect!)

In general, real GC simulates infinite memory virtually, and one
should not rely on finalizers; objects remain virtually forever.
With real GC, finalizers should be used only as the last resort.

BTW, Mr. Matz has invented and implemented his excellent
language, Ruby, which is a sort of Perl's successor with some
intrinsic object-oriented features.  IMHO, we must have respect
for his opinion as such or, at the very least, as the evangelist
of his own language.  See http://www.netlab.co.jp/ruby/

--===-----========------------- Sana esprimo naskas sanan ideon.
SUZUKI Hisao            suzuki611 at okisoft.co.jp, suzuki at acm.org.



From tim_one at email.msn.com  Tue Jun 29 11:56:59 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 29 Jun 1999 11:56:59 -0400
Subject: ANN: Stackless Python 0.2
In-Reply-To: <O3JyuHAD+Ke3Ew2y@jessikat.demon.co.uk>
Message-ID: <000101bec248$0549a960$6d9e2299@tim>

> ...
> Py_BEGIN_ALLOW_THREADS
> ThreadedSpam();
> Py_END_ALLOW_THREADS
>
> it's-only-confusing-if-you-think-about-it-too-much<wink>-ly y'rs  - tim

[Robin Becker]
> so do I then have to poll ThreadedSpam myself to see if it's finished or
> is there a python api version of mutexes/locks etc.

You're in C now -- you do anything you need to do, depending on the
specifics of ThreadedSpam (which was presumed to be a pre-existing
thread-safe C routine).  The snippet above clearly assumes that ThreadedSpam
"is finished" when it returns from the call to it.  If your flavor of
ThreadedSpam doesn't enjoy this property, that's fine too, but then you do
whatever *it* requires you to do.  This is like asking me whether the right
answer is bigger than 5, or less than 4:  how the heck should I know <wink>?

You can certainly use Python's lock abstraction in your own C code, but it's
unlikely a pre-existing C function is using that too.

python-doesn't-restrict-what-c-code-can-do-except-to-insist-that-it-
    acquire-the-lock-before-returning-to-python-ly y'rs  - tim





From MHammond at skippinet.com.au  Wed Jun  9 18:41:08 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Thu, 10 Jun 1999 08:41:08 +1000
Subject: COM dates
References: <816010E2456BD111A48700805FBBE2EEFDF353@ex-quebec-u1.baan.com>
Message-ID: <7jmqev$g8v$1@m2.c2.telstra-mm.net.au>

Check out the win32 extensions help file.  This has a discussion of COM
dates.  In a nutshell, you use the int() function to convert it to a
standard Python time module date, and you can assign a standard Python time
module date to a COM date field and it will do the right thing.

Mark.

Gaetan Corneau wrote in message
<816010E2456BD111A48700805FBBE2EEFDF353 at ex-quebec-u1.baan.com>...
>Hello,
>
>I have a COM object with a field calles "StartDate". When I type:
>
>print Obj.StartDate, I get:
>
><PyTime:1/1/70 12:00:00 AM>
>
>Two questions:
>
>1) What is that PyTime object? Is it documented somewhere?
>
>2) Suppose I want to set StartDate to January 1, 1999, what do I write in
my
>script?






From tjreedy at udel.edu  Sat Jun 12 18:59:01 1999
From: tjreedy at udel.edu (Terry Reedy)
Date: 12 Jun 1999 22:59:01 GMT
Subject: extracting contents from a text file
References: <375B5880.992B0C6C@mediamanager.com.sg>
Message-ID: <7juonl$q1f$1@news.udel.edu>

In article <375B5880.992B0C6C at mediamanager.com.sg>, john at mediamanager.com.sg 
says...

>I would like to extract the text contents residing in another server and
>display it as html format using python. Can anyone help me I am new to
>python.

You need three separate modules to 
1. retrieve text file (1 of url libs or possible FTP lib);
2. process file (hope string or re rather than parser is sufficient);
3. write html (for instance, HTMLgen on starship site -- see python.org).

TJ Reedy




From thomas.heller at ion-tof.com  Thu Jun 24 04:54:10 1999
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Thu, 24 Jun 1999 10:54:10 +0200
Subject: mxCGIPython question (was: Build the Python library into the binary)
References: <376F9518.D3252CEE@interet.com> <7kolnv$gp6$1@nnrp1.deja.com>
Message-ID: <015e01bebe1f$202e9560$4500a8c0@thomasnotebook>

> Hi
> 
> did you have a look at mxCGIPython at
> http://starship.skyport.net/~lemburg/mxCGIPython.html ?
> This is a single executable not shared library with the
> standard Python library embedded.
> 
> Cheers
> --
> Florent Heyworth
> 
Works great, except for one problem (under Windows):
you cannot use any extension modules, because they do
not find python15.dll (or, worse, they find the wrong
python15.dll).

Is it possible to build a python15.dll containing the
standard library or is it only possible to freeze executables?

Thomas Heller




From proteus at cloud9.net  Wed Jun  2 21:19:13 1999
From: proteus at cloud9.net (Michael Muller)
Date: Wed, 02 Jun 1999 21:19:13 -0400
Subject: idle processing
References: <3755bf01@nntp.server.uni-frankfurt.de>
Message-ID: <3755D811.61AEC34@cloud9.net>

mlauer at asmoday-atm.rz.uni-frankfurt.de wrote:
> 
> Hi,
> 
> is there an appropriate place to insert a method
> which does some calculations while the application
> is "idle" (user is not using it or just moving
> the mouse...) - somewhere within the tkinter-
> mainloop ?
> 
> --
>     Regards & Gruesse from Mickey @ http://www.Vanille.de
>   ---------------------------------------------------------
>   How could anyone know me - when I don't even know myself ?

The "after" function allows you to execute code after a given period (in
milliseconds) or in idle time:

   def someFunction():
      " Your code here "

   win = SomeWindowClass()
   win.after('idle', someFunction)
   win.mainloop()

This will cause "someFunction" to be evaluated *once* in the message
loop 
when there are no events left in the event queue.

You should be able to "chain" these by adding another "after idle" call 
_inside_ the function, thus allowing you to have a function that is
performed
whenver the message queue is idle. This doesn't seem to work quite right
for
me, though.  The idle time function ends up blocking the message queue.

It does seem to work if you force a delay between iterations, like so:

   win = SomeWindowClass()

   def otherFunc():      
      win.after('idle', someFunc)
   
   def someFunc():
      "idle time processing here"
      # give the system a second to catch it's breath
      win.after(1000, otherFunc)   

   win.after('idle', someFunc)

=============================================================================
michaelMuller = proteus at cloud9.net | http://www.cloud9.net/~proteus
-----------------------------------------------------------------------------
Mantra for the 60's: Tune in, turn on, drop out
Mantra for the 90's: Turn on, jack in, jerk off
=============================================================================



From Just.van.Rossum at p98.f112.n480.z2.fidonet.org  Wed Jun 30 14:21:37 1999
From: Just.van.Rossum at p98.f112.n480.z2.fidonet.org (Just van Rossum)
Date: Wed, 30 Jun 1999 19:21:37 +0100
Subject: Python Editor
Message-ID: <000015c9@bossar.com.pl>

From: Just van Rossum <just at letterror.com>

>In article <AJqe3.1318$24.76 at news01.t-net.net.ve>,
>Jose A. Antich <jantich at cantv.net> wrote:
>>
>>   Does anyone know if there's any Python Editor for the MAc?

Python on the Mac is shipped with a Mac native IDE.

At 4:39 PM +0000 6/30/99, Aahz Maruch wrote:
>IDLE, a debugger/editor written in Python.  You'll need Python 1.5.2 and
>Tcl/Tk 8.0 (*not* 8.1).

IDLE does not work well on the Mac.

Just




From greg.ewing at compaq.com  Tue Jun 29 16:46:15 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Wed, 30 Jun 1999 08:46:15 +1200
Subject: file.tell problems (1.5.2/Win32)
References: <7krd1r$2oa$1@holly.csv.warwick.ac.uk> <7l3aao$65c$1@nnrp1.deja.com> <3776ED63.35D87F89@compaq.com> <37782C9E.3061AF25@geekspace.com>
Message-ID: <37793097.17D1BDA6@compaq.com>

Joshua Rosen wrote:
> 
> This leads me to believe that `append mode' is intended to `append' at the
> *current* point in the file, which should, for consistency's sake, start at 0.

I just tried an experiment (on Windows) using "a+"
mode. It seems that f.tell() is telling the truth,
but the truth is strange.

What "a"/"a+" modes seem to mean is "open the
file positioned at 0, but if the first operation
is a write, go to the end of file before doing
it". Which seems quite bizarre to me - wouldn't
it be simpler to just go to the end of the
file on opening?

Greg



From tim_one at email.msn.com  Sun Jun 13 16:36:08 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 13 Jun 1999 16:36:08 -0400
Subject: OO (was: Why is tcl broken?)
In-Reply-To: <3763C746.5CA8FE78@iname.com>
Message-ID: <000a01beb5dc$5e88aa80$ae9e2299@tim>

[Fernando Mato Mira]
> Although Dylan never really made it into "Fernando's elegibility list"
> because it was not a serious contender against Scheme and CL yet, it
> definitely got banished the day the Lisp syntax was dropped. I'm still
> waiting for the day when the Dylan fans will reinstate the alternative.

OK!  Now we know the answer Fernando is looking for:  Tcl is fatally broken
because its syntax differs from Lisp's <0.9 wink>.

admit-it-up-front-and-maybe-this-will-go-away?-na-ly y'rs  - tim





From bwarsaw at CNRI.Reston.VA.US  Fri Jun 11 09:55:50 1999
From: bwarsaw at CNRI.Reston.VA.US (Barry A. Warsaw)
Date: Fri, 11 Jun 1999 09:55:50 -0400 (EDT)
Subject: fork()
References: <000c01beb3c1$a3f428a0$329e2299@tim>
	<199906110443.AAA02408@eric.cnri.reston.va.us>
Message-ID: <14177.5478.427212.303048@anthem.cnri.reston.va.us>

>>>>> "Guido" == Guido van Rossum <guido at CNRI.Reston.VA.US> writes:

    Guido> Could we get away with not calling (user) finalizers on
    Guido> objects in trash cycles at all?

    Guido> We'd still get complaints "my __del__ doesn't get called"
    Guido> from some users

It's more mechanism, but if you go this far, it's probably best to
call __del__ exactly once for each object, as Tim suggests.  It's what
Java people know and love, and it seems to make more sense than it
never getting called.

-Barry



From Klaus.Schilling at home.ivm.de  Wed Jun 16 14:53:52 1999
From: Klaus.Schilling at home.ivm.de (Klaus Schilling)
Date: 16 Jun 1999 20:53:52 +0200
Subject: OO (was: Why is tcl broken?)
References: <000801beb745$02768220$979e2299@tim> <871zfdp73f.fsf@home.ivm.de> <wklndk8ly6.fsf@ifi.uio.no> <873dzsma6b.fsf@home.ivm.de> <wk3dzs87mr.fsf@ifi.uio.no>
Message-ID: <87lndkkmda.fsf@home.ivm.de>

Lars Marius Garshol <larsga at ifi.uio.no> writes:

> * Klaus Schilling
> | 
> | No, in the sense `Steele merely rediscovered scheme, not bring it
> | into being'
> 
> I don't understand what you mean by this. The two people who created
> Scheme originally were Sussman and Steele. If that isn't bringing it
> into being I don't know what would be.

No, they discovered Scheme, they did not create it.

Klaus Schilling



From n89553 at hrz.uni-paderborn.de  Mon Jun 28 10:10:12 1999
From: n89553 at hrz.uni-paderborn.de (m liss)
Date: Mon, 28 Jun 1999 16:10:12 +0200
Subject: ssh with telnetlib
Message-ID: <37778244.C4528C5B@hrz.uni-paderborn.de>

Hello,

has someone already made telnetlib.py support ssh ?
I want to use Python for remote administration.

regards, 

Mirko

-- 
M Li?, <n89553 at hrz.uni-paderborn.de>



From suzuki611 at okisoft.co.jp  Wed Jun  2 05:19:21 1999
From: suzuki611 at okisoft.co.jp (Hisao Suzuki)
Date: 02 Jun 1999 18:19:21 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
	<87pv3iq6ss.fsf@ev.netlab.co.jp> <7iv86a$iud@news.dns.microsoft.com>
	<u3r9nwtd9q.fsf@ares.sys1.ngo.okisoft.co.jp>
	<87so8bnsd6.fsf@ev.netlab.co.jp>
Message-ID: <u37lpndko6.fsf@ares.sys1.ngo.okisoft.co.jp>

In article <87so8bnsd6.fsf at ev.netlab.co.jp>,
Yukihiro Matsumoto <matz at netlab.co.jp> wrote:
> |Perhaps it is safe to say that destructors with real GC are more
> |error prone.  (The present Python is safer in this respect!)
> Safer than Java, probably.
> 
> Ruby garantees (well, trying to garantee) that ALL finalizers will be
> called before process termination, so that Ruby is as safe as Python,
> at least theoretically.

This sounds a sort of advocacy of your Ruby...

Any way, the idea of destructors with "real" GC is a conceptual
contradiction.  Such GC emulates infinite memories and objects
are never destroyed *virtually*.  Thus the destructors should
never be invoked with real GC.

Certainly you may make your GC routine retrieve not only
memories but also various resources (such as file descriptors),
but it is highly error prone to make the routine invoke
ARBITRARY destructors specified by arbitrary users.  Such
invocation may introduce serious indeterminism or
unpredictability into the behavior of user's program.

# Well, in this respect, Ruby might be more indeterministic
# than Python.

Even if Python has "real" GC someday, IMHO, the GC should be
used only to imitate infinite resources.  For the sake of
conceptual simplicity, the future Python should still use
reference counting both to invoke __del__ methods and to free
storages; the GC routine should be used as the last resort.

# Or, deprecate both __del__ methods and reference counting?
# --- it will result in another language than Python.

--===-----========------------- Sana esprimo naskas sanan ideon.
SUZUKI Hisao            suzuki611 at okisoft.co.jp, suzuki at acm.org.



From graham at sloth.math.uga.edu  Fri Jun  4 13:07:11 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 4 Jun 1999 17:07:11 GMT
Subject: "The Python Way"
References: <001101beae49$39af7bc0$949e2299@tim> <7j8pra$55v$4@cronkite.cc.uga.edu> <H1T53.19$KR2.41684@news.corridex.com>
Message-ID: <7j913v$6lk$2@cronkite.cc.uga.edu>

Evan Simpson (evan at tokenexchange.com) wrote:
: I get the distinct impression the the reception of *any* proposal to alter
: Python is likely to be vastly improved by the presence of an implementation,
: however primitive, and will allow you de facto last word on a lot of
: arguments.  My own pet suggestions have died for lack of this, apart from
: "and while", for which some kind soul coded a mutant cousin.

You are wrote. People here love to see code. And that would be a great
proof of concept. But the problem with this line of reasoning is that
people have other things to do, and do not necessarily have the time
to provide such code. That should not however be grounds for writing an
idea off.

graham
-- 
                    One day in a nuclear age 
                  They may understand our rage
              We build machines that we can't control
              And bury the waste in a great big hole



From mwh21 at cam.ac.uk  Sun Jun 13 12:11:05 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 13 Jun 1999 17:11:05 +0100
Subject: Module magic methods
References: <WgP83.7966$Yr2.129670@news2.rdc1.on.home.com>
Message-ID: <m3g13wt71h.fsf@atrus.jesus.cam.ac.uk>

arcanedragonNOSPAM at NOSPAMhome.com (Chris Herborth) writes:

> Do modules have magic methods?  Specifically, I'm wondering if there's a
> __del__() or __destroy__() method that I could use to clean up after myself
> a little when the module is about to be unloaded...

Don't think so. However you can fake it by creating a class instance
and storing it in a module global, like so:

class ByeBye:
    def __del__(self):
        print "bye bye"

_byebye = ByeBye()

Then, so long as no other entity has a reference to _byebye, when the
module is unloaded _byebye will be collected and the __del__ method
called.

You know that, in general, modules aren't unloaded, except at
interpreter shutdown time? And then the execution state is so
unpredictable you can't really do anything?

HTH
Michael

FWIW, I've often thought it would be funky to be able to define
__add__ or __call__ etc methods to modules. Pointless, I'd have to
admit, but still funky.
 
> -- 
> ----------================================================----------  _
> Chris Herborth, R&D, Technical Writing                               | \  _
> Arcane Dragon Software                 Arcane Dragon -==(UDIC)==-    | < /_\
> arcane dragon at home dot com          DNRC Holder of Past Knowledge |_/ \_



From rozzin at geekspace.com  Wed Jun 23 21:04:36 1999
From: rozzin at geekspace.com (Joshua Rosen)
Date: Wed, 23 Jun 1999 21:04:36 -0400
Subject: How to obtain an instance's name at runtime?
References: <7kqpu8$7u1$1@nnrp1.deja.com>
Message-ID: <37718424.C70BB9DE@geekspace.com>

"Dinu C. Gherman" wrote:
> Question: How to obtain an instance's name at runtime?

	An *instance's* name? Instances have names?

> >>> c.__name__
> 'c'

	Are you actually looking to have the instance tell you what the variable is?
Wouldn't that be both [0]pointless and [1]meaningless, because [1]you already
know what the variable is, otherwise you can't use it, and [2]the variable-name
isn't necessarily the least bit unique, with regard to the variable to which it
referrs?
	Classes have names, and I don't beleive that instances do, because, well...,
the name of the class tells what the class is (as far as `what kind of object
it is'--I don't really know how to expand on that, except with an analogy like
`what species an animal is'), and it usually goes without saying that `what
kind of object it is', with regard to C(), is C().__class__ (or
C().__class__.__name__, if you want a label rather than an object).


		-Rozzin.




From amaranda at nospam.com  Mon Jun  7 06:32:25 1999
From: amaranda at nospam.com (Alex Maranda)
Date: Mon, 07 Jun 1999 11:32:25 +0100
Subject: threads
References: <000a01beaef4$a53dda20$f19e2299@tim> <Pine.A32.3.90.990605002137.16533A-100000@elvis.med.Virginia.EDU>
Message-ID: <375B9FB9.59D3@nospam.com>

Steven D. Majewski wrote:
>  I recall this was one of the reasons I favored coroutines over threads --
> the explicit yield means that you don't need locks. Sort of like the
> non-preemptive multitasking on the Mac. The Mac, of course, is now
How do coroutines work on MP machines? They don't - just a reminder of
how this thread started :)

> moving towards a fully preemptive multitasking model -- which makes
> better sense for a OS because you can't control all of the code in all
> of the apps and ensure that they all behave like good citizens. However,
> in a single app, that's not a problem, and cooperative multitasking
> with coroutines is a simpler model.
I have to disagree; yielding the processor explicitely (or forgetting to
do it :) is something I can live without. And it very much depends on
what you're doing - suppose you call a blocking syscall in one coroutine
- you're handing out the CPU at least for the duration of the syscall;
not so with a (kernel) threads implementation. The limitations of the
current implementation (which really sucks only on MP machines, for a
heavy-weight app) are just that, current limitations. 

Cheers,
-- 
Alex Maranda         mailto: amaranda at spider dot com
Spider Software Ltd. Tel: +44 (0)131 4757036
Edinburgh, UK        http://members.xoom.com/Alex_Maranda
STREAMS based communications protocols for embedded systems



From rozzin at geekspace.com  Fri Jun  4 08:04:19 1999
From: rozzin at geekspace.com (Joshua Rosen)
Date: Fri, 04 Jun 1999 08:04:19 -0400
Subject: "?:", "a and b or c" or "iif" (eval?)
References: <7ieg4q$fhu$1@elle.eunet.no> <0sV23.12$KR2.13382@news.corridex.com>
Message-ID: <3757C0C3.C56B3522@geekspace.com>

Evan Simpson wrote:

> It would need to be more than built-in (in the sense of being a function
> defined in builtins) to get the semantics that I, at least, would expect
> from this kind of construct; it would need to actually be part of the
> language, so that it could do short-circuit evaluation of its parameters
> (which no function can do).

>
> In other words, your "iif" always evaluates "a" and "b", when it is often
> desirable to only evaluate "b" if "x" is false.  It's true that the
> full-blown "((test1 and [value1]) or (test2 and [value2]) or
> [elsevalue])[0]" is clunky, but it's actually fairly efficient, and not too
> bad once you've seen it twice.

    One could also probably use eval:
        eval(cond0 and "expr0" or "expr1")
    ..., though that's probably not the best way, in terms of execution-speed of
expr0 or expr1, but it does give the effect of lazy evaluation, truly doing the
short-circuit thing. Still, the syntax isn't ugly.

. o O ( Of course, I don't think that the "(cond0 and [expr0] or [expr1])[0]"
syntax is ugly, or even that "(cond0 and (expr0,) or (expr1,))[0]" is too bad,
so who am I to judge?)






From mwh21 at cam.ac.uk  Tue Jun  1 12:00:35 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: Tue, 1 Jun 1999 16:00:35 GMT
Subject: Python 2.0
In-Reply-To: <11A17AA2B9EAD111BCEA00A0C9B4179301E10ED9@forest.origin.ea.com>
References: <11A17AA2B9EAD111BCEA00A0C9B4179301E10ED9@forest.origin.ea.com>
Message-ID: <Pine.LNX.4.10.9906011653470.21013-100000@atrus.jesus.cam.ac.uk>

On Tue, 1 Jun 1999, Stidolph, David wrote:
> Py_TRACE_REFS seems to be defined if Py_DEBUG is defined.
> 
> I cannot find a "getobjects" or any other similar function in the source
> code or docs.  I am using the Win32 version 1.5.2.

Well, Python/sysmodule.c in the source contains this:

static PyMethodDef sys_methods[] = {
        /* Might as well keep this in alphabetic order */
        {"exc_info",    sys_exc_info, 0, exc_info_doc},
        {"exit",        sys_exit, 0, exit_doc},
#ifdef COUNT_ALLOCS
        {"getcounts",   sys_getcounts, 0},
#endif
#ifdef DYNAMIC_EXECUTION_PROFILE
        {"getdxp",      _Py_GetDXProfile, 1},
#endif
#ifdef Py_TRACE_REFS
        {"getobjects",  _Py_GetObjects, 1},
#endif
        {"getrefcount", sys_getrefcount, 0, getrefcount_doc},
#ifdef USE_MALLOPT
        {"mdebug",      sys_mdebug, 0},
#endif
        {"setcheckinterval",    sys_setcheckinterval, 1,
setcheckinterval_doc},
        {"setprofile",  sys_setprofile, 0, setprofile_doc},
        {"settrace",    sys_settrace, 0, settrace_doc},
        {NULL,          NULL}           /* sentinel */
};

and this:

#ifdef Py_TRACE_REFS
/* Defined in objects.c because it uses static globals if that file */
extern PyObject *_Py_GetObjects Py_PROTO((PyObject *, PyObject *));
#endif

and Objects/object.c contains:
 
PyObject *
_Py_GetObjects(self, args)
        PyObject *self;
        PyObject *args;
{
	...

This is all on Linux, but the source is the source, as I understand it. I
mean the C source, not the Python Library source... I'd hoped that was
obvious.

I doubt it is documented anywhere - a quick grep through the docs finds
nowt, though Doc/api/api.tex contains the line:

XXX Explain Py_DEBUG, Py_TRACE_REFS, Py_REF_DEBUG.

I'd heard about it sometime on the newsgroup.

> Any help would be appreciated,

If this is what you were looking for, you're welcome.

Michael

> David Stidolph.
> 
> 
> -----Original Message-----
> From: Michael Hudson [mailto:mwh21 at cam.ac.uk]
> Sent: Tuesday, June 01, 1999 4:55 AM
> To: python-list at cwi.nl
> Subject: Re: Python 2.0
> 
> 
> "Stidolph, David" <stidolph at origin.ea.com> writes:
> 
> > >OK, I'll try to propose something constructive. Maybe we need an object
> > >protocol, that would enumerate
> > >all references held by an object ?  Writing a portable GC would be then
> > much
> > >easier.
> > 
> > Sounds good to me for debugging.  A call that could return a list of
> > everything that holds a reference to an object - that would be cool!
> > 
> > list = GetReferences(object)
> > print 'List of referencest to',object
> > for item in list:
> >   print 'item:',item
> > 
> > Anybody know of a current way to do this?
> 
> If you recompile Python with Py_TRACE_REFS defined, then the sys
> module sprouts a "getobjects" function that returns a list of all
> objects in existence. This could probably be used to implement
> something like this. It would be veeeeery slow, I suspect.
> 
> I've never resorted to this approach, tending always to make cyclic
> references go away be staring at them very, very hard.
> 
> I'm not sure this is at all what is being asked for, but it seems kind
> of relavent.
> 
> Yours,
> Michael
> 




From Klaus.Schilling at home.ivm.de  Thu Jun 17 05:43:31 1999
From: Klaus.Schilling at home.ivm.de (Klaus Schilling)
Date: 17 Jun 1999 11:43:31 +0200
Subject: GNU python? (GNUbe question)
References: <3765B8B4.18DE0257@spots.ab.ca> <37667309.F9BBAE9A@mojam.com> <022c01beb7ec$f9617070$f29b12c2@pythonware.com> <37689AC2.B8F6579A@serop.abb.se>
Message-ID: <87k8t3jh6j.fsf@home.ivm.de>

"Thomas S. Strinnhed" <thstr at serop.abb.se> writes:
> 
> Sounds like a good thing this GNUscript, it'll do almost anything.
> Is there a download site ;-)

ftp://ftp.gnu.org/pub/gnu/guile

Klaus Schilling



From pduffin at mailserver.hursley.ibm.com  Mon Jun 28 08:48:02 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Mon, 28 Jun 1999 13:48:02 +0100
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <z%Q93.4399$5E4.36952@news2.randori.com>
Message-ID: <37776F02.6EEA@mailserver.hursley.ibm.com>

Tom Cloney wrote:
> 
> >   I'm trying to collect a list of all the _current_ issues making tcl
> > a bad language choice...
> 
> Well, the troll aside, I do have quite a lot of experience with Tcl. Among
> other things, I've written a workflow system on the Macintosh for generating
> technical book catalogs using a combination of a relational database (4th
> Dimension), image processing applications (Photoshop, NIH Image), page
> layout and database publishing software (FrameMaker MIF) and other apps. Tcl
> was used extensively as glue for doing things like process control and
> analyzing Encapsulated Postscript files. The Tcl I used is the one embedded
> in Pete Keleher's Alpha text editor, which is a real joy to use. I wish we
> had something like it on NT. I've also used standalone versions of Tcl with
> and without Tk. I worked with it a lot from about 1992 - 1996, writing many
> useful apps. I've kept an eye on the language since then, and while I'm no
> longer very Tclish, I reckon my basic take is still valid.
> 
> My opinions are, of course, shaped by the way I work. I lean towards
> practical applications that one person can write and maintain and still have
> a life. Most of the time I find myself practicing the Mad Max
> post-apocalyptic school of programming, that is, cobbling together
> interesting artifacts that come my way from both commercial and
> non-commercial sources. I'm lazy -- I try to write only what I have to, and
> then I look for the easiest tools to use to do it. I'm a businessman, so I'm
> always balancing cost, productivity, and performance. I'm also aware that an
> attractive solution may carry many hidden costs. I'm not going to get into
> much academic hair-splitting here. For my purposes I judge language
> environments by how easy it is to get things done in them, not by whether
> they have a recursively bi-directed optimizing monkee gland compiler. YMMV.
> 
> Since the late 60's I've done practical programming in over 60 languages,
> from FORTRAN, TECO, and microcode to Java, Python, and *Lisp, with many fun
> stops in between (pause for chest-thumping and vine-swinging ;>), which
> either makes me experienced or a fossil, depending on your point of view.
> That said, here's my take on Tcl.
> 
> What I like:
> 
> Very readable syntax. A great language for writing more or less
> self-documenting "dumb" code that you can come back 6 months or a year later
> and still understand and modify. This is hard to overemphasize: clever
> things _will_ be forgotten!
> 
> It's super when embedded in a text editor. Alpha is a great example.
> 
> It's a very nice hackery language. You can just whack away at things until
> they do what you want. You don't have to be feeling sharp and clever on days
> when you use it.
> 
> The language itself is pretty stable, that is, it doesn't usually crash in
> bad ways.
> 
> It's one of the better languages for text processing. Not as nice as Perl,
> but pretty darn good. Good regexp support. Lots of easy and elegant string
> munging built in.
> 
> Support for lists, while not deep, is elegant and simple as far as it goes.
> I don't like it when languages make list operations a big cumbersome deal,
> so I appreciate Tcl's elegance here.
> 
> Support for hash tables (associative arrays, whatever...) is also elegant.
> 
> These three things: clean support for lists, hash tables, and regular
> expressions are enough in themselves to make a language attractive. They
> just make life much easier for a practical programmer.
> 
> GUI support via Tk is nice. It's a bit ideosyncratic, but good for many
> things. It's not shrinkwrap-quality stuff, but think of all the great
> languages whose inexpensive implementations are crippled by having little,
> if any, support for GUI (Perl until recently), produce GUIs that feel slow
> and clunky (Java), or make it cumbersome (like the joys of COM/IDL).
> 
> Wiser folk than I have said a good language should make doing simple things
> simple. Tcl succeeds in this.
> 
> Good support for file system ops.
> 
> Straightforward interfaces to C etc. are a good thing.
> 
> Intangible fun factor. Tcl puts you in a good mood. It's easy. You feel like
> you're cheating. You concentrate on the problem domain. You still feel good
> when you look at the code six months later.
> 
> Now the cold slimy part:
> 
> Tcl scales very badly. All the same things in Tcl that make simple things
> simple make complex things hard. Tcl is broad and shallow (which is not
> always a bad thing). List structure doesn't really scale. There is no OOP
> support to speak of, and although various people have hacked quick and dirty
> examples, it's not something you would really want to build on. The very
> brainlessness of Tcl is part of its appeal, but this is part of the price.
> This is a _procedural_ language, which is actually nice on days when you are
> too fuzzed out to think like a computer professional. It's not CLOS or
> Eiffel, folks. It's not even Java.
> 

Actually since version 8.0 Tcl has scaled quite nicely thank you, 8.1 has
reduced the scalability of string operations due to a naive implementation
of Unicode / UTF8 support but that will be fixed soon.

> The cute hacks that make string munging kind of cool can also be confusing.
> You tend to spend a fair amount of time (by Tcl standards) finding subtle
> bugs in the way you expressed complex string concatenations and such. It's
> not that the language is buggy in this regard. The rules are relatively
> simple. But the subtleties are hard to keep in mind. Tcl makes string
> processing so easy that you tend to quickly give yourself enough rope to
> hang yourself with. Easy debugging takes some of the sting out.
> 
> Tcl is slow. Oh, I know that's relative. For many applications it will be
> fine. But complex parsing on large amounts of text can really drag. In one
> application I needed to parse FrameMaker MIF files. These are SGML-like
> ASCII files that often stretch to 600k or more, and which give FrameMaker a
> lot of its power as a database publishing platform. Tcl was dog-slow. Perl
> was about an order of magnitude faster than Tcl. C++ using the STL was a bit
> faster than Perl, but still distressingly slow. I ended up using "cheating
> C++", which is C++ with a lot of C standard library calls in it. The C was
> about another order of magnitude faster than Perl. Tcl is great for nabbing
> some text out of a file and doing some clever things with it (like nipping
> the image size data out of an EPS file), but a cruncher it's not.
> 

Tcl can be used to crunch stuff but you just have to be very careful how
you go about doing it. I have written a Tcl program to analyse the whole
Tcl source which is nearly 5MB and it did it very fast indeed.

> Perl can be much better for industrial-strength text processing. Perl has
> better regexp handling than Tcl, in fact it's usually held up as the

Tcl 8.1 now has the same (bar a few weird hacks) regular expression package
that Perl has.

> standard by which other regular expression packages are judged. Perl is much
> faster than Tcl. Perl 5 has rudimentary OOP support, which is a lot better
> than nothing and allows Perl to scale in ways Tcl can't. The comparisons

OOP support doesn't imply scalabity does it ?

> with Perl are important because Perl does a lot of the same things Tcl does
> and more. For a long time the dilemma was that Perl was the better language,
> but Tcl had a GUI. Now Perl has a Tk interface (I haven't used Tk from Perl,
> so I can't vouch for its elegance). After a while, as I developed more
> complex applications in the languages, I found myself abandoning Tcl in
> favor of Perl because of the performance and scalability. Perl code lacks
> the beautiful brainless stickiness of Tcl, though. Perl is a "clever"
> language, with what many would describe as "line noise" syntax. Cleverness
> is dreadful when you have to look at it six months later. Tcl is the glory
> of "stupid" programming, which I lean towards whenever possible. But it will
> drive you crazy the minute you need speed or want to encapsulate something.
> 

It sounds as though you are being a bit unfair with Tcl. You seem to invest
an awful lot of brainpower in Perl (not to deal with the problem but to
deal with the language) but you don't spend any on Tcl because it is so
simple. If you invested as much time learning how to program efficiently
in Tcl as you do in Perl you would soon be able to match performance.

> Tcl, Perl, and Python all hold out the promise of transcending the speed
> problem via extensions written in C or another compiler language. This is
> fine as far as it goes, but in the real world it's often difficult to fully
> encapsulate speed-critical bits in separate modules (this could be the topic
> of a whole thread). It's also somewhat complicated. Tcl makes it as easy as
> anybody, but by the time you are organizing your work to the point where
> you're designing a full component architecture, you're probably going to
> find yourself thinking about using high-level compiler languages that let
> you keep it all under one roof.
> 
> Tk ports are slow to arrive on some platforms and are buggy for a while. It
> took years for heroic volunteers to get a workable Tk on the Mac. I think
> Tcl helped show the way, but many of its enthusiasts have drifted away as
> more alternatives have become available.
> 
> If you want to grind text, why not use Perl? If you want a slow, essentially
> free hacking language with all the features of Tcl and very good OOP
> support, why not use Python? And if you want a faster "scripting" language
> based on the "Boy in the Bubble"* architecture, with built-in GUI, much
> wider support, and a religion, there's always the J-word.
> 


-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From pphalen at teleo.net  Sun Jun 13 14:04:52 1999
From: pphalen at teleo.net (Patrick Phalen)
Date: Sun, 13 Jun 1999 11:04:52 -0700
Subject: Variations on a theme by "pound bang"
References: <3763e75f.3265255@news.iol.ie>
Message-ID: <99061311062410.00557@quadra.teleo.net>

On Sun, 13 Jun 1999, Sean Mc Grath wrote:
> Can anyone shed some light on the benefits of:-
> 
> # !/usr/bin/env python
> 
> as opposed to
> 
> # !/usr/bin/python

Yes:

http://www.python.org/doc/FAQ.html



From Christian.Tismer at p98.f112.n480.z2.fidonet.org  Tue Jun 29 18:08:23 1999
From: Christian.Tismer at p98.f112.n480.z2.fidonet.org (Christian Tismer)
Date: Tue, 29 Jun 1999 23:08:23 +0100
Subject: ANN: Stackless Python 0.2
Message-ID: <0000159c@bossar.com.pl>

From: Christian Tismer <tismer at appliedbiometrics.com>



"Gary D. Duzan" wrote:
...

> =>The cost of a call comes from the slightly complicated
> =>function setup which is done all the time. Frames belong
> =>to the fastest available Python objects already,
> =>like, say dictionaries.
> 
>    I thought that might be the case, which is the point I was making in
> my earlier paragraph. I was just briefly pondering possibilities for
> reducing frame memory management time, if it were necessary.

I hoped it would be possible to save most of the
frame initialization, in cases where the same function
call appears in the same place again and again.
The barrier is that I would have to keep references
longer than necessary, so I'm not sure if that can
be done.

BTW, the continuations which I am just testing
are doing a great job of creating hard to release
references to frames. They are correct, but create
some cycles, quickly. This project is becoming
very time expensive already :-)

cheers - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From mwh21 at cam.ac.uk  Fri Jun 25 03:08:00 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 25 Jun 1999 08:08:00 +0100
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0rr2.2fi.behrends@allegro.cse.msu.edu> <m3aetr2yk9.fsf@atrus.jesus.cam.ac.uk> <7ks38p$l4m$1@brick.cswv.com>
Message-ID: <m3674c3ghb.fsf@atrus.jesus.cam.ac.uk>

neelk at brick.cswv.com (Neel Krishnaswami) writes:

> 
> It's just that writing self.__class__ is so *ugly*! Everything else
> in Python is beautiful, too, which makes this stand out even more. 
>  

You want ugly? I think the whole area of calling base class functions
is far uglier.

You might have a base class like this:

class C:
    def f(self,x):
        print self

then a derived class that needed to call C.f would have to be written:

class D(C):
    def f(self,x):
        C.f(self,x)
        print x

I think needing `self' here sucks. It gets worse if there are keyword
arguments, necessitating use of apply, or if the base class is in a
different module.

This can be combined with another irritation of mine, the lack of
automatic initialization of base classes, to give a situation like
this:

base.py:

class Base:
    def __init__(self,x,y,**kw):
         ...

derived.py:

import base

class Derived(base.Base):
    def __init__(self,x,y,z,**kw):
        apply(base.Base.__init__,(self,x,y),kw)
        print z

which I think is one of the ugliest things I've ever seen in Python
that I couldn't think of a prettier way round. Ideas, anyone?

twisted-ly y'rs
Michael



From xidicone at iname.com  Mon Jun 28 09:20:19 1999
From: xidicone at iname.com (Jussi Jumppanen)
Date: Mon, 28 Jun 1999 23:20:19 +1000
Subject: Zeus Editor Version 3.0 Beta
Message-ID: <37777693.6EFE@iname.com>

Zeus for Windows Version 3.0 Beta Release
-----------------------------------------

The Zeus programmer's editor has been totally re-developed and the 
results of this full re-write are now ready in the form of an
early beta version. The Zeus Beta Version 3.0 should work fine for 
Windows 95, Windows 98 or NT but at this stage it has only been 
rigourously tested against NT.

For anyone interesting in test running the beta just point your 
browser to the following web site:

  http://www.geocities.com/SiliconValley/Monitor/6788/

But please remember it is beta quality software which means it is
guaranteed to still have bugs!

So what is new to Zeus:
    
    1) New syntax highlighting engine should now support almost
          any language.

    2) New plugable scripting modules allows you to write Zeus
          macro scripts using the Python, LUA or Small C macro
          languages.

    3) New configuration model now makes it even easier for you
          to configure Zeus to do just what you want.

    4) Enhanced compiler, project and tool support.

    5) Improved keyboard mapping.

Jussi Jumppanen
Author of: Zeus for Windows, Win32 (Brief, WordStar, Emacs) Text Editor
"The C/C++, Java, HTML, Pascal, Cobol, Fortran programmers text editor" 
Home Page: http://ourworld.compuserve.com/homepages/jussi/



From suzuki611 at okisoft.co.jp  Sun Jun  6 21:30:06 1999
From: suzuki611 at okisoft.co.jp (Hisao Suzuki)
Date: 07 Jun 1999 10:30:06 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7j90va$6lk$1@cronkite.cc.uga.edu>
Message-ID: <u3g1443iht.fsf@ares.sys1.ngo.okisoft.co.jp>

In article <7j90va$6lk$1 at cronkite.cc.uga.edu>,
graham at sloth.math.uga.edu (Graham Matthews) wrote:
> I am completely confused by this statement. The two people that have
> been advocating GC the hardest are myself and the author of Ruby. Since
> Ruby has GC the author has clearly implemented a collector. ...

Regrettably, I'd have to say that Ruby is (was?) suffering from
its insufficient portability.  Once I have used a certain i860
SVR4 box.  On that box, Ruby did not run but dumped immediately
with its twiddling the machine stack, while Python ran perfectly
with no difficulty.  As far as I know, the collector of Ruby has
not changed so much since then.

(I have before told this to the author of Ruby on a Japanese
Pythoneers' mailing-list against his advocacy of Ruby, and now
he appeared here again with just the same opinion before...)

And one more thing: as the author of Ruby said, Ruby's user
cannot specify her/his arbitrary code as destructor.  If so were
Python, it would not be matter what is the semantics of language
in regard to storage management for users.  Ignoring this point,
the whole discussion might be irrelevant to Python.

As you know from reading any Python-1.*.*/Doc/ext/ext.tex, GC
has been taken into account already, and deliberately turned
down from the current Python.  Only a general or theoretical
discussion would be no use here (except for some evangelism).

--===-----========------------- Sana esprimo naskas sanan ideon.
SUZUKI Hisao            suzuki611 at okisoft.co.jp, suzuki at acm.org.



From mwh21 at cam.ac.uk  Sun Jun 13 12:19:47 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 13 Jun 1999 17:19:47 +0100
Subject: Interpolation (was: Keyword calling gotcha ?)
References: <000d01beaa57$c3d2c780$9b9e2299@tim>
Message-ID: <m3d7z0t6n0.fsf@atrus.jesus.cam.ac.uk>

"Tim Peters" <tim_one at email.msn.com> writes: 

> I posted a link earlier to the Best Parsing Book on Earth; read the
> first few chapters of that first.

Which one was that? I think I missed it!

Thanks,
Michael



From eric at unifix-online.com  Wed Jun 30 02:28:38 1999
From: eric at unifix-online.com (Eric Boudaillier)
Date: Wed, 30 Jun 1999 08:28:38 +0200
Subject: Python GUIs: Summary and Conclusion
References: <99FA4A1A1280BEBD.0A1ABA62CC989479.4C2C6022F695BEAC@lp.airnews.net> <377805C8.C6BBD4C1@unifix-online.com> <wPae3.439$4K6.21052@dfiatx1-snr1.gtei.net>
Message-ID: <3779B916.20AF33E6@unifix-online.com>

John Michelsen wrote:
> 
> Philippe, they look very nice. How did you use them with Python
> and Tkinter? Any Python code that you would like to share
> demonstrating BWidgets?  I would be especially interested
> to see use of the tree widget.

  Same as Tkinter: directly call tcl command
  To lauch the demo:

cln46ib:ficus/FicusSrc > cd ../../BWidget/demo/
cln46ib:BWidget/demo > python
Python 1.5.2 (#5, Jun 23 1999, 18:09:42) [C] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import Tkinter
>>> app=Tkinter.Tk()
>>> app.tk.call('source', 'demo.tcl')
''
>>> 

  But, if some volunteers would like to encapsulate
  BWidget in python classes...
  
-- 
Eric Boudaillier
Unifix - http://www.unifix-online.com/



From tim_one at email.msn.com  Fri Jun  4 02:24:15 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 4 Jun 1999 02:24:15 -0400
Subject: A problem with re.split()
In-Reply-To: <87pv3en24q.fsf@pc-hrvoje.srce.hr>
Message-ID: <001c01beae52$dee5b560$949e2299@tim>

[Hrvoje Niksic]
> I understand that the idea with "re" module is for it to behave as
> closely to Perl's regular expressions as possible (which is why the
> order of arguments in string.split and re.split is different,
> string.split being the logical one).

The regexp *language* was meant to be compatible with (a snapshot of
then-current) Perl's, but there wasn't much desire to mimic the rest.

WRT argument ordering, this was deliberate but for a different reason:
every function in the string module takes "the string" as its first argument
because that's the ordering that makes the most sense if you think of the
string as being an object and the function a method of that object.
Likewise every function in the re module takes "the regexp" as its first
argument.  Intra-module consistency was judged to be more important than
inter-module consistency in the case of "split".  Not a pure win, but still
seems the lesser evil.

> The problem is with re.split() in this case:
>
> $ python
> Python 1.5.2 (#3, May 23 1999, 19:57:40)  [GCC 2.8.1] on sunos5
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import re
> >>> re.split('', 'foo')
> ['foo']
>
> Perl splits it to ['f', 'o', 'o']:
>
> $ perl -e 'print join(":", split("", "foo"))'
> f:o:o

Python won't change here (IMO) -- re.split needed to be compatible with the
old regsub.split too, and (IMO) that had a more sensible rule than Perl's:
when splitting on a pattern, never take an empty match as a split point.
That's really all that's going on here; e.g., try

perl -e 'print join(":", split(/b?/, "foobar"))'

Did you expect f:o:o:a:r?  I like Python's ["foo", "ar"] much better.

BTW, Python's string.split also refuses to split on an empty separator.

Also note:

>>> list("foo")
['f', 'o', 'o']
>>>

adopting-perl's-language-didn't-imply-adopting-its-accent<wink>-ly y'rs  -
tim





From vlachid at otenet.gr  Mon Jun  7 00:38:18 1999
From: vlachid at otenet.gr (Vlachidis Costas)
Date: Mon, 07 Jun 1999 07:38:18 +0300
Subject: FREE HARDCORE TEEN PICS  6141
References: <7ja8ou$7j4$4641@nclient5-gui.server.virgin.net>
Message-ID: <375B4CB9.79B220B6@otenet.gr>

how to stop this???

qnmdjb at ffff.com wrote:

> FOR THE BEST FREE TEEN SEX PICTURES VISIT:
>
> http://freespace.virgin.net/eric.johnston/freepics.htm
> gldoqlclctgtcizceuovzqleir
-------------- next part --------------
A non-text attachment was scrubbed...
Name: vlachid.vcf
Type: text/x-vcard
Size: 366 bytes
Desc: Card for Vlachidis Costas
URL: <http://mail.python.org/pipermail/python-list/attachments/19990607/45bc4e82/attachment.vcf>

From Colin.J..Williams at p98.f112.n480.z2.fidonet.org  Wed Jun 30 07:56:16 1999
From: Colin.J..Williams at p98.f112.n480.z2.fidonet.org (Colin J. Williams)
Date: Wed, 30 Jun 1999 12:56:16 +0100
Subject: A Python problem?
Message-ID: <000015d4@bossar.com.pl>

From: "Colin J. Williams" <cjw at connection.com>

This script below illustrates what appears to be a problem with the
identification of a module in a package.

The example is borrowed from the linbot material.

Colin W.
''' A Python problem?   .

  '''
def main():
  import sys
  print 'Start:',dir()
  s0= sys.modules
  import plugins.rptlib
  s1= sys.modules
  if s0 == s1:
    print 'sys.modules is unchanged'
  print 'After attempt to import plugins.rptlib:', dir()
  try:
    print dir(plugins.rptlib)
  except AttributeError:
    import sys
    print sys.exc_info()
  from plugins.rptlib import *
  print 'rptlib names now available:', dir()
main()

'''
------------------------ Resulting Display ----------------------------
>>> Start: ['sys']
sys.modules is unchanged
After attempt to import plugins.rptlib: ['plugins', 's0', 's1', 'sys']
(<class exceptions.AttributeError at 76bbc0>, <exceptions.AttributeError
instance at daa0e0>, <traceback object at dab2b0>)
rptlib names now available: ['Link', 'add_problem', 'check_and_warn', 'config',
'debugio', 'doBotMain', 'doTopMain', 'get_title', 'linbot', 'linkList',
'main_index', 'make_link', 'nav_bar', 'open_file', 'opener', 'os', 'plugins',
'problem_db', 'proxies', 'read_registry', 's0', 's1', 'sort_by_age',
'sort_by_author', 'sort_by_rev_age', 'sort_by_size', 'string', 'stylesheet',
'sys', 'urllib', 'version']

  '''



From Robert.Meegan at wcom.com  Wed Jun  9 10:04:19 1999
From: Robert.Meegan at wcom.com (Robert Meegan)
Date: Wed, 9 Jun 1999 14:04:19 GMT
Subject: Scripting vs. "Real Programming"
Message-ID: <Pine.A41.4.10.9906090836220.65174-100000@barney.mcit.com>

Greetings ---

Even in the relatively short time that I've been using Python, there has
been a notable shift in the focus for the questions and comments on
C.L.P.

In the past, it seems that Python was primarily used as a scripting
language. That is to say, most of the programs written were small and
intended to run for short periods of time, generally serving one well
defined purpose.

Now, an ever greater number of the contributors to this august forum are
discussing issues that are more indicative of Python being used for large
and complex programs. Some of the topics that highlight this change
include:

* Garbage Collection
* Threading
* Forking
* Interprocess Communications

It is a credit to our founding fathers (especially Guido) that Python is
so well equipped with these tools. In general, they are complete and
robust (my desire for more garbage collection does not imply that the
current interpreter leaks memory, but rather that I can't be trusted to
clean up after myself).

What seems to be lacking are guides for creating large, stable programs
using these tools.

My wife is expecting a deliverable in about a month, after which I expect
that there may be a few late nights. During this time, I hope to add a few
entries to the FAQ regarding this topic.

I would be grateful if people would offer suggestions for both questions
and answers. I will attempt to collate the existing knowledge, as well as
adding my own limited contributions.

I am at your service,
--- Robert

--------------------------------
Robert Meegan
MCIWorldCom - Cedar Rapids, Iowa
319.375.2416




From rdev at my-deja.com  Wed Jun 30 15:18:38 1999
From: rdev at my-deja.com (rdev at my-deja.com)
Date: Wed, 30 Jun 1999 20:18:38 +0100
Subject: theads & global namespaces
Message-ID: <000015d0@bossar.com.pl>

From: rdev at my-deja.com

Two suggestions:
- If you're running on Solaris, the thread won't get to run
unless you give up some time.  So try putting a time.sleep(.001)
into your while loop.  On NT, this is not necessary (don't know
why.)
- Your thread will only increment once and then exit, so you'll
never see x>1.  If you want to keep the thread running, put it in
a while loop.  Make sure you give up the time slice there as well.

Roger Dev
dev at opticominc.com

In article <377A5C88.32A1FAB6 at notes.cba.ufl.edu>,
  andersdt at notes.cba.ufl.edu wrote:
> Greetings all,
>
> I am trying to share globals between threads.  Is is possible for me
to
> share globals between threads that call functions from different
> modules?  Thus far I have had little success.  I ask this question
> because  I would prefer not to put all my code in one module.
>
> For example, I have 2 .py files: a.py and b.py.  b.py contains the
> function ('foo()') to be executed  under the new thread, as well as
the
> global variable 'x'.  a.py contains the call to
> start_new_thread(foo,()).  a.py also tries to use the global variable
x.
>
> =========================================
> b.py
>
> x = 0
>
> def foo():
>    global x
>    x  = x + 1
>
> =========================================
> a.py
>
> from b.py import *
>
> if __name__ == '__main__':
>
>     start_new_thread(foo, ())
>     while 1:
>         if x != 0:
>             print x
>
> =========================================
>
> So far, my code in a.py does not detect any change in x.  Any
> suggestions for a workaround?
>
> Many thanks,
>
> Taylor
>
>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From hyoon at bigfoot.com  Mon Jun  7 10:10:02 1999
From: hyoon at bigfoot.com (Hoon Yoon)
Date: Mon, 07 Jun 1999 10:10:02 -0400
Subject: Python an Tcl/Tk 8.1
References: <7jaeof$7ja$1@news.rapidnet.com>
Message-ID: <375BD2BA.9C960A06@bigfoot.com>

Jared,

  First thing you should do is blow away all old tk*.dll and tcl*.dll's
from your computer. If you have mult old copies Python will be confused.
Then you have to set tk_library and tcl_library from autoexec.bat.
  Then use the Tcl/Tk that comes with python.exe installation. You
should have no problem afterward. The docs do say that newest is
problmatic.

Hoon,

Jared Hanson wrote:
> 
> I am new to all of this.  I am running Windows 98, and I just downloaded
> Python and the newest version of Tcl/Tk (8.1).  Unfortunately, it appears
> that Python doesn't like this version.  Doing an import Tkinter command
> prints out an error message saying that its looking for 8.0 dlls.  I was
> wondering what files I need to change to point to 8.1 dlls in order to make
> it work (if that would make it work at all)  Or do I just need to get a copy
> of the 8.0 dlls and stick them in the same folder.  Any help you could offer
> me would be appreciated, just reply to my post.
> 
> Thanks,
> Jared Hanson

-- 
*****************************************************************************
S. Hoon Yoon (Quant) Merrill Lynch Equity Trading, 
yelled at yahoo.com hoon at bigfoot.com(w)
"Miracle is always only few standard deviations away, but so is
catastrophe."
* Expressed opinions are often my own, but NOT my employer's.
"I feel like a fugitive from the law of averages."    Mauldin
*****************************************************************************
-------------- next part --------------
A non-text attachment was scrubbed...
Name: vcard.vcf
Type: text/x-vcard
Size: 202 bytes
Desc: Card for Hoon Yoon
URL: <http://mail.python.org/pipermail/python-list/attachments/19990607/96199a3f/attachment.vcf>

From greg.ewing at compaq.com  Thu Jun 17 19:16:34 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 18 Jun 1999 11:16:34 +1200
Subject: Compiled main program
References: <KyG93.386$dR.22261@newsfeed.avtel.net>
Message-ID: <376981D2.61C7D46D@compaq.com>

Clarence Gardner wrote:
> 
> x = MakeMeMain()
> del x

Wouldn't it just be simpler to do

MakeMeMain()

Greg



From arcege at shore.net  Fri Jun 25 22:14:46 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Sat, 26 Jun 1999 02:14:46 GMT
Subject: load_dynamic problem
References: <7l0j58$cbh$1@nnrp1.deja.com>
Message-ID: <qIWc3.329$Of3.76550@news.shore.net>

John Fisher <jfisher at are.berkeley.edu> wrote:
: Hey folks,

: I'm having a bit of trouble loading a C extension into Python 1.5.2.
: I'm getting the error:

: Traceback (innermost last):
:   File "<stdin>", line 1, in ?
:   File "bpinter.py", line 9, in ?
:     PyBPMPD = imp.load_dynamic('PyBPMPD', 'PyBPMPD.so')
: ImportError: ./PyBPMPD.so: undefined symbol: setdef

: setdef, not written by me, but called from inside PyBPMPD.c, is
: prototyped as:
:  extern int setdef(bpmparam *);

: The module itself compiles without complaint with:
:  gcc -I/usr/include/python1.5/ -shared -o PyBPMPD.so PyBPMPD.c

: I'd appreciate any ideas on the problem.  I'm rather stumped.

: Thanks,
: John

John,

It sounds like "setdef" is in some library that you are not compiling
with.  Or "setdef" was typed incorrectly in your code: resolving
symbols in shared objects is performed at runtime.

When compiling the shared object, you must include library being
referenced (the one that contains "setdef").  Preferably, that
library will be a static library (.a), or you would have to set
LD_LIBRARY_PATH (or LIBPATH on some systems) to point to the
correct location at run-time.

If the library is /usr/local/lib/libdef.so or /usr/local/lib/libdef.a,
then add "-L/usr/local/lib -ldef" to the gcc command you have.

Good luck,
  -Arcege




From nathan at islanddata.com  Mon Jun 21 14:06:48 1999
From: nathan at islanddata.com (Nathan Clegg)
Date: Mon, 21 Jun 1999 11:06:48 -0700 (PDT)
Subject: choosing random numbers with weights/probability?
In-Reply-To: <929987267.6227@www.remarq.com>
Message-ID: <XFMail.990621110648.nathan@islanddata.com>

list = [('one', 25), ('two', 25), ('three', 50)]
l = []
for x, y in list:
        l.extend([x] * y)
item=whrandom.choice(list)

Too bulky?


On 21-Jun-99 kevinsl wrote:
> I've been using the whrandom.choice routine and it's very
> useful. But is there any way to add weights or
> probabilities of being chosen to the items in the list?
> 
> example:
> 
> list=['one','two','three']
> item=whrandom.choice(list)
> 
> Is there any way to say that 'one' and 'two' have a 25%
> chance of being chosen, and 'three' has a 50% chance?
> 
> I'm hoping there's already a module to do this... or else
> I'll be writing my own..


----------------------------------
Nathan Clegg
 nathan at islanddata.com





From MHammond at skippinet.com.au  Thu Jun 10 18:00:33 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Fri, 11 Jun 1999 08:00:33 +1000
Subject: COM dates
References: <816010E2456BD111A48700805FBBE2EEFDF354@ex-quebec-u1.baan.com> <375F77C1.629FED7B@lemburg.com>
Message-ID: <7jpcfb$9bn$1@m2.c2.telstra-mm.net.au>

M.-A. Lemburg wrote in message <375F77C1.629FED7B at lemburg.com>...

>Not sure how you would convert the Win32 PyTime objects into
>DateTime objects though. If they implement the int() APIs, then
>DateTimeFromTicks(pyTime) will do, otherwise there's also a
>constructor DateTimeFromCOMDate(comdate) which takes the raw
>COM date float as argument.

"Win32 PyTime" dates and "COM dates" are the same thing.  Although COM uses
them most heavily, we decided to move the date support into the Win32
extensions, so everyone can use them.  (Specifically, they are implemented
in pywintypes15.dll, along with our Unicode objects, HANDLE objects, UUID
objects, etc)

FWIW, I personally use mxDateTime whenever I hit the many limitations in
Python dates.  As Marc said, I used DateTimeFromCOMDate, then work with the
DateTime object.  Can''t recall exactly how I convert it back, but it is
definately there.

Mark.






From news at dorb.com  Wed Jun 16 16:25:04 1999
From: news at dorb.com (Darrell)
Date: Wed, 16 Jun 1999 16:25:04 -0400
Subject: GC In Python: YAS (Yet Another Summary)
References: <1282571877-108998356@hypernet.com> <zoQ93.198$U6.1646@newsr2.twcny.rr.com> <wku2s86nmb.fsf@ifi.uio.no>
Message-ID: <aNT93.216$U6.1531@newsr2.twcny.rr.com>

Tried the lazy plumbo and it found a cycle in a large application I'm
working on.
And it was easy to fix, once found ::)

It took 5min at least to find the cycle. I'm not complaining because
it would have taken me a lot longer. But maybe there is a way to speed
this up. Maybe the Guido idea of only tracking dictionaries and objects
in dictionaries would be useful ?  Or move find cycles to 'C'

A nice feature might be to allow queries so lazy developers could find out
where all the memory has gone. Such as when you have no cycles and still
consuming 100meg.

Hope no one finds a basic flaw.
Until then I'm dumb, lazy and happy :)
--
--Darrell





From mwh21 at cam.ac.uk  Tue Jun 22 02:36:46 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 22 Jun 1999 07:36:46 +0100
Subject: newbie idiom question
References: <m3emj4yezq.fsf@monsoon.swcp.com>
Message-ID: <m3emj43fnl.fsf@atrus.jesus.cam.ac.uk>

Alex Rice <alex at mindlube.com> writes:

> Something I keep getting tripped up about is that objects being
> iterated in "for" statements cannot be modified.
> 
> I keep trying to do this:
> 
>       >>> eggs = [1,2,3]
>       >>> for spam in eggs:
>       ...	spam = 'cooked'
>       ...
>       >>> eggs
>       [1, 2, 3]
>       >>> 
> 
> The tutorial says this:
> 
>       If you need to modify the list you are iterating over, e.g., duplicate
>       selected items, you must iterate over a copy. The slice notation makes
>       this particularly convenient:
> 
>       >>> for x in a[:]: # make a slice copy of the entire list
>       ...    if len(x) > 6: a.insert(0, x)
>       ... 
>       >>> a
>       ['defenestrate', 'cat', 'window', 'defenestrate']
> 
> Understood, but what if you want to modify each element in a list?
> What's the best way to do this in terms of speed and elegance? I guess
> I just haven't seen a good example of this yet in Python

If I know I'm not going to be altering the length of the list, I
generally do it like this:

for i in range(len(a)):
    if is_cooked(a[i]):
        a[i] = "cooked"

If the list might be changing length, then I generally do

i = 0
while i < len(a):
    ...

and keep track of i by hand.

HTH
Michael

> What I'm unfortunately used to is this, in Perl:
> 
>       @eggs = (1,2,3);
>       foreach $spam (@eggs) {
> 	$spam = 'cooked'; 
>       }
>       print "@eggs";
>       >>> cooked cooked cooked
> 
> Thanks,
> Alex Rice



From bwarsaw at CNRI.Reston.VA.US  Thu Jun  3 14:48:35 1999
From: bwarsaw at CNRI.Reston.VA.US (Barry A. Warsaw)
Date: Thu, 3 Jun 1999 14:48:35 -0400 (EDT)
Subject: Use of StringIO vs cStringIO in standard modules
References: <8790a1620b.fsf@pc-hrvoje.srce.hr>
	<199906031425.KAA06935@eric.cnri.reston.va.us>
	<87aeuh5kqc.fsf@pc-hrvoje.srce.hr>
Message-ID: <14166.52739.423023.826815@anthem.cnri.reston.va.us>

>>>>> "HN" == Hrvoje Niksic <hniksic at srce.hr> writes:

    | try:
    |     from cStringIO import StringIO
    | except:
    |     from StringIO import StringIO

I do this (though I use "except ImportError") in all my new code.
Maybe Guido would take patches to the standard library (most of which
probably pre-dates cStringIO) to effect this change?

-Barry



From skip at mojam.com  Tue Jun 15 11:34:17 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 15 Jun 1999 15:34:17 GMT
Subject: GNU python?
References: <3765B8B4.18DE0257@spots.ab.ca>
Message-ID: <37667309.F9BBAE9A@mojam.com>

Maxwell Sayles wrote:
> 
> Looking to try out python... is there a GNU python for Win9x?  and a
> tutorial maybe? 

Not sure just how a GNU Python would differ from the real thing, but you
can visit the Python Software Association's web site at
http://www.python.org/ to find what you're looking for.  There's a
Windows Topic Guide (upper left corner of the screen) and a tutorial
(click the Documentation link).

-- 
Skip Montanaro	| Mojam: "Uniting the World of Music"
http://www.mojam.com/
skip at mojam.com  | Musi-Cal: http://www.musi-cal.com/
518-372-5583



From greg.ewing at compaq.com  Sun Jun 13 17:47:22 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Mon, 14 Jun 1999 09:47:22 +1200
Subject: Solution to finalisation problem [Re: fork()]
References: <000901beb424$b2878200$a49e2299@tim>
Message-ID: <376426EA.B6C9729E@compaq.com>

Tim Peters wrote:
> 
> the combo of
> finalizers, cycles and resurrection seems to be a philosophical mess even in
> languages that take it all <ahem> seriously.

Indeed, which leads me to believe that the idea of giving objects
a __del__ method is fundamentally flawed in the first place.

Fortunately, there *is* a way to do finalisation that avoids all
these problems. Instead of an object doing its own finalisation,
it designates some *other* object to do its finalisation on its
behalf after it is dead.

The benefits of this are:

(1) The object doing the finalisation is fully alive and operates
in a predictable environment.

(2) The object which triggered the finalisation is fully dead
(its memory has already been reclaimed by the time the finaliser
is called) and there is no possibility of it being resurrected.

In Python, it might look something like this from the
user's point of view:

class FileWrapper:
   # Example of a class needing finalisation.
   # Has an instance variable 'file' which needs
   # to be closed.
   def __init__(self, file):
      self.file = file
      register_finaliser(self, FileWrapperFinaliser(file))

class FileWrapperFinaliser:
   def __init__(self, file):
      self.file = file
   def __finalise__(self):
      self.file.close()

In this example, register_finaliser() is a new built-in
method which stores the object and its finaliser in some
special global dict. Whenever an object is reclaimed (either
by refcount or M&S) the dict is checked for a finaliser for
that object. If one is found, its __finalise__ method is
called and it is removed from the dict.

Note that the finaliser object shouldn't be given a reference
to the original object (doing so would prevent it from ever
becoming unreachable), but only enough information to enable
the finalisation to be carried out.

If a scheme like this were adopted, it should completely
replace the existing __del__ method mechanism, so that there
would be no difference between the finalisation of cyclic
and non-cyclic trash. As such it would have to wait for
Python 2.0.

A-final-solution-to-the-finalisation-problem,
Greg



From holger at phoenix-edv.netzservice.de  Mon Jun 14 05:21:44 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Mon, 14 Jun 1999 09:21:44 GMT
Subject: os.mkdir()
References: <Pine.LNX.4.04.9906101239340.6646-100000@emerald.netskate.ru>
Message-ID: <3764C9A8.E9E33575@phoenix-edv.netzservice.de>

High,


Oleg Broytmann schrieb:
> 
> On Wed, 9 Jun 1999, Holger Jannsen wrote:
> > Look:
> > >>>os.mkdir("c:\\blabla")
> > OSError: [Errno 17] File exists: 'C:\\blabla'
> >
> > or
> > >>>os.makedirs("c:\blabla\nana\gugu")
> > produces
> > OSError: [Errno 2] No such file or directory: 'c:\blabla\nana'
> 
> os.makedirs("c:\\blabla\\nana\\gugu")
>                ^^      ^^    ^^
> 
>    Double your backslashes.

Oh, yes, sorry....;-<)))

 
> Oleg.
> ----
>     Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
>            Programmers don't die, they just GOSUB without RETURN.



From ddainese05x at x25dsi.unive.it  Tue Jun 15 13:33:18 1999
From: ddainese05x at x25dsi.unive.it (Diego Dainese)
Date: 15 Jun 1999 17:33:18 GMT
Subject: Basic python OOPs question
References: <7jot0r$p32$1@nntpd.lkg.dec.com>
Message-ID: <slrn7md3iu.t4.diego@blanka2.blankanet.it>

On Thu, 10 Jun 1999 17:45:35 GMT, Daniel wrote:
> Do I understand correctly that objects are removed
> from the name space when they go out of scope?
> And that they go out of scope based on indentation?
> 

Yes.

> If so is there a default function called on 
> objects that go out of scope?  The __init__
> function is called when an object is created.
> Is there a function that is called when the
> object is destroyed?
> 

The __del__ function.

In the Chapter 3 of the Language Reference there is the whole list of
these `special' methods.

Bye,

-- 
d i e g o
--
To reply remove the numbers and the `x' from my address
--
Sorry for my bad English!



From daniel.mendyke at digital.com  Fri Jun 18 14:02:36 1999
From: daniel.mendyke at digital.com (Daniel)
Date: Fri, 18 Jun 1999 18:02:36 GMT
Subject: wxPython under linux
Message-ID: <7ke110$o4r$1@nntpd.lkg.dec.com>

I've installed the wxPython rpm under RH 6.0
and I'm having the following problems.

First, python can't locate the main module.
wx.py is located in /usr/lib/python1.5/site-packages/wxPython.
I've tried adding this path to PYTHONPATH but it
still has no effect. 

Second, I can copy a simply script to the wxPython
directory and run it from there, however, if I'm 
under KDE it fails with X-errors.  If I switch to 
a generic window manager the script runs fine.

Has anyone else had these problems?
Can you offer me some suggestions to solve them?

-Daniel




From Klaus.Schilling at home.ivm.de  Mon Jun 14 04:47:50 1999
From: Klaus.Schilling at home.ivm.de (Klaus Schilling)
Date: 14 Jun 1999 10:47:50 +0200
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <3762988E.C94AA252@iname.com> <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu> <3763C746.5CA8FE78@iname.com> <m3u2sby4gf.fsf@soma.andreas.org>
Message-ID: <87pv2zjhhl.fsf@home.ivm.de>

Andreas Bogk <andreas at andreas.org> writes:

> Fernando Mato Mira <matomira at iname.com> writes:
> 
> > Although Dylan never really made it into "Fernando's elegibility list"
> > because
> > it was not a serious contender against Scheme and CL yet, it definitely got
> > banished the day the Lisp syntax was dropped. I'm still waiting for the day
> > when the Dylan fans will reinstate the alternative.
> 
> This is an age-old flamewar. It just so happens that there are people
> who like infix syntax and detest prefix, and then there are people who
> love their parens and hate infix.

Prefix rocks, infix is crap.

Klaus Schilling



From billtut at microsoft.com  Wed Jun 16 17:02:18 1999
From: billtut at microsoft.com (Bill Tutt)
Date: Wed, 16 Jun 1999 21:02:18 GMT
Subject: do anonymous pipes normally work on NT?
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8100DB90DE0@RED-MSG-50>


> From: radbit at my-deja.com [mailto:radbit at my-deja.com]
> 
> 
> In article <4D0A23B3F74DD111ACCD00805F31D8100DB90DDC at RED-MSG-50>,
>   Bill Tutt <billtut at microsoft.com> wrote:
> > Well, I finally noticed this thread, and I'll chime in with what I
> found out
> > so far:
> > (No, I haven't figured it all out yet, its time for sleep.)
> >
> > Symptoms:
> >  python -u cat.py < cat.py  is allowed to read from stdin.
> >  python cat.py < cat.py isn't allowed to read from stdin.
> >  python -u cat.py > blah is allowed to write to stdout
> >  python cat.py > blah isn't allowed to wrtie to stdout.
> >  python runproc.py (with -u) cat.py isn't allowed to write to stdout
> >  python runproc.py cat.py isn't allowed to write to stdout.
> >
> > -u does two things:
> > 1) turns stdin/stdout into binary streams
> > 2) turns off stdio buffering
> 
> Hi
> 
> maybe I'm missing something about what you are looking for: here are 2
> test scripts that work as anonymous pipes:
> 

That's not the question... Anonymous pipes on NT work wonderfully. (just not
the way we're expecting from this particular Python context) 
Calling SetStdHandle() is particularly silly as well, you shouldn't need to
overwrite the current stdout just to get this to work.

> win32file.WriteFile(std_out, 'this method works....')
> 

Calling GetStdHandle shouldn't be necessary either. The CRT already grabbed
that information for us when Python started up.

The question is why does sys.stdout.write() not seem to work without -u when
the above line works wonderfully.... I'm not worried about a working example
of anonymous pipes in general, thats the easy part.

Bill



From gerrit at nl.linux.org  Sat Jun 12 11:39:28 1999
From: gerrit at nl.linux.org ('Gerrit Holl)
Date: Sat, 12 Jun 1999 17:39:28 +0200
Subject: Raw Strings!
In-Reply-To: <P3v83.7006$Yb7.1482@news4.mia>; from SunitJoshi on Sat, Jun 12, 1999 at 10:30:16AM -0500
References: <P3v83.7006$Yb7.1482@news4.mia>
Message-ID: <19990612173928.A23568@humbolt.nl.linux.org>

On Sat, Jun 12, 1999 at 10:30:16AM -0500, SunitJoshi wrote:
> From: "SunitJoshi" <smjoshi at bellsouth.net>
> Newsgroups: comp.lang.python
> Subject: Raw Strings!
> Date: Sat, 12 Jun 1999 10:30:16 -0500
> To: python-list at python.org
> 
> Hi All
>         If someone could show me how to use raw-strings with variables, I
> would really appreciate it.
>         For e.g.: if str_path = "C:\\windows\\temp"
> 

var = r"D:\\stupid\\windows\\path"

note the "r" of "raw" before the string. This is always the case.

9 minutes and i reply... hmm, not bad :-)

regards,
Gerrit.



From chega_ at yahoo.com  Tue Jun  1 01:49:46 1999
From: chega_ at yahoo.com (Vadim Chugunov)
Date: Mon, 31 May 1999 22:49:46 -0700
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com><87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net> <87n1ykr0z9.fsf@ev.netlab.co.jp>
Message-ID: <7ivs9q$dc5@news.dns.microsoft.com>

> Remember the original statement is like this:
>
> >As many people have noted before, you would not be able to use
> >destructors to free up resources like memory or file descriptors.
>
>which, as I understand, means destructors can not be used with real
>GC.  That's just not true.

I'm sorry for being unclear, but the key words here were: "to free up
resources".
And I meant "to free up resources the moment I expect them to", not "some time
later".

BTW, this is a common expectation from anyone coming from a C++ background.
In C++, deallocation in destructors is the only way you can use exceptions
without leaking resources.

Vadim






From yopen at my-deja.com  Sat Jun 26 17:47:27 1999
From: yopen at my-deja.com (Bill Wilkinson)
Date: Sat, 26 Jun 1999 21:47:27 GMT
Subject: Environment inheritance under Windows NT
References: <Pine.SGI.3.95q.990621173009.142968O-100000@masox202>
Message-ID: <7l3hpf$84v$1@nnrp1.deja.com>

In article <Pine.SGI.3.95q.990621173009.142968O-100000 at masox202>,
  "Milton L. Hankins" <mlh at swl.msd.ray.com> wrote:
> Python doesn't seem to pick up the environment from its parent process
> under Windows NT.
>
> If I set an EV (environment variable) in the Windows NT System control
> panel applet, then run a command prompt and use the SET command to
change
> the variable's value, then run Python at that same command prompt,
> os.environ reflects the old (control panel) value of the EV.

yeah, bummer huh...
This has nothing to do with python, but with how NT handles environment
variables.

If you take a look after you change the variable using "set",
you will notice that there are actually two copies of your variable.
The new one can be changed on demand in the current window.

You can ease the pain a little by using the setx.exe utility that comes
with the resource kit.

I am sure there is somthing in the pythonwin extensions that would
handle this as well.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From pduffin at mailserver.hursley.ibm.com  Fri Jun 11 08:39:50 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Fri, 11 Jun 1999 13:39:50 +0100
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <wk909syqr4.fsf@ifi.uio.no> <375FF969.C879B4CD@iname.com> <37601668.D9105225@mayo.edu> <3760E130.7456447@iname.com> <3760FA1F.7BCD@mailserver.hursley.ibm.com> <3761021C.8C029142@iname.com>
Message-ID: <37610395.6799@mailserver.hursley.ibm.com>

Fernando Mato Mira wrote:
> 
> Paul Duffin wrote:
> 
> >     In Tcl if you want to pass a "reference" you pass a variable name
> 
> I was refering to this, hence the quotes.

Then I can assure you that "dynamic scoping" is not a problem.

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From tim_one at email.msn.com  Wed Jun 23 01:14:05 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 23 Jun 1999 01:14:05 -0400
Subject: List <-> news hosed again?
Message-ID: <000201bebd37$36eff8e0$cb9e2299@tim>

Seems to me the c.l.py mailing list -> news gateway stopped cold again late
last week.  Less sure of the other direction, but I do believe I've seen
some older recent articles on DejaNews that haven't come thru on the list.

If you're reading this and are not on the mailing list, share the secret
<wink>.

another-day-another-day-ly y'rs  - tim





From jcw at equi4.com  Sat Jun 19 06:56:36 1999
From: jcw at equi4.com (Jean-Claude Wippler)
Date: Sat, 19 Jun 1999 12:56:36 +0200
Subject: Python source code documentation?
References: <7kdv0l$n0g$1@nnrp1.deja.com> <7kenc7$1od$1@m2.c2.telstra-mm.net.au>
Message-ID: <376B7764.D26F56FF@equi4.com>

Mark Hammond wrote:
> 
> Sam Schulenburg wrote in message <7kdv0l$n0g$1 at nnrp1.deja.com>...
> >I have been using a freeware product called Autoduck to document my
> >Python source code.
> 
> Cool!  I use AutoDuck to document all my Python extension C++ code.
[...] 
> However, depending on how far you have gone with AutoDuck,
> pythondoc/gendoc isnt solving the exact same problem.  Well, they all
> attempt to solve the same problem, but have different strengths.

Another package you may want to look at is Mark Peskin's Perceps:
	http://friga.mer.utexas.edu/mark/perl/perceps/
The following site docs for a C++ library were done with it (frames and
all, but I didn't want to activate the Java class hierarchy browser):
	http://www.equi4.com/metakit/info/docs/

Perceps was written in that other p-language, last time I checked.

'nother idea, since we have some many of these tools, and so many
languages (including SWIG, which also disects definitions), why not
define a "language -> XML" defs/docs extractor, plus a package which
turns that into docs, pdf's, help files, web sites (indexed), and such?

-- Jean-Claude



From poinot at onera.fr  Wed Jun 16 03:43:33 1999
From: poinot at onera.fr (Marc POINOT)
Date: Wed, 16 Jun 1999 09:43:33 +0200
Subject: pygtk - clist title button
Message-ID: <376755A5.15621A4@onera.fr>

Hmmm... a question about pygtk, but I guess it's more
gtk than python :}
I make this public instead of sending it to James Henstridge
as a private email (or to gtk list).

How to connect a signal to a clist title button?

Marcvs [alias
\begin{flame}
What about using Gtk instead of Tkinter in the Python "core" built ?
\end{flame}
]






From emile at fenx.com  Thu Jun 10 15:12:38 1999
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 10 Jun 1999 12:12:38 -0700
Subject: Scripting Telnet
Message-ID: <a6U73.1265$VL2.195835@news.direcpc.com>

Hello all,

I've accepted a project and I think I can use Python for part of it.

The project is moving data collected in a warehouse management system
into a proprietary, legacy CHUI application.  The information is to be
processed hourly by a cron job under HPUX.  Most of the legacy app
source code is available, but there are some black boxes in there I
can't get at.  The app doesn't like standard input redirection, but will
run in a telnet session.

My choices appear to be 1) write a utility in the native app's
environment the mimics the result set of the current app (and I _really_
don't want to wade through that crap to figure it out); or 2) script the
input requirements and feed the application while it runs (I've done
this for smaller projects using VB sendkeys).

I feel that python ought to be good for this, and I'm hoping you'll
point me to the appropriate modules to look into or techniques to use.
I did a little browsing through dejanews and saw a reference to perl's
expect that looks like it does this kind of work, so maybe there's a
python tool too?

Trying-to-get-paid-and-learn-more-python-at-it-ly y'rs,

--

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







From rhww at erols.com  Thu Jun 10 09:19:12 1999
From: rhww at erols.com (R Wentworth)
Date: Thu, 10 Jun 1999 09:19:12 -0400
Subject: Debug:  Module to track python objects.
References: <11A17AA2B9EAD111BCEA00A0C9B4179301E10EF0@forest.origin.ea.com>
Message-ID: <375FBB50.D53B7DFC@erols.com>

> I'm open to comments, include those designed to make it more effecient,
> powerful or easier to use.

David,

One thought on ease of use... you could easily set things up so that
classes did not need to initially be defined with your Debug package
in mind.  This might work as follows (this hasn't been debugged):

class "Debug":
	class WrapMethod:
		"Helper class to wrap a class method"
		def __init__(self,appendfunc, method=None):
			self.appendfunc = appendfunc
			self.method = method
		def __call__(self, *largs, **kargs):
			if self.method:
				apply(self.method, (self,)+largs, kargs)
			if __debug__:
				apply(appendfunc, (self,))
	def DebugClass(self, Class):
		"Add debugging functionality to class Class"
		if hasattr(Class, "__del__"):
			Class.__del__ = self.WrapMethod(self.LogDel, Class.__del__)
		else:
			Class.__del__ = self.WrapMethod(self.LogDel)
		if hasattr(Class, "__init__"):
			Class.__init__ = self.WrapMethod(self.LogInit, Class.__init__)
		else:
			Class.__init__ = self.WrapMethod(self.LogInit)
		if hasattr(Class, "__destroy__"):
			Class.__destroy__ = self.WrapMethod(self.LogDestroy, 
				Class.__destroy__)
		else:
			Class.__destroy__ = self.WrapMethod(self.LogDestroy)

	...

class MyClass:
	...

debug=Debug()
debug.DebugClass(MyClass)

In addition, you could write a "DebugModule()" method which, given
a module, would find all the class definitions and apply DebugClass()
to each.
name?

Finally, the name "Debug" is a bit generic for a module.
Maybe something like "ObjLifeDebug"?

FWIW,
Bob Wentworth

"Stidolph, David" wrote:
> 
> First and foremost I want to thank Guido for the traceback system - just
> having made this module possible and my task MUCH easier.
> 
> This is an update to a debugging module that I posted before.  See the end
> of this mail for the module.
> 
> Purpose:  The Debug module tracks python class instances between their
> __init__ and __del__ method calls.  The idea is to provide tracking for
> objects so that a report can be generated with objects that have not been
> freed.  When we first ran this on our code we found almost 300 objects that
> were not being destroyed (more than half were windows resources like graphic
> texture files).
> 
> Asumptions:  The __init__ method of every class calls LogInit method of the
> debug class.  The __del__ method calls LogDel.  We have a special method
> called Destroy for every class that is designed to remove references to
> other objects and handle the circular reference problem.  The Destroy method
> calls LogDel
> 
> Useage (we have a Globals module with specific instance variables to grant
> access to modules like Debug):
> 
>         #sample
>         import Globals
> 
>         class Name:
>           def __del__(self):
>             if __debug__:
>               Globals.debug.LogDel(self)
> 
>           def __init__(self):
>             if __debug__:
>               Globals.debug.LogInit(self)
>             # your code
> 
>           def Destroy(self):
>             if __debug__:
>               Globals.debug.LogDestroy(self)
>             # your code
> 
> Questions:
> 
> 1. We are losing the tracebacks.  At some point the output terminates.  This
> happens after a large output, so I suspect a buffer overrun.
> 2. Can we shrink the tracebacks?  Keeping all the strings while the program
> runs seems a little much.  We also want to modify the output for MSVC.
> 3. Is there any other tracking we can do for catching bugs?  Ways to enhance
> this module?
> 
> 
> Thank you for your help,
> 
> David Stidolph
> Origin.

[detailed code omitted]



From aahz at netcom.com  Wed Jun  2 11:33:44 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Wed, 2 Jun 1999 15:33:44 GMT
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net>
Message-ID: <aahzFCpHw8.Gxt@netcom.com>

In article <375536C2.DD917F99 at prescod.net>,
Paul Prescod  <paul at prescod.net> wrote:
>
>I consider it a compliment that he can't find anything real to point out
>as Python's contribution to language fascism but I also find it
>distressing that he feels the need to sow the seeds of fear, uncertainty
>and doubt in the minds of would-be Python users. The Python feature that
>most of us (in retrospect, unnecessarily) worry about before trying Python
>is the whitespace handling. Larry can only turn people off of Python by
>suggesting that that is Python's central feature or philosophy.

I disagree with your parenthetical comment.  I wouldn't call myself an
"expert" Python programmer by any stretch of the imagination, but I've
written a fair bit of it in connection with a multi-person project over
the past four months.  As a seasoned programmer of many years and many
languages, I've disagreed with almost all the whines that have been
brought against Python (e.g. the stupid "=" vs. "==" thread), but this
is one point that still annoys me.

Oh, I've gotten used to it, no question about that, and I even
appreciate the readability when it comes to working over someone else's
code (which is what I've mostly been doing).  But the whitespace problem
is a real PITA when it comes time to make a change to the structure of a
program -- it's an investment I don't mind making when I'm doing "real"
programming, but when I'm making a bunch of quick, fiddly changes to
debug something it continually gets in my way.

I'm not trying to push a change, mind you.  I'm just pointing out that
IMAO the whitespace issue is a legitimate complaint.  There are plenty
of real issues to be worked on, though.
-- 
                      --- Aahz (@netcom.com)

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

"NRE is a lot like cocaine. Feels great while you're on it, makes you
feel very special and talented, and you become highly annoying to those
who aren't on it."  -- Stef



From aahz at netcom.com  Thu Jun 24 12:31:41 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Thu, 24 Jun 1999 16:31:41 GMT
Subject: class static vs. module global
Message-ID: <aahzFDuB8t.E6t@netcom.com>

I have a module aC, which conveniently holds a class aC.  During some
recent changes, I needed to share information across class instances.
Primarily to localize things, I elected to store the information in the
class.  Anyone want to share opinions on why I should instead have used
a module global variable?  Or perhaps a different approach?

(Some additional information: instances of aC get invoked by aQ; with
the exception of two methods, aQ knows nothing about the actual workings
of aC, so storing the info in aQ would be a bad idea.  If it helps any,
the information I'm storing in aC is essentially a cache that gets
flushed regularly.  Also, I was trying to preserve the structure of
existing code as much as possible to make testing and re-integration
easier.)
-- 
                      --- Aahz (@netcom.com)

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

Javascript: The 90s incarnation of the Commodore 64 Sprite -- SJM



From tismer at appliedbiometrics.com  Tue Jun  8 15:17:52 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Tue, 8 Jun 1999 19:17:52 GMT
Subject: fork()
References: <000601beb1c7$682c6220$559e2299@tim>
		<7jjghd$9il$1@cronkite.cc.uga.edu> <375D5826.23197692@inrialpes.fr>
Message-ID: <375D6C60.FF9998DE@appliedbiometrics.com>


Vladimir Marangozov wrote:
> 
> Graham Matthews wrote:
> >
> > Tim Peters (tim_one at email.msn.com) wrote:
<snip/>

> Graham, you said that you won't contribute a RC+GC scheme to Python, despite
> your positive experience. If you change your mind and consider
> devoting some spare time to give it a try, I'll devote some spare
> time to help you with Python's internals and we'll see whether we
> could come up with something viable, which could compete with the
> actual RC scheme alone. Does this sound constructive enough? :-)

Just an idea here:
If we have knowlegde of the internals of every Python structure,
and if there were no hidden internal references to Python
objects, then we could quite easily build a non-pessimistic
GC, without the need to be hardware specific, scan the stack
and so on.
You know that I have an (early) Python with no stack.
Would this help?

Let's assume that we use our own allocator. Something which
allows to create a couple of heaps which can be identified.
Now, with every new interpreter incarnation, a new heap
could be associated.
Under the assumptions that
- only known types are garbage collected
- only known (well-behaved) functions are called,
it seems to be safe to me to run a garbage collector
for all these objects which were created in the
current interpreter.
And since stackless Python has very few interpreters
(after the imports, there is almost just one),
very many objects would have this nice property.

I don't know how to handle calls to unknown functions.
Perhaps, parameters to such functions must me recorded
in some structure which marks them as "unsafe to collect".

What do you think?

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From jhauser at ifm.uni-kiel.de  Sun Jun 27 13:39:39 1999
From: jhauser at ifm.uni-kiel.de (Janko Hauser)
Date: 27 Jun 1999 19:39:39 +0200
Subject: Python for large projects?
References: <3774b5d1.690922@news.force9.net> <tx3ogi1ahos.fsf@hog.ldgo.columbia.edu>
Message-ID: <87d7yhv8ys.fsf@lisboa.i-did-not-set--mail-host-address--so-shoot-me>

Jody Winston <jody at ldgo.columbia.edu> writes:

> I've used Python for three large distributed applications (each > 100k
> lines of Python).  The only other recommendation that I would add is
> to define interfaces to the code.  If this is done with a product like
> ILU or FNORB, you can then easily call the Python code from other
> languages.  In addition, you can replace critical sections with other
> languages if needed.
> 
I have thought about this as the perfect way to do ``data hiding'',
publish only special interfaces for your data. Has this a significant
overhead, if used on one machiene, or is it overkill just for this
purpose :-)?

__Janko

-- 
  Institut fuer Meereskunde             phone: 49-431-597 3989
  Dept. Theoretical Oceanography        fax  : 49-431-565876
  Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
  24105 Kiel, Germany



From arcege at shore.net  Wed Jun 30 15:04:23 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 30 Jun 1999 19:04:23 GMT
Subject: theads & global namespaces
References: <377A5C88.32A1FAB6@notes.cba.ufl.edu>
Message-ID: <XSte3.389$6M6.125233@news.shore.net>

Taylor Anderson <andersdt at notes.cba.ufl.edu> wrote:
: Greetings all,

: I am trying to share globals between threads.  Is is possible for me to
: share globals between threads that call functions from different
: modules?  Thus far I have had little success.  I ask this question
: because  I would prefer not to put all my code in one module.

: For example, I have 2 .py files: a.py and b.py.  b.py contains the
: function ('foo()') to be executed  under the new thread, as well as the
: global variable 'x'.  a.py contains the call to
: start_new_thread(foo,()).  a.py also tries to use the global variable x.

: =========================================
: b.py

: x = 0

: def foo():
:    global x
:    x  = x + 1

: =========================================
: a.py

: from b.py import *

: if __name__ == '__main__':

:     start_new_thread(foo, ())
:     while 1:
:         if x != 0:
:             print x

: =========================================

: So far, my code in a.py does not detect any change in x.  Any
: suggestions for a workaround?


This isn't just a threads problem, but a namespace problem.

When you create a variable, it is a name bound to an object, a reference
if you will.  When you assign to that same name, often another object
will be bound.  When you import using "from...import *" you create
new bindings to these objects in the new module, but not referencing the
same bindings.

What this means is, when you use "from...import...", assignments within
the module can be hidden from the calling module:

Python 1.5.2 (#14, Jun 11 1999, 16:31:09) [C] on aix4
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import foo
>>> foo.x
1.0
>>> from foo import x
>>> x
1.0
>>> x is foo.x
1
>>> foo.x = "hi"
>>> x
1.0
>>> x is foo.x
0
>>>

When foo.x was changed, x (in the local module) did not change with it,
becuase the variable "foo.x" is not the same as the variable "x".  Even
if both pointed to the same object at first.

Your solution is to import "b" and reference the variable as "b.x".

[If your system is any more complex than what you describe above, you
might have other problems with "x" within a multi-threaded environment
if you do not have some type of synchronization.]

  -Arcege




From Cepl at fpm.cz  Wed Jun 16 11:31:31 1999
From: Cepl at fpm.cz (Matej Cepl)
Date: Wed, 16 Jun 1999 16:31:31 +0100
Subject: Python-list digest, Vol 1 #455 - 20 msgs
Message-ID: <1318D78C9072D11195C9006094EA98A72C2151@ocesrv>

> From: Lars Marius Garshol <larsga at ifi.uio.no>
> Date: 16 Jun 1999 16:59:53 +0200
> 
> You could very well use htmllib and just make an htmllib application
> that writes the HTML back out again, and modify it slightly to do
> heading numbering. I don't think HTMLFormatter would do you any good
> for this sort of thing.
> 
	[MCepl]  That was exactly my question: how to make filter with
htmllib.py doing nothing and just copying HTML input on output?

					Matthew



From mal at lemburg.com  Fri Jun  4 05:18:32 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 04 Jun 1999 11:18:32 +0200
Subject: Changing environment variables into python variables
References: <7j7f0t$165$1@nnrp1.deja.com>
Message-ID: <375799E8.247758A0@lemburg.com>

sragsdale at my-deja.com wrote:
> 
> I'm writing a script that needs to import environment variables into
> python as standard variables.  I.E. if the environment variable USER is
> set to 'soren', after my program starts up and does its thing I want
> there to be a variable named USER with a value of 'soren'.
>
> When I try to write this in python, I have this error:
> 
> -------------------------------
> import posix

You should import os, not posix...

> for var in posix.environ.keys():
>     command = var+" = \""+posix.environ[var]+"\""
>     print command
>     eval(command)

This is dangerous, as someone may easily stick some Python code
into your environ, e.g. WIPEOUT="+os.system('rm -rf /')+".

Better do something like this:

import os
namespace = locals()
for key,value in os.environ.items():
	namespace[key] = value

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   210 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From mark at _spamspamlovelyspam_btweng.krakow.pl  Thu Jun 17 17:10:10 1999
From: mark at _spamspamlovelyspam_btweng.krakow.pl (MK)
Date: Thu, 17 Jun 1999 21:10:10 GMT
Subject: Crawling Python
Message-ID: <37695707.3168269@news.tpnet.pl>

Hello everyone,

While playing with anydbm module using IDLE 0.4, I have made Python to do the
following:

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import anydbm
>>> d=anydbm.open('c:\\any.db','n')
>>> for i in range(1000):
	d[`i`]=`range(i,i+10)`
	
	
>>> d.close()
>>> 
>>> d=anydbm.open('c:\\any.db')
>>> s=''
>>> for i in d.keys():
	s=s+d[i]
	
	
>>> 
>>> len(s)
49035
>>> 


Then I tried was typing s only to see the string. Shell printed it after some 20
seconds, and then came to a crawl. It responded to a command once and then it
hang indefinitely. Techically it does not crash, but it displays a prompt,
accepts command and processes it for minutes for some unfathomable reason.
Curioser and curioser, I have said to myself. Can't I have ~50K strings without
Python crawling? I thought that unlike in other cases crawling was not max
performance for _this_ kind of animal?



v----e----r----y------s---l---o----w-----l----y------y'rs


PS. I have tried the same thing in DOS executable -- above does not happen.
Looks like only IDLE is affected. This is second feature, err, bug I have
found in IDLE. Maybe somebody is willing to give me a job? ;-)))








MK


--------------------------------------------------
Reality is something that does not disappear after 
you cease believing in it - VALIS, Philip K. Dick
--------------------------------------------------

Delete _spamspamlovelyspam_ from address to email me

postmaster at 127.0.0.1
root at 127.0.0.1
webmaster at 127.0.0.1






From Eric.Boudaillier at p98.f112.n480.z2.fidonet.org  Wed Jun 30 06:28:38 1999
From: Eric.Boudaillier at p98.f112.n480.z2.fidonet.org (Eric Boudaillier)
Date: Wed, 30 Jun 1999 11:28:38 +0100
Subject: Python GUIs: Summary and Conclusion
Message-ID: <000015a7@bossar.com.pl>

From: Eric Boudaillier <eric at unifix-online.com>

John Michelsen wrote:
> 
> Philippe, they look very nice. How did you use them with Python
> and Tkinter? Any Python code that you would like to share
> demonstrating BWidgets?  I would be especially interested
> to see use of the tree widget.

  Same as Tkinter: directly call tcl command
  To lauch the demo:

cln46ib:ficus/FicusSrc > cd ../../BWidget/demo/
cln46ib:BWidget/demo > python
Python 1.5.2 (#5, Jun 23 1999, 18:09:42) [C] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import Tkinter
>>> app=Tkinter.Tk()
>>> app.tk.call('source', 'demo.tcl')
''
>>> 

  But, if some volunteers would like to encapsulate
  BWidget in python classes...
  
-- 
Eric Boudaillier
Unifix - http://www.unifix-online.com/



From Klaus.Schilling at home.ivm.de  Mon Jun 28 13:51:07 1999
From: Klaus.Schilling at home.ivm.de (Klaus Schilling)
Date: 28 Jun 1999 19:51:07 +0200
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <37682DD3.43923243@compaq.com> <7ka2be$qja$1@nntp6.u.washington.edu> <37776FFA.2D85@mailserver.hursley.ibm.com> <lw7looxx7m.fsf@copernico.parades.rm.cnr.it>
Message-ID: <876748mcxg.fsf@home.ivm.de>

Marco Antoniotti <marcoxa at copernico.parades.rm.cnr.it> writes:

> Paul Duffin <pduffin at mailserver.hursley.ibm.com> writes:
> 
> 	...
> 
> > Correct. With Tcl you can create new control structures which
> > are indistinguishable from the built in ones, you cannot do
> > that in Python. Python is much more rigid in its syntax although
> > it does have a lot of nice hooks to allow objects to behave
> > in different ways.
> 
> But you can do that much more easily and elegantly in the L-word
> language :)

It can be done best in the Scheme, by means of the almighty 
call-with-current-continuation, the best of all control structures,
and define-syntax on top of it.

Klaus Schilling



From mlh at idt.ntnu.no  Wed Jun 16 18:52:48 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 17 Jun 1999 00:52:48 +0200
Subject: Learning Python on first page at amazon.co.uk
Message-ID: <y0jyahj21xb.fsf@vier.idi.ntnu.no>

Subject says it all :)

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From pduffin at mailserver.hursley.ibm.com  Fri Jun 11 07:50:46 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Fri, 11 Jun 1999 12:50:46 +0100
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <wk909syqr4.fsf@ifi.uio.no> <375FF969.C879B4CD@iname.com>
Message-ID: <3760F816.6B53@mailserver.hursley.ibm.com>

Fernando Mato Mira wrote:
> 
> Lars Marius Garshol wrote:
> 
> > * Fernando Mato Mira
> > |
> > | I'm trying to collect a list of all the _current_ issues making tcl
> > | a bad language choice.
> >
> > You can find three interesting articles on this at
> >
> > <URL: http://itrc.uwaterloo.ca/~papresco/tcl/>
> 
> Thanks. But the Sah article is not accessible,
> and his homepage makes no reference to it.
> 
> Regarding the Stallman article, it's a quite vague,
> and it's not sure one can extrapolate from their TECO experience
> to Tcl.
> 
> About the Connoly one, the namespace issue seems to
> be solved by [incr Tcl]. But what about that "interpreted
> multiple times"? Is it just an efficiency issue, or a semantic
> nightmare (maybe both)?
> 
> What about dynamic scoping?
> 

Dynamic scoping I assume describes the mechanism that Tcl uses
to pass references through to procedures, namely that the calling
procedure passes the name of the variable and the called procedure
creates a link from its stack frame back to the calling procedures
stack frame.

If this is the case then it is just not a problem, unless you use
it badly which of course could be said of just about any programming
construct from "goto" to "multiple inheritance".

Obviously some built in support for passing references would be nice
but the BIG advantage that Tcl has over just about every other language
apart from Lisp is its ability to morph itself (which of course can
also be abused) so it is possible to do this in Tcl itself.

> BTW, I just found:
> 
> http://www.vanderburg.org/~glv/Tcl/war/
> http://www.vanderburg.org/~glv/Tcl/war2/

As with most of these articles they are out of date and the authors
tend to have (although not all) an agenda of their own which REQUIRES
that they prove Tcl (or whatever language they are 'analysing') is 
"bad".

Take for instance the Connolly article.
	"Tcl has one great big huge namespace"
		8.0 introduced support for namespaces

	"Tcl code is interpreted multiple times during the execution of
	 a program. There is no intermediate form as in Python and Perl"
		8.0 introduced a byte compiler

The Stallman article written in 1994 is also way out of date and 
adversely affected by his agenda.

Also considering his complaints about Tcl's syntax I was amazed to find
that he chose Lisp because if you replace the '() with {} and replace 
all other () with newlines you end up with a Tcl program (almost).

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From greg.ewing at compaq.com  Thu Jun 17 20:16:56 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 18 Jun 1999 12:16:56 +1200
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <37682DD3.43923243@compaq.com> <7ka2be$qja$1@nntp6.u.washington.edu>
Message-ID: <37698FF8.BE29BB9@compaq.com>

Donn Cave wrote:
> 
> But I think
> Tcl is syntactically more adaptable, where Python kind of makes
> a virtue of its fixed ways.

Yes, someone who likes languages with highly reconfigurable
syntax would perhaps like Tcl for that reason. Personally
I find that Python provides as much flexibility as I need
or want almost all the time. 

Python's keyword arguments are quite useful for designing 
little sub-languages for certain contexts. Tk's option
settings for widgets map naturally onto keyword arguments,
for example.

Greg



From jlj at cfdrc.com  Thu Jun  3 12:47:05 1999
From: jlj at cfdrc.com (Lyle Johnson)
Date: Thu, 03 Jun 1999 11:47:05 -0500
Subject: pyGTK on Windows
References: <3757D453.A302E757@interar.com.ar>
Message-ID: <3756B189.DAD53068@cfdrc.com>

I'll put in my plug for FOX:

    http://www.cfdrc.com/FOX/fox.html

And though they're not quite ready yet, I'm working on the Python
wrappers as we speak. Should be available within a week.

Nahuel wrote:

> Hi, im about to do a cross-plataform GUI application in Linux and
> Windoze... i must
> choose the GUI toolkit.. my question is .. can i use pyGTK under
> Windows???
>
> (I need a gui that can display a table with about 4000 lines without
> crashing (not like Tk :))
>
> Nahuel Greco.




From pointal at lure.u-psud.fr  Thu Jun 10 06:31:28 1999
From: pointal at lure.u-psud.fr (Laurent POINTAL)
Date: Thu, 10 Jun 1999 10:31:28 GMT
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com>
Message-ID: <375f92b4.618467559@news.u-psud.fr>

[note Follow-up to: set to comp.lang.python]

On Thu, 10 Jun 1999 11:17:59 +0200, Fernando Mato Mira
<matomira at iname.com> wrote:

for Python vs Tcl
>:   If you have references to articles of the sort, that will be useful,
>: too.

See http://www.cwi.nl/~sjoerd/PythonVsTcl.html

I remeber having seen an article of a guy explaining why he switched
from Tcl to Python for a large project... but I loose the URL.

A+

Laurent.

---
Laurent POINTAL - CNRS/LURE - Service Informatique Experiences
Tel/fax: 01 64 46 82 80 / 01 64 46 41 48
email  : pointal at lure.u-psud.fr  ou  lpointal at planete.net 



From tim_one at email.msn.com  Wed Jun  9 22:22:00 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 9 Jun 1999 22:22:00 -0400
Subject: Python 2.0
In-Reply-To: <7jmb47$r22$1@cronkite.cc.uga.edu>
Message-ID: <000801beb2e8$05e98760$af9e2299@tim>

[various people gripe about finalizers in Java
 Graham asks for an explanation
 Tim paraphrases the Java Spec's section 12.6
]

[Graham Matthews]
> There are two obvious answers to almost all the above problems in Python
>
> a) don't call finalisers for circularly referenced collected objects.
>    Instead just collect their memory. This is not perfect but better
>    than current Python which doesn't run finalisers on circularly
>    referenced objects nor collect their memory.

Solve the problem by ignoring it?  Guido would find that more attractive
than I would <wink>.

> b) better still use a two pass collector. If you don't know what that is
>    just think about what you need to do (or not do) in pass 1 to ensure
>    that all the objects needed for a __del__ method to execute in a known
>    environment are there. <wink>

Java does this.  Start the thread from its beginning again <0.5 wink>.

circularly y'rs  - tim





From tim_one at email.msn.com  Fri Jun 11 12:08:51 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 11 Jun 1999 12:08:51 -0400
Subject: fork()
In-Reply-To: <199906110443.AAA02408@eric.cnri.reston.va.us>
Message-ID: <000901beb424$b2878200$a49e2299@tim>

[Tim]
>> ...
>> Guido will also want a doubly-linked list (where the links don't
>> count against the refcount!) so that when a dict goes out of
>> existence via the normal refcount-hits-0 route, the dict can
>> efficiently unlink itself from the list.

[Guido Himself]
> Probably, although everytime this comes up I try to figure out a
> clever scheme to use a singly-linked list, where the zero-refcnt ones
> are recycled at the next mark-and-sweep.

When a class instance (not involved in a cycle) loses its last reference, I
assume __del__ will still get called right away (else we're combining the
worst features of all possible schemes <0.5 wink>).  This is also the time
to clear the instance's __dict__ (provided the instance __del__ doesn't
resurrect the instance), and for the same reason.  Ditto for ordinary dead
objects of other kinds holding dicts.  Provided those are true, it may not
hurt much to keep the stripped dict hanging around (I see that dict.clear()
releases the internal table memory, so all that's left is the dict header).

OTOH, straightforward M&S (without generational etc gimmicks) is quite
expensive, so you won't want to run it often; e.g., I don't think we can
afford to wait until a thread dies to finish cleaning up all the non-cyclic
trash (Perl gets away with waiting that long because it *doesn't* delay
tossing any non-cyclic trash).

>> ...
>> So, ignoring that, you make one pass over the trash list just to run
>> finalizers.

> Bzzt!  We only have a trash list of dictionaries.  The finalizers are
> typically on class instances (or extension type objects) that are somehow
> referenced by those dictionaries!

Ack, yes.  I was regurgitating the (start of the) scheme we (w/ JimF) hashed
out in email a few years ago, which tracked every object.  Ignoring the
wheels means it can't get up enough speed to fly <wink>.

> Could we get away with not calling (user) finalizers on objects in
> trash cycles at all?  Since the finalization order is problematic at
> best, this almost seems acceptable: in the formal semantics, objects
> that are part of cycles would live forever, and as an invisible
> optimization we recycle their memory anyway if we know they are
> unreachable (I've got the feeling that I've seen this rule before
> somewhere.)

The usual dodge is to say that GC is emulating infinite memory, "therefore"
the user has no right to expect finalizers to ever get called.  It's usually
used to justify not adding finalizers to a language, not to castrate the
semantics of languages that already have them <0.7 wink>.

Cycles are irrelevant to that line of argument, and if you introduce a
distinction ("finalizers in cycles don't get called, all others do") then
the "optimization" is not invisible:  cleaning up the cycle can cause
objects that are not *themselves* in a cycle to become trash, so *their*
finalizers will get called (else the distinction is as inexplicable as a
coin toss), but would not have been called had the cycle been left in
unreachable peace.

IOW, Humpty Dumpty sits on a cyclic wall.  If the wall goes away, his poor
little egg butt certainly notices, and especially if the wall doesn't get to
toss out a pillow on its way to reincarnation as an omelet pan <wink>.

> We'd still get complaints "my __del__ doesn't get called" from some
> users, and the answer would still be "it's in a cycle -- use
> an explicit close"; but those who know what they are doing can be
> guaranteed that their memory [gets] recycled.

The last clause should be stronger (if for no other reason than that those
who know what they're doing can already worm around this!):  those who
*don't* know what they're doing get a guarantee that their memory will
(eventually) get recycled.  Certainly value in that.

> In other words it would be no worse than today and for some people
> it would be better.

So far as that goes, yes.  It would, e.g., be better for me -- I like to
create an occasional cycle, and almost never write __del__ methods anyway.
It would be more convenient not to need to write & call dict-clearing
.close() methods.

Hard to weigh that against the probable increases in memory and runtime;
and, philosophically, you gotta admit it's a mess.  OTOH, the combo of
finalizers, cycles and resurrection seems to be a philosophical mess even in
languages that take it all <ahem> seriously.

Don't know, Guido!  I could live happily with it provided it didn't slow
things down much, and especially provided I don't think about the semantics
at all <wink>.  But then I'm not a __del__-lover to begin with.  Anyone
following this who is?

resurrection-should-be-reserved-to-gods-ly y'rs  - tim





From bwarsaw at cnri.reston.va.us  Tue Jun  1 10:56:19 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Tue, 1 Jun 1999 10:56:19 -0400 (EDT)
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7imr6b$1s0$1@nnrp1.deja.com>
	<87pv3iq6ss.fsf@ev.netlab.co.jp>
Message-ID: <14163.62611.886487.694094@anthem.cnri.reston.va.us>

>>>>> "YM" == Yukihiro Matsumoto <matz at netlab.co.jp> writes:

    YM> I don't get it.  There's no relation between destructors and
    YM> real garbage collection.  For example, Java uses destructors
    YM> called finalizers, even though most Java VM use real GC.

Interesting side-note, re finalizers and JPython.  I recall JimH
talking about finalizers during his JPython code review a few months
back.  Here's what I wrote down at the time:

    "PyInstances were 2-5x slower [than now] because __del__() was
    implemented in a finalize method.  But [current] JVMs assume most
    classes don't have a finalize [overriding Object's no-op'd
    version] and having them on everything slows frees down /a lot/
    [emphasis somebody's :) ]"

Apologies to JimH for any transcription errors!  But I think he's
correct as implied by the discussion in section 12.6 of the Java
Language Specification:

    "We encourage implementations to treat such objects [those that
    don't override Object.finalize() or do so only trivially] as
    having a finalizer that is not overridden, and to finalize them
    more efficiently."

And if you look at the diagram and discussion in section 12.6.1, you
can see why this might be so inefficient!

-Barry



From evan at tokenexchange.com  Fri Jun 11 15:37:00 1999
From: evan at tokenexchange.com (Evan Simpson)
Date: Fri, 11 Jun 1999 14:37:00 -0500
Subject: Any Production ZOPE sites we can visit?
References: <37614A32.79FE554E@metricom.com>
Message-ID: <Xyd83.25$KR2.61466@news.corridex.com>

Michael Kersey wrote in message <37614A32.79FE554E at metricom.com>...
> Are there many production Zope sites visible on the www?

> It's a little tough to get my hands around how Zope works and what it
> can do without good examples.
The Zope.org site is its own best example; you can view source for any page.

> Is Zope being used primarily for intranets?
No idea.  I run a whole collection of small sites on Zope, but the most
extensive of them is members-only, so you might not really be able to see
much.  Nevertheless:

http://www.sylvanfoa.com
http://www.littlemotivators.com

Someone on Zope-list recently opened a
super-duper-all-singing-all-dancing-framed-JScripted site running on Zope,
at:
http://www.master.cit.be/c8/default.html





From cwebster at math.tamu.edu  Tue Jun 29 16:43:25 1999
From: cwebster at math.tamu.edu (Corran Webster)
Date: 29 Jun 1999 20:43:25 GMT
Subject: printing with print
References: <377925EF.1CBB640F@starvision.com>
Message-ID: <7lbb5d$121$1@news.tamu.edu>

In article <377925EF.1CBB640F at starvision.com>,
Angus MacKay  <amackay at starvision.com> wrote:
>is there any other way to print that with print?
>perhaps with a real function.

Yes, use sys.stdout, which is a file object:

Python 1.5.2 (#13, Jun 16 1999, 23:19:35)  [GCC egcs-2.91.60 19981201 (egcs-1.1.1  on netbsd1
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import sys
>>> foo = "there"
>>> sys.stdout.write("Hi"); sys.stdout.write(foo);
Hithere>>>

Notice that it doesn't even put a newline in.  Also, you'll need to do
conversion to a string to get the result you want for foo = 2.

>>> sys.stdout.write("Hi"); sys.stdout.write(str(foo));
Hi2>>>

>foo = 2
>print "hi" + foo

You could do the explicit string conversion here too:

print "hi" + str(foo)

Corran




From amackay at starvision.com  Wed Jun 30 15:43:48 1999
From: amackay at starvision.com (Angus MacKay)
Date: Wed, 30 Jun 1999 12:43:48 -0700
Subject: ioctl in python
References: <377A71D7.470AF9DA@starvision.com>
Message-ID: <377A7374.C6919FD3@starvision.com>

well it seems I have sort of solved my problem. if I use a
string for the 3rd arg then it works properly:
>>> fo=open('/tmp')
>>> fcntl.ioctl(fo.fileno(), 0x80047601, "hi")
'\001\000'
>>> fo=open('/')
>>> fcntl.ioctl(fo.fileno(), 0x80047601, "hi")
'\000\000'
>>> fo=open('/proc')
>>> fcntl.ioctl(fo.fileno(), 0x80047601, "hi")
Traceback (innermost last):
  File "<stdin>", line 1, in ?
IOError: (25, 'Inappropriate ioctl for device')

the question is why? it will still fail with only 2 args or an int:
>>> fcntl.ioctl(fo.fileno(), 0x80047601)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
IOError: (14, 'Bad address')
>>> fcntl.ioctl(fo.fileno(), 0x80047601, 1)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
IOError: (14, 'Bad address')

cheers, Angus.

Angus MacKay wrote:
> 
> is there some magic to the fcntl.ioctl() call?
> 
> I can do these ioctl calls in C no problem (with a small wrapper):
> (amackay at phat)~/tmp$ ./ioctltest /tmp 0x80047601
> 0x80047601: 1
> (amackay at phat)~/tmp$ ./ioctltest / 0x80047601
> 0x80047601: 0
> (amackay at phat)~/tmp$ ./ioctltest /proc 0x80047601
> /proc: Inappropriate ioctl for device
> 
> that was the Linux ioctl for EXT2 version.
> 
> but in python:
> >>> a = 0
> >>> req = 0x80047601
> >>> import fcntl
> >>> fo=open('/tmp')
> >>> fcntl.ioctl(fo.fileno(), req, a)
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> IOError: (14, 'Bad address')
> >>>



From gherman at my-deja.com  Fri Jun 25 04:05:36 1999
From: gherman at my-deja.com (Dinu C. Gherman)
Date: Fri, 25 Jun 1999 08:05:36 GMT
Subject: How to obtain an instance's name at runtime?
References: <7kqpu8$7u1$1@nnrp1.deja.com> <3771127C.E40F9BF3@home.com>
Message-ID: <7kvd8b$ueb$1@nnrp1.deja.com>

In article <3771127C.E40F9BF3 at home.com>,
  Jim Meier <fatjim at home.com> wrote:
>
> Well, there sort of is a solution, but I don't think you're going
> to like it. Incidentally, why are you trying to discover this? I'm
> not a super-meister-coder, but I can't imagine a situation where
> knowing this would be useful.

Ok, I knew I was going to look like a dumb camel right after
posting my question... Nevertheless there is a useful case of
knowing such an instance's name, at least I believe I found
one. Have a look at the following page, spot the question mark
and you'll know what I mean:

  http://starship.python.net/crew/gherman/UMLgenTest.pdf

But, PLEASE, don't start to ask me anything about the project
itself!! It's all in a very pre-pre-pre-alpha phase...

There might be some workaround solution in oder to get that
name I want, as in some of the replied in this thread, but
I'll evaluate that later...

Regards,

Dinu



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From Lyle.Johnson at p98.f112.n480.z2.fidonet.org  Wed Jun 30 23:46:02 1999
From: Lyle.Johnson at p98.f112.n480.z2.fidonet.org (Lyle Johnson)
Date: Thu, 01 Jul 1999 04:46:02 +0100
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
Message-ID: <0000160f@bossar.com.pl>

From: Lyle Johnson <jlj at cfdrc.com>

I've already worked up the binary distribution of FXPy for Windows; I just
forgot to mention it in the previous post. The binary installer can be found
at:

    ftp://ftp.cfdrc.com/pub/FOX/FXPy-0.99.42-win32.exe

I haven't worked up a binary distribution of the (C++) FOX distribution
itself, primarily because it changes most every day :) But that is in the
plans as well.

Thanks,

Lyle

Mikael Lyngvig wrote:

> On Wed, 30 Jun 1999 09:07:51 -0500, Lyle Johnson <jlj at cfdrc.com>
> wrote:
>
> >Just to add fuel to the fire, we now have a Python interface to FOX, a
> >cross-platform GUI toolkit:
> >
> >    http://www.cfdrc.com/pub/FOX/fox.html
> >
> >The latest version of the Python wrappers can be picked up at:
> >
> >    ftp://ftp.cfdrc.com/pub/FOX/FXPy-0.99.42.tar.gz
> >
> >I don't know how the speed of FXPy would compare with some of the others
> >mentioned (Tk, wxWindows, Qt, GTK, ...) but it would be worthwhile for
> >some independent opinions.
>
> I'm downloading it now.  The http URL is wrong, it should be
> http://www.cfdrc.com/FOX/fox.html.
>
> The presentation seems promising, but I'd like to suggest a binary
> distribution for PC/Intel computers.  I know this is not common for
> Unix distributions, but it's the norm for PCs.
>
> -- Mikael



From fdrake at cnri.reston.va.us  Tue Jun 29 14:02:26 1999
From: fdrake at cnri.reston.va.us (Fred L. Drake)
Date: Tue, 29 Jun 1999 18:02:26 GMT
Subject: RuntimeError?
In-Reply-To: <000401bec10d$b5bc3a60$e19e2299@tim>
References: <000401bec10d$b5bc3a60$e19e2299@tim>
Message-ID: <14201.2610.316947.169868@weyr.cnri.reston.va.us>

Tim Peters writes:
 > Honestly, I had never seen a RuntimeError before!  I found this one by

  Ooh, lucky you!  ;-)  Actually, I've seen very few.

 > digging thru the source code to see when it gets raised.  There's really no
 > pattern to it, and the RuntimeError docstring in Lib/exceptions.py

  When RuntimeError is raised by the interpreter itself, that seems to 
be the case.  When documenting modules, I've seen that a lot of them
do things like "raise RuntimeError" (note: no parameters) whenever
they need to raise an exception and the programmer hadn't already
defined an exception specific to the module.  I consider this a real
problem, since it becomes hard to catch just the right exception.  The 
ones I've seen in modules didn't seem to be so serious, just lazy
about defining new exceptions the might only get used in one place.
  I've not maintained a list of where I've see this, unfortunately:
fixing them would be a worthwhile activity.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From matomira at iname.com  Tue Jun 15 05:50:07 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Tue, 15 Jun 1999 11:50:07 +0200
Subject: <insert favourite flame-bait here> broken
References: <1282792849-95706075@hypernet.com> <3764D7C3.718E125E@iname.com> <002f01beb64d$58e7b5f0$f29b12c2@pythonware.com> <37651951.ED54894E@acm.org> <005501beb674$c5575890$f29b12c2@pythonware.com>
Message-ID: <376621CF.88C95B5E@iname.com>


Fredrik Lundh wrote:

> btw, while we're on this topic, here's an interesting
> article by kent pitman:
>
> http://world.std.com/~pitman/pfaq/cross-posting.html
>
>     "Why are crossposts generally antisocial?
>     - Physical Burden of Reading Crossposted News
>     - Consumption of Community Resources
>     - Targeted Audience
>     - Inclusivity and Exclusivity"

Well, I have no time for reading that. Obviously crossposts
are useful. I'll tell you how it works:

1.There's one issue A you want to investigate,
   and the people with a balanced view read different newsgroups,
  B, C, D, E. You even know that some people in E are too young and E is
obviously
  a contender with A. As the people in E is quite alike that in D, you
leave out E, even if you think
  it's more probable there's more people that actually know about A in E
than in D.
  You know that B is a contender with A, but there's nothing you can do
about it, as that
  seems to be the most promising group.
  So you _crosspost_ (not multiple post) to A,B,C,D
2. You have to explain that you want to be straight regarding A, so you
give an example of how
    some  thing C and is obviously not in a D tradition is `wrong' but
good. So that the same could
    be true about A.
3. You correct a point regarding C (still on the same line), which
relates to F, that you know
    about and it's good, but where something bothers you, specially
because sometimes you
    _do_ consider F. You obviously have to add F to the discussion.
4. Nobody comes up with anything against A. So you learn that your
preconceptions about
    A are wrong, that all the bad things that were said about A were
religious, and that all you can do is express your stylistic and
philosophical objections against A, but nothing more. The A people is
happy. It has been a great victory, and something has changed from
today. The A people fade off.
5. Suddenly, the silent B people start compaining. What vibe did you
touch? Was it an explicit G vibe (you don't care about telling G because
it's pointless, and you don't care about fixing G either, even if you
`are a G' (by force)), or an implicit B vibe? Is it just some people
that think they own the truth? And that do not like learning? (If
something tells you you're right, and you already believe that, you
learn nothing. If they tell you you're wrong, at least you'll
contemplate the issue) Or is it that they were silent only while
something bad about A was looming?
6. Suddenly, your beliefs about B are changed. It seems that even if B
were good, you don't want to have anything to do about it, because then
you would have to hang around with the Bs. And you think that maybe A is
better just because the As are more mature (of course you've never set
foot in A). Oh, are the As having a party.

In conclusion, when choosing a language one must not underestimate the
style of its community. And actually,
not wanting to hang around in G, even if I have used G for the last 4
years has always struck me as basically perverse (actually, I even went
offline for a couple years, as there was not a compelling reason to be
wired. Occasionals emails and FTPs were enough).

May I have my PhD in Anthropology now? ;-)





From evan at tokenexchange.com  Fri Jun  4 12:51:09 1999
From: evan at tokenexchange.com (Evan Simpson)
Date: Fri, 4 Jun 1999 11:51:09 -0500
Subject: Alternative dictionary output format?
References: <7j8pd8$f01$1@nnrp1.deja.com> <slrn7lg030.i5l.jepler@craie.inetnebr.com>
Message-ID: <WtT53.20$KR2.41493@news.corridex.com>

I don't know if this is 100% bulletproof, but it has simple behavior:

pctsub = re.compile('%(?!%*\()').sub
def pfstr(s, pctsub = pctsub):
  '''Double all '%' except those leading up to a '(', for dict formatting'''
  return pctsub('%%', s)

>>> pfstr("%(href)s/foo/bar?x=%m&y=%n") % {'href': 'baz' }
'baz/foo/bar?x=%m&y=%n'

All percent signs pass unharmed, except for those in front of a left paren.
In that case '%(href)' => 'baz', '%%(href)' => '%(href)', '%%%(href)' =>
'%baz', etc.

Jeff Epler wrote in message ...
On Fri, 04 Jun 1999 14:55:42 GMT, rael at my-deja.com
<rael at my-deja.com> wrote:
>s = "%(href)s/foo/bar?x=%m&y=%n"

(he wants to only substitute the %(href) and not the rest)

s = "%(href)s/foo/bar?x=%%m&y=%%n"

i.e., double the percents you want to have not interpolated at this stage
of the game.

A clever regular expression might be able to do this for you, but the way
to write it 100% bulletproof escapes me.






From MHammond at skippinet.com.au  Wed Jun  9 08:58:55 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Wed, 9 Jun 1999 22:58:55 +1000
Subject: makepy not war
References: <375DD2BE.D918CCCE@yahoo.com>
Message-ID: <7jlobk$85f$1@m2.c2.telstra-mm.net.au>

Flbill Blipf wrote in message <375DD2BE.D918CCCE at yahoo.com>...
>I hope this is not too general of a question.
>
>After running makepy.py on a type library, how do I use
>the resulting file?
>
>I know how to use win32com.client.Dispatch, and I kinda
>sort of know how to use CoCreateInstance, QueryInterface,
>etc. with C++.

You should be able to just use "win32com.client.Dispatch", and it will
magically use the makepy support.  Print the repr() of a COM object with and
without makepy support, and the difference should be obvious.

If it doesnt, let me know (or search deja-news for discussions about how to
use Notes, which is one that doesnt work correctly)

Mark.





From bhosu at my-deja.com  Tue Jun  1 14:04:09 1999
From: bhosu at my-deja.com (bhosu at my-deja.com)
Date: Tue, 01 Jun 1999 18:04:09 GMT
Subject: Handling variable number of arguments.
Message-ID: <7j17dm$pks$1@nnrp2.deja.com>

Hi All,
   I am embedding Python in an application(C++). I would like to pass
variable number of args to the python function. I could do that from
python by using def foo(*args). How do I parse these arguments inside my
application? In PyARg_ParseTuple, I have to know exactly how many args
to read etc. I am using Python v 1.4. I was looking at
PyArg_ParseTupleAndKeywords but that needs the args to be a dictionary.
I am looking a similar thingy for lists. A typeical scenario of the
usage is like,
import foo
foo.set_args('bar','=','1','2','3')

where foo is the module containing the method implementation.

Thanks,
Bhosu.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From graham at sloth.math.uga.edu  Wed Jun  2 16:50:33 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 2 Jun 1999 20:50:33 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7iv86a$iud@news.dns.microsoft.com> <87vhd7nsra.fsf@ev.netlab.co.jp> <3754F5D7.2B6E1954@pop.vet.uu.nl> <aahzFCpG40.EFs@netcom.com>
Message-ID: <7j45ep$908$3@cronkite.cc.uga.edu>

Aahz Maruch (aahz at netcom.com) wrote:
: I'm curious: does GC provide any advantages other than its ability to
: deal with circular references?

You have obviously never worked on a large C project! GC promotes code
modularity. For example say I am using some library, L, and I do a call
to one of it's routines, say f, as follows,

	b = some_temporarily_allocated_thing
	a = L.f(some_arguments_including_b)

Now at this point does my code have to decref b, delete b, incref a?
The answers to all these questions depends on the precise calling
conventions of the routine f in L. Now what say I had called function
g in L, instead of f, then what would have been the answer to those
questions? It depends on the conventions for g, which may of course
be different to those for f. Moreover what happens if I upgrade to
a new version of the library L -- then what are the answers to the
above questions (they might depend on the version number of the
library L!).

In a garbage collection system you have none of these problems. Code
modularity is improved.

graham

PS: before anyone scoffs I should note that I have encountered all of 
the above in large C projects. As well as the usual errors where the
library code forget to incref/decref, or decreffed where they should
have increffed, etc.
-- 
             Like a bird on a wire
        Like a drunk in a midnight choir
            I have tried in my way
                  To be free



From mhagger at blizzard.harvard.edu  Fri Jun 18 16:12:26 1999
From: mhagger at blizzard.harvard.edu (Michael Haggerty)
Date: 18 Jun 1999 16:12:26 -0400
Subject: Crawling Python
References: <37695707.3168269@news.tpnet.pl>
Message-ID: <xj0vhcl5kut.fsf@cyclone.harvard.edu>

mark at _spamspamlovelyspam_btweng.krakow.pl (MK) writes:

> >>> s=''
> >>> for i in d.keys():
> 	s=s+d[i]

Please, before you complain about Python's speed, think twice about
the algorithm that you are using.  The above algorithm is O(n^2),
which means that if you have n keys the time it takes is proportional
to n squared.  That's because each time you add to the end of the
string, the whole string has to be copied.  (Unless there is a
nontrivial optimization built into Python for this special case, but I
don't think so.)

But if instead you do

>>> import string
>>> s = []
>>> for i in d.keys():
>>>   s.append(d[i])
>>> s = string.join(s, '')

then you get exactly the same result in O(n) time, because s.append()
doesn't have to copy each time.  For large n the second algorithm will
beat the first one every time, by an increasingly huge margin.

It may be that idle has problems too, but at least part of this
problem can be blamed on the algorithm.

Michael

-- 
Michael Haggerty
mhagger at blizzard.harvard.edu



From faassen at pop.vet.uu.nl  Tue Jun 15 13:35:20 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Tue, 15 Jun 1999 19:35:20 +0200
Subject: GNU python?
References: <3765B8B4.18DE0257@spots.ab.ca> <37667309.F9BBAE9A@mojam.com>
Message-ID: <37668ED8.1103C514@pop.vet.uu.nl>

Skip Montanaro wrote:
> 
> Maxwell Sayles wrote:
> >
> > Looking to try out python... is there a GNU python for Win9x?  and a
> > tutorial maybe?
> 
> Not sure just how a GNU Python would differ from the real thing, 

* It'd only come with the GNU Public license. 

* GNU Python would use the GNU coding style guidelines (in Python!)

* It'd be used for Emacs extensions.

* It'd do all those functional programming things. Closures and macros
and so on.

* Zope wouldn't have been developed because of the GPL, and then would
never have been opensourced.

* No COM support, unless provided by Cygnus. 

* Cygnus comes with 'peggton', an experimental development adding static
type checking and more optimizations. 

* Any product of <name> would have to be called GNU-Python <name>,
because after all the total size of the library modules imported by that
product would be larger than the product code itself. 

Actually-I-like-the-FSF-ly yours,

Martijn



From tim_one at email.msn.com  Thu Jun  3 00:01:38 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 3 Jun 1999 00:01:38 -0400
Subject: pickle vs .pyc
In-Reply-To: <7j4rf5$5h0$1@mlv.mit.edu>
Message-ID: <000901bead75$c7c04cc0$309e2299@tim>

[Michael Vezie]
> ...
> This is the first I've heard of marshal.  I take it from various
> hints here and there that the .pyc files are handled internally
> by marshal, yes?

Yes, "code objects" in particular are the one type marshal can handle that
pickle can't.  I think we try to sell that as "a feature", though <wink>>

> ...
> One final asside, I've seen disagreements both online and in the
> web docs about the spelling of marshal.

Guido's spelling is the only one we'll mention here, thank you.

> ...
> marshaling-up-the-courage-to-convert-working-code-to-marshal'ly yrs
>
> ... (hoping-this-'ly yrs-tag-isn't-trademarked-or-only-available-
> to-those-who've-achieved-some-level-of-pythondom'ly yrs)

No particular qualifications, fees or exams are needed to use a "-ly y'rs"
closing.  I do own exclusive worldwide rights to it, but grant everyone an
unlimited license in perpetuity.  How can I afford to do this?  Easy:  there
are only so many people out there who want to come across as an ass year
after year <wink>.

it's-not-the-form-it's-the-content-ly y'rs  - tim





From mal at lemburg.com  Tue Jun  1 05:33:07 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Tue, 01 Jun 1999 11:33:07 +0200
Subject: Any Python-friendly web presence providers?
References: <m3vhda8cv7.fsf@ime.net>
Message-ID: <3753A8D3.6B0CAA70@lemburg.com>

Matt Gushee wrote:
> 
> Greetings, Pythoneers--
> 
> I'm working on setting up a (virtual) internet domain ... reserved my
> domain name today ... now I need to find a place to put it. I would
> like to use Python for CGI scripts. Before I wear myself out searching
> the web, I thought maybe someone here could recommend a provider.

You can use basically any decent provider which gives you at least
a cgi-bin dir to execute CGI programs and a FTP connection. No need
to have Python preinstalled; just install it yourself using one
of these easy-install binaries:

	http://starship.skyport.net/~lemburg/mxCGIPython.html

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   213 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From rwentwor at bna.com  Wed Jun 30 15:14:45 1999
From: rwentwor at bna.com (rwentwor at bna.com)
Date: Wed, 30 Jun 1999 15:14:45 -0400
Subject: Reflection in Python?
Message-ID: <852567A0.0069BB98.00@notes1.bna.com>

Assuming you're working with Python class instance objects,
something like the following might work:

     # define class
     class SillyMathTest:
          def testAdd(self, x, y):
               return x+y

     # create class instance object
     MathTest = SillyMathTest()

     # get reference to bound method MathTest.testAdd
     test = getattr(MathTest, "testAdd")

     # effectively do MathTest.testAdd(3,5)
     test(3,5)

As to getting naming info: well, you already have the
name of the method as a string, you can get the name of
the class as MathTest.__class__.__name__, but there
is no way to extract the string "MathTest" from the
MathTest object because "MathTest" is the name of
a variable referencing the object, not something that
is really part of the object.  (You couldn't do this in
Java either, so I'm not sure what you had in mind.)

Bob Wentworth



--original message--
I've just started programming in Python, and I
could really use some help figuring out something.
 I'd like to implement Kent Beck's unit tester in
Python (someone posted a very nice alternative a
few weeks ago, but Junit's approach works better
for my project's needs).  How do I do the
equivalent of what Java calls reflection? I wantto do the
following:

     test = MathTest("testAdd")

where test.run() will run the "testAdd" method on
the MathTest object?  I also want to be able to
tell the user the name of the object (MathTest)
and the method in case the method fails.  Anythoughts?





From fbo at dde.dk  Fri Jun 25 07:24:26 1999
From: fbo at dde.dk (Finn Bock)
Date: Fri, 25 Jun 1999 13:24:26 +0200
Subject: Pb. with JPython & VisiBroker
References: <7kr72p$942$1@news.imaginet.fr>
Message-ID: <377366EA.AA48A43F@dde.dk>

Franck Arnulfo wrote:

> I try to use the scripting facilities of JPython with VisiBroker for Java
> ORB.
> But I've a problem to make an instance of the ORB.
> Here is a copy of the output :
> 
> C:\JPython-1.1beta2>set PATH
> PATH=C:\jdk1.1.7B\bin
> PATHEXT=.COM;.EXE;.BAT;.CMD;.VBS;.JS
> 
> C:\JPython-1.1beta2>set CLASSPATH
> CLASSPATH=C:\jdk1.1.7B\lib\classes.zip;C:\vbroker\lib\vbjorb.jar;C:\vbroker\
> lib\vbjapp.jar
> 
> C:\JPython-1.1beta2>pythonjdk11
> JPython 1.1beta2 on java1.1.7B
> Copyright (C) 1997-1999 Corporation for National Research Initiatives
> >>> from org.omg.CORBA import *
> >>> orb = ORB.init(None, None)
> Traceback (innermost last):
>   File "<console>", line 1, in ?
> NameError: ORB
> >>>

I known, this is not much help, but I can access the ORB just fine.

[c:\storage]jpython
packageManager: processing new jar,
"C:\java\JPython-1.1beta2\jpython.jar"
packageManager: processing new jar, "C:\storage\vbjorb.jar"
packageManager: processing new jar, "C:\java\jdk1.1.8\lib\classes.zip"
JPython 1.1beta2 on java1.1.8
Copyright (C) 1997-1999 Corporation for National Research Initiatives
>>> from org.omg.CORBA import *
>>> ORB.init(None, None)
com.visigenic.vbroker.orb.ORB at 20db91
>>>

I didn't properly install Visibroker. I just added a vbjorb.jar I found
from a altavista search to my CLASSPATH. Consequently, I have no clue to
which version of vbjorb.jar I'm using.

-- 
Regards,
Finn Bock



From phd at emerald.netskate.ru  Tue Jun 22 11:25:38 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Tue, 22 Jun 1999 19:25:38 +0400 (MSD)
Subject: restoring signal handlers
In-Reply-To: <XFMail.990622080335.nathan@islanddata.com>
Message-ID: <Pine.LNX.4.04.9906221924520.249-100000@emerald.netskate.ru>

On Tue, 22 Jun 1999, Nathan Clegg wrote:
> Is it necessary to restore original signal handlers before quitting a

   No.

> ----------------------------------
> Nathan Clegg
>  nathan at islanddata.com

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From n at n.com  Mon Jun  7 13:05:21 1999
From: n at n.com (Jr. King)
Date: Mon, 7 Jun 1999 13:05:21 -0400
Subject: more python to c/c++ questions.
References: <7jgqas$6a6$1@rtpnews.raleigh.ibm.com>
Message-ID: <7jgu2u$19s6$1@rtpnews.raleigh.ibm.com>

Ok, I found some info on it but, it still doesn't work

I changed my c module to this...
<snip>
static PyObject* objPass(PyObject *self, PyObject* args){
    PyObject *result = NULL;
    PyObject *temp;

    if (PyArg_ParseTuple(args, "O", &temp)) {
   PyObject* xarg = Py_BuildValue("(s)","i");
   xarg = PyObject_GetAttr(temp,xarg);  // not getting anything but null
   int val;
   PyArg_ParseTuple(xarg,"i",&val);
 }
    Py_INCREF(Py_None);
    result = Py_None;
    return result;
}

Python module snippet

class Cow:
 i = 10
 x = 0
 str = "string"
 def genie(self):
  print str

testmeth.setCallback(setN)
testmeth.testmeth(13,9)
cow = Cow()
testmeth.objpass(cow);

Thanks
jamarijr at liam_toh.com
                 ^hotmail...  Must spoil spammers






From huy4huy at cs.com  Wed Jun  2 14:47:40 1999
From: huy4huy at cs.com (HUY4HUY)
Date: 2 Jun 1999 18:47:40 GMT
Subject: need scripts...help
Message-ID: <19990602144740.17875.00000120@ng-cf1.news.cs.com>

I need a program that would automate an age verification system, I need the
script to handle all aspects.

It should have easy sign up for webmasters and surfers, anti cheater devices as
well as the real time monitoring of all system activities, email notification
to the webmaster joining to the admin, logs all traffic trough the system and
gathers email addresses, built in password protection for admin area, no SSL's,
no htaccess...this is an example of the format I need.  I will appreciate it,
if anyone can help me out......thanks
--------------------------------------------------------------------
Mark Davidson             inert28 at hotmail.com
please inform with the information to the email address above, thanks.




From NOSPAMbruce_wheeler at hp.com.NOSPAM  Fri Jun  4 15:48:45 1999
From: NOSPAMbruce_wheeler at hp.com.NOSPAM (Bruce Wheeler)
Date: Fri, 04 Jun 1999 12:48:45 -0700
Subject: python active web pages
References: <jonathan-0406991640390001@borg.onegoodidea.net>
Message-ID: <37582D9D.3D7A4B1D@hp.com.NOSPAM>

Jonathan Hogg wrote:
> 
> hi there,
> 
> i'm looking to embed python into web pages in a similar way to ASP and
> PHP3 and so forth.
> 
take a look at <http://starship.python.net/crew/mhammond/>



From salvador at my-deja.com  Wed Jun  2 07:30:43 1999
From: salvador at my-deja.com (salvador at my-deja.com)
Date: Wed, 02 Jun 1999 11:30:43 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net> <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net>
Message-ID: <7j34l3$hbj$1@nnrp1.deja.com>

In article <375372F6.A874554 at prescod.net>,
  Paul Prescod <paul at prescod.net> wrote:
> Yukihiro Matsumoto wrote:
> >
> > >As many people have noted before, you would not be able to use
> > >destructors to free up resources like memory or file descriptors.
> >
> > which, as I understand, means destructors can not be used with real
> > GC.  That's just not true.
>
> Is there a language with real GC where destructors are called as
> predictably as they are in Python?

No, but it could be done in Python, GC isn't incompatible with reference
counting and both can be combined. A simple GC could be added just to
detect ciclical data structures and remove it

Even new special obect methods could be suported by the language to
handle it, say __gc__, that gets called by the GC requesting the object
to liberate its references to other objects so the ciclid data breaks.

> > |But I think that the bigger problem with "real GC" is that C
programs
> > |often hang on to references to Python objects and any kind of
compacting
> > |GC would invalidate those references.
> >
> > It's the common legend especially here in comp.lang.python.
> >
> > >From my experience with Scheme and Ruby, both with real GC, the
real
> > GC is mostly very fast.
>
> Speed wasn't my concern. I asked about references in C code. Do we use
> PyObject **s instead of PyObject *s now?

GC doesn't need to move objects in memory, it's only an optimization
that speeds up memory allocation (it could be implemented with reference
counting too).

The problems associated to GC an foreign modules are usually caused
because the GC doesn't know enough about external objects to determine
its references to native language objects.

 - Salva.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From Gaetan_Corneau at baan.com  Wed Jun  9 22:44:31 1999
From: Gaetan_Corneau at baan.com (Gaetan Corneau)
Date: Thu, 10 Jun 1999 02:44:31 GMT
Subject: COM dates
Message-ID: <816010E2456BD111A48700805FBBE2EEFDF354@ex-quebec-u1.baan.com>

Thanks Mark!

Another question: why no "strptime" function under WIN32? This function is
not, as far as I know, difficult to implement.

Tnanks again,
______________________________________________________
   Gaetan Corneau
   Software Developer (System integration Team)
   BaaN  Supply Chain Solutions  
   E-mail: Gaetan_Corneau at baan.com        
   Compuserve: Gaetan_Corneau at compuserve.com  
   ICQ Number: 7395494             
   Tel: (418) 654-1454 ext. 252          
______________________________________________________
"Profanity is the one language all programmers know best"

> -----Original Message-----
> From:	Mark Hammond [SMTP:MHammond at skippinet.com.au]
> Sent:	Wednesday, June 09, 1999 6:41 PM
> To:	python-list at cwi.nl
> Subject:	Re: COM dates
> 
> Check out the win32 extensions help file.  This has a discussion of COM
> dates.  In a nutshell, you use the int() function to convert it to a
> standard Python time module date, and you can assign a standard Python
> time
> module date to a COM date field and it will do the right thing.
> 
> Mark.
> 



From neelk at brick.cswv.com  Mon Jun 28 23:09:10 1999
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 28 Jun 1999 22:09:10 -0500
Subject: [ANN] DejaGrabber 0.1 -- search Dejanews with Python
References: <7kuu73$1ki$1@brick.cswv.com> <8DF363D15duncanrcpcouk@news.rmplc.co.uk> <al89094u0e3.fsf@myntti.helsinki.fi>
Message-ID: <7l9dcm$an$1@brick.cswv.com>

In article <al89094u0e3.fsf at myntti.helsinki.fi>,
Markus Stenberg  <mstenber at cc.Helsinki.FI> wrote:
>duncan at rcp.co.uk (Duncan Booth) writes:
>> Nice bit of work. Two comments though:
>> Why does get_messages(4) return 5 messages? The doc string implies it 
>> should return only as many as its argument.
>
>It's neat, admittedly (I thought of writing one myself the other day, and
>then found that one ;->). The first time I need to do some serious Deja
>action, I'll be using it :>.
>
>There's some oops in slicing's internal logic at get_messages; following
>patch fixed behavior (for me, anyhow). 

[Code snipped]

Thanks -- I've applied the change and put the new version up. 

>> It would be useful if a test example was included in a "if 
>> __name__=='__main__'" block at the end of the file. At the very least the 
>> example you give above, but perhaps more usefully a full command line 
>> driven grabber.

This is a good idea, but I'm not sure what the right interface would
look like. I suspect that producing a command-line interface useful 
enough to play well with other tools would be a bigger job than the class
itself -- a decent UI must be flexible, and flexibility is always
trickier than it seems.

That said, I'm all for it, and will probably take a shot at it RSN. A
query can return multiple messages, and I think that we can want as
output either:

a) A list of headers,
b) A big file of headers+messages,
c) A lot of little files each with a message in it,
d) A single HTML file with a list of headers with links 
   to the message bodies (using HTML in-page links). 

Any I missed, or any of these fundamentally the wrong thing to do for
a reason I'm not seeing?


Neel



From Christian.Caremoli at der.edf.fr  Tue Jun 22 08:33:42 1999
From: Christian.Caremoli at der.edf.fr (Christian Caremoli)
Date: Tue, 22 Jun 1999 14:33:42 +0200
Subject: creating dictionnaries
Message-ID: <376F82A6.82497D2B@der.edf.fr>

Hi,
Dictionnaries can be created like that :
d={'a':1,'b':2}

By calling a function you create a dictionnary like that :
def f(**d):
   return d

d=f(a=1,b=2)

I would like to be able to create dictionnaries with some similar syntax
like keyed tuples :
d=(a=1,b=2)

Is there a way to do that ?

Thanks

Christian Caremoli




From faassen at pop.vet.uu.nl  Tue Jun  8 16:43:35 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Tue, 08 Jun 1999 22:43:35 +0200
Subject: In what order does Python read dictionary keys?
References: <7jjnt5$snk$1@nnrp1.deja.com>
Message-ID: <375D8077.EA539C6D@pop.vet.uu.nl>

tcpeter at my-deja.com wrote:
> 
> Hello, all.  I've checked the docs and queried DejaNews for this but I
> couldn't come up with an answer.  I wrote a script to convert rgb
> values to hex for when I'm doing colors in HTML pages and don't have
> Photoshop handy.  The code appears below.  My question is, how does
> Python decide in which order to read the keys?  Notice that I have the
> dictionary set up in this order: red, blue, green.  (I mistyped it while
> writing the code).  But when I ran the code, the options were presented
> in the desired order. (A trifle creepy, if you ask me.  Turing would be
> delighted.)  Anyway, how does Python decide the order. Thanks.

Hi there,

The answer is that Python doesn't have any set orders for dictionaries.
The internal hashing algorithm the python dictionary uses has no
guaranteed order. So basically you shouldn't depend on any particular
order it produces.

Regards,

Martijn



From faassen at pop.vet.uu.nl  Wed Jun  9 04:36:55 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Wed, 09 Jun 1999 10:36:55 +0200
Subject: "Programming Python" still worthwhile?
References: <1283218560-70098357@hypernet.com> <7jl750$qsu$1@news.gstis.net>
Message-ID: <375E27A7.4871EA36@pop.vet.uu.nl>

Rick Robino wrote:

> Gordon McMillan <gmcm at hypernet.com> wrote:
> > Python 2 is around a really, really big corner, and then down the
> > block quite a ways. Python 1.6 sometime in 2000 (which will be
> > backwards compatible), so no 2 until 2001 or 2002.
> 
> I didn't do alot of research, but I couldn't get a good feel from
> the huge thread that has been going on here named after 2.0. Appreciate
> the ETA.

Python 2, while one of the favorite topics, is actually (as far as
known) in complete vaporware stage, and no public designs exist. Of
course, Python 2 will also solve all our problems, be resistent to
whitespace eating nanoviruses, and be both garbage collected and
reference counted, automagically. It'll feature the solution to
assignments in expressions too!

For some reason Pythoneers in this newsgroup (me included) like to talk
about improving the language, though it's a very nice language already.
Although some see this as a bad sign, I think the fairly intelligent
discussion that *usually* is the result is quite interesting. Some
discussions excepted. :) 

Thanks to some newsgroup luminaries like Tim Peters (Tim, you're a
newsgroup luminary now!), who seems to have programmed in just about any
language in existence and can compare them all fairly to Python, the
level of discussion remains fairly mature at most times. Though the
latest Python 2 thread became a bit too emotional for my tastes in
places.

The *true* Python 2 things are the rare statements made by Guido on it
(and rumors of his statements :).

Regards,

Martijn



From aahz at netcom.com  Wed Jun 16 00:39:21 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Wed, 16 Jun 1999 04:39:21 GMT
Subject: Newbie: easy filecopy?
References: <376531D3.7E42AC42@phoenix-edv.netzservice.de> <37665D08.267A697C@phoenix-edv.netzservice.de>
Message-ID: <aahzFDEKxL.92I@netcom.com>

In article <37665D08.267A697C at phoenix-edv.netzservice.de>,
Holger Jannsen  <holger at phoenix-edv.netzservice.de> wrote:
>
>The problem with 
>f=open(aFile)
>x=f.read()
>f.close
>f.open(toFile)
>f.write(x)
>f.close
>is, that some special ASCII-codes are not copied as usual 
>under Windows-systems, so e.g. CarriageReturn (tested with normal
>Asccii-text).

Open the file in binary mode.
-- 
                      --- Aahz (@netcom.com)

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

"That doesn't stop me from wanting to work all three of them over with
the clue stick for while, with no safewords allowed."  --abostick



From faassen at pop.vet.uu.nl  Wed Jun  2 14:00:48 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Wed, 02 Jun 1999 20:00:48 +0200
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7iv86a$iud@news.dns.microsoft.com> <87vhd7nsra.fsf@ev.netlab.co.jp> <3754F5D7.2B6E1954@pop.vet.uu.nl> <aahzFCpG40.EFs@netcom.com>
Message-ID: <3755714F.294CCB8D@pop.vet.uu.nl>

Aahz Maruch wrote:
 
> It might make more sense to simply add tools to locate and stomp
> circular references, perhaps augmented by explicit calls to delete
> specific instances of circular references.

Well, those tools would essentially be a form of garbage collecting,
right?

> I'm curious: does GC provide any advantages other than its ability to
> deal with circular references?

I think certain forms of GC can use less memory; I don't think all forms
of GC need extra data for all objects, like you need to store the ref
counter in.

Regards,

Martijn



From gerrit.holl at pobox.com  Sun Jun 27 05:28:41 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sun, 27 Jun 1999 11:28:41 +0200
Subject: Ellipsis?
Message-ID: <19990627112841.A31130@optiplex.palga.uucp>

Hallo,

In the python reference manual, there is this text:

Ellipsis:
This type has a single value. There is a single object with this value.
This object is accessed through the built-in name Ellipsis. It is used
to indicate the presence of the "..." syntax in a slice. Its truth value
is true. 

Can someone please explain that to me? What's so special about '...'?

groeten,
Gerrit.

-- 
The Dutch Linuxgames homepage:	http://linuxgames.nl.linux.org
Personal homepage:		http://www.nl.linux.org/~gerrit/

Discoverb is a python program (in several languages) which tests the words you
learned by asking it. Homepage: http://www.nl.linux.org/~gerrit/discoverb/



From ovidiu at cup.hp.com  Thu Jun 24 20:28:59 1999
From: ovidiu at cup.hp.com (Ovidiu Predescu)
Date: Fri, 25 Jun 1999 00:28:59 +0000
Subject: signal handler
References: <376F87CB.DC3802E6@pk.highway.ne.jp>
Message-ID: <3772CD4B.38677556@cup.hp.com>

Thooney Millennier wrote:
> 
> Hi, I want to use "signal()" in my program.
> 
>   import signal;
>   def func(arg1,arg2):
>       DO_SOMETHING_GOOD;
>       print "OK!";
>   signal.signal(SOME_SIGNAL,func);
> 
> It seems "func()" is not called
> when the program gets signal "SOME_SIGNAL".
> Why?

It looks like you forgot to send the signal. One way you could implement
this code is:

from signal import signal, SIGINT
from os import getpid, kill

def signalHandler (signum, frame):
    print 'signal handler called for signal %d' % (signum,)

signal (SIGINT, signalHandler)
kill (getpid (), 

Take a look at the manpage for signal(7) to get the list of signals you
can send to a process.

Hope this helps,
--
Ovidiu Predescu <ovidiu at cup.hp.com>
http://www.geocities.com/SiliconValley/Monitor/7464/



From garry at sage.att.com  Tue Jun 15 09:43:30 1999
From: garry at sage.att.com (Garrett G. Hodgson)
Date: Tue, 15 Jun 1999 13:43:30 GMT
Subject: lockfile problem
Message-ID: <37665882.9AEA315A@sage.att.com>

i'm having some trouble using mailman, and it appears to
be a problem with lockfiles and either nfs or sgi.
given the following program:

import sys, os, marshal, string, posixfile, time

ou = os.umask(0)
try:
	lock_file = posixfile.open(
		os.path.join( '/home/mailman/locks', '%s.lock' % 'testlock' ),
			'a+')
finally:
	os.umask(ou)
lock_file.lock('w|', 1)

when i run it on a sunos machine, where /home/mailman actually resides,
it works.  when i run it on our sgi web server machine, where it is nfs
mounted, it fails with:

$ testlock.py   
Traceback (innermost last):
  File "testlock.py", line 12, in ?
    lock_file.lock('w|', 1)
  File "/usr/local/lib/python1.5/posixfile.py", line 190, in lock
    flock = fcntl.fcntl(self._file_.fileno(), cmd, flock)
IOError: (22, 'Invalid argument')

any ideas how i can fix this?

-- 
Garry Hodgson			comes a time
garry at sage.att.com		when the blind man
Software Innovation Services	takes your hand, says,
AT&T Labs			"don't you see?"



From DOUGS at oceanic.com  Wed Jun 23 05:17:13 1999
From: DOUGS at oceanic.com (Doug Stanfield)
Date: Wed, 23 Jun 1999 09:17:13 GMT
Subject: Help with String Manipulation
Message-ID: <5650A1190E4FD111BC7E0000F8034D26A0EE6B@HUINA>

> From: wtanksle at dolphin.openprojects.net on Tuesday, June 22, 1999 3:26 PM

> Whew!  Is there any reason to not use: ... ??

The opportunity to challenge with a reference that I consider a good read?

Actually, I was deliberately much more pedantic than you made me out to be.
Doing:
>>> testring = "This is an example"
>>> testlist = map (ord, testring)
>>> testlist
[84, 104, 105, 115, 32, 105, 115, 32, 97, 110, 32, 101, 120, 97, 109, 112,
108, 101]

I thought made it much more obvious that a transformation happened, where a
newbie might confuse:
>>> x = list("This is an example.")
>>> x
['T', 'h', 'i', 's', ' ', 'i', 's', ' ', 'a', 'n', ' ', 'e', 'x', 'a', 'm',
'p', 'l', 'e',
 '.']

as having not done anything.  I know I had that problem as a beginner and
was hoping this would somehow help illustrate mutable and immutable more
clearly.  YMMV.

-Doug-



From Max.M..Stalnaker at p98.f112.n480.z2.fidonet.org  Wed Jun 30 11:11:10 1999
From: Max.M..Stalnaker at p98.f112.n480.z2.fidonet.org (Max M. Stalnaker)
Date: Wed, 30 Jun 1999 16:11:10 +0100
Subject: Using freeze
Message-ID: <000015f9@bossar.com.pl>

From: "Max M. Stalnaker" <stalnaker at acm.org>

Not an expert, but I have done it using vc++5.

In the freeze directory, under tools, which directory you get when to
download and install the source, which you have to do, there is a html file
that is where you need to go next.

My project plan would be :

1) compile source (to prove things are set up and you know what to do)
2) execute freeze.py on your program, with a command line parameter to put
the result in c:\temp
3) goto c:\temp and type nmake
4) test the result

You might have problems with 6.0 since  5.0 is what is supported out of the
box.
The resulting executionable will be dependent on some external programs.

SunitJoshi wrote:

> Hello
>     I'm trying to make exe's from python scripts using VC6++. However I'm
> lost as to how to use freeze to accomplish this. If someone could include
> some details I would really appreciate it.
>
> thanks
> Sunit

--
Max M. Stalnaker  stalnaker at acm.org
Astar Computer Consulting:  Networking and Custom Accounting Software




From wtanksle at dolphin.openprojects.net  Sat Jun 26 13:47:39 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Sat, 26 Jun 1999 17:47:39 GMT
Subject: Why exceptions shouldn't be used for flow control [Re: YAS to the   "Reading line-by-line" Problem]
References: <Pine.SUN.3.95-heb-2.07.990622183356.3454A-100000@sunset.ma.huji.ac.il> <376FDF70.4756AD38@mindspring.com> <slrn7n0chl.gh6.wtanksle@dolphin.openprojects.net> <377173F7.36BCFF6B@compaq.com> <slrn7n31k4.idj.wtanksle@dolphin.openprojects.net> <3772AED9.D9E1110@compaq.com>
Message-ID: <slrn7na4hq.3vt.wtanksle@dolphin.openprojects.net>

On Fri, 25 Jun 1999 10:19:05 +1200, Greg Ewing wrote:
>William Tanksley wrote:

>> "Not exceptional enough".  What does that mean?  (I ain't never had too
>> much exception! ;-)

>Sorry, I was using "exceptional" in a non-technical sense there.
>I meant "not abnormal enough".

>I don't agree that reaching EOF is necessarily "abnormal".
>If I want to read all the data in a file, and I don't know
>in advance how much there is, I write a loop that keeps reading
>until EOF. In that case, reaching EOF is not abnormal at all -
>it's bound to happen eventually!

This is completely true, but a good functional design for an API has each
function doing as little as possible.  By that rule, we'd do well to
remove EOF handling from the contract of the read function, and put it
into an eof function.

I'm also aware that doing so goes against the C API, so that's one
overwhelming strike against doing that to the current library (quite aside
from breaking all software ;-).

>The question, then, is what a general-purpose library routine
>should do on EOF, given that the "right thing" depends on the
>circumstances. You would have it always throw an exception,
>and require the programmer to catch it if it is an "expected"
>exception.

Right.

>Personally, I prefer it the way it is. If I want an exception
>thrown, I can always write a wrapper which does so. Doing it
>the other way around - wrapping exception-throwing code into
>something which doesn't - is trickier to get right; as I've
>shown, it's hard to make sure that it catches only what you
>want to catch.

You're entirely correct there.  Exceptionn-throwing code depends upon
having a good test system.  Without complete tests, it doesn't happen.

OTOH, that particular exceptionn would get thrown only in a truly bizzare
situation, because all code would look somethinng like this:

while not f.eof():
  line = f.readline()
  # ...

Most programmers wouldn't expect to have to catch an exception any mmore
thann they'd have to test for an exceptionnal returnn value.

>> In addition, the complexity of the failure required to produce this is
>> pretty high.  How many files do you need to read to get the next line from
>> this file?

>It's perhaps not the best example of this kind of problem.
>A better one, which I actually have experienced, is from the
>days when it was common to catch KeyError as a way of telling
>when a key wasn't in a dictionary:

>   try:
>     v = d[the_key()]
>   except KeyError:
>     v = some_default_value

Yes, I agree.  And I suspect the problem was that my explanation of my
solution was bad.

>The subtle problem with that piece of code is that if there
>is a bug which causes the_key() to raise a KeyError (which I
>think you'll agree is *not* an unlikely event) it gets
>incorrectly caught instead of triggering a traceback.
>To guard against that, you have to be very careful what
>you put inside the try:

>   i = the_key()
>   try:
>     v = d[i]
>   except KeyError:
>     v = some_default_value

>This is one of the reasons that the get() method was
>added to dictionaries. Using it, you can write

>   v = d.get(the_key(), some_default_value)

>which is not only less error-prone but shorter and
>clearer as well.

This is a very interesting case of a "bad" API replacing what I would have
called a "good" one.  It's obvious that I would have been incorrect.  I
think the distinction here is that a dictionary isn't a stream, and
there's a reasonable expectation of atomic access.

>> If someone else's uncaught EOFError hits this function, something nasty is
>> almost always going to happen... if it fails, it does so loudly so that
>> the tester notices and reports the problem.

>But that's exactly what *doesn't* happen! The misdirected exception
>silently terminates a loop that it wasn't meant to terminate, and
>later on the program fails with some set of symptoms that give little
>or no clue as to what the original cause was. I would much rather get 
>a traceback pinpointing exactly what was thrown and where it was thrown 
>from.

You're right -- and the solution to this is the samme as the solution to
the corresponding problem with exceptional return values (-1 for EOF).
Build the API so that they're never something you'd want to catch.

>> C is famous for noiselessly casting the -1 to a char (look at
>> the old Bash security bug), and Python is even worse -- any return you can
>> possibly make will almost certainly fit in noiselessly with anything you
>> do with it.

>The design of read() is perhaps not the best here -- it might be
>better to return None, which would cause most things expecting a
>string to blow up rather more quickly.

That's also a good idea.  It's been a while since I've used read, so I
don't recall what it actually returns (I've used nothing but readline).

>> You ought to feel lucky.  Some respected programming gurus recommend
>> calling abort() on any odd results

>Raising an exception is Python's equivalent of calling abort(). But
>it only works as such if you can be sure that your exception isn't
>going to be caught by something that makes unwarranted assumptions
>about its cause. The more you use exceptions for "normal" things,
>such as EOF or KeyError or IndexError when they aren't really errors, 
>the more likely that is to happen. In my experience, anyway.

You're entirely right.  And so am I -- in a function designed to get n!=0
more characters, an EOF really is an error.  Unfortunately for me, that's
not the definition of 'read', and never will be.

And it doesn't really matter -- one writes wrappers for everything anyhow
(read is supposed to be primitive).

>Greg

-- 
-William "Billy" Tanksley
Utinam logica falsa tuam philosophiam totam suffodiant!
   :-: May faulty logic undermine your entire philosophy!



From cwebster at math.tamu.edu  Sun Jun 27 20:41:44 1999
From: cwebster at math.tamu.edu (Corran Webster)
Date: 28 Jun 1999 00:41:44 GMT
Subject: ANN: Stackless Python 0.2
References: <37628EAA.C682F16C@appliedbiometrics.com>
Message-ID: <7l6gc8$du4$1@news.tamu.edu>

In article <37628EAA.C682F16C at appliedbiometrics.com>,
Christian Tismer  <tismer at appliedbiometrics.com> wrote:
>ANNOUNCING:
>
>               Stackless Python 0.2
>           A Python Implementation Which
>             Does Not Ese The C Stack
>
>What is it?
>A plugin-replacement for core Python.
>It should run any program which runs under Python 1.5.2 .
>But it does not need space on the C stack.
[snip]

I managed to get this to build eventually (thanks to the hints from Michael
Hudson).  Looking at it, I wonder whether there's the potential here for
more than coroutines and to write an implementation of threading within
Python.  Each thread would presumably need its own Python stack, and a
queueing and locking system would need to be added somehow, but because
the Python and C stacks are no longer intertangled, switching between
threads should be easy (as opposed to impossible <wink>).

This wouldn't be quite as flexible as the current implimentations of
threads at the C level - C extensions would be called in a single block
with no way to swap threads until they return.  On the other hand, this
would allow every platform to have some sort of threading available,
which would be a boon.

Unfortunately I'm not familiar enough with threads to go out there and
implement it right away, but I thought I'd at least raise it as a
possibility and see what people think and what the pros and cons are.

Corran




From jody at ldgo.columbia.edu  Sun Jun 27 19:10:40 1999
From: jody at ldgo.columbia.edu (Jody Winston)
Date: 27 Jun 1999 19:10:40 -0400
Subject: Python for large projects?
References: <3774b5d1.690922@news.force9.net> <tx3ogi1ahos.fsf@hog.ldgo.columbia.edu> <87d7yhv8ys.fsf@lisboa.i-did-not-set--mail-host-address--so-shoot-me>
Message-ID: <tx3aetl9r4f.fsf@hog.ldgo.columbia.edu>

>>>>> "Janko" == Janko Hauser <jhauser at ifm.uni-kiel.de> writes:

    Janko> Jody Winston <jody at ldgo.columbia.edu> writes:
    >> I've used Python for three large distributed applications (each
    >> > 100k lines of Python).  The only other recommendation that I
    >> would add is to define interfaces to the code.  If this is done
    >> with a product like ILU or FNORB, you can then easily call the
    >> Python code from other languages.  In addition, you can replace
    >> critical sections with other languages if needed.
    >>
    Janko> I have thought about this as the perfect way to do ``data
    Janko> hiding'', publish only special interfaces for your
    Janko> data. Has this a significant overhead, if used on one
    Janko> machiene, or is it overkill just for this purpose :-)?

For my data, which are large arrays read from disk, the ILU program is
50% slower on a simple benchmark that just transfers the arrays from
one address space to another on the same machine.  The xml-rpc
program, which transfers the same data from a client to a server, is 2
times slower as the ILU program[1].

However both of these bechmarks, for my case, do not really mean
anything since the real code is compute bound.

To understand the impact of a distributed object system will have on
your design, you'll need to understand how you will access your data
and the possible bottlenecks.  

Footnotes: 
[1]  That's really amazing in my book since all of the xml-rpc data
is ASCII.

-- 
Jody Winston
Lamont-Doherty Earth Observatory
RT 9W, Palisades, NY 10964
jody at ldeo.columbia.edu, 914 365 8526, Fax 914 359 1631 

Under US Code Title 47, Sec.227(b)(1)(C), Sec.227(a)(2)(B) This email
address may not be added to any commercial mail list with out my
permission.  Violation of my privacy with advertising or SPAM will
result in a suit for a MINIMUM of $500 damages/incident, $1500 for
repeats.





From mwh21 at cam.ac.uk  Wed Jun  2 17:34:46 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 02 Jun 1999 22:34:46 +0100
Subject: Bizarre error while ./configure-ing PIL
References: <m3g14dxduq.fsf@ime.net>
Message-ID: <m3n1yii8w9.fsf@atrus.jesus.cam.ac.uk>

Matt Gushee <mgushee at ime.net> writes:
> Hello, folks--
> 
> This may be the weirdest thing I've yet seen happen on a Linux
> box. I'm running RedHat Linux 5.2 and trying to compile PIL 1.0b1. I
> have a feeling this isn't really a PIL- or Python-specific problem,
> but just in case anyone's run into this:
> 
> [matt at crawdad Imaging-1.0b1]$ ls -dl libImaging
> drwxr-xr-x   2 matt     matt         2048 Jun  1 03:26 libImaging
> [matt at crawdad Imaging-1.0b1]$ cd libImaging/
> [matt at crawdad libImaging]$ ls -l configure
> -rwxr-xr-x   1 matt     matt        65845 Jan 10 04:29 configure
> [matt at crawdad libImaging]$ ./configure
> bash: ./configure: Permission denied
> [matt at crawdad libImaging]$ su
> Password: 
> [root at crawdad libImaging]# ./configure
> bash: ./configure: Permission denied
> 
> What the --- ?

Random guess, but check the first line of ./configure. It should
probably read

#!/bin/sh

My guess is either 1) /bin/sh is not executable (very unlikely, all
sorts of things would barf at that) 2) the line's gotten mangled
somehows - maybe a misplaced space?

HTH
Michael
 
> Extremely puzzledly yours,
> Matt Gushee
> Portland, Maine, USA



From Michael.P..Reilly at p98.f112.n480.z2.fidonet.org  Wed Jun 30 14:49:18 1999
From: Michael.P..Reilly at p98.f112.n480.z2.fidonet.org (Michael P. Reilly)
Date: Wed, 30 Jun 1999 19:49:18 +0100
Subject: socket timeouts
Message-ID: <000015cb@bossar.com.pl>

From: "Michael P. Reilly" <arcege at shore.net>

Nathan Clegg <nathan at islanddata.com> wrote:
: Is there any way to set the timeout for a TCP/IP connection? 
: Specifically, I'm writing some apps meant to run on the same machine, so
: if they can't connect *immediately* (i.e. the server app is not running), I
: need to go to plan B.  Any answers or alternate solutions?  I'm also
: interested in a similar issue across the internet, but I'm more generous
: about timeouts in this circumstance.  Thanks.

Since you are using a TCP/IP connection (ie. a socket), using select()
should be cross-platform enough.

  import select
  class TimeoutError(Exception):
    pass
  def wait_for_socket(sock, timeo):
    timeo = float(timeo)
    rdrs, wtrs, excs = select.select([sock], [], [], timeo)
    if not rdrs:
      raise TimeoutError, "no responce"

If the exception is raised, then the connection did not complete.

  -Arcege



From jeffrey at Digicool.com  Fri Jun  4 15:14:52 1999
From: jeffrey at Digicool.com (Jeffrey P Shell)
Date: Fri, 04 Jun 1999 15:14:52 -0400
Subject: python active web pages
References: <jonathan-0406991640390001@borg.onegoodidea.net>
Message-ID: <7j98g1$atj$1@autumn.news.rcn.net>

> i also looked at ZTemplates (a component of Zope),
> 
>    <http://www.zope.org/Download/ZTemplates/>,
>
> and while it doesn't actually embed python, it has a limited embedding
> language of its own, DTML, which gets munged into python. i must say that
> DTML looks a bit hideous (if anyone from digicool is reading, sorry guys).

Well, DTML (especially outside of Zope) is not meant to be
scripting-inside-of-HTML, which most Active Server Page type things are.
Personally, I think that is beyond hideous, the

<HTML>
<%
    document.write('<HEAD>')
    document.write('<TITLE>%s</TITLE>' % someTitleString)
    ...
%>
</HTML>

is terrible (i have seen plenty of PHP source that looks like the above).
DTML Scripting can look quite ugly, but it's (a) safe (which is good
because DTML in Zope and even on its own can be edited through the web and
you have security issues with that), and (b) very powerful depending on the
namespace given it (which is very powerful in Zope due to the inherent
Acquisition model).

Using ZPublisher (the beast previously known as Bobo) and DocumentTemplates
is very powerful for web development.  The DocumentTemplates are meant to
be _views_ on an Object, not just a web page unto themselves.  Thus, DTML
has a lot of very powerful batch processing features (it's basically a
reporting language in its web form) and very powerful namespace features.

This seperation (which was most effective before DTML scripting, aka the
embedding of python expressions in DTML, was implemented into DTML) is very
powerful as it enforces the seperation of Code and Presentation, a rather
MVC (Model-View-Controller) style paradigm.

My first Bobo app was turning a socket server into a web application.
Since I had made my classes and the views on those classes rather seperate
in the socket server implementation, turning into a web app took only a
matter of hours.  I was able to make a very simple DTML page for displaying
results from a search form and focus on implementing some new methods
needed for this new way of accessing data.  Once those methods were
implemented, I was able to spend the rest of the day in web-designer mode
since I already knew all the data that I had available to me and I could
just write DTML and not even have to think about Python.

That was very very _VERY_ powerful to me.

I wouldn't have had such an easy (nor as fun or as flexible) time in an ASP
type environment where all the code is embedded on a page.  Granted, you
can write ASP type pages that import modules and use the beautifully simple
<%= foo %> and have a similar abstraction, but it's not built into the
system.

DTML (esp. DTML Scripting) has gotten rather beyond control in some cases,
but I believe it's the most "right".  But yes, I work for digital creations
so I'm a bit biased.

But all that bias is based on the first time I wrote a major application
and just how _EASY_ (and powerful) that was.

Viva MVC !

> anyone have any experience with active web pages using either of these or
> something else?

yes.  :)

.jPS | jeffrey at digicool.com
zope, through the web. www.zope.org



From jam at newimage.com  Sat Jun 19 09:00:39 1999
From: jam at newimage.com (jam)
Date: Sat, 19 Jun 1999 09:00:39 -0400
Subject: [announce] 'getdatemodule' released. date expression extension for python
Message-ID: <19990619090039.L19541@toast.internal>

greetings,

I've extracted a parser from the 'cvs' sources that knows how to deal with
common date expressions like 'now', 'tomorrow', 'next week', 'next
thursday', etc.

I wrote a simple wrapper around it's one useful function, and packaged it
all up as a tgz file, and uploaded it to starship.

the code works great, but I'd really love it if someone could help come up
with a spec file so it can be released as a binary RPM package, and also
perhaps a perl version.

I realize that SWIG could make quick work of something like this, but I saw
a great deal of benefit in writing a 'pure' interface, in order to get more
practice writing extension modules.

I uploaded the file to <ftp://starship.python.net/pub/crew/jam/> recently.
the latest available snapshot should always be there.

I also run a small *low bandwidth* bbs that I would like to use as a support
site for this module and others as they are released. if anyone is
interested, it can be reached via the url below (quark.newimage.com).

please let me know if there are any questions, concerns, problems, patches
or whathaveyou related to getdatemodule, the bbs, or python and linux in
general.

regards,
Jeff
-- 
|| visit gfd <http://quark.newimage.com/> 
|| psa member #293 <http://www.python.org/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 210 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/19990619/b32483f7/attachment.sig>

From donn at u.washington.edu  Wed Jun  2 20:16:51 1999
From: donn at u.washington.edu (Donn Cave)
Date: 3 Jun 1999 00:16:51 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>  <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>  <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net>  <87hfosquc0.fsf@ev.netlab.co.jp> <3753F121.E4A9BE8B@prescod.net> <87yai3ntdt.fsf@ev.netlab.co.jp> <7j3iif$134e$1@nntp6.u.washington.edu> <7j45iv$908$4@cronkite.cc.uga.edu>
Message-ID: <7j4hhj$118u$1@nntp6.u.washington.edu>

graham at sloth.math.uga.edu (Graham Matthews) writes:
| Donn Cave (donn at u.washington.edu) wrote:
|: You're in good company here, I think the leading lights of the Python
|: world all prefer explicit termination in these cases.  Ref counting sure
|: offers a great convenience here, though.  Files close and flush their
|: buffers, dialogues disappear, all in accordance with their programmatic
|: lifetime.  I personally will miss this finalization more than any resource
|: management if reference counting is taken away.
|
| You seem to believe that garbage collection implies no reference counts.
| Why do you believe that? (a lot of GC schemes use reference counts).
| You also seem to believe that with GC finalisation will disappear. Why
| do you believe that (a lot of GC schemes have finalisation).

Does JPython use reference counts, and invoke delete methods as soon as
reference count drops to 0?

| ... Finally I
| believe you are mis-using finalisation if you use it to close up files,
| sockets, etc. That's what close calls are for.

Well, I think there's something to be said for it.  With Python, we're
free from the obnoxious requirement imposed by the storage model of
languages like C, that forces the programmer to account for each scrap
of malloc'd memory lest there be leaks.  That allows me to export an
object reference (e.g., via function return) and forget about it, and
that has radically beneficial implications for how things can be structured.

Why should I enjoy this freedom with memory resources, but welcome exactly
the same constraints when an object has other termination requirements?

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



From shr at chat.net  Sun Jun  6 15:26:51 1999
From: shr at chat.net (Scott RoLanD)
Date: Sun, 6 Jun 1999 12:26:51 -0700
Subject: Library load err with --prefix build 1.5.2
Message-ID: <19990606122651.A9771@chat.net>

I just did a virgin Mandrake 6.0 install (which comes with Python
1.5.1) and I want to compile and install 1.5.2 in:
  /home/shr/python
I want to leave the original 1.5.1 RPM right where it is and have my
new 1.5.2 install function in place without any problems.

However, it seems that if the 1.5.1 version is at the head of my PATH,
then I have problems using the 1.5.2 version.

I built as follows:
  ./configure --prefix=/home/shr/python
  make
  make test (test_long failed; test_types crashed)
  make install

So now I have a binary at:
  /home/shr/python/bin/python

But the problem comes when I try to use this simple script 'example':
####################
%: unset PYTHONPATH
%: which python
/usr/bin/python

%: python
Python 1.5.1 (#1, Apr 13 1999, 03:12:59)  [GCC pgcc-2.91.60 19981201 (e on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> 

%: cat example
#!/home/shr/python/bin/python
import sys
print sys.path

%: ./example
Built-in exception class not found: EnvironmentError.  Library mismatch?
Warning!  Falling back to string-based exceptions
['.', '/usr/lib/python1.5/', '/usr/lib/python1.5/plat-linux2', '/usr/lib/python1.5/lib-tk', '/usr/lib/python1.5/lib-dynload']

%: /home/shr/python/bin/python example
['', '/home/shr/python/lib/python1.5/', '/home/shr/python/lib/python1.5/plat-linux2', '/home/shr/python/lib/python1.5/lib-tk', '/home/shr/python/lib/python1.5/lib-dynload']

%: PATH=/home/shr/python/bin:$PATH
%: which python
/home/shr/python/bin/python

%: python
Python 1.5.2 (#1, Jun  6 1999, 15:06:23)  [GCC pgcc-2.91.66 19990314 (egcs-1.1.2  on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> 

%: ./example
['.', '/home/shr/python/lib/python1.5/', '/home/shr/python/lib/python1.5/plat-linux2', '/home/shr/python/lib/python1.5/lib-tk', '/home/shr/python/lib/python1.5/lib-dynload']

%: /usr/bin/python example
['', '/usr/lib/python1.5/', '/usr/lib/python1.5/test', '/usr/lib/python1.5/plat-linux-i386', '/usr/lib/python1.5/lib-tk', '/usr/lib/python1.5/lib-dynload']
####################

I don't want to have to set PYTHONPATH in order to get this to
work. Anyone with a vanilla environment should be able to use this
script.

This should work; what is going on?
Thanks In Advance.
-- 
Scott RoLanD
http://www.chat.net/~shr/



From fredrik at pythonware.com  Wed Jun 30 04:37:10 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 30 Jun 1999 08:37:10 GMT
Subject: printing with print
References: <377925EF.1CBB640F@starvision.com>
Message-ID: <011a01bec2d3$c215a900$f29b12c2@secret.pythonware.com>

Angus MacKay wrote:
> is there any other way to print that with print?
> perhaps with a real function.
> 
> I want to print two things without a space being added between them:
> foo = "there"
> print "hi" + foo
> 
> works fine but: 
> foo = 2
> print "hi" + foo
> 
> does not. I like the printing of any object ability of "print" so I
> do not want to use "" % style printing, I just want to be able to
> govern the whitespace that gets output.

note that the "s" formatting operator can handle any non-tuple
object, just like print (both use "str" to convert non-string objects
to strings).

the following two statements give the same result, as long as
foo is not a tuple.

    print foo
    print "%s" % foo

to deal with tuples too, you can call str yourself, or wrap the
object in a singleton tuple:

    print "%s" % str(foo)
    print "%s" % (foo,)

</F>




From news at dorb.com  Fri Jun 25 15:37:55 1999
From: news at dorb.com (Darrell)
Date: Fri, 25 Jun 1999 15:37:55 -0400
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net>
Message-ID: <JWQc3.2620$U6.10996@newsr2.twcny.rr.com>

I know less than nothing about this subject, but have a though to
contribute.

What about using a tool like this:
http://www.ndim.edrc.cmu.edu/dougc/jpi/Home.html
To make use of Java's GUI tools.

Yes speed may be an issue, but it would work on the Mac and book stores are
loaded with Java....
Then Jpython and Cpython would share the same GUI tools.

--
--Darrell





From matz at netlab.co.jp  Wed Jun  2 00:14:33 1999
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 02 Jun 1999 13:14:33 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
 <87pv3iq6ss.fsf@ev.netlab.co.jp> <7iv86a$iud@news.dns.microsoft.com>
Message-ID: <87vhd7nsra.fsf@ev.netlab.co.jp>

"Vadim Chugunov" <chega_ at yahoo.com> writes:

|> I don't get it.  There's no relation between destructors and real
|> garbage collection.  For example, Java uses destructors called
|> finalizers, even though most Java VM use real GC.
|
|Right, but java finalizers will not be called till the moment the memory is
|collected.
|Depending on the GC configuration and the program's memory consumption this
|may be quite a while.

|> I hate ref counting.  It is memory-leak prone, does not reclaim cyclic
|> data without explicit cycle-cut.  I know, by ref counting, object is
|> recycled as soon as it looses last reference to it, but relying on
|> that behavior is error prone, I think.
|
|Can you explain exactly why is it more error prone?
|If something can be done automatically, without writing explicit
|deallocation code,
|why can't we let it be that way?

Well, the reasons why ref counting is more error prone are:

  * ref counting never reclaim cyclic references, resulting memory
    leak.

  * extension module writers are easy to forget to put INCREF(),
    DECREF() properly, resulting memory problems (leak or crash).

These two never be caused by real GC.  From my viewpoint (of cource
biased), avoiding these is more important than predictability.

|OK, I'll try to propose something constructive. Maybe we need an object
|protocol, that would enumerate
|all references held by an object ?  Writing a portable GC would be then much
|easier.

Some languages use both.  Ref counting for usual basis, real GC for
full clean up.  How about adding real GC to Python 2.0, co-working
with ref counting?
						matz.



From marcoxa at copernico.parades.rm.cnr.it  Wed Jun 16 18:28:18 1999
From: marcoxa at copernico.parades.rm.cnr.it (Marco Antoniotti)
Date: 17 Jun 1999 00:28:18 +0200
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <z%Q93.4399$5E4.36952@news2.randori.com>
Message-ID: <lw3dzrn5kt.fsf@copernico.parades.rm.cnr.it>

"Tom Cloney" <muse at viadoc.com> writes:

	...
> 
> It's super when embedded in a text editor. Alpha is a great example.
> 

Emacs is a great example :)

Cheers

Marco

-- 
Marco Antoniotti ===========================================
PARADES, Via San Pantaleo 66, I-00186 Rome, ITALY
tel. +39 - 06 68 10 03 17, fax. +39 - 06 68 80 79 26
http://www.parades.rm.cnr.it/~marcoxa



From hank at prosysplus.com  Mon Jun 28 08:43:31 1999
From: hank at prosysplus.com (Hank Fay)
Date: Mon, 28 Jun 1999 08:43:31 -0400
Subject: ActiveX, IDE
Message-ID: <7cKd3.3449$lr3.133866@news2.jacksonville.net>

I'd like to make Wintel apps using ActiveX controls under Python: is there
an IDE for Python that will support this?

TIA,

Hank Fay

--
http://www.prosysplus.com





From gmcm at hypernet.com  Thu Jun  3 18:12:56 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 3 Jun 1999 17:12:56 -0500
Subject: Where to find samples???
In-Reply-To: <7j6n4g$1jqq$1@rtpnews.raleigh.ibm.com>
Message-ID: <1283674270-42685846@hypernet.com>

Jr. King asks:

> Where can I find some good c/c++ integrated with python, besides the
> 2 or 3 that comes with the download of Python for windows???

The first place to look is at anything in the Modules and Objects 
directories in the source distribution. Or poke around on the contrib 
section of the ftp site. Or download CXX from LLNL. Or use SWIG and 
look at what it generates.

- Gordon



From Darrell at p98.f112.n480.z2.fidonet.org  Wed Jun 30 05:19:24 1999
From: Darrell at p98.f112.n480.z2.fidonet.org (Darrell)
Date: Wed, 30 Jun 1999 10:19:24 +0100
Subject: Reflection in Python?
Message-ID: <000015c1@bossar.com.pl>

From: "Darrell" <news at dorb.com>

    # Get the testAdd method or attribute of MathTest
test=getattr(MathTest,'testAdd')

>>> class A:
...     def myX(self):
...             print 'xxx'
...
>>> a=A()
>>> af=getattr(a,'myX')
>>> apply(af)
xxx
>>> af()
xxx
        # Get the name of 'a'
>>> print a
<__main__.A instance at 7f79d0>
>>> print `a`
<__main__.A instance at 7f79d0>
>>> print str(a)
<__main__.A instance at 7f79d0>
>>> print af
<method A.myX of A instance at 7f79d0>
>>>
--
--Darrell
<anders777 at my-deja.com> wrote in message news:7ldglr$ita$1 at nnrp1.deja.com...
> test = MathTest("testAdd")
>
> where test.run() will run the "testAdd" method on
> the MathTest object?  I also want to be able to
> tell the user the name of the object (MathTest)
> and the method in case the method fails.  Any
> thoughts?
>
> Thanks,
> Anders Schneiderman
> National Journal Daily Briefings Group
>
>
>
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.




From graham at sloth.math.uga.edu  Wed Jun  2 20:47:42 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 3 Jun 1999 00:47:42 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>  <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>  <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net>  <87hfosquc0.fsf@ev.netlab.co.jp> <3753F121.E4A9BE8B@prescod.net> <87yai3ntdt.fsf@ev.netlab.co.jp> <7j3iif$134e$1@nntp6.u.washington.edu> <7j45iv$908$4@cronkite.cc.uga.edu> <7j4hhj$118u$1@nntp6.u.washington.edu>
Message-ID: <7j4jbe$dhh$1@cronkite.cc.uga.edu>

graham at sloth.math.uga.edu (Graham Matthews) writes:
| You seem to believe that garbage collection implies no reference counts.
| Why do you believe that? (a lot of GC schemes use reference counts).
| You also seem to believe that with GC finalisation will disappear. Why
| do you believe that (a lot of GC schemes have finalisation).
Donn Cave (donn at u.washington.edu) wrote:
: Does JPython use reference counts, and invoke delete methods as soon as
: reference count drops to 0?

I don't think reference counting garbage collection is mandated in Java 
(it used to be but I think that has changed). 

But I am not sure I see what the connection is between how JPython does 
memory management and garbage collection in Python in general. 

graham at sloth.math.uga.edu (Graham Matthews) writes:
: | ... Finally I
: | believe you are mis-using finalisation if you use it to close up files,
: | sockets, etc. That's what close calls are for.
Donn Cave (donn at u.washington.edu) wrote:
: Well, I think there's something to be said for it.  With Python, we're
: free from the obnoxious requirement imposed by the storage model of
: languages like C, that forces the programmer to account for each scrap
: of malloc'd memory lest there be leaks.  
: ... 
: Why should I enjoy this freedom with memory resources, but welcome exactly
: the same constraints when an object has other termination requirements?

Because other kinds of objects are different to memory? Also I am not
sure why you think GC can't handle other kinds of objects as well. Maybe
Java doesn't do a good job of resource finalisation, but that doesn't
mean it can't be done.

graham

-- 
       Bang, bang, Maxwell's silver hammer came down on her head
    Bang, bang, Maxwell's silver hammer made sure that she was dead



From otto at olympus.net  Tue Jun  8 23:11:33 1999
From: otto at olympus.net (Otto Smith)
Date: 9 Jun 99 03:11:33 GMT
Subject: Indentation and grouping proposal
References: <000b01beb14d$b502a480$aaa02299@tim>
Message-ID: <01beb225$07ef2600$efed85c6@otto>

# Re: Indentation and grouping proposal
#       You don't need to change python to get
#       the functionality you request..

# nips.py
# Non Intuitive Parsing System...
'''
   Here are some routines that allow you to
   write in a style similar to that proposed
   in the: "Indentation and grouping proposal".

   The routines will take ugly unreadable code
   and translate it to simple, neatly indented
   python code.  The constants below set the
   block grouping symbols to '_{' and '}_'
   and the statement separator to ';', but you
   are welcome to change these to anything
   you want.

   Usage is simple:
      parsenips() takes block delimited code
      and converts it into a list of lists format.

      makeright() takes the output of parsnips
      and makes it into python indented code.

Example:

      #####
      import from nips import makeright, parsenips
      ugly = """ if 1<2: _{ print 'Ho ho'; print '1 < 2' }_
                 else: _{ print "surprise";
				 print '2 < 1!!!' }_
             """
      print makeright(parsenips( ugly ))
      #####

Which produces:

if 1<2:
    print 'Ho ho'
    print '1 < 2'
else:
    print 'surprise'
    print '2 < 1!!!'

'''

import string

SPARSEX = "String parse exception"

MYSTART = '_{'
MYEND   = '}_'
MYLINE  = ';'
BADQUOTES  = "'"+'"'

len_mystart = len(MYSTART)
len_myline  = len(MYLINE)

def match(beg, end, str):
    lbeg = len(beg)
    lend = len(end)
    dex = lbeg
    count = 1
    while dex < len(str):
        if str[dex:dex+lend] == end:
            count = count-1
            if count == 0:
                return (str[lbeg:dex],str[dex+lbeg:])
            dex = dex+lend
        elif str[dex:dex+lbeg] == beg:
            count = count+1
            dex = dex+lbeg
        else:
            dex = dex + 1
            continue
        if count < 0:
            raise SPARSEX, 'badly matched quote brackets'
    raise SPARSEX, 'badly matched quote brackets'


def parsenips( str ):
    str = string.strip(str)
    curinst = ''
    statementlist = []
    dex = 0
    while str:
        if str[dex] in BADQUOTES:
            quote = str[dex]
            val, str = match( quote, quote, str[dex:] )
            curinst = curinst+quote+val+quote
            dex = 0
        elif str[dex:dex+len_mystart] == MYSTART:
            if curinst:
                statementlist.append( string.strip(curinst) )
            val, str = match( MYSTART, MYEND, str[dex:] )
            statementlist.append( parsenips(val))
            curinst = ''
            dex = 0
        elif str[dex:dex+len_myline] == MYLINE:
            statementlist.append( string.strip(curinst) )
            str = string.strip(str[dex+len_myline:])
            curinst = ''
            dex = 0
        else:
            curinst = curinst + str[dex]
            dex = dex+1
    if string.strip(curinst):
        statementlist.append( string.strip(curinst))
    return filter(None, statementlist)


SPACER = '    '
def makeright( list, level=0 ):
    pystrings = []
    for val in list:
        if type(val) == type(""):
            pystrings.append( (SPACER*level)+val )
        else:  ## Assert type(val) == ListType
            pystrings.append( makeright(val,level+1) )
    return string.join( pystrings, '\n')

#Otto
#otto at olympus.net




From phd at emerald.netskate.ru  Mon Jun 28 05:19:15 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Mon, 28 Jun 1999 13:19:15 +0400 (MSD)
Subject: ConfigParser & STRPTIME ????
In-Reply-To: <rNHvjEdhm5Pp-pn2-OUBSngdMEQZP@d185d18f8.rochester.rr.com>
Message-ID: <Pine.LNX.4.04.9906281315320.30533-100000@emerald.netskate.ru>

Hi!

On Mon, 28 Jun 1999, Benjamin Schollnick wrote:
> 	2) Can you use CONFIGPARSER to write a *.INI file that
> 		you have made, or read into config parser, back
> 		out to disk? From reading Dejanews, I believe that
> 		it can't be?!?!?!?!?!? If so, why?  It seems silly to support
> 		read but not write.

   Few years ago i wrote a class library full of config files parsers.
Among other things there is a parser for .INI files. Parsed config can be
written back to disk.

http://members.xoom.com/phd2/Software/Python/
http://www.fortunecity.com/skyscraper/unix/797/Software/Python/

   look for "Flat ASCII Database and config files modules".

   The .INI-part of the library was not extenceivly tested - usually I use
other classes for storing information in text files...

> 			- Benjamin

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From tim_one at email.msn.com  Sat Jun  5 23:25:33 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 5 Jun 1999 23:25:33 -0400
Subject: Idle comments
In-Reply-To: <7jcdcm$19qi$2@hub.org>
Message-ID: <000501beafcc$3c407900$639e2299@tim>

[Ragica <zzz at NO-MAIL.com>, seeking a way to change something about IDLE's
 color decisions]
> I know it's in the code. I have even looked for where (believe it or
> not). In the IDLE readme for example it tells you where and what to
> change to adjust the syntax highlight colours--but not the default
> background colour--, I looked around there. Alas, to no avail. I
> freely admit i find that code confusing. I am not quite the python
> programming genius as yoursly truly.

It's really a question about Tk, whose mysteries are as deep to me as to
anyone.  The thing to do is to grep (or "findstr" on NT) the IDLE source for
"background".  You won't get many hits, and one of them may be the thing you
need to change.  Most likely in EditorWindow's __init__ function (last place
you'd look, right <wink>?).  Look for where it creates a Text widget "... =
Text(...)".  If there are
    background=whatever
    foreground=whatever
keyword arguments already, change them; else add them.  I don't have the
1.5.2 IDLE source anymore, so can't be more specific.

> ...
> I have to tell you, even if the guidos of the world are ever infinitely
> more programatically exalted than insignificant I...
> i simply can't fathom people who hard code stuff into their programs!
> Even at early stages. In my books setting-type variables at very least
> should go somewhere highly visible and up-front... so as to be easy
> to access and find. But that's just me, perhaps.

It's only hard to find the first time <wink>.  In the meantime, a
comprehensive subsystem for managing user preferences is a major piece of
work, and Guido appears to have devoted his limited IDLE time to adding
major new functionality instead.  That's the downside of Open Source, of
course:  unlike as with proprietary software, you don't get to pay for the
privilege of feeling ignored <wink>.

patience-works-better-than-impatience-ly y'rs  - tim





From dalke at bioreason.com  Sat Jun 12 03:13:00 1999
From: dalke at bioreason.com (Andrew Dalke)
Date: Sat, 12 Jun 1999 01:13:00 -0600
Subject: module import, but different
Message-ID: <3762087C.98EA86DA@bioreason.com>

Hello,

  I would like to add a test suite to our software so at the top
level of the distribution I can do a "make check" and run the tests.
(We are planning to do something based to Python's regrtest code.)
I'm running into a problem because I don't know how to deal with
packages; and we do almost everything as packages.  It comes down
to needing a way to import a package when you can't include the
package's parent directory in the PYTHONPATH.

  I'll describe the problem with some detail first.

  Here's an example package layout:

MCS/
  Makefile
  __init__.py
  Graph.py
  ...
  doc/

to which I plan to add
  test/
    br_regrtest.py     ("br_" for "Bioreason's version of")
    test_graph.py
    ...

And in the Makefile add something like

check:
  cd test && $(PYTHON) br_regrtest.py

Inside the test directory will be programs like "test_graph.py".
They will have code like "import MCS.Graph".  This will fail
because MCS isn't a module or package on the PYTHONPATH.  (Setting
PYTHONPATH to ".." will fail for modules which have non-trivial
__init__.py files; and we have a couple of that form.)

One solution would be to include "../.." but that wouldn't be
appropriate because that may include files which get imported
by accident.  For example, I keep all of my development code
under src/, so src/MCS, src/daylight, etc.  But when I'm doing
unit testing I want to import the globally installed "daylight"
and not the ../../daylight imported by adding ../.. to the path.

Of course, I could arbitrarily decide to enforce that ../..
not contain importable code, but that require me to change my
development style, and for everyone using one of these
packages to understand the problem and obey the solution.  Ain't
gonna happen.

Besides, we distribute the packages with version numbers, as in
MCS-1.0.  Including ../.. means we would have to convert the name
somehow.


The Python distribution solves the problem, I think, because in
getpath.c it adds `Lib' to the PYTHONPATH if python was was run
in the build tree.  Seems sneaky to me!

Another possibility is to make a symbolic link in the test/
directory, as in
  ln -s ../ MCS

This will work under unix, but isn't very portable, and I expect
we'll have to support Windows someday.  (Or do the other OSes have
something equivalent to this?)

A third solution is to move the code tree to its own subdirectory
-- what automake calls a "deep" package -- as in:

MCS/
  Makefile
  MCS/
    __init__.py
    Graph.py
    ...
  doc/
  test/

In that way I can do
  cd test && PYTHONPATH=.. $(PYTHON) br_regrtest.py

This is probably the best solution, except that
  1) we already have about 20 modules with code in the top level,
  2) the patches I made for automake assume the code will be of
this form
  3) I will be contributing the patches back to automake, and I
can guarantee that others won't be writing packages of this form.
  4) other reasons I can't currently come up with, other than
"it doesn't feel right/is too complicated" (eg, because there will
always have to be two Makefiles, one in the top directory and one
for the source subdir)

The last solution, and the one I want, is to have some way to
fake the import system so that "../" is treated as a module, say,
by calling
  special_import("MCS", "../")
which fakes out sys.modules.

Here is a unix specific implementation:
 = = = =
import os, sys
# One way to remove the links with Python exits...
class CleanUp:
    def __init__(self, name):
       self.name = name
       self.unlink = os.unlink  # os gets deleted too early!
    def __del__(self):
       self.unlink(self.name)

clean_up = []
def special_import(name, location):
  os.symlink(location, name)
  sys.modules[name] = __import__(name)
  clean_up.append(CleanUp(name))
 = = = =

Suggestions?

						Andrew Dalke
						dalke at bioreason.com



From clarence at silcom.com  Fri Jun 18 11:01:10 1999
From: clarence at silcom.com (Clarence Gardner)
Date: Fri, 18 Jun 1999 15:01:10 GMT
Subject: Compiled main program
References: <KyG93.386$dR.22261@newsfeed.avtel.net> <376981D2.61C7D46D@compaq.com>
Message-ID: <Wata3.485$dR.32335@newsfeed.avtel.net>

Greg Ewing (greg.ewing at compaq.com) wrote:
: Clarence Gardner wrote:
: > 
: > x = MakeMeMain()
: > del x
: 
: Wouldn't it just be simpler to do
: 
: MakeMeMain()
: 
: Greg

Sure would -- I noticed that after I posted and felt a blush of
embarassment.  That came up because I first wrote

	del MakeMeMain()

(don't ask me why), which is illegal (because you del a name, not a value)
so with a quick jerk of the hand I put in 'x'.

By the way, just to compound my initial stupidity, as was pointed out to
me in email by <sorry, don't have your name>, I didn't need the MakeMeMain()
class at all -- this suffices:

	sys.modules['__main__'] = sys.modules[__name__]

I must have had a mental block when I was trying to think of how to get
a reference to the current module, which made me wander off into class-land.

-- 
-=-=-=-=-=-=-=-=
Clarence Gardner
AvTel Communications
Software Products and Services Division
clarence at avtel.com



From Michael.Hudson at p98.f112.n480.z2.fidonet.org  Wed Jun 30 15:32:47 1999
From: Michael.Hudson at p98.f112.n480.z2.fidonet.org (Michael Hudson)
Date: Wed, 30 Jun 1999 20:32:47 +0100
Subject: Reflection in Python?
Message-ID: <000015c3@bossar.com.pl>

From: Michael Hudson <mwh21 at cam.ac.uk>

anders777 at my-deja.com writes:

> I've just started programming in Python, and I
> could really use some help figuring out something.
>  I'd like to implement Kent Beck's unit tester in
> Python (someone posted a very nice alternative a
> few weeks ago, but Junit's approach works better
> for my project's needs).  How do I do the
> equivalent of what Java calls reflection? I want
> to do the following:
> 
> test = MathTest("testAdd")
> 
> where test.run() will run the "testAdd" method on
> the MathTest object?  I also want to be able to
> tell the user the name of the object (MathTest)
> and the method in case the method fails.  Any
> thoughts?
> 
> Thanks,
> Anders Schneiderman
> National Journal Daily Briefings Group
> 
Not entirely sure I follow, but will this do what you want?

class Test:
    def __init__(self,object,meth):
        self.object = object
        self.meth = meth
    def run():
        try:
            getattr(self.object,self.meth)()
        except:
            print self.object.__class__.__name__, "failed!"

then

test = Test(MathTest,"testAdd")
test.run()

does what you ask (I think).

Anyway, I suspect the builtin "getattr" is what you're after.

HTH
Michael



From morse at harborcom.net  Fri Jun 25 14:05:13 1999
From: morse at harborcom.net (Kevin Dahlhausen)
Date: Fri, 25 Jun 1999 18:05:13 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net>
Message-ID: <3773c3a0.102825445@news.oh.verio.com>

m
>I'd like some discussion on whether or not it is a good idea to
>abandon TkInter, before it gets too widely used, and instead adopt
>wxPython as the primary GUI framework for Python.
>
>Having said this, I have to say that I'm in no way a TkInter or a
>wxPython expert.  I've evaluated both of them and found wxPython to be
>the winner - and TkInter without a real chance of getting used by me.
>
>Also, can anyone comment on the features available in TkInter and
>wxPython - perhaps even make a feature comparison chart?
>
>[1] By sluggish I mean: You can see, or at least sense, the window
>update operations.  Windows are not drawn instantly, but in visually
>detectable steps.

I agree with your observations on TK.  I looked at wxPython and some other
GUIs.  I liked the Fast Light Toolkit API the best.  At least Python allows us
to use the best GUI, with 'best' being defined by people according to their own
preferences.

>P.S. Let's not make a war out of this; I just want to know if there's
>anybody else who feels wxPython is performing better than TkInter.

As long as we don't talk about politics,tabs, spaces, and indentation, we should
be fine.



From alex at somewhere.round.here  Tue Jun  8 10:06:17 1999
From: alex at somewhere.round.here (Alex)
Date: 08 Jun 1999 10:06:17 -0400
Subject: Two minor problems with running python in emacs
Message-ID: <etdg1423hye.fsf@m2-225-11.mit.edu>

Hi.  When I run a python script in a py-shell under emacs, sometimes
C-cC-c doesn't return me to the shell for a VERY long time.  Does anyone
know why, and is there a workaround?  In case it's relevant, I think I'm
working under RedHat 5.2.

And now a question from the "there's no harm in asking"-department: if a
script running under a py-shell runs pdb.set_trace (), is there some way
to get emacs to do gud things like point to the debugger's current
execution line in some other buffer?

Alex.



From Greg.Ewing at p98.f112.n480.z2.fidonet.org  Wed Jun 30 16:46:15 1999
From: Greg.Ewing at p98.f112.n480.z2.fidonet.org (Greg Ewing)
Date: Wed, 30 Jun 1999 21:46:15 +0100
Subject: file.tell problems (1.5.2/Win32)
Message-ID: <00001596@bossar.com.pl>

From: Greg Ewing <greg.ewing at compaq.com>

Joshua Rosen wrote:
> 
> This leads me to believe that `append mode' is intended to `append' at the
> *current* point in the file, which should, for consistency's sake, start at
0.

I just tried an experiment (on Windows) using "a+"
mode. It seems that f.tell() is telling the truth,
but the truth is strange.

What "a"/"a+" modes seem to mean is "open the
file positioned at 0, but if the first operation
is a write, go to the end of file before doing
it". Which seems quite bizarre to me - wouldn't
it be simpler to just go to the end of the
file on opening?

Greg



From fleck at informatik.uni-bonn.de  Sat Jun 12 05:06:20 1999
From: fleck at informatik.uni-bonn.de (Markus Fleck)
Date: 12 Jun 1999 09:06:20 GMT
Subject: urllib: an alternative URLOpener
References: <Pine.SUN.3.95-heb-2.07.990611155013.10151A-100000@sunset.ma.huji.ac.il>
Message-ID: <7jt7uc$sh4@news.rhrz.uni-bonn.de>

Moshe Zadka <moshez at math.huji.ac.il> wrote:
> Whom do I contact about adding the following code to urllib:
> -------------->8-------------->8-------------------------
> class QuietFancyURLOpener(FancyURLOpener):
>         def prompt_user_password(self, host, realm):
>                 return None, None
> -------------->8-------------->8-------------------------
> (Yes, I realize I can put it in my own program. But it seems a generally
> useful thing for non-interactive downloaders)

While we're at it... :-) - It might also be nice to have a FancyURLOpener
that raises an exception on an authentication mismatch. And one that takes
a user-defined user/password table dictionary or a callback function.

Yours,
Markus.

-- 
"That isn't the future I have in mind for the information highway: I want to
access more diverse connections and deeper thoughts, not cooler commercials 
and new tolls." --  Geoff Dutton, http://photo.net/philg/humor/media-lab.html



From MHammond at skippinet.com.au  Thu Jun 10 18:26:03 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Fri, 11 Jun 1999 08:26:03 +1000
Subject: Pythonwin - advise sought
References: <375FA468.98976DDF@connection.com>
Message-ID: <7jpdv5$aq0$1@m2.c2.telstra-mm.net.au>

Colin J. Williams wrote in message <375FA468.98976DDF at connection.com>...
>I am looking at Pythonwin as an alternative to Tk and would appreciate
advice on
>the following three questions.
>
>The Pythonwin documentation points to MFC documents as a source of
information.
>The MSDN site has:
>
>http://msdn.microsoft.com/library/devprods/vs6/vc++/vcmfc/_mfc_about_the_mi
crosoft_foundation_classes.htm'''
>
>Q1:  Are there better sources which can be recommended?

Not buy me.  Ive been doing MFC for years now and never owned a book on it
:-(

>The code fragment below:
>
>       code1= "win32ui.CreateFileDialog(1)"
>       fd= eval(code)
>       print(repr(fd))
Or:
  print repr(win32ui.CreateFileDialog(1)) :-)

>     object 'PyCFileDialog' - assoc is 007CC240, vf=False,
>     notify=0,ch/u=0/0, mh=0,kh=0

These are really for debugging:
assoc is: The address of the MFC object wrapped (ie, the CFileDialog)
vf=false: Is there a .py class associated with this C+ object?
notify=0: How many HookNotify() calls have been made on the object
ch/i=0/0: Can't remember - use the source :-)
mh=0: How many HookMessage calls.
kh=0: How many keyboard handlers.

>Q3.  Is there some way that one can test that fd is an instance of some
class,
>any class?

FD is not a class, it is a type.  There are no good examples for a file
dialog, but consider:

>>> ps=win32ui.CreatePropertySheet("Test")
>>> ps
object 'PyCPropertySheet' - assoc is 01102980, vf=False, notify=0,ch/u=0/0,
mh=0, kh=0
>>>

Same as your example.  Now lets create a property sheet using a Python
class:

>>> ps=pywin.mfc.dialog.PropertySheet("Test")
>>> ps
<pywin.mfc.dialog.PropertySheet instance at 1103080>

Now it is a Python class.  We can get the win32ui type object by looking at
_obj_ - ie:
>>> ps._obj_
object 'PyCPropertySheet' - assoc is 01104F10, vf=True, notify=0,ch/u=0/0,
mh=1, kh=0
>>>

Note in this example we have "mh=1" indicating that the
pywin.dialog.PropertySheet class hooked a single windows message.

Hope this helps...

Mark.





From marcoxa at copernico.parades.rm.cnr.it  Tue Jun 29 11:39:41 1999
From: marcoxa at copernico.parades.rm.cnr.it (Marco Antoniotti)
Date: 29 Jun 1999 17:39:41 +0200
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <876748mcxg.fsf@home.ivm.de> <37788BB9.ABD@mailserver.hursley.ibm.com> <lw67471fyi.fsf@copernico.parades.rm.cnr.it> <7laoq0$bvh$1@Starbase.NeoSoft.COM>
Message-ID: <lwso7bdnia.fsf@copernico.parades.rm.cnr.it>

> Delivery-Date: Tue, 29 Jun 1999 17:28:08 +0200
> Date: Tue, 29 Jun 1999 10:30:14 -0500 (CDT)
> From: Cameron Laird <claird at Starbase.NeoSoft.COM>
> Newsgroups: comp.lang.python,comp.lang.lisp,comp.lang.scheme
> Organization: NeoSoft, Inc. +1 713 968 5800
> content-length: 645
> 
> In article <lw67471fyi.fsf at copernico.parades.rm.cnr.it>,
> Marco Antoniotti  <marcoxa at copernico.parades.rm.cnr.it> wrote:
> 			.
> 			.
> 			.
> >spin. :) Finally, AFAIU, a Tcl "macro" must run as an interpreter of
> >the spec. A Common Lisp (Scheme) macro is compiled into regular code
> >by read-time expansion.
> 			.
> 			.
> 			.
> Are you making a semantic point, or arguing
> on the basis of implementation-specific per-
> formance?

It is (mostly) a semantic point.

> If the latter, please be aware
> that the Tcl community is actively investi-
> gating dramatic speedups of [eval].

I have no problems with that.  But the following example is pretty
hard to beat. :)

==============================================================================
* (defmacro zut (a) `(list ,a))
ZUT

* (defun zot (x) (zut x))
ZOT

* (compile 'zot)
ZOT

* (disassemble 'zot)
071B5790:       .ENTRY ZOT(x)                ; (FUNCTION (T) LIST)
     7A8:       ADD   -18, %CODE
     7AC:       ADD   %CFP, 32, %CSP

     7B0:       CMP   %NARGS, 4              ; %NARGS = #:G0
     7B4:       BNE   L0
     7B8:       NOP
     7BC:       MOVE  %A0, %A4               ; %A0 = #:G1
     7C0:       MOVE  %A4, %A0               ; No-arg-parsing entry point
     7C4:       ADD   4, %ALLOC
     7C8:       ANDN  %ALLOC, 7, %A3
     7CC:       OR    3, %A3
     7D0:       MOVE  %A3, %A1
     7D4:       ST    %A0, [%A1-3]
     7D8:       ST    %NULL, [%A1+1]
     7DC:       TADDCCTV 4, %ALLOC
     7E0:       MOVE  %A3, %A0
     7E4:       MOVE  %CFP, %CSP
     7E8:       MOVE  %OCFP, %CFP
     7EC:       J     %LRA+5

     7F0:       MOVE  %LRA, %CODE
     7F4:       UNIMP 0
     7F8: L0:   UNIMP 10                     ; Error trap
     7FC:       BYTE  #x04
     7FD:       BYTE  #x19                   ; INVALID-ARGUMENT-COUNT-ERROR
     7FE:       BYTE  #xFE, #xED, #x01       ; NARGS
     801:       .ALIGN 4
* 
==============================================================================

Of course I could have defined a much more intricated macro.

Cheers

-- 
Marco Antoniotti ===========================================
PARADES, Via San Pantaleo 66, I-00186 Rome, ITALY
tel. +39 - 06 68 10 03 17, fax. +39 - 06 68 80 79 26
http://www.parades.rm.cnr.it/~marcoxa



From bwarsaw at cnri.reston.va.us  Mon Jun  7 09:12:39 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Mon, 7 Jun 1999 09:12:39 -0400 (EDT)
Subject: File locking
References: <199905301615.MAA00996@mira.erols.com>
	<14163.61310.936743.849211@anthem.cnri.reston.va.us>
	<al8ogisbos9.fsf@kruuna.Helsinki.FI>
Message-ID: <14171.50503.221763.561735@anthem.cnri.reston.va.us>

>>>>> "MS" == Markus Stenberg <mstenber at cc.Helsinki.FI> writes:

    >> "Barry A. Warsaw" <bwarsaw at cnri.reston.va.us> writes:
    >> You might want to look at Mailman's flock.py module.  The doc
    >> string says:

    MS> At least on our HP-UX devel box (HP-UX 11.0) it failed
    MS> miserably, didn't have the time to examine why, though.

Darn.  If you get the chance, I'd really like to know more about the
failure.  I don't have an HP-UX box to debug this on and it would help 
make Mailman more robust.

-Barry



From ivnowa at hvision.nl  Sat Jun 12 07:15:12 1999
From: ivnowa at hvision.nl (Hans Nowak)
Date: Sat, 12 Jun 1999 11:15:12 GMT
Subject: Newbie question: DOS console commands?
Message-ID: <199906121113.NAA01576@axil.hvision.nl>

> I'm new to Python this week and was wondering if there are any console
> commands to use on a DOS screen (e.g. ClearScreen, GotoXY, Foreground and
> Background colors, etc.)
> 
> I've got PythonWin running on Windows 95 and am having fun learning the
> language and Tkinter, but I have a need to write a simple DOS interface
> program.  I could pull out my copy of TurboPascal, but I'd like to write it
> in Python to further my education.

I happen to maintain a Python version for DOS, which has *some* 
support for screen manipulation. Due to the little interest in a DOS 
Python, I never really looked into this, I just kludged together a 
few modules and functions. <rant> Now if this stupid little #&#*(@( 
computer vendor had provided me with a well-working box, I would have 
been able to look at it right now and maybe write a few extra 
functions... but since they didn't, I'm unable to update Python-DX 
(and my snippets page) at this moment. I will in the (hopefully near) 
future though. </rant>

Meanwhile, you could check out Python-DX at 
http://www.cuci.nl/~hnowak/html/python-dx.html . Check out modules 
like _conio and _pc. (There are wrapper modules for them, but they're 
part of a separate library, which I have different plans with.)

If you're really wanting to use this extensively I will update my 
modules and docco as soon as I can.

Veel liefs,

--Hans Nowak (ivnowa at hvision.nl)
Homepage: http://fly.to/zephyrfalcon
--Hans Nowak (ivnowa at hvision.nl)
Homepage: http://fly.to/zephyrfalcon



From dick.wall at usa.net  Mon Jun 14 09:12:07 1999
From: dick.wall at usa.net (root)
Date: 14 Jun 1999 09:12:07 -0400
Subject: High performance Python on NT
References: <000301beb3b7$4f973810$6eba0ac8@kuarajy.infosys.com.ar> <37609429.65A7E802@palladion.com> <37611A73.348C6C5C@erols.com>
Message-ID: <zp23exjs.fsf@odin.telinet.com>

R Wentworth <rhww at erols.com> writes:

> Tres Seaver wrote:
> > Run, don't walk, to http://www.zope.org.  You'll think you've died and gone to
> > heaven.

Agreed, I just used Zope for a project here to replace some java servlet code 
which was under-performing. After convincing my bosses to let me demo it,
I got the go ahead. A month later I am running a web based administration
system that is nice and fast, and pretty bullet proof to boot. I actually
used the ZPublisher part of the system and pretty much ignored the templates
and IDE just because I like the hands on approach, and also I wanted to use
the existing Java Servlet templates, which used a different mark-up to the
DTML stuff in Zope.

It was a dream to develop in.

> 
> I looked at Zope a little while ago, and it did seem rather exciting.
> Depends a little bit on your constraints, though, whether it's the
> right solution.  I had a project that needed to be hosted via a commercial
> hosting provider, and nobody was willing to run Zope unless I rented
> a dedicated server.  (Actually, not really nobody, but with web hosting
> providers it can be devastatingly bad to become dependent on a single
> hosting service unless you've got a lot of prior positive experience
> with them.)

Zope can run in a number of ways though. We use it as a back end to an
existing web-server (in fact website pro since we are using NT as the
OS, the work I did was developed in Linux though because I prefer the
taste of it :-). You do need to run it has a process (UNIX) or a 
service (NT) but apart from this and a suitable re-direct on an existing
web server, it has fairly minimal impact.

> 
> I chose instead to go with a solution using the Java "servlet" interface
> (instead of CGI), and writing JPython (http://www.jpython.org/) servlets 
> instead of Java servlets.  Don't know if this could be considered a
> "high-performance" solution; it avoids the process creation and
> DB connection overhead of CGI solutions, but JPython is inherently
> slower than Python.  So, depends what sort of performance matters.
> I expect JPython to be fine as long as I'm not too compute-intensive;
> we'll see.  In this model, if I run into performance bottlenecks
> I can improve things by selectively rewriting code into Java.

Well, the Java servlet stuff did not perform as we wanted it too which was
why we started this in the first place. In fairness it was not the servlet
framework itself that was the problem, but rather some of the support
libraries which would not perform quickly or reliably enough. 

> 
> (I did have a few other concerns about Zope: I wanted to do some
> non-hierarchical design, and found that the limited documentation
> made attempting such a thing difficult.  Still, I hope to have
> an excuse to learn/use Zope someday in the future.)

Agreed, it took me about a week of playing to really "get it" but after
I did, it was smooth sailing from there. I would recommend looking into
using ZPublisher directly, as it helps with understanding the rest of
the system better (at least it did for me).

> 
> Robert Wentworth


Cheers

Dick



From jeremy at cnri.reston.va.us  Mon Jun 14 11:42:14 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Mon, 14 Jun 1999 11:42:14 -0400 (EDT)
Subject: urllib: an alternative URLOpener
In-Reply-To: <14178.38855.323759.698523@bitdiddle.cnri.reston.va.us>
References: <Pine.SUN.3.95-heb-2.07.990611155013.10151A-100000@sunset.ma.huji.ac.il>
	<7jt7uc$sh4@news.rhrz.uni-bonn.de>
	<14178.38855.323759.698523@bitdiddle.cnri.reston.va.us>
Message-ID: <14181.8918.956963.762261@bitdiddle.cnri.reston.va.us>

Ok.  There is now an alpha version of urllib2 available from my Web
page: http://the-tech.mit.edu/~jeremy/python/

I wrote this module because the current urllib is too hard to
customize.  It was difficult to use only a subset of the basic
functionality in FancyURLopener and the error reporting was
problematic.  I hope to have addressed both these issues with urllib2.

This code has not received extensive testing.  I expect that there are
a number of lurking bugs.  There is too much diversity on the Web for
my own testing to have covered all the possible cases.  I'd appreciate
bug reports, comments on the design, requests for new features, etc.
Most of all, I need suggestions for a name; urllib2 is a clunker.

I expect to make one or two more alpha releases, as I finalize the
feature set.  I've got no timetable in mind, although I have a
vacation in August, which would be an appropriate time to make a final 
release.

Jeremy



From radbit at my-deja.com  Sun Jun 20 20:10:01 1999
From: radbit at my-deja.com (radbit at my-deja.com)
Date: Mon, 21 Jun 1999 00:10:01 GMT
Subject: Newbie Q: How to call unmodified VB DLL from Python
References: <3769bcd1.54492213@news.earthlink.net>
Message-ID: <7kjvsl$rtu$1@nnrp1.deja.com>

In article <3769bcd1.54492213 at news.earthlink.net>,
  lull at acm.org (John Lull) wrote:
> As a first Python project, I'd like to use Python (or JPython) to
> automate some hardware testing under Win95/98/NT.  The hardware in
> question is run by a DLL written in VB, and to which I do not have
> source.
>
> "Extending & Embedding the Python interpreter", ch. 3, has
> instructions for building a DLL which will work, but assumes you're
> working in VC++ and have the source.
>
> If anyone could point me toward instructions on how to make this work,
> I'd be grateful.
>
> Thanks.
>
> Regards,
> John
>
>

Hi

since VB has only been able to do in-process OLE (Com) Dlls since VB4.0
I take it that this DLL exposes its methods via COM. In which case it
should be a snip to talk to it via Marc Hammond's Pythonwin COM
support. Find out what it's programmatic id is (progid) from
Pythonwin's COM browser or MS VC++'s COM Viewer. Once you know what
progid it has you can do:

>>> from win32com.client import Dispatch
>>> vbObj=Dispatch('progid')

or you can also look at the methods it exposes via Pythonwin's Com
browser.

If the Dll is not written in VB then you'll probably need Sam Rushing's
CallDll package which lets you invoke the LoadLibray,GetProcAddress, etc
functions.

Hope this helps

--
Florent Heyworth


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From news at dorb.com  Tue Jun 22 10:58:57 1999
From: news at dorb.com (Darrell)
Date: Tue, 22 Jun 1999 10:58:57 -0400
Subject: choosing random numbers with weights/probability?
References: <929987267.6227@www.remarq.com> <lbwb3.3$V4.11640@news-dal.corridex.com> <ERwb3.1452$U6.5885@newsr2.twcny.rr.com> <7knamm$kas$1@news.udel.edu>
Message-ID: <izNb3.1875$U6.6729@newsr2.twcny.rr.com>

Orig:
 Time: 10.28 sec
{'one': 25167, 'two': 25144, 'three': 49689}

Terry's:
 Time: 9.94 sec
{'one': 24983, 'three': 50020, 'two': 24997}

--
--Darrell





From jrush at summit-research.com  Fri Jun 18 03:04:55 1999
From: jrush at summit-research.com (Jeff Rush)
Date: 18 Jun 1999 07:04:55 GMT
Subject: Parsing HTML
References: <376984E2.5055E13D@lockstar.com>
Message-ID: <VrDQRddcbbs9-pn2-E3yjC9kP1xYF@Fortress>

On Thu, 17 Jun 1999 13:27:33, Mordy Ovits <movits at lockstar.com> wrote:

> What is the best way to parse HTML into a Python data structure, allow me change it,
> and output it as HTML?
> -- 
> o Mordy Ovits
> o Cryptographic Engineer
> o LockStar Inc.

One good way is to use the DOM framework of XML.  You can get the module
and docs at www.python.org in the XML-SIG section.  A tiny example follows:

----- cut here -----

#!/usr/bin/env python

from xml.dom.html_builder import HtmlBuilder
from xml.dom.writer       import HtmlWriter

# Read in Original HTML Source
# and Build a DOM Tree Structure
# ----------------------------

htmlstr = open('test.html', 'r').read()
b = HtmlBuilder(ignore_mismatched_end_tags=1)
b.feed(htmlstr)   # Stuff the HTML Source into the HTML Parser
doc = b.document  # Get the Newly Constructed Document Object

# Perform Modifications as Needed
# -------------------------------

text = doc.createTextNode("Additional Title Text")
titlenode = doc.getElementsByTagName('TITLE')[0]
titlenode.appendChild(text)

# Write DOM Tree Back as HTML Source
# --------------------------------

fd = open('output.html', 'w')
w = HtmlWriter(stream=fd)
w.write(doc)

----- cut here -----

There is also an XML/DOM based HTML Pretty Printer posted
in here a week ago, which I find useful to clean up the output
HTML.  If you can't find it, drop me a line.

-Jeff Rush




From pduffin at mailserver.hursley.ibm.com  Fri Jun 11 08:00:56 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Fri, 11 Jun 1999 13:00:56 +0100
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <mtU73.2731$nn.808830@news.shore.net> <7jp5lp$jo1$1@Starbase.NeoSoft.COM>
Message-ID: <3760FA78.63AA@mailserver.hursley.ibm.com>

Cameron Laird wrote:
> 
> In article <mtU73.2731$nn.808830 at news.shore.net>,
> Michael P. Reilly <arcege at shore.net> wrote:
>                         .
>                         .
>                         .
> >I believe that current releases of Tcl can handle null bytes, but I
> >have not verified that.
> Yes.  Tcl8.0 and after handle null bytes.
>                         .
>                         .
>                         .

Although 8.0 did/does have some holes in its null byte support.

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From andrew at andrewcooke.free-online.co.uk  Wed Jun 23 04:22:15 1999
From: andrew at andrewcooke.free-online.co.uk (Andrew Cooke)
Date: Wed, 23 Jun 1999 08:22:15 GMT
Subject: Thread for Tkinter (NT)
Message-ID: <7kq5fh$1g9$1@nnrp1.deja.com>


Hi,

Does the thread running Tk.mainloop() have to be the main (initial)
thread on W32?  This is what I seem to be experiencing, and given
W32's thread based messaging for windows it wouldn't surprise me,
but could someone please confirm?

And if so, is there any way around this - the GUI can't start first,
but is started via call to a subroutine.  If the main thread is
then locked in mainloop is there any way of starting a new thread
with the main thread's stack at that point and returning back to
the rest of the code (OK, I guess not :-)

Thanks,
Andrew


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From jeremy at cnri.reston.va.us  Thu Jun  3 14:31:19 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Thu,  3 Jun 1999 14:31:19 -0400 (EDT)
Subject: threads
In-Reply-To: <3756C078.5D998D3B@icrf.icnet.uk>
References: <3756C078.5D998D3B@icrf.icnet.uk>
Message-ID: <14166.51558.850416.153999@bitdiddle.cnri.reston.va.us>

>>>>> "AM" == Arne Mueller <a.mueller at icrf.icnet.uk> writes:

  AM> Hi All, I'm using the threading module in python 1.5.2b2. I'm
  AM> wondering wheather the different threads are all executed in
  AM> parallel on a multiprocessor machine.

The short answer is no.  The Python interpreter has a lock that
prevents allows only a single thread to execute at a time.  

This is an FAQ that isn't in the FAQ yet.  Here's a more detailed answer:

http://x37.deja.com/getdoc.xp?AN=415182307&CONTEXT=928434593.1326317628&hitnum=0

Jeremy



From bwarsaw at cnri.reston.va.us  Sun Jun  6 19:39:39 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Sun, 6 Jun 1999 19:39:39 -0400 (EDT)
Subject: Using CVS under Win95
References: <yoy63.1157$VL2.171084@news.direcpc.com>
Message-ID: <14171.1723.95797.26115@anthem.cnri.reston.va.us>

>>>>> "EvS" == Emile van Sebille <emile at fenx.com> writes:

    EvS> I'm trying to access the python source per (session paste
    EvS> follows):

    D> cvs -d :pserver:pythoncvs at cvs.python.org:/projects/cvsroot
    D> login
    | (Logging in to pythoncvs at cvs.python.org)
    | CVS.EXE [login aborted]: could not find out home directory

When you log in, CVS tries to write a file called .cvspass in your
home directory, containing authentication information.  Your CVS
client on Win95 apparently can't find your home directory to write
this file.  I have no idea how to set this up on Win95.

-Barry



From delgado at Mathematik.Uni-Bielefeld.DE  Tue Jun 29 13:01:35 1999
From: delgado at Mathematik.Uni-Bielefeld.DE (Olaf Delgado)
Date: Tue, 29 Jun 1999 19:01:35 +0200
Subject: Binary pickle portability
Message-ID: <Pine.OSF.4.05.9906291852100.1376-100000@pinax.Mathematik.Uni-Bielefeld.DE>

Hi folks!

Lately I observed strange things when trying to unpickle a binary pickle
(including floating point numbers) produced on a different architecture.
The Python versions involved were 1.5.1 and 1.5.2, the systems Irix 5
on Mips and Linux on Intel x86.

Explanations I can think of:

a) An incompatibility of binary pickles between 1.5.1 and 1.5.2
b) Pickle stores floats in native format.
c) A problem in my installations.

Any hints? Should binary pickles be avoided when sharing files between
different systems/installations?

Thanks!
Olaf





From thstr at serop.abb.se  Mon Jun  7 03:45:48 1999
From: thstr at serop.abb.se (Thomas S. Strinnhed)
Date: Mon, 07 Jun 1999 09:45:48 +0200
Subject: Programming Python - examples from the book work, no?
Message-ID: <375B78AC.5DB962E4@serop.abb.se>

Hi

Doing everything in MSVC++ on NT I try to run the Example 14-2 (p534-540
) from "Programming Python" by Mark Lutz. I run it just like it comes on
the CD included with the book (CD-DRIVE:/Examples/DOS/Extend/Stacktyp.c)
trying to extend Python with this Stack-type. 
BUT: I get an I-don't-know-what-this-is-about kind of compiler error, 
complaing about this call: PyObject_HEAD_INIT(&PyType_Type)  
with message "initializer is not a constant".

PyObject_HEAD_INIT is in object.h as
#define PyObject_HEAD_INIT(type) 1, type,

and

PyType_Type is (also in object.h) 
extern DL_IMPORT(PyTypeObject) PyType_Type; /* The type of type objects
*/

What's missing in this picture? Is it a version clash??
My sys.version
'1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)]'
(goes for everything but stacktyp.c, which I think is v1.3)

Is it some declaration/definition missing that I need to add?
Has anyone else had problems with this example code?
...and solved it?
maybe even with some MS-developer-environment-point-of-view??

If it is an obvious mistake from my side, please point it out to me,
otherwise just please point it out to me.

Many Thanks
 -- Thomas S. Strinnhed, thstr at serop.abb.se

(Below is the complete initialization of the Stacktype from stacktyp.c,
near the eind of the file.)
...
static PyTypeObject Stacktype = {      /* main python type-descriptor */
  /* type header */                    /* shared by all instances */
      PyObject_HEAD_INIT(&PyType_Type)         
      0,                               /* ob_size */
      "stack",                         /* tp_name */
      sizeof(stackobject),             /* tp_basicsize */
      0,                               /* tp_itemsize */

  /* standard methods */
      (destructor)  stack_dealloc,   /* tp_dealloc  ref-count==0  */
      (printfunc)   stack_print,     /* tp_print    "print x"     */
      (getattrfunc) stack_getattr,   /* tp_getattr  "x.attr"      */
      (setattrfunc) 0,               /* tp_setattr  "x.attr=v"    */
      (cmpfunc)     stack_compare,   /* tp_compare  "x > y"       */
      (reprfunc)    0,               /* tp_repr     `x`, print x  */

  /* type categories */
      0,                             /* tp_as_number  
+,-,*,/,%,&,>>,pow...*/
      &stack_as_sequence,            /* tp_as_sequence +,[i],[i:j],len,
...*/
      0,                             /* tp_as_mapping  [key], len, ...*/

  /* more methods */
      (hashfunc)   0,                /* tp_hash    "dict[x]" */
      (binaryfunc) 0,                /* tp_call    "x()"     */
      (reprfunc)   0,                /* tp_str     "str(x)"  */

};  /* plus others: see Include/object.h */
...



From bmatt at acm.org  Thu Jun 10 23:05:50 1999
From: bmatt at acm.org (Bruno Mattarollo)
Date: Fri, 11 Jun 1999 03:05:50 GMT
Subject: High performance Python on NT
Message-ID: <000301beb3b7$4f973810$6eba0ac8@kuarajy.infosys.com.ar>

Hi!

	I am about to start a project where I will need to do some high performance
Web CGI scripts on WinNT ... I have no choice ... I know it's not the best
combination. It's not my decision ;-) I am looking for some advice. Is it
better if I use some kind of Active Python Scripting Page (ASP but with
Python)? Or just doing CGI with threads it will be fine... ?

	I know this is a very vague question... I just want to know if someone had
to take a similar decision... The app I will be doing should support several
hundred users connecting to the Web server and doing some serious CGI
activity ... Almost all hits that the server will receive will be scripts.
The hardware is no problem in this case... When can scale up to Quad PII
Xeon with 1GB RAM ... U2W SCSI discs...

	I know exactly what to do in a UNIX env... but I am not sure of the
decision to take regarding win32...

	TIA

/B

Bruno Mattarollo <bmatt at acm.org>
... proud to be a PSA member <http://www.python.org/psa>
<http://starship.python.net/crew/bmatt/>




From gmcm at hypernet.com  Thu Jun 17 10:21:51 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 17 Jun 1999 09:21:51 -0500
Subject: wierdness with pythonwin dialogs, threads & refcounts
In-Reply-To: <3768b75e.5469224@NNRP.UK.INSNET.NET>
Message-ID: <1282492928-113747642@hypernet.com>

Dave Kirby writes:
 
> I have found strange behaviour (strange to me at least) when
> creating a pythonwin dialog on a separate thread created using the
> standard python thread module.  The dialog object is never deleted
> since the thread is keeping references to it. 
> 
> The following example uses dlgtest.py that comes with the pythonwin
> distribution:
> 
> #------------- 8< ----------------------------
> C:\dev\PYTHON\Pythonwin\pywin\DEMOS>python
> Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on
> win32 Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> from dlgtest import TestDialog
> >>> import sys, thread
> >>> def test():
> ...   t = TestDialog()
> ...   print "refs:", sys.getrefcount(t)
> ...   t = None
> ...
> >>> test()
> Edit text changed!
> refs: 2
> >>> thread.start_new_thread( test, () )
> >>> refs: 6
> #------------- 8< ----------------------------
> 
> As you can see from the above, creating the dialog in the thread
> gives it an extra 4 references compared with the non-threaded
> invocation. Is this a bug, or is the thread module incompatible with
> pythonwin? If so how would I do the equivalent using pythonwin
> threads?  I have read the available docs, but it is still pretty
> opaque to me.

First, there's no incompatibility with the thread module in 
Pythonwin. 

Second, you should use threading.py, not the low level thread module 
(which is very, very primitive). Or you can use 
win32process.beginthreadex which is a wrapper around MFC threads. 
This lets you manipulate priorities. (As of a couple years ago, there 
was a small memory leak involved with using MFC threads, however).

As for the source of your questions - without doing things in a debug 
build of python / Pythonwin, it's almost impossible to tell what's 
going on, or even if there really is a problem.

Using sys.getrefcount increments the refcount. Printing something in 
the interactive interp binds the thing to a var named "_" (to 
optimize looking at it again). Printing to stdout in Pythonwin 
involves all kinds of hidden manipulations.

If you want to know whether you're leaking TestDialogs you could code 
a loop that creates threading.Thread's that show a TestDialog and 
then joins the Thread. Run it for as many iterations as you can 
stand, and watch the process in TaskManager.

hope-you're-not-prone-to-Carpal-Tunnel-Syndrome-ly y'rs



- Gordon



From avv at quasar.ipa.nw.ru  Thu Jun  3 00:11:53 1999
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Wed, 02 Jun 1999 21:11:53 -0700
Subject: ANN: pyslang-0.2.0
Message-ID: <37560089.83BD84A8@quasar.ipa.nw.ru>

Dear Colleagues,

This release provides more Pythonic way to invoke three SLang functions
which deal with the 'raw' representation of the portions of the screen
as a sequence of pairs (color, character):
    SLsmg_write_color_chars( [(color1, char1), (color2, char2), ...])

    list_length, [(color, char), ...] = SLsmg_read_raw(list_length))
    SLsmg_write_raw([(color, char), ...])

The 'char's in these synopses are represented also pythonically as
strings of the length 1.

An element of the list used as the parameter (but NOT the result) here
may also be just an unsigned short, composed as color << 8 | char. In
that case 'char' is really the character's integer representation
(obtained, e.g., via ord()). This is retained for better compatibility
with the underlying C library, but may be discontinued in future
releases. The result of the function SLsmg_read_raw is always a list of
tuples as depicted above.

When color value is OR'd with 0x80 it represents the alternative
character set, supposed to contain pseudographic characters.

Unfortunately, this part of SLang occur to work differently under Unix
and win32. In the first case, at least under Linux, one switches the
character sets or uses 0x80-d colors and gets pseudographics by
specifying chars as 'q', 'v', etc from a subset of the first 128 ASCII
characters. Under win32, however, both  switching the character sets and
0x80-ing the colors does nothing, and pseudographics is found, in usual
PC manner, in the higher half of the ASCII table.

Certainly one would easily map the win32 behaviour onto the Unix one
within the slang.i interface file. But it would break proper
transparency on the pyslang wrapper. The best would be if the
appropriate mapping was done within the SLang library itself, probably
with some additional 'option-set' function, which would retain standard
win32 behaviour for those who requires best backward compatibility with
the older PC/DOS stuff. Any ideas/comments/desires are welcome to be
reflected in future releases.

In that version the demo_*.py examples are best viewed under Unix. They
run however, under win32, with some minor ugliness.

The distribution is found at the Starship:

ftp://starship.python.net/pub/crew/avv/pyslang/pyslang-0.2.0.tgz

It contains sources for Unices and win32 together with two binaries: for
Linux (latest glibc) and win32.

Alexander





From hniksic at srce.hr  Thu Jun  3 16:00:57 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 03 Jun 1999 22:00:57 +0200
Subject: threads
References: <3756C078.5D998D3B@icrf.icnet.uk> <14166.51558.850416.153999@bitdiddle.cnri.reston.va.us>
Message-ID: <8790a1njeu.fsf@pc-hrvoje.srce.hr>

Jeremy Hylton <jeremy at cnri.reston.va.us> writes:

> >>>>> "AM" == Arne Mueller <a.mueller at icrf.icnet.uk> writes:
> 
>   AM> Hi All, I'm using the threading module in python 1.5.2b2. I'm
>   AM> wondering wheather the different threads are all executed in
>   AM> parallel on a multiprocessor machine.
> 
> The short answer is no.  The Python interpreter has a lock that
> prevents allows only a single thread to execute at a time.

Eek.  That's depressing.  Isn't supporting threads and not allowing
simultaneous execution a bit weird?



From jeremy at cnri.reston.va.us  Wed Jun  2 12:33:17 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Wed,  2 Jun 1999 12:33:17 -0400 (EDT)
Subject: Python 2.0 (perhaps off topic)
In-Reply-To: <3754BAFE.192D7648@prescod.net>
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7imr6b$1s0$1@nnrp1.deja.com>
	<87pv3iq6ss.fsf@ev.netlab.co.jp>
	<3752C76D.604CD474@prescod.net>
	<87n1ykr0z9.fsf@ev.netlab.co.jp>
	<375372F6.A874554@prescod.net>
	<14164.7179.10346.998488@bitdiddle.cnri.reston.va.us>
	<375431E6.6BC64E80@prescod.net>
	<14164.43542.417731.361360@bitdiddle.cnri.reston.va.us>
	<3754BAFE.192D7648@prescod.net>
Message-ID: <14165.22828.439378.233057@bitdiddle.cnri.reston.va.us>

>>>>> "PP" == Paul Prescod <paul at prescod.net> writes:

  PP> Jeremy Hylton wrote:
  >>  Perhaps this is what I get for jumping in in the middle of a
  >> conversation without learning how it got started.  The example I
  >> was responding to depended on file object being immediately
  >> finalized to avoid leaking file descriptors.

  PP> This is a documented feature of Python (documented in the FAQ at
  PP> least).  New users do not currently need to learn otherwise

That doesn't count!  Among other reasons, the FAQ predates JPython.
Guido and Jim spent some time going through the differences between
CPython and JPython, and deciding which of the differences where
legitimate and which were bugs or incompatibilities.  I believe that
memory management is one of those implementation-dependent issues that 
the language semantics do not specify.

  PP> unless they use JPython which is incompatible in this regard. In
  PP> other words, this IS a demonstration of a flaw in GC, or at
  PP> least in Java GC because it required a change in the Python
  PP> language semantics.

Don't think it's fair to call it a change-- just a clarification.  Nor 
would the problem necessarily lie with the garbage collector.  If the
language spec had said that finalizers were always called such that
you could depend on it to free resources like file descriptors, then
it would have been a bug in the spec <0.1 wink>.

  PP> I think that that is all anyone has been saying. Many people
  PP> like the current semantics and don't trust that a full GC would
  PP> deterministically invoke finalizers for arbitrary resources.

I'm sure it wouldn't, but as the FAQ entry you mentioned explains,
there are some exceptions with reference counting, too.  In general,
it's a rather risky to depend on the finalizers to close resources for 
you.  Even with reference counting, it's easy to have one more
reference than you think and not have some files get closed.

Jeremy



From schorsch at schorsch.com  Sun Jun 20 15:54:12 1999
From: schorsch at schorsch.com (Georg Mischler)
Date: Sun, 20 Jun 1999 19:54:12 GMT
Subject: do anonymous pipes normally work on NT?
References: <4D0A23B3F74DD111ACCD00805F31D8100DB90DFB@RED-MSG-50>
Message-ID: <7kjgt1$noj$1@nnrp1.deja.com>

Bill Tutt wrote:
> After digging through this stuff, I went back and looked at the code I
had
> written for win32popen.cpp. I realized that I had forgotten why I had
> written the DuplicateHandle() code, so I went back commented it out
from the
> KB article that I had originally used, and then modified runproc.py
to do
> the same thing.
>
> cat.py is now much happier with life with the new and improved
runproc.py.
[more stuff]

Uh oh, Bill...

If you should ever happen to visit the Munich Oktoberfest,
please remind me that I owe you a few beer!

Actually, I stayed with the inheriting version for the moment,
as it will work with existing versions of the win32 extensions.
And I'm not sure if I got the changes in the sources (or the
compilation) right. At least I didn't see any changes after
replacing my win32process.pyd.

The most important thing is of course that my code works now.
I'll have another look at the esthetically more pleasing
changes later, when the dust has settled a little in my brain...


> Notice I also changed the DETACHED_PROCESS to a zero. This is to not
do
> those wierd things that DETACHED_PROCESS seems to do. Since stdin,
etc.. are
> pipes, it doesn't matter if the process technically has access to
> runproc.py's NT console. (if any)

Another thing I noted here, is that with pythonw.exe (instead
python.exe), the second generation of processes gets those
problems again, no matter what you do (my application starts a
"process wrapper" in python, which will call other programs to
do the real work in the background).


> The other alternative from 0 would be to pass in CREATE_NEW_CONSOLE,
> and tweak win32process_win32module.cpp to add:
> 	{"wShowWindow",	     T_INT,  OFF(m_startupinfo.wShowWindow)},
> //@prop integer|wShowWindow|
>
> onto the end of PySTARTUPINFO::memberlist, and then do the following:
>         StartupInfo = win32process.STARTUPINFO()
>         StartupInfo.hStdInput  = hStdin_r
>         StartupInfo.hStdOutput = hStdout_w
>         StartupInfo.hStdError  = hStderr_w
>         StartupInfo.dwFlags = win32process.STARTF_USESTDHANDLES |
> 		win32process.STARTF_USESHOWWINDOW
> 	  StartupInfo.wShowWindow = win32con.SW_HIDE
>
> Mark: please add that line when you get a chance. :)

Yes, I noticed this attribute missing as well, but wasn't sure
what to insert where... Interestingly, STARTF_USESHOWWINDOW
keeps the console window from opening even without it. The
structure member seems to get initialized to zero somehow.


> '''runproc.py
>
[and more of my code expertively enhanced...]


Thanks a lot for all the tips and bug hunting!

-schorsch

--
Georg Mischler  --  simulation developper  --  schorsch at schorsch.com
+schorsch.com+  --  lighting design tools  --  http://www.schorsch.com/


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From arcege at shore.net  Mon Jun  7 11:52:45 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Mon, 07 Jun 1999 15:52:45 GMT
Subject: bug in os.path.samefile (Python 1.5.1)?
References: <1318D78C9072D11195C9006094EA98A72C20EF@ocesrv> <Pine.LNX.4.10.9906070536020.2300-100000@tbryanpc.arlut.utexas.edu> <22Ms+CA0H7W3EwBA@jessikat.demon.co.uk>
Message-ID: <hVR63.2131$nn.660380@news.shore.net>

Robin Becker <robin at jessikat.demon.co.uk> wrote:
: In article <Pine.LNX.4.10.9906070536020.2300-100000 at tbryanpc.arlut.utexa
: s.edu>, Thomas Bryan <tbryan at tbryanpc.arlut.utexas.edu> writes
:>
:>On Mon, 7 Jun 1999, Matej Cepl wrote:
:>
:>
: ...
:>
:>Aren't filenames case-sensitive on NT?  Look at your traceback.  Notice
: last time I used NT4.0 in November paths were case insensitive

I think case-insensitive is correct; and there is os.path.normcase()
for this I believe.

I've just tried Python 1.5.2 (don't have 1.5.1 installed anymore, sorry)
on WinNt 4.0 SP3 and get case-insensitivity:

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit <Intel>] on win32
Copyright 1991-19995 Stichting Mathematisch Centrum, Amsterdam
>>> import os, win32api
>>> os.getcwd()
'D:\\Program Files\\Python'
>>> os.listdir('d:\\python files\\python')
['DLLs', 'Doc', 'include', INSTALL.LOG', 'Lib', 'libs', 'py.ico', 'pyc.ico',
'pycon.ico', 'python.exe', 'pythonw.exe', 'Tools', 'UNWISE.EXE']
>>> os.stat('doc')
(16895, 0, 3, 1, 0, 0, 0, 957573914, 918516880, 918516864)
>>> win32api.GetVersionEx()
(4, 0, 1381, 2, 'Service Pack 3')
>>> os.path.normcase('D:\\Program files\\Python')
'd:\\program files\\python'
>>>

Not really answering the original posting I know, but... :)

  -Arcege




From tim_one at email.msn.com  Fri Jun  4 21:42:21 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 4 Jun 1999 21:42:21 -0400
Subject: threads
In-Reply-To: <87u2sn4p31.fsf@pc-hrvoje.srce.hr>
Message-ID: <000b01beaef4$a73d0580$f19e2299@tim>

[Hrvoje Niksic]
> ...
> Ironically, when I first saw "tuples" in Python, I assumed they were
> invented to speed up threading because you don't need to lock access
> to immutable structures.  (The same applies to immutable strings.)
> Boy, was I wrong.  :-(

Besides, immutable structures are still mutable "under the covers":  in
particular, their refcounts get boosted and dropped all the time.  That also
needs to be atomic.

OTOH, you weren't wrong!  Guido is well aware of the optimizations that
*may* someday apply thanks to immutability.  That not all coneivable such
have been implemented yet is merely evidence that time has extent <wink>.

the-poor-man's-so-depressed-about-whitespace-he-can-barely-get-
    thru-the-day-ly y'rs  - tim





From mnot at pobox.com  Sat Jun 12 22:59:09 1999
From: mnot at pobox.com (Mark Nottingham)
Date: Sun, 13 Jun 1999 02:59:09 GMT
Subject: urllib: an alternative URLOpener
References: <Pine.SUN.3.95-heb-2.07.990611155013.10151A-100000@sunset.ma.huji.ac.il><7jt7uc$sh4@news.rhrz.uni-bonn.de> <14178.38855.323759.698523@bitdiddle.cnri.reston.va.us>
Message-ID: <016301beb548$b5ef8a80$0301a8c0@cbd.net.au>

FWIW, I've put a fairly complete spidering class together as part of a
project... may be useful for somebody out there...


http://www.mnot.net/cacheability/download.html


----- Original Message -----
From: Jeremy Hylton <jeremy at cnri.reston.va.us>
Newsgroups: comp.lang.python
To: <python-list at cwi.nl>
Sent: Sunday, June 13, 1999 3:24
Subject: Re: urllib: an alternative URLOpener


> I've been working on-and-off on a replacement for urllib that is more
> easily customizable; they basic philosophy has been to avoid
> inheritance as the customization mechanism.  Instead, a single
> director class coordinates the activities of many small handlers that
> implement one feature/protocol.
>
> The resulting architecture makes it much easier to create a UrlOpener
> that handles redirects (which is probably the crucial feature of
> FancyUrlOpener) without getting all of the other cruft (like prompting
> the user to enter a password on stdin).
>
> I've been delayed a bit because I wanted to add better support for
> proxies, e.g. using a proxy only for URLs that match some pattern.
> Since people are asking about this, I'll expedite my release plans
> :-).  I'll try to make a useful release available early next week.
>
> Jeremy
>
>
>
>
>




From Alan.Daniels at p98.f112.n480.z2.fidonet.org  Wed Jun 30 23:55:35 1999
From: Alan.Daniels at p98.f112.n480.z2.fidonet.org (Alan Daniels)
Date: Thu, 01 Jul 1999 04:55:35 +0100
Subject: Python suitable for a game engine?
Message-ID: <000015f7@bossar.com.pl>

From: daniels at mindspring.com (Alan Daniels)

I'm working on a Quake-style game engine, where my goal is to give it
a fully integrated editor, similar to the excellent ones in Starcraft
and Unreal. I'm planning on doing this by giving the game and the
editor as much back-end code in common as possible.

What I want to do is be able to design the level in the editor, and
then sprinkle objects and monsters around the level. All of these will
have default low-level behaviors, so that level designers don't have
to do monster AI from scratch. However, all the objects and monsters
will have simple properties manipulated through a property list,
similar to that in VB. This way, they can set properties such as, is
the monster aggressive or passive, etc.

Here's the thing. I want to have a script as its main property, so that
the monsters can have elaborate behaviors, such as being mournful when
their leader is killed, etc. The events would be things such as
"create", "monster sees player", "clock tick", etc. All the functions
in the script would be callbacks written in C++, to manipulate the
monster to do things such as "patrol area", "play an animation",
"sleep for X seconds", etc. In other words, the scripts would be glue,
not the main thread of execution.

I'm been using Python for a couple of years now, and it is my favorite
scripting language. But, I don't have much experience with embedding
it in a another program, and I don't know enough about the speed
issues. So, my questions are...

- Will Python be fast enough? Even with possibly 100 or 200 active
  items on a level?
- Especially when all of the scripts will be live, and usually in the
  middle of a callback?
- Should I use the interpreter in the library, or should I take an
  approach of crunching the byte-codes myself?

I've seen plenty of cases where Python is used as a "VBA" style
embedded language (load up the script, run it, and be done with it),
but I haven't seen anything like this mentioned before. My other two
candidate languages are Scheme and Perl, and I hate the syntax of
both. So I really REALLY want to use Python if its viable. :=) If
anybody can shed some light on this, I'd appreciate it. Thanks.

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



From kjwolf at online.de  Sun Jun 20 21:43:33 1999
From: kjwolf at online.de (Klaus-Juergen Wolf)
Date: Mon, 21 Jun 1999 03:43:33 +0200
Subject: Problem with Curses
References: <ppgmj7.s7r.ln@solling.seismic.de> <jpeck7.1661.ln@solling.seismic.de> <ve1dk7.1e02.ln@solling.seismic.de> <87ogidzs37.fsf@home.ivm.de> <6Zeb3.526$4e1.32195@iad-read.news.verio.net>
Message-ID: <5c5kk7.fn.ln@solling.seismic.de>

T.E.Dickey <dickey at shell.clark.net> wrote:
 >>> It depends on when the executable is linked to libmytinfo, which only
 >>> takes place on platforms that don't know the terminfo database, like
 >>> BSD and Solaris; there are two competing declarations of tputs().

 >> Why doesn't Freebsd know the terminfo database? Can't the one that is 
 >> integrated into ncurses be used on FreeBSD? 

 > I've been told ncurses 4.2 is incorporated in the FreeBSD 3.1 - but that
 > may still just be a port...

Yes, there is such a port. Do you think, one could release code that
works on several machines, while some of them must install a special
program first?

In the top priority, Python should work. When you compile it, it should
work. When you decide to link libraries you have, it should work. When
you release code for Python, it *must* work.

Therefore, there are some modules that base on old, long forgotten
states, but still, some machines work with these, and only with these.

Curses now is a very old UI, very old, it was old when I first learned
it in the beginning of the 80's. Most people have forgotten that there
was a Curses before NCurses came up. cursesmodule now, it bases on
Curses with NCurses extensions. It is not acceptable to release a
cursesmodule that wouldn't work on all those platforms it formerly
worked on. If you intend that, you'll have to give it a new name.
Furthermore, you can't say, to make cursesmodule work, you have got to
download another one, it's beta, but maybe, it works on your desolated
site. But that's no solution. That's a cramp.

Anyway, that wouldn't have helped in my case, I think, since the name
conflict would have still existed - libtermcap would have been linked
before, since libreadline demands it, and, I'm afraid, since ncurses
originally defines tputs(), it would also been mis-linked. I don't
really know, I guess. I'm not so familiar with the dynamic linking
rules under ELF/BSD.

cu
 k.j.


-- 
 Support Julia Hill - save the Redwood trees: http://www.lunatree.org/



From greg.ewing at compaq.com  Thu Jun 10 20:02:27 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 11 Jun 1999 12:02:27 +1200
Subject: Python 2.0
References: <000b01beb21f$a17ff340$2fa22299@tim>
Message-ID: <37605213.2ACAE4D3@compaq.com>

Tim Peters wrote:
> 
> "Therefore, we recommend that the design of finalize methods be kept simple
> and that they be programmed defensively, so that they will work in all
> cases."

In my experience, this advice applies just as much to Python
__del__ methods, their alleged greater predictability
notwithstanding. 

The only times I can recall using __del__
methods were in shadow classes which needed to free resources
in an extension module. I found that I had to program them
very defensively, because when the interpreter shut down,
it would dismantle the contents of my modules in a random
order, causing my __del__ methods to crash if I wasn't
careful.

So I regard the environment of a __del__ method as
pretty much unknowable as well!

Greg



From news at dorb.com  Wed Jun 16 14:39:44 1999
From: news at dorb.com (Darrell)
Date: Wed, 16 Jun 1999 14:39:44 -0400
Subject: GC In Python: YAS (Yet Another Summary)
References: <1282571877-108998356@hypernet.com> <zoQ93.198$U6.1646@newsr2.twcny.rr.com> <wku2s86nmb.fsf@ifi.uio.no>
Message-ID: <peS93.209$U6.1777@newsr2.twcny.rr.com>

Lars Marius Garshol <larsga at ifi.uio.no> wrote in message
news:wku2s86nmb.fsf at ifi.uio.no...
>
> This may be a perfect world for lazy whiners after all. :)
>
> Check the web site again for a function that has a lazy mode. It
> doesn't print anything unless there actually is a cycle, but other
> than that it should make you happy.
>
> --Lars M.

Wow !
Lars you are the man. That was fast.

I'll give it a try.

Thanks very much.
--Darrell





From siona at chiark.greenend.org.uk  Tue Jun 22 06:50:25 1999
From: siona at chiark.greenend.org.uk (Sion Arrowsmith)
Date: 22 Jun 1999 11:50:25 +0100 (BST)
Subject: Creating a new type with an extension module
References: <Ctp*vl52n@news.chiark.greenend.org.uk>
Message-ID: <rvy*CQ92n@news.chiark.greenend.org.uk>

In article <Ctp*vl52n at news.chiark.greenend.org.uk>,
Sion Arrowsmith  <siona at chiark.greenend.org.uk> wrote:
>http://www.python.org/doc/current/api/newTypes.html is unhelpful
>in the extreme -- do I have to resort to reading source?

It appears that I do, but thanks to everyone who pointed me in
the direction of suitably readable source. I've got what I
wanted working -- it would just be nice to understand why it
works....

-- 
\S -- siona at chiark.greenend.org.uk -- http://www.chaos.org.uk/~sion/
  ___  |  "Frankly I have no feelings towards penguins one way or the other"
  \X/  |    -- Arthur C. Clarke
       |  If music be the food of love, bring me a doggie bag.



From gmcm at hypernet.com  Wed Jun 16 12:25:57 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 16 Jun 1999 11:25:57 -0500
Subject: Xbase files?
In-Reply-To: <1318D78C9072D11195C9006094EA98A72C2149@ocesrv>
Message-ID: <1282571876-108998462@hypernet.com>

> Is there any module for their reading?
> 
>    matthew

At ftp://ftp.python.org/pub/python/contrib/Database/  there should be 
a dbf.py for this purpose.

- Gordon



From adi at i-data.com  Thu Jun 10 05:46:46 1999
From: adi at i-data.com (Adi)
Date: Thu, 10 Jun 1999 09:46:46 +0000
Subject: Asserting object class
Message-ID: <375F8986.73BA3C5@i-data.com>

In a method, I'd like to assert that a parameter is an object which is
of a specific class or has this class as it's base class.

I can test on the __class__ and its __bases__ attributes, but its clumsy
and I'm wondering if there is an easier notation, like:

def method(self, p)
    assert class(p) == class(self)

This is comparable to - in a strongly typed language - specifying a
required base class as the type of the parameter.

Sincerely,
Anders Dinsen

--
i-data is a leading vendor of printing connectivity solutions,
e-forms software, and networking products.
More information is available at www.i-data.com

Disclaimer: I speak for myself, not my employer.






From matomira at acm.org  Mon Jun 14 08:22:22 1999
From: matomira at acm.org (Fernando D. Mato Mira)
Date: Mon, 14 Jun 1999 14:22:22 +0200
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <3762988E.C94AA252@iname.com> <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu> <3763C746.5CA8FE78@iname.com> <m3u2sby4gf.fsf@soma.andreas.org> <87pv2zjhhl.fsf@home.ivm.de>
Message-ID: <3764F3FE.F7CBE6BA@acm.org>

Klaus Schilling wrote:

> Prefix rocks, infix is crap.

Wow, wow, wow. Let's not go there..

--
Fernando D. Mato Mira
Real-Time SW Eng & Networking
Advanced Systems Engineering Division
CSEM
Jaquet-Droz 1                   email: matomira AT acm DOT org
CH-2007 Neuchatel                 tel:       +41 (32) 720-5157
Switzerland                       FAX:       +41 (32) 720-5720

www.csem.ch      www.vrai.com     ligwww.epfl.ch/matomira.html






From larsga at ifi.uio.no  Wed Jun 16 11:53:00 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 16 Jun 1999 17:53:00 +0200
Subject: OO (was: Why is tcl broken?)
References: <000801beb745$02768220$979e2299@tim> <871zfdp73f.fsf@home.ivm.de> <wklndk8ly6.fsf@ifi.uio.no> <873dzsma6b.fsf@home.ivm.de>
Message-ID: <wk3dzs87mr.fsf@ifi.uio.no>

* Klaus Schilling
| 
| No, in the sense `Steele merely rediscovered scheme, not bring it
| into being'

I don't understand what you mean by this. The two people who created
Scheme originally were Sussman and Steele. If that isn't bringing it
into being I don't know what would be.

(See section 2.8 of 'The Evolution of Lisp'.
<URL: http://www.stat.ucla.edu/develop/lisp/common/docs/Evolution-of-Lisp.ps.gz>)

--Lars M.



From see-sig at wavedivision.com  Tue Jun  8 22:05:58 1999
From: see-sig at wavedivision.com (Rick Robino)
Date: 9 Jun 1999 02:05:58 GMT
Subject: "Programming Python" still worthwhile?
Message-ID: <7jki66$qam$1@news.gstis.net>

Greetings,

I've been watching this group for some time and just now getting warm to
Python. Looking at source examples has been great, but for completeness'
sake I'd like to get more of the "why" maybe found in a book. Since Python
is at 1.5.2 and people are talking about 2.0 as if it's around the corner,
I was wondering if anyone here could give me some advice as to whether or
not "Programming Python" is useful, if the current edition is still
current, if I should wait for the "Learning Python", or if I should use
some other resources.

I expect to come up to speed fairly quickly and I'm just looking for
those fundamental insights into OOP and Python that can build up the
kind of inuition and understanding one needs in order to "get
comfortable" connecting/creating the dots. Currently I only program
small - medium programs in C and shell (Korn). Any sage pointers to
resources?

TIA,

--Rick 
                         ^^^^^^^^^^^^^^
Rick Robino                                  rrobino at wavedivision.com



From MHammond at skippinet.com.au  Wed Jun  2 19:19:22 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Thu, 3 Jun 1999 09:19:22 +1000
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net> <aahzFCpHw8.Gxt@netcom.com>
Message-ID: <7j4e2u$8rh$1@m2.c2.telstra-mm.net.au>

Aahz Maruch wrote:
>I disagree with your parenthetical comment.  I wouldn't call myself an
>"expert" Python programmer by any stretch of the imagination, but I've
>written a fair bit of it in connection with a multi-person project over
>the past four months.  As a seasoned programmer of many years and many
>languages, I've disagreed with almost all the whines that have been
>brought against Python (e.g. the stupid "=" vs. "==" thread), but this
>is one point that still annoys me.

Of course, this is your HO.  Many other people (including myself) found that
a few weeks in, they thought "this is great", and many years later, can't
believe that other language still hold on to the "old ways".  When I take a
quick look at some of the new little languages referenced on this newsgroup,
believe it or not, my _first_ reaction is "God those block closers are
stupid and ugly" :-)

Of course, I see and understand you concern, but I dont share it.  My guess,
from newgroup postings and anecdotal evidence, is that most seasoned
programmers also do not share it.

>code (which is what I've mostly been doing).  But the whitespace problem
>is a real PITA when it comes time to make a change to the structure of a
>program -- it's an investment I don't mind making when I'm doing "real"

Not to me.  If I make structural changes to C or other programs, I always
re-indent the reorganised code correctly, and I would bet that you do too.
In almost all languages block indenting is _very_ important to the human
reader.

Further, when indenting the new C/whatever code, I attempt to keep the
tabs/space usage correct (surprise surprise - not :-) so that other editors
still display the code correctly - there isnt much more annoying than seeing
_any_ code (Python or otherwise) incorrectly indented.

So, for me anyway, I tend to do _exactly_ what I do with Python with any
language.  The only difference I can see is that other languages force me to
use redundant block delimiters.

Mark.






From garry at sage.att.com  Thu Jun 17 16:41:25 1999
From: garry at sage.att.com (Garrett G. Hodgson)
Date: Thu, 17 Jun 1999 20:41:25 GMT
Subject: Starting a project: any hints?
References: <19990617154643.A1680@optiplex.palga.uucp> <7kb384$lgs$1@nnrp1.deja.com> <pT9a3.327$U6.2451@newsr2.twcny.rr.com>
Message-ID: <37695D75.CF8B877F@sage.att.com>

Darrell wrote:

> > - Make your program works from the very first day , it doesn't matter
> > if it barely works.
> 
> I'd like to emphasize this point. Don't write it all then try it.
> Its easier to evolve from a functional kernel than to materialize all at
> once.

i second that emotion.  also, don't be afraid to throw it
away and rewrite if you find yourself painted into a corner.
the value of the code is what you learned while writing it;
the rewrite will be all the better for it.

-- 
Garry Hodgson			comes a time
garry at sage.att.com		when the blind man
Software Innovation Services	takes your hand, says,
AT&T Labs			"don't you see?"



From Corran.Webster at p98.f112.n480.z2.fidonet.org  Tue Jun 29 16:43:25 1999
From: Corran.Webster at p98.f112.n480.z2.fidonet.org (Corran Webster)
Date: Tue, 29 Jun 1999 21:43:25 +0100
Subject: printing with print
Message-ID: <00001597@bossar.com.pl>

From: cwebster at math.tamu.edu (Corran Webster)

In article <377925EF.1CBB640F at starvision.com>,
Angus MacKay  <amackay at starvision.com> wrote:
>is there any other way to print that with print?
>perhaps with a real function.

Yes, use sys.stdout, which is a file object:

Python 1.5.2 (#13, Jun 16 1999, 23:19:35)  [GCC egcs-2.91.60 19981201
(egcs-1.1.1  on netbsd1
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import sys
>>> foo = "there"
>>> sys.stdout.write("Hi"); sys.stdout.write(foo);
Hithere>>>

Notice that it doesn't even put a newline in.  Also, you'll need to do
conversion to a string to get the result you want for foo = 2.

>>> sys.stdout.write("Hi"); sys.stdout.write(str(foo));
Hi2>>>

>foo = 2
>print "hi" + foo

You could do the explicit string conversion here too:

print "hi" + str(foo)

Corran



From rdelon at my-deja.com  Mon Jun 21 12:44:20 1999
From: rdelon at my-deja.com (Remi Delon)
Date: Mon, 21 Jun 1999 16:44:20 GMT
Subject: Check for valid date
Message-ID: <7klq51$etp$1@nnrp1.deja.com>

Hi, I'm trying to use the time module to check whether a date is
actually a valid day or not.
Unfortunately, it looks like the strptime function only performs a basic
range check. (for example, it will accept 02/30/199)
Here is a little test I did:

>>> strptime('02/30/1999','%m/%d/%Y)
(1999, 2, 30, 0, 0, 0, 61, 0)
>>> mktime((1999, 2, 30, 0, 0, 0, 61, 0))
920361600.0
>>> ctime(920361600.0)
'Tue Mar  2 00:00:00 1999'

Is there a way to check test whether the day actually exists or not ?


I also have a totally different problem : I'm trying to get python 1.5.2
to run with Tcl/Tk 8.1. Has anyone been able to do that on AIX ?
I found 2 patches for tkinter that were supposed to fix the problems,
but unfortunately, they don't work for AIX.

Thanks

Remi.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From thantos at brimstone.mecha  Fri Jun 25 02:17:35 1999
From: thantos at brimstone.mecha (Alexander Williams)
Date: Fri, 25 Jun 1999 06:17:35 GMT
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <m3aetr2yk9.fsf@atrus.jesus.cam.ac.uk> <7ks38p$l4m$1@brick.cswv.com> <slrn7f68br.2bp.thantos@brimstone.mecha>  <7kuocf$1an$1@brick.cswv.com>
Message-ID: <slrn7f96qm.qg.thantos@brimstone.mecha>

On 24 Jun 1999 21:09:19 -0500, Neel Krishnaswami <neelk at brick.cswv.com> wrote:
>IMO, it's bad style to set properties that should belong to the type
>in the instance initializer -- for example, the sort order is
>something that's a property of the class, rather than the instances.

Here's the crux of the problem, a sense of asthetics at variance with
the usefullness of your code.  Allow me to present a dissenting view:

Setting properties which can vary dynamically over the life of an
instance should be set on the instance itself, since the Class itself
may have multiple instantiations in a multithreaded or
multi-persistance environment.  Ultimately, it becomes easier to note
and manage varying fields on instances as they are easily seen in the
__init__ function (and so satisfy lexical locality in finding them for
maintainance) as well as cuts back on strange circumlocutions in
dealing with them syntactically.

>    sortorder = 'date' # can be anything in Article.comparators.keys()

Let's assume you set this in the __init__:

def __init__(me):
    me.sortorder = 'date'

>    def __cmp__(self, other):
>        return self.__class__.comparators[self.__class__.sortorder](self,
>                                                                    other)

.... then this becomes:

def __cmp__(me, them):
    return me.comparators[me.sortorder](me, them)

Note that comparators is set on the instance by inheritance, so we
don't need to obscure that with a specialized __class__ reference, and
with sortorder being an instance variable, there's no need for
obscuration /there/, either.

>I want the user to be able to determine how a list of Article's can be
>sorted by setting the class's sortorder member. So a statement like
>
>>>> Article.sortorder = 'author'
>
>would change how lists of articles are sorted. This is definitely 
>something that's shouldn't be an instance attribute, imo; it would
>defeat the purpose if each article could compare differently than
>any other. (Oppose standardized testing! :])

In this case, the order in which articles are sorted shouldn't be on
the Article object at all, then!  Remember, encapsulation,
encapsulation.  Sort-order storage should be on the object that you
use to contain references to Article-objects, and thus only needs to
be changed in one place at one time (but still should be an instance
variable, since it does change dynamically and shouldn't be inherited
by all children of ArticleStore, since each may want to be sorted in a
different way, providing different views into the same data).

>It's perfectly reasonable for someone else to subclass Article and
>override the Article.comparators attribute. (Say to add an ordering to
>put Tim Peter's posts at the top of the list, or to use something
>faster than my lambdas.) With the clear way of writing it, the
>NewArticle class would mysteriously fail to do the right thing,
>because the __cmp__ method would look in the Article class dictionary
>without bothering to check NewArticle's.

See above; note that calling me.comparators will /always/ work with
inheritance. 

>This is an unusual glitch in Python; usually the readable way of doing
>something is the right way of doing it. 

Ah, but it /is/ the right way, you're just insisting on the wrong way
out of a twisted sense of asthetics.  :)

-- 
Alexander Williams (thantos at gw.total-web.net)
"In the end ... Oblivion Always Wins."



From mikael at pobox.com  Wed Jun 30 19:51:19 1999
From: mikael at pobox.com (Mikael Lyngvig)
Date: Wed, 30 Jun 1999 23:51:19 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <slrn7nai1j.8ll.zeitlin@seth.lpthe.jussieu.fr>
Message-ID: <64F6457BE0585BB3.1CF2F67BC8A680BD.A19A9FEE23BC9E58@lp.airnews.net>

On 26 Jun 1999 21:37:12 GMT, zeitlin at seth.lpthe.jussieu.fr (Vadim
Zeitlin) wrote:

> While I don't think wxPython can replace TkInter as a de-facto GUI standard
>for Python I still think that it has several advantages compared to TkInter.
>Instead of saying how much better, prettier, faster (this one has really made
>me smile, BTW: I don't think wxWin is currently very fast, if it has been
>cited as its advantage it must really mean that TkInter is extremely slow. 

Well, TkInter is extremely slow, but not because of the Python part.
Even the Tk widget tour, included with Tk, is so slow that you almost
think "Please, draw the next character - you can do it." whenever a
window with some text needs to be redrawn.

In contrast, the wxPython demo is so fast that you probably wouldn't
realize it is implemented in Python, unless you're used to do GUI
programming yourself.

One important point is that I probably experience Tk(Inter) as slow
because I'm used to Win32 desktops with very fast graphics hardware
and no X/Windows style client/server architecture.  I imagine that
many X/Windows users experience Tk(Inter) as fast enough.

>we're working on performance problems too...) - instead, I'd just like to say
>that unlike probably any other GUI toolkit, wxWindows developers are
>interested in making their toolkit useful to the Python community. I don't
>speak only about Robin Dunn, of course, who developed wxPython, but about
>"core" wxWindows developers as well.

The wxWindows performance problems must be more of the "this could be
done better" than "this has to be done better" nature.  

> TkInter has, for the moment, several advantages compared to wxWindows (it
>would be nice if someone did the comparison of 2 toolkits - I'm afraid I don't
>know TkInter well enough to do it), but we're working hard to change this and
>I sincerely believe that wxWin will have all the features we still miss and
>which are present in TkInter really soon (an advanced text widget and
>integrated HTML viewer are in alpha and beta stage; image support in wxWin
>already includes GIF/JPEG/PNG/BMP/XPM; multimedia classes are waiting to
>be integrated in the main distribution; the grid control is being actively
>worked on by several people right now).

A very strong plus for wxWindows is that the Python API and the native
C++ API are so close that it actually makes sense to use Python in the
design and prototyping phase and then rewrite the Python code to C++,
if you want to, when you've reached a stable design.

Personally I feel that it makes most sense to retain TkInter as the
de-facto GUI for Python - at least for a while yet.  But, I don't
think I will ever use TkInter for anything.  (this is probably an
argument for not having a de-facto standard GUI for Python)

As my reply to Michael P. Reilly shows, I don't believe that TkInter
can be boosted significantly unless a *lot* of energy is put into the
task.  I think it would be better to spend this energy on extending
and enhancing wxWindows till it is ready to take over.

P.S. No offense intended to TkInter users.  This is just the way I see
things.


-- Mikael



From s_hubbard at my-deja.com  Tue Jun 15 07:18:20 1999
From: s_hubbard at my-deja.com (s_hubbard at my-deja.com)
Date: Tue, 15 Jun 1999 11:18:20 GMT
Subject: Python CGI on IIS
Message-ID: <7k5cpn$jdk$1@nnrp1.deja.com>

Hello and Help

I'm writing CGI using Python rather than embedding python in ASP, and I
need to know how and if I can read and write to the session object.

Regards and Thanks in Advance
Senior Developer
Simon Hubbard
Rapid Business Solutions


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From Tim.Peters at p98.f112.n480.z2.fidonet.org  Tue Jun 29 03:56:56 1999
From: Tim.Peters at p98.f112.n480.z2.fidonet.org (Tim Peters)
Date: Tue, 29 Jun 1999 08:56:56 +0100
Subject: total idiot question: +=, .=, etc...
Message-ID: <00001594@bossar.com.pl>

From: "Tim Peters" <tim_one at email.msn.com>

>>>     class A(B):
>>>         def __init__(self, x, y, z):
>>>             B.__init__(self, x, y, z)

[Greg Ewing]
> I could live with having to explicitly name the
> superclass, provided something checked that the
> class I named was actually a direct base class
> of the one where the method is defined. That
> way things would be less likely to break
> mysteriously when I rearrange the class
> hierarchy.

[Guido]
> Oh, but that check *is* being made!  (In fact this is the same check
> that prevents "class methods" from working ;-)

Nope, that's a different check.  If C derives from B derives from A, this
check allows either C.__init__ or B.__init__ to be called from A.__init__
with an instance of A.  Greg wants  B.__init__ to be OK here, but not
C.__init__ (C is not a direct base class of the one (A) where the method
(A.__init__) is defined).  IOW, he wants a way to restrict __init__ calls to
immediate bases of the class from which the __init__ is invoked, to e.g.
prevent level-skipping under class rearrangement.

I posted some ugly code to address that, but think it's really a job for a
text editor and a pair of eyeballs <wink>.

although-something-to-be-said-for-automating-the-eyeballs-ly y'rs  - tim






From news at dorb.com  Mon Jun 21 15:57:58 1999
From: news at dorb.com (Darrell)
Date: Mon, 21 Jun 1999 15:57:58 -0400
Subject: choosing random numbers with weights/probability?
References: <929987267.6227@www.remarq.com> <lbwb3.3$V4.11640@news-dal.corridex.com>
Message-ID: <ERwb3.1452$U6.5885@newsr2.twcny.rr.com>

import time, whrandom

list1 = [('one', 0.25), ('two', 0.25), ('three', 0.5)]
def wc(list):
   from whrandom import uniform
   n = uniform(0, 1)
   for item, weight in list:
     if n < weight:
       break
     n = n - weight
   return item


def wc_test1(list, cnt=100000):
 t1=time.time()
 hist={}
 for i in range(cnt):
  k=wc(list)
  if hist.has_key(k):
          hist[k]=hist[k]+1
  else:
          hist[k]=1
        print " Time: %.2f sec"%(time.time()-t1)
 return hist

def popList(l):
 selList=[]
 for i in l:
   for v in range(i[1]):
   selList.append(i[0])
  return selList

list2=popList([('one',25),('two',50),('three',25)])

def wc_test2(list, cnt=100000):
 t1=time.time()
 hist={}
 list=popList([('one',25),('two',50),('three',25)])
 for i in range(cnt):
  k=whrandom.choice(list)
  if hist.has_key(k):
          hist[k]=hist[k]+1
  else:
          hist[k]=1
        print " Time: %.2f sec"%(time.time()-t1)
 return hist


wc_test1(list1)
 Time: 10.36 sec
{'one': 25069, 'three': 50087, 'two': 24844}
wc_test2(list2)
 Time: 4.52 sec
{'one': 24964, 'two': 50278, 'three': 24758}

Selecting from a list is faster.
--
--Darrell





From news at dorb.com  Wed Jun  2 17:33:28 1999
From: news at dorb.com (Darrell)
Date: Wed, 2 Jun 1999 17:33:28 -0400
Subject: Python 2.0
References: <11A17AA2B9EAD111BCEA00A0C9B4179301E10ED6@forest.origin.ea.com>
Message-ID: <Eth53.1798$j7.26742@newsr2.twcny.rr.com>

Stidolph, David <stidolph at origin.ea.com> wrote in message
news:11A17AA2B9EAD111BCEA00A0C9B4179301E10ED6 at forest.origin.ea.com...
> >OK, I'll try to propose something constructive. Maybe we need an object
> >protocol, that would enumerate
> >all references held by an object ?  Writing a portable GC would be then
> much
> >easier.
>
> Sounds good to me for debugging.  A call that could return a list of
> everything that holds a reference to an object - that would be cool!
>
This would be very useful.
When I've needed this, I didn't want to see everything in the system.
In an effort to avoid performance problems and major changes.
Insert a soon to be designed trace object as a member of the instance you
want to trace. It captures file/line number ...
Haven't I seen this sort of thing here before ?

For those times when you have no idea where to begin.
sys.settrace could be used in some way to watch everything. Slowly yes.

The output could be analyzed for cyclic references. Run this like a lint.
Now that you've found the cycles could some code be auto generated to handle
the clean up ?

--
--Darrell





From Robert.Meegan at wcom.com  Fri Jun  4 12:57:38 1999
From: Robert.Meegan at wcom.com (Robert Meegan)
Date: Fri, 4 Jun 1999 16:57:38 GMT
Subject: "The Python Way"
In-Reply-To: <14167.60002.783156.583193@anthem.cnri.reston.va.us>
References: <14167.60002.783156.583193@anthem.cnri.reston.va.us>
Message-ID: <Pine.A41.4.10.9906041155510.41420-100000@barney.mcit.com>

On Fri, 4 Jun 1999, Barry A. Warsaw wrote:

[...]

> 
> Way I see it, making the Timbot code Open Source is the only moral thing
> to do.  Besides, wouldn't it be cool to  check out the Python
> translation of the old Emacs Lisp RMS AI code?
> 

Actually, he's just a version of ELIZA that's stuck in a particularly
cranky loop.

--- Robert


--------------------------------
Robert Meegan
MCIWorldCom - Cedar Rapids, Iowa
319.375.2416




From gvinther at atex.com  Tue Jun  8 14:11:40 1999
From: gvinther at atex.com (Gordon Vinther)
Date: Tue, 8 Jun 1999 14:11:40 -0400
Subject: Tkinter doesn't work with Tcl/Tk8.0.5 part 2
Message-ID: <928865476.429.20@news.remarQ.com>

I forgot to mention that  I rebuilt with Tcl7.6/Tk4.2 and everything is
hunky dory.





From tim_one at email.msn.com  Sat Jun 26 13:48:27 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 26 Jun 1999 13:48:27 -0400
Subject: GC In Python: YAS (Yet Another Summary)
In-Reply-To: <3773EFF7.27D29899@bioreason.com>
Message-ID: <000301bebffc$184fcf60$da9e2299@tim>

[Andrew Dalke, worries about Boehm collection vs Fortran]
> The latter.   Example object creation code might look like:
>
> dt_Handle mol = dt_smilin("CCC", 3);
>
> where dt_Handle is a typedef to unsigned long.  In the SWIGged
> wrapped version, this is mol = dt_smilin("CCC").
>
> Yes, the library has a pointer table to the object.  The problem
> I see is, how does the Boehm collector (or other GC) know what to
> collect from this vendor library?  I can't see how, given that
> my Python code stores it as an integer.
>
> That's why I don't see how you say "It fully supports C and C++"
> when this is a package which doesn't do "pointer XORing or other
> crimes" but cannot be usable with a GC w/o modifications.
>
> But then, I'm a computational biophysicist by training and don't
> know much about GC other than the general concepts.  One of the
> references on the SGI page says Boehm can be used in "uncooperative
> environments" so I'll end by saying that I don't know enough.

You know enough to say this:  no matter *what* Fortran throws at it,
Python's current flavor of GC won't have a problem with it <0.5 wink>.

The BDW collector is a truly wonderful example of the GC art, and can
perform seeming miracles in hostile environments.  But you can't know
whether it will work * a priori*.  It needs to take over all dynamic
allocation (typically "malloc" in C, but your platform's Fortran may not use
that), understand everything about how your platform's compilers and runtime
systems use the machine stack and registers, and implicitly relies on the
compiler not generating "confusing" code (pointer xor'ing is something
compiler-generated code probably never does, but aggressively optimizing
compilers can & do perform transformations equally as confusing to BDW;
Boehm wrote a paper about that, so I'll skip it <wink>).

All of those caveats have to do with BDW being able to find every
bit-pattern that *may* be a live pointer.  It typically doesn't know ints
from doubles from real pointers, and doesn't care:  if a string of bits
"looks like" a heap address, it's treated as one.  This makes it err on the
side of safety, but all assuming that all live pointers are found in their
natural form in the places it knows where to look.

Despite all that, it would probably work fine for you.  I think it's a dead
issue here regardless, though, because Guido isn't going to tell anyone they
have to use a particular routine for memory allocation.  Python could use
its own malloc for its own objects (& Vladimir Marangozov has written a very
nice PyMalloc, available from his Starship page), but that breaks down as
soon as any extension has its own view of how to get memory.

can't-chase-what-you-can't-see-ly y'rs  - tim





From hinsen at cnrs-orleans.fr  Mon Jun  7 04:43:35 1999
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 07 Jun 1999 10:43:35 +0200
Subject: Cannot store Sequences in Numeric arrays of Objects
References: <375453CE.3192@cs.bham.ac.uk>
Message-ID: <m3vhd01juw.fsf@chinon.cnrs-orleans.fr>

Stuart I Reynolds <S.I.Reynolds at cs.bham.ac.uk> writes:

> >>>#Cannot store an array
> >>> a[0,0] = b
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ValueError: array too large for destination

NumPy handles assignments from sequences to arrays correctly in the
"standard" cases, but unfortunately this makes it difficult to assign
sequences to individual elements of general object arrays (which
aren't that well supported anyway).

However, you can assign sequences to elements of one-dimensional arrays:

>>> from Numeric import *
>>> a = zeros((5,), PyObject)
>>> a[0] = [1, 2, 3]
>>> a[1] = ['a', ['b', 'c']]
>>> a
array([[1, 2, 3] , ['a', ['b', 'c']] , 0 , 0 ,
            0 ],'O')
>>> a[2:4] = ['foo', 'bar']
>>> a[4] = array(3*[None], PyObject)
>>> a
array([[1, 2, 3] , ['a', ['b', 'c']] , foo ,
            bar , [None  None  None ] ],'O')

In your case, a[0][0] = b will do what you want. I don't know why
assignment to higher-dimensional arrays fails, but I don't really want
to figure it out either!
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------



From tre17 at cosc.canterbury.ac.nz  Mon Jun 21 22:41:53 1999
From: tre17 at cosc.canterbury.ac.nz (Timothy R Evans)
Date: 22 Jun 1999 14:41:53 +1200
Subject: tkMessageBox returns 'can't invoke "frame" command:' error
References: <7kktvv$550$1@nnrp1.deja.com>
Message-ID: <c75so7lklce.fsf@pc142.cosc.canterbury.ac.nz>

The actuall error is caused by the Tk mainloop cannot be restarted,
obviously Tkinter was never designed to support what you're doing
here.  The cause of your problem seems be partly that you actually
expected this to work, putting lots of code in a module initialisation 
is not good style.  If you post a description of what you are trying
to do I can give you a correct and better style solution.

--
Tim Evans



From fatjim at home.com  Thu Jun  3 13:37:12 1999
From: fatjim at home.com (Jim Meier)
Date: Thu, 03 Jun 1999 17:37:12 GMT
Subject: pyGTK on Windows
References: <3757D453.A302E757@interar.com.ar>
Message-ID: <3756BDC7.5382BE2@home.com>

Nahuel wrote:

> Hi, im about to do a cross-plataform GUI application in Linux and
> Windoze... i must
> choose the GUI toolkit.. my question is .. can i use pyGTK under
> Windows???
>
> (I need a gui that can display a table with about 4000 lines without
> crashing (not like Tk :))

For now, GTK is strictly a unix toolkit. There are some ports to
windows, but most require an X server. I did run across one that didn't
require X, but I lost it's address. It was very alpha anyways.

I don't know what your best bet is, but try searching deja.com for keys
like "toolkit", "tkinter", "pyGTK" and "wxPython". That last one sounds
like it might be your answer, but I've never tried it.


-Jim




From Richard.Jones at fulcrum.com.au  Mon Jun 14 20:21:31 1999
From: Richard.Jones at fulcrum.com.au (Richard Jones)
Date: Tue, 15 Jun 1999 10:21:31 +1000
Subject: python-1.5.2 Alpha Linux RPM?
Message-ID: <199906150021.KAA04106@icarus.fulcrum.com.au>

Does anyone know of the existence of an RPM of 1.5.2 for Alpha Linux?


Thanks,
    Richard





From aa8vb at yahoo.com  Wed Jun  2 12:18:41 1999
From: aa8vb at yahoo.com (Randall Hopper)
Date: Wed, 2 Jun 1999 12:18:41 -0400
Subject: A tkinter kick-off
In-Reply-To: <37544a32.0@news1.mcmail.com>; from Mark Butterworth on Tue, Jun 01, 1999 at 09:58:01PM +0100
References: <37544a32.0@news1.mcmail.com>
Message-ID: <19990602121841.A280136@epa.gov>

Mark Butterworth:
 |I've been using python for a few months now (I've actually managed to
 |persuade my company that freeware is a viable tool) and I've just started
 |looking at tkinter.  However,  my first little project doesn't  require an
 |interactive window or dialog box.  What I do require is a window which
 |displays the result's of my background processing.  The processing will
 |simply be the compression and network transfer of some files.  I'm not
 |looking for a full solution however, I just need a help starting off.  I can
 |create a window containing a listbox alright, I just need to know how to
 |process in the background and display results within the listbox (maybe a
 |cancel button would also be useful).
 |
 |Any help appreciated!?!

For a recent project, I needed exactly what you describe.  See the
ShellWindow.py example in the Python dist.  A little adaptation of this
module was all that was needed.

Randall





From tim_one at email.msn.com  Sun Jun  6 14:04:50 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 6 Jun 1999 14:04:50 -0400
Subject: Using CVS under Win95
In-Reply-To: <yoy63.1157$VL2.171084@news.direcpc.com>
Message-ID: <000001beb047$126c3300$579e2299@tim>

[Emile van Sebille]
> I've downloaded CVS and appear to have successfully compiled it using MS
> C++5.0.

Do yourself a favor and use Cyclic's precompiled Windows distribution.

> I'm trying to access the python source per (session paste follows):
>
> D>cvs -d :pserver:pythoncvs at cvs.python.org:/projects/cvsroot login
> (Logging in to pythoncvs at cvs.python.org)
> CVS.EXE [login aborted]: could not find out home directory

Ya, I vaguely remember this.  Went away after setting a bunch of environment
vrbls for CVS's benefit ...

set HOMEDRIVE=d:
set HOMEPATH=\hugga
set CVSROOT=:local:d:\hugga

Why on earth I installed it into d:\hugga will forever remain a mystery even
to me, but having done so CVS wants to be told about it in multiple
redundant ways <wink>.

linux-isn't-nearly-as-much-fun-ly y'rs  - tim





From news at dorb.com  Wed Jun 23 16:38:40 1999
From: news at dorb.com (Darrell)
Date: Wed, 23 Jun 1999 16:38:40 -0400
Subject: How to obtain an instance's name at runtime?
References: <7kqpu8$7u1$1@nnrp1.deja.com>
Message-ID: <LDbc3.2193$U6.8014@newsr2.twcny.rr.com>

Maybe I'm missing the point here.
I use this approach which shows the class name and an address for the
instance. That's about as good a name as I can imagine possible for that
instance.  Unless you start adding name attributes.

>>> class A:
...     pass
...
>>> a=A()
>>> str(a)
'<__main__.A instance at 7f6350>'

If you want to keep track of the name of the variable that your instance is
contained within. Then I don't think this is possible because it would
require that each instance keep a list of the variables that it was stored
in. Even then you would be faced with a list instead of a single name.
>>> str(b)
'<__main__.A instance at 7f6350>'
--
--Darrell





From velzev at freemail.nl  Tue Jun 15 07:43:49 1999
From: velzev at freemail.nl (Vivienne)
Date: Tue, 15 Jun 1999 13:43:49 +0200
Subject: (wxPython) Double left clicking and right clicking with ListBoxes
Message-ID: <7k5ec1$pl3$1@zonnetje.NL.net>

Hi there,

I've gotten into some trouble over making a selection in a listbox with a
double-left click or a right-click. It doesn't work. At first I used the
available EVT_LISTBOX_DCLICK(id, function). The problem here is that it
forgets the selection that I made, and with event.GetSelection() always
gives the index of the first item (i.e. 0). So I thought, OK, let's try it
with EVT_COMMAND_LEFT_DCLICK(id, function). But now my application doesn't
even respond to the fact that I'm double-clicking.
The problem with right-clicking is twofold :
1. There is no available method for right-clicking in a listbox (that I
could find)
2. The use of EVT_COMMAND_RIGHT_CLICK(id, function) gives the same result as
EVT_COMMAND_LEFT_DCLICK(id, function)

Could anyone please tell me how to use double-left clicking and right
clicking in listboxes (or wxPython in general)?
Thanks alot,

Vivienne
PS. I use Python 1.5.2 and wxPython 2_0b9. Here is the code I've written
that doesn't work:

def InitializeBox(self, ID, StartPoint, BoxSize):
          self.box = wxListBox(panel, ID, StartPoint, BoxSize,
self.namenlijst, wxLB_SINGLE|wxLB_SORT)
          EVT_LISTBOX(panel,ID,self.OnLeftClick)
          EVT_COMMAND_LEFT_DCLICK(panel,ID,self.OnDLeftClick)
          EVT_COMMAND_RIGHT_CLICK(panel, ID, self.OnRightClick)

def OnDLeftClick(self, event):
          p = event.GetSelection()
          print p






From arcege at shore.net  Wed Jun 30 10:07:14 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 30 Jun 1999 14:07:14 GMT
Subject: PySequence_Check
References: <000401bec2a4$3be15200$f39e2299@tim>
Message-ID: <mwpe3.325$6M6.116682@news.shore.net>

Tim Peters <tim_one at email.msn.com> wrote:
: [Reuben Sumner]
:> How can I do the equivalent of a PySequence_Check in Python?
:> GvR told me that the question was illdefined (if I explained it correctly)
:> and to ask here.  It seems that some of the functions in abstract.h
:> are available from python (len(), callable()) and others are not
:> (PySequence_Check, PyNumber_Check).
:>
:> What am I missing?

: For starters, unquestioning faith in Guido's pronouncements <wink>.

:> I have a nested structure of lists and tuple and I just want to know
:> when I have bottomed out.

: Then why not check for that directly and be done with it?

:> I can check for tuple or list indivually but that is ugly and
:> inflexible.

: Not to mention simple, obvious and unsurprising <wink>.

:> I could just try indexing and see if there was an exception but that
:> is not much better.

: OK, you tell me:  what *is* a sequence?  Answer that, and you'll know how to
: check for it <0.5 wink>.  Before answering too fast, note that mapping types
: can be indexed without raising an exception, and so can strings.  Do you
: want your nest of "lists and tuples" *not* to "bottom out" at a string?
: Probably not -- but if so, you're not really interested in testing whether
: it's a sequence!  Note too that user-defined classes can support the
: __getitem__ protocol without supporting the __getslice__ protocol, or vice
: versa.  Are either of those sequences, or do they need to support both?
: Whichever way you answer, half the world will disagree.

: I think that, in most cases, when people say "sequence":  (A) it's
: ill-defined <wink>; and, (B) after five rounds of tedious questioning they
: end up deciding they mean "responds to slicing but isn't a string".  So
: there you go:

: def issequence(x):
:     try:
:         x[0:0]
:     except:
:         return 0
:     return type(x) is not type("")

My basis (in my own programming) has been:
  import types, UserList
  def issequence(x):
    return (isinstance(x, types.ListType) or
            isinstance(x, types.TupleType) or
            isinstance(x, UserList.UserList))

It is very restrictive in some cases, but if I want a user defined
sequence, I usually _try_ to make one from UserList.

  -Arcege




From a.mueller at icrf.icnet.uk  Mon Jun  7 14:29:37 1999
From: a.mueller at icrf.icnet.uk (Arne Mueller)
Date: Mon, 07 Jun 1999 19:29:37 +0100
Subject: waste of resources ?
Message-ID: <375C0F91.1C73E4E3@icrf.icnet.uk>

Hi All

In my program I always get an OSError number 11 (Resource temporarely
not available). It reads in lots of files using child processes
communication with the parent. Each child also uses popen to get data
from lauched 'gunzip' process. After ~1000 processed files the program
raises OSError[11]. I'm pretty sure that all the child processes close
their pipes to the parent and that popen is also closed. The parent
stores the file descriptors connected to the children in a list but uses
the lists 'remove' method to delete descriptors that are finished (those
which the parent has already read from). 

However, I'd like to get sure that's not the number of used
filedescriptors what's causing the error. How can I examine the cause of
error, I mean is there a python method or a variable that tell you the
number of currently used file descriptors or better that lists all the
resources a program is using including it's children? 

Hm, maybe I'll have a hard night of debugging :-(

	thanks,

	Arne

-- 
Arne Mueller
Biomolecular Modelling Laboratory
Imperial Cancer Research Fund
44 Lincoln's Inn Fields
London WC2A 3PX, U.K.
phone : +44-(0)171 2693405      | Fax : +44-(0)171 269 3258
email : a.mueller at icrf.icnet.uk | http://www.icnet.uk/bmm/



From cgw at fnal.gov  Tue Jun  1 13:05:34 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Tue, 1 Jun 1999 17:05:34 GMT
Subject: Looking for a lib fun
In-Reply-To: <3753FDF4.5A13F8B8@zipzap.ch>
References: <3753FDF4.5A13F8B8@zipzap.ch>
Message-ID: <14164.4830.69989.763470@buffalo.fnal.gov>

Boris Borcic writes:
 > There is a function I can't seem to find in the
 > standard library. One that will turn XML/HTML
 > offending characters to corresponding entity- or
 > character references.

is cgi.escape() what you're looking for?



From garry at sage.att.com  Tue Jun  8 10:18:16 1999
From: garry at sage.att.com (Garrett G. Hodgson)
Date: Tue, 8 Jun 1999 14:18:16 GMT
Subject: Teaching python to beginners
References: <7ipbca$4jh$2@news.islandnet.com>
Message-ID: <375D2628.3B3BC188@sage.att.com>

Deryk Barker wrote:
> 
> Does anyone have any experience with teaching Python as a *first*
> programming language.
> 
> I'm looking into the feasbility of this and would like to discuss it
> with anyone who has experience or thoughts in the area.

i've used it to teach my 11 year old to program.
python is just about ideal, in my estimation.
we've just done some simple things, cataloguing
her beanie baby collection in a flat file database.
but she found it very easy to pick up.

-- 
Garry Hodgson			comes a time
garry at sage.att.com		when the blind man
Software Innovation Services	takes your hand, says,
AT&T Labs			"don't you see?"



From marc at oscar.noc.cv.net  Mon Jun 28 02:22:29 1999
From: marc at oscar.noc.cv.net (marc)
Date: Mon, 28 Jun 1999 06:22:29 GMT
Subject: Python for large projects?
References: <3774b5d1.690922@news.force9.net>  <T6bd3.2883$U6.12014@newsr2.twcny.rr.com> <slrn7naup9.jk.thantos@brimstone.mecha> <RGid3.2957$U6.12374@newsr2.twcny.rr.com> <87n1xm85mf.fsf@lisboa.i-did-not-set--mail-host-address--so-shoot-me> <Jhsd3.3090$U6.12410@newsr2.twcny.rr.com> <3776EF65.A07CBEFC@compaq.com>
Message-ID: <slrn7ne533.1rmr.marc@oscar.noc.cv.net>

In article <3776EF65.A07CBEFC at compaq.com>, Greg Ewing wrote:
>Darrell wrote:
>> 
>> I'll give that a look, but they can always get to it if they want. Maybe
>> that's ok, just worries me.
>
>If you clearly warn them in the docs not to do
>it, and they do it anyway and get into trouble,
>it's not your fault.
>
>Greg

But what does management think??  If the user screws something up how
much down time/loss of data will result?  I remember people coming back
from desert storm and telling me about how people were playing with 
unexploded submunitions, aka small bombs with flaky fuses, I am not
kidding on this.  So the question is how much damage can an moron 
do with a screw driver to your program and is that an acceptable
tradeoff for the benifits of python?  If you are realy parinoid
present it to your boss and ask him to decide.

marc

ps just because you are parinoid does not mean you are wrong.



From andermoe at online.no  Wed Jun 16 16:13:51 1999
From: andermoe at online.no (Anders Moe)
Date: Wed, 16 Jun 1999 22:13:51 +0200
Subject: need exit from 'exec' execution
Message-ID: <3768057F.D7473931@online.no>

Hi all -

My python-embedded app frequently makes use of  'exec <program-string>'.
I'd like
to create a Tk Button to halt this execution at some point, with the
option of aborting it. The only command that comes close is the exit
statement, but its to powerful - it
exits from the whole app, whereas I only wanted a local exit from the
exec's
program-string. Anyone know how to make this kind of local exit ? Thanks

in advance,


Anders Moe





From greg.ewing at compaq.com  Wed Jun 23 17:27:16 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Thu, 24 Jun 1999 09:27:16 +1200
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0o3q.toe.bernhard@alpha1.csd.uwm.edu>
Message-ID: <37715134.399AFB2D@compaq.com>

Bernhard Reiter wrote:
> 
> += just saves some space and typing time.

It's not quite as simple as that, because

  a[i] = a[i] + expr

and

  a[i] += expr

would be semantically different if the former
did the index lookup only once. However, there
are serious difficulties with trying to incorporate
such semantics into Python's execution model,
and the consensus among experienced Python
implementation hackers seems to be that the results
would be Not Worth The Bother (tm).

Greg



From arcege at shore.net  Thu Jun  3 14:44:05 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Thu, 03 Jun 1999 18:44:05 GMT
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net> <aahzFCpHw8.Gxt@netcom.com> <19990602132331.A281155@epa.gov>
Message-ID: <V1A53.1778$nn.529906@news.shore.net>

Randall Hopper <aa8vb at yahoo.com> wrote:
: Aahz Maruch:
:  |Paul Prescod  <paul at prescod.net> wrote:
:  |>
:  |>...whitespace handling....
:  |
:  |...this is one point that still annoys me.
:  |
:  |Oh, I've gotten used to it, no question about that, and I even
:  |appreciate the readability when it comes to working over someone else's
:  |code (which is what I've mostly been doing).  But the whitespace problem
:  |is a real PITA when it comes time to make a change to the structure of a
:  |program -- it's an investment I don't mind making when I'm doing "real"
:  |programming, but when I'm making a bunch of quick, fiddly changes to
:  |debug something it continually gets in my way.

: I actually have like Python's whitespace-as-syntax, but I can see your
: perspective.  

: If I was still locked onto vi, it'd be a pain, and I can see that being an
: issue with some editors.  Though with emacs, there's rectangular kill
: (^Xrk) and block indent/dedent (^X^I) -- key sequences I use without even
: thinking anymore.

I don't even see it as a problem in vi.  There are pretty simple block
operations (:'a,.s/^/  /) that make it easy enough (as easy as vi can
be anyway).

Given the problem of multiple coders and "proper whitespace" (insert
your definition here), there is _usually_ going to be an indent/outdent
problem, regardless of the language or the editor.

  -Arcege




From mark.geib at echostar.com  Mon Jun  7 12:22:05 1999
From: mark.geib at echostar.com (mark.geib at echostar.com)
Date: Mon, 07 Jun 1999 16:22:05 GMT
Subject: can't link with python libs
Message-ID: <7jgrj2$rin$1@nnrp1.deja.com>

I have installed python 1.5.2 on Redhat 5.2 Linux, both the python and
python-dev RPMs. The problem is I am writting a C 'main' funtion to
invoke python but cannot link the program because I have no libModules,
libPython, etc. Is there another package I need to install or what...?

THanks,

Mark Geib
Echostar


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From claird at Starbase.NeoSoft.COM  Thu Jun 10 13:29:37 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 10 Jun 1999 12:29:37 -0500
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com>
Message-ID: <7josm1$bgc$1@Starbase.NeoSoft.COM>

In article <375FDCA7.62FF at mailserver.hursley.ibm.com>,
Paul Duffin  <pduffin at mailserver.hursley.ibm.com> wrote:
			.
			.
			.
>Tcl has many OO features, you can have C++ like, Java like, SmallTalk
>like ... it is up to you. A lot of them are implemented in pure Tcl.
With new ones coming on line all the time:  <URL:http://
starbase.neosoft.com/~claird/comp.lang.tcl/tcl.html#ootcl>.
>
>> Has anyone messed with Guile, GNU's extension language?  One specific
>> thing they want to do is be a "universal" back end, that is you can write
>> translators from language A to Guile, and their page[2] mentions Python
>> several times...
>> 
>
>Nice idea but I doubt they will even manage one language completely,
>never mind 2, or 3, or 4 ...
Oh, it appears to be in use already.  Skepticism about
its larger ambitions, though, is widespread <URL:http://
www.sunworld.com/swol-11-1998/swol-11-regex.html>.  I
wish the Guile folk well, and certainly RSM and others
continue to invest energy in it.  From my distant ob-
servation post, it doesn't feel as though it has the
critical mass of resources necessary to "take off".

I believe the time *is* ripe for deeper connections be-
tween different languages.  Perhaps what the original
questioner was after was something like, "What are apt
areas for Tcl, and how can other technologies exploit
its strengths?"

OK, that's too coy.  I recognize that *I* wanted him
to have asked that, but he definitely didn't do so.
Like Paul, I find it hard to think of syntax and seman-
tics which are bad in isolation from context.
			.
			.
			.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From mikey1cal at aol.com  Wed Jun 23 16:59:46 1999
From: mikey1cal at aol.com (Mike Callahan)
Date: Wed, 23 Jun 1999 20:59:46 GMT
Subject: newbie IDLE question
Message-ID: <37714A8C.C6E581@aol.com>

Where can I find documentation on IDLE? Also, is version 0.5 ready to be
downloaded? What about any IDLE vs PTUI comparisons.

Mike Callahan



From vdkoijk at linux01.prc.tno.nl  Wed Jun 16 08:36:12 1999
From: vdkoijk at linux01.prc.tno.nl (John van der Koijk)
Date: 16 Jun 1999 14:36:12 +0200
Subject: DDE link memory leak?
Message-ID: <87vhco726b.fsf@linux01.prc.tno.nl>

Hi There,

I'm using dde under PythonWin to control a data aquisition system
known as Windmill (http://www.windmill.co.uk). This system contains a
program called wmdde (DDE link) which provides an interface to the
hardware via DDE (don't ask me why I use this).

If one is very gentle with the DDE interface, it works. For a
while. Then, it crashes. Interestingly, it is not possible to start
new processes on the Windows machine after that happens: it complains
that there is not enough memory available. I suspect that something is
leaking memory.

Being a Unix afficionado I was allready happy I got it to work in the
first place, and I haven't got the foggiest about what to do now.  For
instance, I can kill PythonWin and all associated processes, and still
have no memory left!? I can't even start a shell window.

Windmill says: use Excell (or VB). It works fine for our
customers. Well...

I've also tried to use calldll, but the IMLKIT dll delivered by
Windmill does not seem to work (16-bit?).

Any pointers are appreciated,
-- 
John van der Koijk.



From " HPeter.Friedrichs" at alliedsignal.com  Tue Jun 22 18:29:53 1999
From: " HPeter.Friedrichs" at alliedsignal.com (H. P. Friedrichs)
Date: Tue, 22 Jun 1999 15:29:53 -0700
Subject: Newbie: How do you "clear" PythonWin between runs?
Message-ID: <37700E61.5388E101@alliedsignal.com>

I've just started playing with python, so all of this is very new to me.
I have been writing some scripts, saving them to *.py files, and running
them under PythonWin 1.5.2 . The process is iterative; I code, save,
run, code, save, run, etc.

What I have noticed is that python doesn't "reset" between runs.
Sometimes error  messages reappear even though I have fixed the problem,
because some artifact of a previous run lingers.

So, my question is this: How can I "flush" or "clear" the python
interpreter between successive runs? Is there a command I can add to my
script to initialize python before running the script?

Thanks,

Pete
hpeter.friedrichs at alliedsignal.com

--

-----------------------------------------------
| The opinions expressed here are my own,     |
| and not necessarily those of my employer.   |
-----------------------------------------------





From abrahams at mediaone.net  Sun Jun 27 13:44:59 1999
From: abrahams at mediaone.net (Dave Abrahams)
Date: Sun, 27 Jun 1999 13:44:59 -0400
Subject: Newbie: How do you "clear" PythonWin between runs?
References: <37700E61.5388E101@alliedsignal.com>
Message-ID: <tntd3.922$kL2.86197@ndnws01.ne.mediaone.net>

In article <37700E61.5388E101 at alliedsignal.com> , "H. P. Friedrichs" <" 
HPeter.Friedrichs"@alliedsignal.com> wrote:

> So, my question is this: How can I "flush" or "clear" the python
> interpreter between successive runs? Is there a command I can add to my
> script to initialize python before running the script?

I haven't tried it yet, but I've been thinking:

# when you start up the IDE
import sys
modules = sys.modules.copy()

# To "clear" the python interpreter
sys.modules = modules.copy()


Just might work,
Dave



From matomira at acm.org  Mon Jun 14 11:41:30 1999
From: matomira at acm.org (Fernando D. Mato Mira)
Date: Mon, 14 Jun 1999 17:41:30 +0200
Subject: <language> broken (was: OO)
References: <000a01beb5dc$5e88aa80$ae9e2299@tim>
Message-ID: <376522AA.9BEE7E5E@acm.org>

Tim Peters wrote:

> OK!  Now we know the answer Fernando is looking for:  Tcl is fatally broken
> because its syntax differs from Lisp's <0.9 wink>.

Response to 0.1 non-wink:

No!  I wanted to know if there were any theoretical reasons for
tcl-bashing (that some people bash it because it lacks data structures,
I can understand, but there the pragmatic evaluation is not the same
in different contexts).

Of all the languages I know that I mentioned, the only one I can say is BAD is
C++. And 200K lines plus a high personal cost entitle me to say it. In spite
of that, the STL taught me a lot about how to program
`ADT style'. However, I still don't know how useful that actually is when you
have a language with type inference. But what I _do_ learned from that once
you find out the proper way to write C++ is that static typing without
inference is BROKEN BROKEN BROKEN (you really wear out your keyboard).
Note that C++ could be fixed to use inference, but that won't save it
(overloading, casts, you name it).

--
Fernando D. Mato Mira
Real-Time SW Eng & Networking
Advanced Systems Engineering Division
CSEM
Jaquet-Droz 1                   email: matomira AT acm DOT org
CH-2007 Neuchatel                 tel:       +41 (32) 720-5157
Switzerland                       FAX:       +41 (32) 720-5720

www.csem.ch      www.vrai.com     ligwww.epfl.ch/matomira.html






From aprasad at magix.com.sg  Sun Jun 20 12:08:55 1999
From: aprasad at magix.com.sg (Ajith Prasad)
Date: Mon, 21 Jun 1999 00:08:55 +0800
Subject: DATE ARITHMETIC
References: <7kejrv$e57$1@violet.singnet.com.sg> <376C08D8.A9032304@callware.com>
Message-ID: <7kj3jv$ssu$1@coco.singnet.com.sg>

Noted and thanks for all the feedback.

The difficulty I had with the mxDateTime module was that results seem to be
expressed as date time objects with timestamps embedded with dates  and,
given my limited knowledge of Python (just a few chapters into "Learning
Python") I have no clue as to how to extract just the dates out. But I am
sure that mxDateTime has all the functionality that justifies further
investment in learning it.

In response to my posting, Tim Peters and Jeff Bauer have pointed out
simpler date-time routines that could be used. Tim's routines (available
together with the Python source code disribution) use the American-style
"month-day-year" format while Jeff's use the less ambiguous
"year-month-date" format which also happens to be the format in which my
date data are in. Jeff's notes at
http://starship.python.net/crew/jbauer/normaldate/ make some useful points
on the need for simple routines which handle the more common date arithmetic
tasks without the added complications of dealing with specific calendars,
time zones, etc. Both Tim's and Jeff's routines are geared towards this. I
will look at the further references that Ivan has suggested for a deeper
understanding of the issues.

Ivan Van Laningham <ivanlan at callware.com> wrote in message
news:376C08D8.A9032304 at callware.com...
> Hi All--
>
> Ajith Prasad wrote:
> >
> > I need to do some simple date arithmetic - finding the number of days
> > between any 2 days, the age of a person on a day given his birthday. The
> > mxDateTime module downloadable from the Python website is more complex
than
> > what I need. Any other suggestions?
>
> No.  Date arithmetic is by nature non-trivial.  Marc-Andre's mxDateTime
> is perfectly suited for your application.
>
> If you *really* *really* don't want to use it, though, look up "Julian
> Date" and Scaliger (Joseph Justus) on the web.  Learn how to reduce
> Gregorian dates to Julian Period dates, and once you do that you can add
> and subtract dates as if they have just become big numbers.
>
> Which they will have done, assuming you follow up on Julian Period
> dates.
>
> <8-K'an-12-Sotz'>-ly y'rs,
> Ivan
> ----------------------------------------------
> Ivan Van Laningham
> Callware Technologies, Inc.
>






From pharaoh at novagate.com  Fri Jun 11 15:14:56 1999
From: pharaoh at novagate.com (Douglas Nichols)
Date: 11 Jun 1999 19:14:56 GMT
Subject: gadfly install problem
Message-ID: <7jrn7g$n2f$0@205.138.136.179>

I am trying to install gadfly for use on my mac with python 1.5.1

The installation seems to have worked.  When I try to test gadfly by
creating a new database, I get the following error message:

Python 1.5.1 (#37, Apr 27 1998, 13:36:04)  [CW PPC w/GUSI w/MSL]
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import gadfly
>>> connection = gadfly.gadfly()
>>> connection.startup("testbase", "testfolder")
>>> cursor = connection.cursor()
>>> cursor.execute("create table ph (nm varchar, ph varchar)")
shift: no mem in addchild
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "Macintosh HD:Development:Python 1.5.1:Lib:lib-gadfly:gadfly.py",
line 320, in execute
    from sqlsem import Parse_Context
MemoryError
>>> 

Any suggestions?  I am very new to python and am still working my way
through the Learning Python book.

Thanks,
Doug Nichols
pharaoh at novagate.com

http://www.ottawahumane.org



From mark at chem.uwa.edu.au  Sun Jun  6 21:44:28 1999
From: mark at chem.uwa.edu.au (Mark C Favas)
Date: 7 Jun 99 01:44:28 GMT
Subject: intractable(?) Tkinter scrollbar problem
References: <19990601001054.36247@nms.otc.com.au> <x7yahxfvzy.fsf@lindm.dm>
Message-ID: <mark.928719868@declan>

Dieter Maurer <dieter at handshake.de> writes:

>Greg McFarlane <gregm at iname.com> writes on Mon, 31 May 1999 14:10:54 GMT:
>> I include a Tkinter script below, modified from one by Mark C Favas,
>> using, I believe, Fredrik Lundh's AutoScrollbar class.  When this
>> script is run, the two scrollbars continuously get mapped and unmapped
>> and the window continuously changes size.

>I think, I now understand the instability and I should have
>a solution.

...<good explanation elided>...

>class AutoScrollbar(Tkinter.Scrollbar): 
>  _active= 0
>  def set(self, lo, hi):
>    self.lo= lo; self.hi= hi
>    #if self._active: print 'active'; return
>    self._active= 1; self.update_idletasks(); self._active= 0
>    self._set()
>  #
>  def _set(self):
>    lo= self.lo; hi= self.hi
>    #print self.id, lo, hi
>    if float(lo) <= 0.0 and float(hi) >= 1.0:
>      self.tk.call("grid", "remove", self)
>    else:
>      self.grid()
>    Tkinter.Scrollbar.set(self, lo, hi)

Thanks for your interesting explanation, Dieter. Interestingly, your fix as
posted, works, even though the test for self._active is commented out (it also
works if the test is uncommented). So, I tried the following, which also
works. The key is saving and restoring the values of lo and hi around the call
to self.update_idletasks()...

class AutoScrollbar(Tkinter.Scrollbar):
    def set(self, lo, hi):
        self.lo = lo
        self.hi = hi
        self.update_idletasks()
        lo = self.lo
        hi = self.hi
        if float(lo) <= 0.0 and float(hi) >= 1.0:
            self.tk.call("grid", "remove", self)
        else:
            self.grid()
        Tkinter.Scrollbar.set(self, lo, hi)

--
Email  - mark at chem.uwa.edu.au      ,-_|\                           Mark C Favas
Phone  - +61 9 380 3482           /     \               Department of Chemistry
Fax    - +61 9 380 1005      ---> *_,-._/   The University of Western Australia
                                       v                               Nedlands
Loc    - 31.97 S, 115.81 E                               Western Australia 6009



From fredrik at pythonware.com  Thu Jun  3 14:39:57 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 3 Jun 1999 18:39:57 GMT
Subject: [tkinter] two questions regarding Canvas
References: <3752a555@nntp.server.uni-frankfurt.de> <37556602.415AFA46@ingr.com> <19990604005922.07590@nms.otc.com.au>
Message-ID: <02fb01beadf0$b358f360$f29b12c2@pythonware.com>

Greg McFarlane <gregm at iname.com> wrote:
> This is small attempt at adding editing bindings to canvas text items. 
> This *must* have been done before, but a search of comp.lang.tcl did
> not come up with anything, except a mention of pg_access in postgresql
> which I could not find.  Does anyone know of it?

http://www.pythonware.com/people/fredrik/fyi/fyi52.htm

</F>




From greg.ewing at compaq.com  Wed Jun 23 20:04:16 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Thu, 24 Jun 1999 12:04:16 +1200
Subject: Windows - Drag & drop on Python script?
Message-ID: <37717600.D076BF10@compaq.com>

How can I create an icon onto which I can drop a file,
and have it run a Python script with the dropped file
as an argument?

I've tried myriad combinations of .bat files, shortcuts,
etc., but nothing seems to work.

The obvious thing would seem to be to use a .bat
file, but the dropped file seems to end up as the
standard input or something, rather than getting
passed as an argument.

There *must* be a way of doing this, surely?

Thanks for any help,
Greg



From mkersey at metricom.com  Fri Jun 11 21:49:22 1999
From: mkersey at metricom.com (Michael Kersey)
Date: Fri, 11 Jun 1999 20:49:22 -0500
Subject: Any Production ZOPE sites we can visit?
References: <37614A32.79FE554E@metricom.com> <7jrou7$p42$1@news.clarkson.edu>
Message-ID: <3761BCA2.29AC4F2E@metricom.com>

It may not be exciting to you, but to me this site is an eye-opener!
Bravo to it's developers.

Especially nice depiction/presentation of regions versus counties and
the handy use of mapping/graphics capabilities throughout. Your ties
into http://tiger.census.gov/ are extremely sharp and well-done, IMO.

When you say "airport lookups", what do you mean? Is part of the search
done with Zope per se, and another part passed to an external database? 
For instance AFAI can tell, the URL
http://www.air.gen.ny.us/Airports/AirportInfo?identifier=NY54
may be (I'm guessing) handled by Zope's own object-oriented database and
DTML, while the URL
http://www.airnav.com/cgi-bin/airport-info?N66
may be handled by significantly different code (not DTML) executing SQL
queries? 

BTW, having worked with it for years, I have a hard time finding
anything exciting about SQL!-))
TIA,
Michael Kersey

Brad Clements wrote:
> http://www.air.gen.ny.us
> Nothing exciting there really, except that the airport lookups are done
> through an SQL query..
> I'm working on an events Calendar, that'll really look nice when it's done
<snipped>



From claird at Starbase.NeoSoft.COM  Wed Jun 16 21:21:31 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 16 Jun 1999 20:21:31 -0500
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <37628EBF.207F7A8F@iname.com> <rnbn1y2n5jk.fsf@nortelnetworks.com> <sFU93.1567$kS.233449@ptah.visi.com>
Message-ID: <7k9iir$8uq$1@Starbase.NeoSoft.COM>

In article <sFU93.1567$kS.233449 at ptah.visi.com>,
David Thornley <thornley at visi.com> wrote:
			.
			.
			.
>I know little of this Qt animal.  Is it free?  Is it open source?
>Does it run on the Macintosh?  Is it easy to set up an interface with?
			.
			.
			.
Qt comes in a handful of flavors.  Some are free of charge.
There's an open-sourced work-alike.  None of these run under
MacOS, to the best of my knowledge.  I'm collecting information
about them at <URL:
http://starbase.neosoft.com/~claird/comp.windows.misc/Qt.html>
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From ivanlan at callware.com  Mon Jun 14 10:49:14 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Mon, 14 Jun 1999 14:49:14 GMT
Subject: Broken 1.5.2 installation? Missing cw3215mt.dll
References: <1282960807-85602620@hypernet.com>
Message-ID: <3765166A.D4351063@callware.com>

Hi All--

Gordon ``Cannibal'' McMillan wrote:
> 
> Ivan Van Laningham wrote:
> 
> > ...  I've also installed Python/Tcl on my three Linux
> > machines--RH5.2 and RH6.0-SMP--but I didn't have to fiddle with
> > anything at all in order to get IDLE to run just fine.  Of course,
> > Python and Tcl etc. have been in my path permanently for a long time
> > on Linux;-)
> 
> [double snort]
> 
> A few months ago you were running some auto-port from a PDP-8. Did
> RH finally convince you no warm-blooded animals were used in user
> testing?
> 

The closest I ever got to a PDP-8 was a MicroVax.  Disgusting box.  As
my mother-in-law says, ``Oh wonderful.  Everything I hate most in one
sandwich!''

And I understood that RH only uses poikilothermic cannibals in their
testing.

<that-explains-the-electrodes-in-gordon's-head>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From cgw at fnal.gov  Wed Jun 30 14:52:15 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Wed, 30 Jun 1999 18:52:15 GMT
Subject: socket timeouts
In-Reply-To: <XFMail.990630103816.nathan@islanddata.com>
References: <XFMail.990630103816.nathan@islanddata.com>
Message-ID: <14202.26463.616657.95654@buffalo.fnal.gov>

Nathan Clegg writes:
 > Is there any way to set the timeout for a TCP/IP connection? 
 > Specifically, I'm writing some apps meant to run on the same machine, so
 > if they can't connect *immediately* (i.e. the server app is not running), I
 > need to go to plan B. 

What type of platform are you running on?  On my Linux 2.2.10 box, a
"connect" to a port on the local host with no server app fails
immediately with a "connection refused".




From fredrik at pythonware.com  Wed Jun 30 04:09:30 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 30 Jun 1999 08:09:30 GMT
Subject: PySequence_Check
References: <slrn7nhjv9.lt6.rasumner@bach.wisdom.weizmann.ac.il>
Message-ID: <00f501bec2cf$e22f1db0$f29b12c2@secret.pythonware.com>

Reuben Sumner wrote:
> How can I do the equivalent of a PySequence_Check in Python?
> GvR told me that the question was illdefined (if I explained it correctly)
> and to ask here.  It seems that some of the functions in abstract.h
> are available from python (len(), callable()) and others are not
> (PySequence_Check, PyNumber_Check).

import operator
operator.isSequenceType
operator.isNumberType
operator.isMappingType

not documented, and should probably be considered as
deprecated (see replies from Paul and Tim for some back-
ground).

> What am I missing?  I have a nested structure of lists and tuple and I
> just want to know when I have bottomed out.  I can check for tuple or
> list indivually but that is ugly and inflexible.

> I could just try indexing and see if there was an exception but
> that is not much better.

given that Python has no support for declared inter-
faces, that's a reasonable approach.  works just fine,
and adds no extra overhead.

on the other hand, it doesn't work if you have strings
in there.  but that's true for most other approaches
as well:

>>> operator.isSequenceType("spam")
1
>>> operator.isSequenceType("spam"[0])
1
>>> operator.isSequenceType("spam"[0][0])
1
>>> operator.isSequenceType("spam"[0][0][0])
1
>>> operator.isSequenceType("spam"[0][0][0][0])
1
>>> operator.isSequenceType("spam"[0][0][0][0][0])
1

</F>




From Andrew_Tanner at mw.3com.com  Wed Jun 23 12:07:12 1999
From: Andrew_Tanner at mw.3com.com (Andrew Tanner)
Date: Wed, 23 Jun 1999 11:07:12 -0500
Subject: Freeze size Question
Message-ID: <86256799.0058B93D.00@mwgate02.mw.3com.com>



When I compiled hello.py using freeze.py on my Solaris 2.6 machine it came out
being 1.5 megs! I then used strip to remove all of the symbols and it came to
about 511k which seems a bit more reasonable. Doesn't 1 megabyte of symbol
information seems a bit excessive? And compiling with out the '-g' switch only
cuts around 50k off the binary.

I'm trying to make the smallest, tightest, easiest binary to distribute, is
there anyway I can poke or prod freeze.py to succumb to my wishes?

Thanks,
Andrew.





From larsga at ifi.uio.no  Tue Jun 29 13:02:42 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 29 Jun 1999 19:02:42 +0200
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <lw67471fyi.fsf@copernico.parades.rm.cnr.it> <7laoq0$bvh$1@Starbase.NeoSoft.COM> <lwso7bdnia.fsf@copernico.parades.rm.cnr.it> <7laqeb$e17$1@Starbase.NeoSoft.COM>
Message-ID: <wkyah35499.fsf@ifi.uio.no>

* Cameron Laird
|
| As it happens, there's very good work going on just now to beef up
| Tcl's introspective capabilities.  Is *that*-- introspection--the
| real content of your preference?

I don't know exactly what Marco had in mind, but this is certainly not
the main value of the Common Lisp macro facility. The main value lies
in being able to extend the language seamlessly with no run-time costs.

Things like the while loop (which does not exist in CL), the Common
Lisp Object System and programming by contract features can be built
on top of the Lisp core using nothing but CL primitives. Personally, I
dream of finding the time to toy with Aspect-Oriented programming in
CL through extension macros.

I know tcl has a similar capability, but I'm unsure whether it is of
comparable power and convenience. 

[From your first posting in this sub-thread]

| Are you making a semantic point, or arguing on the basis of
| implementation-specific performance?

Is there more than one implementation of tcl? If so, does any of them
do macro expansions at compile-time? If not, I think arguments based
on performance also carry considerable weight.

I doubt that you would disagree that C++ has better performance than
does tcl. Given than CL achieves comparable performance to C++ a
conclusion seems near at hand. :) 

(BTW, I don't know if you're aware of this, but Dylan, which you wrote
an excellent article on[1] is essentially Common Lisp bereft of one of
its major features, the S-expression syntax.)

| Incidentally, many of the same ideas and possibilities are available
| to Python, although I don't know of anyone actively pursuing them
| for Python.

Actually, macros, performance and S-expressions are among the things I
miss the most in Python.  bytecodehacks, although certainly cool, are
just a pale shadow of what CL macros provide.

As for introspection, Common Lisp has good support for that as well,
although I very much doubt that that was Marco's point.

| [...] I occasionally argue that 'twould be worth the effort to do
| introspection for Python in a more unified way.

'twould indeed, but I find other issues more pressing.

--Lars M.

[1] <URL: http://www.sunworld.com/swol-03-1999/swol-03-regex.html >



From claird at Starbase.NeoSoft.COM  Wed Jun  2 07:17:01 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 2 Jun 1999 06:17:01 -0500
Subject: A tkinter kick-off
References: <37544a32.0@news1.mcmail.com>
Message-ID: <7j33re$ip0$1@Starbase.NeoSoft.COM>

In article <37544a32.0 at news1.mcmail.com>,
Mark Butterworth <mrkbutty at mcmail.com> wrote:
>I've been using python for a few months now (I've actually managed to
>persuade my company that freeware is a viable tool) and I've just started
>looking at tkinter.  However,  my first little project doesn't  require an
>interactive window or dialog box.  What I do require is a window which
>displays the result's of my background processing.  The processing will
>simply be the compression and network transfer of some files.  I'm not
>looking for a full solution however, I just need a help starting off.  I can
>create a window containing a listbox alright, I just need to know how to
>process in the background and display results within the listbox (maybe a
>cancel button would also be useful).
			.
			.
			.
<URL:http://
Starbase.NeoSoft.COM/~claird/comp.lang.tcl/tcl-examples.html#fileevent>
hints at what you're after.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From l11403 at alunos.uevora.pt  Tue Jun 22 12:45:19 1999
From: l11403 at alunos.uevora.pt (Jose Saias)
Date: Tue, 22 Jun 1999 17:45:19 +0100
Subject: Python scope and parameter passing method
Message-ID: <Pine.LNX.3.96.990622173938.8191A-100000@alunos.uevora.pt>

Hy 


I'm a portuguese student and I am comparing Python to Objective C. Can
some one help me in this?


I would like to ask you about Python:

- the parameter-passing methods (call-by name/value/reference);
- dynamic memory control (garbage collection?, allocation for activation
records);
- the scope of a name (lexical/dynamic);
- innermost or outermost evaluation;
- a small example program that explores the language possibilities.


I appreciate your help!

Jose Saias




From donn at u.washington.edu  Fri Jun 11 18:20:14 1999
From: donn at u.washington.edu (Donn Cave)
Date: 11 Jun 1999 22:20:14 GMT
Subject: ^C and os.system()
References: <lczp2661a9.fsf@twitch.mit.edu>
Message-ID: <7js22u$r7e$1@nntp6.u.washington.edu>

Douglas Alan <nessus at mit.edu> writes:
| Greetings.  Is it considered a bug or a feature that if I type ^C at a
| Python program that is in the middle of executing a call to
| os.system(), the Python program never seems to see the SIGINT signal?
| Instead, only the process created by os.system() receives the signal.
| Consequently, if I have a Python script that executes a program
| repeatedly for a list of files, for instance, I can't kill the Python
| script with ^C.

It's neither bug nor feature, it's just the implementation of the
C system() function.  In the NetBSD source, I see system() ignores
SIGINT and SIGQUIT in the parent fork while calling waitpid(), and
I imagine Solaris 2.6 may do something similar.  Can't explain why.

| Also, when I have the Python program check the return status of a
| command invoked via os.system() and subsequently killed with ^C,
| instead of getting back 0x0002, as would be expected from a process
| killed by SIGINT, I usually get back 0x8200.  Sometimes I get back
| 0xd000.  Only infrequently do I get back the 0x0002 that I would
| expect.

Now that looks like a bug in the Solaris 2.6 system().  I don't get
this effect on Digital UNIX, for example.  You might try a little
C program, it should look the same as Python.

| None of these problems occur if I use commands.getstatusoutput()
| instead of os.system(), so I guess I will stick with getstatusoutput()
| for the time being.

Uses popen(), which doesn't have the same signal handling.  You
can also roll your own system() with fork() and exec*().

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



From WESTER at ilt.fhg.de  Wed Jun  2 04:37:27 1999
From: WESTER at ilt.fhg.de (Rolf Wester)
Date: Wed, 02 Jun 1999 10:37:27 +0200
Subject: Posix: Exec format error
Message-ID: <3754ED47.54C2F0AF@ilt.fhg.de>

Hi,

when doing the following:

import os
id=os.fork()
if id == 0:
    os.execl("/home/wester/test/test","-d", "indata")
else:
    os.wait()

I get the error message:

OSError: [Errno 8] Exec format error

Can aybody tell me how the format should look like?

Thanks in advance

Rolf






From befletch at my-deja.com  Sat Jun 26 01:53:53 1999
From: befletch at my-deja.com (Bruce Fletcher)
Date: Sat, 26 Jun 1999 05:53:53 GMT
Subject: Python socket performance testing
Message-ID: <7l1ptf$ov9$1@nnrp1.deja.com>

I have Python 1.5.1 on a Mac (G3/233) and a Linux box (Celeron/366).
These are connected by 10BT ethernet, and I get good speed with normal
HTTP and FTP clients/servers.  Curiously, I can't seem to get good speed
(yet) in transfering data from Python to Python on the two machines.

Here's some excerpts from my client/server code:

#client: (Mac)
	data = 'a' * 1024

	sock = socket( AF_INET, SOCK_STREAM )
	sock.connect( SERVER, DATA_PORT )

	loops = 1024
	t1 = time.clock()
	while loops > 0:
		sock.send( data )
		loops = loops - 1
	reply = sock.recv(128)
	t2 = time.clock()
	sock.close()

	print 'Test time:', t2 - t1, 'seconds.'

#server: (Linux)
	sock = socket( AF_INET, SOCK_STREAM )
	sock.bind( '', DATA_PORT )
	sock.listen(1)

	conn, addr = sock.accept()

	loops = 1024

	while loops > 0:
		data = conn.recv(1024)
		loops = loops - 1

	conn.send('ok')
	conn.close()

This 1 meg transfer takes 73 seconds!  That works out to 14 kb/sec or so,
which isn't exactly blinding.  Am I doing something silly in my test?  I
suppose I should go test this just on the Linux box, just on the Mac,
etc. but I'm not too confident in my Python code yet so I thought I'd ask
for pointers.

Something else interesting; time.clock() doesn't seem to work on Linux
(RedHat 6.0) in either 1.5.1 or 1.5.2.  The time is always < 1, and it
doesn't change.

Thanks,
- Bruce


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From holger at phoenix-edv.netzservice.de  Wed Jun  9 08:27:58 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Wed, 9 Jun 1999 12:27:58 GMT
Subject: os.mkdir()
Message-ID: <375E5DCE.7005D71A@phoenix-edv.netzservice.de>

Hello again,

I couldn't find a resolution in helpfiles, so please
give a hint to my (easy to solve!?) question:

I'd like to create directories under WinNT. There is 
already a file with the same name as the directory I want
to create.

Look:
>>>os.mkdir("c:\\blabla")
OSError: [Errno 17] File exists: 'C:\\blabla'

or
>>>os.makedirs("c:\blabla\nana\gugu")
produces 
OSError: [Errno 2] No such file or directory: 'c:\blabla\nana'

Don't know if python doing the same under linux, but I need
to solve the problem for every OS!

I think, there must be an easy solution?!

Ciao,
Holger



From cmedcoff at my-deja.com  Sat Jun  5 19:57:34 1999
From: cmedcoff at my-deja.com (cmedcoff at my-deja.com)
Date: Sat, 05 Jun 1999 23:57:34 GMT
Subject: Looking for for some ideas for argument processing
Message-ID: <7jcdhc$htd$1@nnrp1.deja.com>

I'm writing a script with is going to make heavy use of script
arguments.  I'd like to hear some ideas on some ways to support this.

One idea I've had so far is to write a class which is a dictionary
of "directives" to function objects, where a "directive" might be '-
c', '/f', or whatever.  This seems to work well if all arguments are of
this simple form.  However I would also like to support arguments
like "python myscript.py /a /f /x "bip" "bop", /g"; the point here
being that "directives" may take arguments.  This kind of messes up the
simple loop you'd otherwise use to process arguments.

        a = BuildArgumentProcessor()

        # process arguments
        for x in sys.argv[1:]:
            opt = x[1:2]
            optArg = x[2:]
            a.Process(opt, optArg)

This would work if somehow inside the for loop one could "increment
this list iterator more that once" (sorry I'm coming from a C++
background).  I don't see anyway of doing this.  Is it possible?  Any
other ideas?

Regards,
Chuck


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From hs at inferno.nirvananet  Fri Jun 11 18:52:02 1999
From: hs at inferno.nirvananet (Hartmann Schaffer)
Date: Fri, 11 Jun 1999 22:52:02 GMT
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <wk909syqr4.fsf@ifi.uio.no>
    <375FF969.C879B4CD@iname.com> <37601668.D9105225@mayo.edu>
    <3760E130.7456447@iname.com>
    <3760FA1F.7BCD@mailserver.hursley.ibm.com>
Message-ID: <mqg83.23803$%65.45559@tor-nn1.netcom.ca>

In article <3760FA1F.7BCD at mailserver.hursley.ibm.com>,
	Paul Duffin <pduffin at mailserver.hursley.ibm.com> writes:
> ...
> "Call by name" has some very particular characteristics the main one
> being that you cannot implement "swap". I can assure you that you can

This is news to me.  Afair, "call-by-name" was the term for the Algol60
style parameter mechanism, which essentially required each argument
passed by a parameterless function returning a pointer to the passed
value.  Parameter swapping  (and a few pretty usually pointless tricks
like "Jensen's device"(?)) were quite possible.

> ...

Hartmann Schaffer




From andermoe at online.no  Wed Jun 16 15:59:54 1999
From: andermoe at online.no (Anders Moe)
Date: Wed, 16 Jun 1999 21:59:54 +0200
Subject: Raw Strings!
References: <P3v83.7006$Yb7.1482@news4.mia>
Message-ID: <37680239.DFDBB6B1@online.no>


SunitJoshi wrote:

> Hi All
>         If someone could show me how to use raw-strings with variables, I
> would really appreciate it.
>         For e.g.: if str_path = "C:\\windows\\temp"
>
> thanks
> Sunit

Since I suspect you have path-strings like the above in mind, may I suggest
simply
using forward slashes - they will work on win98. Else I recommend using
'os.sep'.

Anders Moe




From perry at ac-copy.net  Tue Jun  8 02:44:25 1999
From: perry at ac-copy.net (Joachim Schmitz)
Date: Tue, 8 Jun 1999 06:44:25 GMT
Subject: mxODBC and MySQL
In-Reply-To: <375C35A0.7754C3BD@lemburg.com>
References: <375C35A0.7754C3BD@lemburg.com>
Message-ID: <Pine.LNX.4.05.9906080840190.12801-100000@ns.ac-copy.net>

I ugraded to 1.5.2 now I get:

Python 1.5.2 (#1, Jun  8 1999, 07:26:20)  [GCC egcs-2.91.66 19990314
(egcs-1.1.2  on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import ODBC.MySQL
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/local/lib/python1.5/ODBC/MySQL/__init__.py", line 7, in ?
    from mxODBC import *
ImportError: /usr/local/lib/python1.5/ODBC/MySQL/mxODBC.so: undefined
symbol: SQLTables

regards


Joachim Schmitz


On Mon, 7 Jun 1999, M.-A. Lemburg wrote:

> Joachim Schmitz wrote:
> > 
> > Hi,
> > 
> > I compiled and installed  mxODBC for MySQL on SuSE 6.1 without problems.
> > I get this:
> > Python 1.5.2b2 (#6, Feb 28 1999, 21:41:19)  [GCC egcs-2.91.60 19981201
> > (egcs-1.1
> > .1  on linux2
> > Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> > >>> import ODBC.MySQL
> > Traceback (innermost last):
> >   File "<stdin>", line 1, in ?
> >   File "/usr/local/lib/python1.5/ODBC/MySQL/__init__.py", line 7, in ?
> >     from mxODBC import *
> > ImportError: /usr/local/lib/python1.5/ODBC/MySQL/mxODBC.so: undefined
> > symbol: PyBuffer_Type
> > >>
> 
> You can either upgrade to 1.5.2 final or turn off the buffer
> type code in mxODBC.c manually (enclosed in HAVE_PYTHON_BUFFEROBJECTS
> #defines). I guess you stumbled into test code that I forgot to remove
> for the distribution; sorry.
> 

------------------------------------------------------------------------------
 P R O F I . N E T  A G    *    Joachim Schmitz   * Senior Network Consultant
------------------------------------------------------------------------------
  profi.net                |    Aindlinger Str. 7   | Voice  +49 821 27062-0
  Hassler & Mair AG        |    86167 Augsburg      | Fax    +49 821 27062-99
                           |    Germany             | Email  js at profi.net  




From aahz at netcom.com  Tue Jun  8 18:43:39 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Tue, 8 Jun 1999 22:43:39 GMT
Subject: dict vs kjBuckets vs ???
Message-ID: <aahzFD15sr.3oD@netcom.com>

I'm currently using a dict for storing references to objects (call them
customer objects, referenced by customer ID (a number)).  I'm using a
dict (instead of SQL or something similar) because we're willing to
trade lots of memory to get speed.  However, I've heard that after about
10K items in a dict, it starts having problems.  Is this a specious
rumor, should I switch to kjBuckets, or should I do something else?
Does the answer change if we have 100K objects?  A million?

(The only operations are object insert, update, delete, and lookup.)
-- 
                      --- Aahz (@netcom.com)

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

"I'm not tense, i'm just terribly, terribly alert"  -- unknown



From larsga at ifi.uio.no  Wed Jun 16 13:50:36 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 16 Jun 1999 19:50:36 +0200
Subject: GC In Python: YAS (Yet Another Summary)
References: <1282571877-108998356@hypernet.com> <zoQ93.198$U6.1646@newsr2.twcny.rr.com>
Message-ID: <wku2s86nmb.fsf@ifi.uio.no>

* news at dorb.com
| 
| At the risk of sounding like a lazy whiner. This would be more
| useful if it could run more like profile or pdb.

Hmmm. This sounds like a useful alternative to the current mode of
operation. 

| Having to add calls like "r.add_object(non_GCed_object)" make it
| less useful.
 
For some uses, perhaps. I can probably still think of cases where the
original behaviour is wanted, but the idea is a good one. The reason I
didn't do it was simply that I wasn't sufficiently ambitious. :)

| Is there a way to capture the creation of any instance and call
| r.add_object?

A little experimentation lead to a trace function that worked. And
since it was so cool I just went ahead and added it to plumbo.
 
| In perfect world for lazy whiners this works, plumbo.run("myStuff")
| and it gives me a report.

This may be a perfect world for lazy whiners after all. :)

Check the web site again for a function that has a lazy mode. It
doesn't print anything unless there actually is a cycle, but other
than that it should make you happy.

--Lars M.



From gerrit.holl at pobox.com  Tue Jun 22 09:44:05 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Tue, 22 Jun 1999 15:44:05 +0200
Subject: Check for valid date
In-Reply-To: <376F7C8A.A809C0A@zxa.basf-ag.de>; from Paolo G. Cantore on Tue, Jun 22, 1999 at 02:07:38PM +0200
References: <7klq51$etp$1@nnrp1.deja.com> <376F7C8A.A809C0A@zxa.basf-ag.de>
Message-ID: <19990622154405.B19393@optiplex.palga.uucp>

On Tue, Jun 22, 1999 at 02:07:38PM +0200, Paolo G. Cantore wrote:
> From: "Paolo G. Cantore" <paolo.cantore at zxa.basf-ag.de>
> Newsgroups: comp.lang.python
> Subject: Re: Check for valid date
> Date: 	Tue, 22 Jun 1999 14:07:38 +0200
> To: python-list at python.org
> 
> Hi Remi,
> 
> I use a two-step procedure:  
> 
> >>>def check_date(year, month, day):
> ...	tup1 = (year, month, day, 0,0,0,0,0,0)
> ...  	try:
> ...    		date = time.mktime (tup1)
> ...    		tup2 = time.localtime (date)
> ...		if tup1[:2] != tup2[:2]:
> ...			return "Date not valid."
> ...		else:
> ...			return "Date valid."
> ...	except OverflowError:
> ...		return "Date not valid."
> ...
> >>> import time
> >>> print date_valid(1999,2,29)
> Date not valid.
> >>> print date_valid(1999,2,28)
> Date valid.
> 

I would return 0 or 1, instead of what you're doing.

groeten,
Gerrit.

-- 
The Dutch Linuxgames homepage:	http://linuxgames.nl.linux.org
Personal homepage:		http://www.nl.linux.org/~gerrit/

Discoverb is a python program in Dutch and English which tests the words you
learned by asking it. Homepage: http://www.nl.linux.org/~gerrit/discoverb/



From graham at sloth.math.uga.edu  Thu Jun  3 14:53:28 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 3 Jun 1999 18:53:28 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <87n1ykr0z9.fsf@ev.netlab.co.jp> <7ivs9q$dc5@news.dns.microsoft.com> <87k8toqv70.fsf@ev.netlab.co.jp> <aahzFCnx28.1Jp@netcom.com> <7j44ql$908$1@cronkite.cc.uga.edu> <7j5rn5$dvs$1@nnrp1.deja.com>
Message-ID: <7j6iv8$58f$1@cronkite.cc.uga.edu>

adjih at technologist.com wrote:
: Hmm... How do you do multi-thread real-time (i.e. incremental) GC being
: a) portable (must run on PalmPilot, WinCE, Unix, Mac,...) b) without
: relying on assumptions [that will fail in some programs] (such as "very
: few writes in shared space"). The supplementary condition would be "c)
: efficient" which would be useful for languages such as C/C++ or compiled
: Java, but I grant you that this is not necessary for the current Python,
: and that we are comparing here to reference counting anyway.

I am no expert on real time garbage collection, but I have read a few
papers which clearly show that it can be done. Moreover done quite
portably and efficiently (indeed if it were not efficient, it can't
be *real-time* can it). I am just suggesting you read such papers
before you right off the whole idea.

graham
-- 
          The fool escaped from paradise
        will look over his shoulder and cry
          Sit and chew on daffodils and
             struggle to answer why?



From mrkbutty at mcmail.com  Wed Jun  2 19:58:28 1999
From: mrkbutty at mcmail.com (Mark Butterworth)
Date: Thu, 3 Jun 1999 00:58:28 +0100
Subject: A tkinter kick-off
References: <37544a32.0@news1.mcmail.com>
Message-ID: <3755c6e1.0@news1.mcmail.com>

First of all thanks for the help.  The pointers to code everyone sent
certainly helped me, and the pipe capturing code will come in handy for
something else I've in mind.  However what I was after was something quick
and dirty (and simple) not to capture command output.  Please find my test
code below, once I span out the process function (it in turn will call many
more functions) it should be Ok.  Anyone got any suggestions for adding a
"Finished" message dialogue at the end and a method of keeping the listbox
scrolled to the end of the list would be appreciated (and save me hunting
through limited documentation).  Any pointers to decent documentation would
come in handy,  the tkinter tutorials are fine but of limited use as
reference.


from Tkinter import *
from tkmessagebox import *
import tkMessageBox,threading

class Logscreen:

    def __init__(self, master):

        frame = Frame(master)
        self.status = StringVar()
        self.status.set("My Test status")
        Label(frame, textvariable=self.status).pack()
        scrollbar=Scrollbar(frame, orient=VERTICAL)
        self.listbox=Listbox(frame, yscrollcommand=scrollbar.set)
        scrollbar.config(command=self.listbox.yview)
        scrollbar.pack(side=RIGHT, fill=Y)
        self.listbox.pack(side=LEFT, fill=BOTH, expand=1)
        frame.pack()

def process(root, listbox, status):
    for x in range(1,500):
        listbox.insert(END,'x is set to %d' % x)
        status.set('x is set to %d' % x)

    status.set('finished')
    root.destroy()

root = Tk()
root.title('My first Tk test')
root.resizable(1,0)
logscreen = Logscreen(root)
t=threading.Thread(target=process, args=(root,
logscreen.listbox,logscreen.status,))
t.start()
root.mainloop()


--

Regards, ]\/[r]{











From timo at bbn.com  Wed Jun  2 14:24:56 1999
From: timo at bbn.com (Tim O'Malley)
Date: Wed, 2 Jun 1999 18:24:56 GMT
Subject: Package for Solaris
In-Reply-To: <3755494B.61C863B2@InfoMation.com>
References: <3755494B.61C863B2@InfoMation.com>
Message-ID: <199906021824.OAA29791@ghost.bbn.com>

hola.

Aron K. Insinga writes:
 > I built 1.5.2 from the sources, but is there a Solaris "package" for it?

http://sunfreeware.com/

-- 
.....................................................................
.   happy daze            .   Age and treachery will beat           .
.        -tim O           .       youth and skill every time.       .
.....................................................................



From akuchlin at mems-exchange.org  Thu Jun 10 10:11:48 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: Thu, 10 Jun 1999 10:11:48 -0400 (EDT)
Subject: Grimoire [was: How do I defect?]
In-Reply-To: <Pine.WNT.4.05.9906092053040.120-100000@david.ski.org>
References: <matt-0906992046290001@scotch.mondoinfo.com>
	<Pine.WNT.4.05.9906092053040.120-100000@david.ski.org>
Message-ID: <14175.51108.398189.219804@amarok.cnri.reston.va.us>

David Ascher writes:
>On Wed, 9 Jun 1999, Matthew Dixon Cowles wrote:
>The tutorial is what I recommend you get started with. However, I'd like
>to also put a little plug for a new document I discovered by accident
>(well, surfing =). Andrew Kuchling is too modest to advertise it, ...

	I don't advertise it because it's nowhere near finished; I
wanted to write a chapter on classes first, because that's the major
missing piece.  It's on my Web pages for the odd person who comes
across it -- some nice people have even sent me corrections already.  

       However, since it's been outed... the Grimoire is a
task-oriented manual, structured along the same lines as the Perl
Cookbook from O'Reilly -- each chapter covers a topic, starting with a
general introduction, and followed by sections that explain various
common questions (derived from FAQs, newsgroup discussions, the tutor
list, etc.)

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
In cyberspace everyone will be anonymous for 15 minutes.
    -- Graham Greenleaf




From holger at phoenix-edv.netzservice.de  Mon Jun 14 06:49:37 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Mon, 14 Jun 1999 10:49:37 GMT
Subject: force sequenz with only one item?
References: <1282887498-90012442@hypernet.com>
Message-ID: <3764DE41.9551EED1@phoenix-edv.netzservice.de>

Thanx again, Gordon!

mg,
Holger



From jlj at cfdrc.com  Wed Jun 30 10:07:51 1999
From: jlj at cfdrc.com (Lyle Johnson)
Date: Wed, 30 Jun 1999 09:07:51 -0500
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773c3a0.102825445@news.oh.verio.com> <xPPc3.535$or1.22866@typhoon-sf.snfc21.pbi.net> <xJRc3.292$Of3.67094@news.shore.net> <7laovo$c4s$1@Starbase.NeoSoft.COM>
Message-ID: <377A24B7.AD6FAA1A@cfdrc.com>

Just to add fuel to the fire, we now have a Python interface to FOX, a
cross-platform GUI toolkit:

    http://www.cfdrc.com/pub/FOX/fox.html

The latest version of the Python wrappers can be picked up at:

    ftp://ftp.cfdrc.com/pub/FOX/FXPy-0.99.42.tar.gz

I don't know how the speed of FXPy would compare with some of the others
mentioned (Tk, wxWindows, Qt, GTK, ...) but it would be worthwhile for
some independent opinions.

Lyle

Cameron Laird wrote:

> In article <xJRc3.292$Of3.67094 at news.shore.net>,
> Michael P. Reilly <arcege at shore.net> wrote:
>                         .
>                         .
>                         .
> >There are other graphical packages for Python out there too, it would
> >be appropriate to contrast them as well.
>                         .
>                         .
>                         .
> I occasionally work on such a comparison, currently available at <URL:
> http://starbase.neosoft.com/~claird/comp.lang.python/python_GUI.html>.
> I welcome updates from readers.
> --
>
> Cameron Laird           http://starbase.neosoft.com/~claird/home.html
> claird at NeoSoft.com      +1 281 996 8546 FAX




From greg.ewing at compaq.com  Wed Jun 23 19:55:35 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Thu, 24 Jun 1999 11:55:35 +1200
Subject: Why exceptions shouldn't be used for flow control [Re: YAS to the 
 "Reading line-by-line" Problem]
References: <Pine.SUN.3.95-heb-2.07.990622183356.3454A-100000@sunset.ma.huji.ac.il> <376FDF70.4756AD38@mindspring.com> <slrn7n0chl.gh6.wtanksle@dolphin.openprojects.net>
Message-ID: <377173F7.36BCFF6B@compaq.com>

William Tanksley wrote:
> 
> Both should be handled by exception, not by returning a special
> value.

I don't agree with that. Reaching the end of a file while
reading it is not exceptional enough to warrant requiring
the use of an exception handler to catch it.

What bothers me about using exceptions for flow control
in situations like this is that the effect of an exception
handler is *non-local*, whereas what you're trying to
catch is really only a local concern.

When you write something like

  while 1:
    try:
      line = read_me_a_line()
    except EOFError:
      break

what you really mean by the try-except is "if THIS
PARTICULAR reading operation reaches the end of
the file". But that's not what it does - it catches
any EOFError raised by any operation invoked by
read_me_a_line() that wasn't caught by something
else first.

In my experience, that sort of behaviour tends to
mask bugs. I prefer results which are not errors to
be returned normally, so that I can be sure where
they came from.

Greg



From bwarsaw at cnri.reston.va.us  Tue Jun 15 17:22:18 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Tue, 15 Jun 1999 17:22:18 -0400 (EDT)
Subject: easy filecopy? - JPython?
References: <PBt93.843$B04.2678@newsfeed.slurp.net>
	<1282653264-104102639@hypernet.com>
	<14182.35687.906156.126592@weyr.cnri.reston.va.us>
Message-ID: <14182.50186.228957.979438@anthem.cnri.reston.va.us>

>>>>> "Fred" == Fred L Drake <fdrake at cnri.reston.va.us> writes:

    Fred> Gordon, This certainly would make sense.  The question
    Fred> becomes "If os.chmod doesn't exist, should shutil.copy()
    Fred> fail?"

I'd say yes, because you can't guarantee the documented semantics,
i.e. that the permission bits are copied.  However, shutil.copyfile()
shouldn't fail though -- and doesn't:

JPython 1.1beta1 on java1.2.1
Copyright (C) 1997-1999 Corporation for National Research Initiatives
>>> import shutil
>>> shutil.copy('/tmp/foo.txt', '/tmp/bar.txt')
Traceback (innermost last):
  File "<console>", line 1, in ?
  File "/home/bwarsaw/projects/python/pristine/Lib/shutil.py", line 53, in copy
  File "/home/bwarsaw/projects/python/pristine/Lib/shutil.py", line 34, in copymode
AttributeError: class 'org.python.modules.os' has no attribute 'chmod'
>>> shutil.copyfile('/tmp/foo.txt', '/tmp/bar.txt')

If you had Finn Bock's jnios module (see my other msg in this thread),
then shutil.copymode() and hence shutil.copy() probably won't fail
because then you /would/ have os.chmod().  I haven't tested this
though...

-Barry



From tim_one at email.msn.com  Tue Jun 29 03:15:15 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 29 Jun 1999 03:15:15 -0400
Subject: ANN: Stackless Python 0.2
In-Reply-To: <19990629121441.A19011@longford.cs.monash.edu.au>
Message-ID: <000301bec1ff$22ac6fe0$719e2299@tim>

[Tim sez Python stores space for each code object's "Python stack" in a
 runtime frame object]

[Toby J Sargeant]
> This might be completely irrelevant, but during the course of my
> masters, I considered doing this kind of thing to java in order to
> allow asynchronous threads to share stack frames (don't ask...).

Don't tell <wink>.

> My supervisor complained bitterly on the grounds that function
> invocations where orders of magnitude more common than object creation,
> and hence associating memory allocation/deallocation with every call was
> considered horrendously inefficient.
>
> It seems that this should affect Stackless Python equally as
> much.

No more so than Stackful Python:  Python has always worked this way;
Stackless Python doesn't change this aspect.

> Does anyone have anything to add on the subject? I would imagine that
> frames could be allocated and managed in chunks to alleviate a lot of
> the memory management load...

The code is in Objects/frameobject.c, and easy to follow.  frameobjects are
recycled  via their own free list.  Typically the total number of
frame-associated mallocs is proportional to the maximum depth of the call
stack, not to the number of calls made.

Setting up a Python frame remains expensive, but for other reasons.

115-lines-of-code-doesn't-run-in-one-cycle-malloc-or-not<wink>-ly y'rs  -
tim





From rfreedm1 at nospam.rochester.rr.com  Wed Jun 23 17:19:01 1999
From: rfreedm1 at nospam.rochester.rr.com (Benjamin Schollnick)
Date: Wed, 23 Jun 1999 21:19:01 GMT
Subject: help?  strptime ....
References: <rNHvjEdhm5Pp-pn2-dCsXnyPPztJy@d185d18f8.rochester.rr.com> <aahzFDsr8L.MI5@netcom.com>
Message-ID: <rNHvjEdhm5Pp-pn2-YHIXRXZXSHDh@d185d18f8.rochester.rr.com>

On Wed, 23 Jun 1999 20:21:56, aahz at netcom.com (Aahz Maruch) wrote:

> In article <rNHvjEdhm5Pp-pn2-dCsXnyPPztJy at d185d18f8.rochester.rr.com>,
> Benjamin Schollnick <rfreedm1 at nospam.rochester.rr.com> wrote:
> >
> >	I'm trying to use strptime (from the time module) and can't....
> 
> What version of Python?  I'm getting AttributeError in 1.5.1, so I
> suspect it's only in 1.5.2.  I'll bet mxDateTime has what you need,
> though, if you don't want to upgrade.

I'm using v1.51 as well.... But STRPTIME is clearly documented
in the v1.51docs.... (I haven't printed the v1.52 docs yet)

I'll try to see if the OS v1.52 python is available yet....

			- Ben



From MHammond at skippinet.com.au  Tue Jun 22 10:11:42 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Wed, 23 Jun 1999 00:11:42 +1000
Subject: [pythonwin] popup menu problems   (was: [Pythonwin] Help needed creating a Window)
References: <37651bde.25372123@NNRP.UK.INSNET.NET> <376f6b7a.760974@nnrp.uk.insnet.net>
Message-ID: <7ko5fd$cp8$1@m2.c2.telstra-mm.net.au>

Dave Kirby wrote in message <376f6b7a.760974 at nnrp.uk.insnet.net>...

>way. I only found it when I enclosed the line in a try/except clause.
>I *think* the exception is getting caught & thrown away somewhere in
>the pythonwin code since I didnt have any try clauses in my code at
>that point.

Eeek.  If you can determine this is true, I would like to know about it.

>level window to process the WM_CANCELMODE message.  Is there a way to
>register my icon window to handle this message, or do I need to go the

win32ui.HookMessage() should do the job?

>[1] well, almost - further experimentation shows that it only closes
>the menu if I click within the Pythonwin window, rather than anywhere
>on the screen.

Well, maybe you should set the owner to the desktop?

Mark.






From jfisher at are.berkeley.edu  Fri Jun 25 14:52:29 1999
From: jfisher at are.berkeley.edu (John Fisher)
Date: Fri, 25 Jun 1999 18:52:29 GMT
Subject: load_dynamic problem
Message-ID: <7l0j58$cbh$1@nnrp1.deja.com>

Hey folks,

I'm having a bit of trouble loading a C extension into Python 1.5.2.
I'm getting the error:

Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "bpinter.py", line 9, in ?
    PyBPMPD = imp.load_dynamic('PyBPMPD', 'PyBPMPD.so')
ImportError: ./PyBPMPD.so: undefined symbol: setdef

setdef, not written by me, but called from inside PyBPMPD.c, is
prototyped as:
	extern int setdef(bpmparam *);

The module itself compiles without complaint with:
	gcc -I/usr/include/python1.5/ -shared -o PyBPMPD.so PyBPMPD.c

I'd appreciate any ideas on the problem.  I'm rather stumped.

Thanks,
John


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From jantich at cantv.net  Wed Jun 30 11:31:14 1999
From: jantich at cantv.net (Jose A. Antich)
Date: Wed, 30 Jun 1999 11:31:14 -0400
Subject: Python Editor
Message-ID: <AJqe3.1318$24.76@news01.t-net.net.ve>

   Does anyone know if there's any Python Editor for the MAc?

Thank's



From mlh at idt.ntnu.no  Sun Jun  6 18:49:03 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 07 Jun 1999 00:49:03 +0200
Subject: Deleting characters from a string
References: <87n1ye3g4v.fsf@pc-hrvoje.srce.hr> <199906052144.XAA15045@axil.hvision.nl>
Message-ID: <y0jso85djxc.fsf@vier.idi.ntnu.no>

"Hans Nowak" <ivnowa at hvision.nl> writes:

> On 5 Jun 99, Hrvoje Niksic wrote:
> 
[...]
> host = filter(lambda c: not c in ' "', host)

Or - IMO more appropriately <wink>:

  host = filter(lambda c: c not in ' "', host)

> 
> Depends on how much you like, or hate, functional idiom, though. :)

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From joeld at anisci.com  Thu Jun 24 17:54:24 1999
From: joeld at anisci.com (Joel Davis)
Date: 24 Jun 1999 14:54:24 PDT
Subject: C++ and Python
Message-ID: <3772AA7D.A7855B93@anisci.com>

I am trying to wrap a bunch of c++ classes so that virtual methods in
derived classes can be written in python, and still callable from c++,

SWIG and friends seem to be one-way, they let you subclass a C++ class
in python via shadow classes or fancy extention types, but the
subclasses behavior only shows up on the python side. I am trying to add
the ability to mix python derived classes into a large c++ application.

I guess what I'm asking is, the ORA Python book says (p626), "it's also
possible to route C++ virtual method calls back down to methods in a
python subclass, but it may require a stub-class on the C++ side of the
interface, too (this is left as a Exercise to the Reader)." Are there
any tools to help make this easier, or any good examples of this?

thanks,
joel
--
----------------------------------------------------------------
                                                joeld at anisci.com





From see-sig at wavedivision.com  Wed Jun  9 04:03:44 1999
From: see-sig at wavedivision.com (Rick Robino)
Date: 9 Jun 1999 08:03:44 GMT
Subject: "Programming Python" still worthwhile?
References: <1283218560-70098357@hypernet.com>
Message-ID: <7jl750$qsu$1@news.gstis.net>

Thanks Gordon and William!

> Rick Robino originally posted:
>> I've been watching this group for some time and just now getting
>> warm to Python. Looking at source examples has been great, but for
>> completeness' sake I'd like to get more of the "why" maybe found in
>> a book. Since Python is at 1.5.2 and people are talking about 2.0 as
>> if it's around the corner, I was wondering if anyone here could give
>> me some advice as to whether or not "Programming Python" is useful,
>> if the current edition is still current, if I should wait for the
>> "Learning Python", or if I should use some other resources.

Gordon McMillan <gmcm at hypernet.com> wrote:
> Python 2 is around a really, really big corner, and then down the 
> block quite a ways. Python 1.6 sometime in 2000 (which will be 
> backwards compatible), so no 2 until 2001 or 2002.

I didn't do alot of research, but I couldn't get a good feel from
the huge thread that has been going on here named after 2.0. Appreciate
the ETA.

I have read the tutorial and it is wonderful, especially the couple
extra smaller ones mentioned in the "Python as a first language" thread.
Just wanted more meat; you know how that is ;)

> Learning Python is on the shelves. Haven't read it. Nobody's 
> flamed it, and that's a _very_ good sign. There's only a few things 
> out of date in PP, and you'll find updates on Mark's web site. I 
> liked PP very much (and the difficult to find Internet Programming 
> with Python, Watters et al, M&T books). When Mark wrote PP, he 
> thought it might be the only Python book ever, so he stuffed in 
> everything he could think of. So don't feel bad if you skip stuff.

Hmm... the Internet Programming book sounds interesting. I'll pick up
that and also the new book to check it out - I'll be a good test ;) I
have one of those weird learning curves where I need things explained
really simply at first and then I can usually jump up pretty fast. I'll
let you all know how I thought the book was. Daring advertisement...  
still feels like there is nothing I can't do in shell that perl can do,
cuz maybe it looks so much like shell? Most of my friends would slap
me silly for such heresy but after poking around with python for a couple
of weeks, I think its quite nice.

--Rick 
                                ^^^^^^^^^^^^^^
Rick Robino                                  rrobino [at] wavedivision.com



From aprasad at magix.com.sg  Fri Jun 18 12:20:18 1999
From: aprasad at magix.com.sg (Ajith Prasad)
Date: Sat, 19 Jun 1999 00:20:18 +0800
Subject: DATE ARITHMETIC
Message-ID: <7kejrv$e57$1@violet.singnet.com.sg>

I need to do some simple date arithmetic - finding the number of days
between any 2 days, the age of a person on a day given his birthday. The
mxDateTime module downloadable from the Python website is more complex than
what I need. Any other suggestions?





From mal at lemburg.com  Tue Jun  8 04:04:53 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Tue, 8 Jun 1999 08:04:53 GMT
Subject: mxODBC and MySQL
References: <Pine.LNX.4.05.9906080840190.12801-100000@ns.ac-copy.net>
Message-ID: <375CCEA5.3EA8C134@lemburg.com>

Joachim Schmitz wrote:
> 
> I ugraded to 1.5.2 now I get:
> 
> Python 1.5.2 (#1, Jun  8 1999, 07:26:20)  [GCC egcs-2.91.66 19990314
> (egcs-1.1.2  on linux2
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import ODBC.MySQL
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "/usr/local/lib/python1.5/ODBC/MySQL/__init__.py", line 7, in ?
>     from mxODBC import *
> ImportError: /usr/local/lib/python1.5/ODBC/MySQL/mxODBC.so: undefined
> symbol: SQLTables

Hmm, the MyODBC driver does define this function. Could be that
you've linked against the wrong lib or that the loader doesn't
find the .so file.

Cheers,
-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   206 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From scott at hk.super.net  Tue Jun 29 07:23:59 1999
From: scott at hk.super.net (Scott Deerwester)
Date: Tue, 29 Jun 1999 19:23:59 +0800
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <59EAFD6B662F0706.86BFF3E906CBF8DE.784D2A219F85BC21@lp.airnews.net> <99062523473500.00769@Vegus> <7l16e5$q4j$1@news.tamu.edu>
Message-ID: <3778ACCF.ACE29713@hk.super.net>

Corran Webster wrote:

> John Travers <jtravs at debian.org> wrote:
>
>> Has anybody tried the PyKDE/Qt package, (you don't have to have KDE), it is
>> absoloutly fantastic, quick, easy, powerful, far far better than Tkinter.
>> I love it!
>
> This is missing the point - although PyKDE/Qt might be a very nice GUI, it's
> not cross-platform in the same way as Tcl/Tk.  The same goes for Fltk,
> gtk or what have you.  It sounds like wxWindows might be getting near
> the same level of cross-platform compatibility - and it certainly seems
> to be one of the groups' objectives - it's not there yet.

That's probably true, but Qt IS cross-platform. Has anybody ever tried running
pyKDE with the Windows version of Qt? I've used pyKDE a bit and it seems to win
in regard to completeness and usability. One "problem" is that pyKDE (unlike Qt)
seems to have tightly coupled KDE (which isn't cross-platform) with Qt (which
is).

Haven't used wxWindows yet. I've used Tkinter a fair bit and have grown to
dislike it. Very awkward windowing and event model, as others have pointed out.
It also tends to be kind of ugly. Qt yields a lot prettier interfaces. and wins
on functionality. PythonWin is okay, but (obviously) Windows only. Sigh.





From moshez at math.huji.ac.il  Wed Jun 16 08:53:52 1999
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 16 Jun 1999 15:53:52 +0300
Subject: GC In Python: YAS (Yet Another Summary)
In-Reply-To: <Pine.A41.4.10.9906160748340.64942-100000@barney.mcit.com>
Message-ID: <Pine.SUN.3.95-heb-2.07.990616155228.15761A-100000@sunset.ma.huji.ac.il>

On Wed, 16 Jun 1999, Robert Meegan wrote:

> On Wed, 16 Jun 1999, Moshe Zadka wrote:
> 
> > Regarding the recend discussion here about garbage collection, I wrote 
> > a small article.
> 
> Moshe ---
> 
> You didn't provide the document or a URL.

You're absolutely right! Shame on me!
http://www.geocities.com/TheTropics/Island/2932/gcpy.html


--
Moshe Zadka <mzadka at geocities.com>. 
QOTD: My own exit is more likely to be horizontal then perpendicular.






From MHammond at skippinet.com.au  Fri Jun 18 20:15:26 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Sat, 19 Jun 1999 10:15:26 +1000
Subject: Python source code documentation?
References: <7kdv0l$n0g$1@nnrp1.deja.com>
Message-ID: <7kenc7$1od$1@m2.c2.telstra-mm.net.au>

Sam Schulenburg wrote in message <7kdv0l$n0g$1 at nnrp1.deja.com>...
>I have been using a freeware product called Autoduck to document my
>Python source code.

Cool!  I use AutoDuck to document all my Python extension C++ code.  All the
helpfiles in the Win32 extensions are generated using autoduck.

As another poster pointed out, you should look at pythondoc/gendoc.  Sorry
in advance for the length of this, but I am in exactly the same predicament
as you, and have been thinking for a number of years how to resolve this.
Im really pleased to finally find someone else with this problem, and really
hope we can solve it in a reasonable way.

However, depending on how far you have gone with AutoDuck, pythondoc/gendoc
isnt solving the exact same problem.  Well, they all attempt to solve the
same problem, but have different strengths.

IMO, AutoDuck generates far superior reference documentation.  Pythondoc's
big strength is the integration with Python.  Using docstrings is an obvious
advantage, and much of what you need to explicitely mark-up using autoduck,
pythondoc already knows.

AutoDuck has 3 strengths over pythondoc that I am having trouble
reconciling:
* pythondoc really doesnt support natural links to objects.  Autoduck can
say "see method <om Foo.Bar>" or "@xref object|method" to generate inline
links or a cross-references section.  Has been some discussion how to expand
Pythondoc to support this, but a) it hasnt been done and b) it still is not
as natural as autoduck.

* Autoduck pulls-apart and re-assembles the output.  For example, one method
you may define entries in the following order: return description, params,
notes and cross-references.  A different method may define them in a
completely different order (as the best place to add the documentation is
right next to the appropriate code.  Autoduck will still always produce a
reference guide formatted as you choose - eg, always "params, return desc,
comments, cross-referneces".

* It can be used to document C extensions damn well.

Some of these problems boil down to an underlying philosophy of pythondoc -
attempting to use raw, unmarked docstrings to produce the doc.  Pythondoc
really doesnt want to introduce ugly mark-ups in the docstring, as this
tends to make the docstring hard to use in native form - it really _must_ be
read via the document generation tool.  This would be a serious bummer given
Python's introspective nature, and that IDE's may use this info on the fly.
(otoh, autoduck info doesnt exist at runtime at all!)

Sooooo.  I have come to the conclusion that we need a _new_ tool -
PythonDuck.  It should work with Pythondoc, and be capable of enhancing its
output.  A new release of pythondoc uses XML I believe, so PythonDuck could
be a tool that can enhance the documentation generated by pythondoc.
Alternatively we just graft it in, and beat up Daniel until he agrees :-)

This tool could possibly still work from comments, or maybe from the
tail-end of the doc strings (thereby stripped from the documentation, and
not messing the raw docstring too much).  #@ comments could be considered
"directives" for the generated output.  Eg, "#@ xref SomeOtherFunction" will
create a cross-references section.  A comment such as "#@ paramtype
param_name <o SomeType>" would direct that the param called "param_name"
should be documented as being of type "SomeType" and a cross-reference to
the SomeType definition be added.  The tool should resolve all links at the
end, and report bad links.

I know I am not making my point that well, and unfortunately dont have time
to fix this.  But in summary, Pythondoc exists and is very cool, but doesnt
quite cover what AutoDuck can do (and vise-versa, but that isnt the point
now).  I believe a real need exists for the functionality AutoDuck provides,
and further I need it personally!  So we should try and make this happen...

Mark.






From dave_kirby at eur.3com.com  Thu Jun 17 05:49:25 1999
From: dave_kirby at eur.3com.com (Dave Kirby)
Date: Thu, 17 Jun 1999 09:49:25 GMT
Subject: wierdness with pythonwin dialogs, threads & refcounts
Message-ID: <3768b75e.5469224@NNRP.UK.INSNET.NET>

I have found strange behaviour (strange to me at least) when creating
a pythonwin dialog on a separate thread created using the standard
python thread module.  The dialog object is never deleted since the
thread is keeping references to it. 

The following example uses dlgtest.py that comes with the pythonwin
distribution:

#------------- 8< ----------------------------
C:\dev\PYTHON\Pythonwin\pywin\DEMOS>python
Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> from dlgtest import TestDialog
>>> import sys, thread
>>> def test():
...   t = TestDialog()
...   print "refs:", sys.getrefcount(t)
...   t = None
...
>>> test()
Edit text changed!
refs: 2
>>> thread.start_new_thread( test, () )
>>> refs: 6
#------------- 8< ----------------------------

As you can see from the above, creating the dialog in the thread gives
it an extra 4 references compared with the non-threaded invocation. Is
this a bug, or is the thread module incompatible with pythonwin? If so
how would I do the equivalent using pythonwin threads?  I have read
the available docs, but it is still pretty opaque to me.

Any help would be greatly appreciated.

	Dave K




From Gaetan_Corneau at baan.com  Thu Jun 10 09:13:14 1999
From: Gaetan_Corneau at baan.com (Gaetan Corneau)
Date: Thu, 10 Jun 1999 09:13:14 -0400
Subject: COM dates
Message-ID: <816010E2456BD111A48700805FBBE2EEFDF357@ex-quebec-u1.baan.com>

I did it the Python way: I looked at what I really needed, looked at the
documentation to see what I had, and wrote a 3 line function that takes
year, month, date, hour, minute and second and returns time in the same
numeric format as time.time().

I knew about mxDateTime and think it's really cool (I have written something
similar in C++ years ago), but for many reasons, I don't want to rely on
extensions for this particular project.


Thanks,
______________________________________________________
   Gaetan Corneau
   Software Developer (System integration Team)
   BaaN  Supply Chain Solutions  
   E-mail: Gaetan_Corneau at baan.com        
   Compuserve: Gaetan_Corneau at compuserve.com  
   ICQ Number: 7395494             
   Tel: (418) 654-1454 ext. 252          
______________________________________________________
"Profanity is the one language all programmers know best"

> -----Original Message-----
> From:	M.-A. Lemburg [SMTP:mal at lemburg.com]
> Sent:	Thursday, June 10, 1999 4:31 AM
> To:	Gaetan Corneau
> Cc:	Python List @ Python.org
> Subject:	Re: COM dates
> 
> 
> You could check out mxDateTime. It has COM date support, strptime()
> and also a much more powerful date/time parser:
> 



From perry at ac-copy.net  Thu Jun  3 17:16:52 1999
From: perry at ac-copy.net (Joachim Schmitz)
Date: Thu, 3 Jun 1999 21:16:52 GMT
Subject: build of python 1.5.2 with threads on SunOs 5.6 fails
Message-ID: <Pine.LNX.4.05.9906032310190.6933-100000@ns.ac-copy.net>

Hi,

I compiled Python 1.5.2 on SunOS 5.6/Solaris 2.6 without threads
successfully. But doing a configure --with-threads, and make results in
the following error:
.....

cd Modules;  make OPT="-g -O2" VERSION="1.5" \
                prefix="/usr/local" exec_prefix="/usr/local" \
                LIBRARY=../libpython1.5.a link
gcc   python.o \
          ../libpython1.5.a   -lsocket -lnsl -ldl  -lthread -lm  -o python
Undefined                       first referenced
 symbol                             in file
Py_Main                             python.o
ld: fatal: Symbol referencing errors. No output written to python
*** Error code 1
make: Fatal error: Command failed for target `link'
Current working directory /u/src/Python-1.5.2/Modules
*** Error code 1
make: Fatal error: Command failed for target `python'

any help is welcome

Joachim Schmitz






From delgado at Mathematik.Uni-Bielefeld.DE  Wed Jun 16 16:54:50 1999
From: delgado at Mathematik.Uni-Bielefeld.DE (Olaf Delgado)
Date: Wed, 16 Jun 1999 22:54:50 +0200
Subject: Newbie again: computing attributes on the fly
In-Reply-To: <1282559115-109766087@hypernet.com>
References: <1282571882-108998091@hypernet.com> <1282559115-109766087@hypernet.com>
Message-ID: <Pine.OSF.4.05.9906162240580.19856-100000@pinax.Mathematik.Uni-Bielefeld.DE>

On Wed, 16 Jun 1999, Gordon McMillan wrote:

> [Olaf on my very simple getattr hack]:
> 
> > Yes, I definitely want to generalize that. The problem is that that
> > method may be defined in a superclass, so "self.__dict__" won't find
> > it. AFAIK (which may be not be very far), I have to use "hasattr"
> > and/or "getattr", which in turn will call "__getattr__".
> 
> Oh, getattr isn't bad. Wait 'till you try setattr! Anyway, this 
> should demonstrate how to control the recursion:
[...]

Yes, that looks much more civilized than my version. So, I'll have to take
a closer look at all these magic methods.

-- 
   ////         Olaf Delgado Friedrichs, Uni Bielefeld, Mathematik
   Olaf         Postfach 100131                  (room:  V4-109)
   `='          D-33501 Bielefeld, Germany       (phone: +49 521 106-4765)
                           http://www.mathematik.uni-bielefeld.de/~delgado




From tbryan at arlut.utexas.edu  Mon Jun 14 23:28:15 1999
From: tbryan at arlut.utexas.edu (Thomas Bryan)
Date: Mon, 14 Jun 1999 22:28:15 -0500
Subject: Looking for modules?  Config files?  Ini files?  in python....
References: <rNHvjEdhm5Pp-pn2-WD108jh22jtO@d185d18f8.rochester.rr.com>
Message-ID: <3765C84F.FC952ACE@arlut.utexas.edu>

Benjamin Schollnick wrote:
> 
> Folks,
> 
>         I'm trying to find a simple method to deal with *.INI or other 
> styles of configuration files....

ConfigParser
http://www.python.org/doc/current/lib/module-ConfigParser.html
user
http://www.python.org/doc/current/lib/module-user.html
site
http://www.python.org/doc/current/lib/module-site.html
rexec
http://www.python.org/doc/current/lib/module-rexec.html

depending on your needs.

-------------------------------------------
Tom Bryan
Applied Research Laboratories
University of Texas at Austin



From borkman at usa.net  Sun Jun 20 22:51:07 1999
From: borkman at usa.net (Lee Borkman)
Date: Mon, 21 Jun 1999 02:51:07 GMT
Subject: XML-toolkit for Windows - binaries available?
References: <376701a3.72288277@newssrv> <wkzp208udt.fsf@ifi.uio.no> <3767fc51.1547455@news.iol.ie> <wkd7yv7t5p.fsf@ifi.uio.no> <37693395.731772@news.iol.ie>
Message-ID: <376da59b.2034245@news>

Thanks very much, but I'm afraid I'm still unsure.

I have copied pyexpat.dll and sgmlop.dll to the DLLs directory

I have modified the PythonPath to include:
C:\Program Files\Python\Lib\xml-0.5.1
and:
C:\Program Files\Python\Lib\xml-0.5.1\expat\bin
	(this is where xmlparse.dll lives)

Now I have to edit  __init__.py to include sgmlop and pyexpat in the
__all__ list, is that right?  Now which __init__.py would that be?  I
assume it's C:\Program Files\Python\Lib\xml-0.5.1\__init__.py

Okay, I've done that.

But none of the demos will work.  I keep getting:
ImportError: No module named xml.sax
or:
ImportError: No module named pyversioncheck


I guess I still don't understand where Python (and PythonWin) look for
their modules, and I guess I don't understand quite a bit of other
stuff.

Thanks very much for all of your help.  I've obviously got a bit more
homework to do before I proceed.

Leeb.



On Thu, 17 Jun 1999 17:43:22 GMT, digitome at iol.ie (Sean Mc Grath)
wrote:

>On 16 Jun 1999 23:05:38 +0200, Lars Marius Garshol <larsga at ifi.uio.no>
>wrote:
>
>>
>>* Sean Mc Grath
>>|
>>| The binaries are there but a bit of fiddling is required to get
>>| things working. (This is from memory but here goes)
>>| 
>>| The pyexpat and sgmlop dlls have to be copied into the xml/parsers
>>| subdirectory. You also need to edit the __init__.py file to include
>>| 'pyexpat' and 'sgmlop' in the __all__ string.
>>
>>Heh. I didn't know you could do this! :)
>>
>>I just put them in the python/dlls directory, and that worked for me.
>>
>Lars, is quite right. python/dlls is a better place to put them than
>xml/parsers.
>
>Sean
>
>
>




From Mike.C..Fletcher at p98.f112.n480.z2.fidonet.org  Wed Jun 30 23:25:28 1999
From: Mike.C..Fletcher at p98.f112.n480.z2.fidonet.org (Mike C. Fletcher)
Date: Thu, 01 Jul 1999 04:25:28 +0100
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
Message-ID: <000015f6@bossar.com.pl>

From: "Mike C. Fletcher" <mfletch at vrtelecom.com>

There is a binary distribution of the python version in the FTP directory,
very nice package.  I'm thinking of rewriting my (few) GUI applications
tonight and ditching tkinter entirely.  Documentation seems rather sparse,
but since everything looks fairly simple, it might be doable.

Note, the OpenGL demos are dependent on PyOpenGL.  This should be mentioned
somewhere.

My only other comment is that the window-dragging border in the demos looks
like static (i.e. noise), instead of a controlled GUI element.  Installation
went flawlessly, was able to run the demos in less than a few minutes after
download.

Enjoy,
Mike

----- Original Message -----
From: Mikael Lyngvig <mikael at pobox.com>
Newsgroups: comp.lang.python
To: <python-list at cwi.nl>
Sent: Wednesday, June 30, 1999 8:48 PM
Subject: Re: Python GUIs: Abandoning TkInter and welcoming wxPython?
...
> http://www.cfdrc.com/FOX/fox.html.
>
> The presentation seems promising, but I'd like to suggest a binary
> distribution for PC/Intel computers.  I know this is not common for
> Unix distributions, but it's the norm for PCs.
...



From mwh21 at cam.ac.uk  Fri Jun 11 13:28:45 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 11 Jun 1999 18:28:45 +0100
Subject: newbie questions
References: <x7emjizs8s.fsf@norden.math.tntech.edu>
Message-ID: <m3u2seaboy.fsf@atrus.jesus.cam.ac.uk>

Jeff Norden <jeff at norden.math.tntech.edu> writes:

> I've been tinkering with python for a few days.  I have to admit that my
> first impression was that this language would turn me into a newt, but I got
> better :-)
> 
> A few questions:
> 
> 1) Is there no equivalent to perl's  "use strict" in which all variables
> have to be declared?  How about perl's "-w" flag, which at least gives a
> warning when a variable is only used once.  These features are nice for
> catching common mistakes, i.e., if you try to increment spam with
> 
>    span = spam + 1
> 
> python will just create a new span variable, and no error message.
> Or, consider the even more insidious error:   vikingl = viking1 + spam

Python has no declarations at all, ever (I think). There's some
kpylint package found off of www.chordate.com that will check for
identifiers that are only used once. It's not incredibly bright (it
doesn't really try to be).

> 2) Is the above the correct way to increment a variable? 

Yes.

> I can't find an
> equivalent for "++" or "+="

That would be because there isn't one. You could use my bytecodehacks
package to fake them, but don't ;-).

> 3) These are rather minor, but is there no way to "forward declare" function
> definitions? 

No.

> I'd rather put them at the end of the file instead of the
> beginning.  Python does parse the whole file before it starts execution,
> doesn't it?  

Yes.

> Also, is there a way to just check for syntax errors without
> execution, like perl's "-c" switch?

Well, you could try this:

def check_file(file):
    import parser
    try:
        parser.suite(open(file).read())
        print "no syntax errors"
    except parser.ParserError:
        print "syntax error"

this was thrown together in about a minute.
seems to work.

HTH
Michael



From tim_one at email.msn.com  Fri Jun  4 01:15:18 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 4 Jun 1999 01:15:18 -0400
Subject: "The Python Way"
In-Reply-To: <Pine.A32.3.90.990603172339.1626A-100000@elvis.med.Virginia.EDU>
Message-ID: <001301beae49$3c88da80$949e2299@tim>

[Steven D. Majewski]
> What makes you think that Tim and Guido would agree on "The Python Way" ?

Just like the old days, Steven:  we would be forced to, just to oppose you
<wink>.

ah-nostalgia-ly y'rs  - tim





From graham at sloth.math.uga.edu  Tue Jun  8 15:39:11 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 8 Jun 1999 19:39:11 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net> <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net> <7j34l3$hbj$1@nnrp1.deja.com> <7j66b7$il5$2@cronkite.cc.uga.edu> <3756C693.2C910B29@usa.net> <7j6jnl$58f$3@cronkite.cc.uga.edu> <3757A820.CE3F195F@usa.net> <7j8obc$55v$2@cronkite.cc.uga.edu> <375BEDF8.DD14C4E6@usa.net> <7jgs5g$q55$1@cronkite.cc.uga.edu> <375CE5BD.3E865EF4@usa.net>
Message-ID: <7jjrgv$h16$1@cronkite.cc.uga.edu>

Graham Matthews wrote:
> I really don't understand this. A lot of languages (eg. Java) have
> garbage collection and finalisers. All these languages manage to delete
> objects in an order that does not cause any problems (all finalisers etc
> work).
Salvador =?iso-8859-1?Q?Fandi=F1o?= (fandino at usa.net) wrote:
: I have been using Java since it was invented and I haven't programed a
: finaliser ever. They should run under so unknown and inconsistent
: environment that they are not usable at all (except for freeing external
: resources).

I am not sure I understand this. Can you explain to me how a finaliser
can run in an unknown environment. Surely everything the finaliser needs
is referenced by the finaliser (either directly or indirectly) and hence 
is still there.

Salvador =?iso-8859-1?Q?Fandi=F1o?= (fandino at usa.net) wrote:
: BTW, how could be acceptable that the same __del__ method could be
: called from two totally different schemes (GC reclaiming objects or
: reference count becoming 0)? it would break all the library classes that
: now use __del__ .

Because anything referenced by a __del__ method will always be there
when the __del__ is called, under any GC scheme.

graham
-- 
           As you grow up and leave the playground
       where you kissed your prince and found your frog
          Remember the jester that showed you tears
                   the script for tears



From neelk at brick.cswv.com  Tue Jun 15 22:19:22 1999
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 15 Jun 1999 21:19:22 -0500
Subject: Module magic methods
References: <WgP83.7966$Yr2.129670@news2.rdc1.on.home.com> <m3g13wt71h.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <7k71ja$i8b$1@brick.cswv.com>

In article <m3g13wt71h.fsf at atrus.jesus.cam.ac.uk>,
Michael Hudson  <mwh21 at cam.ac.uk> wrote:
>
>FWIW, I've often thought it would be funky to be able to define
>__add__ or __call__ etc methods to modules. Pointless, I'd have to
>admit, but still funky.

Not so pointless, imo, because a module is just a singleton class 
with a light dusting of nonstandard syntax and semantics. I've 
been bitten now and again by the differences (don't ask), and I
compiled a list of the differences. Some are superficial, and some
are significant:

o Classes are defined by the block a "class" statement contains; a
  module is defined by the contents of its file (sort of).
 
o Instances of classes are created by the __init__ method of the
  class (mostly), and a module instance is created with the "import"
  statement.

o Type-checking is performed on a call to a method of a class to
  guarantee that the first argument is the same class as the class the
  method belongs to (sigh), whereas no such type checking is performed
  on functions inside a module.

o Assigning to a instance attribute of the same name as a class
  variable will just shadow the class attribute for that instance. 
  For the module, no such shadowing happens. That is,

  >>> class Foo:
          q = 'foo'

  >>> x = Foo()
  >>> x.q = 'bar'
  >>> print x.q
  bar
  >>> print Foo.q
  foo

  >>> import math
  >>> print math.pi
  3.14159265359
  >>> math.pi = 3
  >>> print math.pi 
  3
  >>> del math; import math 
  >>> print math.pi
  3

Right now, since classes and modules are built-in types, so we don't 
have access to what makes them different. So you can't have a module
intializer that you can pass information to, for example. But in the
promised land of Python 2.0, when types and classes are unified, it
may be possible to do things like this:

class WhyWhyWhy(Module, Class):
    [...]

In addition to just being gross, notice how we've snuck a metaclass
into the equation. This would just be too cool, wouldn't it? :) 


Neel



From jkraai at murl.com  Thu Jun  3 09:45:48 1999
From: jkraai at murl.com (jkraai at murl.com)
Date: Thu, 3 Jun 1999 13:45:48 GMT
Subject: starship.python.net
Message-ID: <000601beadc7$75b2ecb0$8b7125a6@cpda6686.mcit.com>

Ok, I don't go to the starship root very often, but went there this morning.

It's _beautiful_.  Who did this excellent work?  Chris T?

Wow.  Clean, elegant, etc.






From news at dorb.com  Wed Jun 16 10:22:55 1999
From: news at dorb.com (Darrell)
Date: Wed, 16 Jun 1999 10:22:55 -0400
Subject: GC In Python: YAS (Yet Another Summary)
References: <Pine.SUN.3.95-heb-2.07.990616133606.13037B-100000@sunset.ma.huji.ac.il>
Message-ID: <FtO93.191$U6.1770@newsr2.twcny.rr.com>

Guido's idea sounds least intrusive.

Is there any merit to building a tool to tell me if I have cycles and where
they are ?
Then I apply one of N strategies to kill them.

How often does the normal Joe have or care that he has cycles ?
I find them hard to track down.

--
--Darrell





From amackay at starvision.com  Fri Jun 18 19:19:21 1999
From: amackay at starvision.com (Angus MacKay)
Date: Fri, 18 Jun 1999 16:19:21 -0700
Subject: looking at function prototypes
Message-ID: <376AD3F9.C51AAED2@starvision.com>

how can one look at a function prototype in python?

the interpreter will tell you the number but what
about the type?

I am using fnorb for some CORBA stuff and would like
to know what types are expected in the methods.

I do realize that this might be impossible as function
arguments are not typed. it so then anyone know how
to get "fnidl" to generate doc strings with the prototype
in it?

cheers, Angus.



From sams at quinta.com  Wed Jun 23 13:08:49 1999
From: sams at quinta.com (Sam Schulenburg)
Date: Wed, 23 Jun 1999 17:08:49 GMT
Subject: Embedded Python Documentation information
Message-ID: <7kr4an$cft$1@nnrp1.deja.com>

While trying to add docinfo strings to my embedded C libraries I found
the folloing documentation issue:

The following structure follows the definition for setting up the 'C'
Python interface in embedded code. What is missing is the information on
the docinfo field parameters.

================== As defined in documentation ======================
// This function is the only exported function from the
// DLL and has to be of type non-static, all other functions must be
static.
__declspec(dllexport) void initASPIPy(void)
    {
        Py_InitModule("ASPIPy",ASPIPyFuncs);
    }

// Define the Methods array
static PyMethodDef ASPIPyFuncs[] =
    { //Python_Func_Name  C_Func_Name   Method_Flags
        {"startaspi"      ,PyStart      ,METH_VARARGS},
        {"cls"            ,PyCls        ,METH_VARARGS},
        {"io6"            ,PyIO6        ,METH_VARARGS},
        {"io10"           ,PyIO10       ,METH_VARARGS},
        {"io12"           ,PyIO12       ,METH_VARARGS},
        {"settid"         ,PySetTID     ,METH_VARARGS},
        {"setasciion"     ,PySetASCIIOn ,METH_VARARGS},
        {"gettid"         ,PyGetTID     ,METH_VARARGS},

=============== End Original documentation =========



The following is the correct definition for the Methods array adding
the docinfo string:
// Define the Methods array
static PyMethodDef ASPIPyFuncs[] =
{
// P_Func_Name  C_Func_Name    Method_Flags  DocInfo
  {"startaspi"  ,PyStart      ,METH_VARARGS,"startaspi(string,flag)"},
  {"cls"        ,PyCls        ,METH_VARARGS,"cls(void) Clears screen"},
  {"io6"        ,PyIO6        ,METH_VARARGS,"io6(cdb0...cdb5}",
  {"io10"       ,PyIO10       ,METH_VARARGS,"io10(cdb0...cdb9)"},
  {"io12"       ,PyIO12       ,METH_VARARGS,"io12(cdb0..cdn11)"},
  {"settid"     ,PySetTID     ,METH_VARARGS,"settid(TID)"},
  {"setasciion" ,PySetASCIIOn ,METH_VARARGS,"setasciion(flag)"},
  {"gettid"     ,PyGetTID     ,METH_VARARGS,"gettid(void)"}
}


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From Paul.Prescod at p98.f112.n480.z2.fidonet.org  Wed Jun 30 09:49:00 1999
From: Paul.Prescod at p98.f112.n480.z2.fidonet.org (Paul Prescod)
Date: Wed, 30 Jun 1999 14:49:00 +0100
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
Message-ID: <000015b9@bossar.com.pl>

From: Paul Prescod <paul at prescod.net>

Lyle Johnson wrote:
> 
> Just to add fuel to the fire, we now have a Python interface to FOX, a
> cross-platform GUI toolkit:

I find it fascinating that there are so many GUI toolkits. The GUI is more
than twenty years right and we haven't agreed on the basic programming
model or API yet??

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for only himself
 http://itrc.uwaterloo.ca/~papresco

"The new revolutionaries believe the time has come for an aggressive 
move against our oppressors. We have established a solid beachhead 
on Friday. We now intend to fight vigorously for 'casual Thursdays.'
  -- who says America's revolutionary spirit is dead?



From greg.ewing at compaq.com  Wed Jun 16 18:22:30 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Thu, 17 Jun 1999 10:22:30 +1200
Subject: Raw Strings!
References: <P3v83.7006$Yb7.1482@news4.mia> <19990612173928.A23568@humbolt.nl.linux.org> <y0j7lp92y0w.fsf@vier.idi.ntnu.no>
Message-ID: <376823A6.E8B2F7AA@compaq.com>

"Magnus L. Hetland" wrote:
> 
> The point with raw strings is that backslashes can be written
> out without being escaped.

Be warned that raw strings are not *quite* completely
raw! For example,

   s = r"\abc\def\"

is a syntax error. On the other hand,

   s = r"\abc\def\\"

is not. I'll leave you with the fun of trying to
guess what the resulting string contains...

They-should-really-be-called-rare-strings,
Greg



From aahz at netcom.com  Mon Jun 14 01:48:42 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Mon, 14 Jun 1999 05:48:42 GMT
Subject: Variations on a theme by "pound bang"
References: <3763e75f.3265255@news.iol.ie>
Message-ID: <aahzFDAyt7.LMI@netcom.com>

In article <3763e75f.3265255 at news.iol.ie>,
Sean Mc Grath <digitome at iol.ie> wrote:
>
>Can anyone shed some light on the benefits of:-
>
># !/usr/bin/env python

ObSmartAss: there's no benefit here, try using "#!" instead.
-- 
                      --- Aahz (@netcom.com)

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

"That doesn't stop me from wanting to work all three of them over with
the clue stick for while, with no safewords allowed."  --abostick



From ovidiu at cup.hp.com  Wed Jun 16 18:43:02 1999
From: ovidiu at cup.hp.com (Ovidiu Predescu)
Date: Wed, 16 Jun 1999 22:43:02 GMT
Subject: Emacs package to the Python debugger? 
In-Reply-To: Your message of "Wed, 16 Jun 1999 16:25:03 CDT."
             <14184.5679.711795.786883@buffalo.fnal.gov> 
References: <14184.5679.711795.786883@buffalo.fnal.gov>
Message-ID: <199906162243.PAA03215@hpcll563.cup.hp.com>

On Wed, 16 Jun 1999 16:25:03 -0500 (CDT), Charles G Waldman <cgw at fnal.gov> 
wrote:

> Ovidiu Predescu writes:
> 
> >   Is there any pac kage that allows debugging Python programs from
> >   within Emacs?  Something that is integrated with Emacs' GUD would
> >   be very convenient.
> 
> Recent versions (21.0 and up) of XEmacs include this in the
> "prog/debug" package.  
> 
> If you don't use the XEmacs package system, you can still get the
> single-file "pdb.el" from
> 
> ftp://ftp.aifb.uni-karlsruhe.de/pub/wwe/pdb.el
> 
> And, yes, it is build on top of the Grand Unified Debugger.

Thanks, I already got the pdb.el from that site. I'm still using Xemacs 20.4 so
I had to install it manually on my machine.

One thing I noticed is that it doesn't have a debugger mode for the source file
similar with gdbsrc-mode. This requires me to use the python debugger buffer to
enter the commands. I cannot simply type 'n' or 's' on the buffer containing
the source. Do you know if such a thing is available?

Thanks,
Ovidiu





From holger at phoenix-edv.netzservice.de  Tue Jun 22 04:26:33 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Tue, 22 Jun 1999 08:26:33 GMT
Subject: glint?
Message-ID: <376F48B9.133D37C6@phoenix-edv.netzservice.de>

Hi,

I'm working on an installation-tool for linux/win without
any graphical interface but file-controlled.
I just got the german translated book about python from 
Himstedt/M?tzel. They tell about RedHat who built their
installation-routine with python. Now I'd like to get more
informations about that. We've got only SuSE-linux-distributions 
here, so I couldn't have a look at myself.
At homepage of RedHat I found:
ftp://ftp.redhat.com/redhat/old-releases/redhat-4.2/i386/RedHat/RPMS/glint-2.1.5-1.i386.rpm
but the server was down (or something else?!)..;-(
Is 'glint' the right thing I'm searching for?
Is it with free source???

Thanx,
Holger



From mnot at pobox.com  Tue Jun 22 07:09:41 1999
From: mnot at pobox.com (Mark Nottingham)
Date: Tue, 22 Jun 1999 11:09:41 GMT
Subject: htmllib: CR in CDATA
References: <00d201bebc5a$adb129a0$0301a8c0@mnot.net> <012601bebc91$f3ded760$0301a8c0@mnot.net>
Message-ID: <015701bebc9f$ba85b7a0$0301a8c0@mnot.net>

And here's a go at a patch. Looking at the DTD, practically all attribute
types are CDATA, and those that aren't shouldn't have tabs or newlines in
them anyway. It uses string.translate; how efficient is this?


*** /opt/local/lib/python1.5/sgmllib.py Thu Apr 15 00:54:11 1999
--- sgmllib.py Tue Jun 22 21:02:05 1999
***************
*** 38,43 ****
--- 38,44 ----
      '[%s]*([a-zA-Z_][-.a-zA-Z_0-9]*)' % string.whitespace
      + ('([%s]*=[%s]*' % (string.whitespace, string.whitespace))
      + r'(\'[^\']*\'|"[^"]*"|[-a-zA-Z0-9./:+*%?!\(\)_#=~]*))?')
+ cdata_tr = string.maketrans('\t\n', '  ')


  # SGML parser base class -- find tags and call handler functions.
***************
*** 251,256 ****
--- 252,258 ----
              elif attrvalue[:1] == '\'' == attrvalue[-1:] or \
                   attrvalue[:1] == '"' == attrvalue[-1:]:
                  attrvalue = attrvalue[1:-1]
+                 attrvalue = string.translate(attrvalue, cdata_tr, '\r')
              attrs.append((string.lower(attrname), attrvalue))
              k = match.end(0)
          if rawdata[j] == '>':





From mark at _spamspamlovelyspam_btweng.krakow.pl  Fri Jun 18 14:00:52 1999
From: mark at _spamspamlovelyspam_btweng.krakow.pl (MK)
Date: Fri, 18 Jun 1999 18:00:52 GMT
Subject: Crawling Python
References: <000601beb91d$e1321ae0$839e2299@tim>
Message-ID: <376a1ac1.10261180@news.tpnet.pl>

On Thu, 17 Jun 1999 20:02:39 -0400, "Tim Peters" <tim_one at email.msn.com> wrote:

>> PS. I have tried the same thing in DOS executable -- above does
>> not happen.

>Sure, and in return none of IDLE's history, undo/redo, searching, replacing,
>scrolling or editing cmds work in a DOS box either <wink>.

True, that's not what I meant, however -- whether it is Python or IDLE that
is affected by the problem. Since Python itself is OK, no problem.

Strangling IDLE continued:

in DOS box I tried

>>> s=''
>>> for i in range(10000):
...     s=s+`range(i,i+20)`
...

>>> len(s)
1178515

and everything is OK.

While the same loop in IDLE crashed it (I mean it hangs for some period, then it
stops responding and has to be killed).







MK


--------------------------------------------------
Reality is something that does not disappear after 
you cease believing in it - VALIS, Philip K. Dick
--------------------------------------------------

Delete _spamspamlovelyspam_ from address to email me

postmaster at 127.0.0.1
root at 127.0.0.1
webmaster at 127.0.0.1






From tavares at connix.com  Mon Jun  7 22:17:58 1999
From: tavares at connix.com (Chris Tavares)
Date: Mon, 07 Jun 1999 22:17:58 -0400
Subject: COM Question - QueryInterface?
Message-ID: <375C7D56.C9744E39@connix.com>

I've got a (hopefully easy) question about python's COM support.

I'm developing a COM object in C++ which implements the COM collection
idiom - an Item method that returns another object. The code in Python
I'm using is something like this:

from win32com.client import Dispatch
drives = Dispatch("AmAccess.Drives")
drive = drives.Item("c:")
dir = drive.Item("some-area")

On this last line, I get an AttributeError for Item.

I think I know the reason why - the interface returned from Drives.Item
isn't a dual, it's an oleautomation compatible vtable interface.
IDispatch on that object is implemented separately. So, PythonCOM sees
the VT_UNKNOWN return type and wraps a PYIUnknown object around it. I
need to be able to do a QueryInterface for IDispatch.

How do I get Python to do that? There's a PyIID type, but apparently it
only works for class or progids.  Anyone got any suggestions?

Thanks for any help,

-Chris





From gerrit.holl at pobox.com  Tue Jun 22 09:36:58 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Tue, 22 Jun 1999 15:36:58 +0200
Subject: signal handler
In-Reply-To: <376F87CB.DC3802E6@pk.highway.ne.jp>; from Thooney Millennier on Tue, Jun 22, 1999 at 09:55:39PM +0900
References: <376F87CB.DC3802E6@pk.highway.ne.jp>
Message-ID: <19990622153658.A19393@optiplex.palga.uucp>

On Tue, Jun 22, 1999 at 09:55:39PM +0900, Thooney Millennier wrote:
> Newsgroups: comp.lang.python
> Subject: signal handler
> Date: 	Tue, 22 Jun 1999 21:55:39 +0900
> To: python-list at python.org
> 
>   import signal;
>   def func(arg1,arg2):
>       DO_SOMETHING_GOOD;
>       print "OK!";
>   signal.signal(SOME_SIGNAL,func);
> 

Why all those ';'s?

groeten,
Gerrit.

-- 
The Dutch Linuxgames homepage:	http://linuxgames.nl.linux.org
Personal homepage:		http://www.nl.linux.org/~gerrit/

Discoverb is a python program in Dutch and English which tests the words you
learned by asking it. Homepage: http://www.nl.linux.org/~gerrit/discoverb/



From Darrell at p98.f112.n480.z2.fidonet.org  Tue Jun 29 06:43:46 1999
From: Darrell at p98.f112.n480.z2.fidonet.org (Darrell)
Date: Tue, 29 Jun 1999 11:43:46 +0100
Subject: run .py not in PATH
Message-ID: <00001588@bossar.com.pl>

From: "Darrell" <news at dorb.com>

#!/usr/bin/env python
"""
The .py you want to run is in the python path but not
in your shell path.

I know everyone but me already has this problem solved
in a better way than this. If so please show me the way.

Assuming this is runIt.py then
runIt.py someScript.py arg1 arg2
"""
import sys
import imp

fp, pathname, description = imp.find_module(sys.argv[1])
sys.argv=sys.argv[2:]
md=imp.load_module('__main__', fp, pathname, description)






From cfelling at iae.nl  Thu Jun 24 18:45:39 1999
From: cfelling at iae.nl (Carel Fellinger)
Date: 25 Jun 1999 00:45:39 +0200
Subject: tkMessageBox returns 'can't invoke "frame" command:' error
References: <7kktvv$550$1@nnrp1.deja.com> <7ku1l5$13q$1@vvs.superst.iae.nl>
Message-ID: <7kucej$1aq$1@vvs.superst.iae.nl>

Carel Fellinger <cfelling at iae.nl> wrote:
> though I'm new to python and Tkinter myself, I think I know what the

so, I'm new and reacted to soon

> As for your error message complaining about the frame command not working
> because the application has been destroyed, that's a little trickier.
> Just where should I start...

> First, Tkinter keeps track of the current Tk invocation using the var
> Tkinter._default_root, initially this equals None, and upon invocing
> Tk() it gets initialised. Unfortunately Tkinter doesn't reset this
> var to None upon termination of Tk through an appropriate destroy()!
> (maybe this should be considered a bug? who should know about it?)

I know, I should have read the source. Or the docs.
In Tkinter.py I found that Tk.destroy() *does* reset Tkinter._default_root
when appropriate. But elas, this function doesn't get called when you
press the window-close-button without having redefined the window-close
protocol:(. Still a bug in my opinion, but a subtle one. The Tkinter C
module (_tk) provides its own window-close handler. And this one, as can
be read in Lund's excellent albeit slowly growing Tkinter Introduction
docs, doesn't tidy-up the python/tk widget hiearchy. But even if it did,
and according to Lund's docs in future versions it will, this wouldn't
reset Tkinter._default_root to None.

So all in all, just use rootWindow.destroy instead of rootWindow.quit
as command for buttons used to end your script. And use rootWindow.destroy
as the command for the window-close protocol (see Lund's docs).

I sure hope I have it right this time:)

-- 
groetjes, carel



From robin at jessikat.demon.co.uk  Fri Jun 11 13:52:18 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Fri, 11 Jun 1999 18:52:18 +0100
Subject: Embedded Tcl for Python
References: <lJT73.2677$nn.808830@news.shore.net>
 <3760F0A1.1DDC@mailserver.hursley.ibm.com>
Message-ID: <B57RNPASzUY3EwGq@jessikat.demon.co.uk>

In article <3760F0A1.1DDC at mailserver.hursley.ibm.com>, Paul Duffin
<pduffin at mailserver.hursley.ibm.com> writes
>Michael P. Reilly wrote:
>> 
>> A few weeks ago, someone asked if there was a Tcl module for Python.
>> People responded (including myself) that Tkinter uses a Tcl/Tk
>> interpreter, but you couldn't get rid of the annoying window.
>> 
>> I've since written an extension ("pytcl") that creates Tcl interpreter
>> objects.  As part of that, I have also rewritten _tkinter to use these
>> new objects.
>> 
>> I've had a short discussion with Guido and he's interested in the
>> idea.  Now I need some alpha testers; I only have access to UNIX boxes,
>> and very few types at that.
>> 
>> The _tkinter module has the same functionality as before, including the
>> same threading issues.  I've include Deiter's Tcl/Tk8.1 patch but have
>> not had the opportunity to test that yet (right not it is just do-no-
>> damage).
>> 
>> Contact me at arcege at shore.net if you are interested in helping me test
>> these modules.
>> 
>
>I was interested in doing something similar in Tcl (e.g. creating a
>python interpreter object) and I was wondering if someone could give
>me some pointers as to how to go about doing it. 
>
Paul I created a socket linked python interpreter and have used it with
reasonable success from Tcl. Latest code is at

http://www.jessikat.demon.co.uk in the file pyserver.zip

this may not be as efficient as havin the in core thing, but gets round
the problem of threading model problems etc. 
-- 
Robin Becker



From news at dorb.com  Thu Jun 24 14:09:53 1999
From: news at dorb.com (Darrell)
Date: Thu, 24 Jun 1999 14:09:53 -0400
Subject: Environment inheritance under Windows NT 4, Attempt 2
References: <Pine.SGI.3.95q.990623204139.149495A-100000@masox202>
Message-ID: <aBuc3.2343$U6.9529@newsr2.twcny.rr.com>

Works fine for me.
Use the same case for the python version.
> C:\>python -c "import os; print os.environ['WINDIR']"

--
--Darrell

Milton L. Hankins <mlh at swl.msd.ray.com> wrote in message
news:Pine.SGI.3.95q.990623204139.149495A-100000 at masox202...
> Since I've not yet gotten a response to my first post, here's an
> elaboration.
>
> Windows NT Workstation 4.00.1381, SP5.
>
> 1.  Right click on My Computer, choose Properties, Environment tab.  Note
> the value of the System Variable "windir" and press Cancel.  (On my
> system, this is "C:\WINNT".)
>
> 2.  Open a Command prompt.
>
> 3.  Here's a transscript of my CMD session, you can follow along:
>
> -----------------------------------------------------
> C:\>echo %WINDIR%
> C:\WINNT
>
> C:\>python -c "import os; print os.environ['windir']"
> C:\WINNT
>
> C:\>set WINDIR=foo
>
> C:\>echo %WINDIR%
> foo
>
> C:\>python -c "import os; print os.environ['windir']"
> C:\WINNT
>
> -----------------------------------------------------
>
> What is going on here?  I wrote a program using VC5 to extract a variable
> (1) directly from environ and _environ and (2) by calling getenv().  All
> three reported "foo".  (Source provided below.)
>
> After rummaging through the Python source, it appears that os.environ is
> provided through the posix module, which in turn gets the environment data
> from the "extern char** environ".
>
> Could this have something to do with Microsoft's run-time DLLs?  I noticed
> that Cygwin B19 used to exhibit this behavior as well, but it seemed to go
> away in B20.1.
>
> Python picks up the environment correctly if I go into the Control Panel,
> System, Environment, and click OK.  But I have to do this each time I log
> in to NT.
>
>
> ----------------------------------------------------------------------
> #include <stdlib.h>
> #include <stdio.h>
> #include <assert.h>
> #include <string.h>
>
> extern char** environ;
>
> #define BIG 80
>
>
> int envIndex(const char** envArray, const char* key) {
> int i;
> char keyPrime[BIG];
>
> strcpy(keyPrime, key);
> strcat(keyPrime, "=");
>
>
> for (i = 0; envArray[i] != NULL; i++) {
> if (strstr(envArray[i], keyPrime) == envArray[i]) {
> break;
> }
> }
>
> return i;
> }
>
>
> // Don't screw with my return value...
> const char* myGetEnv(const char** envArray, const char* key) {
> int i;
>
> i = envIndex(envArray, key);
>
> if (envArray[i] == NULL) return NULL;
> else return (envArray[i]+strlen(key)+1);
> }
>
>
> void printenvMulti(const char* key) {
> char* s;
>
> s = myGetEnv(_environ, key);
> if (s == NULL) printf("%s is not in _environ\n", key);
> else printf("myGetEnv(_environ, \"%s\") --> %s\n", key, s);
>
> s = myGetEnv(environ, key);
> if (s == NULL) printf("%s is not in environ\n", key);
> else printf("myGetEnv(environ, \"%s\") --> %s\n", key, s);
>
> s = getenv(key);
> if (s == NULL) printf("%s is not available to getenv()\n", key);
> else printf("getenv(\"%s\") --> %s\n", key, s);
> }
>
>
> int main(int argc, char** argv) {
> assert(argv[1] != NULL);
>
> printenvMulti(argv[1]);
> return 0;
> }
> ----------------------------------------------------------------------
>
> --
>               Milton L. Hankins              \\  ><> Ezekiel 18:21 ><>
> Software Engineer, Raytheon Systems Company  //  <mlh at swl.msd.ray.com>
>       http://amasts.msd.ray.com/~mlh         \\  RayComNet  7-225-4728
>
>
>





From graham at sloth.math.uga.edu  Thu Jun  3 11:21:44 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 3 Jun 1999 15:21:44 GMT
Subject: Python 2.0 (perhaps off topic)
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
Message-ID: <7j66i8$il5$4@cronkite.cc.uga.edu>

Paul Prescod (paul at prescod.net) wrote:
: Personally, I wouldn't mind if Python ran on the JVM or had real GC. I
: just think that both sides need to be told. Python is the way it is not by
: accident but through design decisions intended to keep it simple to use
: and embed.

Why does having GC or implementing on top of the JVM make things more
complex, or restrict embeddability?

graham

-- 
                          So I turned around
                    And 40,000 headmen bit the dirt
                      Firing twenty shotguns each
                        And man it really hurt



From schorsch at schorsch.com  Wed Jun 16 19:17:26 1999
From: schorsch at schorsch.com (Georg Mischler)
Date: Wed, 16 Jun 1999 23:17:26 GMT
Subject: do anonymous pipes normally work on NT?
References: <4D0A23B3F74DD111ACCD00805F31D8100DB90DE0@RED-MSG-50>
Message-ID: <7k9b9u$2ti$1@nnrp1.deja.com>

Bill Tutt wrote:
>
>
> radbit at my-deja.com wrote:
> >
> > maybe I'm missing something about what you are looking for:
> > here are 2 test scripts that work as anonymous pipes:
> >
>
> That's not the question...
> Anonymous pipes on NT work wonderfully. (just not
> the way we're expecting from this particular Python context)

Well, at least he solved *my* problem! And after fiddling a
little, it even works with my cat.py as well, and should
work with arbitrary child processes. I'll just have to steal
a readline() method somewhere to make it practical...


The Documentaion provided with VC++ about this stuff
it not very helpful, so you have to try and guess.
Here's what I figured by now: The CreateProcess()
function has several ways to pass handles to child
processes. The way I tried it (passing through
StartupInfo.hStdXxx) only seems to work with disk files,
and not with pipes.

The other way is to assign the handles to the stdxxx
of the current process, and inherit them to the
child. This does work with pipes, and is fundamentally
incompatible with the first approach. In fact, as soon
as you just tell it you want to pass any handles in the
StartupInfo, by assigning STARTF_USESTDHANDLES to it's
dwFlags attribute, inheritance is dead. Still, if
you *want* to pass (file) handles through the
StartupInfo object, you nonetheless *need* to set the
bInheritHandles parameter of CreateProcess() to TRUE.
Go figure...

The least obvious thing though, which I only found
by comparing the working code with mine very carefully,
is that for inheritance to work, the dwCreationFlags
must *not* contain win32process.DETACHED_PROCESS.
The VisualStudio docs (and consequently the win32proc
help file) tell us that this would deny acess to the
*console* of the parent process. In fact, it appears
to cut all connections between the two processes.

many thanks for everybody's assistance in figuring
this out!

> Calling SetStdHandle() is particularly silly as well,
> you shouldn't need to overwrite the current stdout just
> to get this to work.

In fact, This is the way I saw it in several C++ examples
by now, and my own observations make me think that it's
just needed if you want to inherit *pipe* handles.


> > win32file.WriteFile(std_out, 'this method works....')
>
> Calling GetStdHandle shouldn't be necessary either. The CRT
> already grabbed that information for us when Python started up.
>
> The question is why does sys.stdout.write() not seem to work
> without -u when

In my latest experiments this *does* work if the parent
process manages to get the handles across. CreateProcess()
has so many mutually exclusive combinations of parameters
that it's just hard to find the right ones for a particular
purpose.

I don't mind setting and resetting the std handles in the
parent, as long as I get the result I want. But then, I'm
also trying to get along with as little knowledge as possible
about Windows programming. I assume you look at it from a
slightly different angle.


Thanks again to everybody!

-schorsch

--
Georg Mischler  --  simulation developper  --  schorsch at schorsch.com
+schorsch.com+  --  lighting design tools  --  http://www.schorsch.com/


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From rozzin at geekspace.com  Thu Jun 24 14:23:51 1999
From: rozzin at geekspace.com (Joshua Rosen)
Date: Thu, 24 Jun 1999 14:23:51 -0400
Subject: How to obtain an instance's name at runtime?
References: <7kqpu8$7u1$1@nnrp1.deja.com> <37718424.C70BB9DE@geekspace.com> <7ksgad$nvp$1@news.udel.edu>
Message-ID: <377277B7.AC477B79@geekspace.com>

Terry Reedy wrote:
> As near as I can remember, only three types of objects have definition
> names (in readonly? __name__ attributes): modules, functions, and
> classes.  All others are anonymous, though they may or may not have one
> or more namespace references.
> 
> While named objects usually get a namespace (reference) name equal to
> the definition name, there is no necessary connection.  Consider:
> 
> def f(): pass
> g=f
> del f
> 
> Now the function which internally thinks of itself as 'f' is referred to
> externally as 'g' and only as 'g'.

	Yes, it makes sense for functions, classes, and modules to all have names,
because they are all -definers-, which is what I think I was trying to say.
	Instances aren't `definers'--they don't act as the root of the definition for
`what something is'.

		-Rozzin.




From mwh21 at cam.ac.uk  Fri Jun 11 09:44:44 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 11 Jun 1999 14:44:44 +0100
Subject: Python as a shared library on Unix
References: <376102C9.5648@mailserver.hursley.ibm.com>
Message-ID: <m3vhcuam2b.fsf@atrus.jesus.cam.ac.uk>

Paul Duffin <pduffin at mailserver.hursley.ibm.com> writes:

> Is it possible to build Python itself as a small executable and a
> shared library, rather than an archive library which of
> course results in a monolithic executable. 

Not really. For a start look at 

http://x35.deja.com/[ST_rn=ps]/getdoc.xp?AN=468116075&CONTEXT=929108272.1543635130&hitnum=0

HTH
Michael

> I know how to get the modules built as shared libraries but cannot get
> Python built as a shared library.
> 
> -- 
> Paul Duffin
> DT/6000 Development	Email: pduffin at hursley.ibm.com
> IBM UK Laboratories Ltd., Hursley Park nr. Winchester
> Internal: 7-246880	International: +44 1962-816880



From hinsen at cnrs-orleans.fr  Mon Jun  7 04:31:17 1999
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 07 Jun 1999 10:31:17 +0200
Subject: problem with static linking
References: <Pine.SGI.3.92.990602171041.19399A-100000@sea30>
Message-ID: <m3yahw1kfe.fsf@chinon.cnrs-orleans.fr>

Xiaohong Yuan <xhyuan at oe.fau.edu> writes:

> I am working on Python c extensions on QNX. Since QNX doesn't support
> dynamic linking. I have to link my c extension module staticly with python
> interpreter. But in my c extension module, I have external variable
> declaration, which makes it fail to be linked with Python interpreter.
> There must be some way to solve this, could anybody help me with this?

Why do you need the external declaration? Where is this external variable
ultimately defined?
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------



From m01ymu00 at cwcom.net  Sat Jun 12 15:41:27 1999
From: m01ymu00 at cwcom.net (Daniel Faulkner)
Date: Sat, 12 Jun 1999 20:41:27 +0100
Subject: Browsers
Message-ID: <3762B7E7.CB001E3D@cwcom.net>

I have just been messing around with some network modules and I was
wondering how you would go about turning HTML into a basic picture (you
know just get rid of the HTML tabs make some of the writing bold and
include some spaces/center it nothing to hard or complicated) I have
messed around with some of the parsers but they don't seem to make a
huge difference. Any tips?

Dan




From hniksic at srce.hr  Wed Jun  2 17:37:09 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 02 Jun 1999 23:37:09 +0200
Subject: pickle vs .pyc
References: <7j43pn$55l$1@mlv.mit.edu> <m3pv3eian1.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <87vhd69tdm.fsf@pc-hrvoje.srce.hr>

Michael Hudson <mwh21 at cam.ac.uk> writes:

> Random aside: something fishy's going on when I try to try to marshal
> *arrays* (as opposed to mere lists):

The documentation for marshal says:

    Not all Python object types are supported; in general, only
    objects whose value is independent from a particular invocation of
    Python can be written and read by this module. The following types
    are supported: `None', integers, long integers, floating point
    numbers, strings, tuples, lists, dictionaries, and code objects
    (...)



From thstr at serop.abb.se  Wed Jun  9 09:45:07 1999
From: thstr at serop.abb.se (Thomas S. Strinnhed)
Date: Wed, 09 Jun 1999 15:45:07 +0200
Subject: Python/C API and setattr
References: <Pine.SUN.3.95-heb-2.07.990609160058.9041A-100000@sunset.ma.huji.ac.il>
Message-ID: <375E6FE3.A5CDA87@serop.abb.se>

Hi,

Thank you for a fast response.

Moshe Zadka wrote:
[my original message]
> You should return Py_None from a function. Every function called by
> Python has the following prototype:
> PyObject *f(PyObject *self, PyObject *args)
> (Actually, this is a lie. But a small white Knuth one)
> --
> Moshe Zadka <mzadka at geocities.com>.
> QOTD: My own exit is more likely to be horizontal then perpendicular.

Well I thought so too, but actually it must be something else,
cause it doesn't go away!! 

static PyObject *counter_setattr(counterobject *self, char *name,
PyObject *v)
{
	if(strcmp(name, "value") == 0)
	{
		self->value = (int)PyInt_AsLong(v);
		Py_INCREF(Py_None); 
		return Py_None;	
	}
	Py_INCREF(Py_None);
	return Py_None; 
}
Just doesn't help!!

Could there be something about this line;
	self->value = (int)PyInt_AsLong(v); 
(self->value is an int and v is PyObject*)

And about the prototype:
The few examples I've found about setattr are just like I've used it:
compare my prototype
static void counter_setattr(counterobject *self, char *name, PyObject
*v)
with this one from Objects\xxobject.c
static int xx_setattr(xxobject *xp, char *name, PyObject *v)

I think my main question would be:
 Where to find detailed documentation/information about this?
and subquestion asked standing on my bare knees (?? :-)
 What's happening here?? (Here beeing my code)

Admit it's tricky??

Many Thanks
 -- Thomas S. Strinnhed, thstr at serop.abb.se



From akuchlin at mems-exchange.org  Fri Jun  4 13:37:16 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: Fri, 4 Jun 1999 13:37:16 -0400 (EDT)
Subject: Python 2.0
In-Reply-To: <7j90va$6lk$1@cronkite.cc.uga.edu>
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7j90va$6lk$1@cronkite.cc.uga.edu>
Message-ID: <14168.3788.596777.267373@amarok.cnri.reston.va.us>

Graham Matthews writes:
>I have no plans to implement a GC for Python. I have already implemented
>enough standard collectors for one lifetime. Besides which someone far
>more intimate with the implementation details of Python would take less
>time. That's not a cop out. I just don't have the time.

       Someone once added full GC to Python by adding a separate
garbage collector thread, at the cost of a sizable speed hit and
requiring threads.  That might at least prove a good starting point,
and perhaps some optimization work could make it a practical
proposition.

	  Unfortunately I'm unable to pull up any postings in DejaNews
with a pointer to the code (though I can get the previous GC
flamewar). I vaguely recall the code was on a host in .hr, but can't
track it down.  Anyone remember that

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
I want to know the truth, however perverted that may sound.
    -- Stephen Wolfram



From arcege at shore.net  Tue Jun 22 09:43:12 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 22 Jun 1999 13:43:12 GMT
Subject: Creating a new type with an extension module
References: <Ctp*vl52n@news.chiark.greenend.org.uk> <rvy*CQ92n@news.chiark.greenend.org.uk>
Message-ID: <QpMb3.908$7X1.231730@news.shore.net>

Sion Arrowsmith <siona at chiark.greenend.org.uk> wrote:
: In article <Ctp*vl52n at news.chiark.greenend.org.uk>,
: Sion Arrowsmith  <siona at chiark.greenend.org.uk> wrote:
:>http://www.python.org/doc/current/api/newTypes.html is unhelpful
:>in the extreme -- do I have to resort to reading source?

: It appears that I do, but thanks to everyone who pointed me in
: the direction of suitably readable source. I've got what I
: wanted working -- it would just be nice to understand why it
: works....

Mark Lutz's _Programming Python_ book explains this quite well (I still
use these parts today in fact).  He has two chapters on embedding and
extending Python.

But the documentation on extending Python is pretty well known
deficiency.  Luckily it is far easier than most languages, both to
accomplish and to understand (yet more kudos to Guido).  Actually, in
responce to your earlier posting, I started writing a _SHORT_ webpage
explaining what is needed, which I hope to have available later today.
I plan (and hope) to make it into a How-To later.

  -Arcege




From graham at sloth.math.uga.edu  Wed Jun  2 16:39:49 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 2 Jun 1999 20:39:49 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <87n1ykr0z9.fsf@ev.netlab.co.jp> <7ivs9q$dc5@news.dns.microsoft.com> <87k8toqv70.fsf@ev.netlab.co.jp> <aahzFCnx28.1Jp@netcom.com>
Message-ID: <7j44ql$908$1@cronkite.cc.uga.edu>

Aahz Maruch (aahz at netcom.com) wrote:
: In addition to Paul's excellent comments, if one is writing an
: interactive process, one cannot afford the unpredictability of random
: garbage collection.  Garbage collection could run in a thread, but then
: one is restricted to environments where threads are well-supported.

You appear to be making an argument like "when a GC occurs the system 
pauses, and this is bad in an interactive environment". If that's the
case then this is an old chestnut. Modern GC implementations do not
have such pauses. Look up papers on real-time garbage collection for
example.

graham
-- 
             Like a bird on a wire
        Like a drunk in a midnight choir
            I have tried in my way
                  To be free



From hjstein at bfr.co.il  Tue Jun 15 17:46:56 1999
From: hjstein at bfr.co.il (Harvey J. Stein)
Date: 16 Jun 1999 00:46:56 +0300
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <3762988E.C94AA252@iname.com> <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu> <3763C746.5CA8FE78@iname.com> <m3u2sby4gf.fsf@soma.andreas.org> <87pv2zjhhl.fsf@home.ivm.de> <3764F3FE.F7CBE6BA@acm.org> <0Sgs3BA5FPZ3Ew6E@jessikat.demon.co.uk> <3765B3B8.E45F373B@spots.ab.ca> <slrn7mbse1.mb7.wtanksle@dolphin.openprojects.net>
Message-ID: <m2wvx5b0hb.fsf@blinky.bfr.co.il>

wtanksle at dolphin.openprojects.net (William Tanksley) writes:

 > On Tue, 15 Jun 1999 01:54:53 GMT, Maxwell Sayles wrote:
 > >Robin Becker wrote:
 > 
 > >> postfix, prefix, infix etc etc are isosemantic
 > 
 > >for myself and the others who might not know... can we get an example of
 > >each?
 > >and i remember someone mentioned parenthesized vs non-parenthesized...
 > 
 > Infix (ML and Prolog):
 > 
 >   3 + (4 minus 10) == 1
 > (parenthesis are not optional)
 > 
 > Prefix (C, Lisp, Scheme):
 > 
 >   function(that(x(1)))
 > or:
 >   (do-this (do-that (x 1)))
 > (Parenthesis are not optional)
 > 
 > Postfix (Forth, Postscript):
 > 
 >   3 4 + SWAP MOD do-that
 >   rinse on   agitate  10 seconds   rinse off
 > (Parenthesis are not optional)
 > 
 > As a common thread, note that in none of the languages are parenthesis
 > optional.  In Postfix parenthesis are not optional because they have nno
 > meaning; in the other languages they're assigned arbitrary meaning.

This isn't a completely fair example.  Postfix can get away without
parentheses only when a) everything that's not a number is a function
and b) the arity (# of arguments) of each function is known and fixed.

Given such restrictions one can also drop the parentheses in prefix
notation.

Infix still requires parentheses even given the above restrictions.

-- 
Harvey J. Stein
Bloomberg LP
hjstein at bfr.co.il



From wtanksle at dolphin.openprojects.net  Mon Jun 14 19:07:06 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Mon, 14 Jun 1999 23:07:06 GMT
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <3762988E.C94AA252@iname.com> <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu> <3763C746.5CA8FE78@iname.com> <m3u2sby4gf.fsf@soma.andreas.org> <87pv2zjhhl.fsf@home.ivm.de> <ekvn1y2efz0.fsf@gnoll.pvv.ntnu.no>
Message-ID: <slrn7mb2oq.lpd.wtanksle@dolphin.openprojects.net>

On 14 Jun 1999 21:31:47 +0200, Stig Hemmer wrote:
>> Prefix rocks, infix is crap.

>(think I (meant you (and (rocks prefix) (is-crap infix))))

prefix rocks infix crap is   and false =

...as long as we're fighting...  :-)

>Stig Hemmer,
>Jack of a Few Trades.

(I like all the languages I know, except Fortran.  I even like C++,
although I wish they could break backwards compatibility.)

-- 
-William "Billy" Tanksley
Utinam logica falsa tuam philosophiam totam suffodiant!
   :-: May faulty logic undermine your entire philosophy!



From catlee at my-deja.com  Wed Jun 23 14:04:10 1999
From: catlee at my-deja.com (catlee at my-deja.com)
Date: Wed, 23 Jun 1999 18:04:10 GMT
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0o3q.toe.bernhard@alpha1.csd.uwm.edu>
Message-ID: <7kr7ii$dss$1@nnrp1.deja.com>

> |>> vim 5.x version.)
> |
> |:map! += ^[:s/\([a-zA-Z_][a-zA-Z0-9_]*\)$/\1=\1+/^M:noh^MA
> |
> |Where   ":" means go into vim Commandmode
> |        "^[" means Esc, you have to enter it pressing Ctrl-v and
then Esc
> |        "^M" means "Return', you have to enter pressing Ctrl-v and
then Return

I've got the following in my .vimrc file:
au BufReadPost * if b:current_syntax == "python"
au BufReadPost * 	map! += ^[:s/\(\S*\)$/\1=\1+/^M:noh^MA
au BufReadPost * endif

What this does is check to see if the current syntax is python.  The
mapping hopefully covers more complex
expressions than yours does, it takes all the previous non-whitespace
characters as the expression to copy.  I wonder if there's a way to use
the vim syntax parsing mechanism to determine what an expression is?

Cheers,
Chris


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From gherman at my-deja.com  Wed Jun  2 05:49:34 1999
From: gherman at my-deja.com (Dinu C. Gherman)
Date: Wed, 02 Jun 1999 09:49:34 GMT
Subject: Pictures of JPF1 available
Message-ID: <7j2und$fp9$1@nnrp1.deja.com>

Hello,

please have a look at a few pictures I've taken during the JPF1
French Python meeting last week-end in Paris. Guido and Aaron
were there, plus others. The page below will also provide some
comments about how the pictures were taken.

  http://starship.python.net/crew/gherman/photography/jpf1/

Enjoy,

Dinu

--
Dinu C. Gherman       :  Mit Berlin kannste mir jagen!
................................................................
LHS International AG  :  http://www.lhsgroup.com
8050 Zurich           :  http://www.zurich.ch
Switzerland           :  http://pgp.ai.mit.edu
                      :  mobile://49.172.3060751
......................:.........................................
Vote against SPAM on http://www.politik-digital.de/spam


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From Lyle.Johnson at p98.f112.n480.z2.fidonet.org  Wed Jun 30 23:43:05 1999
From: Lyle.Johnson at p98.f112.n480.z2.fidonet.org (Lyle Johnson)
Date: Thu, 01 Jul 1999 04:43:05 +0100
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
Message-ID: <0000160e@bossar.com.pl>

From: Lyle Johnson <jlj at cfdrc.com>

Kevin,

Right now each instance has its own copy of the message table, and this is
set up in the class constructor. But I would prefer that the message table
data belonged to the class itself (as it is in the C++ FOX, static class
data).

Lyle

Kevin Dahlhausen wrote:

> Lyle Johnson <jlj at cfdrc.com> wrote:
>
> >Just to add fuel to the fire, we now have a Python interface to FOX, a
> >cross-platform GUI toolkit:
> >
> >    http://www.cfdrc.com/pub/FOX/fox.html
>
> How did you get around the requirement of  preprocessor
> table definitions for message handlers?
>
> (I'm downloading it now....)



From paul at polariscomm.com  Thu Jun 10 18:56:48 1999
From: paul at polariscomm.com (Paul Stillwell)
Date: Thu, 10 Jun 1999 15:56:48 -0700
Subject: tkinter and events
Message-ID: <7oX73.898$q3.114540@news21b.ispnews.com>

Ok, I am missing some critical piece of information so maybe someone out
there can help me.  I want to be able to click on an object on the canvas
and reference variables within that object, but I am not sure how to do it.
When I try it, I always get the values from the last object of that type
that I created.  For example, I have the following code:

from Tkinter import *
from Canvas import *

class MyTestObject:

    def __init__(self, Parent, Domain, State):
        self.Parent = Parent
        self.Domain = Domain
     self.State = State
     self.Canvas = Parent
     Distance = 5
     self.x = 370
     self.y = 30 + (Distance * Domain)
     self.Begx = self.x - 320
     self.Begy = self.y - 10
     self.Endx = self.x + 320
     self.Endy = self.y + 10
        self.Box = Rectangle(Parent, self.Begx, self.Begy, self.Endx,
self.Endy)
        self.text = CanvasText(Parent, self.x, self.y, text = "Domain = %x,
State = %s" % (Domain, State))
        self.Canvas.bind('<Double-1>', self.DoubleClickHandler)

    def DoubleClickHandler(self, event):
        print "Clicked on domain %x" % self.Domain

class Demo(Frame):

    def __init__(self, parent = None):
        Frame.__init__(self, parent)
        self.pack(fill = BOTH, expand = YES)
        self.Canvas = Canvas(self, width = "7.5i", height = "6i", background
= "white", scrollregion = (0, 0, "7.5i", "100i"))
        self.Canvas.ScrollY = Scrollbar(self, orient = VERTICAL)
        self.Canvas['yscrollcommand'] = self.Canvas.ScrollY.set
        self.Canvas.ScrollY['command'] = self.Canvas.yview
        self.Canvas.ScrollY.pack(side = RIGHT, fill = Y)
        self.Canvas.pack(side = RIGHT, fill = BOTH, expand = YES)

and I do the following at the Python command prompt (tester is the module I
created that has the above code):

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> from tester import *
>>> root = Tk()
>>> app = Demo(root)
>>> o1 = MyTestObject(app.Canvas, 0x10, 50)
>>> o2 = MyTestObject(app.Canvas, 0x20, 70)

So, when I run the code and double click on o1, I get:

>>> Clicked on domain 20

which is the last object I created even though I double clicked on the first
object.  What am I doing wrong?

Please help!

Paul
paul at polariscomm.com





From panakal at bigfoot.com  Fri Jun  4 19:31:30 1999
From: panakal at bigfoot.com (ADP)
Date: Fri, 4 Jun 1999 16:31:30 -0700
Subject: Python 1.5 help
References: <7j6uht$q1tk@eGroups.com>
Message-ID: <7j9nhk$6sd$1@news.fm.intel.com>

O'Reilly publishing has a fairly recent book called 'Learning Python' that
is an extremely readable introduction to the language.  It also has good
further references.







From hyoon at bigfoot.com  Thu Jun 10 15:03:31 1999
From: hyoon at bigfoot.com (Hoon Yoon)
Date: Thu, 10 Jun 1999 15:03:31 -0400
Subject: IDLE interacting w Shell
Message-ID: <37600C03.A5B4FE28@bigfoot.com>

Hi,

   I am trying to use IDLE somewhat like Emacs. I can't use Emacs
because some of Numpy Code will stop and kill the shell on Emacs when
certain errors occur. Not to mention IDLE is a lot lighter.
  Every time I hit F5 or 'Edit|Run module' I get an output window. I
want to get the run results and direct to shell like Emacs' C-c C-c,
instead of getting an output window.
  for exaple: if I have following in Editor and press F5
import os
  I would like to see following working on the shell:
>> dir(os)

  Any help would be much appreciated!

-- 
*****************************************************************************
S. Hoon Yoon (Quant) Merrill Lynch Equity Trading, 
yelled at yahoo.com hoon at bigfoot.com(w)
"Miracle is always only few standard deviations away, but so is
catastrophe."
* Expressed opinions are often my own, but NOT my employer's.
"I feel like a fugitive from the law of averages."    Mauldin
*****************************************************************************
-------------- next part --------------
A non-text attachment was scrubbed...
Name: vcard.vcf
Type: text/x-vcard
Size: 202 bytes
Desc: Card for Hoon Yoon
URL: <http://mail.python.org/pipermail/python-list/attachments/19990610/a14edf19/attachment.vcf>

From arcege at shore.net  Mon Jun 21 17:39:31 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Mon, 21 Jun 1999 21:39:31 GMT
Subject: while(s=infile.readline()):     # reading input
References: <376EA6E2.4E5EE527@starvision.com>
Message-ID: <niyb3.814$7X1.211285@news.shore.net>

Angus MacKay <amackay at starvision.com> wrote:

: why can't I have syntax like this?
: python is written in C after all isn't it?

: this works fine:
:   while infile.readline():
:     print "Hi!"

: this doe not:
:   while s = infile.readline():
:     print s

: which is okay, but alas this does not work:
:   while(s=infile.readline()):
:     print s

: is there any way to avoid using the ugly:
:   while 1:
:     s = infile.readline()
:     if not s: break
:     print s

Please read the FAQ 6.30 "Why can't I use an assignment in an expression?"
(http://grail.cnri.reston.va.us/cgi-bin/faqw.py?req=show&file=faq06.030.htp).
You can also read numerous discussions on this question from past years
on Dejanews (http://www.dejanews.com/).

As for your other question about chomp; this question was also asked in
this newsgroup a year or two ago, you can probably find that discussion
on Dejanews as well.  One basic answer is the difference between CISC
and RISC: CISC stands for "complex instruction set computer" and RISC
is "reduced instruction set computer".  Perl is CISC-oriented, and has
specialized functions and operations like chomp();  Python is
RISC-oriented, giving you a simplier, more flexible language, at the
perceived expense of not having "nifty functions."

For that reason and many, many others, "things" like assignment
expressions and specialized core language constructs just aren't there
in Python (there are always exceptions tho).  Then there is Python 2.0
coming in the future (don't ask when); and we trust Guido not to make
the mistake of Perl5.

  -Arcege




From tcpeter at my-deja.com  Tue Jun  8 19:27:13 1999
From: tcpeter at my-deja.com (tcpeter at my-deja.com)
Date: Tue, 08 Jun 1999 23:27:13 GMT
Subject: In what order does Python read dictionary keys?
References: <7jjnt5$snk$1@nnrp1.deja.com> <375D8077.EA539C6D@pop.vet.uu.nl>
Message-ID: <7jk8s8$3ma$1@nnrp1.deja.com>

In article <375D8077.EA539C6D at pop.vet.uu.nl>,
  M.Faassen at vet.uu.nl wrote:
> tcpeter at my-deja.com wrote:
>>  when I ran the code, the options were
presented
> > in the desired order. (A trifle creepy, if you ask me.  Turing would
be
> > delighted.)  Anyway, how does Python decide the order. Thanks.
>
> Hi there,
>
> The answer is that Python doesn't have any set orders for
dictionaries.
> The internal hashing algorithm the python dictionary uses has no
> guaranteed order. So basically you shouldn't depend on any particular
> order it produces.

OK.  So is there a way to ensure a specific order?  In this instance
I've given it's not particularly critical, but there are times where it
might be.  Any ideas?


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From cmedcoff at my-deja.com  Sun Jun  6 10:30:47 1999
From: cmedcoff at my-deja.com (cmedcoff at my-deja.com)
Date: Sun, 06 Jun 1999 14:30:47 GMT
Subject: Looking for for some ideas for argument processing
References: <7jcdhc$htd$1@nnrp1.deja.com> <3759A481.135D19CF@arlut.utexas.edu>
Message-ID: <7je0mn$v7d$1@nnrp1.deja.com>

I didn't even know that was there.  Still a newbie I guess.

I also just realized I can simulate a C-style for loop using a while
loop in Python.  Silly me..

Regards,
Chuck


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From tim_one at email.msn.com  Tue Jun 15 01:28:57 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 15 Jun 1999 01:28:57 -0400
Subject: fork()
In-Reply-To: <u3d7yzn1np.fsf@ares.sys1.ngo.okisoft.co.jp>
Message-ID: <000901beb6ef$f73c0f60$069e2299@tim>

[Hisao Suzuki]
> ...
> [quoting Stroustrup] and in the C++ 3rd Ed. (1997):
>    "When an object is about to be recycled by a garbage
>     collector, two alternatives exist:
>       [1] Call the destructor (if any) for the object.
>       [2] Treat the object as raw memory (don't call its
>           destructor).
>     By default, a garbage collector should choose option (2)
>     because objects created using _new_ and never _delete_d are
>     never destroyed.  Thus, one can see a garbage collector as a
>     mechanism for simulating an infinite memory." (C.9.1.3
>     Destructors)
>
> Python's __del__ is conceptually equivalent to C++'s destructor.

This is true, but Python has nothing equivalent to the C++ "delete"
operator, and *everything* in Python acts as if it had been created by the
C++ "new" operator. Stroustrup is only talking about C++ objects explicitly
created by "new" but never explicitly destroyed by "delete".  C++ has
several other ways for objects to come in and out of existence that don't
involve "new" or "delete" (block-scope auto; static; anonymous temps in
expressions), and rigidly defines when destructors are invoked for those
(more rigidly than Python, which actually opens some uses for destructors in
C++ that can't quite be done in Python).

> (Otherwise we would not rely on self.close() at __del__ in
> Lib/tempfile.py!)  The above phrase `(be) never deleted' for C++
> can be safely translated as `(be) part of cycles' for Python.
> Thus, Guido's idea is quite consistent and orthodox in this
> regard.

If there's an analogy to be drawn here, it would be that since everything in
Python acts as if created by C++ "new" and can never by explicitly
destroyed, Stroustrup is arguing that Python should never invoke a
destructor period.

The contexts are so different, though, that this analogy is as silly as it
sounds <wink>.

As a practical matter, there's a universe of difference between a user
explicitly forcing heap-based storage via C++ "new" and a Python programmer
who happens to create a cycle by design or accident.  For example, the
former is apparent via static inspection of program text, while the latter
is impossible to determine statically.

> It will be acceptable to most of the current working Pythoneers,
> including me, even if they are not fans of the Lots of Intolerable Stupid
> Parentheses ;-)

But now you're getting *really* practical <wink>.  I'm still most concerned
that (besides it being an incoherent mess <0.9 wink>) objects that aren't
themselves in a cycle may nevertheless by reachable only from an unreachable
cycle, and in that case it seems intolerable to say the (in Evan's nice turn
of phrase) "dangly bits" won't have destructors invoked either.  "Well,
sure, open file objects get closed by magic -- unless they happen to be
buried in some object reachable only via an unreachable cycle."  That's a
true stmt today, and not sure there's real gain if it continues to be true
tomorrow.

> By the way, C++'s destructor and Python's __del__ have been
> proved useful in practice, particularly in conjunction with the
> `resource acquisition is initialization' technique.  On the
> other hand, such a finalizer as found in Java is almost useless
> and unreliable because of its unpredictability.  I have never
> made use of finalize() in Java for years, and I don't know any
> skilled Java programmer substantially making use of finalize().

All different mechanisms, though.  Useful in C++ because block-scope autos
are reliably destructed upon block exit, and at least current C++ rigidly
defines the order in which they're destructed (so cyclic autos have *some*
ground to stand on).  Useful in Python because RC gets close to the same
effect (but less flexibly & less reliably).

> Further the C++ 3rd Ed. says:
>    "It is possible to design a garbage collector to invoke the
>     destructors for objects that have been specifically
>     `registered' with the collector.  However, there is no
>     standard way of `registering' objects.  Note that it is
>     always important to destroy objects in an order that ensures
>     that the destructor for one object doesn't refer to an
>     object that has been previously destroyed.  Such ordering
>     isn't easily achieved by a garbage collector without help
>     from the programmer."

Java solved that one -- for all the good it did <wink>.

> All in all, we need a sort of control or programmability over
> the behavior of our programs when a fixed, built-in mechanism of
> the language cannot solely address the problem very well.
> Suppose that Python recycles the memory of unreachable cycles of
> objects without calling __del__.  What if cycles that contain
> `registered' objects are saved from deallocation so that
> clean-up or other arbitrary actions can be performed?
>
> Such an action will break, say, an unreachable doubly-linked
> tree into separated nodes.  Then the nodes will be destroyed by
> a normal process including invocations of __del__.  To make this
> occur,
>  (a) the programmer must register the root of the tree with a
>      function to break up the tree, which will be invoked later
>      automatically by the garbage collector, or
>
>  (b) the programmer must register the root of the tree, and at
>      certain point of the program she/he must explicitly invoke
>      a function which retrieves the trees unreachable except
>      from the `registration' server and breaks them up.
>
>  (In any case, unreachable trees that contain no `registered'
>   objects will be recycled with no invocations of __del__.)
>
> I am not sure which option is more useful in practice, but
> either of them would be acceptable since they are entirely
> optional and compatible with the current semantics of Python.
> With no zapping in the formal semantics, they are conceptually
> clean and their implementations will be straightforward (thus
> they are also very Pythonic ;-)

I'm getting very keen on *some* scheme that gives motivated users control
over what happens when dead cycles are found, but Guido really wants to
track only dicts and handing the user a list/enumeration of disembodied
dicts isn't giving them much to work with <wink>.

> As for a reference to option (b), see, say, R. Kent Dybvig's
> ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
> which I have mentioned before in this newsgroup, though it is
> written for Scheme and actually the coming version of GNU guile
> implements guardians, i.e. `registration' servers described in
> it.

Indeed a very general mechanism!  Combining it with sensible __del__ rules
for non-cycles could well lead to something better than Scheme and Python
<0.5 wink>.

> Note that the guardians may be implemented somehow in 100%
> pure Python even if it doesn't have a garbage collector.  The
> sys.getrefcount built-in function will be used then.

Don't think so.  Knowing the refcount doesn't tell you whether a thing is
trash, e.g.

class A: pass
a = A()
a.seq = [a] * 1000000
del a

and there's no general way to trace the objects reachable from a given
object without a protocol for the reachable objects to cooperate in
revealing who is directly reachable from them (it can be laboriously
accomplished for objects of  builtin types now by exploiting deep knowledge
of the way they happen to be implemented, but toss in an extension type and
you're dead).

In any case, implementing M&S in Python is even slower than implementing it
in C <wink>.

much-ado-over-garbage-ly y'rs  - tim





From jan at q-bus.de  Tue Jun  8 06:59:35 1999
From: jan at q-bus.de (Jan Walter)
Date: Tue, 8 Jun 1999 10:59:35 GMT
Subject: I'll pay $40 in e-gold if you'll fix this bug.
References: <7jigrv$ppo$1@xs3.xs4all.nl>
Message-ID: <375CF797.FF0D6436@q-bus.de>

Hi,

Maybe you get confused a bit by the bash!!!

Try following:
bash$ tcsh
tcsh: cd /tmp/
tcsh: mkdir dirOne
tcsh: mkdir dirTwo
tcsh: mkdir dirTwo/dirThree
tcsh: ln -s dirTwo/dirThree/ .
tcsh: cd dirThree/
tcsh: pwd
/tmp/dirTwo/dirThree

Do the same with the bash (removing the directories of course):
bash$ cd /tmp/
bash$ mkdir dirOne
bash$ mkdir dirTwo
bash$ mkdir dirTwo/dirThree
bash$ ln -s dirTwo/dirThree/ .
bash$ cd dirThree/
bash$ pwd
/tmp/dirThree

The tcsh gives the right answer. You are in the /tmp/dirTwo/dirThree directory. If you try
to move a file with "mv file.txt ../dirOne/file.txt" you can?t do that because there is no
directory dirOne !!!

Python works like tcsh. If you try the following you get the right answer:
bash$ cd /tmp/dirThree/
bash$ python
Python 1.5.1 (#1, May  6 1998, 01:48:27)  [GCC 2.7.2.3] on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import os
>>> os.getcwd()
'/tmp/dirTwo/dirThree'

My suggestion:

You can test if there are symbolic links in your path (e.g. os.path.islink("/tmp/dirThree")
within Python) or you can prove that /tmp/dirThree is in reality /tmp/dirTwo/dirThree and
life with that fact ...

Cheers,

Jan






From Klaus.Schilling at home.ivm.de  Tue Jun 22 04:05:12 1999
From: Klaus.Schilling at home.ivm.de (Klaus Schilling)
Date: 22 Jun 1999 10:05:12 +0200
Subject: need some fast balanced-tree datastructure
References: <376ED372.FA73994F@cs.ucsd.edu>
Message-ID: <874sk0k6dj.fsf@home.ivm.de>

Pavel Velikho <pvelikho at cs.ucsd.edu> writes:

> Hi,
>     I am looking for a fast implementation of incremental container that
> 
> maintains an ordering on the keys... (i.e. I want an AVL, Red-Black, or
> anything similar). I have tried the b+tree in Python but its not fast
> enough.
> Is there anything like that available as a C extension module?

There is an avltree implementation in C by Ben Pfaff on the gnu
ftp site ftp.gnu.org/pub/gnu . Maybe this can be swigged to python.

Klaus Schilling



From Angus.MacKay at p98.f112.n480.z2.fidonet.org  Tue Jun 29 01:57:39 1999
From: Angus.MacKay at p98.f112.n480.z2.fidonet.org (Angus MacKay)
Date: Tue, 29 Jun 1999 06:57:39 +0100
Subject: clobbering
Message-ID: <0000158f@bossar.com.pl>

From: Angus MacKay <amackay at starvision.com>

is there any way to get python to warn you when you clobber
a function refrence with a variable?

I was writing this and it vexed me for some time:

for dir in sys.path:
   dir(...)
...

it is silly user error but a flag to say warn if clobbering functions
with variables (or at lease builtin functions).

cheers, Angus.



From Newbie at p98.f112.n480.z2.fidonet.org  Tue Jun 29 20:55:49 1999
From: Newbie at p98.f112.n480.z2.fidonet.org (Newbie)
Date: Wed, 30 Jun 1999 01:55:49 +0100
Subject: Decent Tutorial
Message-ID: <000015af@bossar.com.pl>

From: "Newbie" <8499 at greenhead.ac.uk>


Where can I find a decent tutorial which is 
for beginners to Python, preferably including 
some documented source code and avaliable for 
download.

Thanks.



From wtanksle at dolphin.openprojects.net  Thu Jun  3 23:07:53 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Fri, 04 Jun 1999 03:07:53 GMT
Subject: All this self dot is making me...
References: <Vn653.468$HD.99317@typhoon-sf.snfc21.pbi.net>
Message-ID: <slrn7lego8.932.wtanksle@dolphin.openprojects.net>

On Wed, 2 Jun 1999 01:59:35 -0700, Jim Rudnicki wrote:
>To the beginner, having to type all these self. 's in class methods seems a
>bit painful.  So all three of these methods below do the same thing.  That

>For those of us who have less cartilage left in their hands, would it be
>acceptable to use "my." or better "me."  Or is self. strongly ingrained?
>The two less characters add up and the keys are on stronger fingers.

It's absolutely acceptable to use something else.  It's good to make it
clear, so "self" is the best, "this" works well, "me" and "my" are fine,
and "snowman" is abominable.

The reason you don't want to use external functions is that object
orientation doesn't work on them -- you can't override them.

>Jim

-- 
-William "Billy" Tanksley
Utinam logica falsa tuam philosophiam totam suffodiant!
   :-: May faulty logic undermine your entire philosophy!



From zigron at jps.net  Sun Jun 20 23:05:41 1999
From: zigron at jps.net (Zigron)
Date: Sun, 20 Jun 1999 20:05:41 -0700
Subject: python15.dll
Message-ID: <1%hb3.804$o7.114638@typhoon-sf.snfc21.pbi.net>

    Can it be obtained /alone/, without the rest of the distribution? I
goofed, and it
got screwed. I'm lazy, don't wanna download the whole 5meg distrib :)

Thanks!

--Stephen, who goes back to desperately trying to uncross his eyes as he
works
with non-existing c/c++ knowledge to build a wxWindows add-on into it, then
SWIGify it into Python. ugh! Shoot me?







From rdev at my-deja.com  Wed Jun 30 14:02:46 1999
From: rdev at my-deja.com (rdev at my-deja.com)
Date: Wed, 30 Jun 1999 19:02:46 +0100
Subject: Reflection in Python?
Message-ID: <000015c8@bossar.com.pl>

From: rdev at my-deja.com

In article <7ldglr$ita$1 at nnrp1.deja.com>,
  anders777 at my-deja.com wrote:
>  ...How do I do the
> equivalent of what Java calls reflection? I want
> to do the following:
>
> test = MathTest("testAdd")
>
> where test.run() will run the "testAdd" method
on
> the MathTest object?  I also want to be able to
> tell the user the name of the object (MathTest)
> and the method in case the method fails.  Any
> thoughts?
>
If I understand what you are trying to do, it's
easy in Python.  Assuming MathTest is an
object instance:
test = MathTest.testAdd
will return you a bound method object which you
can then call as:
test()
Getting the name of the object class and method
is also easy.  There are several ways to do this:
1) Define a doc string within the method by
   putting a quoted string as the first line
   under the def line.  You can then refer to it
   as "test.__doc__".  This could say, for
   example, "Method bar of class foo".
2) "test.__name__" will return the name of the
   method.  "test.im_class.__name__" will
   return the name of the class that the method
   is defined within.

Example:
class MathTest:
  def testAdd(self, parm1, parm2):
      "Method testAdd of Class MathTest"
      result = parm1 + parm2
      return result
<<< mt = MathTest()
<<< test = mt.testAdd
<<< test.run(2,2)
4
<<< print test.__doc__
Method testAdd of Class MathTest
<<< print test.__name__
testAdd
<<< print test.im_class.__name__
MathTest


I hope that helps.

- Roger




Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From Johann.Spies at p98.f112.n480.z2.fidonet.org  Wed Jun 30 04:14:49 1999
From: Johann.Spies at p98.f112.n480.z2.fidonet.org (Johann Spies)
Date: Wed, 30 Jun 1999 09:14:49 +0100
Subject: Please help me to understand classes
Message-ID: <000015ab@bossar.com.pl>

From: Johann Spies <jhspies at futurenet.co.za>

On Tue, 29 Jun 1999, Stidolph, David wrote:

> Finally, a simple question I can answer!
> 
> First of all, class methods like "druk" need to have a parameter declared in
> them called self (or whatever you want to call it).  Secondly you need to
> call the method with open/close parens:
> 
> class A:
>   def druk(self):
>     print 'Jy is in Class A'
> 
> class B:
>   def druk(self):
>     print 'Jy is in Class B'
> 
> a = A()
> c = B()
> c.druk()
> Jy is in Class B
> a.druk()
> Jy is in Class A
> 
> I hope this helps,

Thanks David.  It helped.  After your message I went back and read the
tutorial again and saw what you mean.

Johann



 --------------------------------------------------------------------------
| Johann Spies                                 Windsorlaan 19              |
| jhspies at futurenet.co.za                3201 Pietermaritzburg             |
| Tel/Faks Nr. +27 331-46-1310		       Suid-Afrika (South Africa)  |
 --------------------------------------------------------------------------

     "Because he himself suffered when he was tempted, he
      is able to help those who are being tempted."       
                                        Hebrews 2:18 



From arcege at shore.net  Wed Jun 30 11:04:54 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 30 Jun 1999 15:04:54 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773c3a0.102825445@news.oh.verio.com> <xPPc3.535$or1.22866@typhoon-sf.snfc21.pbi.net> <xJRc3.292$Of3.67094@news.shore.net> <00f701bec212$f6a83050$f29b12c2@secret.pythonware.com>
Message-ID: <qmqe3.337$6M6.119017@news.shore.net>

Fredrik Lundh <fredrik at pythonware.com> wrote:
: Michael P. Reilly <arcege at shore.net> wrote:
:> But.. the argument over (these two) GUIs is not exactly a valid one.
:> Tkinter was written using the Tcl interface not the Tk/C API.  If
:> Tkinter was rewritten to use the C API, it probably wouldn't be slower
:> than wxPython.

: The interface is only part of the story; many complaints
: seem to come from the fact that Tk uses idle tasks to
: redraw things.  When an "expose" event arrives from
: the windowing system, Tk calls "after_idle", rather
: than redrawing the widget at once.  And idle tasks
: are only executed when there are no other events
: to take care of...

Hmm... something about that argument makes it circularly invalid.
Non-idle event a comes in, but must be handled after idle event d.
Event d can only execute when event a is taken care of.  Sounds like
deadlock to me. ;)

Events (from Tk_MainLoop) are handled by Tcl_QueueEvent and
Tcl_ServiceEvent (handled by Tcl_DoOneEvent).  Yes, it handles idle
events IF those are the only ones being requested to be serviced:
(taken from tcl8.0.5/generic/tclNotify.c, lines 668-672, in the
function Tcl_DoOneEvent)

        /*
         * If idle events are the only things to service, skip the
         * main part of the loop and go directly to handle idle
         * events (i.e. don't wait even if TCL_DONT_WAIT isn't set).
         */

Otherwise the first thing called is Tcl_ServiceEvent.  Now yes, this
means that when it is servicing idle events, it will not check to see
if there are any pending events, but I see that as a different issue.

In fact, what might be objectionable, is there are two Tcl_ServiceEvent
calls (one on either side of the Tcl_WaitForEvent) before the
Tcl_ServiceIdle call, delaying the "do_idle" even more.

: Tk also uses double buffering for all widgets, which is
: quite expensive, especially since Tk allocates a new
: buffer for each update :-(

Yes, that could be a problem, but see what I mean down below about
additional layers.

:   In fact, since wxPython has to go thru extra layers, it
:> would probably be slower (Tk/C is written on the native API on all the
:> platforms).

: What native API?  As of 8.0, Tk still emulates all widgets
: except the scrollbar, and use a somewhat ineffective
: X emulation layer for graphics.  AFAIK, they haven't
: changed that in 8.1...

Exactly, Tk is using the native API: X11 (not Motif, not Xt, no
emulation other than its own), Windows, MacOS.  wxWindows requires GTK
or Motif on UNIX, requiring another level of software.  Motif is
licensed (but most vendors have gotten around that.  GTK+ 1.2.3 is 2.58
Mb with monthly updates (they do kindly give patch updates tho).

Also, Tk was developed on X11, where the window object created is
stored in the X server (another process), not in the application.  It
is standard practice to keep your own information about the objects you
request creation of.  This is no different from what wxWindows would
have to do using GTK or Motif, both using X11 (except that wxWindows
would probably have its own information as well as similar information
would be stored for widgets created from either GTK or Motif).

I'm not saying that Tk is the greatest thing since sliced bread, just
that it may not be as off-kilter as people think.  Tkinter is off in
that is uses the Tcl/Tk interpreter instead of the Tk/C API (but I
for one am not complaining ;).

  -Arcege




From hniksic at srce.hr  Sun Jun  6 09:27:53 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 06 Jun 1999 15:27:53 +0200
Subject: Looking for for some ideas for argument processing
References: <7jcdhc$htd$1@nnrp1.deja.com> <3759A481.135D19CF@arlut.utexas.edu>
Message-ID: <87iu91zcfa.fsf@pc-hrvoje.srce.hr>

"Thomas A. Bryan" <tbryan at arlut.utexas.edu> writes:

> cmedcoff at my-deja.com wrote:
> > 
> > I'm writing a script with is going to make heavy use of script
> > arguments.  I'd like to hear some ideas on some ways to support this.
> 
> I usually just use Python's getopt or DPyGetOpt.
> 
> The first is in the standard library.
> 
> The second is modeled after Perl's GetOpt::Long, which is very  
> flexible and supports many options.

Hey!  I've been meaning to write such a thing -- glad it's already
been written.

The chief problem with Python's getopt is that it doesn't support some 
nice features of GNU getopt, such as naming the options after the
arguments (e.g. wget URL -o outfile).



From aahz at netcom.com  Sat Jun 26 22:26:45 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 27 Jun 1999 02:26:45 GMT
Subject: Advice for Newbie
References: <377588E9.82E8F567@dsuper.net>
Message-ID: <7l4255$c22@dfw-ixnews13.ix.netcom.com>

In article <377588E9.82E8F567 at dsuper.net>, Dave  <redsand at dsuper.net> wrote:
>
>I read the Python homepage. I sounds promising. What books do you
>commend for a beginner?

If you already have experience with other programming languages, some
combination of the following: _Programming Python_, _Learning Python_,
and the on-line tutorial/documentation.  If you're not a programmer and
you're not in a hurry, wait for _Teach Yourself Python in 24 Hours_.
--
                      --- Aahz (@netcom.com)

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



From MHammond at skippinet.com.au  Tue Jun 29 06:27:45 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Tue, 29 Jun 1999 20:27:45 +1000
Subject: PythonWin VSS Integration
References: <002a01bec203$1fc90230$a9807018@yec1.on.wave.home.com>
Message-ID: <7la6vj$rh0$1@m2.c2.telstra-mm.net.au>

Mike C. Fletcher wrote in message <002a01bec203$1fc90230
[Laments about Pythonwin and Source Safe]

>So, any hints?  What does visual source safe integration mean?

Well, it _should_ give you some clues.  Note that you need to restart
Pythonwin before the option will take effect.  Also note that the file must
be marked read-only on the disk.

Once enabled, and assuming VSS is installed correctly, the most likely error
message you will see is something like "this directory is not configured for
VSS".  [Once the option is checked, you should get _some_ message on a
read-only file]

In this case, you need a "mssccprj.scc" file either in the file's directory
or a parent directory.  This is the same name and format as VB uses -
basically a Windows INI file.  A section [Python] with entry
"Project=ProjectName" is referenced.  The project name is the name of the
VSS project the Python code attempts to check the file out from.  If the
.scc file is in a parent directory, the correct relative VSS path is built -
so if your file system matches your VSS structure, you only need a single
.scc file in the VSS "root" directory.

Hope this makes sense.

Mark.
>
>Thanks,
>Mike
>





From evan at tokenexchange.com  Mon Jun 14 11:35:02 1999
From: evan at tokenexchange.com (Evan Simpson)
Date: Mon, 14 Jun 1999 10:35:02 -0500
Subject: fork()
References: <000601beb56d$829f6720$319e2299@tim>
Message-ID: <Ri993.1$V4.6985@news-dal.corridex.com>

Tim Peters wrote in message <000601beb56d$829f6720$319e2299 at tim>...
> [Evan Simpson]
>> ...
>> Perhaps I'm being naive, but I have the impression that when cycles
>> arise it's usually a simple accident and easy to fix once you know
>> you've done it.

> Exactly so:  you're being naive <wink>.

I rather thought so, although I did know about Guido's "native" cycles
already.

> Stuff "like that" pops up in all sorts of apps.  With practice you learn
> ways around it most often.  I've noted more than once that the resulting
> cycle-free design is often more robust and comprehensible than its cyclic
>ancestor, so I think it's a fruitful exercise to do this; doesn't always
> succeed, though, and when you're in a rush it's a pain in the ass to have
to
> bother.

"easy to fix" was a bit of an overstatement, but this is essentially what I
meant.

>> Would it be possible (and not too expensive) to distinguish actual cycle
>> members from dangly bits (maximal non-cyclic subgraphs)?
[Interesting SCC approach snipped]

> What you suggest is weaker, effectively lumping the SCCs into two gross
> classes (singletons and non-singletons), and nuking the latter first.

Sorry if I was unclear, but that's not what I meant by "dangly bits".
Rather, take the set of nodes from which you cannot reach a cycle.  Their
subgraph is a union of DAGs, which can be released normally.  In your
example, handing "c" a reference to "d<->e" converts it from a dangly bit to
a cycle-bridge.

On the other hand, the SCC approach does look stronger than this, since it
can handle cycle-bridges as well.  I'll happily take your word that
SCC-counting is likely to be too much trouble, but how about just doing an
SCC-analysis once M&S has discovered a cycle?  If this is all happening in
the background, with no deadlines, and with rare, naughty data structures,
then it doesn't have to be terribly efficient.

and-my-non-existent-code-can-be-maximally-inefficient-if-it-wants-ly y'rs
Evan Simpson





From greg.ewing at compaq.com  Wed Jun 16 19:15:08 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Thu, 17 Jun 1999 11:15:08 +1200
Subject: Module magic methods
References: <WgP83.7966$Yr2.129670@news2.rdc1.on.home.com> <m3g13wt71h.fsf@atrus.jesus.cam.ac.uk> <dkuhlmanFDA99M.6AH@netcom.com>
Message-ID: <37682FFC.FF1A59B7@compaq.com>

"G. David Kuhlman" wrote:
> 
>     import mymodule
>     #
>     # Do some stuff with mymodule
>     #
>     mymodule = None
> 
> Won't reference counting cause the space taken up by mymodule to be
> garbage collected?

Nope, because references all imported modules are kept
on a global list in the bowels of the interpreter, so
that the next time the module is imported you get a
reference to the previously loaded instance.

> this is what
> I'm going to tell them to do when they want to reclaim space in
> long running programs.

It won't work.

>     anInstance = mymodule.MyClass()
>     mymodule = None
>     anInstance.show()
> 
> Something must stay around so that I can call the methods defined
> MyClass.

The instance has a reference to the class, so as long as the
instance is alive the class will be too, even if you wipe
out the reference to it in the module where it was defined.

Greg



From gmcm at hypernet.com  Tue Jun  8 12:18:47 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 8 Jun 1999 11:18:47 -0500
Subject: more python to c/c++ questions.
In-Reply-To: <7jgu2u$19s6$1@rtpnews.raleigh.ibm.com>
Message-ID: <1283263515-67394082@hypernet.com>

Jr. King is asking something or other:

> Ok, I found some info on it but, it still doesn't work
> 
> I changed my c module to this...
> <snip>
> static PyObject* objPass(PyObject *self, PyObject* args){
>     PyObject *result = NULL;
>     PyObject *temp;
> 
>     if (PyArg_ParseTuple(args, "O", &temp)) {
>    PyObject* xarg = Py_BuildValue("(s)","i");

You've just built a tuple. Probably not what you want.

>    xarg = PyObject_GetAttr(temp,xarg);  // not getting anything but
>    null int val; PyArg_ParseTuple(xarg,"i",&val);
>  }
>     Py_INCREF(Py_None);
>     result = Py_None;
>     return result;
> }
> 
> Python module snippet
> 
> class Cow:
>  i = 10
>  x = 0
>  str = "string"
>  def genie(self):
>   print str
> 
> testmeth.setCallback(setN)
> testmeth.testmeth(13,9)
> cow = Cow()
> testmeth.objpass(cow);

- Gordon



From n at n.com  Tue Jun  1 10:45:25 1999
From: n at n.com (Jr. King)
Date: Tue, 1 Jun 1999 10:45:25 -0400
Subject: Why won't this work???
Message-ID: <7j0rkr$1eeu$1@rtpnews.raleigh.ibm.com>

I know you are getting tired of me, but I have tried repeatedly to solve
this on my own and I can't get it to work.
I am trying to do a python callback to call from a c/c++ program.
Here is the c/c++ code

// learnPyth.cpp : Defines the entry point for the console application.
//

#include "stdafx.h"
#include "python.h"
#include "stdio.h"
#include <iostream.h>
#include <windows.h>

static PyObject *my_callback = NULL;

static PyObject* Test(PyObject *self, PyObject* args){
 cout<<"Test"<<endl;
 int command;
 int aoe;
    if (!PyArg_ParseTuple(args, "ii", &command,&aoe))
        return NULL;

 int xs(5);
 PyObject* xarg = Py_BuildValue("(i)",xs);

 PyEval_CallObject(my_callback, xarg);//python returns the PYNone thingie
here
 Py_DECREF(xarg);

 return args;
}

static PyObject* setCallback(PyObject *self, PyObject* args){
    PyObject *result = NULL;
    PyObject *temp;

    if (PyArg_ParseTuple(args, "O:set_callback", &temp)) {
        if (!PyCallable_Check(temp)) {
            PyErr_SetString(PyExc_TypeError, "parameter must be callable");
            return NULL;
        }
        Py_XINCREF(temp);         /* Add a reference to new callback */
        Py_XDECREF(my_callback);  /* Dispose of previous callback */
        my_callback = temp;       /* Remember new callback */
        /* Boilerplate to return "None" */
        Py_INCREF(Py_None);
        result = Py_None;
    }
    return result;
}

static PyMethodDef meths[]={
 {"testmeth", Test, METH_VARARGS},
 {"setCallback", setCallback, METH_VARARGS},
 {NULL,  NULL}  /* sentinel */
};

int main(int argc, char* argv[])
{
 Py_Initialize();
 char input[200];
 int ret;
 FILE *fp;
 cout<<Py_GetPath()<<endl;
 fp = fopen("beer.py","r+");
 Py_InitModule("testmeth", meths);
 Py_InitModule("setCallback",meths);
 if(fp == NULL)
  cout<<"Didn't open file";
 else{
  ret = PyRun_AnyFile(fp,"beer.py");
  cout<<ret<<endl;
  fclose(fp);
  Py_Exit(0);
 }
 return 0;
}

Here is the python code

#! /usr/bin/env python
# By GvR, demystified after a version by Fredrik Lundh.
#import sys
import testmeth
import setCallback
n = 20
#if sys.argv[1:]: n = int(sys.argv[1])

def setN(x):
 x = n
 print n

def bottle(n):
    if n == 0: return "no more bottles of beer"
    if n == 1: return "one bottle of beer"
    return str(n) + " bottles of beer"

setCallback.setCallback(setN)
testmeth.testmeth(13,9)
for i in range(n):
    #print bottle(n-i), "on the wall,"
    #print bottle(n-i) + "."
    #print "Take one down, pass it around,"
    print bottle(n-i-1), "on the wall."

print n

thanks again.







From discman_64 at hotmail.com  Sun Jun 20 22:07:52 1999
From: discman_64 at hotmail.com (John Smith)
Date: Sun, 20 Jun 1999 22:07:52 -0400
Subject: Better Online Docs?
Message-ID: <7kk6tj$b42$1@news.igs.net>

Hi.. I was just wondering if there are any other online docs for python on
internet for free?





From proteus at cloud9.net  Sat Jun  5 14:48:40 1999
From: proteus at cloud9.net (Michael Muller)
Date: Sat, 05 Jun 1999 14:48:40 -0400
Subject: ask help for problem with static linking (fwd)
References: <Pine.SGI.3.92.990604230514.11561A-100000@sea30>
Message-ID: <37597107.5FB5324C@cloud9.net>

Xiaohong Yuan wrote:
[snip]
> I am sorry I didn't make it clear. In my C extension module, I declare
> some extern variables. Those extern variables are for my own application.
> They can not be found in Python interpreter source code. So when I do
> static linking, my c extension module fails to be linked with the python
> interpreter. I don't want to add into Python source code the definition of
> my extern variables. There might be some way to wrap my c extensions such
> that extern variables would not cause a problem. My platform is QNX, my
> compiler is watcom compiler. I don't think platform and compiler matter,
> the problem might be how should I deal with extern variable when writing
> python c extensions.
> 
> --Xiaohong Yuan

I am assuming that you mean that your externs are defined in some main
program that your extension is normally linked to, and are presumably
used
to share information between your extension and your main program.  If
this 
is the case syou have two options:

1) redesign your extension so that the information shared by your
external
   variables is shared through an API instead.  If necessary, this
   information can be stored statically in the extension library.

2) Create a stub file which is linked to your extension that provides
   bogus (but useable) definitions for these variables.

Method 2 is a hack.  Information that needs to be shared between
components
should almost always be shared using an API.  This allows a "loose
coupling" of components, so that you can make implementation changes 
in one without affecting the other.

=============================================================================
michaelMuller = proteus at cloud9.net | http://www.cloud9.net/~proteus
-----------------------------------------------------------------------------
If you are not willing to control your own mind, there are plenty of
other
people who are willing to do it for you.
=============================================================================



From aahz at netcom.com  Fri Jun 11 14:02:49 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Fri, 11 Jun 1999 18:02:49 GMT
Subject: Any Production ZOPE sites we can visit?
References: <37614A32.79FE554E@metricom.com>
Message-ID: <aahzFD6Csp.236@netcom.com>

In article <37614A32.79FE554E at metricom.com>,
Michael Kersey  <mkersey at metricom.com> wrote:
>
>Are there many production Zope sites visible on the www?

http://www.searchbutton.com/

We're still looking for Python programmers, BTW....
-- 
                      --- Aahz (@netcom.com)

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

"That doesn't stop me from wanting to work all three of them over with
the clue stick for while, with no safewords allowed."  --abostick



From gmcm at hypernet.com  Sat Jun 12 20:45:43 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 12 Jun 1999 19:45:43 -0500
Subject: force sequenz with only one item?
In-Reply-To: <375FA9AE.6299E9DD@phoenix-edv.netzservice.de>
Message-ID: <1282887498-90012442@hypernet.com>

Holger Jannsen asks:
> 
> I think it's easy to solve, but I couldn't get it right now.
> Take a look at the command-line source below:
> 
> >>> A = 0
> >>> B = 1
> >>> seq=raw_input("?")
> ?A,B
> >>> print seq
> A,B
> >>> print eval(seq)
> (0, 1)
> >>> seq=raw_input("?")
> ?A
> >>> print eval(seq)
> 0
> >>> A in seq
> Traceback (innermost last)
:...
> TypeError: 'in' or 'not in' needs sequence right argument
> >>>

The reason "A,B" gets evaluated to the sequence (0,1) is because of 
the comma. "(A,)" would get evaluated to (0,), while "[A]" would 
evaluate to [0]. If you want to always end up with sequences, you 
should probably eval it and test it:

  if type(evaledinput) not in (type(()), type([])):
      evaledinput = (evaledinput,)

Single element tuples need the extra comma to distinguish them from
a simple grouping, as in  4 * ( 3 + 7 ). It would cause great
wailing and gnashing of teeth if it yeilded (10, 10, 10, 10).

even-worse-than-while-1-ly y'rs

- Gordon



From pavlos at gaaros.com  Sat Jun 19 15:04:58 1999
From: pavlos at gaaros.com (pavlos at gaaros.com)
Date: Sat, 19 Jun 1999 19:04:58 GMT
Subject: HTML bettering?
In-Reply-To: <14186.24371.249981.437742@weyr.cnri.reston.va.us>
References: <14186.24371.249981.437742@weyr.cnri.reston.va.us>
Message-ID: <7kgpkq$kma7@eGroups.com>

 <14186.24371.249981.43774- at weyr.cnri.reston.va.us> wrote: 
Original Article: http://www.egroups.com/group/python-list/?start=61289
> 
> Matej Cepl writes:
>  > goodwill, is it? But, to make Pythonians more happy, the story is as
> 
>   A worthy goal!  ;-)
> 
>  > html2ps which has much better output, but it numbers headings in
>  > different way (still unresolved issue, what it the top level -- H2 or

Or if your HTML is well structured you can use the html2pdf converter in the Zpdf (it is an alpha release) package 
http://starship.python.net/crew/pavlos

It does heading numbering too.

Pavlos




From aahz at netcom.com  Sun Jun  6 02:18:57 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Sun, 6 Jun 1999 06:18:57 GMT
Subject: Newbie question
References: <3759FF72.E393CAEA@massey.ac.nz>
Message-ID: <aahzFCw6vL.3Hn@netcom.com>

In article <3759FF72.E393CAEA at massey.ac.nz>,
yourname  <yourname at massey.ac.nz> wrote:
>
>I'm getting this error when I try to call a function
>
>Traceback (innermost last):
>  File "ftpclient", line 229, in ?
>    FTPLogin(oFtp,site,port,sUserID,sUserPassword) #site,port)
>NameError: FTPLogin

Try ftplib.FTPLogin() (or something like that).  In general, you need to
explicitly specify the namespace you're using.
-- 
                      --- Aahz (@netcom.com)

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

"Why do you like my boyfriend to tie you up and beat me.?"
    -- ELIZA generates a poly moment



From orlov at diasoft.ru  Wed Jun 23 09:48:16 1999
From: orlov at diasoft.ru (Oleg Orlov)
Date: Wed, 23 Jun 1999 17:48:16 +0400
Subject: set im_self
Message-ID: <7kqokb$7as$1@ns.mtu.ru>

Hello.
Is there any way to set im_self attribute (ie bind method object)?
Thanks.

--
====================
Oleg Orlov
orlov at diasoft.ru





From mukundd at my-deja.com  Wed Jun 30 15:17:28 1999
From: mukundd at my-deja.com (mukundd at my-deja.com)
Date: Wed, 30 Jun 1999 20:17:28 +0100
Subject: Problem embedding C++ code
Message-ID: <000015d1@bossar.com.pl>

From: mukundd at my-deja.com

Hi,

I am trying to write a wrapper around a library written in C++ and I am
running into lots of problems.

I am running Python 1.5.2 on SunOS 5.5.1 (compiled with SUN's cc and
linked with Sun's CC)

The wrapper is written in a file called link-wrap.c and currently it
contains only one very simple function.

CC -c -KPIC link-wrap.c -I/home/mdeshpan/SW/include/python1.5
"link-wrap.c", line 13: Warning: String literal converted to char* in
formal argument 1 in call to Py_BuildValue(char*, ...).
1 Warning(s) detected.
CC -G link-wrap.o -lapic -lpthread -o link.so

libapic.so is the library which I am trying to write the wrapper for,
it was compiled using Sun's CC compiler

when I try to import link module in python I get the following error..

daffy:~/Swig$python
Python 1.5.2 (#3, Jun 29 1999, 11:17:23) [C] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import link
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ImportError: ld.so.1: python: fatal: relocation error: file
./libapic.so: symbol _ex_rethrow_q: referenced symbol not found

any help would be greatly appreciated..

Thanks in advance

Mukund


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From mcfletch at home.com  Thu Jun 10 23:21:05 1999
From: mcfletch at home.com (Mike C. Fletcher)
Date: Thu, 10 Jun 1999 23:21:05 -0400
Subject: Broken 1.5.2 installation? Missing cw3215mt.dll
References: <000901beb3b2$baa6d980$329e2299@tim>
Message-ID: <01cb01beb3b9$7130fc70$a9807018@yec1.on.wave.home.com>

Of course, should have done that in the first place (been too long since a
reformat).  Second thing I should have done was to think a few seconds.  I'd
renamed the cygnus install's directory but hadn't searched the system32
directory where some ancient version of the TK/TCL dlls was hanging
around...

Thanks Tim, all working fine now, evil old TCL dlls banished to the aether,
Mike
______________________________________
   M i k e   C .  F l e t c h e r

       mcfletch at vrtelecom.com

        Designer, Researcher

----- Original Message -----
From: Tim Peters <tim_one at email.msn.com>
To: Mike C. Fletcher <mcfletch at home.com>; <python-list at python.org>
Sent: Thursday, June 10, 1999 22:33
Subject: RE: Broken 1.5.2 installation? Missing cw3215mt.dll


> > Problem is encountered when loading command-line Python or IDLE.  IDLE
is
> > non-functional, command line works fine after acknowledging the
> > missing DLLs (as far as I can see).
> > ...
> > Could be the Cygnus TCL is interfering I suppose.
>
> Aha!  Stop your grinnin' and drop your linen, Mike.  Having more than one
> version of tk*.dll or tcl*.dll anywhere in your entire *city* is like
> putting a male brain-eater in your left ear and a female brain-eater in
your
> right ear at the peak of brain-eater mating season, then betting you can
> keep them apart by force of moral suasion.  You should reformat your
drives
> and reinstall NT to be *sure* <wink>, but before that try uninstalling
every
> Tcl/Tk you have, "dir/s" your drives from the roots to make sure nothing
> matching tk*.dll or tcl*.dll survived, uninstall Python, and reinstall
> Python with the One True Tcl/Tk that comes with it.
>
> followups-to-cameron-laird<wink>-ly y'rs  - tim
>
>




From ivnowa at hvision.nl  Sat Jun  5 17:45:38 1999
From: ivnowa at hvision.nl (Hans Nowak)
Date: Sat, 5 Jun 1999 21:45:38 GMT
Subject: Deleting characters from a string
In-Reply-To: <87n1ye3g4v.fsf@pc-hrvoje.srce.hr>
References: <87n1ye3g4v.fsf@pc-hrvoje.srce.hr>
Message-ID: <199906052144.XAA15045@axil.hvision.nl>


On 5 Jun 99, Hrvoje Niksic wrote:

> string.translate can delete characters from a string, as well as
> translate them.  But what if I want *only* to delete them?  Currently
> I do this:
> 
> dummytable = string.maketrans('', '')
> 
> ...
>   host = string.translate(host, dummytable, ' "')
> ...
> 
> What I would really like to be able to do is write:
> 
>   host = string.translate(host, None, ' "')
> 
> Am I missing a more elegant solution?

Hmm... how about

host = filter(lambda c: not c in ' "', host)

Depends on how much you like, or hate, functional idiom, though. :)

--Hans Nowak (ivnowa at hvision.nl)
Homepage: http://fly.to/zephyrfalcon



From tgallati at mainnet.com  Tue Jun 15 20:58:27 1999
From: tgallati at mainnet.com (Tim Gallati)
Date: Wed, 16 Jun 1999 00:58:27 GMT
Subject: Cut down on writing repetitive code
Message-ID: <3766f6a8.19474000@news2.mainnet.net>

If you do a lot of repetitive typing, check out Perfect Keyboard Lite.
I use it everyday and consider it my most useful program.

It launches when you boot your computer and creates a file of
frequently used phrases, paragraphs, e-mail addresses, URLs and
macros.  They are activated in any application by typing a short
abbreviation.

Free 30 day trial. 
http://www.tricksandtrinkets.com/pk



From Evan.Simpson at p98.f112.n480.z2.fidonet.org  Tue Jun 29 06:00:14 1999
From: Evan.Simpson at p98.f112.n480.z2.fidonet.org (Evan Simpson)
Date: Tue, 29 Jun 1999 11:00:14 +0100
Subject: Please help me to understand classes
Message-ID: <00001590@bossar.com.pl>

From: "Evan Simpson" <evan at tokenexchange.com>

Your Pascal experience is showing <wink>.  While you can call a
parameterless function or procedure without parentheses in Pascal, Python
always requires them, since functions are first-class objects.

So when you write "a.druk", you are fetching the method "druk" from "a", but
then doing nothing with it.  You should write "a.druk()", which will call
the method with no parameters.

One reason for the required parentheses is the ability to do this:

f = a.druk  # get the method and make f refer to it.
f()  # call the method.

Johann Spies wrote in message ...
I have been playing around with python for some time now, but I am a
relative newbie as it is my hobby and normally I have very little time for
programming.  I am on leave now and want to write a program but I find
python a very difficult language to comprehend despite the fact that I
have a few years of experience in programming a language like Pascal.
Maybe it is because I have little experience in OOP.

Can somebody explain to me why the following program produces no output:
---------------------------------------------
class A:

    def druk():
        print 'Jy is in Class A'

class B:

    def druk():
        print 'Jy is in Class B'

a = A()
c = B()
c.druk
a.druk





From ivanlan at callware.com  Fri Jun 18 11:01:11 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Fri, 18 Jun 1999 15:01:11 GMT
Subject: Newbie: Truth values (three-valued logic)
References: <000f01beb7b8$2722a540$2ca22299@tim>
		<Pine.OSF.4.05.9906161106470.295-100000@pinax.Mathematik.Uni-Bielefeld.DE> <37697DFD.91C2B8F6@compaq.com>
Message-ID: <376A5F37.F13DB3CA@callware.com>

Hi All--

Greg Ewing wrote:
> 
> Olaf Delgado wrote:
> >
> > Next time I'll want to redefine 'and' and 'or'.
> 
> You could use '*' and '+', if you always know
> when you're dealing with a 3-valued boolean
> (troolean? trillian?).
> 

Triskelion!

> By the way, I think Python 2.0 should let you define
> new infix operators -- and design new characters to
> represent them! We all have bitmapped screens now --
> there's no reason why we should be restricted to a
> fixed character set any more...
> 

Ah, yes.  I can see it now.  Little Mayan glyphs peppering everyone's
screens.  ``The "'Ahaw" operator is logically equivalent to the 'doit'
command in GNUScript.''  World domination is close!

<pop-goes-the-ahaw>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From moshez at math.huji.ac.il  Fri Jun 11 08:52:14 1999
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 11 Jun 1999 15:52:14 +0300
Subject: urllib: an alternative URLOpener
Message-ID: <Pine.SUN.3.95-heb-2.07.990611155013.10151A-100000@sunset.ma.huji.ac.il>

Whom do I contact about adding the following code to urllib:

-------------->8-------------->8-------------------------
class QuietFancyURLOpener(FancyURLOpener):
        def prompt_user_password(self, host, realm):
                return None, None
-------------->8-------------->8-------------------------

(Yes, I realize I can put it in my own program. But it seems a generally
useful thing for non-interactive downloaders)
--
Moshe Zadka <mzadka at geocities.com>. 
QOTD: My own exit is more likely to be horizontal then perpendicular.





From MHammond at skippinet.com.au  Fri Jun 18 01:18:42 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Fri, 18 Jun 1999 15:18:42 +1000
Subject: do anonymous pipes normally work on NT?
References: <4D0A23B3F74DD111ACCD00805F31D8100DB90DE4@RED-MSG-50> <7kbq78$vgm$1@nnrp1.deja.com>
Message-ID: <7kckoh$ith$1@m2.c2.telstra-mm.net.au>

Inside win32_src.zip you should find win32process.i and
win32processmodule.cpp - the latter is generated by SWIG from the former.

Bill is working from the same archive, so it must be there!

The only thing I can think of is that you didnt grab the archives from my
starship page...

Mark.

Georg Mischler wrote in message <7kbq78$vgm$1 at nnrp1.deja.com>...
>But then, I just downloaded win32_src.zip and Pythonwin_src.zip
>and didn't find anything resembling a win32processmodule in there.
>Guess I am looking in the wrong place. Where is this module defined?






From tgm at math.uni-bonn.de  Mon Jun 28 05:03:34 1999
From: tgm at math.uni-bonn.de (Thomas Ackermann)
Date: Mon, 28 Jun 1999 11:03:34 +0200
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <Pine.GSO.3.96.990625170056.28322B-100000@saga3.Stanford.EDU>
Message-ID: <6pd7l7.9d.ln@final.math.uni-bonn.de>

When i need to do some GUI, i always prefered GTK (and the formidable
pygtk from James Henstridge) ...
As it uses gtk, i hope it will soon be available on all mayor
platforms.
Is anybody capable of comparing Tkinter, pygtk and wxWindows?

   Byebye,
-- 
  Thomas Ackermann | Tel. +49-(0)228/631369|73-7773 | <tgm at math.uni-bonn.de>
             finger tgm at rhein.math.uni-bonn.de for public key
               GNU LINUX Python gtk pygtk MySQL FUDGE GURPS



From claird at Starbase.NeoSoft.COM  Wed Jun 16 21:17:58 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 16 Jun 1999 20:17:58 -0500
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <37628EBF.207F7A8F@iname.com> <rnbn1y2n5jk.fsf@nortelnetworks.com> <m1909lpipq.fsf@brightstar.swin.de>
Message-ID: <7k9ic6$8oh$1@Starbase.NeoSoft.COM>

In article <m1909lpipq.fsf at brightstar.swin.de>,
Julian Einwag  <julian at brightstar.swin.de> wrote:
			.
			.
			.
>Do you know where you can get the Qt bindings? Im quite interested in 
>it.

<URL:http://
starbase.neosoft.com/~claird/comp.lang.python/python_GUI.html#PythonQt>
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From holger at phoenix-edv.netzservice.de  Fri Jun 18 12:01:40 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Fri, 18 Jun 1999 16:01:40 GMT
Subject: NT
Message-ID: <376A6D64.DF881D27@phoenix-edv.netzservice.de>

Hi,

another problem under NT: I'd like to message the user a longer description about the
programm with sys.stdout.write(usagetxt).
The user should have possibility to pipe the output like he
wants, e.g. with 'less' or 'more'.
But I get an error:

C:\myPython>installation15.py /? |more
Traceback (innermost last):
  File "C:\myPython\installation15.py", line 642, in ?
    switcher()
  File "C:\myPython\installation15.py", line 629, in switcher
    __usage()
  File "C:\myPython\installation15.py", line 189, in __usage
    sys.stdout.write(usagetxt)
IOError: [Errno 9] Bad file descriptor

If I start with 'C:\myPython>python installation15.py /? |more'
there's no problem and it works allright.

I've read the messages about NT and piping in this group, but I think
it's something differnt, ouh?!

hand (='have a nice day'!),
Holger



From morse at harborcom.net  Wed Jun 30 15:12:15 1999
From: morse at harborcom.net (Kevin Dahlhausen)
Date: Wed, 30 Jun 1999 19:12:15 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773c3a0.102825445@news.oh.verio.com> <xPPc3.535$or1.22866@typhoon-sf.snfc21.pbi.net> <xJRc3.292$Of3.67094@news.shore.net> <7laovo$c4s$1@Starbase.NeoSoft.COM> <377A24B7.AD6FAA1A@cfdrc.com>
Message-ID: <377a6bb0.10820509@news.oh.verio.com>

Lyle Johnson <jlj at cfdrc.com> wrote:

>Just to add fuel to the fire, we now have a Python interface to FOX, a
>cross-platform GUI toolkit:
>
>    http://www.cfdrc.com/pub/FOX/fox.html


How did you get around the requirement of  preprocessor 
table definitions for message handlers?

(I'm downloading it now....)





From gmcm at hypernet.com  Sat Jun  5 23:14:51 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 5 Jun 1999 22:14:51 -0500
Subject: Sockets and lemon curry
In-Reply-To: <qDf63.375$os.20566@typ42b.nn.bcandid.com>
Message-ID: <1283483353-54170129@hypernet.com>

Ian King writes:
 
> I'm trying to implement an rshd server in Python.  The problem seems
> to be with establishing the secondary socket for stderr.  I'm using
> a Linux 2.0 box as client, and a Windows machine as server

> ...  In experimenting, I've gotten things to work with utilities
> that don't require the second socket (e.g. rcp).  But the moment I
> try to establish the secondary socket, the client says, "protocol
> violation" and gives up.  Ideas?

/snip code/

> ...  This seems like a simple handshake, and if I use (for
> instance) the Windows machine as a client against the Linux server
> (and do a network trace), everything works and looks great.

Yes. On *nix, it takes root authority to create sockets on ports 
below 1024. No such limit on NT. You didn't say what 'port2' was, but 
it sure fits the symptoms...

- Gordon



From gmcm at hypernet.com  Fri Jun 18 18:48:53 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 18 Jun 1999 17:48:53 -0500
Subject: Installing ODBC on PythonWin
In-Reply-To: <7ke9kq$3c2$1@holly.csv.warwick.ac.uk>
Message-ID: <1282376107-120775232@hypernet.com>

Andrew Clover has Access problems:

>  I've got an Access database accessed by Python scripts under CGI.
>  However,
> sometimes, calls to odbc.odbc(dsn) hang up and never return.
> (Eventually the web server times out and kills the process.) This
> seems to happen randomly, but increasingly. Does anyone have any
> idea why? A fault with the odbc module, or Access, or what?

Repeat after me: "Access is NOT a server, Access is NOT a server".

It is a file access method. If you access it through Jet (ie, DAO or 
one of those) it is sort-of safe for multiuser usage. If accessed 
through ODBC it is most definitely absolutely NOT safe. It's not even 
safe for mutiple connections from one process.

Access is dandy on the desktop as an overgrown spreadsheet. It is the 
LAST thing you should use for CGI.

- Gordon



From misty at pao.fiba.tampere.fi  Thu Jun 17 02:16:55 1999
From: misty at pao.fiba.tampere.fi (Juhani Rantanen)
Date: 17 Jun 1999 06:16:55 GMT
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <rnbn1y2n5jk.fsf@nortelnetworks.com> <sFU93.1567$kS.233449@ptah.visi.com>
Message-ID: <slrn7mh4sb.b7u.misty@pao.fiba.tampere.fi>

On Wed, 16 Jun 1999 21:27:52 GMT, David Thornley <thornley at visi.com> wrote:
>I know little of this Qt animal.  Is it free?  Is it open source?
>Does it run on the Macintosh?  Is it easy to set up an interface with?

Qt will be soon licensed with an open-source license, though it will
be practically very restrictive (all changes must be distributed as
patches). Current version of Qt is not free software, and I am
surprised to find one who doesn't know about the flamewars... Try "qt
gtk free" in Altavista or Dejanews to find out:) 

I also suggest looking at Gtk+ <http://www.gtk.org/> that is used in
GIMP (free Photoshop clone for Unix/Linux http://www.gimp.org/), Gnome
(a free desktop environment for GNU project http://www.gnome.org/) .
Gnome folks have made a version of Guile with Gtk and Gnome bindings,
but this is not very widely tested AFAIK, and some higher-level
interface would be preferable (though now we have C-level API mapped
to scheme, the higher-level interface could be written in scheme). 

Gtk+ is available on Unix and Win32. Win32 support is not in the official
sources, but I even occasionally use a version of GIMP on Windows NT
and it is quite robust compared to some commercial programs on the
same operating system (though both Gtk+ and Gimp on that machine are
developer's versions from CVS). 

-- 
Juhani Rantanen
L?hderanta 20 G 55 Espoo  
 This message is best read on a 80x24 tty with 2 colors and a
 non-proportional font



From hniksic at srce.hr  Fri Jun  4 11:50:52 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 04 Jun 1999 17:50:52 +0200
Subject: threads
References: <3756C078.5D998D3B@icrf.icnet.uk> <14166.51558.850416.153999@bitdiddle.cnri.reston.va.us> <8790a1njeu.fsf@pc-hrvoje.srce.hr> <14166.60091.353043.829598@amarok.cnri.reston.va.us> <87k8tks2y5.fsf@pc-hrvoje.srce.hr> <14167.61710.929549.586641@anthem.cnri.reston.va.us>
Message-ID: <87aeug7yn7.fsf@pc-hrvoje.srce.hr>

"Barry A. Warsaw" <bwarsaw at cnri.reston.va.us> writes:

> You might be interesting in digging around for Greg Stein's
> free-threading patches.  Greg did a lot of work, I think in the
> Python 1.4 days, to remove the central interpreter lock and make all
> the data structures and such thread safe.

Why would I want to do that?  If Greg's patches haven't been applied
to the base, surely mine would not be either.  I certainly have no
wish to maintain a parallel version of Python.



From arcege at shore.net  Sat Jun 12 21:51:14 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Sun, 13 Jun 1999 01:51:14 GMT
Subject: tkinter and signals
References: <3762392C.AE94C44D@saiph.com>
Message-ID: <m8E83.3879$nn.924422@news.shore.net>

Pierre Imbaud <pierre at saiph.com> wrote:
: Hi everyone.
: Im quite new to tkinter. I try to make a program using tkinter, that
: is able to listen to some signals: read a pipe, perform periodic
: actions. I first tried with SIGALRM, with a widget changing fg color
: every second. I dont wait() for SIGALRM, just return in mainloop().
: the behavior I get is: I need the mouse to hover on the app, to get
: the SIGALRM signal. My diagnostic, right or wrong is: in the mainloop,
: the idle state is a wait() for a signal (dunno which) related to an X
: event. In this state, an unwaited signal just wont wake up the
: process.
: It just occurs to me threads might be the solution: for every possible
: signal source, spawn a thread that will explicitly wait for this
: signal? Im not familiar with thread either, so I better wait for a
: confirmation from someone who knows....
: Oh, btw, Im using Solaris 2.5.1 on X86 (its an Unix V.4)
: Thanks for answering!

I would suggest first looking into the after() and after_idle() methods.
They are designed for this type of operation.

There is also a lower-level interface, the _tkinter.createtimerhandler
function.

The after() and _tkinter.createtimerhandler() functions both take
a number of milliseconds and a function.

  f = Frame(None)
  oneminute = 60000
  def change_red(f=f):
    f.config(fd="red")
    f.after(oneminute, change_green)
  def change_green(f=f):
    f.config(fg="green")
    f.after(oneminute, change_red)
  f.after(oneminute, change_red)

  -Arcege





From mwh21 at cam.ac.uk  Wed Jun  9 06:57:44 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 09 Jun 1999 11:57:44 +0100
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> 	<7iqhkb$d7s@wiscnews.wiscnet.net> 	<7iu7ma$d33$1@cronkite.cc.uga.edu> 	<02hfonzsx2.fsf@rtp.ericsson.se> 	<02909zzou0.fsf@rtp.ericsson.se> 	<87zp2f4qrz.fsf@pc-hrvoje.srce.hr> 	<wkd7z858c3.fsf@turangalila.harmonixmusic.com> <14172.6652.550882.854800@bitdiddle.cnri.reston.va.us>
Message-ID: <m3yahtabfb.fsf@atrus.jesus.cam.ac.uk>

Jeremy Hylton <jeremy at cnri.reston.va.us> writes:

> >>>>> "DS" == Dan Schmidt <dfan at harmonixmusic.com> writes:
> 
>   DS> Hrvoje Niksic <hniksic at srce.hr> writes: | Kumar Balachandran
>   DS> <kumar*xspam*@*xspam*rtp.ericsson.se> writes: | | > What you
>   DS> said is quite wrong. Since Python provides closures, I can | >
>   DS> change my programming style to be completely different from the
>   DS> | > standard C or Fortran style and use a functional interface.
>   DS> | | Since when does Python provide closures?
> 
>   DS> The bytecodehacks module allows the creation of closures as I
>   DS> understand them.
> 
> Yes and no.  The bytecodehacks closures work right most of the time,
> but there some problems with them.  Hence, I would guess, the name
> "hacks."  

Quite. PLease note that the bytecodehacks resulted from me having an
idea that noone else seems to have had before and running away with it
as far as I could get it. The modules and functions that resulted from
this were really just illustrations of what the technique can do,
rather than solutions to any particular problem.

Having said that, if someone does find them useful, then that makes me
happy.

[explanantion of why "BCH closures" aren't quite like scheme closures]

> 
> def works():
>     sum = [0]
>     def add(x):
>         sum[0] = sum[0] + x
>     def sub(x):
>         sum[0] = sum[0] - x
>     cadd = closure.bind_locals(add)
>     csub = closure.bind_locals(sub)
>     cadd(2)
>     cadd(3)
>     csub(1)
>     return sum[0]

It probably wouldn't be all that hard to make bytecodehacks.closure
generate code like this automatically if it sees the substituted name
being mutated via STORE_FAST/STORE_GLOBAL.

Also note that the above code puts a mutable value into a code object,
something that isn't supposed to happen: (from the Language Reference)

| Unlike function objects, code objects are immutable and contain no
| references (directly or indirectly) to mutable objects.

Also Python expects to ba able to marshal code objects, so if you use
clsure.bind with an unmarshallable value there might be trouble.

Having said all that, it shouldn't cause any real problems (at least
not in 1.5.2).

Have a nice day!
Michael



From matomira at iname.com  Tue Jun 15 09:11:22 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Tue, 15 Jun 1999 15:11:22 +0200
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <3762988E.C94AA252@iname.com> <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu> <3763C746.5CA8FE78@iname.com> <m3u2sby4gf.fsf@soma.andreas.org> <87pv2zjhhl.fsf@home.ivm.de> <ekvn1y2efz0.fsf@gnoll.pvv.ntnu.no> <slrn7mb2oq.lpd.wtanksle@dolphin.openprojects.net> <37660FEC.D4F3371B@iname.com> <37661C02.9CA6E258@cfmu.eurocontrol.be>
Message-ID: <376650FA.1AED1606@iname.com>


Ian Wild wrote:

> Fernando Mato Mira wrote:
> >
> > William Tanksley wrote:
> >
> > > (I like all the languages I know, except Fortran.  I even like C++,
> > > although I wish they could break backwards compatibility.)
> >
> > Quiz: What do Scheme, Common Lisp, HPF and Java have in common?
>
> And when's Java*, the massively parallel version, coming out?

I seem to have forgotten C, too..






From gerrit.holl at pobox.com  Thu Jun 10 09:34:23 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Thu, 10 Jun 1999 15:34:23 +0200
Subject: Why is tcl broken?
In-Reply-To: <375FA62E.CCEF5079@cfmu.eurocontrol.be>; from Ian Wild on Thu, Jun 10, 1999 at 01:49:02PM +0200
References: <375F82C7.CF0F0477@iname.com> <375FA62E.CCEF5079@cfmu.eurocontrol.be>
Message-ID: <19990610153423.A23327@optiplex.palga.uucp>

On Thu, Jun 10, 1999 at 01:49:02PM +0200, Ian Wild wrote:
> From: Ian Wild <ian at cfmu.eurocontrol.be>
> Subject: Re: Why is tcl broken?
> Date: Thu, 10 Jun 1999 13:49:02 +0200
> To: python-list at python.org
> 
> Fernando Mato Mira wrote:
>  
> >   I'm trying to collect a list of all the _current_ issues making tcl
> > a bad language choice. I'd like concrete examples,
> 
> (I mean, it's theoretically possible that, in some
> parallel universe, there's a even problem for which
> C++ is the ideal choice.)
> 

Can I conclude that C++ isn't a good language? I'm a starting programmer: why
isn't C++ good for many problems?

-- 
The Dutch Linuxgames website.		De Nederlandse Linuxgames pagina.
Everything about games on Linux.	Alles over spelletjes onder Linux.
Site address:				http://linuxgames.nl.linux.org
Personal homepage:			http://nl.linux.org/~gerrit/



From ivanlan at callware.com  Sat Jun 19 17:17:12 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Sat, 19 Jun 1999 21:17:12 GMT
Subject: DATE ARITHMETIC
References: <7kejrv$e57$1@violet.singnet.com.sg>
Message-ID: <376C08D8.A9032304@callware.com>

Hi All--

Ajith Prasad wrote:
> 
> I need to do some simple date arithmetic - finding the number of days
> between any 2 days, the age of a person on a day given his birthday. The
> mxDateTime module downloadable from the Python website is more complex than
> what I need. Any other suggestions?

No.  Date arithmetic is by nature non-trivial.  Marc-Andre's mxDateTime
is perfectly suited for your application.

If you *really* *really* don't want to use it, though, look up "Julian
Date" and Scaliger (Joseph Justus) on the web.  Learn how to reduce
Gregorian dates to Julian Period dates, and once you do that you can add
and subtract dates as if they have just become big numbers.

Which they will have done, assuming you follow up on Julian Period
dates.

<8-K'an-12-Sotz'>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From matomira at acm.org  Mon Jun 14 11:46:23 1999
From: matomira at acm.org (Fernando D. Mato Mira)
Date: Mon, 14 Jun 1999 17:46:23 +0200
Subject: <language> broken
References: <1282792849-95706075@hypernet.com> <3764D7C3.718E125E@iname.com> <002f01beb64d$58e7b5f0$f29b12c2@pythonware.com>
Message-ID: <376523CF.607E59A@acm.org>

Fredrik Lundh wrote:

> as far as I can tell, you haven't even mentioned Python in any
> of your posts to comp.lang.python and the python language
> mailing list.  sure looks like the usual cross-posting time-wasting
> usenet troll to me, but maybe someone is abusing your user
> account?

No. The vague impression I got from python tells me it's a good
language, and their folk seem pretty much on track. As many of its users

seem to have the same kinds of needs the tcl people do, I thought
it would probably be the place to find knowledgeable people with
a balanced opinion.

>the other hand, if you're not interested in Python, maybe you
>could stop wasting our time?

If the subject does not interest you, you don't need to read.
Myself, I find it very enriching to meet people from other
`cults'. That's how you can learn new things.

You've been polite, but the last line is pretty ambiguous to me in tone,
given the limitations of this electronic
medium. And the same happened with some other person in the newsgroup.

Beware. That does not do Python any good. Suddenly, I have
the impression the group is populated with very excitable and
all-knowing `Linux kiddies' (obviouly, why didn't I think so? - I also
use Linux),
and that it's not one some place one should want to visit very often
(like comp.lang.c++, for example).
Just think about it.

And no, I don't see why I should use Python instead of other things.
Although seeing it everywhere (ILU, GNOME, etc.) has intrigued me,
although not
from a practical perspective, so I haven't really looked at it.
For me, it would be interesting to study it if there was something
`new', like Occam has CSP, Eiffel has assertions, Ada has tasking (and
ADTs were
`new', at the time I studied it, in1986), Haskell has monads, etc.

Regards,

--
Fernando D. Mato Mira
Real-Time SW Eng & Networking
Advanced Systems Engineering Division
CSEM
Jaquet-Droz 1                   email: matomira AT acm DOT org
CH-2007 Neuchatel                 tel:       +41 (32) 720-5157
Switzerland                       FAX:       +41 (32) 720-5720

www.csem.ch      www.vrai.com     ligwww.epfl.ch/matomira.html






From peterh at sapros.com  Wed Jun  2 22:37:51 1999
From: peterh at sapros.com (Peter Haight)
Date: Wed, 02 Jun 1999 19:37:51 -0700
Subject: idle processing 
Message-ID: <199906030237.TAA08709@wartch.sapros.com>

>is there an appropriate place to insert a method
>which does some calculations while the application
>is "idle" (user is not using it or just moving
>the mouse...) - somewhere within the tkinter-
>mainloop ? 

One way is to not use mainloop and call update and whatever other functions
that you want in your own loop like this:

root = Tkinter.Tk()
while 1:
    root.update()
    idle_processing()





From dalke at bioreason.com  Mon Jun 14 23:09:44 1999
From: dalke at bioreason.com (Andrew Dalke)
Date: Mon, 14 Jun 1999 21:09:44 -0600
Subject: Bug in `imp' module?
Message-ID: <3765C3F8.D3F8A4CD@bioreason.com>

Summary: imp.find_file returns the .py file before the .pyc file
when I think it should be the other way around.  Solution should be
to swap two lines in Python/importdl.c 


  I've been learning how to use the imp module.  There is a function
called find_module which looks up the initialization information
needed for a given module name.  So I tried it out:

val> cat > norm.py
bar = "Cheers"
val> python
Python 1.5.1 (#21, Nov 23 1998, 15:04:47) [C] on irix6
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import norm
>>> 
val> ls -l norm*
-rw-r--r--    1 dalke    staff         15 Jun 14 20:10 norm.py
-rw-r--r--    1 dalke    staff         98 Jun 14 20:10 norm.pyc
>>> import imp
>>> imp.find_module("norm")
(<open file 'norm.py', mode 'r' at 100a4a90>, 'norm.py', ('.py', 'r', 1))


--- PROBLEM ---

I expected to get something like:
(<open file 'norm.pyc', mode 'r' at 100a4a90>, 'norm.py', ('.py', 'r', 2))

 My expectation was that it would act like a normal import and try
to get the .pyc (or .pyo) file first, before getting the source file.
Hence, a ".pyc" for the filename extension and a 2 (==imp.PY_COMPILED)
instead of a 1 (==imp.PY_SOURCE).

  If I remove the .py file, it does get the .pyc file with the right
description information:

val> python
Python 1.5.1 (#21, Nov 23 1998, 15:04:47) [C] on irix6
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import imp
>>> imp.find_module("norm")
(<open file 'norm.pyc', mode 'rb' at 100a4a90>, 'norm.pyc', ('.pyc',
'rb', 2))

  I believe the problem boils down to the list returned from
imp.get_suffixes.  The standard list is:

[('.so', 'rb', 3), ('module.so', 'rb', 3), ('.py', 'r', 1),
 ('.pyc', 'rb', 2)]

so .py files are found before .pyc files.  (BTW, the .pyc will be
replaced with a .pyo if -O is used.)

I tested this possibility by changing Python/importdl.c to swap the
 .py and the .pyc lines when building _PyImport_Filetab:

>         {".pyc", "rb", PY_COMPILED},    /* added */
>         {".py", "r", PY_SOURCE},
> /*      {".pyc", "rb", PY_COMPILED},*/  /* original */


After recompiling, I get that the .pyc file is indeed found first:

val> ls -l norm*
-rw-r--r--    1 dalke    staff         15 Jun 14 20:10 norm.py
-rw-r--r--    1 dalke    staff         98 Jun 14 20:10 norm.pyc
val> ~/Python-1.5.1/python 
Python 1.5.1 (#32, Jun 14 1999, 20:37:01) [C] on irix6
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import imp
>>> imp.find_module("norm")
(<open file 'norm.pyc', mode 'rb' at 100a8fc8>, 'norm.pyc', ('.pyc',
 'rb', 2))
>>> 

and if the .pyc isn't present it will back down to the .py file.


  Is this really a bug?  Oh, and the most recent version I have,
1.5.2a2, has the same problem, and no one in Deja News has mentioned
it.

						Andrew Dalke
						dalke at bioreason.com



From scott at chronis.pobox.com  Tue Jun  1 12:21:34 1999
From: scott at chronis.pobox.com (scott cotton)
Date: Tue, 1 Jun 1999 12:21:34 -0400
Subject: Python 2.0
In-Reply-To: <7j0urf$2dm$3@cronkite.cc.uga.edu>; from Graham Matthews on Tue, Jun 01, 1999 at 03:39:27PM +0000
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <7j0urf$2dm$3@cronkite.cc.uga.edu>
Message-ID: <19990601122134.A30248@chronis.pobox.com>

kaffe, the gnu java implementation, has some interesting GC
features which might be explored for python 2.0:

0601 12:11 bort:src/kaffe-1.0b4# kaffe --help
Unknown flag: --help
usage: kaffe [-options] class
Options are:
[...]
        -ss <size>              Maximum native stack size
        -mx <size>              Maximum heap size
        -ms <size>              Initial heap size
        -as <size>              Heap increment
        -verbosegc              Print message during garbage collection
        -noclassgc              Disable class garbage collection
[...]
        -noasyncgc *            Do not garbage collect asynchronously
        -oss <size> *           Maximum java stack size

0601 12:11 bort:src/kaffe-1.0b4#

Additionally, it's garbage collection takes place through an
interface which can be implemented any number of ways,
including using reference counting.

import gc; gc.refcounting(1)

or

gc.markandsweep(1)

or

class MyGC: pass

gc.setgc(MyGC)

seems like they could come in handy. 

memory management has no single ideal for any language, it
seems best to let the programmer manipulate it a bit.

in the land of python-2.0 anything is possible ;)

scott



On Tue, Jun 01, 1999 at 03:39:27PM +0000, Graham Matthews wrote:
| Paul Boddie (paulb at infercor.no) wrote:
| : As has been discussed before on this newsgroup/list, such resources get
| : overlooked in many a garbage collection scheme. I'm sure a perusal of various
|                 ----
| : news/list archives would provide a substantial amount of relevant literature.
| 
| That word "many" is your key Paul. Just because many a garbage collection
| scheme works that way doesn't mean all do, or all have to. It's entirely
| an implementation dependent thing.
| 
| You are creating a strawman argument that GC cannot properly free resources
| like files. But some GC do just that so what is your point?
| 
| graham
| 
| 
| -- 
|       This mountain is not visible from any inhabitable place. 
|        It has no name, and since it's discovery in 1856 by a 
|       British survey team, it has mysteriously resisted every
|     attempt to name it. The mountain is known simply and starkly
|      by it's original designation on the surveyor's map ... K2.
| 
| -- 
| http://www.python.org/mailman/listinfo/python-list



From tismer at appliedbiometrics.com  Thu Jun 17 06:03:55 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Thu, 17 Jun 1999 10:03:55 GMT
Subject: ref counting, access from python
References: <7kad31$o7f$1@ns.mtu.ru>
Message-ID: <3768C80B.14B245E6@appliedbiometrics.com>


Oleg Orlov wrote:
> 
> Hello.
> How can i get objects current ref count from my python code?
> Thanks.

>>> import sys
>>> sys.getrefcount(None)
1583
>>> print "ciao - chris"
ciao - chris
>>> 

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From mgushee at havenrock.com  Sun Jun 20 12:48:11 1999
From: mgushee at havenrock.com (Matt Gushee)
Date: 21 Jun 1999 01:48:11 +0900
Subject: newbie tkinter -- radio buttons
References: <376C672C.F817E0C0@pixar.com>
Message-ID: <m3hfo2q0ms.fsf@havenrock.com>

Hi, Cynthia--

Welcome to the great Tkinter adventure!

Cynthia Pettit <kiki at pixar.com> writes:

> Okay, you get your choice of answering any single one of these questions
> to solve my woes! :)
> 
> I'm making a color menu in a drawing program.  It allows you to add
> named colors as you go.
> 
> At first, I was adding colors this way:
> 
>   self.color_menu.add_radiobutton(label=color_name,
>    value = color_name,
>    variable = self.current_color)
> 
> [Note: a couple other issues: value and variable don't seem to work.
> When I choose the button, it does not change the value of
> self.current_color.]

I bet you're trying to do this:

	new_color = self.current_color

I wish the documentation were clearer on this inherently confusing
point. The 'variable' attribute of a Tkinter widget does *not* refer
to a Python variable. It refers to a Python *object* which acts as the 
interface to a Tk variable. So you need to

1) Create an instance of IntVar, StringVar, DoubleVar, or BooleanVar,
e.g.:

	self.current_color = StringVar()

and

2) Use the get() and set() methods to access the value:

	new_color = self.current_color.get()

the menu constructor command is fine as you wrote it.

>   temp = Radiobutton(self.color_menu, text=color_name,
>    value = color_name,
>    variable = self.current_color)
>   temp.pack(anchor=W)
>   temp.select()

> ...

> But now, of course, this conflicts with other calls, like add_command
> and add_separator.  [eg "Add Color"]  The buttons look to be on top of
> eachother.  I think this is due to the .pack() calls not working with
> the add_radiobutton calls.

Well, they're not supposed to. The Menu class isn't designed to
contain other widgets; the radiobuttons, separators, etc. within menus
that are created with add_radiobutton are not class instances, they're
just components of the Menu instance. Hence they don't have their own
methods, and must be manipulated through the methods of the Menu
class.

Unless you want to go to the trouble of manually creating a menu from 
a Frame() class instance -- but unless you need to do something
extremely sophisticated and non-standard, I don't see why you would
need to do that.

> Here's a choice of things that if solved would save me [in order of
> preference]:
> 
> Call .select() on an add_radiobutton [and just how *does* one ever
> access these buttons ever again?!?!]

Haven't worked with Tkinter in a while, but IIRC, calling the set()
method of your radiobutton variable should do what you want.

Have you read the Tk manual page for the Menu class? If not, I'd highly
recommend doing so. It's a shame to have to learn two languages in
order to use one, but the Tkinter documentation has a ways to go
yet. Personally I find that understanding Tk widgets from a Tcl
perspective has saved me a lot of unproductive flailing around w/
Tkinter.

Hope this helps a bit.

Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com




From nickb at earth.ox.ac.uk  Mon Jun  7 07:57:31 1999
From: nickb at earth.ox.ac.uk (Nick Belshaw)
Date: Mon, 07 Jun 1999 12:57:31 +0100
Subject: event handling in classes?
References: <7ivg74$rau$1@nnrp1.deja.com>
Message-ID: <375BB3AA.9CC679A6@earth.ox.ac.uk>

bgue at my-deja.com wrote:

> Hi, I'm stumped on how to accomplish event handling with a Tkinter app
> that's a class. I didn't find anything on a Dejanews search either...can
> anybody tell me what the following example is doing wrong?
>
> from Tkinter import *
>
> class gui(Frame):
>   def __init__(self, parent=None):
>     Frame.__init__(self, parent)
>     self.grid()
>     self.bind("<Key>", self.callback)
>     self.str = ""
>     self.mainloop()
>
>   def callback(self, event):
>     print event.char
>     self.str = self.str + event.char
>
> g = gui()
>
> -----

I don't see any replies so I'll try a little clumsy help.

Your code seems a bit confused so if I modify it slightly....
#---------------------------------------
from Tkinter import *

class gui(Frame):
  def __init__(self, parent=None):
    self.test_widget1 = Canvas(parent,width=50,height=50)
    self.test_widget1.pack()
    self.test_widget1.bind('<1>', self.callback)

    self.test_widget2 = Text(parent,width=10,height=10)
    self.test_widget2.pack()
    self.test_widget2.bind('<Key>', self.callback)

  def callback(self, event):
    print event

Main = Tk()
g = gui(Main)
Main.mainloop()
#----------------------------------------
Same class approach but I have initialised a master-window with Tk() and
passed
it to the instance on creation and finally started the Tk.event-loop
outside the instance with Main.mainloop()

Your main problem is with the widget bindings. You can only bind things to
a widget that the widget can recognise.

So the Canvas will recognise the mouse '<1>' press but if you bind a
<KeyPress> it will not see it.
A Text widget will then see the <Key> event ok of course.

You had a Frame which might take entry and motion bindings (not sure -
maybe none) but a Frame is a holder for other widgets.

Anyway - hope that sets you on the right track.

For more info try

http://www.pythonware.com/library.htm

cheers
Nick/Oxford





From mlh at swl.msd.ray.com  Mon Jun 28 10:03:30 1999
From: mlh at swl.msd.ray.com (Milton Hankins)
Date: Mon, 28 Jun 1999 14:03:30 GMT
Subject: Environment inheritance under Windows NT
References: <Pine.SGI.3.95q.990621173009.142968O-100000@masox202> <7l3hpf$84v$1@nnrp1.deja.com>
Message-ID: <7l7vba$hil$1@nnrp1.deja.com>

In article <7l3hpf$84v$1 at nnrp1.deja.com>,
Bill Wilkinson <yopen at my-deja.com> wrote:

>...
> If you take a look after you change the variable using "set",
> you will notice that there are actually two copies of your variable.
> The new one can be changed on demand in the current window.
>
> You can ease the pain a little by using the setx.exe utility that
comes
> with the resource kit.

Yes, I tried installing the resource kit, but it required Internet
Explorer 4.  So, off I went to find a copy of IE4 to download.  No luck.
So, I tried installing IE5.  I tried for several hours; the install kept
failing.

Besides, doesn't setx change the "global" environment -- the same as in
the System control panel?  That's not really the effect we want.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From news at dorb.com  Tue Jun 15 15:48:40 1999
From: news at dorb.com (Darrell)
Date: Tue, 15 Jun 1999 15:48:40 -0400
Subject: Newbie: Truth values (three-valued logic)
References: <Pine.OSF.4.05.9906152037050.23333-100000@pinax.Mathematik.Uni-Bielefeld.DE>
Message-ID: <29y93.24$U6.625@newsr2.twcny.rr.com>

Not exactly what you want I know. But None is close.

>>> not 0
1
>>> not None
1
>>>
... def myNot(x):
...     if x != None:
...             return not x
...     return x
...
>>>





From yopen at my-deja.com  Sat Jun 26 15:40:09 1999
From: yopen at my-deja.com (Bill Wilkinson)
Date: Sat, 26 Jun 1999 19:40:09 GMT
Subject: file.tell problems (1.5.2/Win32)
References: <7krd1r$2oa$1@holly.csv.warwick.ac.uk>
Message-ID: <7l3aao$65c$1@nnrp1.deja.com>

hmmm.... not too sure what your asking, here is a session on my NT box.

>>> g = open("c:\\temp\\trick.ini","a")
>>> g.tell()
0
>>> g.seek(5)
>>> g.tell()
5


In article <7krd1r$2oa$1 at holly.csv.warwick.ac.uk>,
  esuzm at primrose.csv.warwick.ac.uk (Andrew Clover) wrote:
> I've been meeting some surprises using the seek and tell on file
objects
> recently. I don't know if these are bugs in Python or stdio, in Win32
or
> in general, or are simply broken as designed, but:
>
>  - If a file is opened in 'a'ppend mode, but nothing has yet been
written to
>    it, tell() returns zero.
>
>  - Values returned by tell() - possibly only on a append file, I
don't know -
>    seem to be randomly a bit wrong unless flush() is used before
reading.
>
>  Can anyone confirm this?
>
> --
> This posting was brought to you by And Clover.
> (Sorry.)
>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From MHammond at skippinet.com.au  Wed Jun  2 08:35:41 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Wed, 2 Jun 1999 22:35:41 +1000
Subject: need help checking for stack overflow in Win32
References: <5lk8to3su3.fsf@eric.cnri.reston.va.us> <7j1pcc$256$1@m2.c2.telstra-mm.net.au> <375502C7.39C4FD0E@appliedbiometrics.com>
Message-ID: <7j38bs$ipe$1@m2.c2.telstra-mm.net.au>

Christian Tismer wrote in message
<375502C7.39C4FD0E at appliedbiometrics.com>...

>Nope. The trap is:
>You compiled with VC++ 6.0 which has better register spilling.
>Therefore, recusion costs only 100 bytes, not 104.
>Not calculate 104 * 10000 and you see why 1 MB stack
>lets it crash.

Cool.  Thanks.

But doesnt the fact remain that counting the recursion depth is dangerous?
Im thinking of embedded environments, where a) the stack size of the hosting
.exe can't be known and b) the stack used by the app before Python is even
called can't be determined.  Add to this callbacks in GUI and COM type
applications, and we seem to be applying more band-aids.

The idea of counting space used seems better, but still assumes a known
stack size.

>I submitted a patch program for this kind of
>problems, so I think there is no need to change
>Python 1.5.2 at all.

Huh?  Are you suggesting people continue to build from sources, then run
your script to get an "official" version?  Wouldnt it be better to get Guido
to apply these changes to the build process?

You have provided an excellent interim fix, but IMO the problem remains.

Mark.






From vmas at servicom2000.com  Tue Jun 22 05:15:11 1999
From: vmas at servicom2000.com (vmas)
Date: Tue, 22 Jun 1999 11:15:11 +0200
Subject: file system quota
Message-ID: <Pine.LNX.4.10.9906221110230.308-100000@madre.servicom2000.com>

Hi,
I'm using Python for writing system administration scripts with a Unix
system and I'd like to set quota limits for a given user. Do you know how
to manipulate the file system quotas using Python?
Maybe you know some module for managing file system quotas ...
Thanks.


-- 
=========================================================================
Vicente Mas                       ||
Dpto. Tecnico, Servicom 2000      ||    Do or do not ... There is no try.
e-mail: vmas at servicom2000.com     ||    Yoda (The Empire Strikes Back)
Tel: +34963618776                 ||    
Fax: +34963605508                 ||
=========================================================================




From alan.gauld at gssec.bt.co.uk  Wed Jun  9 12:41:39 1999
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Wed, 09 Jun 1999 17:41:39 +0100
Subject: "Programming Python" still worthwhile?
References: <1283218560-70098357@hypernet.com>
Message-ID: <375E9943.F9FEA7BA@gssec.bt.co.uk>

Gordon McMillan wrote:
> Rick Robino writes:
> > me some advice as to whether or not "Programming Python" is useful,

Yes it is. If you are used to the O'Reilly style and are 
happy with it PP is an excellent book. I'm currently jusat 
over half way thru' and it has helped me a lot.

At work we have Python 1.3 (don't ask!) and at home I use 
1.5.2. The vast majority of things work on both systems.

> > if the current edition is still current, if I should wait for the
> > "Learning Python", or if I should use some other resources.

I looked at LP but couldn't see a good reason to buy it at 
the stage I'm now at - basic Tkinter and CGI programs 
without recourse to the books.

> liked PP very much (and the difficult to find Internet Programming
> with Python, Watters et al, M&T books). 

I had no problem getting the IPwithP book via Amazon :-)

I didn't find it too useful, and still wasn't clear how 
much of the code applied to the standard distribution of 
Python and how much was for the book only. It reminded me of
being at uni' where the lecturers insist on deriving 
everything from first principles before telling you that 
you can buy a book of tables... (Actually, come to think 
of it I'm still not that clear on some of it!) 
[ The socket examples are OK. ]

The general layout of the language reference part was 
confusing to me. Sorry Guido, but I prefer PP....

YMMV of course.

> > I expect to come up to speed fairly quickly and I'm just looking for
> > those fundamental insights into OOP and Python that can build up the
> > kind of inuition and understanding one needs in order to "get
> > comfortable" connecting/creating the dots. 

I'd say thats exactly where the PP book fits. Its not a hold 
hands tutorial, neither is it a function by function reference
(you get that with the online documents!)

There are another 2 books on Python around and at least one 
other on its way. Browse away.

Alan g.

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



From andersdt at notes.cba.ufl.edu  Wed Jun 30 14:06:01 1999
From: andersdt at notes.cba.ufl.edu (Taylor Anderson)
Date: Wed, 30 Jun 1999 14:06:01 -0400
Subject: theads & global namespaces
Message-ID: <377A5C88.32A1FAB6@notes.cba.ufl.edu>

Greetings all,

I am trying to share globals between threads.  Is is possible for me to
share globals between threads that call functions from different
modules?  Thus far I have had little success.  I ask this question
because  I would prefer not to put all my code in one module.

For example, I have 2 .py files: a.py and b.py.  b.py contains the
function ('foo()') to be executed  under the new thread, as well as the
global variable 'x'.  a.py contains the call to
start_new_thread(foo,()).  a.py also tries to use the global variable x.

=========================================
b.py

x = 0

def foo():
   global x
   x  = x + 1

=========================================
a.py

from b.py import *

if __name__ == '__main__':

    start_new_thread(foo, ())
    while 1:
        if x != 0:
            print x

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

So far, my code in a.py does not detect any change in x.  Any
suggestions for a workaround?

Many thanks,

Taylor




From qnmdjb at ffff.com  Sat Jun  5 00:23:58 1999
From: qnmdjb at ffff.com (qnmdjb at ffff.com)
Date: 5 Jun 1999 04:23:58 GMT
Subject: FREE HARDCORE TEEN PICS  6141
Message-ID: <7ja8ou$7j4$4641@nclient5-gui.server.virgin.net>

FOR THE BEST FREE TEEN SEX PICTURES VISIT:

http://freespace.virgin.net/eric.johnston/freepics.htm
gldoqlclctgtcizceuovzqleir




From fredrik at pythonware.com  Thu Jun  3 15:16:03 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 3 Jun 1999 19:16:03 GMT
Subject: pyGTK on Windows
References: <3757D453.A302E757@interar.com.ar>
Message-ID: <037b01beadf5$8b5d1a80$f29b12c2@pythonware.com>

Nahuel <nahuel at interar.com.ar> wrote:
> Hi, im about to do a cross-plataform GUI application in Linux and
> Windoze... i must
> choose the GUI toolkit.. my question is .. can i use pyGTK under
> Windows???
> 
> (I need a gui that can display a table with about 4000 lines without
> crashing (not like Tk :))

you do mean 4,000,000 lines, don't you?

or did you use a seriously broken version of Tk(inter)?

after all, if it's a bug in your program, changing the UI
toolkit won't help...

</F>




From smigowsky at dspace.de  Tue Jun  8 08:01:51 1999
From: smigowsky at dspace.de (Stefan Migowsky)
Date: Tue, 8 Jun 1999 12:01:51 GMT
Subject: COM Question - QueryInterface?
Message-ID: <137053459BA7D211B3930090272A62CA07B4EC@newsintern.dspace.de>

Hi, 

you get the IDispatch interface with this call prtoviding
you have an IUnknown.

>>>import pythoncom
>>>IUnk
   <PyIUnknown at 0x3bf86a4 with obj at 0x837ab0>
>>>IUnk.QueryInterface(pythoncom.IID_IDispatch)    # Query IDispatch
   <PyIDispatch at 0x3bf4d54 with obj at 0x837ab0>
>>>IUnk.QueryInterface(pythoncom.IID_IDispatch).QueryInterface(pythoncom.IID
_IUnknown) # Query IUnkown again
   <PyIUnknown at 0x3bf86a4 with obj at 0x837ab0>

You just have to get hold off at least one interface object.
If you use the wrapper classes generated by makepy you can get the 
interface by using the _oleobj_ attribute e.g:

>>>from win32com.client import Dispatch
>>>drives = Dispatch("AmAccess.Drives")
>>>drives._oleobj_
   <PyIDispatch at XXX with obj at XXX>

Hope this helps

   Stefan



> ----------
> Von: 	Chris Tavares[SMTP:]
> Gesendet: 	Dienstag, 8. Juni 1999 03:17:58
> An: 	python-list at cwi.nl
> Betreff: 	COM Question - QueryInterface?
> 
> I've got a (hopefully easy) question about python's COM support.
> 
> I'm developing a COM object in C++ which implements the COM collection
> idiom - an Item method that returns another object. The code in Python
> I'm using is something like this:
> 
> from win32com.client import Dispatch
> drives = Dispatch("AmAccess.Drives")
> drive = drives.Item("c:")
> dir = drive.Item("some-area")
> 
> On this last line, I get an AttributeError for Item.
> 
> I think I know the reason why - the interface returned from 
> Drives.Item
> isn't a dual, it's an oleautomation compatible vtable interface.
> IDispatch on that object is implemented separately. So, PythonCOM sees
> the VT_UNKNOWN return type and wraps a PYIUnknown object around it. I
> need to be able to do a QueryInterface for IDispatch.
> 
> How do I get Python to do that? There's a PyIID type, but 
> apparently it
> only works for class or progids.  Anyone got any suggestions?
> 
> Thanks for any help,
> 
> -Chris
> 
> 
> 



From matomira at acm.org  Sun Jun 13 21:31:57 1999
From: matomira at acm.org (Fernando D. Mato Mira)
Date: Mon, 14 Jun 1999 03:31:57 +0200
Subject: <language> broken
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com>
Message-ID: <37645B8D.1A267771@acm.org>

Fernando Mato Mira wrote:

> [For example, I can think the Eiffel syntax is `philosophically' broken, but
> as `infix' syntaxes go, it is not. And it's a good language (I programmed

Correction. There's one thing broken. The Subject Oriented syntax, but as Eiffel
was started in '85, I guess that is understandable.

Ada 95 got this one right. Unfortunately, the semantics qualify as `multiply
broken' regarding this issue:

1. They thought about multiple dispatch, and they broke it on purpose.
2. If you specialize on more than one parameter, it should be the same type
specifier
     for all of them, i.e. only applicable to predicates, copying, arithmetic and
things like that,
     not useful most of the times one wants several specializers.
3. Worse of all, instead of just adopting a left-to-right precedence rule, one that
looks very
     ugly was added, increasing the complexity of a document that is already big
enough.

In '95, at VRAI we refocused our application from AI-oriented to (soft) real-time.
I grabbed the 9X specs, and the initial excitement turned to dissapointment (the
prototype was in CLOS). Moving would have not only meant losing a lot in
flexibility, but, worse of all, we would move only from being third-class to
second-class citizens (SGI platform).
And I would have rather dedicate the extra effort to CL. So after a little
successful experience, and given
the simpler requirements, we went with C++.
Needless to say, the project became a death march. You don't go with C++ unless you
can afford paying 5-10 times more for a lot less functionality.

--
Fernando D. Mato Mira
Real-Time SW Eng & Networking
Advanced Systems Engineering Division
CSEM
Jaquet-Droz 1                   email: matomira AT acm DOT org
CH-2007 Neuchatel                 tel:       +41 (32) 720-5157
Switzerland                       FAX:       +41 (32) 720-5720

www.csem.ch      www.vrai.com     ligwww.epfl.ch/matomira.html






From tavares at connix.com  Thu Jun 10 21:43:59 1999
From: tavares at connix.com (Chris Tavares)
Date: Thu, 10 Jun 1999 21:43:59 -0400
Subject: COM Question - QueryInterface?
References: <375C7D56.C9744E39@connix.com> <375fe0b1.1117810@news.freeserve.net> <375EFF29.2C1D0FB7@connix.com> <3760ffc3.2905460@news.freeserve.net>
Message-ID: <376069DF.399FD81D@connix.com>

Toby Dickenson wrote:

> (posted and cced to Chris)
>
> Chris Tavares <tavares at connix.com> wrote:
>
> >Thanks. The reason that the interface in question isn't a dual is because my
> >object design depends quite heavily on having multiple interfaces. If you have
> >multiple duals, you end up with lots of problems with IDispatch, since you can't
> >safely implement more than one version of the same interface on a single object
> >(and lots of people have tried, REALLY HARD).
>
> This is a fallacy about IDispatch that I've heard from many sources, but it
> certainly is not true.
>
> One true statement is that you can only have one interface known as
> IID_IDispatch. This causes a problem for scripting languages that can not
> QueryInterface (or, worse, which QueryInterface behind your back).[
>

I would disagree that this is a fallacy. It DOES cause problems for scripting languages
that won't do QI, but there's another client that can do a QI behind your back - the
COM remoting layer. You've got the possibility of some really wierd bugs that don't
show up in inproc that jump up and bite you when you go oop or remote.

>
> If you need to support this kind of client (VBScript, JScript, what else?
> definitely not Python) then you need to implement IID_IDispatch with the union
> of the methods from all your other interfaces. The page Chris mentions (below)
> has a great summary of some solutions to this.
>
> >So what I'm doing is having multiple oleautomation compatible interfaces, and then
> >using a typelib-driven IDispatch implementation that I got from
> >http://www.sellsbrothers.com/tools/multidisp/index.html. This automagically merges
> >all my separate interfaces into a pseudo-object model. Really neat stuff, and
> >saves a TON of work.
>
> Follow that advice if you need to support those lesser scripting clients, or
> don't if you don't.
>
> In any case, there is still no reason not to make all your other interfaces dual
> too. This should be an easy step. I am assuming you already have some IDL for
> these interfaces? Change the oleautomation line to dual, and hook up the same
> typelib-driven IDispatch implementation you are already using for IID_Dispatch.
> If you are using ATL for this implementation then it all comes for free. The
> advantages? You get easier access from Python, and in VB you get compile time
> type safety.
>

I understand the argument you're making. To be honest, I don't expect anyone other than
me to call these objects from Python - heck, my coworkers are still doing database apps
in C++!

I don't want to make the interfaces dual. Just chalk it up to stubborness, I guess. <G>
And in VB I've already got type safety.

>
> In your specific example, the type library would say your collection object's
> Item method returns an ITavaresDrive (or whatever) and python would use an
> appropriate makepy-generated wrapper.
>
> >Having tried that, I find that I now have access to IDispatch::Invoke. Well, what
> >I really wanted was win32com.client.dynamic.Dispatch. How do I get to the wrapped
> >IDispatch rather than the raw one?
>
> drive = drives.Item("c:")._oleobj_.QueryInterface(pythoncom.IID_IDispatch)
> win32com.client.Dispatch(drive)
>
> (from memory)
>

Thanks, I think this will do the trick!

>
> I hope this helps,
>
> Toby Dickenson

Sure does!

-Chris





From tim_one at email.msn.com  Sat Jun  5 15:59:58 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 5 Jun 1999 15:59:58 -0400
Subject: Proper Radix Formatting of Longs?
In-Reply-To: <7j4vp6$je8$1@bashir.ici.net>
Message-ID: <000201beaf8d$fde14100$8c9e2299@tim>

[Tom Vrankar]
> I use Python a lot for computer hardware testing and communications. It
> generally works great.
>
> However,
> [a number of irritations boiling down to that Python has no unsigned
>  integral type & checks for overflow on signed integer arithmetic,
>  so unsigned 32-bit int arithmetic is clumsy on a 32-bit platform]
> ...
> Is there a standard pythonic idiom for using 32-bit integers as
> if they were unsigned, so that 0x8000_0000 +0x8000_0000 =0, just
> like in C or my hardware ALU?

Curiously, I first started using Python when working for a HW vendor trying
to use 32-bit workstations to design a 64-bit CPU with a 64x64->128
multiplier.  All the same irritations applied there -- but since *no* form
of 32-bit arithmetic was adequate (in Python or C), it never occurred to us
that faking it with Python longs was a burden <wink>.

Which is the easiest all-Python answer there is:  Python doesn't have an
unsigned integral type, period.  So you convert everything to long on input,
live happily for the duration, and cut it back to whatever you need on
output.

If that's not fast enough (it was for me -- in fact, even wrapped stuff in
classes to hide the I/O conversions), you'll have to write a C extension
module.  Note that Python doesn't gripe about 32-bit bit-fiddling ops (~ & |
^ << >>), so maybe you can get away with writing just a handful of C
functions.

> Is there a standard pythonic way to format longs into
> arbitrary radices (sort of like an inverse string.atol)?

str(), hex() and oct() are "it".  Feed hex a long, and it will always have
"0x" at the front and "L" at the end, so hex(along)[2:-1] always works to
strip the decorations.

> Is there a secret handshake required to at least get string.atoi()
> to happily convert "0x8000_0000"  symmetrically with "%x" %0x8000_0000?

No.  Python allows 0x notation with the MSB set in numeric program literals
for the convenience of programmers building bit masks; atoi is stricter
because it's used to crack user (as opposed to programmer) input.

Note that there's a real distinction here:  as a *bitmask*, 0xffffffff
yields an int with exactly the least-significant 32 bits set, regardless of
whether you're running an a platform where Python's ints are 32 bits or 64
bits.  But as an *integer* (which is atoi's view of the world), 0xffffffff
would be -1 on a 32-bit platform but 4294967295 on a 64-bit one.

So the dual view of bounded signed ints as either bitmasks or integers isn't
symmetric "even in theory", and Python picked a compromise.  It's not
convenient for everyone, but so it goes.  The dual view of *un*bounded
signed ints doesn't suffer this asymmetry, and is independent of platform.

> Yes, I know I could probably write some elaborate python
> functions to manually convert back and forth,

If you can afford Python speed, "elaborate" is inaccurate -- these kinds of
functions are generally trivial to knock off; e.g.,

>>> def last32(i, _mask32=0xffffffffL):
	try:
		return int(i)
	except OverflowError:
		tail = i & _mask32 # now positive long
		return (int(tail >> 1) << 1) | int(tail & 1)

>>> def uns32_add(i, j):
	try:
		return i + j
	except OverflowError:
		return last32(long(i) + j)

>>> uns32_add(0x80000000, 0x80000000)
0
>>>

> but when handling megabytes, the performance would be unacceptable.

That's what you get for working in a field 99.9999% of the world doesn't
care about <wink>.  Seriously, Python isn't going to grow an unsigned 32-bit
type overnight, so if it really is too slow in Python you're going to have
to avail yourself of the extension facilities.  You can write a complete
uns32 type as an extension type (a doable but non-trivial project), or just
the handful of functions you need most.

general-purpose-vs-all-purpose-ly y'rs  - tim





From da at ski.org  Tue Jun 29 00:35:41 1999
From: da at ski.org (David Ascher)
Date: Mon, 28 Jun 1999 21:35:41 -0700 (Pacific Daylight Time)
Subject: Python GUIs: Summary and Conclusion
In-Reply-To: <377845DD.23AD193A@compaq.com>
Message-ID: <Pine.WNT.4.05.9906282126560.188-100000@david.ski.org>

On Tue, 29 Jun 1999, Greg Ewing wrote:

> Wasn't there a Tcl-less Tk interface once before, that wasn't kept up
> to date? Called something beginning with R that I can't remember.

Rivet.

> Maybe you could use that as a starting point if the Tk C interface
> hasn't changed too radically.

I wouldn't -- I don't think the rivet webpage is around anymore!  My
Trinket.py is still around, but certainly not up to date.

I thought that Max Skaller had done some work more recently along the
same lines, but I can't find a URL for it.

--david





From jepler at inetnebr.com  Wed Jun  2 10:36:03 1999
From: jepler at inetnebr.com (Jeff Epler)
Date: Wed, 02 Jun 1999 14:36:03 GMT
Subject: A problem with re.split()
References: <87pv3en24q.fsf@pc-hrvoje.srce.hr>
Message-ID: <slrn7lage9.3de.jepler@craie.inetnebr.com>

import re
def resplit(pat, str):
	"""Emulate perl braindamage at the cost of an extra function
	call"""
	if not pat: return map(None, str)
	else: return re.split(pat, str)

Jeff
-- 
\/ http://www.redhat.com/                        Jeff Epler jepler at inetnebr.com
Hey, what do you expect from a culture that *drives* on *parkways* and
*parks* on *driveways*?
		-- Gallagher



From tim_one at email.msn.com  Tue Jun 29 03:15:19 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 29 Jun 1999 03:15:19 -0400
Subject: total idiot question: +=, .=, etc...
In-Reply-To: <377847EE.D5B97A23@compaq.com>
Message-ID: <000501bec1ff$2517f1a0$719e2299@tim>

>>     class A(B):
>>         def __init__(self, x, y, z):
>>             B.__init__(self, x, y, z)
>>
>> can be grating too

[Greg Ewing]
> I could live with having to explicitly name the
> superclass, provided something checked that the
> class I named was actually a direct base class
> of the one where the method is defined. That
> way things would be less likely to break
> mysteriously when I rearrange the class
> hierarchy.
>
> I'm still thinking about how to implement
> this...

Hard!  Class attrs (like __init__) don't point back to the class they belong
to.  In the absence of sharing, though, the class chain can be searched for
the class that owns a particular attr value.  I'll attach something that
does that; it's ugly; I'll never use it <wink>; could speed it a lot by
caching the search result.

seems-a-lot-easier-to-do-it-by-eye-ly y'rs  - tim

class BaseInit:
    """A mixin class for "safe" superclass __init__ calls.

    class X(Y, Z, ..., BaseInit):
        def __init__(self, whatever):
            self.baseinit(K, whatever)

    self.baseinit(K, whatever) invokes K.__init__(self, whatever), but
    raises an error unless K is a direct base class of X (the class
    in which the __init__ calling baseinit is defined).
    """

    def baseinit(self, base, *args, **kwargs):
        import sys
        if not isinstance(self, base):
            raise TypeError(`self` + " isn't even an instance of " +
                            `base`)
        # init <- code object of __init__ we were called from
        try:
            raise 'bogus'
        except 'bogus':
            tb = sys.exc_info()[2]
            init = tb.tb_frame.f_back.f_code  # back to the __init__
            if init.co_name != "__init__":
                raise ValueError("baseinit must be called from __init__")
        # search the class hierarchy for that __init__, setting
        # c to the class it belongs to
        candidates = [self.__class__]
        while candidates:
            c = candidates.pop(0)
            if c.__dict__.has_key("__init__"):
                cinit = c.__init__.im_func.func_code
                if init is cinit:
                    break
            candidates = list(c.__bases__) + candidates
        else:
            raise SystemError("didn't find __init__ in class or bases!")
        if base not in c.__bases__:
            raise TypeError("attempt to init superclass " + `base` +
                            " from " + `c` + " but the "
                            "superclass isn't an immediate base class")
        apply(base.__init__, (self,) + args, kwargs)

class A:
    def __init__(self):
        print "in A's init"

class B(A, BaseInit):
    def __init__(self):
        print "in B's init"
        self.baseinit(A)

class C(B, BaseInit):
    def __init__(self):
        print "in C's init"
        self.baseinit(B)
        print "next one should be bogus"
        self.baseinit(A)

b = B()
c = C()





From larsga at ifi.uio.no  Wed Jun 16 10:59:53 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 16 Jun 1999 16:59:53 +0200
Subject: HTML bettering?
References: <1318D78C9072D11195C9006094EA98A72C214D@ocesrv>
Message-ID: <wk7lp48a3a.fsf@ifi.uio.no>

* Matej Cepl
|
| How to make as simple as tool which would make some simple operation
| over HTML (e.g., headings numbering)? If I would use htmllib.py, I
| would have to build HTMLformatter or I am wrong and it is easy?

You could very well use htmllib and just make an htmllib application
that writes the HTML back out again, and modify it slightly to do
heading numbering. I don't think HTMLFormatter would do you any good
for this sort of thing.

However, HTML isn't a very good SGML application for this sort of
thing, since the following is entirely legal:

<!DOCTYPE ...>
<HTML>
  ...

<BODY>
<H5>On the topic of nesting</H5>

<H2>Why it's good</H2>

<H4>And why HTML doesn't have it</H4>

...
</HTML>


ISO HTML provides a better starting point for this sort of thing, as
do most other SGML applications intended for articles.

Personally, I use a private SGML DTD for my articles and then use
DSSSL stylesheets to produce the HTML, complete with a table of
contents. Adding heading numbering should take all of 5 minutes. :)

And today I've put together the infrastructure to produce PDF from the
same SGML source as well.

Here's an example:

<URL: http://www.stud.ifi.uio.no/~larsga/download/artikler/perl.html>

--Lars M.



From tismer at appliedbiometrics.com  Wed Jun 16 10:09:27 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Wed, 16 Jun 1999 14:09:27 GMT
Subject: Python Webserver
References: <3767A69D.50DC7FBF@serop.abb.se>
Message-ID: <3767B017.BA69E220@appliedbiometrics.com>


"Thomas S. Strinnhed" wrote:
> 
> Hi
> 
> I don't know if I've dreamt, but I think I saw something about
> a webserver written in Python. (It's not the embedded-in-Apache-thing
> I'm thinking of)
> Now I just can't seem to find it anywhere, so does someone know anything
> about this. Hipshots are welcome. I thought it'd be something fun to
> play with.

go to http://www.nightmare.com/
and read about Medusa.
That's it.
-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From joe at strout.net  Wed Jun 16 16:58:44 1999
From: joe at strout.net (Joe Strout)
Date: Wed, 16 Jun 1999 13:58:44 -0700
Subject: BiBTeX module?
Message-ID: <joe-160619991358444033@chinacat.salk.edu>

Is there a BiBTeX module for Python?  I did a search at python.org, but
apart from a Perl module (ick!), I didn't find anything.

Thanks,
-- Joe

-- 
,------------------------------------------------------------------.
|    Joseph J. Strout           Biocomputing -- The Salk Institute |
|    joe at strout.net             http://www.strout.net              |
`------------------------------------------------------------------'
Check out the Mac Web Directory!    http://www.strout.net/macweb.cgi



From bernhard at alpha1.csd.uwm.edu  Thu Jun  3 14:17:08 1999
From: bernhard at alpha1.csd.uwm.edu (Bernhard Reiter)
Date: 3 Jun 1999 18:17:08 GMT
Subject: Why won't this work???
References: <7j0rkr$1eeu$1@rtpnews.raleigh.ibm.com> <7j1qc5$3rj$1@m2.c2.telstra-mm.net.au> <7j3e5q$nrq$1@rtpnews.raleigh.ibm.com> <slrn7lbh5v.j9o.bernhard@alpha1.csd.uwm.edu> <7j5pp9$68c$1@rtpnews.raleigh.ibm.com>
Message-ID: <slrn7ldhl4.6nu.bernhard@alpha1.csd.uwm.edu>

On Thu, 3 Jun 1999 07:44:21 -0400, Jr. King <n at n.com> wrote:
> Most of
> the time people look for email answer they get the response Post here, Read
> Here, I thought that was a rule.
It depends.  Generally it is true.
Of course there are several occasions, when it is better to 
mail the poster and not clutter of the newsgroup with messages
which are only interesting to the author.

Right now I don't feel like looking it up, but AFAIR not giving a
email address is broadly considered unappropriate in newsgroups.

> I can say he was wrong all I want, and you can say he wasn't.  
> You will not be changing my opinion.
I just wanted to give you some insights in views of other persons,
if you decide not to use it, that is entirely up to you.
This is not about opinion.

	Bernhard



From news at dorb.com  Thu Jun 24 20:41:24 1999
From: news at dorb.com (Darrell)
Date: Thu, 24 Jun 1999 20:41:24 -0400
Subject: How do can you make a variable static?
References: <3772AE33.AA764DFA@alliedsignal.com>
Message-ID: <ghAc3.2466$U6.9897@newsr2.twcny.rr.com>

> from foo import foo
>
> SymHanDict={0:0}
>
> foo()                            #each call to foo() should add items to
> the dictionary
> foo()
> foo()
>
> *** the following contained in module foo.py ************************
> def foo():
>     global SymHanDict

Global means look for it in the foo.py module

In this case SymHanDict is in another file and your best bet is to pass it
to foo(). If you just don't want to do that try something like this in the
module where you plan to call foo().

import foo

SymHanDict={0:0}

def myFoo():
    foo.foo(SymHanDict)


Passing a parm is much cleaner than globals anyway.
Might I suggest using a class:

class Foo:
    def __init__(self, dict):
            self._myDict=dict

    def __call__(self):
            """Do foo stuff here"""
            pass

myFoo = Foo(SymHanDict)

    # Call the foo thing here.
myFoo()


--Darrell







From news at dorb.com  Thu Jun 17 13:00:35 1999
From: news at dorb.com (Darrell)
Date: Thu, 17 Jun 1999 13:00:35 -0400
Subject: Starting a project: any hints?
References: <19990617154643.A1680@optiplex.palga.uucp> <7kb384$lgs$1@nnrp1.deja.com>
Message-ID: <pT9a3.327$U6.2451@newsr2.twcny.rr.com>

> - Make your program works from the very first day , it doesn't matter
> if it barely works.

I'd like to emphasize this point. Don't write it all then try it.
Its easier to evolve from a functional kernel than to materialize all at
once.
--
--Darrell





From proteus at cloud9.net  Wed Jun  2 08:37:48 1999
From: proteus at cloud9.net (Michael Muller)
Date: Wed, 02 Jun 1999 08:37:48 -0400
Subject: Windows Stack Size
References: <3752C225.65C98B52@cloud9.net> <3752F154.6C444928@ricochet.net> <375413D8.C3B02D2@cloud9.net> <7j1q41$3fu$1@m2.c2.telstra-mm.net.au>
Message-ID: <3755259C.3513D254@cloud9.net>

Mark Hammond wrote:
[snip]
> statements in __getattr__ would make this immediately obvious.  Also, the
> death should take a few seconds to occur - try hitting Ctrl+C in this
> period - you may get a traceback telling you exactly where the problem is.
> 

As I recall, the trap was immediate.  I'll try the "print"s under linux,
though, just to see if there's anything funky going on.

> >I'd debug it myself except that I'm happily Windows-free (only creates
> >problems when I have to write code for Windows users).  If there is no
> >obvious answer, I have a work-around.
> 
> And what would the work-around be?
> 
> Mark.

Using the raw Tkinter 'Text' control instead of my ScrollingText class. 
It
doesn't seem to have this problem.  Unfortunately, it doesn't have
scrollbars
either.

=============================================================================
michaelMuller = proteus at cloud9.net | http://www.cloud9.net/~proteus
-----------------------------------------------------------------------------
We are explorers in the further reaches of experience: demons to some,
angels
to others.  - "Pinhead" from "Hellraiser"
=============================================================================



From MHammond at skippinet.com.au  Tue Jun  1 19:13:30 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Wed, 2 Jun 1999 09:13:30 +1000
Subject: need help checking for stack overflow in Win32
References: <5lk8to3su3.fsf@eric.cnri.reston.va.us>
Message-ID: <7j1pcc$256$1@m2.c2.telstra-mm.net.au>

Actually, I was about to post this yesterday, and to prove my point I wrote:

def die():
    die()

Expecting to see the crash.  I was then going to go into a rant about how
win32api.Apply would actually let you catch this (and any other Win32)
exception as a temporary solution.

But it worked correctly - ie, Python raised a RuntimeError: Max Recursion.

So I guessed that 1.5.2 had fixed this, presumably by reducing the recursion
count for Win32.  Apparently not!

Thomas' reply is correct, and indeed a number of years ago I did play with
this.  I even reported my results, but no one seemed interested, so I
dropped it (waiting for someone else to complain :-)

It still seems to me that this is not the ideal good solution.  In my simple
example above, we did not die.  But make anything less contrived, and we
have C frames with larger allocations on the stack, and we will die.
Whatever number we pick, we would need to make it quite a bit lower than the
"normal" case, to handle the fact that some C functions may allocate from
bytes to kilobytes on the stack.

A better mechanism would be to use Win32 exception handlers.  Then you dont
need artificial recursion checks at all - leave it to the OS.  However, it
isnt clear to me where this exception handler should be placed, nor indeed
if setting up a Win32 exception handler inside the VM would have a
significant performance hit.  A plus would be we could also catch Access
Violations (Im not sure what we would do with them - but a traceback would
be nicer than Dr Watson when dealing with buggy extensions!)

Thomas' solution is simple, and requires either a simple source file change,
or using the MSVC GUI to adjust the stack size.  If you would like to
experiment with a Win32 exception handler then I would be happy to see where
that leads us...

Mark.

Guido van Rossum wrote in message <5lk8to3su3.fsf at eric.cnri.reston.va.us>...
>I could use some help from someone with a little time, some knowledge
>of Python internals, and some Win32 C expertise.  On Win32 platforms,
>infinite recursion causes an Application Error (a fatal windows
>error).  There is some machinery in Python to check for this
>condition, but apparently it doesn't work on Win32.  Please help!
>
>--Guido van Rossum (home page: http://www.python.org/~guido/)





From gerrit.holl at pobox.com  Sun Jun 27 05:24:33 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sun, 27 Jun 1999 11:24:33 +0200
Subject: RuntimeError?
Message-ID: <19990627112432.A29097@optiplex.palga.uucp>

Hallo,

can someone please give me an example when a RuntimeError is raised?

I don't understand: if none of the others are raised.

groeten,
Gerrit.

-- 
The Dutch Linuxgames homepage:	http://linuxgames.nl.linux.org
Personal homepage:		http://www.nl.linux.org/~gerrit/

Discoverb is a python program (in several languages) which tests the words you
learned by asking it. Homepage: http://www.nl.linux.org/~gerrit/discoverb/



From mark.butterworth at cis.co.uk  Wed Jun 23 04:25:52 1999
From: mark.butterworth at cis.co.uk (Mark Butterworth - U-Net)
Date: Wed, 23 Jun 1999 09:25:52 +0100
Subject: pkzip and using popen on Windows 98 ???
Message-ID: <2H1c3.180$T3.411@newsr2.u-net.net>

Im having trouble using os.system or popen (both win32pipe & os) under
windows 98.  It will not return the exit value of the command Im running
(which is infozips - zip).  Also when attempting to use win32pipe.popen3 -
my pipe read's hang after one or two reads.

Also does anyone know of any python modules which can read / write pkzip (or
infozip) format files as I don't like calling external commands.

Any help would be appreciated?






From matomira at iname.com  Sat Jun 12 13:07:36 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Sat, 12 Jun 1999 19:07:36 +0200
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com>
Message-ID: <376293D8.163C6C15@iname.com>


Fernando Mato Mira wrote:

> [For example, I can think the Eiffel syntax is `philosophically' broken, but
> as `infix' syntaxes go, it is not. And it's a good language (I programmed

Actually, it's a _great_ language to learn OO (although of course in the restricted
single-dispatch way most people know about. I like to call that "Subject Oriented").

Occan taught me about concurrency.
Eiffel taught me OO.
CLOS taught me what OO really means.

--

"Object Oriented is to not think only about your `subject' and forget about your
`arguments' - the syntactic objects of your sentence".

Fernando D. Mato Mira
Real-Time SW Eng & Networking
Advanced Systems Engineering Division
CSEM
Jaquet-Droz 1                   email: matomira AT acm DOT org
CH-2007 Neuchatel                 tel:       +41 (32) 720-5157
Switzerland                       FAX:       +41 (32) 720-5720

www.csem.ch      www.vrai.com     ligwww.epfl.ch/matomira.html





From kas at maps.magnetic-ink.dk  Wed Jun 30 05:39:04 1999
From: kas at maps.magnetic-ink.dk (Klaus Alexander Seistrup)
Date: Wed, 30 Jun 1999 11:39:04 +0200
Subject: I am v. stupid, please help
References: <3779d5f5@discussions>
Message-ID: <3779E5B8.3CDB36CB@maps.magnetic-ink.dk>

Newbie wrote:

> name = input('What is your name : ')
> 
> This works for integers but not for strings,

Use raw_input() instead.


  //Klaus

-- 
???[ Magnetic Ink ]??????????????????????????????????????????????????????????
???[ http://www.magnetic-ink.dk/ ][ maps == mail abuse prevention system ]???



From arcege at shore.net  Thu Jun  3 14:13:50 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Thu, 03 Jun 1999 18:13:50 GMT
Subject: Use of StringIO vs cStringIO in standard modules
References: <8790a1620b.fsf@pc-hrvoje.srce.hr> <199906031425.KAA06935@eric.cnri.reston.va.us> <87aeuh5kqc.fsf@pc-hrvoje.srce.hr>
Message-ID: <yBz53.1777$nn.529906@news.shore.net>

Hrvoje Niksic <hniksic at srce.hr> wrote:
: Guido van Rossum <guido at CNRI.Reston.VA.US> writes:

:> Hrvoje Niksic <hniksic at srce.hr>:
:> 
:> > I noticed that many standard modules use StringIO and not
:> > cStringIO, although they don't need subclassing.  Is this
:> > intentional?
:> > 
:> > For example, base64.py uses StringIO to implement encodestring()
:> > and decodestring().  Since both functions write to output line by
:> > line, I imagine the performance hit of StringIO vs cStringIO might
:> > be non-negligible.
:> 
:> Have you noticed any speed difference?

: Yes, quite a bit.  Trivially replacing StringIO with cStringIO in
: base64.py makes encoding 2.3 times and decoding 3.6 times faster.
: That's on my system (Ultra 2 under Solaris 2.6), measured repeatedly
: with time.clock() and an approx. 1M sample string.  I can post the
: script if there is interest.

: Maybe the correct solution for base64.py would be to use something
: like this at top-level:

: try:
:     from cStringIO import StringIO
: except:
:     from StringIO import StringIO

:> cPickle, because calling it from C is much faster than calling
:> StringIO from C; however I believe that for calls from Python,
:> StringIO isn't that much slower.

: I've looked at the code, and to me it seems that the slowness comes
: from creating new strings on each write, where cStringIO just resizes
: its internal buffer and creates the string only at the end.

:> > Furthermore, is there a particular reason for maintaining two
:> > parallel StringIO implementations?  If subclassing is the reason,
:> > I assume it would be trivial to rewrite StringIO to encapsulate
:> > cStringIO the same way that UserDict encapsulates dictionary
:> > objects.
:> 
:> That's not the reason; it's got more to do with not requiring a C
:> extension where plain Python code will do.  Also to have a reference
:> implementation.

: But then you have to maintain both, *and* you get much slower code.
: Is it worth it?

There are some interface differences between calls to cStringIO.StringIO
and StringIO.StringIO which I think makes StringIO more "usable" in some
cases.  Specifically, if you create an instances of cStringIO.StringIO
with initializeing data, then the instance becomes unwritable:

  >>> import cStringIO, StringIO
  >>> f = StringIO.StringIO("Hi there")
  >>> f.seek(0, 2)
  >>> f.write("\n")
  >>> f.getvalue()
  'Hi there\012'
  >>> g = cStringIO.StringIO("Hi there")
  >>> g.seek(0, 2)
  >>> g.write("\n")
  Traceback (innermost last):
    File "<stdin>", line 1, in ?
  AttributeError: write
  >>>

This could surely break some of the existing code.  But I will mention
that:
  f = StringIO.StringIO(str)
is equivalent to:
  f = cStringIO.StringIO()
  f.write(str)
  f.seek(0)

I would want to see cStringIO mimic StringIO before the two are merged.

And yes, Guido, I have noticed some differences at times.

  -Arcege




From tom at spirit.gcrc.upenn.edu  Mon Jun  7 13:01:38 1999
From: tom at spirit.gcrc.upenn.edu (Tom Fenn)
Date: Mon, 07 Jun 1999 13:01:38 -0400
Subject: Giving talk on Python. Need references.
Message-ID: <375BFAF2.C543178B@spirit.gcrc.upenn.edu>

Hello all,

I will be giving a talk on using Python at the ASMS Conference on Mass
Spectrometry next week.  I'm discussing a "Cross Platform solution to
data analysis applications" with an emphasis on chromatography and mass
spectrometry.  I'd like some references to other scientific/numerical
applications that I can reference.  If anyone has a relevent document
they'd like cited, please
send me the reference. (And a copy I can look at if it's not readily
available.) Thanks.

Tom Fenn



From harv at hub.org  Tue Jun  1 23:38:22 1999
From: harv at hub.org (Michael Harvey)
Date: 2 Jun 1999 03:38:22 GMT
Subject: ax25 module?
Message-ID: <7j28ve$2k8u$1@hub.org>

Hi, I know that this question is probably answered in documentation somewhere
that I can't find, but I'm going to ask anyway.

Are there any ax25 modules for Python?  There might not be, but I just started
using it yesterday, and am wondering if there are.

Thanks in advance

Mike Harvey
harv at hub.no_spam.org




From guido at CNRI.Reston.VA.US  Thu Jun  3 10:25:03 1999
From: guido at CNRI.Reston.VA.US (Guido van Rossum)
Date: Thu, 03 Jun 1999 10:25:03 -0400
Subject: Use of StringIO vs cStringIO in standard modules
In-Reply-To: Your message of "03 Jun 1999 11:56:36 +0200."
             <8790a1620b.fsf@pc-hrvoje.srce.hr> 
References: <8790a1620b.fsf@pc-hrvoje.srce.hr> 
Message-ID: <199906031425.KAA06935@eric.cnri.reston.va.us>

Hrvoje Niksic <hniksic at srce.hr>:

> I noticed that many standard modules use StringIO and not cStringIO,
> although they don't need subclassing.  Is this intentional?
> 
> For example, base64.py uses StringIO to implement encodestring() and
> decodestring().  Since both functions write to output line by line, I
> imagine the performance hit of StringIO vs cStringIO might be
> non-negligible.

Have you noticed any speed difference?  cStringIO is important for
cPickle, because calling it from C is much faster than calling
StringIO from C; however I believe that for calls from Python,
StringIO isn't that much slower.  (This as opposed to cPickle
vs. Pickle, where the performance difference is 2-3 orders of
magnitude.)

> Furthermore, is there a particular reason for maintaining two parallel
> StringIO implementations?  If subclassing is the reason, I assume it
> would be trivial to rewrite StringIO to encapsulate cStringIO the same 
> way that UserDict encapsulates dictionary objects.

That's not the reason; it's got more to do with not requiring a C
extension where plain Python code will do.  Also to have a reference
implementation.  (Ditto for pickle.)

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



From Reimer.Behrends at p98.f112.n480.z2.fidonet.org  Wed Jun 30 09:14:09 1999
From: Reimer.Behrends at p98.f112.n480.z2.fidonet.org (Reimer Behrends)
Date: Wed, 30 Jun 1999 14:14:09 +0100
Subject: total idiot question: +=, .=, etc...
Message-ID: <000015b4@bossar.com.pl>

From: behrends at cse.msu.edu (Reimer Behrends)

Bernhard Herzog (herzog at online.de) wrote:
[...]
> I think a better solution that better fits the python syntax is to
> simply omit the self but not the dot:
> 
> 	.last = .lines[-1]
>  	.lines = .lines[:-1]

Such a syntax was proposed a few years ago for private variables,
IIRC, but was discarded because it had various problems.

And personally, I'd rather live with the inconvenience of the explicit
self rather than having Python's design obfuscated by some strange
syntax; syntax highlighting can go a long way towards making the current
state of affairs more manageable.

			Reimer Behrends



From bbaetz at ug.cs.usyd.edu.au  Tue Jun  1 19:38:02 1999
From: bbaetz at ug.cs.usyd.edu.au (Bradley Baetz)
Date: Wed, 02 Jun 1999 09:38:02 +1000
Subject: Reproducable Python core dump
Message-ID: <37546EDA.20825D43@ug.cs.usyd.edu.au>

Spam detection software, running on the system "albatross.python.org", has
identified this incoming email as possible spam.  The original message
has been attached to this so you can view it (if it isn't spam) or label
similar future email.  If you have any questions, see
the administrator of that system for details.

Content preview:  Hi, My CS project for this semester involves modifying Grail
   to support navigation through a keypad, similarly to Lynx. Whilst testing
   my handing of nested tables, I came across this bug: The following will cause
   a reproducable core dump in Python: [...] 

Content analysis details:   (5.9 points, 5.0 required)

 pts rule name              description
---- ---------------------- --------------------------------------------------
 2.0 FH_DATE_IS_19XX        The date is not 19xx.
-0.0 NO_RELAYS              Informational: message was not relayed via SMTP
 0.0 NORMAL_HTTP_TO_IP      URI: Uses a dotted-decimal IP address in URL
 1.3 URI_HEX                URI: URI hostname has long hexadecimal sequence
 2.2 HTML_IMAGE_ONLY_12     BODY: HTML: images with 800-1200 bytes of words
 3.2 HTML_OBFUSCATE_10_20   BODY: Message is 10% to 20% HTML obfuscation
 0.0 HTML_MESSAGE           BODY: HTML included in message
 1.1 HTML_SHORT_LINK_IMG_1  HTML is very short with a linked image
-0.0 NO_RECEIVED            Informational: message has no Received headers
-3.9 AWL                    AWL: From: address is in the auto white-list

The original message was not completely plain text, and may be unsafe to
open with some email clients; in particular, it may contain a virus,
or confirm that your address can receive spam.  If you wish to view
it, it may be safer to save it to a file and open it with an editor.

-------------- next part --------------
An embedded message was scrubbed...
From: Bradley Baetz <bbaetz at ug.cs.usyd.edu.au>
Subject: Reproducable Python core dump
Date: Wed, 02 Jun 1999 09:38:02 +1000
Size: 3497
URL: <http://mail.python.org/pipermail/python-list/attachments/19990602/c0af74cc/attachment.mht>

From a.mueller at icrf.icnet.uk  Mon Jun  7 11:33:37 1999
From: a.mueller at icrf.icnet.uk (Arne Mueller)
Date: Mon, 07 Jun 1999 16:33:37 +0100
Subject: fork()
Message-ID: <375BE64A.EB997F24@icrf.icnet.uk>

Hi All,

I've switched from 'threading' to 'fork' (remember the discussion some
days ago) since that gives me real parallel processing. But (off course)
there are other problems when forking a process ;-( :

1. The parent process reads in a discitonary with about 250000 keys,
this is a sort of static database which never changes.

2. A bidirectional pipe is created for each child process

3. The parent forks

4. The children claculate some for which they've to access the large
dictionary.

5. All the children are able to send pickled objects to the parent (via
the pipe)

All that woks fine now (after I spend a long time fiddling with the pipe
and select stuff to synchronize I/O). However the forking is a waste of
time and memory because all the children get their own large dictionary!
Is there a way to use shared memory in python, I mean how can different
processes access (for read only) one object without passing the keys of
the dictionary object through a pipe (the children need _quick_ access
to that dictionary)? 

	Thanks alot (again) for your suggestions,

	Arne

-- 
Arne Mueller
Biomolecular Modelling Laboratory
Imperial Cancer Research Fund
44 Lincoln's Inn Fields
London WC2A 3PX, U.K.
phone : +44-(0)171 2693405      | Fax : +44-(0)171 269 3258
email : a.mueller at icrf.icnet.uk | http://www.icnet.uk/bmm/



From suzuki611 at okisoft.co.jp  Mon Jun  7 20:18:13 1999
From: suzuki611 at okisoft.co.jp (Hisao Suzuki)
Date: 08 Jun 1999 09:18:13 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7iqhkb$d7s@wiscnews.wiscnet.net> <7iu7ma$d33$1@cronkite.cc.uga.edu>
	<02hfonzsx2.fsf@rtp.ericsson.se> <02909zzou0.fsf@rtp.ericsson.se>
	<87zp2f4qrz.fsf@pc-hrvoje.srce.hr>
	<wkd7z858c3.fsf@turangalila.harmonixmusic.com>
	<14172.6652.550882.854800@bitdiddle.cnri.reston.va.us>
Message-ID: <u3u2sjtuii.fsf@ares.sys1.ngo.okisoft.co.jp>

In article <14172.6652.550882.854800 at bitdiddle.cnri.reston.va.us>,
Jeremy Hylton <jeremy at cnri.reston.va.us> wrote:
| >>> (let ((sum 0))
|        (let ((add (lambda (x) (set! sum (+ sum x))))
|              (sub (lambda (x) (set! sum (- sum x)))))
|                 (add 2)
|                 (add 3)
|                 (sub 1)
|              sum))
| 4

A bound method is equivalent to a closure (or rather a function
with its own environment) in a certain sense.  The `self'
parameter serves as the environment.

The traditional (or good old :-) translation of the above code
would be like this:

def alsoWorks():
    class _alsoWorks_t:
	def add(self, x): self.sum = self.sum + x
	def sub(self, x): self.sum = self.sum - x
    i = _alsoWorks_t()
    i.sum = 0
    i.add(2)
    i.add(3)
    i.sub(1)
    return i.sum

--===-----========------------- Sana esprimo naskas sanan ideon.
SUZUKI Hisao            suzuki611 at okisoft.co.jp, suzuki at acm.org.



From mwh21 at cam.ac.uk  Wed Jun 23 13:28:34 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 23 Jun 1999 18:28:34 +0100
Subject: set im_self
References: <7kqokb$7as$1@ns.mtu.ru>
Message-ID: <m3909a3jy5.fsf@atrus.jesus.cam.ac.uk>

"Oleg Orlov" <orlov at diasoft.ru> writes:

> Hello.
> Is there any way to set im_self attribute (ie bind method object)?
> Thanks.

The `new' module can do this:

>>> class C:
...  def m(self):
...   print self
...
>>> c=C()
>>> c.m()
<__main__.C instance at 8111318>
>>> import new
>>> cm=new.instancemethod(C.m.im_func,c,C.m.im_class)
>>> cm()
<__main__.C instance at 8111318>

Be aware that RedHat up to version 5.2 (I think) shipped a Python
without the new module compiled in, so you'll need to recompile or
update from an Andrich rpm if you're lumbered with one of those.

HTH
Michael



From paul at prescod.net  Wed Jun 30 19:53:38 1999
From: paul at prescod.net (Paul Prescod)
Date: Wed, 30 Jun 1999 23:53:38 GMT
Subject: Mailbox, tell and read
Message-ID: <377AAE02.E8591727@prescod.net>

The mailbox message objects simulate a file interface by seeking through a
mailbox. A read() returns the entire message. It does this by seeking to
the start, reading to the message end, and returning the data. The problem
is that message offsets are derived by file.tell() which I presume is in
terms of bytes. But then it does a read() based on the offsets and read()
is in terms of characters. 

These don't match up on Windows because of CR/LF pairs. I think that the
right fix is to never use tell() and instead keep track of the location by
counting characters. Does that make sense?
-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for only himself
 http://itrc.uwaterloo.ca/~papresco

"I don't want you to describe to me -- not ever -- what you were doing
to that poor boy to make him sound like that; but if you ever do it
again, please cover his mouth with your hand," Grandmother said.
        -- John Irving, "A Prayer for Owen Meany"



From johnga at primenet.com  Sat Jun 19 11:47:10 1999
From: johnga at primenet.com (John E. Garrott)
Date: Sat, 19 Jun 1999 08:47:10 -0700
Subject: Newbie question
Message-ID: <376BBB7E.A82DA789@primenet.com>

I am have a problem with the program below.
The intent is, eventually, to get a list and display it,
in a list box, then print the selected row
from the listbox, but I can't get the callback
to work.

The function dbs.databases() works, putting,
in this case, the databases owned by myself
into the listbox.  When I click on the listbox
items they highlight as expected, but, apparently,
list_cb() is never called.

I've read all the available literature on 
binding and signals, tried numerous variations
on what is here.  No results.  This program
shows no errors, so is syntactically(sp?) 
correct.  Just doesn't work.

What am I doing wrong?

Thanks in advance,

John
--------------------------------------------------------
import dbs
import string
from Tkinter import *

class SelectionBox:

	def __init__(self, master=None):
		self.frame = Frame(master, name="frame")
		self.frame.pack(expand=1, fill=BOTH)

		self.frame.listbox = Listbox(self.frame, name='listbox')
		self.frame.listbox.pack(expand=1, fill=BOTH)
		self.frame.listbox.bind('Button-1', self.list_cb)

		for item in dbs.databases():
			self.frame.listbox.insert(END, item)

	def list_cb(event):
		print "Here we are!"

def main():
	root = Tk()
	sp = SelectionBox(root)
	root.minsize(1, 1)
	root.mainloop()

main()



From arcege at shore.net  Wed Jun  9 10:38:57 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 09 Jun 1999 14:38:57 GMT
Subject: Python/C API and setattr
References: <375E5257.8579541B@serop.abb.se>
Message-ID: <50v73.2457$nn.746290@news.shore.net>

Thomas S. Strinnhed <thstr at serop.abb.se> wrote:
: Hi

: I'm building an extention type for Python in C using NT, MSVC++.
: It's built as a dll to link in at Runtime. 

: I'm doing this for practice so the type is quite simple, it's
: a CounterType, with an attribute called value (an int)
: and the simple operations/methods inc() and dec().

: I've looked pretty much at Mark Lutz's example StackType  from
: Programming Python and I think I'm doing the same, but I've added
: setattrfunc to be able to directly manipulate value as c.value = 23
: and when I use it something goes wrong.
: (inc() and dec() work fine)

: Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
: Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
:>>> from CounterType import *
:>>> c = Counter()
:>>> c.value = 23
: Traceback (innermost last):
:   File "<stdin>", line 1, in ?
: SystemError: error return without exception set
:>>> c.value
: 23
:>>>
: The value is set properly but I can't figure out the system error, it
: seems to be something wrong with my setattrfunc. 

: static void counter_setattr
: 	(counterobject *self, char *name, PyObject *v)
: {
: 	if(strcmp(name, "value") == 0)
: 	{
: 		self->value = (int)PyInt_AsLong(v);
: 	}
: }

: But what would that be??
: I have a hard time finding examples of how to do setattr so
: any pointers would be great. Also if someone understands this error,
: please post an answer.

: My counterobject is this:
: typedef struct
: {
: 	PyObject_HEAD
: 	int value;
: } counterobject;

: Thanks 
:  -- Thomas S. Strinnhed, thstr at serop.abb.se

Thomas, the prototype for functions in the setattr slot of a Python type
is:
  typedef int (*setattrfunc)(PyObject *, char *, PyObject *);

Your counter_setattr function should return an int, true (0) for
success, false (-1) for error.  I believe the source code and Mark's
book (somewhere in Ch 15? but I don't have the book with me) are about
the only places this is documented, and it is very obscure.

"This looks like a job for Documentation Man!"

  -Arcege




From alan.gauld at gssec.bt.co.uk  Thu Jun 24 08:50:01 1999
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Thu, 24 Jun 1999 13:50:01 +0100
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0rr2.2fi.behrends@allegro.cse.msu.edu>
Message-ID: <37722979.2FCD9A5F@gssec.bt.co.uk>

Reimer Behrends wrote:
> _am_ getting pretty sick of having to type self.whatever countless
> times, where pretty much any other language allows me to discard the
> "self." part. Not only is it annoying to type, it also reduces the
> readability of OO code a lot, much like the gratuitious use of
> punctuation characters as variable prefixes in Perl does.

I used to think that, but after some time in python land I 
just finished reviewing the code of a C++ project (about 
5,000 new lines, 7 new classes in a system with 200,000 
lines and about 200 classes.).

I was amazed to discover how difficult it was to understand 
the code because I couldn't identify which variables were 
local to the function, which were instance variables and 
which were global.

This gets worse when dealing with multiple inheritance, 
I wound up spend significant time in the headers directory 
doing grep... The self notation makes it clear and pretty 
unambiguous.

Alan G.

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



From amk at mira.erols.com  Mon Jun 28 19:21:21 1999
From: amk at mira.erols.com (A.M. Kuchling)
Date: 28 Jun 1999 23:21:21 GMT
Subject: SWIG and binary strings?
References: <al8pv2nwnny.fsf@myntti.helsinki.fi> <3773a918.0@news.cyberway.com.sg> <al8bte0udlr.fsf@myntti.helsinki.fi>
Message-ID: <slrn7ng18h.du.amk@207-172-38-191.s191.tnt8.ann.va.dialup.rcn.com>

On 28 Jun 1999 07:57:04 +0300, Markus Stenberg <mstenber at cc.Helsinki.FI> wrote:
>Ah, I've concentrated on SSL itself (TLS actually, but still; full TLS/SSL
>functionality, no ciphers whatsoever wrapped). I wish someone wrapped whole
>OpenSSL and maintained it.. ;-) (I'd do it, but I'm too lazy by half)

mxCrypto (http://www.cs.uni-duesseldorf.de/~lemburg/mxCrypto.html)
wraps the cipher implementations in OpenSSL; talk to M.A. Lemburg, the
author of mxCrypto, about incorporating your SWIG interfaces into
mxCrypto.  I believe he's been wanting to wrap the SSL interfaces for
a while; you may be able to save him some time.

	(mxCrypto + SSL socket support would be a wonderful thing to
have, particularly if it's developed outside of the US.)

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
It would be nice to be unfailingly, perpetually, remorselessly funny, day in
and day out, year in and year out until somebody murdered you, now wouldn't it?
    -- Robertson Davies, _The Diary of Samuel Marchbanks_
 



From landrum at foreman.ac.rwth-aachen.de  Fri Jun 25 09:51:32 1999
From: landrum at foreman.ac.rwth-aachen.de (Gregory A. Landrum)
Date: 25 Jun 1999 15:51:32 +0200
Subject: version of python?
References: <3773866C.9D29C071@phoenix-edv.netzservice.de>
Message-ID: <szp1oqtgb.fsf@foreman.ac.rwth-aachen.de>

>How could I get the version-number of the installed python-interpreter?

Python 1.5.2 (#3, May 23 1999, 10:48:24) [C] on aix4
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import sys
>>> sys.version
'1.5.2 (#3, May 23 1999, 10:48:24) [C]'

-greg

---------------------
Dr. Greg Landrum  (landrumSPAM at foreman.ac.rwth-aachen.de)
Institute of Inorganic Chemistry
Aachen University of Technology



From chad at vision.arc.nasa.gov  Wed Jun  2 14:00:28 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Wed, 02 Jun 1999 11:00:28 -0700
Subject: All this self dot is making me...
References: <Vn653.468$HD.99317@typhoon-sf.snfc21.pbi.net>
Message-ID: <3755713C.E59F86AF@vision.arc.nasa.gov>

Jim Rudnicki wrote:

> For those of us who have less cartilage left in their hands, would it be
> acceptable to use "my." or better "me."  Or is self. strongly ingrained?
> The two less characters add up and the keys are on stronger fingers.

First of all, you should probably look into an alternate keybard or keyboard
layout, such as Dvorak, if this is a chronic problem.  Also, drink lots of
water; I've found it to be the best defense against keyboard/mouse related
hand problems.  Also, use an abbreviation expansion mode of your editor,
or find a better editor; these are VERY helpful when using descriptive names.

> def Func( this, a, b ):
>   this.x = a + b
>
> class Foo:
>   def __init__( self ):
>     self.x = 0
>     self.f = Func
>   def fooFunc( me, a, b ):
>     me.x = a + b
>
> # who needs class methods?
> z1 = Foo()
> Func( z1, 5, 7 )
> print z1.x

Numeric Python uses this approach (mainly), since these functions can then be
made to work on more than one type of object.  So, if Func() is meant to apply
to a number of objects, this is probably the way to go.  However, it would be
better written as:

def Func(a,b):
   return a + b

So that is doesn't rely on the behavior of "this", which would need to support
attribute assignment.  If you are going to alter an object's attributes, it is often
best to use a method on that object, to "encapsulate" its behavior.

But, if you really just want to save typing, I'd suggest letting the computer help
you, or change some habits, but don't make obfuscated code.  Using "this" instead
of "self" should work; but use it consistently.  I'd suggest the abbrev expansion
is a better solution for you.

Chad Netzer





From doughellmann at mindspring.com  Wed Jun 23 10:18:11 1999
From: doughellmann at mindspring.com (Doug Hellmann)
Date: Wed, 23 Jun 1999 14:18:11 GMT
Subject: Tk "save as" dialog
References: <7kp071$peg$1@oceanite.cybercable.fr>
Message-ID: <3770ED6A.8F173340@mindspring.com>

BoB wrote:
> 
> For the moment, I'm using Tk "save as" dialog box. But it's plenty of buggs,
> and I really can't use it any more.

What sort(s) of problems are you having with it?

Doug



From claird at Starbase.NeoSoft.COM  Tue Jun 29 11:58:03 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 29 Jun 1999 10:58:03 -0500
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <lw67471fyi.fsf@copernico.parades.rm.cnr.it> <7laoq0$bvh$1@Starbase.NeoSoft.COM> <lwso7bdnia.fsf@copernico.parades.rm.cnr.it>
Message-ID: <7laqeb$e17$1@Starbase.NeoSoft.COM>

In article <lwso7bdnia.fsf at copernico.parades.rm.cnr.it>,
Marco Antoniotti  <marcoxa at copernico.parades.rm.cnr.it> wrote:
			.
			.
			.
>I have no problems with that.  But the following example is pretty
>hard to beat. :)
>
>==============================================================================
>* (defmacro zut (a) `(list ,a))
>ZUT
>
>* (defun zot (x) (zut x))
>ZOT
>
>* (compile 'zot)
>ZOT
>
>* (disassemble 'zot)
>071B5790:       .ENTRY ZOT(x)                ; (FUNCTION (T) LIST)
>     7A8:       ADD   -18, %CODE
>     7AC:       ADD   %CFP, 32, %CSP
>
>     7B0:       CMP   %NARGS, 4              ; %NARGS = #:G0
>     7B4:       BNE   L0
>     7B8:       NOP
>     7BC:       MOVE  %A0, %A4               ; %A0 = #:G1
>     7C0:       MOVE  %A4, %A0               ; No-arg-parsing entry point
>     7C4:       ADD   4, %ALLOC
>     7C8:       ANDN  %ALLOC, 7, %A3
>     7CC:       OR    3, %A3
>     7D0:       MOVE  %A3, %A1
>     7D4:       ST    %A0, [%A1-3]
>     7D8:       ST    %NULL, [%A1+1]
>     7DC:       TADDCCTV 4, %ALLOC
>     7E0:       MOVE  %A3, %A0
>     7E4:       MOVE  %CFP, %CSP
>     7E8:       MOVE  %OCFP, %CFP
>     7EC:       J     %LRA+5
>
>     7F0:       MOVE  %LRA, %CODE
>     7F4:       UNIMP 0
>     7F8: L0:   UNIMP 10                     ; Error trap
>     7FC:       BYTE  #x04
>     7FD:       BYTE  #x19                   ; INVALID-ARGUMENT-COUNT-ERROR
>     7FE:       BYTE  #xFE, #xED, #x01       ; NARGS
>     801:       .ALIGN 4
>* 
>==============================================================================
>
>Of course I could have defined a much more intricated macro.
			.
			.
			.
As it happens, there's very good work going on just now
to beef up Tcl's introspective capabilities.  Is *that*--
introspection--the real content of your preference?
Incidentally, many of the same ideas and possibilities
are available to Python, although I don't know of anyone
actively pursuing them for Python.  While Tim Peters im-
presses me with, among much else, his ability to code
clever little methods that tease all sorts of informa-
tion from a Python interpreter, I occasionally argue that
'twould be worth the effort to do introspection for Py-
thon in a more unified way.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From hinsen at cnrs-orleans.fr  Tue Jun  1 04:52:48 1999
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 01 Jun 1999 10:52:48 +0200
Subject: Compiling netCDFmodule on NT: Error
References: <374D90FD.51356F8D@bigfoot.com>
Message-ID: <m3ogj0e1zz.fsf@chinon.cnrs-orleans.fr>

Hoon Yoon <hyoon at bigfoot.com> writes:

>   After rediscovering compile.py by David, I am on the way to compiling
> some addin modules.
>   One of the things that I really want to check out is Konrad's
> netCDFmodule. When I attempt to compile it, I get following error.

Please try the alpha release of ScientificPython 2.0, which contains
a modified netCDF interface. This has been compiled under Windows.
The URL is ftp://dirac.cnrs-orleans.fr/pub/ScientificPython-2.0a5.tar.gz
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------



From arcege at shore.net  Tue Jun  8 10:48:57 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 08 Jun 1999 14:48:57 GMT
Subject: I'll pay $40 in e-gold if you'll fix this bug.
References: <7jigrv$ppo$1@xs3.xs4all.nl> <375CF797.FF0D6436@q-bus.de>
Message-ID: <t3a73.2244$nn.704759@news.shore.net>

Jan Walter <jan at q-bus.de> wrote:
: Hi,

: Maybe you get confused a bit by the bash!!!

: Try following:
: bash$ tcsh
: tcsh: cd /tmp/
: tcsh: mkdir dirOne
: tcsh: mkdir dirTwo
: tcsh: mkdir dirTwo/dirThree
: tcsh: ln -s dirTwo/dirThree/ .
: tcsh: cd dirThree/
: tcsh: pwd
: /tmp/dirTwo/dirThree

: Do the same with the bash (removing the directories of course):
: bash$ cd /tmp/
: bash$ mkdir dirOne
: bash$ mkdir dirTwo
: bash$ mkdir dirTwo/dirThree
: bash$ ln -s dirTwo/dirThree/ .
: bash$ cd dirThree/
: bash$ pwd
: /tmp/dirThree

: The tcsh gives the right answer. You are in the /tmp/dirTwo/dirThree directory. If you try
: to move a file with "mv file.txt ../dirOne/file.txt" you can at t do that because there is no
: directory dirOne !!!

: Python works like tcsh. If you try the following you get the right answer:
: bash$ cd /tmp/dirThree/
: bash$ python
: Python 1.5.1 (#1, May  6 1998, 01:48:27)  [GCC 2.7.2.3] on linux-i386
: Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
:>>> import os
:>>> os.getcwd()
: '/tmp/dirTwo/dirThree'

: My suggestion:

: You can test if there are symbolic links in your path (e.g. os.path.islink("/tmp/dirThree")
: within Python) or you can prove that /tmp/dirThree is in reality /tmp/dirTwo/dirThree and
: life with that fact ...

: Cheers,

: Jan

It is not a matter of "correctness", but of perceived current
pathname.  In that same tcsh, if you echo $cwd, you will get the same
value that bash returns,  there are some systems where using this
"faked" value is required (Automounter V1 systems for example).  Both
csh and tcsh have common, documentated add-ons (alias pwd 'echo $cwd')
to give the same functionality as bash.

But, the getcwd(3) C function performs a call to traverse up the tree
to retrieve the pathname.  This is because a process only holds the
inode of the current directory, not the path that got it there.

Using relative pathnames with symbolic links is problematic at times,
but they aren't "incorrect."  Symbolic links have to be organized
properly; if they are not, you have odd side-effects.  These are the
Facts of Life (I always like "Jo" myself).

  -Arcege




From llwo at dbtech.net  Mon Jun 14 08:07:15 1999
From: llwo at dbtech.net (Karl Putland)
Date: Mon, 14 Jun 1999 07:07:15 -0500
Subject: New to threading. Right idea???
References: <Pine.WNT.4.05.9906132220070.103-100000@david.ski.org>
Message-ID: <3765775d.0@news.dbtech.net>

Thanks David.
>
> If you must call EXPENSIVE for each directory, then you must -- I don't
> think there's any way around that.  Loading the same DLL multiple times
> won't help if you have to call EXPENSIVE for each (especially if
> unexpensive is, well, unexpensive =).  If EXPENSIVE is VERY_EXPENSIVE,
> where VERY_EXPENSIVE >> cost of loading the DLL, then even a fork()
> wouldn't help much, even if it were available on your OS.
>
> --david ascher

I experimented with creating a copy of the dll with a diferent name and
loading it.  This worked.  I could call EXPENSIVE for the dll and the
copy_of_dll and both succeed.

The other interesting part of this is that EXPENSIVE executes over twenty
times faster the second time around.  That is if dll and copy_of_dll are
both loaded, calling dll.EXPENSIVE takes between 1.5 and 4.8 seconds, then
calling copy_of_dll.EXPENSIVE takes .022 - .22.  EXPENSIVE is really called
load_bde().

I guess that I will just manage the files to load one ofr more copies of the
dll since that seems to work like I need it to.

>>> from gm4s32 import *
>>> load_bde()
(1, 'Startup Time: 0.99999997139')
>>> copy_load_bde()
(1, 'Startup Time: 0.0499999523163')
>>>


***snippet from module gm4s32.py***
import windll
import time
import os
import string
from types import *

gm=windll.module('GM4S32')
gma=windll.module('GM4S3a')

##if raw_input('(W)ork or (H)ome: ')=='W':
##    SysDir='G:\\'
##    GoldDir='G:\\'
##    CommonDir='g:\\merged'
##    User='MASTER'
##    Password='ACCESS'
##else:

SysDir='d:\\goldmine'
GoldDir='d:\\goldmine\\gmbase'
CommonDir='d:\\goldmine\\demo'
User='KARL'
Password='KAPPASIG'

def load_bde(sysDir='',
      goldDir='',
      commonDir='',
      user='',
      password=''):

    global SysDir, GoldDir, CommonDir, User, Password

    if not sysDir: sysDir=SysDir
    if not goldDir: goldDir=GoldDir
    if not commonDir: commonDir=CommonDir
    if not user: user=User
    if not password: password=Password

    SysDir=sysDir
    GoldDir=goldDir
    CommonDir=commonDir
    User=user
    Password=password

    start=time.time()
    (sysDir, goldDir, commonDir,
     user, password)=map(windll.cstring,(sysDir, goldDir, commonDir,
        user, password))
    return (gm.GMW_LoadBDE(sysDir, goldDir, commonDir, user, password),
"Startup Time: " + str(time.time()-start))

def unload_bde():
    return gm.GMW_UnloadBDE()

def copy_load_bde(sysDir='',
      goldDir='',
      commonDir='',
      user='',
      password=''):

    global SysDir, GoldDir, CommonDir, User, Password

    if not sysDir: sysDir=SysDir
    if not goldDir: goldDir=GoldDir
    if not commonDir: commonDir=CommonDir
    if not user: user=User
    if not password: password=Password

    SysDir=sysDir
    GoldDir=goldDir
    CommonDir=commonDir
    User=user
    Password=password

    start=time.time()
    (sysDir, goldDir, commonDir,
     user, password)=map(windll.cstring,(sysDir, goldDir, commonDir,
        user, password))
    return (gma.GMW_LoadBDE(sysDir, goldDir, commonDir, user, password),
"Startup Time: " + str(time.time()-start))

def copy_unload_bde():
    return gma.GMW_UnloadBDE()





From mrkbutty at mcmail.com  Tue Jun  1 16:58:01 1999
From: mrkbutty at mcmail.com (Mark Butterworth)
Date: Tue, 1 Jun 1999 21:58:01 +0100
Subject: A tkinter kick-off
Message-ID: <37544a32.0@news1.mcmail.com>

I've been using python for a few months now (I've actually managed to
persuade my company that freeware is a viable tool) and I've just started
looking at tkinter.  However,  my first little project doesn't  require an
interactive window or dialog box.  What I do require is a window which
displays the result's of my background processing.  The processing will
simply be the compression and network transfer of some files.  I'm not
looking for a full solution however, I just need a help starting off.  I can
create a window containing a listbox alright, I just need to know how to
process in the background and display results within the listbox (maybe a
cancel button would also be useful).

Any help appreciated!?!





From graham at sloth.math.uga.edu  Thu Jun 10 11:05:37 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 10 Jun 1999 15:05:37 GMT
Subject: Python 2.0
References: <000a01beb2e8$087fc2a0$af9e2299@tim>
Message-ID: <7jok81$hih$2@cronkite.cc.uga.edu>

[Hisao Suzuki]
> Me too.  And I am afraid that some evangelists would do nothing
> but _only_ point out Python's some so-called fault in order to
> drive us into convert to their favorite or invented language...
[Yukihiro Matsumoto]
> Sorry for bothering the newsgroup.
Tim Peters (tim_one at email.msn.com) wrote:
: Hmm.  The level of paranoia in this thread is quite remarkable!  I don't
: think Suzuki was talking about you (or Ruby) here, and your contributions to
: the thread have been civil and helpful.

Well my reading of Hisao's posts (especially the most recent ones) is
indeed that he was talking about Yukihiro.

Now two data points.

graham
-- 
           As you grow up and leave the playground
       where you kissed your prince and found your frog
          Remember the jester that showed you tears
                   the script for tears



From stidolph at origin.ea.com  Tue Jun 29 16:05:22 1999
From: stidolph at origin.ea.com (Stidolph, David)
Date: Tue, 29 Jun 1999 20:05:22 GMT
Subject: Please help me to understand classes
Message-ID: <11A17AA2B9EAD111BCEA00A0C9B4179301E10F81@molach.origin.ea.com>

Finally, a simple question I can answer!

First of all, class methods like "druk" need to have a parameter declared in
them called self (or whatever you want to call it).  Secondly you need to
call the method with open/close parens:

class A:
  def druk(self):
    print 'Jy is in Class A'

class B:
  def druk(self):
    print 'Jy is in Class B'

a = A()
c = B()
c.druk()
Jy is in Class B
a.druk()
Jy is in Class A

I hope this helps,

David.


-----Original Message-----
From: Johann Spies [mailto:jhspies at futurenet.co.za]
Sent: Tuesday, June 29, 1999 2:04 PM
To: Python-poslys
Subject: Please help me to understand classes


I have been playing around with python for some time now, but I am a
relative newbie as it is my hobby and normally I have very little time for
programming.  I am on leave now and want to write a program but I find
python a very difficult language to comprehend despite the fact that I
have a few years of experience in programming a language like Pascal.
Maybe it is because I have little experience in OOP.

Can somebody explain to me why the following program produces no output:
---------------------------------------------
class A:

    def druk():
        print 'Jy is in Class A'

class B:

    def druk():
        print 'Jy is in Class B'
        
a = A()
c = B()
c.druk
a.druk
---------------------------------------------


 --------------------------------------------------------------------------
| Johann Spies                                 Windsorlaan 19              |
| jhspies at futurenet.co.za                3201 Pietermaritzburg             |
| Tel/Faks Nr. +27 331-46-1310		       Suid-Afrika (South Africa)  |
 --------------------------------------------------------------------------

     "One thing have I desired of the LORD, that will I seek
      after; that I may dwell in the house of the LORD all
      the days of my life, to behold the beauty of the LORD,
      and to inquire in his temple."            
                               Psalms 27:4       



From jim at interet.com  Tue Jun 22 09:52:24 1999
From: jim at interet.com (James C. Ahlstrom)
Date: Tue, 22 Jun 1999 09:52:24 -0400
Subject: Build the Python library into the binary
Message-ID: <376F9518.D3252CEE@interet.com>

Now that Python 1.5.2 is here I need to upgrade from
1.5.  We use Python internally and also ship Python
applications.  Our environment is mixed PC's, Suns and
linux boxes.  Our customers all have PC's.  I always
find upgrades difficult, especially at customer sites since
we have no access, and have to depend on our Wise install
script.

I keep thinking how much easier this would be if the
Python library were built into the binary.  I mean that
on PC's, the library were part of python15.dll for example.
No more worrys about a bad PYTHONPATH, the registry, my
1.5.2 binary using 1.5 libraries, the different ways all
this works on different platforms, etc.  You just replace
python15.dll and you are at 1.5.2.  Simple.

So I wrote some code.  I don't consider it finished but
it works.  It makes the Python library part of the binary.

1) The same code works for Unix and Windows.  No "#ifdefs".
2) It uses the existing "frozen module" logic which has been
   used as a way to ship Python main programs.
3) It extends frozen modules so that multiple frozen modules
   can be linked together at run time.  Currently there can only
   be one frozen module.
4) Multiple frozen modules are built as regular Python extension
   modules, and they install themselves when they are initialized.
   You request them in Setup as usual.  This was the only way I knew
   to do the initialization in a portable way (suggestions?).
5) The C code is stupidly simple and very short.  New code goes
   into frozen.c, and few other changes are required.
6) There is a Python program in Tools/freeze which creates the Python
   extension module with the Python libraries.  It uses the new
   "modulefinder.py" program.  Thanks to whoever wrote it!  It can
   also be used to create other frozen modules, thus solving the
   problem of how to make a frozen module.
7) The extension module need not be imported itself.  It installs
   the built-in frozen modules jusy by being initialized.  But if
   you import it, it has methods to turn the frozen modules on and
   off, and to print the names in the modules.

So my questions are:

1) Am I the only one having trouble keeping the Python binary and
   library files in sync?
2) Is any of this of interest?  Should it be added to Python?
3) What features should be included?  Does this do enough, too much?
4) Is this the right way to do it.

More technical details are available on request.

Jim Ahlstrom



From bwarsaw at cnri.reston.va.us  Tue Jun 15 17:11:05 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Tue, 15 Jun 1999 17:11:05 -0400 (EDT)
Subject: easy filecopy? - JPython?
References: <376531D3.7E42AC42@phoenix-edv.netzservice.de>
	<aahzFDCq4D.Do1@netcom.com>
	<PBt93.843$B04.2678@newsfeed.slurp.net>
	<14182.28241.949065.590839@weyr.cnri.reston.va.us>
Message-ID: <14182.49513.956557.398784@anthem.cnri.reston.va.us>

>>>>> "Fred" == Fred L Drake <fdrake at cnri.reston.va.us> writes:

    Fred>   shutil.copyfile() works just fine (JPython 1.0.3 on
    Fred> java1.2), but shutil.copymode() relies on os.chmod(), which
    Fred> isn't there.  I'm sure Barry Warsaw (bwarsaw at python.org)
    Fred> would love to receive a patch for JPython to support that!
    Fred> ;-)

Probably not possible in 100% Java.  If os.chmod() is really important
to you, then you should check out Finn Bock's JNI implementation of os
and posix.

http://pip.dknet.dk/~pip1848/jpython/Readme_jnios.html

-Barry



From mesak1234 at my-deja.com  Sun Jun  6 11:25:15 1999
From: mesak1234 at my-deja.com (mesak1234 at my-deja.com)
Date: Sun, 06 Jun 1999 15:25:15 GMT
Subject: Newbie question
Message-ID: <7je3sk$46$1@nnrp1.deja.com>

Hi!
Well, i get problems with getting date from OS....(win32 at this
moment)... All what i need that to find out date attribute of a file.
But i even cannot find out what the function i can use to get the real
time (anything like Date date = new java.util.Date(); ???? :( )
All functions of 'time library' seem fine but don't help so much...
Any suggestions?
Thanks,
  Daniel


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From riftweaver at hotmail.com  Thu Jun  3 11:40:48 1999
From: riftweaver at hotmail.com (riftweaver at hotmail.com)
Date: Thu, 03 Jun 1999 15:40:48 GMT
Subject: unable to access Tk from Python152/WinNT
Message-ID: <7j67lu$ibe$1@nnrp1.deja.com>

Hi,

I am trying to run the tk interface from the Python 1.5.2 release on
WindowsNT. I have installed Python with both a pre-existing Tcl/TK
installation, and with the included Tcl/Tk installation.

When I couldn't start idle, I followed the troubleshooting selection
from python.org. First Pyhon couldn't find a lib/Tcl8.0 directory; so I
copied that from the Tcl install to Python/lib. Then it couldn't find a
lib/Tk8.0 directory; again I copied from Tcl install to Python/lib.

Then I got a stack trace with an error about binding to a <MouseWheel>.

Any help or suggestions are appreciated. Thanks.

--->Jason


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From sams at quinta.com  Fri Jun 18 15:34:03 1999
From: sams at quinta.com (Sam Schulenburg)
Date: Fri, 18 Jun 1999 19:34:03 GMT
Subject: Crawling Python
References: <000601beb91d$e1321ae0$839e2299@tim> <376a1ac1.10261180@news.tpnet.pl>
Message-ID: <7ke6va$qa9$1@nnrp1.deja.com>

In article <376a1ac1.10261180 at news.tpnet.pl>,
  mark at _spamspamlovelyspam_btweng.krakow.pl (MK) wrote:
> On Thu, 17 Jun 1999 20:02:39 -0400, "Tim Peters"
<tim_one at email.msn.com> wrote:
>  -- clip clip --
> Strangling IDLE continued:
>
> in DOS box I tried
>
> >>> s=''
> >>> for i in range(10000):
> ...     s=s+`range(i,i+20)`
> ...
>
> >>> len(s)
> 1178515
>
> and everything is OK.
>
> While the same loop in IDLE crashed it (I mean it hangs for some
period, then it
> stops responding and has to be killed).
>
> MK
> -- clip  clip --

I just tried both environments on a WindowsNT system and obtained
130000 for len(s) under IDLE and Console mode python.
IDLE took approx. 25 seconds and Console took approx 20 seconds.


Sam Schulenburg


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From guido at cnri.reston.va.us  Tue Jun  1 10:20:52 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 01 Jun 1999 10:20:52 -0400
Subject: need help checking for stack overflow in Win32
Message-ID: <5lk8to3su3.fsf@eric.cnri.reston.va.us>

I could use some help from someone with a little time, some knowledge
of Python internals, and some Win32 C expertise.  On Win32 platforms,
infinite recursion causes an Application Error (a fatal windows
error).  There is some machinery in Python to check for this
condition, but apparently it doesn't work on Win32.  Please help!

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



From arcege at shore.net  Wed Jun 30 13:22:07 1999
From: arcege at shore.net (arcege at shore.net)
Date: Wed, 30 Jun 1999 18:22:07 +0100
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
Message-ID: <000015c2@bossar.com.pl>

From: arcege at shore.net

Fredrik Lundh <fredrik at pythonware.com> wrote:
: Michael P. Reilly wrote:
:> Hmm... something about that argument makes it circularly invalid.
:> Non-idle event a comes in, but must be handled after idle event d.
:> Event d can only execute when event a is taken care of.  Sounds like
:> deadlock to me. ;)

: hmm.  I thought I'd meant that idle tasks kept piling up
: when the windowing system generates a lot of events
: at once (like when you're redrawing an entire system).
: instead of redrawing everything at once, Tkinter waits
: until the window event queue is empty...

Yes, indefinate postponement is a problem with the current definition
of Tcl_DoOneEvent.  But idle tasks are for when the event loop is idle,
and indefinate postponement is the standard, accepted trade-off for
that.  Timer handlers exist for handling more time critical actions.

You had made the comment(s) earlier in the thread:
> The interface is only part of the story; many complaints
> seem to come from the fact that Tk uses idle tasks to
> redraw things.  When an "expose" event arrives from
> the windowing system, Tk calls "after_idle", rather
> than redrawing the widget at once.  And idle tasks
> are only executed when there are no other events
> to take care of...

This lead me to think that you meant windowing events (such as expose)
were handled as idle tasks and I don't see this (MotionNotify events,
yes, not all others are handled by Tcl_QueueEvent).  Hence my comment
about the deadlock.

  -Arcege



From mnot at pobox.com  Tue Jun 22 05:31:04 1999
From: mnot at pobox.com (Mark Nottingham)
Date: Tue, 22 Jun 1999 09:31:04 GMT
Subject: htmllib: CR in CDATA
References: <00d201bebc5a$adb129a0$0301a8c0@mnot.net>
Message-ID: <012601bebc91$f3ded760$0301a8c0@mnot.net>

Whooops, nevermind, I misread the spec -- carriage returns are turned into
spaces (which is what htmllib does) - *line feeds* should be ignored...

--
"Get me the phone book."
  "Which one?"
"Doesn't matter."


----- Original Message -----
From: Mark Nottingham <mnot at pobox.com>
To: Python <python-list at cwi.nl>
Sent: Tuesday, June 22, 1999 12:55
Subject: htmllib: CR in CDATA


> It appears that htmllib doesn't ignore returns in CDATA fields, as HTML
4.0
> says it should:
> http://www.w3.org/TR/REC-html40/types.html#type-cdata
> http://www.w3.org/TR/REC-html40/sgml/dtd.html
>
> As a result, htmllib improperly parses any CDATA element that wraps across
a
> line; this affects elements like
>
> <A href="foo.
> gif">
>
> I'm happy to work up a patch, but I thought I'd ask around first. It may
be
> a bit involved to fix it properly; every CDATA should be handled this way,
> which practically means almost every tag attribute.
>
> Regards,
>
>
> Mark Nottingham, Melbourne Australia
> mnot at pobox.com  http://www.mnot.net/
>
>
>




From wc at zyan.com  Thu Jun 17 11:36:10 1999
From: wc at zyan.com (John B. Williston)
Date: Thu, 17 Jun 1999 15:36:10 GMT
Subject: Newbie: Truth values (three-valued logic)
References: <000f01beb7b8$2722a540$2ca22299@tim>
Message-ID: <376915d7.2445700@news.dslspeed.com>

On Wed, 16 Jun 1999 01:21:57 -0400, "Tim Peters"
<tim_one at email.msn.com> wrote:

>we don't need no stinking excluded middle!

Cool! Not cool! You're so right. That's completely wrong.

John



From movits at lockstar.com  Wed Jun 23 16:06:49 1999
From: movits at lockstar.com (Mordy Ovits)
Date: 23 Jun 1999 13:06:49 PDT
Subject: Thanks for Python...
References: <376ead64@anonymous.newsfeeds.com> <VrDQRddcbbs9-pn2-QdftOXYk6Hdd@Fortress>
Message-ID: <37716901.43CB604@lockstar.com>

Jeff Rush wrote:
> 
> Welcome to the Python community, John.  Yes, Python is a truly
> wonderful language for cranking out quick prototypes or test jigs.
> I prototyped a complete web-based general ledger system in Python,
> figuring I'd downcode it to C for speed later, but so far I haven't needed
> to.  Python is also just plain fun...  Now if I could just get time to port
> it to the PalmPilot, I'd never be away from it.
> 
> -Jeff Rush

Get a Windows CE Palm-sized PC if you want to take Python with you whereever you
go.  Python runs stunningly well on my Casio E-100, and has nifty GUI
libraries.  It's not polished by any sense of the word but it works, and the
source is a barely changed version of the standard windows Python.  This makes
for easy improvement.

see:
http://starship.python.net/crew/mhammond/ce/

It sure is nifty to be coding during my commute :-)

Enjoy!
Mordy

-- 
o Mordy Ovits
o Cryptographic Engineer
o LockStar Inc.
---------------------------------------------------------------------------
#!/usr/local/bin/python 
from sys import*;from string import*;a=argv;[s,p,q]=filter(lambda x:x[:1]!=
'-',a);d='-d'in a;e,n=atol(p,16),atol(q,16);l=(len(q)+1)/2;o,inb=l-d,l-1+d
while s:s=stdin.read(inb);s and map(stdout.write,map(lambda i,b=pow(reduce(
lambda x,y:(x<<8L)+y,map(ord,s)),e,n):chr(b>>8*i&255),range(o-1,-1,-1)))



From matz at netlab.co.jp  Mon Jun  7 21:47:36 1999
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 08 Jun 1999 10:47:36 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
 <7jgk4j$s8p$1@cronkite.cc.uga.edu>
 <u3vhcztwwo.fsf@ares.sys1.ngo.okisoft.co.jp>
Message-ID: <87hfoj4g5j.fsf@ev.netlab.co.jp>

Hisao Suzuki <suzuki611 at okisoft.co.jp> writes:

|In article <u3g1443iht.fsf at ares.sys1.ngo.okisoft.co.jp>,
|I also wrote:
|| ... On that box, Ruby did not run but dumped immediately
|| with its twiddling the machine stack, while Python ran perfectly
|| with no difficulty.  ...
|
|We have been enjoying Python's portability for many years.

You are opposing aginst nobody.  No one (even me) claims Python should
abandon its great portability.
						matz.



From a.mueller at icrf.icnet.uk  Mon Jun  7 12:45:21 1999
From: a.mueller at icrf.icnet.uk (Arne Mueller)
Date: Mon, 07 Jun 1999 17:45:21 +0100
Subject: fork()
References: <375BE64A.EB997F24@icrf.icnet.uk> <87so840zcc.fsf@pc-hrvoje.srce.hr>
Message-ID: <375BF721.4B1864F2@icrf.icnet.uk>

Hrvoje Niksic wrote:
> 
> Arne Mueller <a.mueller at icrf.icnet.uk> writes:
> 
> > All that woks fine now (after I spend a long time fiddling with the
> > pipe and select stuff to synchronize I/O). However the forking is a
> > waste of time and memory because all the children get their own
> > large dictionary!
> 
> If you don't change the dictionary, the memory will not be copied.
> Most modern Unixes support COW (copy on write) which means that the
> actual fork()ing does surprisingly little.  Only when the memory is
> written to, a new copy is spawned off.
> 
> If the children are modifying the dictionary, that's well something
> different...

Huh, that's realy good news! But I don't understand how that works, when
the child changes the dictionary it quickly gets it's own copy, so the
operating system is watching the child's activity and copies memory when
it's accessed for writing? However my children don't change anything in
the dictionary ;-)


	Thanks,
	Arne



From asynthREMOVE at THISio.com  Mon Jun 14 16:02:29 1999
From: asynthREMOVE at THISio.com (James McCartney)
Date: Mon, 14 Jun 1999 14:02:29 -0600
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <3762988E.C94AA252@iname.com> <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu> <3763C746.5CA8FE78@iname.com> <m3u2sby4gf.fsf@soma.andreas.org> <87pv2zjhhl.fsf@home.ivm.de>
Message-ID: <asynthREMOVE-ya023180001406991402290001@news.io.com>

In article <87pv2zjhhl.fsf at home.ivm.de>, Klaus Schilling
<Klaus.Schilling at home.ivm.de> wrote:


>Prefix rocks, infix is crap.

No, you mean:

Rocks prefix? Is infix crap?

James McCartney  asynth <at> io <dot> com



From evan at tokenexchange.com  Wed Jun 16 13:14:16 1999
From: evan at tokenexchange.com (Evan Simpson)
Date: Wed, 16 Jun 1999 12:14:16 -0500
Subject: fork()
References: <000b01beb6ef$fcaba960$069e2299@tim>
Message-ID: <9XQ93.2$V4.8796@news-dal.corridex.com>

Tim Peters wrote in message <000b01beb6ef$fcaba960$069e2299 at tim>...
> So my plan for the hour <wink> is:  if M&S finds a cycle, it will clean it
> all up iff no involved object has a __del__.  If an object in a trash
cycle
> does have a __del__, tough, the cycle won't be reclaimed.  Instead we'll
> supply a function you can call if you care, that will return those objects
> (one at a time?  in a list?  whatever), and you do with them what you
will;
> if you don't clean them up, you'll keep getting them back every time you
> call that function until you resurrect them or break the cycle by hand in
> the way that makes most sense to your algorithm.

Presuming the hour isn't up yet, I've a handful of scattered comments:

. How about a callback instead of/in addition to a probe function?  That way
you could flip a switch and get called even on cycles which are "del-clean",
if you wanted to check for unintentional cycle creation.
. "del-cleanness" only requires that actual cycle participants lack __del__,
not dangly bits, right?  Or does this re-introduce analysis that you
intended to throw out?
. (somewhat rhetorical) How the heck do you recognize what you've got when
the machinery hands you an object which isn't otherwise reachable?  I
realize that there's no general answer to this, and that in some specific
cases it would be enough to check __bases__, but the only answer which
occurs to me off the bat is to uniquely mark objects which you expect to
have to handle, or provide them with special __cleanup__ methods, or the
like.  Now we have wide-open policy, which I recall you decrying <wink>.  I
guess that no matter what mechanism is implemented, certain badly-handled
cases will appear and need to be handled with "well, don't do that!"

finalization-is-like-life-you-can't-win-and-you-can't-break-even-ly y'rs
Evan Simpson






From ovidiu at cup.hp.com  Wed Jun 30 20:57:23 1999
From: ovidiu at cup.hp.com (Ovidiu Predescu)
Date: Thu, 01 Jul 1999 00:57:23 +0000
Subject: strange behaviour at termination time
References: <377765F0.38230773@steria.fr> <37783685.C7BB5B2E@compaq.com>
Message-ID: <377ABCF3.3C222F37@cup.hp.com>

Greg Ewing wrote:

> To avoid problems like this, you have to design
> your __del__ methods so that they can do their
> work without having to refer to any module-level
> names at the time they are called. One way is
> to use the default-argument trick:
> 
> class B (A) :
>     def __del__ (self, A = A) :
>         A.__del__ (self)

Can you explain why this trick works? When are the default arguments
stored in the method instance? I guess this happens when the method
instance for __del__ is created by the interpreter, right?

Thanks,
-- 
Ovidiu Predescu <ovidiu at cup.hp.com>
http://www.geocities.com/SiliconValley/Monitor/7464/



From jerome.dannoville at bull.net  Tue Jun 22 08:51:50 1999
From: jerome.dannoville at bull.net (Jerome d'Annoville)
Date: Tue, 22 Jun 1999 14:51:50 +0200
Subject: [JPYTHON] Installation Problems
References: <0D1FF10FB191D45F.A19ABABF1E01F5C3.C3A5123AAD9506A3@lp.airnews.net>
Message-ID: <376F86E6.D17314A4@bull.net>

> ...
>
> D:\Downloading\jpython>jpython
> Bad command or file name
>
> I'm supposing this is because the MakeScriptAction didn't run.
>
> Suggestions welcome!
>

  You must update your PATH, add something like C:\JPython-1.0 depending
where you download JPython. Then jpython.bat might be call from dos
prompt. --Jerome




From gherman at my-deja.com  Wed Jun 23 10:11:30 1999
From: gherman at my-deja.com (Dinu C. Gherman)
Date: Wed, 23 Jun 1999 14:11:30 GMT
Subject: How to obtain an instance's name at runtime?
Message-ID: <7kqpu8$7u1$1@nnrp1.deja.com>

This is a nasty question, perhaps, so I appologize in
advance if its stupidity level is far above the accepted
average in this newsgroup...

I would like to do the following:

>>> class C:
...    pass
...
>>>
>>> c = C()
>>> C.__class__.__name__ # ok
'C'
>>> c.__name__           # not ok
Traceback (innermost last):
  File "<interactive input>", line 0, in ?
AttributeError: __name__
>>>

Question: How to obtain an instance's name at runtime?
That is I'd like to see this happen (one was or the other):

>>> c.__name__
'c'

I've checked the reference manual, but it's silent about
this topic or I'm too blind...

Fearing-there's-no-such-solution'ly,

Dinu

PS: Do not use my Deja email-address.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From jonathan at onegoodidea.com  Fri Jun  4 18:11:29 1999
From: jonathan at onegoodidea.com (Jonathan Hogg)
Date: Fri, 04 Jun 1999 23:11:29 +0100
Subject: python active web pages
References: <jonathan-0406991640390001@borg.onegoodidea.net> <7j98g1$atj$1@autumn.news.rcn.net>
Message-ID: <jonathan-0406992311300001@borg.onegoodidea.net>

In article <7j98g1$atj$1 at autumn.news.rcn.net>, "Jeffrey P Shell"
<jeffrey at Digicool.com> wrote:

>Well, DTML (especially outside of Zope) is not meant to be
>scripting-inside-of-HTML, which most Active Server Page type things are.
>Personally, I think that is beyond hideous, the
>
><HTML>
><%
>    document.write('<HEAD>')
>    document.write('<TITLE>%s</TITLE>' % someTitleString)
>    ...
>%>
></HTML>
>
>is terrible (i have seen plenty of PHP source that looks like the above).

i couldn't agree more. this is what i'm trying to avoid. the most simple
model is to execute python scripts from CGI and build web-pages using
something like HTMLgen or just writing out bits of HTML as strings and
sticking in the dynamic bits in between. this is basically what the ASP
you show above does (hence making the use of ASP quite pointless in my
book).

>This seperation (which was most effective before DTML scripting, aka the
>embedding of python expressions in DTML, was implemented into DTML) is very
>powerful as it enforces the seperation of Code and Presentation, a rather
>MVC (Model-View-Controller) style paradigm.

it's the separation that i'm not sure i want. i have skimmed through the
DTML manual on your web pages and believe i understand (to a first
approximation) how it operates. the problem is that i'm not using the rest
of Zope and if i want to execute setup code, rather than just evaluate
expressions within <#var expr> tags, then i'd need to stick this into a
separate script which sucks in the template and passes in whatever data as
variables.

what i'd really like to be able to write is stuff along the lines of:

   <%
   import foobar

   your_name = foobar.frib( frob = "all" )
   nee = foobar.noneshallpass( what is your_name )
   %>

   <html>
   <head>
      <title>Wotnot</title>
   </head>
   <body>
      "We are the knights who say... <%= nee %>
   </body>
   </html>

all well and good. looks like "asp" can do this fine. but unfortunately it
won't allow me to do the *really* useful bits that ZTemplates supports:

   <h1>How do I love thee?</h1>
   <ol>
      <!--#in ways -->
         <li> <!--#var way -->
      <!--#\in -->
   </ol>

don't get me wrong, while i think the above is a *bit* hideous, it is
undeniably powerful and very close to what i'm looking for.

>My first Bobo app was turning a socket server into a web application.
>Since I had made my classes and the views on those classes rather seperate
>in the socket server implementation, turning into a web app took only a
>matter of hours.  I was able to make a very simple DTML page for displaying
>results from a search form and focus on implementing some new methods
>needed for this new way of accessing data.  Once those methods were
>implemented, I was able to spend the rest of the day in web-designer mode
>since I already knew all the data that I had available to me and I could
>just write DTML and not even have to think about Python.

a very valid point of view. to be honest, how much the integration of
control and interaction will be a pain in the arse is still to be tested.

>DTML (esp. DTML Scripting) has gotten rather beyond control in some cases,
>but I believe it's the most "right".  But yes, I work for digital creations
>so I'm a bit biased.
>
>But all that bias is based on the first time I wrote a major application
>and just how _EASY_ (and powerful) that was.

's ok, we all have our biases. these are still the sort of opinions i was
hoping to garner. thanks for your comments.

does anyone else use active web pages? surely all you python/web hackers
don't just use CGI scripts?

:-j

-- 
jonathan hogg, one good idea limited, 25 clouston street, glasgow g20 8qr, uk
jonathan at onegoodidea.com www.onegoodidea.com t:(0)976-614338 f:(0)7970-537451



From peterh at sapros.com  Fri Jun 11 22:20:02 1999
From: peterh at sapros.com (Peter Haight)
Date: Fri, 11 Jun 1999 19:20:02 -0700
Subject: Higher level FTP lib? 
Message-ID: <199906120220.TAA32246@wartch.sapros.com>

>I wonder if there is a FTP module for python that allows for higher
>level operations than ftplib. With ftplib, I have to parse LIST lines
>myself (footnote: is there a module for *that* like perl's
>File::Listine?) if I want to know file size or date.

You could use urllib instead. It works like this:

import urllib
file = urllib.urlopen('ftp://ftp.python.org/pub/python/src/py152.tgz')

The file object you get back is pretty much the same as a file object you
would get from an open() command.





From fatjim at home.com  Thu Jun  3 12:23:58 1999
From: fatjim at home.com (Jim Meier)
Date: Thu, 03 Jun 1999 16:23:58 GMT
Subject: Handling variable number of arguments.
References: <7j17dm$pks$1@nnrp2.deja.com>
Message-ID: <3756AC9E.649A4D0D@home.com>


bhosu at my-deja.com wrote:

> Hi All,
>    I am embedding Python in an application(C++). I would like to pass
> variable number of args to the python function. I could do that from
> python by using def foo(*args). How do I parse these arguments inside my
> application? In PyARg_ParseTuple, I have to know exactly how many args
> to read etc. I am using Python v 1.4. I was looking at
> PyArg_ParseTupleAndKeywords but that needs the args to be a dictionary.
> I am looking a similar thingy for lists. A typeical scenario of the
> usage is like,
> import foo
> foo.set_args('bar','=','1','2','3')
>
> where foo is the module containing the method implementation.

PyArg_ParseTuple supports optional arguments:

int a,b,c,d,e,f,how_many;
if(!PyArg_ParseTuple("ii|iiii",a,b,c,d,e,f)) return NULL;
how_many=PyTuple_Size(args);

So you could use that if you simply have a known number of optional extra
arguments. See "Extending and Embedding", section 1.7 "Format Strings for
PyArg_ParseTuple". Otherwise, if you need a varying number of arguments
(only really possible if you are calling the function with apply()), do as
Michael P. Reilly's response suggests (access args as a tuple directly).

Hope that helps,
-Jim.




From fredrik at pythonware.com  Thu Jun  3 14:48:13 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 3 Jun 1999 18:48:13 GMT
Subject: idle processing
References: <3755bf01@nntp.server.uni-frankfurt.de> <3755D811.61AEC34@cloud9.net>
Message-ID: <032401beadf1$a30f11f0$f29b12c2@pythonware.com>

Michael Muller <proteus at cloud9.net> wrote:
>
>    win = SomeWindowClass()
>
>    def otherFunc():      
>       win.after('idle', someFunc)

that's win.after_idle(someFunc), btw.

also see the section titled "Alarm handlers
and other non-event callbacks" on

http://www.pythonware.com/library/tkinter/introduction/widget.htm

(now back to single smurf mode).

</F href="http://www.pythonware.com/people/fredrik/busy.htm">




From donn at u.washington.edu  Wed Jun  2 11:42:56 1999
From: donn at u.washington.edu (Donn Cave)
Date: 2 Jun 1999 15:42:56 GMT
Subject: Posix: Exec format error
References: <3754ED47.54C2F0AF@ilt.fhg.de>
Message-ID: <7j3je0$11gk$1@nntp6.u.washington.edu>

Rolf Wester <WESTER at ilt.fhg.de> writes:
| when doing the following:
|
| import os
| id=os.fork()
| if id == 0:
|     os.execl("/home/wester/test/test","-d", "indata")
| else:
|     os.wait()
|
| I get the error message:
|
| OSError: [Errno 8] Exec format error
|
| Can aybody tell me how the format should look like?

Probably it should have a first line like
#!/bin/sh

In other words, the format error is in /home/wester/test/test, not
in your Python statements here.  That file needs a code in its first
16 bits, that tells the OS how to execute it.  If that code is "#!",
it also uses the rest of the that line as an interpreter file path.
So I guess your file lacks its interpreter header, though it could
also be a compiled binary from the wrong operating system.

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



From jo at delorges.in-berlin.de  Wed Jun  2 20:38:47 1999
From: jo at delorges.in-berlin.de (Jo Meder)
Date: 03 Jun 1999 02:38:47 +0200
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net> <aahzFCpHw8.Gxt@netcom.com> <7j4e2u$8rh$1@m2.c2.telstra-mm.net.au> <aahzFCq5CC.KFK@netcom.com>
Message-ID: <m3d7zeglt4.fsf@delorges.in-berlin.de>

aahz at netcom.com (Aahz Maruch) writes:
> if I need to add a quick "if" statement
> during debugging, I don't like having to make sure I clean up the block
> structure formatting.  With just about any other language, I don't have
> to.  I do, as I think I did make clear, clean up the formatting after
> the debugging.

This seems to be more like a request for editing features than a gripe 
about syntax. 

I use the python-mode for emacs, where I can C-c > to indent a region
and C-c < to dedent it. Couldn't be easier.


	Jo.






From Cepl at fpm.cz  Wed Jun 16 10:25:18 1999
From: Cepl at fpm.cz (Matej Cepl)
Date: Wed, 16 Jun 1999 15:25:18 +0100
Subject: HTML bettering?
Message-ID: <1318D78C9072D11195C9006094EA98A72C214D@ocesrv>

How to make as simple as tool which would make some simple operation
over HTML (e.g., headings numbering)? If I would use htmllib.py, I would
have to build HTMLformatter or I am wrong and it is easy?

		Thanks

				Matthew



From greg.ewing at compaq.com  Sun Jun 27 23:50:43 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Mon, 28 Jun 1999 15:50:43 +1200
Subject: ANN: Stackless Python 0.2
References: <37628EAA.C682F16C@appliedbiometrics.com> <7l6gc8$du4$1@news.tamu.edu>
Message-ID: <3776F113.8573DAA5@compaq.com>

Corran Webster wrote:
> In article <37628EAA.C682F16C at appliedbiometrics.com>,
> Christian Tismer  <tismer at appliedbiometrics.com> wrote:
> >ANNOUNCING:

> >               Stackless Python 0.2
> >           A Python Implementation Which
> >             Does Not Ese The C Stack

This sounds interesting, but how do you deal with the 
issue of C extensions which call back to Python code?
Or is this simply disallowed in Stackless Python?

Greg



From jam at newimage.com  Sun Jun 13 14:14:14 1999
From: jam at newimage.com (jam)
Date: Sun, 13 Jun 1999 14:14:14 -0400
Subject: Variations on a theme by "pound bang"
In-Reply-To: <3763e75f.3265255@news.iol.ie>; from Sean Mc Grath on Sun, Jun 13, 1999 at 05:17:33PM +0000
References: <3763e75f.3265255@news.iol.ie>
Message-ID: <19990613141414.D723@toast.internal>

On Sun, Jun 13, 1999 at 05:17:33PM +0000, Sean Mc Grath wrote:
> 
> Can anyone shed some light on the benefits of:-
> 
> # !/usr/bin/env python
> 
> as opposed to
> 
> # !/usr/bin/python
> 
> Sean
> 

yes. 

the first one is more generic, and independent of the actual location of the
command in question (in this case, the python interpreter). there's a man
page for 'env'-- it basically reads the PATH variable, and looks for it's
argument on it. a danger is that it might leave your scripts open to attack
(if someone manages to place a trojan program somewhere in the path *before*
the actual 'python' interpreter, say in $HOME/bin).

the 2nd one is specific to finding the python interpreter in '/usr/bin',
which might be fine for you, but I might install mine in /usr/local/bin
instead, for example. it's a bit more secure in that someone will have to
compromise /usr/bin, which is (theoretically) more difficult.

hope that helps.

regards,
J
-- 
|| visit gfd <http://quark.newimage.com/> 
|| psa member #293 <http://www.python.org/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>



From mlh at idt.ntnu.no  Sat Jun  5 15:51:05 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 05 Jun 1999 21:51:05 +0200
Subject: Deleting characters from a string
References: <87n1ye3g4v.fsf@pc-hrvoje.srce.hr>
Message-ID: <y0jyahye89i.fsf@vier.idi.ntnu.no>

Hrvoje Niksic <hniksic at srce.hr> writes:

> string.translate can delete characters from a string, as well as
> translate them.  But what if I want *only* to delete them?  Currently
> I do this:
> 
> dummytable = string.maketrans('', '')
> 
> ...
>   host = string.translate(host, dummytable, ' "')
> ...
> 
> What I would really like to be able to do is write:
> 
>   host = string.translate(host, None, ' "')

Just to butt in with my private opinion... :) I think it would be more
natural if bot the last two arguments were optional... Thus you could
do:

    host = string.translate(host, deletechars=' "')

To me that is more readable, and seems more like standard Python than
defining None to mean an ID-mapping... Just use an ID-mapping as a
default argument in string.translate.

> 
> Am I missing a more elegant solution?

result = ""
for char in host:
    if char not in ' "':
        result = result+char

host = result


What!? To me that's more elegant... <wink>

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From akuchlin at mems-exchange.org  Tue Jun  1 13:33:17 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: Tue, 1 Jun 1999 13:33:17 -0400 (EDT)
Subject: Whats missing?
In-Reply-To: <37541767.AB2B250B@cwcom.net>
References: <37541767.AB2B250B@cwcom.net>
Message-ID: <14164.6493.227317.96564@amarok.cnri.reston.va.us>

Daniel Faulkner writes:
>from urllib import urlopen
>Data = urlopen("http://www.python.org/").read()
>print Data
 ...
>        Data = urlopen('"' + UrlCurr + '"').read()

	 Why are you adding quotes here?  In the first example, you're 
passing the string 'http://www.python.org'; the double or single
quotes are just part of the language notation for a string literal.
UrlCurr is, I assume, already a string, so Data =
urlopen(UrlCurr).read() should work fine.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
Luckily, every good story is allowed one astounding coincidence. The writer
has decided to cash his in now.
    -- In Phil Foglio's STANLEY AND HIS MONSTER #2



From graham at sloth.math.uga.edu  Mon Jun 14 23:14:20 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 15 Jun 1999 03:14:20 GMT
Subject: Solution to finalisation problem [Re: fork()]
References: <000301beb6c9$a01d2960$b2a22299@tim>
Message-ID: <7k4gec$6vg$1@cronkite.cc.uga.edu>

Graham Matthews:
> If you run *all* finalisers of *all* collected objects before you
> actually delete or partially delete any of those objects then all
> finalisers run in well defined contexts and there are no problems
> with finalisers.
Tim Peters (tim_one at email.msn.com) wrote:
: Sure there are:  finalizers in cycles remain largely useless in practice
: without a way to specify the order in which they're invoked.  Go back to the
: Java part of the thread.  The language can't know whether a cycle needs to
: be finalized in a particular order, or, if it does, what that order may be.
: In Java this has nothing to do with partial deletion:  the spec is very
: clear that the implementation must guarantee that all objects reachable from
: a finalizer are wholly intact when the finalizer is run.  What it doesn't
: say-- because it can't --is that all objects reachable from a finalizer will
: be in a "useful" state.  

I don't understand this phrase "useful" state. If you run a two pass
collector then the order in which you finalise doesn't matter. All
objects will be in a state in which they can be finalised in pass 1 since
nothing is deleted in pass 1. I don't see the problem here at all.

Tim Peters (tim_one at email.msn.com) wrote:
: Understood the first time; but if finalizers are to be invoked on cycle
: members, that's just necessary, not sufficient.
: 
At the risk of sounding completely brain-dead ... huh?!

graham

-- 
           As you grow up and leave the playground
       where you kissed your prince and found your frog
          Remember the jester that showed you tears
                   the script for tears



From ivnowa at hvision.nl  Sat Jun 19 15:30:11 1999
From: ivnowa at hvision.nl (Hans Nowak)
Date: Sat, 19 Jun 1999 19:30:11 GMT
Subject: Newbie question
In-Reply-To: <376BBB7E.A82DA789@primenet.com>
References: <376BBB7E.A82DA789@primenet.com>
Message-ID: <199906191928.VAA20205@axil.hvision.nl>

On 19 Jun 99, John E. Garrott wrote:

> I am have a problem with the program below.
> The intent is, eventually, to get a list and display it,
> in a list box, then print the selected row
> from the listbox, but I can't get the callback
> to work.
> 
> The function dbs.databases() works, putting,
> in this case, the databases owned by myself
> into the listbox.  When I click on the listbox
> items they highlight as expected, but, apparently,
> list_cb() is never called.
> 
> I've read all the available literature on 
> binding and signals, tried numerous variations
> on what is here.  No results.  This program
> shows no errors, so is syntactically(sp?) 
> correct.  Just doesn't work.
> 
> What am I doing wrong?
> 
> Thanks in advance,
> 
> John
> --------------------------------------------------------
> import dbs
> import string
> from Tkinter import *
> 
> class SelectionBox:
> 
> 	def __init__(self, master=None):
> 		self.frame = Frame(master, name="frame")
> 		self.frame.pack(expand=1, fill=BOTH)
> 
> 		self.frame.listbox = Listbox(self.frame, name='listbox')
> 		self.frame.listbox.pack(expand=1, fill=BOTH)
> 		self.frame.listbox.bind('Button-1', self.list_cb)
> 
> 		for item in dbs.databases():
> 			self.frame.listbox.insert(END, item)
> 
> 	def list_cb(event):
> 		print "Here we are!"

I don't know for sure, but shouldn't this be

    def list_cb(self, event):

?

I don't know if that's the solution for your problem, though... It's 
not like *I* know much about Tkinter. :o)

Veel liefs,

--Hans Nowak (ivnowa at hvision.nl)
Homepage: http://fly.to/zephyrfalcon



From arcanedragonNOSPAM at NOSPAMhome.com  Sun Jun 13 10:31:18 1999
From: arcanedragonNOSPAM at NOSPAMhome.com (Chris Herborth)
Date: Sun, 13 Jun 1999 14:31:18 GMT
Subject: Module magic methods
Message-ID: <WgP83.7966$Yr2.129670@news2.rdc1.on.home.com>

Do modules have magic methods?  Specifically, I'm wondering if there's a
__del__() or __destroy__() method that I could use to clean up after myself
a little when the module is about to be unloaded...

-- 
----------================================================----------  _
Chris Herborth, R&D, Technical Writing                               | \  _
Arcane Dragon Software                 Arcane Dragon -==(UDIC)==-    | < /_\
arcane dragon at home dot com          DNRC Holder of Past Knowledge |_/ \_



From greg.ewing at compaq.com  Thu Jun 10 19:52:05 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 11 Jun 1999 11:52:05 +1200
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7j90va$6lk$1@cronkite.cc.uga.edu> <u3g1443iht.fsf@ares.sys1.ngo.okisoft.co.jp> <87pv384koy.fsf@ev.netlab.co.jp> <u3909slwbo.fsf@ares.sys1.ngo.okisoft.co.jp> <slrn7lud8t.g2c.bwinton@tor.dhs.org> <7jokdo$hih$3@cronkite.cc.uga.edu>
Message-ID: <37604FA5.BEA4A11C@compaq.com>

Graham Matthews wrote:
> 
> Actually Blake I have a feeling that I have read a paper somewhere that
> says that a mark sweep collector with refcounts can in fact patch up
> refcount errors,

Obviously this is true if the M&S can be sure of finding
all the roots, since the refcount information is then
redundant.

But if there are roots that the M&S can miss (e.g. on
the C stack or in data structures managed by an extension
module) then it's still just as important as ever to keep
the refcounts correct, since the M&S needs them to make
sure it doesn't make a mistake.

Greg



From andreas at andreas.org  Sun Jun 13 21:08:00 1999
From: andreas at andreas.org (Andreas Bogk)
Date: 14 Jun 1999 03:08:00 +0200
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <3762988E.C94AA252@iname.com> <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu> <3763C746.5CA8FE78@iname.com>
Message-ID: <m3u2sby4gf.fsf@soma.andreas.org>

Fernando Mato Mira <matomira at iname.com> writes:

> Although Dylan never really made it into "Fernando's elegibility list"
> because
> it was not a serious contender against Scheme and CL yet, it definitely got
> banished the day the Lisp syntax was dropped. I'm still waiting for the day
> when the Dylan fans will reinstate the alternative.

This is an age-old flamewar. It just so happens that there are people
who like infix syntax and detest prefix, and then there are people who
love their parens and hate infix.

It's a matter of preferences, I happen to like infix, and I'm not
gonna change.

> But my dream is a `parenthesized Cecil'

You're obviously one of THEM :-).

Andreas

-- 
Reality is two's complement. See:
ftp://ftp.netcom.com/pub/hb/hbaker/hakmem/hacks.html#item154



From ivnowa at hvision.nl  Thu Jun 10 15:38:29 1999
From: ivnowa at hvision.nl (Hans Nowak)
Date: Thu, 10 Jun 1999 19:38:29 GMT
Subject: Date formats
In-Reply-To: <375FEA53.95E3D3BC@ausinfo.com>
References: <375FEA53.95E3D3BC@ausinfo.com>
Message-ID: <199906101937.VAA05181@axil.hvision.nl>


On 10 Jun 99, mike milliger wrote:

> I read the documentation in the Python Reference Library (6.3) on time
> access and conversions and chapter 8 in programming python, yet I can't
> figure out if i need ( or how to use) a tuple to format the return from
> localtime().  i get a string now (wed jun 10 1999 xx:xx:xx) but i need
> to format it to yyyymmdd_hhmmss.  Is there any switch ( %) or function
> that can be useful?  Any help is appreciated.

You can roll your own:

#

import time

t = time.localtime(time.time())
# prints the tuple
print t

# prints a formatted tuple
print "%d%02d%02d_%02d%02d%02d" % t[0:6]

Disclaimer: This is untested, maybe the format will appear weird... 
but you'll get the idea.

Veel liefs,

--Hans Nowak (ivnowa at hvision.nl)
Homepage: http://fly.to/zephyrfalcon



From matomira at iname.com  Sat Jun 12 13:27:42 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Sat, 12 Jun 1999 19:27:42 +0200
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com>
Message-ID: <3762988E.C94AA252@iname.com>

Fernando Mato Mira wrote:

> [For example, I can think the Eiffel syntax is `philosophically' broken, but
> as `infix' syntaxes go, it is not. And it's a good language (I programmed

Actually, it's a _great_ language to learn SE (for OO pretty much too,  but of
course only in the restricted
single-dispatch way most people know about. I like to call that "Subject Oriented").

Smalltalk taught me OO.
Occam taught me about concurrency.
Eiffel taught me SE.
CLOS taught me what OO really means.

--

"Object Oriented is to not think only about your `subject' and forget about your
`arguments' - the syntactic objects of your sentence"

Fernando D. Mato Mira
Real-Time SW Eng & Networking
Advanced Systems Engineering Division
CSEM
Jaquet-Droz 1                   email: matomira AT acm DOT org
CH-2007 Neuchatel                 tel:       +41 (32) 720-5157
Switzerland                       FAX:       +41 (32) 720-5720

www.csem.ch      www.vrai.com     ligwww.epfl.ch/matomira.html







From gmcm at hypernet.com  Thu Jun 17 12:56:25 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 17 Jun 1999 11:56:25 -0500
Subject: Why don't my global variables work?
In-Reply-To: <ttbd7yvho9h.fsf@talon.i-did-not-set--mail-host-address--so-shoot-me>
Message-ID: <1282483655-114305528@hypernet.com>

Nathan J Froyd writes:

> Can somebody explain this point to me? I have the following program

> Could somebody look at this, perhaps running the program and explain
> to me what fundamental fact I am missing here? Because it looks like
> my program should run without errors. Thanks!
/snip/
> # I HAVE DEFINED `count' HERE, PYTHON!
> count = 0
/snip/
> def topic_read(regmatch, evilcount):
>   # if you comment this line, the program will not work
>   count = evilcount
>   print "\t" * count, "<topic name=\"%s\">" % regmatch.group(1)
>   count = count + 1

Alas, the FAQ is down, so I can't point you there.

Changing the binding of a global var requires a "global" statment.

def topic_read(...):
   global count
   count = count + 1

Without that, Python (at the time the "def" is scanned) decides you 
are creating a local var named count. Then you go and reference said 
local var before you've given it a value. Boom. NameError. 
Non-intuitive, perhaps, but think of the strange bugs you'd have if 
locals didn't shadow globals of the same name.

- Gordon



From john at mediamanager.com.sg  Mon Jun 28 07:37:50 1999
From: john at mediamanager.com.sg (john)
Date: Mon, 28 Jun 1999 19:37:50 +0800
Subject: Extracting Text file contents using Python
Message-ID: <37782f9a.0@news.smartnet.com.sg>

Hi,

I am trying to extract the following text file using python and the program
& text file are as shown below.

But unfortunately I am unable to get every separate data displayed except
for the first data and after the separation line which is
"====================" the next data is not getting picked up by the
program.
Could anyone kindly let me know what could be wrong with the program.

Example Text File - But will be similar
----------------------------------------------------

Title:Meet Me
Http Host:www.www.www
Remote Address:10.2.0.1
Remote Host:10.2.0.1
Rating:8
Date:06/25/99
Time:17.15.28
=================================
Title:Meet Me
Http Host:www.www.www
Remote Address:10.2.0.1
Remote Host:10.2.0.1
Rating:8
Date:06/25/99
Time:17.15.28
=================================
Title:Meet Me
Http Host:www.www.www
Remote Address:10.2.0.1
Remote Host:10.2.0.1
Rating:8
Date:06/25/99
Time:17.15.28
=================================
Title:Meet Me
Http Host:www.www.www
Remote Address:10.2.0.1
Remote Host:10.2.0.1
Rating:8
Date:06/25/99
Time:17.15.28
=================================
Title:Meet Me
Http Host:www.www.www
Remote Address:10.2.0.1
Remote Host:10.2.0.1
Rating:8
Date:06/25/99
Time:17.15.28
=================================

The python program is as follows:

#!c:/progra~1/python

import string
import sys
import cgi
#import urllib

data_file = open('c:/xitami/cgi-bin/textfiles/FilmFestReview.txt', 'r')
form = cgi.FieldStorage()
header = \
"""
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN">
<html>
<head>
<title>Extract & Write</title>
</head>
<body>
"""

footer = \
"""
</body>
</html>
"""

out = sys.stdout

begin_table = 1
out.write(header)
for line in data_file.readlines():
        if begin_table:
                out.write('<table border = "1">')
                begin_table = 0
        if line[:5] == "=====":
                out.write('</table>')
                out.write('<br>')
                begin_table = 1
                continue
        field_name, field_value = string.split(line, ':')
        out.write('\t<tr>\n')
        out.write('\t\t<td>\n')
        out.write('\t\t\t%s\n' % field_name)
        out.write('\t\t</td>\n')
        out.write('\t\t<td>\n')
        out.write('\t\t\t%s\n' % field_value)
        out.write('\t\t</td>\n')
        out.write('\t</tr>\n')



out.write(footer)






From suzuki611 at okisoft.co.jp  Mon Jun 14 23:04:43 1999
From: suzuki611 at okisoft.co.jp (Hisao Suzuki)
Date: 15 Jun 1999 12:04:43 +0900
Subject: fork()
References: <000c01beb3c1$a3f428a0$329e2299@tim>
	<199906110443.AAA02408@eric.cnri.reston.va.us>
	<Pine.SUN.3.95-heb-2.07.990611075346.4942G-100000@sunset.ma.huji.ac.il>
	<u3d7yzn1np.fsf@ares.sys1.ngo.okisoft.co.jp>
	<7k35bf$13iq$1@nntp6.u.washington.edu>
Message-ID: <u3674q88qc.fsf@ares.sys1.ngo.okisoft.co.jp>

In article <7k35bf$13iq$1 at nntp6.u.washington.edu>,
donn at u.washington.edu (Donn Cave) wrote:
> I want to ask more about this, partly I'm not sure I understand what
> the trees and their roots are or what registration would require.

I'm sorry for my poor explanation.  The following is a simple
example of a doubly-linked tree.

	class Node: pass
	root, node1, node2 = Node(), Node(), Node()
	root.child1, node1.parent = node1, root
	root.child2, node2.parent = node2, root

Having circular references, the nodes are never deleted even if
you `del root, node1, node2'.  The registration requires at
least one node: root, node1, or node2.  You can traverse the
entire tree from any node, but you would probably register the
root since it is the most convenient node to start the traverse.

In option (A), the registration also requires a function which
will be applied to the root (or the node you registered) by the
garbage collector. (The function may be a method of a certain
name and thus the registration may require only the root.)

In general, so-called `aggregations' in OO terms are implemented
with trees.  If the association of an aggregation is
bi-directional, the result is a doubly-linked tree.  I hope you
imagine more practical examples.

> Could something like a standard class method accomplish this?  Like,
> 
>    def __garbagecollected__(self):
>        # author knows self.parent is a circular reference.
>        self.parent = None
> 
>    def __del__(self):
>        if self.parent:
>            ...

Perhaps.  But I am afraid it may be a little difficult, in
practice, to write a correct __garbagecollected__ method which
must work in any order of invocations determined arbitrarily by
the garbage collector.  On the other hand, if you register the
root explicitly, you have a guarantee that the root will be
accessed first, thus the whole tree is guaranteed to be in a
more predictable state.

--===-----========------------- Sana esprimo naskas sanan ideon.
SUZUKI Hisao            suzuki611 at okisoft.co.jp, suzuki at acm.org.



From news at dorb.com  Tue Jun 22 20:27:24 1999
From: news at dorb.com (Darrell)
Date: Tue, 22 Jun 1999 20:27:24 -0400
Subject: Newbie: How do you "clear" PythonWin between runs?
References: <37700E61.5388E101@alliedsignal.com>
Message-ID: <cUVb3.1952$U6.6930@newsr2.twcny.rr.com>

http://grail.cnri.reston.va.us/cgi-bin/faqw.py?query=reload&querytype=simple
&casefold=yes&req=search


--
--Darrell
H. P. Friedrichs <" HPeter.Friedrichs"@alliedsignal.com> wrote in message
news:37700E61.5388E101 at alliedsignal.com...
> I've just started playing with python, so all of this is very new to me.
> I have been writing some scripts, saving them to *.py files, and running
> them under PythonWin 1.5.2 . The process is iterative; I code, save,
> run, code, save, run, etc.
>
> What I have noticed is that python doesn't "reset" between runs.
> Sometimes error  messages reappear even though I have fixed the problem,
> because some artifact of a previous run lingers.
>
> So, my question is this: How can I "flush" or "clear" the python
> interpreter between successive runs? Is there a command I can add to my
> script to initialize python before running the script?
>
> Thanks,
>
> Pete
> hpeter.friedrichs at alliedsignal.com
>
> --
>
> -----------------------------------------------
> | The opinions expressed here are my own,     |
> | and not necessarily those of my employer.   |
> -----------------------------------------------
>
>





From cwebster at math.tamu.edu  Mon Jun 28 21:06:49 1999
From: cwebster at math.tamu.edu (Corran Webster)
Date: 29 Jun 1999 01:06:49 GMT
Subject: ANN: Stackless Python 0.2
References: <000c01bec11f$8dfbf800$e19e2299@tim>
Message-ID: <7l9679$3od$1@news.tamu.edu>

In article <000c01bec11f$8dfbf800$e19e2299 at tim>,
Tim Peters <tim_one at email.msn.com> wrote:
>[Corran Webster, commenting on Christian Tismer's "stackless Python"]
>> ...
>> I wonder whether there's the potential here for more than coroutines
>> and to write an implementation of threading within Python.
>
>It's been mentioned a few times.  So far Guido isn't keen on any of this,
>but he's been known to buckle after a few short years of incessant whining
><wink>.  BTW, a coroutine pretty much *is* a thread in a time-sliced world,
>just lacking a "transfer" function that invokes implicitly whenever it
>bloody well feels like it <wink>.

Indeed - and it was all this talk of coroutines that made me wonder
why not go that extra bit.

>> Each thread would presumably need its own Python stack,
>
>Nope!  It's a beauty of the implementation already that each code object
>knows exactly how much "Python stack space" it needs, and (just) that much
>is allocated directly into the code object's runtime frame object.  IOW,
>there isn't "a Python stack" as such, so there's nothing to change here --
>the stack is implicit in the way frames link up to each other.

Ah - my impression of the way that the stack works was wrong (hadn't
gotten that far in the reading of the source).  So each code object
has a little block of stack space set aside (which is precalculated
to be enough), together with a pointer to its own top of the stack
inside that block?

>> and a queueing and locking system would need to be added somehow,
>
>Yes, and that would require some changes to the core.  Sounds doable,
>anyway.

More work than coroutines, I suspect, and I don't think it can be
done easily to use the same framework as the current C threads.

>> but because the Python and C stacks are no longer intertangled, switching
>> between threads should be easy (as opposed to impossible <wink>).
>
>I think Christian's approach moves lots of crazy ideas from impossible to
>plausible.

Indeed.  A very nice piece of code.

>> This wouldn't be quite as flexible as the current implimentations of
>> threads at the C level - C extensions would be called in a single block
>> with no way to swap threads until they return.
>
>That's mostly true today too:  Python threads run serially now, one at a
>time, and if a thread calling out to C doesn't release the global lock no
>other thread will run until it returns.

OK.  And presumably most extensions don't bother to release the lock
unless they specifically want to take advantage of threading somehow.

>> On the other hand, this would allow every platform to have some sort
>> of threading available, which would be a boon.
>
>It's even quite possible that "fake threads" would-- for programs that
>aren't doing true multiprocessing --run significantly more efficiently than
>today's scheme of creating OS-level threads and then choking them into
>taking strict turns; e.g., because "the Python stack" grows only to the
>exact size it needs, there's almost certainly much less memory overhead that
>way than by letting the OS allocate a mostly unused Mb (whatever) to each
>real thread's stack.  This opens the possibility to create thousands &
>thousands of fake threads.

Although any machine which is being used that intensively probably already
has some sort of C level threading available.

>> Unfortunately I'm not familiar enough with threads to go out there and
>> implement it right away, but I thought I'd at least raise it as a
>> possibility and see what people think and what the pros and cons are.
>
>It's sure worth pondering!
>
>write-some-code-anyway-&-get-famous-one-way-or-another<wink>-ly y'rs  - tim

Well, I won't be writing it this week<wink> but I have a better idea of the
ins and outs now.

Thanks.

Corran




From billtut at microsoft.com  Thu Jun 24 22:27:18 1999
From: billtut at microsoft.com (Bill Tutt)
Date: Fri, 25 Jun 1999 02:27:18 GMT
Subject: pkzip and using popen on Windows 98 ???
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8100DB90E3F@RED-MSG-50>

There was a bug in Win95 that prevents win32pipe from working. KB: Q150956
The same problem may not have been fixed in Windows 98. I don't have a local
Win98 box to test it on.

The current URL for this information is:
http://support.microsoft.com/support/kb/articles/q150/9/56.asp

I also just added this using Guido's cool FAQ wizard..
Maybe I won't have to keep digging the KB article up anymore. :)

Bill



From nathan at islanddata.com  Tue Jun 22 11:03:35 1999
From: nathan at islanddata.com (Nathan Clegg)
Date: Tue, 22 Jun 1999 08:03:35 -0700 (PDT)
Subject: restoring signal handlers
Message-ID: <XFMail.990622080335.nathan@islanddata.com>

Is it necessary to restore original signal handlers before quitting a
program?  I can see saving them if I want to restore them at some middle
point of execution, but if I want my own handlers in place for the entire
life of my app, do I need to bother saving and restoring the originals?


----------------------------------
Nathan Clegg
 nathan at islanddata.com





From john.michelsen at gte.net  Wed Jun  2 23:52:27 1999
From: john.michelsen at gte.net (John Michelsen)
Date: Thu, 03 Jun 1999 03:52:27 GMT
Subject: More Tkinter menu problems
References: <3755C714.526B7994@pixar.com>
Message-ID: <%Zm53.899$S15.46310@dfiatx1-snr1.gtei.net>

>  self.menu.add_command(label = new_name,
>   command = lambda temp_color = new_color:
>   self.select_color(temp_color))
>
>but I get a Name Error on "self".  So now "self" is out of scope when
>the menu call happens.


try:
self.menu.add_command(label = new_name,
    command = lambda c=new_color, f=self.select_color: f(c))

John





From nessus at mit.edu  Fri Jun 11 14:27:10 1999
From: nessus at mit.edu (Douglas Alan)
Date: 11 Jun 1999 14:27:10 -0400
Subject: ^C and os.system()
Message-ID: <lczp2661a9.fsf@twitch.mit.edu>

Greetings.  Is it considered a bug or a feature that if I type ^C at a
Python program that is in the middle of executing a call to
os.system(), the Python program never seems to see the SIGINT signal?
Instead, only the process created by os.system() receives the signal.
Consequently, if I have a Python script that executes a program
repeatedly for a list of files, for instance, I can't kill the Python
script with ^C.

Personally, I find this kind of annoying.  This is using Python 1.5.1
under Solaris 2.6.  Please forgive me if this a bug that has already
been fixed under 1.5.2.

Also, when I have the Python program check the return status of a
command invoked via os.system() and subsequently killed with ^C,
instead of getting back 0x0002, as would be expected from a process
killed by SIGINT, I usually get back 0x8200.  Sometimes I get back
0xd000.  Only infrequently do I get back the 0x0002 that I would
expect.

None of these problems occur if I use commands.getstatusoutput()
instead of os.system(), so I guess I will stick with getstatusoutput()
for the time being.

|>oug
   http://space.mit.edu/~nessus



From amp26494 at ammex001.amp.com  Thu Jun  3 11:01:17 1999
From: amp26494 at ammex001.amp.com (Solenberger, Keith)
Date: Thu, 3 Jun 1999 11:01:17 -0400 
Subject: PythonWin wish list
References: <7iv11t$nb0$1@nnrp1.deja.com>
Message-ID: <56AEBCBEB11CD211BC9800805FA766FDA06FDD@ammex006.amp.com>


> -----Original Message-----
> From:	cmedcoff at my-deja.com [SMTP:cmedcoff at my-deja.com]
> Posted At:	Monday, May 31, 1999 6:05 PM
> Posted To:	python
> Conversation:	PythonWin wish list
> Subject:	PythonWin wish list
> 
> One of the nice things about python is the free development
> environment
> PythonWin.  I certainly don't get my other development environments
> for
> free.
> 
> Having used PyhonWin a couple of weeks now I have a couple of comments
> in the form of a wish list.  Comments are welcome.
> 
> 1)It would be really nice of the interactive window was dockable.  I
> find myself constanly searching for that window.  I like to have my
> edit windows maximized. An example of this in another application
> would
> be Microsoft Visual Studio's output window where once can see error
> messages from compiles.  It is never covered up even if my edit
> buffers/windows are maximized and I can hide it if I want even more
> space.
> 
> 2)PythonWin conforms to the Windows MDI interface except for the
> floating window that comes up during debugging.  It would be nice if
> that window could be incorported into the MDI environment.  This would
> be another candidate for a dockable window.
> 
> 3)One other thing that would make the PyhonWin environment even nicer
> is hot-keys or accelerators.  Its always nice to have the mouse when
> you are learning an application but hot-keys are always faster once
> you've got them memorized.  Good candidates here are keys for stepping
> in, out, over during debugging and access to the interactive window.
> 
> Regards,
> Chuck
> 
> 
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.




From gmcm at hypernet.com  Mon Jun  7 19:52:23 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 7 Jun 1999 18:52:23 -0500
Subject: Tkinter problem in Python 1.5.2 on Solaris 2.5.1
In-Reply-To: <928792539.297.78@news.remarQ.com>
Message-ID: <1283322699-63834024@hypernet.com>

Gordon Vinther wrote:

> I seem to be having the same problem.  Where and how do I find these
> patches??

> > Gordon McMillan wrote:
> > >
> > > Jan Zegers wrote:
> > > >
> > > > I have the following problem with Tkinter in Python 1.5.2.
> > > /snip/
> > > >   This probably means that Tcl wasn't installed properly.
> > > >
> > > > Note the whitespace between the two last test lines. The
> > > > installation is as follows:
> > > >   Tcl/Tk 8.1 with --enable-shared so we have libraries
> > > >     /usr/local/lib/libtcl8.1.so and /usr/local/lib/libtk8.1.so
> > >
> > > Tcl / Tk 8.0 is still the official version for Python.
> > >
> > > On Jun 1, Dieter Maurer posted patches to use 8.1 ("Tkinter-patch for
> > > use with Tk8.1...").
> > >
> > > - Gordon
> > Thank you. I applied the patches and it worked!
> > Kind regards,
> > Jan

You go to www.python.org.

You click the "Search" link.

You use "Search the Python Newsgroups" and type in the article title 
I quoted.

You click the link for the article written by Dieter Maurer.

Now you never need ask this question again.

but-give-a-man-a-fishing-pole-and
 -he'll-never-get-any-work-done-again-ly y'rs

- Gordon



From chapman at bioreason.com  Wed Jun 16 10:10:55 1999
From: chapman at bioreason.com (Mitch Chapman)
Date: Wed, 16 Jun 1999 08:10:55 -0600
Subject: pygtk - clist title button
References: <376755A5.15621A4@onera.fr>
Message-ID: <3767B06F.70E70EDB@bioreason.com>

Marc POINOT wrote:
> 
> Hmmm... a question about pygtk, but I guess it's more
> gtk than python :}
> I make this public instead of sending it to James Henstridge
> as a private email (or to gtk list).
> 
> How to connect a signal to a clist title button?

Here's what works for me:

            # Get the title button for column i
            columnHeader = clist.get_column_widget(i)
            button = columnHeader["parent"]

The trick is that gtkclist_get_column_widget(), instead of 
giving you the button widget, gives you the GtkAlignment
widget contained by the button.  Dunno why.

-- 
Mitch Chapman
chapman at bioreason.com



From ivanlan at callware.com  Mon Jun 14 10:11:21 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Mon, 14 Jun 1999 14:11:21 GMT
Subject: Variations on a theme by "pound bang"
References: <3763e75f.3265255@news.iol.ie>
		<3763F77E.6094AEE9@appliedbiometrics.com> <3764185B.7BCBC488@Lugoj.Com>
Message-ID: <37650D89.52D687F8@callware.com>

Hi All--

James Logajan wrote:
> 
> Christian Tismer wrote:
> >
> > Sean Mc Grath wrote:
> > >
> > > Can anyone shed some light on the benefits of:-
> > >
> > > # !/usr/bin/env python
> >
> > Uses the env program which is allways there, to figure out
> > from the environment settings which python to use.
> 
> Alas, this is a myth. Not all Unixen put env in /usr/bin, some put it in
> /bin. I can't recall if it was Solaris (probably not) or LynxOS (probably)
> that put env in /bin. Anyway, I got bit by it.

UnixWare doesn't put it in /usr/bin either; since I have successfully
TRASHED my UnixWare system, AT LAST!, in favor of RedHat 6.0 SMP, I
can't tell you where it was.  Someplace bizarre, I think.

</usr/sbin/etc/man/bin>-ly y'rs,
Ivan 
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From Mikael.Lyngvig at p98.f112.n480.z2.fidonet.org  Wed Jun 30 19:51:16 1999
From: Mikael.Lyngvig at p98.f112.n480.z2.fidonet.org (Mikael Lyngvig)
Date: Thu, 01 Jul 1999 00:51:16 +0100
Subject: Python GUIs: Summary and Conclusion
Message-ID: <000015e4@bossar.com.pl>

From: mikael at pobox.com (Mikael Lyngvig)

On Mon, 28 Jun 1999 18:09:48 GMT, "Michael P. Reilly"
<arcege at shore.net> wrote:

>I've already rewritten _tkinter to use a new pytcl module, but the
>functionality is still the same, ie. executing Tk calls from within an
>interpreter.  But I'd like to see it using the TK/C API (so much for
>my hard word on rewriting _tkinter *wink*).
>
>The main issue would be to make it extensible for adding the new
>widgets and other extensions (BTL, PIL, etc.).  Also, the interface
>between Python and C would not be as "simple" as the existing module.
>It's doable tho.
>
>Note that using the "native" Tk API would still require linking with
>the Tcl library.
>
>I could scratch my _tkinter rewrite and work on a native solution.
>(I could still publish my pytcl module.)  Any thoughts?

I've downloaded the Tcl 8.1.1 and Tk 8.1.1 sources, rebuilt them, and
studied them for a bit.

I think we're looking at a major task.  As far as I can determine from
the Tk sources, the parser is an integral part of the Tk library.
E.g., the Tk sources appear to be high-level down to the lowest level.

Even the lowest level functions takes argv[] style parameters and,
aassisted by a horde of strncmp()s (350+ in the generic directory),
parse the parameter/option list into a command, which is executed by
code which is interleaved with the parsing code.

I incorrectly assumed the design would be something like:

	[Tcl parser layer]
	[Tk parser layer]
	[Tk implementation layer]

But it appears to be:

	[Tcl parser layer]
	[Tk parser and implementation layer]

My hope was to make a Python interface run on top of the
implementation layer, but that appears to be impossible with the
current Tk design.

So, in order to make a native Tk layer for Python, we'd have to create
the implementation layer ourselves, and convince the Tk maintainers to
adopt this (if we'd want their future to be part of our TkInter
automatically).

I can illustrate what I mean by looking at the Tk bell command's
implementation:

int
Tk_BellObjCmd(clientData, interp, objc, objv)
    ClientData clientData;      /* Main window associated with
interpreter. */
    Tcl_Interp *interp;         /* Current interpreter. */
    int objc;                   /* Number of arguments. */
    Tcl_Obj *CONST objv[];      /* Argument objects. */
{
    static char *bellOptions[] = {"-displayof", (char *) NULL};
    Tk_Window tkwin = (Tk_Window) clientData;
    char *displayName;
    int index;

    if ((objc != 1) && (objc != 3)) {
        Tcl_WrongNumArgs(interp, 1, objv, "?-displayof window?");
        return TCL_ERROR;
    }

    if (objc == 3) {
        if (Tcl_GetIndexFromObj(interp, objv[1], bellOptions,
"option", 0,
                &index) != TCL_OK) {
            return TCL_ERROR;
        }
        displayName = Tcl_GetStringFromObj(objv[2], (int *) NULL);
        
        tkwin = Tk_NameToWindow(interp, displayName, tkwin);
        if (tkwin == NULL) {
            return TCL_ERROR;
        }
    }
    XBell(Tk_Display(tkwin), 0);
    XForceScreenSaver(Tk_Display(tkwin), ScreenSaverReset);
    XFlush(Tk_Display(tkwin));
    return TCL_OK;
}

The way I see it, this should have been something like this:

int Tki_WindowBell(Window)
   Tk_Window Window;
{
    XBell(Tk_Display(Window), 0);
    XForceScreenSaver(Tk_Display(Window), ScreenSaverReset);
    XFlush(Tk_Display(Window));
    return TCL_OK;
}

int
Tk_BellObjCmd(clientData, interp, objc, objv)
    ClientData clientData;      /* Main window associated with
interpreter. */
    Tcl_Interp *interp;         /* Current interpreter. */
    int objc;                   /* Number of arguments. */
    Tcl_Obj *CONST objv[];      /* Argument objects. */
{
    static char *bellOptions[] = {"-displayof", (char *) NULL};
    Tk_Window tkwin = (Tk_Window) clientData;
    char *displayName;
    int index;

    if ((objc != 1) && (objc != 3)) {
        Tcl_WrongNumArgs(interp, 1, objv, "?-displayof window?");
        return TCL_ERROR;
    }

    if (objc == 3) {
        if (Tcl_GetIndexFromObj(interp, objv[1], bellOptions,
"option", 0,
                &index) != TCL_OK) {
            return TCL_ERROR;
        }
        displayName = Tcl_GetStringFromObj(objv[2], (int *) NULL);
        
        tkwin = Tk_NameToWindow(interp, displayName, tkwin);
        if (tkwin == NULL) {
            return TCL_ERROR;
        }
    }

    Tki_WindowBell(tkwin)
    return TCL_OK;
}

That way we could easily add a very fast Python layer on top of the
Tki_Xxx() functions.

So, in order to accomplish the task we'd have to restructure the
entire Tk source distribution - basically modify all functions in all
source files.

>We already know that Python 1.6 won't be until the new year, so any
>changes to Tkinter would be as a patch.

>From what I know now, I suspect that Python 3.0 could be done by the
time we're through :)

Also, if we did separate the parser and the implementation, we'd end
up with something which is directly incompatible with add-on widgets
and so on.  Although we could probably make a TkParser module, which
could interface to the parser layer.

My view, given the latest info, is that the task of making a native
TkInter is so huge that it doesn't pay off, unless your goal is to use
Tk in some form (not to get a fast, portable GUI).  The quickest way
to make TkInter faster is to wait a couple of years till your (then)
computer has quadrubled in speed.

What are your thoughts?  

P.S. Can anyone tell me how to ring a bell using the native Tk API?  I
don't see any way, except to do what Tk_BellObjCmd() does yourself.



-- Mikael



From alex at somewhere.round.here  Sun Jun  6 17:22:43 1999
From: alex at somewhere.round.here (Alex)
Date: 06 Jun 1999 17:22:43 -0400
Subject: Suggestion: Dutch Python mailinglist
References: <000101beb058$c992bc00$e39e2299@tim>
Message-ID: <etdogitvxb0.fsf@m2-225-11.mit.edu>

this_notion = 'one-day-the-whole-world-will-speak-python'
scary = 0

this_notion is scary




From MHammond at skippinet.com.au  Thu Jun 17 17:57:31 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Fri, 18 Jun 1999 07:57:31 +1000
Subject: do anonymous pipes normally work on NT?
References: <4D0A23B3F74DD111ACCD00805F31D8100DB90DE4@RED-MSG-50>
Message-ID: <7kbqt5$bep$1@m2.c2.telstra-mm.net.au>

Bill Tutt wrote in message
<4D0A23B3F74DD111ACCD00805F31D8100DB90DE4 at RED-MSG-50>...
>

>Alter the PyHANDLE_Check() branch of win32processmodule_win32.cpp:sethandle
>to:
> *pobHandle = v;
> if (PyWinObject_AsHANDLE(v, ph))
> {
> Py_INCREF(v);
> }
> else
> {
> rc = -1;
> }

Ahh - excellent analysis.  I made the mistake of attempting to use some
fancy C++ operators, and, like many C++ features, turned around and bit me
in the bum.  I will also seek out and destroy any similar casts.

Good stuff!

Mark.






From H..P..Friedrichs at p98.f112.n480.z2.fidonet.org  Wed Jun 30 22:30:31 1999
From: H..P..Friedrichs at p98.f112.n480.z2.fidonet.org (H. P. Friedrichs)
Date: Thu, 01 Jul 1999 03:30:31 +0100
Subject: Help: Can you explain this error? (contains code snip)
Message-ID: <00001628@bossar.com.pl>

From: "H. P. Friedrichs" <" HPeter.Friedrichs"@alliedsignal.com>

Thanks!

Pete

Jeff Epler wrote:

> On Wed, 30 Jun 1999 16:49:58 -0700, H. P. Friedrichs
>  <@alliedsignal.com> wrote:
> >Hello:
> >
> >This is week 2 for me in python...What fun!
> >
> >I have stumbled across an interesting problem. When the code below is
> >run under Win95 (python 1.5.2),it works without problems. If the same
> >code is run under Linux (python 1.4) I get the error:
>
> Make things simpler:
>
> >>> int("123")
>
> this works in Python 1.5.x and fails in 1.4.  I believe this feature is
> present in all 1.5.x and no standard 1.4 versions.
>
> Use
>
> >>> string.atoi("123")
> to work on both versions.
>
> Jeff

--

-----------------------------------------------
| The opinions expressed here are my own,     |
| and not necessarily those of my employer.   |
-----------------------------------------------




From sdm7g at Virginia.EDU  Fri Jun 25 20:02:01 1999
From: sdm7g at Virginia.EDU (Steven D. Majewski)
Date: Fri, 25 Jun 1999 20:02:01 -0400 (EDT)
Subject: How to obtain an instance's name at runtime?
In-Reply-To: <7kvd8b$ueb$1@nnrp1.deja.com>
References: <7kqpu8$7u1$1@nnrp1.deja.com>
	<3771127C.E40F9BF3@home.com> <7kvd8b$ueb$1@nnrp1.deja.com>
Message-ID: <Pine.A32.3.90.990625185053.18511B-100000@elvis.med.Virginia.EDU>

On Fri, 25 Jun 1999, Dinu C. Gherman wrote:

> In article <3771127C.E40F9BF3 at home.com>,
>   Jim Meier <fatjim at home.com> wrote:
> >
> > Well, there sort of is a solution, but I don't think you're going
> > to like it. Incidentally, why are you trying to discover this? I'm
> > not a super-meister-coder, but I can't imagine a situation where
> > knowing this would be useful.
> 
> Ok, I knew I was going to look like a dumb camel right after
> posting my question... Nevertheless there is a useful case of
> knowing such an instance's name, at least I believe I found
> one. Have a look at the following page, spot the question mark
> and you'll know what I mean:
> 
>   http://starship.python.net/crew/gherman/UMLgenTest.pdf
> 
> But, PLEASE, don't start to ask me anything about the project
> itself!! It's all in a very pre-pre-pre-alpha phase...
> 
> There might be some workaround solution in oder to get that
> name I want, as in some of the replied in this thread, but
> I'll evaluate that later...


It's easy to get *A* name -- 

 Classes and functions have cannonical names in Python because Python
distinguishes between definition and assignment for classes and
functions. They also have a cannonical namespace -- for classes,
their __module__ attribute, for Python functions, their func_globals
attribute (which is the __dict__ of the __module__ they were defined
in.)  That's why their names are "sticky" . 

 Classes and functions can be aliased and assigned to other names,
but for debugging, tracebacks and printing, you want that cannonical
defining name. 

 There are other types of objects that could benefit from "sticky" 
names: constants and enumns for example. But Python doesn't 
have a different syntax to indicate a definition separate from 
any other assignment. 


You can get a name by searching for the thing in all of the namespaces:

>>> def findname(thing):
... 	for mod in sys.modules.values():
... 		for name,val in mod.__dict__.items():
... 			if val is thing : return name 

>>> findname( sys.stdin )
'stdin'
>>> findname( findname )
'findname'

>>> findname( lambda x: x+1 )	# anonymous function doesn't have a name 
				# returns None 

>>> xxx = sys.stdin		# alias sys.stdin to another name
>>> findname( sys.stdin )	# finds *a* name (whichever is first)
'xxx'


# Try the __name__ attribute first before searching... 

>>> def nameof( THING ):
... 	return getattr( THING, '__name__', None ) or findname( THING )
... 


>>> nameof( lambda x: x+1 )
'<lambda>'
>>> mod = sys
>>> nameof( mod )
'sys'
 

[1] You could do a smarter seach of the name space, but I'm not sure
if that would be any better for your application.

[2] You could also append the module name to the name and return 
  "sys.stdin" rather than "stdin" above. 

[3] Might there be cases where you want the '__repr__' rather than
 the name ?  i.e. "<open file '<stdin>', mode 'r' at 5a28500>" rather
 than "stdin" 



---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---

 "IA-64 looks just about like what you would expect from a PA-RISC
  and IA-32 train wreck with a little VLIW thrown in for spice."
 * Thomas J. Merritt  <tjm at spam.codegen.com> in <news:comp.arch> *




From moshez at math.huji.ac.il  Thu Jun 10 01:47:42 1999
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Thu, 10 Jun 1999 08:47:42 +0300
Subject: Python 2.0
In-Reply-To: <slrn7ludnh.g2c.bwinton@tor.dhs.org>
References: <000801beb2e8$05e98760$af9e2299@tim> <slrn7ludnh.g2c.bwinton@tor.dhs.org>
Message-ID: <Pine.SUN.3.95-heb-2.07.990610084653.17920B-100000@sunset.ma.huji.ac.il>

On Thu, 10 Jun 1999, Blake Winton wrote:

> >Solve the problem by ignoring it?  Guido would find that more
> >attractive than I would <wink>.
> 
> If so, then he would have given up on this thread a long time ago...

Why? He seems to ignore it just fine!

quoting-out-of-context-ly y'rs, Z.
--
Moshe Zadka <mzadka at geocities.com>. 
QOTD: My own exit is more likely to be horizontal then perpendicular.





From adjih at technologist.com  Wed Jun  9 19:55:14 1999
From: adjih at technologist.com (Cedric Adjih)
Date: Thu, 10 Jun 1999 01:55:14 +0200
Subject: Python CPU
References: <14171.17113.39933.904980@lrz.de>
Message-ID: <375EFEE2.4910D99D@technologist.com>

Eugene Leitl wrote:

> Have there been any attempts in creating an ASIC capable of
> interpreting Python opcodes directly? Of course, with
> soon-to-arrive Java chips and JPython there might be not
> that much need for it...

  Well Python opcode is so complex that you will need microcode,
and many microcode instructions for each Python bytecode.

  Fortunatly the computer industry has anticipated the fact that
Python will become the Next Big Language, so they made some
processors so as:
- microcode is downloadable.
- microcode is pipelined (a must for speed).

  Actually, they already made a big architectural optimisation:
they added some microcode cache, so that you wouldn't be
limited by some total microcode memory limit. Instead you
continuously download the microcode for the bytecode ;
except when it is already in the cache.
  Because of program locality, most of the microcode will
end-up in the cache ; so the penalty will be small compared
to downloading the microcode only once ; while the flexibility
will be much higher.
  The good thing is that compilers generating directly
microcode have been released, and Guido has already
managed  to compile Python on one this processors.
  These Python-aware processors are also known as
"RISC microprocessors".

-- Cedric





From dkirby at bigfoot.com  Mon Jun 14 11:53:03 1999
From: dkirby at bigfoot.com (Dave Kirby)
Date: Mon, 14 Jun 1999 15:53:03 GMT
Subject: [pythonwin] popup menu problems   (was: [Pythonwin] Help needed creating a Window)
Message-ID: <37651bde.25372123@NNRP.UK.INSNET.NET>

Thanks to Mark Hammond's help I now have an app that displays an icon
in the system tray which I can process messages from. However when I
try to create a popup menu menu for it nothing happens. The call to
PyCMenu.TrackPopupMenu doesnt display anything and never returns.  I
have tried both using it to send a notify message to the window and
returning a command, but it didnt make any difference. I am unsure
whether it is a failure of my understanding of the way Windows does
these things, a bug in my code, or a problem with pythonwin. 


I also have the minor problem that when I destroy the icon it doesnt
disappear until I move the mouse pointer over it. Does anyone know how
to force the system tray to redraw?

Any help would be greatly appreciated.

   Dave K


Here is the relevant code:

#--------------- 8< --------------------------------------
class Icon:
    #constructor
    # param: dll - handle to the dll containing the icon resources
    # param: startIcon - the initial icon resource ID to display
    # param: tooltip = the string to display when the cursor is over
    # the icon
    def __init__( self, dll, startIcon, tooltip ):
        self.iconMap = {}
        self.msgID = win32con.WM_USER+20
        launchID = win32con.WM_USER+21
        quitID = win32con.WM_USER+22
        message_map = {
           win32con.WM_DESTROY: self.OnDestroy,
           self.msgID : self.OnTaskbarNotify,
           win32con.WM_CONTEXTMENU : self.OnContextMenu,
        }
        #the menu_map contains the information on the popup menu
        #entries -
        # message ID, callback function, text string
        menu_map = (
            ( launchID, self.OnLaunch, "Launch Browser" ),
            ( quitID, self.OnExit, "Exit program" )
        )
        
        self.dll = dll
        
        #add menu commands to the message_map dictionary
        for entry in menu_map:
            message_map[entry[0]] = entry[1]
            
        #define a window class for the icon handler
        wc = win32gui.WNDCLASS()
        hinst = wc.hInstance = win32gui.GetModuleHandle(None)
        wc.lpszClassName = "BaxterTaskbar"
        wc.style = win32con.CS_VREDRAW | win32con.CS_HREDRAW
        wc.hCursor = win32gui.LoadCursor( 0, win32con.IDC_ARROW )
        wc.hbrBackground = win32con.COLOR_WINDOW
        wc.lpfnWndProc = message_map
        classAtom = win32gui.RegisterClass(wc)
        
        #create the window
        style = win32con.WS_OVERLAPPED | win32con.WS_SYSMENU
        self.hwnd = win32gui.CreateWindow( classAtom, "title", style,
0,0, win32con.CW_USEDEFAULT, win32con.CW_USEDEFAULT, 0, 0, hinst,
None)
        win32gui.UpdateWindow(self.hwnd)
        self.cWnd = win32ui.CreateWindowFromHandle( self.hwnd )
        
        #initialise the icon display
        icon = win32gui.LoadIcon( 0, win32con.IDI_APPLICATION )
        flags =   win32gui.NIF_ICON \
		| win32gui.NIF_MESSAGE \
                | win32gui.NIF_TIP
        nid = (self.hwnd, 0, flags, self.msgID, icon, tooltip)
        win32gui.Shell_NotifyIcon(win32gui.NIM_ADD, nid)
        self.changeIcon( startIcon )

        #initialise the popup menu
        self.cMenu = win32ui.CreateMenu()
        for entry in menu_map:
            self.cMenu.AppendMenu( win32con.MF_STRING, entry[0],
entry[2] )

    ####################################
    # this is the bit that actually calls the popup menu
    ####################################
    def OnTaskbarNotify(self, hwnd, msg, wparam, lparam):
        if lparam==win32con.WM_LBUTTONDBLCLK:
            print "double clicked"
        elif lparam==win32con.WM_RBUTTONUP:
            pos = win32api.GetCursorPos()
            print "opening menu at", pos
            self.cMenu.TrackPopupMenu( pos, owner=self.cWnd )

            #alternative attempt returning a command:
            #cmd = self.cMenu.TrackPopupMenu( pos,
            #       win32con.TPM_LEFTALIGN
            #       |win32.TPM_LEFTBUTTON
            #       |win32con.TPM_NONOTIFY
            #       |win32con.TPM_RETURNCMD, None )

            print "finished popup menu"
        return 1

    #the following methods are included for completeness, but are not
    # part of the problem
        
    #change to a new icon - if hasnt been loaded then do so, otherwise

    #use cached icon
    def changeIcon( self, iconResource ):
        if self.iconMap.has_key( iconResource ):
            icon = self.iconMap[iconResource]
        else:
            icon = win32gui.LoadIcon(self.dll, iconResource)
            self.iconMap[iconResource] = icon
            
        flags = win32gui.NIF_ICON
        nid = (self.hwnd, 0, flags, 0, icon)
        win32gui.Shell_NotifyIcon(win32gui.NIM_MODIFY, nid)
        
    def changeTooltip( self, tip ):
        flags = win32gui.NIF_TIP
        nid = (self.hwnd, 0, flags, 0, 0, tip)
        win32gui.Shell_NotifyIcon(win32gui.NIM_MODIFY, nid)
        
    
    #destructor - remove icon
    def __del__( self ):
        nid = (self.hwnd, 0, win32gui.NIF_MESSAGE, self.msgID )
        win32gui.Shell_NotifyIcon(NIM_DELETE, nid)
        win32gui.UpdateWindow(self.hwnd)

    def OnDestroy(self, hwnd, msg, wparam, lparam):
        nid = (self.hwnd, 0, win32gui.NIF_MESSAGE, self.msgID )
        win32gui.Shell_NotifyIcon(NIM_DELETE, nid)
        win32gui.UpdateWindow(self.hwnd)

    def OnContextMenu(self, *args):
        print "Icon.OnContextMenu", args

    def OnLaunch(self,  *args ):
        print "Icon.OnLaunch", args

    def OnExit(self, *args ):
        print "Icon.OnExit", args




From annis at biostat.wisc.edu  Wed Jun  9 12:14:07 1999
From: annis at biostat.wisc.edu (William Annis)
Date: 09 Jun 1999 11:14:07 -0500
Subject: [ANN] PyKstat 0.01 - Solaris kstat(3k) access
Message-ID: <jtk8td8i7k.fsf@wazor.biostat.wisc.edu>

        I have finally decided to release PyKstat, a Python interface
to the Solaris kstat(3k) API.  The package can be found at

        http://www.biostat.wisc.edu/~annis/creations/pykstat.html

        Here is the README file which comes with the package...

This is the initial release of a Python interface to the Solaris kstat
API.  I recommend checking out the man pages for information on the C
API.  This will only work for Solaris, and has been tested on Solaris
2.5, 2.5.1, and 2.6 for SPARC Solaris, and 2.6 for the Intel version
of Solaris, using Python 1.5.1 and 1.5.2.


* Why did I do this?

    I am currently re-implementing a machine monitoring system.  The
original version, written in Perl, has to fork off and parse the
output from various system data tools, such as iostat, vmstat, uptime,
etc.  As it turns out, the vast majority of this data -- and a whole
lot more -- is available via the kstat(3k) interface.  I wanted to be
able to get all this data without forking.

* How did I do this?

    The PyKstat package has several parts, starting with a helper
library in C wrapping kstat into something more friendly.  The next
layer is the lowest Python layer, and is generated by SWIG.  I
distribute the SWIG generated wrappers, so you do not need SWIG
installed to use this package... unless you want to.  The SWIG layer
is little more than a Python version of my wrappers.

    The next layer is in kstat.py.  This provides an object oriented
interface, and nicely hides the chain-following you have to do if you
use the simplest interface.  kstat.py provides a single class, Kstat,
which does all the interesting work.  This interface does however
assume some familiarity with the kstat(3k) API.

    The final layer of PyKstat, and the interface I suspect most
people will want to use, is in kstats.py.  This interface is *not*
OOPy, but it does use the Kstat class as its engine.  I strongly
recommend looking at the doc strings in kstats.py and playing with the
many functions provided to get a feel for the API.

    For the perversely curious, the dependencies are: kstat API <-
wkstat.c (my helpers) <- wkstatmodule.so (SWIG generated interface to
wkstat.c) <- kstat.py (OOPy interface to wkstatmodule.so) <- kstats.py
(the most friendly interface).

* What about bugs?

    There are doubtless some bugs.  However, I don't believe there are
any major ones.  I have been using this software to check machine
statistics for a few months now, and in fact some single Python
programs have been running *continuously* for a month and reporting on
machine statistics.

    If you discover any bugs, certainly let me know, but try to
include as much information as possible, and a snippet of code which
causes the problem.

    The install process could be more elegant.

* Cool!  How do I install this.

    Take a look at the Makefile.  It's pretty simple.  You should only
need to change your CC, though I recommend using gcc, and the two
lines saying where your Python is installed.  Then run 'make', then
'make install'.  The install process is crude in the extreme, and
simply copies a few files into your Python library area.

    Check out the few examples in the examples directory to get an
idea of how this package can be used.  I've included the program netd
as an example.  It will *not* work for you, since it requires
libraries I've not made public yet.  It does, however, provide an
interesting example.

William S. Annis
99jun09

-- 
William Annis - System Administrator - Biomedical Computing Group
annis at biostat.wisc.edu                       PGP ID:1024/FBF64031
Mi parolas Esperanton - La Internacia Lingvo    www.esperanto.org



From mss at kepler.de  Mon Jun 14 12:33:48 1999
From: mss at kepler.de (Michael S. Schliephake)
Date: Mon, 14 Jun 1999 18:33:48 +0200
Subject: TKinter modal dialog
References: <018301beb673$8fdd6ca0$a9807018@yec1.on.wave.home.com>
Message-ID: <37652EEC.27373AF1@kepler.de>

Mike C. Fletcher wrote:

> The win32 extension dialogs have a "doModal", which is exactly what
> I'm looking for.  Anyone care to point out what section of the docs I
> didn't read that describes a TKinter doModal :) .

I'm using a subclass of Toplevel to do this. However the window will
appear in the Windows Task Bar.

Michael.
--------
Short listing:
--------------
DEFBTN_OK     = 0x00000001
DEFBTN_CANCEL = 0x00000002

class aDlgBox (Tkinter.Toplevel):
  def __init__( self, parent, title, defBtns):
    Tkinter.Toplevel.__init__(self, parent)
    self.protocol( 'WM_DELETE_WINDOW', self._setAbort)
    self.State = 0
    self.title(title)
    self.BtnFrame = Tkinter.Frame(self, bd=0)
    self.BtnFrame.pack(side='bottom', fill='x', expand=1)
    if defBtns & DEFBTN_CANCEL:
      self.Cancel = Tkinter.Button(self.BtnFrame, text="Abbrechen",
                                   underline=0, command =
self._setAbort)
      self.Cancel.pack(side='right')
    if defBtns & DEFBTN_OK:
      self.Ok = Tkinter.Button(self.BtnFrame, text="Best?tigen",
underline=0,
                               command = self._setOk)
      self.Ok.pack(side='right')

  def AddChild( self, wdg, **kw ):
      wdg.pack(kw, after=self.BtnFrame)

  def Control( self ):
    oldFoc = self.focus_get()
    self.focus_set()
    self.grab_set()
    self.mainloop()
    self.grab_release()
    if oldFoc:
      oldFoc.focus_set()

    return self.State

  def _setAbort( self ):
    self.State = 0
    self.quit()

  def _setOk( self ):
    self.State = 1
    self.quit()



From news at dorb.com  Thu Jun 17 11:20:16 1999
From: news at dorb.com (Darrell)
Date: Thu, 17 Jun 1999 11:20:16 -0400
Subject: GC play
Message-ID: <mp8a3.324$U6.2508@newsr2.twcny.rr.com>

Playing with finding cycles and have a problem.
If I'm passed a reference is there a way to get its vital information like
name, filename, line number.... ?

--
--Darrell





From arcege at shore.net  Tue Jun 22 12:31:55 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 22 Jun 1999 16:31:55 GMT
Subject: Creating a new type with an extension module
References: <Ctp*vl52n@news.chiark.greenend.org.uk> <rvy*CQ92n@news.chiark.greenend.org.uk> <QpMb3.908$7X1.231730@news.shore.net> <14191.44065.236852.546922@weyr.cnri.reston.va.us>
Message-ID: <%TOb3.936$7X1.238390@news.shore.net>

Fred L. Drake <fdrake at cnri.reston.va.us> wrote:

: Michael P. Reilly writes:
:  > But the documentation on extending Python is pretty well known
:  > deficiency.  Luckily it is far easier than most languages, both to
:  > accomplish and to understand (yet more kudos to Guido).  Actually, in
:  > responce to your earlier posting, I started writing a _SHORT_ webpage
:  > explaining what is needed, which I hope to have available later today.
:  > I plan (and hope) to make it into a How-To later.

:   If you'd like to make it a patch to the standard documentation, I'm
: sure many people would be incredible grateful, including myself.  The
: time I have available to extend the documentation in all the ways that 
: are requested seems a bit fleeting at the moment.
:   Seriously, if anyone would like to extend the standard documents,
: I'll be glad to lend a hand with the markup.  If you write the
: content, I'll accept it as text if I need to!

I can do that, Fred.  I like the Python Tex style, but I do have
problems making the necessary DVIs from your tools.  I may need your
help on that (sun-i386-solaris).  I do have to download your newest
version tho.

First things first: get the little HTML ref kit out. :)

  -Arcege




From kiki at pixar.com  Sat Jun 19 23:59:40 1999
From: kiki at pixar.com (Cynthia Pettit)
Date: Sat, 19 Jun 1999 20:59:40 -0700
Subject: newbie tkinter -- radio buttons
Message-ID: <376C672C.F817E0C0@pixar.com>

Okay, you get your choice of answering any single one of these questions
to solve my woes! :)

I'm making a color menu in a drawing program.  It allows you to add
named colors as you go.

At first, I was adding colors this way:

  self.color_menu.add_radiobutton(label=color_name,
   value = color_name,
   variable = self.current_color)

[Note: a couple other issues: value and variable don't seem to work.
When I choose the button, it does not change the value of
self.current_color.]

This works, but the problem is I can't set the first one to being
"selected", nor can I change the new color to be "selected"...  it looks
really dorky to have no colors selected when you start and when you add
new colors.  You have to *select* a color to start the radio buttons
working [after that, they're fine and switch properly among them.]

So I changed to this, so I could call the select() method:

  temp = Radiobutton(self.color_menu, text=color_name,
   value = color_name,
   variable = self.current_color)
  temp.pack(anchor=W)
  temp.select()

But now, of course, this conflicts with other calls, like add_command
and add_separator.  [eg "Add Color"]  The buttons look to be on top of
eachother.  I think this is due to the .pack() calls not working with
the add_radiobutton calls.

Here's a choice of things that if solved would save me [in order of
preference]:

Call .select() on an add_radiobutton [and just how *does* one ever
access these buttons ever again?!?!]
Integrate add_command and .pack() so they don't stomp on eachother
Be able to temp = Separator(self.color_menu)

I'm sure each of these have very simple answers...

Thanx!

Kiki




From luisroco at hotmail.com  Fri Jun 11 18:31:48 1999
From: luisroco at hotmail.com (Luis Roco Arriagada)
Date: Fri, 11 Jun 1999 22:31:48 GMT
Subject: Using PIL with CGI scripts
Message-ID: <19990611223149.63326.qmail@hotmail.com>

Mark:

Sorry for bothering you again, but I can't figure out how can I do this 
work.

I can`t write the image to the disk because all of the images are inside a 
single file where they are compressed using pickle. What I'm trying to do is 
to get some images from this file according to a word search perform by the 
user of the web site. This part is already done, so the images are extracted 
from the file but they are in a PIL format, ie they are an Image type, and 
for that i cannot display them in a web page using the HTML tag <IMG SRC = > 
because this tag only accepts a file name as input.

I try to use cStringIO to emulate a file, but I don't know how I must use 
this variable with the HTML tag.

This part of the code is something like this:
from cStringIO import StringIO

print "Content-type: text/html"
print
print "<HEAD><TITLE>WEBIMAGE</TITLE></HEAD>"
print "<BODY BGCOLOR=\"#FFFFFF\" TEXT=\"#000000\" LINK=\"#0000FF\" 
VLINK=\"#000080\" ALINK=\"#FF0000\">"
print "<center><H1>WEBIMAGE</h1>"
print "<p>Version 0.01</p>"
print "<hr>"
.
.
.
temp = StringIO()
im.save(temp,"JPEG") #im is an Image variable
temp.seek(0)
print "<p><IMG SRC='%s'>"%temp

print "</center>"
print "</body>"


Hope you can help me and thanks in advance
Luis Roco

>From: "Mark Nottingham" <mnot at pobox.com>
>To: <luisroco at hotmail.com>,<python-list at cwi.nl>
>Subject: Re: Using PIL with CGI scripts
>Date: Wed, 9 Jun 1999 08:27:23 +1000
>
>See the StringIO module (actually, cStringIO).
>
>One thing to keep in mind - unless the image is *really* dynamic (i.e., it
>changes EVERY time someone accesses it), you're much better off if you 
>write
>it to disk.
>
>See genpix.py at http://www.mnot.net/python/  for an example of this (it's 
>a
>command line utility that HTMLifys a directory of images, but it does use
>cStringIO in this manner).
>
>Cheers,
>
>
>----- Original Message -----
>From: <luisroco at hotmail.com>
>To: <python-list at cwi.nl>
>Sent: Wednesday, June 09, 1999 7:10
>Subject: Using PIL with CGI scripts
>
>
> > Hi all
> >
> > How can a CGI scripts outputs graphics images through Python code using
>the Python Imaging Library package?
> >
> > I just want to produce some images which has been processed using PIL 
>and
>show them in a web site
> >
> >
>


______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com



From bgue at my-deja.com  Sat Jun  5 22:58:08 1999
From: bgue at my-deja.com (bgue at my-deja.com)
Date: Sun, 06 Jun 1999 02:58:08 GMT
Subject: format of event.char?
Message-ID: <7jco3t$kpj$1@nnrp1.deja.com>

Hi everyone. I'm wondering what the format of event.char is...how can I
append it to a string? I've got a string, self.str, and a callback
bound to keystrokes:

def callback(self, event):
  self.str = self.str + event.char


...but this doesn't work. The exact error is (ommitting trace):

TypeError: neither _and_ nor _rand_ defined for these operands

So how does one use event.char for things other than printing?


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From rozzin at geekspace.com  Wed Jun 23 22:32:43 1999
From: rozzin at geekspace.com (Joshua Rosen)
Date: Wed, 23 Jun 1999 22:32:43 -0400
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0rr2.2fi.behrends@allegro.cse.msu.edu> <8L_b3.1213$4s.136166@typhoon-sf.snfc21.pbi.net>
Message-ID: <377198CB.5FE62D0D@geekspace.com>

Zigron wrote:
> 
> > Among all the suggestions for Python 2.0, removing the necessity of
> > prefixing methods and instance variables with "self." (and not having
> > to list it as an argument, either) would be my #1 wish.
> 
>     Whyfor? I personally like the seperate scopes, beween the class
> and the method. It makes it so I don't have to clean up as much. I can
> use as many temporary variables as I want in a method, and as soon
> as that method is gone...... its gone.

	The other reason is that explicitely using a self-reference for instances goes
along with the method of calling unbound class-methods, and allows one to
easily `fake' membership of an instance in a class.

> I thought pv2 was going to
> objectify eeeeverything. (I can't tell you how much I want strings
> turned into objects! "blah".append('hi') should return 'blahhi'! :-))

	Last I checked, everything in Python already was an object;)
	Strings -are- objects--they just don't belong to a class or have any
properties/members/whateveryouwanttocallthem (some non-class--`top-level' I
suppose they should be called--objects do have members, like lists and
imaginary numbers).
	I rather like the var-name-subsitution that PERL and Ruby and some other
languages do (ie: $something="bar"; "foo$something" == "foobar"), because it
shortens code, and sometimes makes it much prettier (having stripped out all of
the `end string, use an operator, reference something, resume new string'
procedures); Python's %-replacement is nice, too, though a bit more lengthy.

	One thing to remember: languages all have their strengths and weaknesses, and
-rarely- is a language best for -everything-.

(back on track with the start of the thread: yeah, the shortcut operators
wouldn't really work well with Pythonic `assingment'/name-binding and immutable
objects).

		-Rozzin.




From wallenborn at phys.chem.ethz.ch  Tue Jun 29 09:02:40 1999
From: wallenborn at phys.chem.ethz.ch (Ernst-Udo Wallenborn)
Date: 29 Jun 1999 15:02:40 +0200
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <7l4nht$jl7$1@nnrp1.deja.com> <9DDBCC52A7B26002.649D772F7232D418.CE2709D4B08E7401@lp.airnews.net>
Message-ID: <ui6747qhvz.fsf@bacon.ethz.ch>


To ask a completely different question: What about
threads and GUI? With Tkinter you have to rewirte the
mainloop like this

    def mainloop(self):
        while 1:
            Frame.update(self)
            time.sleep(0.02)
            if self.quitting == 1:
                sys.exit(0)


to convince it to run with thread or threading.  I remember having
read a comment from Guido van Rossum, citing Jan Osterhout that Tk
will never be thread-safe and that therefore this problem will not go
away.

Would this be any different with a different GUI kit?

-- 
Ernst-Udo Wallenborn
Laboratorium fuer Physikalische Chemie
ETH Zuerich



From tim_one at email.msn.com  Sun Jun 27 22:26:51 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 27 Jun 1999 22:26:51 -0400
Subject: GC In Python: YAS (Yet Another Summary)
In-Reply-To: <3776A025.53B35103@inrialpes.fr>
Message-ID: <000001bec10d$aeca9440$e19e2299@tim>

[Tim, in a thread about BDW-style conservative GC]
> ...
> Python could use its own malloc for its own objects (& Vladimir
> Marangozov has written a very nice PyMalloc, available from his
> Starship page), but that breaks down as soon as any extension has
> its own view of how to get memory.

[Vladimir Himself]
> I can't let you say this (heh, it's too late ;-) because you seem to
> overlook one of the nicest features of pymalloc. If the extension module
> works in current Python + libc malloc, it will work with Python +
> pymalloc because of the dynamic block management.
> ...

Hmm!  I at least agree I didn't say what I meant <wink>.

I realized PyMalloc plays peacefully with other mallocs as-is.  What "breaks
down" is extension to BDW-style GC, where the collector wants to perform
everybody's allocations.  IOW, PyMalloc pushes a Python-specific allocation
scheme as far as it's ever likely to go, but that falls short of what BDW
requires.

You're not required to solve this.  Heck, you're not even encouraged to view
it as "a problem" <wink>.

plenty-of-others-will-ly y'rs  - tim





From arielle at adams.patriot.net  Tue Jun 15 10:40:47 1999
From: arielle at adams.patriot.net (A.K. Masters, Frumious Bandersnatch)
Date: Tue, 15 Jun 1999 14:40:47 GMT
Subject: easy filecopy? - JPython?
References: <376531D3.7E42AC42@phoenix-edv.netzservice.de> <aahzFDCq4D.Do1@netcom.com>
Message-ID: <PBt93.843$B04.2678@newsfeed.slurp.net>

Aahz Maruch (aahz at netcom.com) wrote:
: Holger Jannsen  <holger at phoenix-edv.netzservice.de> wrote:
: >For sure, I could write a thing like open-read-write-close...
: Look at the shutil module.

Copy, copyfile, copymode in shutil apparently haven't been implemented
under jpython.  Any copy ideas for jpython (Unix, of course)?

Thanks!
Arielle
--
Arielle K. Masters: arielle at bigfoot.com; http://adams.patriot.net/~arielle/
Lucy dog:                           http://adams.patriot.net/~arielle/lucy/
Herndon Area Dog Coalition (HADC):  http://adams.patriot.net/~arielle/hadc/
Genealogy:                        http://adams.patriot.net/~arielle/geneal/



From abc at def.ghi.jkl  Thu Jun 10 15:00:45 1999
From: abc at def.ghi.jkl (DORAEMON.HK)
Date: Fri, 11 Jun 1999 03:00:45 +0800
Subject: Serial programming
Message-ID: <7jp1jt$8hg$1@imsp009a.netvigator.com>

I am looking for some example in python about send and receive data from
serial port.  Like send "ATZ" and receive "OK".  Any help?? Thanks a lot!!


Doraemon









From gerrit.holl at pobox.com  Fri Jun 18 16:50:02 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Fri, 18 Jun 1999 22:50:02 +0200
Subject: Starting a project: any hints?
In-Reply-To: <19990617154643.A1680@optiplex.palga.uucp>; from Gerrit Holl on Thu, Jun 17, 1999 at 03:46:43PM +0200
References: <19990617154643.A1680@optiplex.palga.uucp>
Message-ID: <19990618225002.A23494@optiplex.palga.uucp>

On Thu, Jun 17, 1999 at 03:46:43PM +0200, Gerrit Holl wrote:
> Date: Thu, 17 Jun 1999 15:46:43 +0200
> From: Gerrit Holl <gerrit.holl at pobox.com>
> To: python-list at python.org
> Subject: Starting a project: any hints?
> 

[cut ask for hints]

thanks for all of your hints!

The project homepage is now at http://www.nl.linux.org

groeten/regards,
Gerrit.

--
The Dutch Linuxgames website.		De Nederlandse Linuxgames pagina.
Site address:				http://linuxgames.nl.linux.org
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
Discoverb: learn words and definitions	Discoverb: leer woordjes en definities
Site address:				http://nl.linux.org/~gerrit/discoverb
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
Personal homepage:			http://nl.linux.org/~gerrit/



From jwpolley at collins.rockwell.com  Sun Jun 27 16:43:11 1999
From: jwpolley at collins.rockwell.com (Jonathan Polley)
Date: Sun, 27 Jun 1999 15:43:11 -0500
Subject: Python 1.52/Tkinter/IDLE Problem
Message-ID: <37768CDF.F8A3947F@collins.rockwell.com>

I have just finished building version 1.5.2 of the python interpreter
for Solaris 2.1 and have discovered a problem that I do not get with the
Windows version.  Let me first say that IDLE runs just fine, with one
exception, and that makes me think that my build of the interpreter went
correctly.  When I open a module in IDLE and use the right mouse button
to set a breakpoint, I get the following traceback in the "Python Shell"
window:

Exception in Tkinter callback
Traceback (innermost last):
  File "../../Lib/lib-tk/Tkinter.py", line 764, in __call__
    return apply(self.func, args)
  File "EditorWindow.py", line 217, in right_menu_event
    rmenu.tk_popup(event.x_root, event.y_root)
  File "../../Lib/lib-tk/Tkinter.py", line 1506, in tk_popup
    self.tk.call('tk_popup', self._w, x, y, entry)
TclError: invalid command name "tk_menuSetFocus(.3314608.text.3461936)"

The popup is displayed correctly but I get the above traceback when it
appears. If I have the debugger running, the line that I selected is
highlighted, so it looks as if the breakpoint was set properly (it
behaves the same way as python for Windows).

Any ideas as to what is happening????


Thanks for your assistance.

Jonathan Polley
jwpolley at collins.rockwell.com





From donn at u.washington.edu  Fri Jun 25 19:03:53 1999
From: donn at u.washington.edu (Donn Cave)
Date: 25 Jun 1999 23:03:53 GMT
Subject: C++ and Python
References: <3772AA7D.A7855B93@anisci.com> <3773761C.1B822655@cfdrc.com>
Message-ID: <7l11sp$8m6$1@nntp6.u.washington.edu>

Lyle Johnson <jlj at cfdrc.com> writes:
| This is indeed a tricky area. I recommend you start by reading this paper,
| several times:
|
|     http://www.python.org/workshops/1995-12/papers/ahlstrom1.html
|
| Then to see this implemented in a real extension, take a look at WPY itself,
| wxPython or FXPy (and probably others?).

I'll second that.  I looked at this paper to get started with an interface
to BeOS API classes, which use virtual functions to give the programmer
hooks into the system event dispatching, for example.

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



From claird at Starbase.NeoSoft.COM  Tue Jun 29 11:27:05 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 29 Jun 1999 10:27:05 -0500
Subject: Python GUIs: Summary and Conclusion
References: <99FA4A1A1280BEBD.0A1ABA62CC989479.4C2C6022F695BEAC@lp.airnews.net> <377805C8.C6BBD4C1@unifix-online.com>
Message-ID: <7laok9$bpi$1@Starbase.NeoSoft.COM>

In article <377805C8.C6BBD4C1 at unifix-online.com>,
Philippe Chassany  <philc at unifix-online.com> wrote:
>> ....
>> More widgets should be added to Tk(Inter).
>> ....
>
>Our company has made a 100% Tcl/Tk Toolkit named "BWidget".
>BWidgets are "well known" in the Tcl community because
>they are free, flexible, powerful and completly interpreted (no C).
>More than 25 widgets are available.
>
>We have tested BWidget with Tkinter the result is quite fine :
>Quick and good looking. It's a good alternative to GTk or Qt
>compiled programming to have professional application GUIs.
>
>Try it to make your own idea (free download, screenshots,
>full HTML manuals)
>
>http://www.unifix-online.com/BWidget
			.
			.
			.
Philippe's right:  Tcl-ers have received Unifix's
BWidgets *very* warmly.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From guido at CNRI.Reston.VA.US  Mon Jun 14 15:24:38 1999
From: guido at CNRI.Reston.VA.US (Guido van Rossum)
Date: Mon, 14 Jun 1999 15:24:38 -0400
Subject: [meta-sig] WHY are all those sigs ending??
In-Reply-To: Your message of "Mon, 14 Jun 1999 20:47:54 +0200."
             <19990614204754.A26177@optiplex.palga.uucp> 
References: <19990614204754.A26177@optiplex.palga.uucp> 
Message-ID: <199906141924.PAA01494@eric.cnri.reston.va.us>

> I was very happy when I saw all thos SIGs, but they are all retiring? WHY?
> XML-sig, DB-sig, Doc-sig, String-sig: if it's because there isn't interest:
> I'm interested!

Ah, this happens twice a year.  In order to be able to recycle
inactive SIGs, we regularly determine a new expiration date for each
SIG.  Typically, the expiration date is about half a year in the
future.  At the expiration date, the SIG's archives are inspected.  If
the SIG has been comatose or dead since the expiration time was last
set, it is retired; otherwise, the expiration date is moved another 6
months.  Some SIGs (e.g. the META SIG, which should never be retired)
have different rules.

While some of the SIGs are a bit slow, none look comatose or dead at
the moment, so I propose to renew all SIGs expiring this month for
another half year.

Any objections?

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



From avv at quasar.ipa.nw.ru  Mon Jun 28 15:28:05 1999
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Mon, 28 Jun 1999 12:28:05 -0700
Subject: [FYI] Future and Status of the RPM based Linux distribution
References: <7kqcpl$u4l$1@news.rhein-zeitung.DE>
Message-ID: <3777CCC5.B76EE975@quasar.ipa.nw.ru>

Dear Oliver,

Thank you for your great effort.

olli at rhein-zeitung.de wrote:

>     - RedHat Linux 5.2 (glibc 2.0)
>     - RedHat Linux 6   (glibc 2.1)
>     - Mandrake Linux 6 (glibc 2.1 and pentium optimized)
>     - (maybe) SuSE 6.x (glibc 2.x)

Please do not forget SuSE, if at all possible...

BTW, I succeeded in installing your current RPMs under SuSE 6.1, but it required to
convert manually their libreadline.a to libreadline.so (which they do not ship).

Alexander





From dickey at shell.clark.net  Sun Jun 20 19:39:46 1999
From: dickey at shell.clark.net (T.E.Dickey)
Date: Sun, 20 Jun 1999 23:39:46 GMT
Subject: Problem with Curses
References: <ppgmj7.s7r.ln@solling.seismic.de> <jpeck7.1661.ln@solling.seismic.de> <ve1dk7.1e02.ln@solling.seismic.de> <87ogidzs37.fsf@home.ivm.de>
Message-ID: <6Zeb3.526$4e1.32195@iad-read.news.verio.net>

Klaus Schilling <Klaus.Schilling at home.ivm.de> wrote:
> kjwolf at online.de (Klaus-Juergen Wolf) writes:
>> It depends on when the executable is linked to libmytinfo, which only
>> takes place on platforms that don't know the terminfo database, like
>> BSD and Solaris; there are two competing declarations of tputs().

> Why doesn't Freebsd know the terminfo database? Can't the one that is 
> integrated into ncurses be used on FreeBSD? 

I've been told ncurses 4.2 is incorporated in the FreeBSD 3.1 - but that
may still just be a port...

-- 
Thomas E. Dickey
dickey at clark.net
http://www.clark.net/pub/dickey



From anders777 at my-deja.com  Wed Jun 30 12:30:04 1999
From: anders777 at my-deja.com (anders777 at my-deja.com)
Date: Wed, 30 Jun 1999 16:30:04 GMT
Subject: Reflection in Python?
Message-ID: <7ldglr$ita$1@nnrp1.deja.com>

I've just started programming in Python, and I
could really use some help figuring out something.
 I'd like to implement Kent Beck's unit tester in
Python (someone posted a very nice alternative a
few weeks ago, but Junit's approach works better
for my project's needs).  How do I do the
equivalent of what Java calls reflection? I want
to do the following:

test = MathTest("testAdd")

where test.run() will run the "testAdd" method on
the MathTest object?  I also want to be able to
tell the user the name of the object (MathTest)
and the method in case the method fails.  Any
thoughts?

Thanks,
Anders Schneiderman
National Journal Daily Briefings Group




Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From C.Kotsokalis at ece.ntua.gr  Fri Jun 18 19:29:51 1999
From: C.Kotsokalis at ece.ntua.gr (Constantinos A. Kotsokalis)
Date: 18 Jun 1999 23:29:51 GMT
Subject: Tkinter & socket question
References: <19990618215411.23898.qmail@nwcst277.netaddress.usa.net>
Message-ID: <slrn7mllgk.dh.C.Kotsokalis@ppp02.softlab.ece.ntua.gr>

On Fri, 18 Jun 1999 21:54:11 GMT, Def P wrote:
>2) I need to poll the connection periodically, say, every second. This would
>not have been a problem in console mode... I don't know how to do this in a
>Tkinter environment though. So, how can I do something every second (or every
>n milliseconds, or whatever the time span may be)? Are threads the way to go?
>I haven't used them before.

Well you can always check for a recv() that returns a zero-length
buffer. That means that the other end has closed connection.

  --C

-- 
Constantinos A. Kotsokalis || C.Kotsokalis at ece.ntua.gr
National Technical University of Athens - GREECE
Electrical and Computer Engineering Department
"Bus error -- driver executed."



From markusk at bidra241.bbn.hp.com  Mon Jun 28 03:58:28 1999
From: markusk at bidra241.bbn.hp.com (Markus Kohler)
Date: 28 Jun 1999 09:58:28 +0200
Subject: GC In Python: YAS (Yet Another Summary)
References: <Pine.SUN.3.95-heb-2.07.990616155228.15761A-100000@sunset.ma.huji.ac.il> <p53dzrfcea.fsf@bidra241.bbn.hp.com> <930267199.139037@clint.waikato.ac.nz> <37730BAA.28A7C738@bioreason.com> <930340397.465868@clint.waikato.ac.nz> <3773EFF7.27D29899@bioreason.com>
Message-ID: <p5iu88n4d7.fsf@bidra241.bbn.hp.com>

Andrew Dalke <dalke at bioreason.com> writes:

> Stuart Yeates <syeates at cs.waikato.ac.nz> said to me:
> > I'm not quite sure what you mean by 'integer handle,' but if you
> > do a straight cast then it shouldn't be a problem (since the 
> > collector operates on the binaries representation). If you mean
> > that it uses an index to an array of pointers (or some such), 
> > then that shouldn't be a problem either, since the pointer still
> > exists somewhere (i.e. the array). 
> 
> The latter.   Example object creation code might look like:
> 
> dt_Handle mol = dt_smilin("CCC", 3);
> 
> where dt_Handle is a typedef to unsigned long.  In the SWIGged
> wrapped version, this is mol = dt_smilin("CCC").
> 
> Yes, the library has a pointer table to the object.  The problem
> I see is, how does the Boehm collector (or other GC) know what to
> collect from this vendor library?  I can't see how, given that
> my Python code stores it as an integer.

As long as you have a pointer to the array it will not be reclaimed by the GC. 
The integer doesn't count as a pointer. It wouldn't count in any GC I know. 

The Boehm Weiser collector also has special support for interior
pointers,  which are pointers  into an array. 

> 
> That's why I don't see how you say "It fully supports C and C++"
> when this is a package which doesn't do "pointer XORing or other
> crimes" but cannot be usable with a GC w/o modifications.

The claim is right. Even in a cooperative environment your integer
wouldn't count as a reference to an object.  

> 
> But then, I'm a computational biophysicist by training and don't
> know much about GC other than the general concepts.  One of the
> references on the SGI page says Boehm can be used in "uncooperative
> environments" so I'll end by saying that I don't know enough.
> 

The Boehm collector is a conservative collector, which means it just
scans the memory and if there's an integer that points to a valid
address it wouldn't reclaim the object at this address. 

Your integer wouldn't probably be a valid address and therefore
wouldn't be counted  by the GC. 


Markus
-- 
Markus Kohler  mailto:markus_kohler at hp.com



From doctorwes at mindspring.com  Tue Jun  8 04:24:19 1999
From: doctorwes at mindspring.com (Wesley Phoa)
Date: Tue, 8 Jun 1999 01:24:19 -0700
Subject: Using the Windows clipboard in Python
Message-ID: <7jik35$esd$1@nntp6.atl.mindspring.net>

This is a preliminary version of a module which lets you exchange data via
the Windows clipboard, either as a string or as a 2D array. For example,
this lets you copy and paste data between Python programs and Excel.

ftp://ftp.mindspring.com/users/doctorwes/clipboard.py

It uses the Win32 extensions and Sam Rushing's calldll module. There must be
an easier way - I'm a Windows novice, so suggestions are very welcome.

If you use this module (or calldll generally) you may notice mysterious but
apparently harmless errors when exiting Python; something to do with
libraries not unloading. If you know why this happens, please tell me!





From Timothy.R.Evans at p98.f112.n480.z2.fidonet.org  Wed Jun 30 19:10:51 1999
From: Timothy.R.Evans at p98.f112.n480.z2.fidonet.org (Timothy R Evans)
Date: Thu, 01 Jul 1999 00:10:51 +0100
Subject: printing with print
Message-ID: <0000159f@bossar.com.pl>

From: Timothy R Evans <tre17 at pc142.cosc.canterbury.ac.nz>

Angus MacKay <amackay at starvision.com> writes:

> is there any other way to print that with print?
> perhaps with a real function.
> 
> I want to print two things without a space being added between them:
> foo = "there"
> print "hi" + foo
> 
> works fine but: 
> foo = 2
> print "hi" + foo
> 
> does not. I like the printing of any object ability of "print" so I
> do not want to use "" % style printing, I just want to be able to
> govern the whitespace that gets output.
> 
> cheers, Angus.

import string

def print_without_space(*args):
    print string.join(map(str, args), '')


>>> foo = 2
>>> print_without_space('foo', '2', foo)
foo22

This does what you want, but wouldn't the % operator be easier?

--
Tim Evans



From MHammond at skippinet.com.au  Wed Jun 16 18:27:46 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Thu, 17 Jun 1999 08:27:46 +1000
Subject: how to play waves?
References: <19990616210844.A8571@optiplex.palga.uucp>
Message-ID: <7k98ba$3o6$1@m2.c2.telstra-mm.net.au>

Gerrit Holl wrote in message <19990616210844.A8571 at optiplex.palga.uucp>...
>Hello all,
>
>is it possible to _play_ a wave file with python? So not importing a wave
>module which is much to large and searching hours for the way to do it... I
>just want to give the name of a file and the module plays it for me... is
it
>possible?

What OS?  If I guessed Windows, I would say "use the builtin winsound
module".

>The Dutch Linuxgames website. De Nederlandse Linuxgames pagina.
>Everything about games on Linux. Alles over spelletjes onder Linux.

But it appears I would have guessed wrong :-)

Mark.






From tfb at tfeb.org  Wed Jun 30 09:26:00 1999
From: tfb at tfeb.org (Tim Bradshaw)
Date: 30 Jun 1999 14:26:00 +0100
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <37682DD3.43923243@compaq.com> <7ka2be$qja$1@nntp6.u.washington.edu> <37776FFA.2D85@mailserver.hursley.ibm.com> <lw7looxx7m.fsf@copernico.parades.rm.cnr.it> <876748mcxg.fsf@home.ivm.de> <37788BB9.ABD@mailserver.hursley.ibm.com> <lw67471fyi.fsf@copernico.parades.rm.cnr.it> <3779ED0F.2C67@mailserver.hursley.ibm.com>
Message-ID: <nkjzp1hizvb.fsf@tfeb.org>

 Paul Duffin <pduffin at mailserver.hursley.ibm.com> writes:

> I am surprised at this because Tcl's [upvar] is simply an explicit form
> of dynamic binding which I seem to remember is what Lisp uses. (That
> is how the 'let' function can be (is) implemented as a lambda function).

No, (Common) Lisp doesn't use dynamic binding by default, and you
don't need anything like upvar to implement LET.  In fact LET is
really the most common case of a lisp design pattern -- `WITH-x ->
CALL-WITH-x:

	(let ((a b)) . body)
     -> ((lambda (a) . body) b)

(Writing the macro for this is a little fiddly if you want to get it
right in all cases, but not more than 10 minutes work.)

> I find that Lisp macros (while a very powerful and necessary mechanism)
> are sooo confusing. They are in essence another language inside Lisp,
> and as such introduce inconsistencies. Tcl on the other hand doesn't
> need a macro language and as such is much more consistent than Lisp.

But the language of Lisp macros is Lisp, that's really the whole
point!  Without knowing TCL, I find it hard to see how you can
introduce new constructs to the language *without* a macro language,
even if that language is TCL.

--tim






From tom7654 at my-deja.com  Fri Jun 18 16:03:22 1999
From: tom7654 at my-deja.com (tom7654 at my-deja.com)
Date: Fri, 18 Jun 1999 20:03:22 GMT
Subject: MySQL connection problem with linux
Message-ID: <7ke8m2$r0p$1@nnrp1.deja.com>

I am trying to use the MySQL module on pythonwin to acess a mysql
database on linux zues server. I cannot establish a connection with the
server, is there a a certain way i need to login, to it. If anyone has
encountered this problem and knows a way around it I would apreciate
it.




Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From tim_one at email.msn.com  Sun Jun 13 14:55:27 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 13 Jun 1999 14:55:27 -0400
Subject: dict vs kjBuckets vs ???
In-Reply-To: <7k0nl5$bgu$1@brick.cswv.com>
Message-ID: <000201beb5ce$4cfa11e0$ae9e2299@tim>

[Tim]
> Let's do a little math <wink>:  10M * 100 = ?, a lower bound on
> what you're contemplating.  Do you have gigabytes of RAM?

[Neel Krishnaswami]
> That might actually be reasonable, if you do a little more math. RAM
> is going for about a dollar a meg right now, so 2 gigs of RAM will
> cost around $2000. This is price-competitive with tying up a
> programmer for a week or two.

Alas, you still need to throw in somebody else to find and install a
motherboard capable of using that much RAM, and an OS capable of exposing
it -- we're bashing into the limits of 32-bit addressing here.  I had the
feeling the original questioner didn't have a 64-bit supercomputer just
sitting around waiting to be applied <wink>.

> I'm always surprised by how effective brute force and ignorance can
> be. :)

Oh yes, the above is temporary.  When the barbarians come knocking on your
door with an appetite for rape and pillage, it's rarely effective to point
out that this simply isn't the way it's done in your country <wink>.

the-shortest-distance-between-two-points-is-a-hammer-with-a-head-
    that-wide-ly y'rs  - tim





From Mordy.Ovits at p98.f112.n480.z2.fidonet.org  Wed Jun 30 21:13:11 1999
From: Mordy.Ovits at p98.f112.n480.z2.fidonet.org (Mordy Ovits)
Date: Thu, 01 Jul 1999 02:13:11 +0100
Subject: Python suitable for a game engine?
Message-ID: <00001616@bossar.com.pl>

From: Mordy Ovits <movits at lockstar.com>

Alan Daniels wrote:
> 
> I'm working on a Quake-style game engine, where my goal is to give it
> a fully integrated editor, similar to the excellent ones in Starcraft
> and Unreal. I'm planning on doing this by giving the game and the
> editor as much back-end code in common as possible.

The game Extreme Warfare was a realtime 3D videogame that used Python as its
scripting language for AI and game logic.  Take a good look at it.

Does this project of yours have a homepage I could look at?

Mordy
-- 
o Mordy Ovits
o Cryptographic Engineer
o LockStar Inc.
---------------------------------------------------------------------------
#!/usr/local/bin/python 
from sys import*;from string import*;a=argv;[s,p,q]=filter(lambda x:x[:1]!=
'-',a);d='-d'in a;e,n=atol(p,16),atol(q,16);l=(len(q)+1)/2;o,inb=l-d,l-1+d
while s:s=stdin.read(inb);s and map(stdout.write,map(lambda i,b=pow(reduce(
lambda x,y:(x<<8L)+y,map(ord,s)),e,n):chr(b>>8*i&255),range(o-1,-1,-1)))



From mal at lemburg.com  Fri Jun 11 09:07:47 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 11 Jun 1999 15:07:47 +0200
Subject: Help: MIME Attachments
References: <7jqlc1$bk0$1@nnrp1.deja.com>
Message-ID: <37610A23.6A90DB35@lemburg.com>

k_mcdermott at my-deja.com wrote:
> 
> Hi,
> 
> I am trying to send a plain text file, as an attachment through e-mail.
> 
> This is a hacked version of my test code below, basically the
> program interprets a dump from a diary system, and then writes
> Lotus Notes importable Calendar entries.  I want to then mail
> them to users with an accompanying instruction header to allow them
> to import them into their calendars...
> 
> The only problem is the mailing them to users bit, I either get
> the plain text in the file (which I don't want, I just want them
> to detach the text file and then import) or I get the Base64 encoded
> text, either way there is no attachment...
> 
> My Problem is that the MIME headers indicating Base64 encoding
> appears a line after the content type, and I believe that this blank
> line terminates the headers...
> 
> So my question is, at what point should I be introducing these
> headers...?
> 
> TIA
> 
> Kevin
> 
> ps Python has brought a project that was originally quoted at "tens of
> thousands of pounds and hundreds of man-hours" down to around 4 hours
> so far, and if I can crack this, ZERO outlay :-)
> 
> import base64
> import sys
> import StringIO
> import MimeWriter
> from smtplib import SMTP
> 
> # Mails the Diary Import file to the user in MIME format
> # Based upon code by GVR
> def mailFileToUser(_userName, _fileName):
>         outputfp = StringIO.StringIO()
>         w = MimeWriter.MimeWriter(outputfp)
>         w.addheader("subject", "Diary Entries from Mainframe")
>         w.flushheaders()
>         w.startmultipartbody("mixed")
>         subwriter = w.nextpart()
>         f = subwriter.startbody('application/octet-stream;
> name="DiaryFile.notes"')
> 
> >>> This is the puzzling bit...
> 
>         subwriter.addheader("Content-Transfer-Encoding", "base64")
>         subwriter.addheader("Content-Disposition", 'attachment;
> filename="DiaryFile.notes"')
> >>>

Reading the source in MimeWrite.py I'd suggest moving these lines
up just behind the subwrite = ... line. The .startbody()
call flushes the headers for you and also terminates the headers
(by inserting the blank line you already noticed).

Should work then, I guess.
 
>         subwriter.flushheaders()
>         base64.encode(open('./DiaryFile.notes', 'r'), f)
>         w.lastpart()
>         # s = SMTP("localhost")
>         # s.sendmail("diary at system.com", ["my at email.address.com"],
> outputfp.getvalue())
>         # s.close()
>         print outputfp.getvalue()
> if __name__=='__main__':
>         mailFileToUser('Kevin McDermott', 'TestFile.out')
> 
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   203 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From ivanlan at callware.com  Fri Jun  4 14:46:14 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Fri, 4 Jun 1999 18:46:14 GMT
Subject: "The Python Way"
References: <Pine.A41.4.10.9906041158070.41420-100000@barney.mcit.com>
Message-ID: <37581EF6.8E903744@callware.com>

Hi All--

Robert Meegan wrote:
> 
> On Fri, 4 Jun 1999, Graham Matthews wrote:
> 
> >

[snip]

> 
> One problem that exists with suggestions for improvements to Python is that
> different people have different ideas about what features are essential to
> the language. Most people seem to agree that objects and whitespace are
> core features of Python, but others have made (more or less) well-reasoned
> arguments pointing out the weaknesses inherent in each of these.
> 

[...]

> The important thing is neither to quash dissent nor to debate the number of
> angels that can dance on the head of an object, but to freely share ideas
> in the hope of improving both Python and our own pool of knowledge.
> 
> This learned opinion is, of course, not worth the bits it's written on.
> 

Ah, but it is.  Very nice summary of how the Python community works and
gets things done, Robert.  Thank you.

<sorry-for-the-me-too-post-but-roberts-rules-dont-apply-here-except-when-they-do>-ly
y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From mgm at unpkhswm04.bscc.bls.com  Thu Jun 24 09:49:17 1999
From: mgm at unpkhswm04.bscc.bls.com (Mitchell Morris)
Date: 24 Jun 1999 13:49:17 GMT
Subject: Thanks for Python...
References: <376ead64@anonymous.newsfeeds.com> <VrDQRddcbbs9-pn2-QdftOXYk6Hdd@Fortress> <37716901.43CB604@lockstar.com> <14193.16609.554364.194800@lrz.de>
Message-ID: <slrn7n4dqt.cge.mgm@unpkhswm04.bscc.bls.com>

In article <14193.16609.554364.194800 at lrz.de>, Eugene Leitl wrote:
[snip]
>
>*Is* there a Python for the PalmPilot out there? 'cause I would be
>buying one if there was. I've got a Garmin 25 OEM GPS board I'd like
>to put to some use...
>
>-- Eugene
>

I suspect there won't be one running native ... the memory model of the Palm
is sufficiently restrictive that it may be fundamentally incompatible with
the Python memory model. On the other hand, I'd love to be proven wrong.

That said, you can write all the host-side components to your Palm app
in Python:
	http://www.concentric.net/~n9mtb/cq/pyrite/pyrite.html

+Mitchell


-- 
Mitchell Morris

Microsoft via the genius of Bill Gates invented the mouse-windows user
interface, reliable operating systems, affordable computing, and the Internet;
if you don't think all that is true, ask someone who has never used a computer
and whose only exposure to the industry is through mass media.
	-- Philip Greenspun



From aahz at netcom.com  Tue Jun 22 11:44:49 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Tue, 22 Jun 1999 15:44:49 GMT
Subject: newbie idiom question
References: <m3emj4yezq.fsf@monsoon.swcp.com> <m3emj43fnl.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <aahzFDqJqp.82w@netcom.com>

In article <m3emj43fnl.fsf at atrus.jesus.cam.ac.uk>,
Michael Hudson  <mwh21 at cam.ac.uk> wrote:
>
>If I know I'm not going to be altering the length of the list, I
>generally do it like this:
>
>for i in range(len(a)):
>    if is_cooked(a[i]):
>        a[i] = "cooked"

And, just as a reminder for anyone reading this, this does not work for
tuples (as I got forcibly reminded yesterday).
-- 
                      --- Aahz (@netcom.com)

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

Disciple of Uncle Chuck



From rdev at my-deja.com  Wed Jun 30 14:02:46 1999
From: rdev at my-deja.com (rdev at my-deja.com)
Date: Wed, 30 Jun 1999 18:02:46 GMT
Subject: Reflection in Python?
References: <7ldglr$ita$1@nnrp1.deja.com>
Message-ID: <7ldm3s$led$1@nnrp1.deja.com>

In article <7ldglr$ita$1 at nnrp1.deja.com>,
  anders777 at my-deja.com wrote:
>  ...How do I do the
> equivalent of what Java calls reflection? I want
> to do the following:
>
> test = MathTest("testAdd")
>
> where test.run() will run the "testAdd" method
on
> the MathTest object?  I also want to be able to
> tell the user the name of the object (MathTest)
> and the method in case the method fails.  Any
> thoughts?
>
If I understand what you are trying to do, it's
easy in Python.  Assuming MathTest is an
object instance:
test = MathTest.testAdd
will return you a bound method object which you
can then call as:
test()
Getting the name of the object class and method
is also easy.  There are several ways to do this:
1) Define a doc string within the method by
   putting a quoted string as the first line
   under the def line.  You can then refer to it
   as "test.__doc__".  This could say, for
   example, "Method bar of class foo".
2) "test.__name__" will return the name of the
   method.  "test.im_class.__name__" will
   return the name of the class that the method
   is defined within.

Example:
class MathTest:
  def testAdd(self, parm1, parm2):
      "Method testAdd of Class MathTest"
      result = parm1 + parm2
      return result
<<< mt = MathTest()
<<< test = mt.testAdd
<<< test.run(2,2)
4
<<< print test.__doc__
Method testAdd of Class MathTest
<<< print test.__name__
testAdd
<<< print test.im_class.__name__
MathTest


I hope that helps.

- Roger




Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From bwarsaw at cnri.reston.va.us  Wed Jun 16 10:23:59 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Wed, 16 Jun 1999 10:23:59 -0400 (EDT)
Subject: python-mode question
References: <m3hfo8an9s.fsf@monsoon.swcp.com>
Message-ID: <14183.45951.429301.652913@anthem.cnri.reston.va.us>

>>>>> "AR" == Alex Rice <alex at mindlube.com> writes:

    AR> When I run python-mode in Emacs 20, I see a second Python menu
    AR> that always shows me a heirarchy of my method and class names
    AR> in the current buffer. In XEmacs, it's not there. How do I
    AR> display that?  Using version 3.90.

Newer XEmacsen support Imenu just like FSF Emacs.  With XEmacs 21 I
see a menu entry called "IM-Python" which contains the methods and
classes.  This should all work by default, but you may need to upgrade 
the XEmacs you're using.

-Barry



From mikael at pobox.com  Fri Jun 25 14:00:22 1999
From: mikael at pobox.com (Mikael Lyngvig)
Date: Fri, 25 Jun 1999 18:00:22 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
Message-ID: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net>

Hi,

Python v1.5 adopted TkInter as its standard GUI framework, even to the
extent of including TkInter in the (Intel/Windows) installation
package.  Unfortunately, TkInter appears to run sluggish [1] on all
the Intel/Windows computers I've evaluated it on (which includes
several Pentium II 400+ computers).

However, the wxPython GUI, available from
http://alldunn.com/wxPython/, does not appear to suffer from similar
performance problems, at least under Windows.  As a matter of fact, it
appears to be very easy to make commercial grade GUIs using wxPython. 

I'd like some discussion on whether or not it is a good idea to
abandon TkInter, before it gets too widely used, and instead adopt
wxPython as the primary GUI framework for Python.

Having said this, I have to say that I'm in no way a TkInter or a
wxPython expert.  I've evaluated both of them and found wxPython to be
the winner - and TkInter without a real chance of getting used by me.

Also, can anyone comment on the features available in TkInter and
wxPython - perhaps even make a feature comparison chart?

[1] By sluggish I mean: You can see, or at least sense, the window
update operations.  Windows are not drawn instantly, but in visually
detectable steps.

P.S. Let's not make a war out of this; I just want to know if there's
anybody else who feels wxPython is performing better than TkInter.


-- Mikael




From terocr at mysolution.com  Sat Jun 19 00:21:00 1999
From: terocr at mysolution.com (KP)
Date: Fri, 18 Jun 1999 23:21:00 -0500
Subject: simple text file 'parsing' question
Message-ID: <376B1AAC.19FE8BCE@mysolution.com>

Here's my dilema: a directory filled (200+) with small emails. My goal
is to strip all the headers and combine them into one file. I can read
all the files just fine and write them all to one file, but I cannot
discern how to strip the headers. The answer must be very simple, yet
I cannot see it. Can anyone give a few pointers on how to do it, our
what module might be best? Thank you.
Ken




From greg.ewing at compaq.com  Wed Jun 23 23:35:16 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Thu, 24 Jun 1999 15:35:16 +1200
Subject: Help with String Manipulation
References: <5650A1190E4FD111BC7E0000F8034D26A0EE67@HUINA> <slrn7n0dt9.gh6.wtanksle@dolphin.openprojects.net>
Message-ID: <3771A774.E0DC1BEE@compaq.com>

Or, if you'd like something a bit more creative:

def gnirts(s):
    n = len(s)
    if n < 2:
        return s
    else:
        return gnirts(s[n/2:]) + gnirts(s[:n/2])

print gnirts("a")
print gnirts("ab")
print gnirts("abc")
print gnirts("abcd")
print gnirts("It's not dead, it's just... resting!")

(Hmmm... the output from that last test case looks
a bit like Klingon...)

Greg



From hinsen at cnrs-orleans.fr  Mon Jun 28 10:09:37 1999
From: hinsen at cnrs-orleans.fr (hinsen at cnrs-orleans.fr)
Date: Mon, 28 Jun 1999 14:09:37 GMT
Subject: math domain error
Message-ID: <7l7vmp$hmi$1@nnrp1.deja.com>

SC Zhong <sai at dft.chem.cuhk.edu.hk> writes:

> I used the Numerical modules to find the square root of a matrix and
> get the "math domain error", what is it?  The matrix I have is
> 1096*1096 dimension.  When I tested the program with smaller problem,
> it seems to work.  Please advice.
>
> /var/tmp/chk_overlap-> Traceback (innermost last):
>   File ".//pop", line 412, in ?
>     popMatrix = LdinAnalysis(alphacoefficientsMatrix)
>   File ".//pop", line 345, in LdinAnalysis
>     M(overlapMatrix, eigvecsT))),eigvecsTInverse)))
> ValueError: math domain error

You don't show the code that being executed, so I'll have to guess a bit. I
suppose you transform your matrix to diagonal form, take the square root of
the eigenvalues, and then transform back, right? Then you can get a math
domain error if any eigenvalue becomes negative. First of course you must
make sure that no eigenvalue can be *really* negative (i.e. that your matrix
is positive definite). But you could still get very small negative
eigenvalues due to numerical errors in the calculation. If that's your
problem, it is safe in most cases to set those eigenvalues to zero (e.g. with
eigenvalues = Numeric.maximum(eigenvalues, 0.)). --
-----------------------------------------------------------------------------
-- Konrad Hinsen  | E-Mail: hinsen at cnrs-orleans.fr Centre de Biophysique
Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69 Rue Charles Sadron  | Fax: 
+33-2.38.63.15.17 45071 Orleans Cedex 2  | Deutsch/Esperanto/English/ France 
| Nederlands/Francais
-----------------------------------------------------------------------------
--


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From greg.ewing at compaq.com  Sun Jun 20 17:22:42 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Mon, 21 Jun 1999 09:22:42 +1200
Subject: Python source code documentation?
References: <7kdv0l$n0g$1@nnrp1.deja.com> <7kenc7$1od$1@m2.c2.telstra-mm.net.au>
Message-ID: <376D5BA2.70D845C4@compaq.com>

Mark Hammond wrote:
> 
> Pythondoc
> really doesnt want to introduce ugly mark-ups in the docstring

What about non-ugly markups? Surely it's possible to
design a form of markup which doesn't look ugly.

Greg



From gmcm at hypernet.com  Mon Jun  7 09:22:49 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 7 Jun 1999 08:22:49 -0500
Subject: bug in os.path.samefile (Python 1.5.1)?
In-Reply-To: <1318D78C9072D11195C9006094EA98A72C20EF@ocesrv>
Message-ID: <1283360446-61563238@hypernet.com>

Matej Cepl writes:
/snip/
> Now, I need to check whether the processed file is in the "current"
> directory or I should make steps to make recursive calling on
> another directory. I guess, that the best way how to do it, is by
> calling os.path.samefile(os.path.dirname(full_name),base) {see below
> for context). However, Python 1.5.1 (I cannot use 1.5.2 not having
> administrator's rights on WinNT 4.0 WK) complains by following
> report: 

No, ntpath.samefile was broken in 1.5.1 and does not exist in 1.5.2. 
Can't compare inodes of things that don't have inodes.

You don't say how the name of the "processed file" was gathered. You 
could get tripped up by case, forward vs backward slashes and the 
presence or absence of a drive letter. If it was done in a "sane" 
way, then os.path.normpath on both operands will probably let a 
string compare work properly.


- Gordon



From tom at spirit.gcrc.upenn.edu  Tue Jun  8 11:25:04 1999
From: tom at spirit.gcrc.upenn.edu (Tom Fenn)
Date: Tue, 08 Jun 1999 11:25:04 -0400
Subject: Need advice on application architecture
Message-ID: <375D35D0.CA115D77@spirit.gcrc.upenn.edu>

I have an application that is required to:

A) run on windows NT and UNIX
B) access an activeX control when running on NT

I'm running into problems with the GUI interface.  If I use Tkinter, I
can
write my GUI for both NT and UNIX no problem, but I need pythonwin to
use
the activeX control and Tkinter and pythonwin don't like each other.
So, one possiblility is to use a different tool for writing the GUI.
Does wxWindows get along with pythonwin?  Any other suggestions?

BTW, The activeX control doesn't provide any user interface elements.
Rather, it provides access to a proprietary data file format used by a
mass spectrometer we have.  Access to the data files is implemented in
shared libraries on the UNIX platform, so I don't have any problems
there.

I also tried moving the access to the  activeX control into a C++
extension for python. The idea was to use MFC to create the extension,
and during the initalization of the module, an invisible, modeless
dialog would be created that contained the ActiveX control. 
Unfortunately, the call to CDialog::Create always fails.  I'd guess this
has something to do with missing message queues, since there's no MFC
main window providing the message queue, but I don't for sure, and even
if I am right, I don't know how to fix it.  Anyone have any ideas on
what might be causing this problem and how to fix it?  Getting this
working would probably be the easiest approach for me, because then I
wouldn't have to learn a new GUI toolkit.

Tom Fenn



From johnm at magnet.com  Tue Jun 29 17:51:58 1999
From: johnm at magnet.com (John Mitchell)
Date: Tue, 29 Jun 1999 21:51:58 GMT
Subject: BOFs (Birds-Of-a-Feather) sessions at the Monterey Python
 conference
In-Reply-To: <Pine.WNT.4.04.9906291133130.243-100000@rigoletto.ski.org>
References: <Pine.WNT.4.04.9906291133130.243-100000@rigoletto.ski.org>
Message-ID: <Pine.SGI.4.05.9906291745100.5438-100000@lemur.magnet.com>

On Tue, 29 Jun 1999, David Ascher wrote:

> As you may know, O'Reilly & Associates is organizing a Python Conference
> as part of its Open Source Convention in Monterey in August. [...]
>
> I'd like to solicit discussion on what BOFs would be good to organize, and
> find some organizers.  [...]
> http://conferences.oreilly.com/bof/


I'd love to see "Python and Java BOF", and maybe one on CORBA.  If we
could hornswaggle Jim Hugunin or Barry Warsaw to talk a bit, that would be
great.  Whoa -- just found Jim's JPython slides[1].  For the
record, his talk at JavaOne was excellent -- thanks!

Basically, Java can bite me, but it has tons of libraries, everything
talks to it, and it has local fatass support.  Switching to JPython sounds
like a win/win, and I need to prove it.

CORBA I havent messed with since Fnorb used to be unstable, I'd love to
hear how it's changed.


- j


[1] http://www.jpython.org/jpython-talk-1.ppt




From gerrit.holl at pobox.com  Sun Jun 20 15:27:09 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sun, 20 Jun 1999 21:27:09 +0200
Subject: max(s) does not what I expect it to do.
In-Reply-To: <19990620202121.A22076@optiplex.palga.uucp>; from Gerrit Holl on Sun, Jun 20, 1999 at 08:21:21PM +0200
References: <19990620202121.A22076@optiplex.palga.uucp>
Message-ID: <19990620212709.A16367@optiplex.palga.uucp>

On Sun, Jun 20, 1999 at 08:21:21PM +0200, Gerrit Holl wrote:
> Date: 	Sun, 20 Jun 1999 20:21:21 +0200
> From: Gerrit Holl <gerrit.holl at pobox.com>
> To: python-list at python.org
> Subject: max(s) does not what I expect it to do.
> 
> So I did the following:
> 
> Python 1.5.2 (#1, Apr 18 1999, 00:16:12)  [GCC 2.7.2.3] on linux2
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> list = [ ['a'], ['a', 'b'], ['a', 'b', 'c'] ]
> >>> max(list)
> ['a', 'b', 'c']
> >>> list = [ ['a', 'b', 'c'], ['a', 'b'], ['a'] ]
> >>> print max(list)
> ['a', 'b', 'c']
> 

Huh?
This is right!

But if I run this script:

#!/usr/bin/python

list = [ ['a', 'b', 'c'], ['d', 'e'], ['f'] ]

print max(list)

I get:
['f']

groeten,
Gerrit.

-- 
The Dutch Linuxgames homepage:	http://linuxgames.nl.linux.org
Personal homepage:		http://www.nl.linux.org/~gerrit/

Discoverb is a python program in Dutch and English which tests the words you
learned by asking it. Homepage: http://www.nl.linux.org/~gerrit/discoverb/



From peter_carl at my-deja.com  Thu Jun 17 19:08:22 1999
From: peter_carl at my-deja.com (peter_carl at my-deja.com)
Date: Thu, 17 Jun 1999 23:08:22 GMT
Subject: Problems with CGI, genCGI and a multipart form
Message-ID: <7kbv4u$1ck$1@nnrp1.deja.com>

I'm having trouble getting cgi.FieldStorage to read data posted as
multipart/form-data.

Right now I am testing a script using stdin piped from a test file.  The
test file has everything after the header from a posted form, e.g.,:

    -----------------------------172981055621895
    Content-Disposition: form-data; name="section"

    production
    -----------------------------172981055621895
    Content-Disposition: form-data; name="title"

    This is the title

etc...

Environment variables for Content-type, Content-Length, and
REQUEST_METHOD (POST) are set correctly.

When the script is run, cgi.FieldStorage is called with stdin to
populate self.formdict, but the results are always empty...

A similar script that is not a multipart POST (just a standard POST)
works fine, where stdin is:
MessageType=Suggestion&Subject=Web+Site&SubjectOther=&Comments=&UserName
=&UserEmail=&UserTel=&UserFAX=

Am I missing an environment variable?  Is there a trick I'm missing?

I've tried including the header; not including the header; and other
variations on the test input.

Help!

Thanks in advance,

pcc


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From Robert.Meegan at wcom.com  Fri Jun  4 13:56:43 1999
From: Robert.Meegan at wcom.com (Robert Meegan)
Date: Fri, 4 Jun 1999 17:56:43 GMT
Subject: "The Python Way"
In-Reply-To: <7j8pra$55v$4@cronkite.cc.uga.edu>
References: <7j8pra$55v$4@cronkite.cc.uga.edu>
Message-ID: <Pine.A41.4.10.9906041158070.41420-100000@barney.mcit.com>

On Fri, 4 Jun 1999, Graham Matthews wrote:

> 
> Oh thank you Fredrik for posting the above. Does anyone else consider
> that the above sounds slightly paranoid? 
>

Paranoid is a little harsh, Graham. May I suggest parochial?

> Can I make a suggestion. When
> someone, for example myself, makes a suggestion for adding something
> to Python or changing something, or dare I say it "improving" Python,
> perhaps you could consider the faint possibility that they are not
> trying to "find a way to bend Python into uncomfortable positions", 
> but rather they are earnest Python users who happen to have some
> expertise/experience in the new idea being suggested, and think that
> it could not only greatly benefit Python, but also fit in rather well
> with the mystical "Python way". 
> 

One problem that exists with suggestions for improvements to Python is that
different people have different ideas about what features are essential to
the language. Most people seem to agree that objects and whitespace are
core features of Python, but others have made (more or less) well-reasoned
arguments pointing out the weaknesses inherent in each of these.

My opinion (which will get you a cup of coffee, if you ever meet me and
profess to share it) is that I would prefer to avoid syntax changes and
keyword additions. I find the following construction to be the height of
clarity:

while 1:
	[...]
	if x:
		break
	else:
		[...]
		
and I would hate to see some of the proposed solutions become part of the
language. On the other hand, I think that garbage collection would be a
wonderful feature to add. (I don't like to clean up after myself.)

The decision making process for what gets into Python seems to resemble a
parliamentary government with a strong president. The president runs the
show while the parliment debates. He can't make any major changes that are
directly in conflict with the parliamentary will, but without a concerted
and organized effort parliament doesn't have much say in the way things are
run.

The way to get significant changes incorporated into Python is to discuss
them in this august forum, until a consenus emerges (preferably including a
few key figures). Of course, implementing the change yourself and making it
available to others helps to demonstrate the superiority of your ideas.

Whether this is a good mention of defining a computer language is also
debatable, but without spinning off multiple versions of Python, I can't
foresee any other way.

> 
> In short when someone posts some suggested "improvement" to Python don't
> be so defensive. Open your mind to the possibility that the poster may
> have a point, that they are not trying to bend Python into uncomfortable
> positions, that they may know something about what they are positing 
> about, and may even be positing something that would be good for Python. 
> 

Unless there are saboteurs from the Perl community lurking amongst us, it's
probable that every suggestion offered is intended to improve Python. With
this in mind, every idea deserves a hearing. Looking at the archives shows
that if an idea has any merit at all a spirited discuss will ensue. The
ideas that fail to attract much attention (either pro or con) are either
self-evidently crucial or much more likely not of interest to anyone but
the author.

To choose an (admittedly self-serving) example, the issue of garbage
collection is obviously very important to a lot of people. In the last
year, hardly a month has passed without a motion being brought to the
floor. Perhaps what should be done at this point is for some of the more
vocal and knowledgable members (hopefully not a mutually exclusive
requirement) to create a garbage collection SIG. In that environment a
"best choice" method could be selected and implemented as a patch against
1.5.2. Once released, people could test it and determine whether the
advantages outweigh the disadvantages in the real world. Under those
circumstances it might be possible to decide whether GC belongs as a core
piece of Python 2.0.

The important thing is neither to quash dissent nor to debate the number of
angels that can dance on the head of an object, but to freely share ideas
in the hope of improving both Python and our own pool of knowledge.

This learned opinion is, of course, not worth the bits it's written on.

--- Robert

--------------------------------
Robert Meegan
MCIWorldCom - Cedar Rapids, Iowa
319.375.2416




From radbit at my-deja.com  Tue Jun 22 14:47:31 1999
From: radbit at my-deja.com (radbit at my-deja.com)
Date: Tue, 22 Jun 1999 18:47:31 GMT
Subject: Build the Python library into the binary
References: <376F9518.D3252CEE@interet.com>
Message-ID: <7kolnv$gp6$1@nnrp1.deja.com>

In article <376F9518.D3252CEE at interet.com>,
  "James C. Ahlstrom" <jim at interet.com> wrote:
> Now that Python 1.5.2 is here I need to upgrade from
> 1.5.  We use Python internally and also ship Python
> applications.  Our environment is mixed PC's, Suns and
> linux boxes.  Our customers all have PC's.  I always
> find upgrades difficult, especially at customer sites since
> we have no access, and have to depend on our Wise install
> script.
>
> I keep thinking how much easier this would be if the
> Python library were built into the binary.  I mean that
> on PC's, the library were part of python15.dll for example.
> No more worrys about a bad PYTHONPATH, the registry, my
> 1.5.2 binary using 1.5 libraries, the different ways all
> this works on different platforms, etc.  You just replace
> python15.dll and you are at 1.5.2.  Simple.
>
> So I wrote some code.  I don't consider it finished but
> it works.  It makes the Python library part of the binary.
>
Hi

did you have a look at mxCGIPython at
http://starship.skyport.net/~lemburg/mxCGIPython.html ?
This is a single executable not shared library with the
standard Python library embedded.

Cheers
--
Florent Heyworth


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From nathan at islanddata.com  Wed Jun 30 11:24:51 1999
From: nathan at islanddata.com (Nathan Clegg)
Date: Wed, 30 Jun 1999 08:24:51 -0700 (PDT)
Subject: printing with print
In-Reply-To: <011a01bec2d3$c215a900$f29b12c2@secret.pythonware.com>
Message-ID: <XFMail.990630082451.nathan@islanddata.com>

> Angus MacKay wrote:
>> foo = 2
>> print "hi" + foo

How about:

print "hi" + `foo`


----------------------------------
Nathan Clegg
 nathan at islanddata.com





From matomira at iname.com  Thu Jun 10 13:44:10 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Thu, 10 Jun 1999 19:44:10 +0200
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <wk909syqr4.fsf@ifi.uio.no>
Message-ID: <375FF969.C879B4CD@iname.com>


Lars Marius Garshol wrote:

> * Fernando Mato Mira
> |
> | I'm trying to collect a list of all the _current_ issues making tcl
> | a bad language choice.
>
> You can find three interesting articles on this at
>
> <URL: http://itrc.uwaterloo.ca/~papresco/tcl/>

Thanks. But the Sah article is not accessible,
and his homepage makes no reference to it.

Regarding the Stallman article, it's a quite vague,
and it's not sure one can extrapolate from their TECO experience
to Tcl.

About the Connoly one, the namespace issue seems to
be solved by [incr Tcl]. But what about that "interpreted
multiple times"? Is it just an efficiency issue, or a semantic
nightmare (maybe both)?

What about dynamic scoping?

BTW, I just found:

http://www.vanderburg.org/~glv/Tcl/war/
http://www.vanderburg.org/~glv/Tcl/war2/







From olberger at club-internet.fr  Tue Jun  1 19:14:01 1999
From: olberger at club-internet.fr (Olivier Berger)
Date: Tue, 01 Jun 1999 23:14:01 +0000
Subject: A tkinter kick-off
References: <37544a32.0@news1.mcmail.com>
Message-ID: <37546939.E9CE0264@club-internet.fr>

Mark Butterworth wrote:
> 
> I've been using python for a few months now (I've actually managed to
> persuade my company that freeware is a viable tool) and I've just started
> looking at tkinter.  However,  my first little project doesn't  require an
> interactive window or dialog box.  What I do require is a window which
> displays the result's of my background processing.  The processing will
> simply be the compression and network transfer of some files.  I'm not
> looking for a full solution however, I just need a help starting off.  I can
> create a window containing a listbox alright, I just need to know how to
> process in the background and display results within the listbox (maybe a
> cancel button would also be useful).
> 
> Any help appreciated!?!

Look at the FAQ ... it explains how registering a "callback" that can be
linked to the processing of characters coming on an input to your
ptogram.

My suggestion is then to have a finite state machine (simple to code if
you use disctionnaries, lists and eval() ...) to interprete these
inputs, then rendering them as you like in the window...

I'm myself writing such a tool, so this is only my actual experiments
results... Maybe you can find allmost complete application frmaework
elsewhere for such a "simple" tool...

Olivier.

-- 
  Olivier Berger - olberger at club-internet.fr - membre de PARINUX 
       Paris Linux User Group : http://www.parinux.org
   Vive Linux (www.linux.org) - Vive python (www.python.org)



From tre17 at cosc.canterbury.ac.nz  Mon Jun 14 18:43:25 1999
From: tre17 at cosc.canterbury.ac.nz (Timothy R Evans)
Date: 15 Jun 1999 10:43:25 +1200
Subject: Problem with Lists & Index command
References: <rNHvjEdhm5Pp-pn2-zOTjsVjHebb4@d185d18f8.rochester.rr.com> <37618424.8B8F3116@mindspring.com> <rNHvjEdhm5Pp-pn2-XcIGYOCMcDxI@d185d18f8.rochester.rr.com> <37650BEB.564B2B20@mindspring.com>
Message-ID: <c757lp68ktu.fsf@pc142.cosc.canterbury.ac.nz>

Doug Hellmann <doughellmann at mindspring.com> writes:

> 
> wordcounts = {}
> for word in document.words():
>   try:
>     wordcounts[word] = wordcounts[word] + 1
>   except KeyError:
>     wordcounts[word] = 1

Alternaively, use the `get' method on the dictionary, which will
return a default value if the key is not found.  In this case you want 
a default value of 0, so the lines become:

wordcounts = {}
for word in documents.words():
    wordcounts[word] = wordcounts.get(word,0) + 1

Its not actually any different, just smaller.

--
Tim Evans




From doctorwes at mindspring.com  Sat Jun  5 02:10:12 1999
From: doctorwes at mindspring.com (Wesley Phoa)
Date: Fri, 4 Jun 1999 23:10:12 -0700
Subject: calldll and floats/doubles (newbie q)
Message-ID: <7jaih5$gk1$1@nntp8.atl.mindspring.net>

I'd like to write a small library of option pricing functions in C, and test
it from Python using calldll or windll. Does anyone know how to invoke
functions that take float or double arguments and return float or double
results?

Wesley Phoa.





From thstr at serop.abb.se  Wed Jun  9 10:37:42 1999
From: thstr at serop.abb.se (Thomas S. Strinnhed)
Date: Wed, 09 Jun 1999 16:37:42 +0200
Subject: Python/C API and setattr
References: <Pine.SUN.3.95-heb-2.07.990609171623.11656A-100000@sunset.ma.huji.ac.il>
Message-ID: <375E7C36.B323D091@serop.abb.se>

Hi again

Gee, you guys are so fast :-)

Moshe Zadka wrote:
> > Could there be something about this line;
> >       self->value = (int)PyInt_AsLong(v);
> > (self->value is an int and v is PyObject*)
> 
> Err......defensive programing says: check it is a Python int before
> converting (there's a PyInt_Check() function somewhere -- check with
> the docs).
Ok, you're right about that one, but PyInt_AsLong() *does* deliver and
int and the value *is* correct. 

> Are you sure there shouldn't be a Py_ParseTuple(..) function somewhere
> too?
No I'm not, but my Objects\xxobject.c inspiration doesn't use one
so I thought I'd get away without one.
 
> What exactly "doesn't fly?" (i.e., what error do you get?)

>>> c.value = 23
Traceback (innermost last):
  File "<stdin>", line 1, in ?
SystemError: error return without exception set
>>> c.value
23

This is the error I get and according to Gordon McMillan this
is (just like you pointed out) related to the fact that C
returns NULL to Python. But the value is, as you can see
from the printout, just what I wanted.

And I now always return a Py_None (carefully incref'd),
but the error remains :-/

Many Thanks
 -- Thomas S. Strinnhed, thstr at serop.abb.se



From fatjim at home.com  Wed Jun 23 13:02:24 1999
From: fatjim at home.com (Jim Meier)
Date: Wed, 23 Jun 1999 17:02:24 GMT
Subject: How to obtain an instance's name at runtime?
References: <7kqpu8$7u1$1@nnrp1.deja.com>
Message-ID: <3771127C.E40F9BF3@home.com>


"Dinu C. Gherman" wrote:

> This is a nasty question, perhaps, so I appologize in
> advance if its stupidity level is far above the accepted
> average in this newsgroup...
>
> I would like to do the following:
>
> >>> class C:
> ...    pass
> ...
> >>>
> >>> c = C()
> >>> C.__class__.__name__ # ok
> 'C'
> >>> c.__name__           # not ok
> Traceback (innermost last):
>   File "<interactive input>", line 0, in ?
> AttributeError: __name__
> >>>
>
> Question: How to obtain an instance's name at runtime?
> That is I'd like to see this happen (one was or the other):
>
> >>> c.__name__
> 'c'
>
> I've checked the reference manual, but it's silent about
> this topic or I'm too blind...
>
> Fearing-there's-no-such-solution'ly,
>
> Dinu

Well, there sort of is a solution, but I don't think you're going to
like it. Incidentally, why are you trying to discover this? I'm not a
super-meister-coder, but I can't imagine a situation where knowing this
would be useful.

There are some problems in the way of what you want. First of all, there
is not actual name for the instance. The instance is a nameless object.
It only gets a 'name' by being referenced by a variable. And that name
can change. For example,

c=C()
d=c

At this point there are two references to the instance; which one is
it's 'name'? The same thing happens during function calls. For example,

def func(f):
    print 'this is my f:' + str(f)

c=C()
func(c)

The function gives it it's own new name. So you already have given the
instance a name!

But if you are just running in the interactive loop, and understand the
above problems, you could try this:

names=[]
for i in locals().items():
    if id(i[1]) == id(var):
        names.append(id[0])

Which should give you a list of the names of all the references to the
object IN THE CURRENT SCOPE.




-Jim




From edcjones at erols.com  Wed Jun  2 11:20:29 1999
From: edcjones at erols.com (Edward C. Jones)
Date: Wed, 02 Jun 1999 11:20:29 -0400
Subject: When I run a module written in C, core dumps at Python exit.
References: <37529504.669DE3D8@erols.com> <000b01beac96$1262f6a0$d89e2299@tim> <19990601220516.A608@chronis.pobox.com>
Message-ID: <37554BBD.9F33B8FD@erols.com>

scott cotton wrote:

> On Tue, Jun 01, 1999 at 09:20:16PM -0400, Tim Peters wrote:
> | [Edward C. Jones]
> | > I wrote an extension module written in C. It seems to work OK.
> | > But core dumps when exiting Python. When I run the code from the
> | > interpreter, Python works fine until I try to exit with ^D. Then
> | > there is a core dump. What are the most common causes of such
> | > problems?
> |
> | Buggy extension modules written in C.
> |
> | suspecting-that's-obvious-ly y'rs  - tim
>
> More specifically <0.314 ;>, since python core dumps on exit,
> it is most likely a deallocation function that is freeing
> memory that it shouldn't.  Of course, that may be caused by
> almost anything in the extension module.
>
> scott

It turned out that the program was writing over the end of an array.
This usually dumps core well before Python exits but this time took
longer.

Thanks,
  Ed Jones





From mark at _spamspamlovelyspam_btweng.krakow.pl  Sat Jun 19 05:38:03 1999
From: mark at _spamspamlovelyspam_btweng.krakow.pl (MK)
Date: Sat, 19 Jun 1999 09:38:03 GMT
Subject: Green or Red for stderr under IDLE?
References: <7kea0m$rig$1@nnrp1.deja.com>
Message-ID: <376b5fcf.7493987@news.tpnet.pl>

On Fri, 18 Jun 1999 20:26:11 GMT, Sam Schulenburg <sams at quinta.com> wrote:

>I notice that stderr for the IDLE environmnet is defaulted to green for
>it's color. Is this a standard anywhere or should it be defaulted to
>red?

>I find that on all the releases that I have used, I have to reset its
>value to red as I use this feature to print out error messages on the
>status of the scripts that I run.

IDLE | Help | Help:


Tips:
	To change the font on Windows, open EditorWindow.py and change
	    text['font'] = ("verdana", 8)
	to, e.g.,
	    text['font'] = ("courier new", 10)

	To change the Python syntax colors, edit the tagdefs table in
	ColorDelegator.py; to change the shell colors, edit the tagdefs
	table in PyShell.py.

Pyshell.py:

    tagdefs.update({
        ##"stdin":   {"background": "yellow"},
        "stdout":  {"foreground": "blue"},
        "stderr":  {"foreground": "#007700"},
        "console": {"foreground": "#770000"},
        "ERROR":   {"background": "#FF7777"},
        None:      {"foreground": "purple"}, # default
    })





MK


--------------------------------------------------
Reality is something that does not disappear after 
you cease believing in it - VALIS, Philip K. Dick
--------------------------------------------------

Delete _spamspamlovelyspam_ from address to email me

postmaster at 127.0.0.1
root at 127.0.0.1
webmaster at 127.0.0.1






From gmcm at hypernet.com  Thu Jun  3 11:53:41 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 3 Jun 1999 10:53:41 -0500
Subject: mailing list to news gateway
In-Reply-To: <7j4asp$v4a$1@nnrp1.deja.com>
Message-ID: <1283697026-41316958@hypernet.com>

Jeremy writes:

> I have noticed what appear to be problems with the gateway between
> the mailing list (python-list at python.org) and the newsgroup -- and
> more specifically with the view of the newsgroup at dejanews.

> So two questions: First, has anyone else noticed problems with
> messages sent to the mailing list not appearing in the newsgroup? 
> Trying to establish the scope of the problem.  Perhaps it's just
> weirdness related to my messages.  Second, has anyone seen posts
> from me on the newsgroup lately -- say in the Python 2.0 thread? 
> (I'm sure everybody keeps track of all the witty and insightful
> comments I make <1.0 wink>.)

I mooned everybody, too, and got no response. Hard to sustain a good 
flame war with just the mailies.

will-moon-for-food-ly y'rs

- Gordon



From graham at sloth.math.uga.edu  Thu Jun  3 15:06:29 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 3 Jun 1999 19:06:29 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net> <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net> <7j34l3$hbj$1@nnrp1.deja.com> <7j66b7$il5$2@cronkite.cc.uga.edu> <3756C693.2C910B29@usa.net>
Message-ID: <7j6jnl$58f$3@cronkite.cc.uga.edu>

Graham Matthews wrote:
> You don't need a new method call -- all you need is a mark and sweep
> collector with refcounts. I really am amazed that this has generated so
> much traffic. It's very simple. You have ref counts just like you do
> now in Python. When an objects refcount goes to 0 its __del__ method
> is called. On top of this you add a mark and sweep collector to handle
> circular references. During a sweep if it finds dead objects it calls
> the __del__ method for that object.
Salvador =?iso-8859-1?Q?Fandi=F1o?= (fandino at usa.net) wrote:
: GC usually deletes objects in random order which is not a very good
: behavior.

I am sorry but this comment is all myth. Think about what I wrote
above. Imagine Joe average Python program P. If P contains no
circular references then *under the scheme I propose above*
P's behaviour will be EXACTLY the same as it is in current Python
(no circular refs means the mark sweep collector will never find
any dead wood and hence never need to call any __del__ methods -- all
the __del__ calls will be via the ref count mechanism as it is now!). 
The calls to the __del__ methods will happen EXACTLY as they do now. If 
P contains circular refs then current python will never reclaim the 
resources associated with these refs. *Under the scheme I propose above* 
these resources will be reclaimed, *when* the mark sweep collector runs. 
Certainly one has no control over when exactly that will be, but at least 
the resources will be reclaimed, unlike under the current Python where
they will never be reclaimed.

Why is it so hard for people to see that adding a mark sweep collector
*ON TOP OF* the existing ref count scheme essentially changes nothing.
All the current ref count behaviour that people love will stay. The
only additional behaviour will be that circular refs are mopped up
correctly.

graham

-- 
          The fool escaped from paradise
        will look over his shoulder and cry
          Sit and chew on daffodils and
             struggle to answer why?



From " HPeter.Friedrichs" at alliedsignal.com  Thu Jun 24 18:16:19 1999
From: " HPeter.Friedrichs" at alliedsignal.com (H. P. Friedrichs)
Date: Thu, 24 Jun 1999 15:16:19 -0700
Subject: How do can you make a variable static?
Message-ID: <3772AE33.AA764DFA@alliedsignal.com>

Hello:
I hope somebody can help me with this problem (python newbie, <<week)

I have a dictionary called SymHanDict. It's entries are comprised of a
"symbol handle" number (the key), and a numeric value. A function foo()
resides in another module. When called, it's supposed to look at
SymHanDict, and create a new entry comprised of an incremented handle
number (the highest handle number in the existing dictionary plus one),
and a default data value of 0.

I keep getting name errors on SymHanDict, even though I've specified it
as global. I've seen the FAQ and thumbed through a borrowed copy of
Programming Python (M. Lutz). I don't see anything that allows me to
declare variables as static, and I don't understand the explanation of
globals in python.

Can anyone clarify? What's the workaround? For a variety of reasons I
won't go into right now, I really can't pass the dictionary as an
argument.

Thanks

Pete


**** the following contained in main module test.py
*****************************
from foo import foo

SymHanDict={0:0}

foo()                            #each call to foo() should add items to
the dictionary
foo()
foo()

*** the following contained in module foo.py ************************
def foo():
    global SymHanDict

    handlelist=SymHanDict.keys()          # get the list of existing
handles
    handlelist.sort()                     # sort the list, then

    handle=handlelist[len(handlelist)-1]        # find largest handle
    handle=handle+1                         # generate new handle, and
    SymHanDict[handle]=0                  # add it to dictionary along
with
                                          # a default value of zero.

    print SymHanDict




--

-----------------------------------------------
| The opinions expressed here are my own,     |
| and not necessarily those of my employer.   |
-----------------------------------------------





From gregm at iname.com  Fri Jun 25 02:29:38 1999
From: gregm at iname.com (Greg McFarlane)
Date: Fri, 25 Jun 1999 06:29:38 GMT
Subject: Tkinter: why does an empty frame not shrink?
Message-ID: <19990625162938.39336@nms.otc.com.au>

If you have several child widgets packed in a frame and you unpack the
children one by one, the frame will gradually get smaller, except when
the last child is unpacked, in which case the frame does not resize. 
Does anyone know why the frame does not shrink to 0x0 when you unpack
the last child?  Any soluton?

Example script:
============================================================
import Tkinter

root = Tkinter.Tk()
root.geometry('400x400')

frame = Tkinter.Frame(background = 'red')
frame.pack(fill = 'x')

for i in range(5):
    button = Tkinter.Button(frame, text = 'Unpack me')
    button.configure(command = button.pack_forget)
    button.pack()

root.mainloop()
============================================================

-- 
Greg McFarlane:    INMS Telstra Australia (gregm at iname.com)
Today's forecast:  Sunny, with occasional cloudy periods and a chance
		   of precipitation in some areas.



From fandino at usa.net  Fri Jun  4 06:19:12 1999
From: fandino at usa.net (Salvador =?iso-8859-1?Q?Fandi=F1o?=)
Date: Fri, 04 Jun 1999 12:19:12 +0200
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net> <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net> <7j34l3$hbj$1@nnrp1.deja.com> <7j66b7$il5$2@cronkite.cc.uga.edu> <3756C693.2C910B29@usa.net> <7j6jnl$58f$3@cronkite.cc.uga.edu>
Message-ID: <3757A820.CE3F195F@usa.net>

Graham Matthews wrote:

> Salvador =?iso-8859-1?Q?Fandi=F1o?= (fandino at usa.net) wrote:
> : GC usually deletes objects in random order which is not a very good
> : behavior.
> 
> I am sorry but this comment is all myth.
No, it's a real problem I had in my current project. You can't delete
complex data structures just deleting component objects in random order.

> Why is it so hard for people to see that adding a mark sweep collector
> *ON TOP OF* the existing ref count scheme essentially changes nothing.
> All the current ref count behaviour that people love will stay. The
> only additional behaviour will be that circular refs are mopped up
> correctly.
I am just suggesting a way to improve the mark sweet collector you
talked about. Calling the __del__ method could be the default behavior
for it, but a new method hook could add better functionality...

BTW, usual __del__ methods free resources like sockets, files, windows,
etc. but they don't break circular references so calling __del__ will
not free any object at all. 

 -Salva.



From jhouchin at texoma.net  Fri Jun 18 09:24:07 1999
From: jhouchin at texoma.net (Jimmie Houchin)
Date: Fri, 18 Jun 1999 13:24:07 GMT
Subject: IPwP books available, 99cents
Message-ID: <376a479b.131911984@news.texoma.net>

Hello,

I was visiting a bookstore in Plano, TX yesterday and saw two new
copies of Internet Programming with Python on their shelves for $.99,
yes 99 cents.

The store is called 75 Off and is at Spring Creek and Avenue K. in the
Garden Ridge Shopping Center.

The books were in the very back of the store on their most discounted
shelves.

For directions to the store or other questions just email me.

Jimmie Houchin



From phd at emerald.netskate.ru  Wed Jun 23 06:47:28 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Wed, 23 Jun 1999 14:47:28 +0400 (MSD)
Subject: Using C to execute Python
In-Reply-To: <7kqcds$4nu$1@kopp.stud.ntnu.no>
Message-ID: <Pine.LNX.4.04.9906231445140.8942-100000@emerald.netskate.ru>

Hi!

On Wed, 23 Jun 1999, Thomas R?berg wrote:
> I have a C program and I want to call certain Python methods / functions
> passing quite a few parameters.
> How can I do this, in general fashion?

   You need to embed python interpreter into your program. That is: you
link your program with libpython.a and call some functions (initilize, evel
string, etc.)
   More details in the standard docs on Extending and Embedding Python.

> I would have to include python.h, I reckon - where can I find this, I've
> been searching all over for it (is it just me???)

   It is in the source distribution.

> I'm no expert, far from, in Python...
> 
> Thomas
> email: thomar at ifi.ntnu.no

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From robin at jessikat.demon.co.uk  Wed Jun  2 05:12:25 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Wed, 2 Jun 1999 10:12:25 +0100
Subject: need help checking for stack overflow in Win32
References: <5lk8to3su3.fsf@eric.cnri.reston.va.us>
 <7j1pcc$256$1@m2.c2.telstra-mm.net.au>
Message-ID: <kSso3CA5VPV3EwYB@jessikat.demon.co.uk>

In article <7j1pcc$256$1 at m2.c2.telstra-mm.net.au>, Mark Hammond
<MHammond at skippinet.com.au> writes
>Actually, I was about to post this yesterday, and to prove my point I wrote:
>
>def die():
>    die()
>
>Expecting to see the crash.  I was then going to go into a rant about how
>win32api.Apply would actually let you catch this (and any other Win32)
>exception as a temporary solution.
>
>But it worked correctly - ie, Python raised a RuntimeError: Max Recursion.
>
...
I just tried this with my bog standard 1.5.2 under win95 osr2 and it
faults to the system. Strange :)
-- 
Robin Becker



From pduffin at mailserver.hursley.ibm.com  Thu Jun 10 11:17:23 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Thu, 10 Jun 1999 16:17:23 +0100
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com>
Message-ID: <375FD703.C6D@mailserver.hursley.ibm.com>

Fernando Mato Mira wrote:
> 
> Hello,
> 
>   I'm trying to collect a list of all the _current_ issues making tcl
> a bad language choice. I'd like concrete examples, and not just
> vague assertions that cannot be put porward honestly. I think
> that as is usual in this newsgroups, there's going to be people that
> actually know what they are talking about (and even better, experienced)
> 
> when referring to a language that is not their favorite.
>   If you have references to articles of the sort, that will be useful,
> too.
>   Note that I've not listed comp.lang.tcl for increased productivity,
> and to avoid starting a flame war. Hopefully some misconceptions
> will be filtered down here, and _then_, the summary can be presented
> in that newsgroup for their defense.
> 

I will start by saying that your choice of title and opening statement
was bad and certainly not conducive to a reasoned debate about the
pros and cons of Tcl.

What you are going to end up with from this sort of article is a list of
outdated problems from people who have not touched it for years, plus
a lot of myths and possibly mingled with a few nuggets of useful 
information.

Whether the choice of language is a good or bad idea depends entirely on
the context. i.e. environment and job to be done.

You cannot avoid taking into account the context even if you only look
at the language from a semantic / syntactic view point.

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From ivanlan at callware.com  Wed Jun  9 10:26:12 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Wed, 9 Jun 1999 14:26:12 GMT
Subject: Looking for for some ideas for argument processing
References: <7jcdhc$htd$1@nnrp1.deja.com> <375BCEA1.F5B59623@mindspring.com>
Message-ID: <375E7984.85CD9E30@callware.com>

Hi All--

(I've been off email for a week, just getting it back today, so forgive
me for posting replies to letters that might be out of date. ...)

Doug Hellmann wrote:
> 
> cmedcoff at my-deja.com wrote:
> >
> > I'm writing a script with is going to make heavy use of script
> > arguments.  I'd like to hear some ideas on some ways to support this.
> >
> > One idea I've had so far is to write a class which is a dictionary
> > of "directives" to function objects, where a "directive" might be '-
> > c', '/f', or whatever.  This seems to work well if all arguments are of
> > this simple form.  However I would also like to support arguments
> > like "python myscript.py /a /f /x "bip" "bop", /g"; the point here
> > being that "directives" may take arguments.  This kind of messes up the
> > simple loop you'd otherwise use to process arguments.
> >
> >         a = BuildArgumentProcessor()
> >
> >         # process arguments
> >         for x in sys.argv[1:]:
> >             opt = x[1:2]
> >             optArg = x[2:]
> >             a.Process(opt, optArg)
> >
> > This would work if somehow inside the for loop one could "increment
> > this list iterator more that once" (sorry I'm coming from a C++
> > background).  I don't see anyway of doing this.  Is it possible?  Any
> > other ideas?
> >

Check out
	
	ftp://www.pauahtun.org/pub/getargspy.zip

I think you may find it gets pretty close to what you want.  Arguments
provided on the command-line can be ints, strings, lists, tuples; IIRC
it can do dictionaries, too.  Documentation is sketchy, so you'll have
to read the code & look at the docstrings, but if it can't do what you
want I may have time to fix it so it will.  Some of these days, in my
copious free time, I will HTML-ize the documentation;-)  The short
version is that you make a dictionary out of the command-line args, and
by the time you instantiate the dictionary describing the args, most
processing has been done.  I.e., if one of the args is a filename ("-f
foofile"), when you access the "f" key in the dict, the value of it is
the opened file.  And so on.

<yes-you-can-change-the-leadin-char-from-dash-to-/-or-whatever>-ly y'rs,
Ivan

----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From holger at phoenix-edv.netzservice.de  Fri Jun 25 09:38:52 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Fri, 25 Jun 1999 13:38:52 GMT
Subject: version of python?
Message-ID: <3773866C.9D29C071@phoenix-edv.netzservice.de>

Hi!

How could I get the version-number of the installed python-interpreter?

I need this because I got an error under Python 1.51 (on my WinNT-test-machine)
when running a script for linking files:

def createLink(filePath, linkPath, args, descr, wrkngDir):
        "creates a link to a file"
        import pythoncom
        from win32com.shell import shell
        #get the shell interface
        sh = pythoncom.CoCreateInstance(shell.CLSID_ShellLink, None,
pythoncom.CLSCTX_INPROC_SERVER, shell.IID_IShellLink)
        #get an IPersist interface
        persist = sh.QueryInterface(pythoncom.IID_IPersistFile)
        #set the data
        sh.SetPath( filePath )
        sh.SetDescription(descr)
        sh.SetWorkingDirectory(wrkngDir)
        sh.SetArguments(args)
        #save the link itself.
        persist.Save(linkPath, 1)  


  File "C:\myPython\phoenix\installation.py", line 886, in ?
    startInstallation()
  File "C:\myPython\phoenix\installation.py", line 839, in startInstallation
    ins._runInstallation()
  File "C:\myPython\phoenix\installation.py", line 615, in _runInstallation
    self._linkFiles(anOrder)
  File "C:\myPython\phoenix\installation.py", line 808, in _linkFiles
    createLink(pFilePath, pLinkPath, pArguments , pDescription, pWrkngDir)
  File "C:\myPython\phoenix\installation.py", line 334, in createLink
    sh = pythoncom.CoCreateInstance(shell.CLSID_ShellLink, None, pythoncom.CLSCT
X_INPROC_SERVER, shell.IID_IShellLink)
AttributeError: CLSID_ShellLink


I tested with Python 1.52 on my working-place: Everything OK!
But, I wonder a little bit about that. win32com-modules havn't changed
since "Sat Dec 05 19:30:34 1998" as mentioned in win32com.txt (this one 
I found on my harddisk...;-)))
And I'm fortunaly sure that this version of 'Python 1.5 combined Win32 extensions'
is the same as on the other computer.

So what? How could this be?
(sorry, couldn't have a look source of win32com, because it's not within
the windows-version)

ciao,
Holger



From bwarsaw at cnri.reston.va.us  Tue Jun 29 20:52:46 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Tue, 29 Jun 1999 20:52:46 -0400 (EDT)
Subject: A Certifiable Very Cool Trick
Message-ID: <14201.27231.1003.466027@anthem.cnri.reston.va.us>

Just a quick note because this is way too cool not to share.  When
python.org came back up today, it almost immediately got cpu slammed
(and has been for at least the last 3 hours).  I know what's going on
-- it has to do with performance problems in Mailman that in this case 
was triggered by a flood of incoming messages to all the mailing
lists.  This Mailman problem has been seen by a few other users, but
it's dang hard to track down.  Guido, Jeremy, and I got to talking
about what tools we'd like to have to help debug this problem.

My model was gdb's attach feature where you can attach to, stop and
trace through a running external C program.  I would love to be able
to do the same thing at the Python level.  We started talking about
remote pdb's with threads listening on sockets, etc. (still cool ideas
but more work).  I mentioned that I'd love to be able to just get a
traceback of where the program is currently stopped.  Even if it hoses
the interpreter afterwards, it would still be incredibly useful.
Guido and I spent some fun after hours hacking time and came up with
the following trick.

[This is described in terms of gdb, but you might be able to do
something similar with dbx or other Unix and non-Unix debuggers, I
dunno]

First, get the pid of the running Python process.  Start up gdb on the
python executable and at the gdb prompt type "attach XXX" where XXX is
the pid.  Now, at the gdb prompt, type the following:

(gdb) call PyRun_SimpleString("import sys, traceback; sys.stderr=open('/tmp/tb','w',0); traceback.print_stack()")

Sitting in /tmp/tb will be the stack trace of where the Python program
was when you stopped it!  There's reason to believe this will not
always work, and it could seriously confuse your interpreter depending
on where it stops.  But every time I've tried it it /does/ work, and
you can even detach the program and let it continue on!

Chris Tismer, somehow I think your stackless Python might either make
this trick harder or much easier :)

makes-this-geek-quiver-with-joy-but-my-mom-would-just-scratch-her-head-ly y'rs,
-Barry



From tim_one at email.msn.com  Mon Jun 28 00:34:48 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 28 Jun 1999 00:34:48 -0400
Subject: ANN: Stackless Python 0.2
In-Reply-To: <7l6gc8$du4$1@news.tamu.edu>
Message-ID: <000c01bec11f$8dfbf800$e19e2299@tim>

[Corran Webster, commenting on Christian Tismer's "stackless Python"]
> ...
> I wonder whether there's the potential here for more than coroutines
> and to write an implementation of threading within Python.

It's been mentioned a few times.  So far Guido isn't keen on any of this,
but he's been known to buckle after a few short years of incessant whining
<wink>.  BTW, a coroutine pretty much *is* a thread in a time-sliced world,
just lacking a "transfer" function that invokes implicitly whenever it
bloody well feels like it <wink>.

> Each thread would presumably need its own Python stack,

Nope!  It's a beauty of the implementation already that each code object
knows exactly how much "Python stack space" it needs, and (just) that much
is allocated directly into the code object's runtime frame object.  IOW,
there isn't "a Python stack" as such, so there's nothing to change here --
the stack is implicit in the way frames link up to each other.

> and a queueing and locking system would need to be added somehow,

Yes, and that would require some changes to the core.  Sounds doable,
anyway.

> but because the Python and C stacks are no longer intertangled, switching
> between threads should be easy (as opposed to impossible <wink>).

I think Christian's approach moves lots of crazy ideas from impossible to
plausible.

> This wouldn't be quite as flexible as the current implimentations of
> threads at the C level - C extensions would be called in a single block
> with no way to swap threads until they return.

That's mostly true today too:  Python threads run serially now, one at a
time, and if a thread calling out to C doesn't release the global lock no
other thread will run until it returns.

> On the other hand, this would allow every platform to have some sort
> of threading available, which would be a boon.

It's even quite possible that "fake threads" would-- for programs that
aren't doing true multiprocessing --run significantly more efficiently than
today's scheme of creating OS-level threads and then choking them into
taking strict turns; e.g., because "the Python stack" grows only to the
exact size it needs, there's almost certainly much less memory overhead that
way than by letting the OS allocate a mostly unused Mb (whatever) to each
real thread's stack.  This opens the possibility to create thousands &
thousands of fake threads.

> Unfortunately I'm not familiar enough with threads to go out there and
> implement it right away, but I thought I'd at least raise it as a
> possibility and see what people think and what the pros and cons are.

It's sure worth pondering!

write-some-code-anyway-&-get-famous-one-way-or-another<wink>-ly y'rs  - tim





From zessin at my-deja.com  Sat Jun 19 11:10:45 1999
From: zessin at my-deja.com (Uwe Zessin)
Date: Sat, 19 Jun 1999 15:10:45 GMT
Subject: Installing ODBC on PythonWin
References: <7k4grk$b19$1@nnrp1.deja.com> <7ke9kq$3c2$1@holly.csv.warwick.ac.uk>
Message-ID: <7kgbtg$sh5$1@nnrp1.deja.com>

In article <7ke9kq$3c2$1 at holly.csv.warwick.ac.uk>,
  esuzm at primrose.csv.warwick.ac.uk (Andrew Clover) wrote:
> eeghead84 at yahoo.com wrote:
>
> > It is written that the ODBC module is included in the pythonwin
> > distribution but >>>import ODBC says otherwise. I've d/e'd
> > (downloaded/executed) Win32, but it died and said that it was
> > looking for version 1.4!
>
>  If you download the latest Python and Win32 from
> http://www.python.org/download/download_windows.html it should be
> fine. You'll want to write "import dbi, odbc" though.
>
>  Now for my problems.
>
>  I've got an Access database accessed by Python scripts under CGI.
> However, sometimes, calls to odbc.odbc(dsn) hang up and never return.
> (Eventually the web server times out and kills the process.) This
> seems to happen randomly, but increasingly. Does anyone have any
> idea why? A fault with the odbc module, or Access, or what?

Oh, that is interesting. I had a similar problem with the driver
to a remote Navision database and was suspecting it was limited
to that configuration - I do not use ODBC very much.
Inserting a delay after the import seemed to "fix" my problem:

import dbi, odbc
import time
time.sleep (3)

odbc_nav = odbc.odbc (dsn_nav)

--
Uwe Zessin


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From quango at watervalley.net  Mon Jun  7 02:58:15 1999
From: quango at watervalley.net (Chris Lawrence)
Date: Mon, 7 Jun 1999 06:58:15 GMT
Subject: os.path.realpath() on POSIX
Message-ID: <19990607015814.A18254@quango2.watervalley.net>

This patch implements a realpath() function (cfr GNU libc) that gives
the canonical location of a file on the filesystem, resolving any
symlinks in the middle.

For example, /usr/bin/X11/xterm is often actually
/usr/X11R6/bin/xterm, even though xterm itself is not a symbolic link
(the link is usually /usr/bin/X11 -> /usr/X11R6/bin or ../../X11R6/bin).

This sort of thing is useful when you're trying to track down what
package includes a certain file, etc.

This patch also gives abspath() a docstring.

[os.path.realpath() should probably act like os.path.abspath() on most
non-POSIX systems.]

--- /usr/lib/python1.5/posixpath.py	Fri May 21 10:26:23 1999
+++ posixpath.py	Mon Jun  7 01:48:16 1999
@@ -371,6 +371,28 @@
 
 # Return an absolute path.
 def abspath(path):
+    """Return an absolute path.  If the path does not start with a /, assume
+it is relative to os.getcwd()."""
     if not isabs(path):
         path = join(os.getcwd(), path)
     return normpath(path)
+
+# Return a canonical path (i.e. the absolute location of a file on the
+# filesystem).
+def realpath(filename):
+    """Return the canonical path of the specified filename, eliminating any
+symbolic links encountered in the path."""
+    import string
+    filename = abspath(filename)
+
+    bits = ['/'] + string.split(filename, '/')[1:]
+    for i in range(2, len(bits)+1):
+        component = apply(join, bits[0:i])
+        if islink(component):
+            resolved = os.readlink(component)
+            (dir, file) = split(component)
+            resolved = normpath(join(dir, resolved))
+            newpath = apply(join, [resolved] + bits[i:])
+            return realpath(newpath)
+
+    return filename


Chris
-- 
=============================================================================
|          Chris Lawrence         |       Get your Debian 2.1 CD-ROMs       |
|     <quango at watervalley.net>    |        http://www.lordsutch.com/        |
|                                 |                                         |
|     Grad Student, Pol. Sci.     |     Do you want your bank to snoop?     |
|    University of Mississippi    |    http://www.defendyourprivacy.com/    |
=============================================================================



From avv at quasar.ipa.nw.ru  Fri Jun 25 19:46:07 1999
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Fri, 25 Jun 1999 16:46:07 -0700
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773CA67.7057F797@quasar.ipa.nw.ru> <3773D55B.8C966174@spacenet.tn.cornell.edu>
Message-ID: <377414BF.B40B8302@quasar.ipa.nw.ru>

Tom Loredo wrote:

> Well, one reason not to rush to wxWindows is that there is
> no "real" Mac version.  I realize that a Mac version is "underway",
> but the wxWindows site has said that for a *long* time now.
> A beta version was posted about a month ago, but I've not
> heard much about it, and most of the wx web site doesn't
> seem to know it exists (e.g., the FAQ still says the beta
> is forthcoming, and the dowload section appears to still
> have the old preview alpha release).  I develop on both Unix
> and Mac, and although the Unix part of me finds wx appealing,
> the Mac side of me feels like a 2nd class citizen in the wx world.

I have just visited wxWindows site, and picked up from the "News" page:

May 27th, 1999

    Beta 1 of wxWindows 2 for Mac has been released, thanks to Stefan
Csomor's amazing
    efforts.

Alexander






From greg.ewing at compaq.com  Thu Jun 17 22:00:32 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 18 Jun 1999 14:00:32 +1200
Subject: GC play
References: <mp8a3.324$U6.2508@newsr2.twcny.rr.com>
Message-ID: <3769A840.A5AE6778@compaq.com>

Darrell wrote:
> 
> If I'm passed a reference is there a way to get its vital information like
> name, filename, line number.... ?

If x is an instance of a class, then sys.modules[x.__class__.__module__]
will give you the module that the class was defined in; from there,
__file__ gives you the file name (but not the line number, as far as I
know).

If f is a function, then f.func_code has attributes co_filename and
co_firstlinenumber.

I don't think there's anything you can find out about the source
of any other kind of object.

Greg



From adjih at technologist.com  Fri Jun  4 08:33:37 1999
From: adjih at technologist.com (adjih at technologist.com)
Date: Fri, 04 Jun 1999 12:33:37 GMT
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net> <aahzFCpHw8.Gxt@netcom.com>
Message-ID: <7j8h30$bhn$1@nnrp1.deja.com>

In article <aahzFCpHw8.Gxt at netcom.com>,
  aahz at netcom.com (Aahz Maruch) wrote:
> In article <375536C2.DD917F99 at prescod.net>,
> Paul Prescod  <paul at prescod.net> wrote:
> >
> >I consider it a compliment that he can't find anything real to point
out
> >as Python's contribution to language fascism but I also find it
> >distressing that he feels the need to sow the seeds of fear,
uncertainty
> >and doubt in the minds of would-be Python users. The Python feature
that
> >most of us (in retrospect, unnecessarily) worry about before trying
Python
> >is the whitespace handling. Larry can only turn people off of Python
by
> >suggesting that that is Python's central feature or philosophy.
>
> I disagree with your parenthetical comment.  I wouldn't call myself an
> "expert" Python programmer by any stretch of the imagination, but I've
> written a fair bit of it in connection with a multi-person project
over
> the past four months.  As a seasoned programmer of many years and many
> languages, I've disagreed with almost all the whines that have been
> brought against Python (e.g. the stupid "=" vs. "==" thread), but this
> is one point that still annoys me.
>
> Oh, I've gotten used to it, no question about that, and I even
> appreciate the readability when it comes to working over someone
else's
> code (which is what I've mostly been doing).  But the whitespace
problem
> is a real PITA when it comes time to make a change to the structure of
a
> program -- it's an investment I don't mind making when I'm doing
"real"
> programming, but when I'm making a bunch of quick, fiddly changes to
> debug something it continually gets in my way.
>
> I'm not trying to push a change, mind you.  I'm just pointing out that
> IMAO the whitespace issue is a legitimate complaint.  There are plenty
> of real issues to be worked on, though.

  Well I just quickly browsed the Python source code, and Guido seems
to have heard your complaint today , since block-delimiters are now
supported.
Actually he seems to have finally found that so necessary, that he took
his usual time travel machine, and has modified the code of already
released python verions.

  Two problems though:
- 1) You need a magic sequence to enable them, see code below.
- 2) The block delimiters are completly advisory, so you need a an
auxiliary program to ensure proper indentation: a patch of
'examples/Tools/scripts/pindent.py' from Python-1.5.2 is included
below.


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

# Magic code to trigger hooks added to the Python interpreter by Guido
# in his recent time travel. Reverse engineered from interpreter
# source code.

enddef, endclass, endif, endfor, endwhile, endtry= \
        not SyntaxError, Ellipsis is not SyntaxError, \
        range is locals and compile, eval or input, \
        4/6/1999, "--Guido"

#-----
# Example of use

def printFactor(i):
    initial=i
    factor=[]
    current=3
    while i>1:
        if i%current==0:
            factor.append(current)
            i=i/current
        else:
            current=current+1
        endif
    endwhile
    print "%d factors=%s" % (initial, factor)
enddef

for i in range(0,1000):
    if i%2==0:
        print "even=",i
    else:
        printFactor(i)
    endif
endfor

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

Patch for pindent.py (only reformating, the patch isn't enough for the
'-c(omplete)' operation, i.e. adding block delimiters).

--- pindent-orig.py     Fri Jun  4 16:39:17 1999
+++ pindent.py  Fri Jun  4 17:06:00 1999
@@ -105,7 +105,7 @@
                        r'(\s+(?P<id>[a-zA-Z_]\w*))?'
                        r'[^\w]')
                self.endprog = re.compile(
-                       r'^\s*#?\s*end\s+(?P<kw>[a-z]+)'
+                       r'^\s*#?\s*end(?P<kw>[a-z]+)'
                        r'(\s+(?P<id>[a-zA-Z_]\w*))?'
                        r'[^\w]')
                self.wsprog = re.compile(r'^[ \t]*')
@@ -156,7 +156,9 @@
                        if not line: break      # EOF
                        # end if
                        m = self.endprog.match(line)
-                       if m:
+                       if (m and string.find(line,"enddef,
endclass")!=0
+                           and m.group('kw') in ['def','class','if',
+                                                 'for','while','try']):
                                kw = 'end'
                                kw2 = m.group('kw')
                                if not stack:


(Ok, I guess one could easily do better, and it won't help much, but
I couldn't resist)

-- Cedric


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From addicted at cybercable.fr  Tue Jun 22 17:51:48 1999
From: addicted at cybercable.fr (BoB)
Date: Tue, 22 Jun 1999 23:51:48 +0200
Subject: Tk "save as" dialog
Message-ID: <7kp071$peg$1@oceanite.cybercable.fr>

Hello world
I m french, which explains my so bad english... sorry...

here is my request :

For the moment, I'm using Tk "save as" dialog box. But it's plenty of buggs,
and I really can't use it any more.
Is anybody here has already made in Python TKinter a save as dialog which
works better than the basic Tk one.
Please answer me, those buggs are really ruinning my applications...

thanx in advance for your answers.

                        Laurent








From arcege at shore.net  Thu Jun 24 16:29:14 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Thu, 24 Jun 1999 20:29:14 GMT
Subject: Python & Tkinter problems...
References: <7ku177$f42$1@nnrp1.deja.com>
Message-ID: <uywc3.119$Of3.27661@news.shore.net>

alltube at netscape.net wrote:
: Hi there,

: I am trying to run Tkinter (tcl/tk 8.1 I compiled and installed it..)
: on Solaris 2.6.  Python version 1.5.2.  The tcl/tk seems to work
: fine, I was able to run a couple of demos, the TCL_LIBRARY is also set
: ok, my environment stuff:

: /usr/local/lib/tcl8.1 $ls
: auto.tcl      http2.0       msgcat1.0     safe.tcl
: encoding      init.tcl      opt0.4        tclAppInit.c
: history.tcl   ldAix         package.tcl   tclIndex
: http1.0       ldAout.tcl    parray.tcl    word.tcl
: /usr/local/lib/tcl8.1 $echo $TCL_LIBRARY
: /usr/local/lib/tcl8.1
: /usr/local/lib/tcl8.1 $

: The error I get from Python is:

: /usr/local/lib/tcl8.1 $python
: Python 1.5.2 (#5, Jun 24 1999, 14:52:16)  [GCC 2.8.1] on sunos5
: Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
:>>> import Tkinter
:>>> Tkinter._test()
: Traceback (innermost last):
:   File "<stdin>", line 1, in ?
:   File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 1947, in _test
:     root = Tk()
:   File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 886, in
: __init__
:     self.tk = _tkinter.create(screenName, baseName, className)
: TclError: Can't find a usable init.tcl in the following directories:




: This probably means that Tcl wasn't installed properly.

:>>>


: The init.tcl is in fact in /usr/local/lib/tcl8.1 and it appears to be
: ok, I have run a number of tk demo programs...

: Thanks for any help...

: Alex.
: Toronto

There are some incompatibilities to Tcl 8.1, which came out after Python
1.5.2.

Dieter Maurer published some patches to this newsgroup about a month ago.

The URL on DejaNews for that thread is:
  http://www.deja.com/viewthread.xp?search=thread&svcclass=dncurrent&ST=PS&group=comp.lang.python&recnum=%3cx7g14g9a37.fsf at lindm.dm%3e%231/1

  -Arcege




From fatjim at home.com  Mon Jun  7 05:05:34 1999
From: fatjim at home.com (Jim Meier)
Date: Mon, 07 Jun 1999 09:05:34 GMT
Subject: CGI module misbehaving?
Message-ID: <375B8BCF.C55B1F31@home.com>

Hi folks! I'm making my first foray into web programming, so my question
_might_ be silly; however, I've checked it out and seems to be an actual
bug.

At the end of this message is a python program that is supposed to
handle displaying and adding players to a gadfly database. The database
part is working fine. The program is there so that the bug can be
demonstrated..

The problem comes at the end of the file (the start of the program) when
it attempts to use the CGI module to read some form values in. A
traceback is generated on the line:

if form.has_key["name"]:
    ...

the traceback is:
Traceback (innermost last):
File "c:\httpd\cgi-bin\ahdb\add-player.py",line 178, in ?
if form.has_key["name"]:
TypeError: expected integer index

(formatting faked, as it was copied of the browser window)

I've searched through the source for the CGI module, and it does not
seem to generate this exception anywhere!

Is this a problem that is know, or is there something wrong with my
setup here?

Platform info:
Windows 98 (sigh)
Python 1.5.2 from PythonWare's download section (
http://www.pythonware.com/downloads/py15-980706.exe )
omniHTTPd v2.02

-Jim

code exhibiting the bug: (yes, pretty ugly I know. I take suggestions..
:)


# add-player.py ::sun jun 6::Jim Meier
# display forms for adding or deleting players

from os import environ
import cgi
from gadfly import gadfly

# find environment variables:
if environ.has_key('AH_HOME'):
  ah_dir=environ['AH_HOME']
else:
  ah_dir="c:\\py15\\src\\ahdb"
ahdb_dir=ah_dir+"\\ahdb"

connection=gadfly("ahdb", ahdb_dir)
curs=connection.cursor()

def get_players():
  curs.execute("select * from players")
  return curs.fetchall()

def player_table_row(player,sel_col):
  name='<a href="/cgi-bin/show-player?%s">%s</a>' %
(player[0],player[1])
  email='<a href="mailto:%s>%s</a>' % (player[2],player[2])
  web='<a href="%s">%s</a>' % (player[3],player[3])

  print "<tr>"
  print "  <td>%s</td>" % sel_col
  print "  <td>%s</td>" % name
  print "  <td>%s</td>" % email
  print "  <td>%s</td>" % web
  print "</tr>"

def player_table():
  print '<table border=0 width="90%" cellspacing=3 bgcolor="000000">'

  print '<tr bgcolor="6495ed">'
  print "  <td>Select</td>"
  print "  <td>Name</td>"
  print "  <td>E-Mail</td>"
  print "  <td>Home/AH Page</td>"
  print "</tr>"

  for i in get_players():
    sel_col='<input type="checkbox" name="player_sel" value="%s">' %
str(i[0])
    player_table_row(i, sel_col)

  print "</table>"


def new_player_form():
  print '<form method="POST" action="/cgi-bin/ahdb/add-player.py">'
  print '<table border=0 width="70%" cellspacing=3>'
  print '<tr><td>Name:</td> <td><input name="name"
type="text"></td></tr>'
  print '<tr><td>E-Mail:</td> <td> <input name="email"
type="text"></td></tr>'
  print '<tr><td>Home/AH Page:</td> <td> <input name="web"
type="text"></td></tr>'
  print '<tr><td></td><td><input type="submit" value="Add
Player.."></td></tr>'
  print '</table>'
  print '</form>'


def add_player():
  # make sure all required fields are filled..
  global mylog
  global name,email,web

  form_ok=1
  if name is None:
    form_ok, reason = None, "The Name field is required. Please go back
and fill it in."
  if name is None:
    form_ok, reason = None, "The E-Mail field is required. Please go
back and fill it in."
  if not form_ok:
    mylog.write("form doesn't have all we need, returning '%s'\n" %
reason)
    mylog.flush()
    return reason

  # get a unique id for him/her
  mylog.write("getting log id number..\n")
  mylog.flush()
  import whrandom
  id=whrandom.randint(0,sys.maxint)
  while 1: # no way in hell this'll fill up..
    mylog.write("have %i, check to see if it's used..\n" % id)
    mylog.write("executing the select statement..\n")
    curs.execute("select id from players where id=%s" % str(id))
    mylog.write("done execing select..\n")
    if len(curs.fetchall())==0:
      mylog.write("fetchall returned [] so we're breaking the loop..\n")

      break
    try:
      id=id+1
    except OverflowError:
      id=whrandom.randint(0,sys.maxint)

  # add 'em up!
  if web is None:
    web=""
  res=curs.execute("""insert into players (id,name,email,web)
                  values (%s,'%s','%s','%s')"""
                  % str(id), name, email, web)
  if res!=1:
    connection.abort()
    return "Couldn't update the database!"

  connection.commit()


def main():
  global mylog
  global name,email,web

  form=cgi.FieldStorage()
  print repr(form)
  name,email,web=None,None,None
  if form.has_key["name"]:
    name=form['name'].value
  if form.has_key['email']:
    email=form['email'].value
  if form.has_key['web']:
    web=form['web'].value


  mylog=open("c:\windows\profiles\jim\desktop\log.txt","w")
  mylog.write("add-player.py run..\n")
  mylog.write("cgi arguments:\n")
  mylog.write("name=%s\nemail=%s\nweb=%s\n"%(name,email,web))
  mylog.write("\n")
  mylog.write("enviroment:\n")
  for i in environ.keys():
    mylog.write("%s: '%s'\n" % (i, environ[i]))
  mylog.write("\nthat's all, total.\n")


  if len(form.keys()) != 0:
    # should add player, and redirect back to list unless an error
occurs.
    mylog.write("about to call add_player()..\n")
    mylog.flush()
    res=add_player()
    mylog.write("add_player() returned %s\n" % `res`)
    mylog.flush()
    if res!=None:
      mylog.write("res was %s so we're sending an error page..\n" %
`None`)
      print 'content-type: text/html'
      print
      print '<html><head><title>Whoops!</title></head>'
      print "<body><H1>Something's Amiss!</h1>"
      print """Either you screwed up or I did. Something happened while
               processing your form request that caused the operation to
fail.
               They tell me that the reason it failed was:"""
      print '<br>'
      print '<h3><p fgcolor="ff0000">%s</p></h3>' % res
      print '<hr>'
      print '<p align="left"> <a href="/cgi-bin/ahdb/add-player.py">'
      print 'Go Back..</a></p>'
      print '</body></html>'
      mylog.flush()
      mylog.close()
      return

  mylog.write("didn't call add_player()..\n")
  print "content-type: text/html"
  print
  print "<html><body>"
  player_table()
  new_player_form()
  print "</body></html>"

  #..and have a nice day.

  mylog.flush()
  mylog.close()



form=cgi.FieldStorage()
print repr(form)
name,email,web=None,None,None
if form.has_key["name"]:
  name=form['name'].value
if form.has_key['email']:
  email=form['email'].value
if form.has_key['web']:
  web=form['web'].value

main()





From none at microsoft.com  Thu Jun 10 00:35:19 1999
From: none at microsoft.com (plz_reply_to_group)
Date: Wed, 09 Jun 1999 21:35:19 -0700
Subject: Sory I meant '/'
References: <375F21B3.D23A5EA9@west.boeing.com>
Message-ID: <375F4087.AC2D60AD@microsoft.com>

No, no, I meant '/' -- I want to match on the _foreward_ slash

"Jeff P." wrote:
> 
> I've been messing with this for a little while now, but I can't get this
> to work. How do I match the last occurence of '\' in a string? I want to
> remove the file name from a url (e.g. news/breakingnews/981001b.html -->
> news/breakingnews/ ).
> 
> Here is the simplest version (that doesn't work). It matches the _first_
> backslash.
> 
> import regex
> 
> file_in = open( 'dir.txt', 'r+' )
> for line in file_in.readlines():
>         print regex.search( '/', line )
> 
> Thanks for any help!
> 
> Jeff P.




From belbo_j at my-deja.com  Fri Jun 25 08:14:43 1999
From: belbo_j at my-deja.com (belbo_j at my-deja.com)
Date: Fri, 25 Jun 1999 12:14:43 GMT
Subject: How to use Perfmon Library
Message-ID: <7kvrrc$2d3$1@nnrp1.deja.com>

Hello,

Is it anybody that can give me some information
about the usage of Perfmon library ?

I want to monitor a process on a remote NT server


Thanks,

Jacobo


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From bhoel at server.python.net  Thu Jun 24 04:00:18 1999
From: bhoel at server.python.net (Berthold =?iso-8859-1?Q?H=F6llmann?=)
Date: Thu, 24 Jun 1999 10:00:18 +0200
Subject: generating valid random SMILES, water ice modelling
References: <14193.17579.54805.701994@lrz.de>
Message-ID: <3771E592.5730D930@starship.python.net>

Eugene Leitl wrote:
> 
> Does anybody have (preferably Python, but I'd be thankful for
> anything) code to generate random but valid SMILES? Thanks a bunch.
> 
Hello,

I've ported David W. Sanderson's smiley program to Python. It also works
as a package. I have'nt released it yet because it needs the original
smiley definition by David W. Sanderson. I don'nt wanted to ship this
file without permission of Mr. Sanderson, but havn't receved any anwer
to my request. Though it is some month now since I asked, I dont't
expect an answer anymore. So I change my prepared distribution to
include the original C version and put it to my starship homepage with
the next update. In the meantime, if you are interested, drop me a mail.

Cheers

Berthold
-- 
bhoel at starship.python.net / http://starship.python.net/crew/bhoel/
        It is unlawful to use this email address for unsolicited ads
        (USC Title 47 Sec.227). I will assess a US$500 charge for
        reviewing and deleting each unsolicited ad.



From fredrik at pythonware.com  Fri Jun  4 12:44:17 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 4 Jun 1999 16:44:17 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu><7imr6b$1s0$1@nnrp1.deja.com><7j8o26$55v$1@cronkite.cc.uga.edu> <14167.62078.226416.757006@anthem.cnri.reston.va.us>
Message-ID: <011901beaea9$7da7af40$f29b12c2@pythonware.com>

Barry A. Warsaw wrote:
> What we need is CODE.  Somebody's gotta step up to the plate and
> generate some patches that people can play with.

    http://www.python.org/cgi-bin/todo.py?req=show&file=faq02.012.htp

    "Some experiments have been tried (e.g.
    http://starship.skyport.net/crew/gandalf/gc-ss.html
    and http://puli.lib.bke.hu/~xr/) but they are a bit
    disappointing: it ends up being slower and using
    more memory."

M: No it doesn't. 

</F>




From mnot at pobox.com  Mon Jun 21 22:55:24 1999
From: mnot at pobox.com (Mark Nottingham)
Date: Tue, 22 Jun 1999 02:55:24 GMT
Subject: htmllib: CR in CDATA
Message-ID: <00d201bebc5a$adb129a0$0301a8c0@mnot.net>

It appears that htmllib doesn't ignore returns in CDATA fields, as HTML 4.0
says it should:
http://www.w3.org/TR/REC-html40/types.html#type-cdata
http://www.w3.org/TR/REC-html40/sgml/dtd.html

As a result, htmllib improperly parses any CDATA element that wraps across a
line; this affects elements like

<A href="foo.
gif">

I'm happy to work up a patch, but I thought I'd ask around first. It may be
a bit involved to fix it properly; every CDATA should be handled this way,
which practically means almost every tag attribute.

Regards,


Mark Nottingham, Melbourne Australia
mnot at pobox.com  http://www.mnot.net/





From graham at sloth.math.uga.edu  Thu Jun 10 11:03:44 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 10 Jun 1999 15:03:44 GMT
Subject: Python 2.0
References: <000801beb2e8$05e98760$af9e2299@tim>
Message-ID: <7jok4g$hih$1@cronkite.cc.uga.edu>

[Graham Matthews]
> a) don't call finalisers for circularly referenced collected objects.
>    Instead just collect their memory. This is not perfect but better
>    than current Python which doesn't run finalisers on circularly
>    referenced objects nor collect their memory.
Tim Peters (tim_one at email.msn.com) wrote:
: Solve the problem by ignoring it?  Guido would find that more attractive
: than I would <wink>.

Even if you do this you still end up in a better situation than current
Python, since you can at least reclaim the memory of circularly reffed
objects. Moreover for builtin objects (like files) you can do better
and also reclaim their resources. So it is not perfect but still an
improvement on current Python.

[Graham Matthews]
> b) better still use a two pass collector. If you don't know what that is
>    just think about what you need to do (or not do) in pass 1 to ensure
>    that all the objects needed for a __del__ method to execute in a known
>    environment are there. <wink>
Tim Peters (tim_one at email.msn.com) wrote:
: Java does this.  Start the thread from its beginning again <0.5 wink>.
: 
I am sure you can come up with something better than this Tim! <wink>

graham
-- 
           As you grow up and leave the playground
       where you kissed your prince and found your frog
          Remember the jester that showed you tears
                   the script for tears



From bkc at Murkworks.com  Wed Jun  9 19:17:00 1999
From: bkc at Murkworks.com (Brad Clements)
Date: Wed, 9 Jun 1999 19:17:00 -0400
Subject: AFS module for python?
Message-ID: <7jmskp$e6d$1@news.clarkson.edu>

Anyone swig AFS routines for handling pts, kas and vos entries?

Thanks..

--
Brad Clements,
bkc at murkworks.com





From mlauer at amalia-atm.rz.uni-frankfurt.de  Sun Jun 13 19:46:54 1999
From: mlauer at amalia-atm.rz.uni-frankfurt.de (mlauer at amalia-atm.rz.uni-frankfurt.de)
Date: 14 Jun 1999 00:46:54 +0100
Subject: Dumb key-binding question
Message-ID: <376434de@nntp.server.uni-frankfurt.de>

Pardon, but I can not get binding
<Control-KeyPress> to work correctly.
<Control-KeyRelease> works fine, but
<Control-KeyPress> only functions while
the holding the key. This is not exactly
what I was expecting... Hmm ?


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





From matz at netlab.co.jp  Wed Jun  2 00:01:02 1999
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 02 Jun 1999 13:01:02 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
 <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>
 <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net>
 <87hfosquc0.fsf@ev.netlab.co.jp> <3753F121.E4A9BE8B@prescod.net>
Message-ID: <87yai3ntdt.fsf@ev.netlab.co.jp>

Paul Prescod <paul at prescod.net> writes:

|> I said nothing about the destructor invocation predictability.
|
|I consider that a part of the definition of destructor.

Well,.. is it?  That's fine, anyway.  Now I understand you.

|> I don't understand and really interested in why you guys require
|> to predict destructor invocation so much, but it's another story.
|
|What if I am holding GUI "device context" objects and the GUI has a
|limited number of them. I don't think that there is a way to tell the GC
|that there is a limited number of them. It doesn't know that device
|contexts are a limited resource, just as memory is.

Of cource you can tell the GC about limitaion, if you can detect
outage of the resource somehow.  And you can detect outage for most of
the resources on the machine.

e.g.

  fd = open(path, mode);
  if (fd < 0 && (errno == EMFILE || errno == ENFILE)) {
    call_gc();  
    fd = open(path, mode);
  }
  if (fd < 0) {
    raise("can't help it");
  }

Ruby uses this scheme for most of the resources like fd, sockets.  So
you can write something like:

  txt = open('blahbalh.txt','r').read()

close() will be called sometime later (not immediately after), but
called for sure.  If not called, it's a bug in interpreter.

I don't mean real GC does everything ref counting does.  For
exapmle, with real GC:

  dialog = GUItoolkit.dialog(params)
  ... do something with the dialog ...
  dialog = None

will close the dialog at unpredictable timing, which is not
desirable for everyone.  But I prefer explicit termination of the
dialog for this case.

  dialog = GUItoolkit.dialog(params)
  ... do something with the dialog ...
  dialog.close

|What GC algorithm does Ruby use and how much overhead does it add?

Ruby uses conservative mark and sweep GC.  It never moves the objects.
The overhead tends to depend on applications, but memory management
overhead is normally less than 5% of the execution (Ruby often runs
faster than Python).  And as I said before, the hang is not noticable
for interactive processes.

|> Second, do you think that optimization for basic users compensates the
|> risk of potential memory leak, by cyclic structures or missing DECREF
|> in extension modules?  Do you mean it's easy for experts to find
|> cyclic structure and cut their reference to destruct?  For me, it's
|> not.  I hate that.  Is this because I'm not a Python expert?
|
|It's pretty much a toss-up for me. In general I prefer optimizations that
|make the language easy for new users to those that are for complex, long
|running programs but full garbage collection would make complex programs a
|fair bit easier.

So you think ref counting is nice for new users, maybe by its
predictability.  I don't think so because even new users can make
cyclic references.  But it's a toss-up indeed.

						matz.



From jbauer at rubic.com  Fri Jun 18 20:29:18 1999
From: jbauer at rubic.com (Jeff Bauer)
Date: Sat, 19 Jun 1999 00:29:18 GMT
Subject: DATE ARITHMETIC
References: <7kejrv$e57$1@violet.singnet.com.sg>
Message-ID: <376AE45E.596B746D@rubic.com>

Ajith Prasad wrote:
> I need to do some simple date arithmetic - finding the number 
> of days between any 2 days, the age of a person on a day given
> his birthday. The mxDateTime module downloadable from the 
> Python website is more complex than what I need. Any other
> suggestions?

Here's a simple one:

  http://starship.python.net/crew/jbauer/normaldate/

Best regards,

Jeff Bauer
Rubicon, Inc.



From Angus.MacKay at p98.f112.n480.z2.fidonet.org  Tue Jun 29 02:00:47 1999
From: Angus.MacKay at p98.f112.n480.z2.fidonet.org (Angus MacKay)
Date: Tue, 29 Jun 1999 07:00:47 +0100
Subject: printing with print
Message-ID: <0000158e@bossar.com.pl>

From: Angus MacKay <amackay at starvision.com>

is there any other way to print that with print?
perhaps with a real function.

I want to print two things without a space being added between them:
foo = "there"
print "hi" + foo

works fine but: 
foo = 2
print "hi" + foo

does not. I like the printing of any object ability of "print" so I
do not want to use "" % style printing, I just want to be able to
govern the whitespace that gets output.

cheers, Angus.



From hniksic at srce.hr  Sat Jun  5 15:56:05 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 05 Jun 1999 21:56:05 +0200
Subject: Deleting characters from a string
References: <87n1ye3g4v.fsf@pc-hrvoje.srce.hr> <y0jyahye89i.fsf@vier.idi.ntnu.no>
Message-ID: <87vhd2zaju.fsf@pc-hrvoje.srce.hr>

mlh at idt.ntnu.no (Magnus L. Hetland) writes:

> result = ""
> for char in host:
>     if char not in ' "':
>         result = result+char
> 
> host = result
> 
> What!? To me that's more elegant... <wink>

To me it would be if Python actually had chars.  <handshake>

:-)



From mstenber at cc.Helsinki.FI  Wed Jun 23 00:23:13 1999
From: mstenber at cc.Helsinki.FI (Markus Stenberg)
Date: 23 Jun 1999 07:23:13 +0300
Subject: SWIG and binary strings?
Message-ID: <al8pv2nwnny.fsf@myntti.helsinki.fi>

I've swig'ed OpenSSL library for my Python needs. That was a cakewalk,
until I ran at the 'char *, int'-defined strings with nulls embedded in
them used in, among other things, SSL_write(). 

Is there any elegant way to make SWIG generate s# automatically (as opposed
to s, and requiring unneccessary length parameter and much whining if
string has 0's?)

My (ugly) solution was following:

- Define each function in wrong fashion in the .i
(SSL_write as (SSL*, char *) while it should be SSL_write(SSL *, char *,
int), etc)

- Run SWIG

- Run patch after SWIG, to do following fixes:
@@ -1108,10 +1110,11 @@
     int  _result;
     SSL * _arg0;
     char * _arg1;
+    int _arg2;
     char * _argc0 = 0;
     self = self;
-    if(!PyArg_ParseTuple(args,?s:SSL_write",&_argc0,&_arg1))
+    if(!PyArg_ParseTuple(args,?s#:SSL_write",&_argc0,&_arg1,&_arg2))
         return NULL;
     if (_argc0) {
         if (SWIG_GetPtr(_argc0,(void **) &_arg0,"_SSL_p")) {
@@ -1119,7 +1122,7 @@
         return NULL;
         }
     }
-    _result = (int )SSL_write(_arg0,_arg1);
+    _result = (int )SSL_write(_arg0,_arg1,_arg2);
     _resultobj = Py_BuildValue(?",_result);
     return _resultobj;
 }

.. and obviously same for other functions as well.
I do not find this elegant, however, is there any other way? (Beyond
writing ugly wrappers in pure Python-C, and I am too lazy to maintain those
;->)

- Academically-curious-ly, Markus

-- 
		        Markus Stenberg
Finger fingon at mpoli.fi for PGP key --  PGP key id: 1024/5DAC7D21



From dkuhlman at netcom.com  Tue Jun 22 20:27:58 1999
From: dkuhlman at netcom.com (G. David Kuhlman)
Date: Wed, 23 Jun 1999 00:27:58 GMT
Subject: Help with String Manipulation
References: <0UUb3.1337$el4.44932@ord-read.news.verio.net>
Message-ID: <dkuhlmanFDr7yM.LC1@netcom.com>

 
>>> import array
>>> a = 'abcdef'
>>> a
'abcdef'
>>> b = array.array('c', a)
>>> b
array('c', 'abcdef')
>>> b.reverse()
>>> b
array('c', 'fedcba')
>>> c = b.tostring()
>>> c
'fedcba'
>>>


See the Python Library Reference, Sect. 5.6 "array -- Efficient
arrays of numeric values"

  - Dave



Samuel G. Williams (samw at masu.wwa.com) wrote:
> I know this is a very stupid question and one that has been addressed
> within available documentation, however I need a quick hand. I am in need
> of a routine to reverse the order of a string. I have tried several things
> that have resulted in failures. It is really stupid, but I am just
> learning python. I can do this readily in C, perl, tcl, but am at a loss.
> I would appreciate any help I can get.
> 
> Thanks in advance.
> 
> -- 
> 
> 



From matz at netlab.co.jp  Wed Jun  9 02:14:58 1999
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 09 Jun 1999 15:14:58 +0900
Subject: fork()
References: <000001beb23b$9f17bc20$a69e2299@tim>
Message-ID: <877lpd528t.fsf@ev.netlab.co.jp>

"Tim Peters" <tim_one at email.msn.com> writes:

|Context:  this offshoot of the thread was talking about memory use after a
|fork() call in a copy-on-write implementation of fork.  So long as
|bookkeeping info is stored in object headers, "real GC" is at a real
|disadvantage, in that respect, compared to RC.  Simply because RC won't
|touch something unless *you* do.

Indeed.  Though some "real GC" keeps mark bits in separated pages to
avoid copying bunch of pages in such situation.

BTW, ref counting modifies objects for mere referencing, it causes
copying too.  Also, it would be performance pain if Python's global
thread lock is removed someday, since ref counting requires mutual
lock to protect ref count update.  I'm not sure this problem can be
solved or not.
						matz.



From brainwav at hotmail.com  Sun Jun 13 17:35:54 1999
From: brainwav at hotmail.com (Denis Lamarche)
Date: Sun, 13 Jun 1999 21:35:54 GMT
Subject: Python Newbie
Message-ID: <37644D2B.CC7D0BB3@hotmail.com>

Where can I find "Python.h"?  I am using dos.  I cant find it anywhere.

Denis Lamarche



From akuchlin at mems-exchange.org  Mon Jun 21 09:15:09 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 21 Jun 1999 09:15:09 -0400
Subject: BiBTeX module?
References: <joe-160619991358444033@chinacat.salk.edu> <7k950a$hsb@cs.vu.nl> <slrn7miqcg.tpi.bernhard@alpha1.csd.uwm.edu> <slrn7mk9ui.dic.gobry@blanc.idiap.ch> <wk1zf9wo5y.fsf@ifi.uio.no> <slrn7moad8.4l5.bernhard@alpha1.csd.uwm.edu>
Message-ID: <3dg13litk2.fsf@amarok.cnri.reston.va.us>

I'm surprised Greg Ward hasn't jumped into this thread yet, but take a
look at his tech report on his BibTeX parser btOOL, particularly the
first 3 sections, which contain an interesting examination of BibTeX's
design.  Later sections dive into the Perl code and are less
interesting, but the C code for the parser could probably be wrapped
as a Python module without too much difficulty.

	btOOL's page is at http://starship.python.net/~gward/btOOL/ ;
the tech report is btOOL.ps.gz or btOOl.dvi.gz.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
Brigadier, a straight line may be the shortest distance between two points,
but it is by no means the most interesting.
    -- The Doctor, in "The Time Monster"





From mnot at pobox.com  Mon Jun 14 23:44:51 1999
From: mnot at pobox.com (Mark Nottingham)
Date: Tue, 15 Jun 1999 03:44:51 GMT
Subject: Looking for modules?  Config files?  Ini files?  in python..
References: <1282710643-100650854@hypernet.com>
Message-ID: <00ac01beb6e1$6c541860$0301a8c0@cbd.net.au>

Another possibility is at
http://www.mnot.net/python/Conf.py

Each has its ups and downs - this one doesn't do sections.


----- Original Message ----- 
From: Gordon McMillan <gmcm at hypernet.com>
Newsgroups: comp.lang.python
To: <python-list at cwi.nl>
Sent: Tuesday, June 15, 1999 11:53
Subject: Re: Looking for modules? Config files? Ini files? in python..


> Benjamin Schollnick writes:
> > 
> >  I'm trying to find a simple method to deal with *.INI or other
> >  styles
> > of
> > configuration files....
> > 
> >  I haven't seen anything, and would hate to re-invent the wheel....
> > 
> >  Any suggestions?
> 
> How 'bout ConfigParser.py? It's in the standard distribution.
> 
> - Gordon
> 
> 
> 
> 




From smjoshi at bellsouth.net  Sun Jun 27 22:47:55 1999
From: smjoshi at bellsouth.net (SunitJoshi)
Date: Sun, 27 Jun 1999 21:47:55 -0500
Subject: Using freeze
Message-ID: <EoBd3.6644$rZ4.56779@news4.mia>

Hello
    I'm trying to make exe's from python scripts using VC6++. However I'm
lost as to how to use freeze to accomplish this. If someone could include
some details I would really appreciate it.

thanks
Sunit





From vbykovsk at sdsp.mc.xerox.com  Fri Jun 25 10:44:11 1999
From: vbykovsk at sdsp.mc.xerox.com (Val Bykoski)
Date: Fri, 25 Jun 1999 10:44:11 -0400
Subject: [ANN] DejaGrabber 0.1 -- search Dejanews with Python
References: <7kuu73$1ki$1@brick.cswv.com>
Message-ID: <7l0520$qe4$1@news.wrc.xerox.com>

Neel:
This looks really great..
But I have this:

Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File "C:\Program Files\Py152\Lib\httplib.py", line 51, in __init__
    if host: self.connect(host, port)
  File "C:\Program Files\Py152\Lib\httplib.py", line 79, in connect
    self.sock.connect(host, port)
  File "<string>", line 1, in connect
error: (10065, 'winsock error')

when i am trying to reach www.python.org thru a proxy (www.mc.xerox.com).
How to handle this proxy-related situation?

Thanx -- val

Neel Krishnaswami <neelk at brick.cswv.com> wrote in message
news:7kuu73$1ki$1 at brick.cswv.com...
>
> If you'd like a convenient way of searching the DejaNews archive using
> Python, I'm happy to present the DejaGrabber module. This is a module
> to make searching Dejanews easier. There are two classes, DejaGrabber
> and Article. I've tried to add a reasonable number of useful
> docstrings, too.
>
> The interface to DejaGrabber could probably use a bit of work. If I
> ever use it enough to find out what the flaws are, I will fix
> it. Patches are welcome. :)
>
> Here's an example:
> >>> from DejaGrabber import *
> >>> d = DejaGrabber(group='comp.lang.python', author='Tim Peters')
> >>> l = d.get_messages(4)
> >>> print l
> [99/06/23  30 comp.lang.python          RE: Python Exes!      Tim Peters
> , 99/06/18  28 comp.lang.python RE: Newbie: Truth values (th      Tim
Peters
> , 99/06/18  28 comp.lang.python       RE: DATE ARITHMETIC      Tim Peters
> , 99/06/18  28 comp.lang.python     RE: NameError Problem      Tim Peters
> , 99/05/18  25 comp.lang.python     RE: while (a=b()) ...      Tim Peters
> ]
>
> You can grab a copy at
>
> http://www.sff.net/people/neelk/free-software/DejaGrabber.py
>
>
> Neel





From arcege at shore.net  Thu Jun 24 11:40:06 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Thu, 24 Jun 1999 15:40:06 GMT
Subject: Tkinter: radiobutton
References: <3771B52E.80F3FA46@pixar.com>
Message-ID: <qjsc3.77$Of3.12274@news.shore.net>

kiki <kiki at pixar.com> wrote:
: Newbie speaks: I'm sure there's an elegant solution to this, but I just
: can't find it, and my head hurts from reading man pages...

: I have a color menu that has "Add Color" and "Change Color", and an
: appended list of user-created colors as radiobuttons.

: What I would like to do is when the user chooses "Change Color", the new
: color overwrites the current color selected in the radiobutton list.

: The problem is I don't know how to find out which button is selected.
: self.color_menu.entryconfig("active"...) configures the "Change Color"
: button!!  Arg!  The index is otherwise implicit...

: A group of radiobuttons are associated by the fact that they share a
: variable.  When that variable changes, the states of the radio buttons
: change.  Certainly there must be a way to find out the index of the
: selected radiobutton in a group of radiobuttons and modify the contents
: [eg "value", "activebackground", etc.] of it...

: Thanx, and sorry if this was an obvious question.

: Kiki

No, it's not an obvious question at all.  Unfortunately, you need to
track the index when you create the menu.  By default, the first call
starts at 1, because the "tear-off" is 0.  If the menu is created with
the -tearoff=0 option, then the index starts at 0

This example creates a menu where the first item is a button (which 
when pressed doesn't do anything) used as a Label to hold the current
color, the other items (radiobuttons) after it change the label.

  menu = Menu(menubutton)
  def change_red(m=menu):
    m.entryconfig(1, label='Color: red')
  menubutton['menu'] = menu
  # the first is effectively a "Label"
  menu.add_command(label='Color: ')                             # index 1
  menu.add_radiobutton(label='Change red', command=change_red)  # index 2

Maybe it's time I give those menu entry classes to Guido (that does
this for you as class instances).

  -Arcege




From pphalen at teleo.net  Thu Jun  3 16:17:01 1999
From: pphalen at teleo.net (Patrick Phalen)
Date: Thu, 3 Jun 1999 13:17:01 -0700
Subject: "The Python Way"
References: <045501beadf8$610e62e0$f29b12c2@pythonware.com>
Message-ID: <9906031407150D.00552@quadra.teleo.net>

On Thu, 03 Jun 1999, Fredrik Lundh wrote:

> sure looks like the "community" thinks that changing the
> language is more important than using it...

Yeah; now that you mention it, the "let's fix Python" discourse
has been approaching my pain threshold, as it does once in awhile.

(Have you noticed that, during such periods, both Guido and Tim
seem to become singularly quiet?)

Not that Python shouldn't be open to critique, but the more I use
and learn about the language, the more I find myself appreciating the
nice balance and heft Guido gave to it. Yet there doesn't seem to be a
single document that sums up that "aesthetic," but rather it
tends to appear piecemeal, over time, mostly in the Wisdom of Chairman
Tim.

I'd like to suggest something as a sort of balm for those of us who
come here to try to learn to work with, rather than against, the grain
of Python:

Would both Guido and TIm Peters be willing to collaborate on a short
paper -- call it "The Python Way" for lack of a better title -- which
sets out the 10-20 prescriptives they might offer to those who come to
Python from other languages and immediately want to find a way to bend
it into uncomfortable positions -- (implement closures, etc.).

What I have in mind is sort of a very brief Strunk-&-White-like
"Elements of Style" for Python, which suggests fundamental idiomatic
recommendations for operating within the spirit of the language. A
distillation of Python Zen is what I'm talking about -- something to go
off and contemplate when the "fix Python now" decibels become a bit
much.

Tim? Guido?







From druid at princeton.crosswinds.net  Thu Jun  3 20:43:18 1999
From: druid at princeton.crosswinds.net (The Well of Latis)
Date: Fri, 04 Jun 1999 00:43:18 GMT
Subject: newbie trouble with raw_input and idle
Message-ID: <37581eb4.2704370@netnews.worldnet.att.net>

If I enter the following code (or any other using raw_input for that
matter) into idle's shell it works, when trying to run it from a file
(new window, edit it, save it then run it, nada)

def f():
    x=raw_input("ladida:")
    print x

f()

systems involved are a win98 and an nt4 machines, tcl version 8.0.5,
Python 1.5.2. The same code works under the plain Python interactive
shell, and pythonwin, but not idle. Any ideas?

--------------------------
Bull Run ACD Rescue
Several low mileage cow dogs available, visit the "barking lot" at
http://www.crosswinds.net/~bullrun/rescue.html
 



From Nikolai.Kirsebom at p98.f112.n480.z2.fidonet.org  Tue Jun 29 18:15:42 1999
From: Nikolai.Kirsebom at p98.f112.n480.z2.fidonet.org (Nikolai Kirsebom)
Date: Tue, 29 Jun 1999 23:15:42 +0100
Subject: Python, ASP - Namespace (again)
Message-ID: <00001584@bossar.com.pl>

From: "Nikolai Kirsebom" <nikolai at micon.no>

Trying again - since I'm rather stuck, and I would like to stick to Python.


I'm trying to use Python as scripting language on the web-server, however -
have problems to see how the namespace scope is handled.

FILE - GenDb.py
=============
class GenDb:
   def __init__(self):
      self.ConnStr = ""

   def SetDbAndTable(self, db, table):
      self.Db = db
      self.Table = table
      self.ConnStr = "Driver={Microsoft Access Driver (*.mdb)};
DBQ=f:\\Databases\\%s" % db

   def Attr(self, attrName):
      if attrName == "abc":
         return(self.ConnStr)
      else:
         return("-non-")

FILE - KSdebatt.py
===============
from GenDb import *

class KSdebatt(GenDb):
   def __init__(self):
      self.SetDbAndTable("KSdebatt.mdb", "BillMrk")

FILE - SCRIPT1.ASP
=================
<%@ LANGUAGE = Python %>
<%
import sys
sys.path.append("f:\\PythonScripts")
from KSdebatt import *
o=KSdebatt()
%>
<HTML>
<INPUT TYPE="text" NAME="abc" VALUE="<%= o.Attr("abc") %>" SIZE=20>
</HTML>


when I run the statements in the script1.asp file in an interactive window,
I can create the KSdebatt object, however when I run the script on the
server - I  get the following message:

Python ActiveX Scripting Engine error '80020009'
Traceback (innermost last): File "<Script Block >", line 4, in ?
o=KSdebatt() NameError: KSdebatt
/ksdebatt/pages/Script1.asp, line 6
o=KSdebatt()

If it is possible I would like to 'hide' the object (o) inside the KSdebatt
module - that is to have the instanciation inside the module itself - but
still be able to access it from the ASP-code in the script (o.Attr....).  Is
this possible ?


I'm running on NT 4.0 SP 3 and python 1.5.2b2


Thanks for any help

Nikolai Kirsebom






From guido at CNRI.Reston.VA.US  Thu Jun 24 10:52:18 1999
From: guido at CNRI.Reston.VA.US (Guido van Rossum)
Date: Thu, 24 Jun 1999 10:52:18 -0400
Subject: List <-> news hosed again?
In-Reply-To: Your message of "Thu, 24 Jun 1999 10:22:06 EDT."
             <199906241422.KAA03838@python.org> 
References: <199906241422.KAA03838@python.org> 
Message-ID: <199906241452.KAA27732@eric.cnri.reston.va.us>

> > [Guido]
> > > We're looking into this; uunet changed their configuration and we
> > > neglected to make a matching change.  It should be fixed now
> > > (presuming this message shows up on the uunet side of things).
[Tim]
> > Alas, don't think it worked -- DejaNews doesn't have your msg yet, nor my
> > ISP's (admittedly flaky) c.l.py feed.  Appears to have stopped working very
> > early AM (EDT) on 18-June-99.
[me]
> Indeed, it didn't work.  Inspection of the logs indicates that the
> uunet news software raises a "500 command not understood" error
> whenever our news server tries to upload anything; this suggests a
> serious version skew.  I think we may have to install a different
> version of the news software (unless it is caused by bogus data sent
> in a corrupted article that somehow upsets the command parser).

Hello, it's me again, interim webmaster :-)

Just like the previous time we experienced this, the uunet news server
mysteriously started accepting our posts again after several
complaints, *without* further action on our side.  Thanks to Tim for
alerting us to this problem and thanks to Roger & Stanley at CNRI for
taking action (even if that action only was a couple of mails and
phone calls to uunet :-).

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



From pduffin at mailserver.hursley.ibm.com  Mon Jun 28 08:48:28 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Mon, 28 Jun 1999 13:48:28 +0100
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <z%Q93.4399$5E4.36952@news2.randori.com> <m2n1xsmh10.fsf@desk.crynwr.com>
Message-ID: <37776F1C.5656@mailserver.hursley.ibm.com>

Russell Nelson wrote:
> 
> >   I'm trying to collect a list of all the _current_ issues making tcl
> > a bad language choice...
> 
> Because it necessarily parses strings.  Been there, done that, have
> the scars to prove it.
> 

What do you mean by "necessarily parses strings" ?


-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From greg.ewing at compaq.com  Thu Jun 17 18:13:00 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 18 Jun 1999 10:13:00 +1200
Subject: <language> broken
References: <1282792849-95706075@hypernet.com> <3764D7C3.718E125E@iname.com> <002f01beb64d$58e7b5f0$f29b12c2@pythonware.com> <376523CF.607E59A@acm.org>
Message-ID: <376972EC.85FCB52B@compaq.com>

"Fernando D. Mato Mira" wrote:
> 
> Suddenly, I have
> the impression the group is populated with very excitable and
> all-knowing `Linux kiddies'

I can reassure you that is *not* the typical kind of
person who frequents this group. The vast majority
of the time, discussions here are very polite and
well-reasoned, as I'm sure you'll find if you read
the group for a while!

Don't-be-put-off-by-the-occasional-heated-remark,
Greg



From ajung at sz-sb.de  Wed Jun 30 11:32:49 1999
From: ajung at sz-sb.de (Andreas Jung)
Date: Wed, 30 Jun 1999 15:32:49 GMT
Subject: looking for cookie.py
In-Reply-To: <7lcu6r$6p$1@minus.oleane.net>; from Mathieu Lecarme on Wed, Jun 30, 1999 at 01:15:19PM +0200
References: <7lcu6r$6p$1@minus.oleane.net>
Message-ID: <19990630173249.A15110@sz-sb.de>

On Wed, Jun 30, 1999 at 01:15:19PM +0200, Mathieu Lecarme wrote:
> the link on python web site doesn't work!
> where can i find it?
> 
ftp://ftp.bbn.com/pub/timo/python/Cookie.py


-- 
                                _\\|//_
                               (' O-O ')
------------------------------ooO-(_)-Ooo--------------------------------------
   Andreas Jung, Saarbr?cker Zeitung Verlag und Druckerei GmbH
   Saarbr?cker Daten-Innovations-Center
   Gutenbergstr. 11-23, D-66103 Saarbr?cken, Germany
   Phone: +49-(0)681-502-1563, Fax: +49-(0)681-502-1509
   Email: ajung at sz-sb.de (PGP key available)
-------------------------------------------------------------------------------



From drcursor at leirianet.pt  Mon Jun 28 00:42:30 1999
From: drcursor at leirianet.pt (DrCursor)
Date: 28 Jun 1999 04:42:30 GMT
Subject: New Site about East Timor
Message-ID: <930544952.196921@gorgoroth.esoterica.pt>

  Visite http://www.leirianet.pt/~drcursor , 
a nova pagina sobre Timor, 25 de Abril e
politica nacional em geral.
  Nesta pagina podera livremente deixar textos
, e/ou imagens que deseje ver on-line.

  See http://www.leirianet.pt/~drcursor, 
a new page about East-Timor, and politics...
  You can publish your own texts or images 
in this site.

drcursor at leirianet.pt
 



From webmaster at epls.com  Mon Jun 21 03:45:31 1999
From: webmaster at epls.com (webmaster at epls.com)
Date: Mon, 21 Jun 1999 15:45:31 +0800
Subject: Problems with CGI, genCGI and a multipart form
Message-ID: <199906210745.PAA00722@yangtze.dev.epls.com>

On Jun 17, 1999, peter_carl at my-deja.com wrote:

>I'm having trouble getting cgi.FieldStorage to read data posted as
>multipart/form-data.

>Right now I am testing a script using stdin piped from a test file.  The
>test file has everything after the header from a posted form, e.g.,:
>
>    -----------------------------172981055621895
>    Content-Disposition: form-data; name="section"
>
>    production
>    -----------------------------172981055621895
>    Content-Disposition: form-data; name="title"
>
>    This is the title
>
>etc...
>
>Environment variables for Content-type, Content-Length, and
>REQUEST_METHOD (POST) are set correctly.

1. Did you end the multipart data with an additional "--" to your last boundary:

   -----------------------------172981055621895--
   
   This is required to end the data portion correctly.

2. BTW, is there any reason you test your cgi script with a file via stdin 
   (plus somehow set your environment vars)? Isn't it simpler to just set up
   a httpd on your development host to test your scripts? Preparing the http headers
   and data file manually for testing is, in our opinion, error-prone.

3. We have used multipart POST cgi (with cgi.py) without problem in our 
   application. We always test directly with a httpd.

yours-common-gateway-ly

alan kang
http://www.epls.com - free groupware online




From KAnton at gmx.de  Thu Jun 17 12:54:36 1999
From: KAnton at gmx.de (Konrad Anton)
Date: Thu, 17 Jun 1999 18:53:36 +0159
Subject: HTML bettering?
Message-ID: <slrn7mia0g.18q.KAnton@casimbo.minus1.de>

Spam detection software, running on the system "albatross.python.org", has
identified this incoming email as possible spam.  The original message
has been attached to this so you can view it (if it isn't spam) or label
similar future email.  If you have any questions, see
the administrator of that system for details.

Content preview:  Matej Cepl <Cepl at fpm.cz> wrote: >How to make as simple as
  tool which would make some simple operation >over HTML (e.g., headings numbering)?
   If I would use htmllib.py, I would >have to build HTMLformatter or I am wrong
   and it is easy? > Does it have to be a selfmade Python solution? If you just
   want to number your headlines (and if you speak Perl), have a look at WML[1]
   which combines embedded Perl, m4, MetaHTML, an HTML optimizer and five other
   passes your input script is fed into. It's under GPL and fairly well documented.
   [...] 

Content analysis details:   (6.5 points, 5.0 required)

 pts rule name              description
---- ---------------------- --------------------------------------------------
 4.5 L_SPAM_TOOL_13         L_SPAM_TOOL_13
 2.0 FH_DATE_IS_19XX        The date is not 19xx.
-0.0 NO_RELAYS              Informational: message was not relayed via SMTP
-0.0 NO_RECEIVED            Informational: message has no Received headers


-------------- next part --------------
An embedded message was scrubbed...
From: KAnton at gmx.de (Konrad Anton)
Subject: Re: HTML bettering?
Date: Thu, 17 Jun 1999 18:53:36 +0159
Size: 1512
URL: <http://mail.python.org/pipermail/python-list/attachments/19990617/28159bdd/attachment.mht>

From janz at easics.be  Fri Jun  4 04:35:41 1999
From: janz at easics.be (Jan Zegers)
Date: Fri, 04 Jun 1999 10:35:41 +0200
Subject: Tkinter problem in Python 1.5.2 on Solaris 2.5.1
Message-ID: <37578FDD.179459EE@easics.be>

Hello,

I have the following problem with Tkinter in Python 1.5.2.
  Python 1.5.2 (#6, Jun  4 1999, 09:42:04)  [GCC 2.8.1] on sunos5
  Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
  >>> import Tkinter
  >>> Tkinter._test()
  Traceback (innermost last):
    File "<stdin>", line 1, in ?
    File "./Lib/lib-tk/Tkinter.py", line 1947, in _test
      root = Tk()
    File "./Lib/lib-tk/Tkinter.py", line 886, in __init__
      self.tk = _tkinter.create(screenName, baseName, className)
  TclError: Can't find a usable init.tcl in the following directories: 
    



  This probably means that Tcl wasn't installed properly.

Note the whitespace between the two last test lines. The installation
is as follows:
  Tcl/Tk 8.1 with --enable-shared so we have libraries 
    /usr/local/lib/libtcl8.1.so and /usr/local/lib/libtk8.1.so
  Python 1.5.2 without any configure options but using *shared*
    in the Modules/Setup (and some extra modules, including Tk).

Setting TCL_LIBRARY and TK_LIBRARY doesn't change anything.

Any suggestions ?

Kind regards,
Jan
-- 
===================================================================
Jan Zegers                ===              Easics               ===
General Manager           ===  VHDL-based ASIC design services  ===
Tel: +32-16-395 601          ===================================
Fax: +32-16-395 619      Interleuvenlaan 86, B-3001 Leuven, BELGIUM
mailto:janz at easics.be              http://www.easics.com



From tavares at connix.com  Wed Jun  9 19:56:25 1999
From: tavares at connix.com (Chris Tavares)
Date: Wed, 09 Jun 1999 19:56:25 -0400
Subject: COM Question - QueryInterface?
References: <375C7D56.C9744E39@connix.com> <375fe0b1.1117810@news.freeserve.net>
Message-ID: <375EFF29.2C1D0FB7@connix.com>

Toby Dickenson wrote:

> Chris Tavares <tavares at connix.com> wrote:
>
> >I've got a (hopefully easy) question about python's COM support.
> >
> >I'm developing a COM object in C++ which implements the COM collection
> >idiom - an Item method that returns another object. The code in Python
> >I'm using is something like this:
> >
> >from win32com.client import Dispatch
> >drives = Dispatch("AmAccess.Drives")
> >drive = drives.Item("c:")
> >dir = drive.Item("some-area")
> >
> >On this last line, I get an AttributeError for Item.
> >
> >I think I know the reason why - the interface returned from Drives.Item
> >isn't a dual, it's an oleautomation compatible vtable interface.
> >IDispatch on that object is implemented separately. So, PythonCOM sees
> >the VT_UNKNOWN return type and wraps a PYIUnknown object around it. I
> >need to be able to do a QueryInterface for IDispatch.
>
> Your analysis sounds about right, this should give you the IDispatch interface
> returned by querying for IID_IDispatch:
>
> drive = drives.Item("c:")._oleobj_.QueryInterface(pythoncom.IID_IDispatch)
>
> A better approach might be to make the interface dual? It sounds like you have
> already tackled the painful bits of making your interface dual; implementing
> IDispatch (under IID_IDispatch) and sticking to the oleautomation compatible
> restrictions. Making it dual should be an easy step, and will avoid this and
> several different non-python problems too.
>
> Toby Dickenson

Toby,

Thanks. The reason that the interface in question isn't a dual is because my
object design depends quite heavily on having multiple interfaces. If you have
multiple duals, you end up with lots of problems with IDispatch, since you can't
safely implement more than one version of the same interface on a single object
(and lots of people have tried, REALLY HARD).

So what I'm doing is having multiple oleautomation compatible interfaces, and then
using a typelib-driven IDispatch implementation that I got from
http://www.sellsbrothers.com/tools/multidisp/index.html. This automagically merges
all my separate interfaces into a pseudo-object model. Really neat stuff, and
saves a TON of work.

Having tried that, I find that I now have access to IDispatch::Invoke. Well, what
I really wanted was win32com.client.dynamic.Dispatch. How do I get to the wrapped
IDispatch rather than the raw one?

-Chris





From sams at quinta.com  Thu Jun  3 20:09:15 1999
From: sams at quinta.com (Sam Schulenburg)
Date: Fri, 04 Jun 1999 00:09:15 GMT
Subject: newbie trouble with raw_input and idle
References: <37581eb4.2704370@netnews.worldnet.att.net>
Message-ID: <7j75f5$u49$1@nnrp1.deja.com>

In article <37581eb4.2704370 at netnews.worldnet.att.net>,
  druid at princeton.crosswinds.net wrote:
> If I enter the following code (or any other using raw_input for that
> matter) into idle's shell it works, when trying to run it from a file
> (new window, edit it, save it then run it, nada)
>
> def f():
>     x=raw_input("ladida:")
>     print x
>
> f()
>
> systems involved are a win98 and an nt4 machines, tcl version 8.0.5,
> Python 1.5.2. The same code works under the plain Python interactive
> shell, and pythonwin, but not idle. Any ideas?
>
> --------------------------
> Bull Run ACD Rescue
> Several low mileage cow dogs available, visit the "barking lot" at
> http://www.crosswinds.net/~bullrun/rescue.html
>
>

I just tried your sample under windows NT with IDLE 0.4 and it worked OK
Sam Schulenburg


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From mlh at swl.msd.ray.com  Thu Jun 24 12:54:21 1999
From: mlh at swl.msd.ray.com (Milton L. Hankins)
Date: Thu, 24 Jun 1999 12:54:21 -0400
Subject: Environment inheritance under Windows NT 4, Attempt 2
Message-ID: <Pine.SGI.3.95q.990623204139.149495A-100000@masox202>

Since I've not yet gotten a response to my first post, here's an
elaboration.

Windows NT Workstation 4.00.1381, SP5.

1.  Right click on My Computer, choose Properties, Environment tab.  Note
the value of the System Variable "windir" and press Cancel.  (On my
system, this is "C:\WINNT".)

2.  Open a Command prompt.

3.  Here's a transscript of my CMD session, you can follow along:

-----------------------------------------------------
C:\>echo %WINDIR%
C:\WINNT

C:\>python -c "import os; print os.environ['windir']"
C:\WINNT

C:\>set WINDIR=foo

C:\>echo %WINDIR%
foo

C:\>python -c "import os; print os.environ['windir']"
C:\WINNT

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

What is going on here?  I wrote a program using VC5 to extract a variable
(1) directly from environ and _environ and (2) by calling getenv().  All
three reported "foo".  (Source provided below.)

After rummaging through the Python source, it appears that os.environ is
provided through the posix module, which in turn gets the environment data
from the "extern char** environ".

Could this have something to do with Microsoft's run-time DLLs?  I noticed
that Cygwin B19 used to exhibit this behavior as well, but it seemed to go
away in B20.1.

Python picks up the environment correctly if I go into the Control Panel,
System, Environment, and click OK.  But I have to do this each time I log
in to NT.


----------------------------------------------------------------------
#include <stdlib.h>
#include <stdio.h>
#include <assert.h>
#include <string.h>

extern char** environ;

#define BIG 80


int envIndex(const char** envArray, const char* key) {
	int i;
	char keyPrime[BIG];

	strcpy(keyPrime, key);
	strcat(keyPrime, "=");


	for (i = 0; envArray[i] != NULL; i++) {
		if (strstr(envArray[i], keyPrime) == envArray[i]) {
			break;
		}
	}

	return i;
}


// Don't screw with my return value...
const char* myGetEnv(const char** envArray, const char* key) {
	int i;

	i = envIndex(envArray, key);

	if (envArray[i] == NULL) return NULL;
	else return (envArray[i]+strlen(key)+1);
}


void printenvMulti(const char* key) {
	char* s;

	s = myGetEnv(_environ, key);
	if (s == NULL) printf("%s is not in _environ\n", key);
	else printf("myGetEnv(_environ, \"%s\") --> %s\n", key, s);

	s = myGetEnv(environ, key);
	if (s == NULL) printf("%s is not in environ\n", key);
	else printf("myGetEnv(environ, \"%s\") --> %s\n", key, s);

	s = getenv(key);
	if (s == NULL) printf("%s is not available to getenv()\n", key);
	else printf("getenv(\"%s\") --> %s\n", key, s);
}


int main(int argc, char** argv) {
	assert(argv[1] != NULL);

	printenvMulti(argv[1]);
	return 0;
}
----------------------------------------------------------------------

--
              Milton L. Hankins              \\  ><> Ezekiel 18:21 ><>
Software Engineer, Raytheon Systems Company  //  <mlh at swl.msd.ray.com>
      http://amasts.msd.ray.com/~mlh         \\  RayComNet  7-225-4728






From aahz at netcom.com  Thu Jun  3 15:45:32 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Thu, 3 Jun 1999 19:45:32 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7j4jbe$dhh$1@cronkite.cc.uga.edu> <37564A02.F115B9C5@infercor.no> <045501beadf8$610e62e0$f29b12c2@pythonware.com>
Message-ID: <aahzFCro7w.E9v@netcom.com>

In article <045501beadf8$610e62e0$f29b12c2 at pythonware.com>,
Fredrik Lundh <fredrik at pythonware.com> wrote:
>
>sure looks like the "community" thinks that changing the
>language is more important that using it...

People doing Real Work don't talk.
-- 
                      --- Aahz (@netcom.com)

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

"NRE is a lot like cocaine. Feels great while you're on it, makes you
feel very special and talented, and you become highly annoying to those
who aren't on it."  -- Stef



From wtanksle at dolphin.openprojects.net  Sat Jun  5 01:47:53 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Sat, 05 Jun 1999 05:47:53 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7iqhkb$d7s@wiscnews.wiscnet.net> <7iu7ma$d33$1@cronkite.cc.uga.edu> <02hfonzsx2.fsf@rtp.ericsson.se>
Message-ID: <slrn7lheg7.civ.wtanksle@dolphin.openprojects.net>

On 04 Jun 1999 15:07:05 -0400, Kumar Balachandran wrote:

>>>>>> "Graham" == Graham Matthews <graham at sloth.math.uga.edu> writes:

>    Graham> Graham Matthews wrote in message
>    Graham> <7ik6mi$lbk$1 at cronkite.cc.uga.edu>...
>    Graham> You are envangelising (a not uncommon response when
>    Graham> someone remotely criticises Python). Stop evangelising and
>    Graham> start considering the technical issues involve (read (!) 
>    Graham> other posts for what those issues are).

>Here is some useful evangelizing (methinks). One of the irritating things
>about Python is the use of whitespace in syntax. Agreed, the code is
>readable without parentheses or braces, but why not have optional
>syntactic sugar such as

You're kidding, right?  I mean, you posted twice about this, but they were
both jokes.

Yes.  I'm sure you're making a satire on how many other people suggest
making Python exactly like some other language, to the point that it would
be just as good to program in that other language.

I don't think it's an amusing satire -- the other people are discussing a
technical issue, and you're discussing an esthetic one.

>if
>...
>elif
>...
>else
>...
>fi

(Urk.)

>def
>...
>...
>fed

(Turns away, covers mouth.)

>while
>...
>elihw or wend

(Technicolor yawn.)

>etc. It makes the language more elegant.

Oh.  It's hard to argue esthetics, but can I recommend that you find a
language which uses that type of syntax instead of talking about Python
like that?

Reversed words are a really, really sick practice.  Ugh (shudder).  I can
handle 'END' (as in Oberon), and I sorta like having an optional qualifier
on the terminator (as in Ada), but those unpronouncable (shudder) things...

Anyhow, Python is block-structured, and that's all you're gonna get.

>When I see code using indentation or blank likes to achieve blocking of
>constructs, it reminds me of an old language (FORTRAN I think it was
>called:-).

Really?  Why?  Fortran didn't use indentation to achieve blocking.
Neither did Cobol.  Both used indentation to help in parsing.  And for
that matter, Fortran didn't use whitespace indentation -- it used
punchcard indentation.  I don't know anything about Cobol, but I bet it's
the same basic way.

>The change is simple to achieve if backward compatibility is given to
>accomodate people that grew up with FORTRAN.

Ah, I see.  A troll.  Of course.  :-)

-- 
-William "Billy" Tanksley
Utinam logica falsa tuam philosophiam totam suffodiant!
   :-: May faulty logic undermine your entire philosophy!



From che at debian.org  Mon Jun 14 22:42:35 1999
From: che at debian.org (Ben Gertzfield)
Date: 14 Jun 1999 19:42:35 -0700
Subject: Created a Python module for ID3 MP3 tag manipulation
References: <yttaeu7z4o8.fsf@gilgamesh.cse.ucsc.edu> <c753dzv8tel.fsf@pc142.cosc.canterbury.ac.nz> <87g13vy05d.fsf@slip-32-100-55-159.ma.us.ibm.net>
Message-ID: <yttaeu2b2w4.fsf@gilgamesh.cse.ucsc.edu>

>>>>> "greg" == greg andruk <meowing at banet.net> writes:

    greg> The fix for this is to treat your data as an MPEG stream,
    greg> not a simple file.  If the marker appears inside the bounds
    greg> of a frame, then it's not a tag.  This does imply walking
    greg> through the whole file to parse headers, but that's a much
    greg> faster operation than one might think.

Is there a simple way to do this? What kind of scanning does it
involve?

-- 
Brought to you by the letters Y and P and the number 2.
"Porcoga daisuki!"
Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/



From jim at digicool.com  Tue Jun 29 07:27:29 1999
From: jim at digicool.com (Jim Fulton)
Date: Tue, 29 Jun 1999 07:27:29 -0400
Subject: Extension Class patch for 1.5.2 (and C++)
References: <gregc.930618999@cgl.ucsf.edu>
Message-ID: <3778ADA1.2F1FFA12@digicool.com>

"Greg Couch %CGL" wrote:
> 
> Enclosed is a patch file for Jim Fulton's Extension Class package that
> makes it work with the new PyTypeObject structure in Python 1.5.2.
> I've also included changes to ExtensionClass.h that make it work with
> C++.  The source that the patch is based on is the 1.2 release that
> is available from http://www.digicool.com/releases/ExtensionClass/.

I'm very sorry, you must have started from a pretty old version of ExtensionClass.
Unfortunately, the information at: http://www.digicool.com/releases/ExtensionClass/
was sorely out of date. I just corrected this.  The current version of ExtensionClass
is included in the Zope distribution, www.zope.org and works fine with Python 1.5.2.

Jim

--
Jim Fulton           mailto:jim at digicool.com   Python Powered!        
Technical Director   (888) 344-4332            http://www.python.org  
Digital Creations    http://www.digicool.com   http://www.zope.org    

Under US Code Title 47, Sec.227(b)(1)(C), Sec.227(a)(2)(B) This email
address may not be added to any commercial mail list with out my
permission.  Violation of my privacy with advertising or SPAM will
result in a suit for a MINIMUM of $500 damages/incident, $1500 for
repeats.



From zigron at jps.net  Wed Jun 23 02:00:43 1999
From: zigron at jps.net (Zigron)
Date: Tue, 22 Jun 1999 23:00:43 -0700
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0rr2.2fi.behrends@allegro.cse.msu.edu>
Message-ID: <8L_b3.1213$4s.136166@typhoon-sf.snfc21.pbi.net>

Reimer Behrends <behrends at cse.msu.edu> wrote in message
news:slrn7n0rr2.2fi.behrends at allegro.cse.msu.edu...
> Preston Landers (planders at mail.utexas.edu) wrote:
> Personally, I can easily live without a construct that I can use on
> average once every three screens of code or so. On the other hand, I
> _am_ getting pretty sick of having to type self.whatever countless
> times, where pretty much any other language allows me to discard the
> "self." part. Not only is it annoying to type, it also reduces the
> readability of OO code a lot, much like the gratuitious use of
> punctuation characters as variable prefixes in Perl does.
>
> Among all the suggestions for Python 2.0, removing the necessity of
> prefixing methods and instance variables with "self." (and not having
> to list it as an argument, either) would be my #1 wish.

    Whyfor? I personally like the seperate scopes, beween the class
and the method. It makes it so I don't have to clean up as much. I can
use as many temporary variables as I want in a method, and as soon
as that method is gone...... its gone.
    The self.attribute thing makes sense. Each object has its own scope,
er, namespace, whatever. Methods are considered objects, as are
classes. Removing the 'self' requirement would, to me, change that.
Methods wouldn't be objects anymore. :) At least it seems so to me,
.. they'd  be almost-objects, maybe. I thought pv2 was going to
objectify eeeeverything. (I can't tell you how much I want strings
turned into objects! "blah".append('hi') should return 'blahhi'! :-))

    I think that would have further reprocussions, maybe,..maybe not,
dunno. I'm just a newbie.

--Stephen






From dkuhlman at netcom.com  Sat Jun 12 19:48:26 1999
From: dkuhlman at netcom.com (G. David Kuhlman)
Date: Sat, 12 Jun 1999 23:48:26 GMT
Subject: module import, but different
References: <000301beb508$6fcca140$ce9e2299@tim>
Message-ID: <dkuhlmanFD8nGq.n0A@netcom.com>

Tim Peters (tim_one at email.msn.com) wrote:


> Ack, enough already.  If your packages don't match your directories, the
> builtin importer simply doesn't do what you want.  Look in the Library
> manual's Built-in Functions section for __import__:  that's the builtin
> function an "import" stmt invokes.  You can replace it with whatever you
> like.  The std imp module packages many functions that will be helpful when
> doing this.  As I said at the start, I don't understand this stuff (never
> had reason to use it!), but Greg Stein is widely reputed to have developed a
> much more flexible import mechanism; maybe DejaNews will turn up something
> useful.

> the-distutils-sig-may-also-be-your-friend-ly y'rs  - tim

If you decide to follow Tim's suggestion and implement your own
import, then search DejaNews for "imputil".  I'm not at a machine
where I have the URL, but if you can't find it, email me.

I've used imputil.py to modify the search behavior to
extract/import scripts from our own script library/database before
searching directories on disk.  imputil makes it easy enough to do
so that even I was able to use it -- a strong recommendation. 
Our special import seems to work, although I'll have to admit that
I have not made it "production grade" yet.

  - Dave







From movits at lockstar.com  Thu Jun 24 10:55:38 1999
From: movits at lockstar.com (Mordy Ovits)
Date: 24 Jun 1999 07:55:38 PDT
Subject: Thanks for Python...
References: <376ead64@anonymous.newsfeeds.com> <VrDQRddcbbs9-pn2-QdftOXYk6Hdd@Fortress> <37716901.43CB604@lockstar.com> <slrn7n2gkr.i3k.wtanksle@dolphin.openprojects.net>
Message-ID: <37727194.D9CD6B7@lockstar.com>

William Tanksley wrote:
> Sure, but we've got Pilots, so that's not an open option.  Fortunately,
> the Pilot has Quartus Forth available for it -- it's not Python, but
> honestly, Forth fits the limitations of the device better.

I had a Pilot for a year before I realized that it wasn;t what I needed.  It's a
glorified organizer, when what I wanted was a pocket computer.  So I shelled out
for the E-100 and I've never looked back.  The difference is light years, not
miles.
 
> I used Quartus on my Crypto exam to solve a modular equation after I
> forgot how to solve modular equations (I had it use brute force).

Wow!  I wish I had had my E100 when I was in college.  I just realized how
awesome that would have been. :-)
 
> I'd still like to have Python, but dang, Quartus is good at what it does.
> Python and Forth -- for two different reasons, pretty much my favorite
> languages.

Best tool for the job wins.  Agreed.
 
> >o Mordy Ovits
> >o Cryptographic Engineer
> >o LockStar Inc.
> 
> >#!/usr/local/bin/python
> >from sys import*;from string import*;a=argv;[s,p,q]=filter(lambda x:x[:1]!=
> >'-',a);d='-d'in a;e,n=atol(p,16),atol(q,16);l=(len(q)+1)/2;o,inb=l-d,l-1+d
> >while s:s=stdin.read(inb);s and map(stdout.write,map(lambda i,b=pow(reduce(
> >lambda x,y:(x<<8L)+y,map(ord,s)),e,n):chr(b>>8*i&255),range(o-1,-1,-1)))
> 
> You know, the first time I saw this it was in ESR's .sig, and I tried to
> remember how long the Perl version had been.  A few minutes of searching
> revealed that it was three lines long -- but also revealed that it was
> using bc rather than Perl to do most of the work!  I spent about half an
> hour searching for the real Perl version, only to find that it was five
> lines long.
> 
> The Python for this .sig is shorter than the Perl!  Cool, eh?

The SECOND time I saw that sig was in ESR's mail.  The first was mine.  I doubt
ESR copied mine, but who knows? :-)
Actually there IS a 3 line PERL version, but it does use dc for the heavy
lifting.  I consider that cheating.

Python *is* my favorite language,
Mordy


-- 
o Mordy Ovits
o Cryptographic Engineer
o LockStar Inc.
---------------------------------------------------------------------------
#!/usr/local/bin/python 
from sys import*;from string import*;a=argv;[s,p,q]=filter(lambda x:x[:1]!=
'-',a);d='-d'in a;e,n=atol(p,16),atol(q,16);l=(len(q)+1)/2;o,inb=l-d,l-1+d
while s:s=stdin.read(inb);s and map(stdout.write,map(lambda i,b=pow(reduce(
lambda x,y:(x<<8L)+y,map(ord,s)),e,n):chr(b>>8*i&255),range(o-1,-1,-1)))



From scott at chronis.pobox.com  Tue Jun  1 22:05:16 1999
From: scott at chronis.pobox.com (scott cotton)
Date: Tue, 1 Jun 1999 22:05:16 -0400
Subject: When I run a module written in C, core dumps at Python exit.
In-Reply-To: <000b01beac96$1262f6a0$d89e2299@tim>; from Tim Peters on Tue, Jun 01, 1999 at 09:20:16PM -0400
References: <37529504.669DE3D8@erols.com> <000b01beac96$1262f6a0$d89e2299@tim>
Message-ID: <19990601220516.A608@chronis.pobox.com>

On Tue, Jun 01, 1999 at 09:20:16PM -0400, Tim Peters wrote:
| [Edward C. Jones]
| > I wrote an extension module written in C. It seems to work OK.
| > But core dumps when exiting Python. When I run the code from the
| > interpreter, Python works fine until I try to exit with ^D. Then
| > there is a core dump. What are the most common causes of such
| > problems?
| 
| Buggy extension modules written in C.
| 
| suspecting-that's-obvious-ly y'rs  - tim

More specifically <0.314 ;>, since python core dumps on exit,
it is most likely a deallocation function that is freeing
memory that it shouldn't.  Of course, that may be caused by
almost anything in the extension module.

scott



From gerrit.holl at pobox.com  Sun Jun 20 15:00:52 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sun, 20 Jun 1999 21:00:52 +0200
Subject: mysql vs pickle
Message-ID: <19990620210052.A7187@optiplex.palga.uucp>

Hello,

I recently downloaded the mysql module, but the pickle module seems to be
much easier!

So why use mysql?

groeten,
Gerrit.

-- 
The Dutch Linuxgames homepage:	http://linuxgames.nl.linux.org
Personal homepage:		http://www.nl.linux.org/~gerrit/

Discoverb is a python program in Dutch and English which tests the words you
learned by asking it. Homepage: http://www.nl.linux.org/~gerrit/discoverb/



From Jose.A..Antich at p98.f112.n480.z2.fidonet.org  Wed Jun 30 03:31:14 1999
From: Jose.A..Antich at p98.f112.n480.z2.fidonet.org (Jose A. Antich)
Date: Wed, 30 Jun 1999 08:31:14 +0100
Subject: Python Editor
Message-ID: <000015bb@bossar.com.pl>

From: "Jose A. Antich" <jantich at cantv.net>

   Does anyone know if there's any Python Editor for the MAc?

Thank's



From hongkongmail at iname.com  Sun Jun 13 15:00:01 1999
From: hongkongmail at iname.com (HANG CHEONG)
Date: 13 Jun 1999 19:00:01 GMT
Subject: A new successful marketing:
Message-ID: <7k0v3h$bob$11125@hfc.pacific.net.hk>

              HANG CHEONG INTERNATIONAL

A new successful marketing tactic to increase your sales :

    Do you always have problems to find channels to increase your sales?
The advertise in newspapers, magazines and doing some Direct Mailing, but
still can not achieve you goal, so why still dumping your money to a media
that could not help your sales and not using the most efficient and modern
way of selling stragies E-MAIL.Now many big companies are using this,
because it is economical, fast and efficient, and the other way is selling
by fax through the internets, and the result are remarkable.

  According to report pointed out that the ratio of internet selling is
1,000:1.5 that is 20,000,000 names x 0.0015 = 30,000.  These 30,000 clients
will buy from you and if your net profit is $50 that is to say your total
profit is:

           HK$50 x 30,000 = HK$1,500,000.  1.5 Million Dollars !!!

How to make your first 1 million in your life, now it is quite possible that
you can make your first 1 million within one month.  But the thing is ??Are 
you selling the right products, through the right media.??

***************************************************************************
A full set of sales tool with 20 million of client??s names will make you a SUPER SALES.  

A professional mail software (no need to go through any ISP Mail Server,making your computer
as the professional Mail Server, its super speed.  Easy to use.

* 500,000 E-mail addresses in Hong Kong
* 1,000,000 E-mail address in Taiwan
* 20,000,000 E-mail address in the World
 (latest revised on 1-6-1999.)

* A free revision one year (around 100,000 every month)

* Free of charge of 200,000 names of fax addresses in Hong Kong for 1999.

* Video Teaching  * Technical Support  * Software Update  * Data Update.

* Using CD-R dish and CD-AutoRun Menu.

(Only HK$680 for one full set) for unlimited use.

HK$680 is very small amount, you may not need it now but you will sure to use it when you 
have every thing ready.

***************************************************************************
If you need any E-mail address, call us any time!!
***************************************************************************
Don??t miss this chance, you may make lots of money with this new marketing technique.

*********************************************************************************************************
PLEASE TRANSFER THE MONEY $680(HK DOLLAR) TO: 

BANK NAME    : (HONG KONG) HANG SENG BANK. 
ACCOUNT NO   :  269-226981-882. 
ACCOUNT NAME :  WONG CHANG SANG. 

AFTER I GET THE MONEY WILL SEND THE E-Mail Marketing CD TO
YOU AT SOON(IN 7 DAY) 
**********************************************************************************************************

Booking Hotline : (852) 23488365 Mr Wong

2ND FL.FRONT BLOCK HING YIP HOUSE.24,SAI YEE ST,HK



From stefan_radke at csi.com  Fri Jun 25 04:17:16 1999
From: stefan_radke at csi.com (Stefan Radke)
Date: Fri, 25 Jun 1999 10:17:16 +0200
Subject: Does ODBC support 'tables()' function ?
Message-ID: <37733B0B.98127E7B@csi.com>

Looking for a database solution that is portable between Windows 95 and another
OS I thought to use the ODBC interface. I played around with several options,
including myQDBC for MySQL database, but a MS Access databse as well. Everything
went pretty smooth. Marc-Andre Lemburg's python module mxODBC works fine.

But now I have the idea to be able to find out which tables are available in a
database connected. When trying to use the <cursor>.tables() function I get the
following error message on an Access database:(db is database handle, c is
cursor):

> >>> c.tables('','','','')
> Traceback (innermost last):
>   File "<interactive input>", line 0, in ?
> NotSupportedError: ('S1C00', 84, '[Microsoft][ODBC Microsoft Access 97 Driver]
> Driver not capable ', 3383)
>

On mySQL databases I only get the number returned by 'c.tables('','','','').
Then, trying:

> >>> c.tableprivileges('','','')
> Traceback (innermost last):
>   File "<interactive input>", line 0, in ?
> NotSupportedError: ('IM001', 0, '[Microsoft][ODBC Driver Manager]
> Driver does not support this function', 3431)
>
How can I get the names of the tables ???

I spent almost a whole day in the MSDN Library to get information on the ODBC
interface implemented; I got the impression that it should work ... I checked
the ODBC Drivers installed; all are version 3.5.xxx .

On my crusade through the web I came accross a post in this newsgroup referring
to My SQL.pyd together with a DLL which now gives me direct access to MySQL
databases. Fine, and are there similar Python modules for other OSs ???

Many questions at once, but a wrapup of several days od trial.
I will be happy about answers !

Thank you, Stefan.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: stefan_radke.vcf
Type: text/x-vcard
Size: 162 bytes
Desc: Card for Stefan Radke
URL: <http://mail.python.org/pipermail/python-list/attachments/19990625/370760a7/attachment.vcf>

From gherman at my-deja.com  Wed Jun 16 07:19:28 1999
From: gherman at my-deja.com (Dinu C. Gherman)
Date: Wed, 16 Jun 1999 11:19:28 GMT
Subject: PPSI's "Quick Ref." and Lutz' "Pocket Ref."
Message-ID: <7k817q$hk8$1@nnrp1.deja.com>

Having never seen PPSI's "Python Quick Reference" as advertised
on pythonpros.com on paper I wonder how it compares to Mark Lutz'
"Python Pocket Reference"? I understand the former is much shorter,
but I wonder if it will be continued, filling a different niche,
updated to 1.5.2 (if need be) or swallowed one day by O'Reilly,
too? I think they are already selling or giving away reference
cards for GNU Emacs and the like.

Thanks,

Dinu

---
Dinu C. Gherman       :  Women inspire us to great things and
                      :  prevent us from achieving them. (Dumas)
................................................................
LHS International AG  :  http://www.lhsgroup.com
8050 Zurich           :  http://www.zurich.ch
Switzerland           :  http://pgp.ai.mit.edu
                      :  mobile://49.172.3060751
......................:.........................................
Vote against SPAM on http://www.politik-digital.de/spam



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From hniksic at srce.hr  Sun Jun  6 09:09:39 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: Sun, 6 Jun 1999 13:09:39 GMT
Subject: Deleting characters from a string
In-Reply-To: "Hans Nowak"'s message of "Sat, 5 Jun 1999 23:45:38 +0200"
References: <199906052144.XAA15045@axil.hvision.nl>
Message-ID: <87so85zd9o.fsf@pc-hrvoje.srce.hr>

"Hans Nowak" <ivnowa at hvision.nl> writes:

> Hmm... how about
> 
> host = filter(lambda c: not c in ' "', host)
> 
> Depends on how much you like, or hate, functional idiom, though. :)

I adore functional idioms (I also use the various Lisps), but I don't
adore their slowness in Python so I avoid using them.



From tbryan at arlut.utexas.edu  Sun Jun 13 00:49:41 1999
From: tbryan at arlut.utexas.edu (Thomas Bryan)
Date: Sat, 12 Jun 1999 23:49:41 -0500
Subject: Python on HP-UX
References: <zTD83.1322$VL2.213635@news.direcpc.com>
Message-ID: <37633865.632B662@arlut.utexas.edu>

Emile van Sebille wrote:

> Is there a
> way to get around this, or might binary executables be available
> somewhere someone could point me to? 

I don't know anything about HP-UX or its versioning scheme, 
but there are a couple of HP-UX builds listed on mxCGIPython.

http://starship.python.net/~lemburg/mxCGIPython.html

If one of them works, then you'll at least be able to start 
working with Python until your Sys Admin comes back.

-------------------------------------------
Tom Bryan
Applied Research Laboratories
University of Texas at Austin



From geek+ at cmu.edu  Fri Jun 11 10:39:51 1999
From: geek+ at cmu.edu (Brian E Gallew)
Date: 11 Jun 1999 10:39:51 -0400
Subject: Python as a shared library on Unix
References: <376102C9.5648@mailserver.hursley.ibm.com> <m3vhcuam2b.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <emacs-smtp-532-14177-8119-350099@export.andrew.cmu.edu>

Then <mwh21 at cam.ac.uk> spoke up and said:
> Paul Duffin <pduffin at mailserver.hursley.ibm.com> writes:
> 
> > Is it possible to build Python itself as a small executable and a
> > shared library, rather than an archive library which of
> > course results in a monolithic executable. 

Sure it is.  I've done it.  The problem is, doing this is even more
platform dependent than just building shared libraries.  When I did
it, it was a "by-hand" process, where I manually modified some files,
forced all object files to be relocatable code, manually converted
libpython.a to libpython.so.1.5.2, and so on.  What makes this
difficult is that some Modules want to reference things in libpython,
and sometimes it means linking the modules against -lpython, and
sometimes not.  It's ugly, and not for the faint of heart.  Further,
the utility of such a hack quickly became questionable for me,
considering that I'm never embedding python itself.

-- 
=====================================================================
| JAVA must have been developed in the wilds of West Virginia.      |
| After all, why else would it support only single inheritance??    |
=====================================================================
| Finger geek at cmu.edu for my public key.                            |
=====================================================================
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 266 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/19990611/1c1095d3/attachment.sig>

From fdrake at cnri.reston.va.us  Mon Jun 21 16:56:43 1999
From: fdrake at cnri.reston.va.us (Fred L. Drake)
Date: Mon, 21 Jun 1999 16:56:43 -0400 (EDT)
Subject: Python curses in non-Unix environments
Message-ID: <14190.42763.51741.190064@weyr.cnri.reston.va.us>

  I'd like to determine if the Python curses module has been
successfully used in non-Unix environments.
  I understand that various ports of the curses library have been made 
for DOS and OS/2 systems.  If Python has been used with these, or if
you have non-Unix systems to add to the list, I'd like to hear about
it.
  Thanks!


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From digitome at iol.ie  Sat Jun 19 05:05:53 1999
From: digitome at iol.ie (Sean Mc Grath)
Date: Sat, 19 Jun 1999 09:05:53 GMT
Subject: simple text file 'parsing' question
References: <376B1AAC.19FE8BCE@mysolution.com> <d%Ga3.498$dR.35932@newsfeed.avtel.net>
Message-ID: <376b5b51.1570328@news.iol.ie>

Unfortunately, Eudora's .mbx files are not consistent
in how a message starts. The sentinel string is always
the same. from memory it is something like
	"From ???@@???"

The problem is that it sometimes occurs in the middle of a line.
As long as you allow the sentinel to occur anywhere on a line
and keep the bit to the left of the sentinel, you can skip from
their to the blank line -- it will be all headers.

BTW, two weeks ago a new programmer with two years
college joined my company. No experience in Python. No
experience in XML. Two weeks later he has:-

	A python parser for Eudora .mbx mail archives that
	uses rfc822.py to tease out the headers
	An XML transformation script in Python
	Used Python reporting scripts to gelp create
	a DTD for rfc822 e-mail
	The beginnings of a down-translate to Folio Views
	in Python.

Does this language make programmers productive or what!!!!!



On Fri, 18 Jun 1999 23:44:11 -0700, "Phil Mayes"
<nospam at bitbucket.com> wrote:

>KP wrote in message <376B1AAC.19FE8BCE at mysolution.com>...
>>Here's my dilema: a directory filled (200+) with small emails. My goal
>>is to strip all the headers and combine them into one file. I can read
>>all the files just fine and write them all to one file, but I cannot
>>discern how to strip the headers. The answer must be very simple, yet
>>I cannot see it. Can anyone give a few pointers on how to do it, our
>>what module might be best? Thank you.
>>Ken
>
>
>A raw email always has a blank line between the header and the body.
>(To be pendantic, it should also have all its lines ending in CRLF.)
>So you can read it in and find the gap by looking for 2 EOLs:
>
>import string
>f = open('c:\\apps\\eudora\\in.mbx', 'r')
>all = f.read()
>x = string.find(all, '\n\n')
>body = all[x+2:]
># append body to output file
>--
>Phil Mayes    pmayes AT olivebr DOT com
>
>
>
>




From cmyanko at bigfoot.com  Thu Jun 24 13:35:32 1999
From: cmyanko at bigfoot.com (Curtis Yanko)
Date: Thu, 24 Jun 1999 17:35:32 GMT
Subject: Default printing behavior for classes
References: <7ktoe5$bbl$1@nnrp1.deja.com>
Message-ID: <7ktq93$c7i$1@nnrp1.deja.com>

In article <7ktoe5$bbl$1 at nnrp1.deja.com>,
  Ben Glazer <glazer at scicomp.com> wrote:
> I'm looking for a way to set default printing behavior for a class
I've
> defined.  Lists are printed in a certain format by default.  I'd like
to
> be able to define this for any type of class.  It'd be really nice if
> there were something I could include in my class definition, e.g.
>
> 	class Board:
> 	    def __init__(self, width=2, height=4, length=20):
> 		self.width = width
> 		self.height = height
> 		self.length = length
>
> 	    def __print__:
> 		print "[%s,%s,%s]" % (width,height,length)
>
> such that
>
> 	b = Board(8,8,2)
> 	print b
>
> would print "[8,8,2]".
>
> Is there anything in Python that encapsulates this functionality, or
> should I resign this desire to a feature req for Python 2.0?
>
> Thanks,
> Ben
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
>

Isn't that what __repr__ is for?
--
-Curtis Yanko


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From wtanksle at dolphin.openprojects.net  Tue Jun 22 21:02:45 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Wed, 23 Jun 1999 01:02:45 GMT
Subject: YAS to the "Reading line-by-line" Problem
References: <Pine.SUN.3.95-heb-2.07.990622183356.3454A-100000@sunset.ma.huji.ac.il> <376FDF70.4756AD38@mindspring.com>
Message-ID: <slrn7n0chl.gh6.wtanksle@dolphin.openprojects.net>

On Tue, 22 Jun 1999 15:09:36 -0400, Jesse D. Sightler wrote:
>Ack is raising an exception for a normally occuring event really a
>_good_ idea?  Generally exceptions are only used for things that
>actually are error conditions, and it would seem to me that using them
>as nothing more than a fancy flow-control construct is a sure path to
>nasty spaghetti code.

Exceptions may be used for errors, but they're actually intended for
_exceptions_ to normal operation.

Normal operation for a function named "readline" is to return the next
char in the file and advance the file pointer.  There are two exceptions
to this behavior: when the file pointer is invalid, and when it can't be
advanced.  Both should be handled by exception, not by returning a special
value.

-- 
-William "Billy" Tanksley
Utinam logica falsa tuam philosophiam totam suffodiant!
   :-: May faulty logic undermine your entire philosophy!



From jim at interet.com  Wed Jun 23 12:01:41 1999
From: jim at interet.com (James C. Ahlstrom)
Date: Wed, 23 Jun 1999 12:01:41 -0400
Subject: Build the Python library into the binary
References: <376F9518.D3252CEE@interet.com> <TMPb3.1886$U6.6789@newsr2.twcny.rr.com>
Message-ID: <377104E5.23D76E86@interet.com>

Darrell wrote:
> 
> I thought these guys were working on these problems.
> http://www.python.org/sigs/distutils-sig/

I was not aware of this.  Thanks for pointing it out.

> This also seems close to what you want.

Close but not quite.

Jim Ahlstrom



From mcfletch at home.com  Thu Jun 10 22:08:23 1999
From: mcfletch at home.com (Mike C. Fletcher)
Date: Thu, 10 Jun 1999 22:08:23 -0400
Subject: Broken 1.5.2 installation? Missing cw3215mt.dll
References: <000101beb3a7$2ea18440$329e2299@tim>
Message-ID: <014501beb3af$48f2e840$a9807018@yec1.on.wave.home.com>

Problem is encountered when loading command-line Python or IDLE.  IDLE is
non-functional, command line works fine after acknowledging the missing DLLs
(as far as I can see).  I'm beginning to think there must be something
screwy with my configuration that messes with 1.5.2 where it didn't with 1.5
(which I believe was the last version on this machine).  Could be the Cygnus
TCL is interfering I suppose.

Oh, of course most people would have figured it out by my calling it a DLL,
but this is on NT with the standard win32 installer from www.python.org .

Pythonwin works fine (no error messages at all).  Python running a file as a
script also works fine (no error messages).

Must get back to work one of these days :) , enjoy,
Mike
______________________________________
   M i k e   C .  F l e t c h e r

       mcfletch at vrtelecom.com

        Designer, Researcher

----- Original Message -----
From: Tim Peters <tim_one at email.msn.com>
To: Mike C. Fletcher <mcfletch at home.com>; <python-list at python.org>
Sent: Thursday, June 10, 1999 21:10
Subject: RE: Broken 1.5.2 installation? Missing cw3215mt.dll
...
> BTW, *when* did this error pop up?  It's unclear.  That is, did the
> installation complete (in which case what does the error have to do with
the
> installation?), or did it occur during the install (in which case exactly
> when during the install)?  There's no mention of cw*.dll in either the
> Tcl/Tk or Python install logs left behind by my 1.5.2 installation.  Makes
> me wonder whether the installer itself needs this DLL.
...




From fredrik at pythonware.com  Thu Jun  3 14:41:29 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 3 Jun 1999 18:41:29 GMT
Subject: problems using Tkinter
References: <928350546.726.80@news.remarQ.com>
Message-ID: <02fc01beadf0$b3c3ea30$f29b12c2@pythonware.com>

Gordon Vinther <gvinther at atex.com> wrote:
> I get the following error whenever I attempt to use Tk
> 
> Os AIX4.2.1
> Python 1.5.2
> Tk/TCL 8.0.1
> 
> from Tkinter import  *  .............this works
> root=Tk() ........... this errors with the following message
> 
> File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 886, in __init__
> self.tk = _tkinter.create(screenName, baseName, className)
> 
> TclError: invalid command name "tcl_findLibrary"

looks like tkinter is picking up the wrong Tcl/Tk
library files.   try setting the TCL_LIBRARY and
TK_LIBRARY environment variables.

</F>




From avv at quasar.ipa.nw.ru  Fri Jun 25 15:28:58 1999
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Fri, 25 Jun 1999 12:28:58 -0700
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773CA67.7057F797@quasar.ipa.nw.ru> <3773D55B.8C966174@spacenet.tn.cornell.edu>
Message-ID: <3773D87A.52FDF121@quasar.ipa.nw.ru>

Tom Loredo wrote:

> Well, one reason not to rush to wxWindows is that there is
> no "real" Mac version.  I realize that a Mac version is "underway",
> but the wxWindows site has said that for a *long* time now.
> A beta version was posted about a month ago, but I've not
> heard much about it, and most of the wx web site doesn't
> seem to know it exists (e.g., the FAQ still says the beta
> is forthcoming, and the dowload section appears to still
> have the old preview alpha release).  I develop on both Unix
> and Mac, and although the Unix part of me finds wx appealing,
> the Mac side of me feels like a 2nd class citizen in the wx world.

That's exactly I wished to append to my message, but hesistated because
I do not know Mac enough to judge on this. Well, this is true, but how
does it apply to those who have never seeen a Mac? :-) And probably
never see it in the future. At least until LinuxPPC gets mature enough.

Alexander






From petrilli at trump.amber.org  Wed Jun  2 19:48:09 1999
From: petrilli at trump.amber.org (Christopher Petrilli)
Date: 02 Jun 1999 16:48:09 PDT
Subject: Python & IRC?
References: <rNHvjEdhm5Pp-pn2-tyCQ80XVscVt@d185d18f8.rochester.rr.com> <7U853.1589$nn.478377@news.shore.net>
Message-ID: <7j4frp$4e9@journal.concentric.net>

Michael P. Reilly <arcege at shore.net> wrote:

> You might also want to look at the SocketServer module (which includes
> ThreadingTCPServer and ThreadingUDPServer classes).

I'll throw in my two cents which is that at some point I hacked out a quick
implementation of the protocol using asynchat, and probably could do so
again, but would rather just say use it :-)  It's really a great way to
do this, since IRC has no pipelining, there's really not a concern there,
and async sockets are REALLY low overhead---at least on any real OS :-)

Chris

-- 
| Christopher Petrilli                      ``Television is bubble-gum for
| petrilli at amber.org                          the mind.''-Frank Lloyd Wright



From doughellmann at mindspring.com  Mon Jun  7 09:48:54 1999
From: doughellmann at mindspring.com (Doug Hellmann)
Date: Mon, 07 Jun 1999 13:48:54 GMT
Subject: Looking for for some ideas for argument processing
References: <7jcdhc$htd$1@nnrp1.deja.com>
Message-ID: <375BCEA1.F5B59623@mindspring.com>

Several people have suggested getopt as a solution, and if you think it
will help, you might want to check out a class library I've written to
make it easier for me to use getopt.  Basically, you create a class to
define your application.  By defining methods of the form:

def optionHandler_<option_name>(self[, <arg_name>]):
	"""
	<help message for option_name>
	"""
	<process arg_name>

You can automatically specify which arguments your application calls,
and quickly identify which code processes which argument.  Your script
will also have automatic -h and --help arguments which produce usage
messages (short and long, respectively).  The optionHandlers are called
in order based on how the arguments come in on the command line, so if
your app takes "commands" as arguments, you option handlers can assume
that state is maintained.  Your options can also take multiple
arguments, as any strings with commas will be split into multiple args
before being passed into the option handler.

Check out CommandLineApp.py in:

http://www.mindspring.com/~doughellmann/Projects/pybox/pybox.r1_3.tar.gz

Let me know if you find it useful,
Doug


cmedcoff at my-deja.com wrote:
> 
> I'm writing a script with is going to make heavy use of script
> arguments.  I'd like to hear some ideas on some ways to support this.
> 
> One idea I've had so far is to write a class which is a dictionary
> of "directives" to function objects, where a "directive" might be '-
> c', '/f', or whatever.  This seems to work well if all arguments are of
> this simple form.  However I would also like to support arguments
> like "python myscript.py /a /f /x "bip" "bop", /g"; the point here
> being that "directives" may take arguments.  This kind of messes up the
> simple loop you'd otherwise use to process arguments.
> 
>         a = BuildArgumentProcessor()
> 
>         # process arguments
>         for x in sys.argv[1:]:
>             opt = x[1:2]
>             optArg = x[2:]
>             a.Process(opt, optArg)
> 
> This would work if somehow inside the for loop one could "increment
> this list iterator more that once" (sorry I'm coming from a C++
> background).  I don't see anyway of doing this.  Is it possible?  Any
> other ideas?
> 
> Regards,
> Chuck
> 
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.



From phd at emerald.netskate.ru  Sun Jun 20 11:57:04 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Sun, 20 Jun 1999 19:57:04 +0400 (MSD)
Subject: Questions?  Time Module?
In-Reply-To: <rNHvjEdhm5Pp-pn2-5b4dbg1xd7dt@d185d18f8.rochester.rr.com>
Message-ID: <Pine.LNX.4.04.9906201954561.25630-100000@emerald.netskate.ru>

On Sun, 20 Jun 1999, Benjamin Schollnick wrote:
> Sure, I can use the "IMPORT TIME" and access everything from there
> but heaven help me, if I can't find a TIME.PY or TIME.PYC file 
> anywhere...

   There is no time.py - time is not "stabdard" module - it is "builtin"
module. There is timemodule.c source file - not in Python, but in C!
   In python, low-level functions are implemented in C (actually, any
programming language do it this way, not only python).

> 			Benjamin

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From sz at zmail.ru  Mon Jun  7 14:08:44 1999
From: sz at zmail.ru (Zaur)
Date: Mon, 07 Jun 1999 22:08:44 +0400
Subject: To add OCCAM-like construction to Python  
Message-ID: <375C0AAB.CCE95F44@zmail.ru>

This message for people which working on Python 2.
Can Python "inherit" <par> statement from OCCAM for parallelizing.

For example:
   par
      <statement1>
      <statement2>
   <statement3>
where interpreter (internally) run <statement1> and <statement2> in
parallel
(for example, in different threads) , wait while they complete and then
execute <statement3>.

Then use something like:
   X = ?Y
where if Y is empty (None) current thread wait while Y become nonempty.
   X! = Y
where current thread wait while X is busy and then assign: X = Y

Or something else ...  in OCCAM-style.
Sintax in above statements may be adopt for Python, of cause.

I think and hope that it is easy for Python.
Python inherits many features from programming languages (C, Pascal,
...),
scripts (Perl, Tcl, ...), shells (Rexx, ...), object orientation. I
believe that it can also inherit parallel processing from OCCAM, which
also has clear semantics and easy sintax.











From mrkbutty at mcmail.com  Wed Jun  2 15:36:03 1999
From: mrkbutty at mcmail.com (Mark Butterworth)
Date: Wed, 2 Jun 1999 20:36:03 +0100
Subject: A tkinter kick-off
References: <37544a32.0@news1.mcmail.com>
Message-ID: <37558878.0@news1.mcmail.com>

Many thanks to all!

--

Regards, ]\/[r]{





From delgado at Mathematik.Uni-Bielefeld.DE  Fri Jun 18 07:24:21 1999
From: delgado at Mathematik.Uni-Bielefeld.DE (Olaf Delgado)
Date: Fri, 18 Jun 1999 13:24:21 +0200
Subject: Newbie: Truth values (three-valued logic)
In-Reply-To: <7kd3ke$d1n$1@nnrp1.deja.com>
References: <Pine.OSF.4.05.9906161106470.295-100000@pinax.Mathematik.Uni-Bielefeld.DE> <000201beb946$b9c99e00$969e2299@tim> <7kd3ke$d1n$1@nnrp1.deja.com>
Message-ID: <Pine.OSF.4.05.9906181317100.413-100000@pinax.Mathematik.Uni-Bielefeld.DE>

On Fri, 18 Jun 1999 mgutierrez at dss.es wrote:

> "Not maybe" equals to "not".
> Imagine this case:
> Maybe, I'll go to the cinema tomorrow.
> If you say, not "maybe, I'll go to the cinema tomorrow", then you are
> saying not to go the cinema. Imagine:
> Johan: Maybe, we could go to the cinema tomorrow.
> Anthony: No
> Johan : ok, let's go to the zoo, to see that python.

Okay, but this kind of intuitive semantics is not what I need here. The
negation of "Maybe, I'll go to the cinema tomorrow." should be "Maybe, I
wan't go to the cinema tomorrow.". By the way, this is IMO another,
stronger argument against hardwiring multi-valued logics into a
programming language. There may be different, equally justified views on
what the natural behaviour would be.

Undecidably yours,
Olaf

-- 
   ////         Olaf Delgado Friedrichs, Uni Bielefeld, Mathematik
   Olaf         Postfach 100131                  (room:  V4-109)
   `='          D-33501 Bielefeld, Germany       (phone: +49 521 106-4765)
                           http://www.mathematik.uni-bielefeld.de/~delgado




From mlh at idt.ntnu.no  Sat Jun 12 12:18:07 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 12 Jun 1999 18:18:07 +0200
Subject: Raw Strings!
References: <P3v83.7006$Yb7.1482@news4.mia> <19990612173928.A23568@humbolt.nl.linux.org>
Message-ID: <y0j7lp92y0w.fsf@vier.idi.ntnu.no>

"'Gerrit Holl" <gerrit at nl.linux.org>, '@humbolt.nl.linux.org writes:

> On Sat, Jun 12, 1999 at 10:30:16AM -0500, SunitJoshi wrote:
[...]
> > 
> > Hi All
> >         If someone could show me how to use raw-strings with variables, I
> > would really appreciate it.
> >         For e.g.: if str_path = "C:\\windows\\temp"
> > 
> 
> var = r"D:\\stupid\\windows\\path"
> 
> note the "r" of "raw" before the string. This is always the case.

I don't think this is what is wanted here... Try

  r"C:\windows\temp"

instead. The point with raw strings is that backslashes can be written
out without being escaped.

> 
> 9 minutes and i reply... hmm, not bad :-)

Well - not too good either, considering that you gave a misleading
answer <wink>

> 
> regards,
> Gerrit.
> 
> 

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From alrice at swcp.com  Thu Jun  3 16:08:21 1999
From: alrice at swcp.com (Alex Rice)
Date: Thu, 03 Jun 1999 14:08:21 -0600
Subject: PythonWin wish list
References: <7iv11t$nb0$1@nnrp1.deja.com>
Message-ID: <3756E0B5.955FB1A9@swcp.com>

cmedcoff at my-deja.com wrote:
 
> One of the nice things about python is the free development environment
> PythonWin.  I certainly don't get my other development environments for
> free.
> 
> Having used PyhonWin a couple of weeks now I have a couple of comments
> in the form of a wish list.  Comments are welcome.

Looks good, I agree :-)

If you register or buy support for PythonWin with Mark Hammond he would
probably pop those features right in there.

http://starship.python.net/crew/mhammond/
http://starship.python.net/crew/mhammond/Support.html

"Support will be limited to installation issues only. Once the software
has been installed, all people asking questions or for feature requests
etc will politely be referred to how they can become a registered user."



From perry at ac-copy.net  Fri Jun  4 02:33:15 1999
From: perry at ac-copy.net (Joachim Schmitz)
Date: Fri, 4 Jun 1999 06:33:15 GMT
Subject: build of python 1.5.2 with threads on SunOs 5.6 fails
In-Reply-To: <3756FACF.D62ABE2A@rubic.com>
References: <3756FACF.D62ABE2A@rubic.com>
Message-ID: <Pine.LNX.4.05.9906040828480.12456-100000@ns.ac-copy.net>

On Thu, 3 Jun 1999, Jeff Bauer wrote:

> Joachim Schmitz wrote:
> > I compiled Python 1.5.2 on SunOS 5.6/Solaris 2.6 without threads
> > successfully. But doing a configure --with-threads, and make results
> > in the following error:
> > .....
> 
> Joachim,
> 
> I'd guess this problem is caused by first configuring
> Python without threads.  Configure is probably using a
> stale config.status -- you can check Modules/Setup.thread.
> 
> A 'make clean' might be the easy way to fix your problem.
> 

That obviously didn?t remove libpython1.5.a. After removing it and doing a
make again, it works now.

thanks for the hint.

Joachim Schmitz




From matt at mondoinfo.com  Wed Jun  9 21:46:29 1999
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Wed, 09 Jun 1999 20:46:29 -0500
Subject: How do I defect?
References: <375EEBE2.3F20390F@lucent.com>
Message-ID: <matt-0906992046290001@scotch.mondoinfo.com>

In article <375EEBE2.3F20390F at lucent.com>, Burton Kent <burton at lucent.com>
wrote:

> I use Perl because of the nifty regex.  But
> I hate Perl because of the data structures.

Python's re module has Perl-like regexes so you should be happy.

> If you ever need any kind of data structure
> in perl, move into a padded cubicle.  Check
> into the funny farm. Flop into the loony bin.
> Perl is like LSD.  It's cool and pretty and
> will slowly drive you insane.

Agreed.

> So how do I defect from Perl to Python?  I 
> have the Programming Python book on order.  
> What can I read to get started ASAP?

Read

http://www.python.org/doc/current/tut/tut.html

(it's short). Then skim

http://www.python.org/doc/current/lib/lib.html

Then start writing code. Check

http://www.python.org/doc/FAQ.html

if you run into problems. I doubt you'll have any serious trouble getting
started.

Regards,
Matt



From tjreedy at udel.edu  Thu Jun 24 01:39:25 1999
From: tjreedy at udel.edu (Terry Reedy)
Date: 24 Jun 1999 05:39:25 GMT
Subject: How to obtain an instance's name at runtime?
References: <7kqpu8$7u1$1@nnrp1.deja.com> <37718424.C70BB9DE@geekspace.com>
Message-ID: <7ksgad$nvp$1@news.udel.edu>

In article <37718424.C70BB9DE at geekspace.com>,  says...
>
>"Dinu C. Gherman" wrote:
>> Question: How to obtain an instance's name at runtime?

rozzin at geekspace.com responded:
>        An *instance's* name? Instances have names?

and others similarly.

As near as I can remember, only three types of objects have definition 
names (in readonly? __name__ attributes): modules, functions, and 
classes.  All others are anonymous, though they may or may not have one 
or more namespace references.

While named objects usually get a namespace (reference) name equal to 
the definition name, there is no necessary connection.  Consider:

def f(): pass
g=f
del f

Now the function which internally thinks of itself as 'f' is referred to 
externally as 'g' and only as 'g'.

Terry J. Reedy




From jrush at summit-research.com  Wed Jun 23 06:43:42 1999
From: jrush at summit-research.com (Jeff Rush)
Date: 23 Jun 1999 10:43:42 GMT
Subject: Thanks for Python...
References: <376ead64@anonymous.newsfeeds.com>
Message-ID: <VrDQRddcbbs9-pn2-QdftOXYk6Hdd@Fortress>

Welcome to the Python community, John.  Yes, Python is a truly
wonderful language for cranking out quick prototypes or test jigs.
I prototyped a complete web-based general ledger system in Python,
figuring I'd downcode it to C for speed later, but so far I haven't needed
to.  Python is also just plain fun...  Now if I could just get time to port
it to the PalmPilot, I'd never be away from it.

-Jeff Rush



On Mon, 21 Jun 1999 21:30:44, Anonymous <nobody at newsfeeds.com> wrote:

> Dear Python Developers,
> 
> I must say how thankful I am for Python. It has saved me at work once so
> far.

<snip>

> This afternoon, at work (in about a 20 minute time period) I rewrote the
> mach server that I had originally written in Perl and it works much nicer
> and more reliably than the Perl version which I was using before. The server
> works quite nicely and was surprisingly written in a few less lines than the
> Perl server was written in. I hope that my boss is impressed.
> 
> Sincerely,
> John Brainard




From cmyanko at bigfoot.com  Wed Jun  9 09:33:47 1999
From: cmyanko at bigfoot.com (Curtis Yanko)
Date: Wed, 09 Jun 1999 13:33:47 GMT
Subject: "Programming Python" still worthwhile?
References: <7jki66$qam$1@news.gstis.net> <00c301beb260$09e94680$f29b12c2@pythonware.com>
Message-ID: <7jlqfq$jgs$1@nnrp1.deja.com>

Isn't PP due for 2nd edition soon?

--
-Curtis Yanko


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From Jon.K.Hellan at item.ntnu.no  Wed Jun  9 09:33:33 1999
From: Jon.K.Hellan at item.ntnu.no (Jon K Hellan)
Date: 09 Jun 1999 15:33:33 +0200
Subject: Windows 95/98: Python uses console, not stdio
Message-ID: <87r9nl33de.fsf@parus.itea.ntnu.no>

I am unable to run Python in a shell window in emacs on Windows
95/98. I use the bash shell, but don't think this makes any
difference.

If Python reads/writes from/to console, not stdio, this would be an
explanation. 

If my guess is right, would it be possible to switch to using stdio?

Jon




From bernhard at alpha1.csd.uwm.edu  Thu Jun 17 19:04:29 1999
From: bernhard at alpha1.csd.uwm.edu (Bernhard Reiter)
Date: 17 Jun 1999 23:04:29 GMT
Subject: DLL prob win32com, old, resolved
References: <slrn7gr6o1.7kb.bernhard@alpha1.csd.uwm.edu> <7em639$fto$1@m2.c2.telstra-mm.net.au> <370F78E2.8EA7433E@connix.com>
Message-ID: <slrn7mivnr.tpi.bernhard@alpha1.csd.uwm.edu>

way back on the 9th of April, I reported the following problem:

| Get strange errors, when using PythonWin (win32all-124)
| with Python 1.5.2b2 or 1.5.2c1 on this Windows 95 machine here.
| The installprogram said that
|         AXScript, Python Interpreter and Python Dictionary
| couldn't be registered. Should be registered manually.
| 
| When importing COM stuff, I get:
| import pythoncom
| Traceback (innermost last):  File "<interactive input>", line 1,
| in ?ImportError:
| DLL load failed: A device attached to the system is not functioning.
| 
| Reading documentation I know about possible old COM dlls as noted on
| http://starship.python.net/crew/mhammond/bigpond/Downloads.html
| but my error message seems to be different.

I updated a few Dlls and the one's that Mark suggested, but it didn't help.
Now I had the same problem on a different machine again and
Chris Tavares <tavares at connix.com> in <370F78E2.8EA7433E at connix.com>
was right (on the 10th of April):

| Another option is to download DCOM for Win95 - that'll get the user up to date
| and then some!

I tried that this time and it worked! Hurray!
Part of the problem, why I didn't tried that the last time was, that
I couldn't find DCOM on the support.microsoft.com site. (The search
enginge is really bad.)
Stupid me, it is no support it can be reached from: 
	http://www.microsoft.com/msdownload/
search for "DCOM95".

Problem could be solved with that, finally.
Mark: you might want to add that hint for other desperate users now.
I recall at least one additional question regarding that.

	Bernhard
-- 
Research Assistant, Geog Dept UM-Milwaukee, USA.         (www.uwm.edu/~bernhard)
Funding/Promoting Association for a Free Informational Infrastructure (ffii.org)



From faassen at pop.vet.uu.nl  Mon Jun  7 11:15:58 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Mon, 07 Jun 1999 17:15:58 +0200
Subject: Python CPU
References: <14171.17113.39933.904980@lrz.de>
Message-ID: <375BE22E.CF5AC82D@pop.vet.uu.nl>

Eugene Leitl wrote:
> 
> Have there been any attempts in creating an ASIC capable of
> interpreting Python opcodes directly? Of course, with
> soon-to-arrive Java chips and JPython there might be not
> that much need for it...

Except for the April 1 breakthrough at Stichting Mathematisch Centrum I
posted about on April 1 (also featuring high speed nanovirus resistant
whitespace circuitry :), I don't think anyone's worked on something like
that. I think the consensus is that the Python opcodes are too high
level to make this feasible.

Regards,

Martijn



From graham at sloth.math.uga.edu  Fri Jun  4 10:37:32 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 4 Jun 1999 14:37:32 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net> <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net> <7j34l3$hbj$1@nnrp1.deja.com> <7j66b7$il5$2@cronkite.cc.uga.edu> <3756C693.2C910B29@usa.net> <7j6jnl$58f$3@cronkite.cc.uga.edu> <3757A820.CE3F195F@usa.net>
Message-ID: <7j8obc$55v$2@cronkite.cc.uga.edu>

Salvador =?iso-8859-1?Q?Fandi=F1o?= (fandino at usa.net) wrote:
: GC usually deletes objects in random order which is not a very good
: behavior.
Graham Matthews wrote:
> I am sorry but this comment is all myth.
Salvador =?iso-8859-1?Q?Fandi=F1o?= (fandino at usa.net) wrote:
: No, it's a real problem I had in my current project. You can't delete
: complex data structures just deleting component objects in random order.

The proposal I gave for adding mark sweep collection on top of reference
counts will delete non circularly referred objects in precisely the order
they are now deleted in current Python. That order doesn't appear to
random for you, so I don't understand the above comment? As for circularly
referred data it will be deleted in the order determined by the collector,
but that's better than current Python which doesn't delete it at all.

Graham Matthews wrote:
> Why is it so hard for people to see that adding a mark sweep collector
> *ON TOP OF* the existing ref count scheme essentially changes nothing.
> All the current ref count behaviour that people love will stay. The
> only additional behaviour will be that circular refs are mopped up
> correctly.
Salvador =?iso-8859-1?Q?Fandi=F1o?= (fandino at usa.net) wrote:
: BTW, usual __del__ methods free resources like sockets, files, windows,
: etc. but they don't break circular references so calling __del__ will
: not free any object at all. 

Yes I know this. My point was that when the collector collects a circularly
referred object it can still call its __del__ *finaliser* just like now.
I was not suggestinf that __del__ be used to break circular refs.

graham
-- 
                  A cat named Easter
              He say "will you ever learn"
              Its just an empty cage girl
                 If you kill the bird



From holger at phoenix-edv.netzservice.de  Wed Jun 30 10:11:28 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Wed, 30 Jun 1999 14:11:28 GMT
Subject: Why must path be in args in os.execv?
Message-ID: <377A2590.D95D72FE@phoenix-edv.netzservice.de>

High,

there's again something I don't understand. Hope you do.;-)

I'd like to start a windows-executable with some arguments.
Ok. 'os.execv' or anything like that is the thing.

But if I don't repeat the path with the arguments, the execution fails.
Better: Something happens, but not the correct execution.

For testing I constructed an easy batch named test.bat with only one command:
echo %0 %1 %2 %3 %4 %5 >result.log

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import os
>>> os.chdir("c:\\mypython")
>>> os.execv("test.bat",["test.bat","1","2","3","4"])

When I start with:
>>> os.execv("test.bat",["1","2","3","4"])
only the interpreter ends. No Log-File has been written...

I really spent a long time since I found out. Perhaps the documentation
have to be changed?
>>>
execv (path, args) 
Execute the executable path with argument list args, replacing the current process (i.e.,
the Python interpreter). The argument list may be a tuple or list of strings.
Availability: Unix, Windows. 
>>>

Ciao,
Holger



From matomira at iname.com  Sat Jun 12 13:08:38 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Sat, 12 Jun 1999 19:08:38 +0200
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com>
Message-ID: <37629416.79DE9E4A@iname.com>


Fernando Mato Mira wrote:

> [For example, I can think the Eiffel syntax is `philosophically' broken, but
> as `infix' syntaxes go, it is not. And it's a good language (I programmed

Actually, it's a _great_ language to learn OO (although of course in the restricted
single-dispatch way most people know about. I like to call that "Subject Oriented").

Occam taught me about concurrency.
Eiffel taught me OO.
CLOS taught me what OO really means.

--

"Object Oriented is to not think only about your `subject' and forget about your
`arguments' - the syntactic objects of your sentence".

Fernando D. Mato Mira
Real-Time SW Eng & Networking
Advanced Systems Engineering Division
CSEM
Jaquet-Droz 1                   email: matomira AT acm DOT org
CH-2007 Neuchatel                 tel:       +41 (32) 720-5157
Switzerland                       FAX:       +41 (32) 720-5720

www.csem.ch      www.vrai.com     ligwww.epfl.ch/matomira.html





From hniksic at srce.hr  Mon Jun  7 16:38:08 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 07 Jun 1999 22:38:08 +0200
Subject: waste of resources ?
References: <375C0F91.1C73E4E3@icrf.icnet.uk> <375C16A9.5ADC4990@icrf.icnet.uk> <375C2003.B01BEC3@arlut.utexas.edu>
Message-ID: <87yahvzqz3.fsf@pc-hrvoje.srce.hr>

Bryan VanDeVen <bryanv at arlut.utexas.edu> writes:

> On sysV or Unix98 systems you may set the disposition of SIGCHLD to
> SIG_IGN (i.e., ignore death or a child signals) The watertight
> method is to catch SIGCHLD and have the handler call waitpid(...) 
> with WNOHANG in a loop.

Except that when a signal arrives during an IO operation, it will be
canceled due to the EINTR brain-damage, and Python will throw an
exception.  So if you expect your program to accept SIGCHLD signals at
random intervals, you'd better encase all your IO operation in a loop
like this:

while 1:
    try:
        read(...)             # or whatever
    except (IOError, os.error), detail:
        if detail.args[0] == errno.EINTR:
            continue
        else:
            raise

This problem with signals has been my long-standing gripe with Python.



From neelk at brick.cswv.com  Thu Jun 24 22:09:19 1999
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 24 Jun 1999 21:09:19 -0500
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <m3aetr2yk9.fsf@atrus.jesus.cam.ac.uk> <7ks38p$l4m$1@brick.cswv.com> <slrn7f68br.2bp.thantos@brimstone.mecha>
Message-ID: <7kuocf$1an$1@brick.cswv.com>

In article <slrn7f68br.2bp.thantos at brimstone.mecha>,
Alexander Williams <thantos at gw.total-web.net> wrote:
>On 23 Jun 1999 20:56:41 -0500, Neel Krishnaswami <neelk at brick.cswv.com> wrote:
>>It's just that writing self.__class__ is so *ugly*! Everything else
>>in Python is beautiful, too, which makes this stand out even more. 
>
>Er, that could be because you're doing it wrong?  Try:
>
>class Foo:
>    def __init__(me):
>        me.x = 9
>
>    def setX(me, val):
>        me.x = val

IMO, it's bad style to set properties that should belong to the type
in the instance initializer -- for example, the sort order is
something that's a property of the class, rather than the instances.

Here's an example from some code of mine (it's from a programmatic
front-end to Dejanews):

class Article:
    [...]
    #
    # To set the sort order, set Article.sortorder to one of the
    # keys of Article.comparators.
    #
    comparators = {'date':      lambda x,y: cmp(x.date,y.date),
                   'score':     lambda x,y: cmp(x.score,y.score),
                   'author':    lambda x,y: cmp(x.author,y.author),
                   'subject':   lambda x,y: cmp(x.subject,y.subject),
                   'newsgroup': lambda x,y: cmp(x.newsgroup,y.newsgroup)
                   }
    sortorder = 'date' # can be anything in Article.comparators.keys()

    [...]

    def __cmp__(self, other):
        return self.__class__.comparators[self.__class__.sortorder](self,
                                                                    other)

I want the user to be able to determine how a list of Article's can be
sorted by setting the class's sortorder member. So a statement like

>>> Article.sortorder = 'author'

would change how lists of articles are sorted. This is definitely 
something that's shouldn't be an instance attribute, imo; it would
defeat the purpose if each article could compare differently than
any other. (Oppose standardized testing! :])

But look how ugly that __cmp__ method is; it would be a lot clearer to
write something like

    def __cmp__(self, other):
        return Article.comparators[Article.sortorder](self, other)

because that would make it clearer what class state is being used, and
why. However, though it would be clearer to the programmer, it would
profoundly mess up inheritance.

It's perfectly reasonable for someone else to subclass Article and
override the Article.comparators attribute. (Say to add an ordering to
put Tim Peter's posts at the top of the list, or to use something
faster than my lambdas.) With the clear way of writing it, the
NewArticle class would mysteriously fail to do the right thing,
because the __cmp__ method would look in the Article class dictionary
without bothering to check NewArticle's.

This is an unusual glitch in Python; usually the readable way of doing
something is the right way of doing it. 

Maybe we could re-use the class keyword and let people write class.foo
inside class definitions, by analogy to writing self.foo to refer to
the current instance? And the class.foo syntax would get translated to
the equivalent of self.__class__? 

Hmm. Might hairify the grammar a bit too much, though. 


Neel



From ian at cfmu.eurocontrol.be  Tue Jun 15 05:25:22 1999
From: ian at cfmu.eurocontrol.be (Ian Wild)
Date: Tue, 15 Jun 1999 11:25:22 +0200
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <3762988E.C94AA252@iname.com> <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu> <3763C746.5CA8FE78@iname.com> <m3u2sby4gf.fsf@soma.andreas.org> <87pv2zjhhl.fsf@home.ivm.de> <ekvn1y2efz0.fsf@gnoll.pvv.ntnu.no> <slrn7mb2oq.lpd.wtanksle@dolphin.openprojects.net> <37660FEC.D4F3371B@iname.com>
Message-ID: <37661C02.9CA6E258@cfmu.eurocontrol.be>

Fernando Mato Mira wrote:
> 
> William Tanksley wrote:
> 
> > (I like all the languages I know, except Fortran.  I even like C++,
> > although I wish they could break backwards compatibility.)
> 
> Quiz: What do Scheme, Common Lisp, HPF and Java have in common?

And when's Java*, the massively parallel version, coming out?



From Fredrik.Lundh at p98.f112.n480.z2.fidonet.org  Wed Jun 30 04:37:10 1999
From: Fredrik.Lundh at p98.f112.n480.z2.fidonet.org (Fredrik Lundh)
Date: Wed, 30 Jun 1999 09:37:10 +0100
Subject: printing with print
Message-ID: <000015ac@bossar.com.pl>

From: "Fredrik Lundh" <fredrik at pythonware.com>

Angus MacKay wrote:
> is there any other way to print that with print?
> perhaps with a real function.
> 
> I want to print two things without a space being added between them:
> foo = "there"
> print "hi" + foo
> 
> works fine but: 
> foo = 2
> print "hi" + foo
> 
> does not. I like the printing of any object ability of "print" so I
> do not want to use "" % style printing, I just want to be able to
> govern the whitespace that gets output.

note that the "s" formatting operator can handle any non-tuple
object, just like print (both use "str" to convert non-string objects
to strings).

the following two statements give the same result, as long as
foo is not a tuple.

    print foo
    print "%s" % foo

to deal with tuples too, you can call str yourself, or wrap the
object in a singleton tuple:

    print "%s" % str(foo)
    print "%s" % (foo,)

</F>



From mfletch at vrtelecom.com  Wed Jun 30 23:25:28 1999
From: mfletch at vrtelecom.com (Mike C. Fletcher)
Date: Thu, 1 Jul 1999 03:25:28 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net><3773c3a0.102825445@news.oh.verio.com><xPPc3.535$or1.22866@typhoon-sf.snfc21.pbi.net><xJRc3.292$Of3.67094@news.shore.net><7laovo$c4s$1@Starbase.NeoSoft.COM><377A24B7.AD6FAA1A@cfdrc.com> <3A174E56AD89F7F6.47941772B9D545B9.852EB0CCFCA68614@lp.airnews.net>
Message-ID: <006f01bec371$5f44ae90$a9807018@yec1.on.wave.home.com>

There is a binary distribution of the python version in the FTP directory,
very nice package.  I'm thinking of rewriting my (few) GUI applications
tonight and ditching tkinter entirely.  Documentation seems rather sparse,
but since everything looks fairly simple, it might be doable.

Note, the OpenGL demos are dependent on PyOpenGL.  This should be mentioned
somewhere.

My only other comment is that the window-dragging border in the demos looks
like static (i.e. noise), instead of a controlled GUI element.  Installation
went flawlessly, was able to run the demos in less than a few minutes after
download.

Enjoy,
Mike

----- Original Message -----
From: Mikael Lyngvig <mikael at pobox.com>
Newsgroups: comp.lang.python
To: <python-list at cwi.nl>
Sent: Wednesday, June 30, 1999 8:48 PM
Subject: Re: Python GUIs: Abandoning TkInter and welcoming wxPython?
...
> http://www.cfdrc.com/FOX/fox.html.
>
> The presentation seems promising, but I'd like to suggest a binary
> distribution for PC/Intel computers.  I know this is not common for
> Unix distributions, but it's the norm for PCs.
...




From mstenber at cc.Helsinki.FI  Mon Jun 28 05:42:28 1999
From: mstenber at cc.Helsinki.FI (Markus Stenberg)
Date: 28 Jun 1999 12:42:28 +0300
Subject: [ANN] DejaGrabber 0.1 -- search Dejanews with Python
References: <7kuu73$1ki$1@brick.cswv.com> <8DF363D15duncanrcpcouk@news.rmplc.co.uk>
Message-ID: <al89094u0e3.fsf@myntti.helsinki.fi>

duncan at rcp.co.uk (Duncan Booth) writes:
> Nice bit of work. Two comments though:
> Why does get_messages(4) return 5 messages? The doc string implies it 
> should return only as many as its argument.

It's neat, admittedly (I thought of writing one myself the other day, and
then found that one ;->). The first time I need to do some serious Deja
action, I'll be using it :>.

There's some oops in slicing's internal logic at get_messages; following
patch fixed behavior (for me, anyhow). 

Index: DejaGrabber.py
===================================================================
RCS file: /home/fingon/cvs/projects/python/DejaGrabber.py,v
retrieving revision 1.1
diff -u -r1.1 DejaGrabber.py
--- DejaGrabber.py      1999/06/28 09:35:31     1.1
+++ DejaGrabber.py      1999/06/28 09:40:18
@@ -208,6 +208,5 @@
             message_list.extend(l)
         message_list.sort()
         message_list.reverse()
-        if len(message_list) > n:
-            del message_list[n:-1]
+       message_list = message_list[:n]
         return message_list


> It would be useful if a test example was included in a "if 
> __name__=='__main__'" block at the end of the file. At the very least the 
> example you give above, but perhaps more usefully a full command line 
> driven grabber.



-Markus

-- 
I consider "Anthill Inside" to be much more comforting sticker than
"Intel Inside".



From behrends at cse.msu.edu  Sat Jun 26 01:17:26 1999
From: behrends at cse.msu.edu (Reimer Behrends)
Date: 26 Jun 1999 05:17:26 GMT
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0rr2.2fi.behrends@allegro.cse.msu.edu>  <37722979.2FCD9A5F@gssec.bt.co.uk> <slrn7n66t8.rh7.behrends@allegro.cse.msu.edu>  <37736958.5BED2666@gssec.bt.co.uk>
Message-ID: <slrn7n8okn.ji.behrends@allegro.cse.msu.edu>

Alan Gauld (alan.gauld at gssec.bt.co.uk) wrote:
[...]
> Not so. If the method is long(as these were - over 200 lines 
> in some cases, mostly initialising huge DCE Data structures) 
> you can't see the local definitions. C++ compounds this with 
> its ability to declare local variables anywhere in the code
> (so a split screen editor doesn't help).

A program shouldn't have methods that are 200 lines long; as a rule of
thumb, any method should fit on a single screen (and I mean 80x24, not a
maximized xterm on a 21" monitor). And most methods should be much
shorter than that. If the code you have is screwed up like that, you
have much worse problems than scoping problems. If you need to
initialize large data structures, the initialization process needs to
be broken down into manageable entities (which can then be composed
and reused).

				Reimer Behrends




From krp02078 at my-deja.com  Wed Jun 30 05:10:51 1999
From: krp02078 at my-deja.com (krp02078 at my-deja.com)
Date: Wed, 30 Jun 1999 09:10:51 GMT
Subject: announcement: DIALOG.py
Message-ID: <7lcmun$9ih$1@nnrp1.deja.com>

Hi everybody!

Just finished this module, and thought it could be useful:
DIALOG.py - an interface to the famous dialog program
get it from http://www.brigadoon.de/peter/DIALOG.py

ciao,

	pete

--
-rw-r--r--   1 kruse    users          66 Nov  2 11:50 .signature


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From pmai at acm.org  Tue Jun 29 20:07:31 1999
From: pmai at acm.org (Pierre R. Mai)
Date: 30 Jun 1999 02:07:31 +0200
Subject: The Power of Common Lisp Macros (was Re: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <lw67471fyi.fsf@copernico.parades.rm.cnr.it> <7laoq0$bvh$1@Starbase.NeoSoft.COM> <lwso7bdnia.fsf@copernico.parades.rm.cnr.it> <7laqeb$e17$1@Starbase.NeoSoft.COM> <lwr9mvdlqk.fsf@copernico.parades.rm.cnr.it>
Message-ID: <873dzah7p8.fsf_-_@orion.dent.isdn.cs.tu-berlin.de>

Marco Antoniotti <marcoxa at copernico.parades.rm.cnr.it> writes:

> When "proc" is used within some code, its code is executed at
> runtime.  This is not what happens in Lisp, Scheme and Dylan, where
> macros are evaluated at read-time.  This is pure code transformation,

Only reader-macros are evaluated at read-time.  Normal macros are
evaluated at compile time.  (I know Marco knows this, just to make
sure those without CL knowledge aren't confused.).

> which when fed in the Common Lisp *compiler* produces inlined code.

What I find more interesting than the fact that you get inlined code,
is the possibility that compile-time evaluation affords you: In effect
macros (and compiler-macros) let you extend and use the normal CL
compiler to do most of the dirty work for you in many situations.
IMHO CL's macros are _the_ most important reason that makes CL so good 
at embedding domain-specific languages.

Take for example Harlequin's Common SQL, which embeds SQL in CL
(without any ugly pre-processors, or other evil stuff, such as C
needs):

(do-query ((name) [select [ename] :from [emp]])
  (print name))

This maps name over the tuples returned by the query and prints them.
The important wrinkle:  The query expression gets optimized and
compiled as part of the normal CL compilation process.  Here do-query
is implemented as a macro, the special SQL-syntax using [] is
implemented as reader-macros on [ (and ]).

Or take this silly macro definition (yes, this would need to pass
around a tcl environment object and probably do a number of other
things, but for simplicity):

(defun compile-tcl-to-lisp (exp)
  "Compiles a TCL expression in Lisp syntax into optimized Common Lisp 
code for run-time execution."
  ...)

(defmacro in-tcl (&body expressions)
  (let ((compiled-code (mapcar #'compile-tcl-to-lisp expressions)))
    `(progn , at compiled-code)))

Given this (and a suitably lispified TCL syntax, i.e. one which used sexps
as basic units), I could now write embedded TCL in my CL programs, which
would get compiled to optimized native code along with my normal code.
And this works for any language for which you can dream up an apropriate
sexp syntax.  If done correctly, I can let the embedded language access
and modify the normal CL environment.

And all of this is done at the level of CL itself, that is all of this
can be written as portable ANSI Common Lisp code, without ever having
to open up the implementation, or resort to things like C...

Regs, Pierre.

PS:  Since I still think that this thread should never have been
started, especially as a cross-posted one, I've changed the subject
line and set a follow-up to comp.lang.lisp, where the possibilities of 
Common Lisp in this area can be suitably discussed.  Anyone who
disagrees with this judgement may of course ignore the F'up header...

-- 
Pierre Mai <pmai at acm.org>         PGP and GPG keys at your nearest Keyserver
  "One smaller motivation which, in part, stems from altruism is Microsoft-
   bashing." [Microsoft memo, see http://www.opensource.org/halloween1.html]



From milliger at ausinfo.com  Mon Jun  7 11:49:43 1999
From: milliger at ausinfo.com (mike milliger)
Date: Mon, 07 Jun 1999 10:49:43 -0500
Subject: Calling .exe's from module
Message-ID: <375BEA17.B03B528D@ausinfo.com>

Shouldn't the following module call test.exe? Could someone point me to
the section in the programming python book which covers this
functionality?   No objects persist from my executable, so I don't have
to worry about ref counts and obj's returning to my module. i'm
basically going to pkzip a list of filenames passed to my module as
argv's.

#start
import string,sys,os

# --------------------------------------------------------------------
# run test

if __name__ == "__main__":

    os.system ("D:\Archiver\test.exe")





From gerrit.holl at pobox.com  Sun Jun 20 14:21:21 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sun, 20 Jun 1999 20:21:21 +0200
Subject: max(s) does not what I expect it to do.
Message-ID: <19990620202121.A22076@optiplex.palga.uucp>

Hello all,

I have troubles with the max() function. From the python documentation:

max (s[, args...]) 
    With a single argument s, return the largest item of a non-empty sequence
    (e.g., a string, tuple or list). With more than one argument, return the
    largest of the arguments. 

So I did the following:

Python 1.5.2 (#1, Apr 18 1999, 00:16:12)  [GCC 2.7.2.3] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> list = [ ['a'], ['a', 'b'], ['a', 'b', 'c'] ]
>>> max(list)
['a', 'b', 'c']
>>> list = [ ['a', 'b', 'c'], ['a', 'b'], ['a'] ]
>>> print max(list)
['a', 'b', 'c']

It seems to return the item with the highest index, the latest item. But that
sounds weird, becuase one can also reach that with list[-1].

And than I've got another question: is there an equivalent to
string.ljust, string.rjust and string.center for lists? e.g., I've the
following list:
['a']
and I want to make it:
['a', None, None]
by doing:
['a'].ljust('None', 3)	# or something similar.

Actually, this is the thing I want to reach with the above, I've got a lists
of lists and they have to be the same lenght. Does anyone have better
algorithms than the above for that?

By the way, If you're Dutch and reply me, than please reply in Dutch.

groeten,
Gerrit.

-- 
The Dutch Linuxgames homepage:	http://linuxgames.nl.linux.org
Personal homepage:		http://www.nl.linux.org/~gerrit/

Discoverb is a python program in Dutch and English which tests the words you
learned by asking it. Homepage: http://www.nl.linux.org/~gerrit/discoverb/



From tgm at math.uni-bonn.de  Sun Jun  6 09:43:31 1999
From: tgm at math.uni-bonn.de (Thomas Ackermann)
Date: Sun, 6 Jun 1999 15:43:31 +0200
Subject: Designing Large Systems with Python
References: <m37lqz0yoa.fsf@solo.david-steuber.com> <372599D6.C156C996@pop.vet.uu.nl> <7g4a3l$atk$1@news.worldonline.nl> <3725DD04.4452128F@aachen.heimat.de>
Message-ID: <3utdj7.5a.ln@final.math.uni-bonn.de>

Christian Scholz <ruebe at aachen.heimat.de> wrote:
> Ilja Heitlager wrote:
>> The online-documentation and examples were sufficient for me, but I love
>> paper refs and
>> everbody should have Programming Python

> Well, I found Programming Python a bit too narrating.. I usually just
> want
> a quick overview of function or who I do this or that. Thus the library
> reference + the FAQ + The Python Quick Reference was quite enough for
> me..

I found "Internet Programming with Python" the better book for this.
It has not much to do with CGI or HTML stuff, but is an extremly good
reference for many things about Python (better than "Programming Python" IMO).

	Byebye,
-- 
  Thomas Ackermann | Tel. +49-(0)228/631369|73-7773 | <tgm at math.uni-bonn.de>
             finger tgm at rhein.math.uni-bonn.de for public key
               GNU LINUX Python gtk pygtk MySQL FUDGE GURPS



From David.Arnold at p98.f112.n480.z2.fidonet.org  Wed Jun 30 01:46:00 1999
From: David.Arnold at p98.f112.n480.z2.fidonet.org (David Arnold)
Date: Wed, 30 Jun 1999 06:46:00 +0100
Subject: A Certifiable Very Cool Trick
Message-ID: <000015a5@bossar.com.pl>

From: David Arnold <arnold at dstc.edu.au>

-->"BAW" == Barry A Warsaw <bwarsaw at cnri.reston.va.us> writes:

  BAW> My model was gdb's attach feature where you can attach to, stop
  BAW> and trace through a running external C program.  I would love
  BAW> to be able to do the same thing at the Python level.  We
  BAW> started talking about remote pdb's with threads listening on
  BAW> sockets, etc. (still cool ideas but more work).

on a (vaguely) related note, i was talking to Mark Hammond the other
day (hi Mark) about MS's ActiveDebugging(TM) (sp?) stuff.

as i understood it, each interpreter presents a set of COM APIs which
provide an abstract view of stack frames, etc, suitable for use by a
symbolic script debugger.  you can thus debug across scripting
languages using a single tool, watching calls between interpreters,
etc etc.

all very cool, and so i started thinking about doing the same thing
for Unix via gdb (for C) and the Python interpreter (what other
languages are there?  oh -- ELisp, yeah -- we can do that!)

of course, attaching to a running process should still be possible,
and thus the round-about relationship to the thread.

ahhhhh.  now, where's that spare time again?




d



From jeremy at cnri.reston.va.us  Thu Jun  3 17:13:24 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Thu,  3 Jun 1999 17:13:24 -0400 (EDT)
Subject: Python 2.0
In-Reply-To: <045501beadf8$610e62e0$f29b12c2@pythonware.com>
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7imr6b$1s0$1@nnrp1.deja.com>
	<87pv3iq6ss.fsf@ev.netlab.co.jp>
	<3752C76D.604CD474@prescod.net>
	<87n1ykr0z9.fsf@ev.netlab.co.jp>
	<375372F6.A874554@prescod.net>
	<87hfosquc0.fsf@ev.netlab.co.jp>
	<3753F121.E4A9BE8B@prescod.net>
	<87yai3ntdt.fsf@ev.netlab.co.jp>
	<7j3iif$134e$1@nntp6.u.washington.edu>
	<7j45iv$908$4@cronkite.cc.uga.edu>
	<7j4hhj$118u$1@nntp6.u.washington.edu>
	<7j4jbe$dhh$1@cronkite.cc.uga.edu>
	<37564A02.F115B9C5@infercor.no>
	<045501beadf8$610e62e0$f29b12c2@pythonware.com>
Message-ID: <14166.61071.201719.405639@bitdiddle.cnri.reston.va.us>

>>>>> "FL" == Fredrik Lundh <fredrik at pythonware.com> writes:

  FL> sure looks like the "community" thinks that changing the
  FL> language is more important that using it...

To the contrary, the community is too busy using Python to bother with 
all these silly discussions about changing Python.  The fact that I'm
posting in this threads is... ummm... the exception that proves the
rule.

Most of the volume has been a vigorous but unenlightening back and
forth about garbage collection.  I've for one have concluded that
there have been enough papers written about garbage collection that,
for any particular argument against GC, there exists at least one
paper describing a system for which that argument doesn't apply.  

Of course, none of that has anything to do with Python.  I'm tired of
hearing what GC can do in theory.  I'd be happy to hear about some
actual experience implementing a GC system for Python.  Otherwise,
let's put this on to bed.

Jeremy






From David.Ascher at p98.f112.n480.z2.fidonet.org  Tue Jun 29 00:50:35 1999
From: David.Ascher at p98.f112.n480.z2.fidonet.org (David Ascher)
Date: Tue, 29 Jun 1999 05:50:35 +0100
Subject: BOFs (Birds-Of-a-Feather) sessions at the Monterey Python conference
Message-ID: <00001591@bossar.com.pl>

From: David Ascher <da at ski.org>

As you may know, O'Reilly & Associates is organizing a Python Conference
as part of its Open Source Convention in Monterey in August.  (For more
information, see http://conferences.oreilly.com).  Early word from
O'Reilly is that the Python conference is getting quite a few
early registrants eager to get the discount & free t-shirts!

In addition to the tutorials, talks and panels, there will be informal
"Birds-Of-a-Feather" sessions in the tradition of Usenix BOFs.  These meet
at "off" times:  Saturday and Sunday from 7pm to 10pm, and Monday from 8pm
to 10pm -- meeting spaces will be made available by ORA.  The format for
these is deliberately left up to the birds -- however, one or more
organizers are expected to show up and help lead discussions.

I'd like to solicit discussion on what BOFs would be good to organize, and
find some organizers.  Topics that come to mind include field-specific
BOFs such as a Zope BOF, an image-processing BOF, a database BOF, etc.,
tool-BOFs (Tkinter, wxPython, ...) or "market segment" BOFs such as
"Python in Education", "Selling Python programs", etc.  Other ideas are
welcome.  Anyone can register a BOF through ORA's website at
http://conferences.oreilly.com/bof/, but it might make sense to talk about
it on the mailing list first to minimize overlaps and work out scheduling.

If a BOF is scheduled by July 15, it will be included in the conference
program given to attendees.  BOFs can be scheduled later, and will be
available on the conference website (as well as onsite, I imagine).

Looking forward to seeing many of you there,

--David Ascher & the rest of the Python Conference Organizing Committee

PS: If you have any questions about the Python conference, you can contact
    the conference staff at oscon at oreilly.com or you can contact me at
    da at python.net.  In addition, the folks at oscon at oreilly.com can mail
    you a nifty glossy printed brochure showing the entire schedule of the
    convention, all the talk descriptions, etc.






From velzev at freemail.nl  Mon Jun 14 06:59:16 1999
From: velzev at freemail.nl (Vivienne)
Date: Mon, 14 Jun 1999 12:59:16 +0200
Subject: Trouble with ListBoxes in wxPyhton
References: <7jr72m$s72$1@zonnetje.NL.net>
Message-ID: <7k2nca$pqr$1@zonnetje.NL.net>

Never mind, I've solved the problem. I've declared x and y in class MyApp
global. Because of this, I no longer have to create a new CList or PList,
but use the same one throughout the application.

Thanks anyway,

Vivienne







From hr at deming.com  Fri Jun 18 18:19:53 1999
From: hr at deming.com (calouro)
Date: Fri, 18 Jun 1999 22:19:53 GMT
Subject: Examples of usage for win32pdh.ExpandCounterPath()?
Message-ID: <376ac08a.13773775@news.worldtalk.com>

The win32 extensions work well once I figure out how to call them
correctly -- which is much easier if I have some examples to study and
experiment with.

Does anyone have a working python script that calls
win32pdh.ExpandCounterPath()?  If so please send me some fragments,
with enough preamble and results processing code to show how to set up
the arguments and interpret the results.

I'll forward a compilation of the  responses to anyone who asks for
them.

If more than two or three ask me to, I'll post the compilation to this
group.  

Thanks,
Dan Knierim



From gerrit.holl at pobox.com  Thu Jun 17 14:49:08 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Thu, 17 Jun 1999 20:49:08 +0200
Subject: Starting a project: any hints?
In-Reply-To: <7kb384$lgs$1@nnrp1.deja.com>; from mgutierrez@dss.es on Thu, Jun 17, 1999 at 03:12:14PM +0000
References: <19990617154643.A1680@optiplex.palga.uucp> <7kb384$lgs$1@nnrp1.deja.com>
Message-ID: <19990617204908.A14176@optiplex.palga.uucp>

On Thu, Jun 17, 1999 at 03:12:14PM +0000, mgutierrez at dss.es wrote:
> From: mgutierrez at dss.es
> Newsgroups: comp.lang.python
> Subject: Re: Starting a project: any hints?
> Date: Thu, 17 Jun 1999 15:12:14 GMT
> 
> In article <19990617154643.A1680 at optiplex.palga.uucp>,
>   Gerrit Holl <gerrit.holl at pobox.com> wrote:
> > Hello python-aholics,
> >
> > I'm a 13-years-old-boy and I've about half a year programming
> > experience.
> >
> [snip]
> I started a similar project called Lritaunas Peki , currently this
> project is quite ''frozen'' ( or dead) : www.ctv.es/USERS/irmina
> 
> I have started nearly a dozen of small (freeware) projects and I
> can give you some pieces of advice:
> - Make your program works from the very first day , it doesn't matter
> if it barely works.
> - Don't be ambitious, be humble and try easier/simpler things always.
> - Involve other people in the project, keeping an open mind and a lot
> of patience!
> - Post a lot about it in newsgroups and freshmeat.
> - Work a lot in your project, if you don't nobody will it do for you!
> - Use python, it's by far the best thing you can use: objects and
> readibility.
> I wish you luck, boy.
> And remember if you don't succeed at the first time,... next will be
> Manolo
> 

thanks for all of your hints. I'll think about it when programming...

groeten/regards,
Gerrit.

--
The Dutch Linuxgames website.		De Nederlandse Linuxgames pagina.
Everything about games on Linux.	Alles over spelletjes onder Linux.
Site address:				http://linuxgames.nl.linux.org
Personal homepage:			http://nl.linux.org/~gerrit/



From moshez at math.huji.ac.il  Fri Jun 11 00:55:58 1999
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 11 Jun 1999 07:55:58 +0300
Subject: fork()
In-Reply-To: <199906110443.AAA02408@eric.cnri.reston.va.us>
References: <000c01beb3c1$a3f428a0$329e2299@tim>  <199906110443.AAA02408@eric.cnri.reston.va.us>
Message-ID: <Pine.SUN.3.95-heb-2.07.990611075346.4942G-100000@sunset.ma.huji.ac.il>

On Fri, 11 Jun 1999, Guido van Rossum wrote:

<snip/>
> ...this almost seems acceptable: in the formal semantics, objects
> that are part of cycles would live forever, and as an invisible
> optimization we recycle their memory anyway if we know they are
> unreachable.  (I've got the feeling that I've seen this rule before
> somewhere.)  
<snip/>

No, that can't be. If we did, it were in Scheme, and after you put it in,
you'd be morally obliged to add full-fledged lambda, tail-recursion and 
user visible continuations. 

you-asked-for-it-ly y'rs, Z.
--
Moshe Zadka <mzadka at geocities.com>. 
QOTD: My own exit is more likely to be horizontal then perpendicular.





From holger at phoenix-edv.netzservice.de  Mon Jun 14 05:24:20 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Mon, 14 Jun 1999 09:24:20 GMT
Subject: os.mkdir()
References: <Pine.LNX.4.04.9906101240310.6646-100000@emerald.netskate.ru>
Message-ID: <3764CA44.4C053675@phoenix-edv.netzservice.de>

High,

ok, that has been one other fault.
I came even on  myself on it, but I was so confused, that I
must have missed that bitch!;-)

Ciao,
Holger

Oleg Broytmann schrieb:
> 
> On Wed, 9 Jun 1999, Holger Jannsen wrote:
> > os.makeDirs(thePath, aMode)
> > AttributeError: makeDirs
> 
> os.makedirs(thePath, aMode)
>        ^
> 
>    Lowercase...
> 
> Oleg.
> ----
>     Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
>            Programmers don't die, they just GOSUB without RETURN.



From delgado at Mathematik.Uni-Bielefeld.DE  Wed Jun 16 13:58:44 1999
From: delgado at Mathematik.Uni-Bielefeld.DE (Olaf Delgado)
Date: Wed, 16 Jun 1999 19:58:44 +0200
Subject: Newbie again: computing attributes on the fly
In-Reply-To: <1282571882-108998091@hypernet.com>
References: <1282571882-108998091@hypernet.com>
Message-ID: <Pine.OSF.4.05.9906161916040.10750-100000@pinax.Mathematik.Uni-Bielefeld.DE>

Hi and thanks for answering!

On Wed, 16 Jun 1999, Gordon McMillan wrote:

> Olaf Delgado writes:
> > 
> > class x:
> >     def __getattr__(self, attr):
> >         import string
> >         if string.find(attr, "get_") != 0:
> >             val = ( lambda x = getattr(self, "get_"+attr)(): x )
> >             self.__dict__[attr] = val
> >             return val
> >         else:
> >             raise AttributeError("no such attribute")
> > [...]


> You're making things too complicated. __getattr__ is only called if 
> the attribute is not found. So something like this will suffice:
> [...]
> If you want to generalize that, I'd see if we have an attribute 
> "calc_%s" % nm and use that method (vs the inline computation).

Yes, I definitely want to generalize that. The problem is that that method
may be defined in a superclass, so "self.__dict__" won't find it. AFAIK
(which may be not be very far), I have to use "hasattr" and/or "getattr",
which in turn will call "__getattr__".

Recursively yours,
Olaf

-- 
   ////         Olaf Delgado Friedrichs, Uni Bielefeld, Mathematik
   Olaf         Postfach 100131                  (room:  V4-109)
   `='          D-33501 Bielefeld, Germany       (phone: +49 521 106-4765)
                           http://www.mathematik.uni-bielefeld.de/~delgado




From paulb at infercor.no  Thu Jun  3 05:25:22 1999
From: paulb at infercor.no (Paul Boddie)
Date: Thu, 03 Jun 1999 11:25:22 +0200
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>  <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>  <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net>  <87hfosquc0.fsf@ev.netlab.co.jp> <3753F121.E4A9BE8B@prescod.net> <87yai3ntdt.fsf@ev.netlab.co.jp> <7j3iif$134e$1@nntp6.u.washington.edu> <7j45iv$908$4@cronkite.cc.uga.edu> <7j4hhj$118u$1@nntp6.u.washington.edu> <7j4jbe$dhh$1@cronkite.cc.uga.edu>
Message-ID: <37564A02.F115B9C5@infercor.no>

Graham Matthews wrote:

[...]

> But I am not sure I see what the connection is between how JPython does
> memory management and garbage collection in Python in general.

[...]

> Donn Cave (donn at u.washington.edu) wrote:
> : Well, I think there's something to be said for it.  With Python, we're
> : free from the obnoxious requirement imposed by the storage model of
> : languages like C, that forces the programmer to account for each scrap
> : of malloc'd memory lest there be leaks.
> : ...
> : Why should I enjoy this freedom with memory resources, but welcome exactly
> : the same constraints when an object has other termination requirements?
> 
> Because other kinds of objects are different to memory? Also I am not
> sure why you think GC can't handle other kinds of objects as well. Maybe
> Java doesn't do a good job of resource finalisation, but that doesn't
> mean it can't be done.

So, we have established that it can indeed be done, but reaching back into the
depths of time and pulling out the original proposal - being that Python should
have its primary (or only) implementation running in the Java environment - we
can see that regardless of the existence of satisfactory GC schemes, the typical
Java implementation surely falls short of what many people require.

Therefore, whilst there may well be a need for more sophisticated GC in Python
2.0, we must conclude that the Java environment may not be the appropriate place
in which such technologies are to be found.

Paul



From Robert.Meegan at wcom.com  Mon Jun 14 11:34:05 1999
From: Robert.Meegan at wcom.com (Robert Meegan)
Date: Mon, 14 Jun 1999 15:34:05 GMT
Subject: GC SIG Response
Message-ID: <Pine.A41.4.10.9906141029140.64942-100000@barney.mcit.com>

Greetings ---

The interest level in a garbage collection SIG was, to a first order
approximation, zero (sorry, Moshe!). This leaves GC in the weather
category, with much discussion and little action.

--- Robert

--------------------------------
Robert Meegan
MCIWorldCom - Cedar Rapids, Iowa
319.375.2416




From suzuki611 at okisoft.co.jp  Thu Jun 10 02:26:19 1999
From: suzuki611 at okisoft.co.jp (Hisao Suzuki)
Date: 10 Jun 1999 15:26:19 +0900
Subject: Python 2.0
References: <000c01beb2e8$0b825440$af9e2299@tim>
Message-ID: <u3zp28a7w4.fsf@ares.sys1.ngo.okisoft.co.jp>

In article <000c01beb2e8$0b825440$af9e2299 at tim>,
"Tim Peters" <tim_one at email.msn.com> wrote:
| [Hisao Suzuki]
| > ...
| > It would be an issue of distributed garbage collection.  And it
| > is a difficulty for automatic GC with finalization.
| Yes, there are real problems here, and Python certainly isn't the first to
| face them.

Certainly.  And many languages either give up them to
programmers entirely (---"Please make your own finalization
mechanism from scratch" from ???), or provide rather
insufficient ones (---"However, it is not a good idea to rely on
finalization for the timely freeing of resources." from Java
Programmers FAQ).

| > (The solution is another story.  It is not so obvious for
| > Python.)
| > # Certainly the current reference counting system of Python
| > # reduces the issues to fairly (though _not_ fully) tolerable
| > # ones...
| It automatically deals with the order-of-finalization problem in a correct
| way, by following a topological sort of the "references" graph.  Which says
| almost nothing, since that's *natural* for RC.  In return it can't deal with
| cycles at all, and indeed it's still easy to write __del__ methods that
| trigger mysterious exceptions at system shutdown time, when Python *does*
| clean up *some* kinds of cyclic trash (namely the cycles the interpreter
| itself creates for its own nefarious purposes).

Surely the shutdown time is the most critical for the current
Python and it is the main reason why I wrote "though _not_
fully".  You might know that

in article <u36756gcoc.fsf at ares.sys1.ngo.okisoft.co.jp>,
I wrote:
< As for the issue of circular references, in my opinion, the
< future Python should use non-compacting GC at the end of the
< last thread and destroy the remained objects by invoking their
< __del__ methods --- it is THE time to destroy all the objects
< remained, so this is conceptually clean (thus very Pythonic :-),
< though it is the way of memory management used in the present
< Perl in fact...

It is only a casual idea and I am not sure what sort of
non-compacting garbage collector (or garbage detector, in this
case?) is implementable and acceptable as part of Python system.

For example, I have once thought what if each instance has two
additional pointers, which make up the object list, i.e. oblist.
(PyInstance_New and others will need modifying then.)  After
folding up the stack(s), Python checks the referred instances
from, say sys.modules (see my article
<u31zfszath.fsf at ares.sys1.ngo.okisoft.co.jp> for a half-baked,
or rather a-tenth-baked quick-and-incomplete prototype), detects
possible circular references from the oblist, invokes their
__del__s first and then shuts down the same way as the current
Python does.  Well, it may correct some faults, but I know it is
neither perfect nor efficient anyway.

--===-----========------------- Sana esprimo naskas sanan ideon.
SUZUKI Hisao            suzuki611 at okisoft.co.jp, suzuki at acm.org.



From achrist at easystreet.com  Tue Jun 22 13:04:35 1999
From: achrist at easystreet.com (Al Christians)
Date: Tue, 22 Jun 1999 10:04:35 -0700
Subject: choosing random numbers with weights/probability?
References: <929987267.6227@www.remarq.com> <lbwb3.3$V4.11640@news-dal.corridex.com> <ERwb3.1452$U6.5885@newsr2.twcny.rr.com> <7knamm$kas$1@news.udel.edu> <izNb3.1875$U6.6729@newsr2.twcny.rr.com>
Message-ID: <376FC223.D34403BC@easystreet.com>

It looks like generating the random number takes much longer than
searching a short list, so that the method I described, which takes
two or more random numbers per selection, is a big loser on speed 
unless one is searching a list of 100 elements or more.  I  think
it might be better for statistical properties of the results, 
depending on lots of things (the magnitudes of the weights and
the quality of the rng), but in simple applications, that probably 
doesn't matter much, and IDK for sure.

Al 


Darrell wrote:
> 
> Orig:
>  Time: 10.28 sec
> {'one': 25167, 'two': 25144, 'three': 49689}
> 
> Terry's:
>  Time: 9.94 sec
> {'one': 24983, 'three': 50020, 'two': 24997}
> 
> --
> --Darrell



From fatjim at home.com  Thu Jun  3 13:34:01 1999
From: fatjim at home.com (Jim Meier)
Date: Thu, 03 Jun 1999 17:34:01 GMT
Subject: starship.python.net
References: <000601beadc7$75b2ecb0$8b7125a6@cpda6686.mcit.com>
Message-ID: <3756BD08.EC98BF93@home.com>


jkraai at murl.com wrote:

> Ok, I don't go to the starship root very often, but went there this morning.
>
> It's _beautiful_.  Who did this excellent work?  Chris T?
>
> Wow.  Clean, elegant, etc.

The page says "Web site design by Robin Friedrich", and links an email address
where you can applaud him if he doesn't read this group.

I agree - It's an excellent page. Exactly as fancy/simple as it needs to be, and
beauty packed in as well.



-Jim




From tdc0331 at hti.com  Tue Jun 29 11:55:55 1999
From: tdc0331 at hti.com (David Stokes)
Date: Tue, 29 Jun 1999 10:55:55 -0500
Subject: Tkinter slide question
Message-ID: <3778EC8B.833DEBC4@hti.com>

I'm trying to implement a slide to control how often a function restarts
but it appears that I'm not getting the result back from the 'variable'
variable.  Could some kind soul out there on the proper way to read the
slide 'position'?

Code snipit:
Scale(label="Rescan",variable=self.yspeed,orient=HORIZONTAL,).grid(column=0)

currently returns PY_VAR12 instead of the slider value.


Thanks!




From mlh at idt.ntnu.no  Thu Jun 24 04:21:03 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 24 Jun 1999 10:21:03 +0200
Subject: interactive-logging
References: <376FF09A.8170CB@icrf.icnet.uk> <y0jso7jghez.fsf@vier.idi.ntnu.no> <jt3dzjjaa4.fsf@wazor.biostat.wisc.edu>
Message-ID: <y0jiu8euhzk.fsf@vier.idi.ntnu.no>

William Annis <annis at biostat.wisc.edu> writes:

> mlh at idt.ntnu.no (Magnus L. Hetland) writes:
> 
> > Hm. Isn't there a UN*X command that does that? (That logs everything
> > going on in a terminal...)
> > 
> > I'm damned if I can remember what it was, though...
> 
>         /bin/script

Right! Thanks.

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From tre17 at cosc.canterbury.ac.nz  Tue Jun 22 18:21:37 1999
From: tre17 at cosc.canterbury.ac.nz (Timothy R Evans)
Date: 23 Jun 1999 10:21:37 +1200
Subject: creating dictionnaries
References: <376F82A6.82497D2B@der.edf.fr>
Message-ID: <c75hfnzanbi.fsf@pc142.cosc.canterbury.ac.nz>

Christian Caremoli <Christian.Caremoli at der.edf.fr> writes:

> Hi,
> Dictionnaries can be created like that :
> d={'a':1,'b':2}
> 
> By calling a function you create a dictionnary like that :
> def f(**d):
>    return d
> 
> d=f(a=1,b=2)
> 
> I would like to be able to create dictionnaries with some similar syntax
> like keyed tuples :
> d=(a=1,b=2)
> 
> Is there a way to do that ?
> 
> Thanks
> 
> Christian Caremoli

I don't think so, there's no equivalent operator to Perl's =>.  You
cound just use your function, but this introduces some restrictions on
the dictionary keys, the most important of which is that they must be
valid identifiers and will always be strings.  There tend to be many
situations where you want the key to cotain spaces or other characters
that are not allowed in identifiers, or where you want the key to be
an integer or tuple etc.

--
Tim Evans



From bbailey08 at my-deja.com  Tue Jun  8 14:53:10 1999
From: bbailey08 at my-deja.com (Brenda)
Date: Tue, 08 Jun 1999 18:53:10 GMT
Subject: Using perl in python
Message-ID: <7jjoqh$t56$1@nnrp1.deja.com>

I have a bunch of perl libraries that I would like to call in my python
scripts.  Is this possible?


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From planders at mail.utexas.edu  Tue Jun 22 22:54:42 1999
From: planders at mail.utexas.edu (Preston Landers)
Date: 22 Jun 1999 21:54:42 -0500
Subject: total idiot question: +=, .=, etc...
Message-ID: <877lov62z1.fsf@pbl.ddns.org>

Hi there,

another question in this popular series of foolish questions...

Can someone *please* give me a good reason why Python doesn't yet have 
+=, -=, and .= and other such operators?  I know it's just syntactic
sugar, but damn, it is such sweet sugar!

Also, if this is in 1.5.2+, please forgive me, as I'm not up to speed
on the new versions yet.  And I have searched the FAQ for this
question but have not found it.

many thanks,

---Preston



From tom at spirit.gcrc.upenn.edu  Wed Jun 23 16:30:01 1999
From: tom at spirit.gcrc.upenn.edu (Tom Fenn)
Date: Wed, 23 Jun 1999 16:30:01 -0400
Subject: More Variant Difficulties
References: <377135DF.311437@spirit.gcrc.upenn.edu>
Message-ID: <377143C8.56E7DC82@spirit.gcrc.upenn.edu>

I'm also having trouble with a bunch of functions that take VARIANT FAR*
arguments.  Any hints on how to deal with this?

Tom Fenn



From dkuhlman at netcom.com  Tue Jun  1 15:26:05 1999
From: dkuhlman at netcom.com (G. David Kuhlman)
Date: Tue, 1 Jun 1999 19:26:05 GMT
Subject: Browsers
References: <37530E1B.CAC2991C@cwcom.net>
Message-ID: <dkuhlmanFCnxzH.72@netcom.com>

Lynx is a text mode browser:

    http://lynx.browser.org/

For fancier stuff, look at Mozilla:

      http://www.mozilla.org/

Also go to http://www.w3.org/,  and look for both "amaya" and
"arena".

I believe that source is available for most, if not all of these.

  - Dave



Daniel Faulkner <m01ymu00 at cwcom.net> wrote:
> Is there a basic browser some where that I can look at to see how it
> works? (not grail)
> As I can't understand much of the python internet software and don't
> understand how to parse the HTML once I've got it.

> Thanks

> Dan




From akuchlin at mems-exchange.org  Fri Jun 18 20:18:51 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: Fri, 18 Jun 1999 20:18:51 -0400 (EDT)
Subject: NameError Problem
In-Reply-To: <7kels4$9tv$1@eskinews.eskimo.com>
References: <7kels4$9tv$1@eskinews.eskimo.com>
Message-ID: <14186.57835.612462.716239@amarok.cnri.reston.va.us>

Jonathon writes:
>	    if ( the_file_is == "john_www.db" ) :
   ...
>            elif ( the_file_is == "jan_www.db " ):
   ...
>            print temp_record_id

Hmm...  try adding 
	     else:
	         print "EEK! Got unexpected filename:", repr(the_file_is)

You could also raise an exception, instead of just printing a message.
I'll wager that the_file_is winds up with some sort of unexpected
value, and causes your variable to never be assigned a value,. because 
none of the conditions trigger.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
"Since the invasion of Grenada," a military source informed me, "we call it
C^5. That's Command, Control, Communications, Computers and Confusion."
    -- Barbara Garson, _The Electronic Sweatshop_ (1988)




From stephensong at iname.com  Sat Jun 19 21:36:01 1999
From: stephensong at iname.com (Gary Stephenson)
Date: Sun, 20 Jun 1999 11:36:01 +1000
Subject: Embed Python in Xbase++??
Message-ID: <376C4581.2A2D1BA1@iname.com>

How difficult/feasible would it be to embed Python in ALASKA's Xbase++
language (WIN32), so as to make the Python Dictionary and ImmutableTuple
types (we already have Lists - but we call them arrays) usable as
primitive classes in the Xbase++ language.

Xbase++ has the following types: string, date, numeric, logical, array,
codeblock, object-reference and NIL.  Instances of these types would
need to be storable in the dictionaries and tuples, along with the
native Python types.  There is a published API for interfacing with
these types from C.

Can anyone describe or direct me to examples of how this might best be
accomplished? 

I would also consider hiring someone to have a go at this - (hope I'm
allowed to "call for quotes" in UseNet?? If so, and you're interested,
send me e-mail directly.  If not, humblest apologies, and forget I
asked!! )

All suggestions welcomed. 

gary



From mnot at pobox.com  Tue Jun  8 18:27:23 1999
From: mnot at pobox.com (Mark Nottingham)
Date: Tue, 8 Jun 1999 22:27:23 GMT
Subject: Using PIL with CGI scripts
References: <7jk0rm$o9ar@eGroups.com>
Message-ID: <000a01beb1fe$14e34200$0301a8c0@cbd.net.au>

See the StringIO module (actually, cStringIO).

One thing to keep in mind - unless the image is *really* dynamic (i.e., it
changes EVERY time someone accesses it), you're much better off if you write
it to disk.

See genpix.py at http://www.mnot.net/python/  for an example of this (it's a
command line utility that HTMLifys a directory of images, but it does use
cStringIO in this manner).

Cheers,


----- Original Message -----
From: <luisroco at hotmail.com>
To: <python-list at cwi.nl>
Sent: Wednesday, June 09, 1999 7:10
Subject: Using PIL with CGI scripts


> Hi all
>
> How can a CGI scripts outputs graphics images through Python code using
the Python Imaging Library package?
>
> I just want to produce some images which has been processed using PIL and
show them in a web site
>
>




From arcege at shore.net  Tue Jun  8 22:30:57 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 09 Jun 1999 02:30:57 GMT
Subject: In what order does Python read dictionary keys?
References: <7jjnt5$snk$1@nnrp1.deja.com> <c754skitdv4.fsf@pc142.cosc.canterbury.ac.nz>
Message-ID: <Blk73.2394$nn.730426@news.shore.net>

Timothy R Evans <tre17 at cosc.canterbury.ac.nz> wrote:

: To get the keys sorted alter your loop to be:

: for key in entries.keys().sort():

The right idea, but the wrong implimentation.  Unfortunately, sort()
does not return the sorted list, it sorts in-place.

  keys = entries.keys()
  keys.sort()
  for key in keys:

: From the look of this example, you probably would be better off using
: a nested tuple like this:

: """
: entries = (('red', 'Type the red value, please: '),
:            ('green', 'Type the green value, please: '),
:            ('blue', 'Type the blue value, please: '))

: for key,value in entries:
: ...
: """

: or even:

: """
: entries = ('red', 'green', 'blue')

: for key in entries:
:     value = 'Type the ' + key + ' value, please: '
: """

: This way you can control the order exactly and it might be faster as
: you avoid the cost of a dictionary lookup.

If this was only for these three values, tuples would be the most
efficient.  

But in general (outside of the question now) tuples can't be searched
easily, I would suggest either the dictionary or a list (with the
'index' method).

You can also make a subclass of UserDict with the proper ordering:

class Colors(UserDict):
  ordered_keys = ('red', 'green', blue')
  def keys(self):
    l = []
    for key in self.ordered_keys:
      if self.has_key(key):
        l.append(key)
    return l
  def items(self):
    l = []
    for key in self.keys():
      l.append( (key, self[key]) )
    return l
  def values(self):
    l = []
    for key in self.keys():
      l.append( self[key] )
    return l

This makes it all work like you desire, at the expense of a little more
processing.

  -Arcege




From tismer at appliedbiometrics.com  Sat Jun 12 12:45:30 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Sat, 12 Jun 1999 16:45:30 GMT
Subject: ANN: Stackless Python 0.2
Message-ID: <37628EAA.C682F16C@appliedbiometrics.com>

ANNOUNCING:

               Stackless Python 0.2
           A Python Implementation Which
             Does Not Ese The C Stack

What is it?
A plugin-replacement for core Python.
It should run any program which runs under Python 1.5.2 .
But it does not need space on the C stack.

Why did I write it?
Stackless Python was never written before (afaik), since it
was said to be impossible without major rewrites of core Python.
I am proving the controverse:
It is easy to write, just hard to think.

Who needs it?
At the moment, this is only useful for C programmers
who want to try certain new ideas. Hardcore stuff.
It allows to modify the current execution state by
changing the frame stack chain without restictions,
and it allows for pluggable interpreters on a per-frame-basis.

The possibilities are for instance:

Coroutines, Continuations, Generators

Restartable exceptions and Persistent execution state
might be possible.

Stackless extension modules can be built. The new builtin
stackless "map" function is a small example for this.

Coroutines will be able to run at the speed of
a single C function call, which makes them a considerable
alternative in certain algorithms.

Status of the project:
Stackless-ness has been implemented and tested with pystone.
pystone works correctly and is about 4-5 % slower than
with standard Python.

What I need at the moment is
- time to build a sample coroutine extension
- your input, your testing, critics and hints.

Some still rough documentation is available at
http://www.pns.cc/stackless/stackless.htm

Source code and a VC++6.0 build for Windows can be found
from the document or directly from
ftp://ftp.pns.cc/pub/stackless_990611.zip

cheers - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From webmaster at epls.com  Wed Jun 23 00:31:38 1999
From: webmaster at epls.com (webmaster at epls.com)
Date: Wed, 23 Jun 1999 12:31:38 +0800
Subject: W95 binary output from CGI script
Message-ID: <199906230431.MAA00798@yangtze.dev.epls.com>

On Jun 22, 1999, tlhall at blackhole.nyx.net wrote:

>I'm working on a CGI script for Windows 95, and trying to send
>binary data out stdout. Every time I try to send a 
 it spits
>out 
. This really messes up a JPEG, for example.
>
>How can I get python to print just 
 ?
>I've tried
>   sys.stdout.write("
")
>and
>....

Tom, 

use a raw string -- observe the "r" before the string. See eg below:

>>> print r"abc
xyz"
abc
xyz
>>> print "abc
xyz"
abc
xyz

The first string with 
 was printed without being interpreted as a 
line feed character. (note: eg pasted from Linux, not W95)

BTW, why do you use sys.stdout.write() instead of print?

yours-CRLF-or-LF-ly

alan kang
http://www.epls.com - free groupware online




From hniksic at srce.hr  Thu Jun  3 15:55:33 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 03 Jun 1999 21:55:33 +0200
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net> <aahzFCpHw8.Gxt@netcom.com> <19990602132331.A281155@epa.gov>
Message-ID: <87emjtnjnu.fsf@pc-hrvoje.srce.hr>

Randall Hopper <aa8vb at yahoo.com> writes:

> If I was still locked onto vi, it'd be a pain, and I can see that
> being an issue with some editors.  Though with emacs, there's
> rectangular kill (^Xrk) and block indent/dedent (^X^I) -- key
> sequences I use without even thinking anymore.

Same here.  It's a nice coincidence that a single C-u means a prefix
argument of four, which makes `C-u C-x C-i' indent a Python block.



From fredrik at pythonware.com  Mon Jun 28 09:48:56 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 28 Jun 1999 13:48:56 GMT
Subject: strange behaviour at termination time
References: <377765F0.38230773@steria.fr>
Message-ID: <00de01bec16c$f9fc68d0$f29b12c2@secret.pythonware.com>

Fabien COUTANT <fabien.coutant at steria.fr> wrote:
> Basically there is a base class name that I found bound to None,
> which prevents me to call base methods.
> 
> I tested it with 1.5.1 on Linux and 1.5.2 on Windows. Was working
> fine with 1.4 in Debian/Linux 1.3.1 till I upgraded to Debian 2.1...
> 
> Bug or feature ? Quite annoying if feature...

feature.

1.5.2 does a more aggressive global clean-up when the
interpreter is shut down.  this includes zapping modules
and module globals, and replacing them with references
to None.

> More that it, I discovered during testing that sometimes it is correct
> if I change the base class name !

namespaces are dictionaries.  changing the name may or may
not change the clean-up order...

> Here is sample code to show it:
> 
> ----------------------------------
> class A :
>     def __del__ (self) :
>         print "A.__del__", self
> 
> class B (A) :
>     def __del__ (self) :
>         print "B.__del__", self, A
>         # can't do A.__del__ (self) for example
> 
> x = A ()
> y = B ()
> ----------------------------------
> 
> In this configuration it should show ok.
> If you change the name of the first class from A to X everywhere
> then X is bound to None in B.__del__ !!!

the only reasonable way to get around this is to explicitly
bind all the names you want in your destructor:

>     def __del__ (self, A=A) :
>         print "B.__del__", self, A
>         A.__del__(self)

(btw, 1.4 probably didn't call B's destructor at all...)

</F>




From graham at sloth.math.uga.edu  Thu Jun  3 11:17:59 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 3 Jun 1999 15:17:59 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net> <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net> <7j34l3$hbj$1@nnrp1.deja.com>
Message-ID: <7j66b7$il5$2@cronkite.cc.uga.edu>

Paul Prescod <paul at prescod.net> wrote:
> Is there a language with real GC where destructors are called as
> predictably as they are in Python?
salvador at my-deja.com wrote:
: No, but it could be done in Python, GC isn't incompatible with reference
: counting and both can be combined. A simple GC could be added just to
: detect ciclical data structures and remove it
: 
: Even new special obect methods could be suported by the language to
: handle it, say __gc__, that gets called by the GC requesting the object
: to liberate its references to other objects so the ciclid data breaks.

You don't need a new method call -- all you need is a mark and sweep
collector with refcounts. I really am amazed that this has generated so
much traffic. It's very simple. You have ref counts just like you do
now in Python. When an objects refcount goes to 0 its __del__ method
is called. On top of this you add a mark and sweep collector to handle
circular references. During a sweep if it finds dead objects it calls
the __del__ method for that object. Why is this approach to memory
management any more problematic than the current approach? Of course
you need to have a way of registering roots and traversing foreign
objects, and of allocating non compactable blocks, but these are all
totally standard things.

graham
-- 
                          So I turned around
                    And 40,000 headmen bit the dirt
                      Firing twenty shotguns each
                        And man it really hurt



From greg.ewing at compaq.com  Mon Jun 28 22:05:50 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Tue, 29 Jun 1999 14:05:50 +1200
Subject: Python for large projects?
References: <3774b5d1.690922@news.force9.net>  <T6bd3.2883$U6.12014@newsr2.twcny.rr.com> <slrn7naup9.jk.thantos@brimstone.mecha> <RGid3.2957$U6.12374@newsr2.twcny.rr.com> <87n1xm85mf.fsf@lisboa.i-did-not-set--mail-host-address--so-shoot-me> <Jhsd3.3090$U6.12410@newsr2.twcny.rr.com> <3776EF65.A07CBEFC@compaq.com> <slrn7ne533.1rmr.marc@oscar.noc.cv.net> <xeLd3.3224$U6.13605@newsr2.twcny.rr.com> <q6Od3.3254$U6.12368@newsr2.twcny.rr.com>
Message-ID: <377829FE.EA457FDB@compaq.com>

Darrell wrote:
> 
> Read up on RExec and don't think it's the solution.

I think the only way to make an exposed data object
completely tamper-proof in Python is, as was suggested,
implement it as a C extension.

Maybe it could maybe be a thin C layer that has Python on
the other side. Would it be feasible to design a
general-purpose "rinstance" object that does for
data structures what rexec does for namespaces?

Greg



From sdm7g at Virginia.EDU  Thu Jun  3 17:26:41 1999
From: sdm7g at Virginia.EDU (Steven D. Majewski)
Date: Thu, 3 Jun 1999 17:26:41 -0400 (EDT)
Subject: "The Python Way"
In-Reply-To: <9906031407150D.00552@quadra.teleo.net>
References: <045501beadf8$610e62e0$f29b12c2@pythonware.com> <9906031407150D.00552@quadra.teleo.net>
Message-ID: <Pine.A32.3.90.990603172339.1626A-100000@elvis.med.Virginia.EDU>

What makes you think that Tim and Guido would agree on "The Python Way" ? 

Not that such an argument wouldn't be worth while: the signal to noise
ratio would be much better than recent threads and the disagreements
would be over much more interesting and subtle points than 
"White Space: Bad or Evil?" threads. 


---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---

 "IA-64 looks just about like what you would expect from a PA-RISC
  and IA-32 train wreck with a little VLIW thrown in for spice."
 * Thomas J. Merritt  <tjm at spam.codegen.com> in <news:comp.arch> *




From arcege at shore.net  Tue Jun  1 15:31:23 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 01 Jun 1999 19:31:23 GMT
Subject: Help for beginner
References: <7ipplg$31d7@enews3.newsguy.com> <37530B50.6FB1D114@cwcom.net>
Message-ID: <fyW43.1494$nn.453396@news.shore.net>

Daniel Faulkner <m01ymu00 at cwcom.net> wrote:
: Chuck & Amy Hoskins wrote:

:> I just flat out need to learn how to use Python.  I am really interested in
:> getting into programming and I don't find many sites to help out the common
:> Moron (myself in this case).  Please help!!!
:>
:> Chuck

: This is about the best choice you could of made as Python is a very easy
: language but I suppose this is why you decided to use Python well first down
: load python from www.python.org (use version1.5.2b2 as this includes Tkinter,
: graphical interface)

Just a note: you should download version 1.5.2, not 1.5.2b2 which is a
beta version.  Python 1.5.2 was released 7 weeks ago (April 13th).  All
releases of Python include Tkinter, but none include Tcl/Tk (but there
are links to download that as well).

: and read through the tutorial and just start creating some programs as a
: project and if you come into any problems just ask some one at the python
: newsgroup.
:  The graphical part is some thing not to use at the beginning (also there is
: not proper manual for it but there is a survival manual at the python home page
: some where) but do look at some of the code used in example programs. To show
: you what Python does try downloading grail from
: http://monty.cnri.reston.va.us/grail/ and pysol
: which you will find some where. There is a quick tutorial at
: http://www.troubleshooters.com/codecorn/python/pptut.htm
: I hope this helps if you want any help feel free to email me (even though i
: have only a few months of experience)

There is also the http://www.python.org/Html.html which has a number of
links to useful beginner information.  Two mailing lists exist for people
learning Python as well: python-help (help at python.org) and the tutor
mailing list (http://www.python.org/mailman/listinfo/tutor).

The book _Programming Python_ contains a good lengthy discussion on using
Tkinter.  And Aaron Watters' _Internet Programming with Python_ explains
another Python windowing system, WPY.  Both are very good books but
are a little dated.  _Learning Python_ is more up-to-date.

  -Arcege




From arcege at shore.net  Mon Jun 21 07:15:19 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Mon, 21 Jun 1999 11:15:19 GMT
Subject: Better Online Docs?
References: <7kk6tj$b42$1@news.igs.net>
Message-ID: <b9pb3.756$7X1.186681@news.shore.net>

John Smith <discman_64 at hotmail.com> wrote:
: Hi.. I was just wondering if there are any other online docs for python on
: internet for free?

Hi John,

Other than what?  Python's online documentation is at
http://www.python.org/doc/.

Other network accessible documentation is referred to in old postings
to this newsgroup available thru www.dejanews.com.

  -Arcege




From dwelton at linuxcare.com  Mon Jun  7 13:22:52 1999
From: dwelton at linuxcare.com (David N. Welton)
Date: 07 Jun 1999 10:22:52 -0700
Subject: python active web pages
References: <jonathan-0406991640390001@borg.onegoodidea.net> <37582D9D.3D7A4B1D@hp.com.NOSPAM> <jonathan-0406992318350001@borg.onegoodidea.net>
Message-ID: <877lpgc4cz.fsf@smtp.linuxcare.com>

jonathan at onegoodidea.com (Jonathan Hogg) writes:

> sorry, i should perhaps have made myself more clear. i was wondering
> as i posted whether mention of ASP would confuse the matter.
 
> i'm not implementing on Windows or using "Active Server Pages" as
> defined by Microsoft. i'm interested in dynamic web pages on
> unix/apache.

> (and yes, i know about PyApache - but that's just a
> convenience/optimisation for CGI, not an embedding of Python in web
> pages)

If someone were interested, it probably wouldn't take *too* much work
to convert my mod_dtcl to mod_pysomething.

The url is http://comanche.com.dtu.dk/dave/

I'd actually be interested in this myself, but I don't have any free
time:-(

Ciao,
-- 
David N. Welton, Web Engineer, Linuxcare, Inc.
415.354.4878 x241 tel, 415.701.7457 fax
dwelton at linuxcare.com, http://www.linuxcare.com/
Linuxcare. At the center of Linux.



From hniksic at srce.hr  Thu Jun  3 15:36:32 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 03 Jun 1999 21:36:32 +0200
Subject: Python 2.0
In-Reply-To: "Barry A. Warsaw"'s message of "Thu, 3 Jun 1999 14:55:47 -0400 (EDT)"
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net> <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net> <87hfosquc0.fsf@ev.netlab.co.jp> <3753F121.E4A9BE8B@prescod.net> <87yai3ntdt.fsf@ev.netlab.co.jp> <7j3iif$134e$1@nntp6.u.washington.edu> <7j45iv$908$4@cronkite.cc.uga.edu> <7j4hhj$118u$1@nntp6.u.washington.edu> <14166.46756.866947.64007@anthem.cnri.reston.va.us> <87g14942co.fsf@pc-hrvoje.srce.hr> <14166.53171.289824.873109@anthem.cnri.reston.va.us>
Message-ID: <87yai1nkjj.fsf@pc-hrvoje.srce.hr>

"Barry A. Warsaw" <bwarsaw at cnri.reston.va.us> writes:

> The right approach would be to create a subclass of PyInstance (say
> called PyFinalizingInstance) which /does/ call __del__() and then
> create one of these when building a class that has a __del__
> defined.

Would this right approach ever work in Python?  Given Python's
dynamism, it's perfectly legal to write:

    x.__del__ = FN

where x is an instance of a class without a __del__() method.  But
then again, maybe it would be OK to silently ignore such __del__
methods...



From arcege at shore.net  Wed Jun  9 07:26:58 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 09 Jun 1999 11:26:58 GMT
Subject: Using perl in python
References: <7jjoqh$t56$1@nnrp1.deja.com>
Message-ID: <6cs73.2432$nn.739920@news.shore.net>

Brenda <bbailey08 at my-deja.com> wrote:
: I have a bunch of perl libraries that I would like to call in my python
: scripts.  Is this possible?

It is possible;  Tkinter utilizes an embedded Tk interpreter.  You
would have to write a Perl interpreter.  There may be some people
around who might help you with that.  I'm in the middle of embedding
another interpreter myself.

I've found that it is more efficient (and more fun) to convert the
libraries to Python.  The code is invariably clearer.

  -Arcege




From jrush at summit-research.com  Tue Jun 29 06:48:45 1999
From: jrush at summit-research.com (Jeff Rush)
Date: 29 Jun 1999 10:48:45 GMT
Subject: signal handler
References: <376F87CB.DC3802E6@pk.highway.ne.jp> <3772CD4B.38677556@cup.hp.com> <37778711.A2752414@pk.highway.ne.jp>
Message-ID: <VrDQRddcbbs9-pn2-m5ZtPHTdlluI@Fortress>

To catch the Control-C keyboard interrupt, you can just do the following,
depending upon what you are trying to accomplish:

   try:
       pass # Do Whatever
   except KeyboardInterrupt:
       print "*** BREAK ***"

It may be the Python interpreter is interfering with your handler,
since it catches the keyboard interrupt and turns it into an exception.

-Jeff Rush


On Mon, 28 Jun 1999 14:30:41, Thooney Millennier <too at pk.highway.ne.jp> 
wrote:

> I am sorry for my late reply,
> and thank for your suggestion.
> I did't know functions such as
> "getpid()","kill()".
> 
> Well, I tried and realized...
> If you use "kill(getpid(),SIGINT)",
> it works. (ie. callback ok!)
> Instead,if you send "Control-C" by keyboard input,
> it does not. (nothing happens,signal ignored)




From arcege at shore.net  Wed Jun 23 10:08:26 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 23 Jun 1999 14:08:26 GMT
Subject: Creating a new type with an extension module
References: <Ctp*vl52n@news.chiark.greenend.org.uk> <rvy*CQ92n@news.chiark.greenend.org.uk> <QpMb3.908$7X1.231730@news.shore.net> <faSb3.968$7X1.247240@news.shore.net> <ugu*8md3n@news.chiark.greenend.org.uk>
Message-ID: <uT5c3.1060$7X1.271982@news.shore.net>

Sion Arrowsmith <siona at chiark.greenend.org.uk> wrote:
: In article <faSb3.968$7X1.247240 at news.shore.net>,
: Michael P. Reilly <arcege at shore.net> wrote:
:>  http://starship.python.net/crew/arcege/extwriting/pyext.html
:>
:>Let me know how well it reads (I doubt it reads well since it was
:>started this morning).

: It's a lot more readable than just source code, and a lot more
: informative with respect to the type and method structures. Any
: chance of you borrowing Guido's time machine so I can read it
: 48 hours ago?

Guido I'm not. ;)  Besides, I've had a cold for the last four days -
I'm not sure if it's a good idea for you to get some of my jumbled
thoughts from that far back.

  -Arcege




From greg.ewing at compaq.com  Thu Jun 17 18:08:06 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 18 Jun 1999 10:08:06 +1200
Subject: Solution to finalisation problem [Re: fork()]
References: <000401beb886$6a7f2b00$099e2299@tim>
Message-ID: <376971C6.90976DD4@compaq.com>

Tim Peters wrote:
> 
> Greg, this is kind of like asking "if each loop terminates, how can we have
> an infinite loop?".  That is, sure, if you write code that works, it works
> <wink>.

What I meant was, does anyone have a real example of a
case where there would be serious inconvenience if
finalisers couldn't play with each other's data?

Greg



From pduffin at mailserver.hursley.ibm.com  Wed Jun 30 06:10:23 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Wed, 30 Jun 1999 11:10:23 +0100
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <37682DD3.43923243@compaq.com> <7ka2be$qja$1@nntp6.u.washington.edu> <37776FFA.2D85@mailserver.hursley.ibm.com> <lw7looxx7m.fsf@copernico.parades.rm.cnr.it> <876748mcxg.fsf@home.ivm.de> <37788BB9.ABD@mailserver.hursley.ibm.com> <lw67471fyi.fsf@copernico.parades.rm.cnr.it>
Message-ID: <3779ED0F.2C67@mailserver.hursley.ibm.com>

Marco Antoniotti wrote:
> 
> Nope. *best* is the Common Lisp Macro system, with Scheme/Dylan syntax
> stuff a second best.  The notion of 'upvar' in Tcl makes my head

I am surprised at this because Tcl's [upvar] is simply an explicit form
of dynamic binding which I seem to remember is what Lisp uses. (That
is how the 'let' function can be (is) implemented as a lambda function).

> spin. :) Finally, AFAIU, a Tcl "macro" must run as an interpreter of
> the spec. A Common Lisp (Scheme) macro is compiled into regular code
> by read-time expansion.
> 

I find that Lisp macros (while a very powerful and necessary mechanism)
are sooo confusing. They are in essence another language inside Lisp,
and as such introduce inconsistencies. Tcl on the other hand doesn't
need a macro language and as such is much more consistent than Lisp.

If I needed a macro language in Tcl I can just write one.

Lisp and Tcl have a lot more in common than Lispers seem to want to 
acknowledge.

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From sams at quinta.com  Fri Jun 18 13:18:25 1999
From: sams at quinta.com (Sam Schulenburg)
Date: Fri, 18 Jun 1999 17:18:25 GMT
Subject: Python source code documentation?
Message-ID: <7kdv0l$n0g$1@nnrp1.deja.com>

I have been using a freeware product called Autoduck to document my
Python source code.

=============== Example source ====================================
# ######################################
# DATE:         June 17,1999
# PROGRAMMER:   Samuel W. Schulenburg
# This function will print out the command time statistics from<nl>
# getcommandtime(tid) function<nl>
#
#//@pyfunc void | printcommandtime | This function will take a <nl>
#//                          tid and print out the following<nl>
#//                          commandtime structure statictics:<nl>
#//                          1) Total command time counts<nl>
#//                          2) Start time for last command<nl>
#//                          3) Minimum time for all commands<nl>
#//                          4) Maximum time for all commands<nl>
#//                          5) End time for last command<nl>
#//                          6) Time for last command<nl>
#//                          7) Total time for all commands<nl>
#//                          8) Average time for all commands<nl>
#//@parm   integer  | Tid | Target id to use
#//@rdesc  void
#
######################################################################
def printcommandtime(tid):
        s=getcommandtime(tid)
        print "Total command time counts     = ",s[0]
        print "Start time for last command   = ",s[1]
        print "Minimum time for all commands = ",s[2]
        print "Maximum time for all commands = ",s[3]
        print "End time for last command     = ",s[4]
        print "Time for last command         = ",s[5]
        print "Total time for all commands   = ",s[6]
        print "Average time for all commands = ",s[7]
================= End Example ======================================
This product will then generate a windows help file based upon the
information provided above.

I would prefer to have a Python based solution that would generate HTML
files. Can anyone point me towards any information that may start me
down this path, If a solution does not exist, then I may have to write
my own Python parser to handle this.

Sam Schulenburg


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From gmcm at hypernet.com  Wed Jun  2 18:05:01 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 2 Jun 1999 17:05:01 -0500
Subject: pickle vs .pyc
In-Reply-To: <7j43pn$55l$1@mlv.mit.edu>
Message-ID: <1283761144-37459760@hypernet.com>

Michael Vezie writes:
[saving / loading complex structures]
> The obvious choice is, of course pickle, or some flavor thereof. But
> can someone tell me why this wouldn't be faster:
> 
> In the code that does the "pickling", simply do:
> f = open("cache.py", "w")
> f.write("# cache file for fast,slow\n")
> f.write("fast = "+`fast`+'\n')
> f.write("slow = "+`slow'+'\n')
> f.close()
> import cache
> 
> Then, later, when I want the data, I just do:
> 
> from cache import fast,slow

> Am I missing something here?  This sounds like an obvious, and fast,
> way to do things.  True, the caching part may take longer.  But I
> really don't care about that, since it's done only once, and in the
> background.  

Not at all. Where x == eval(repr(x)), this is dandy. You can even use 
pprint to dump into a humanly digestible format. You can then edit 
and reload.

But marshall should be even faster.

- Gordon



From garry at sage.att.com  Fri Jun 11 14:03:55 1999
From: garry at sage.att.com (Garrett G. Hodgson)
Date: Fri, 11 Jun 1999 18:03:55 GMT
Subject: newbie questions
References: <x7emjizs8s.fsf@norden.math.tntech.edu>
Message-ID: <37614F8B.CFFEE5A2@sage.att.com>

Jeff Norden wrote:

> 1) Is there no equivalent to perl's  "use strict" in which all variables
> have to be declared?

nope.

> 2) Is the above the correct way to increment a variable?  I can't find an
> equivalent for "++" or "+="

var = var + 1

> 3) These are rather minor, but is there no way to "forward declare" function
> definitions?  I'd rather put them at the end of the file instead of the
> beginning.

def main():
	func1()
	func2()

def func1():
	print 'func1 called'

def func2():
	print 'func2 called'


if __name__ == '__main__':
	main()



-- 
Garry Hodgson			comes a time
garry at sage.att.com		when the blind man
Software Innovation Services	takes your hand, says,
AT&T Labs			"don't you see?"



From jeff at ollie.clive.ia.us  Wed Jun 16 10:26:25 1999
From: jeff at ollie.clive.ia.us (Jeffrey C. Ollie)
Date: Wed, 16 Jun 1999 09:26:25 -0500
Subject: Support for mSQL?
References: <3765b5ee@news.mdc.net>
Message-ID: <slrn7mfd0n.su1.jeff@worf.netins.net>

-----BEGIN PGP SIGNED MESSAGE-----

On 14 Jun 1999 22:09:50 -0500, Daniel <daniel at virtualvoices.com> wrote:
>
>From the 'python.org' web site the mSQL module is
>no longer being supported.  Or has someone else picked 
>up it's support?

No, no one has picked up support for the mSQL module.  If you aren't
already committed to mSQL I'd suggest checking out PostgreSQL or
MySQL.


[A copy of the headers and the PGP signature follow.]

Date: Wed, 16 Jun 1999 09:26:25 -0500
Newsgroups: comp.lang.python
Organization: netINS, Des Moines, IA, USA
References: <3765b5ee at news.mdc.net>
Subject: Re: Support for mSQL?

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2
Comment: AnySign 1.4 - A Python tool for PGP signing e-mail and news.

iQCVAwUBN2e0FJwkOQz8sbZFAQHnHwQArJ8M2bE9hWW2XhKEOkZo224QZulLOm8z
Jf95sbyy2Aw0Stth1IBy1k3EErd0yg4lo0SLaWj/leDF1DTYYCzhGZ14Q0SdUkZ9
8qqAQP4LTx2nKBzjMcMvWW6xYa9WpCkTdkdWbr7KYUb+k62/C2t79a2uLWgDjekj
C4iaOW+7UHU=
=wrPN
-----END PGP SIGNATURE-----
-- 
Jeffrey C. Ollie                     |            Should Work Now (TM)



From jeremy at cnri.reston.va.us  Thu Jun  3 11:13:55 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Thu, 3 Jun 1999 15:13:55 GMT
Subject: Python 2.0
In-Reply-To: <7j4hhj$118u$1@nntp6.u.washington.edu>
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7imr6b$1s0$1@nnrp1.deja.com>
	<87pv3iq6ss.fsf@ev.netlab.co.jp>
	<3752C76D.604CD474@prescod.net>
	<87n1ykr0z9.fsf@ev.netlab.co.jp>
	<375372F6.A874554@prescod.net>
	<87hfosquc0.fsf@ev.netlab.co.jp>
	<3753F121.E4A9BE8B@prescod.net>
	<87yai3ntdt.fsf@ev.netlab.co.jp>
	<7j3iif$134e$1@nntp6.u.washington.edu>
	<7j45iv$908$4@cronkite.cc.uga.edu>
	<7j4hhj$118u$1@nntp6.u.washington.edu>
Message-ID: <14166.39460.846897.694749@bitdiddle.cnri.reston.va.us>

>>>>> "DC" == Donn Cave <donn at u.washington.edu> writes:

  DC> Does JPython use reference counts, and invoke delete methods as
  DC> soon as reference count drops to 0?

No.  It relies on Java to do memory management.  And even that doesn't 
always work as expected :-).  I was going to show you the example from 
the Python FAQ, where files are opened in a loop and implicitly closed 
when the reference disappears.  That won't happen in JPython, because
the finalization isn't guaranteed to happen.

But I got a completely different error:

JPython 1.0.3 on java1.2
Copyright 1997-1998 Corporation for National Research Initiatives
>>> import os
>>> for file in os.listdir("."):
...     f = open(file)
...     buf = f.read()
... 
Out of Memory
You might want to try the -mx flag to increase heap size

Sheesh!  And people want to use Java as the sole platform for Python
<0.5 wink>.

Jeremy



From MHammond at skippinet.com.au  Mon Jun 28 21:16:29 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Tue, 29 Jun 1999 11:16:29 +1000
Subject: RuntimeError?
References: <19990627112432.A29097@optiplex.palga.uucp> <000401bec10d$b5bc3a60$e19e2299@tim> <19990628105921.A26266@optiplex.palga.uucp>
Message-ID: <7l96lm$62h$1@m2.c2.telstra-mm.net.au>

Gerrit Holl wrote in message <19990628105921.A26266 at optiplex.palga.uucp>...

>> can be <wink>.  It generally seems to mean something bad and unexpected
>> happened, but not bad enough to shut down the interpreter (things *that*
bad
>> raise SystemError).

Actually, I dare to correct Tim on this point.  SystemErrors typically will
_not_ shutdown the interpreter (they used to, but not any more!)

>Ah, I understand. But what kind of things are bad enough to raise
SystemError?
>core dump?

My take is this:  RuntimeError is raised in a situation that could not
reasonably have been predicted by the programmer (eg, someone deleting
sys.stdout).  A SystemError OTOH usually means an internal Python error
(Python or an extension).  The most common example of a SystemError that I
see is when an extension module returns NULL but doesnt set a Python
exception - ie, a bug that needs to be fixed by the original programmer, and
nothing the end user could do would prevent this. (roughly speaking, of
course :-)

A core dump is unfortunately (or fortunately depending on your POV) not
mapped to a Python exception.

Mark.






From Vladimir.Marangozov at inrialpes.fr  Tue Jun  8 18:01:20 1999
From: Vladimir.Marangozov at inrialpes.fr (Vladimir Marangozov)
Date: Wed, 09 Jun 1999 00:01:20 +0200
Subject: fork()
References: <000601beb1c7$682c6220$559e2299@tim> <7jjghd$9il$1@cronkite.cc.uga.edu> <375D5826.23197692@inrialpes.fr> <7jjrot$h16$2@cronkite.cc.uga.edu>
Message-ID: <375D92B0.A67930A1@inrialpes.fr>

Graham Matthews wrote:
> ...
> I was actually wondering if anyone had done any work putting the Boehm
> collector under Python. The Boehm collector was designed to run in a
> C environment.
> 

FAQ 6.14

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



From graham at sloth.math.uga.edu  Thu Jun  3 11:20:06 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 3 Jun 1999 15:20:06 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>  <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>  <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net>  <87hfosquc0.fsf@ev.netlab.co.jp> <3753F121.E4A9BE8B@prescod.net> <87yai3ntdt.fsf@ev.netlab.co.jp> <7j3iif$134e$1@nntp6.u.washington.edu> <7j45iv$908$4@cronkite.cc.uga.edu> <7j4hhj$118u$1@nntp6.u.washington.edu> <7j4jbe$dhh$1@cronkite.cc.uga.edu> <37564A02.F115B9C5@infercor.no>
Message-ID: <7j66f6$il5$3@cronkite.cc.uga.edu>

Paul Boddie (paulb at infercor.no) wrote:
:So, we have established that it can indeed be done, but reaching back into the
:depths of time and pulling out the original proposal - being that Python should
:have its primary (or only) implementation running in the Java environment - we
:can see that regardless of the existence of satisfactory GC schemes, the 
:typical Java implementation surely falls short of what many people require.
: 
:Therefore, whilst there may well be a need for more sophisticated GC in Python
:2.0, we must conclude that the Java environment may not be the appropriate 
: place in which such technologies are to be found.

Can I suggest you read the Java spec. As someone else has pointed out
Java memory management is done via an interface. Hence you can implement
any memory management scheme you like in Java.

graham
-- 
                          So I turned around
                    And 40,000 headmen bit the dirt
                      Firing twenty shotguns each
                        And man it really hurt



From notepower at surfnetusa.com  Sun Jun 27 14:31:55 1999
From: notepower at surfnetusa.com (Notepower)
Date: Sun, 27 Jun 1999 11:31:55 -0700
Subject: Organize those Code Snippets!
Message-ID: <L7ud3.94920$_m4.939999@news2.giganews.com>

I have placed a fully functional preview version of our software,
Notepower, which is a windows based browser/editor specifically
optimized for storing and organizing code snippets and other types of
notes...

As such it is the perfect tool for organizing code and HTML
snippets! Try it out today at:

http://surfnetusa.com/notepower








From jdrudnicki at yahoo.com  Wed Jun  2 04:06:13 1999
From: jdrudnicki at yahoo.com (Jim Rudnicki)
Date: Wed, 2 Jun 1999 01:06:13 -0700
Subject: need help checking for stack overflow in Win32
References: <5lk8to3su3.fsf@eric.cnri.reston.va.us>
Message-ID: <TB553.466$HD.96226@typhoon-sf.snfc21.pbi.net>

> I could use some help from someone with a little time, some knowledge
> of Python internals, and some Win32 C expertise.  On Win32 platforms,
> infinite recursion causes an Application Error (a fatal windows
> error).  There is some machinery in Python to check for this
> condition, but apparently it doesn't work on Win32.  Please help!

My vague memory is that page fault exceptions are thrown as the stack grows.
Unfortunately, everything is throwing page faults keeping the virtual memory
manager busy.  Hooking into that exception handler seems like real work.

Not knowing what is going on inside, let me ask this: how do you count
recursion depth?  I presume a counter somewhere (no duh).  How about instead
of counting calls, you count cumulative space pushed.  Now, the recursion
limit would be in bytes.  This allows many calls for functions with small
parameter lists, but also accomodates the program that trys to pass bitmaps
on the stack.  You set it to 800,000 and it always works?  Now of course the
SP itself is this count.  If we could only get that...

I'll try looking through Richter tommorrow and see if anything pops out (no
pun intended)

Jim





From jrush at summit-research.com  Thu Jun 24 07:03:54 1999
From: jrush at summit-research.com (Jeff Rush)
Date: 24 Jun 1999 11:03:54 GMT
Subject: An Adventure in Python and Xanadu Docuverse Tumblers
References: <376895db.114275851@news.teleport.com> <376F3E50.855741DA@mincom.com>
Message-ID: <VrDQRddcbbs9-pn2-rgQPhX15d43M@Fortress>

On Tue, 22 Jun 1999 07:42:08, Martin Pool <martinp at mincom.com> wrote:

> Kirby Urner wrote:
> > ...
> > Like C (and unlike Java), Python permits operator overloading,
> > so if we have A and B defined as quadray objects, then we can
> > write C = A+B and get a new quadray object that's the sum of
> > the two, just as we'd expect in a textbook math notation.  
> > ...

I missed the beginning of this thread, but if you are talking about
some form of tuple addition, I've implemented Ted Nelson's (of
Xanadu Hypertext fame) transfinite tumbler arithmetic in Python at:

    http://starship.python.net/crew/jrush/xanadu/

If you aren't talking about that, well, never mind... <grin>

-Jeff Rush




From htrd90 at zepler.org  Mon Jun 28 17:23:47 1999
From: htrd90 at zepler.org (Toby Dickenson)
Date: Mon, 28 Jun 1999 21:23:47 GMT
Subject: Pythonwin and Office 2000
References: <377648A7.9B9BE59F@prescod.net> <7l672r$5k0$1@m2.c2.telstra-mm.net.au> <3776D1E6.45E1AAF7@prescod.net> <7l7bbt$42u$1@m2.c2.telstra-mm.net.au>
Message-ID: <377cdaaa.6963555@news.freeserve.net>

"Mark Hammond" <MHammond at skippinet.com.au> wrote:

>[CC'd from the newsgroup to the python-com developers list]
>
>Paul Prescod wrote in message <3776D1E6.45E1AAF7 at prescod.net>...
>
>>Is there an equivalent to EnsureModule that works from the textual library
>>name? I know that the textual library name still changes from version to
>>version but I don't think it is Pythonish (easy, clean and readable) to
>>put UUIDs in code at all.
>
>I agree.  Worse, as you found out, it breaks when upgrading the COM object.
>However, the only way to do it textually would be to search the registry for
>the name.  Depending on the system, this can be significant.  The IID and
>version info give us a direct key into the registry.  On my development
>system, searching for an IID name could take a few seconds, and (IMO) we
>can't afford this hit just to check.

If you need to ensure that the makepy wrapper is always available, independent
of type library, then I have found the most convenient approach is to copy the
appropriate module out of the genpy directory and import it directly. No UUIDs
in sight, and it works with freeze. This is effectively what Mark describes
below.

makepy should always generate the same wrapper for the same version of each type
library, so you don't lose anything after this initial inconvenience. 

In theory, the old wrapper should continue to work after upgrading the com
server. In practice, any uprade involves some risk but this strategy seems to
minimise it.

>I can't think of a better solution.  The Python equivilent to the VB system
>would be for the developer of a script to nominate the type libraries they
>reference, and to generate the necessary makepy support into a single file.
>This single file would then be shipped with the script.

(this is possible to hack today, described above)

> (VB puts the
>generated code in the same .exe, hiding this extra distribution - in Python
>terms, this would be integrating with freeze etc).  To go this route would
>really require that makepy no longer generate source code, but instead
>generate a tight binary format (which may be desirable anyway).

Why single out the makepy wrapper for this compacting effort? Several other
modules lead to embarrasingle big .pyc and frozen .exe files, but I have never
found this a practical problem.

VB and C++ (and maybe Delphi) gain an advantage from the compile-time approach
since, in combination with static typing, they can optimise into vtbl method
calls. Today, Python is firmly tied to Invoke. We could gain this same
_performance_ advantage by extending makegw to understand type libraries, but
that just changes the problem into having big .dlls. And you also need a
compiler.

(Note: this may be a practical approach if you need to run many python processes
on the one machine; the dll images could be shared between processes. However I
don't see it as a general solution)

>VBScript, for example, does not have the concept - it only understands (in
>Python parlance) "dynamic dispatch".  Delphi, AFAIK, also makes it a
>build-time step, rather than a run-time step.
>However, it is worth noting that speed is no longer a good justification for
>using makepy.  Recent changes to dynamic dispatch mean that we are now
>comparable speed wise to VB without makepy support.



>  The reason you often
>now need makepy support is to get decent parameter support, but the problem
>is worse than that.

Is the parameter support still a continuing problem? I had the impression that
in general only older servers had this problem, with newer servers being much
better behaved?

>  In many non-trivial cases, the code you write is simply
>not compatible between makepy supported and dynamic dispatch.  So, if you
>want to ensure your code will run when installed, you must _insist_ on
>makepy support.  (ie, silent failure of makepy is usually not a good option,
>as you are simply delaying the inevitable error, but in a less obvious way).

Agreed.



Toby Dickenson



From jepler at inetnebr.com  Wed Jun 30 20:53:08 1999
From: jepler at inetnebr.com (Jeff Epler)
Date: Thu, 01 Jul 1999 00:53:08 GMT
Subject: Help: Can you explain this error? (contains code snip)
References: <377AAD25.96947EBA@alliedsignal.com>
Message-ID: <slrn7nlf4u.5ng.jepler@craie.inetnebr.com>

On Wed, 30 Jun 1999 16:49:58 -0700, H. P. Friedrichs
 <@alliedsignal.com> wrote:
>Hello:
>
>This is week 2 for me in python...What fun!
>
>I have stumbled across an interesting problem. When the code below is
>run under Win95 (python 1.5.2),it works without problems. If the same
>code is run under Linux (python 1.4) I get the error:

Make things simpler:

>>> int("123")

this works in Python 1.5.x and fails in 1.4.  I believe this feature is
present in all 1.5.x and no standard 1.4 versions.

Use

>>> string.atoi("123")
to work on both versions.

Jeff



From andrew at andrewcooke.free-online.co.uk  Sun Jun 27 04:31:56 1999
From: andrew at andrewcooke.free-online.co.uk (Andrew Cooke)
Date: Sun, 27 Jun 1999 08:31:56 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net>
Message-ID: <7l4nht$jl7$1@nnrp1.deja.com>


Tk isn't the standard GUI for Python - it's the standard GUI.  It's
been around for aeons and I, for one, would be a tad annoyed if I
had to learn a new GUI each time I learnt a new language.

There seem to be two problems with it:

1. Speed.  For most applications, is this an issue?  Python is, above
all, a scripting language.  Speed is not important.  What you can do -
and do fairly simply with Tk, once you know it, is build a GUI front-
end to whatever programs you are tying together.  The text widget, for
example, is amazingly flexible with very little code.

2. Widgets.  Tk certainly doesn't have a modern set of widgets.  There
are extensions, however.  And with Idle, it looks as though the range
included in standard Python is increasing (does anyone complain Idle
is slow?!)

If you are programming games or graphics-intensive code then sure, use
another (C-based) GUI - Python is designed to make this easy, just
as you can drop into C if non-graphics code is CPU intensive.

For others, I suspect the lack of good docs for Tk is a more serious
problem.  There is a pretty steep learning curve involved if you're
meeting Tk for the first time (and if you do know it, finding the
right way to express it in Python can still mean reading the Tkinter
code).

I admit - I haven't used the alternatives (like the vast majority of
people who have been using computers for some time, I suspect - we
were brought up on Tk).  But I have read all this thread, and when
people start saying that wxWindows is better for "simple" graphics I
start worrying...

Andrew



In article
<52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162 at lp.airnews.net>,
  mikael at pobox.com (Mikael Lyngvig) wrote:
> Python v1.5 adopted TkInter as its standard GUI framework, even to the
> extent of including TkInter in the (Intel/Windows) installation
> package.  Unfortunately, TkInter appears to run sluggish [1] on all
> the Intel/Windows computers I've evaluated it on (which includes
> several Pentium II 400+ computers).
[...]
> I'd like some discussion on whether or not it is a good idea to
> abandon TkInter, before it gets too widely used, and instead adopt
> wxPython as the primary GUI framework for Python.
>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From j_belbo at hotmail.com  Tue Jun 29 04:13:39 1999
From: j_belbo at hotmail.com (Jacobo Belbo)
Date: Tue, 29 Jun 1999 10:13:39 +0200
Subject: How to use Perfmon Library
References: <7kvrrc$2d3$1@nnrp1.deja.com> <7l10a5$3bt$1@m2.c2.telstra-mm.net.au>
Message-ID: <37788032.B6B6F04E@hotmail.com>

Thanks for this info
My target is to monitor for certain process on a remote NT Server
  - Memory usage
  - Processor usage
  - Is it alive

with killProcName.py I can just working with a local machine

After a search on MSDN it seems that I must work with
KHEY_PERFORMANCCE_DATA + other specific call
I have found this in the Python Perfmon module from a certain Mark H.
You know this guy ? ;-)
But without example is not easy to use

Jacobo

Mark Hammond wrote:

> There are 2 ways to use the performance monitor:
>
> * If you just want to collect performance data from an existing process that
> is "performance monitor aware" (or you just want to see data that NT
> provides, such as memory usage, processor utilization etc.  This is quite
> simple to use - you use the win32pdhutil module - there is a sample
> "killProcName.py" that shows how to use these functions to query the local
> machine - a remote machine is not much harder.
>
> * If you want your application to provide custom perfmon data that another
> process can read, then this is significantly harder - you really need to
> read the NT documentation on how to do this; Python just wraps the standard
> NT APIs.
>
> Mark.
>
> belbo_j at my-deja.com wrote in message <7kvrrc$2d3$1 at nnrp1.deja.com>...
> >Hello,
> >
> >Is it anybody that can give me some information
> >about the usage of Perfmon library ?
> >
> >I want to monitor a process on a remote NT server




From ragica at news.vex.net  Sat Jun  5 19:55:02 1999
From: ragica at news.vex.net (Ragica)
Date: 5 Jun 1999 23:55:02 GMT
Subject: Idle comments
References: <000d01bead82$2a2564c0$309e2299@tim>
Message-ID: <7jcdcm$19qi$2@hub.org>

On Thu, 3 Jun 1999 01:30:17 -0400, Tim Peters <tim_one at email.msn.com> wrote:
>[Ragica, zzz at NO-MAIL.com, which means he'll probably never see this since
> the mail->news gateway is down]

I see it, i see it.

>Well, it's *Python* code:  you look in the source, find where it sets the
>colors, and change 'em to whatever you want.  If you grab the development

I know it's in the code. I have even looked for where (believe it or
not). In the IDLE readme for example it tells you where and what to
change to adjust the syntax highlight colours--but not the default
background colour--, I looked around there. Alas, to no avail. I
freely admit i find that code confusing. I am not quite the python
programming genius as yoursly truly. 

So i have watched in vain for months hoping someone might let slip
this apparently closely guarded secret as to the where and what to
edit to make this miraculous IDLE environmental change.

>IDLE off the CVS repository at python.org, you'll find that it now has a
>IdlePrefs.py file containing a ColorPrefs class you can fiddle more directly

Thank God for that! I have to tell you, even if the guidos of the world
are ever infinitely more programatically exalted than insignificant I...
i simply can't fathom people who hard code stuff into their programs! 
Even at early stages. In my books setting-type variables at very least
should go somewhere highly visible and up-front... so as to be easy 
to access and find. But that's just me, perhaps. 

>start-changing-one-color-and-you'll-end-changing-'em-all-ly y'rs  - tim

Precisely (not to mention concisely, and certainly not to mention
dashingly) what my plan would be... 





From " HPeter.Friedrichs" at alliedsignal.com  Wed Jun 23 11:42:37 1999
From: " HPeter.Friedrichs" at alliedsignal.com (H. P. Friedrichs)
Date: Wed, 23 Jun 1999 08:42:37 -0700
Subject: Newbie: How do you "clear" PythonWin between runs?
References: <37700E61.5388E101@alliedsignal.com> <cUVb3.1952$U6.6930@newsr2.twcny.rr.com>
Message-ID: <3771006D.22A9EC2D@alliedsignal.com>

Darrell:

What exactly is this URL? What document were you referencing? Is there another
way to get it?

I can't seem to connect to this URL...I keep getting "server busy or offline"
type errors.

Thanks,
Pete


Darrell wrote:

> http://grail.cnri.reston.va.us/cgi-bin/faqw.py?query=reload&querytype=simple
> &casefold=yes&req=search
>
> --
> --Darrell
> H. P. Friedrichs <" HPeter.Friedrichs"@alliedsignal.com> wrote in message
> news:37700E61.5388E101 at alliedsignal.com...
> > I've just started playing with python, so all of this is very new to me.
> > I have been writing some scripts, saving them to *.py files, and running
> > them under PythonWin 1.5.2 . The process is iterative; I code, save,
> > run, code, save, run, etc.
> >
> > What I have noticed is that python doesn't "reset" between runs.
> > Sometimes error  messages reappear even though I have fixed the problem,
> > because some artifact of a previous run lingers.
> >
> > So, my question is this: How can I "flush" or "clear" the python
> > interpreter between successive runs? Is there a command I can add to my
> > script to initialize python before running the script?
> >
> > Thanks,
> >
> > Pete
> > hpeter.friedrichs at alliedsignal.com
> >

-----------------------------------------------
| The opinions expressed here are my own,     |
| and not necessarily those of my employer.   |
-----------------------------------------------





From dtyler at easynet.co.uk  Sat Jun 26 19:53:31 1999
From: dtyler at easynet.co.uk (Derek Tyler)
Date: Sat, 26 Jun 1999 23:53:31 GMT
Subject: COM Client - WithEvents ?
Message-ID: <37755d0b.1310959@news.easynet.co.uk>

I would like to make a client for Internet Explorer 3.02 in Python
with event handlers like VB WithEvents - is this possible ?

I've succeeded in making a client without events.

I've not found anything in the PythonCom  documentation about using
events.




Derek Tyler
dtyler at easynet.co.uk



From news at dorb.com  Wed Jun 16 12:34:01 1999
From: news at dorb.com (Darrell)
Date: Wed, 16 Jun 1999 12:34:01 -0400
Subject: GC In Python: YAS (Yet Another Summary)
References: <1282571877-108998356@hypernet.com>
Message-ID: <zoQ93.198$U6.1646@newsr2.twcny.rr.com>

> Check out Lars's plumbo.py:
>
> <URL:http://www.stud.ifi.uio.no/~larsga/download/python/plumbo.py>
>
> - Gordon
>
Thanks Gordon.

At the risk of sounding like a lazy whiner. This would be more useful if it
could run more like profile or pdb. Having to add calls like
"r.add_object(non_GCed_object)" make it less useful.

Is there a way to capture the creation of any instance and call r.add_object
?
>>> import sys
>>> def p(frame, even, arg):
...     print frame, even, arg
...
>>> sys.settrace(p)
>>> l=[]
<frame object at 7c9410> call None
>>> d={}
<frame object at 7c9410> call None
>>>

In perfect world for lazy whiners this works,  plumbo.run("myStuff") and it
gives me a report.

--
--Darrell





From mlauer at asmoday-atm.rz.uni-frankfurt.de  Wed Jun  2 20:32:17 1999
From: mlauer at asmoday-atm.rz.uni-frankfurt.de (mlauer at asmoday-atm.rz.uni-frankfurt.de)
Date: 3 Jun 1999 01:32:17 +0100
Subject: idle processing
Message-ID: <3755bf01@nntp.server.uni-frankfurt.de>

Hi,

is there an appropriate place to insert a method
which does some calculations while the application
is "idle" (user is not using it or just moving
the mouse...) - somewhere within the tkinter-
mainloop ? 

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





From nbornstein at plr.com  Tue Jun 29 19:54:27 1999
From: nbornstein at plr.com (Niel M. Bornstein)
Date: Tue, 29 Jun 1999 19:54:27 -0400
Subject: FTPlib - how to optimize transfer?
Message-ID: <nbornstein-ya02408000R2906991954270001@news.atl.bellsouth.net>

I am using FTPlib on the Mac, and I am noticing binary uploads seem to go
very slowly.  Is there any way to optimize the block size or something to
speed it up?  Should the block size match that of the open() call used to
open the file?  What's a good block size to use, anyway?

Email replies preferred but I'll check back here too.

Thanks,
Niel



From alex at mindlube.com  Tue Jun 22 01:01:54 1999
From: alex at mindlube.com (Alex Rice)
Date: 21 Jun 1999 23:01:54 -0600
Subject: Installing ODBC on PythonWin
References: <7kf84e$fog$1@holly.csv.warwick.ac.uk>
Message-ID: <m3g13kygjh.fsf@monsoon.swcp.com>

esuzm at primrose.csv.warwick.ac.uk (Andrew Clover) writes:

>  Oh well. I downloaded and installed MySQL and though the process wasn't
> terribly pleasant, at least the damned thing *works* now. Anyone want to
> criticise this choice of server/suggest something else?

MySQL rocks. Although I've never used the ODBC layer or the windows
version of it though.

-- 
Alex Rice    |    alrice at swcp.com    |    http://www.swcp.com/~alrice
    Current Location: N. Rio Grande Bioregion, Southwestern USA



From vlachid at otenet.gr  Fri Jun 25 17:01:56 1999
From: vlachid at otenet.gr (Vlachidis Costas)
Date: Sat, 26 Jun 1999 00:01:56 +0300
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <JWQc3.2620$U6.10996@newsr2.twcny.rr.com>
Message-ID: <3773EE44.F231F1E4@otenet.gr>

Hi
Ok, but why somebody thar starts now to learm Python MUST learn Java too?
Costas

Darrell wrote:

> I know less than nothing about this subject, but have a though to
> contribute.
>
> What about using a tool like this:
> http://www.ndim.edrc.cmu.edu/dougc/jpi/Home.html
> To make use of Java's GUI tools.
>
> Yes speed may be an issue, but it would work on the Mac and book stores are
> loaded with Java....
> Then Jpython and Cpython would share the same GUI tools.
>
> --
> --Darrell
-------------- next part --------------
A non-text attachment was scrubbed...
Name: vlachid.vcf
Type: text/x-vcard
Size: 366 bytes
Desc: Card for Vlachidis Costas
URL: <http://mail.python.org/pipermail/python-list/attachments/19990626/d81cd889/attachment.vcf>

From che at debian.org  Fri Jun 11 01:32:35 1999
From: che at debian.org (Ben Gertzfield)
Date: 10 Jun 1999 22:32:35 -0700
Subject: Created a Python module for ID3 MP3 tag manipulation
References: <yttaeu7z4o8.fsf@gilgamesh.cse.ucsc.edu>
Message-ID: <ytt7lpbz4i4.fsf@gilgamesh.cse.ucsc.edu>

>>>>> "Ben" == Ben Gertzfield <che at debian.org> writes:

    Ben> A simple example of using the module:

    Ben> id3info = ID3('/some/file/moxy.mp3');

Of course I have to make a typo in my example. *BONK* No semicolon
there of course. I blame Perl-fingers. :)

Ben

-- 
Brought to you by the letters V and T and the number 19.
"I choose YOU! Pikachu!"
Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/



From c.evans at clear.net.nz  Sat Jun 12 19:43:48 1999
From: c.evans at clear.net.nz (Carey Evans)
Date: 13 Jun 1999 11:43:48 +1200
Subject: Extending Python with C
References: <7jp409$qu5$1@nnrp1.deja.com>
Message-ID: <87ogil9e8b.fsf@psyche.evansnet>

John Fisher <jfisher at are.berkeley.edu> writes:

> I'm a relative Python newbie having some trouble getting a module to
> load.  I wrote a simple one, using as a model the example given on the
> Python website, just to get the hang of it; however, when I try to load
> the compiled file I get a rather baffling error:
> 
> ImportError: ./echo.o: ELF file's phentsize not the expected size

Can you tell use what you're actually doing?  I've never encountered
this error, so I can't suggest how to fix it.  It does seem strange
that you're importing a .o file, though.

FWIW, this is my procedure for compiling one of my own extensions:

$ cat Setup.in
*shared*
iconv iconvmodule.c
$ cp /usr/lib/python1.5/config/Makefile.pre.in .
$ make -f Makefile.pre.in boot
[...]
$ make
gcc -fpic  -g -O2 -I/usr/include/python1.5 -I/usr/include/python1.5 -DHAVE_CONFIG_H -c ./iconvmodule.c
gcc -shared -lc  iconvmodule.o  -o iconvmodule.so
$ python
Python 1.5.2 (#0, Apr 21 1999, 15:19:36)  [GCC egcs-2.91.66 Debian GNU/Linux (egcs-1.1.2 release)] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import iconv
>>> iconv
<module 'iconv' from './iconvmodule.so'>
>>> dir(iconv)
['INCOMPLETE', 'Iconv', 'IconvType', '__doc__', '__file__', '__name__', 'convert', 'error']

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

	     "I'm not a god.  I've just been misquoted."



From hniksic at srce.hr  Mon Jun  7 15:25:29 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 07 Jun 1999 21:25:29 +0200
Subject: waste of resources ?
References: <375C0F91.1C73E4E3@icrf.icnet.uk> <375C16A9.5ADC4990@icrf.icnet.uk>
Message-ID: <87emjnzuc6.fsf@pc-hrvoje.srce.hr>

Arne Mueller <a.mueller at icrf.icnet.uk> writes:

> Im running python 1.5.2b2 on SGI power challange irix64 version 6.5.
> Everytime a child process is created via fork(), it realy does it's
> job but never dies, instead the number of zombie processes (as
> reported by the 'top' program) steadily increases, though I call the
> sys.exit(0) method for the child. Any idea how exit a process
> without creating a zombie?

The parent should wait() for its children.  See a Unix programming
manual for more details.



From matomira at iname.com  Wed Jun 16 11:06:23 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Wed, 16 Jun 1999 17:06:23 +0200
Subject: <insert favourite flame-bait here> broken
References: <1282792849-95706075@hypernet.com> <3764D7C3.718E125E@iname.com> <002f01beb64d$58e7b5f0$f29b12c2@pythonware.com> <37651951.ED54894E@acm.org> <005501beb674$c5575890$f29b12c2@pythonware.com> <376621CF.88C95B5E@iname.com> <150619990851173299%pereira@research.att.com> <3766653A.D9F6E99B@iname.com> <014301beb7e3$94707610$f29b12c2@pythonware.com>
Message-ID: <3767BD6F.7BEA65E9@iname.com>

<WASTE OF BANWIDTH>

Fredrik Lundh wrote:

> > "cult" "kiddie" "flamer" "dogmatic"
>
> interesting.  a lot of new cool things to add
> to my business card.  given that fernando has
> never met me in person, it's amazing what he
> could figure out just by reading a usenet post

Just to clarify:

"cult" not used in any negative sense, and I never called
anyone "kiddie". I only wondered exists(k) / reads-python(k) ?

As you can see from the tags, you can see I how understand
what the problem is. Sorry if I've never had less than T1 for free,
and never stepped out from `expensive' newsgroups. For some of
us `old guys' this thing about people paying to take even the subjects
home is new (after 10 years wired, who reads guidelines, if they have
always been basically the same?).

</WASTE OF BANWIDTH>

Constructive bit:

Put in your FAQ and guidelines to warn about this to any new guy
before somehow making him feel he is being flamed.

I also have to go through Kent's piece and see with him, because the
`intersection'
part is very misleading. The people interested is that in
intesect(A,I) U intesect(B,I) where I is the set of ALL people
interested in issue I
which will in general be very different from intersect(A,B)






From fatjim at home.com  Thu Jun  3 22:36:31 1999
From: fatjim at home.com (Jim Meier)
Date: Fri, 04 Jun 1999 02:36:31 GMT
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net> <aahzFCpHw8.Gxt@netcom.com> <7j4e2u$8rh$1@m2.c2.telstra-mm.net.au> <37564132.2D01B7C3@pop.vet.uu.nl> <3756BAFC.2066A636@home.com> <3756F071.E717D0FA@pop.vet.uu.nl>
Message-ID: <37573C16.D0233810@home.com>


Martijn Faassen wrote:

> Jim Meier wrote:
> [snip]
>
> You're the guy who did the GGI wrapper, right? I've been playing with it
> a little, now I need more docs... Perhaps I'll go write some.

Please feel free - I'm a very poor documentor. You might also want to check out David
Schere's python wrapper, available from the same place as mine. His wrapping is very
complete - he used SWIG to generate it. I'm in the process of adding the functionality
of my wrapper to his code - things like the indexable color objects, etc.. His module
will provide a complete, mostly direct wrapper for the GGI library functions, while
mine will try to be a more 'Pythonic' interface; hopefully with features that make it
a little more efficient to do graphics work in Python. Ideas on how this will be
accomplished are being auditioned now, so speak up!
One idea I'm toying with is adding a GGI display target to PIL..

>
> >
> > Wow, talking about old H/W makes me sound old - I'm only 19!
>
> Well, computer tech moves quickly (I'd say 'in internet time' but it
> already did before the net was popular).
>
> I remember the days before the great internet explosion, 1991 and such,
> where I could talk about the marvels of the net and get looks 'Martijn
> is talking about computer stuff again'. These days I sit in the bus and
> hear people talk about 'getting email'. And I'm only 25. :) I can go on
> and on about the MSX-2, its VDP chip and Z80 assembler. :)

Oh, I remember those days(Not the MSX-2, though!). Back when the dream hadn't been
ruined by being implemented for the masses! I wish I could still amaze people with
computer trivia - now they just yawn, like you were talking about repairing a TV.
*sigh* :)

> This made me come up with the Quantum-Martijn-Solipsist theory of
> Reality Alteration: 'If Martijn rambles about something exciting to
> people that people don't get, reality will change within a few years so
> that suddenly everybody is using it.'
>
> I'm rambling about Python these days. Yesterday I read a paper from
> Guido about how he was thinking about bringing Python (or some successor
> language) to *billions* of users. :)

Just think of all the nasty code _that_ would generate... I can't beleive some of the
English that some people speak, and am horrified to think that when I'm 30, I might
end up teaching "Python Literacy" to a group of bored and whiny (and by that point,
negative-attention-spanned) high school students and trying to explain the design
philosophy of "The-Right-Way"... !!

Putting that thought out of mind..
- Ole' (and facetious) long-in-the-Tooth Jim.




From greg.ewing at compaq.com  Thu Jun 10 21:41:56 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 11 Jun 1999 13:41:56 +1200
Subject: dict vs kjBuckets vs ???
References: <000e01beb21f$a4c5a860$2fa22299@tim> <375ffb1d.26189164@news.cyf-kr.edu.pl>
Message-ID: <37606964.9E320111@compaq.com>

MK wrote:
> 
> In some book on algorithms I've read that after inserting limited
> number of items performance of operating on hash tables
> drops dramatically.

My understanding is that Python dicts expand their
hash tables before they become dangerously full,
so this behaviour doesn't occur.

Greg



From matomira at iname.com  Sat Jun 12 12:45:51 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Sat, 12 Jun 1999 18:45:51 +0200
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM>
Message-ID: <37628EBF.207F7A8F@iname.com>


Cameron Laird wrote:

> www.sunworld.com/swol-11-1998/swol-11-regex.html>.  I
> wish the Guile folk well, and certainly RSM and others
> continue to invest energy in it.  From my distant ob-
> servation post, it doesn't feel as though it has the
> critical mass of resources necessary to "take off".

Maybe I'm dreaming, but I have the impression we've just passed the global
minimum in the Lisp usage history. With guile as a unification theme, suddenly
it becomes
very interesting to learn scheme, as a lot of folks end up using one or another
GNU tool. As a side effect, some newcomers to Tk that would have normally used
tcl could end up going  the STk way
(unfortunately, STk is not publicized enough). Things like Kawa also increase
the feel of security regarding code reusability.
And some of those people will end up wanting to write full-fledged applications
"with parenthesis".
But it's probably more the fact that Java hype has quieted down, which make me
hopeful sooner than expected (although not in the same league, the fact that
surprises
_can_ happen, like the Linux/NT "battle for the enterprise" arriving so fast, is
reassuring).

> OK, that's too coy.  I recognize that *I* wanted him
> to have asked that, but he definitely didn't do so.

Well, a hot headline in a classic press trick to attract readers.
Also, the premise was that _I_ believe that it's philosophically broken,
and I wanted to find out whether the disgust for tcl I've seen many people
express had some basis on some generally accepted principles.
[For example, I can think the Eiffel syntax is `philosophically' broken, but
as `infix' syntaxes go, it is not. And it's a good language (I programmed
in it for 2 years  (years ago).. But just because I didn't have a CL compiler
then, and I wanted to do OO :-> )].






From m01ymu00 at cwcom.net  Tue Jun  1 13:24:56 1999
From: m01ymu00 at cwcom.net (Daniel Faulkner)
Date: Tue, 01 Jun 1999 18:24:56 +0100
Subject: Whats missing?
Message-ID: <37541767.AB2B250B@cwcom.net>

I was just messing about and found out that if I made some thing like
this it would not work but if I did it by typing at the python prompt

from urllib import urlopen
Data = urlopen("http://www.python.org/").read()
print Data

It would work but why is this?

from Tkinter import *
from urllib import urlopen

class HTMLBrowser(Frame):
    def Create_Widgets(self):
        # frames
        butFrame = Frame()
        butFrame.pack(side='top')
        # normal widgets
        self.QuitB = Button(butFrame, text='Quit', fg='red',
command=self.quit)
        self.UrlEnt = Entry(self, relief=SUNKEN)
        self.source_box = Text(self)
        self.startB = Button(butFrame, text='Start',
command=self.UrlGeting)
        # Pack it
        self.QuitB.pack(side='left')
        self.startB.pack(side='left')
        self.UrlEnt.pack(side='top', fill='x')
        self.source_box.pack(side='bottom')
    def lets_see_it(self):
        self.source_box.insert('0.0', self.Data)
    def UrlGeting(self):
        UrlCurr = self.UrlEnt.get()
        Data = urlopen('"' + UrlCurr + '"').read()
        self.source_box.insert('0.0', Data)
    def __init__(self, master=None):
        Frame.__init__(self, master)
        self.pack()
        self.Create_Widgets()

# Test
Dat = urlopen("http://www.python.org/")
# print Dat

Browser = HTMLBrowser()
Browser.mainloop()




From matomira at iname.com  Sat Jun 19 14:30:15 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Sat, 19 Jun 1999 20:30:15 +0200
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <rnbn1y2n5jk.fsf@nortelnetworks.com> <sFU93.1567$kS.233449@ptah.visi.com> <rnbpv2v5ejh.fsf@nortelnetworks.com> <qjQa3.2280$kS.310807@ptah.visi.com>
Message-ID: <376BE1B7.A2CCDB5E@iname.com>


David Thornley wrote:

> besides Tk that meets the requirements I've specified:  free, open source,
> on MacOS, Microsoft Windows and X Windows.

wxWindows.

DrScheme (MzScheme) uses that.







From delgado at Mathematik.Uni-Bielefeld.DE  Fri Jun 18 04:55:13 1999
From: delgado at Mathematik.Uni-Bielefeld.DE (Olaf Delgado)
Date: Fri, 18 Jun 1999 10:55:13 +0200
Subject: Newbie: Truth values (three-valued logic)
In-Reply-To: <000201beb946$b9c99e00$969e2299@tim>
References: <000201beb946$b9c99e00$969e2299@tim>
Message-ID: <Pine.OSF.4.05.9906181025150.6087-100000@pinax.Mathematik.Uni-Bielefeld.DE>

On Fri, 18 Jun 1999, Tim Peters wrote:

> [Olaf Delgado]
> > ...
> > It came to my mind, though, that redefining negation alone will not
> > help. Next time I'll want to redefine 'and' and 'or'. So, unless
> > python 2 implements three-valued logic <hint, hint :)>, I'll have
> > to come up with a routine for myself to handle these truth values
> > consistently.
> 
> Well, "and" and "or" do short-ciruit evaluation (i.e., they don't evaluate
> their right-hand operand at all if the value of their left-hand operand
> suffices to determine the result).  For that reason, they're more properly
> viewed as control structures than operators, and so unlikely to get "opened
> up" to user intervention.

I understand that. A new meaning for "and" and "or" would have to be built
into the language, but I don't think the merits would justify the efforts.
The simplest way I see would be to introduce a new type with a single
value "Maybe", similar to "None", which would have to be handled specially
by "if", "not", "and" and "or". It should be relatively simple to
implement this, as far as I can see, but of course it would slow down
things by having to consider all those special cases.

> But in the spirit of overloading "~" instead of "not", consider overloading
> "&" and "|" instead of "and" and "or".  Then you can do whatever you want.

Yes, I was thinking about something like this.


> If I were you I'd consider not overloading anything, and using ~ & |
> directly with a naming trick, like so:
> 
> f, t, m = 0, -1, 42
> name = {0: 'false', -1: 'true', 42: 'maybe', ~42: 'maybe'}
[...]
> This hacks around assigning two values to "maybe", just so that it's closed
> under negation.

That's neat! It's quite intuitive, also. But let's say not overloading
anything but "__nonzero__()", which will return "1" only if there's not
a single bit of doubt. :)

> Or use false=0, true=1, maybe=0.5; map __invert_(x)=1-x;
> __and__(x,y)=min(x,y); and __or__(x,y)=max(x,y).  Then you get a nice little
> MV logic with a whole bunch of possible "maybe" values.  Once you go beyond
> the first distinction (true/false), I don't know why you'd ever want to stop
> <wink>.

Or I could set __and__(x,y)=x*y and __or__(x,y)=x+y-x*y. Then I only would
have to invent a way to partially execute a block of code. :)

Thanks for all the suggestions!

Truely (or maybe falsely?) yours,
Olaf

-- 
   ////         Olaf Delgado Friedrichs, Uni Bielefeld, Mathematik
   Olaf         Postfach 100131                  (room:  V4-109)
   `='          D-33501 Bielefeld, Germany       (phone: +49 521 106-4765)
                           http://www.mathematik.uni-bielefeld.de/~delgado




From kodela at freenet.co.uk  Sat Jun 26 22:48:59 1999
From: kodela at freenet.co.uk (Sindh)
Date: Sun, 27 Jun 1999 02:48:59 +0000
Subject: Great Python : CGI Upload Help please.
Message-ID: <3775911B.A156BE7B@freenet.co.uk>

Dear all

I am in the process of making an ambitious web site[ not too big].
However python is helping a great deal. I am able to proceed as planned
which was unthinkable for my knowledge with perl / c etc.

Now the problem is I want to allow users to be able to upload file to my
site through a form. I checked out the upload.py but looks pretty
complex. Isn't there an easy way to allow this !

If not can somebody give me a sample html file and the python script
being used to give me an idea.

Please help
Thanks
sreekant





From olli at rhein-zeitung.de  Wed Jun 23 06:27:01 1999
From: olli at rhein-zeitung.de (olli at rhein-zeitung.de)
Date: 23 Jun 1999 10:27:01 GMT
Subject: [FYI] Future and Status of the RPM based Linux distribution
Message-ID: <7kqcpl$u4l$1@news.rhein-zeitung.DE>

Hi,

as I am receiving more and more emails when I will release a RedHat 6
compatible ready to use version of my distribution and when which module is
available for the distribution, I like to comment on this in the newsgroup,
cause all of the senders seem to read this group. And so I can reach some
other people too.

Well, you can certainly be sure, that there will be a release available soon
cause some kind users have supplied patches. But, of course there is a but, I
have to deal with the following flavours of RPM based Linux servers and
workstations now, so that I have to do some changes to be able to create all
packages from a common source.

    - RedHat Linux 5.2 (glibc 2.0)
    - RedHat Linux 6   (glibc 2.1)
    - Mandrake Linux 6 (glibc 2.1 and pentium optimized)
    - (maybe) SuSE 6.x (glibc 2.x)

As you can see, there is pretty much incompatibility available. ;-)) But I am
willing to deal with it. To further complicate this, two kind people have
supplied non x86 platforms, I have to integrate these versions too. Ok, so the
roadmap to this is:

On 1.7.1999 I plan to release a first RedHat 6 and Mandrake 6 release that is more
uptodate then the stuff available on the ftp server at the moment, which is a
RedHat 5.2 compatible version. This is somewhat an interim release until I
have a final version available. Just do cover the requests. ;-)) Of course, I
will provide optimized versions for all Mandrake architectures, cause Mandrake
is the distribution of choice at the moment (just a little personal opinion).
A first release of a packaged pyKDE module will be available at this time too.
(Only for RedHat 6 and Mandrake 6 at this time.)

But I also have some plans concerning the distribution and the non x86
versions. I like to setup some group writeable FTP directory on Starship
Python so that I can give the contributors a password to upload their packages
in a certain directory. That is just an idea, and I have to talk to the
Starship admins about that. Otherwise I will offer this access on my own
server and rsync the Starship dir from here.

Best regards,

    Oliver Andrich

-- 
Oliver Andrich, RZ-Online, Schlossstrasse Str. 42, D-56068 Koblenz
Telefon: +49-261-39210-40 Fax: -33 |       http://rhein-zeitung.de 
                                   |            http://andrich.net
GMC d- s++:++ a- C++ UL++++ P-- L+++ E--- W++ N+ w-- O- M-- 
V- PS++ PE Y+ PGP+ t 5+ X R+ tv+ b++ DI++ D- G e h- r z+



From tbryan at arlut.utexas.edu  Wed Jun  2 10:10:17 1999
From: tbryan at arlut.utexas.edu (Thomas Bryan)
Date: Wed, 02 Jun 1999 09:10:17 -0500
Subject: Problems with Install of LLNLDistribution11 (beginner)
References: <02k8ttz0ep.fsf@rtp.ericsson.se>
Message-ID: <37553B49.24624DA7@arlut.utexas.edu>

Kumar Balachandran wrote:
> 
> I am trying to install the Numeric and Graphics packages and I have
> the follwing problems
> 1. The numeric package was installed as
>         $PREFIX/lib/python1.5/site-packages/numeric
>    However, I am not able to import numeric using
>         from numeric import *

The package is called Numeric.

Python 1.5.1 (#1, Mar 21 1999, 22:49:36)  [GCC egcs-2.91.66 
19990314/Li on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> from Numeric import *
>>> array
<built-in function array>
>>> 


-------------------------------------------
Tom Bryan
Applied Research Laboratories
University of Texas at Austin



From mwh21 at cam.ac.uk  Wed Jun  2 16:57:06 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 02 Jun 1999 21:57:06 +0100
Subject: pickle vs .pyc
References: <7j43pn$55l$1@mlv.mit.edu>
Message-ID: <m3pv3eian1.fsf@atrus.jesus.cam.ac.uk>

Michael Vezie <mlv at pobox.com> writes:

> I need to be able to read a couple very complex (dictionary of arrays 
> of dictionaries, and array of dictionaries of array of dictionaries) 
> data structures into python.  To generate it by hand takes too long, 
> so I want to generate it once, and read it each time (the data doesn't 
> change).
> 
> The obvious choice is, of course pickle, or some flavor thereof.
> But can someone tell me why this wouldn't be faster:
> 
> In the code that does the "pickling", simply do:
> f = open("cache.py", "w")
> f.write("# cache file for fast,slow\n")
> f.write("fast = "+`fast`+'\n')
> f.write("slow = "+`slow'+'\n')
> f.close()
> import cache
> 
> Then, later, when I want the data, I just do:
> 
> from cache import fast,slow
> 
> and it's right there.  It's compiled, and seems really fast (loading a 
> 50k file in .12 seconds).  I just tried the same data using cPickle, and 
> it took 1.4 seconds.  It's also not as portable.  There is a space savings 
> with pickle, but it's only 5% (well, 56% if you count both the .py and 
> .pyc files), but that doesn't really matter to me.
> 
> Am I missing something here?  This sounds like an obvious, and fast, 
> way to do things.  True, the caching part may take longer.  But I 
> really don't care about that, since it's done only once, and in the 
> background.  
> 
> Michael

Hmm, you're relying on all the data you're storing having faithful
__repr__ methods. This certainly isn't universally true. I'd regard
this method as too fragile.

If you're only storing simple data (by which I mean simple types of
data, not that the data is simple) (and I think you must be for the
approach you're using to work) give the marshal module a whirl.

I think it will be substantially faster than your repr-based method
(cryptic hint: if it wasn't, the marshal module probably wouldn't
exist).

Eg:

import marshal

complex_data_structure = {'key1':['nested list'],9:"mixed types"}

marshal.dump(complex_data_structure,open('/tmp/foo','w'))

print marshal.load(open('/tmp/foo'))

HTH
Michael

Random aside: something fishy's going on when I try to try to marshal
*arrays* (as opposed to mere lists):

>>> import array,marshal
>>> marshal.loads(marshal.dumps(array.array('f',[0,1])))
'\000\000\000\000\000\000\200?'
>>> 

That shouldn't be happening should it? Surely that should be raising
an unmarshalable object exception? Oh well...





From tim_one at email.msn.com  Mon Jun 21 11:46:55 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 21 Jun 1999 11:46:55 -0400
Subject: Creating a new type with an extension module
In-Reply-To: <Ctp*vl52n@news.chiark.greenend.org.uk>
Message-ID: <000201bebbfd$4a310880$4fa22299@tim>

[Sion Arrowsmith]
> Is there any documentation anywhere on how to go about writing an
> extension which implements a new type? The extending/embedding
> tutorial give me, very quickly and painlessly, a module with
> functions, but I could really do with them being object methods.
> http://www.python.org/doc/current/api/newTypes.html is unhelpful
> in the extreme

Everything important is there -- it's just missing all the words <wink>.

> -- do I have to resort to reading source?

Better, copy the source for some existing type and edit it.  The
implementations of the builtin types are in the source distribution's
Objects directory; e.g., floatobject.c implements Python floats, and
listobject.c implements (no, I'm not telling <wink>).  Browse those to get a
feel for it, then look at xxobject.c.  That's a skeleton object
implementation you can start from.

the-source-is-documentation-too-&-even-better-you-can-compile-it-ly
    y'rs  - tim





From travel at mailcity.com  Sat Jun  5 10:11:30 1999
From: travel at mailcity.com (Travel Now!)
Date: Sat, 5 Jun 1999 14:11:30 GMT
Subject: Complimentary hotel package!
Message-ID: <419.436316.42509907travel@mailcity.com>

Your name has been received from this list.
http://www.maxnetpromo.com 

Free Vacation!!
***********************************************************
 Just for stopping by our site and reviewing some
material  We will send you and your companion for
a 2 night stay with luxury hotel accomodations-
absolutely no strings attached!  ($175 value)  You
will need to fill out the form completely with honest, 
accurate information. It is very short, only ten questions.  
You will then receive your travel voucher via email 
after the form is submitted and information verified.
They will ask for a $14.95 shipping and handling
fee for the vacation vouchers.
http://www.inficad.com/~dreams2000

The business material will let you look at how to 
take advantage of a Mult-Billion dollar travel industry-
working from your own home. Featuring a huge 
compensation plan, it is not uncommon to make 
six figures!  This is not MLM or a travel industry 
franchise.You will want to review this information.
Visit & take advantage of this free mini vacation
for a limited time only!
http://www.inficad.com/~dreams2000
*************************************************************




From sdm7g at virginia.edu  Tue Jun  1 15:06:38 1999
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Tue, 1 Jun 1999 19:06:38 GMT
Subject: Zope in the news
Message-ID: <Pine.A32.3.90.990601150224.5679B-100000@elvis.med.Virginia.EDU>

Linux Weekly News <http://lwn.net> has a link to this article
about open source software, which mentions Zope prominently: 

<http://www.amcity.com/washington/stories/1999/05/24/focus4.html>


| The Washington Business Journal has run an article about free software 
| with an emphasis on Zope. "'Zope is an excellent product, but its market 
| niche is dominated by big players with deep pockets,' said Hadar 
| Pedhazur, a principal in VIG who is also chairman of Digital Creations' 
| board of directors. 'We couldn't outspend those companies. Going open 
| source was a way of showing the world what the company can do in terms of 
| innovation and ideas.'" 


---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---

 "IA-64 looks just about like what you would expect from a PA-RISC
  and IA-32 train wreck with a little VLIW thrown in for spice."
 * Thomas J. Merritt  <tjm at spam.codegen.com> in <news:comp.arch> *




From nobody at newsfeeds.com  Mon Jun 21 17:30:44 1999
From: nobody at newsfeeds.com (Anonymous)
Date: Mon, 21 Jun 1999 17:30:44 -0400
Subject: Thanks for Python...
Message-ID: <376ead64@anonymous.newsfeeds.com>

Dearn Python Developers,

I must say how thankful I am for Python. It has saved me at work once so
far. I am working on a program to poll data from a server, but I don't have
the actual server. I was fortunate enough to get sample data files from the
server and fairly decent specification. So, I embarked on my mission to
write a mach-server that emulated the real server. I was successful at
writing one in Perl until today when I found out that the data I was sending
to it was not the right data. I was having a difficult time parsing the data
in Perl (not too familiar with the regular expressions & perl syntax seems
to be a bit buggy to me).

Well, this weekend I happened to decide to learn Python because I know C++,
Java, Visual Basic and a few other minor languages (or so called languages
like HTML & SQL). I needed to know a good scripting language for UNIX
Administration and simple utilities that I like to maintain my computers.
Python looked like an excellent candidate because of it's object oriented
features and it also appeard to be quite simple, which has turned out to be
the case.

This afternoon, at work (in about a 20 minute time period) I rewrote the
mach server that I had originally written in Perl and it works much nicer
and more reliably than the Perl version which I was using before. The server
works quite nicely and was surprisingly written in a few less lines than the
Perl server was written in. I hope that my boss is impressed.

Sincerely,

John Brainard



  --------== Posted Anonymously via Newsfeeds.Com ==-------
     Featuring the worlds only Anonymous Usenet Server
    -----------== http://www.newsfeeds.com ==----------



From akuchlin at mems-exchange.org  Thu Jun 24 10:58:28 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: Thu, 24 Jun 1999 10:58:28 -0400 (EDT)
Subject: Thanks for Python...
In-Reply-To: <7ktfn0$5q$1@aurwww.aur.alcatel.com>
References: <376ead64@anonymous.newsfeeds.com>
	<VrDQRddcbbs9-pn2-QdftOXYk6Hdd@Fortress>
	<37716901.43CB604@lockstar.com>
	<slrn7n2gkr.i3k.wtanksle@dolphin.openprojects.net>
	<7ktfn0$5q$1@aurwww.aur.alcatel.com>
Message-ID: <14194.18324.582362.94739@amarok.cnri.reston.va.us>

John Klassa writes:
>It doesn't run for me...  What's it supposed to do?  When I run it, I get:

See http://starship.python.net/crew/amk/writing/crypto-curiosa.html
for an explanation.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
Once you accept that the world is a giant computer run by white mice, all
other movies fade into insignificance.
    -- Mutsumi Takahashi, on _The Hitch-hiker's Guide to the Galaxy_




From suzuki611 at okisoft.co.jp  Sun Jun 13 19:03:54 1999
From: suzuki611 at okisoft.co.jp (Hisao Suzuki)
Date: 14 Jun 1999 08:03:54 +0900
Subject: fork()
References: <000c01beb3c1$a3f428a0$329e2299@tim>
	<199906110443.AAA02408@eric.cnri.reston.va.us>
	<Pine.SUN.3.95-heb-2.07.990611075346.4942G-100000@sunset.ma.huji.ac.il>
Message-ID: <u3d7yzn1np.fsf@ares.sys1.ngo.okisoft.co.jp>

In article <199906110443.AAA02408 at eric.cnri.reston.va.us>,
Guido van Rossum <guido at CNRI.Reston.VA.US> wrote:
> Could we get away with not calling (user) finalizers on objects in
> trash cycles at all?  Since the finalization order is problematic at
> best, this almost seems acceptable: in the formal semantics, objects
> that are part of cycles would live forever, and as an invisible
> optimization we recycle their memory anyway if we know they are
> unreachable.  (I've got the feeling that I've seen this rule before
> somewhere.)  We'd still get complaints "my __del__ doesn't get called"
> from some users, and the answer would still be "it's in a cycle -- use
> an explicit close"; but those who know what they are doing can be
> guaranteed that their memory cycles get recycled.  In other words it
> would be no worse than today and for some people it would be better.

In article <Pine.SUN.3.95-heb-2.07.990611075346.4942G-100000 at sunset.ma.huji.ac.il>,
Moshe Zadka <moshez at math.huji.ac.il> wrote:
| No, that can't be. If we did, it were in Scheme, and after you put it in,
| you'd be morally obliged to add full-fledged lambda, tail-recursion and 
| user visible continuations. 

Certainly the R5RS (1998) says:
   "All objects created in the course of a Scheme computation,
    including procedures and continuations, have unlimited
    extent.  No Scheme object is ever destroyed.  The reason
    that implementations of Scheme do not (usually!) run out of
    storage is that they are permitted to reclaim the storage
    occupied by an object if they can prove that the object
    cannot possibly matter to any future computation. ..."

However, (strangely enough :-) this idea is found also in C++
literature.  B. Stroustrup wrote in the C++ Programming Language
2nd Ed. (1991):
   "Garbage collection can be seen as a way of simulating an
    infinite memory in a limited memory.  With this in mind, we
    can answer a common question:  Should a garbage collector
    call the destructor for an object it recycles?  The answer
    is no, because an object placed on free store and never
    deleted is never destroyed. ..." (13.10.1 Garbage
    Collection)

and in the C++ 3rd Ed. (1997):
   "When an object is about to be recycled by a garbage
    collector, two alternatives exist:
      [1] Call the destructor (if any) for the object.
      [2] Treat the object as raw memory (don't call its
          destructor).
    By default, a garbage collector should choose option (2)
    because objects creatd using _new_ and never _delete_d are
    never destroyed.  Thus, one can see a garbage collector as a
    mechanism for simulating an infinite memory." (C.9.1.3
    Destructors)

Python's __del__ is conceptually equivalent to C++'s destructor.
(Otherwise we would not rely on self.close() at __del__ in
Lib/tempfile.py!)  The above phrase `(be) never deleted' for C++
can be safely translated as `(be) part of cycles' for Python.
Thus, Guido's idea is quite consistent and orthodox in this
regard.  It will be acceptable to most of the current working
Pythoneers, including me, even if they are not fans of the Lots
of Intolerable Stupid Parentheses ;-)

By the way, C++'s destructor and Python's __del__ have been
proved useful in practice, particularly in conjunction with the
`resource acquisition is initialization' technique.  On the
other hand, such a finalizer as found in Java is almost useless
and unreliable because of its unpredictability.  I have never
made use of finalize() in Java for years, and I don't know any
skilled Java programmer substantially making use of finalize().

Further the C++ 3rd Ed. says:
   "It is possible to design a garbage collector to invoke the
    destructors for objects that have been specifically
    `registered' with the collector.  However, there is no
    standard way of `registering' objects.  Note that it is
    always important to destroy objects in an order that ensures
    that the destructor for one object doesn't refer to an
    object that has been previously destroyed.  Such ordering
    isn't easily achieved by a garbage collector without help
    from the programmer."

All in all, we need a sort of control or programmability over
the behavior of our programs when a fixed, built-in mechanism of
the language cannot solely address the problem very well.
Suppose that Python recycles the memory of unreachable cycles of
objects without calling __del__.  What if cycles that contain
`registered' objects are saved from deallocation so that
clean-up or other arbitrary actions can be performed?

Such an action will break, say, an unreachable doubly-linked
tree into separated nodes.  Then the nodes will be destroyed by
a normal process including invocations of __del__.  To make this
occur,
 (a) the programmer must register the root of the tree with a
     function to break up the tree, which will be invoked later
     automatically by the garbage collector, or

 (b) the programmer must register the root of the tree, and at
     certain point of the program she/he must explicitly invoke
     a function which retrieves the trees unreachable except
     from the `registration' server and breaks them up.

 (In any case, unreachable trees that contain no `registered'
  objects will be recycled with no invocations of __del__.)

I am not sure which option is more useful in practice, but
either of them would be acceptable since they are entirely
optional and compatible with the current semantics of Python.
With no zapping in the formal semantics, they are conceptually
clean and their implementations will be straightforward (thus
they are also very Pythonic ;-)

As for a reference to option (b), see, say, R. Kent Dybvig's
ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
which I have mentioned before in this newsgroup, though it is
written for Scheme and actually the coming version of GNU guile
implements guardians, i.e. `registration' servers described in
it.  Note that the guardians may be implemented somehow in 100%
pure Python even if it doesn't have a garbage collector.  The
sys.getrefcount built-in function will be used then.

--===-----========------------- Sana esprimo naskas sanan ideon.
SUZUKI Hisao            suzuki611 at okisoft.co.jp, suzuki at acm.org.



From e_l_green at hotmail.com  Wed Jun 30 19:20:15 1999
From: e_l_green at hotmail.com (Eric Lee Green)
Date: Wed, 30 Jun 1999 16:20:15 -0700
Subject: theads & global namespaces
References: <377A5C88.32A1FAB6@notes.cba.ufl.edu>
Message-ID: <02373AD04B719342.C0FD90108DFF9F56.4A29D47664BD287B@lp.airnews.net>

Taylor Anderson wrote:
> I am trying to share globals between threads.  Is is possible for me to
> share globals between threads that call functions from different
> modules? 

I solved that problem by introducing a module "depository" that has my
"global" stuff that I want shared between threads. It is full of lines like
"hash_table={}" and so forth. Then in my modules I just

import depository

and down further, to access my "global" hash_table,

names=depository.hash_table.keys()   # get list of keys in hash table for
expire purposes

and voila. 

Yes, it works across threads. Just make sure to properly use semaphors if you
are doing operations on multiple structures (e.g., if I had two tables
"hash_table" and "hash_table_weights" and needed to keep them in sync). You
may wish to create some helper functions in your depository module to handle
the locking/setting/retrieving of such structures. 

-- 
Eric Lee Green                         eric at estinc.com
SysAdmin/Software Engineer             Visit our Web page:
Enhanced Software Technologies, Inc.   http://www.estinc.com/
(602) 470-1115 voice                   (602) 470-1116 fax



From bwinton at tor.dhs.org  Thu Jun 10 00:38:01 1999
From: bwinton at tor.dhs.org (Blake Winton)
Date: Thu, 10 Jun 1999 00:38:01 -0400 (EDT)
Subject: How do I match the last occurence of a character?
In-Reply-To: <375F3B81.78438F5C@netscape.com> from "Steve Halliburton" at Jun 09, 1999 09:13:53 PM
Message-ID: <199906100438.AAA16802@tor.dhs.org>

> I'm pretty sure that it does basically the same thing your code did...
> just under the covers.  I haven't used it on NT, but I know on unix it
> handles everything gracefully:

One of the nice things about Unices is that a whole lot of specs are
written to accomodate them.  Like using "/" for the separator in URLs.
I hypothesize that if the Mr. Berners-Lee was a Mac addict, the Weeb
would look a lot different today.

> It just splits off the last part and calls that the filename and the
> rest is the basename.

The Macintosh version of os.split (which is accessed as os.split, and
can be grabbed from macpath.py), looks like:

# Split a pathname in two parts: the directory leading up to the final
# bit, and the basename (the filename, without colons, in that
# directory).  The result (s, t) is such that join(s, t) yields the
# original argument.
def split(s):
        if ':' not in s: return '', s
        colon = 0
        for i in range(len(s)):
                if s[i] == ':': colon = i+1
        path, file = s[:colon-1], s[colon:]
        if path and not ':' in path:
                path = path + ':'
        return path, file

And when we use it to split, we get results like

>>> name = "http://www.foo.com/~bwinton/test.html"
>>> name2 = "/home/bwinton/test.html"
>>> split( name )
('http:', '//www.foo.com/~bwinton/test.html')
>>> split( name2 )
('', '/home/bwinton/test.html')

Which I think you will agree is not the desired behaviour.
(ntpath.split seems to do the right thing, though.)

Later,
Blake.




From greg.ewing at compaq.com  Thu Jun 24 20:50:06 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 25 Jun 1999 12:50:06 +1200
Subject: Windows - Drag & drop on Python script?
References: <37717600.D076BF10@compaq.com> <3771908b.1913288@news.gte.net>
Message-ID: <3772D23E.10C89EC@compaq.com>

Well, I don't know what went wrong yesterday, but
I tried the same thing again today, and it worked
fine. Drag a file onto a .bat and it comes in as
%1.

Another question - what is the recommended way of
getting the effect of a unix #!? I came up with:

python -x %0 %1
exit
<python code here>

which only works by the serendipitous fact that
there is an 'exit' defined in the python builtins.
Without the exit, after the python finishes, the
shell goes on to try and interpret it as msdos
commands!

Is there a more elegant way to do this?

Another other question - what is the "Batch File"
box in the Properties dialog of a .bat file for?
If you put the name of another .bat file there,
it gets executed before the main file. But what
is its purpose? Just curious.

Greg



From sams at quinta.com  Fri Jun 18 13:27:45 1999
From: sams at quinta.com (Sam Schulenburg)
Date: Fri, 18 Jun 1999 17:27:45 GMT
Subject: Python source code documentation Repost?
References: <7kdv0l$n0g$1@nnrp1.deja.com>
Message-ID: <7kdvi3$n78$1@nnrp1.deja.com>

In article <7kdv0l$n0g$1 at nnrp1.deja.com>,
  Sam Schulenburg <sams at quinta.com> wrote:
> I have been using a freeware product called Autoduck to document my
> Python source code.
>
> =============== Example source ====================================
> # ######################################
> # DATE:         June 17,1999
> # PROGRAMMER:   Samuel W. Schulenburg
> # This function will print out the command time statistics from<nl>
> # getcommandtime(tid) function<nl>
> #
> #//@pyfunc void | printcommandtime | This function will take a <nl>
> #//                          tid and print out the following<nl>
> #//                          commandtime structure statictics:<nl>
> #//                          1) Total command time counts<nl>
> #//                          2) Start time for last command<nl>
> #//                          3) Minimum time for all commands<nl>
> #//                          4) Maximum time for all commands<nl>
> #//                          5) End time for last command<nl>
> #//                          6) Time for last command<nl>
> #//                          7) Total time for all commands<nl>
> #//                          8) Average time for all commands<nl>
> #//@parm   integer  | Tid | Target id to use
> #//@rdesc  void
> #
> #####################################################################
> def printcommandtime(tid):
>         s=getcommandtime(tid)
>         print "Total command time counts     = ",s[0]
>         print "Start time for last command   = ",s[1]
>         print "Minimum time for all commands = ",s[2]
>         print "Maximum time for all commands = ",s[3]
>         print "End time for last command     = ",s[4]
>         print "Time for last command         = ",s[5]
>         print "Total time for all commands   = ",s[6]
>         print "Average time for all commands = ",s[7]
> ================= End Example =====================================
> This product will then generate a windows help file based upon the
> information provided above.
>
> I would prefer to have a Python based solution that would generate
HTML
> files. Can anyone point me towards any information that may start me
> down this path, If a solution does not exist, then I may have to write
> my own Python parser to handle this.
>
> Sam Schulenburg
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.

The previous post had formatting problems, If this post does not
correct it I give up on deja news.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From Klaus.Schilling at home.ivm.de  Tue Jun 15 16:00:20 1999
From: Klaus.Schilling at home.ivm.de (Klaus Schilling)
Date: 15 Jun 1999 22:00:20 +0200
Subject: OO (was: Why is tcl broken?)
References: <000801beb745$02768220$979e2299@tim>
Message-ID: <871zfdp73f.fsf@home.ivm.de>

"Tim Peters" <tim_one at email.msn.com> writes:

> [Fernando Mato Mira]
> > Quiz: What do Scheme, Common Lisp, HPF and Java have in common?
> 
> Guy Steele's loving ministrations -- but even he gave up on Scheme <wink>.

Scheme doesn't need G.Steele. Steele just rediscovered Scheme.

Klaus Schilling



From cgw at fnal.gov  Mon Jun 28 16:07:16 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Mon, 28 Jun 1999 20:07:16 GMT
Subject: [PSA MEMBERS] Stuck trying to build some Python predecessor software
In-Reply-To: <199906281951.PAA07090@cm-24-29-94-19.nycap.rr.com>
References: <199906281951.PAA07090@cm-24-29-94-19.nycap.rr.com>
Message-ID: <14199.54772.884928.89818@buffalo.fnal.gov>

You shouldn't need to install binutils.  Just make sure /usr/ccs/bin
is in your path, you should be able to pick up a working 'ar' and
'make' there.




From jhspies at futurenet.co.za  Wed Jun 30 04:14:49 1999
From: jhspies at futurenet.co.za (Johann Spies)
Date: Wed, 30 Jun 1999 08:14:49 GMT
Subject: Please help me to understand classes
In-Reply-To: <11A17AA2B9EAD111BCEA00A0C9B4179301E10F81@molach.origin.ea.com>
References: <11A17AA2B9EAD111BCEA00A0C9B4179301E10F81@molach.origin.ea.com>
Message-ID: <Pine.LNX.3.96.990630101325.912D-100000@Johann>

On Tue, 29 Jun 1999, Stidolph, David wrote:

> Finally, a simple question I can answer!
> 
> First of all, class methods like "druk" need to have a parameter declared in
> them called self (or whatever you want to call it).  Secondly you need to
> call the method with open/close parens:
> 
> class A:
>   def druk(self):
>     print 'Jy is in Class A'
> 
> class B:
>   def druk(self):
>     print 'Jy is in Class B'
> 
> a = A()
> c = B()
> c.druk()
> Jy is in Class B
> a.druk()
> Jy is in Class A
> 
> I hope this helps,

Thanks David.  It helped.  After your message I went back and read the
tutorial again and saw what you mean.

Johann



 --------------------------------------------------------------------------
| Johann Spies                                 Windsorlaan 19              |
| jhspies at futurenet.co.za                3201 Pietermaritzburg             |
| Tel/Faks Nr. +27 331-46-1310		       Suid-Afrika (South Africa)  |
 --------------------------------------------------------------------------

     "Because he himself suffered when he was tempted, he
      is able to help those who are being tempted."       
                                        Hebrews 2:18 




From cyrius at usc-interactive.com  Fri Jun 11 15:07:49 1999
From: cyrius at usc-interactive.com (CyRiuS)
Date: Fri, 11 Jun 1999 15:07:49 -0400
Subject: Embedding Question from a Newbie
Message-ID: <00b801beb43d$b3efe240$a515bfd1@EVILDEAD>

i have a question. i am new to python and am trying to use it as a scripting language for a game i am working on. I have the interpreter embedded and running scripts but i have a few questions.

1) can i return values from the python script to the C/C++ program?
for example say at the end of my script, it returned 'character' and character was equal to say 'hi'
now is there anyway my C/C++ program could find out that the script just returned the value 'hi'?

2) is there anyway I can pass values from the C/C++ program into the python script?

and

3) say i had a python script that had several functions (defs). is there anyway i could call just one of these defs from the C/C++ program without calling the entire script?

Thanks for your help
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/19990611/2ae3929b/attachment.html>

From tismer at appliedbiometrics.com  Fri Jun 11 11:10:34 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Fri, 11 Jun 1999 15:10:34 GMT
Subject: fork()
References: <000601beb1c7$682c6220$559e2299@tim>
		<7jjghd$9il$1@cronkite.cc.uga.edu>
		<375D5826.23197692@inrialpes.fr>
		<375D6C60.FF9998DE@appliedbiometrics.com> <5laeu8zmza.fsf@eric.cnri.reston.va.us>
Message-ID: <376126EA.4C5FC6E9@appliedbiometrics.com>


Guido van Rossum wrote:
> 
> Christian Tismer <tismer at appliedbiometrics.com> writes:
> 
> > If we have knowlegde of the internals of every Python structure,
> > and if there were no hidden internal references to Python
> > objects, then we could quite easily build a non-pessimistic
> > GC, without the need to be hardware specific, scan the stack
> > and so on.
> > You know that I have an (early) Python with no stack.
> > Would this help?
> 
> I don't see how: the roots you are looking for are all in the frame
> objects anyway.  (There are a few variables in the C stack of
> eval_code() but they point to objects that are also in the frame's list
> of local variables or on the eval stack in the frame.)

Correct. You are right. My version has some slight advantages
since it can release some objects earlier,
but that's no point here.

The problem which I was thinking about was how to detect
that a cycle can be safely removed. eval_code() is well-behaved,
but it cannot tell how it was called, and how many other
functions from extensions are sitting on the stack which might
have alive references.
My interpre^H^H^H^H^H^H^H^Hevaluator knows a bit more
about the objects which are created during its lifettime,
since its lifetime is so much longer.

I don't see it yet, but I guess that there comes some
helpful extra information for free.
By now I'm just to dumb too dumb to find it out.

> But here's another dumb idea based on the same premise (since this
> thread won't die anyway, I might as well join in :-).

...

> - Almost all cycles involve a dictionary, in practice (probably also
> an instance, but those cycles involving an instance always also
> involve a dictionary, since all references from instances go through
> dictionaries).  Why does this matter?  Because now I only need to put
> a mark bit in dioctionary objects, and I can maintain a list of all
> dictionaries instead of a list of all objects, and apply
> mark-and-sweep to it, without occurring overhead for all objects.

Considering a "typed" allocator:
Would be equivalent to know all heap pages which hold
dictionaries? Or is the mark bit needed alsewhere?

> (You can create a cycle using only lists, but I bet that happens only
> for demonstration purposes, as in ``a = []; a.append(a)''.  So I can
> live with not freeing such cycles.)

Would come for free, but well.

<snipped the best, but reserved for later thoughts/>

> Why this compromise instead of "true GC" like Java?  Because one of
> Python's main reasons of existence is its good integration with
> external C code -- in a glue language, you can't set the rules for
> management of all memory!

Very good. Like conservative for the outside (by not touching it)
and collaborative for the inside.

> May this thread spurn some code,

:-) chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     not coding at all is proven to be the best time saver



From akuchlin at mems-exchange.org  Thu Jun  3 09:38:58 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: Thu, 3 Jun 1999 09:38:58 -0400 (EDT)
Subject: Checking for Python modules in autoconf
Message-ID: <199906031338.JAA28564@amarok.cnri.reston.va.us>

I'm working on packaging some software for distribution, and would
like the configure script (generated using GNU autoconf) to check
whether various Python modules are installed.  Does anyone have an
autoconf macro for doing this?

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
Thanks. The sooner I get discouraged and quit, the more time I'll save overall.
    -- Frank Sergeant, 28 Mar 1999




From jblake at eskimo.com  Sat Jun 19 18:46:47 1999
From: jblake at eskimo.com (Jonathon)
Date: 19 Jun 1999 22:46:47 GMT
Subject: NameError Problem
References: <7kels4$9tv$1@eskinews.eskimo.com> <14186.57835.612462.716239@amarok.cnri.reston.va.us>
Message-ID: <7kh6kn$6cr$1@eskinews.eskimo.com>

Andrew M. Kuchling (akuchlin at mems-exchange.org) wrote:
: I'll wager that the_file_is winds up with some sort of unexpected
	
	It appears top have the correct value.

: value, and causes your variable to never be assigned a value,. because 

	But the variables I am initializing in the loop don't get
	initialized.  It doesn't look like the file goes into the loop.

	For those who would like to see the script, I have it
	at http://www.stamp-coin.com/python/problem.html.
	That page also includes a sample of the database that it
	is looking at.

	Thanks.

	jonathon



	
-- 
	I'm still looking for a good book on
		3:	The Recent Unpleasantness
		1:	The War Of Northern Aggression.
		2:	The War of Southern Rebellion.



From mnot at pobox.com  Tue Jun 22 18:39:07 1999
From: mnot at pobox.com (Mark Nottingham)
Date: Tue, 22 Jun 1999 22:39:07 GMT
Subject: htmllib: CR in CDATA
References: <00d201bebc5a$adb129a0$0301a8c0@mnot.net> <012601bebc91$f3ded760$0301a8c0@mnot.net> <014401bebc9a$1963a9e0$0301a8c0@mnot.net> <000b01bebca0$8a022c20$f29b12c2@pythonware.com> <017501bebcad$830a3180$0301a8c0@mnot.net> <003701bebcb4$a6d33fb0$f29b12c2@pythonware.com>
Message-ID: <002901bebd00$0a35cbe0$0301a8c0@mnot.net>

> imho, it's pretty reasonable for an SGML parser to
> behave like an XML parser: split the document up
> into pieces, but pass them all to the application as
> untouched as possible. if you wish to implement
> additional behaviour, do that on the application
> level. otherwise, you'll end up in a situation where
> some users cannot use the standard library...

I see your point. I suppose in a perfect world, the standards would be
coherent enough that these kinds of choices wouldn't have to be made; this
shouldn't have to be additional behaviour.

Cheers,




From rfreedm1 at nospam.rochester.rr.com  Tue Jun 22 22:33:56 1999
From: rfreedm1 at nospam.rochester.rr.com (Benjamin Schollnick)
Date: Wed, 23 Jun 1999 02:33:56 GMT
Subject: help?  strptime ....
Message-ID: <rNHvjEdhm5Pp-pn2-dCsXnyPPztJy@d185d18f8.rochester.rr.com>

Folks,

	I'm trying to use strptime (from the time module) and can't....

Could someone please post a example, so that I can try to figure out 
what
I'm doing wrong?

			- Benjamin



From sk at fyemf.fys.ntnu.no  Sun Jun 20 16:47:58 1999
From: sk at fyemf.fys.ntnu.no (Steinar Knutsen)
Date: 20 Jun 1999 20:47:58 -0000
Subject: Alg Problem (new to Python)
References: <7k38gf$khr$1@holly.prod.itd.earthlink.net>
Message-ID: <7kjk1u$8b6$1@fyemf.fys.ntnu.no>

In article <7k38gf$khr$1 at holly.prod.itd.earthlink.net>, Steve Hunter
<Steve at HOSYS.com> wrote:

> The out put of a program is below....the problem is the tasks are
>being added to every project vs. to the specific project....it sees
>like an alg program but I can't see it....brand new to Python...have
>mercy.
>
> [...]
>
>the program below is in on .py file
>
>
>import sys
>
>projectDict = {} #MAPS PROJECT id string to PROJECT  object
>taskDict = {}  #MAPS TASK id string to TASK object
>
>class Project:
> _id = ''
> _name = ''
> _descr = ''
> _tasks = {}    #list of Tasks I contain

The dictionary Project._tasks is a mutable object, so when your method
for adding a task adds a new task to the dictionary, it adds the task to
the class attribute (Instead of making a new object as with strings and
integers, for instance.). Putting something like
   self.tasks = {}
into the __init__-method of the class Project will solve your problem.
Don't worry, the behaviour of mutable versus immutable objects is one of
the few things that I think often bites people new to Python.
-- 
Steinar



From mal at lemburg.com  Tue Jun  8 15:27:37 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Tue, 08 Jun 1999 21:27:37 +0200
Subject: mxCGIPython project update
Message-ID: <375D6EA9.128FD735@lemburg.com>

Hi everybody,

I just give you an update of the current list of available one-file
easy-to-install Python 1.5.2 interpreters. There are many more
versions and even one for Windows NT4 thanks to Florent Heyworth (*).

	cgipython-1.5.2-AIX-2-000771064C00-.gz
	cgipython-1.5.2-FreeBSD-2.2.8-RELEASE-i386-i386.gz
	cgipython-1.5.2-FreeBSD-3.0-RELEASE-i386-i386.gz
	cgipython-1.5.2-FreeBSD-3.1-19990426-STABLE-i386-i386.gz
	cgipython-1.5.2-HP-UX-A.09.03-9000_715-unknown.gz
	cgipython-1.5.2-HP-UX-B.10.20-9000_782-unknown.gz
	cgipython-1.5.2-IRIX-5.3-IP19-mips.gz
	cgipython-1.5.2-IRIX-6.5-IP32-mips.gz
	cgipython-1.5.2-IRIX64-6.5-IP30-mips.gz
	cgipython-1.5.2-Linux-2.0.34-i586-RedHat-5.1.gz
	cgipython-1.5.2-Linux-2.0.35-i586-unknown.gz
	cgipython-1.5.2-Linux-2.0.35-i686-unknown.gz
	cgipython-1.5.2-Linux-2.0.36-i586-Debian-2.1.gz
	cgipython-1.5.2-Linux-2.0.36-i586-unknown.gz
	cgipython-1.5.2-Linux-2.2.7-i586-SuSE60.gz
	cgipython-1.5.2-NetBSD-1.3.2-alpha.gz
	cgipython-1.5.2-NetBSD-1.3.3-i386.gz
	cgipython-1.5.2-OpenBSD-2.4-i386-Intel_Pentium_II.gz
	cgipython-1.5.2-SunOS-4.1.4-sun4m.gz
	cgipython-1.5.2-SunOS-5.5.1-sun4u-sparc-alternate.gz
	cgipython-1.5.2-SunOS-5.5.1-sun4u-sparc.gz
	cgipython-1.5.2-Winnt-4.0.zip

The current place to look for contributed binaries is 

     ftp://starship.skyport.net/pub/cgipython/ 

This will probably change to starship.python.net in the near
future. 

The above binaries contain the Python 1.5.2 interpreter
plus the whole standard lib (except for Tkinter) in one
single file: this reduces installation time/problems and
startup time as well.

More information about the project and how to contribute
can be found here:

        http://starship.skyport.net/~lemburg/mxCGIPython.html

(*) I'm curious if the cgipython EXE works out of the box for
    MS IIS. Would be really cool, IMHO, and yet another argument
    for using Python for CGI scripting.

Cheers and thanks to everyone who contributed,
-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   206 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From jbauer at rubic.com  Thu Jun 17 17:06:40 1999
From: jbauer at rubic.com (Jeff Bauer)
Date: Thu, 17 Jun 1999 21:06:40 GMT
Subject: MySQL module compiling c in win32
References: <7kbf8l$qtn$1@nnrp1.deja.com>
Message-ID: <37696360.12C1E930@rubic.com>

tom at hrmanager.com wrote:
> I have been trying to compile the MYSQL module for python 
> with MSVC++ 5.0.

I have a precompiled client .dll on the starship.

  ftp://starship.python.net/pub/crew/jbauer/

Look for MySQL.pyd and the accompanying libmySQL.dll
file.

Best regards,

Jeff Bauer
Rubicon, Inc.



From hniksic at srce.hr  Sat Jun  5 12:43:27 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 05 Jun 1999 18:43:27 +0200
Subject: Deleting characters from a string
References: <87n1ye3g4v.fsf@pc-hrvoje.srce.hr> <aahzFCv4EL.7HC@netcom.com>
Message-ID: <87wvxi1tu8.fsf@pc-hrvoje.srce.hr>

aahz at netcom.com (Aahz Maruch) writes:

> In article <87n1ye3g4v.fsf at pc-hrvoje.srce.hr>,
> Hrvoje Niksic  <hniksic at srce.hr> wrote:
> >
> >What I would really like to be able to do is write:
> >
> >  host = string.translate(host, None, ' "')
> >
> >Am I missing a more elegant solution?
> 
> I dunno, but I'd tend to use re.sub().

I should have made it clear, sorry.  I don't want want to import re
because I don't need it anywhere else in my program.  It's even less
elegant and probably much slower than string.translate() anyway.



From arcege at shore.net  Fri Jun 25 07:53:43 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Fri, 25 Jun 1999 11:53:43 GMT
Subject: Tkinter: radiobutton
References: <3771B52E.80F3FA46@pixar.com> <qjsc3.77$Of3.12274@news.shore.net> <3772EEE3.2DCCE7C0@pixar.com>
Message-ID: <b5Kc3.209$Of3.47843@news.shore.net>

kiki <kiki at pixar.com> wrote:
: "Michael P. Reilly" wrote:

:>
:> No, it's not an obvious question at all.  Unfortunately, you need to
:> track the index when you create the menu.  By default, the first call
:> starts at 1, because the "tear-off" is 0.  If the menu is created with
:> the -tearoff=0 option, then the index starts at 0
:>

: Bleh.  Oh well.  Kind of a shame, since I'd rather just have the color value
: put into the color variable, not an index with my own list I have to maintain
: myself.

: Well, cool, at least I have a solution.

: Thanx so much for your help!

Kiki,

I just uploaded a module I made for another project (and never remembered
to give to Guido).

The URL is ftp://starship.python.net/pub/crew/arcege/modules/MenuEntry.py.

In that module, you get five usable classes (and one super class):
MenuCascade, MenuCommand, MenuCheckbutton, MenuRadiobutton, MenuSeparator.

You get the following methods:
  item.activate()
  item.cget(option)
  item.configure([cnf={...},] [key=value,]...)
  item.config  - same as configure
  item.invoke()
  item.yposition()

These map to methods in the Menu class, but the MenuEntry object knows
which index it is.

  -Arcege




From stephan at pcrm.win.tue.nl  Tue Jun  8 09:00:32 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 08 Jun 1999 15:00:32 +0200
Subject: fork()
References: <375BE64A.EB997F24@icrf.icnet.uk> <87so840zcc.fsf@pc-hrvoje.srce.hr> <375BF721.4B1864F2@icrf.icnet.uk> <m31zfn99td.fsf@greebo.nodomain.de> <87hfojzudh.fsf@pc-hrvoje.srce.hr> <375C26D9.327BA92@icrf.icnet.uk>
Message-ID: <m3aeuac0en.fsf@pcrm.win.tue.nl>

Arne Mueller <a.mueller at icrf.icnet.uk> writes:

> Hrvoje Niksic wrote:
> > 
> > Bernhard Herzog <herzog at online.de> writes:
> > 
> > > I fear that in reality the memory is copied after all. If they use
> > > the dictionary in any way they are modifying refcounts of at least
> > > some of the objects in the dictionary. From the OS's point of view
> > > this counts as modifying the memory.
> > 
> > Eek.  You're right.  Refcounting truly sucks.  :-(
> 
> Hm, that means as long as I realy don't look at the data of the parent
> the data is shared between parent and child. Thats funny - it's a
> useless feature. As soon as a child reference that dictionary  it gets
> it's own copy. 

It's still not completely useless; you don't have to explicitely say
which memory pages will be copied. 

> What about shared memory, is it possible to put objects into a shared
> memory segment, most advanced language can do that?

Don't know if anyone wrote a Python wrapper around the POSIX shared memory
stuff. However, you probably don't want/need to do that. The reason is
that there's basically no (portable) way to get malloc() to allocate
new objects in the shared memory, so you can't create Python objects
in the shared memory. So you can probably only access the shared 
memory as a special array a la the "array" module.

In which case this buys you little, since if you stored the data in
an "array" module array, the data would be shared automagically.
So that's the approach I recommend:

1. Build your shared data, painfully, in one or more "raw" arrays as provided
   by the "array" module.
2. Don't change your arrays after that.
3. Do the fork and access the arrays from all the child processes.

This should work. (I think)

Greetings,

Stephan



From phd at emerald.netskate.ru  Thu Jun 10 04:41:06 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Thu, 10 Jun 1999 12:41:06 +0400 (MSD)
Subject: os.mkdir()
In-Reply-To: <375E6DAA.DD279E10@phoenix-edv.netzservice.de>
Message-ID: <Pine.LNX.4.04.9906101240310.6646-100000@emerald.netskate.ru>

On Wed, 9 Jun 1999, Holger Jannsen wrote:
> os.makeDirs(thePath, aMode)
> AttributeError: makeDirs

os.makedirs(thePath, aMode)
       ^

   Lowercase...

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From julian at brightstar.swin.de  Tue Jun 15 11:49:21 1999
From: julian at brightstar.swin.de (Julian Einwag)
Date: 15 Jun 1999 17:49:21 +0200
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <rnbn1y2n5jk.fsf@nortelnetworks.com>
Message-ID: <m1909lpipq.fsf@brightstar.swin.de>

Chris Ebenezer <chriseb at nortelnetworks.com> writes:

> Why use TK at all ?  Once you are using scheme anyway there are nicer
> widget toolkits to use (from a point of view of more "normal" looking
> toolkits that don't suffer from color allocation problems),  there are
> bindings for guile and gtk and guile and Qt. 

Do you know where you can get the Qt bindings? Im quite interested in 
it.



From jfisher at are.berkeley.edu  Mon Jun 28 17:58:32 1999
From: jfisher at are.berkeley.edu (John Fisher)
Date: Mon, 28 Jun 1999 21:58:32 GMT
Subject: load_dynamic problem
References: <7l0j58$cbh$1@nnrp1.deja.com> <qIWc3.329$Of3.76550@news.shore.net>
Message-ID: <7l8r62$tpa$1@nnrp1.deja.com>

In article <qIWc3.329$Of3.76550 at news.shore.net>,
  "Michael P. Reilly" <arcege at shore.net> wrote:

> John,
>
> It sounds like "setdef" is in some library that you are not compiling
> with.  Or "setdef" was typed incorrectly in your code: resolving
> symbols in shared objects is performed at runtime.
[...]
>
> Good luck,
>   -Arcege

Thanks, Arcege.  I was not aware that I needed to compile with the
libraries mentioned explicitly.  Silly me though I could just #include
and have gcc look at the dirs.  Anyway, I appreciate the pointer.

John


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From ivnowa at hvision.nl  Sat Jun 19 15:26:34 1999
From: ivnowa at hvision.nl (Hans Nowak)
Date: Sat, 19 Jun 1999 19:26:34 GMT
Subject: Curses for MSDOS
In-Reply-To: <376BD33B.5A6F54BB@erols.com>
References: <376BD33B.5A6F54BB@erols.com>
Message-ID: <199906191925.VAA20167@axil.hvision.nl>


On 19 Jun 99, Swartz wrote:

> Is there anything like curses for MS-DOS?  Thanx.

Python-DX has curses built-in. I got it from a public domain library 
called pdcurses. I've never really tried it though, so you're on your 
own here.

Check out Python-DX (only compiled binaries...) at

http://www.cuci.nl/~hnowak/html/python-dx.html


--Hans Nowak (ivnowa at hvision.nl)
Homepage: http://fly.to/zephyrfalcon



From tlng at phileo.com.my  Mon Jun 21 04:43:44 1999
From: tlng at phileo.com.my (tlng at phileo.com.my)
Date: Mon, 21 Jun 1999 08:43:44 GMT
Subject: tkMessageBox returns 'can't invoke "frame" command:' error
Message-ID: <7kktvv$550$1@nnrp1.deja.com>

import Tkinter
import tkMessageBox

def askyesno():
	if tkMessageBox.askyesno("Testing Askyesno",
				 "Click on something!"):
		print 'You clicked on Yes!';

rootWindow = Tkinter.Tk()

button1 = Tkinter.Button(rootWindow,
			 text="Askyesno",
			 command=askyesno)
button1.pack()

button2 = Tkinter.Button(rootWindow,
			 text="QUIT",
                         command=rootWindow.quit)
button2.pack()

rootWindow.mainloop()



I wrote a script similar to the one shown above. Here are my questions:

1) If I import my script above for the first time, my script runs fine.
However, after I quit my script and try to reload my script at the
Python interpreter by typing reload(myscriptname), I get the error
message: "TclError: Can't invoke "frame" command:". Actually, it is
quite a long-winded error message, I'm just giving the last line.
What's going on and how do I fix this?

2) I even tried including a frame in def askyesno(), like this:

	frame1 = Tkinter.Frame(rootWindow)
	if tkMessageBox.askyesno("Testing Askyesno",
				 "Click on something!"):
		print 'You clicked on Yes!';
	frame1.pack()

However, I still get the same problem. I'm using Tkinter for the first
time and this "Can't invoke "frame" command" error message has been
frustrating me tremendously whenever I try to use one of tkMessageBox's
standard dialogs. I must be doing something wrong, and I would
appreciate it if someone could give some pointers on how to use
tkMessageBox.

3) Also, how do you quit a Tkinter script? Whenever I try to close a Tk
window, I can't get back to the Python interpreter ">>>" prompt.
Instead, my Python shell hangs and I have to do a CTRL-ALT-DEL and
end-task. Surely, there are better ways of closing a Tk window.

I'm running Windows 95 on a Pentium class computer. My Python
interpreter is version 1.5.2 and my Tcl/Tk interpreter is the same one
that is installed automatically by the Python 1.5.2 installer.

Thanks in advance.

Raymond Tong Leng Ng




Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From moshez at math.huji.ac.il  Tue Jun  8 22:40:20 1999
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 9 Jun 1999 05:40:20 +0300
Subject: In what order does Python read dictionary keys?
In-Reply-To: <7jk8s8$3ma$1@nnrp1.deja.com>
References: <7jjnt5$snk$1@nnrp1.deja.com> <375D8077.EA539C6D@pop.vet.uu.nl> <7jk8s8$3ma$1@nnrp1.deja.com>
Message-ID: <Pine.SUN.3.95-heb-2.07.990609053849.27545A-100000@sunset.ma.huji.ac.il>

On Tue, 8 Jun 1999 tcpeter at my-deja.com wrote:

<about the order of keys()>
> 
> OK.  So is there a way to ensure a specific order?  In this instance
> I've given it's not particularly critical, but there are times where it
> might be.  Any ideas?

Subclass UserDict.UserDict.
--
Moshe Zadka <mzadka at geocities.com>. 
QOTD: My own exit is more likely to be horizontal then perpendicular.





From proteus at cloud9.net  Tue Jun 29 11:56:35 1999
From: proteus at cloud9.net (Michael Muller)
Date: Tue, 29 Jun 1999 11:56:35 -0400
Subject: high-level "read" function?
References: <3777D967.484F0B6D@lbl.gov>
Message-ID: <3778ECB3.573B19A7@cloud9.net>

Christine Celata wrote:
> 
> I am just starting to use Python.  It is foreign to me-- I'm a physicist
> used to Fortran, not C.  I have succeeded in getting Python to read
> program user input, but as far as I can see, in order to get it to read
> numbers separated by commas from the terminal and store them as floating
> point variables, I have to use a sequence of low-level call to:  read
> the input as one character string, then strip off the carriage return,
> then split the string into several strings using the commas as markers
> for the end of the string, then convert the separated strings into
> floating point numbers.  In the case where I am reading in strings
> separated by commas the last step is omitted, but I have to add a step
> to eliminate leading whitespace.  I am about to try to write a function
> for myself that does all this, and a function to read integers, and a
> function to read lists of strings, but--  Since lots of people must be
> inputting numbers, isn't this stuff already written and lying around
> somewhere?  I can't find it in the documentation for modules.   Thanks.

You want to look at the string and regular expression modules.  This
should pretty much do what you want:

 >>> import re
 >>> line = raw_input()
 >>> map(float, re.split('\s*,\s*', line))

=============================================================================
michaelMuller = proteus at cloud9.net | http://www.cloud9.net/~proteus
-----------------------------------------------------------------------------
Government, like dress, is the badge of lost innocence; the palaces of
kings
are built on the ruins of the bowers of paradise. - Thomas Paine
=============================================================================



From news at dorb.com  Sat Jun 26 16:52:44 1999
From: news at dorb.com (Darrell)
Date: Sat, 26 Jun 1999 16:52:44 -0400
Subject: Python for large projects?
References: <3774b5d1.690922@news.force9.net>
Message-ID: <T6bd3.2883$U6.12014@newsr2.twcny.rr.com>

Python is great fun, with positive qualities that go on and on.

Large projects come with special problems of their own.
For C++ they have books about this subject.
Large-Scale C++ Software Design by John Lakos
Effective C++ by Scott Meyers.

I can recommend Python for large projects but watch out for the following.
IMHO.

1. Memory management. Search for keywords GC, cycles, reference counting in
this news group.
One problem we had is reading a very large file in and passing the buffer
down to be ripped apart. We got stuck holding the original buffer for this
file until the program exited. At least I think we did, it's hard to know
sometimes. This was a problem because as we ripped it apart our memory needs
grew very fast. We should have kept track of this from the start, it was a
pain to fix later.

This isn't a memory cycle but a problem still.
Don't:
    buf=open(file).read()
    doTheBigJob(buf)
        # buf won't be freed until this function exits. Which might be the
entire run of the program.
        # The idea is don't leave a reference at the top level
        # Even worse is if at a lower level you want to change buf.  Try
this and monitor memory usage.
>>> s="        "*999999
>>> b=s
>>> b=b+" "

Do something like this:
   parm={}
   parm['buf']=open(file).read()
   doTheBigJob(parm)
        # Now doTheBigJob can "del parm['buf']" to free the memory.
        # Or pass the buffer with in an object of some kind.

Look out for:
    dict1={}
    dict1['dict']=dict1
    # This and variations on this theme become immortal. Unless you kill it
off carefully.

Useful memory leak finding tool. But it won't find the above example unless
dict1 is in an object.
http://www.stud.ifi.uio.no/~larsga/download/python/plumbo.py

2. To use packages or not.
    We had multiple directories for the different components of our system.
So we had to setup sys.path and or use packages.  I like packages, but get
everyone to accept this up front. We fought about it because no one wanted
to deal with it. Using just the python path is simple but sooner or later
you end up with a module name the same as another on your system. The user
calls up describing a screen full of traceback that makes no sense because
you've just grabbed someone else's config.py file.

3. Decide upon exception handling strategy up front.
Don't mindlessly catch all errors, unless it's your last line of defense:
    try:
        xxx()
    except:
        yyyy
This hides errors of all kinds under a single description or handler.
Check out http://lwn.net/1999/0610/devel.phtml "The Python Way" by Tim
Peters

Do something like this:
import projectExceptions
    try:
        xxx()
    except projectExceptions.category1 , msg:
              # Our users don't want to see huge tracebacks.
          if debug:
               traceback.print_exc()
               log(sys.exc_type, msg)
          else:
               log(msg)
    except:
            # Unknown error
          if debug:
               log(sys.exc_type)
          else:
               log(projectExceptions.defaultMsg())

4. How should you handle internationalization.
    I don't know.

5. Python doesn't care about types.
    Good and bad this. Good you pass me an object and all I care about is
that it has a write method. So much nicer than C++.

    Bad, you pass me an object with out a write() and it's not detected
until runtime.
    I'm not sure but I think freeze can detect this. But don't write a boat
load of code then try a tool like freeze on it. You'll probably do something
it can't handle.

6. No const or private.
    You have to trust everyone.

Had enough ?
Do you think I have some opinions on this ?
--
--Darrell





From Christian.Tismer at p98.f112.n480.z2.fidonet.org  Tue Jun 29 15:15:25 1999
From: Christian.Tismer at p98.f112.n480.z2.fidonet.org (Christian Tismer)
Date: Tue, 29 Jun 1999 20:15:25 +0100
Subject: ANN: Stackless Python 0.2
Message-ID: <0000158a@bossar.com.pl>

From: Christian Tismer <tismer at appliedbiometrics.com>



Gary Duzan wrote:
> 
> Toby J Sargeant wrote:
> >
...
>    This appears to be the conventional wisdom. The optimization class I
> took used an ML-like language, and the first thing we did was to move
> heap frames, which was the default, to the stack (i.e. closure
> optimization.)

Loosing much of flexibility by doing that, btw.

> It might be less of an issue in an interpreted language if it already has
> heap
> allocation (or other sorts of) overhead to deal with when making calls.
> There may be other allocation tricks which can be used, as well.
> 
> > It seems that this should affect Stackless Python equally as much. Does
anyone
> > have anything to add on the subject? I would imagine that frames could be
> > allocated and managed in chunks to alleviate a lot of the memory management
> > load...
> 
>    I would think that as well (at least at first.) Maybe pools of different
> sizes of frame to limit the space overhead.

Maybe you should really look into the sources.
You are discussing things which are not Python
related, since this simply does not apply.

The Python "stacks" are so tiny that you can forget about them.
Allocations always never occour, unless your program runs
astray in recursion.

The cost of a call comes from the slightly complicated
function setup which is done all the time. Frames belong
to the fastest available Python objects already, 
like, say dictionaries.

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From tim_one at email.msn.com  Sat Jun  5 12:56:44 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 5 Jun 1999 12:56:44 -0400
Subject: Deleting characters from a string
In-Reply-To: <87n1ye3g4v.fsf@pc-hrvoje.srce.hr>
Message-ID: <000101beaf74$646413e0$8c9e2299@tim>

[Hrvoje Niksic]
> string.translate can delete characters from a string, as well as
> translate them.  But what if I want *only* to delete them?  Currently
> I do this:
>
> dummytable = string.maketrans('', '')
>
> ...
>   host = string.translate(host, dummytable, ' "')
> ...
>
> What I would really like to be able to do is write:
>
>   host = string.translate(host, None, ' "')

That's easy; change your assignment to

None = string.maketrans('', '')

and you can write exactly that <wink>.

It's (yours, not mine ...) a decent idea; in the meantime how about writing
a little wrapper

def deletechars(s, todelete, id=string.maketrans('', '')):
    return string.translate(s, id, todelete)

Then you can use an interface that doesn't bother you with arguments you
don't care about.  Or you can borrow the string module's private spelling,

    host = string.translate(host, string._idmap, ' "')

Maybe good too if _idmap lost its leading underscore -- it's been in the
module forever, and is exactly the "don't translate anything" argument you
need.

tweakingly y'rs  - tim





From daniel.mendyke at digital.com  Thu Jun 10 13:45:35 1999
From: daniel.mendyke at digital.com (Daniel)
Date: Thu, 10 Jun 1999 17:45:35 GMT
Subject: Basic python OOPs question
Message-ID: <7jot0r$p32$1@nntpd.lkg.dec.com>


Do I understand correctly that objects are removed
from the name space when they go out of scope?
And that they go out of scope based on indentation?

If so is there a default function called on 
objects that go out of scope?  The __init__
function is called when an object is created.
Is there a function that is called when the
object is destroyed?

Daniel




From mwh21 at cam.ac.uk  Wed Jun 30 13:32:47 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 30 Jun 1999 18:32:47 +0100
Subject: Reflection in Python?
References: <7ldglr$ita$1@nnrp1.deja.com>
Message-ID: <m3k8sld268.fsf@atrus.jesus.cam.ac.uk>

anders777 at my-deja.com writes:

> I've just started programming in Python, and I
> could really use some help figuring out something.
>  I'd like to implement Kent Beck's unit tester in
> Python (someone posted a very nice alternative a
> few weeks ago, but Junit's approach works better
> for my project's needs).  How do I do the
> equivalent of what Java calls reflection? I want
> to do the following:
> 
> test = MathTest("testAdd")
> 
> where test.run() will run the "testAdd" method on
> the MathTest object?  I also want to be able to
> tell the user the name of the object (MathTest)
> and the method in case the method fails.  Any
> thoughts?
> 
> Thanks,
> Anders Schneiderman
> National Journal Daily Briefings Group
> 
Not entirely sure I follow, but will this do what you want?

class Test:
    def __init__(self,object,meth):
        self.object = object
        self.meth = meth
    def run():
        try:
            getattr(self.object,self.meth)()
        except:
            print self.object.__class__.__name__, "failed!"

then

test = Test(MathTest,"testAdd")
test.run()

does what you ask (I think).

Anyway, I suspect the builtin "getattr" is what you're after.

HTH
Michael



From gmcm at hypernet.com  Thu Jun 17 20:15:42 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 17 Jun 1999 19:15:42 -0500
Subject: wierdness with pythonwin dialogs, threads & refcounts
In-Reply-To: <7kbsn1$dno$1@m2.c2.telstra-mm.net.au>
Message-ID: <1282457297-115891134@hypernet.com>

Mark Hammond writes:

> There could be a number of leaks in Pythonwin, but I dont think
> dialogs do.
> 
> The ref count is likely to be 6, as the pywin.mfc.Dialog base class
> hooks a number of commands and messages.  Each of these hooks is a
> circular reference, as a bound method is saved away.  These
> references are deleted when a window recieves the WM_DESTROY
> message.
> 
> So in a nutshell, there is nothing obvious as to why the dialog is
> leaking. The very first step would be to ensure it does not leak
> without threads.
> 
> Otherwise, it may come down to me tracking this down, as there are
> some complicated interactions with MFC going on...

I played with this enough to find that he's onto something. Without
any threading, the dialog gets a WM_DESTROY and goes thru __del__.
When executed from a threading.Thread, it never goes thru OnDestroy.
I tried doing DoModal myself, then un-hooking the messages myself.
No luck. Perhaps because there's no parent window, and the trhead
does not have a Windows message queue?

> Dave Kirby wrote in message <3768b75e.5469224 at NNRP.UK.INSNET.NET>...
> >
> >I have found strange behaviour (strange to me at least) when creating
> >a pythonwin dialog on a separate thread created using the standard
> 
> 
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list

- Gordon



From holger at phoenix-edv.netzservice.de  Thu Jun 10 08:03:58 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Thu, 10 Jun 1999 12:03:58 GMT
Subject: force sequenz with only one item?
Message-ID: <375FA9AE.6299E9DD@phoenix-edv.netzservice.de>

High,

I think it's easy to solve, but I couldn't get it right now.
Take a look at the command-line source below:

>>> A = 0
>>> B = 1
>>> seq=raw_input("?")
?A,B
>>> print seq
A,B
>>> print eval(seq)
(0, 1)
>>> seq=raw_input("?")
?A
>>> print eval(seq)
0
>>> A in seq
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: string member test needs char left operand
>>> A in eval(seq)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: 'in' or 'not in' needs sequence right argument
>>>

Ok? Check it out?
That's what I want to do with it:
There's a file I'm parsing over special keywords per line, e.g.:
Action = CREATE, VERBOSE, APPEND
When read the line I put the right term to a variable, e.g:
anAction = eval(rightTerm)
Then I want to test for that keywords (global integer variables in 
source, e.g.: CREATE = 0):
if CREATE in anAction:
	pass #do something

But how could I force python to accept one-item sequences (see below!)?

Thanx for comin' with me...

Ciao,
Holger



From Olaf.Delgado at p98.f112.n480.z2.fidonet.org  Tue Jun 29 18:25:16 1999
From: Olaf.Delgado at p98.f112.n480.z2.fidonet.org (Olaf Delgado)
Date: Tue, 29 Jun 1999 23:25:16 +0100
Subject: Binary pickle portability
Message-ID: <00001585@bossar.com.pl>

From: Olaf Delgado <delgado at Mathematik.Uni-Bielefeld.DE>

On Tue, 29 Jun 1999, Olaf Delgado wrote:

> Lately I observed strange things when trying to unpickle a binary pickle
> (including floating point numbers) produced on a different architecture.

Sorry, that was too quick.

After looking a bit closer, I found that the problem is probably related
to the Numeric module (i.e., pickling matrices) and doesn't depend on
whether binary or ordinary mode is used. So please don't bother now!
Still, I am interested in whether people use binary pickles.

Olaf




From aahz at netcom.com  Mon Jun 14 21:15:04 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Tue, 15 Jun 1999 01:15:04 GMT
Subject: do anonymous pipes normally work on NT?
References: <7juatc$fef$1@nnrp1.deja.com> <7k43sk$6vd$1@nnrp1.deja.com>
Message-ID: <aahzFDCGt4.2Dn@netcom.com>

In article <7k43sk$6vd$1 at nnrp1.deja.com>,
Georg Mischler  <schorsch at schorsch.com> wrote:
>
>I simplified my class to the absolute minimum I could.  It now does
>nothing but create three pipes, save the parent ends to self.xxx, and
>create a process with the child ends of the pipes. After that it tries
>to write to the stdin pipe of the child, which goes without error, and
>tries to read from stdout, which blocks.

How about sockets?
-- 
                      --- Aahz (@netcom.com)

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

"That doesn't stop me from wanting to work all three of them over with
the clue stick for while, with no safewords allowed."  --abostick



From faassen at pop.vet.uu.nl  Mon Jun  7 04:28:04 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Mon, 07 Jun 1999 10:28:04 +0200
Subject: Suggestion: Dutch Python mailinglist
References: <slrn7lli2q.ia.gerrit@gerrit.palga.uucp> <aahzFCxG59.D2@netcom.com>
Message-ID: <375B8294.F3A8C382@pop.vet.uu.nl>

Aahz Maruch wrote:
[snip]
> (Gerrit, in case the double-talk above wasn't clear, your English is
> very, very good, and I wish you would stop beating yourself.)

I frequently get embarrassed by the English of native speakers. My
English grammar may not be optimal, and I make a few spelling mistakes
here and there, but Cmdr_Taco of Slashdot makes far more. :)

That said, if there's a Dutch Python mailing list I'll probably join it.
I don't know how talkative I'll be, though -- I'm spread across a few
too many mailing lists already.

Note that there is a cosmic conspiracy involving the Dutch language, by
the way, which I uncovered some months. All kinds of people in
comp.lang.python that look like normal Usenet denizens turn out to speak
Dutch!

Regards,

Martijn



From eugene.leitl at lrz.uni-muenchen.de  Wed Jun  9 17:55:46 1999
From: eugene.leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Wed,  9 Jun 1999 14:55:46 -0700 (PDT)
Subject: Using perl in python
In-Reply-To: <6cs73.2432$nn.739920@news.shore.net>
References: <7jjoqh$t56$1@nnrp1.deja.com>
	<6cs73.2432$nn.739920@news.shore.net>
Message-ID: <14174.58040.951136.437087@lrz.de>

Michael P. Reilly writes:
 
 > I've found that it is more efficient (and more fun) to convert the
 > libraries to Python.  The code is invariably clearer.

You have got some spare time to do the whole of CPAN? Much appreciated.



From petrilli at trump.amber.org  Mon Jun 28 18:35:43 1999
From: petrilli at trump.amber.org (Christopher Petrilli)
Date: 28 Jun 1999 15:35:43 PDT
Subject: ConfigParser.py?
References: <rNHvjEdhm5Pp-pn2-ob9OW6dCFSRN@d185d18f8.rochester.rr.com> <7l876r$nrs@chronicle.concentric.net> <rNHvjEdhm5Pp-pn2-obELMofjPdsy@d185d18f8.rochester.rr.com>
Message-ID: <7l8tbv$62m@journal.concentric.net>

Benjamin Schollnick <rfreedm1 at nospam.rochester.rr.com> wrote:
> On Mon, 28 Jun 1999 09:17:31, Christopher Petrilli 
> <petrilli at trump.amber.org> wrote:

>> > 	I'm trying to use ConfigParser... And I believe I'm doing everything
>> > write, except I don't see any documentation (including in the 
>> > configparser.py file)
>> > telling my what function(s) to use to write the file out to disk???
>> 
>> That's simple... it doesn't exist :-)

> What's the rationale behind this???  It seems to be silly to allow 
> read, and not some
> sort of output???

Um, laziness is the most powerful of rationale :-) Write it and I'm sure
Guido will be happy to include it :-)  I've just never needed it, which
is why I don't use it.  I use config files for a bunch of scripts that
run, but there's no "GUI" way to edit them.

Chris
-- 
| Christopher Petrilli                      ``Television is bubble-gum for
| petrilli at amber.org                          the mind.''-Frank Lloyd Wright



From aahz at netcom.com  Sat Jun  5 12:27:57 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Sat, 5 Jun 1999 16:27:57 GMT
Subject: Deleting characters from a string
References: <87n1ye3g4v.fsf@pc-hrvoje.srce.hr>
Message-ID: <aahzFCv4EL.7HC@netcom.com>

In article <87n1ye3g4v.fsf at pc-hrvoje.srce.hr>,
Hrvoje Niksic  <hniksic at srce.hr> wrote:
>
>What I would really like to be able to do is write:
>
>  host = string.translate(host, None, ' "')
>
>Am I missing a more elegant solution?

I dunno, but I'd tend to use re.sub().
-- 
                      --- Aahz (@netcom.com)

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

"Why do you like my boyfriend to tie you up and beat me.?"
    -- ELIZA generates a poly moment



From aahz at netcom.com  Sun Jun  6 18:36:44 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Sun, 6 Jun 1999 22:36:44 GMT
Subject: Suggestion: Dutch Python mailinglist
References: <slrn7lli2q.ia.gerrit@gerrit.palga.uucp>
Message-ID: <aahzFCxG59.D2@netcom.com>

In article <slrn7lli2q.ia.gerrit at gerrit.palga.uucp>,
Gerrit Holl <gerrit at nl.linux.org> wrote:
>
>I'm a 13-years-old Dutch boy, and my English is not good. Now I have
>the chance to set up a Dutch python mailinglist. I know there are many
>Dutch python-users, but most of them know English very well. So will
>there be any users who'll subscribe to such a mailinglist, or will it
>remain very quiet?

Geeze, I send Gerrit an e-mail telling him that his English is better
than many native speakers twice his age, and this is the thanks I get?

;-)

(Gerrit, in case the double-talk above wasn't clear, your English is
very, very good, and I wish you would stop beating yourself.)
-- 
                      --- Aahz (@netcom.com)

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

"Why do you like my boyfriend to tie you up and beat me.?"
    -- ELIZA generates a poly moment



From morse at harborcom.net  Mon Jun 28 08:28:01 1999
From: morse at harborcom.net (Kevin Dahlhausen)
Date: Mon, 28 Jun 1999 12:28:01 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773c3a0.102825445@news.oh.verio.com> <930340248.9300@www.remarq.com>
Message-ID: <37776874.4629426@news.oh.verio.com>

Voynow, Sasha <anonymous at web.remarq.com> wrote:

>
>Has anyone SWIG'd or otherwise wrapped FLTK?

It's 'a work in progress.'   Check out:  http://fltk.netpedia.net





From claird at Starbase.NeoSoft.COM  Sat Jun 12 14:23:36 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 12 Jun 1999 13:23:36 -0500
Subject: Broken 1.5.2 installation? Missing cw3215mt.dll
References: <1282960807-85602620@hypernet.com>
Message-ID: <7ju8j8$2hi$1@Starbase.NeoSoft.COM>

In article <1282960807-85602620 at hypernet.com>,
Gordon McMillan <gmcm at hypernet.com> wrote:
>Ivan Van Laningham wrote:
			.
			.
			.
>[double snort]
>
>A few months ago you were running some auto-port from a PDP-8. Did
>RH finally convince you no warm-blooded animals were used in user
>testing?
>
>
>
>- Gordon
>
>

Do it with a feathered snake, and he'll follow you an...
Whoops!  This is in public.


Still rolling up those paper tapes, 
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From goran.allerbo at carmenta.se  Tue Jun  1 04:26:13 1999
From: goran.allerbo at carmenta.se (Goran Allerbo)
Date: Tue, 01 Jun 1999 10:26:13 +0200
Subject: works: hjälpprylar (ignore!)
References: <002201beab54$66218d30$f29b12c2@pythonware.com> <003301beab56$88cb3af0$f29b12c2@pythonware.com>
Message-ID: <37539925.31FF8DCC@carmenta.se>

Fredrik Lundh wrote:
> 
> I wrote:
> > makedoc.py bygger hj?lppackarna fr?n XHTML-
> > dokumentation under manual/...
> 
> but didn't mean to send it to this forum,
> of course...
> 
> (fast det ?r klart, om ni har n?gon nytta av
> informationen s? ta g?rna f?r er ;-)
> 
> </F>

det ?r ok med lite l?ttf?rst?eligt spr?k i gruppen
tycker jag ...

;) ga



From mgm at unpkhswm04.bscc.bls.com  Tue Jun 22 14:30:19 1999
From: mgm at unpkhswm04.bscc.bls.com (Mitchell Morris)
Date: 22 Jun 1999 18:30:19 GMT
Subject: It's just magic, and I hate it ...
References: <slrn7mss5c.f90.mgm@unpkhswm04.bscc.bls.com> <14190.35530.772547.166720@buffalo.fnal.gov>
Message-ID: <slrn7mvlhr.4is.mgm@unpkhswm04.bscc.bls.com>

In article <14190.35530.772547.166720 at buffalo.fnal.gov>, Charles G Waldman
wrote:
>
>Try running Python in unbuffered mode - with a "-u" commandline flag,
>or by setting the PYTHONUNUFFERED environment variable to a non-empty
>string.

Are you saying that this worked for you? Or was this a guess?

I ask because it makes no difference when I try it.


-- 
Mitchell Morris

We think the new [Nintendo] system will use floppy disks and will be powered
by a steam engine.
	-- Next Generation Online



From jeffrey.polaski at west.boeing.com  Wed Jun  9 19:19:30 1999
From: jeffrey.polaski at west.boeing.com (Jeff P.)
Date: Wed, 9 Jun 1999 23:19:30 GMT
Subject: Windows Printing Question
Message-ID: <375EF682.CCAC69D1@west.boeing.com>

I have a set of Internet shortcuts on a Windows 95 machine that I'd like
to print out. I know Windows can do this, after all, I can right-click
on a shortcut and select "Print" and it'll do it. The question is: How
to do this programaticaly?

The problem is Windows grabs the files too fast, and if it doesn't get
the web page fast enough, just prints out a blank page with the url at
the top. Not really useful to me, as I need to have a copy of the page,
as seen by the browser. There _should_ be a delay of 2-5 seconds between
each one being printed. That way there won't be so many blank pages
after all of this is done. 

I have 4,605 shortcuts that I created with a little python script and a
directory listing of the site. The site has enough links that use
Javascript so if I use a site-grabber to grab the pages, it'll skip a
bunch of pages.

How would I use Python to print each of these pages pointed to by the
shortcuts?

Thanks in advance!

Jeff Polaski

W. jeffrey.polaski at west.boeing.com
H. nrrd at earthlink.net



From adjih at technologist.com  Fri Jun  4 12:49:24 1999
From: adjih at technologist.com (adjih at technologist.com)
Date: Fri, 04 Jun 1999 16:49:24 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <7j450j$908$2@cronkite.cc.uga.edu> <7j5sr7$eb4$1@nnrp1.deja.com> <7j6j8d$58f$2@cronkite.cc.uga.edu>
Message-ID: <7j902g$hob$1@nnrp1.deja.com>

In article <7j6j8d$58f$2 at cronkite.cc.uga.edu>,
  graham at sloth.math.uga.edu (Graham Matthews) wrote:
> adjih at technologist.com wrote:
> :   But there will be cases where full garbage collection will be
> : definitly a nuisance.
> :   First, you must scan all the address space of the C code, which
can be
> : a problem if you're just using the Python interpreter in a library
(i.e.
> : don't have the ability to control or modifying the main program) ;
or if
> : your address space is too big (I use python for instance in a
program
> : that does an mmap of a 350 MB file).
>
> a) you can turn gc off

Yes this is a good solution. But then you must be careful that very few
of the modules publicly released for Python rely on GC  unless they
are already very specific.

> b) I routinely use garbage collection on 1 Gig memory spaces with no
> 	 speed problems.

The problem is not the size of address space as such, but the ratio
[non-paged out memory] / [total memory accessible by the program]
compared to the speed at which your application generate cyclic
structures. If you have to read 350 MB from disk each time you do
a GC, then you will have a huge pause, but still may be forced
to GC often if your application create many cycles at a high rate.

Disabling the GC is the solution.

>
> adjih at technologist.com wrote:
> :   Second you kill the portability: you could use a subset of Python
with
> : ANSI-C alone ; but you can't make a incremental GC in ANSI-C.
>
> Then don't use an incremental collector! Use a mark sweep collector.
> I use such a thing almost every day -- written entirely in ANSI-C.

  It is true that it shouldn't be hard to make a mark-sweep collector
that is working in practice on many platforms.

  Pedantically I doubt that ANSI-C tells you the current stack top
address, or guarantees that the whole address space between the current
stack pointer and this top stack address is stack space. But I agree
that actual C compilers wouldn't do something too stupid, we
shouldn't care about this :-)

>
> :   You'd get problems when embedding Python in unusual environments ;
for
> : instance if you embed Python in the Linux kernel, then the GC should
> : also trace all the pages in the swap, or be rewritten specially for
that
> : case.
>
> Why do you need to do any of these things?

Use Python in the Linux kernel or in an unusual environment ? Why not ?
Mark-and-sweep pages in swap ? Well only if you could have references
from pages in swap (which is the case if you use the generic vmalloc in
2.2+).

> adjih at technologist.com wrote:
> :   The advantages of GC must be traded with its drawbacks:
> : - advantages: reclaims cyclic structures.
> : - drawbacks: non-portable, requires a special handling when linking
with
> : C, high complexity for decent GC (multi-thread incremental with
> : finalization and support for C/C++ structures referencing Python
> : structures).
>
> The drawbacks you list are all misleading. Mark sweep collection (with
> refcounts) is highly portable. Non compacting (or even partially
> compacting) collectors require no special code to link with C.
> Complexity of a mark sweep collector is not very high at all
>(registering
> roots and traversing foreign objects being the main areas of
> complexity).

Registering the roots isn't necessarily straightforward: you need
control and information about all the allocators that could be used in
the program.
You'd have problems if you created a shared library that created
internally an instance a Python interpreter and offered objects
with internal pointers to Python objects.

> I am sure that for *some* collection schemes some of
> the drawbacks you list are true, but not for all.

  Non incremental mark-sweep collection should be portable, but
then of course you'll have to cope with the GC pauses. This is
another reason to disable GC by default.

  What I would object to, is a mandatory GC (such as in many Lisps),
or the existence of an important proportion of C or Python code
relying on GC.

-- Cedric


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From bernhard at alpha1.csd.uwm.edu  Fri Jun  4 00:54:28 1999
From: bernhard at alpha1.csd.uwm.edu (Bernhard Reiter)
Date: 4 Jun 1999 04:54:28 GMT
Subject: Exmaples of re module
References: <7j78eq$s1dt@eGroups.com>
Message-ID: <slrn7len04.ice.bernhard@alpha1.csd.uwm.edu>

On Fri, 4 Jun 1999 01:00:10 GMT, air at apex.net <air at apex.net> wrote:
>I am looking for some examples of the re module, mainly I need to be able to do what I used to do with awk, grab part of a file from Point a to point b, in awk you could do it like this
>/--Begin--/,/--End--/!d
>or something similar to this, any help would be appreciated.

Hmm here is my zero-dot-second shot at doing a typical awk task
with python, it might be a crude example, because I just played with 
some elements and the main() part is a bit lengthy.
Methods from the string and re modules are used as needed.
For some purposes the string methods seem to be sufficient.

I wanted to filter lines which my tirc here spits out, and they
look like (with ^C being real Control-Cs, I replaced them for this posting):

### Opening logfile (channel #jzz), [Mon Feb 16 09:55:44 1999]
[Mon Feb 16 09:57:12 1999]<ROW^C7,0;> da ist er ja 
[Mon Feb 16 09:57:18 1999]#jzz> Wer? ;-)
[Mon Feb 16 09:57:32 1999]<ROW^C7,0;> Cliff (bernhard at alpha1.csd.uwm.edu)
[Mon Feb 16 09:57:47 1999]#jzz> Ja! 17:00 war doch abgemacht, oder?
[Mon Feb 16 09:59:12 1999]* Cliff bemerkt, dass wir etwa 2 Sekunden "ping Zeit" haben.

After filtering, they look like:

### Opening logfile (channel #jzz), [Mon Feb 15 09:55:44 1999]
[09:57:12] <JOW> da ist er ja 
[09:57:18] <Cliff> Wer? ;-)
[09:57:32] <JOW> Cliff (bernhard at alpha1.csd.uwm.edu)
[09:57:47] <Cliff> Ja! 17:00 war doch abgemacht, oder?
[09:59:12]* Cliff bemerkt, dass wir etwa 2 Sekunden "ping Zeit" haben.


Usually you would want to overwrite the body of process().
I tried to minic a few AWK variables, they are not all used.
They could be commented out for speed, if unneeded.

You can use a global variable to record a state, like
when you have encoundered your beginning line.

I inserted a few comments to show you where the BEGIN and END awk
actions have to take place, so this is untested coded ahead:
	Bernhard


#! /bin/env python
"""Format and clean tirc normal logfile for xxx IRC session logs.  v%(version)s

USAGE:
 %(progname)s [inputfilename(s)]  >outputfilename

The default is to use <Cliff> as user.
The time entries on each line will be cutted to include only the time
and not the day or year, which easily can be seen from the beginning
and end of a consecutive session.

Of course this is dependend on the time format tirc uses.
"""
__version__="0.1demo"
# initial 2.6.1999 Bernhard Reiter
# 3.6.1999 demo for comp.lang.python


import fileinput
import sys
import string
import re


nick="Cliff"

def process(line,fileinputobject,write):
    """Process one inputline and spit out what is wanted."""

    # share some things between subsequent calls
    global channel
    global nickre

    f=string.split(line,";")
    f=map(string.strip,f)

    # imitate gawk a bit   ;-)
    NF=len(f)
    FILENAME=fileinputobject.filename()
    NR=fileinputobject.lineno()
    FNR=fileinputobject.filelineno()


    if line[0:3]=="###":
	# opening or closing of logfile

	# grab channelname
	matchobj=re.search('\(channel (#[^)]+)\)',line)
	channel=matchobj.group(1)
	sys.stderr.write("Found channel: " + channel + "\n" )
	
	# prepare nick replacement procedure
	nickre=re.compile(re.escape("]" +channel+">"))
	
    if line[0]=="[":
	# normal line

	# replace funny control string
	line=string.replace(line,chr(3)+"7,0;","",1)
	line=nickre.sub("]<"+nick+">",line,1)
	
	# distance between time and normal text
	if line[25:27]=="]<":
		line=line[:26]+" "+line[26:]

	# cut date and year out 
	line="["+line[12:20]+"]"+line[26:]

    write(line)
	


def main():
    """Check script arguments and do the work."""

    if len(sys.argv)==1:
	sys.stderr.write(__doc__ % 
	    {"progname":sys.argv[0], "version":__version__})

	if sys.platform=="win32":
	    import msvcrt
	    msvcrt.getch()
	    sys.stderr.write("\nPress any key to start reading from stdin.\n")
	else:
	    sys.stderr.write("Now reading from stdin.\n")

    #AWK's BEGINs {}

    #prepare for AWK middelpart in process()

    fileinputobject=fileinput.input()
    outputwritefunction=sys.stdout.write

    try:
	for line in fileinputobject:
	    process(line,fileinputobject,outputwritefunction)
    finally:
    	fileinputobject.close()

    #AWK's ENDs {}
	

if __name__=="__main__":
	main()





From just at letterror.com  Wed Jun 30 14:21:37 1999
From: just at letterror.com (Just van Rossum)
Date: Wed, 30 Jun 1999 18:21:37 GMT
Subject: Python Editor
In-Reply-To: <7ldh84$rle@dfw-ixnews12.ix.netcom.com>
References: <AJqe3.1318$24.76@news01.t-net.net.ve>
            <7ldh84$rle@dfw-ixnews12.ix.netcom.com>
Message-ID: <l03102803b3a0104edd47@[193.78.237.121]>

>In article <AJqe3.1318$24.76 at news01.t-net.net.ve>,
>Jose A. Antich <jantich at cantv.net> wrote:
>>
>>   Does anyone know if there's any Python Editor for the MAc?

Python on the Mac is shipped with a Mac native IDE.

At 4:39 PM +0000 6/30/99, Aahz Maruch wrote:
>IDLE, a debugger/editor written in Python.  You'll need Python 1.5.2 and
>Tcl/Tk 8.0 (*not* 8.1).

IDLE does not work well on the Mac.

Just





From dan_knierim at my-deja.com  Mon Jun 28 16:02:49 1999
From: dan_knierim at my-deja.com (dan_knierim at my-deja.com)
Date: Mon, 28 Jun 1999 20:02:49 GMT
Subject: getting memory information under NT
References: <002201beab54$66218d30$f29b12c2@pythonware.com> <003301beab56$88cb3af0$f29b12c2@pythonware.com> <375E88A2.68C1F118@opticominc.com>
Message-ID: <7l8kd0$r18$1@nnrp1.deja.com>

In article <375E88A2.68C1F118 at opticominc.com>,
  Eric Renouf <erenouf at opticominc.com> wrote:
> Does anyone know of an easy way to get the the total physical memory
> and/or available memory on an NT system using Python 1.5.2?  Is there
a
> module that has functions for things like this already?
>
> Thanks,
> Eric Renouf
>
I earlier posted an incorrect response to your inquiry, sorry.

A better answer is: look at Sam Rushing's windll and structob modules,
part of the DynWin software available from the python.org download site,
or from http://nightmare.com/software.html.  These use Mr. Rushing's
calldll and npstruct utilities, which are also available from those
sites.

The downloads include some very helpful demo code, but the demos I
got seem not quite in sync with the release of DynWin that I have.  In
particular the module dlldemo.py has an example of how to call the
Windows function 'GlobalMemoryStatus' as follows (edited for brevity):

# -*- Mode: Python; tab-width: 4 -*-
# Author: Sam Rushing <rushing at nightmare.com>

import windll, structob

kernel32 = windll.module ('kernel32')

memstat = structob.Oracle ('memory status structure', 'Nllllllll',
                           ('Length', 'MemoryLoad', 'TotalPhys',
                            'AvailPhys', 'TotalPageFile',
                            'AvailPageFile', 'TotalVirtual',
                            'AvailVirtual') )

def global_memory_status():
	buffer = windll.membuf (memstat.size)
	result = kernel32.GlobalMemoryStatus (buffer.address())
	dict, size = memstat.unpack (buffer.read())
	memstat.describe (dict)

----------- end of code fragment

I haven't checked the following with the author or any Python wizes, but
I'll explain what I think based on my experiments and reading bits of
the code.

The second parameter in the call to structob.Oracle, 'Nllllllll', is a
format code, and the first character 'N' is supposed to indicate whether
the data structure uses big-endian or little-endian byte order.  However
this value must be for a different version of npstruct, because the
release I have of npstruct.py (which interprets the format code) expects
either an 'L' (for little-endian) or a 'B' (for big-endian) as the first
character.  After I replaced the 'N' with 'L', the assignment to memstat
looks like

memstat = structob.Oracle ( 'memory status structure', 'Lllllllll',
etc.

The demo now runs fine.

The Win32 extensions from Mark Hammond (also available from
www.python.org in the win32all download) provide an interface to the
Performance Monitor counters, in modules win32pdh*.py.  This gives
access to several detailed measures related to memory usage (the same
ones available through the NT Performance Monitor admin tool).

Good luck,
Dan K.

--------------------------------------------------------------
Corporation: n. An ingenious device for obtaining individual profit
without individual responsibility. -- Ambrose Bierce


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From arcege at shore.net  Sat Jun  5 19:16:56 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Sat, 05 Jun 1999 23:16:56 GMT
Subject: smtp module
References: <0gf63.28$Vv4.277@news.megsinet.net>
Message-ID: <Idi63.2035$nn.613205@news.shore.net>

Philip C. <pjc_2 at geocities.com> wrote:
: I have pretty much figured out how to use the module but I can't seem
: set the subject or the from of the bcc etc....

: please tell me how to do it?

You need to include the SMTP headers in the message itself.

  from smtplib import SMTP
  fromaddr = 'arcege at shore.net'
  toaddr = 'python-list at python.org'
  message = '''\
  From: %(fromaddr)s
  To: %(toaddr)s
  Subject: Re: smtp module
  Bcc: %(fromaddr)s
  Reply-to: %(fromaddr)s

  Philip C. <pjc_2geocities.com> wrote:
  : I have pretty much figured out how to use the module but I can't seem
  : set the subject or the from of the bcc etc....

  : please tell me how to do it?
  ''' % locals()

  SMTP('mailhost').sendmail(fromaddr, [toaddr], message)

The protocol for SMTP requires that the headers are included.

  -Arcege




From alex at mindlube.com  Tue Jun  8 03:45:23 1999
From: alex at mindlube.com (alex at mindlube.com)
Date: 08 Jun 1999 01:45:23 -0600
Subject: python active web pages
References: <jonathan-0406991640390001@borg.onegoodidea.net> <37582D9D.3D7A4B1D@hp.com.NOSPAM> <jonathan-0406992318350001@borg.onegoodidea.net> <877lpgc4cz.fsf@smtp.linuxcare.com>
Message-ID: <m3ogircezw.fsf@monsoon.swcp.com>

One word: Zope!

www.zope.org


dwelton at linuxcare.com (David N. Welton) writes:

> jonathan at onegoodidea.com (Jonathan Hogg) writes:
> > i'm not implementing on Windows or using "Active Server Pages" as
> > defined by Microsoft. i'm interested in dynamic web pages on
> > unix/apache.
> 




From tim_one at email.msn.com  Fri Jun 11 00:19:46 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 11 Jun 1999 00:19:46 -0400
Subject: fork()
In-Reply-To: <14175.51022.577016.265867@amarok.cnri.reston.va.us>
Message-ID: <000c01beb3c1$a3f428a0$329e2299@tim>

[Andrew M. Kuchling]
> ...
> Perhaps lists and dictionaries could contain prev/next pointers that
> are used to maintain them in a doubly-linked list, and a linear-time
> traversal would be done to find unmarked objects, but there has to
> be a better way.

In this kind of approach, I doubt there's a substantially better way.  Guido
will also want a doubly-linked list (where the links don't count against the
refcount!) so that when a dict goes out of existence via the normal
refcount-hits-0 route, the dict can efficiently unlink itself from the list.

A worthwhile variation of the mark/sweep scheme you sketched is to move
objects into a new doubly-linked list *as* they're marked.  Then when the
mark phase ends, that new doubly-linked list is automatically the new list
of live dicts, and what remains in the old doubly-linked list is
automatically the list of cyclic dict trash.

At this point finalizers can still be run with everything intact, although
(as the Java folk keep saying) not particularly *usefully* in the absence of
a way for the user to force the order they're called in.

So, ignoring that, you make one pass over the trash list just to run
finalizers.

The next problem is resurrection:  running a finalizer can magically make
one or more trash items reachable again (e.g., "global somelist;
somelist.append(self)".  So at the end of the finalizer pass, you have no
idea what's still trash!

That's where variants of Graham's "two pass" come in.  If you're Java, you
mark each object whose finalizer has been run with a secret sticky "already
ran the finalizer" flag, and never run the finalizer on such an object again
even if it resurrects and becomes trash again a thousand times over.

Conceptually, you need to run the mark phase all over again to determine
what's still trash after finalization, and so on until you reach a steady
state.  In practice, you get clever instead.

The reward in the end is that you get something like the Java Spec's
(section 12.6.1) 8-state transition network for the possible paths of an
object thru the "is it trash yet?" space, and users figure the only way to
preserve their sanity in the face of this mess is to avoid finalizers
altogether <0.4 wink>.

resurrection-is-a-bitch-ly y'rs  - tim





From gregm at iname.com  Thu Jun  3 10:59:22 1999
From: gregm at iname.com (Greg McFarlane)
Date: Thu, 3 Jun 1999 14:59:22 GMT
Subject: [tkinter] two questions regarding Canvas
In-Reply-To: <37556602.415AFA46@ingr.com>; from Joseph Robertson on 02 Jun 1999 at 12:12:34PM
References: <3752a555@nntp.server.uni-frankfurt.de> <37556602.415AFA46@ingr.com>
Message-ID: <19990604005922.07590@nms.otc.com.au>

This is small attempt at adding editing bindings to canvas text items. 
This *must* have been done before, but a search of comp.lang.tcl did
not come up with anything, except a mention of pg_access in postgresql
which I could not find.  Does anyone know of it?

======================================================================
import Tkinter 

def click(event):
    canvas.delete('rect')
    region = canvas.bbox('current')
    if region is None:
        canvas.focus('')
    else:
        canvas.create_rectangle(region, outline = 'blue', tags = 'rect')
        canvas.focus('current')
        canvas.icursor('current', '@%d,%d' % (event.x, event.y))

def double(event):
    canvas.select_from('current', 0)
    canvas.select_to('current', 'end')

root = Tkinter.Tk()
canvas = Tkinter.Canvas(root)
canvas.pack()

canvas.bind('<Button-1>', click)
item1 = canvas.create_text(100, 100, text = 'Hello\nthere', tags = 'text')
item2 = canvas.create_text(200, 100, text = 'Good evening', tags = 'text')

canvas.focus_set()

# Need lots more bindings: button motion (for changing the selection),
# keys (for adding and deleting text), etc (same as text widget).
canvas.tag_bind('text', '<Double-1>', double)

root.mainloop()   
======================================================================

On 2 Jun, Joseph Robertson wrote:
> Sure does,  heres some relevant clips.  Sorry I couldn't send the whole
> thing but its too big.
> I have a class that creates the canvas, then I set the binding:
> 
> def __init__(self)
> ...
> self.canvas.bind('<Button-1>', self.CellClick)
> ...
> 
> Heres the working method:
> def CellClick(self, event):
>         self.canvas.delete('rect')
>         cellbbox = self.canvas.bbox('current')
>         self.canvas.create_rectangle(cellbbox, outline='blue',
> tags='rect')
> 
> 
> Now I don't have it editable yet, I am using this,
>     item = self.canvas.create_text(xpos, ypos,  text=field, font=self.f)
> 
> in a loop to put data onto the canvas.
> 
> I'm sure that creating a hidden Entry and the moving it and making it
> visible would work fine, or don't move it and make a 'status line' type
> of data editing, like Lotus did.
> 
> Good Luck,
> Joe
> 
> 
> 
> mlauer at amalia-atm.rz.uni-frankfurt.de wrote:
> 
> > Howdy, though I'm gaining knowledge through
> > browsing through all the available documentation
> > (that is, the tkinter life-preserver, tkinter.py
> > and the tk/tcl documentation) some things
> > remain misterious...
> >
> > For instance:
> >
> > a) Does the tk Canvas support "inplace" editing of
> > text items or what are the functions Canvas.icursor
> > and Canvas.focus for ? If so, how ? For instance,
> > binding the <Any-Enter> to a text item function,
> > which sets icursor and focus doesn't work for me
> > (yes, the canvas has the current focus)
> >
> > b) Does the Canvas support a visible selection
> > of items (such as a boundaring rectangle or so)
> > on its own or have I to do this via binding <Any-Enter>
> > and <Any-Leave> ?
> >
> > Have a nice day!
> >
> > --
> >     Regards & Gruesse from Mickey @ http://www.Vanille.de
> >   ---------------------------------------------------------
> >   How could anyone know me - when I don't even know myself ?
> 
> 
> 
> 
> 

-- 
Greg McFarlane:    INMS Telstra Australia (gregm at iname.com)
Today's forecast:  Sunny, with occasional cloudy periods and a chance
		   of rain in some areas.



From emile at fenx.com  Sat Jun 12 21:34:03 1999
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 12 Jun 1999 18:34:03 -0700
Subject: Python on HP-UX
Message-ID: <zTD83.1322$VL2.213635@news.direcpc.com>

Hi,

I'm trying to get Python 1.52 going on HP-UX 10.20.  I put the source
on, compiled, and found out about the ansi-c issue.  I tried to continue
after taking out pcre support per a post I found from Guido, but it then
stalled on "posixmodule.c, line 666: error 1705: Function prototypes are
an ANSI feature.".    I'm guessing that I shouldn't force it further.

I then put gcc on, learned about depot files and swinstall, and...
swinstall's either not on the machine, or not in my path!  Is there a
way to get around this, or might binary executables be available
somewhere someone could point me to?  I'd like to continue this
week-end, and not have to wait for the sysadmin to return on Tuesday.

Reminded-of-the-unbundled-SCO-dazed-ly yr's

--

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







From aahz at netcom.com  Sun Jun  6 12:42:04 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Sun, 6 Jun 1999 16:42:04 GMT
Subject: Newbie question
References: <7je3sk$46$1@nnrp1.deja.com>
Message-ID: <aahzFCwzq4.5v@netcom.com>

In article <7je3sk$46$1 at nnrp1.deja.com>,  <mesak1234 at my-deja.com> wrote:
>
>Well, i get problems with getting date from OS....(win32 at this
>moment)... All what i need that to find out date attribute of a file.

I think you want os.stat()
-- 
                      --- Aahz (@netcom.com)

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

"Why do you like my boyfriend to tie you up and beat me.?"
    -- ELIZA generates a poly moment



From j_belbo at hotmail.com  Fri Jun 25 08:18:22 1999
From: j_belbo at hotmail.com (j_belbo at hotmail.com)
Date: Fri, 25 Jun 1999 12:18:22 GMT
Subject: How to use perfmon ?
Message-ID: <7kvs28$2fq$1@nnrp1.deja.com>

Hello,

Is anybody that can help me to use this library.
I want to monitor the CPU ressources used by a remote process on a NT
server.
I'm looking for a starting example.

Thanks,

Jacobo


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From akuchlin at mems-exchange.org  Fri Jun  4 13:52:31 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: Fri, 4 Jun 1999 13:52:31 -0400 (EDT)
Subject: Alternative dictionary output format?
In-Reply-To: <WtT53.20$KR2.41493@news.corridex.com>
References: <7j8pd8$f01$1@nnrp1.deja.com>
	<slrn7lg030.i5l.jepler@craie.inetnebr.com>
	<WtT53.20$KR2.41493@news.corridex.com>
Message-ID: <14168.4703.163579.140741@amarok.cnri.reston.va.us>

Evan Simpson writes:
>I don't know if this is 100% bulletproof, but it has simple behavior:
>pctsub = re.compile('%(?!%*\()').sub
>def pfstr(s, pctsub = pctsub):
>  '''Double all '%' except those leading up to a '(', for dict formatting'''
>  return pctsub('%%', s)

An alternative is re.sub('%([^(]|\Z)', r'%%\1', <string>) .  The [^(]
| \Z will match either any character that's not a (, or at the end of
the string.  The \Z is needed in order to handle a trailing '%' --
regexes are tricky.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
You mustn't kill me. You don't love me. You d-don't even know me.
    -- The Furies kill Abel, in SANDMAN #66: "The Kindly Ones:10"





From gherron at aw.sgi.com  Mon Jun  7 15:38:57 1999
From: gherron at aw.sgi.com (Gary Herron)
Date: Mon, 7 Jun 1999 19:38:57 GMT
Subject: waste of resources ?
References: <375C0F91.1C73E4E3@icrf.icnet.uk> <375C16A9.5ADC4990@icrf.icnet.uk>
Message-ID: <375C1FD1.A85565C@aw.sgi.com>

Arne Mueller wrote: 
> Im running python 1.5.2b2 on SGI power challange irix64 version 6.5.
> Everytime a child process is created via fork(), it realy does it's job
> but never dies, instead the number of zombie processes (as reported by
> the 'top' program) steadily increases, though I call the sys.exit(0)
> method for the child. Any idea how exit a process without creating a
> zombie?
> 
>         thanks alot,
> 
>         Arne

Yes.  The parent process must execute a os.wait() or os.waitpid() in
order for the child process's return code to be sent back to the parent
program.  Only after that does the child process (now a zombie process)
actually get eliminated.

Hope this helps.

-- 
Dr. Gary Herron <gherron at aw.sgi.com>
206-287-5616
Alias | Wavefront
1218 3rd Ave, Suite 800, Seattle WA 98101



From tim_one at email.msn.com  Fri Jun 25 02:38:10 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 25 Jun 1999 02:38:10 -0400
Subject: class static vs. module global
In-Reply-To: <aahzFDuB8t.E6t@netcom.com>
Message-ID: <000b01bebed5$4a89aa00$d29e2299@tim>

[Aahz Maruch]
> I have a module aC, which conveniently holds a class aC.  During some
> recent changes, I needed to share information across class instances.
> Primarily to localize things, I elected to store the information in the
> class.  Anyone want to share opinions on why I should instead have used
> a module global variable?  Or perhaps a different approach?

There's no prohibition, either technical or cultural, against using class
data in Python.  Thoroughly appropriate!  It's "class static" *methods* that
get you in trouble -- but that's all in the FAQ.

class-data-is-classy-ly y'rs  - tim







From suzuki611 at okisoft.co.jp  Wed Jun 16 19:25:55 1999
From: suzuki611 at okisoft.co.jp (Hisao Suzuki)
Date: 17 Jun 1999 08:25:55 +0900
Subject: fork()
References: <000901beb6ef$f73c0f60$069e2299@tim>
Message-ID: <u3k8t3ohh8.fsf@ares.sys1.ngo.okisoft.co.jp>

In article <u3d7yzn1np.fsf at ares.sys1.ngo.okisoft.co.jp>, I wrote:
| Python's __del__ is conceptually equivalent to C++'s destructor.

In article <000901beb6ef$f73c0f60$069e2299 at tim>,
"Tim Peters" <tim_one at email.msn.com> wrote:
> This is true, but Python has nothing equivalent to the C++ "delete"
> operator, and *everything* in Python acts as if it had been created by the
> C++ "new" operator. Stroustrup is only talking about C++ objects explicitly
> created by "new" but never explicitly destroyed by "delete".  C++ has
> several other ways for objects to come in and out of existence that don't
> involve "new" or "delete" (block-scope auto; static; anonymous temps in
> expressions), and rigidly defines when destructors are invoked for those
> (more rigidly than Python, which actually opens some uses for destructors in
> C++ that can't quite be done in Python).

You miss the point.

| (Otherwise we would not rely on self.close() at __del__ in
| Lib/tempfile.py!)  The above phrase `(be) never deleted' for C++
| can be safely translated as `(be) part of cycles' for Python.
| Thus, Guido's idea is quite consistent and orthodox in this
| regard.  It will be acceptable to most of the current working
| Pythoneers, including me, even if they are not fans of the Lots
| of Intolerable Stupid Parentheses ;-)

> If there's an analogy to be drawn here, it would be that since everything in
> Python acts as if created by C++ "new" and can never by explicitly
> destroyed, Stroustrup is arguing that Python should never invoke a
> destructor period.

It sounds like a theoretical nonsense.

> The contexts are so different, though, that this analogy is as silly as it
> sounds <wink>.

By your saying, object construction in Python corresponds to
`new' operation in C++, doesn't it?  When you program something
in C++, you use `delete' to destroy the object that you have
`new'ed, don't you?

And _what_ do you do to destroy the object that you have
constructed in Python?  You remove all the references to the
object by assigning another object to each variable or attribute
that refers to the object, or by `del'ing such variables, etc.,
don't you?

Certainly C++'s `auto' variable disappears automatically upon a
block exit.  It corresponds to disappearance of a (sole, in
typical case) reference from a local variable to the object upon
a function return in Python.  In typical case, the object loses
all the references and is destroyed.  Whether it is the typical
case or not is usually obvious when you program in Python.

Thus, on programmers' view, `delet'ion of a `new'ed object and
disappearance of an `auto' variable in C++ are conceptually
equivalent to removing all the references to an object in
Python.  Indeed the Python Reference Manual specifies a fairly
weaker semantics in this regard, but some (or rather most?) of
the current working Python programmers _rely_ on the current
implementation's semantics, you know.

If an object is part of a cycle, the object will never lose the
reference(s) to it.  So the phrase `(be) never deleted' for C++
can be safely translated as `(be) part of cycles' for Python.
You see?


| Further the C++ 3rd Ed. says:
|    "It is possible to design a garbage collector to invoke the
|     destructors for objects that have been specifically
|     `registered' with the collector.  However, there is no
|     standard way of `registering' objects.  Note that it is
|     always important to destroy objects in an order that ensures
|     that the destructor for one object doesn't refer to an
|     object that has been previously destroyed.  Such ordering
|     isn't easily achieved by a garbage collector without help
|     from the programmer."

> Java solved that one -- for all the good it did <wink>.

Are you referring to java.lang.ref.PhantomReference of Java2?
Otherwise you wholly miss the point.  The traditional mechanism
of finalization in Java serves as only a safety net at _best_.


| As for a reference to option (b), see, say, R. Kent Dybvig's
| ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
| which I have mentioned before in this newsgroup, though it is
| written for Scheme and actually the coming version of GNU guile
| implements guardians, i.e. `registration' servers described in
| it.  Note that the guardians may be implemented somehow in 100%

> Indeed a very general mechanism!  Combining it with sensible __del__ rules
> for non-cycles could well lead to something better than Scheme and Python
> <0.5 wink>.

The idea of guardians is not so particular to Scheme.  It might
be safe to say that Scheme is a language that happened to be
used.  Needless to say, it is important to keep _Python_ simple,
handy, and consistent.

| it.  Note that the guardians may be implemented somehow in 100%
| pure Python even if it doesn't have a garbage collector.  The
| sys.getrefcount built-in function will be used then.

> Don't think so.  Knowing the refcount doesn't tell you whether a thing is
> trash, e.g.
> 
> class A: pass
> a = A()
> a.seq = [a] * 1000000
> del a
> 
> and there's no general way to trace the objects reachable from a given
> object without a protocol for the reachable objects to cooperate in
> revealing who is directly reachable from them (it can be laboriously
> accomplished for objects of  builtin types now by exploiting deep knowledge
> of the way they happen to be implemented, but toss in an extension type and
> you're dead).

However, the programmer often knows the expected refcount of the
object or the aggregation object of several objects when it
becomes unreachable from any other (aggregation of) objects.
Usually (though not always) he/she can pass the expected
refcount to the guardian.

> In any case, implementing M&S in Python is even slower than implementing it
> in C <wink>.

Of course.

And it is our usual practice to make prototypes in Python first,
if possible, to experiment on some ideas, isn't it? :-)

--===-----========------------- Sana esprimo naskas sanan ideon.
SUZUKI Hisao            suzuki611 at okisoft.co.jp, suzuki at acm.org.



From jeffrey.polaski at west.boeing.com  Wed Jun  9 22:23:47 1999
From: jeffrey.polaski at west.boeing.com (Jeff P.)
Date: Thu, 10 Jun 1999 02:23:47 GMT
Subject: How do I match the last occurence of a character?
Message-ID: <375F21B3.D23A5EA9@west.boeing.com>

I've been messing with this for a little while now, but I can't get this
to work. How do I match the last occurence of '\' in a string? I want to
remove the file name from a url (e.g. news/breakingnews/981001b.html -->
news/breakingnews/ ).

Here is the simplest version (that doesn't work). It matches the _first_
backslash.

import regex

file_in = open( 'dir.txt', 'r+' )
for line in file_in.readlines():
	print regex.search( '/', line )



Thanks for any help!

Jeff P.



From MHammond at skippinet.com.au  Mon Jun 28 20:38:27 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Tue, 29 Jun 1999 10:38:27 +1000
Subject: Pythonwin and Office 2000
References: <377648A7.9B9BE59F@prescod.net>	<7l672r$5k0$1@m2.c2.telstra-mm.net.au>	<3776D1E6.45E1AAF7@prescod.net> <7l7bbt$42u$1@m2.c2.telstra-mm.net.au> <3777C93C.F0B5E236@prescod.net>
Message-ID: <7l94ec$2kn$1@m2.c2.telstra-mm.net.au>

[This is getting too esoteric for the newsgroup.  I'll continue this on the
pycom-dev at pythonpros.com list]

Paul Prescod wrote in message <3777C93C.F0B5E236 at prescod.net>...

>Would that be the same speed as .Dispatch("..."), or ar you saying it
>would be slower? Anyhow, could we store the mapping from string names to
>UUIDs in the gencache directory so that the second time is fast?

Significantly slower.  Using Dispatch(), there are 2 indexed lookups in the
registry to locate the object.  Using a typelibrary description, there will
need to be a search down "HKCR\Typelib" - have a quick look at this sub-tree
(and the reason for the EnsureModule() params should become more obvious :-)

>
>Okay, I don't understand what stuff works with makepy and what stuff

Neither do I most of the time :-(  This often depends on the COM object
itself.

>doesn't. Let's put it this way: if we did everything dynamically the same
>stuff VBScript does dynamically would we slow down so much that makepy
>would be significantly faster again?

If I understand the question, then "No".  But again, it depends on the COM
object, and what type information it provides at runtime.  Some provide all
info both in a typelib and at runtime.  Some provide all info in only one
place or another.  Some (like Notes :-) provide half the info in half the
places!

PythonCOM does the worst job without _any_ typeinfo.  Then it guesses, and
byref or other complex parameters will be wrong.  The level of type info
available determines the likely success.

>Part of the reason I was using makepy was because I needed access to
>constants. How does JScript/VBscript handle this? Do you just have to use
>numbers?

I believe so.

>Even if makepy is 10 times faster than dynamic dispatch, if code at least
>behaved the same with and without it, then we could just try to do the

Yes.  But as I mentioned, speed is no longer the driver.  The type info and
constants are.  And there is no way we can _guarantee_ the same level of
support will be available.  Best we can do is get this working for a class
of "nice" servers (where "nice" may be defined as providing all type info
both statically (ie, type library) and as runtime type info.  Most MS
servers, and indeed most servers written with VB or ATL, will conform.


>You're suggesting a volunteer take your big, scary code and improve upon
>it? :)
Actually, most of this code did not originate from me (is Curt Hagenlocher
still out there:-)  But I have certainly hacked it beyond recognition since
then!


>Is this just Python code that needs fixing or a mix of Python and C++?
>(i.e. are the C++ primatives you provide enough to build the cool new
>Python system on top?)

Just Python.  And just in the win32com.client directory.

Mark.






From skip at mojam.com  Wed Jun 30 23:02:22 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 01 Jul 1999 03:02:22 GMT
Subject: [Q] Static type checker for Python
References: <7lee1g$sod$1@news.etri.re.kr>
Message-ID: <377ADAD7.EA105C50@mojam.com>

> Can I make static type checker for Python?
> or Is there any previous work about this?

In theory, you could do some type inference and generate some warning
messages, but since you don't declare variables to be of a particular
type in Python, it's unlikely it would be very useful.  The same
variable can point to many different types at runtime.

For example, the following code may be just what the author intended:

    tmp = 1.7 * math.sin(a)
    tmp = `tmp`
    print tmp
    tmp = len(tmp)

After the first statement you might *presume* that tmp holds a floating
point number, but there's no guarantee, since you don't know that
math.sin returns a float.  You can infer that tmp holds a string after
the second statement because its builtin to the language that ``
converts what it encloses to a string (though the timbot will probably
chime in with a case where I'm wrong about that <wink>).  After the last
statement you no longer know what tmp holds because you can't be sure
that len() returns an int.

The topic of static type declarations has consumed more than its fair
share of heat and light on the newsgroup and at Python conferences.  For
one recent example, check out:

   
http://www.foretec.com/python/workshops/1998-11/proceedings/papers/masse/masse.html
    http://www.foretec.com/python/workshops/1998-11/dd-rmasse-sum.html
    http://www.python.org/~rmasse/presentations/python-types/

You can scan the comp.lang.python archives for "static types" using the
form at

    http://www.python.org/search/

to browse around the previous discussions about this topic on the
newsgroup.

Also, it's worth taking a look at Guido's most recent "Python Futures"
talk to get an idea where he things Python might be headed:

   
http://www.foretec.com/python/workshops/1998-11/proceedings/guido/index.html

Note, however, that the above talk was given last November.  Things have
probably changed a bit since then.

<hungarian-notation-anyone-l'yrs?>

-- 
Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
518-372-5583



From phd at emerald.netskate.ru  Thu Jun 24 03:55:48 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Thu, 24 Jun 1999 11:55:48 +0400 (MSD)
Subject: tarlib.py
In-Reply-To: <376F65D2.DC2E2B36@phoenix-edv.netzservice.de>
Message-ID: <Pine.LNX.4.04.9906241153210.18268-100000@emerald.netskate.ru>

Hi!

On Tue, 22 Jun 1999, Holger Jannsen wrote:
> Sorry for asking newbie-questions again:
> How could I start extern Gnu-tar inside of python,
> or any other program outside of python?
> I think 'execfile' is not the thing...

   execfile executes (actaully, interprets) python files. To run a program,
you need the os module and its system() or exec*() functions. Please
rememeber, Mr Newbie, on UNIX exec*() call stops current program and
repplaces it with new one, where system() just runs a program, waits for
completion and returns...

> thanx,
> Holger

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From n at n.com  Mon Jun  7 12:01:17 1999
From: n at n.com (Jr. King)
Date: Mon, 7 Jun 1999 12:01:17 -0400
Subject: more python to c/c++ questions.
Message-ID: <7jgqas$6a6$1@rtpnews.raleigh.ibm.com>

Can you pass a user defined type in python to a c/c++ program.  I didn't see
this in the docs and I would assume it isn't possible.  I have the questions
commented inside my code

I snipped (=-)) this from my code...

pythond.py
class Cow:
 i = 10
 x = 0
 str = "string"

testmeth.setCallback(setN)  //fine
testmeth.testmeth(13,9)  //fine
cow = Cow()  //fine
testmeth.objpass(cow);   //problem with this and the c code.

code.cpp
static PyObject* objPass(PyObject *self, PyObject* args){
    PyObject *result = NULL;
    PyObject *temp;

    if (PyArg_ParseTuple(args, "O", &temp)) {  /*this works ok*/
                                            /*But how Am I supposed to
parseTuple the arg???*/
        }
        Py_INCREF(Py_None);
        result = Py_None;
    }
    return result;
}


Do I need to make a C++ class or struct that can map directly to the python
object ?

Thanx





From aa8vb at yahoo.com  Wed Jun  2 12:31:27 1999
From: aa8vb at yahoo.com (Randall Hopper)
Date: Wed, 2 Jun 1999 12:31:27 -0400
Subject: SWIG'in python question
In-Reply-To: <37542E8B.ABEAA531@kw.igs.net>; from JB on Tue, Jun 01, 1999 at 07:00:56PM +0000
References: <37542E8B.ABEAA531@kw.igs.net>
Message-ID: <19990602123127.B280136@epa.gov>

JB:
 |I've got a .so that I built with swig for python. Everything builds fine
 |but when I import i get this message from python "_mymod.so: undefined
 |symbol: __eh_pc". I wondered if anyone had seen this before and could
 |give me a clue. I think I'm missing a library somewhere. I'm building on
 |RedHat 5.2.

Does your code in _mymod.so depend on any user code another .o/.so?  

I'll assume not.  That being the case, it's probably a symbol defined in
some system library.  Use "nm" to find out which one, then add a "-l"
reference to it on the link line (the command line where you create
_mymod.so).

Check your man page for nm for the right options to print the filename and
the symbol on one line.  For example:

    (FreeBSD:)   nm -o  /usr/lib/*.so* | grep <mysymbol>
    (IRIX:)      nm -Bo /usr/lib/*.so* | grep <mysymbol>
    (Solaris:)   nm -Rh /usr/lib/*.so* | grep <mysymbol>

Randall



From matomira at iname.com  Tue Jun 15 04:29:28 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Tue, 15 Jun 1999 10:29:28 +0200
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com>  <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com>  <375FDCA7.62FF@mailserver.hursley.ibm.com>  <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com>  <3762988E.C94AA252@iname.com>  <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu>  <3763C746.5CA8FE78@iname.com> <m3u2sby4gf.fsf@soma.andreas.org>  <87pv2zjhhl.fsf@home.ivm.de> <3764F3FE.F7CBE6BA@acm.org> <0Sgs3BA5FPZ3Ew6E@jessikat.demon.co.uk> <3765B3B8.E45F373B@spots.ab.ca> <slrn7mbse1.mb7.wtanksle@dolphin.openprojects.net>
Message-ID: <37660EE8.D010DA6E@iname.com>


William Tanksley wrote:

> My preference?  I use C at my job.  I prefer postfix (Forth) for its lack
> of punctuation.  I like ML and Prolog because they let me create my own
> punctuation (but then I don't use them much).

BTW, you can do infix stuff in Common Lisp
http://www.cs.cmu.edu/afs/cs/project/ai-repository/ai/lang/lisp/code/syntax/0.html

eg.

(defun bar (x y)
   (whatever))

(defun foo (u v x y z)
  (typical-lisp-code)
  (lets-just-copy-a-formula-from-a-book-and-use-it-here #I( u + v * bar(x,y)
^^ z))
  (blah-blah))





From matz at netlab.co.jp  Thu Jun 10 21:55:36 1999
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 11 Jun 1999 10:55:36 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
 <7j90va$6lk$1@cronkite.cc.uga.edu>
 <u3g1443iht.fsf@ares.sys1.ngo.okisoft.co.jp>
 <87pv384koy.fsf@ev.netlab.co.jp>
 <u3909slwbo.fsf@ares.sys1.ngo.okisoft.co.jp>
 <slrn7lud8t.g2c.bwinton@tor.dhs.org> <7jokdo$hih$3@cronkite.cc.uga.edu>
 <37604FA5.BEA4A11C@compaq.com>
Message-ID: <87so7z8prb.fsf@ev.netlab.co.jp>

Greg Ewing <greg.ewing at compaq.com> writes:

|Obviously this is true if the M&S can be sure of finding
|all the roots, since the refcount information is then
|redundant.

M&S without knowing all roots is not only imperfect, but also hell. It
may free living objects.  On the other hand, mark and DETECT method to
report cyclic reference would have no problem, though not amusing
really.

With proper M&S, refcount infomation appears redundant.  But M&S makes
reclamation and finalization behavior incompatible, so I doubt we can
remove refcounting from future Python.

						matz.



From amackay at starvision.com  Tue Jun 29 15:57:39 1999
From: amackay at starvision.com (Angus MacKay)
Date: Tue, 29 Jun 1999 12:57:39 -0700
Subject: clobbering
Message-ID: <37792533.95E412BD@starvision.com>

is there any way to get python to warn you when you clobber
a function refrence with a variable?

I was writing this and it vexed me for some time:

for dir in sys.path:
   dir(...)
...

it is silly user error but a flag to say warn if clobbering functions
with variables (or at lease builtin functions).

cheers, Angus.



From annis at biostat.wisc.edu  Wed Jun 23 09:50:43 1999
From: annis at biostat.wisc.edu (William Annis)
Date: 23 Jun 1999 08:50:43 -0500
Subject: interactive-logging
References: <376FF09A.8170CB@icrf.icnet.uk> <y0jso7jghez.fsf@vier.idi.ntnu.no>
Message-ID: <jt3dzjjaa4.fsf@wazor.biostat.wisc.edu>

mlh at idt.ntnu.no (Magnus L. Hetland) writes:

> Hm. Isn't there a UN*X command that does that? (That logs everything
> going on in a terminal...)
> 
> I'm damned if I can remember what it was, though...

        /bin/script

        :)
        

-- 
William Annis - System Administrator - Biomedical Computing Group
annis at biostat.wisc.edu                       PGP ID:1024/FBF64031
Mi parolas Esperanton - La Internacia Lingvo    www.esperanto.org



From rasumner at bach.wisdom.weizmann.ac.il  Tue Jun 29 09:50:19 1999
From: rasumner at bach.wisdom.weizmann.ac.il (Reuben Sumner)
Date: 29 Jun 1999 13:50:19 GMT
Subject: patch to ease reference debugging
Message-ID: <slrn7nhjna.lt6.rasumner@bach.wisdom.weizmann.ac.il>

Attached you will find a patch designed to make debugging references in
Python a bit easier.

Python 1.5.2 comes with the following flags to enable certain debugging
features
	Py_REF_DEBUG keeps a global counter of the total number of references
	in the system.  Very low overhead, but useful for detecting leaks
	due to circular references or incorrect C modules
	
	Py_TRACE_REFS implies Py_REF_DEBUG and keeps more track of things.
	I haven't really used it so I can't tell you what it does.
	
	Py_DEBUG implies both of the above and other paranoid checks meant
	for debugging the interpreter itself.
	
	COUNT_ALLOCS keeps track of how many object of a certain type have
	been created (not number of references created).
	
Unfortunately all of these result in the compiler spitting out debug
information unconditionally, and in the case of Py_TRACE_REF asking an
interactive question after running all scripts.  The included patch
simply makes most of this behavior dependent on a new command line argument
or environment variable.

These defines should be put in config.h.  It is important that all of
python and all binary modules be compiled with the same flags.  With
Py_TRACE_REFS this is absolutely critical.

We have a new environment variable PYTHONREFDEBUG.  Multiple values are
possible:
	0 output is like a non-debug interpreter
	1 print total number of references at end of execution or after an
	  interactive step
	2 If Py_TRACE_REFS ask before printing "left references"
	3 If Py_TRACE_REFS don't even ask, just print "left references"
	
On the command line you just say "-r" multiple times, each time increasing
the value by 1.

The final feature is the sys.getrefcount(obj) which returns the total number
of references to obj can now be caled as sys.getrefcount() in which case
it returns the grand total number of references.

I will put any updates on my homepage at
http://www.wisdom.weizmann.ac.il/~rasumner/


Reuben


diff -ur Python-1.5.2/Include/pydebug.h pydbg/Include/pydebug.h
--- Python-1.5.2/Include/pydebug.h	Fri Dec  4 20:48:15 1998
+++ pydbg/Include/pydebug.h	Fri Jun 25 18:20:03 1999
@@ -36,6 +36,7 @@
 #endif
 
 extern DL_IMPORT(int) Py_DebugFlag;
+extern DL_IMPORT(int) Py_RefDebug;
 extern DL_IMPORT(int) Py_VerboseFlag;
 extern DL_IMPORT(int) Py_InteractiveFlag;
 extern DL_IMPORT(int) Py_OptimizeFlag;
diff -ur Python-1.5.2/Modules/main.c pydbg/Modules/main.c
--- Python-1.5.2/Modules/main.c	Tue Feb  9 20:36:51 1999
+++ pydbg/Modules/main.c	Fri Jun 25 18:20:09 1999
@@ -71,8 +71,12 @@
 -O     : optimize generated bytecode (a tad; also PYTHONOPTIMIZE=x)\n\
 -OO    : remove doc-strings in addition to the -O optimizations\n\
 -S     : don't imply 'import site' on initialization\n\
--t     : issue warnings about inconsistent tab usage (-tt: issue errors)\n\
-";
+-t     : issue warnings about inconsistent tab usage (-tt: issue errors)\n"
+#ifdef Py_REF_DEBUG
+"-r     : debug reference counting\n"
+#endif
+;
+
 static char *usage_mid = "\
 -u     : unbuffered binary stdout and stderr (also PYTHONUNBUFFERED=x)\n\
 -v     : verbose (trace import statements) (also PYTHONVERBOSE=x)\n\
@@ -119,7 +123,11 @@
 	if ((p = getenv("PYTHONUNBUFFERED")) && *p != '\0')
 		unbuffered = 1;
 
+#ifdef Py_REF_DEBUG	
+	while ((c = getopt(argc, argv, "c:diOStuvxXr")) != EOF) {
+#else
 	while ((c = getopt(argc, argv, "c:diOStuvxX")) != EOF) {
+#endif
 		if (c == 'c') {
 			/* -c is the last option; following arguments
 			   that look like options are left for the
@@ -170,6 +178,9 @@
 
 		case 'X':
 			Py_UseClassExceptionsFlag = 0;
+			break;
+		case 'r':
+			Py_RefDebug++;
 			break;
 
 		/* This space reserved for other options */
diff -ur Python-1.5.2/Python/pythonrun.c pydbg/Python/pythonrun.c
--- Python-1.5.2/Python/pythonrun.c	Wed Apr  7 21:32:51 1999
+++ pydbg/Python/pythonrun.c	Sun Jun 27 00:21:15 1999
@@ -73,6 +73,7 @@
 static void call_ll_exitfuncs Py_PROTO((void));
 
 int Py_DebugFlag; /* Needed by parser.c */
+int Py_RefDebug; /* needed by sysmodule.c and main.c */
 int Py_VerboseFlag; /* Needed by import.c */
 int Py_InteractiveFlag; /* Needed by Py_FdIsInteractive() below */
 int Py_NoSiteFlag; /* Suppress 'import site' */
@@ -119,6 +120,10 @@
 		Py_VerboseFlag = 1;
 	if ((p = getenv("PYTHONOPTIMIZE")) && *p != '\0')
 		Py_OptimizeFlag = 1;
+#ifdef Py_REF_DEBUG
+	if ((p = getenv("PYTHONREFDEBUG")) && *p != '\0')
+		Py_RefDebug = atoi(p);
+#endif
 
 	interp = PyInterpreterState_New();
 	if (interp == NULL)
@@ -214,11 +219,13 @@
 #endif
 
 #ifdef Py_REF_DEBUG
-	fprintf(stderr, "[%ld refs]\n", _Py_RefTotal);
+	if (Py_RefDebug)
+		fprintf(stderr, "[%ld refs]\n", _Py_RefTotal);
 #endif
 
 #ifdef Py_TRACE_REFS
-	if (_Py_AskYesNo("Print left references?")) {
+	if (Py_RefDebug > 2
+		|| Py_RefDebug > 1 && _Py_AskYesNo("Print left references?")) {
 		_Py_PrintReferences(stderr);
 	}
 #endif /* Py_TRACE_REFS */
@@ -471,7 +478,8 @@
 	for (;;) {
 		ret = PyRun_InteractiveOne(fp, filename);
 #ifdef Py_REF_DEBUG
-		fprintf(stderr, "[%ld refs]\n", _Py_RefTotal);
+		if (Py_RefDebug)
+			fprintf(stderr, "[%ld refs]\n", _Py_RefTotal);
 #endif
 		if (ret == E_EOF)
 			return 0;
diff -ur Python-1.5.2/Python/sysmodule.c pydbg/Python/sysmodule.c
--- Python-1.5.2/Python/sysmodule.c	Wed Jan 27 18:33:19 1999
+++ pydbg/Python/sysmodule.c	Sun Jun 27 00:24:47 1999
@@ -225,17 +225,36 @@
 	PyObject *self;
 	PyObject *args;
 {
+#ifdef Py_REF_DEBUG
+	PyObject *arg = NULL;
+	if (!PyArg_ParseTuple(args, "|O", &arg))
+		return NULL;
+	if (arg == NULL)
+		return PyInt_FromLong(_Py_RefTotal);
+	else
+		return PyInt_FromLong((long) arg->ob_refcnt);
+#else
 	PyObject *arg;
-	if (!PyArg_Parse(args, "O", &arg))
+	if (!PyArg_ParseTuple(args, "O", &arg))
 		return NULL;
 	return PyInt_FromLong((long) arg->ob_refcnt);
+#endif
 }
 
 static char getrefcount_doc[] =
+#ifdef Py_REF_DEBUG
+"getrefcount([object]) -> integer\n\
+\n\
+Return the current reference count for the object.  This includes the\n\
+temporary reference in the argument list, so it is at least 2.\n\
+\n\
+If no object is supplied, returns the total number of references";
+#else
 "getrefcount(object) -> integer\n\
 \n\
 Return the current reference count for the object.  This includes the\n\
 temporary reference in the argument list, so it is at least 2.";
+#endif
 
 #ifdef COUNT_ALLOCS
 static PyObject *
@@ -273,7 +292,7 @@
 #ifdef Py_TRACE_REFS
 	{"getobjects",	_Py_GetObjects, 1},
 #endif
-	{"getrefcount",	sys_getrefcount, 0, getrefcount_doc},
+	{"getrefcount",	sys_getrefcount, 1, getrefcount_doc},
 #ifdef USE_MALLOPT
 	{"mdebug",	sys_mdebug, 0},
 #endif



From gmcm at hypernet.com  Wed Jun  2 18:05:01 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 2 Jun 1999 17:05:01 -0500
Subject: metaclasses and setting __class__
In-Reply-To: <19990602133443.A3846@chronis.pobox.com>
Message-ID: <1283761142-37459855@hypernet.com>

scott cotton writes:

> Last night i was playing with meta classes, and found out
> that we are not able to assign an instance's __class__
> attribute to a metaclass instance.  was just wondering if
> this is probably always going to be the case, or if there
> are any plans to make .__class__ assignable to such beasts.

It wouldn't do any good. By the time you have an instance, it's too 
late to add metaclass behaviors. All the magic happens at the time 
the "class MyClass(MyMetaClassInstance):" is encountered. At this 
point, MyMetaClassInstance can do all kinds of evil things to the way 
MyClass works.

Assigning to an instance's __class__ will alter the instance's 
behavior and appearance, but not to the same extent that using a 
metaclass will. 

It's-like-changing-your-name-to-"Crazy"
 -instead-of-being-born-that-way-ly y'rs

- Gordon



From tim_one at email.msn.com  Tue Jun  8 11:56:03 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 8 Jun 1999 11:56:03 -0400
Subject: Indentation and grouping proposal
In-Reply-To: <375BCAD1.BE6E391B@erols.com>
Message-ID: <000701beb1c7$6983b4c0$559e2299@tim>

[Tim, lying thru his teeth again, claims ...]
>> [R Wentworth, suggests "_(" and ")_" as alternative grouping constructs]
>> Note that this won't fly, since "_(" already has a meaning in
>> the grammar (call the function whose name is _).

[R Wentworth, the offended party, retaliates]
> Actually, the suggested tokens were _{ and }_, with braces rather than
> parentheses.  Admittedly braces and parenthesis can be hard to distinguish
> on a computer screen.  I'm well aware that "_(" is syntactically
> legal and meaningful, and therefore not a candidate.

Blame it on old eyes and a terrible email font -- both on my end.  OK, it's
"_{" and "}_" that will never make it in <wink>.

[notes that pindent.py & one of Donald's old patches ...]

> [do] not allow for 1-line programs.

Let's skip to:

> So, I guess having this "blessed" is unlikely. :(

Exceedingly, yes.  That was the point in trotting out (just a few of the)
previous attempts -- not that their details are in any way superior to your
proposal, but that this kind of suggestion is as old as Python and has never
made progress.

> (I do suspect that the "embedding" issue is more important now than it
> was five years ago, what with the rise of HTML templating systems and
> the like.  To my mind that makes the issue somewhat less frivolous.  Oh,
> well.)

Christian (Tismer) once invented a cute scheme, sticking a number of periods
at the start of each stmt equal to the stmt's intended indent level.  This
preserved his Python programs on a server that stripped leading whitespace,
at the cost of a trivial preprocessing step.  pyindent can also be used in a
similar fashion.  IIRC,

[
>> IIRC,

> "IIRC" ?

If I Remember/Recall Correctly
]

that flavor of server later stopped destroying whitespace, rendering the
scheme moot.  The Perlish urge to write a bunch of stuff on one line anyway
can be resisted, with practice <wink>.  And if codegen is properly
abstracted at the start (via open_block() & close_block() calls, etc),
generating appropriate whitespace is really no harder than generating
visible bracketing glyphs.

Granted that some people don't-- and never will --*like* whitespace, the
notion that there's a killer *technical* argument (on either side) is pretty
much hogwash.  Python's decision was inherited from ABC, where usability
studies with newbies concluded that grouping via indentation was easier to
learn and less likely to confuse.  That's where the colon after
block-opening stmts comes from too, despite that the grammar almost never
really *needs* it.

>> Steven Majewski christened the alternatives "frowning Guidos", in
>> honor of everyone's favorite some 5 years ago:
>>
>>     if whatever :(
>>     print "no whitespace!" ):

> Plausible, if inauspicious.

More than plausible, I believe it got fully implemented.  Wasn't able to
find the patches, though.

although-did-not-suffer-a-coronary-occlusion-from-trying-ly y'rs  - tim





From hniksic at srce.hr  Sun Jun  6 16:09:40 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 06 Jun 1999 22:09:40 +0200
Subject: Deleting characters from a string
References: <87n1ye3g4v.fsf@pc-hrvoje.srce.hr> <375AD334.6E45A6C3@mojam.com>
Message-ID: <87ogitqeez.fsf@pc-hrvoje.srce.hr>

Skip Montanaro <skip at mojam.com> writes:

> > What I would really like to be able to do is write:
> > 
> >   host = string.translate(host, None, ' "')
> > 
> > Am I missing a more elegant solution?
> 
> Probably less elegant, but you can use string.replace.  You'll have to
> call it twice to get the spaces and the double quotes:
> 
>     host = string.replace(string.replace(host, '"', ""), " ", "")

The problem with this solution is that it requires calling
string.replace() as many times as there are characters to be deleted.
This would not be a problem if not for the fact that string.replace()
creates a new string every time around.  So for many characters
string.translate() should be much more efficient.



From rozzin at geekspace.com  Thu Jun 24 15:32:21 1999
From: rozzin at geekspace.com (Joshua Rosen)
Date: Thu, 24 Jun 1999 15:32:21 -0400
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0rr2.2fi.behrends@allegro.cse.msu.edu>  <8L_b3.1213$4s.136166@typhoon-sf.snfc21.pbi.net> <slrn7n3gi0.44d.behrends@allegro.cse.msu.edu>
Message-ID: <377287C5.4E096380@geekspace.com>

Reimer Behrends wrote:
> To remind you, pretty much every single OO language other
> than Python does not require you to add a self; yet at the same time
> they have per-method and per-object data. It's a matter of how you
> declare it.

	Of course, Python doesn't have pre-assignment declarations of variables....

		-Rozzin.




From fandino at usa.net  Tue Jun  8 05:43:26 1999
From: fandino at usa.net (Salvador =?iso-8859-1?Q?Fandi=F1o?=)
Date: Tue, 08 Jun 1999 11:43:26 +0200
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net> <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net> <7j34l3$hbj$1@nnrp1.deja.com> <7j66b7$il5$2@cronkite.cc.uga.edu> <3756C693.2C910B29@usa.net> <7j6jnl$58f$3@cronkite.cc.uga.edu> <3757A820.CE3F195F@usa.net> <7j8obc$55v$2@cronkite.cc.uga.edu> <375BEDF8.DD14C4E6@usa.net> <7jgs5g$q55$1@cronkite.cc.uga.edu>
Message-ID: <375CE5BD.3E865EF4@usa.net>

Graham Matthews wrote:

> I really don't understand this. A lot of languages (eg. Java) have
> garbage collection and finalisers. All these languages manage to delete
> objects in an order that does not cause any problems (all finalisers etc
> work).
I have been using Java since it was invented and I haven't programed a
finaliser ever. They should run under so unknown and inconsistent
environment that they are not usable at all (except for freeing external
resources).

> If a collector can do that in these languages why can't it do in
> Python (__del__ methods are Python's finalisers)?
Python __del__ methods are always called under consistent environments
so they can do really interesting thinks. I know a lot of cases where
you need complex __del__ methods: file buffers, persistent objects, etc.
.

> I am very very confused by your question. You seem to have a misconception
> about __del__ methods. They are finalisers, used only to close resources
> like files, etc. In a garbage collection scheme the __del__ methods are
> not used to implement the actual collection.
You are describing the usual java finaliser. Under Python or C++,
finalisers can do (and they usually do) better thinks than just freeing
resources.

BTW, how could be acceptable that the same __del__ method could be
called from two totally different schemes (GC reclaiming objects or
reference count becoming 0)? it would break all the library classes that
now use __del__ .

 - Salva.



From mgutierrez at dss.es  Thu Jun 17 04:05:08 1999
From: mgutierrez at dss.es (mgutierrez at dss.es)
Date: Thu, 17 Jun 1999 08:05:08 GMT
Subject: Help me with Java doc programm in awk!
Message-ID: <7kaa7i$d8f$1@nnrp1.deja.com>

I'm rewriting some Java classes into a more structured and decent form.
Part of this rewriting is an exhaustive/advanced documentation ( that
lacks now). I'm a great fan of python/awk used for documentation
purposes.

This is what I'm doing :
/***************
 * Name:
 * ControlPanel
 * <p>
 * Taxonomy
 * Control Process Gui
 *
....
 */
import StateMachine;

This is done for javadoc.
What I want to is an awk and python version of this:
- For every class/function documented this way, be included in a
query system, which enables the user ask using regexp for the classes.
- For every attribute in the Taxonomy field, let the query system ask
for those clases that hold that attribute.
- Parse the import lines to enrich the attributes every class/function
has.
After all this is done. Anyone could query for functionalities and
for general stuff of my classes. Learning very fast the design of
class I've done.

I like from you, suggestions and tips for doing this in awk/python.

I know in python this would be much, much easier to be done, but
awk is more common in the place I work at.

Thanks
Manolo


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From gmcm at hypernet.com  Wed Jun 16 12:25:57 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 16 Jun 1999 11:25:57 -0500
Subject: Compiled main program
In-Reply-To: <KyG93.386$dR.22261@newsfeed.avtel.net>
Message-ID: <1282571879-108998272@hypernet.com>

> [Clarence, posting a request for help managed to find the solution
> before committing himself -- but he's posting it anyway (with his
> solution) just to see what people think...]

Strangely enough,

>python test.pyc

seems to work!
 
> 
> One of the applications that I've written in Python (in fact, the
> first big one, a billing system for ISPs in operation for about two
> years now) consists of about 120 main programs, all running via CGI.
>  There is one that is particularly larger than the rest (the main
> program is about 900 lines; all the rest are less than 400) and,
> unlike the others, is used in a very 'interactive' way; that is,
> you'll have hundreds or maybe thousands of http transactions between
> your browser and it over the course of a half hour or an hour.  
> 
> Today I finally got around to something that's bugged me for a long
> time, namely, the fact that Python is compiling that program every
> one of those hundreds of times.  So I tried the obvious:
> 
> $ cat nph-pmtbatch
> #!/usr/local/bin/python
> import PmtBatchGuts
> $
> 
> This worked fine (and with a noticeable speed increase) until the
> execution path happened to intersect with one of various spots in my
> various modules that reference a name in __main__ (it's an object
> that gives access to the context of a login session).  Hoping
> against hope, I made sure nobody was looking and put this at the top
> of PmtBatchGuts.py:
> 
> # Because this is really a main, but pre-compiled, program
> __name__ = '__main__'
> 
> Didn't think it would work, and it didn't.  I figured that what I
> really wanted to do was mess with sys.modules somehow, but I don't
> know how to get a reference to the current module.  I assume that
>  sys.modules['__main__'] = <this module>
> would do it.  As distasteful as that is (do I hear retching in the
> back of the room?), I like it much more than changing
>  S = Session.Session()
> to
>  __main__.S = Session.Session()
> 
> 
> 
> Remembering that people have on occasion mentioned attributes that
> seem to be invisible unless you know they're there, I've just been
> trying random things and came up with this:
> 
> $ cat PmtBatchGuts.py
> import sys
> 
> class MakeMeMain:
>  def __init__(self):
>         sys.modules['__main__'] = sys.modules[self.__module__]
> x = MakeMeMain()
> del x
> 
> import Session
> S = Session.Session()
> ...
> 
> 
> This works, but I'm interested in the what the community thinks of
> it. TIA
> 
> -- 
> -=-=-=-=-=-=-=-=
> Clarence Gardner
> AvTel Communications
> Software Products and Services Division
> clarence at avtel.com
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list

- Gordon



From sdm7g at Virginia.EDU  Wed Jun 16 15:52:29 1999
From: sdm7g at Virginia.EDU (Steven D. Majewski)
Date: Wed, 16 Jun 1999 15:52:29 -0400 (EDT)
Subject: OO (was: Why is tcl broken?)
In-Reply-To: <87lndkkmda.fsf@home.ivm.de>
References: <000801beb745$02768220$979e2299@tim>
	<871zfdp73f.fsf@home.ivm.de>
	<wklndk8ly6.fsf@ifi.uio.no>
	<873dzsma6b.fsf@home.ivm.de>
	<wk3dzs87mr.fsf@ifi.uio.no> <87lndkkmda.fsf@home.ivm.de>
Message-ID: <Pine.A32.3.90.990616155011.17815A-100000@elvis.med.Virginia.EDU>

On 16 Jun 1999, Klaus Schilling wrote:

> Lars Marius Garshol <larsga at ifi.uio.no> writes:
> 
> > * Klaus Schilling
> > | 
> > | No, in the sense `Steele merely rediscovered scheme, not bring it
> > | into being'
> > 
> > I don't understand what you mean by this. The two people who created
> > Scheme originally were Sussman and Steele. If that isn't bringing it
> > into being I don't know what would be.
> 
> No, they discovered Scheme, they did not create it.
> 
> Klaus Schilling
> 

*If* you're trying to be enigmatic, PLEASE STOP IT!
-- you're only managing to be obscure. 

( If, on the other hand, you're trying to be ignorant, then
  by all means, go right ahead! That, we can handle! ;-) 


---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---

 "IA-64 looks just about like what you would expect from a PA-RISC
  and IA-32 train wreck with a little VLIW thrown in for spice."
 * Thomas J. Merritt  <tjm at spam.codegen.com> in <news:comp.arch> *




From jhanson at hcpd.com  Sat Jun  5 02:09:14 1999
From: jhanson at hcpd.com (Jared Hanson)
Date: Sat, 5 Jun 1999 01:09:14 -0500
Subject: Python an Tcl/Tk 8.1
Message-ID: <7jaeof$7ja$1@news.rapidnet.com>

I am new to all of this.  I am running Windows 98, and I just downloaded
Python and the newest version of Tcl/Tk (8.1).  Unfortunately, it appears
that Python doesn't like this version.  Doing an import Tkinter command
prints out an error message saying that its looking for 8.0 dlls.  I was
wondering what files I need to change to point to 8.1 dlls in order to make
it work (if that would make it work at all)  Or do I just need to get a copy
of the 8.0 dlls and stick them in the same folder.  Any help you could offer
me would be appreciated, just reply to my post.

Thanks,
Jared Hanson





From marcoxa at copernico.parades.rm.cnr.it  Tue Jun 29 06:03:33 1999
From: marcoxa at copernico.parades.rm.cnr.it (Marco Antoniotti)
Date: 29 Jun 1999 12:03:33 +0200
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <37682DD3.43923243@compaq.com> <7ka2be$qja$1@nntp6.u.washington.edu> <37776FFA.2D85@mailserver.hursley.ibm.com> <lw7looxx7m.fsf@copernico.parades.rm.cnr.it> <876748mcxg.fsf@home.ivm.de> <37788BB9.ABD@mailserver.hursley.ibm.com>
Message-ID: <lw67471fyi.fsf@copernico.parades.rm.cnr.it>

Nope. *best* is the Common Lisp Macro system, with Scheme/Dylan syntax
stuff a second best.  The notion of 'upvar' in Tcl makes my head
spin. :) Finally, AFAIU, a Tcl "macro" must run as an interpreter of
the spec. A Common Lisp (Scheme) macro is compiled into regular code
by read-time expansion.

Cheers

-- 
Marco Antoniotti ===========================================
PARADES, Via San Pantaleo 66, I-00186 Rome, ITALY
tel. +39 - 06 68 10 03 17, fax. +39 - 06 68 80 79 26
http://www.parades.rm.cnr.it/~marcoxa



From bwinton at tor.dhs.org  Thu Jun 10 00:09:02 1999
From: bwinton at tor.dhs.org (Blake Winton)
Date: Thu, 10 Jun 1999 04:09:02 GMT
Subject: How do I match the last occurence of a character?
References: <375F21B3.D23A5EA9@west.boeing.com> <slrn7luc4m.g2c.bwinton@tor.dhs.org> <375F3380.1379B1CA@netscape.com>
Message-ID: <slrn7lueqq.g7p.bwinton@tor.dhs.org>

On Wed, 09 Jun 1999, Steve Halliburton <shallib at netscape.com> wrote:
>Why not just use the os module:
>>>> import os
>>>> name = 'news/breakingnews/981001b.html'
>>>> base = os.path.split(name)[0]
>>>> base
>'news/breakingnews'
>>>> file = os.basename(name)
>>>> file
>'981001b.html'

Would that work on NT, where the path separator is "/", but the
separator for the strings he wants to split is "\"?  Does it put the
drive in [0], and then the directory in [1]?  Unfortunately, I left my
NT box at the office, so I can't check and tell you.  And how would that
handle a full URL (i.e. "http://tor.dhs.org/~bwinton/index.html")?  What
about the Mac, where the separator is ":"?  Couldn't you have a basename
of "http", and a filename of "//tor.dhs.org/~bwinton/index.html"?  For
that matter, how does that handle something like "home/bwinton/"?  I'm
pretty sure my code can deal with both in a reasonable way, but I have
no idea what the os.path.split code would do, or what it would do on
different machines.

Later,
Blake.

P.S. Sorry about mailing this to you as well as replying in the group...
     I didn't realize that you were posting as well, and perhaps it's a
     point worth sharing with the group.

-- 
One Will.  One Dream.  One Truth.  One Destiny.  One Love.



From k_mcdermott at my-deja.com  Fri Jun 11 11:02:09 1999
From: k_mcdermott at my-deja.com (k_mcdermott at my-deja.com)
Date: Fri, 11 Jun 1999 15:02:09 GMT
Subject: Help: MIME Attachments
References: <7jqlc1$bk0$1@nnrp1.deja.com> <37610A23.6A90DB35@lemburg.com>
Message-ID: <7jr8da$htf$1@nnrp1.deja.com>

Thanks Marc-Andre,

> Reading the source in MimeWrite.py I'd suggest moving these lines
> up just behind the subwrite = ... line. The .startbody()
> call flushes the headers for you and also terminates the headers
> (by inserting the blank line you already noticed).
>
> Should work then, I guess.
>

It works great (good guess), Guido had posted a message almost a year
ago (which the code I wrote is a modification of) but I couldn't quite
figure it out...

If anybody wants the code, probably the only interesting bits are the
generation of calendar entries for Notes from a template file (using a
subclassed HTMLgen.TemplateDocument) and the bit for appending an
instruction file as e-mail body with a base64 encoded attachment let me
know and I'll post it (it's around 166 lines including extensive
documentation).

Thanks again

Kevin


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From wtanksle at dolphin.openprojects.net  Tue Jun 15 20:08:55 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Wed, 16 Jun 1999 00:08:55 GMT
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <3762988E.C94AA252@iname.com> <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu> <3763C746.5CA8FE78@iname.com> <m3u2sby4gf.fsf@soma.andreas.org> <87pv2zjhhl.fsf@home.ivm.de> <3764F3FE.F7CBE6BA@acm.org> <0Sgs3BA5FPZ3Ew6E@jessikat.demon.co.uk> <3765B3B8.E45F373B@spots.ab.ca> <slrn7mbse1.mb7.wtanksle@dolphin.openprojects.net> <m2wvx5b0hb.fsf@blinky.bfr.co.il>
Message-ID: <slrn7mdqon.nv7.wtanksle@dolphin.openprojects.net>

On 16 Jun 1999 00:46:56 +0300, Harvey J. Stein wrote:
>wtanksle at dolphin.openprojects.net (William Tanksley) writes:

> > On Tue, 15 Jun 1999 01:54:53 GMT, Maxwell Sayles wrote:
> > >Robin Becker wrote:

> > >> postfix, prefix, infix etc etc are isosemantic

> > >for myself and the others who might not know... can we get an example of
> > >each?
> > >and i remember someone mentioned parenthesized vs non-parenthesized...

> > Prefix (C, Lisp, Scheme):

> >   function(that(x(1)))
> > or:
> >   (do-this (do-that (x 1)))
> > (Parenthesis are not optional)

> > Postfix (Forth, Postscript):

> >   3 4 + SWAP MOD do-that
> >   rinse on   agitate  10 seconds   rinse off
> > (Parenthesis are not optional)

> > As a common thread, note that in none of the languages are parenthesis
> > optional.  In Postfix parenthesis are not optional because they have nno
> > meaning; in the other languages they're assigned arbitrary meaning.

>This isn't a completely fair example.  Postfix can get away without
>parentheses only when a) everything that's not a number is a function
>and b) the arity (# of arguments) of each function is known and fixed.

Not true -- in Forth part (a) holds, but part (b) doesn't.  The number of
arguments and number of returns can vary arbitrarily; they're put onto and
taken off of a stack.

And I don't see the relevance of restriction (a).  It seems that with or
without it we would get the same result -- postfix doesn't need
parenthesis.

Oh, for example, Forth "immediate" words.  These aren't the same as other
functions, since they execute at compile time (in all other respects they
are functions, though).

Also, HP RPN uses syntactic elements which are neither numbers nor
functions.

So therefore, neither of your conditions are necessary.

>Given such restrictions one can also drop the parentheses in prefix
>notation.

You're right here, though.

>Infix still requires parentheses even given the above restrictions.

Yes.

We see that postfix never requires parenthesis, and prefix can skip its
parenthesis under certain conditions.  Are there any conditions under
which infix can ignore parenthesis?

>Harvey J. Stein

-- 
-William "Billy" Tanksley
Utinam logica falsa tuam philosophiam totam suffodiant!
   :-: May faulty logic undermine your entire philosophy!



From gmcm at hypernet.com  Sat Jun  5 23:14:52 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 5 Jun 1999 22:14:52 -0500
Subject: Looking for for some ideas for argument processing
In-Reply-To: <7jcdhc$htd$1@nnrp1.deja.com>
Message-ID: <1283483354-54170045@hypernet.com>

Chuck <cmedcoff at my-deja.com> writes:

> I'm writing a script with is going to make heavy use of script
> arguments.  I'd like to hear some ideas on some ways to support
> this.
> 
> One idea I've had so far is to write a class which is a dictionary
> of "directives" to function objects, where a "directive" might be '-
> c', '/f', or whatever.  This seems to work well if all arguments are
> of this simple form.  However I would also like to support arguments
> like "python myscript.py /a /f /x "bip" "bop", /g"; the point here
> being that "directives" may take arguments.  This kind of messes up
> the simple loop you'd otherwise use to process arguments.

Try getopts from the standard distribution. If that doesn't fit your 
needs, there's at least one enhanced version in the contrib section 
(follow the links from python.org).

- Gordon



From n at n.com  Thu Jun  3 07:44:21 1999
From: n at n.com (Jr. King)
Date: Thu, 3 Jun 1999 07:44:21 -0400
Subject: Why won't this work???
References: <7j0rkr$1eeu$1@rtpnews.raleigh.ibm.com> <7j1qc5$3rj$1@m2.c2.telstra-mm.net.au> <7j3e5q$nrq$1@rtpnews.raleigh.ibm.com> <slrn7lbh5v.j9o.bernhard@alpha1.csd.uwm.edu>
Message-ID: <7j5pp9$68c$1@rtpnews.raleigh.ibm.com>

I don't feel I have the right to get an answer.  I would have been ok if the
question was ignored and not even looked at.  I don't need Mark questioning
my decency.  And the wording and use of an exclamation point doesn't sound
very friendly to me.  c.l.c stands comp.lang.c++.

I felt Mark's answer would be better served if he would have just said don't
post so much code.  Try to drill down to the problem.  And encode your email
in such a way it can be humanly read and not by machine.  That would have
been cool, but as you can see I will not put my email up anyway.  Most of
the time people look for email answer they get the response Post here, Read
Here, I thought that was a rule.

Lets not waste any more time with this battle of opinions.  I can say he was
wrong all I want, and you can say he wasn't.  You will not be changing my
opinion.

Happy coding.





From dieter at handshake.de  Sun Jun  6 06:45:21 1999
From: dieter at handshake.de (Dieter Maurer)
Date: 06 Jun 1999 12:45:21 +0200
Subject: intractable(?) Tkinter scrollbar problem
References: <19990601001054.36247@nms.otc.com.au>
Message-ID: <x7yahxfvzy.fsf@lindm.dm>

Hello Greg

Greg McFarlane <gregm at iname.com> writes on Mon, 31 May 1999 14:10:54 GMT:
> I include a Tkinter script below, modified from one by Mark C Favas,
> using, I believe, Fredrik Lundh's AutoScrollbar class.  When this
> script is run, the two scrollbars continuously get mapped and unmapped
> and the window continuously changes size.

I think, I now understand the instability and I should have
a solution.

What happens:

  When the grid is changed in "AutoScrollbar.set",
  Tk initiates two waves that propagate this information.

  The waves are initiated and propagated through
  successive "Tcl_DoWhenIdle(ArrangeGrid,...)"

  One wave goes upward towards the top level. Once
  the top level is reached, the window manager is told
  to adjust the toplevel size. Tk waits locally
  for a ConfigureNotify event, telling that the
  update is completed.
  Then, "Tcl_DoWhenIdle(UpdateGeometryInfo,...)" initiates
  a new wave down towards the "scrollFrame" informing
  about the new window geometry. This wave, too,
  is propagated through successive "Tcl_DoWhenIdle(ArrangeGrid,...)
  In some sense, the wave from the grid change is reflected
  at the top level and now goes back towards the scrollFrame.

  The second wave goes downward towards the grid slaves.
  When it reaches the canvas, "CanvasUpdateScrollbars"
  is called in "DisplayCanvas" which in turn calls
  "AutoScrollbar.set".

  At this time, the toplevel has already been adjusted
  and the wave informing about the change is going downward,
  but it is behind the first downward wave.
  Based on the outdated information, a new grid change
  is performed, resulting again in two waves upward
  and downward.

  Shortly later, the geometry propagation wave from
  the toplevel reaches the canvas. In "CanvasUpdateScrollbars",
  the grid structure is changed a third time,
  two new waves are initiated ....


The solution
  We must prevent the first downward wave to do bad
  things before the upward (and eventually reflected)
  wave has reached the canvas.

  The following code should do this:

class AutoScrollbar(Tkinter.Scrollbar): 
  _active= 0
  def set(self, lo, hi):
    self.lo= lo; self.hi= hi
    #if self._active: print 'active'; return
    self._active= 1; self.update_idletasks(); self._active= 0
    self._set()
  #
  def _set(self):
    lo= self.lo; hi= self.hi
    #print self.id, lo, hi
    if float(lo) <= 0.0 and float(hi) >= 1.0:
      self.tk.call("grid", "remove", self)
    else:
      self.grid()
    Tkinter.Scrollbar.set(self, lo, hi)


  The first wave downward reaches the canvas and there
  calls "AutoScrollbar.set".
  It sets "_active=1" and then waits in "update_idletasks"
  for the arrival of the reflected wave.

  When the reflected wave reaches the canvas, it provides
  new updated values for "lo" and "hi".
  It then sees the "_active == 1" and returns.

  The first wave takes over and updates the grid
  with up to date "lo" and "hi" values.


Potential risks:
  "update_idletasks" is a powerful mechanism of which the use
  may have nasty side effects.
  For example, it may result in flickering.

  However, this is probably better than the instability,
  it avoids.


- Dieter




From gmcm at hypernet.com  Thu Jun 10 09:54:21 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 10 Jun 1999 08:54:21 -0500
Subject: Why is tcl broken?
In-Reply-To: <375F82C7.CF0F0477@iname.com>
Message-ID: <1283099377-77267484@hypernet.com>

Fernando Mato Mira writes:
> 
>   I'm trying to collect a list of all the _current_ issues making
>   tcl
> a bad language choice.

Why? Convincing your boss to use Python over TCL for your next 
project is one thing, but in general this is a zero-sum (or worse) 
game.

The TCLers I know value its simplicity and accept its limitations. 
Telling them they're wrong will only piss them off.

- Gordon



From syeates at cs.waikato.ac.nz  Fri Jun 25 15:53:18 1999
From: syeates at cs.waikato.ac.nz (Stuart Yeates)
Date: 25 Jun 1999 19:53:18 GMT
Subject: GC In Python: YAS (Yet Another Summary)
References: <Pine.SUN.3.95-heb-2.07.990616155228.15761A-100000@sunset.ma.huji.ac.il> <p53dzrfcea.fsf@bidra241.bbn.hp.com> <930267199.139037@clint.waikato.ac.nz> <37730BAA.28A7C738@bioreason.com>
Message-ID: <930340397.465868@clint.waikato.ac.nz>

Andrew Dalke (dalke at bioreason.com) wrote:
| Stuart Yeates <syeates at cs.waikato.ac.nz> said:
| > Of course, if anything does pointer XORing or other crimes,
| > tools such as these are unavaliable.
| 
| Here's one which isn't a crime but would still have problems.
| One of the C libraries we use also supports FORTRAN.  To do this,
| it internally maps all pointers to integer handles, and it is
| these integers which are used in the client apps.

I'm not quite sure what you mean by 'integer handle,' but if you
do a straight cast then it shouldn't be a problem (since the 
collector operates on the binaries representation). If you mean
that it uses an index to an array of pointers (or some such), 
then that shouldn't be a problem either, since the pointer still
exists somewhere (i.e. the array). 

stuart
--    stuart yeates <s.yeates at cs.waikato.ac.nz> aka `loam'
If you're not part of the solution, then you're part of the precipitate.



From greg.ewing at compaq.com  Mon Jun 28 22:59:17 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Tue, 29 Jun 1999 14:59:17 +1200
Subject: strange behaviour at termination time
References: <377765F0.38230773@steria.fr>
Message-ID: <37783685.C7BB5B2E@compaq.com>

Fabien COUTANT wrote:
> 
> If you change the name of the first class from A to X everywhere
> then X is bound to None in B.__del__ !!!

When the Python interpreter is shutting down, it
goes through all the loaded modules and sets all
the names defined in them to None. It does this in
an arbitrary order, and it appears that in this
case it happens to delete the name "A" after
"y", but "X" before "y".

To avoid problems like this, you have to design
your __del__ methods so that they can do their
work without having to refer to any module-level
names at the time they are called. One way is
to use the default-argument trick:

class B (A) :
    def __del__ (self, A = A) :
        A.__del__ (self)

Greg



From Hank.Fay at p98.f112.n480.z2.fidonet.org  Wed Jun 30 23:47:45 1999
From: Hank.Fay at p98.f112.n480.z2.fidonet.org (Hank Fay)
Date: Thu, 01 Jul 1999 04:47:45 +0100
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
Message-ID: <00001609@bossar.com.pl>

From: "Hank Fay" <hank at prosysplus.com>

The current reference is: http://www.cfdrc.com/FOX/fox.html

Hank Fay

--
http://www.prosysplus.com
Lyle Johnson <jlj at cfdrc.com> wrote in message
news:377A24B7.AD6FAA1A at cfdrc.com...
> Just to add fuel to the fire, we now have a Python interface to FOX, a
> cross-platform GUI toolkit:
>
>     http://www.cfdrc.com/pub/FOX/fox.html
>
> The latest version of the Python wrappers can be picked up at:
>





From news at dorb.com  Thu Jun 10 14:08:54 1999
From: news at dorb.com (Darrell)
Date: Thu, 10 Jun 1999 14:08:54 -0400
Subject: Fixed; not elegant, but if interested...
References: <375FEA53.95E3D3BC@ausinfo.com> <375FF6E5.105F9B5E@ausinfo.com>
Message-ID: <CdT73.1440$h14.6666@newsr2.twcny.rr.com>

>>> strftime("%y%m%d_%H%M%S",localtime(time()))
'990610_140535'



--
--Darrell





From gmcm at hypernet.com  Wed Jun  2 10:08:18 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 2 Jun 1999 09:08:18 -0500
Subject: Python 2.0
In-Reply-To: <3754F5D7.2B6E1954@pop.vet.uu.nl>
Message-ID: <1283789750-35739043@hypernet.com>

Martijn Faassen wrote:

> Yukihiro Matsumoto wrote:
> 
> > Some languages use both.  Ref counting for usual basis, real GC for
> > full clean up.  How about adding real GC to Python 2.0, co-working
> > with ref counting?
> 
> This sounds like an interesting idea; I believe I saw references to
> garbage collecting schemes that help clean up circular references
> caused by referencing counting. I forget where, though. :)

That would probably be Ken Thompson discussing Inferno. William  
Tanksley mentioned this recently. You can read the interview at
 http://computer.org/computer/thompson.htm

> * In the complex case, garbage collecting is nice as it cleans up
> circular references. There are unwanted side-effects of its
> unpredictability, though; a possible leaking of resources (files
> which aren't closed), and some difficulties interfacing with C.

Unpredictability has been well-thrashed recently. Two posters have 
mentioned that GC doesn't *have* to be unpredictable, but I've never 
seen one that wasn't. I can attest to GC not necessarily imposing 
difficulties in writing extensions. Writing Java native methods in 
MS's RNI is very straightforward. No pointers to pointers. They use 
generational GC. In the context of a native call, nothing will be 
moved. You can "fix" an object before returning to Java, and 
otherwise control GC. Not as slick as using CXX, but easier than 
Python's C API.

- Gordon



From evan at tokenexchange.com  Fri Jun 11 01:12:35 1999
From: evan at tokenexchange.com (Evan Simpson)
Date: Fri, 11 Jun 1999 00:12:35 -0500
Subject: Newbie class questions
References: <tN13wIASNEY3EwAl@puckish.demon.co.uk>
Message-ID: <PU083.23$KR2.60595@news.corridex.com>

Toby Kelsey wrote in message ...
[example snipped]
> (1) Is there a reason why I cannot call MyClass.q()?  Why require a
> meaningless instance arg to a class-static method?  Alternatively
> shouldn't it be an error at parse time?
The answer to all three questions is this:  MyClass.__dict__['q'] is a
perfectly good function with no parameters, and you can call it.  MyClass.q
is that same function magically wrapped as an unbound method of MyClass,
which can't be called since it lacks the required arg.  You could achieve
the same effect by defining 'otherq' elsewhere, then saying 'MyClass.q =
otherq', since the magic lives in the class-attribute access, not the
function itself.

> (2) Why doesn't t.y = g1 work?  How am I meant to access instance
> attributes in g0 without an instance argument?
't.y = g1' works just fine, but it doesn't do what you expected.  Since
method-wrapping magic lives in the class-attribute access rather than the
instance-attribute access, you'll get the same unwrapped g1 out of 't.y'
that you put in, and it will want an arg, while g0 is happy with no args.
If you really want them to act as class methods, you need to 'MyClass.y =
g1'.  Then 't.y' will pass through to 'MyClass.y', then the
instance-attribute access method-binding magic will kick in.  Clear? <wink>

> (3) Some mention was made in the tutorial of the scope rules being in
> flux.  Are any impending changes likely to affect the output of this
> class?
Not that I've heard.

should-be-safe-till-2.0-then-all-bets-are-off-ly y'rs






From emile at fenx.com  Sun Jun  6 22:16:54 1999
From: emile at fenx.com (Emile van Sebille)
Date: Sun, 6 Jun 1999 19:16:54 -0700
Subject: Using CVS under Win95
References: <yoy63.1157$VL2.171084@news.direcpc.com> <m3u2slcf9y.fsf@deneb.cygnus.stuttgart.netsurf.de>
Message-ID: <00G63.1175$VL2.173122@news.direcpc.com>

Thanks all.

Setting the HOME variables did it.  I saw the CVSROOT info but not these.

--

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






From bwarsaw at cnri.reston.va.us  Wed Jun 30 11:12:56 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Wed, 30 Jun 1999 11:12:56 -0400 (EDT)
Subject: A Certifiable Very Cool Trick 
References: <14201.27231.1003.466027@anthem.cnri.reston.va.us>
	<199906300546.PAA31064@piglet.dstc.edu.au>
Message-ID: <14202.13304.411475.346417@anthem.cnri.reston.va.us>

>>>>> "DA" == David Arnold <arnold at dstc.edu.au> writes:

    DA> as i understood it, each interpreter presents a set of COM
    DA> APIs which provide an abstract view of stack frames, etc,
    DA> suitable for use by a symbolic script debugger.  you can thus
    DA> debug across scripting languages using a single tool, watching
    DA> calls between interpreters, etc etc.

Another thing that would be really cool, would be a debugging tool
that could cross the implementation language barrier.  E.g. stepping
through Python, I could then drop down into the C or Java at that
point.  Having the ability to move between even those two abstraction
levels would be pretty useful.  Really cool would be the ability to
also drop into the bytecode level (either under the Java source in the
case of JPython, or between the Python and C code in the case of
CPython).

    DA> all very cool, and so i started thinking about doing the same
    DA> thing for Unix via gdb (for C) and the Python interpreter
    DA> (what other languages are there?  oh -- ELisp, yeah -- we can
    DA> do that!)

Ah, so our dream of scripting XEmacs with Python isn't so far away...

    DA> ahhhhh.  now, where's that spare time again?

...or is it? :)  Fun to think about at least.

-Barry



From paulb at infercor.no  Tue Jun  1 05:28:53 1999
From: paulb at infercor.no (Paul Boddie)
Date: Tue, 01 Jun 1999 11:28:53 +0200
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
	 <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>
	 <87n1ykr0z9.fsf@ev.netlab.co.jp> <7ivs9q$dc5@news.dns.microsoft.com> <87k8toqv70.fsf@ev.netlab.co.jp>
Message-ID: <3753A7D5.5DAC2E4F@infercor.no>

Yukihiro Matsumoto wrote:
> 
> "Vadim Chugunov" <chega_ at yahoo.com> writes:

[...]

> |I'm sorry for being unclear, but the key words here were: "to free up
> |resources".
> |And I meant "to free up resources the moment I expect them to", not "some
> |time later".
> 
> Well, OK.  I understand now.  Now only thing I can say is, why do you
> emphasize so much on reclaiming object at the time you exepcted?

Because, and I have seen this happen in real-life situations, there's no
guarantee that as the last reference to an object is released all the resources
associated with that object are going to be released at that time. Now, if one
has some kind of object containing a reference to a socket object (for
simplicity let us say that this is a basic resource), then it means that the
socket may for an indefinite amount of time be floating around in a reserved,
but unusable state. Now, if the garbage collector doesn't get around to this
object before one instantiates more of them it is possible that one runs out of
resources: too many sockets in use.

As has been discussed before on this newsgroup/list, such resources get
overlooked in many a garbage collection scheme. I'm sure a perusal of various
news/list archives would provide a substantial amount of relevant literature.

> Because of your C++ background?

It is amusing to note that in (probably most implementations of) Java one has to
write code as one would in C++ in situations like that outlined above, using
explicit destructor-like "close" methods on all resource objects. It isn't
sufficient to let the last reference slip away through one's fingers with a
happy feeling. ;-)

> |BTW, this is a common expectation from anyone coming from a C++ background.
> |In C++, deallocation in destructors is the only way you can use exceptions
> |without leaking resources.
> 
> Yes, but it's not true for HIGHER languages like Python or Java (or Ruby).

Absolutely, as far as memory is concerned, but for resources there is a kind of
temporary "resource leak" which is critical, and about which the collector
doesn't apparently care.

Paul



From bwinton at tor.dhs.org  Thu Jun 17 22:19:37 1999
From: bwinton at tor.dhs.org (Blake Winton)
Date: Fri, 18 Jun 1999 02:19:37 GMT
Subject: ref counting, access from python
References: <7kad31$o7f$1@ns.mtu.ru> <3768C80B.14B245E6@appliedbiometrics.com> <wq7a3.223$7X1.49549@news.shore.net>
Message-ID: <slrn7mj7nm.1un.bwinton@tor.dhs.org>

On Thu, 17 Jun 1999 14:15:56 GMT, Michael P. Reilly <arcege at shore.net> wrote:
>Christian Tismer <tismer at appliedbiometrics.com> wrote:
>:>>> import sys
>:>>> sys.getrefcount(None)
>: 1583
>:>>> print "ciao - chris"
>: ciao - chris
>:>>> 
>Wow, 1583 seems a little high to me. ;)
>Anyone wonder what chris has been doing? ;)

A whole lotta nothin'?

Later,
Blake.

-- 
One Will.  One Dream.  One Truth.  One Destiny.  One Love.



From tom at spirit.gcrc.upenn.edu  Wed Jun 23 15:30:39 1999
From: tom at spirit.gcrc.upenn.edu (Tom Fenn)
Date: Wed, 23 Jun 1999 15:30:39 -0400
Subject: Trouble with DATE Variants under Pythonwin
Message-ID: <377135DF.311437@spirit.gcrc.upenn.edu>

Hello:

I'm accessing an ActiveX control under Pythonwin, and having trouble
when
the method returns a DATE variant:

>>> df.dialog.xrf.GetCreationDate()
Traceback (innermost last):
  File "<interactive input>", line 0, in ?
  File
"C:\Python\win32com\gen_py\B5562CF4-56EC-11D2-8102-00104B304896x0x1x0.py", 
	line 97, in GetCreationDate
    return self._ApplyTypes_(0x6, 1, (3, 0), (((26, 7), 0),),
'GetCreationDate', None, 	pCreationDate)
  File
"C:\Python\win32com\gen_py\B5562CF4-56EC-11D2-8102-00104B304896x0x1x0.py", 
	line 35, in _ApplyTypes_
    return self._get_good_object_(apply(self._oleobj_.InvokeTypes,
(dispid, LCID, wFlags, 	retType, argTypes) + args), user, resultCLSID)
TypeError: The VARIANT type is unknown.
>>> 

The prototype of the function call in C is : long GetCreationDate(DATE
FAR* pCreationDate)

Couldn't find anything in particular about DATE variants in Pythonwin
docs or searching the newsgroup.  Does anyone know a workaround so that
I can use this function?

Tom Fenn



From matz at netlab.co.jp  Tue Jun  1 02:41:23 1999
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 01 Jun 1999 15:41:23 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
 <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>
 <87n1ykr0z9.fsf@ev.netlab.co.jp> <7ivs9q$dc5@news.dns.microsoft.com>
Message-ID: <87k8toqv70.fsf@ev.netlab.co.jp>

Hi.

"Vadim Chugunov" <chega_ at yahoo.com> writes:

|> Remember the original statement is like this:
|>
|> >As many people have noted before, you would not be able to use
|> >destructors to free up resources like memory or file descriptors.
|>
|>which, as I understand, means destructors can not be used with real
|>GC.  That's just not true.
|
|I'm sorry for being unclear, but the key words here were: "to free up
|resources".
|And I meant "to free up resources the moment I expect them to", not "some time
|later".

Well, OK.  I understand now.  Now only thing I can say is, why do you
emphasize so much on reclaiming object at the time you exepcted?
Because of your C++ background?

|BTW, this is a common expectation from anyone coming from a C++ background.
|In C++, deallocation in destructors is the only way you can use exceptions
|without leaking resources.

Yes, but it's not true for HIGHER languages like Python or Java (or Ruby).

						matz.



From air at apex.net  Thu Jun  3 21:00:10 1999
From: air at apex.net (air at apex.net)
Date: Fri, 4 Jun 1999 01:00:10 GMT
Subject: Exmaples of re module
Message-ID: <7j78eq$s1dt@eGroups.com>

I am looking for some examples of the re module, mainly I need to be able to do what I used to do with awk, grab part of a file from Point a to point b, in awk you could do it like this
/--Begin--/,/--End--/!d
or something similar to this, any help would be appreciated.

Thanks, Todd




From evan at tokenexchange.com  Fri Jun  4 12:21:00 1999
From: evan at tokenexchange.com (Evan Simpson)
Date: Fri, 4 Jun 1999 11:21:00 -0500
Subject: "The Python Way"
References: <001101beae49$39af7bc0$949e2299@tim> <7j8pra$55v$4@cronkite.cc.uga.edu>
Message-ID: <H1T53.19$KR2.41684@news.corridex.com>

Graham Matthews [objects to the tone taken by respondents to his
suggestions, especially since he made a point of being as "Pythonic" as
possible]

Your GC-for-cycles-refcounts-for-everyday-business proposal is, indeed,
quite reasonable.  Personally, I would prefer it as a debugging tool used to
discover code which *produces* cycles rather than a standard feature of
Python, but hey, I'm not the one implementing it <wink>.

I get the distinct impression the the reception of *any* proposal to alter
Python is likely to be vastly improved by the presence of an implementation,
however primitive, and will allow you de facto last word on a lot of
arguments.  My own pet suggestions have died for lack of this, apart from
"and while", for which some kind soul coded a mutant cousin.

and-I-still-like-"and if"-ly y'rs
Evan Simpson







From thantos at brimstone.mecha  Sat Jun 26 21:22:34 1999
From: thantos at brimstone.mecha (Alexander Williams)
Date: Sun, 27 Jun 1999 01:22:34 GMT
Subject: Python for large projects?
References: <3774b5d1.690922@news.force9.net>  <T6bd3.2883$U6.12014@newsr2.twcny.rr.com>
Message-ID: <slrn7naup9.jk.thantos@brimstone.mecha>

On Sat, 26 Jun 1999 16:52:44 -0400, Darrell <news at dorb.com> wrote:
>This isn't a memory cycle but a problem still.
>Don't:
>    buf=open(file).read()
>    doTheBigJob(buf)
>        # buf won't be freed until this function exits. Which might be the
>entire run of the program.
>        # The idea is don't leave a reference at the top level
>        # Even worse is if at a lower level you want to change buf.  Try
>this and monitor memory usage.
>>>> s="        "*999999
>>>> b=s
>>>> b=b+" "
>
>Do something like this:
>   parm={}
>   parm['buf']=open(file).read()
>   doTheBigJob(parm)
>        # Now doTheBigJob can "del parm['buf']" to free the memory.
>        # Or pass the buffer with in an object of some kind.

Actually, this isn't fully necessary.  Borrowing from your first
example above, all you need is:

  buf = open(file).read()
  doTheBigJob(buf)
  del buf

No need to introduce new passing schema or wrap the buffer object in
another object at all; this will clean it up for you.

Mind you, ideally (for a number of reasons) you want to avoid opening
a file in the top level of a module /anyway/, better to embed the
opening and data-extraction in its own function which you then exit,
returning the space used by the variables therein to the pool
naturally.

If the only thing you're doing with the content of file is passing it
to doTheBigJob, then its even /better/ to simply call it as:

  doTheBigJob(open(file).read())

.... since that will only contain the content for the temporal extent
of the local its bound to within the function.

>Look out for:
>    dict1={}
>    dict1['dict']=dict1
>    # This and variations on this theme become immortal. Unless you kill it
>off carefully.

Cycles in any structure are dangerous and usually indicate a failure
of the author to understand the actual structure of the data he's
trying to model.  On the other hand, some data representations really
/do/ require recursive representations and, as such, probably bear
watching closely whether you're in a strictly reference-counting
environ or in a language implimentation with GC.

>6. No const or private.
>    You have to trust everyone.

This always struck me as a really /odd/ concern.  If you don't trust
the people on your programming team, why are they on your team?  If
you're afraid the hooks into your inner-code are visible, then rethink
why you're concerned.  Mathods/slots on Python objects can be prefixed
with _ and __ to make them harder to casually examine, but I've always
equated 'reference hiding' with 'security through obscurity,' it
doesn't protect you and it lends a false sense of hope.

-- 
Alexander Williams (thantos at gw.total-web.net)
"In the end ... Oblivion Always Wins."



From thomas.heller at ion-tof.com  Wed Jun 23 10:33:23 1999
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Wed, 23 Jun 1999 16:33:23 +0200
Subject: Freeze?
References: <7kqcj4$3bd$1@nnrp1.deja.com>
Message-ID: <002901bebd85$59221b20$4500a8c0@thomasnotebook>

> I have loaded 1.5.2 and the Win32All-125 but dont see the freeze tool
> anywhere. Searching the FTP site I only see Freeze for older versions.
> Hopefully I'm just confused but if somebody could give me a nudge I'd
> appreciate it.
> 
You need the source distribution to use freeze.
This also includes the freeze tool itself.

Thomas Heller
ION-TOF GmbH




From jfisher at are.berkeley.edu  Fri Jun 11 16:17:18 1999
From: jfisher at are.berkeley.edu (John Fisher)
Date: Fri, 11 Jun 1999 20:17:18 GMT
Subject: Python and C, arrays
Message-ID: <7jrqs5$otb$1@nnrp1.deja.com>

Hey folks,

I've run into a problem with a module I'm writing in C.  Basically, I
need to be able to convert Python tuples or lists (it's unimportant
which) into C arrays.  I know at compile-time the type of the array (and
that the tuple or list will be homogeneous, and of compatible type), but
I don't know the size.

I then need to find a way to take a C array and stuff it back into a
Python tuple or list.  I have some ideas involving using "stack-like"
lists or tuples in Python, with a "pop" function which I could call from
C, repeated in a loop to get everything.  But this seems overly complex,
and indeed like it has the potential to become truly hideous.

I would imagine these two tasks are actually quite common.  Is there
some simpler way to do either?  Dare I hope that PyArg_ParseTuple and
Py_BuildValue can do it?

Thanks,
John


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From fdrake at cnri.reston.va.us  Tue Jun 22 11:30:41 1999
From: fdrake at cnri.reston.va.us (Fred L. Drake)
Date: Tue, 22 Jun 1999 15:30:41 GMT
Subject: Creating a new type with an extension module
In-Reply-To: <QpMb3.908$7X1.231730@news.shore.net>
References: <Ctp*vl52n@news.chiark.greenend.org.uk>
	<rvy*CQ92n@news.chiark.greenend.org.uk>
	<QpMb3.908$7X1.231730@news.shore.net>
Message-ID: <14191.44065.236852.546922@weyr.cnri.reston.va.us>

Michael P. Reilly writes:
 > But the documentation on extending Python is pretty well known
 > deficiency.  Luckily it is far easier than most languages, both to
 > accomplish and to understand (yet more kudos to Guido).  Actually, in
 > responce to your earlier posting, I started writing a _SHORT_ webpage
 > explaining what is needed, which I hope to have available later today.
 > I plan (and hope) to make it into a How-To later.

  If you'd like to make it a patch to the standard documentation, I'm
sure many people would be incredible grateful, including myself.  The
time I have available to extend the documentation in all the ways that 
are requested seems a bit fleeting at the moment.
  Seriously, if anyone would like to extend the standard documents,
I'll be glad to lend a hand with the markup.  If you write the
content, I'll accept it as text if I need to!


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From tim_one at email.msn.com  Sat Jun 12 13:35:32 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 12 Jun 1999 13:35:32 -0400
Subject: regular expression question: last occurence
In-Reply-To: <19990612180242.A6075@optiplex.palga.uucp>
Message-ID: <000001beb4f9$f9937700$ce9e2299@tim>

[Gerrit Holl]
> I want to get the _last_ occurence of some string in some other
> string.

The best thing is to use string.rfind ("reverse find"):

>>> import string
>>> string.rfind("abcabcabc", "abc")
6
>>> string.rfind("abcabcabc", "xyz")
-1
>>>

Note that (like string.find, its forward-searching relative) it returns -1
if it can't find the string it's looking for.

This is faster and much more obvious than regexp tricks.

> What I want exactly, is having the current directory (for example,
> /home/gerrit/linuxgames/links/) and than put everything after the last
> "/linuxgames" in a string.

>>> have = "/home/gerrit/linuxgames/links/"
>>> want = "/linuxgames"
>>> i = string.rfind(have, want)
>>> have[i+len(want):]
'/links/'
>>>

> But when someone is in the directory
> "/home/gerrit/linuxgames/cvs/linuxgames",
> my current regexp is getting /cvs/linuxgames, and that's not what I want.

What you do want is an empty string?  Continuing the above,

>>> have = "/home/gerrit/linuxgames/cvs/linuxgames"
>>> i = string.rfind(have, want)
>>> have[i+len(want):]
''
>>>


> Now, I have:
>
> currentdir = os.getcwd()
> mo = re.search('/linuxgames', currentdir)
> eind = mo.end()
> subdir = currentdir[eind:] + '/'
>
> But that doesn't solve my problem.

Right, each part of a regexp matches at the leftmost position possible:

>>> m = re.search("a", "aaaaaaaaaaaa")
>>> m.span()
(0, 1)
>>>

The way to *trick* it is to stick ".*" at the front:

>>> m = re.search(".*a", "aaaaaaaaaaaa")
>>> m.span()
(0, 12)
>>>

First the ".*" part matches at the leftmost position possible (which is the
start of the string!).  Then the other obscure part of regexps kicks in:
each part of a regexp matches the *longest* string possible such that the
*rest* of the regexp is still able to match.  So, above, ".*" matches the
first 11 "a"s, and the "a" in the regexp matches the last "a" in the string.

This is subtle!  That's why I recommend string.rfind <wink>.  Still, you can
trick regexps into working for this:

>>> pattern = re.compile(".*/linuxgames")
>>> for current in ("/home/gerrit/linuxgames/links/",
                    "/home/gerrit/linuxgames/cvs/linuxgames"):
        mo = pattern.search(current)
        eind = mo.end()
        subdir = current[eind:] + '/'
        print current, "->", subdir

/home/gerrit/linuxgames/links/ -> /links//
/home/gerrit/linuxgames/cvs/linuxgames -> /
>>>

regexps-the-international-corrupter-of-youth-ly y'rs  - tim





From claird at Starbase.NeoSoft.COM  Tue Jun 29 11:47:12 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 29 Jun 1999 10:47:12 -0500
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <7l4nht$jl7$1@nnrp1.deja.com> <slrn7nglvv.g9.behrends@allegro.cse.msu.edu>
Message-ID: <7lapq0$d78$1@Starbase.NeoSoft.COM>

In article <slrn7nglvv.g9.behrends at allegro.cse.msu.edu>,
Reimer Behrends <behrends at cse.msu.edu> wrote:
>Andrew Cooke (andrew at andrewcooke.free-online.co.uk) wrote:
>[...]
>
>> For others, I suspect the lack of good docs for Tk is a more serious
>> problem.  There is a pretty steep learning curve involved if you're
>> meeting Tk for the first time (and if you do know it, finding the
>> right way to express it in Python can still mean reading the Tkinter
>> code).
>
>I always thought that Fredrik Lundh's introduction to Tkinter was
>excellent (and easily accessible from the Tkinter page on
>www.python.org, too). What problems do you see with it?
>
>[...]
I agree that Fredrik's writing is valuable.  Moreover,
two books on Tkinter are scheduled to appear this summer.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From Cepl at fpm.cz  Wed Jun 16 05:54:03 1999
From: Cepl at fpm.cz (Matej Cepl)
Date: Wed, 16 Jun 1999 10:54:03 +0100
Subject: Xbase files?
Message-ID: <1318D78C9072D11195C9006094EA98A72C2149@ocesrv>

Is there any module for their reading?

			matthew



From skip at mojam.com  Mon Jun 14 08:28:27 1999
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 14 Jun 1999 12:28:27 GMT
Subject: Variations on a theme by "pound bang"
References: <3763e75f.3265255@news.iol.ie> <19990613141414.D723@toast.internal>
Message-ID: <3764F5FA.A18BE1CE@mojam.com>

on the merits of "#!/usr/bin/env python" vs. "#!/usr/bin/python", jam
wrote:

> the first one is more generic, and independent of the actual location of the
> command in question (in this case, the python interpreter). 

There is one situation where the second form is preferred.  If you are
invoking a script in an environment where your normal PATH-setting
mechanisms haven't been executed and you're running with a default PATH,
you may find that "/usr/bin/env python" either finds no interpreter or
finds the wrong one.  In that case, you need to hard-code the path to
the program you want to execute.  This often happens with CGI scripts or
scripts that are executed via rsh/ssh.

-- 
Skip Montanaro	| Mojam: "Uniting the World of Music"
http://www.mojam.com/
skip at mojam.com  | Musi-Cal: http://www.musi-cal.com/
518-372-5583



From ivnowa at hvision.nl  Tue Jun 22 16:49:27 1999
From: ivnowa at hvision.nl (Hans Nowak)
Date: Tue, 22 Jun 1999 20:49:27 GMT
Subject: creating dictionnaries
In-Reply-To: <376F82A6.82497D2B@der.edf.fr>
References: <376F82A6.82497D2B@der.edf.fr>
Message-ID: <199906222048.WAA08720@axil.hvision.nl>

On 22 Jun 99, Christian Caremoli wrote:

> Hi,
> Dictionnaries can be created like that :
> d={'a':1,'b':2}
> 
> By calling a function you create a dictionnary like that :
> def f(**d):
>    return d
> 
> d=f(a=1,b=2)
> 
> I would like to be able to create dictionnaries with some similar syntax
> like keyed tuples :
> d=(a=1,b=2)
> 
> Is there a way to do that ?

Nope. :-) I wonder why you would want this? As far as I'm concerned, 
there's not screamingly much difference between

d=f(a=1,b=2)    # valid

and

d=(a=1,b=2)     # invalid


--Hans Nowak (ivnowa at hvision.nl)
Homepage: http://fly.to/zephyrfalcon



From thomas at xs4all.nl  Fri Jun  4 01:27:28 1999
From: thomas at xs4all.nl (Thomas Wouters)
Date: Fri, 4 Jun 1999 07:27:28 +0200
Subject: Changing environment variables into python variables
In-Reply-To: <7j7f0t$165$1@nnrp1.deja.com>; from sragsdale@my-deja.com on Fri, Jun 04, 1999 at 02:52:16AM +0000
References: <7j7f0t$165$1@nnrp1.deja.com>
Message-ID: <19990604072728.I22133@xs4all.nl>

On Fri, Jun 04, 1999 at 02:52:16AM +0000, sragsdale at my-deja.com wrote:

> I'm writing a script that needs to import environment variables into
> python as standard variables.  I.E. if the environment variable USER is
> set to 'soren', after my program starts up and does its thing I want
> there to be a variable named USER with a value of 'soren'.
> 
> I can write this program with little effort in perl:

spare me the whitespace-impaired comic-character-cussword-language, pleeease
:-)

> When I try to write this in python, I have this error:

> -------------------------------
> import posix
> for var in posix.environ.keys():
>     command = var+" = \""+posix.environ[var]+"\""
>     print command
>     eval(command)
> -------------------------------
> > ./importer.py
> Traceback (innermost last):
>   File "./importer.py", line 12, in ?
>     eval(command)
>   File "<string>", line 1
>     WF_VC_PLUG_INS = "/usr/local/aw/userconfig/composer/plug-ins"
>                    ^
> SyntaxError: invalid syntax

> What am I doing wrong here?  How can I, at runtime, define a variable
> and assign it a value?

You need to use exec(), not eval(). This is probably a FAQ, but i'm not sure
if it's in the FAQ. From the Python Documentation however:

eval (expression[, globals[, locals]]) 
	The arguments are a string and two optional dictionaries. The
	expression argument is parsed and evaluated as a Python expression
	(technically speaking, a condition list) using the globals and
	locals dictionaries as global and local name space. If the locals
	dictionary is omitted it defaults to the globals dictionary. If both
	dictionaries are omitted, the expression is executed in the
	environment where eval is called.  The return value is the result of
	the evaluated expression. Syntax errors are reported as exceptions.

Note the use of the word 'expression' instead of 'code' or 'statement'.
Also, below the example:

Hints: dynamic execution of statements is supported by the exec statement. 

Below eval() is the documentation on execfile, with more hints. Curiously enough, the
documentation for the exec statement seems to be missing... Anyone know if
that's intentional or oversight ?

Anyway, for completeness:

>>> list = {'knight':'ni', 'answer':'42', 'range':'african or european'}
>>> for var in list.keys():
...     command = var + ' = "' + list[var] + '"'
...     print command
...     eval(command)
... 
range = "african or european"
Traceback (innermost last):
  File "<stdin>", line 4, in ?
  File "<string>", line 1
    range = "african or european"
          ^
SyntaxError: invalid syntax
>>> for var in list.keys():
...     command = var + ' = "' + list[var] + '"'
...     print command
...     exec(command)
... 
range = "african or european"
knight = "ni"
answer = "42"

-- 
Thomas Wouters <thomas at xs4all.net>

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



From jesse at student.usyd.edu.au  Fri Jun 25 02:16:13 1999
From: jesse at student.usyd.edu.au (Jesse Sweeney)
Date: Fri, 25 Jun 1999 06:16:13 GMT
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0rr2.2fi.behrends@allegro.cse.msu.edu> <m3aetr2yk9.fsf@atrus.jesus.cam.ac.uk> <7ks38p$l4m$1@brick.cswv.com>
Message-ID: <37731810.4819561@news.usyd.edu.au>

On 23 Jun 1999 20:56:41 -0500, neelk at brick.cswv.com (Neel Krishnaswami) wrote:

[ snip -- complaints about having to write 'self' over and over ]

>I don't mind writing self.foo = bar, for the reason you describe, but
>I know I hate having to write self.__class__.member = foo in order to
>keep inheritance from breaking:
>
>class Foo:
>    x = 9
>    def munge_x(self, x):
>        Foo.x = x
>
>class Bar(Foo):
>    x = 12
     ^^^
     This creates an attribute Bar.x

>
>>>> a = Foo()
>>>> a.munge_x(15)
>>>> print a.x
>15
>
>>>> y = Bar()
>>>> y.munge_x(15)
>>>> print y.x
>12
^^^
Here, y.x evaluates to Bar.x

The problem is that a call to munge_x is munging Foo.x no matter what the self
parameter turns out to be. So, if self is an object with it's own x attribute,
whether as an instance attribute, or as a class attribute lower down in the
heirarchy, that attribute will mask the attribute of the same name in Foo (sorry
about that long sentence). ie: self.x isn't always the same as Foo.x.
    I don't really see what's unintuitive about this. You could just write:

        class Bar(Foo):
            pass

>
>But if I change Foo to:
>
>class Foo:
>    x = 9
>    def munge_x(self, x):
>        self.__class__.x = x
>
>and try the same thing, I get the more intuitive:
>
>>>> a = Foo()
>>>> a.munge_x(15)
>>>> print a.x
>15
>
>>>> y = Bar()
>>>> y.munge_x(15)
>>>> print y.x
>15

This is a different kettle of fish. Note:

>>> a = Foo()
>>> a.munge_x(15)
>>> a.x
15
>>> y = Bar()
>>> y.munge_x(20)
>>> y.x
20
>>> a.x
15

>
>It's just that writing self.__class__ is so *ugly*! Everything else
>in Python is beautiful, too, which makes this stand out even more. 

	Cheers, Jesse




From arcege at shore.net  Mon Jun 28 11:57:29 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Mon, 28 Jun 1999 15:57:29 GMT
Subject: Embedded Tcl for Python
References: <lJT73.2677$nn.808830@news.shore.net>  <3760F0A1.1DDC@mailserver.hursley.ibm.com> <B57RNPASzUY3EwGq@jessikat.demon.co.uk> <37776F4E.6113@mailserver.hursley.ibm.com>
Message-ID: <JXMd3.116$6M6.40546@news.shore.net>

Paul Duffin <pduffin at mailserver.hursley.ibm.com> wrote:
: Robin Becker wrote:
:> 
:> In article <3760F0A1.1DDC at mailserver.hursley.ibm.com>, Paul Duffin
:> <pduffin at mailserver.hursley.ibm.com> writes
:> >Michael P. Reilly wrote:
:> >>
:> >> A few weeks ago, someone asked if there was a Tcl module for Python.
:> >> People responded (including myself) that Tkinter uses a Tcl/Tk
:> >> interpreter, but you couldn't get rid of the annoying window.
:> >>
:> >> I've since written an extension ("pytcl") that creates Tcl interpreter
:> >> objects.  As part of that, I have also rewritten _tkinter to use these
:> >> new objects.
:> >>
:> >> I've had a short discussion with Guido and he's interested in the
:> >> idea.  Now I need some alpha testers; I only have access to UNIX boxes,
:> >> and very few types at that.
:> >>
:> >> The _tkinter module has the same functionality as before, including the
:> >> same threading issues.  I've include Deiter's Tcl/Tk8.1 patch but have
:> >> not had the opportunity to test that yet (right not it is just do-no-
:> >> damage).
:> >>
:> >> Contact me at arcege at shore.net if you are interested in helping me test
:> >> these modules.
:> >>
:> >
:> >I was interested in doing something similar in Tcl (e.g. creating a
:> >python interpreter object) and I was wondering if someone could give
:> >me some pointers as to how to go about doing it.
:> >
:> Paul I created a socket linked python interpreter and have used it with
:> reasonable success from Tcl. Latest code is at
:> 
:> http://www.jessikat.demon.co.uk in the file pyserver.zip
:> 
:> this may not be as efficient as havin the in core thing, but gets round
:> the problem of threading model problems etc.

: Thanks Robin, I will take a look at that, have you got any information
: on the threading model problems as I have not yet looked into Python
: threading. I must admit I was amazed that Python does not support
: being built as a shared library and can think of no practical reason why
: not as Python does support dynamically loadable extensions.

: Have any Pythoneers out there done this as it is very useful ?
: If you have any questions as to how useful it is you just need to look
: at Tkinter which could not be dynamically loadable (it is isn't it)
: without Tcl and Tk being dynamically loadable.

Python supports dynamically loadable extensions.  Simply uncomment the
"*shared*" line in Modules/Setup and recompile.  Tkinter (or more
correctly _tkinter) can be dynamically loadable (ie a shared object)
with libtcl.a and libtk.a - in fact, I would suggest it; you do not
have to set the LD_LIBRARY_PATH at runtime if compiled with the static
libraries.

You might be referring to the fact that Python does not create a shared
object of its library, libpython1.5.a.  It could be accomplished fairly
easily (I've never needed to try), but it is often better to make it a
static library.

Static libraries are created from ar(1) (yes, the parent of tar).  The
contents of libraries are the object files themselves, unaltered.  When
linking to create an executable, if a symbol is found in one object
file, then that object file is linked in.  If only one function is
referenced in an object file with one hundred functions, all the
functions are included in the executable, even if none are ever
called.

Since a shared object is a special form of an object file, it follows
the same rules.  If one function or global variable is referenced in
the shared object, then all of the shared object is included in the
executable.  Again even if functions or variables will never be
accessed, they will be included in the program.

Right now, the static library for Python 1.5.2 on AIX 4.2.1 is about
3 Mb.  If you were embedding Python with that library, you would not
be including _all_ of that 3 Mb: if you only used PyIntegerObjects,
then you would probably not link with the longobject.o, complexobject.o,
etc. (except on misbehaving systems like AIX).  But if you linked
with a dynamic library (libpython1.5.so), then the entire 3 Mb would
be linked.

Yes, there are reasons to want shared objects, but there are also
good reasons to use static libraries.

  -Arcege




From digitome at iol.ie  Mon Jun 21 15:47:10 1999
From: digitome at iol.ie (Sean Mc Grath)
Date: Mon, 21 Jun 1999 19:47:10 GMT
Subject: XML-toolkit for Windows - binaries available?
References: <376701a3.72288277@newssrv> <wkzp208udt.fsf@ifi.uio.no> <3767fc51.1547455@news.iol.ie> <wkd7yv7t5p.fsf@ifi.uio.no> <37693395.731772@news.iol.ie> <376da59b.2034245@news>
Message-ID: <376e957c.1307620@news.iol.ie>

On Mon, 21 Jun 1999 02:51:07 GMT, borkman at usa.net (Lee Borkman) wrote:

>Thanks very much, but I'm afraid I'm still unsure.
>
>I have copied pyexpat.dll and sgmlop.dll to the DLLs directory
>
>I have modified the PythonPath to include:
>C:\Program Files\Python\Lib\xml-0.5.1
>and:
>C:\Program Files\Python\Lib\xml-0.5.1\expat\bin
>	(this is where xmlparse.dll lives)
>
Ah! I think this is part of your problem. the XML package
is a python module and uses commands like:-

	from xml.parsers import pyexpat

To get this to work you need the xml package to be in a
subdirectory called xml. I renamed my xml-0.5.1 to be
xml and put in beneath my main Python directory. i,e,
c:\program files\Python\xml

The directory "c:\program files\python" is in my PythonPath variable
and so "import xml.parsers" finds what it wants in
	c:\program files\Python\xml\parsers


>Now I have to edit  __init__.py to include sgmlop and pyexpat in the
>__all__ list, is that right?  Now which __init__.py would that be?  I
>assume it's C:\Program Files\Python\Lib\xml-0.5.1\__init__.py
>

The one to edit is the one in the parsers subdirectory.





From nospam at bitbucket.com  Sun Jun 27 01:07:03 1999
From: nospam at bitbucket.com (Phil Mayes)
Date: Sat, 26 Jun 1999 22:07:03 -0700
Subject: Why exceptions shouldn't be used for flow control [Re: YAS to the   "Reading line-by-line" Problem]
References: <Pine.SUN.3.95-heb-2.07.990622183356.3454A-100000@sunset.ma.huji.ac.il> <376FDF70.4756AD38@mindspring.com> <slrn7n0chl.gh6.wtanksle@dolphin.openprojects.net> <377173F7.36BCFF6B@compaq.com> <slrn7n31k4.idj.wtanksle@dolphin.openprojects.net> <3772AED9.D9E1110@compaq.com> <slrn7na4hq.3vt.wtanksle@dolphin.openprojects.net>
Message-ID: <6kid3.780$dR.68151@newsfeed.avtel.net>

William Tanksley wrote in message ...
>On Fri, 25 Jun 1999 10:19:05 +1200, Greg Ewing wrote:
>>William Tanksley wrote:


> [long discussion of f.eof() vs. exceptions excised]

>OTOH, that particular exceptionn would get thrown only in a truly bizzare
>situation, because all code would look somethinng like this:
>
>while not f.eof():
>  line = f.readline()
>  # ...


I have sometimes had to read non-atomic objects from a stream, eg:
  while 1:
    name = f.readline()
    addr1 = f.readline()
    addr2 = f.readline()
    # combine name, addr1, addr2 into something
    # put it somewhere
Other examples: Pascal-style strings; mixed-length objects.
Doing this properly in C is a pain because of the requirement to test
error/eof after reading each fragment.  Assuming low overhead, exceptions
strike me as a significant improvement.
--
Phil Mayes    pmayes AT olivebr DOT com
Olive Branch Software - home of Arranger






From skip at mojam.com  Wed Jun  2 13:16:35 1999
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 02 Jun 1999 17:16:35 GMT
Subject: where best to define __USE_XOPEN or include <features.h>?
Message-ID: <3755677C.F384C747@mojam.com>

My Linux system (RH 5.2) is warning me about strptime when I build
Python 1.5.2.  I saw the comment in timemodule.c:

    /* extern char *strptime(); /* Enable this if it's not declared in
<time.h> */

I then took a look at <time.h> and saw:

    #ifdef __USE_XOPEN
    /* Parse S according to FORMAT and store binary time information in
TP.
       The return value is a pointer to the first unparsed character in
S.  */
    extern char *strptime __P ((__const char *__s, __const char *__fmt,
				struct tm *__tp));
    #endif

This set me looking for the definition of __USE_XOPEN, which I found in
<features.h>, guarded by #ifdef _XOPEN_SOURCE.

Short-term, I can take the easy way out and simply modify timemodule.c,
however that's likely to be forgotten.  I was hoping there was some way
already in place (perhaps via configure) that would allow me to easily
set up the environment necessary to get strptime defined.  Being able to
set this stuff from the configure command line means my settings get
remembered at the top of config.status.

Any suggestions?

-- 
Skip Montanaro	| Mojam: "Uniting the World of Music"
http://www.mojam.com/
skip at mojam.com  | Musi-Cal: http://www.musi-cal.com/
518-372-5583



From Robert.Meegan at wcom.com  Tue Jun  8 16:39:04 1999
From: Robert.Meegan at wcom.com (Robert Meegan)
Date: Tue, 8 Jun 1999 20:39:04 GMT
Subject: GC SIG
Message-ID: <Pine.A41.4.10.9906081526430.65174-100000@barney.mcit.com>

Greetings ---

Are there enough potential Sanitation Engineers to create a Garbage
Collection SIG?

This would have several benefits:

1) A crack team of dedicated professionals could quickly decide on the
optimum solution, implement it, and live forever in fame and glory.
<huzzah!>

2) The GC discussions would be moved out of the general mailing
list/newsgroup, freeing them of the constant debate. <huzzah again!>

3) When someone new shows up with a scheme for garbage collection, the
regulars on the ML/NG can bombard him or her with FTSS (Find The Stinking
SIG) messages. <quiet murmur as future satisfaction is anticipated>


Since it's nice to put your money where your mouth is, I'm willing to
serve as the SIG coordinator as long as everyone understands that:

a) I have only token knowledge of such things as compiler or interpreter
design.

b) What I know about garbage collection schemes could fill a very small
wastebasket.

If anyone even remotely more qualified is willing to take the job, I'll
*gladly* remove my hat from the ring.

Your humble servant,
--- Robert 

--------------------------------
Robert Meegan
MCIWorldCom - Cedar Rapids, Iowa
319.375.2416




From robin at jessikat.demon.co.uk  Mon Jun 28 12:46:51 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 28 Jun 1999 17:46:51 +0100
Subject: Embedded Tcl for Python
References: <lJT73.2677$nn.808830@news.shore.net>
 <3760F0A1.1DDC@mailserver.hursley.ibm.com>
 <B57RNPASzUY3EwGq@jessikat.demon.co.uk>
 <37776F4E.6113@mailserver.hursley.ibm.com>
Message-ID: <rHU7yEA7b6d3EwJI@jessikat.demon.co.uk>

In article <37776F4E.6113 at mailserver.hursley.ibm.com>, Paul Duffin
<pduffin at mailserver.hursley.ibm.com> writes
>Robin Becker wrote:
...
>> Paul I created a socket linked python interpreter and have used it with
>> reasonable success from Tcl. Latest code is at
>> 
>> http://www.jessikat.demon.co.uk in the file pyserver.zip
>> 
>> this may not be as efficient as havin the in core thing, but gets round
>> the problem of threading model problems etc.
>
>Thanks Robin, I will take a look at that, have you got any information
>on the threading model problems as I have not yet looked into Python
>threading. I must admit I was amazed that Python does not support
>being built as a shared library and can think of no practical reason why
>not as Python does support dynamically loadable extensions.
>
The python15 dll/so is usable directly by other c programs; I was just
too new to python internals to attempt the creation of a python
interpreter object for use by tcl. Using the socket interface allows the
standard python interpreter to handle all the textual commands directly
and also decouples the two script interpreters.

As I understand it python is not multithreaded out of the box. Python
has it's own form of 'threads' to avoid a large per thread data set. tcl
8.0.x isn't multi-threaded so to achieve my target of having event
driven i/o for my http server + a long running calculation (in python) I
used tclHTTPD + a socket interface to the python cpu hog. The result is
similar to having a proper multithread thing.
>Have any Pythoneers out there done this as it is very useful ?
>If you have any questions as to how useful it is you just need to look
>at Tkinter which could not be dynamically loadable (it is isn't it)
>without Tcl and Tk being dynamically loadable.
>
>

-- 
Robin Becker



From paulb at infercor.no  Fri Jun  4 05:28:36 1999
From: paulb at infercor.no (Paul Boddie)
Date: Fri, 04 Jun 1999 11:28:36 +0200
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>  <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>  <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net>  <87hfosquc0.fsf@ev.netlab.co.jp> <3753F121.E4A9BE8B@prescod.net> <87yai3ntdt.fsf@ev.netlab.co.jp> <7j3iif$134e$1@nntp6.u.washington.edu> <7j45iv$908$4@cronkite.cc.uga.edu> <7j4hhj$118u$1@nntp6.u.washington.edu> <7j4jbe$dhh$1@cronkite.cc.uga.edu> <37564A02.F115B9C5@infercor.no> <7j66f6$il5$3@cronkite.cc.uga.edu>
Message-ID: <37579C44.FE0DCC3@infercor.no>

Graham Matthews wrote:
> 
> Can I suggest you read the Java spec. As someone else has pointed out
> Java memory management is done via an interface. Hence you can implement
> any memory management scheme you like in Java.

I'm not too bothered about what *can* be done in Java. There are surely decent
enough real-time garbage collection schemes in some implementations of Java, for
example, and that would appease at least a few people, but what worries me more
is this sequence of events:

1. The suggestion that Java should bear the principal implementation of Python.
   One of the reasons given is Java's "superior" garbage collection mechanisms.
2. The expression of doubts that such garbage collection mechanisms are "good
   enough". These doubts are founded on "real world" experiences.
3. The assertion that good enough schemes exist which would not necessarily go
   against people's Python expectations, and that Java will get better over
   time (as if we haven't heard that enough in the last 4-5 years).
4. The repeated suggestion that there are "real world" problems with Java.
5. The repeated assertion that Java doesn't have to be done the way it
   frequently is.

Now we end up stuck in steps 4 and 5.

I really don't care what "hosts" Python in the future as long as most people are
sufficiently happy with the implementation, are not caught out by weird
goings-on which go against their expectations, and can at least run it on their
preferred platform. The C implementation might have some shortcomings, although
I am not that affected by them in practice, and it is right to draw attention to
them - I have seen users of certain computing platforms bury their heads in the
sand over some major problems in the systems that they use, with the result of
such overly-defensive behaviour being the severe decline in the usage of and
development for such platforms. However, given the less-than-rapturous reception
of previous "true GC in Python" work, I don't think that I am out of line in
saying that it isn't necessarily as high a priority as some contributors to this
group/list insist.

Paul



From emile at fenx.com  Sun Jun  6 13:05:11 1999
From: emile at fenx.com (Emile van Sebille)
Date: Sun, 6 Jun 1999 10:05:11 -0700
Subject: Using CVS under Win95
Message-ID: <yoy63.1157$VL2.171084@news.direcpc.com>

I've downloaded CVS and appear to have successfully compiled it using MS
C++5.0.

I'm trying to access the python source per (session paste follows):

D>cvs -d :pserver:pythoncvs at cvs.python.org:/projects/cvsroot login
(Logging in to pythoncvs at cvs.python.org)
CVS.EXE [login aborted]: could not find out home directory

D>ping cvs.python.org

Pinging sweetpea.python.org [132.151.1.85] with 32 bytes of data:

Reply from 132.151.1.85: bytes=32 time=596ms TTL=248
Reply from 132.151.1.85: bytes=32 time=634ms TTL=248
Reply from 132.151.1.85: bytes=32 time=587ms TTL=248
Reply from 132.151.1.85: bytes=32 time=605ms TTL=248

I've looked through dejanews, and didn't see/recognize anything that
applies.

Any help?

I-should-probably-be-doing-this-on-linux-ly y'rs

--

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







From news at dorb.com  Thu Jun 10 16:43:07 1999
From: news at dorb.com (Darrell)
Date: Thu, 10 Jun 1999 16:43:07 -0400
Subject: Serial programming
References: <7jp1jt$8hg$1@imsp009a.netvigator.com>
Message-ID: <cuV73.1455$h14.6642@newsr2.twcny.rr.com>

Assuming your in Windows.
http://www.python.org/windows/
Although I expect there is a better way using win32 stuff.
http://msdn.microsoft.com/isapi/msdnlib.idc?theURL=/library/sdkdoc/portals/m
ainport.htm

--
--Darrell
DORAEMON.HK <abc at def.ghi.jkl> wrote in message
news:7jp1jt$8hg$1 at imsp009a.netvigator.com...
> I am looking for some example in python about send and receive data from
> serial port.  Like send "ATZ" and receive "OK".  Any help?? Thanks a lot!!
>
>
> Doraemon
>
>
>
>
>
>





From rceci at my-deja.com  Mon Jun  7 23:06:34 1999
From: rceci at my-deja.com (rceci at my-deja.com)
Date: Tue, 08 Jun 1999 03:06:34 GMT
Subject: Question about envvars
References: <7jgl9n$p13$1@nnrp1.deja.com> <375BFE38.3FB3E83C@ix.netcom.com> <7jh7pc$as$1@nnrp1.deja.com> <375C3891.1D11E1A4@ix.netcom.com>
Message-ID: <7ji1bl$955$1@nnrp1.deja.com>

Another thought,

If I specify a -S, can I compensate by specifically saying at the
appropriate time, "import site" ?

Rob

In article <375C3891.1D11E1A4 at ix.netcom.com>,
  "Stephen J. Turner" <sjturner at ix.netcom.com> wrote:
> Hi Rob,
>
> > Boy I was excited to get a reasonable explaination so quickly, but
alas,
> > it still does not work.  Here is my test script:
> >
> > #!/usr/local/bin/python
> >
> > import MDIreg
> >
> > reg = MDIreg.FileRegistry(1,1)
> > reg.exportEnvVars()
> >
> > import os
> > print "child has it : "
> > os.system("env | grep MDI_REG_VERSION")
> > print "but do I have it? :"
> > print os.environ['MDI_REG_VERSION']
> >
> > ---
> >
> > I am interested in this "MDI_REG_VERSION" envvar which gets exported
by
> > my library.  Here is the output:
> >
> > child has it :
> > MDI_REG_VERSION=10.0
> > but do I have it? :
> > Traceback (innermost last):
> >   File "regtest.py", line 12, in ?
> >     print os.environ['MDI_REG_VERSION']
> >   File "/usr/local/lib/python1.5/UserDict.py", line 12, in
__getitem__
> >     def __getitem__(self, key): return self.data[key]
> > KeyError: MDI_REG_VERSION
>
> Oops!  I just realized that the site module, which is imported by
> default when the Python interpreter starts, imports the os module --
so
> by the time your script begins to execute, it's already too late.
Just
> for grins, you could try disabling the site module by adding the '-S'
> option to the Python command line, as in:
>
> #!/usr/local/bin/python -S
>
> I'm not advocating this approach, since it's generally a good idea to
> import the site module.  I'd just like to know if it works for you.
>
> > Any other ideas?  Any way to tell the os/posix module to "refresh"?
>
> None that I can see from looking in Modules/posixmodule.c.  Sorry.
>
> > By the way, thanks for the library implementation suggestion, but I
must
> > keep it the way it is, because other c, and c++ (at the class level)
> > programs access this library.
>
> OK, how 'bout this?  Suppose you added a virtual "putenv" method to
your
> FileRegistry C++ class whose default implementation was to call the
> global putenv function.  Then in your Python C extension, you could
> subclass FileRegistry and override the putenv method with one that
> modifies os.environ as described in the prior post.
>
> Regards,
> Stephen
>
> --
> Stephen J. Turner <sjturner at ix.netcom.com>
>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From john at mediamanager.com.sg  Tue Jun 15 05:56:27 1999
From: john at mediamanager.com.sg (John Francis)
Date: Tue, 15 Jun 1999 17:56:27 +0800
Subject: Time display without colon
Message-ID: <3766234B.3B347856@mediamanager.com.sg>

Hi,

I want to display time without colon separting them

I am using strftime(format,tuple) function.

But the display is like this HH:MM:SS

But I am trying to put it in a table form by extracting from the text
file

TIME: HH:MM:SS

My extraction of each line is based on the colon splitting the field
name and the value in the field, but my script goes heywire when I have
the above format (TIME:HH:MM:SS) is there a way to extract this or to
display TIME in different format.

Thanking you

Regards
John




From tuna at delirious.fsnnet  Thu Jun 24 19:05:32 1999
From: tuna at delirious.fsnnet (Lloyd Weehuizen)
Date: 24 Jun 1999 23:05:32 GMT
Subject: Embedding Question
References: <slrn7n4699.g36.tuna@delirious.fsnnet> <Frrc3.63$Of3.12274@news.shore.net>
Message-ID: <slrn7n5eds.5ns.tuna@delirious.fsnnet>

On Thu, 24 Jun 1999 14:40:37 GMT, Michael P. Reilly <arcege at shore.net> wrote:
>Lloyd Weehuizen <tuna at delirious.fsnnet> wrote:
>: Hi
>
>: If I use the PyRun_String method, with a string that defines a method called

>You can use any number of the "del" based operations.  But from the
>high-level (using the PyRun_* functions), you would want:

Ok, that would work, but I don't really know what functions are being
defined, as the user can edit them. Is there no way to flush everything and
start with a clean slate except doing a Finalize() and Initialize()?

Hmm I spose I could search through the strings looking for all the "def"
keyworks ;)

Also, with the PyRun_String method, what is the difference between the
locals and globals definition? I've set them to be the same, like the
PyRun_SimpleString method does, could this be causing the problem? Are
globals remembered the entire time the interpreter is running and locals
only for that PyRun_String method? Into what namespace do "def" statements
get loaded? Globals or locals? 

Thanks
Lloyd
Lloyd




From Tim.Peters at p98.f112.n480.z2.fidonet.org  Tue Jun 29 03:56:59 1999
From: Tim.Peters at p98.f112.n480.z2.fidonet.org (Tim Peters)
Date: Tue, 29 Jun 1999 08:56:59 +0100
Subject: ANN: Stackless Python 0.2
Message-ID: <00001593@bossar.com.pl>

From: "Tim Peters" <tim_one at email.msn.com>

> ...
> Py_BEGIN_ALLOW_THREADS
> ThreadedSpam();
> Py_END_ALLOW_THREADS
>
> it's-only-confusing-if-you-think-about-it-too-much<wink>-ly y'rs  - tim

[Robin Becker]
> so do I then have to poll ThreadedSpam myself to see if it's finished or
> is there a python api version of mutexes/locks etc.

You're in C now -- you do anything you need to do, depending on the
specifics of ThreadedSpam (which was presumed to be a pre-existing
thread-safe C routine).  The snippet above clearly assumes that ThreadedSpam
"is finished" when it returns from the call to it.  If your flavor of
ThreadedSpam doesn't enjoy this property, that's fine too, but then you do
whatever *it* requires you to do.  This is like asking me whether the right
answer is bigger than 5, or less than 4:  how the heck should I know <wink>?

You can certainly use Python's lock abstraction in your own C code, but it's
unlikely a pre-existing C function is using that too.

python-doesn't-restrict-what-c-code-can-do-except-to-insist-that-it-
    acquire-the-lock-before-returning-to-python-ly y'rs  - tim






From jam at newimage.com  Mon Jun  7 15:03:44 1999
From: jam at newimage.com (jam)
Date: Mon, 7 Jun 1999 15:03:44 -0400
Subject: Online/Offline Manual on Python
In-Reply-To: <928782341.193714@nurn.esoterica.pt>; from dgap on Mon, Jun 07, 1999 at 07:06:01PM +0200
References: <928782341.193714@nurn.esoterica.pt>
Message-ID: <19990607150344.F27840@toast.internal>

On Mon, Jun 07, 1999 at 07:06:01PM +0200, dgap wrote:
> 
> Hey, guys...
> 
> Anyone knows where can I get an on-line (or off-line...) manual of python??
> 
> Thanks
> 
> Paulo
> 

there is *lots* of documentation available.. check
<http://www.python.org/doc/> as a first stab.. you can print out the
postscript or pdf versions if you want to read them offline.. also, do a
search on amazon.com for 'python'.. there are at least two books I know of.
there are also lots of 'white papers' available, written by guido and others
available on the python website.. 

if you have further questions after that, please feel free to let us know.

-- 
|| visit gfd <http://quark.newimage.com/> 
|| psa member #293 <http://www.python.org/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>



From emile at fenx.com  Fri Jun 25 08:57:57 1999
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 25 Jun 1999 05:57:57 -0700
Subject: How to obtain an instance's name at runtime?
References: <7kqpu8$7u1$1@nnrp1.deja.com> <3771127C.E40F9BF3@home.com> <7kvd8b$ueb$1@nnrp1.deja.com>
Message-ID: <w1Lc3.1850$VL2.286812@news.direcpc.com>

I'm new at this, but it seems I've read that the tracestack is
available.  Might it be possible to read in the calling line number,
then parse the source code for your 'name'?

--

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





From kjwolf at online.de  Sun Jun 20 20:48:43 1999
From: kjwolf at online.de (Klaus-Juergen Wolf)
Date: Mon, 21 Jun 1999 02:48:43 +0200
Subject: Problem with Curses
References: <ppgmj7.s7r.ln@solling.seismic.de> <jpeck7.1661.ln@solling.seismic.de> <ve1dk7.1e02.ln@solling.seismic.de> <87ogidzs37.fsf@home.ivm.de>
Message-ID: <b52kk7.gj.ln@solling.seismic.de>

Klaus Schilling <Klaus.Schilling at home.ivm.de> wrote:
 > kjwolf at online.de (Klaus-Juergen Wolf) writes:
 >> It depends on when the executable is linked to libmytinfo, which only
 >> takes place on platforms that don't know the terminfo database, like
 >> BSD and Solaris; there are two competing declarations of tputs().

 > Why doesn't Freebsd know the terminfo database? Can't the one that is 
 > integrated into ncurses be used on FreeBSD? 

Some versions in the past still knew it, but, it seems, nowadays,
termcap is the only valid standard for declaring terminal interactions.
In fact, NCurses _is_ integrated into FreeBSD, the bridge between the
two systems is libmytinfo, it seems. mytinfo now declares tputs(),
which termcap also also declares, therefore, there is a small conflict,
if you don't link them libraries in the correct order.

You could, anyway, install another NCurses "on-the-head" of the old
one. But that is Linux behavior, we won't do that.

cu
 k.j.


-- 
 Support Julia Hill - save the Redwood trees: http://www.lunatree.org/



From fredrik at pythonware.com  Wed Jun 16 07:40:01 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 16 Jun 1999 11:40:01 GMT
Subject: GNU python?
References: <3765B8B4.18DE0257@spots.ab.ca> <37667309.F9BBAE9A@mojam.com>
Message-ID: <022c01beb7ec$f9617070$f29b12c2@pythonware.com>

Skip Montanaro <skip at mojam.com> wrote:
> > Looking to try out python... is there a GNU python for Win9x?  and a
> > tutorial maybe? 
> 
> Not sure just how a GNU Python would differ from the real thing.

a bit more like GNUscript, I suppose:

http://art.net/Studios/Hackers/Hopkins/Don/text/gnuscript.html

</F>




From news at dorb.com  Sat Jun  5 21:59:05 1999
From: news at dorb.com (Darrell)
Date: Sat, 5 Jun 1999 21:59:05 -0400
Subject: pyscript help
References: <BRS53.233$h14.1175@newsr2.twcny.rr.com>
Message-ID: <xEk63.805$_U4.27184@newsr1.twcny.rr.com>

Found it.

http://support.microsoft.com/support/kb/articles/q201/3/64.asp

comcat.dll version 5.0 got put on my system some how.
comcat.dll version 4.75 is what you need.

That was a painful experience !

D:\python\win32comext\axscript\client>pyscript.py
Registered: Python
Registration of Scripting Engine complete.

None of these demos work though ??
Should they ?
D:\python\win32comext\axscript\demos\client\ie\demo.htm

--Darrell







From aaron_watters at my-deja.com  Thu Jun 17 14:59:32 1999
From: aaron_watters at my-deja.com (aaron_watters at my-deja.com)
Date: Thu, 17 Jun 1999 18:59:32 GMT
Subject: Newbie: Truth values (three-valued logic)
References: <Pine.OSF.4.05.9906152037050.23333-100000@pinax.Mathematik.Uni-Bielefeld.DE> <Pine.WNT.4.04.9906151331390.296-100000@rigoletto.ski.org>
Message-ID: <7kbgid$ref$1@nnrp1.deja.com>


> Actually, after looking at the documentation for the operator module,
I
> think that there's a documentation problem -- reading it, one would
think
> that there is a __not__ special method, but it is not supported by the
> core.  I think a special comment should be added to the doc to that
> effect.

Ah but should this change?  This is not all that different
from rich comparisons!  Shouldn't

    not A<B

give the same result as

    A>=B

(at least for appropriate A, B)?  Of course then

    A>B or A==B

should also give the same result.

Hey, if we're gonna get rich comparisons, let's go all
the way.  But, as you say, you can't do it *now*.

   -- Aaron Watters http://www.chordate.com

===

no plan.



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From redhog at lysator.liu.se  Wed Jun  9 16:23:07 1999
From: redhog at lysator.liu.se (redhog at lysator.liu.se)
Date: Wed, 09 Jun 1999 22:23:07 +0200
Subject: A small module question
Message-ID: <375ECD2A.71AA9751@lysator.liu.se>

I am having a problem, which is reducible to the following, which I
would be happy if someone could explain to me:

[redhog at k35 Python]$ ls -R test/
test:
__init__.py  a.py  b.py
[redhog at k35 Python]$ cat test/__init__.py
__all__ = ["a", "b"]
[redhog at k35 Python]$ cat test/a.py
from test import b
[redhog at k35 Python]$ cat test/b.py
from test import a
[redhog at k35 Python]$ python
Python 1.5.1 (#1, Mar 21 1999, 22:49:36)  [GCC egcs-2.91.66 19990314/Li
on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> from test import a
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "test/a.py", line 1, in ?
    from test import b
  File "test/b.py", line 1, in ?
    from test import a
ImportError: cannot import name a
>>>




From brainwav at hotmail.com  Tue Jun 15 22:58:35 1999
From: brainwav at hotmail.com (Denis Lamarche)
Date: Wed, 16 Jun 1999 02:58:35 GMT
Subject: Python And Dos
Message-ID: <37673BBB.8E85D41E@hotmail.com>

Has anyone ever embeded pyhton into a dos c++?  I've been trying with
DJGPP and cant seen to do it.  I got the python source, but I think all
the files are for unix or something. I tried building the demo sources
of python and it didnt work.  Has anyone out there ever get it to work
in dos?



From digitome at iol.ie  Wed Jun 16 15:37:35 1999
From: digitome at iol.ie (Sean Mc Grath)
Date: Wed, 16 Jun 1999 19:37:35 GMT
Subject: XML-toolkit for Windows - binaries available?
References: <376701a3.72288277@newssrv> <wkzp208udt.fsf@ifi.uio.no>
Message-ID: <3767fc51.1547455@news.iol.ie>

On 16 Jun 1999 09:41:34 +0200, Lars Marius Garshol <larsga at ifi.uio.no>
wrote:

>
>* Lee Borkman
>| 
>| Can anyone help me out in installing from the standard distribution?
>| I don't have a C-compiler on my system, but I do have nmake/dmake.
>
>I think the web page is out of date, since the /xml/windows directory
>in the CVS does contain the .dll files for pyexpat and sgmlop (and has
>done for quite a while). So just download the ZIP file and they should
>be there.
>
>If not, send me an email and I'll send them to you.
>
The binaries are there but a bit of fiddling is required to
get things working. (This is from memory but here goes)

The pyexpat and sgmlop dlls have to be copied into the
xml/parsers subdirectory. You also need to edit the 
__init__.py file to include 'pyexpat' and 'sgmlop' in
the __all__ string.

Pyexpat requires xmlparse.dll. This dll needs to be
on your path.

I *think* that is what I did to get it all working:-)

Hope it helps,
Sean





From mlv at pobox.com  Thu Jun  3 00:14:32 1999
From: mlv at pobox.com (Michael L Vezie)
Date: Thu, 03 Jun 1999 00:14:32 -0400
Subject: pickle vs .pyc
In-Reply-To: <000901bead75$c7c04cc0$309e2299@tim>
References: <7j4rf5$5h0$1@mlv.mit.edu>
Message-ID: <4.1.19990603001148.00c73690@mlv.mit.edu>

At 12:01 AM 6/3/99 -0400, Tim Peters wrote:
>> This is the first I've heard of marshal.  I take it from various
>> hints here and there that the .pyc files are handled internally
>> by marshal, yes?
>
>Yes, "code objects" in particular are the one type marshal can handle that
>pickle can't.  I think we try to sell that as "a feature", though <wink>>

And code objects can't be handled by my repr/pyc method.  There is one place 
where I would have need to store a code object.

Sounds like I'm going to be spending some quality time with marshal...


>> One final asside, I've seen disagreements both online and in the
>> web docs about the spelling of marshal.
>
>Guido's spelling is the only one we'll mention here, thank you.

And 'marshalling' being with two 'l's doesn't help.


>> ...
>> marshaling-up-the-courage-to-convert-working-code-to-marshal'ly yrs
>>
>> ... (hoping-this-'ly yrs-tag-isn't-trademarked-or-only-available-
>> to-those-who've-achieved-some-level-of-pythondom'ly yrs)
>
>No particular qualifications, fees or exams are needed to use a "-ly y'rs"
>closing.  I do own exclusive worldwide rights to it, but grant everyone an
>unlimited license in perpetuity.  How can I afford to do this?  Easy:  there
>are only so many people out there who want to come across as an ass year
>after year <wink>.

Of course I need to learn how to spell "y'rs"...

Michael




From larsga at ifi.uio.no  Wed Jun 16 09:58:00 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 16 Jun 1999 15:58:00 +0200
Subject: Python Webserver
References: <3767A69D.50DC7FBF@serop.abb.se>
Message-ID: <wkaeu08cyf.fsf@ifi.uio.no>

* Thomas S. Strinnhed
| 
| I don't know if I've dreamt, but I think I saw something about a
| webserver written in Python.

There's a very simple one in the standard distribution (actually, more
than one). And then there's Medusa:

<URL: http://www.nightmare.com/medusa/>

| I thought it'd be something fun to play with.

It is. Very much so, in fact. :)

--Lars M.



From graham at sloth.math.uga.edu  Sun Jun  6 20:21:02 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 7 Jun 1999 00:21:02 GMT
Subject: "The Python Way"
References: <7j8pra$55v$4@cronkite.cc.uga.edu> <Pine.A41.4.10.9906041158070.41420-100000@barney.mcit.com>
Message-ID: <7jf39e$4t4$2@cronkite.cc.uga.edu>

Robert Meegan (Robert.Meegan at wcom.com) wrote:
: Paranoid is a little harsh, Graham. May I suggest parochial?

Yes I agree that "parochial" is a much better choice of word. Less
"aggresive" and more descriptive of what I meant. Thanks.

Robert Meegan (Robert.Meegan at wcom.com) wrote:
: My opinion (which will get you a cup of coffee, if you ever meet me and
: profess to share it) is that I would prefer to avoid syntax changes and
: keyword additions.
: 
: while 1:
: 	[...]
: 	if x:
: 		break
: 	else:
: 		[...]

To be perfectly honest I consider most arguments about syntax to be a
pointless waste of time (notice that I haven't posted in either the
'whitespace noise' thread or the 'while' thread). Language semantics
are a much more interesting kettle of fish and for my money really
"define" a language.

Robert Meegan (Robert.Meegan at wcom.com) wrote:
: The way to get significant changes incorporated into Python is to discuss
: them in this august forum, until a consenus emerges (preferably including a

The key word is "discuss". To have a "discussion" one must have participants
who don't simply want to defend "the python way". That's where that word
"parachial" comes in!
: 
graham
-- 
                    One day in a nuclear age 
                  They may understand our rage
              We build machines that we can't control
              And bury the waste in a great big hole



From gmcm at hypernet.com  Fri Jun  4 16:37:03 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 4 Jun 1999 15:37:03 -0500
Subject: ask help for problem with static linking (fwd)
In-Reply-To: <Pine.SGI.3.92.990604062558.15744A-100000@sea30>
Message-ID: <1283593623-47537420@hypernet.com>

Xiaohong Yuan wrote:

> I am trying to link my c extension module staticly with python
> interpreter. But in my c extension module, I have external variable
> declaration, which makes it fail to be linked with Python
> interpreter. There must be some way to solve this, could anybody
> help me with this?

Please forgive the lack of response to your query, but I think people 
are confused about what you are asking.

A variable declared as "extern" will have to be found in some object 
when you link. Is this variable something you expected the Python 
interpreter to expose? If so, what is it? What version of Python? 
What platform? What compiler?

- Gordon



From fatjim at home.com  Mon Jun  7 06:03:49 1999
From: fatjim at home.com (Jim Meier)
Date: Mon, 07 Jun 1999 10:03:49 GMT
Subject: CGI module misbehaving?
References: <Pine.LNX.4.04.9906071320360.5443-100000@emerald.netskate.ru>
Message-ID: <375B9974.2FFF42FA@home.com>


Oleg Broytmann wrote:

> On Mon, 7 Jun 1999, Jim Meier wrote:
> > if form.has_key["name"]:
>
> if form.has_key("name"):
>
>    has_key is a function!

(I was asked to post this to the group so it would show up in a deja search..)

On Mon, Jun 07, 1999 at 03:26:44AM -0600, Jim Meier wrote:
>
> Whoops! That'll teach me to try working this late. Thanks a lot - Now my only
> problem will be recovering from the embarrassment!
>
> -Jim
>




From arcege at shore.net  Mon Jun 21 12:08:56 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Mon, 21 Jun 1999 16:08:56 GMT
Subject: Using perl in python
References: <7jjoqh$t56$1@nnrp1.deja.com> <6cs73.2432$nn.739920@news.shore.net> <376D7F38.D1500244@pobox.com>
Message-ID: <sstb3.772$7X1.197160@news.shore.net>

Ashish Goyal <ashishg at pobox.com> wrote:
: hi,
:     try and use swig (www.swig.org) - that is an interface generator that i
: had come across some time back. Since i don't remember things clearly,  it
: might not be able to glue python and perl together (i used it to glue C and
: python code).

Correct, swig probably can't.  The Perl C API is too complex to do this
with SWIG (of course); the interface file would probably be larger than
the C code if you didn't use SWIG.  It's doable in C.  Also, unless
specially compiled, Perl will only allow one interpreter per process
(cf. -DMULTIPLICITY).  A new release of SWIG came out a week or two ago
that might make things easier, but I doubt the interface file would be
shorter.

I'm never surprised at how overly complicated Perl makes things.

  -Arcege




From greg.ewing at compaq.com  Sun Jun 27 20:24:39 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Mon, 28 Jun 1999 12:24:39 +1200
Subject: getstate, setstate and pickle
References: <X%vc3.1830$VL2.283108@news.direcpc.com> <43Ac3.1839$VL2.284144@news.direcpc.com>
Message-ID: <3776C0C7.1723CB78@compaq.com>

Emile van Sebille wrote:
> 
> The script below illustrates that dumps(r) ==dumps(loads(dumps(r))) but
> that r <> loads(dumps(r)).

If you haven't given your class a __cmp__ method,
the == will be comparing object references, which
will be different even if the two objects have an
identical structure.

To do a "deep comparison" of two instances, they
need a __cmp__ method which does the appropriate
things -- and so on recursively for any other
objects which are part of the structure.

If you really do want dumps(loads(dumps(r))) to
return a reference to the original r, then you're
out of luck -- pickle won't do that. It will
give you a deep copy of r instead.

Greg



From muse at viadoc.com  Wed Jun 16 13:27:50 1999
From: muse at viadoc.com (Tom Cloney)
Date: Wed, 16 Jun 1999 13:27:50 -0400
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com>
Message-ID: <z%Q93.4399$5E4.36952@news2.randori.com>

>   I'm trying to collect a list of all the _current_ issues making tcl
> a bad language choice...

Well, the troll aside, I do have quite a lot of experience with Tcl. Among
other things, I've written a workflow system on the Macintosh for generating
technical book catalogs using a combination of a relational database (4th
Dimension), image processing applications (Photoshop, NIH Image), page
layout and database publishing software (FrameMaker MIF) and other apps. Tcl
was used extensively as glue for doing things like process control and
analyzing Encapsulated Postscript files. The Tcl I used is the one embedded
in Pete Keleher's Alpha text editor, which is a real joy to use. I wish we
had something like it on NT. I've also used standalone versions of Tcl with
and without Tk. I worked with it a lot from about 1992 - 1996, writing many
useful apps. I've kept an eye on the language since then, and while I'm no
longer very Tclish, I reckon my basic take is still valid.

My opinions are, of course, shaped by the way I work. I lean towards
practical applications that one person can write and maintain and still have
a life. Most of the time I find myself practicing the Mad Max
post-apocalyptic school of programming, that is, cobbling together
interesting artifacts that come my way from both commercial and
non-commercial sources. I'm lazy -- I try to write only what I have to, and
then I look for the easiest tools to use to do it. I'm a businessman, so I'm
always balancing cost, productivity, and performance. I'm also aware that an
attractive solution may carry many hidden costs. I'm not going to get into
much academic hair-splitting here. For my purposes I judge language
environments by how easy it is to get things done in them, not by whether
they have a recursively bi-directed optimizing monkee gland compiler. YMMV.

Since the late 60's I've done practical programming in over 60 languages,
from FORTRAN, TECO, and microcode to Java, Python, and *Lisp, with many fun
stops in between (pause for chest-thumping and vine-swinging ;>), which
either makes me experienced or a fossil, depending on your point of view.
That said, here's my take on Tcl.


What I like:

Very readable syntax. A great language for writing more or less
self-documenting "dumb" code that you can come back 6 months or a year later
and still understand and modify. This is hard to overemphasize: clever
things _will_ be forgotten!

It's super when embedded in a text editor. Alpha is a great example.

It's a very nice hackery language. You can just whack away at things until
they do what you want. You don't have to be feeling sharp and clever on days
when you use it.

The language itself is pretty stable, that is, it doesn't usually crash in
bad ways.

It's one of the better languages for text processing. Not as nice as Perl,
but pretty darn good. Good regexp support. Lots of easy and elegant string
munging built in.

Support for lists, while not deep, is elegant and simple as far as it goes.
I don't like it when languages make list operations a big cumbersome deal,
so I appreciate Tcl's elegance here.

Support for hash tables (associative arrays, whatever...) is also elegant.

These three things: clean support for lists, hash tables, and regular
expressions are enough in themselves to make a language attractive. They
just make life much easier for a practical programmer.

GUI support via Tk is nice. It's a bit ideosyncratic, but good for many
things. It's not shrinkwrap-quality stuff, but think of all the great
languages whose inexpensive implementations are crippled by having little,
if any, support for GUI (Perl until recently), produce GUIs that feel slow
and clunky (Java), or make it cumbersome (like the joys of COM/IDL).

Wiser folk than I have said a good language should make doing simple things
simple. Tcl succeeds in this.

Good support for file system ops.

Straightforward interfaces to C etc. are a good thing.

Intangible fun factor. Tcl puts you in a good mood. It's easy. You feel like
you're cheating. You concentrate on the problem domain. You still feel good
when you look at the code six months later.


Now the cold slimy part:

Tcl scales very badly. All the same things in Tcl that make simple things
simple make complex things hard. Tcl is broad and shallow (which is not
always a bad thing). List structure doesn't really scale. There is no OOP
support to speak of, and although various people have hacked quick and dirty
examples, it's not something you would really want to build on. The very
brainlessness of Tcl is part of its appeal, but this is part of the price.
This is a _procedural_ language, which is actually nice on days when you are
too fuzzed out to think like a computer professional. It's not CLOS or
Eiffel, folks. It's not even Java.

The cute hacks that make string munging kind of cool can also be confusing.
You tend to spend a fair amount of time (by Tcl standards) finding subtle
bugs in the way you expressed complex string concatenations and such. It's
not that the language is buggy in this regard. The rules are relatively
simple. But the subtleties are hard to keep in mind. Tcl makes string
processing so easy that you tend to quickly give yourself enough rope to
hang yourself with. Easy debugging takes some of the sting out.

Tcl is slow. Oh, I know that's relative. For many applications it will be
fine. But complex parsing on large amounts of text can really drag. In one
application I needed to parse FrameMaker MIF files. These are SGML-like
ASCII files that often stretch to 600k or more, and which give FrameMaker a
lot of its power as a database publishing platform. Tcl was dog-slow. Perl
was about an order of magnitude faster than Tcl. C++ using the STL was a bit
faster than Perl, but still distressingly slow. I ended up using "cheating
C++", which is C++ with a lot of C standard library calls in it. The C was
about another order of magnitude faster than Perl. Tcl is great for nabbing
some text out of a file and doing some clever things with it (like nipping
the image size data out of an EPS file), but a cruncher it's not.

Perl can be much better for industrial-strength text processing. Perl has
better regexp handling than Tcl, in fact it's usually held up as the
standard by which other regular expression packages are judged. Perl is much
faster than Tcl. Perl 5 has rudimentary OOP support, which is a lot better
than nothing and allows Perl to scale in ways Tcl can't. The comparisons
with Perl are important because Perl does a lot of the same things Tcl does
and more. For a long time the dilemma was that Perl was the better language,
but Tcl had a GUI. Now Perl has a Tk interface (I haven't used Tk from Perl,
so I can't vouch for its elegance). After a while, as I developed more
complex applications in the languages, I found myself abandoning Tcl in
favor of Perl because of the performance and scalability. Perl code lacks
the beautiful brainless stickiness of Tcl, though. Perl is a "clever"
language, with what many would describe as "line noise" syntax. Cleverness
is dreadful when you have to look at it six months later. Tcl is the glory
of "stupid" programming, which I lean towards whenever possible. But it will
drive you crazy the minute you need speed or want to encapsulate something.

Tcl, Perl, and Python all hold out the promise of transcending the speed
problem via extensions written in C or another compiler language. This is
fine as far as it goes, but in the real world it's often difficult to fully
encapsulate speed-critical bits in separate modules (this could be the topic
of a whole thread). It's also somewhat complicated. Tcl makes it as easy as
anybody, but by the time you are organizing your work to the point where
you're designing a full component architecture, you're probably going to
find yourself thinking about using high-level compiler languages that let
you keep it all under one roof.

Tk ports are slow to arrive on some platforms and are buggy for a while. It
took years for heroic volunteers to get a workable Tk on the Mac. I think
Tcl helped show the way, but many of its enthusiasts have drifted away as
more alternatives have become available.

If you want to grind text, why not use Perl? If you want a slow, essentially
free hacking language with all the features of Tcl and very good OOP
support, why not use Python? And if you want a faster "scripting" language
based on the "Boy in the Bubble"* architecture, with built-in GUI, much
wider support, and a religion, there's always the J-word.

* Baby with the baboon heart sold separately.

-Tom






From tismer at appliedbiometrics.com  Tue Jun 29 18:08:23 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Tue, 29 Jun 1999 22:08:23 GMT
Subject: ANN: Stackless Python 0.2
References: <199906291939.PAA24426@lion.gte.com>
Message-ID: <377943D7.47D0BFFE@appliedbiometrics.com>


"Gary D. Duzan" wrote:
...

> =>The cost of a call comes from the slightly complicated
> =>function setup which is done all the time. Frames belong
> =>to the fastest available Python objects already,
> =>like, say dictionaries.
> 
>    I thought that might be the case, which is the point I was making in
> my earlier paragraph. I was just briefly pondering possibilities for
> reducing frame memory management time, if it were necessary.

I hoped it would be possible to save most of the
frame initialization, in cases where the same function
call appears in the same place again and again.
The barrier is that I would have to keep references
longer than necessary, so I'm not sure if that can
be done.

BTW, the continuations which I am just testing
are doing a great job of creating hard to release
references to frames. They are correct, but create
some cycles, quickly. This project is becoming
very time expensive already :-)

cheers - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From meowing at banet.net  Sun Jun 13 22:41:02 1999
From: meowing at banet.net (greg andruk)
Date: 13 Jun 1999 22:41:02 -0400
Subject: Created a Python module for ID3 MP3 tag manipulation
References: <yttaeu7z4o8.fsf@gilgamesh.cse.ucsc.edu> <c753dzv8tel.fsf@pc142.cosc.canterbury.ac.nz>
Message-ID: <87g13vy05d.fsf@slip-32-100-55-159.ma.us.ibm.net>

Timothy R Evans <tre17 at cosc.canterbury.ac.nz> writes:

> === from the `id3' manpage
> BUGS
>        The ID3 format has some shortcomings.  There  is  a  1  in
>        16777216  chance of your mp3 having a certain bit-pattern,
>        causing id3 to write over the last 128 bytes of  your  mp3
>        upon  tag  creation.

> This is more of a problem with the id3 `standard' than with anything
> else.  There is a small chnace that your mp3 file will contain the
> data "TAG" at just the wrong place.

The fix for this is to treat your data as an MPEG stream, not a simple
file.  If the marker appears inside the bounds of a frame, then it's
not a tag.  This does imply walking through the whole file to parse
headers, but that's a much faster operation than one might think.



From greg.ewing at compaq.com  Sun Jun 27 17:45:54 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Mon, 28 Jun 1999 09:45:54 +1200
Subject: How to obtain an instance's name at runtime?
References: <7kqpu8$7u1$1@nnrp1.deja.com> <3771127C.E40F9BF3@home.com> <7kvd8b$ueb$1@nnrp1.deja.com>
Message-ID: <37769B92.47EBE99C@compaq.com>

"Dinu C. Gherman" wrote:
> 
> Have a look at the following page, spot the question mark
> and you'll know what I mean:

I've looked at the page, and it helps, but still
doesn't provide enough information to tell exactly
what you want the "name" of the object to be.

The best solution by far would be to give the
object a name explicitly when you create it.
Is this possible, or do you want to retrofit
it onto existing code?

It looks like you want it to be something like
"the name of the first variable to be assigned
a reference to the object". Does this apply to
local variables as well as globals? Instance
variables of another object? What if the object
is never assigned to a variable? What if it
is, but somewhere far removed from where it
is created?

Greg



From tony at lsl.co.uk  Wed Jun 16 09:22:04 1999
From: tony at lsl.co.uk (Tony J Ibbs (Tibs))
Date: Wed, 16 Jun 1999 13:22:04 GMT
Subject: PPSI's "Quick Ref." and Lutz' "Pocket Ref."
In-Reply-To: Message from "Dinu C. Gherman" <gherman@my-deja.com> of Wed, 16
    Jun 1999 11:19:28 GMT <7k817q$hk8$1@nnrp1.deja.com>
References: <gherman@my-deja.com>
Message-ID: <199906161322.OAA19822@lsls4k.lsl>

"Dinu C. Gherman" <gherman at my-deja.com> wrote:
> Having never seen PPSI's "Python Quick Reference" as advertised
> on pythonpros.com on paper I wonder how it compares to Mark Lutz'
> "Python Pocket Reference"?

*Drumroll* - they're different. There, that was a surprise.

But seriously, I do think they address different sorts of minds/niches,
really. 

The PPSI quick ref is a "traditional" reference card. It is suitable 
for standing up on end on one's desk, or for losing amidst the 
paperwork on same when it falls over. It doesn't fit in my pocket
(or at least not without risk of crumpling, which I don't want to do).
It was evidently written with care, and it is a reference I find very 
useful as it is formatted in a way I find easy to look at and scan.
Compare it to the "traditional" Emacs ref card.

Lutz's pocket reference is a little book, which DOES fit in my pocket, 
and because it is a book it is rather more robust. It is, of course, 
also harder to lay down (open) on the desk without destroying it. I 
find the typesetting to be a bit "loose" for my taste, with a 
corresponding spread of information over pages. It contains extra 
information about non-directly-Python things (e.g., [X]Emacs Python 
mode, etc.), and a lot more text attempting to give short explanations 
(which PPSI QR does not). There are some inaccuracies, but that is 
inevitable. Probably true of the QR as well.

Personally, I think I would prefer the "tutorial" section out of the 
back of Programming Python if I had to have one pocket reference from 
him, but I suspect that would be harder to do "politically".

I am happy to have both - I will carry the Lutz in my pocket if I want 
a portable ref, and keep the QR in one place so it doesn't get bent 
(damn, I should have bought 2 for myself). If I had to have one I'd go
for the QR, but that may just be me.

As to futures - I hope the PPSI will bring out another edition some 
day, but I doubt it is necessary yet.

Tibs
-- 
Tony J Ibbs (Tibs)      http://www.tibsnjoan.demon.co.uk/
My views! Mine! Mine! (Unless Laser-Scan ask nicely to borrow them.)




From aahz at netcom.com  Wed Jun  2 20:00:11 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Thu, 3 Jun 1999 00:00:11 GMT
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <375536C2.DD917F99@prescod.net> <aahzFCpHw8.Gxt@netcom.com> <7j4e2u$8rh$1@m2.c2.telstra-mm.net.au>
Message-ID: <aahzFCq5CC.KFK@netcom.com>

In article <7j4e2u$8rh$1 at m2.c2.telstra-mm.net.au>,
Mark Hammond <MHammond at skippinet.com.au> wrote:
>
>Not to me.  If I make structural changes to C or other programs, I always
>re-indent the reorganised code correctly, and I would bet that you do too.
>In almost all languages block indenting is _very_ important to the human
>reader.

I think you misunderstood what I meant by "structural" (and I wasn't
particularly clear -- my bad): if I need to add a quick "if" statement
during debugging, I don't like having to make sure I clean up the block
structure formatting.  With just about any other language, I don't have
to.  I do, as I think I did make clear, clean up the formatting after
the debugging.
-- 
                      --- Aahz (@netcom.com)

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

"NRE is a lot like cocaine. Feels great while you're on it, makes you
feel very special and talented, and you become highly annoying to those
who aren't on it."  -- Stef



From rfreedm1 at nospam.rochester.rr.com  Tue Jun 15 05:12:24 1999
From: rfreedm1 at nospam.rochester.rr.com (Benjamin Schollnick)
Date: Tue, 15 Jun 1999 09:12:24 GMT
Subject: Looking for modules?  Config files?  Ini files?  in python..
References: <1282710643-100650854@hypernet.com>
Message-ID: <rNHvjEdhm5Pp-pn2-DlC4do9ULYWY@d185d18f8.rochester.rr.com>

On Tue, 15 Jun 1999 01:53:17, "Gordon McMillan" <gmcm at hypernet.com> 
wrote:

> Benjamin Schollnick writes:
> > 
> >  I'm trying to find a simple method to deal with *.INI or other
> >  styles
> > of
> > configuration files....
> > 
> >  I haven't seen anything, and would hate to re-invent the wheel....
> > 
> >  Any suggestions?
> 
> How 'bout ConfigParser.py? It's in the standard distribution.
> 
> - Gordon
> 
> 

Thanks.... I think that'll do it.... I wonder why I never thought to 
look
in the standard dist.????  I just never thought about it!!!

		- Benjamin



From tim_one at email.msn.com  Tue Jun 29 03:15:07 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 29 Jun 1999 07:15:07 GMT
Subject: RuntimeError?
In-Reply-To: <19990628105921.A26266@optiplex.palga.uucp>
References: <19990628105921.A26266@optiplex.palga.uucp>
Message-ID: <000101bec1ff$1df57780$719e2299@tim>

[RuntimeError ... generally seems to mean something bad and unexpected
 happened, but not bad enough to shut down the interpreter (things *that*
 bad raise SystemError).
]

[Gerrit Holl]
> Ah, I understand. But what kind of things are bad enough to raise
> SystemError?  core dump?

>>> print SystemError.__doc__
Internal error in the Python interpreter.

    Please report this to the Python maintainer, along with the traceback,
    the Python version, and the hardware/OS platform and version.
>>>

That means something has happened Python can't figure out -- it's a problem
in Python itself (or in some C code you wrote to extend Python), not a
problem in your Python code.  This is the kind of thing that *could* cause a
core dump (or other random damage) if Python didn't check for it first.
It's hard to give you an example, since if I could I'd have to report it to
the Python maintainer and he'd fix it before you could reproduce it <wink>.

Core dumps are even worse -- that's something so bad Python didn't even
check for it first, or didn't check for it correctly.  For example, if you
have unbounded recursion:

def f():
    f()
f()

Python does check how many levels deep f calls itself, and raises an
exception if it gets "too deep".  Under Windows, though, the check in 1.5.2
doesn't quite work, and Python dies with a stack overflow in the operating
system.  Python never wants that to happen!  There's no chance for your
Python program to recover from it.

read-Lib/exceptions.py-for-more-entertainment-ly y'rs  - tim





From morse at harborcom.net  Tue Jun 15 07:42:32 1999
From: morse at harborcom.net (Kevin Dahlhausen)
Date: Tue, 15 Jun 1999 11:42:32 GMT
Subject: is there a Gantt chart toolkit?
References: <7in1y2imtt.fsf@betty.artisan.com>
Message-ID: <37663b0c.520391733@news.oh.verio.com>

alt at artisan.com wrote:

>
>subject says it all.

Kind of.  If you search freshmeat for 'Opensched' you will find a 
text-based toolkit for project schedules, resources, and Gantt charts. It
generates Latex output.

It currently has no gui.   I'm fixing that.   It will soon (couple of weeks)
have a Python gui based on the Fast-Light-Toolkit (http://fltk.netpedia.net).





From Lee_Borkman at rta.nsw.gov.au  Tue Jun 15 21:49:06 1999
From: Lee_Borkman at rta.nsw.gov.au (Lee Borkman)
Date: Wed, 16 Jun 1999 01:49:06 GMT
Subject: XML-toolkit for Windows - binaries available?
Message-ID: <376701a3.72288277@newssrv>

Hi,

I'm trying to install the XML-toolkit on Python 1.5.2 / Win95.  I see
on the "Installing the XML Toolkit" page
(http://www.python.org/doc/howto/xml/node4.html), that a pre-compiled
version is soon-to-be-available.

Can anyone help me out in installing from the standard distribution?
I don't have a C-compiler on my system, but I do have nmake/dmake.

TIA,
Leeb.
 



From wlfraed at ix.netcom.com  Sun Jun 13 19:23:56 1999
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Sun, 13 Jun 1999 23:23:56 GMT
Subject: Broken 1.5.2 installation? Missing cw3215mt.dll
References: <1282960807-85602620@hypernet.com> <7ju8j8$2hi$1@Starbase.NeoSoft.COM>
Message-ID: <37643af3.1453829@nntp.ix.netcom.com>

On 12 Jun 1999 13:23:36 -0500, claird at Starbase.NeoSoft.COM (Cameron
Laird) declaimed the following in comp.lang.python:


> 
> Do it with a feathered snake, and he'll follow you an...
> Whoops!  This is in public.
>
	Pity... That was starting to sound... interesting... <G>

	Now there is a cover for the next Python book -- Quetzalcoatl! 

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <



From dickey at shell.clark.net  Sun Jun 20 16:04:04 1999
From: dickey at shell.clark.net (T.E.Dickey)
Date: Sun, 20 Jun 1999 20:04:04 GMT
Subject: Curses for MSDOS
References: <376BD33B.5A6F54BB@erols.com> <e8rgk7.hj81.ln@solling.seismic.de> <376C82FB.4B7F4350@quasar.ipa.nw.ru>
Message-ID: <UObb3.503$4e1.30776@iad-read.news.verio.net>

Alexander V. Voinov <avv at quasar.ipa.nw.ru> wrote:
> Hi,

> Klaus-Juergen Wolf wrote:

>> Swartz <swartz at erols.com> wrote:
>>  > Is there anything like curses for MS-DOS?  Thanx.
>>
>> I know of PC-Curses. Written in the early 80s. Maybe someone has
>> extended it, but in the state I had seen it had no NCurses extensions
>> and wouldn't work with Python.
>>
>> Probably very few people used it, because that screen access mechanism
>> was relatively slow on MS-DOS.

> I learned that the emulation of curses in SLang
> (ftp://space.mit.edu/pub/davis/slang/) is sometimes even better than the
> original curses, ncurses, etc. I may be valid also for MS-DOS. If you can
> build the SLang library for MS-DOS and link your curses-demanding
> application with it you may have solved your problems.

um.  Not exactly.  The two have comparable performance, and slang's curses
interface differs subtly from the normal curses implementations, so
actually only trivial applications can be ported to it without
modification.

Once in a while I am briefly tempted to do side-by-side comparisons (I have
some programs that work with BSD and SYSV curses), but since slang's curses
imitation is ad hoc - contains only the functions that were necessary for
one or another of the sample programs, then either I'd have to butcher the
program or fill in the gaps in slang's code - and then it still wouldn't
function the same.

PDcurses on the other hand, does implement curses...

-- 
Thomas E. Dickey
dickey at clark.net
http://www.clark.net/pub/dickey



From bryanv at arlut.utexas.edu  Mon Jun  7 15:39:47 1999
From: bryanv at arlut.utexas.edu (Bryan VanDeVen)
Date: Mon, 07 Jun 1999 14:39:47 -0500
Subject: waste of resources ?
References: <375C0F91.1C73E4E3@icrf.icnet.uk> <375C16A9.5ADC4990@icrf.icnet.uk>
Message-ID: <375C2003.B01BEC3@arlut.utexas.edu>

Arne Mueller wrote:

> 
> Im running python 1.5.2b2 on SGI power challange irix64 version 6.5.
> Everytime a child process is created via fork(), it realy does it's job
> but never dies, instead the number of zombie processes (as reported by
> the 'top' program) steadily increases, though I call the sys.exit(0)
> method for the child. Any idea how exit a process without creating a
> zombie?

On sysV or Unix98 systems you may set the disposition of SIGCHLD to
SIG_IGN (i.e., ignore death or a child signals)  The watertight method
is to catch SIGCHLD and have the handler call waitpid(...) with WNOHANG
in a loop.

-- 
Bryan Van de Ven
Applied Research Labs
University of Texas, Austin



From rael at my-deja.com  Fri Jun  4 10:55:42 1999
From: rael at my-deja.com (rael at my-deja.com)
Date: Fri, 04 Jun 1999 14:55:42 GMT
Subject: Alternative dictionary output format?
Message-ID: <7j8pd8$f01$1@nnrp1.deja.com>

I would like to be able to take a string which contains C-format
extension variables (e.g., "%(foo)s") and format it with a dictionary in
the standard way, except I would like the format to only substitute for
the recognized variables and to ignore other instances of percent signs.

For example, given this program:

s = "%(href)s/foo/bar?x=%m&y=%n"
d = {'href': "http://www.gomer.com"}
u = s % d
print s

running it produces:

% python t.py
Traceback (innermost last):
  File "t.py", line 3, in ?
    u = s % d
TypeError: not enough arguments for format string

And this program:

s = "%m foobar %(href)s/foo/bar?x=%m&y=%n"
d = {'href': "http://www.gomer.com"}
u = s % d
print s

produces:

% python t.py
Traceback (innermost last):
  File "t.py", line 3, in ?
    u = s % d
ValueError: unsupported format character 'm' (0x6d)

What I would like to write:

s = "%m foobar %(href)s/foo/bar?x=%m&y=%n"
d = {'href': "http://www.gomer.com"}
u = format_ignoring_non_variable_percent_signs(s, d)
print s

And what I would like the behavior to be:

% python t.py
%m foobar http://www.gomer.com/foo/bar?x=%m&y=%n

Is there any way to surface the C interface for the '%' format operator
and to allow an option to effectively wrap each format attempt with a
try/except/pass block?

I realize there are other ways to do this (regular expression
substitution, write my own C extension, etc.), but formatting a string
with multiple variables using a dictionary and the format operator is
extremely convenient (and I imagine relatively effiecient), and I
thought others may have come across this as well.


Bill
--
r a e l @
d e j a .
c o m


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From arcege at shore.net  Thu Jun  3 08:31:46 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Thu, 03 Jun 1999 12:31:46 GMT
Subject: Handling variable number of arguments.
References: <7j17dm$pks$1@nnrp2.deja.com>
Message-ID: <SAu53.1742$nn.518460@news.shore.net>

bhosu at my-deja.com wrote:
: Hi All,
:    I am embedding Python in an application(C++). I would like to pass
: variable number of args to the python function. I could do that from
: python by using def foo(*args). How do I parse these arguments inside my
: application? In PyARg_ParseTuple, I have to know exactly how many args
: to read etc. I am using Python v 1.4. I was looking at
: PyArg_ParseTupleAndKeywords but that needs the args to be a dictionary.
: I am looking a similar thingy for lists. A typeical scenario of the
: usage is like,
: import foo
: foo.set_args('bar','=','1','2','3')

: where foo is the module containing the method implementation.

: Thanks,
: Bhosu.

It looks like you already know how to use PyArg_ParseTuple, so you are
halfway there.  You process the args as a tuple itself.  For example,

static int foo_argc = 0;
static PyObject *foo_argv[];

static PyObject *
foo_set_args(PyObject *self, PyObject *args)
  { register int count, length;
    foo_argc = length = PyTuple_Size(args);
    for (count = 0; count < length; count++) {
      item = PyTuple_GetItem(args, count);
      Py_INCREF(item);
      foo_argv[count] = item;
    }
    foo_argv[length] = NULL;
    Py_INCREF(Py_None);
    return Py_None;
  }

Is this kind of what you are looking for?
  -Arcege




From simngolda at voyager.net  Wed Jun 30 20:14:15 1999
From: simngolda at voyager.net (Sim & Golda Zacks)
Date: Wed, 30 Jun 1999 20:14:15 -0400
Subject: MySQL interface
Message-ID: <7lebn4$m1i@newsops.execpc.com>

My ISP has a MySQL server and I want to use it through Python 1.4. The
version of MySQL they have is:  mysql  Ver 9.16 Distrib 3.21.33, for
pc-linux-gnu (i586).
I looked at the Python database site and they have a module for MySQL,
but it is a C file and the instructions are to build it with the
distribution. I am not building anything, all I want to do is use the
database server. Can someone help me figure out what I need to use, and
how to get it. I was going to use Gadfly, but I'm also going to need to
use a database with JAVA JDBC so I want to use the same product for all
my stuff.
The ISP only has Python 1.4.

Thanks

Sim




From shallib at netscape.com  Wed Jun  9 23:39:44 1999
From: shallib at netscape.com (Steve Halliburton)
Date: Wed, 09 Jun 1999 20:39:44 -0700
Subject: How do I match the last occurence of a character?
References: <375F21B3.D23A5EA9@west.boeing.com> <slrn7luc4m.g2c.bwinton@tor.dhs.org>
Message-ID: <375F3380.1379B1CA@netscape.com>

Why not just use the os module:

>>> import os
>>> name = 'news/breakingnews/981001b.html'
>>> base = os.path.split(name)[0]
>>> base
'news/breakingnews'
>>> file = os.basename(name)
>>> file
'981001b.html'


Steve H.

Blake Winton wrote:

> On Thu, 10 Jun 1999, Jeff P. <jeffrey.polaski at west.boeing.com> wrote:
> >How do I match the last occurence of '\' in a string?
> >import regex
>
> This is your first problem.  Why use a tool more powerful than you
> really want?  Check out the string module.
>
> >>> import string
> >>> name = "test\\a\\b\\c.html"
> >>> string.rfind(name,"\\")
> 8
> >>> name2 = name[0:string.rfind(name,"\\" )]
> >>> name2
> 'test\\a\\b'
>
> Note that I have to use "\\" to match a single "\".
> Oh, and to get output similar to the one in your example, you would have
> to use something like
>
> baseurl = url[ 0 : string.rfind( url, "\\" )+1 ]




From stidolph at origin.ea.com  Thu Jun  3 18:58:06 1999
From: stidolph at origin.ea.com (Stidolph, David)
Date: Thu, 3 Jun 1999 17:58:06 -0500 
Subject: Debug:  Module to track python objects.
Message-ID: <11A17AA2B9EAD111BCEA00A0C9B4179301E10EF0@forest.origin.ea.com>

First and foremost I want to thank Guido for the traceback system - just
having made this module possible and my task MUCH easier.

This is an update to a debugging module that I posted before.  See the end
of this mail for the module.

Purpose:  The Debug module tracks python class instances between their
__init__ and __del__ method calls.  The idea is to provide tracking for
objects so that a report can be generated with objects that have not been
freed.  When we first ran this on our code we found almost 300 objects that
were not being destroyed (more than half were windows resources like graphic
texture files).

Asumptions:  The __init__ method of every class calls LogInit method of the
debug class.  The __del__ method calls LogDel.  We have a special method
called Destroy for every class that is designed to remove references to
other objects and handle the circular reference problem.  The Destroy method
calls LogDel

Useage (we have a Globals module with specific instance variables to grant
access to modules like Debug):  

	#sample
	import Globals

	class Name:
	  def __del__(self):
	    if __debug__:
	      Globals.debug.LogDel(self)

	  def __init__(self):
	    if __debug__:
	      Globals.debug.LogInit(self)
	    # your code

	  def Destroy(self):
	    if __debug__:
	      Globals.debug.LogDestroy(self)
	    # your code

Questions:

1. We are losing the tracebacks.  At some point the output terminates.  This
happens after a large output, so I suspect a buffer overrun.
2. Can we shrink the tracebacks?  Keeping all the strings while the program
runs seems a little much.  We also want to modify the output for MSVC.
3. Is there any other tracking we can do for catching bugs?  Ways to enhance
this module?

I'm open to comments, include those designed to make it more effecient,
powerful or easier to use.

Thank you for your help,

David Stidolph
Origin.

#Debug Module
#
#This class is designed to track Python objects between their __init__() and
__del__() method calls.
#There are different 'levels' to the tracking.  If we do it at all, we will
track the objects by the object
#id() using a dictionary.  If we have 'holdTraceback' active, then in
LogInit we get the current
#stack, translate it into a list of strings and use that list as the value
and the object reference as
#the key in the dictionary.
#
#As a dictionary we keep only one copy and don't allow for
LogInit/LogDestroy to be called more than
#once (have setting to throw exception if it occurs).  This means you put
the calls to LogInit and
#LogDestroy in the base class .. not the derived classes.  Proper class name
of the object is used
#by __class__.__name__.
#
#TODO:
#       Fix output for MSVC IDE to use absolute paths to Python files for
double-clicking
#       (Note that this works now for Windows 95, but not 98)
#       Add Python object for output in LogReport to allow output to file or
sending across net
#       Add print formatted output to wrap output (allow to send across
net).
#       Optimize use of stack trace to reduce memory requirements.
#       Clipping problem on the traceback - gets cut-off.

from sys import exc_info
from traceback import extract_stack

class Debug:
  """Only one instance allowed"""

  __refDebugCount = 0 # class variable

  def __del__(self):
    Debug.__refDebugCount = Debug.__refDebugCount - 1

  def __init__(self, trackObjects = 1, holdTraceback = 1,
raiseExceptionOnDuplicate = 0, displayInit = 0, displayDestroy = 0):
    if Debug.__refDebugCount != 0:
      raise "Only one copy of debug allowed"

    if __debug__:
      self.trackObjects = trackObjects
      self.holdTraceback = holdTraceback
      self.raiseExceptionOnDuplicate = raiseExceptionOnDuplicate
      self.displayInit = displayInit
      self.displayDestroy = displayDestroy
    else:
      self.trackObjects = 0               #if we don't have debug mode, then
it doesn't matter what
      self.holdTraceback = 0              #settings we use since nothing
SHOULD be calling us.
      self.raiseExceptionOnDuplicate = 0  #Just in case we ignore and set
for nothing.
      self.displayInit = 0
      self.displayDestroy = 0

    self.activeObjects = {}     #clear dictionary
    Debug.__refDebugCount = 1    #we are initialized

  def Destroy(self):
    Debug.__refDebugCount = 0
    self.activeObjects = None

  def LogDel(self,object):
    #We wrap the destruction of the key inside a try..except block in case
the LogDestroy is called without
    #calling the LogInit (rare) or LogDestroy is being called for a second
time (far more likely).
    try:
      objectID = id(object)
      value = self.activeObjects[objectID]
      if 0 == value[1]:
        if self.raiseExceptionOnDuplicate:
          raise "MEM: LogDel called on "+object.__class__.__name__+" without
calling Destroy()"
        else:
          print
          print "MEM: LogDel called on "+object.__class__.__name__+" without
calling Destroy()"
          print
      del self.activeObjects[objectID]
    except:
      if self.raiseExceptionOnDuplicate:
        raise "MEM: LogDel called twice on "+object.__class__.__name__+".
See if base class doing LogInit/LogDestroy/LogDel (only one should)"
      else:
        print
        print "MEM: LogDel called twice on "+object.__class__.__name__+".
See if base class doing LogInit/LogDestroy/LogDel (only one should)"
        print

  #Called from each base object Destroy() method.
  def LogDestroy(self,object):
    if self.trackObjects:
      if self.displayDestroy:
        print 'MEM: ',repr(object),'of class',object.__class__.__name__,'is
being Destroyed'
      #We wrap the destruction of the key inside a try..except block in case
the LogDestroy is called without
      #calling the LogInit (rare) or LogDestroy is being called for a second
time (far more likely).
      self.activeObjects[id(object)][1] = 1

  #Called from base object __init__(self) routines
  def LogInit(self,object):
    if self.trackObjects:
      if self.displayInit:
        #display that the object has been constructed and its __init__ has
been called
        print 'MEM: ',repr(object),'of class',object.__class__.__name__,'is
being initialized'

      #traceback of stack is generated using extract_stack() and used as the
value in the
      #dictionary with the object reference used as the key.
      if self.holdTraceback:
        #We check for the object already being in the dictionary ONLY if we
will throw an 
        #exception, silent otherwise
        if self.raiseExceptionOnDuplicate:
          if self.activeObjects.haskey(object):
            raise "Object has already called LogInit"
        tb = extract_stack()
        self.activeObjects[id(object)] = [object.__class__.__name__,0,tb]
      else:
        self.activeObjects[id(object)] = [object.__class__.__name__,0,1]

  #Called at program end to report on undestroyed Python objects.
  def LogReport(self):
    # For formatting that Visual Studio can recognize, we're adjusting
    # output to look like:
    # c:\test\gagob.cpp(236) : error C2065: 'Clear' : undeclared identifier
    if self.trackObjects:
      if len(self.activeObjects) > 0:
        print 
        print 'MEM: UnDestroyed Python Objects Report:'
        print '     %s python objects total.' % len(self.activeObjects)
        print
        print 'Summary:  These objects were not released.  Individual stack
traces follow'
        for objectID,value in self.activeObjects.items():
          print ' class ' + value[0]
        print
        print
        #Get Object reference and stack traceback for each undestroyed
Python object
        for objectID,value in self.activeObjects.items():
          className = value[0]
          destroyCalled = value[1]
          tb = value[2]
          if destroyCalled == 1:
            print 'MEM:  class ' + className + ' Destroy() called, but not
__del__.'
          else:
            print 'MEM:  class ' + className + ' Destroy() and __del__
methods not called.'
          if self.holdTraceback:
            for tuple in tb[:-2]: #-2 is to eliminate report of call to
LogInit()
              print str(tuple[0])+'('+str(tuple[1])+'): in
'+str(tuple[2])+','
              print '  '+str(tuple[3])
          print #blank line between
      else:
        print 'MEM: Congratulations!  No Python object leaks!'




From heiland at ncsa.uiuc.edu  Fri Jun 11 12:47:06 1999
From: heiland at ncsa.uiuc.edu (Randy Heiland)
Date: Fri, 11 Jun 1999 11:47:06 -0500
Subject: Embedding question
Message-ID: <37613D8A.BD4E07FA@ncsa.uiuc.edu>

While I wade thru some docs, maybe someone can answer this straightaway.

Python 1.5.2
IRIX 6.4

In a C++ pgm, I can successfully do this:

  PyRun_SimpleString("import string");

and this:
  PyRun_SimpleString("from math import *");

however, I cannot do this:
  PyRun_SimpleString("from myStuff import *")

where "myStuff" is a shared lib (.so) found in my PYTHONPATH.

I can do this just fine from the interpreter, i.e.:

Python 1.5.2 (#9, Jun  3 1999, 14:44:16) [C] on irix646
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> from myStuff import *

Any tips?

thanks,
--Randy  (heiland at ncsa.uiuc.edu)




From farnulfo at cpr.fr  Wed Jun 23 13:59:00 1999
From: farnulfo at cpr.fr (Franck Arnulfo)
Date: Wed, 23 Jun 1999 19:59:00 +0200
Subject: Pb. with JPython & VisiBroker
Message-ID: <7kr72p$942$1@news.imaginet.fr>

Hi,

I try to use the scripting facilities of JPython with VisiBroker for Java
ORB.
But I've a problem to make an instance of the ORB.
Here is a copy of the output :

C:\JPython-1.1beta2>set PATH
PATH=C:\jdk1.1.7B\bin
PATHEXT=.COM;.EXE;.BAT;.CMD;.VBS;.JS

C:\JPython-1.1beta2>set CLASSPATH
CLASSPATH=C:\jdk1.1.7B\lib\classes.zip;C:\vbroker\lib\vbjorb.jar;C:\vbroker\
lib\vbjapp.jar

C:\JPython-1.1beta2>pythonjdk11
JPython 1.1beta2 on java1.1.7B
Copyright (C) 1997-1999 Corporation for National Research Initiatives
>>> from org.omg.CORBA import *
>>> orb = ORB.init(None, None)
Traceback (innermost last):
  File "<console>", line 1, in ?
NameError: ORB
>>>

And my pythonjdk11.bat file is:
----------------------------------------------------------------------------
----------------
@echo off
REM This file generated by JPython installer
REM Created on Mon Jun 21 12:36:59 CEST 1999 by ARNULFO_F
"C:\jdk1.1.7B\bin\java.exe" -Dpython.home=C:\JPython-1.1beta2 -classpath
"C:\JPython-1.1beta2\jpython.jar;%CLASSPATH%" org.python.util.jpython %1 %2
%3 %4 %5 %6 %7 %8 %9
----------------------------------------------------------------------------
----------------

Can you help me out of this trouble?
Thanks!!







From bwarsaw at cnri.reston.va.us  Thu Jun  3 14:55:47 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Thu, 3 Jun 1999 14:55:47 -0400 (EDT)
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7imr6b$1s0$1@nnrp1.deja.com>
	<87pv3iq6ss.fsf@ev.netlab.co.jp>
	<3752C76D.604CD474@prescod.net>
	<87n1ykr0z9.fsf@ev.netlab.co.jp>
	<375372F6.A874554@prescod.net>
	<87hfosquc0.fsf@ev.netlab.co.jp>
	<3753F121.E4A9BE8B@prescod.net>
	<87yai3ntdt.fsf@ev.netlab.co.jp>
	<7j3iif$134e$1@nntp6.u.washington.edu>
	<7j45iv$908$4@cronkite.cc.uga.edu>
	<7j4hhj$118u$1@nntp6.u.washington.edu>
	<14166.46756.866947.64007@anthem.cnri.reston.va.us>
	<87g14942co.fsf@pc-hrvoje.srce.hr>
Message-ID: <14166.53171.289824.873109@anthem.cnri.reston.va.us>

>>>>> "HN" == Hrvoje Niksic <hniksic at srce.hr> writes:

    HN> Isn't that a case of sacrificing (Python) correctness for
    HN> (Java) efficiency?  The code that relies on __del__() getting
    HN> called eventually will lose in JPython.

You're right.  In this specific case it's probably worth it, since Jim
reported a 10x performance hit when PyInstance.finalize() was defined.

Given that JVMs probably won't make finalize() an order of magnitude
more efficient any time soon, and given that most Python classes don't 
use __del__() it makes sense that PyInstance has no finalize()
method.  The right approach would be to create a subclass of
PyInstance (say called PyFinalizingInstance) which /does/ call
__del__() and then create one of these when building a class that has
a __del__ defined.

I haven't thought it through completely, and its definitely not on my
personal radar scope for near-term JPython development, but I do at
least plan to document this difference between CPython and JPython.

-Barry



From tjreedy at udel.edu  Mon Jun 14 02:45:30 1999
From: tjreedy at udel.edu (Terry Reedy)
Date: 14 Jun 1999 06:45:30 GMT
Subject: The Best Parsing Book on Earth (was RE: Interpolation)
References: <000001beb5ef$eaec04a0$f09e2299@tim>
Message-ID: <7k28ea$8pq$1@news.udel.edu>

In article <000001beb5ef$eaec04a0$f09e2299 at tim>, tim_one at email.msn.com 
says...

>Bears repeating, then (I've gotten many email inquiries too).  It's a
>wonderful intro, now out of print, but thanks to the authors' 
generosity
>downloadable for free:
>
>    http://www.cs.vu.nl/~dick/PTAPG.html
>

PTAPG = Parsing Techniques: a practical Guide; Grune and Jacobs.

I just finished rereading the copy from our library.  Recommend it 
highly to anyone interested in the subject.  A free copy is indeed a 
gift.  TJReedy





From John.Mitchell at p98.f112.n480.z2.fidonet.org  Tue Jun 29 17:51:58 1999
From: John.Mitchell at p98.f112.n480.z2.fidonet.org (John Mitchell)
Date: Tue, 29 Jun 1999 22:51:58 +0100
Subject: BOFs (Birds-Of-a-Feather) sessions at the Monterey Python
Message-ID: <00001599@bossar.com.pl>

From: John Mitchell <johnm at magnet.com>

On Tue, 29 Jun 1999, David Ascher wrote:

> As you may know, O'Reilly & Associates is organizing a Python Conference
> as part of its Open Source Convention in Monterey in August. [...]
>
> I'd like to solicit discussion on what BOFs would be good to organize, and
> find some organizers.  [...]
> http://conferences.oreilly.com/bof/


I'd love to see "Python and Java BOF", and maybe one on CORBA.  If we
could hornswaggle Jim Hugunin or Barry Warsaw to talk a bit, that would be
great.  Whoa -- just found Jim's JPython slides[1].  For the
record, his talk at JavaOne was excellent -- thanks!

Basically, Java can bite me, but it has tons of libraries, everything
talks to it, and it has local fatass support.  Switching to JPython sounds
like a win/win, and I need to prove it.

CORBA I havent messed with since Fnorb used to be unstable, I'd love to
hear how it's changed.


- j


[1] http://www.jpython.org/jpython-talk-1.ppt



From janssen at parc.xerox.com  Wed Jun  2 22:35:41 1999
From: janssen at parc.xerox.com (Bill Janssen)
Date: Thu, 3 Jun 1999 02:35:41 GMT
Subject: linking C++ extensions for Python & autoconf?
Message-ID: <99Jun2.193548pdt."3612"@watson.parc.xerox.com>

I've got an autoconf issue that perhaps some of you can shed some
light on.  I'd like to generate a shared library (DLL) containing C++
subroutines that would then be loaded into a non-C++ program, like a
Python or Java VM.  To do this, the shared library must be linked with
with not only the application C++ code, but also any C++ libraries
that the system may use, so that we don't get "unresolved symbol"
errors when the VM loads the shared library.

How can I find out, in an autoconf configure script, what these
libraries are?

Any ideas?  Or is this perhaps well-known?  I'm not a C++ guru.

Bill



From bbaetz at ug.cs.usyd.edu.au  Wed Jun  2 19:39:59 1999
From: bbaetz at ug.cs.usyd.edu.au (Bradley Baetz)
Date: Thu, 03 Jun 1999 09:39:59 +1000
Subject: Reproducable Python core dump
References: <37546EDA.20825D43@ug.cs.usyd.edu.au> <5lg14azdh8.fsf@eric.cnri.reston.va.us>
Message-ID: <3755C0CF.74EAE15A@ug.cs.usyd.edu.au>

Guido van Rossum wrote:
> 
> Since Tk_GetGC() is a huge function, I can't say more without
> recompiling Tk with -g.  Perhaps you can try this yourself?
> 
Not for a while. I've got to hand in my assignment on Monday. :) I don;t
have to use the mouse for my demonstratios though.

For anyone who is looking into this, it only happens when using the
mouse - not when I use the keypad to highlight the link, and then call a
slightly modified version of Viewer.do_link(). It still happens in my
version when I use the mouse (That's how I noticed it).

In case its relevent, data is read from the keypad from an fd to a C
program. Python then selects the appropriate action.

The versions of libtk used were:

Stock RedHat 5.2 (tk-8.0.3-20)
SunOS tk8.0.?? (How can I find out which subversion it is?)
	tk4 is also installed, so it may be this one
Debain 2.0 (8.0p2-4)

Bradley



From tim_one at email.msn.com  Fri Jun 18 20:33:17 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 18 Jun 1999 20:33:17 -0400
Subject: NameError Problem
In-Reply-To: <7kels4$9tv$1@eskinews.eskimo.com>
Message-ID: <000401beb9eb$534904e0$a69e2299@tim>

[Jonathon]
> 	I'm writing a long << 25K line >> script to verify data from a
>  	database.
>
> 	The following bit of script is giving me a NameError, and
> 	I can't figure out why.
>
> 	datafile = open ( the_data_file_is, "r" )
> 	new_datafile = open ( the_temp_data_file_is, "w" )
> 	line_kounter = 0
> 	while line_kounter < 50000
> 	    line_kounter = line_kounter + 1
> 	    templine = datafile.readline(line_kounter)
>             validate_line = templine
> 	    if ( the_file_is == "john_www.db" ) :
> 	         temp_record_id = int(validate_line[2:18])
> 		 temp_country  = strip(validate_line[20:50])
>             elif ( the_file_is == "jan_www.db " ):
>                  temp_record_id = int(validate_line[2:10])
>                  temp_country = strip(validate_line[12:20])
>             print temp_record_id
>
> 	At that point I get a NameError message.
> 	All I'm trying to do is have it print to screen
> 	the record number it is working on.
>
> 	I've double checked the white spaces, and they are
> 	all lined up at 0, 4, 8, 12, 16 etc.
>
> 	What am I missing here?  << And how do I correct it. >>
> ...

1) Don't mix tabs and spaces.

2) Use "python -t" or tabnanny.py (not your eyes) to check for consistent
indentation if you can't live with #1.

3) Post the code you're actually running (e.g., this version is missing a
":" after the "while" stmt, so Python won't run it at all).

4) After the "new_datafile = ..." line, insert

        temp_record_id = "Oops!  I never gave this a value!!  Silly me."

and then see what your program prints <wink>.

if/elif-without-an-else-ly y'rs  - tim





From jeremy at cnri.reston.va.us  Wed Jun  2 18:23:31 1999
From: jeremy at cnri.reston.va.us (jeremy at cnri.reston.va.us)
Date: Wed, 02 Jun 1999 22:23:31 GMT
Subject: mailing list to news gateway
Message-ID: <7j4asp$v4a$1@nnrp1.deja.com>

I have noticed what appear to be problems with the gateway between
the mailing list (python-list at python.org) and the newsgroup -- and
more specifically with the view of the newsgroup at dejanews.

Since about last Thursday, messages I have sent to
python-list at python.org haven't been showing up at Deja News. I
would assume they also aren't showing up in the newsgroup, but
I'm not sure.

So two questions: First, has anyone else noticed problems with messages
sent to the mailing list not appearing in the newsgroup?  Trying
to establish the scope of the problem.  Perhaps it's just weirdness
related to my messages.  Second, has anyone seen posts from me on the
newsgroup lately -- say in the Python 2.0 thread?  (I'm sure everybody
keeps track of all the witty and insightful comments I make <1.0 wink>.)

Jeremy


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From dgap at esoterica.pt  Mon Jun  7 13:06:01 1999
From: dgap at esoterica.pt (dgap)
Date: Mon, 7 Jun 1999 19:06:01 +0200
Subject: Online/Offline Manual on Python
Message-ID: <928782341.193714@nurn.esoterica.pt>

Hey, guys...

Anyone knows where can I get an on-line (or off-line...) manual of python??

Thanks

Paulo





From guido at cnri.reston.va.us  Wed Jun  2 14:03:47 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 02 Jun 1999 14:03:47 -0400
Subject: Reproducable Python core dump
References: <37546EDA.20825D43@ug.cs.usyd.edu.au>
Message-ID: <5lg14azdh8.fsf@eric.cnri.reston.va.us>

Bradley Baetz <bbaetz at ug.cs.usyd.edu.au> writes:

> My CS project for this semester involves modifying Grail to support
> navigation through a keypad, similarly to Lynx. Whilst testing my
> handing of nested tables, I came across this bug:
> 
> The following will cause a reproducable core dump in Python:
[...]
> This is reproducable on two different RedHat 5.2 Intel boxes, and on a
> SunOS SPARC machine. One of the Redhat boxes and the SPARC box were
> running 1.5.2, the other RedHat box 1.5.1.

Thos will probably be hard to debug.  I looked briefly at the core
dump and it seems to be deep in Tk code -- perhaps the problem is with 
Tcl/Tk?  (That would explain why it's independent of the Python
version you use.)

For whoever wants to look into this more, here's part of the stack
trace from gdb:

#0  0xeeb50094 in Tk_GetGC ()
#1  0xeeb37c10 in GetStyle ()
#2  0xeeb37db4 in LayoutDLine ()
#3  0xeeb38488 in UpdateDisplayInfo ()
#4  0xeeb3b6bc in TkTextPixelIndex ()
#5  0xeeb401d8 in TkTextPickCurrent ()
#6  0xeeb400b0 in TkTextBindProc ()
#7  0xeeb4b9e8 in Tk_HandleEvent ()
#8  0xeeb4bd88 in WindowEventProc ()
#9  0xeeb99d9c in Tcl_ServiceEvent ()
#10 0xeeb9a00c in Tcl_DoOneEvent ()

Since Tk_GetGC() is a huge function, I can't say more without
recompiling Tk with -g.  Perhaps you can try this yourself?

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



From jeremy at cnri.reston.va.us  Tue Jun  1 13:55:28 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Tue,  1 Jun 1999 13:55:28 -0400 (EDT)
Subject: Gzip.py
In-Reply-To: <1318D78C9072D11195C9006094EA98A72C2081@ocesrv>
References: <1318D78C9072D11195C9006094EA98A72C2081@ocesrv>
Message-ID: <14164.7764.546060.19420@bitdiddle.cnri.reston.va.us>

>>>>> "MC" == Matej Cepl <Cepl at fpm.cz> writes:

  MC> Hi, I tried to play with gzip.py and it doesn't seem to
  MC> work. Enclosed module produced output which Windows Commander as
  MC> well as gzip refused to accept as correct gzip archive.

I created my own file, fre-en_2.htm, and ran your script.  It worked
correctly; gzip uncompressed it.  What exactly was your problem? Can
you send me a copy of the .gz file produced?   

Jeremy



From wtanksle at dolphin.openprojects.net  Tue Jun 29 19:15:57 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Tue, 29 Jun 1999 23:15:57 GMT
Subject: Why exceptions shouldn't be used for flow control [Re: YAS to the    "Reading line-by-line" Problem]
References: <Pine.SUN.3.95-heb-2.07.990622183356.3454A-100000@sunset.ma.huji.ac.il> <376FDF70.4756AD38@mindspring.com> <slrn7n0chl.gh6.wtanksle@dolphin.openprojects.net> <377173F7.36BCFF6B@compaq.com> <slrn7n31k4.idj.wtanksle@dolphin.openprojects.net> <3772AED9.D9E1110@compaq.com> <slrn7na4hq.3vt.wtanksle@dolphin.openprojects.net> <3776B0ED.C1ABBEC1@compaq.com>
Message-ID: <slrn7niktd.3rr.wtanksle@dolphin.openprojects.net>

On Mon, 28 Jun 1999 11:17:01 +1200, Greg Ewing wrote:
>William Tanksley wrote:

>> while not f.eof():
>>   line = f.readline()
>>   # ...

>I'd love to be able to write reading loops that way.
>Unfortunately, it's a fact of life that some of the
>things read() has to deal with are unable to detect
>eof without trying to read something. I think the
>existing definition of read() and friends is the best 
>that can be done in those circumstances.

I'd guess that you're talking about things like text file parsing and
such.  I would use readline there, but that's fine.  Yes, unstructured
data is more difficult to handle.

>> It's been a while since I've used read, so I
>> don't recall what it actually returns

>It returns an empty string if it can't read more
>than 0 characters, same as readline().

Ah.  There's a justification for that, as you observed.  I don't like the
fact that readline is stuck doing the same thing (record delimiters are
not really part of the record).

>> in a function designed to get n!=0
>> more characters, an EOF really is an error.  Unfortunately for me, that's
>> not the definition of 'read', and never will be.

>Perhaps there should be a 2-parameter version of read:

>   read(min, max)

>which would raise an exception if it couldn't read at
>least min characters. Setting min=0 would give the
>current behaviour, and setting min=max would allow
>reading a fixed-length record without having to check
>for errors. Then everyone would be happy!

A facinating solution.  Well, since I don't use read, I probably wouldn't
use it, and I'm the only person whining.  Therefore, it's probably not
worth implementing (even though it is a cool solution).

>Greg

-- 
-William "Billy" Tanksley



From graham at sloth.math.uga.edu  Thu Jun 10 22:58:10 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 11 Jun 1999 02:58:10 GMT
Subject: fork()
References: <000601beb1c7$682c6220$559e2299@tim> <7jjghd$9il$1@cronkite.cc.uga.edu> <375D5826.23197692@inrialpes.fr> <375D6C60.FF9998DE@appliedbiometrics.com> <5laeu8zmza.fsf@eric.cnri.reston.va.us> <14175.51022.577016.265867@amarok.cnri.reston.va.us> <5lzp27yf89.fsf@eric.cnri.reston.va.us>
Message-ID: <7jpu02$hsc$1@cronkite.cc.uga.edu>

Guido van Rossum (guido at cnri.reston.va.us) wrote:
: Of course finalizers are still screwed, since if an instance is
: involved in a cycle, I may have zapped its instance variables before
: calling its __del__.  (To break the cycles, the collector would call
: the clear() method of all unmarked dicts, which would break the
: cycles, and then the ref counting would free the depending objects.)

This is why I suggested two pass collection. Don't zap things in pass
one. Zapping things is exactly what removes the context in which __del__
methods can run predictably and without error. Do the zapping in pass
2.

graham
-- 
              Where do you think you're going
              Don't you know it's dark outside
              Where do you think you're going
              Don't you care about my pride



From matomira at iname.com  Mon Jun 14 06:13:24 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Mon, 14 Jun 1999 12:13:24 +0200
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <3762988E.C94AA252@iname.com> <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu> <3763C746.5CA8FE78@iname.com> <m3u2sby4gf.fsf@soma.andreas.org>
Message-ID: <3764D5C4.68DB9A66@iname.com>

Andreas Bogk wrote:
> 
> Fernando Mato Mira <matomira at iname.com> writes:

> This is an age-old flamewar. It just so happens that there are people
> who like infix syntax and detest prefix, and then there are people who
> love their parens and hate infix.

I have no idea what flamewars have been going on in comp.lang.dylan,
as it's the first time I set foot here, so FWIW, here's what I think:

1. The Dylan case is different. It was supposed to have 2 syntaxes.
2. I was not excited about the Dylan effort when it came out, as it the
main   motivation was to create a new language for some Apple palmtop,
and this consumed resources that could have been used to improve CL (it
could be argued that it was an `improved Scheme' but how about
supporting EULisp instead?).
3. CMU shuts down CMUCL and starts to work on Dylan. Bad
3. Apple decides to drop the Lisp syntax. Bad.
4. Apple drops Dylan. 
5. Open Source community embraces Dylan. Great. Multimethods to the
mainstream.

The question is: what prevents you now from to going back to the
original idea?
[I'm sorry, if I was funded to spend time myself in creating a Lisp
frontend
for something, it would be for Cecil. If CLOS were too heavy for
something,
there's much more to gain over there. And good prototype-based dialect
would
be a more interesting addition to the Lisp family.]

Regards,

-- 
Fernando D. Mato Mira                    
Real-Time SW Eng & Networking            
Advanced Systems Engineering Division
CSEM                             
Jaquet-Droz 1                   email: matomira AT acm DOT org
CH-2007 Neuchatel                 tel:       +41 (32) 720-5157
Switzerland                       FAX:       +41 (32) 720-5720

www.csem.ch      www.vrai.com     ligwww.epfl.ch/matomira.html



From dalke at bioreason.com  Fri Jun 25 00:55:06 1999
From: dalke at bioreason.com (Andrew Dalke)
Date: Thu, 24 Jun 1999 22:55:06 -0600
Subject: GC In Python: YAS (Yet Another Summary)
References: <Pine.SUN.3.95-heb-2.07.990616155228.15761A-100000@sunset.ma.huji.ac.il> <p53dzrfcea.fsf@bidra241.bbn.hp.com> <930267199.139037@clint.waikato.ac.nz>
Message-ID: <37730BAA.28A7C738@bioreason.com>

Stuart Yeates <syeates at cs.waikato.ac.nz> said:
> Of course, if anything does pointer XORing or other crimes,
> tools such as these are unavaliable.

Here's one which isn't a crime but would still have problems.
One of the C libraries we use also supports FORTRAN.  To do this,
it internally maps all pointers to integer handles, and it is
these integers which are used in the client apps.

On the other hand, maybe supporting FORTRAN is a crime, but
that's the perfered language of the vendor's customer base;
mostly chemists.

So if Python switches to these sorts of GC, I'll have to stick
with an older (pre-GC) version.

						Andrew Dalke
						dalke at bioreason.com



From tre17 at pc142.cosc.canterbury.ac.nz  Tue Jun 29 19:05:42 1999
From: tre17 at pc142.cosc.canterbury.ac.nz (Timothy R Evans)
Date: 30 Jun 1999 11:05:42 +1200
Subject: Arbitrary precision calculations
References: <7lb60c$ot1$1@nnrp1.deja.com>
Message-ID: <c75k8sm620p.fsf@pc142.cosc.canterbury.ac.nz>

catlee at my-deja.com writes:

> Hi,
> 
> I'm looking for a package that will let me do arbitrary precision
> arithmatic in python.  I've come across the real.py module written by
> Jurjen N.E. Bos, and it seems like a really nice module except for the
> fact that you can't compare two numbers for equality,inequality,etc.
> 
> Cheers,
> Chris
> 
> P.S. I found the real.py package by doing a search on www.deja.com for
> arbirary precision in comp.land.python
> 
> 
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.

If I'm reading this right you want arbitrary precision integers right?
Just use the Python `Long' integer, created using the `long' function
or by placing an L after a literal number (e.g. 10L).  All standard
numerical operators ork on long intergers (if you're doing crypto
stuff then don't forget that `pow(X,Y,Z)' calucates X to the power of
Y, modulo Z efficiently.

--
Tim Evans




From mwh21 at cam.ac.uk  Wed Jun 23 13:41:07 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 23 Jun 1999 18:41:07 +0100
Subject: os.kill()
References: <37708560.BF23BA36@dlr.de>
Message-ID: <m37lou3jd8.fsf@atrus.jesus.cam.ac.uk>

Sven Drescher <Sven.Drescher at dlr.de> writes:

> Hi!
> I have any problems to kill a process started with execv().
> The process is right on and the pid is ok. But I tested :
>         os.kill(pid, 0)
> or    os.kill(pid, 15)
> or    os.kill(pid, signal.SIGTERM)
> an nothing happens.
> I looked for my problem in the newgroups and on the python homepage, but
> I haven't found anything.
> I killed my process on console with " kill -9 pid".
> Is there a stronger version of os.kill() like under UNIX?

os.kill(pid,9)?

> 
> Thanks for help and hints!
> 
> --
> Sven



From zigron at jps.net  Fri Jun 25 22:51:55 1999
From: zigron at jps.net (Zigron)
Date: Fri, 25 Jun 1999 19:51:55 -0700
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <JWQc3.2620$U6.10996@newsr2.twcny.rr.com> <3773EE44.F231F1E4@otenet.gr> <_ASc3.2630$U6.11110@newsr2.twcny.rr.com>
Message-ID: <5gXc3.1573$or1.80853@typhoon-sf.snfc21.pbi.net>

    Neither tkinter nor wxPython are a language, actually, they're just a
series of "classes" etc that you access/implement in Python.
    You wouldn't ever have to learn TCL to learn Tkinter, nor C/C++ to
learn wxWindows.

-S


Darrell <news at dorb.com> wrote in message
news:_ASc3.2630$U6.11110 at newsr2.twcny.rr.com...
> Is this more trouble than learning tkinter and or wxPython ?
> I just installed and gave the jpi thing a try. It's not ready for prime
time
> yet.
>
> --
> --Darrell
>
>
>





From dave_kirby at eur.3com.com  Tue Jun 22 07:25:54 1999
From: dave_kirby at eur.3com.com (Dave Kirby)
Date: Tue, 22 Jun 1999 11:25:54 GMT
Subject: [pythonwin] popup menu problems   (was: [Pythonwin] Help needed creating a Window)
References: <37651bde.25372123@NNRP.UK.INSNET.NET>
Message-ID: <376f6b7a.760974@nnrp.uk.insnet.net>

On Mon, 14 Jun 1999 15:53:03 GMT, I wrote:

>Thanks to Mark Hammond's help I now have an app that displays an icon
>in the system tray which I can process messages from. However when I
>try to create a popup menu menu for it nothing happens. The call to
>PyCMenu.TrackPopupMenu doesnt display anything and never returns.  I
>have tried both using it to send a notify message to the window and
>returning a command, but it didnt make any difference. I am unsure
>whether it is a failure of my understanding of the way Windows does
>these things, a bug in my code, or a problem with pythonwin. 
> [ example code snipped ]

I have discovered what the problem was - the line 
            self.cMenu.TrackPopupMenu( pos, owner=self.cWnd )
was failing with an exception since it didnt like the use of named
parameters, but the exception was geting discarded somewhere along the
way. I only found it when I enclosed the line in a try/except clause.
I *think* the exception is getting caught & thrown away somewhere in
the pythonwin code since I didnt have any try clauses in my code at
that point.

I now have the popup menu working, except for one small problem. When
I run it using python(w).exe the menu appears in the correct place and
the user can select a menu item ok, but clicking outside of the menu
doesn't dismiss it. There is no way to get rid of the menu without
selecting a menu item.  When I run it from within Pythonwin.exe it
works fine[1], so I presume that the problem is that there is no top
level window to process the WM_CANCELMODE message.  Is there a way to
register my icon window to handle this message, or do I need to go the
full PythonWin route? My app doesnt use the doc/view architecture so I
want to avoid unnecessary complication if possible. From looking at
the source for win32ui.dll the menu routines call the C API directly,
bypassing MFC, so I would have thought it was possible. 

	Dave K


[1] well, almost - further experimentation shows that it only closes
the menu if I click within the Pythonwin window, rather than anywhere
on the screen. 




From kumar*xspam* at *xspam*rtp.ericsson.se  Fri Jun  4 16:03:40 1999
From: kumar*xspam* at *xspam*rtp.ericsson.se (Kumar Balachandran)
Date: 04 Jun 1999 16:03:40 -0400
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7iqhkb$d7s@wiscnews.wiscnet.net> <7iu7ma$d33$1@cronkite.cc.uga.edu> <02hfonzsx2.fsf@rtp.ericsson.se>
Message-ID: <02btevzqar.fsf@rtp.ericsson.se>

But it does allow you to forget an indentation or a blank line,
something that can happen very easily. The point I am trying to make
is to allow optional use if the user desires more readability, and to
accept whitespace if the user is from the planet Vulcan :-)

>>>>> "Charles" == Charles G Waldman <cgw at fnal.gov> writes:

    Charles> Kumar Balachandran writes:
    >> why not have optional syntactic sugar such as <snip>

    Charles> Because this introduces new keywords into the language,
    Charles> which nobody wants to do, especially if it introduces no
    Charles> new functionality.





>>>>> "Kumar" == Kumar Balachandran <kumar*xspam*@*xspam*rtp.ericsson.se> writes:

>>>>> "Graham" == Graham Matthews <graham at sloth.math.uga.edu> writes:

    Graham> Graham Matthews wrote in message
    Graham> <7ik6mi$lbk$1 at cronkite.cc.uga.edu>...  You are
    Graham> envangelising (a not uncommon response when someone
    Graham> remotely criticises Python). Stop evangelising and start
    Graham> considering the technical issues involve (read (!) other
    Graham> posts for what those issues are).
 
    Graham> graham
    Kumar> Here is some useful evangelizing (methinks). One of the
    Kumar> irritating things about Python is the use of whitespace in
    Kumar> syntax. Agreed, the code is readable without parentheses or
    Kumar> braces, but why not have optional syntactic sugar such as

    Kumar> if ...  elif ...  else ...  fi

    Kumar> def ...  ...  fed


    Kumar> while ...  elihw or wend



    Kumar> for ...  rof

    Kumar> etc. It makes the language more elegant. When I see code
    Kumar> using indentation or blank likes to achieve blocking of
    Kumar> constructs, it reminds me of an old language (FORTRAN I
    Kumar> think it was called:-).

    Kumar> The change is simple to achieve if backward compatibility
    Kumar> is given to accomodate people that grew up with FORTRAN.





From aaron_watters at my-deja.com  Sun Jun  6 14:59:42 1999
From: aaron_watters at my-deja.com (aaron_watters at my-deja.com)
Date: Sun, 06 Jun 1999 18:59:42 GMT
Subject: JPF1 pictures available
References: <mt2.0-23205-928649985@news.informatik.uni-bonn.de>
Message-ID: <7jeger$42s$1@nnrp1.deja.com>

In article <mt2.0-23205-928649985 at news.informatik.uni-bonn.de>,
  gherman at europemail.dropthis.com wrote:
> Hello,
>
> please have a look at a few pictures I've taken during the JPF1...

Thanks! This inspired me to document "after JPF" using my scanner
and photoshop enhancements in some cases.  Thanks to Marc Poinot
again for making all the arrangements for a highly successful
French Python Day, and of course for inviting me :) :) :).

  http://www.chordate.com/jpf/

   -- Aaron Watters


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From greg.ewing at compaq.com  Wed Jun 16 17:46:08 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Thu, 17 Jun 1999 09:46:08 +1200
Subject: waste of resources ?
References: <375C0F91.1C73E4E3@icrf.icnet.uk> <87r9nl398o.fsf@psyche.evansnet> <375E86C1.73B2CF80@icrf.icnet.uk> <37606565.5A93BCE9@compaq.com> <aahzFD6CI1.1qD@netcom.com>
Message-ID: <37681B20.BDCB6062@compaq.com>

Aahz Maruch wrote:
> 
> I can't speak to Arne's situation, but I'm about to rewrite a chunk of
> code to fork off a specific single child process precisely because I
> *don't* want the parent to block while this code executes, but I still
> want to reap the child later (and perform some final cleanup).

If you have some way of knowing when the child has finished
doing its thing (such as having received results from it),
you can do a blocking wait for that particular pid at that
point, because you know the child has either exited already
or is about to do so very soon.

If you have no idea when the child is expected to exit, your
only portable option is a SIGCHLD handler which loops doing non-blocking
waits until no more children are reaped. The loop is necessary
because if more than one child exits before the handler is
entered, you won't get any extra signals.

Greg



From arcege at shore.net  Tue Jun 22 10:04:34 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 22 Jun 1999 14:04:34 GMT
Subject: More Problems With Curses
References: <376EC592.A97D117D@sdofi.com>
Message-ID: <SJMb3.912$7X1.231730@news.shore.net>

ATS Public Labs <xcvx at sdofi.com> wrote:
: Hello,

: I'm trying to write an IRC client in Python using the curses and
: threading modules.  Basically one thread will read user input while the
: other threads print the messages received from the servers the user is
: connected to.   This works fine if I use print() and input() instead of
: curses, but it's annoying because the other threads will print stuff to
: the screen in the middle of what you are typing.  However, when I try to
: use curses instead, only the user-input thread seems to have control of
: the screen; the other threads will hang and not print anything until you
: press return.

: I wrote the following to try to figure out what's going on:


: import time
: import threading
: import curses

: def repd():
:   x=0
:   while 1:
:     time.sleep(1)
:     print 'x=%d' % x
:     x=x+1

: thr=threading.Thread(target=repd)
: thr.start()
: while 1:
:   time.sleep(1)
:   str=raw_input()
:   print 'got %s' % str


: This prints out 'x=0', 'x=1' etc. at one second intervals, and in
: addition prints whatever you type as soon as you press return.  However,
: if I add the following two lines:

: stdscr=curses.initscr()
: curses.endwin()

: right before I start the thread, then what happens is that 'x=0' gets
: printed, and then nothing will get printed again until you type
: something and press return, at which point all the 'x=n' messages which
: you should have seen get printed all at once.  So, just the act of
: initializing curses causes this annoying blocking behavior which I don't
: want.

: So, what am I doing wrong?  How can I get curses and threading to play
: nice with each other?

Curses was designed before threading (and before the "modern thought"
of event loops).  Curses is based on reading and writing from the same
file descriptor.  So it is better to "block" on a poll to see if there
is data there before calling the curses routines (with select).

You might want to look at how I did it in my own alpha version of an
IRC client:
  ftp://starship.python.net/pub/crew/arcege/Pyirc-1.5.1.tar.gz

  -Arcege




From news at dorb.com  Tue Jun 15 00:17:00 1999
From: news at dorb.com (Darrell)
Date: Tue, 15 Jun 1999 00:17:00 -0400
Subject: NEWBIE: opaque error message
References: <14181.50740.800169.537860@lrz.de>
Message-ID: <Evk93.2122$h14.10498@newsr2.twcny.rr.com>

A hint.

>>> import copy
>>> class X:
...     def __deepcopy__(self):
...             pass
...
>>> x=X()
>>> copy.deepcopy(x)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "d:\python\Lib\copy.py", line 151, in deepcopy
    y = copierfunction(x, memo)
  File "d:\python\Lib\copy.py", line 221, in _deepcopy_inst
    return x.__deepcopy__(memo)
TypeError: too many arguments; expected 1, got 2
>>>

--
--Darrell





From Holger.Jannsen at p98.f112.n480.z2.fidonet.org  Wed Jun 30 10:11:28 1999
From: Holger.Jannsen at p98.f112.n480.z2.fidonet.org (Holger Jannsen)
Date: Wed, 30 Jun 1999 15:11:28 +0100
Subject: Why must path be in args in os.execv?
Message-ID: <000015f4@bossar.com.pl>

From: Holger Jannsen <holger at phoenix-edv.netzservice.de>

High,

there's again something I don't understand. Hope you do.;-)

I'd like to start a windows-executable with some arguments.
Ok. 'os.execv' or anything like that is the thing.

But if I don't repeat the path with the arguments, the execution fails.
Better: Something happens, but not the correct execution.

For testing I constructed an easy batch named test.bat with only one command:
echo %0 %1 %2 %3 %4 %5 >result.log

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import os
>>> os.chdir("c:\\mypython")
>>> os.execv("test.bat",["test.bat","1","2","3","4"])

When I start with:
>>> os.execv("test.bat",["1","2","3","4"])
only the interpreter ends. No Log-File has been written...

I really spent a long time since I found out. Perhaps the documentation
have to be changed?
>>>
execv (path, args) 
Execute the executable path with argument list args, replacing the current
process (i.e.,
the Python interpreter). The argument list may be a tuple or list of strings.
Availability: Unix, Windows. 
>>>

Ciao,
Holger



From yopen at my-deja.com  Sat Jun 26 13:44:11 1999
From: yopen at my-deja.com (Bill Wilkinson)
Date: Sat, 26 Jun 1999 17:44:11 GMT
Subject: Python for large projects?
References: <3774b5d1.690922@news.force9.net>
Message-ID: <7l33ha$458$1@nnrp1.deja.com>

In article <3774b5d1.690922 at news.force9.net>,
  joechan25 at hotmail.com (Joe Chan) wrote:
> Hi, I've been doing mostly desktop and Client/Server development work
> but I'm considering how best to develop web-based and distributed
> projects.
>
> I read a paper on the Python.org site comparing Java and Python and
> was impressed. I'm interested in other peoples views and experiences
> of using Python for: large projects; team developments; distributed
> objects; intranet development etc.
>
> TIA

Hello Joe

I am using Python for a large  web enabled simulation/planning model.
The project is going wonderfully.  Keeping it brief, here are the
features that my model would not have if I had chosen other tools I was
considering:

Multiple Interfaces: Microsoft Access, Web , Command line,TK

Multiple Platform: WinNT and SGI Origin (Irix) ->
     There will also be a smaller version that will run on WinCE

Numerous Options:  The productivity boost allows me to focus on
business logic. This allows me to add more features and options to my
model.

Increased Accuracy:  The reduced development time allows me and my co-
workers to do a great deal of testing without disrupting our schedule.

I have also found that using Python softens some of the blows during
the requirement gathering phase.  It is nice to be able to put together
a quick mock up of an application component to help your customers
clarify what their requirments are.  Also when you find that you have
perhaps erred in your direction, it hurts alot less to throw away two
hours of work than twenty.


Hope this helps,

Bill Wilkinson


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From morse at harborcom.net  Wed Jun  2 08:17:08 1999
From: morse at harborcom.net (Kevin Dahlhausen)
Date: Wed, 02 Jun 1999 12:17:08 GMT
Subject: All this self dot is making me...
References: <Vn653.468$HD.99317@typhoon-sf.snfc21.pbi.net> <m33e0a52x2.fsf@chinon.cnrs-orleans.fr>
Message-ID: <37551fa6.90445994@news.oh.verio.com>

>Calling this parameter "self" is just convention, you can use whatever
>you like. However, it's a convention that just about everybody follows,
>so I'd think twice before doing something different.

A couple of options come to mind:

Or call it something short - s. ? - and use a (careful) global search and
replace when  you're done with the coding.    I sometimes use this
for long identifiers.  

Or in VIM set up a mapping:
 :map! s. self.




From claird at Starbase.NeoSoft.COM  Tue Jun 29 11:37:52 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 29 Jun 1999 10:37:52 -0500
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <xPPc3.535$or1.22866@typhoon-sf.snfc21.pbi.net> <xJRc3.292$Of3.67094@news.shore.net> <00f701bec212$f6a83050$f29b12c2@secret.pythonware.com>
Message-ID: <7lap8g$cfj$1@Starbase.NeoSoft.COM>

In article <00f701bec212$f6a83050$f29b12c2 at secret.pythonware.com>,
Fredrik Lundh <fredrik at pythonware.com>, who is very trustworthy on
these matters, wrote:
>Michael P. Reilly <arcege at shore.net> wrote:
			.
			.
			.
>What native API?  As of 8.0, Tk still emulates all widgets
>except the scrollbar, and use a somewhat ineffective
>X emulation layer for graphics.  AFAIK, they haven't
>changed that in 8.1...
			.
			.
			.
No, and no upgrades from Scriptics seem imminent.  See, however,
<URL:http://www.sunworld.com/swol-05-1999/swol-05-regex.html> for
an "outsider's" initiative in this area.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From bkxuof at hotmail.com  Thu Jun 10 03:21:50 1999
From: bkxuof at hotmail.com (bkxuof at hotmail.com)
Date: Thu, 10 Jun 1999 07:21:50 GMT
Subject: Free Sex Links!!  9824
Message-ID: <iIJ73.55784$pl3.56342@newreader.ukcore.bt.net>

For Nasty Sex links visit:

http://website.lineone.net/~simonsi/sexlinks.htm
hnrwcyylfkqvcrdnz




From phxken at netzone.com  Mon Jun 21 21:22:17 1999
From: phxken at netzone.com (Farrell Kenimer)
Date: Tue, 22 Jun 1999 01:22:17 GMT
Subject: grail browser failure
References: <376D4A9E.B9B90C3D@netzone.com> <14190.20502.944277.743491@weyr.cnri.reston.va.us>
Message-ID: <376EE3E9.A8307239@netzone.com>

"Fred L. Drake" wrote:

> Farrell Kenimer writes:
>  > fails with something like isabs not available
>  > I can import grail and after a try or two I can also import Browser
>  > How would one try to run grail after it's imported?
>
>   There's a patch for this on the Grail Web site.
>
>   -Fred
>
> --
> Fred L. Drake, Jr.           <fdrake at acm.org>
> Corporation for National Research Initiatives

PLease excuse previous complaint.  Read the April 99 archive of the grail
maillist and you had already
found that grailutil.py was incorrect in two places where the os.path had
been left off isals() and
normpath().

It's running now.
Is there a way to establish a display outside of X in Linux to run grail
from the console or does it require an
X server?  Thanks.

--
Farrell Kenimer of Phoenix, Arizona  USA


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/19990622/29ab4b95/attachment.html>

From MHammond at skippinet.com.au  Tue Jun 29 19:00:20 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Wed, 30 Jun 1999 09:00:20 +1000
Subject: Python, ASP - Namespace (again)
References: <O18e3.449$U%.3201@news1.online.no>
Message-ID: <7lbj2t$2es$1@m2.c2.telstra-mm.net.au>

Nikolai Kirsebom wrote in message ...
...
>when I run the statements in the script1.asp file in an interactive window,
>I can create the KSdebatt object, however when I run the script on the
>server - I  get the following message:
>
>Python ActiveX Scripting Engine error '80020009'
>Traceback (innermost last): File "<Script Block >", line 4, in ?
>o=KSdebatt() NameError: KSdebatt
>/ksdebatt/pages/Script1.asp, line 6
>o=KSdebatt()
>
>If it is possible I would like to 'hide' the object (o) inside the KSdebatt
>module - that is to have the instanciation inside the module itself - but
>still be able to access it from the ASP-code in the script (o.Attr....).
Is
>this possible ?

There is no obvious reason to me why this wont work.  Note that if you are
not running win32all-125, then I recommend you upgrade - there have been
some fixes (but none that adddress this, as far as I can tell).

The best I can suggest is that you use the win32trace module and print
statements to work out what the hell is going on.

Eg, making your first line "import win32traceutil" will mean that all print
statement go to a debugging collector.  You can view the print statements by
starting Pythonwin on the same machine, and selecting "Tools/Trace Collector
Debugging Output".

Then use lots of print statements - specifically print dir() etc, to see
exactly what the namespace is...

Mark.






From eric at userland.com  Thu Jun 17 17:25:29 1999
From: eric at userland.com (Eric Kidd)
Date: Thu, 17 Jun 1999 17:25:29 -0400
Subject: XML-RPC for Zope
Message-ID: <19990617172529.E14374@moebius.dartmouth.edu>

Hello! Dave Winer of UserLand software (my boss) has generously allowed me
to add XML-RPC support to Zope. Since I'm lazy--and Fredrik Lundh has
already written a wonderful XML-RPC module for Python--my code simply
hooks Fredrik's routines into Zope's server architecture. The hard work
was done by the following people:

 * Jim Fulton of Digital Creations, who added XML-RPC hooks to ZPublisher
   and answering my stupid questions.
 * Fredrik Lundh of PythonWare, who wrote an excellent Python XML-RPC module.
 * Dave Winer of UserLand Software, who wrote the XML-RPC spec (and who
   pays my salary).

Please give all your thanks to these folks; they really *did* do all of the
work. However, I may have plugged their components together incorrectly, so
send all the blame and hate mail to me. :-)

What is XML-RPC?
----------------

XML-RPC is a very simple RPC protocol that runs over HTTP. The function
name and arguments are encoded as XML. XML-RPC isn't the fastest way to
call a function on remote server, but it might be one of the easier to
understand. (See http://www.xmlrpc.com/ for details.)

An XML-RPC request looks like this:

  POST /z2/ 1.0
  Content-Type: text/xml
  Content-Length: 170

  <?xml version='1.0'?>
  <methodCall>
    <methodName>test.string_length</methodName>
      <params>
        <param>
          <value><string>foo</string></value>
        </param>
      </params>
  </methodCall>

The response looks like this:

  HTTP/1.1 200 OK
  Content-Length: 122
  Content-Type: text/xml

  <?xml version='1.0'?>
  <methodResponse>
    <params>
      <param>
        <value><int>3</int></value>
      </param>
    </params>
  </methodResponse>

Fredrik Lundh of PythonWare wrote a very nice XML-RPC library for
Python. (See http://www.pythonware.com/products/xmlrpc/ to get a copy.)
Using his code, I could call the above function as:

  >>> import xmlrpclib
  >>> s = xmlrpclib.Server("http://localhost/z2/")
  >>> s.test.string_length("foo")
  3

In other words, XML-RPC lets you treat a remote object almost like a local
Python object. Think of it as a super-simple cousin to CORBA and DCOM, and
you'll get the right idea.

XML-RPC for Zope
----------------

Writing XML-RPC methods is just as easy, thanks to Zope. Simply load a
Python function as an External Method, and you're ready to go:

  def string_length(string):
    return len(string)

Easy, huh? To use XML-RPC for Zope, you'll need to get the latest version
of Zope from CVS and read the release notes on UserLand's site:

  http://linux.userland.com/stories/storyReader$18

Future Directions
-----------------

Zope XML-RPC still has a few rough edges. If you discover any bugs which
aren't mentioned on the site, or have any ideas for improvement, drop me a
line.

Thank you to all the people who helped out!

Cheers,
Eric Kidd
UserLand Software



From schorsch at schorsch.com  Sat Jun 12 15:03:11 1999
From: schorsch at schorsch.com (Georg Mischler)
Date: Sat, 12 Jun 1999 19:03:11 GMT
Subject: do anonymous pipes normally work on NT?
Message-ID: <7juatc$fef$1@nnrp1.deja.com>

'''
Hi all,

I seem to be unable to do anonymous pipes right on NT4sp3.
This may be more of a Windows than a Python related
question, but I'll try it here for a start anyway...

The code below was created with the help of an example
posted here some time ago by Bill Tut, a C++ example by
someone else, and after studying the docs to the win32
extensions by Mark Hammond, as well as the documentation
in VisualStudio for VC++5.

To my uninitiated eyes, everything looks as if it would
make sense, and corresponds to similar code for unix,
using fork() and execv().

The problem is, that in the case here, I can write
to the stdin pipe, but reading from stdout or stderr
will get no results at all. All I see is an exception
eventually, telling me:
  '(109, 'ReadFile', 'Die Pipe wurde beendet')
'The pipe was terminated'.

When calling a program from cygwin (b19), instead
of listening to lines of input and answering with
its results, it will just die and nothing else
will happen. This is normally the same if I call a
program from windows, except for example with ftp.exe,
which starts an infinite loop and eats all cpu. I don't
know if this is a related problem or just a ftp.exe
weirdness.

An example invocation would be:

d:\somewhere\> python
>>> import ntprocess
>>> p = ntprocess.Coprocess('c:\\winnt\\system32\\ping.exe -t
localhost')
>>> print p.read()
# nothing happens until I kill ping from the task manager:
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "ntprocess.py", line 111, in read
    res, str = win32file.ReadFile(self.hStdout, 0)
pywintypes.api_error: (109, 'ReadFile', 'Die Pipe wurde beendet.')

Now it may be that the nt console programs don't
write to stdout but just directly to the console.
But my cygwin stuff certainly *does* write to stdout
and stderr. Are there any problems known when
combining native NT Python with cygwin binaries?
Am I doing aything else wrong?

Since I have *no* idea where to look, I can't help
but just include the complete source below.
If anyone knowing more about this stuff could
take a short look at it and point me to the blatant
error I made, I'd be extremely happy.


Thanks for any insights!

-schorsch

'''
# ntprocess.py

import types
import os
import sys
import string

import msvcrt
import winnt
import win32api
import win32pipe
import win32file
import win32process
import win32event
import win32security


class Coprocess:
    def __init__ (self, cmd, env={}):
        self.running = 0
        self.exitstatus = None

        if type(cmd) == types.StringType:
            cmd = string.split(cmd)
        executable = cmd[0]
        self.args = cmd[1:]
        self.executable = os.path.normpath(executable)

        # security attributes for created pipes
        sAttrs = win32security.SECURITY_ATTRIBUTES()
        sAttrs.bInheritHandle = 1

        # redirecting child STDOUT:
        # XXX redirection should actually be checked before
        # cmd is split.
        # XXX Right now, the '>' must be framed by whitespace
        # in a string.
        if '>' in self.args:
            redirpos = self.args.index('>')
            if len(self.args) < redirpos + 2:
                raise IOError, 'Empty Redirect.'
            if len(self.args) > redirpos + 2:
                raise IOError, 'Ambiguous Redirect.'
            outfn = self.args[-1]
            outfd = os.open(outfn,
                    os.O_WRONLY | os.O_CREAT | os.O_TRUNC, 0664)
            hStdout_w = msvcrt.get_osfhandle(outfd)
            self.args = self.args[:-2]
            hStdout_r = None
            self.hStdout = hStdout_r
        elif '>>' in self.args:
            redirpos = self.args.index('>>')
            if len(self.args) < redirpos + 2:
                raise IOError, 'Empty Redirect.'
            if len(self.args) > redirpos + 2:
                raise IOError, 'Ambiguous Redirect.'
            outfn = self.args[-1]
            outfd = os.open(outfn,
                    os.O_WRONLY | os.O_CREAT | os.O_APPEND, 0664)
            hStdout_w = msvcrt.get_osfhandle(outfd)
            self.args = self.args[:-2]
            hStdout_r = None
            self.hStdout = hStdout_r
        else:
            hStdout_r, hStdout_w = win32pipe.CreatePipe(sAttrs, 0)
            self.hStdout = hStdout_r

        # child STDIN:
        hStdin_r, hStdin_w = win32pipe.CreatePipe(sAttrs, 0)
        self.hStdin = hStdin_w

        # child STDERR:
        hStderr_r, hStderr_w = win32pipe.CreatePipe(sAttrs, 0)
        self.hStderr = hStderr_r

        # create environment for new process.
        newenv = os.environ.copy() # use parent environment.
        newenv.update(env) # add/override from caller data.

        # set up the command line.
        sCmdLine = self.executable + ' ' + string.join(self.args)

        # set the info structure for the new process.
        StartupInfo = win32process.STARTUPINFO()
        StartupInfo.hStdInput  = hStdin_r
        StartupInfo.hStdOutput = hStdout_w
        StartupInfo.hStdError  = hStderr_w
        StartupInfo.dwFlags = win32process.STARTF_USESTDHANDLES

        # start the process.
        hProcess, hThread, dwPid, dwTid = win32process.CreateProcess(
                self.executable, sCmdLine, # what to start
                None,    # process security attributes
                None,    # thread attributes
                1,       # inherit handles, or USESTDHANDLES won't work.
                         # creation flags. Don't access the console.
                win32process.DETACHED_PROCESS,
                newenv,  # new environment
                None,    # current directory (stay where we are)
                StartupInfo)
        #print hProcess, hThread, dwPid, dwTid
        self.hProcess = hProcess
        self.hThread = hThread
        self.pid = dwPid
        self.tid = dwTid
        self.running = 1

    def write(self, str):
        err, bytes = win32file.WriteFile(self.hStdin, str)
        return bytes

    def read(self):
        if not self.hStdout: raise IOError, 'Error: output redirected.'
        res, str = win32file.ReadFile(self.hStdout, 0)
        if res == 0: return str
        else: return ''

    def stderr_read(self):
        res, str = win32file.ReadFile(self.hStderr, 0)
        if res == 0: return str
        else: return ''

    #XXX we'd probably need to implement line buffering ourselves
    #def readline(self):
    #    return ''

    #def stderr_readline(self):
    #    return ''

    def flush(self):
        win32file.FlushFileBuffers(self.hStdin)

    def waitpid (self, opt=0):
        wres = win32event.WaitForSingleObject(self.hProcess, opt)
        # XXX do we want to check the status if we timed out?
        status = win32process.GetExitCodeProcess(self.hProcess)
        self.exitstatus = status/256
        self.running = 0
        return (self.pid, status)

    def kill (self, signal=9):
        self.close()
        if self.running:
            win32process.TerminateProcess(self.hProcess, -9)
            return self.waitpid(0)
        else:
            return 0

    def __del__(self):
        '''close all handles when our instance is garbage collected.'''
        self.close()

    def close(self):
        # close handles defensively, in case we fail early...
        if hasattr(self, 'self.hStdin'):
            win32api.CloseHandle(self.hStdin)
        if hasattr(self, 'self.hStdout'):
            win32api.CloseHandle(self.hStdout)
        if hasattr(self, 'self.hStderr'):
            win32api.CloseHandle(self.hStderr)
        if hasattr(self, 'self.hProcess'):
            win32api.CloseHandle(self.hProcess)
        if hasattr(self, 'self.hThread'):
            win32api.CloseHandle(self.hThread)


--
Georg Mischler  --  simulation developper  --  schorsch at schorsch.com
+schorsch.com+  --  lighting design tools  --  http://www.schorsch.com/


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From kas at maps.magnetic-ink.dk  Mon Jun 21 04:47:25 1999
From: kas at maps.magnetic-ink.dk (Klaus Alexander Seistrup)
Date: Mon, 21 Jun 1999 10:47:25 +0200
Subject: Linking an Interpreter dynamically
References: <slrn7mrpes.94.tuna@delirious.fsnnet>
Message-ID: <376DFC1C.57BAB8AA@maps.magnetic-ink.dk>

Lloyd Weehuizen wrote:

> I was just wondering if it is possible to create a dynamically linkable
> python interpreter? I've looked in the docs, but can't seem to find
> anything, and "make sharedinstall" doesn't seem to give me any linkable
> libraries??

So what you want is a shared python library?  If so, I managed to do it
sometime back, although in a kludgy manner:

First I made sure that Python was compiled using gcc's -fPIC flag and then
I manually made a shared library out of the resulting libpython1.5.a using
the --whole-archive option along with whatever it takes to make a shared
library with gcc (-shared -Wl,-soname,...), and I ended up with a shared
libpython.so.1.5.x that I installed manually.

Sorry for the clumsy description but it's quite a while since I made the
experiment (which succeeded, I ended up with a python executable less than
3 kbytes in size dynamically linked to the shared python library).

Hope this helps you a bit...


  //Klaus

-- 
???[ Magnetic Ink ]?????????????????????????????????????????????????????????
???[ http://www.magnetic-ink.dk/ ]??????????????????????????????????????????



From tim_one at email.msn.com  Tue Jun 29 22:57:04 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 29 Jun 1999 22:57:04 -0400
Subject: PySequence_Check
In-Reply-To: <slrn7nhjv9.lt6.rasumner@bach.wisdom.weizmann.ac.il>
Message-ID: <000401bec2a4$3be15200$f39e2299@tim>

[Reuben Sumner]
> How can I do the equivalent of a PySequence_Check in Python?
> GvR told me that the question was illdefined (if I explained it correctly)
> and to ask here.  It seems that some of the functions in abstract.h
> are available from python (len(), callable()) and others are not
> (PySequence_Check, PyNumber_Check).
>
> What am I missing?

For starters, unquestioning faith in Guido's pronouncements <wink>.

> I have a nested structure of lists and tuple and I just want to know
> when I have bottomed out.

Then why not check for that directly and be done with it?

> I can check for tuple or list indivually but that is ugly and
> inflexible.

Not to mention simple, obvious and unsurprising <wink>.

> I could just try indexing and see if there was an exception but that
> is not much better.

OK, you tell me:  what *is* a sequence?  Answer that, and you'll know how to
check for it <0.5 wink>.  Before answering too fast, note that mapping types
can be indexed without raising an exception, and so can strings.  Do you
want your nest of "lists and tuples" *not* to "bottom out" at a string?
Probably not -- but if so, you're not really interested in testing whether
it's a sequence!  Note too that user-defined classes can support the
__getitem__ protocol without supporting the __getslice__ protocol, or vice
versa.  Are either of those sequences, or do they need to support both?
Whichever way you answer, half the world will disagree.

I think that, in most cases, when people say "sequence":  (A) it's
ill-defined <wink>; and, (B) after five rounds of tedious questioning they
end up deciding they mean "responds to slicing but isn't a string".  So
there you go:

def issequence(x):
    try:
        x[0:0]
    except:
        return 0
    return type(x) is not type("")

may-not-match-what-you-mean-by-"sequence"-ly y'rs  - tim





From hinsen at cnrs-orleans.fr  Wed Jun  2 06:10:01 1999
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 02 Jun 1999 12:10:01 +0200
Subject: All this self dot is making me...
References: <Vn653.468$HD.99317@typhoon-sf.snfc21.pbi.net>
Message-ID: <m33e0a52x2.fsf@chinon.cnrs-orleans.fr>

"Jim Rudnicki" <jdrudnicki at yahoo.com> writes:

> So what has me perplexed is why all the self. 's?  Is this a namespace issue
> or maybe just to expensive to do at runtime?

First, I find the immediate visual distinction between local variables
and object attributes helpful. Of course your opinion can differ!

Second, in the absence of variable declarations, Python has no way
to find out what is local and what is an attribute. And this isn't made
easier by the fact that attributes can be created dynamically at any
time.

> For those of us who have less cartilage left in their hands, would it be
> acceptable to use "my." or better "me."  Or is self. strongly ingrained?

Calling this parameter "self" is just convention, you can use whatever
you like. However, it's a convention that just about everybody follows,
so I'd think twice before doing something different.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------



From tim_one at email.msn.com  Sun Jun 13 22:32:18 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 13 Jun 1999 22:32:18 -0400
Subject: Solution to finalisation problem [Re: fork()]
In-Reply-To: <376426EA.B6C9729E@compaq.com>
Message-ID: <000001beb60e$1fd4f500$559e2299@tim>

[Tim]
> the combo of finalizers, cycles and resurrection seems to be a
> philosophical mess even in languages that take it all <ahem> seriously.

[Greg Ewing]
> Indeed, which leads me to believe that the idea of giving objects
> a __del__ method is fundamentally flawed in the first place.

Except it works so nice in "the usual" case (no cycle, no resurrection):
everyone agrees on what should be done then, and it's remarkably
surprise-free!

Toss resurrection in, and it's still not *much* of a debate:  Python invokes
__del__ each time the refcount hits 0, Java does it only once, and the
choice seems arbitrary in the sense that there's no apparent killer argument
either way.

It's cycles that break its back, in Python and Java:  the language is then
in the business of choosing an order in a context where it can't possibly be
smart enough to choose wisely.  Java defines just enough to keep its own
internals sane, explicitly warning the programmer that it's not going to be
of much use to them.  Guido wisely picked reference-counting so Python
didn't have to disappoint users similarly <wink>.

Anyway, I'm not sure __del__ deserves the blame for not being able to handle
every insane complication a user can throw at it.  Like many another
feature, it's *fine* for what it's designed for, and simply can't be pushed
beyond that.

Hisao Suzuki earlier posted a link to an excellent paper describing proposed
"guardians" in Scheme.  Like many things Schemeish, it defines a perfectly
general mechanism and absolutely no policy.  Want to finalize cycles?  Fine,
here's a hammer, build it yourself.  Want to clean up I/O channels?  Fine,
be our guest.  20,000 nails, and one hammer to pound them all in one at a
time <wink>.

> Fortunately, there *is* a way to do finalisation that avoids all
> these problems. Instead of an object doing its own finalisation,
> it designates some *other* object to do its finalisation on its
> behalf after it is dead.
>
> The benefits of this are:
>
> (1) The object doing the finalisation is fully alive and operates
> in a predictable environment.
>
> (2) The object which triggered the finalisation is fully dead
> (its memory has already been reclaimed by the time the finaliser
> is called) and there is no possibility of it being resurrected.

This appears isomorphic to the "register-for-finalization" approach dissed
in the aforementioned Scheme paper.

I'll leave it to Schemers to argue the Scheme case, but at least in Python
#2 is a drawback to people who use resurrection on purpose, e.g. to avoid
high initialization costs for large objects connecting to the outside world.
A __del__ in that case can consist of just enough to close the connection,
then put the expensively constructed object on a global "free list" for
immediate reuse (one reason to prefer Python's treatment of resurrection
over Java's -- in this case you definitely *want* __del__ to get called each
time the object is about to die).

> In Python, it might look something like this from the
> user's point of view:
>
> class FileWrapper:
>    # Example of a class needing finalisation.
>    # Has an instance variable 'file' which needs
>    # to be closed.
>    def __init__(self, file):
>       self.file = file
>       register_finaliser(self, FileWrapperFinaliser(file))
>
> class FileWrapperFinaliser:
>    def __init__(self, file):
>       self.file = file
>    def __finalise__(self):
>       self.file.close()
>
> In this example, register_finaliser() is a new built-in
> method which stores the object and its finaliser in some
> special global dict. Whenever an object is reclaimed (either
> by refcount or M&S) the dict is checked for a finaliser for
> that object. If one is found, its __finalise__ method is
> called and it is removed from the dict.

Presumably this needs to be some form of weak dict, else the very act of
registering would render the object immortal ... OK, that's easily doable
under the covers.

> Note that the finaliser object shouldn't be given a reference
> to the original object (doing so would prevent it from ever
> becoming unreachable), but only enough information to enable
> the finalisation to be carried out.

In an object with mutable state this can be quite inconvenient, don't you
think?  Seems the "interesting state" would need to be factored out into
another container, shared by the object and its finalizer, referenced
indirectly by both.

> If a scheme like this were adopted, it should completely
> replace the existing __del__ method mechanism, so that there
> would be no difference between the finalisation of cyclic
> and non-cyclic trash. As such it would have to wait for
> Python 2.0.
>
> A-final-solution-to-the-finalisation-problem,
> Greg

Seems OK and doable so far as it goes.  Doubt people would like giving up
the easy convenience of __del__ in the vast bulk of cases where it's
non-problematic.  In a sense, Python (& Java) get in trouble now by trying
to provide policy as well as mechanism; I've become convinced that there is
no sensible policy in the presence of cycles, so punting on policy there is
fine; but don't want the language to make users "build it all themselves" in
the simpler cases where a reasonable policy is clear.

don't-throw-the-del-out-with-the-cycle-water-ly y'rs  - tim





From graham at sloth.math.uga.edu  Tue Jun  1 11:39:27 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 1 Jun 1999 15:39:27 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
Message-ID: <7j0urf$2dm$3@cronkite.cc.uga.edu>

Paul Boddie (paulb at infercor.no) wrote:
: As has been discussed before on this newsgroup/list, such resources get
: overlooked in many a garbage collection scheme. I'm sure a perusal of various
                ----
: news/list archives would provide a substantial amount of relevant literature.

That word "many" is your key Paul. Just because many a garbage collection
scheme works that way doesn't mean all do, or all have to. It's entirely
an implementation dependent thing.

You are creating a strawman argument that GC cannot properly free resources
like files. But some GC do just that so what is your point?

graham


-- 
      This mountain is not visible from any inhabitable place. 
       It has no name, and since it's discovery in 1856 by a 
      British survey team, it has mysteriously resisted every
    attempt to name it. The mountain is known simply and starkly
     by it's original designation on the surveyor's map ... K2.



From dieter at handshake.de  Tue Jun  1 15:21:53 1999
From: dieter at handshake.de (Dieter Maurer)
Date: 01 Jun 1999 21:21:53 +0200
Subject: Tkinter-patch for use with Tk8.1 (Windows)
References: <x7g14g9a37.fsf@lindm.dm>
Message-ID: <x7so8bbub2.fsf@lindm.dm>

The attached patch is necessary (together with the patch
posted on 28 May 1999) for Tkinter to work with Tk8.1
under Windows (tested under Windows/NT with Tk8.1.1).

- Dieter
-------------- next part --------------
A non-text attachment was scrubbed...
Name: tkinter8.1_2.pat
Type: application/x-patch
Size: 359 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/19990601/2f58cd36/attachment.bin>

From mikael at pobox.com  Mon Jun 28 13:08:39 1999
From: mikael at pobox.com (Mikael Lyngvig)
Date: Mon, 28 Jun 1999 17:08:39 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <7l4nht$jl7$1@nnrp1.deja.com>
Message-ID: <9DDBCC52A7B26002.649D772F7232D418.CE2709D4B08E7401@lp.airnews.net>

On Sun, 27 Jun 1999 08:31:56 GMT, Andrew Cooke 

>1. Speed.  For most applications, is this an issue?  Python is, above
>all, a scripting language.  Speed is not important.  What you can do -
>and do fairly simply with Tk, once you know it, is build a GUI front-
>end to whatever programs you are tying together.  The text widget, for
>example, is amazingly flexible with very little code.

I don't agree that speed is not important, because Python is a
scripting language.  The really great thing about Python, IMHO, is
that it actually performs well enough that you're rarely affected by
the fact that it's an interpreted language - especially because the
standard libraries offers native implementations of many CPU intensive
tasks.  When this is an issue, you have the option of creating C
support modules.

I believe that Python is fast enough that you can use it to create
"commercial grade" applications, if only the GUI part does not reveal
that the application wasn't written in C. 

(I guess this depends on the tasks you do with Python; for me it is
usually I/O bound tasks, so the speed of the interpreter is of little
concern to me)

>If you are programming games or graphics-intensive code then sure, use
>another (C-based) GUI - Python is designed to make this easy, just
>as you can drop into C if non-graphics code is CPU intensive.

I agree, but I'd never even think of doing a graphics-intensive
application in Python, except perhaps to prototype it.


-- Mikael



From emile at fenx.com  Tue Jun  8 18:04:13 1999
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 8 Jun 1999 15:04:13 -0700
Subject: In what order does Python read dictionary keys?
References: <7jjnt5$snk$1@nnrp1.deja.com>
Message-ID: <vqg73.1213$VL2.183595@news.direcpc.com>


Tim,

"Note that the printed representation for the dictionary lists the entries
in a different order than in the assignment because dictionaries are
unordered "sets" of pairs, and the implementation reserves the right to
print out the content in any order it pleases (so there!)."
    -From "Internet Programming with Python"
    -Watters, van Rossum & Ahlstrom

What you want to do is use sort:

>>> entries = {'red':'Type the red value, please: ',
'blue':'Type the blue value, please: ',
'green':'Type the green value, please: '}

>>> entrieskeys=entries.keys()
>>> entrieskeys.sort()

>>> for key in entrieskeys: # Now get a loop going

A related question for those in the know, is there a way to combine these
two steps into a single statement, along the lines of:

entrieskeys=entries.keys().sort()


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


----- Original Message -----
From: <tcpeter at my-deja.com>
Newsgroups: comp.lang.python
Sent: Tuesday, June 08, 1999 11:37 AM
Subject: In what order does Python read dictionary keys?


> Hello, all.  I've checked the docs and queried DejaNews for this but I
> couldn't come up with an answer.  I wrote a script to convert rgb
> values to hex for when I'm doing colors in HTML pages and don't have
> Photoshop handy.  The code appears below.  My question is, how does
> Python decide in which order to read the keys?  Notice that I have the
> dictionary set up in this order: red, blue, green.  (I mistyped it while
> writing the code).  But when I ran the code, the options were presented
> in the desired order. (A trifle creepy, if you ask me.  Turing would be
> delighted.)  Anyway, how does Python decide the order. Thanks.
>
...








From robert.yodlowski at intel.com  Fri Jun 11 14:12:20 1999
From: robert.yodlowski at intel.com (Robert Yodlowski)
Date: Fri, 11 Jun 1999 14:12:20 -0400
Subject: Love Python? Wanna $$ job? Have I got a deal for you!
Message-ID: <37615184.BC762479@intel.com>

Hello, I'm Bob Yodlowski.   robert.yodlowski at intel.com     (978)
568-4477

I work at Intel Corp.'s Network Communications Group in Hudson, MA, I
hope you'll want to also when I tell you about what they're foolish
enough to actually PAY me to do!

For the past 2 years, I've been using Python to create and run a
simulation environment to design a new network controller chip (sa1200).
The the second generation of the chip (sa2200) will be starting design
shortly.

The chips are BIG! - The sa1200 consists of a Strongarm RISC processor,
SIX! special purpose, microcoded communications processors, a PCI bus
interface, highly optimized (i.e. convoluted) interfaces to SDRAM, SRAM
and flash memories and a high speed, 64 bit bus interfacing to up to 6
MAC (fancy, ultra high speed UART's) that whip data back and forth to
and from stuff like FDDI, ATM, Ethernet, LAN, WAN or any other kind of
serial data transfer gadget.

I've had a ball! Thanks to the power of Python, just 3 of us have so far
been able to keep up with the needs of the project but, now, we need
more help! I've been able to use almost every aspect of Python:
multithreading, extending, embedding, pickling, run-time compiling even
a little Tkinter. We have LOTS more work that needs to be done and we
need the right people who love Python to help do it.

Intel is SERIOUS about networking! To see how serious, I refer you to
the May 3, 1999 issue of "Forbes" magazine - the one with Intel's
president, Craig Barrett, on the cover whose lead article "Reinventing
Intel"  starts on p155. That article, especially the comments on p157,
158 has led one of our marketing people to say "We intend to be the
moral equivalent of the Pentium (r) in the networking silicon market."

Even if you're just mildly interested, give me a call or send me email,
Central Massachusetts (30mi from Boston) is especially beautiful in the
summer!

Bob Yodlowski.   robert.yodlowski at intel.com     (978) 568-4477

P.S.

A colleague and I will be giving talks about our work at the OReilly
Open Source Software Convention in August. Go to the OReilly web site to
see the abstracts - if they're there at the moment - (they tend to
disappear from time to time).
http://conference.oreilly.com/python/overview.html#conf




From jhspies at futurenet.co.za  Tue Jun 29 15:03:55 1999
From: jhspies at futurenet.co.za (Johann Spies)
Date: Tue, 29 Jun 1999 19:03:55 GMT
Subject: Please help me to understand classes
Message-ID: <Pine.LNX.3.96.990629205752.3228A-100000@Johann>

I have been playing around with python for some time now, but I am a
relative newbie as it is my hobby and normally I have very little time for
programming.  I am on leave now and want to write a program but I find
python a very difficult language to comprehend despite the fact that I
have a few years of experience in programming a language like Pascal.
Maybe it is because I have little experience in OOP.

Can somebody explain to me why the following program produces no output:
---------------------------------------------
class A:

    def druk():
        print 'Jy is in Class A'

class B:

    def druk():
        print 'Jy is in Class B'
        
a = A()
c = B()
c.druk
a.druk
---------------------------------------------


 --------------------------------------------------------------------------
| Johann Spies                                 Windsorlaan 19              |
| jhspies at futurenet.co.za                3201 Pietermaritzburg             |
| Tel/Faks Nr. +27 331-46-1310		       Suid-Afrika (South Africa)  |
 --------------------------------------------------------------------------

     "One thing have I desired of the LORD, that will I seek
      after; that I may dwell in the house of the LORD all
      the days of my life, to behold the beauty of the LORD,
      and to inquire in his temple."            
                               Psalms 27:4       




From Sven.Drescher at dlr.de  Tue Jun 15 07:42:44 1999
From: Sven.Drescher at dlr.de (Sven Drescher)
Date: Tue, 15 Jun 1999 13:42:44 +0200
Subject: Tk Scrollbar
Message-ID: <37663C34.2957622F@dlr.de>

Hi!
I have only an easy question. I think it is possible to bind a scrollbar
and an other widget e.g. a frame with buttons. But I have not found any
literature or programming examples. I want to bind the scrollbar to an
frame and I don't know how. Please help me.
Thanks for links and Python-code or other help.

Sven




From loredo at spacenet.tn.cornell.edu  Wed Jun 23 17:02:09 1999
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Wed, 23 Jun 1999 17:02:09 -0400
Subject: Auto (as in *Car*) Python?
Message-ID: <37714B50.CE3F440F@spacenet.tn.cornell.edu>

Hi folks-

Just learned of this MP3 player for your car:

http://www.empeg.com/

Visit the "software" and "tech" pages:  It actually runs Linux, and uses Python
for the UI!

Cool!
Tom Loredo



From bgue at my-deja.com  Sat Jun  5 22:58:16 1999
From: bgue at my-deja.com (bgue at my-deja.com)
Date: Sun, 06 Jun 1999 02:58:16 GMT
Subject: format of event.char?
Message-ID: <7jco46$kpm$1@nnrp1.deja.com>

Hi everyone. I'm wondering what the format of event.char is...how can I
append it to a string? I've got a string, self.str, and a callback
bound to keystrokes:

def callback(self, event):
  self.str = self.str + event.char


...but this doesn't work. The exact error is (ommitting trace):

TypeError: neither _and_ nor _rand_ defined for these operands

So how does one use event.char for things other than printing?


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From mal at lemburg.com  Thu Jun 10 04:30:57 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Thu, 10 Jun 1999 10:30:57 +0200
Subject: COM dates
References: <816010E2456BD111A48700805FBBE2EEFDF354@ex-quebec-u1.baan.com>
Message-ID: <375F77C1.629FED7B@lemburg.com>

Gaetan Corneau wrote:
> 
> Another question: why no "strptime" function under WIN32? This function is
> not, as far as I know, difficult to implement.

You could check out mxDateTime. It has COM date support, strptime()
and also a much more powerful date/time parser:

	http://starship.skyport.net/~lemburg/mxDateTime.html

Not sure how you would convert the Win32 PyTime objects into
DateTime objects though. If they implement the int() APIs, then
DateTimeFromTicks(pyTime) will do, otherwise there's also a
constructor DateTimeFromCOMDate(comdate) which takes the raw
COM date float as argument.

Mark ?

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   204 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From gerrit.holl at pobox.com  Sat Jun 12 12:02:42 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sat, 12 Jun 1999 18:02:42 +0200
Subject: regular expression question: last occurence
Message-ID: <19990612180242.A6075@optiplex.palga.uucp>

Hello all,

I want to get the _last_ occurence of some string in some other string. What
I want exactly, is having the current directory (for example,
/home/gerrit/linuxgames/links/)
and than put everything after the last "/linuxgames" in a string. But when
someone is in the directory "/home/gerrit/linuxgames/cvs/linuxgames", my current
regexp is getting /cvs/linuxgames, and that's not what I want. Now, I have:

currentdir = os.getcwd()
mo = re.search('/linuxgames', currentdir)
eind = mo.end()
subdir = currentdir[eind:] + '/'

But that doesn't solve my problem.

kind regards,
Gerrit.


-- 
The Dutch Linuxgames website.		De Nederlandse Linuxgames pagina.
Everything about games on Linux.	Alles over spelletjes onder Linux.
Site address:				http://linuxgames.nl.linux.org
Personal homepage:			http://nl.linux.org/~gerrit/



From wtanksle at dolphin.openprojects.net  Thu Jun 17 18:53:30 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Thu, 17 Jun 1999 22:53:30 GMT
Subject: Solution to finalisation problem [Re: fork()]
References: <000401beb886$6a7f2b00$099e2299@tim> <376971C6.90976DD4@compaq.com>
Message-ID: <slrn7miv3a.s5g.wtanksle@dolphin.openprojects.net>

On Fri, 18 Jun 1999 10:08:06 +1200, Greg Ewing wrote:
>Tim Peters wrote:

>> Greg, this is kind of like asking "if each loop terminates, how can we have
>> an infinite loop?".  That is, sure, if you write code that works, it works
>> <wink>.

>What I meant was, does anyone have a real example of a
>case where there would be serious inconvenience if
>finalisers couldn't play with each other's data?

Good question.

It seems to me that the only time we want finalizers is when we're the
only link to an external resource, and furthermore it seems that the
finalizer should do nothing more than close that external resource.

Wait.  Perhaps I should have said "when we _represent_ an external
resource".

It sure seems to me that one basic definition of an external resource is
that it contains no references to Python objects (except for _perhaps_ a
table of lazy references).

If I'm correct, and we can claim that finalizers are only needed in this
case, perhaps we can get away with making some simplifying assumptions.

>Greg

-- 
-William "Billy" Tanksley
Utinam logica falsa tuam philosophiam totam suffodiant!
   :-: May faulty logic undermine your entire philosophy!



From milliger at ausinfo.com  Thu Jun 10 13:33:25 1999
From: milliger at ausinfo.com (mike milliger)
Date: Thu, 10 Jun 1999 12:33:25 -0500
Subject: Fixed; not elegant, but if interested...
References: <375FEA53.95E3D3BC@ausinfo.com>
Message-ID: <375FF6E5.105F9B5E@ausinfo.com>

date2= time.localtime(time.time())

year = repr(date2[0])
month= repr(date2[1])
day  = repr(date2[2])
hour = repr(date2[3])
min  = repr(date2[4])
sec  = repr(date2[5])

f = year + month + day
b = hour+ min+ sec

os.system (r"pkzip25 -add blah"+ f + r"_" + b +r".zip data_in.dat")

to get this file: blahyyyymmss_hhmmss.zip
although i am not guaranteed to get 2 spots if the integer value is less
than 10.


mike milliger wrote:

> I read the documentation in the Python Reference Library (6.3) on time
> access and conversions and chapter 8 in programming python, yet I can't
> figure out if i need ( or how to use) a tuple to format the return from
> localtime().  i get a string now (wed jun 10 1999 xx:xx:xx) but i need
> to format it to yyyymmdd_hhmmss.  Is there any switch ( %) or function
> that can be useful?  Any help is appreciated.
> --mike




From gmcm at hypernet.com  Sat Jun 12 00:23:53 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 11 Jun 1999 23:23:53 -0500
Subject: dict vs kjBuckets vs ???
In-Reply-To: <3760c9d2.1106146@news.cyf-kr.edu.pl>
Message-ID: <1282960806-85602695@hypernet.com>

Mark R. This:
> >> I plan to write a program that would store lots (in range of 10M or even
> >> more) of relatively small objects (a few hundred bytes at most), so what
> >> do you think I should use?
[Tim]
> >Let's do a little math <wink>:  10M * 100 = ?, a lower bound on what you're
> >contemplating.  Do you have gigabytes of RAM?
[Brian, er, Mark]
> I'm opening a boutique.
[Tim]
> >...Memory-based data structures aren't
> >going to work for the size of thing you have in mind.  If you can make it
> >fly it all, you'll likely require a powerful database, so of those choices
> >Metakit is the only approach that's not dead on arrival.
[Mark]
> A few additional informations: items stored would be natural
> language text fragments (several sentences at most, several words
> typically)
> + binary descriptions, primary operation would be lots of searching. 
> Is there anything else that would be better for this kind of
> program? Object database?

Searching 10M items is going to strain just about anything. I get 
wonderful response times out of MetaKit, but I've never tried 
anything approaching that size. The state of the art for searching 
big amounts of data are the big boys of SQL databases. But even 
there you'll have an index of 6 or 10 levels, and unless you have 
huge amounts of RAM, only a few of them will be in memory. So the 
disk will grind and grind for each search. I'd try MetaKit first, 
since it's a whole lot simpler and lighter wieght.

The best solution would be to exploit any interrelationships in your 
items. If you could reduce it to 100K items, you could burn a whole 
lot of CPU and still come out ahead.

and-good-luck-with-the-boutique-Brian-ly y'rs

- Gordon



From matz at netlab.co.jp  Thu Jun  3 21:53:00 1999
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 04 Jun 1999 10:53:00 +0900
Subject: Python 2.0
References: <1283872943-30734874@hypernet.com>
Message-ID: <874skobukj.fsf@ev.netlab.co.jp>

"Gordon McMillan" <gmcm at hypernet.com> writes:

|As someone used to GC, you would probably never write:
|
| txt = open('blahbalh.txt','r').read()
|
|This is a common idiom for those of us used to ref counting.
|You would certainly take a reference to the file object so you could 
|explicitly close it. 

We often do that, at least in Ruby.  Ruby will close the file sometime
in the future.  To ensure immediate close of the file, you need to
close it explicitly in GC system.  Some may consider this as drawback,
though I don't.
						matz.



From larsga at ifi.uio.no  Wed Jun 16 06:43:45 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 16 Jun 1999 12:43:45 +0200
Subject: OO (was: Why is tcl broken?)
References: <000801beb745$02768220$979e2299@tim> <871zfdp73f.fsf@home.ivm.de>
Message-ID: <wklndk8ly6.fsf@ifi.uio.no>

* Klaus Schilling
| 
| Steele just rediscovered Scheme.

What, do you mean recently? 

--Lars M.



From Nathan.Clegg at p98.f112.n480.z2.fidonet.org  Tue Jun 29 23:38:16 1999
From: Nathan.Clegg at p98.f112.n480.z2.fidonet.org (Nathan Clegg)
Date: Wed, 30 Jun 1999 04:38:16 +0100
Subject: socket timeouts
Message-ID: <000015c4@bossar.com.pl>

From: Nathan Clegg <nathan at islanddata.com>

Is there any way to set the timeout for a TCP/IP connection? 
Specifically, I'm writing some apps meant to run on the same machine, so
if they can't connect *immediately* (i.e. the server app is not running), I
need to go to plan B.  Any answers or alternate solutions?  I'm also
interested in a similar issue across the internet, but I'm more generous
about timeouts in this circumstance.  Thanks.


----------------------------------
Nathan Clegg
 nathan at islanddata.com






From claird at Starbase.NeoSoft.COM  Thu Jun 10 07:44:19 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 10 Jun 1999 06:44:19 -0500
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com>
Message-ID: <7jo8ej$gnp$1@Starbase.NeoSoft.COM>

In article <375F82C7.CF0F0477 at iname.com>,
Fernando Mato Mira  <matomira at iname.com> wrote:
>Hello,
>
>  I'm trying to collect a list of all the _current_ issues making tcl
>a bad language choice. I'd like concrete examples, and not just
			.
			.
			.
I don't entirely understand your intent.  <URL:http://
starbase.neosoft.com/~claird/comp.lang.tcl/fmm.html>
details confusions that persistently arise in use of
Tcl.  Perhaps that will serve you.

Please give a concrete example of a concrete example.
Commitment to PalmOS for your deliverable is one of
the first issues that comes to my mind; in that case,
Tcl is "a bad language choice", because it's not cur-
rently available for PalmOS.  Is that the sort of
thing you have in mind?
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From burton at lucent.com  Wed Jun  9 18:34:10 1999
From: burton at lucent.com (Burton Kent)
Date: Wed, 09 Jun 1999 17:34:10 -0500
Subject: How do I defect?
Message-ID: <375EEBE2.3F20390F@lucent.com>

I use Perl because of the nifty regex.  But
I hate Perl because of the data structures.

If you ever need any kind of data structure
in perl, move into a padded cubicle.  Check
into the funny farm. Flop into the loony bin.
Perl is like LSD.  It's cool and pretty and
will slowly drive you insane.

So how do I defect from Perl to Python?  I 
have the Programming Python book on order.  
What can I read to get started ASAP?

Thanks.

B



From skip at mojam.com  Fri Jun 11 15:47:36 1999
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 11 Jun 1999 19:47:36 GMT
Subject: Date formats
References: <375FEA53.95E3D3BC@ausinfo.com>
Message-ID: <37616867.80E1C2EF@mojam.com>

mike milliger wrote:
> 
> I read the documentation in the Python Reference Library (6.3) on time
> access and conversions and chapter 8 in programming python, yet I can't
> figure out if i need ( or how to use) a tuple to format the return from
> localtime(). 

Mike,

You want to use time.strftime if it's available on your platform. 
Here's a simple example:

  >>> import time
  >>> now = time.localtime(time.time())
  >>> now
  (1999, 6, 11, 15, 48, 40, 4, 162, 1)
  >>> time.strftime("%m/%d/%y", now)
  '06/11/99'
  >>> time.strftime("%Y-%m-%d", now)
  '1999-06-11'

-- 
Skip Montanaro	| Mojam: "Uniting the World of Music"
http://www.mojam.com/
skip at mojam.com  | Musi-Cal: http://www.musi-cal.com/
518-372-5583



From arcege at shore.net  Wed Jun  2 11:12:04 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 02 Jun 1999 15:12:04 GMT
Subject: All this self dot is making me...
References: <Vn653.468$HD.99317@typhoon-sf.snfc21.pbi.net> <m33e0a52x2.fsf@chinon.cnrs-orleans.fr> <37551fa6.90445994@news.oh.verio.com>
Message-ID: <8Rb53.1622$nn.483094@news.shore.net>

Kevin Dahlhausen <morse at harborcom.net> wrote:
:>Calling this parameter "self" is just convention, you can use whatever
:>you like. However, it's a convention that just about everybody follows,
:>so I'd think twice before doing something different.

: A couple of options come to mind:

: Or call it something short - s. ? - and use a (careful) global search and
: replace when  you're done with the coding.    I sometimes use this
: for long identifiers.  

: Or in VIM set up a mapping:
:  :map! s. self.

I think abbreviations would be better:
  :abbrev s. self.

The convention is to use "self".  It was the adoption of the original
object-oriented language (SmallTalk) in the early '70s.  Why did C++
change it to "this"?  I don't know.. probably just to be different.

People who are very well-versed in Python (Guido, Timbot, etc.)
probably wouldn't have a problem if someone used something other than
"self", but the average user might not.

Also from a psychological view, it might be better to start using
"self" instead of "this" to get your out of the "C++" thinking and into
a Python mode.. at least when you are programming in Python. :)

  -Arcege




From rfreedm1 at nospam.rochester.rr.com  Fri Jun 11 20:22:19 1999
From: rfreedm1 at nospam.rochester.rr.com (Benjamin Schollnick)
Date: Sat, 12 Jun 1999 00:22:19 GMT
Subject: Problem with Lists & Index command
References: <rNHvjEdhm5Pp-pn2-zOTjsVjHebb4@d185d18f8.rochester.rr.com> <37617DFE.9BE0E03F@weihenstephan.org>
Message-ID: <rNHvjEdhm5Pp-pn2-qhHq6lGpn6DZ@d185d18f8.rochester.rr.com>

On Fri, 11 Jun 1999 21:22:06, "Dr. Peter Stoehr" 
<peter.stoehr at weihenstephan.org> wrote:

> Hi Benjamin,
> 
> maybe I missed something, but why don't you a dictionary ?

Because I'm still learning and haven't really read up up on
dictionary's yet....

I'm only about 1/3 (somewhere around pg 150) in the 
Python book I have >g<  (Out of somewhere around
page 500 or so).  And dictionaries haven't been covered
yet....

			 - Benjamin



From killes at geneseo.net  Sat Jun 26 08:21:19 1999
From: killes at geneseo.net (Steven L. Kille)
Date: Sat, 26 Jun 1999 13:21:19 +0100
Subject: Memory leaks
Message-ID: <7l35rf$dl3$1@news3.infoave.net>

I'm embedding python in an MFC application.  When I close the application I
get a bunch memory leak messages (16).  Im sure this is just memory
allocated on startup and does not accumulate during the execution of my
program so it's not big deal.  It's just kind of annoying, especially if
your looking for memory leaks in your own program.  Does anyone know of any
other function I need to call other than Py_Finalize() when closing down my
app?  If I just place the following two lines in my InitInstance() function
I get the memory leak messages:

BOOL CFormulasApp::InitInstance()
{

  Py_Initialize();
  Py_Finalize();
return FALSE;
}









From bh at cellware.de  Tue Jun 22 04:50:31 1999
From: bh at cellware.de (bernard)
Date: Tue, 22 Jun 1999 10:50:31 +0200
Subject: TkInter callback - how to ?
Message-ID: <376F4E57.6A23A66F@cellware.de>

I have an array of buttons (a keypad) all using the same callback
function - is there any way of finding out which button invoked
the callback ? I suppose I am looking for a tcl/tk solution
something like
  -command "myCallback $thisButton"
but tkInter only seems to provide
  -command myCallback

Any help welcome

bernard



From nospam at bitbucket.com  Sat Jun 19 02:44:11 1999
From: nospam at bitbucket.com (Phil Mayes)
Date: Fri, 18 Jun 1999 23:44:11 -0700
Subject: simple text file 'parsing' question
References: <376B1AAC.19FE8BCE@mysolution.com>
Message-ID: <d%Ga3.498$dR.35932@newsfeed.avtel.net>

KP wrote in message <376B1AAC.19FE8BCE at mysolution.com>...
>Here's my dilema: a directory filled (200+) with small emails. My goal
>is to strip all the headers and combine them into one file. I can read
>all the files just fine and write them all to one file, but I cannot
>discern how to strip the headers. The answer must be very simple, yet
>I cannot see it. Can anyone give a few pointers on how to do it, our
>what module might be best? Thank you.
>Ken


A raw email always has a blank line between the header and the body.
(To be pendantic, it should also have all its lines ending in CRLF.)
So you can read it in and find the gap by looking for 2 EOLs:

import string
f = open('c:\\apps\\eudora\\in.mbx', 'r')
all = f.read()
x = string.find(all, '\n\n')
body = all[x+2:]
# append body to output file
--
Phil Mayes    pmayes AT olivebr DOT com







From pphalen at teleo.net  Thu Jun  3 23:14:43 1999
From: pphalen at teleo.net (Patrick Phalen)
Date: Thu, 3 Jun 1999 20:14:43 -0700
Subject: "The Python Way"
References: <14167.3323.707498.300295@anthem.cnri.reston.va.us>
Message-ID: <99060320192502.01312@quadra.teleo.net>

On Thu, 03 Jun 1999, Barry A. Warsaw wrote:
> >>>>> "SDM" == Steven D Majewski <sdm7g at Virginia.EDU> writes:
> 
>     SDM> What makes you think that Tim and Guido would agree on "The
>     SDM> Python Way" ?
> 
> What makes you think /Guido/ would agree on "The Python Way"? :)


For that matter, what makes me think that Guido and Timbot are two
different entities?



From arcege at shore.net  Mon Jun 28 14:44:58 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Mon, 28 Jun 1999 18:44:58 GMT
Subject: Python script background execution...
References: <7l628i$f2e$1@snatch.hq.eso.org>
Message-ID: <KoPd3.145$6M6.44310@news.shore.net>

NO_SPAMnpirzkal at hq.eso.org wrote:
: Hi,

:  I do not seem to be able to run my python script in the background (test.py 
: &..). After a couple of second, the OS (Solaris 2.5.1) reports that the 
: process has been suspended. 
:  Does anyone know what I am missing?

:   Nor

Solaris has signals for reading from and writing to the terminal while
in the background.  If your script has "print" statements or calls to
sys.stdout.write or sys.stderr.write (or write to /dev/tty), then
you'll get a SIGTTOU signal.  Likewise if you read with "input()" or
"raw_input()" or make calls to sys.stdin.read*() (or read from
/dev/tty), then you will get a SIGTTIN signal.  You will want to
open a log file and write to that and not read from the terminal.

Check if you are getting either of these signals.

  -Arcege




From dfan at harmonixmusic.com  Mon Jun  7 11:38:52 1999
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 07 Jun 1999 11:38:52 -0400
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7iqhkb$d7s@wiscnews.wiscnet.net> <7iu7ma$d33$1@cronkite.cc.uga.edu> <02hfonzsx2.fsf@rtp.ericsson.se> <02909zzou0.fsf@rtp.ericsson.se> <87zp2f4qrz.fsf@pc-hrvoje.srce.hr>
Message-ID: <wkd7z858c3.fsf@turangalila.harmonixmusic.com>

Hrvoje Niksic <hniksic at srce.hr> writes:

| Kumar Balachandran <kumar*xspam*@*xspam*rtp.ericsson.se> writes:
|
| > What you said is quite wrong. Since Python provides closures, I can
| > change my programming style to be completely different from the
| > standard C or Fortran style and use a functional interface.
| 
| Since when does Python provide closures?

The bytecodehacks module allows the creation of closures as I understand
them.

I'd love to see the bytecodehacks stuff become standard in some way.
The facilities that it has added for closures and better functional-type
programming closed the only hole in Python that I (personally) felt
needed filling.

-- 
                 Dan Schmidt -> dfan at harmonixmusic.com, dfan at alum.mit.edu
Honest Bob & the                http://www2.thecia.net/users/dfan/
Factory-to-Dealer Incentives -> http://www2.thecia.net/users/dfan/hbob/
          Gamelan Galak Tika -> http://web.mit.edu/galak-tika/www/



From bwarsaw at cnri.reston.va.us  Mon Jun  7 22:43:49 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Mon, 7 Jun 1999 22:43:49 -0400 (EDT)
Subject: fork()
References: <375BE64A.EB997F24@icrf.icnet.uk>
	<87so840zcc.fsf@pc-hrvoje.srce.hr>
	<375BF721.4B1864F2@icrf.icnet.uk>
	<m31zfn99td.fsf@greebo.nodomain.de>
	<87hfojzudh.fsf@pc-hrvoje.srce.hr>
Message-ID: <14172.33637.119185.613719@anthem.cnri.reston.va.us>

>>>>> "HN" == Hrvoje Niksic <hniksic at srce.hr> writes:

    >> I fear that in reality the memory is copied after all. If they
    >> use the dictionary in any way they are modifying refcounts of
    >> at least some of the objects in the dictionary. From the OS's
    >> point of view this counts as modifying the memory.

    HN> Eek.  You're right.  Refcounting truly sucks.  :-(

It needn't though.  IIRC, NextStep didn't keep the refcounts in the
objects.  Also, since most objects in that system were transient and
only had a refcount of 1, such objects didn't have entries in the
object refcount dictionary until their refcounts were increased to 2.

-Barry



From mwh21 at cam.ac.uk  Sun Jun 13 12:18:48 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 13 Jun 1999 17:18:48 +0100
Subject: Checking for Python modules in autoconf
References: <199906031338.JAA28564@amarok.cnri.reston.va.us>
Message-ID: <m3emjgt6on.fsf@atrus.jesus.cam.ac.uk>

"Andrew M. Kuchling" <akuchlin at mems-exchange.org> writes:

> I'm working on packaging some software for distribution, and would
> like the configure script (generated using GNU autoconf) to check
> whether various Python modules are installed.  Does anyone have an
> autoconf macro for doing this?

Checking through a pile of messages I've flagged over the last however
long...

James Henstridge has written some extensions to automake/autoconf as
part of his Python GTK+ bindings. Apparently they'll be part of
automake 1.5 whever that surfaces, but you can get them as part of
gnome-python or pygtk. Go see

ftp://ftp.gtk.org/pub/gtk/python/

HTH
Michael
 
> -- 
> A.M. Kuchling			http://starship.python.net/crew/amk/
> Thanks. The sooner I get discouraged and quit, the more time I'll save overall.
>     -- Frank Sergeant, 28 Mar 1999



From jfisher at are.berkeley.edu  Thu Jun 10 15:34:34 1999
From: jfisher at are.berkeley.edu (John Fisher)
Date: Thu, 10 Jun 1999 19:34:34 GMT
Subject: Extending Python with C
Message-ID: <7jp409$qu5$1@nnrp1.deja.com>

I'm a relative Python newbie having some trouble getting a module to
load.  I wrote a simple one, using as a model the example given on the
Python website, just to get the hang of it; however, when I try to load
the compiled file I get a rather baffling error:

ImportError: ./echo.o: ELF file's phentsize not the expected size

Any ideas on what this is?  I'm using Python 1.5.2, and the .c file
compiled without complaint on gcc 2.91.66.  I imagine I'm missing
something simple, like a compile option for gcc, but I just can't figure
it out.

Thanks,
John


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From fysx at spots.ab.ca  Mon Jun 14 22:16:09 1999
From: fysx at spots.ab.ca (Maxwell Sayles)
Date: Tue, 15 Jun 1999 02:16:09 GMT
Subject: GNU python?
Message-ID: <3765B8B4.18DE0257@spots.ab.ca>

Looking to try out python... is there a GNU python for Win9x?  and a
tutorial maybe?  thx
Maxwell Sayles




From pduffin at mailserver.hursley.ibm.com  Thu Jun 10 10:20:04 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Thu, 10 Jun 1999 15:20:04 +0100
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <375f92b4.618467559@news.u-psud.fr>
Message-ID: <375FC994.586@mailserver.hursley.ibm.com>

Laurent POINTAL wrote:
> 
> [note Follow-up to: set to comp.lang.python]
> 
> On Thu, 10 Jun 1999 11:17:59 +0200, Fernando Mato Mira
> <matomira at iname.com> wrote:
> 
> for Python vs Tcl
> >:   If you have references to articles of the sort, that will be useful,
> >: too.
> 
> See http://www.cwi.nl/~sjoerd/PythonVsTcl.html
> 
> I remeber having seen an article of a guy explaining why he switched
> from Tcl to Python for a large project... but I loose the URL.
> 
> A+
> 

The article mentioned is VERY out of date wrt Tcl. This is probably
because the author is not keeping up to date with the developments in
Tcl because he has already chosen Python.

I would recommend that you go to the home pages of all of the languages
that you are interested in to get the pro-language biased stuff and then
try and make the decision for yourself about which language is best for
you.

BTW I cannot see the original document so I do not know the context of
the subject but Tcl is most certainly not broken.

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From cgw at fnal.gov  Wed Jun 16 17:25:03 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Wed, 16 Jun 1999 21:25:03 GMT
Subject: Emacs package to the Python debugger?
In-Reply-To: <3767FEB1.87D18512@cup.hp.com>
References: <3767FEB1.87D18512@cup.hp.com>
Message-ID: <14184.5679.711795.786883@buffalo.fnal.gov>


Ovidiu Predescu writes:

>   Is there any pac kage that allows debugging Python programs from
>   within Emacs?  Something that is integrated with Emacs' GUD would
>   be very convenient.

Recent versions (21.0 and up) of XEmacs include this in the
"prog/debug" package.  

If you don't use the XEmacs package system, you can still get the
single-file "pdb.el" from

ftp://ftp.aifb.uni-karlsruhe.de/pub/wwe/pdb.el

And, yes, it is build on top of the Grand Unified Debugger.






From bhosu at my-deja.com  Tue Jun  8 09:41:16 1999
From: bhosu at my-deja.com (bhosu at my-deja.com)
Date: Tue, 08 Jun 1999 13:41:16 GMT
Subject: Using PyArg_VaParse
Message-ID: <7jj6hr$l1g$1@nnrp1.deja.com>

Hi All,
   I want to use PyArg_VaParse to retrieve variable number of arguments.
I am not sure if this is the correct function to use. I have embedded
python in a C++ application and implemented python methods in C++. I
have a situation where, the method may pass any number of string
parameters. I would like to get these parameters in my C++ code. Here is
a scenario of the whole thing.

(C++ implementation.. for method bar in the module foo.)

static PyObject *bar(PyObject *self,PyObject *args)
{
....
if(PyArg_VaParse(args,...) /* What is that I need to use here...? */
...
}

and a pythod call may be like,
import foo
foo.bar('a','b','c')
foo.bar('a','b','c','d'.....)

I need to know the Python API call to use to get the arguments. As I
said before I donot know how many arguments are being passed.

Thanks for any help...

Sundar.R


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From gmcm at hypernet.com  Tue Jun 15 13:58:01 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 15 Jun 1999 12:58:01 -0500
Subject: idle tasks
In-Reply-To: <XFMail.990615092752.nathan@islanddata.com>
Message-ID: <1282652759-104133036@hypernet.com>

Nathan Clegg writes:

> I'm writing a multi-threaded application with the threading module. 
> I would like one of the threads to run as the lowest priority,
> effectively an idle function.  I typically only see idle functions
> in GUI toolkits, however, and this needs to be a console app.  Is
> there a way to lower the priority of a thread or create an idle
> function without using tkinter or pygtk??

A GUI's "idle" function is usually run from the GUI's thread when it 
finds that there hasn't been a GUI event in awhile. So your function 
needs to only do small snippets of work, or it will block the GUI.

No priorities on Python threads. You might just put in plenty of 
small sleep()s, which will ensure that other threads get plenty of 
chances to run. But since the PVM swaps threadstates every 10 (unless 
you've fiddled with it) byte code instructions, you probably don't 
need it.


- Gordon



From tim_one at email.msn.com  Mon Jun 28 00:34:50 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 28 Jun 1999 00:34:50 -0400
Subject: total idiot question: +=, .=, etc...
In-Reply-To: <7kuocf$1an$1@brick.cswv.com>
Message-ID: <000d01bec11f$8f966e20$e19e2299@tim>

[Neel Krishnaswami]
> ...
> Here's an example from some code of mine (it's from a programmatic
> front-end to Dejanews):
>
> class Article:
>     [...]
>     #
>     # To set the sort order, set Article.sortorder to one of the
>     # keys of Article.comparators.
>     #
>     comparators = {'date':      lambda x,y: cmp(x.date,y.date),
>                    'score':     lambda x,y: cmp(x.score,y.score),
>                    'author':    lambda x,y: cmp(x.author,y.author),
>                    'subject':   lambda x,y: cmp(x.subject,y.subject),
>                    'newsgroup': lambda x,y: cmp(x.newsgroup,y.newsgroup)
>                    }
>     sortorder = 'date' # can be anything in Article.comparators.keys()
>
>     [...]
>
>     def __cmp__(self, other):
>         return self.__class__.comparators[self.__class__.sortorder](self,
>                                                                     other)
>
> I want the user to be able to determine how a list of Article's can be
> sorted by setting the class's sortorder member. So a statement like
>
> >>> Article.sortorder = 'author'
>
> would change how lists of articles are sorted. This is definitely
> something that's shouldn't be an instance attribute, imo; it would
> defeat the purpose if each article could compare differently than
> any other. (Oppose standardized testing! :])
>
> But look how ugly that __cmp__ method is; it would be a lot clearer to
> write something like
>
>     def __cmp__(self, other):
>         return Article.comparators[Article.sortorder](self, other)
>
> because that would make it clearer what class state is being used, and
> why. However, though it would be clearer to the programmer, it would
> profoundly mess up inheritance.
>
> It's perfectly reasonable for someone else to subclass Article and
> override the Article.comparators attribute. (Say to add an ordering to
> put Tim Peter's posts at the top of the list,

Yes, an excellent idea!

> or to use something faster than my lambdas.)

Oh, yawn <wink>.

> With the clear way of writing it, the NewArticle class would
> mysteriously fail to do the right thing, because the __cmp__ method
> would look in the Article class dictionary without bothering to check
> NewArticle's.

The trap is hard to avoid, though:  if a direct instance of Article is
"self", and a direct instance of NewArticle is "other", even keying off
self.__class__.etc is a surprise (well, at least to "other"!).  In part
you're taking a single-dispatch approach to a problem with multiple-dispatch
headaches.  Face that head-on and you'll write a multiple-argument
dispatcher with which the various classes need to register their intentions
("OK, if arg1 and arg2 are both Articles, I need to compare such-&-such a
way; if arg1 is Article but arg2 is NewArticle, that-&-that a way;" etc).
Then it's factored out of the xxxArticle classes.

> This is an unusual glitch in Python; usually the readable way of doing
> something is the right way of doing it.

Guido long ago confessed he had static class hierarchies in mind, where
everyone knew who they were inheriting from for all time so that "super" etc
weren't needed.  That should probably be re-thought for Python2, since it
has come up more than once <wink>.  Writing

    class A(B):
        def __init__(self, x, y, z):
            B.__init__(self, x, y, z)

can be grating too (hmm ... I see Michael and Greg are whining about that
too, so I'll skip it).

In your particular case, the solution that plays along with the current
design is to forget direct access *and* __class__, instead indirecting
through a method:

class Article:
    comparators = { ... }
    sortorder = { ... }
    def get_class_comparators(self):
        return Article.comparators
    def get_class_sortorder(self):
        return Article.sortorder
    def __cmp__(self, other):
        return self.get_class_comparators() ...

and override get_class_comparators/sortorder() in NewArticle.  Perfectly
clear and ridiculously involved <wink>.  The fact is that self.comparators
does the search that you want, so that's the best way to do it -- live with
a comment.

> Maybe we could re-use the class keyword and let people write class.foo
> inside class definitions, by analogy to writing self.foo to refer to
> the current instance? And the class.foo syntax would get translated to
> the equivalent of self.__class__?

Except that you want class.foo to act like self.__class__.foo <wink>.  Guido
certainly won't change this in Python1 regardless; take it up on the
Types-SIG for Python2!

more-dead-ideas-there-than-in-kosovo-ly y'rs  - tim





From m.greiner at t-online.de  Mon Jun 21 15:35:49 1999
From: m.greiner at t-online.de (Markus A. Greiner)
Date: Mon, 21 Jun 1999 21:35:49 +0200
Subject: How to call a Java superclass from jpython ??
Message-ID: <376E9415.8E002ACA@t-online.de>

Hello,

I?m trying to port some java applications into jpython.
Now I got a problem. I?m not able to call a java superclass from
jpython.
Did anybody know how to call them? Here the code I?m trying to convert:

It?s from PropertiesMetalTheme.java of the JDK Metalworks example.
In method initcolors:

	...
	primary1 = super.getPrimary1()

Everything I tried in jpython failed. In some other cases replacing the
super call into a construct like:

	...
	ParentClass.method(self) 

works fine, but in this case

	...
	DefaultMetalTheme.getPrimary1()

and all other constructs failed. I?m using JPython 1.1 beta 2 so
calls like self.super__method are obsolte and didn?t work anymore.
(I think :-)) or better I tried without success)

Thanks Markus



From jblake at eskimo.com  Tue Jun 15 18:03:42 1999
From: jblake at eskimo.com (Jonathon)
Date: 15 Jun 1999 22:03:42 GMT
Subject: Python Paterns
References: <7k5nk0$nc0$1@nnrp1.deja.com>
Message-ID: <7k6iju$75v$1@eskinews.eskimo.com>

Dinu C. Gherman (gherman at my-deja.com) wrote:
: jblake at eskimo.com (Jonathon) posted the following about
: a week ago:
: > Searching Python.org, I didn't come across
: > anything _recent_ about
: > patterns in Python.  Did that SIG die, get swallowed up

: Well, I guess the question can be answered like this: the SIG
: died as you correctly noticed, the reasons being a lack of
: contributions as well as an ill-defined (or at least vague)
: purpose of the SIG itself.

	Aaarggh.

	Given that Python is Object Orientated, and that
	Patterns help design better object orientated
	software, it seemmms to be to be natural that
	there would be more than a few random things on
	Patterns In Python.

	<< For various reasons, I suspect that teaching Patterns
	in Python to people as they learn Python would be a very
	good thing.  >>

	xan

	Jonathon

-- 
	I'm still looking for a good book on
		3:	The Recent Unpleasantness
		1:	The War Of Northern Aggression.
		2:	The War of Southern Rebellion.



From tseaver at palladion.com  Fri Jun 11 00:44:25 1999
From: tseaver at palladion.com (Tres Seaver)
Date: Thu, 10 Jun 1999 23:44:25 -0500
Subject: High performance Python on NT
References: <000301beb3b7$4f973810$6eba0ac8@kuarajy.infosys.com.ar>
Message-ID: <37609429.65A7E802@palladion.com>

Bruno Mattarollo wrote:
> 
> Hi!
> 
>         I am about to start a project where I will need to do some high performance
> Web CGI scripts on WinNT ... I have no choice ... I know it's not the best
> combination. It's not my decision ;-) I am looking for some advice. Is it
> better if I use some kind of Active Python Scripting Page (ASP but with
> Python)? Or just doing CGI with threads it will be fine... ?
> 
>         I know this is a very vague question... I just want to know if someone had
> to take a similar decision... The app I will be doing should support several
> hundred users connecting to the Web server and doing some serious CGI
> activity ... Almost all hits that the server will receive will be scripts.
> The hardware is no problem in this case... When can scale up to Quad PII
> Xeon with 1GB RAM ... U2W SCSI discs...
> 
>         I know exactly what to do in a UNIX env... but I am not sure of the
> decision to take regarding win32...
> 
>         TIA
> 
> /B
> 
> Bruno Mattarollo <bmatt at acm.org>
> ... proud to be a PSA member <http://www.python.org/psa>
> <http://starship.python.net/crew/bmatt/>

Run, don't walk, to http://www.zope.org.  You'll think you've died and gone to
heaven.

-- 
=========================================================
Tres Seaver         tseaver at palladion.com    713-523-6582
Palladion Software  http://www.palladion.com



From herzog at online.de  Mon Jun  7 13:53:18 1999
From: herzog at online.de (Bernhard Herzog)
Date: 07 Jun 1999 19:53:18 +0200
Subject: fork()
References: <375BE64A.EB997F24@icrf.icnet.uk> <87so840zcc.fsf@pc-hrvoje.srce.hr> <375BF721.4B1864F2@icrf.icnet.uk>
Message-ID: <m31zfn99td.fsf@greebo.nodomain.de>

Arne Mueller <a.mueller at icrf.icnet.uk> writes:

> Hrvoje Niksic wrote:
> > 
> > Arne Mueller <a.mueller at icrf.icnet.uk> writes:
> > 
> > > All that woks fine now (after I spend a long time fiddling with the
> > > pipe and select stuff to synchronize I/O). However the forking is a
> > > waste of time and memory because all the children get their own
> > > large dictionary!
> > 
> > If you don't change the dictionary, the memory will not be copied.
> > Most modern Unixes support COW (copy on write) which means that the
> > actual fork()ing does surprisingly little.  Only when the memory is
> > written to, a new copy is spawned off.
> > 
> > If the children are modifying the dictionary, that's well something
> > different...
> 
> Huh, that's realy good news! But I don't understand how that works, when
> the child changes the dictionary it quickly gets it's own copy, so the
> operating system is watching the child's activity and copies memory when
> it's accessed for writing? However my children don't change anything in
> the dictionary ;-)

I fear that in reality the memory is copied after all. If they use the
dictionary in any way they are modifying refcounts of at least some of
the objects in the dictionary. From the OS's point of view this counts
as modifying the memory.

The same applies to all python objects that are referenced in the child
or the parent after the fork(). If I'm not mistaken this includes some
object categories that take a fair amount of your program's memory like
identifiers (represented by string objects) and function/code objects.

-- 
Bernhard Herzog	  | Sketch, a python based drawing program
herzog at online.de  | http://www.online.de/home/sketch/



From rfreedm1 at nospam.rochester.rr.com  Sun Jun 20 11:32:18 1999
From: rfreedm1 at nospam.rochester.rr.com (Benjamin Schollnick)
Date: Sun, 20 Jun 1999 15:32:18 GMT
Subject: Questions?  Time Module?
Message-ID: <rNHvjEdhm5Pp-pn2-5b4dbg1xd7dt@d185d18f8.rochester.rr.com>

Folks,

	I'm trying to figure out where the "time" functions are
stored, but GREP isn't finding anything in the \lib directory??!?!?

	I'm have some questions on the STRPTIME function, and
thought I'd examine the code, but can't find it in the 1.51 
installation?
Sure, I can use the "IMPORT TIME" and access everything from there
but heaven help me, if I can't find a TIME.PY or TIME.PYC file 
anywhere...

			Benjamin



From guido at cnri.reston.va.us  Tue Jun 29 12:08:06 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: Tue, 29 Jun 1999 12:08:06 -0400
Subject: total idiot question: +=, .=, etc...
In-Reply-To: Your message of "Tue, 29 Jun 1999 11:56:56 EDT."
             <000001bec248$040ac0c0$6d9e2299@tim> 
References: <000001bec248$040ac0c0$6d9e2299@tim> 
Message-ID: <199906291608.MAA02892@eric.cnri.reston.va.us>

> >>>     class A(B):
> >>>         def __init__(self, x, y, z):
> >>>             B.__init__(self, x, y, z)
> 
> [Greg Ewing]
> > I could live with having to explicitly name the
> > superclass, provided something checked that the
> > class I named was actually a direct base class
> > of the one where the method is defined. That
> > way things would be less likely to break
> > mysteriously when I rearrange the class
> > hierarchy.
> 
> [Guido]
> > Oh, but that check *is* being made!  (In fact this is the same check
> > that prevents "class methods" from working ;-)
> 
> Nope, that's a different check.  If C derives from B derives from A, this
> check allows either C.__init__ or B.__init__ to be called from A.__init__
> with an instance of A.  Greg wants  B.__init__ to be OK here, but not
> C.__init__ (C is not a direct base class of the one (A) where the method
> (A.__init__) is defined).  IOW, he wants a way to restrict __init__ calls to
> immediate bases of the class from which the __init__ is invoked, to e.g.
> prevent level-skipping under class rearrangement.

Sigh.  Indeed.

> I posted some ugly code to address that, but think it's really a job for a
> text editor and a pair of eyeballs <wink>.

Hm -- he may be right!  I was reading some IDLE code the other day and
noticed that PyShell's close() method was calling
PyShellEditorWindow.close(self) while it didn't inherit from that
class at all...  Turns out the actual method was defined in a common
base class -- a silent (and in this case harmless) bug, caused by,
indeed, a class restructuring.  So there is a point to his wish.  On
the other hand Python explicitly allows you to ignore base class
methods in overriding ones, and even constructors...  (And then again,
there is something to say for requiring constructors to play be
slightly different rules -- certainly they don't have to duplicate the
signature of the base class constructor like regular methods need to
do.)

Enough,

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




From syeates at cs.waikato.ac.nz  Thu Jun 24 19:33:20 1999
From: syeates at cs.waikato.ac.nz (Stuart Yeates)
Date: 24 Jun 1999 23:33:20 GMT
Subject: GC In Python: YAS (Yet Another Summary)
References: <Pine.SUN.3.95-heb-2.07.990616155228.15761A-100000@sunset.ma.huji.ac.il> <p53dzrfcea.fsf@bidra241.bbn.hp.com>
Message-ID: <930267199.139037@clint.waikato.ac.nz>

Markus Kohler (markusk at bidra241.bbn.hp.com) wrote:

| The main problem with introducing a GC in Python is that a lot of C-code
| might directly point to Python objects which could be moved at runtime 
| by a state of the art GC  (generational mark and sweep + copying GC). 

[i'm coming at this from a GC point of view not a Python point of view]

There are two ways to handle this:
(a) instrument the python/c++ calling interface
(b) use a 'language independant' garbage collector such as the 
Boehm collector (http://reality.sgi.com/boehm_mti/gc.html). 

The Boehm collector is fast, efficient, portable across
hardware, portable across operating systems and portable 
across target languages. It does threads, finalisation etc

It fully supports C and C++ (and i believe one of the Java
implementation uses it). Because it allows continued use of
malloc/free new/delete it should be possible to build a 
system that recognises pointers from C++ to Python code.

Of course, if anything does pointer XORing or other crimes,
tools such as these are unavaliable.

stuart
--    stuart yeates <s.yeates at cs.waikato.ac.nz> aka `loam'
If you're not part of the solution, then you're part of the precipitate.



From claird at Starbase.NeoSoft.COM  Thu Jun 10 23:15:38 1999
From: claird at Starbase.NeoSoft.COM (Cameron Laird)
Date: 10 Jun 1999 22:15:38 -0500
Subject: Scripting Telnet
References: <a6U73.1265$VL2.195835@news.direcpc.com> <37606A0D.6332A9C2@horvath.com>
Message-ID: <7jpv0q$90v$1@Starbase.NeoSoft.COM>

In article <37606A0D.6332A9C2 at horvath.com>,
Bob Horvath  <bob at horvath.com> wrote:
			.
			.
			.
>Well, the tcl based expect is probably the most appropriate tool for such a
>thing, but there is a pythonized version(s) of it as well.  The one I can
>think of is ExpectPy, (http://www1.shore.net/~arcege/python/ExpectPy/), but
>I seem to remember seeing at least one other one.
>

You're probably thinking of <URL:ftp://ftp.bbn.com/pub/timo/python/exploop.py>.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From hinsen at cnrs-orleans.fr  Tue Jun 15 09:49:20 1999
From: hinsen at cnrs-orleans.fr (hinsen at cnrs-orleans.fr)
Date: Tue, 15 Jun 1999 13:49:20 GMT
Subject: Python and C, arrays
Message-ID: <7k5lkr$mht$1@nnrp1.deja.com>

John Fisher <jfisher at are.berkeley.edu> writes:

> I've run into a problem with a module I'm writing in C.  Basically, I
> need to be able to convert Python tuples or lists (it's unimportant
> which) into C arrays.  I know at compile-time the type of the array
(and
> that the tuple or list will be homogeneous, and of compatible type),
but
> I don't know the size.

The easiest solution by far is to use neither tuples nor lists, but arrays,
as implemented in the Numerical Python extension. The internal data structure
of such an array object *is* a C array. See
http://starship.python.net/crew/hinsen/NumPyExtensions.html for a tutorial on
using arrays in C extension modules. --
-----------------------------------------------------------------------------
-- Konrad Hinsen  | E-Mail: hinsen at cnrs-orleans.fr Centre de Biophysique
Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69 Rue Charles Sadron  | Fax: 
+33-2.38.63.15.17 45071 Orleans Cedex 2  | Deutsch/Esperanto/English/ France 
| Nederlands/Francais
-----------------------------------------------------------------------------
--


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From kas at maps.magnetic-ink.dk  Wed Jun 23 02:20:15 1999
From: kas at maps.magnetic-ink.dk (Klaus Alexander Seistrup)
Date: Wed, 23 Jun 1999 08:20:15 +0200
Subject: glint?
References: <376F48B9.133D37C6@phoenix-edv.netzservice.de>
Message-ID: <37707C9F.7A8CD609@maps.magnetic-ink.dk>

Holger Jannsen wrote:

> I'm working on an installation-tool for linux/win without
> any graphical interface but file-controlled.
> 
> Is 'glint' the right thing I'm searching for?

Glint is a GUI-thing, but it's written in Python allright and I believe
it's GPL'ed, although I'm not sure.


  //Klaus

-- 
???[ Magnetic Ink ]?????????????????????????????????????????????????????????
???[ http://www.magnetic-ink.dk/ ]??????????????????????????????????????????



From duncan at rcp.co.uk  Mon Jun 28 04:39:04 1999
From: duncan at rcp.co.uk (Duncan Booth)
Date: 28 Jun 1999 08:39:04 GMT
Subject: [ANN] DejaGrabber 0.1 -- search Dejanews with Python
References: <7kuu73$1ki$1@brick.cswv.com> <7l0520$qe4$1@news.wrc.xerox.com>
Message-ID: <8DF36264Aduncanrcpcouk@news.rmplc.co.uk>

Val Bykoski <vbykovsk at sdsp.mc.xerox.com> wrote in 
<7l0520$qe4$1 at news.wrc.xerox.com>:

>Neel:
>when i am trying to reach www.python.org thru a proxy (www.mc.xerox.com).
>How to handle this proxy-related situation?
>
>Thanx -- val
>
You need to set an environment variable HTTP_PROXY to point to the proxy server 
before using DejaGrabber.

-- 
Duncan Booth                                    duncan at dales.rmplc.co.uk
int month(char *p){return(124864/((p[0]+p[1]-p[2]&0x1f)+1)%12)["\5\x8\3"
"\6\7\xb\1\x9\xa\2\0\4"];} // Who said my code was obscure?
http://dales.rmplc.co.uk/Duncan



From zigron at jps.net  Tue Jun 22 20:44:39 1999
From: zigron at jps.net (Zigron)
Date: Tue, 22 Jun 1999 17:44:39 -0700
Subject: TCL/Tk
References: <B3wb3.67$Rj1.29977@typhoon-sf.snfc21.pbi.net> <m2zp1sdeda.fsf@intergate.bc.ca>
Message-ID: <W6Wb3.730$4s.84712@typhoon-sf.snfc21.pbi.net>

    I'm on a Windows machine, and don't have Visual C++, and can't actually
get Borland C++ 5.02 to compile it, there's a couple rather weird errors
that
I can't fix.. most I did, b ut anyways.

    So the answer is no :)

--Stephen


Gerald Gutierrez <gutier at intergate.bc.ca> wrote in message
news:m2zp1sdeda.fsf at intergate.bc.ca...
>
> You can always compile it yourself, can't you? Make your own distribution.
>
> "Zigron" <zigron at jps.net> writes:
>
> >     I personally don't like that TCL/Tk is now included in the main
> > distribution,
> > even if Tk is the defacto GUI library for it. And, even if you can't use
> > IDLE
> > without it.
> >     But, lots of people do, I'd expect.. so I've a sugguestion :)
> >
> >     Have a seperate distribution that /doesn't/ have TCL/Tk in it? :-)
> >
> > --Stephen





From Michael.P..Reilly at p98.f112.n480.z2.fidonet.org  Wed Jun 30 11:04:54 1999
From: Michael.P..Reilly at p98.f112.n480.z2.fidonet.org (Michael P. Reilly)
Date: Wed, 30 Jun 1999 16:04:54 +0100
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
Message-ID: <000015b7@bossar.com.pl>

From: "Michael P. Reilly" <arcege at shore.net>

Fredrik Lundh <fredrik at pythonware.com> wrote:
: Michael P. Reilly <arcege at shore.net> wrote:
:> But.. the argument over (these two) GUIs is not exactly a valid one.
:> Tkinter was written using the Tcl interface not the Tk/C API.  If
:> Tkinter was rewritten to use the C API, it probably wouldn't be slower
:> than wxPython.

: The interface is only part of the story; many complaints
: seem to come from the fact that Tk uses idle tasks to
: redraw things.  When an "expose" event arrives from
: the windowing system, Tk calls "after_idle", rather
: than redrawing the widget at once.  And idle tasks
: are only executed when there are no other events
: to take care of...

Hmm... something about that argument makes it circularly invalid.
Non-idle event a comes in, but must be handled after idle event d.
Event d can only execute when event a is taken care of.  Sounds like
deadlock to me. ;)

Events (from Tk_MainLoop) are handled by Tcl_QueueEvent and
Tcl_ServiceEvent (handled by Tcl_DoOneEvent).  Yes, it handles idle
events IF those are the only ones being requested to be serviced:
(taken from tcl8.0.5/generic/tclNotify.c, lines 668-672, in the
function Tcl_DoOneEvent)

        /*
         * If idle events are the only things to service, skip the
         * main part of the loop and go directly to handle idle
         * events (i.e. don't wait even if TCL_DONT_WAIT isn't set).
         */

Otherwise the first thing called is Tcl_ServiceEvent.  Now yes, this
means that when it is servicing idle events, it will not check to see
if there are any pending events, but I see that as a different issue.

In fact, what might be objectionable, is there are two Tcl_ServiceEvent
calls (one on either side of the Tcl_WaitForEvent) before the
Tcl_ServiceIdle call, delaying the "do_idle" even more.

: Tk also uses double buffering for all widgets, which is
: quite expensive, especially since Tk allocates a new
: buffer for each update :-(

Yes, that could be a problem, but see what I mean down below about
additional layers.

:   In fact, since wxPython has to go thru extra layers, it
:> would probably be slower (Tk/C is written on the native API on all the
:> platforms).

: What native API?  As of 8.0, Tk still emulates all widgets
: except the scrollbar, and use a somewhat ineffective
: X emulation layer for graphics.  AFAIK, they haven't
: changed that in 8.1...

Exactly, Tk is using the native API: X11 (not Motif, not Xt, no
emulation other than its own), Windows, MacOS.  wxWindows requires GTK
or Motif on UNIX, requiring another level of software.  Motif is
licensed (but most vendors have gotten around that.  GTK+ 1.2.3 is 2.58
Mb with monthly updates (they do kindly give patch updates tho).

Also, Tk was developed on X11, where the window object created is
stored in the X server (another process), not in the application.  It
is standard practice to keep your own information about the objects you
request creation of.  This is no different from what wxWindows would
have to do using GTK or Motif, both using X11 (except that wxWindows
would probably have its own information as well as similar information
would be stored for widgets created from either GTK or Motif).

I'm not saying that Tk is the greatest thing since sliced bread, just
that it may not be as off-kilter as people think.  Tkinter is off in
that is uses the Tcl/Tk interpreter instead of the Tk/C API (but I
for one am not complaining ;).

  -Arcege



From tim_one at email.msn.com  Thu Jun 10 21:54:52 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 10 Jun 1999 21:54:52 -0400
Subject: IDLE interacting w Shell
In-Reply-To: <37600C03.A5B4FE28@bigfoot.com>
Message-ID: <000401beb3ad$65ab6cc0$329e2299@tim>

[Hoon Yoon]
>    I am trying to use IDLE somewhat like Emacs. I can't use Emacs
> because some of Numpy Code will stop and kill the shell on Emacs when
> certain errors occur.  Not to mention IDLE is a lot lighter.

Guido is working hard to close the gap <wink>.

>   Every time I hit F5 or 'Edit|Run module' I get an output window. I
> want to get the run results and direct to shell like Emacs' C-c C-c,
> instead of getting an output window.

The current CVS IDLE has distinct "run script" and "import module" cmds.
The former is much closer to pymode C-c C-c, although you still need to
explicitly save the file before running it.  Eventually I suppose it will
grow "run region" etc subcommands too.

>   for exaple: if I have following in Editor and press F5
> import os
>   I would like to see following working on the shell:
> >> dir(os)

Yes, works fine with the new "run script" cmd.  Any output produced by the
script is also written to the shell window -- the separate output windows
are gone.

time-machine-working-overtime-ly y'rs  - tim





From tismer at appliedbiometrics.com  Mon Jun 28 07:12:19 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Mon, 28 Jun 1999 11:12:19 GMT
Subject: ANN: Stackless Python 0.2
References: <37628EAA.C682F16C@appliedbiometrics.com>
		<7l6gc8$du4$1@news.tamu.edu> <3776F113.8573DAA5@compaq.com>
Message-ID: <37775893.AB8A0ED@appliedbiometrics.com>


Greg Ewing wrote:
> 
> Corran Webster wrote:
> > In article <37628EAA.C682F16C at appliedbiometrics.com>,
> > Christian Tismer  <tismer at appliedbiometrics.com> wrote:
> > >ANNOUNCING:
> 
> > >               Stackless Python 0.2
> > >           A Python Implementation Which
> > >             Does Not Ese The C Stack
> 
> This sounds interesting, but how do you deal with the
> issue of C extensions which call back to Python code?
> Or is this simply disallowed in Stackless Python?

No, everything is allowed in Stackless Python.
Just grab the source and look at the implementation
of stackless map. That is exactly what a conformant
extension needs to do.
Extensions which don't are not much of a problem,
they just create incompatible frames which will not
be allowed jump targets.

ciao - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From johnm at magnet.com  Thu Jun 10 11:21:36 1999
From: johnm at magnet.com (John Mitchell)
Date: Thu, 10 Jun 1999 15:21:36 GMT
Subject: Why is tcl broken?
In-Reply-To: <1283099377-77267484@hypernet.com>
References: <1283099377-77267484@hypernet.com>
Message-ID: <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com>

On Thu, 10 Jun 1999, Gordon McMillan wrote:

> Fernando Mato Mira writes:
> > I'm trying to collect a list of all the _current_ issues making tcl
> > a bad language choice.
> 
> Why? Convincing your boss to use Python over TCL for your next 
> project is one thing, but in general this is a zero-sum (or worse) 
> game.
> 
> The TCLers I know value its simplicity and accept its limitations. 
> Telling them they're wrong will only piss them off.
> 
> - Gordon

For what it's worth, I've tried convincing a friend at AOL to start using
Python.  They mostly use TCL with AOLServer[1] (webserver with TCL built
in at a very low level, quite fast); with a bit of Perl.

He's not a trained programmer, and really likes TCL.   Since they've more
or less standardized on *one* language across many developers, they have a
large library of code to steal from, which makes nearly all new projects
trivial to whip up.  Switching to a "real" (OO, scripting) language really
wouldnt help them out.

Ironically, TCL *does* have some object-oriented features.  When I tried
it out many years ago, there was a [itcl] extension to basically add
C++-style object suppport.  But, all the TCL types I've talked to since
have never heard of it.  Odd.

Has anyone messed with Guile, GNU's extension language?  One specific
thing they want to do is be a "universal" back end, that is you can write
translators from language A to Guile, and their page[2] mentions Python
several times...


- j


[1] http://www.aolserver.com/

[2] http://www.gnu.org/software/guile/guile.html



John Mitchell        "It's not gentlemanly, but this industry lost
                     any pretense of sportsmanship long ago."
johnm at magnet.com     -- Charles Cooper, "Why it's easy to hate Microsoft"





From abc at def.ghi.jkl  Fri Jun 11 10:10:55 1999
From: abc at def.ghi.jkl (DORAEMON.HK)
Date: Fri, 11 Jun 1999 22:10:55 +0800
Subject: Serial programming
References: <7jp1jt$8hg$1@imsp009a.netvigator.com> <cuV73.1455$h14.6642@newsr2.twcny.rr.com>
Message-ID: <7jr509$912$1@imsp009a.netvigator.com>

Oh! Sorry! I am using Linux

Darrell <news at dorb.com> wrote in message
news:cuV73.1455$h14.6642 at newsr2.twcny.rr.com...
> Assuming your in Windows.
> http://www.python.org/windows/
> Although I expect there is a better way using win32 stuff.
>
http://msdn.microsoft.com/isapi/msdnlib.idc?theURL=/library/sdkdoc/portals/m
> ainport.htm
>
> --
> --Darrell
> DORAEMON.HK <abc at def.ghi.jkl> wrote in message
> news:7jp1jt$8hg$1 at imsp009a.netvigator.com...
> > I am looking for some example in python about send and receive data from
> > serial port.  Like send "ATZ" and receive "OK".  Any help?? Thanks a
lot!!
> >
> >
> > Doraemon
> >
>





From jso at boxer.no.---  Tue Jun 29 03:24:54 1999
From: jso at boxer.no.--- (Jostein =?iso-8859-1?Q?S=F8iland?=)
Date: Tue, 29 Jun 1999 09:24:54 +0200
Subject: Uploading files
Message-ID: <377874C6.6941CCFF@boxer.no.--->

Hi, I'm writing a script that needs to upload files from the user's
machine to a server, and it's used in combination with a form and
HTML-file that lies on the server side. I wrote a piece of code that I
thought would work, but it turned out that when i specify c:\foo.exe, it
is the file on the server side that is uploaded and not the foo.exe on
the client side. Evidently since the script is run on the server side.

So my question is: How do I specify that it is the client's c:\foo.exe
and not the server file that gets uploaded?


Thanks

Jostein S?iland




From pavlos at gaaros.com  Wed Jun 30 18:41:27 1999
From: pavlos at gaaros.com (pavlos at gaaros.com)
Date: Wed, 30 Jun 1999 22:41:27 GMT
Subject: Uploading files
In-Reply-To: <377874C6.6941CCFF@boxer.no.--->
References: <377874C6.6941CCFF@boxer.no.--->
Message-ID: <7le6en$jta6@eGroups.com>

 <377874c6.6941ccf- at boxer.no.---> wrote: 
original article:http://www.egroups.com/group/python-list/?start=61969
> Hi, I'm writing a script that needs to upload files from the user's
> machine to a server, and it's used in combination with a form and
> HTML-file that lies on the server side. I wrote a piece of code that I
> thought would work, but it turned out that when i specify c:\foo.exe, it
> is the file on the server side that is uploaded and not the foo.exe on
> the client side. Evidently since the script is run on the server side.
> 
> So my question is: How do I specify that it is the client's c:\foo.exe
> and not the server file that gets uploaded?
> 

A bit unrelated. Digital Creation ZPublisher is one of the easiest ways to
web publish python objects. Coincidentally I have a ZPublisher app 
that does exactly what you want. You can download it from:
http://starship.python.net/crew/pavlos

Pavlos




From fatjim at home.com  Tue Jun 22 16:50:34 1999
From: fatjim at home.com (Jim Meier)
Date: Tue, 22 Jun 1999 20:50:34 GMT
Subject: Win32 Question
References: <7koiib$fdu$1@nnrp1.deja.com>
Message-ID: <376FF68A.5698AF68@home.com>


sjoshi at ingr.com wrote:

>  I want to find out if a certaing application is releasing memor or not
> when it terminates and was wondering how to go about doing that using
> win32api.
>   I would appreciate if someone could point me in the right direction

I've never worked with NT, but a simple batch file containing something
like:

python myscript.py %1 %2 %3 %4 %5

does file on my system. I'd like to see more elegant solutions, though!

-Jim





From belbo_j at my-deja.com  Fri Jun 25 10:54:19 1999
From: belbo_j at my-deja.com (belbo_j at my-deja.com)
Date: Fri, 25 Jun 1999 14:54:19 GMT
Subject: ODBC connectivity to Access!
References: <37723F25.31033307@wipsys.soft.net>
Message-ID: <7l056k$64l$1@nnrp1.deja.com>

kvbalaji at wipsys.soft.net wrote:

> Hi ,
>   Can any body tell me how to connect and retrieve data stored in a
> MS Access database using ODBC in python?
> I am new to this language, Hints, pointers welcome!
>
> regards,
> bala

In my case I'm using the ODBC module that you can find in Python Win
See http://www.python.org/windows/win32/odbc.html
Unfortunately this module is not supported by his author :-(

Jacobo


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From olipt at mayo.edu  Fri Jun 11 18:22:41 1999
From: olipt at mayo.edu (Travis Oliphant)
Date: Fri, 11 Jun 1999 17:22:41 -0500
Subject: Python and C, arrays
In-Reply-To: <7jrqs5$otb$1@nnrp1.deja.com>
References: <7jrqs5$otb$1@nnrp1.deja.com>
Message-ID: <Pine.LNX.4.10.9906111712380.19664-100000@us2.mayo.edu>

> Hey folks,
> 
> I've run into a problem with a module I'm writing in C.  Basically, I
> need to be able to convert Python tuples or lists (it's unimportant
> which) into C arrays.  I know at compile-time the type of the array (and
> that the tuple or list will be homogeneous, and of compatible type), but
> I don't know the size.
> 
> I then need to find a way to take a C array and stuff it back into a
> Python tuple or list.  I have some ideas involving using "stack-like"
> lists or tuples in Python, with a "pop" function which I could call from
> C, repeated in a loop to get everything.  But this seems overly complex,
> and indeed like it has the potential to become truly hideous.

There are lots of ways to do this.  I'll let others respond with more
"standard" Python methods.

But, my favorite way to interface with homogeneous C arrays is using the C
API of the Numeric extension module:

PyArrayObject *c;

c = PyArray_ContiguousFromObject(some_sequence,the_type,mindim,maxdim);

c->data is now a pointer to the C array.

PyArray_Return(c) returns the Numeric object to Python.

Note that the Numeric array can be just like any other sequence object.
But, if you must have a list you can use the tolist() method of the
Numeric object.

Good luck,

Travis Oliphant





From ovidiu at cup.hp.com  Wed Jun 16 18:20:19 1999
From: ovidiu at cup.hp.com (Ovidiu Predescu)
Date: Wed, 16 Jun 1999 22:20:19 +0000
Subject: Emacs package for the Python debugger?
References: <3767FEB1.87D18512@cup.hp.com>
Message-ID: <37682323.99768146@cup.hp.com>

Ovidiu Predescu wrote:

> Is there any package that allows debugging Python programs from within
> Emacs? Something that is integrated with Emacs' GUD would be very
> convenient.

I found one, so no response is necessary. For those interested the
package could be found on:

http://www.python.org/emacs/

-- 
Ovidiu Predescu <ovidiu at cup.hp.com>
http://www.geocities.com/SiliconValley/Monitor/7464/



From matomira at iname.com  Tue Jun 15 04:33:49 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Tue, 15 Jun 1999 10:33:49 +0200
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <3762988E.C94AA252@iname.com> <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu> <3763C746.5CA8FE78@iname.com> <m3u2sby4gf.fsf@soma.andreas.org> <87pv2zjhhl.fsf@home.ivm.de> <ekvn1y2efz0.fsf@gnoll.pvv.ntnu.no> <slrn7mb2oq.lpd.wtanksle@dolphin.openprojects.net>
Message-ID: <37660FEC.D4F3371B@iname.com>


William Tanksley wrote:

> (I like all the languages I know, except Fortran.  I even like C++,
> although I wish they could break backwards compatibility.)

Quiz: What do Scheme, Common Lisp, HPF and Java have in common?





From William.Tanksley at p98.f112.n480.z2.fidonet.org  Tue Jun 29 19:27:42 1999
From: William.Tanksley at p98.f112.n480.z2.fidonet.org (William Tanksley)
Date: Wed, 30 Jun 1999 00:27:42 +0100
Subject: [Python|Crypto] on handhelds (was Re: Thanks for Python...)
Message-ID: <000015a0@bossar.com.pl>

From: wtanksle at dolphin.openprojects.net (William Tanksley)

On 29 Jun 1999 07:35:42 PDT, Mordy Ovits wrote:
>William Tanksley wrote:

>To each his own. :-)  Don't you wish all flame were required to end like that
>(i.e. before they started?)

Yes, it is kinda nice.  Can we please argue about something now?  This IS
Usenet, after all -- I can't pass the Turing test on other topics.

>> >> I used Quartus on my Crypto exam to solve a modular equation after I
>> >> forgot how to solve modular equations (I had it use brute force).

>> >Wow!  I wish I had had my E100 when I was in college.  I just realized how
>> >awesome that would have been. :-)

>> Actually, I sometimes miss my old OmniGo, which was a pocket computer, in
>> spite of being slower than the Pilot for some tasks.  I really needed the
>> glorified organizer (and had been fooled into thinking that the OmniGo was
>> one), but the OmniGo had some abilties that the Pilot doesn't.

>I'm not familiar with the OmniGo.  Got a link for me?

Sure.  It's discontinued, because HP got touched by Microsoft, but it was
a decent little tool, just a little slow to handle things that an
organizer needs to do quickly (in spite of the fact that it's faster than
a Pilot in actual speed).

http://www.geoworks.com/devices/omnigo/ seems to be the best thing I can
find -- there was a really good user site, but I lost the link and Google
can't find it (there are too many Omnigo mailing lists).

>> >The SECOND time I saw that sig was in ESR's mail.  The first was mine.  I
doubt
>> >ESR copied mine, but who knows? :-)

>> You mean you wrote it?  If so, there seems to have crept into it a bug --
>> one of the implicit unpackings don't work.

>No, I didn't write it.  It was written by the all-talented AMK.  I did write a
>nifty version of Bruuce Schneier's Solitaire algorithm, though.  check out:
>http://www.counterpane.com/solitaire.html

I've read that -- I always read Counterpane's site :-).

>Mordy

-- 
-William "Billy" Tanksley



From bwarsaw at cnri.reston.va.us  Mon Jun 28 14:43:24 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Mon, 28 Jun 1999 14:43:24 -0400 (EDT)
Subject: python.org scheduled downtime
Message-ID: <14199.49740.267357.848009@anthem.cnri.reston.va.us>

Folks,

python.org will be taken off-line for some system maintenance.

Scheduled downtime is for Tuesday, 29-June-1999, starting at 10am EDT.
I don't know how long the machine will be down for, but it will
probably be a few hours at least.

This will affect Web, ftp, email and mailing lists for the python.org
and jpython.org domains.

Apologies for any inconveniences this will cause.
-Barry



From William.H.Duquette at jpl.nasa.gov  Wed Jun 30 19:07:23 1999
From: William.H.Duquette at jpl.nasa.gov (William H. Duquette)
Date: Wed, 30 Jun 1999 23:07:23 GMT
Subject: Language extensibility (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <37682DD3.43923243@compaq.com> <7ka2be$qja$1@nntp6.u.washington.edu> <37776FFA.2D85@mailserver.hursley.ibm.com> <lw7looxx7m.fsf@copernico.parades.rm.cnr.it> <876748mcxg.fsf@home.ivm.de> <37788BB9.ABD@mailserver.hursley.ibm.com> <lw67471fyi.fsf@copernico.parades.rm.cnr.it> <3779ED0F.2C67@mailserver.hursley.ibm.com> <nkjzp1hizvb.fsf@tfeb.org> <377A3383.D453468@iname.com>
Message-ID: <377aa228.31380423@news.jpl.nasa.gov>

On Wed, 30 Jun 1999 17:10:59 +0200, Fernando Mato Mira
<matomira at iname.com> wrote:

>
>
>Tim Bradshaw wrote:
>
>> But the language of Lisp macros is Lisp, that's really the whole
>> point!  Without knowing TCL, I find it hard to see how you can
>> introduce new constructs to the language *without* a macro language,
>> even if that language is TCL.
>
>Err.. What about Forth keywords?

Tcl is a lot like Forth, too. :-)

A Tcl command is just a procedure that can process its white-space
delimited arguments however it likes, including evaluating them as Tcl
code. There's nothing in the language but Tcl commands; the standard
control structures like "if", "for", "foreach", and "while" can
be written in Tcl.  By using the same kinds of tricks, you can
add new constructs of whatever kind you like, *provided* that
your new constructs have standard Tcl syntax.

You can do more interesting things in Lisp with readmacros,
from what I'm told, including making it look like an entirely
different language.  Tcl code always looks like Tcl code.

Will

--------------------------------------------------------------------------
Will Duquette, JPL  | William.H.Duquette at jpl.nasa.gov
But I speak only    | http://eis.jpl.nasa.gov/~will (JPL Use Only)
for myself.         | It's amazing what you can do with the right tools.



From vlachid at otenet.gr  Sat Jun 26 01:35:44 1999
From: vlachid at otenet.gr (Vlachidis Costas)
Date: Sat, 26 Jun 1999 08:35:44 +0300
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <JWQc3.2620$U6.10996@newsr2.twcny.rr.com> <3773EE44.F231F1E4@otenet.gr> <_ASc3.2630$U6.11110@newsr2.twcny.rr.com>
Message-ID: <377466B0.CABE051E@otenet.gr>

Hi
with Tkinter you have right, it's an other language.But wx is just a C-library
with well defined functions that can be called like black boxes, and is very
compact.The problem withthis is thwt it is not completed and it can't cnvert all
modern GUI'.s functions yet.There'is no image processing,no sound functions, no
multimedia at all.
I think that insteed to learning another language like Java,we must develop
wxPython modules as a built in library.
Costas

Darrell wrote:

> Is this more trouble than learning tkinter and or wxPython ?
> I just installed and gave the jpi thing a try. It's not ready for prime time
> yet.
>
> --
> --Darrell
-------------- next part --------------
A non-text attachment was scrubbed...
Name: vlachid.vcf
Type: text/x-vcard
Size: 366 bytes
Desc: Card for Vlachidis Costas
URL: <http://mail.python.org/pipermail/python-list/attachments/19990626/88cd91f3/attachment.vcf>

From eugene.leitl at lrz.uni-muenchen.de  Sun Jun  6 23:57:49 1999
From: eugene.leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Sun,  6 Jun 1999 20:57:49 -0700 (PDT)
Subject: Python CPU
Message-ID: <14171.17113.39933.904980@lrz.de>

Have there been any attempts in creating an ASIC capable of
interpreting Python opcodes directly? Of course, with 
soon-to-arrive Java chips and JPython there might be not 
that much need for it...



From mark.butterworth at cis.co.uk  Fri Jun 25 09:21:35 1999
From: mark.butterworth at cis.co.uk (Mark Butterworth)
Date: Fri, 25 Jun 1999 14:21:35 +0100
Subject: pkzip and using popen on Windows 98 ???
References: <2H1c3.180$T3.411@newsr2.u-net.net> <37734D77.1979CBD7@phoenix-edv.netzservice.de>
Message-ID: <7cMc3.533$T3.1070@newsr2.u-net.net>

>
>Tested module GZIP?
>
>f = gzip.GzipFile(fromPath , 'rb')
>var = f.read()
>f.close()
>writeFile(toPath, var)
>
>..works fine for unzipping file.

Yes but gzip only works with single files,  the existing data I need to
manipulate is stored in pkzipped files with a password! therefore I need
full pkzip format support.





From jeremy at cnri.reston.va.us  Wed Jun  2 12:01:12 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Wed,  2 Jun 1999 12:01:12 -0400 (EDT)
Subject: HTTPlib.py error
In-Reply-To: <1318D78C9072D11195C9006094EA98A72C20C4@ocesrv>
References: <1318D78C9072D11195C9006094EA98A72C20C4@ocesrv>
Message-ID: <14165.21296.2050.181547@bitdiddle.cnri.reston.va.us>

>        [MCepl]  Perhaps I am behind firewall, which doesn't allow me to
>do this. Is this stupid or not considering that I am using WWW all the
>time?

If you are using WWW all the time, then you should be able to make
httplib and linbot work.  You may need to do a little configuration.
I had forgotten that firewalls require some manual configuration!

(In general, you can get some more debugging information by running
python httplib.py -d.  In this particular case, I don't think it's
going to help any.  The problem occurs before httplib.py has a chance
to make use of the -d option.)

Is your browser configured in some special way to pass through the
firewall?  I use Netscape, and I can configure proxies from the
Preferences dialog, under Advanced->Proxies.  If you do have a proxy
configured for your browser, you are going to have problems using
httplib.  However, urllib will work if you set up the http_proxy
environment variable.

Can you load www.python.org from your Web browser?  It would be odd if 
you could load it from the browser and not from a Python script.

Jeremy



From a.mueller at icrf.icnet.uk  Mon Jun  7 14:59:53 1999
From: a.mueller at icrf.icnet.uk (Arne Mueller)
Date: Mon, 07 Jun 1999 19:59:53 +0100
Subject: waste of resources ?
References: <375C0F91.1C73E4E3@icrf.icnet.uk>
Message-ID: <375C16A9.5ADC4990@icrf.icnet.uk>

Arne Mueller wrote:
> 
> Hi All
> 
> In my program I always get an OSError number 11 (Resource temporarely
> not available). It reads in lots of files using child processes
> communication with the parent. Each child also uses popen to get data
> from lauched 'gunzip' process. After ~1000 processed files the program
> raises OSError[11]. I'm pretty sure that all the child processes close
> their pipes to the parent and that popen is also closed. The parent
> stores the file descriptors connected to the children in a list but uses
> the lists 'remove' method to delete descriptors that are finished (those
> which the parent has already read from).
> 
> However, I'd like to get sure that's not the number of used
> filedescriptors what's causing the error. How can I examine the cause of
> error, I mean is there a python method or a variable that tell you the
> number of currently used file descriptors or better that lists all the
> resources a program is using including it's children?

I think I found the source of error:

Im running python 1.5.2b2 on SGI power challange irix64 version 6.5.
Everytime a child process is created via fork(), it realy does it's job
but never dies, instead the number of zombie processes (as reported by
the 'top' program) steadily increases, though I call the sys.exit(0)
method for the child. Any idea how exit a process without creating a
zombie?

	thanks alot,

	Arne



From tuna at delirious.fsnnet  Thu Jun 24 07:40:25 1999
From: tuna at delirious.fsnnet (Lloyd Weehuizen)
Date: 24 Jun 1999 11:40:25 GMT
Subject: Embedding Question
Message-ID: <slrn7n4699.g36.tuna@delirious.fsnnet>

Hi

If I use the PyRun_String method, with a string that defines a method called
"xyz", is there any way to remove this method from the Python interpreters
"memory" without doing a Finalize() and Initialize(), because I am running
multiple Python scripts, and I don't want there methods getting all mixed
up, and I presume Finalize() and Initialize() are not very speedy ;)

Thanks
Lloyd 
Lloyd 




From fredrik at pythonware.com  Thu Jun  3 15:36:29 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 3 Jun 1999 19:36:29 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>  <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>  <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net>  <87hfosquc0.fsf@ev.netlab.co.jp> <3753F121.E4A9BE8B@prescod.net> <87yai3ntdt.fsf@ev.netlab.co.jp> <7j3iif$134e$1@nntp6.u.washington.edu> <7j45iv$908$4@cronkite.cc.uga.edu> <7j4hhj$118u$1@nntp6.u.washington.edu> <7j4jbe$dhh$1@cronkite.cc.uga.edu> <37564A02.F115B9C5@infercor.no>
Message-ID: <045501beadf8$610e62e0$f29b12c2@pythonware.com>

Paul Boddie <paulb at infercor.no> wrote:
> So, we have established that it can indeed be done, but reaching back into the
> depths of time and pulling out the original proposal - being that Python should
> have its primary (or only) implementation running in the Java environment - we
> can see that regardless of the existence of satisfactory GC schemes, the typical
> Java implementation surely falls short of what many people require.
> 
> Therefore, whilst there may well be a need for more sophisticated GC in Python
> 2.0, we must conclude that the Java environment may not be the appropriate place
> in which such technologies are to be found.

does this mean that there are no cyclic references in this
thread, and it will be garbage collected soon even without
us adding "true" gc to c.l.^h^h^h^hpython?

or are there two different Grahams in this thread, pointing
to each other?  ouch.

maybe Gordon could hypnotize Graham?

...

btw, has anyone looked at

http://www.egroups.com/GroupTopStatsPage?listName=python-list&show=topThreads

lately?  here's a summary:

Top Subjects

    Python 2.0 (99)
    while (a=b()) ... (95)
    Re: Is Python dying? (87)
    Python IS slow ! [was] Re: Python too slow for re... (67)
    Re: Python 2.0 (59)
    Defining VCL-like framework for Python (47)
    Python too slow for real world (46)
    Bytecode optimisation (35)
    Concrete Proposal: while ... and while ... (34)
    Python vs. VB (32)
    Python T-shirt (32)
    "?:", "a and b or c" or "iif" (29)
    The way to a faster python [was Python IS slow !]... (27)

sure looks like the "community" thinks that changing the
language is more important that using it...

</F>




From tim_one at email.msn.com  Tue Jun  1 23:01:05 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 1 Jun 1999 23:01:05 -0400
Subject: python extension problem (newbie)
In-Reply-To: <37548FA5.526A8E48@crystal.uwa.edu.au>
Message-ID: <001301beaca4$28a887a0$d89e2299@tim>

[Douglas du Boulay]
> I've been trying to write a python extension in C to interface to
> a Fortran program and a class wrapper for the extension ...
> It seems that the __getattr__  method of the wrapper class cannot
> find the getattr method of the extension library ...

The good news is that it has nothing to do with your extension library, as
you can see by running this all-Python one-file equivalent:

class X:
    pass

class Grasp:
    def __init__(self, start=None):
        self._base = start or X()
        print "doing done"                  # which is never executed
    def __getattr__(self, name):
        print "called getattr"              # executed  endlessly
        return getattr( self._base , name)
    def __setattr__(self, name, value):
        print "called setattr"
        return setattr(self._base, name, value)

i = Grasp()

__init__ gets back from X() fine, and tries to bind self._base to the
result.

But you defined __setattr__, so trying to bind self._base calls __setattr__
(note that I put a "called setattr" print in there -- which, had you done
that, would have provided the clue you're missing!  __setattr__ is your real
problem, not __getattr__).

The first thing __setattr__ does is try to look up self._base.  But _base
hasn't been bound yet, so __getattr__ is called.

The first thing __getattr__ does is also to try to look up self._base.  But
_base still hasn't been bound yet, so __getattr__ is called.

The first thing __getattr__ does is also to try to look up self._base.  But
_base still hasn't been bound yet, so __getattr__ is called.

The first thing __getattr__ does is also to try to look up self._base.  But
_base still hasn't been bound yet, so __getattr__ is called.

The first thing __getattr__ does is also to try to look up self._base.  But
_base still hasn't been bound yet, so __getattr__ is called.

The attentive reader will detect a subtle pattern <wink>.

Defining __setattr__ is powerful but delicate, and you fell into a common
trap (well, common among people who muck with this stuff at all ...).  The
easiest way out is to bypass the killer __setattr__ by changing the first
__init__ line to

        self.__dict__['_base'] = start or X()

the-attribute-that-cannot-be-set-is-not-the-true-attribute-ly y'rs  - tim





From faassen at pop.vet.uu.nl  Thu Jun  3 17:17:22 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Thu, 03 Jun 1999 23:17:22 +0200
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl>
			<375536C2.DD917F99@prescod.net>
			<aahzFCpHw8.Gxt@netcom.com>
			<7j4e2u$8rh$1@m2.c2.telstra-mm.net.au>
			<37564132.2D01B7C3@pop.vet.uu.nl> <3756BAFC.2066A636@home.com> <3756C2F6.9D7E6EFD@appliedbiometrics.com>
Message-ID: <3756F0E2.91E24FD9@pop.vet.uu.nl>

Christian Tismer wrote:
> > Wow, talking about old H/W makes me sound old - I'm only 19!
> 
> Are you sure? How did you get your hands at a VIC-20, then? :-))

Perhaps he's related to BIFF. He posts from his big brothers' VIC-20.
No, no, only kidding, you're not related to BIFF at all, sorry Jim! :)

Regards,

Martijn



From emile at fenx.com  Sun Jun  6 13:39:02 1999
From: emile at fenx.com (Emile van Sebille)
Date: Sun, 6 Jun 1999 10:39:02 -0700
Subject: [pickle] Different version of a class
References: <375a9760@nntp.server.uni-frankfurt.de>
Message-ID: <Goy63.1158$VL2.171084@news.direcpc.com>

It seems I've seen references ( a quick scan of dejanews didn't locate them
however) to keeping a version variable in your data that you inspect upon
loading.  Based on this, you can initialize just the new fields.

--

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


<mlauer at asmoday-atm.rz.uni-frankfurt.de> wrote in message
news:375a9760 at nntp.server.uni-frankfurt.de...
> Hi,
>
> my application uses pickle to store and load data.
>
> During developing cycle, many variables are added
> to the classes which are pickled. Whenever I load
> a file pickled with a previous version, some
> variables are not existent (naturally, because
> __init__ is not called). Is there an easy way
> to initialize the "new" variables or must I
> call __init__ - which would overwrite some of
> my saved variables.
>
> Any idea ?
>
> --
>     Regards & Gruesse from Mickey @ http://www.Vanille.de
>   ---------------------------------------------------------
>   How could anyone know me - when I don't even know myself ?
>
>





From donn at u.washington.edu  Mon Jun 14 10:58:55 1999
From: donn at u.washington.edu (Donn Cave)
Date: 14 Jun 1999 14:58:55 GMT
Subject: fork()
References: <000c01beb3c1$a3f428a0$329e2299@tim> 	<199906110443.AAA02408@eric.cnri.reston.va.us> 	<Pine.SUN.3.95-heb-2.07.990611075346.4942G-100000@sunset.ma.huji.ac.il> <u3d7yzn1np.fsf@ares.sys1.ngo.okisoft.co.jp>
Message-ID: <7k35bf$13iq$1@nntp6.u.washington.edu>

Hisao Suzuki <suzuki611 at okisoft.co.jp> writes:
...
| Such an action will break, say, an unreachable doubly-linked
| tree into separated nodes.  Then the nodes will be destroyed by
| a normal process including invocations of __del__.  To make this
| occur,
|  (a) the programmer must register the root of the tree with a
|      function to break up the tree, which will be invoked later
|      automatically by the garbage collector, or
|
|  (b) the programmer must register the root of the tree, and at
|      certain point of the program she/he must explicitly invoke
|      a function which retrieves the trees unreachable except
|      from the `registration' server and breaks them up.
|
|  (In any case, unreachable trees that contain no `registered'
|   objects will be recycled with no invocations of __del__.)

I want to ask more about this, partly I'm not sure I understand what
the trees and their roots are or what registration would require.

Could something like a standard class method accomplish this?  Like,

   def __garbagecollected__(self):
       # author knows self.parent is a circular reference.
       self.parent = None

   def __del__(self):
       if self.parent:
           ...


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



From mwh21 at cam.ac.uk  Wed Jun 23 02:58:14 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 23 Jun 1999 07:58:14 +0100
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0rr2.2fi.behrends@allegro.cse.msu.edu>
Message-ID: <m3aetr2yk9.fsf@atrus.jesus.cam.ac.uk>

behrends at cse.msu.edu (Reimer Behrends) writes:

> Preston Landers (planders at mail.utexas.edu) wrote:
> > Can someone *please* give me a good reason why Python doesn't yet have 
> > +=, -=, and .= and other such operators?  I know it's just syntactic
> > sugar, but damn, it is such sweet sugar!
> 
> Few people, I think, have ever tried to objectively count how often such
> a construct is used. The Python 1.5.2 source, for instance, has
> approximately one of ++, --, +=, -=, *=, /= every 40 lines. If you do
> not count the occurrences if ++ and -- in for loops, for which there's a
> construct in Python anyway, you can roughly double that number. Now, C
> is not Python, but it's a good first-order approximation of how often
> you need it--the Python source is one of the cleanest C programs I know
> of.
> 
> Personally, I can easily live without a construct that I can use on
> average once every three screens of code or so. On the other hand, I
> _am_ getting pretty sick of having to type self.whatever countless
> times, where pretty much any other language allows me to discard the
> "self." part. Not only is it annoying to type, it also reduces the
> readability of OO code a lot, much like the gratuitious use of
> punctuation characters as variable prefixes in Perl does.

I sometimes think this when writing Python; but in my summer job which
I've just started, I'm frequently faced with large screenfulls of C++
and not being able to tell whether a given variable is local, static,
class or global at a glance is mightily annoying. 

Just my humble opinion.
Michael
 
> Among all the suggestions for Python 2.0, removing the necessity of
> prefixing methods and instance variables with "self." (and not having
> to list it as an argument, either) would be my #1 wish.
> 
> [...]
> 
> 				Reimer Behrends



From hniksic at srce.hr  Fri Jun  4 17:09:04 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 04 Jun 1999 23:09:04 +0200
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7iqhkb$d7s@wiscnews.wiscnet.net> <7iu7ma$d33$1@cronkite.cc.uga.edu> <02hfonzsx2.fsf@rtp.ericsson.se> <02909zzou0.fsf@rtp.ericsson.se>
Message-ID: <87zp2f4qrz.fsf@pc-hrvoje.srce.hr>

Kumar Balachandran <kumar*xspam*@*xspam*rtp.ericsson.se> writes:

> What you said is quite wrong. Since Python provides closures, I can
> change my programming style to be completely different from the
> standard C or Fortran style and use a functional interface.

Since when does Python provide closures?



From klassa at aur.alcatel.com  Thu Jun 24 10:35:12 1999
From: klassa at aur.alcatel.com (John Klassa)
Date: 24 Jun 1999 14:35:12 GMT
Subject: Thanks for Python...
References: <376ead64@anonymous.newsfeeds.com> <VrDQRddcbbs9-pn2-QdftOXYk6Hdd@Fortress> <37716901.43CB604@lockstar.com> <slrn7n2gkr.i3k.wtanksle@dolphin.openprojects.net>
Message-ID: <7ktfn0$5q$1@aurwww.aur.alcatel.com>

On Wed, 23 Jun 1999 20:24:59 GMT, wtanksle at dolphin.openprojects.net wrote:
  > >#!/usr/local/bin/python 
  > >from sys import*;from string import*;a=argv;[s,p,q]=filter(lambda x:x[:1]!=
  > >'-',a);d='-d'in a;e,n=atol(p,16),atol(q,16);l=(len(q)+1)/2;o,inb=l-d,l-1+d
  > >while s:s=stdin.read(inb);s and map(stdout.write,map(lambda i,b=pow(reduce(
  > >lambda x,y:(x<<8L)+y,map(ord,s)),e,n):chr(b>>8*i&255),range(o-1,-1,-1)))
  > 
  > You know, the first time I saw this it was in ESR's .sig, and I tried to
  > remember how long the Perl version had been.  A few minutes of searching
  > revealed that it was three lines long -- but also revealed that it was
  > using bc rather than Perl to do most of the work!  I spent about half an
  > hour searching for the real Perl version, only to find that it was five
  > lines long.
  > 
  > The Python for this .sig is shorter than the Perl!  Cool, eh?

It doesn't run for me...  What's it supposed to do?  When I run it, I get:

Traceback (innermost last):
  File "/tmp/foo", line 2, in ?
    from sys import*;from string import*;a=argv;[s,p,q]=filter(lambda
x:x[:1]!=
ValueError: unpack list of wrong size

-- 
John Klassa / Alcatel USA / Raleigh, NC, USA



From mgushee at havenrock.com  Sun Jun 20 12:04:35 1999
From: mgushee at havenrock.com (Matt Gushee)
Date: 21 Jun 1999 01:04:35 +0900
Subject: Newbie question
References: <376BBB7E.A82DA789@primenet.com> <199906191928.VAA20205@axil.hvision.nl> <376C4579.C7B38E25@primenet.com>
Message-ID: <m3lndeq2ng.fsf@havenrock.com>

"John E. Garrott" <johnga at primenet.com> writes:

> Hans Nowak wrote:
> > 
> > On 19 Jun 99, John E. Garrott wrote:
> > 
> > > I am have a problem with the program below.
> > >
> > > The function dbs.databases() works, putting,
> > > in this case, the databases owned by myself
> > > into the listbox.  When I click on the listbox
> > > items they highlight as expected, but, apparently,
> > > list_cb() is never called.

> > >               self.frame.listbox.bind('Button-1', self.list_cb)

The Tcl/Tk binding syntax uses angle brackets to designate events. Try 
this:

               self.frame.listbox.bind('<Button-1>', self.list_cb)
                                        ^        ^

Hope this helps.

Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com



From greg.ewing at compaq.com  Sun Jun 20 16:48:48 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Mon, 21 Jun 1999 08:48:48 +1200
Subject: Newbie: Truth values (three-valued logic)
References: <000201beb946$b9c99e00$969e2299@tim>
Message-ID: <376D53B0.18B9F003@compaq.com>

Tim Peters wrote:
> 
> Well, "and" and "or" do short-ciruit evaluation (i.e., they don't evaluate
> their right-hand operand at all if the value of their left-hand operand
> suffices to determine the result).  For that reason, they're more properly
> viewed as control structures than operators, and so unlikely to get "opened
> up" to user intervention.

It could be done if Python 2.0 gets lexical closures.
The __and__ and __or__ methods would get the value of
the left argument and a function for evaluating the
right argument. Then they could be lazy or not as
they pleased.

Greg



From fredrik at pythonware.com  Tue Jun 29 05:37:09 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 29 Jun 1999 09:37:09 GMT
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net> <3773c3a0.102825445@news.oh.verio.com> <xPPc3.535$or1.22866@typhoon-sf.snfc21.pbi.net> <xJRc3.292$Of3.67094@news.shore.net>
Message-ID: <00f701bec212$f6a83050$f29b12c2@secret.pythonware.com>

Michael P. Reilly <arcege at shore.net> wrote:
> But.. the argument over (these two) GUIs is not exactly a valid one.
> Tkinter was written using the Tcl interface not the Tk/C API.  If
> Tkinter was rewritten to use the C API, it probably wouldn't be slower
> than wxPython.

The interface is only part of the story; many complaints
seem to come from the fact that Tk uses idle tasks to
redraw things.  When an "expose" event arrives from
the windowing system, Tk calls "after_idle", rather
than redrawing the widget at once.  And idle tasks
are only executed when there are no other events
to take care of...

Tk also uses double buffering for all widgets, which is
quite expensive, especially since Tk allocates a new
buffer for each update :-(

  In fact, since wxPython has to go thru extra layers, it
> would probably be slower (Tk/C is written on the native API on all the
> platforms).

What native API?  As of 8.0, Tk still emulates all widgets
except the scrollbar, and use a somewhat ineffective
X emulation layer for graphics.  AFAIK, they haven't
changed that in 8.1...

(We've fixed all this in uiToolkit, of course, but that's
another story ;-)

</F>




From alex at mindlube.com  Tue Jun 15 22:32:15 1999
From: alex at mindlube.com (Alex Rice)
Date: 15 Jun 1999 20:32:15 -0600
Subject: python-mode question
Message-ID: <m3hfo8an9s.fsf@monsoon.swcp.com>

When I run python-mode in Emacs 20, I see a second Python menu that
always shows me a heirarchy of my method and class names in the
current buffer. In XEmacs, it's not there. How do I display that?
Using version 3.90.

TIA

-- 
Alex Rice    |    alrice at swcp.com    |    http://www.swcp.com/~alrice
    Current Location: N. Rio Grande Bioregion, Southwestern USA




From m01ymu00 at cwcom.net  Fri Jun  4 11:42:54 1999
From: m01ymu00 at cwcom.net (Daniel Faulkner)
Date: Fri, 04 Jun 1999 16:42:54 +0100
Subject: problems using Tkinter
References: <928350546.726.80@news.remarQ.com>
Message-ID: <3757F3FD.3BA3804@cwcom.net>

Have you tried making you programme like this:

class tester(Frame):
    def Create_Widgets(self):
        self.QuitB = Button(butFrame, text='Quit', fg='red',
command=self.quit)
        self.QuitB.pack(side='left')
    def __init__(self, master=None):
        Frame.__init__(self, master)
        self.pack()
        self.Create_Widgets()

test = tester()
test.mainloop()

I know it is not much help but if this works it is better than not being
able to do anything.
Dan

Gordon Vinther wrote:

> I get the following error whenever I attempt to use Tk
>
> Os AIX4.2.1
> Python 1.5.2
> Tk/TCL 8.0.1
>
> from Tkinter import  *  .............this works
> root=Tk() ........... this errors with the following message
>
> File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 886, in __init__
> self.tk = _tkinter.create(screenName, baseName, className)
>
> TclError: invalid command name "tcl_findLibrary"






From heller at ms.tlk.com  Tue Jun  1 13:18:40 1999
From: heller at ms.tlk.com (Thomas Heller)
Date: Tue, 1 Jun 1999 19:18:40 +0200
Subject: need help checking for stack overflow in Win32
References: <5lk8to3su3.fsf@eric.cnri.reston.va.us>
Message-ID: <7j14fp$otv$1@mars.ms.tlk.com>

Guido van Rossum <guido at cnri.reston.va.us> wrote in message
news:5lk8to3su3.fsf at eric.cnri.reston.va.us...
> I could use some help from someone with a little time, some knowledge
> of Python internals, and some Win32 C expertise.  On Win32 platforms,
> infinite recursion causes an Application Error (a fatal windows
> error).  There is some machinery in Python to check for this
> condition, but apparently it doesn't work on Win32.  Please help!
>
> --Guido van Rossum (home page: http://www.python.org/~guido/)

I've just took a first look after this (under NT4 SP4, will also
have to check under Win95):

In principle the machinery seems to work, but the default reserved
stack size of 1MB is too small for the recursion limit of 10000
(a stack frame is 332 byte, and 10000 * 332 = 3.32 MB).
The stack will grow up to the reserved limit, but not beyond...
AFAIK the options are:
- Reserve a stack of about 4MB, or
- Decrease MAX_RECURSION_DEPS to 2500

Regards

Thomas Heller
ION-TOF GmbH






From cwebster at math.tamu.edu  Sun Jun 27 20:22:44 1999
From: cwebster at math.tamu.edu (Corran Webster)
Date: 28 Jun 1999 00:22:44 GMT
Subject: Advice for Newbie
References: <377588E9.82E8F567@dsuper.net>
Message-ID: <7l6f8k$cr5$1@news.tamu.edu>

In article <377588E9.82E8F567 at dsuper.net>, Dave  <redsand at dsuper.net> wrote:
>I read the Python homepage. I sounds promising. What books do you
>commend for a beginner?

People have suggested books, and I won't add to the comments because I
basically agree.  You might also like to subscribe to the Tutor mailing
list (details available at the Python website).  This'll give you access
to a group of more experienced folks who can answer your questions.

Corran




From cgw at fnal.gov  Wed Jun 23 11:38:18 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Wed, 23 Jun 1999 15:38:18 GMT
Subject: How to obtain an instance's name at runtime?
In-Reply-To: <7kqpu8$7u1$1@nnrp1.deja.com>
References: <7kqpu8$7u1$1@nnrp1.deja.com>
Message-ID: <14192.65386.504016.711375@buffalo.fnal.gov>

Please ignore my previous reply to this question,  I replied *way* too 
quickly without properly understanding what Dinu was looking for.

sometimes-I-wish-there-were-a-delay-on-outgoing-mail-ly y'rs,

							-cgw



From jon at rdt.monash.edu.au  Sat Jun  5 00:49:29 1999
From: jon at rdt.monash.edu.au (Jonathan Giddy)
Date: 5 Jun 1999 04:49:29 GMT
Subject: File locking
References: <199905301615.MAA00996@mira.erols.com>
Message-ID: <7jaa8p$rec$1@towncrier.cc.monash.edu.au>

"A.M. Kuchling" <amk1 at erols.com> writes:

>What is the correct solution for doing file locking in Python
>programs?  The docs for the posixfile module say:

After recent experimentation and reading of the source, I discovered that
Python's fcntl.lockf() accepts the flock() arguments, providing a portable 
and useful interface to fcntl locking.

So, the answer is:
fcntl.lockf(fd, FCNTL.LOCK_EX) -- set write (exclusive) lock
fcntl.lockf(fd, FCNTL.LOCK_SH) -- set read (shared) lock
fcntl.lockf(fd, FCNTL.LOCK_UN) -- unlock

with FCNTL.LOCK_NB available for non-blocking lock attempts.

The only trick is that FCNTL doesn't contain these values on all platforms, 
so you need something like:

from FCNTL import *

try:
    LOCK_SH
except NameError:
    LOCK_SH = 1
    LOCK_EX = 2
    LOCK_NB = 4 
    LOCK_UN = 8




From radbit at my-deja.com  Wed Jun 16 15:08:11 1999
From: radbit at my-deja.com (radbit at my-deja.com)
Date: Wed, 16 Jun 1999 19:08:11 GMT
Subject: do anonymous pipes normally work on NT?
References: <4D0A23B3F74DD111ACCD00805F31D8100DB90DDC@RED-MSG-50>
Message-ID: <7k8smh$t1p$1@nnrp1.deja.com>

In article <4D0A23B3F74DD111ACCD00805F31D8100DB90DDC at RED-MSG-50>,
  Bill Tutt <billtut at microsoft.com> wrote:
> Well, I finally noticed this thread, and I'll chime in with what I
found out
> so far:
> (No, I haven't figured it all out yet, its time for sleep.)
>
> Symptoms:
>  python -u cat.py < cat.py  is allowed to read from stdin.
>  python cat.py < cat.py isn't allowed to read from stdin.
>  python -u cat.py > blah is allowed to write to stdout
>  python cat.py > blah isn't allowed to wrtie to stdout.
>  python runproc.py (with -u) cat.py isn't allowed to write to stdout
>  python runproc.py cat.py isn't allowed to write to stdout.
>
> -u does two things:
> 1) turns stdin/stdout into binary streams
> 2) turns off stdio buffering

Hi

maybe I'm missing something about what you are looking for: here are 2
test scripts that work as anonymous pipes:

file runproc.py
---------------

'''runproc.py

start a process with three inherited pipes.
Try to write to and read from those.
'''

import win32api
import win32pipe
import win32file
import win32process
import win32security
import win32event
import os

#Constants for stdandard handles
STD_ERR_HANDLE=-12
STD_OUTPUT_HANDLE=-11
STD_INPUT_HANDLE=-10

class Process:
    def run(self, cmdline):

        # security attributes for pipes
        sAttrs = win32security.SECURITY_ATTRIBUTES()
        sAttrs.bInheritHandle = 1

        # create pipes
        hStdin_r,self.hStdin_w=win32pipe.CreatePipe(sAttrs, 0)
        self.hStdout_r,hStdout_w=win32pipe.CreatePipe(sAttrs,0)
        self.hStderr_r,hStderr_w=win32pipe.CreatePipe(sAttrs, 0)

        #associate Standard output to previously open handle
        win32api.SetStdHandle(STD_OUTPUT_HANDLE,hStdout_w)

        # set the info structure for the new process.
        StartupInfo = win32process.STARTUPINFO()

        # start the process.
        hProcess, hThread, dwPid, dwTid=win32process.CreateProcess(
                None, #program
                cmdline,# command line
                sAttrs,  # process security attributes
                sAttrs,  # thread attributes
                1,      # inherit handles,
                win32process.NORMAL_PRIORITY_CLASS,
                None,  # no new environment
                None,  # current directory (stay where we are)
                StartupInfo)
        # normally, we would save the pid etc. here...


        res, str = win32file.ReadFile(self.hStdout_r,100)
        if res == 0: print 'read:', str
        else: 'read nothing.'


if __name__ == '__main__':
    p = Process()
    p.run('..\\python.exe testpipe.py')
-----------------
end of runproc.py

file testpipe.py
----------------
import win32api
import win32file

STD_ERR_HANDLE=-12
STD_OUTPUT_HANDLE=-11
STD_INPUT_HANDLE=-10

std_in=win32api.GetStdHandle(STD_INPUT_HANDLE)
std_err=win32api.GetStdHandle(STD_ERR_HANDLE)
std_out=win32api.GetStdHandle(STD_OUTPUT_HANDLE)

win32file.WriteFile(std_out, 'this method works....')

end of testpipe.py
------------------


Cheers

Florent Heyworth


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From jenglish at flightlab.com  Wed Jun 30 16:03:05 1999
From: jenglish at flightlab.com (Joe English)
Date: 30 Jun 1999 13:03:05 -0700
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <lw67471fyi.fsf@copernico.parades.rm.cnr.it> <3779ED0F.2C67@mailserver.hursley.i <lwoghyexy5.fsf@copernico.parades.rm.cnr.it>
Message-ID: <7ldt5p$srh$1@dragon.flightlab.com>

Marco Antoniotti wrote:

>BTW. Is TCL statically or dynamically scoped?

Neither, really, since Tcl doesn't have nested procedures, lambda
expressions, "let/letrec", local bindings or the like.

I'd describe Tcl as "namespace-scoped": there is an open-ended set
of top-level "namespace" environments, plus a distinguished "global"
environment.  In addition, each Tcl "proc" has its own local environment,
into which variables from other environments can be explicitly imported
via "variable", "global", explicit namespace reference, or "upvar".
"upvar" gives you something *sort* of like dynamic scope, but not
exactly.


--Joe English

  jenglish at flightlab.com



From spamfranke at bigfoot.de  Fri Jun  4 13:11:13 1999
From: spamfranke at bigfoot.de (Stefan Franke)
Date: Fri, 04 Jun 1999 17:11:13 GMT
Subject: pickle vs .pyc
References: <7j43pn$55l$1@mlv.mit.edu>
Message-ID: <375902d4.10650955@news.omnilink.de>

On 2 Jun 1999 16:22:15 -0400, Michael Vezie <mlv at pobox.com> wrote:

>from cache import fast,slow

Note that importing compiles the cache module only if the .py file is newer than
the .pyc. 
While this should be no problem on a single machine, there are (unlikely) cases
where it could hurt you when working with multiple machines ot transporting your
files.  At least it is an external depency you may want to avoid. 

Stefan




From aaron_watters at my-deja.com  Tue Jun  8 08:53:57 1999
From: aaron_watters at my-deja.com (aaron_watters at my-deja.com)
Date: Tue, 08 Jun 1999 12:53:57 GMT
Subject: CGI module misbehaving?
References: <375B8BCF.C55B1F31@home.com> <Pine.LNX.4.04.9906071320360.5443-100000@emerald.netskate.ru>
Message-ID: <7jj3p0$jvq$1@nnrp1.deja.com>

In article <Pine.LNX.4.04.9906071320360.5443-
100000 at emerald.netskate.ru>,
  phd at sun.med.ru wrote:
> Hi!
>
> On Mon, 7 Jun 1999, Jim Meier wrote:
> > if form.has_key["name"]:
>
> if form.has_key("name"):
>
>    has_key is a function!

This is one of my most common typos for some reason.

Also I'd note that for cgi use you really should use a gadfly
server for safety.  It is conceivable that two cgi clients could
access the database at once otherwise and the updates from
one of them would probably get lost.  It will also probably run
faster if the database gets big.

  --  Aaron Watters http://www.chordate.com

===

May your hemorroids heal without surgery.
   -- blessing from "Bored of the Rings"


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From tim_one at email.msn.com  Fri Jun 18 00:55:01 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 18 Jun 1999 00:55:01 -0400
Subject: Newbie: Truth values (three-valued logic)
In-Reply-To: <Pine.OSF.4.05.9906161106470.295-100000@pinax.Mathematik.Uni-Bielefeld.DE>
Message-ID: <000201beb946$b9c99e00$969e2299@tim>

[Olaf Delgado]
> ...
> It came to my mind, though, that redefining negation alone will not
> help. Next time I'll want to redefine 'and' and 'or'. So, unless
> python 2 implements three-valued logic <hint, hint :)>, I'll have
> to come up with a routine for myself to handle these truth values
> consistently.

Well, "and" and "or" do short-ciruit evaluation (i.e., they don't evaluate
their right-hand operand at all if the value of their left-hand operand
suffices to determine the result).  For that reason, they're more properly
viewed as control structures than operators, and so unlikely to get "opened
up" to user intervention.

But in the spirit of overloading "~" instead of "not", consider overloading
"&" and "|" instead of "and" and "or".  Then you can do whatever you want.
If I were you I'd consider not overloading anything, and using ~ & |
directly with a naming trick, like so:

f, t, m = 0, -1, 42
name = {0: 'false', -1: 'true', 42: 'maybe', ~42: 'maybe'}

for x in f, t, m:
   print "not", name[x], "=", name[~x]

for x in f, t, m:
    for y in f, t, m:
        print name[x], "and", name[y], "=", name[x & y]

for x in f, t, m:
    for y in f, t, m:
        print name[x], "or", name[y], "=", name[x | y]

This hacks around assigning two values to "maybe", just so that it's closed
under negation.

Or use false=0, true=1, maybe=0.5; map __invert_(x)=1-x;
__and__(x,y)=min(x,y); and __or__(x,y)=max(x,y).  Then you get a nice little
MV logic with a whole bunch of possible "maybe" values.  Once you go beyond
the first distinction (true/false), I don't know why you'd ever want to stop
<wink>.

the-notion-that-something-is-true-is-as-silly-as-the-notion-that-
    something-is-false-ly y'rs  - tim





From blipf at yahoo.com  Fri Jun 11 12:33:14 1999
From: blipf at yahoo.com (Flbill Blipf)
Date: Fri, 11 Jun 1999 09:33:14 -0700
Subject: search and then replace
Message-ID: <37613A4A.F5D81F3E@yahoo.com>

Hi,

I can't figure out how to do search, then something, then
replace. I'm hoping someone will tell me I'm being too
complicated with this code. I should be using a parser for
this, but that's beside the point.

The idea is that I want to put <xref rid=...> inside,
<name>, but only when <name> is in <front>. RID is
grabbed from a dictionary keyed on names. So, I grab all
of <front> and do search and replace for each <name> in
<front>:

# I can assume that all of <front> will be on a single
# line.
#
# lines and ids_by_name are global
def wiggle_snoot():
  name_rex = re.compile('<name>(.*?)</name>')
  front_rex = re.compile('<front>(.*?)</front>')
  split_line = [0,0,0]
  split_sub = [0,0,0]
  for i in range(0, len(lines)):
    m = front_rex.search(lines[i])
    if not m: continue
    
    split_line[0] = lines[i][:m.start()]
    split_line[1] = lines[i][m.start():m.end()]
    split_line[2] = lines[i][m.end():]
      
    start = 0
    while 1:
      m = name_rex.search(split_line[1], start)
      if not m: break

      split_sub[0] = split_line[1][:m.start()]
      split_sub[1] = split_line[1][m.start():m.end()]
      split_sub[2] = split_line[1][m.end():]
      
      if ids_by_name.has_key(m.group(1)):
        split_sub[1] = ('%s%s%s' %
                ('<name><xref rid="%s">' %
                 (ids_by_name[m.group(1)]),
                 m.group(1),
                 '</xref></name>'))
      split_line[1] = join(split_sub, '')
      start = len(split_sub[0]) + len(split_sub[1])
    lines[i] = join(split_line, '')



From behrends at cse.msu.edu  Tue Jun 29 00:33:38 1999
From: behrends at cse.msu.edu (Reimer Behrends)
Date: 29 Jun 1999 04:33:38 GMT
Subject: total idiot question: +=, .=, etc...
References: <000001bebffc$12060200$da9e2299@tim>
Message-ID: <slrn7ngj6b.g9.behrends@allegro.cse.msu.edu>

Tim Peters (tim_one at email.msn.com) wrote:
[...]
> Let me ammend that, to "any OO language with declarations <that isn't
> Eiffel> has to ..." <wink>.  Eiffel allows inherited features to be accessed
> under local alias names as if the original name never existed (its "rename"
> clauses), so it has *a* way to recover locally from trivial superclass name
> changes.  Java (& most other languages) take a different approach.

These are two different things, actually. If you forbid shadowing of
instance variables by local variables, then simply renaming the local
variable in the subclass can fix the problem. This is applicable to
any language.

The related problem of adding a new method or attribute to a superclass
that already exists in a subclass, while most easily solved by a renaming
mechanism, is also amenable to a minor rewrite of the subclass. I am
assuming, of course, that you have a system where the compiler notifies
you if you accidentally redefine a method.

> > Yes, changing stuff in a superclass can thus break code in a subclass; but
> > this can always happen (like if you add a new method that also exists in
> > a subclass).  Doing a major internal change to a superclass requires at
> > least rechecking subclasses, anyway.
> 
> I did say "trivial".

Adding methods or instance variables to a superclass is never "trivial".
It always has the potential of breaking existing subclasses.

> A purely local way of addressing this is vital:  the
> subclass author may not be able to change the superclass code, or may not
> even have read access to its source code; in the other direction, a
> superclass author may have no idea who subclasses from them.

Yes. This is why changes to a superclass are always critical. However,
a well-designed subclass should always be able to recover by simply
changing the names of conflicting entities. It will always entail a
change to the subclass, however.

[...]
> >> Readability is more important to me over the long haul.
> 
> > Huh? Where have I said otherwise? I just happen to think that these two
> > goals are not mutually exclusive. And there should be a better solution
> > than the language ordaining a strange kind of Hungarian notation.
> 
> The C++ conventions I mentioned were a strange kind of Hungarian;

I am aware of that. That's why I suggested that the "self." notation is
not too much different. Because more often than not it _is_ entirely
clear from the context what you are doing. And in these cases the
required qualification is needless visual overhead, just like Hungarian
notation (in fact, it is probably even worse).

> "x.y" is
> just the way Python *always* spells "give me the value of attribute y in x's
> namespace", whether x is a module, class or instance object.  Consistent,
> explicit & obvious, given Python's modeling of modules, classes and
> instances *as* namespaces.

Then why do we have "from module import name"? :)

The point I am trying to make is that sometimes making a qualification is
redundant, if the context is already known. In this case the unambiguity
of the reference comes at the cost of cluttering the code with extraneous
tokens. This is exacerbated by the fact that "self" does not stand out
and that you easily get more tokens per line than your short-term memory
is comfortable with.

Come to think of it, why don't we require that _local_ variables be
qualified, too? For instance:

	for local.item in local.list:
	  local.sum = local.sum + local.item

It is, I think, the assumption that local variables are the default
case; that _most_ of the time you operate on local variables, and
you don't want to clutter your code. But in an OO language that
assumption no longer is true; you are going to operate on instance
variables just as often. The same reasoning that leads to local
variables not being qualified can be applied to methods operating
on instance variables.

For instance, I noticed that I much preferred the way Ruby denotes
attributes, even though it uses a superficially similar approach (Ruby
prefixes attributes with an "@" symbol rather than with "self."). So,
for instance, instead of

	self.last = self.lines[-1]
	self.lines = self.lines[:-1]

we have something like:

	@last = @lines[-1]
	@lines = @lines[:-1]

While I'm not that happy with this kind of notation, either, I find it
much easier to pick out what exactly is going on.

Ideally, of course, the code would look like:

	last = lines[-1]
	lines = lines[:-1]

and it should be made obvious from the context whether last/lines are
local or instance variables.

[Perl's prefix symbols.]
> Seriously, languages rarely start out doing things that are insane within
> their own worldview.  Perl would be *much* harder to work with in the
> absence of those $@%& thingies!

I am fully aware of this and I agree that they are a very necessary evil
in Perl. The point, however, remains that many people do regard Perl as
being less legible because of this.

[...]

> So the decorations solve a problem in Perl that Python doesn't have, so they
> make good sense in Perl but not in Python.

Pardon me, but I disagree here; these symbols provide annotations that
tell you which type a certain variable has. Since Python does not allow
you to declare that a variable is of a certain type, you have to rely
on conventions to ensure that there is no misunderstanding. In a mailing
list manager, is "list" really a list, or instead a string that contains
the name of the mailing list?

But apparently this isn't a big problem, which implies that it is
possible to deal with the absence of declarations in a way that doesn't
require you to qualify each and every use of an entity and still come
up with maintaineable code.

> Python's object.attr notation
> solves Python's problem:  with respect to which namespace is attr to be
> resolved?  Christian Tismer once suggested that Python take that another
> step, and require that e.g.
> 
>     def f(x, y):
>         global sum1, sum2, total
>         sum1 = x
>         sum2 = y
>         total = x + y
> 
> instead be written
> 
>     def f(x, y):
>         global.sum1 = x
>         global.sum2 = y
>         global.total = x + y
> 
> I think that's a fine idea too.  I'd even go a step toward Perl and make
> mapping objects use "{}" for indexing instead of "[]".  Stop me before I sin
> <wink>.

I would prefer to go into the opposite direction, actually. Using
something like global to denote attributes. For instance:

	attr last, lines

	last = lines[-1]
	lines = lines[:-1]

			Reimer Behrends



From wware-nospam at world.std.com  Tue Jun  8 18:28:43 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Tue, 8 Jun 1999 22:28:43 GMT
Subject: In what order does Python read dictionary keys?
References: <7jjnt5$snk$1@nnrp1.deja.com> <vqg73.1213$VL2.183595@news.direcpc.com>
Message-ID: <FD153v.26B@world.std.com>

Emile van Sebille (emile at fenx.com) wrote:
: ... is there a way to combine these
: two steps into a single statement, along the lines of:
: entrieskeys=entries.keys().sort()

That won't work as written, since sort() doesn't return anything. But
you can easily do it with a function.

def sortedDictionaryKeys(d):
  k = d.keys()
  k.sort()
  return k

print sortedDictionaryKeys({'r':1, 'g':2, 'b':3})
     ===> ['b', 'g', 'r']
-- 
 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Will Ware		email: wware[at]world[dot]std[dot]com
PGP fp (new key 07/15/97) 67683AE2 173FE781 A0D99636 0EAE6117



From pkoren at hex.net  Wed Jun 16 21:12:22 1999
From: pkoren at hex.net (Peter A. Koren)
Date: Wed, 16 Jun 1999 20:12:22 -0500
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <37682DD3.43923243@compaq.com>
Message-ID: <37684B76.AA8D0CD5@hex.net>

Greg Ewing wrote:

snip

>
> Tcl proponents will tell you that you can emulate all of
> that using tcl lists and arrays, which is true, but misses
> the point. Why bother with emulation when you can have the
> real thing at no extra cost?

snip

In the spirit of abductio ad absurdum, I can do everything you can do in
python or Tcl with a Turing machine.  I don't.  I am also not into whips
and chains.

Python is very nice, though I also want to take a look at a good OO scheme.

Regards,

Peter Koren

------------------------------------------------------------------
Modify the anti spam string in my email address to contact me.






From see-sig at wavedivision.com  Wed Jun  9 19:12:18 1999
From: see-sig at wavedivision.com (Rick Robino)
Date: 9 Jun 1999 23:12:18 GMT
Subject: "Programming Python" still worthwhile?
References: <7jki66$qam$1@news.gstis.net> <00c301beb260$09e94680$f29b12c2@pythonware.com>
Message-ID: <7jmsci$t5m$1@news.gstis.net>

Fredrik Lundh <fredrik at pythonware.com> wrote:
> Rick Robino <see-sig at wavedivision.com> wrote:
>> I've been watching this group for some time and just now getting warm to
>> Python. Looking at source examples has been great, but for completeness'
>> sake I'd like to get more of the "why" maybe found in a book.

> Programming Python was written to given you the "why";
> it's as much about the process of programming (and how
> python affects that), as about the python language itself.
> and most of it will work just fine with 1.6.  you'll find the
> necessary updates (up until 1.5.2) at:

> http://home.rmi.net/~lutz/

Sounds like I'll have to check this out no matter what, but I can
borrow it. I appreciate the confirmation of its currency.

> also note that Programming Python is *not* a tutorial
> (though you'll find one in an appendix), and it's *not*
> a reference manual.  many PP critics seems to have
> missed that.

> but you may find Learning Python a better choice,
> though.  see below.

A tutorial would be Ok, if (like the amazon description says) it's
a self guided one and doesn't stop at simple arrays at the end of
the book. All the posters to this thread seem to be quite happy 
with Learning Python, so I will certainly get it.

>> Since Python is at 1.5.2 and people are talking about 2.0
>> as if it's around the corner.

> it's not.  forget everything you've ever heard about it.
> it only exists as a few ideas in Guido's head, and he's
> about to embark on a "programming for the masses"
> project, so those ideas are likely to change...

> next stop is 1.6.

> probably early next year, or so. expect unicode, a
> more modular interpreter, string methods, and pro-
> bably lots of changes under the hood.  hardly any
> major changes on the surface.

> if I'd were you, I'd stop worrying about the future,
> and learn to love what Python can offer you today!

I will. I just don't know how many times I've bought books
about a week before O'Reilly makes a press release about
the next edition (sendmail, esa, et al.) And then again,
I have books for some things that are too new to be much
use outside of experimentation because the product hasn't
penetrated yet (eg., ksh93). This time I thought I would
break down and ask ;)

> (btw, there are LOTS of books coming out now, so
> changing Python in a way that renders them useless
> would be very, very stupid...  and if Guido suddenly
> loses his mind, I can guarantee that others will step
> in and maintain the 1.X series for another 30 years
> or so...)

>> I was wondering if anyone here could give me some advice as to whether or
>> not "Programming Python" is useful, if the current edition is still
>> current, if I should wait for the "Learning Python", or if I should use
>> some other resources.

> Learning Python is available, since long.  if you're a new-
> comer to python, it's by far your best choice.

> the five-star rating at amazon isn't there by accident...

> btw, If you're going to buy it via amazon, consider
> supporting the Python Software Association:

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

"Your order has been submitted. Thank you for shopping at Amazon.com."

Done via the PSA link, filled out the survey too - cost was $23.99 plus
shipping ($6.00 + $1.95). Not too bad.

> (and this concludes my weekly broadcast to c.l.py ;-)

Thanks to everyone who replied. One of the reasons python appeals to
me are the people in this group. Even when they disagree its _so_
civil! And always intelligent.

Cheers all,

--Rick 
                                ^^^^^^^^^^^^^^
Rick Robino                                  rrobino [at] wavedivision.com



From thstr at serop.abb.se  Wed Jun 16 09:29:01 1999
From: thstr at serop.abb.se (Thomas S. Strinnhed)
Date: Wed, 16 Jun 1999 15:29:01 +0200
Subject: Python Webserver
Message-ID: <3767A69D.50DC7FBF@serop.abb.se>

Hi

I don't know if I've dreamt, but I think I saw something about
a webserver written in Python. (It's not the embedded-in-Apache-thing
I'm thinking of) 
Now I just can't seem to find it anywhere, so does someone know anything
about this. Hipshots are welcome. I thought it'd be something fun to
play with.

Thanks
 -- Thomas S. Strinnhed, thstr at serop.abb.se



From faassen at pop.vet.uu.nl  Mon Jun  7 10:38:53 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Mon, 07 Jun 1999 16:38:53 +0200
Subject: "The Python Way"
References: <001101beae49$39af7bc0$949e2299@tim>
Message-ID: <375BD97D.B86DE72A@pop.vet.uu.nl>

Tim Peters wrote:

> [Patrick Phalen]
> > (Have you noticed that, during such periods, both Guido and Tim
> > seem to become singularly quiet?)
 
> It's just old.  python.org has archives of the pre-c.l.py days, and you'll
> find many of the old-timers arguing vigorously about all the "current
> issues" 8 years ago already.  There have been few notable advances in the
> art of GC-- let alone whitespace --since then, and repetition is tiring.

What? Then what do you call the invention of alien whitespace eating
nanoviruses, and hardware whitespace processors (though still buggy)? :)

> That doesn't mean people to whom it's all fresh shouldn't have fun clawing
> each others' eyes out, though.  OTOH, doesn't mean they should, either.

But we want to claw the old-timers' eyes out!
 
> > Not that Python shouldn't be open to critique,
> 
> It's far too late to complain about 1.5.2 <0.9 wink>.
> 
> > but the more I use and learn about the language, the more I find myself
> > appreciating the nice balance and heft Guido gave to it. Yet there doesn't
> > seem to be a single document that sums up that "aesthetic," but rather it
> > tends to appear piecemeal, over time, mostly in the Wisdom of Chairman
> > Tim.

> Eh?  I make shit up as I go along -- & Guido often appears to think that my
> fabrications would be good positions for him to have pretended he had all
> along too.   OTOH, after all these years, I *still* have to lecture him
> about his true vision of what Python should be ... although he learns pretty
> fast, for a Dutch guy <wink>.

Watch it, pal!

[snip]
 
[proposed elements of style document for Python]
> > Tim? Guido?
 
> Clearly a job for Guido alone -- although I doubt it's one he'll take on
> (fwiw, I wish he would too!).  Here's the outline he would start from,
> though <wink>:

[snip outline list]
>     There should be one-- and preferably only one --obvious way to do it.
>     Although that way may not be obvious at first unless you're Dutch.

At least I have one advantage here. :) (if anyone wants to know the
obvious way to do it in Python, just contact me, Genuine Dutch (tm), low
consulting fees, etc)

Regards,

Martijn



From rfreedm1 at nospam.rochester.rr.com  Thu Jun 24 16:41:18 1999
From: rfreedm1 at nospam.rochester.rr.com (Benjamin Schollnick)
Date: Thu, 24 Jun 1999 20:41:18 GMT
Subject: MSQL.py?
Message-ID: <rNHvjEdhm5Pp-pn2-biDGBOSsigWN@d185d18f8.rochester.rr.com>

Folks,

	I'm looking to interface with a SQL server (MS SQL, actually), and 
found
some notes on this module (MSQL.py).

	But it appears to not be supported anymore?  Anyone know where I can
find the "latest" versions, and if anyone else has taken up the mantle
of supporting it?

		- Benjamin



From kjwolf at online.de  Fri Jun 18 22:16:11 1999
From: kjwolf at online.de (Klaus-Juergen Wolf)
Date: Sat, 19 Jun 1999 04:16:11 +0200
Subject: Curses in Colors (binary)
Message-ID: <bhuek7.0c51.ln@solling.seismic.de>

Dear people,

I have just tried to improve cursesmodule with colors. I need that
feature for an urgent project.

It is known to me that there are several far more extended versions of
cursesmodule by Oliver Andrich, but still, it has been beyond my
capabilities to compile them under FreeBSD 3.2.

I have asked Guido about integrating my changes into Python, and he
told me ask here again, since he couldn't do the testing.  I have
developed a small testing program which will show a lot of colors if
your terminal allows it.  It is included together with the patches for
cursesmodule as of Python 1.5.2.

If your Python only dumps the core, please remove the "-ltermcap" from
the "readline" line and link, if appliable, libmytinfo or such alike
_before_ libtermcap - see the "curses" line in Modules/Setup.  This is
not my fault, this bug was in there before I came around.

Please excuse if I annoy you with this binary posting.

Best wishes & thanx in advance

cu
 k.j.



 
begin 644 colorcurses.tar.gz
M'XL(`._X:C<``^T::W/:2#)?K5_18;=<DI$QXNE'^38$XXTK!+L`;RY55Z42
M,(!L(5&2".&<_/?KGM$3,"$;[^Y]4)<-,U)W3T\_1RV&"]=CWLP9+2Q6&)GC
M\:N7!ZU8K%4J\`H`M.`;2L$WCK1:60.HU<NE:KU2+I41JURJU5]!\2^090,6
MGF^X`*\>'Y:.M6/WB/:7*.<?AN/C8Q at F?6!XH)V=G9T4:R=:';3Z>:5Z7CD]
MT`J:E,_GGT4]@Z)V7JV=E\K2FS=PK*D5R&MJ#=Z\D>#DZ,^#E/_U9G2^MJSZ
M&;1"J8`RE6%- at B+^@3`EM+[,X5<I+T'3F:]<<S+U";L"@Q6T#7O(H&59ICV9
M. at N/J8AE^%,V<@T+FJ:_4G%NF6/'M4T#NFR^&%CF4(5[V_39"'J^X3,/G#$T
M9LPUAT9!`K[Q2ADWG:]45>V4[[WGN[C$[>"!#7WA[I^9ZYF.#2EH@,<1P65S
MEWG,]FF"\M#.79Q*L">$[%$T?VIZ$.A, at H^F/7*624E,W(LW=&4E2=\G(KIC
MXO;_BWLD(1"+,1M0&X"ZDO(=QV;I92F(?'WH6$Z"WP8O8D#\D`D**.7WW!-G
MZTGYOKM@)]>&Y;%P&U/#$XMZB5TT#1N68A%Q[[=MI$/#UH=3PYZPA-@!J;@N
M]!]RN+']I/((^;9]V]7O&C==>6Z8KA"@R_R%:WM at P,SP'L$R'QDT]$*AL(T#
MT>J=^P]O6UW9\'W!XL9.V#!>8YO62;<ZK<T%4%&];.(Z"WND#HSAHQ at J>ZNY
MQWP4FSC1RF+CA6>7%4KCGZJK3M3!_ at LUA7H-L08N(0L&"11^!]>P?71^L8I"
MNG5-]CE%.4;"29*2Y(\(0[NL4W(L#(K-O3%[M#3M=$Q`TW(P_0#&CYV,!\,>
M8<!R>^\=GQB4]@3=@P?6"IW-93!@9#GT2)Z/]F85Q:](N35-K6/2K5>#M.NO
MYFS$QI19%NAS3Q(<W*V$`^KO6HTKG'^\Z5S=?H0CW/&%=!Q-YP:EG`L)OL'=
MJLE3;S)[X(TMW+D,IV<D0ZE84D7ZP[3 at FT.*)Q>.AH:/`\M"CV^WX1)R(JO#
M>&$/??)WH4M,KX20X\N<'$&?698'RRE#A;F)I(6QCYI#,Y#>D,AWHESCL:!B
M`!R=2,>!%*;M)S!&*,%UH]UK74CYW0BJE#\@$*ZSE0-*^@MJPT1O:G1_UYNW
M]YV^_&\%_H-*EF4:75[R72GP&Q3A'.2[57\QMYC>G++A(R'\!N&E'N9*NG(.
MFA+8]NR4RDI)*ZEG7*]2&$-M9W*.T_P?0<K0"N5SB,JW_-XR%AY\Q&JHT"8:
MHQ%*+2+`6\SGCNL7J$#&U"6DKIX42R=X#)-[/H5,$\N at KPBOO#:_(`,/+U.=
MG#L49*@,9.FRX]%B-J>R-5A,,&]P_"Y;.NXCDACN!&,.5\'[:$KT]L^&:QH#
MB_%;J$X?@\+E67M4X)LNE8K<F<IU_!*;_HF#1``GR.8XY96AAX<Z0+]$->1B
MK]B!5 at Z=%'=[!-R4/#\(1T:MC!B,76=&:6?-U^/<(9*0V'*MQ.U</U,U8>AO
MQ#_IH%(^E*1)9KR)O1'S0/X)K6R.Y1`E=9=[@#D&>=.-+Z'?O6]QA(-`*.V"
M9 at P+)CSQZW>KENOJ6"/$D29:`Z]B!8#<#`_2/!37S@)CT_7\G'*19%[DLV]2
M'O_(_S#*CY,`O6:WU>H`%C`\"*':CM,`W([PBSFF%-2Y[>N?6GVAP;,R.0T^
M1019D.]XZ43QAU="*6B.]CM8.L?_P at R()@WR(%T0.1"OR6%65$+\@%P.LRG>
M6CH7PE8D0[G.LV"E>!K(0)!>$P at 9+_JV+E:6MR5:9)R^K'?84D8"U.9Q1)V0
MU6/6F.^%+[%%SH`D(6RM6L=<G:^=5H,HX[!=&F)_$6&$5S%^/;SZ1")1!K4\
M%5S\]_%_0-\6_J.'#/![X'+),9]?(@Y^^/0QX".+/ER:TFA`(^XG at N=ED;C2
MI\\_!V)L\4^77^'C`8ZYC&3WUW>KACO1[S#Q,)Y;91)6S7TU`[A`85%6%!4E
M14%13A1SX.;4YZOP(6WPD'9X2%L\'/"111\N36DT<!5EB]UY+B^5N'MHI6HM
M]-&[U96)51F#Z\9GLR#`/C at CNJJ*)P/U;H7G1_T:LTG;P;NRA9\*!O-4413:
M,++!,'I'J97RJ.?,&$Q<8S[%O.&M9@,'E;YR%G3NY4=C*HSAEG!A6 at H7D'.-
M9D^_;S=ONYU6-Z?*R2E?Z."GD-MIY/9NSMTTYVZ,+#19(PVB)L]*@?,^K5F^
MX^"GQZVNP->O\/J9S+AIJOS!EFJ_O?9'E%%FID+0ZG;E\#")Z5$,*0]&H:>5
MBW7N">5R)0H^V^()[Q+LH</#1'BSMS3Q\`1R?+K`/2F*V/#00'N6SH4]U_V>
M$-6<;)K*Q8!-,'?H*Y4/]"\Y]5",5N'@2UPE=O'X'HO`K=9S'F!T,N/Q(I*Y
M<BX443G3N"&KVJE*'2*ABDBK(B"X[P>Q(3_:\^21*I?#TQ)>"]4;%LPX2\5U
M4^]1?>+54Z:,II(F:=<);!"I;NTB(EX$-19SDLN3TS9?V^EJ^;7ZL]71J!S3
M+?<R54S%BK)+6[]]#X>'Z0?@1!WG-3L93_&#9`_#*3%31&G>P(W0`HQO),XS
M?D[U/R$HK_:\M.\RPSM#'&&\R`K>WF;P0COL$>J;YZ1M1B!.Z\KD`:Z(\\_=
M2K_IX*GD&LNISAL)0F^12L3%4%,I=&H^<.P8F2[MHZ2F88N3?MI?_VE-;6F>
M_)_HBZ2YHX[(3Z@*@YOZ`XGX_E']K"=//'.8%\03\R7O26PCC'9+!9YGNWO;
M,R?X0"PJ_B*8`2_]B>[3^BVQPEY!2*KJ+&8#]K,*H_;52RN,>*+">&=LI\*V
MG(R2K3716]M/(23M"_F/"HF&',0=N9=5$U9E+,O/]_\"CU,/$W</$^W!G8I=
MR_-[]1NQ%D1X^U4"VOI/YS=4>J(9^?(Z1B4G^*-6Q>P0_R:HT!_7XV8#-=3<
M#Q11OJUFT.U\`>T1DC=U7#S>O+P:3:'!4'?/D,I;>K_TP(7_$WJTXJ4&M1@=
M<[[7F4CS4^(V(S+)I:I2*$3JK*._79C6Z`_#6C!91%L.'T550&$&>^?8ES%1
M6)6$A<8O;Y_OE2BTSD9_'6TR_E-V27'Z:;.05<8)DZQWAV##1/C('-FHPY9]
MYLZ$?4`8B#^:5,M%_FA2.ZNJ6E4\KC]AF/KNV%IX4UPT!MKA==#@4Q+9S7>O
M"?>;RFEGS#=29#MH/R!N0/;(5K8Q8VN4V\G>LU4'<;]1Y_HI=3)7OT,9/QL%
MQ/&9.+GR=N+H1!_0KI\2!8?MM.F#;L`@/N-\?_'HY!?0)@X`W]UU?`H*B.,"
MMH]]Q8DA2;JN[!VDR?VF,E5$OF._05H)R),!%2_^_)9CZHU at 06XV6_H8$GLY
M7!`^@:MZS-<W29]Q..8'E"+<:J(34"]7 at U=(VYMC>*#*Z7KPHE37<:7/O(6$
M>>&JQ9\C^)P_63Q#GNQP]E=SBJQDJQ0.UWJOA$/)95NCC0Z8YF!!OPD(>VS&
MB+^2HK<''@WF+F5P:ERG^UQZY[;[H='.J0<'X7BS&:;W^HW.U>U]']'B2=@'
MJ]9%CJI7:E&.2I-?W7P0"^!@"_>WM^TK<9]&6Q`:[7[S7:/;:PD!XBE72#Z%
M_;;=:+Y'98K at Y3.>S9,XW=95A('CC?N_4_L_PN"S#9Q/K7;[]F.$)*8;6&_;
M]ZV$,/>M#8P/C=];G7XC0 at KF&WC-3XU8))IL8'Q\=]./%^,S[H3<83K.$FN\
M#_3N;,7?`L-BSG\1\;[U28]>AWG"19ZDO^GW/SSEC-C,*<Q7?]4:6K%8>?[W
M7Z`5*[7H]U\:_19,*Y7*6O;[K[\#S!F]!0Y>3TJ2YX^\H7LIIH7X%P;B>F%H
M,8.FP?U4<U(2&2ZX)5J':G(F^HT2'CY3.'@02\]?7\(FV;ETX!KT8K]E?S9=
MQYYAX0H/DH+NZ*@$>#!>6",8,$CV.R6*,A-,&UPZ8<C%E%@]!7D3QL-.##HT
MF_RU[0/-#A(J$H_$YE&*)O^@@JG"@_(RJP<&P*HRG,H/JJG6JOER:5._FV(H
M\#749Y#?0_.-G,5\9/@LMN^$^<@^0 at C?7_Q=R2B###+(((,,,L@@@PPRR""#
I##+(((,,,L@@@PPRR""###+(((,,,L@@@PPRR""#O>%_XP\C6`!0```R
`
end




From phd at emerald.netskate.ru  Wed Jun  2 05:24:42 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Wed, 2 Jun 1999 13:24:42 +0400 (MSD)
Subject: Posix: Exec format error
In-Reply-To: <3754ED47.54C2F0AF@ilt.fhg.de>
Message-ID: <Pine.LNX.4.04.9906021322540.8860-100000@emerald.netskate.ru>

On Wed, 2 Jun 1999, Rolf Wester wrote:
>     os.execl("/home/wester/test/test","-d", "indata")

os.execl("/home/wester/test/test", "/home/wester/test/test", "-d", "indata")
         ^ ^ prog to run ^ ^       ^ pass this as argv[0] ^
                                       for that program

> Can aybody tell me how the format should look like?
> 
> Thanks in advance
> 
> Rolf

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




From cmyanko at bigfoot.com  Wed Jun 23 06:23:37 1999
From: cmyanko at bigfoot.com (Curtis Yanko)
Date: Wed, 23 Jun 1999 10:23:37 GMT
Subject: Freeze?
Message-ID: <7kqcj4$3bd$1@nnrp1.deja.com>

I have loaded 1.5.2 and the Win32All-125 but dont see the freeze tool
anywhere. Searching the FTP site I only see Freeze for older versions.
Hopefully I'm just confused but if somebody could give me a nudge I'd
appreciate it.

--
-Curtis Yanko


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From neelk at brick.cswv.com  Tue Jun 15 21:16:43 1999
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 15 Jun 1999 20:16:43 -0500
Subject: ANN: Stackless Python 0.2
References: <37628EAA.C682F16C@appliedbiometrics.com> <7k4fd4$c04$1@brick.cswv.com> <376636EA.84240129@appliedbiometrics.com>
Message-ID: <7k6ttr$i5g$1@brick.cswv.com>

In article <376636EA.84240129 at appliedbiometrics.com>,
Christian Tismer  <tismer at appliedbiometrics.com> wrote:
>
>
>Neel Krishnaswami wrote:
>
>[Stackless Python 0.2]
>
>> This is very neat, and you are completely deranged. I know just
>> enough to know that I should cheer you on, mind, but I'll try to
>> cheer loudly. Feel free to make an announce when you add coroutine
>> support. Please? :)
>
>Thanks :-)
>
>Sure I will. I'd like to know wether my binary works ok for you.
>You might also try a recursive function, it should raise an
>exception after 29999 recursions (just to have *a* limit).

Unfortunately, I couldn't get it to build (I'm on a Linux machine).
I tried copying the files from the stackless distribution into the
appropriate places in the Python source distribution, but I get 
the following error:

$ gcc -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H   -c myreadline.c -o myreadline.o
In file included from ../Include/Python.h:38,
                 from myreadline.c:42:
../Include/patchlevel.h:69: parse error before `5'
../Include/patchlevel.h:69: stray '\' in program
../Include/patchlevel.h:70: stray '\' in program
../Include/patchlevel.h:71: stray '\' in program

[plus more cascading errors as a result of this]

Apparently gcc is not liking this:

/* Version as a single 4-byte hex number, e.g. 0x010502B2 == 1.5.2b2.
   Use this for numeric comparisons, e.g. #if PY_VERSION_HEX >= ... */
#define PY_VERSION_HEX ((PY_MAJOR_VERSION << 24) | \   
			(PY_MINOR_VERSION << 16) | \   
			(PY_MICRO_VERSION <<  8) | \
			(PY_RELEASE_LEVEL <<  4) | \
			(PY_RELEASE_SERIAL << 0))

The '5' in the "parse error before '5'" is the PY_MICRO_VERSION; the
macro looks ok to me, but I haven't used C to any great extent in
literally years. Am I making some very naive mistake? 

>I'm just undecided on the design. There are a couple of
>different coroutine interfaces. You find them for instance in
>Scheme, in Modula II, some Oberons, Icon, and in Tim Peter's
>example which I added to my stackless archive.
>Do you have a proposal?

Not really; all I know is a little Scheme, and that part of my brain
is telling me that "coroutines are just closures over continuations." 
Fortunately, I'm still able to realize that that's probably the
wrong interface. :)


Neel



From MHammond at skippinet.com.au  Fri Jun 25 18:38:38 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Sat, 26 Jun 1999 08:38:38 +1000
Subject: How to use Perfmon Library
References: <7kvrrc$2d3$1@nnrp1.deja.com>
Message-ID: <7l10a5$3bt$1@m2.c2.telstra-mm.net.au>

There are 2 ways to use the performance monitor:

* If you just want to collect performance data from an existing process that
is "performance monitor aware" (or you just want to see data that NT
provides, such as memory usage, processor utilization etc.  This is quite
simple to use - you use the win32pdhutil module - there is a sample
"killProcName.py" that shows how to use these functions to query the local
machine - a remote machine is not much harder.

* If you want your application to provide custom perfmon data that another
process can read, then this is significantly harder - you really need to
read the NT documentation on how to do this; Python just wraps the standard
NT APIs.

Mark.

belbo_j at my-deja.com wrote in message <7kvrrc$2d3$1 at nnrp1.deja.com>...
>Hello,
>
>Is it anybody that can give me some information
>about the usage of Perfmon library ?
>
>I want to monitor a process on a remote NT server






From che at debian.org  Fri Jun 11 01:28:55 1999
From: che at debian.org (Ben Gertzfield)
Date: 10 Jun 1999 22:28:55 -0700
Subject: Created a Python module for ID3 MP3 tag manipulation
Message-ID: <yttaeu7z4o8.fsf@gilgamesh.cse.ucsc.edu>

Hey, Pythoners! Just a quick announcement that I've created a Python
module for easy manipulation of ID3 tags on MP3 files.  These tags
generally involve the title, artist, and other information about the
file. I've mailed webmaster at python.org and submitted it to
Freshmeat, so I hope it can be of use to you all.

The home-page for the ID3.py module is at:

http://csl.cse.ucsc.edu/~ben/python/

A simple example of using the module:

from ID3 import * 

try: 
  id3info = ID3('/some/file/moxy.mp3'); 
  print id3info 
  id3info.title = "Green Eggs and Ham" 
  id3info.artist = "Moxy Fruvous"

except InvalidTagError, message:
  print "Invalid MP3 tag:', message

More documentation is available in the source code to the module.

Enjoy!

Ben Gertzfield

-- 
Brought to you by the letters K and S and the number 2.
"I don't want the world.. I just want your half."
Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/



From bruises at my-deja.com  Mon Jun 14 23:20:07 1999
From: bruises at my-deja.com (bruises at my-deja.com)
Date: Tue, 15 Jun 1999 03:20:07 GMT
Subject: Installing ODBC on PythonWin
References: <3760D550.2071295E@lemburg.com>
Message-ID: <7k4gov$avi$1@nnrp1.deja.com>

Before I go wack and take it out on my computer
like this guy
(http://www.web42.com/badday/badday_small.mpg)
did, I'd like to ask for some help.

I'm trying to install an ODBC module into
PythonWin, but it still *won't* work!

It is written that the ODBC module is included in
the pythonwin distribution but >>>import ODBC
says otherwise. I've d/e'd (downloaded/executed)
Win32, but it died and said that it was looking
for version 1.4!

I was able to partially install mxODBC by copying
all the files into the right place but I'm having
trouble compiling the modules in vc++.

I'm going to pull my hair out!!! Could someone
show me an easier way of installing ODBC? (I
don't want to mess with DAO since I'm expecting
to migrate data from a database.)

Hhelp!!

SL


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From Stidolph  Tue Jun 29 16:05:22 1999
From: Stidolph (Stidolph, David)
Date: Tue, 29 Jun 1999 21:05:22 +0100
Subject: Please help me to understand classes
Message-ID: <00001592@bossar.com.pl>

From: "Stidolph, David" <stidolph at origin.ea.com>

Finally, a simple question I can answer!

First of all, class methods like "druk" need to have a parameter declared in
them called self (or whatever you want to call it).  Secondly you need to
call the method with open/close parens:

class A:
  def druk(self):
    print 'Jy is in Class A'

class B:
  def druk(self):
    print 'Jy is in Class B'

a = A()
c = B()
c.druk()
Jy is in Class B
a.druk()
Jy is in Class A

I hope this helps,

David.


-----Original Message-----
From: Johann Spies [mailto:jhspies at futurenet.co.za]
Sent: Tuesday, June 29, 1999 2:04 PM
To: Python-poslys
Subject: Please help me to understand classes


I have been playing around with python for some time now, but I am a
relative newbie as it is my hobby and normally I have very little time for
programming.  I am on leave now and want to write a program but I find
python a very difficult language to comprehend despite the fact that I
have a few years of experience in programming a language like Pascal.
Maybe it is because I have little experience in OOP.

Can somebody explain to me why the following program produces no output:
---------------------------------------------
class A:

    def druk():
        print 'Jy is in Class A'

class B:

    def druk():
        print 'Jy is in Class B'
        
a = A()
c = B()
c.druk
a.druk
---------------------------------------------


 --------------------------------------------------------------------------
| Johann Spies                                 Windsorlaan 19              |
| jhspies at futurenet.co.za                3201 Pietermaritzburg             |
| Tel/Faks Nr. +27 331-46-1310		       Suid-Afrika (South Africa)  |
 --------------------------------------------------------------------------

     "One thing have I desired of the LORD, that will I seek
      after; that I may dwell in the house of the LORD all
      the days of my life, to behold the beauty of the LORD,
      and to inquire in his temple."            
                               Psalms 27:4       



From Jeff.Raven at p98.f112.n480.z2.fidonet.org  Tue Jun 29 15:31:00 1999
From: Jeff.Raven at p98.f112.n480.z2.fidonet.org (Jeff Raven)
Date: Tue, 29 Jun 1999 20:31:00 +0100
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
Message-ID: <000015a2@bossar.com.pl>

From: Jeff Raven <jraven at psu.edu>

Scott Deerwester wrote:
> 
> That's probably true, but Qt IS cross-platform. Has anybody ever
> tried running pyKDE with the Windows version of Qt? I've used
> pyKDE a bit and it seems to win in regard to completeness and
> usability. One "problem" is that pyKDE (unlike Qt) seems to have
> tightly coupled KDE (which isn't cross-platform) with Qt (which
> is).
> 

Well, although Qt is cross-platform, it isn't cross platform
_free_software_. Only the X11/Unix version of Qt is freely
available for use in free software. To develop with Qt for
Windows you need a developer's license -- regardless of whether
your program is open source or not (the Windows version of the
library is not, to my knowledge, available for download). At
$1500+ a seat, that's going to cost you just a wee bit more
than using Tk or wxWindows as your cross-platform toolkit.

And God forbid your software not be open source -- then you'll
need to shell out for a license for both Unix and Windows...

Jeff Raven



From Mordy.Ovits at p98.f112.n480.z2.fidonet.org  Wed Jun 30 21:15:11 1999
From: Mordy.Ovits at p98.f112.n480.z2.fidonet.org (Mordy Ovits)
Date: Thu, 01 Jul 1999 02:15:11 +0100
Subject: sms tests
Message-ID: <00001618@bossar.com.pl>

From: Mordy Ovits <movits at lockstar.com>

It depend on how you plan on sending the SMS.  If you're doing it raw, it's
helpful to have a library for it already written.  If you're doing it via some
web gateway, it'll be a breeze in Python.
Also, if you have a good C library for SMS, you might consider SWIGing it.  See
www.swig.org
Mordy

marilena_kolia at my-deja.com wrote:
> 
> Hi,
> 
> I am trying to test the SMS performance and get some results about the
> time it takes a message to be transfered, but also the delays of each
> part of this procedure. I am planning to use Tcl programming. Does
> anybody know how to do it or have some suggestions about it?
> 
> Thank you
> 
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.

-- 
o Mordy Ovits
o Cryptographic Engineer
o LockStar Inc.
---------------------------------------------------------------------------
#!/usr/local/bin/python 
from sys import*;from string import*;a=argv;[s,p,q]=filter(lambda x:x[:1]!=
'-',a);d='-d'in a;e,n=atol(p,16),atol(q,16);l=(len(q)+1)/2;o,inb=l-d,l-1+d
while s:s=stdin.read(inb);s and map(stdout.write,map(lambda i,b=pow(reduce(
lambda x,y:(x<<8L)+y,map(ord,s)),e,n):chr(b>>8*i&255),range(o-1,-1,-1)))



From sjturner at ix.netcom.com  Mon Jun  7 13:15:36 1999
From: sjturner at ix.netcom.com (Stephen J. Turner)
Date: Mon, 07 Jun 1999 13:15:36 -0400
Subject: Question about envvars
References: <7jgl9n$p13$1@nnrp1.deja.com>
Message-ID: <375BFE38.3FB3E83C@ix.netcom.com>

Hi Rob,

> I have built a shared module that reads flat files for environment
> variable information, and makes subsequent putenv calls for those
> pertinent ones.  This functionality is encapsulated in a c++ class,
> FileRegistry, which I think I have working correctly, using shadow
> classes under python.  All I expose in the class is the ctor, the dtor,
> and the method called exportEnvVars():
> 
> %module MDIreg
> %{
> #include "fileregistry.h"
> %}
> 
> class FileRegistry
> {
> public:
>   FileRegistry(bool, bool);
>  ~FileRegistry();
>   void exportEnvVars();
> };
> 
> ...
> 
> However, I can't seem to get the envvars to persist into the current
> process space.  If I fork off a child, then I can see the envvars that I
> expect.

The problem is that, although putenv does indeed modify the current
process's environment, Python only reads the environment when the posix
module (imported by os) is first imported, so as to create the environ
dictionary.  (Try confirming this behavior by postponing "import os"
until after your call to MDIreg.exportEnvVars.  You should see your
environment changes reflected in os.environ.)

The proper solution is, I think, to avoid the direct use of the putenv
function, and instead to use the Python C API to modify the os.environ
mapping (which calls putenv as a side effect).  Here is a crude C module
to do just such a thing:

-------------------- START modenv.c --------------------
#include <Python.h>

#include <stdio.h>
#include <string.h>

static PyObject *modenv_load(PyObject *self, PyObject *args)
{
  FILE *fp;
  PyObject *os, *environ, *v;
  char buf[256];
  char *file;
  char *value;

  if (!PyArg_ParseTuple(args, "s", &file))
    return NULL;

  os = PyImport_ImportModule("os");
  environ = PyObject_GetAttrString(os, "environ");

  if (fp = fopen(file, "r")) {
    while (fgets(buf, sizeof(buf), fp), !feof(fp)) {
      /* skip lines that do not contain equals sign */
      if (!(value = strchr(buf, '=')))
	continue;

      /* remove trailing newline */
      buf[strlen(buf)-1] = '\0';

      *value++ = '\0';
      v = PyString_FromString(value);
      PyMapping_SetItemString(environ, buf, v);
      Py_DECREF(v);
    }  
    fclose(fp);
  }

  Py_DECREF(environ);
  Py_DECREF(os);

  Py_INCREF(Py_None);
  return Py_None;
}

static struct PyMethodDef modenv_methods[] = {
  {"load", (PyCFunction)modenv_load, METH_VARARGS},
  {NULL, (PyCFunction)NULL, 0, NULL}
};

void initmodenv()
{
  if (!Py_InitModule4("modenv", modenv_methods, NULL, NULL,
		      PYTHON_API_VERSION))
    Py_FatalError("can't initialize module modenv");
}
-------------------- END modenv.c --------------------

Hope it helps.

Regards,
Stephen

--
Stephen J. Turner <sjturner at ix.netcom.com>



From garry at sage.att.com  Wed Jun  9 13:58:10 1999
From: garry at sage.att.com (Garrett G. Hodgson)
Date: Wed, 9 Jun 1999 17:58:10 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7iqhkb$d7s@wiscnews.wiscnet.net> <7iu7ma$d33$1@cronkite.cc.uga.edu> <02hfonzsx2.fsf@rtp.ericsson.se>
Message-ID: <375EAB32.FA8B4F45@sage.att.com>

Kumar Balachandran wrote:

> One of the irritating things about
> Python is the use of whitespace in syntax.

no, that is one of the nicest things about python.

> why not have optional syntactic sugar such as
> 
> if
> ...
> fi
...
> etc. It makes the language more elegant.

no, it clutters it up with syntactic noise.

> The change is simple to achieve if backward compatibility is given to
> accomodate people that grew up with FORTRAN.

last i heard, fortran is still around, and people are welcome to use it.

> In conclusion, 
> lets face the fact that that design decision was flawed, and people
> have to conjure up excuses to save face. 

no, it was a fine decision.  people defend it because they like it,
not to save face.  if you don't like it, don't use it.  there is no
shortage of languages that insist on overstating the obvious.

-- 
Garry Hodgson			comes a time
garry at sage.att.com		when the blind man
Software Innovation Services	takes your hand, says,
AT&T Labs			"don't you see?"



From icwpcs at aol.com  Tue Jun 29 18:57:40 1999
From: icwpcs at aol.com (icwpcs at aol.com)
Date: Tue, 29 Jun 1999 15:57:40 PDT
Subject: Save 40-60% on Cellular & Digital Phone Accessories
Message-ID: <7lbbv7$96c$10208@mercury.hiline.net>

  Need a new battery, car charger or leather protective carrying case for your

cellular or digital phone? We offer package prices on acessories for the most

popular phones of today as well as older models at 40-60% savings over retail

stores. Don't pay $30 for a car charger, $70 for a battery or $20 for a leather 

case. Our package prices for battery, car charger and leather case start at $29.99. 

EXAMPLE OF COMPLETE STARTER KIT W/BATTERY, CAR CHARGER AND CASE:
        
       Qualcomm 1920/2700 , Sale Price $29.99, Regularly $79.99

       Nokai 5000/6000 series, Sale Price $54.99, Regularly $99.99 with vib battery

       Motorola 550/660 Flip phone, Sale Price $29.99, Regularly $89.99

 
Just need one item, not the entire package? Most leather case are just $9.99 and

 vehicle chargers are only $14.99 on most models. Battery's start at $24.99. 

Hard to find item? We can special order your accessories for older models at 

no additional cost.

           PARTIAL LIST OF PHONE WE HAVE ACCESSORIES FOR:       

Sprint QUALCOMM 1920/2700

SprintSAMSUNG 1000/1900/2000

SONY 1201,1301,CMZ-100,CM-777

MOTOROLA MICRO TAC 550-650 ,TELETAC, STAR TAC, MICRO TAC ELITE, BAG PHONES, FLIP STYLE

NOKAI 5000-6000 SERIES, 100, 232-252 , 1000 , 638-636, 918, 2160-2170-2190

NEC 700/800

ERICCISON  - MOST MODELS

AUDIOVOX -MOST MODELS

SANYO 3000

 CREDIT CARD ORDERS ONLY, NO COD'S, MUST HAVE EXACT MODEL NUMBER 


                                               1-800-686-8988

                            Monday-Friday 9am-6pm, Saturday 12n-5pm
 

Shipping and handling charge of $7.99 for starter kits and $3.99 for individual items. 

Most items shipped  in 5 business days. This offer is brought to you by ICW, a large 

wireless dealer in the Southwestern US. If you no longer wish to receive e-mail please

 notify us as we make every effort  to clean our marketing list on a regular basis. 











From thstr at serop.abb.se  Wed Jun  9 09:55:28 1999
From: thstr at serop.abb.se (Thomas S. Strinnhed)
Date: Wed, 09 Jun 1999 15:55:28 +0200
Subject: Python classes in C
References: <7jlqtp$o8e$1@rtpnews.raleigh.ibm.com>
Message-ID: <375E7250.CDC35CED@serop.abb.se>

Hi

I haven't used them myself, but the FAQ could maybe solve some problems
(it've solved some for me)
 
http://www.python.org/doc/FAQ.html#5.9
http://www.python.org/doc/FAQ.html#5.5

Good Luck
 -- Thomas S. Strinnhed
Arinte wrote:
> 
> How can I get the member variable's values from a python class from inside
> my c .dlls???
> I have seen PyObject_GetAttr C function, but I am unsure on how to use it.
> Could someone post a small example showing both c and python side or point
> me to one.  I didn't see on in the docs.



From tismer at appliedbiometrics.com  Wed Jun  2 06:13:05 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Wed, 2 Jun 1999 10:13:05 GMT
Subject: Patch: Win32 stack overflow
Message-ID: <375503B1.9893E25B@appliedbiometrics.com>

Patch for python15.dll

stack152.py

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home
-------------- next part --------------
######################################
# patching Python 1.5.2 dll file
# to adjust maximum recusion depth.
# CT 990602

"""
What is this patch for?

Python 1.5.2 calls eval_code recursively for every
call of a python function. The current maximum
recursion depth is compiled into python15.dll
to be 10000.

My measures have shown that each recursive call
costs 104 bytes, when Python is compiled with MS VC++ 5.0.
The default stack size for a Windows program is 1 MB, and
this is slightly too few for 10000 calls.
The secure limit seems to be below 9610.

Note that this is no problem with MS VC++ 6.0, since
due to better register spilling, the stack growth per
recursion is only 100 byte :-)

Note also that for certain applications with different
stack requirements/compilation options, this patch might
be used to set the maximum recusion depth to any
arbitrary limit.

Test program:

def f():return f()

f()

After applying the patch, this sequence should give
a stack overflow error, but no general protection fault
any longer.

good luck - chris

"""

import string, struct, win32api

ORIG_OFFSET = 49785  # offset of signature in 1.5.2 orig dll
SIGNATURE = "\x89\x57\x0c\x3d"

DEF_FNAME = win32api.GetModuleFileName(win32api.GetModuleHandle("python15.dll"))

LIMIT = 9610 # secure under VC++ 5.0

def _get_stacksize(fname):
    """determine stack size of a Python 1.5.2 dll file"""
    f = open(fname, "rb")
    bin = f.read()
    if not string.find(bin, SIGNATURE) or len(string.split(bin, SIGNATURE)) != 2 :
        raise SystemError, "this is no python 1.5.2 dll file"
    else:
        offset = string.find(bin, SIGNATURE)
    off2 = offset+4
    word = bin[off2:off2+4]
    return offset, struct.unpack("i", word)[0]
    
def stacksize(fname):
    offset, stack = _get_stacksize(fname)
    if offset != ORIG_OFFSET:
        print "this is no standard Python 1.5.2 dll, but it seems to be valid"
    return stack
    
def patch(fnold, fnnew, size):
    bin = open(fnold, "rb").read()
    left, right = string.split(bin, SIGNATURE)
    right = struct.pack("i", size) + right[4:]
    bin = left + SIGNATURE + right
    open(fnnew, "wb").write(bin)

if __name__ == "__main__":
    ssize = stacksize(DEF_FNAME)
    print "Your current python recursion depth is %d" % ssize
    newsize = raw_input("stacksize = %d. Enter new size (to be sure, below %d)" \
       % (ssize, LIMIT))
    try:
        newsize = string.atoi(string.strip(newsize))
        old = DEF_FNAME
        new = old+".NEW"
        patch(old, new, newsize)
        print "File %s created with recusion depth of %d." % (new, newsize)
        print "please quit python, delete %s\n and rename %s to %s" % \
            (old, new, old)
    except ValueError:
        print "Recursion depth not changed."
        
# simply run this program.
# (c) Christian Tismer
# Professional Net Service GmbH

From tim_one at email.msn.com  Thu Jun 10 02:48:02 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 10 Jun 1999 02:48:02 -0400
Subject: "Programming Python" still worthwhile?
In-Reply-To: <375E27A7.4871EA36@pop.vet.uu.nl>
Message-ID: <001901beb30d$3138b7e0$af9e2299@tim>

[Martijn Faassen]
> ...
> For some reason Pythoneers in this newsgroup (me included) like to talk
> about improving the language, though it's a very nice language already.
> Although some see this as a bad sign, I think the fairly intelligent
> discussion that *usually* is the result is quite interesting. Some
> discussions excepted. :)

It's not because we loathe Python?  Hey -- you're right!

> Thanks to some newsgroup luminaries like Tim Peters (Tim, you're a
> newsgroup luminary now!),

It's better than "resident asshole", I guess.

> who seems to have programmed in just about any language in existence

There are at least 107 variants of BASIC I missed, and I only used REXX on
IBM mainframes <wink>.

> and can compare them all fairly to Python,

Oh, sure.  That's easy!  Say 5 nice things about each, then fairly conclude
that Python wins overall.  It's like filling in a tax form, except there's
no penalty for cheating.  "Perl is highly expressive, succinct, offers
exceptionally complete access to Unixisms, has the best-integrated regular
expressions in the world, and is very responsive to the ever-changing needs
of a fiercely loyal programming community.  Python blah blah blah blah blah.
Overall, then, Python is the clear winner."  Couldn't be any fairer than
that <wink>.

> the level of discussion remains fairly mature at most times. Though the
> latest Python 2 thread became a bit too emotional for my tastes in
> places.

Did someone actually write "if brains were dynamite, you couldn't blow your
nose", or did I just think it?  Now let's see who thinks I thought it about
them <0.5 wink>.

> The *true* Python 2 things are the rare statements made by Guido on it
> (and rumors of his statements :).

Overall, Python2 is a real improvement over Python.  That statement is
inoperative, though, if it turns out Python2 sucks.

for-all-we-know-now-it-will-turn-gc-on-its-head-and-throw-away-the-
    live-objects-ly y'rs  - tim





From hniksic at srce.hr  Thu Jun  3 13:32:07 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 03 Jun 1999 19:32:07 +0200
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net> <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net> <87hfosquc0.fsf@ev.netlab.co.jp> <3753F121.E4A9BE8B@prescod.net> <87yai3ntdt.fsf@ev.netlab.co.jp> <7j3iif$134e$1@nntp6.u.washington.edu> <7j45iv$908$4@cronkite.cc.uga.edu> <7j4hhj$118u$1@nntp6.u.washington.edu> <14166.46756.866947.64007@anthem.cnri.reston.va.us>
Message-ID: <87g14942co.fsf@pc-hrvoje.srce.hr>

"Barry A. Warsaw" <bwarsaw at cnri.reston.va.us> writes:

> JPython used to call __del__() when the Java object (PyInstance) was
> finalized.  See my previous message on the subject for why this has
> been commented out.  I don't believe __del__() is ever invoked now
> in JPython.

Isn't that a case of sacrificing (Python) correctness for (Java)
efficiency?  The code that relies on __del__() getting called
eventually will lose in JPython.



From doughellmann at mindspring.com  Fri Jun 11 17:44:47 1999
From: doughellmann at mindspring.com (Doug Hellmann)
Date: Fri, 11 Jun 1999 21:44:47 GMT
Subject: Problem with Lists & Index command
References: <rNHvjEdhm5Pp-pn2-zOTjsVjHebb4@d185d18f8.rochester.rr.com>
Message-ID: <37618424.8B8F3116@mindspring.com>

> import string
> 
> entry   = 'ABCD'
> 
> tcmlist = [('EFGHI',1),('JKLMN',2),('ABCD',3)]
> found   = 0
> entry   = string.upper(string.strip(entry))
> 
> try:
>         found = tcmlist.index (entry)
>         tcmlist[found][1] = tcmlist[found][1] + 1
>         print 'found'
> except ValueError:
>         tcmlist.append (entry,1)
>         print 'Sorry could not find it'
> else:
>         pass
> 
> print tcmlist
> 
> I don't see anything that looks wrong, and getting
> no syntax errors, but....
> 
> It appears that INDEX doesn't work on a multiple
> dimension list????

Using index() on tcmlist is only going to search through the values in
tcmlist.  It won't look inside the tuples that tcmlist contains.  So,
you either have to construct a tuple to pass to index, or use a
dictionary instead of a list:

tcmvalues = { 'EFGHI':1,
              'JKLMN':2,
              'ABCD':3,
            }
entry = 'ABCD'
try:
    found = tcmvalues[ entry ]
except KeyError:
    tcmvalues[ entry ] = 1
else:
    pass

This will be much faster than a list lookup anyway, so if you're trying
to count occurances of entries or something similar you should use the
dictionary.

Doug



From hniksic at srce.hr  Mon Jun  7 15:24:42 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 07 Jun 1999 21:24:42 +0200
Subject: fork()
References: <375BE64A.EB997F24@icrf.icnet.uk> <87so840zcc.fsf@pc-hrvoje.srce.hr> <375BF721.4B1864F2@icrf.icnet.uk> <m31zfn99td.fsf@greebo.nodomain.de>
Message-ID: <87hfojzudh.fsf@pc-hrvoje.srce.hr>

Bernhard Herzog <herzog at online.de> writes:

> I fear that in reality the memory is copied after all. If they use
> the dictionary in any way they are modifying refcounts of at least
> some of the objects in the dictionary. From the OS's point of view
> this counts as modifying the memory.

Eek.  You're right.  Refcounting truly sucks.  :-(



From jbkerr at sr.hp.com  Wed Jun 30 17:33:11 1999
From: jbkerr at sr.hp.com (Jim Kerr)
Date: Wed, 30 Jun 1999 14:33:11 -0700
Subject: problem with Fnorb and module new
Message-ID: <377A8D17.9B4D17C8@sr.hp.com>

  I just tried installing Fnorb on HP-UX 10.20... the install went
smoothly, but when I tried stepping through the Hello World example,
python complained about not being able to find module new. The
complaint
is certainly valid -- I can't find new.py anywhere either. Anyone know
what's going on here?

$ fnidl HelloWorld.idl
Traceback (innermost last):
  File
"/users/jbkerr/python/Python-1.5.1/site-packages/Fnorb/script/fnidl",
line 2, in ?
    import sys; from Fnorb.script import fnidl;
sys.exit(fnidl.main(sys.argv))
  File
"/users/jbkerr/python/Python-1.5.1/site-packages/Fnorb/script/fnidl.py",
line 41, in ?
    from Fnorb.orb      import CORBA, Util
  File
"/users/jbkerr/python/Python-1.5.1/site-packages/Fnorb/orb/CORBA.py",
line 38, in ?
    import fnorb_thread, TypeManager, Util
  File
"/users/jbkerr/python/Python-1.5.1/site-packages/Fnorb/orb/Util.py",
line 38, in ?
    import keyword, new, string, UserList
ImportError: No module named new

-- 
Jim Kerr
Hewlett-Packard
Santa Rosa Systems Division
1400 Fountaingrove Pkwy, MS 3USZ
Santa Rosa, CA 95403
Phone: 707-577-5575
Telnet: 577-5575



From gmcm at hypernet.com  Wed Jun  9 00:48:00 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 8 Jun 1999 23:48:00 -0500
Subject: makepy not war
In-Reply-To: <375DD2BE.D918CCCE@yahoo.com>
Message-ID: <1283218562-70098280@hypernet.com>

Flbill Blipf writes:
 
> After running makepy.py on a type library, how do I use
> the resulting file?
> 
> I know how to use win32com.client.Dispatch, and I kinda
> sort of know how to use CoCreateInstance, QueryInterface,
> etc. with C++.

You use win32com.client.Dispatch. If he finds you've been bad and 
haven't run makepy, he'll use win32com.client.dynamic.Dispatch. 
Otherwise he'll use the makepy generated stuff. Either way, you 
should never have to CoCreateInstance again.

unless-you-really-want-to-ly y'rs

- Gordon



From landrum at foreman.ac.rwth-aachen.de  Wed Jun  2 02:46:27 1999
From: landrum at foreman.ac.rwth-aachen.de (Gregory A. Landrum)
Date: 02 Jun 1999 08:46:27 +0200
Subject: A newbie learns... [Part 1]
Message-ID: <syai32j7g.fsf@foreman.ac.rwth-aachen.de>

Howdy,

I'm in the process of learning python, Tkinter and all that good
stuff. I'm doing this by writing an application which, hopefully, will
actually be useful.  Along the way, I'm experimenting and playing and
trying to figure out how to make things go faster/better.

What I figure is that as I find tricks which I think might be useful
to other people trying to learn python, I'll post them.  Mabye other
newbies will also find them useful.  

If people find this a waste of time/space/electrons, let me know and I
won't continue.

---------

I (Tkinter) Drawing lines/polygons made up of many points.

The easy way to do this is to create a 1D array of x,y pairs (i.e
something like this: [x1, y1, x2, y2, ... xn, yn]) and then call
Canvas.create_lines:

  canvas.create_line(data,width=width,fill=color,tags="foo");

unforunately, that's also pretty slow.  Tkinter takes your nice array
of x,y pairs and, using Tkinter._flatten, turns it into the tuple
which actually gets passed to Tk.  This is wonderfully general, but
pretty slow.  Luckily, since we know the structure of the data, we can
bypass this by forming the tuple ourselves and making the Tk call
directly: 

  tpl = tuple(data);
  apply(canvas.tk.call,
        (canvas._w,'create','line') + tpl +
        ('-width',width,'-fill',color,'-tags',"foo"));
  
This results in a huge speed up when the line contains a lot of
points.  For example, on my machine when the number of points in the
line is 4000, it takes ~10 seconds to flatten the list of points prior
to drawing (first method).  The second method takes less than half a
second.  The relative difference increases dramatically with
increasing numbers of points.

The same exact trick can be used with polygons:

  tpl = tuple(data);
  apply(canvas.tk.call,
        (canvas._w,'create','polygon') + tpl +
        ('-width',width,'-fill',color,'-tags',"foo"));

is way faster than:

  canvas.create_polygon(data,width=width,fill=color,tags="foo");

---------

II Building lists

This isn't related to Tk at all, but it is something that makes a huge
difference in speed.  I naively assumed that a good way to grow a list
was to just do something like:

    fooarray = fooarray + [data];

This is, once again, completely wrong.  This is the *slow* way of
doing things.  It's much better to do:

    fooarray.append(data);

For example, this function:

def test1(num):
  fooarr = [];
  for i in xrange(num):
    fooarr = fooarr + [i];

takes almost 14 seconds on my machine when num=10000, whereas this
one:

def test2(num):
  fooarr = [];
  for i in xrange(num):
    fooarr.append(i);

takes less than a tenth of a second.  Now *that* is a speedup.

---------

-greg

---------------------
Dr. Greg Landrum  (landrumSPAM at foreman.ac.rwth-aachen.de)
Institute of Inorganic Chemistry
Aachen University of Technology




From tim_one at email.msn.com  Thu Jun 17 01:58:19 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 17 Jun 1999 01:58:19 -0400
Subject: fork()
In-Reply-To: <9XQ93.2$V4.8796@news-dal.corridex.com>
Message-ID: <000201beb886$66348900$099e2299@tim>

[Tim, rambling]
> So my plan for the hour <wink> is:  if M&S finds a cycle, it
> will clean it all up iff no involved object has a __del__.  If
> an object in a trash cycle does have a __del__, tough, the cycle
> won't be reclaimed.  Instead we'll supply a function you can call
> if you care, that will return those objects (one at a time?  in
> a list?  whatever), and you do with them what you will; if you don't
> clean them up, you'll keep getting them back every time you call that
> function until you resurrect them or break the cycle by hand in
> the way that makes most sense to your algorithm.

[Evan Simpson, scrambling <wink>.]
> Presuming the hour isn't up yet, I've a handful of scattered comments:
>
> . How about a callback instead of/in addition to a probe function?

Dunno.  Not keen on burdening it with features before I think I know whether
it makes sense at *all* <wink>.

> That way you could flip a switch and get called even on cycles which
> are "del-clean", if you wanted to check for unintentional cycle creation.

A worthy goal indeed.

> . "del-cleanness" only requires that actual cycle participants lack
> __del__, not dangly bits, right?

"refcount rules" seem to work fine in the absence of cycles (at least nobody
has complained about them yet!), and once a cycle is purged the dangly bits
can follow those rules without further intervention.  So, yes, your reading
seems to make good sense.  For this hour, anyway <wink>.

> Or does this re-introduce analysis that you intended to throw out?

I'm trying to build on what Guido has in mind -- tracking only the dicts in
cycles is very attractive for time and space reasons, and there's sure no
analysis I could even dream of throwing out from *that* bare bones approach
<wink>.  But Guido is resorting to "well, let's ignore __del__ entirely
then" because it doesn't collect enough info to do anything more ambitious
than that.

What it leaves behind is a list of dicts definitely alive, and another list
of dicts definitely unreachable *and* definitely involved in a cycle.  What
I want to explore is whether we can build on that, by now making a deeper
analysis of *just* the stuff in the latter "doomed" list.  By running a more
expensive M&S with the dead dicts as "the roots", the structure of the trash
can be deduced and the objects containing those dicts identified (if an
object contains a dead dict, the object must itself be trash -- else the
dict could not be dead).

That would take additional time roughly proportional to the amount of cyclic
trash, and influenced by the complexity of its structure.  Which should be
"not much" and "not very" most often.  Then I wave my hands a lot and
everyone is happy <snort>.

> . (somewhat rhetorical) How the heck do you recognize what you've got when
> the machinery hands you an object which isn't otherwise reachable?

If you intend to use __del__ in objects involved in cycles, the obvious
answer is that it's your problem to make each object store enough info about
itself in itself so that your code knows what to do with it when it gets it
back; i.e., it's akin to asking what the heck you're supposed to do with a
floating-point number <wink>.

> I realize that there's no general answer to this, and that in some
> specific cases it would be enough to check __bases__, but the only answer
> which occurs to me off the bat is to uniquely mark objects which you
> expect to have to handle, or provide them with special __cleanup__
methods,
> or the like.

Whatever works for you is fine by me.  I personally expect I would use
__del__ in a cycle only to torture newbies on c.l.py with the obscure
possibilities <wink>.

> Now we have wide-open policy, which I recall you decrying <wink>.

When a language *can* do everything "the right way" on my behalf & without
bothering me, I appreciate it when it does.  refcounting is like that.

But when it *can't* do "the right thing" by magic, it's obligated first to
do no harm, and second to give me a way to do it myself.  That's where the
wide-open policy comes in; e.g., if objects X and Y are in a dead cycle, and
X's finalizer happens to need the services of an active Y in order to shut
itself down cleanly, the language does harm if it runs Y's finalizer
first -- and may also do harm if it refuses to finalize either.  But the
language simply can't guess the intent of your code, so it should refuse to
act in ignorance, instead letting you do it explicitly however you require.
I don't think Python should fill in your "def" bodies with code it makes up
either <wink>.

> I guess that no matter what mechanism is implemented, certain
badly-handled
> cases will appear and need to be handled with "well, don't do that!"

No, no, that's never ever happened in *Python* <wink>.

> finalization-is-like-life-you-can't-win-and-you-can't-break-even-ly y'rs
> Evan Simpson

luckily-though-you-do-get-to-die-in-the-end-ly y'rs  - tim





From news at dorb.com  Mon Jun 28 09:58:33 1999
From: news at dorb.com (Darrell)
Date: Mon, 28 Jun 1999 09:58:33 -0400
Subject: Python for large projects?
References: <3774b5d1.690922@news.force9.net>  <T6bd3.2883$U6.12014@newsr2.twcny.rr.com> <slrn7naup9.jk.thantos@brimstone.mecha> <RGid3.2957$U6.12374@newsr2.twcny.rr.com> <87n1xm85mf.fsf@lisboa.i-did-not-set--mail-host-address--so-shoot-me> <Jhsd3.3090$U6.12410@newsr2.twcny.rr.com> <3776EF65.A07CBEFC@compaq.com> <slrn7ne533.1rmr.marc@oscar.noc.cv.net>
Message-ID: <xeLd3.3224$U6.13605@newsr2.twcny.rr.com>

> >Darrell wrote:
> >>
> >> I'll give that a look, but they can always get to it if they want.
Maybe
> >> that's ok, just worries me.
> >
> >If you clearly warn them in the docs not to do
> >it, and they do it anyway and get into trouble,
> >it's not your fault.
> >
> >Greg
>
> But what does management think??  If the user screws something up how
> much down time/loss of data will result?  I remember people coming back
> from desert storm and telling me about how people were playing with
> unexploded submunitions, aka small bombs with flaky fuses, I am not
> kidding on this.  So the question is how much damage can an moron
> do with a screw driver to your program and is that an acceptable
> tradeoff for the benifits of python?  If you are realy parinoid
> present it to your boss and ask him to decide.
>
> marc
>
> ps just because you are parinoid does not mean you are wrong.

Yes I'm paranoid. Got trained that way in the medical instrument business.
But now I'm in the publishing business where I'm sure they wonder about me.

I'm glad I brought this up because Michael Hudson suggested using RExec.
This sounds like a perfect solution to my concerns. Even better than C++
const.

--Darrell






From tismer at appliedbiometrics.com  Tue Jun 29 15:15:25 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Tue, 29 Jun 1999 19:15:25 GMT
Subject: ANN: Stackless Python 0.2
References: <000c01bec11f$8dfbf800$e19e2299@tim>
		<19990629121441.A19011@longford.cs.monash.edu.au> <37790FE1.4F778B4C@gte.com>
Message-ID: <37791B4D.43ADC401@appliedbiometrics.com>


Gary Duzan wrote:
> 
> Toby J Sargeant wrote:
> >
...
>    This appears to be the conventional wisdom. The optimization class I
> took used an ML-like language, and the first thing we did was to move
> heap frames, which was the default, to the stack (i.e. closure
> optimization.)

Loosing much of flexibility by doing that, btw.

> It might be less of an issue in an interpreted language if it already has
> heap
> allocation (or other sorts of) overhead to deal with when making calls.
> There may be other allocation tricks which can be used, as well.
> 
> > It seems that this should affect Stackless Python equally as much. Does anyone
> > have anything to add on the subject? I would imagine that frames could be
> > allocated and managed in chunks to alleviate a lot of the memory management
> > load...
> 
>    I would think that as well (at least at first.) Maybe pools of different
> sizes of frame to limit the space overhead.

Maybe you should really look into the sources.
You are discussing things which are not Python
related, since this simply does not apply.

The Python "stacks" are so tiny that you can forget about them.
Allocations always never occour, unless your program runs
astray in recursion.

The cost of a call comes from the slightly complicated
function setup which is done all the time. Frames belong
to the fastest available Python objects already, 
like, say dictionaries.

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From bkc at Murkworks.com  Fri Jun 11 15:44:30 1999
From: bkc at Murkworks.com (Brad Clements)
Date: Fri, 11 Jun 1999 15:44:30 -0400
Subject: Any Production ZOPE sites we can visit?
References: <37614A32.79FE554E@metricom.com>
Message-ID: <7jrou7$p42$1@news.clarkson.edu>

http://www.air.gen.ny.us

Nothing exciting there really, except that the airport lookups are done
through an SQL query..

I'm working on an events Calendar, that'll really look nice when it's done

--
Brad Clements,
bkc at murkworks.com
Michael Kersey <mkersey at metricom.com> wrote in message
news:37614A32.79FE554E at metricom.com...
> Are there many production Zope sites visible on the www?
>
> Of the links from the Zope home page (Case Studies), only Musi-Cal at
> http://www.musi-cal.com/ _appears_ (to the untrained eye!-)) to be a
> Zope site.
>
> It's a little tough to get my hands around how Zope works and what it
> can do without good examples.
>
> Is Zope being used primarily for intranets?





From robin at alldunn.com  Tue Jun 29 03:41:43 1999
From: robin at alldunn.com (Robin Dunn)
Date: Tue, 29 Jun 1999 00:41:43 -0700
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
References: <52B9C15E4136A939.DC364288A600C018.AB3BFEAACFF2E162@lp.airnews.net>
Message-ID: <9N_d3.31$oY.3801295@news.randori.com>

Gee, I'm away from the newsgroup for just a few days and look what happens.
People everywhere start talking about me and my kid, comparing him to the
other kids on the block and trying to figure out whose kid is better,
faster, cuter, etc.  Personally, I'm happy if they all just play together
nicely.

Seriously though, I guess it's time for me to put in my 2 cents...  I don't
think that wxPython is ready to replace tkinter as the defacto standard,
(though it would be nice if it was someday.)  Quite simply it just doesn't
have broad enough platform support.  Now if somebody could convince Steve
Jobs to give me a Mac then maybe I could help do something about that
problem, but until then I can't.

I also don't think that wxPython is ready yet to be included with the Python
distribution.  (Notice I said "yet.")  wxPython (and wxWindows itself) is
still changing too fast to be stable enough for Guido to not lay awake at
night worrying about it.  On the other hand, I expect that to change over
the next few months so maybe by the time 1.6 is close to being ready we
shoudl address this again...

One real good idea to come out of this thread is wxTk, building a wxWindows
port on top of the Tk C API.  That would certainly help get the platform
coverage, and help reduce some library dependencies.  A fair amount of
wxWindows is written genericly using wxWindows and no platform code so it
would not be as big of a job as it seems...

Well enough of my rambling.

Have fun and Happy Pythoning!

--
Robin Dunn
robin at AllDunn.com
http://AllDunn.com/robin/
http://AllDunn.com/wxPython/  Check it out!
Try http://AllDunn.com/laughworks/ for a good laugh.








From janssen at parc.xerox.com  Mon Jun 28 22:55:15 1999
From: janssen at parc.xerox.com (Bill Janssen)
Date: Tue, 29 Jun 1999 02:55:15 GMT
Subject: Python for large projects?
In-Reply-To: <87d7yhv8ys.fsf@lisboa.i-did-not-set--mail-host-address--so-shoot-me>
References: <3774b5d1.690922@news.force9.net>
	<tx3ogi1ahos.fsf@hog.ldgo.columbia.edu>
	<87d7yhv8ys.fsf@lisboa.i-did-not-set--mail-host-address--so-shoot-me>
Message-ID: <0rS3KHAB0KGWBho21G@holmes.parc.xerox.com>

Excerpts from ext.python: 27-Jun-99 Re: Python for large projects? Janko
Hauser at ifm.uni-kie (881)

> Jody Winston <jody at ldgo.columbia.edu> writes:

> > I've used Python for three large distributed applications (each > 100k
> > lines of Python).  The only other recommendation that I would add is
> > to define interfaces to the code.  If this is done with a product like
> > ILU or FNORB, you can then easily call the Python code from other
> > languages.  In addition, you can replace critical sections with other
> > languages if needed.
> > 
> I have thought about this as the perfect way to do ``data hiding'',
> publish only special interfaces for your data. Has this a significant
> overhead, if used on one machiene, or is it overkill just for this
> purpose :-)?

I've done this a fair amount with Python and ILU.  I specify the
`public' ILU interface, which anyone can call, then also specify various
other interfaces, private to one more specific management concern or
another.  Typically the private interfaces contain subtypes of the
object types declared in the public interface, with additional methods
that allow more access.  Then I implement only the more-derived object
types in my Python code.  So there's no real overhead in the code or in
the method calls.  The users of the public interface only get to do the
public operations; the more specialized clients get to do the fancier
operations.

Bill




From mfletch at vrtelecom.com  Tue Jun 29 12:30:39 1999
From: mfletch at vrtelecom.com (Mike C. Fletcher)
Date: Tue, 29 Jun 1999 16:30:39 GMT
Subject: PythonWin VSS Integration
References: <002a01bec203$1fc90230$a9807018@yec1.on.wave.home.com> <7la6vj$rh0$1@m2.c2.telstra-mm.net.au>
Message-ID: <Pw6e3.16902$o6.286713@news2.rdc1.on.home.com>

Spent the morning playing with this.  The mssccprj.scc file was not there,
adding it to the directory allowed errors to occur.  The errors were of the
type can't find this item.  After verifying that I was using the proper
format for project name, spent some time spelunking on MSDN to discover how
SourceSafe decides which database to load.  Since PythonWin does not check
for a non-default database, SourceSafe searches in this manner (quoted from
MSDN, copyright Microsoft):

1  Search for srcsafe.ini in the directory where ssapi.dll is located.

2  Search for srcsafe.ini in each directory of the path to ssapi.dll. In
other words, if ssapi.dll is located in
C:\Folder1\Folder2\Folder3\SSAPI.DLL, Folder3, Folder2, Folder1 and C:\ are
searched (in that order).

3  Search for srcsafe.ini in the location indicated by the named value "API
Current Database" in the registry key
HKEY_LOCAL_MACHINE\Software\Microsoft\SourceSafe.

4  Search for srcsafe.ini in the location indicated by the named value
"SCCProviderPath" in the registry key
HKEY_LOCAL_MACHINE\Software\Microsoft\SourceSafe.

I have as of yet been unable to get the SourceSafe Open automation command
to accept a non-default argument, so I cannot offer a patch to vss.py.  I've
worked around problem by renaming the default database's INI file, and
putting the path to my non-default database into the key mentioned in step
three above.  Everything now works as expected (that is, attempts to alter
the file prompt for check out and do perform check).

Okay, have spent far more time on this than I was looking to, enjoy
yourselves,
Mike

Mark Hammond <MHammond at skippinet.com.au> wrote in message
news:7la6vj$rh0$1 at m2.c2.telstra-mm.net.au...
> Mike C. Fletcher wrote in message <002a01bec203$1fc90230
> [Laments about Pythonwin and Source Safe]
>
> >So, any hints?  What does visual source safe integration mean?
>
> Well, it _should_ give you some clues.  Note that you need to restart
> Pythonwin before the option will take effect.  Also note that the file
must
> be marked read-only on the disk.
>
> Once enabled, and assuming VSS is installed correctly, the most likely
error
> message you will see is something like "this directory is not configured
for
> VSS".  [Once the option is checked, you should get _some_ message on a
> read-only file]
>
> In this case, you need a "mssccprj.scc" file either in the file's
directory
> or a parent directory.  This is the same name and format as VB uses -
> basically a Windows INI file.  A section [Python] with entry
> "Project=ProjectName" is referenced.  The project name is the name of the
> VSS project the Python code attempts to check the file out from.  If the
> .scc file is in a parent directory, the correct relative VSS path is
built -
> so if your file system matches your VSS structure, you only need a single
> .scc file in the VSS "root" directory.
>
> Hope this makes sense.
>
> Mark.
> >
> >Thanks,
> >Mike
> >
>
>
>





From mike at proteus.dialup.cloud9.net  Fri Jun  4 10:42:35 1999
From: mike at proteus.dialup.cloud9.net (Michael Muller)
Date: Fri, 4 Jun 1999 10:42:35 -0400
Subject: Threaded Tkinter
In-Reply-To: <14167.23200.724983.451158@tiana.kli.prv>
References: <m3btex4vxu.fsf@tiana.kli.prv>
	<375690B9.37C1B8B6@cloud9.net> <14167.23200.724983.451158@tiana.kli.prv>
Message-ID: <Pine.LNX.3.91.990604103109.242D-100000@proteus.dialup.cloud9.net>

On Thu, 3 Jun 1999 gutier at intergate.bc.ca wrote:

> 
> Hi Michael.
> 
> I just went through your code and tried it out. I think I understand
> what it does now. Thanks for posting it.
> 
> For what I would really like to eventually do (processing network
> activity on a separate thread and updating it on a UI), I still need
> to signal a change in UI from another thread. The poll works but, as
> you obviously know, it is suboptimal. What would be best is if one
> could inject a message into Tk's event queue to cause it to
> immediately call a function, which would read a queue and do
> something.
> 
> I understand there is an "event" command in Tk, but I know nothing
> about it. Do you think it would be helpful?
[snip]

You can generate custom events under Tkinter, but I think you're going to run
into the same problem with the Tkinter lock: the function that sends the
event is going to block on the Tkinter lock for as long as the main thread
is in the message queue.

>From what I've seen of the Tkinter main loop, there are facilities for
obtaining events from sources other than the window event queue,
unfortunately I don't know how to use them.

=============================================================================
michaelMuller = proteus at cloud9.net | http://www.cloud9.net/~proteus
-----------------------------------------------------------------------------
If you are not willing to control your own mind, there are plenty of other
people who are willing to do it for you.
=============================================================================




From Kevin.Dahlhausen at p98.f112.n480.z2.fidonet.org  Wed Jun 30 15:12:15 1999
From: Kevin.Dahlhausen at p98.f112.n480.z2.fidonet.org (Kevin Dahlhausen)
Date: Wed, 30 Jun 1999 20:12:15 +0100
Subject: Python GUIs: Abandoning TkInter and welcoming wxPython?
Message-ID: <000015ce@bossar.com.pl>

From: morse at harborcom.net (Kevin Dahlhausen)

Lyle Johnson <jlj at cfdrc.com> wrote:

>Just to add fuel to the fire, we now have a Python interface to FOX, a
>cross-platform GUI toolkit:
>
>    http://www.cfdrc.com/pub/FOX/fox.html


How did you get around the requirement of  preprocessor 
table definitions for message handlers?

(I'm downloading it now....)




From hniksic at srce.hr  Fri Jun  4 05:57:38 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 04 Jun 1999 11:57:38 +0200
Subject: threads
References: <3756C078.5D998D3B@icrf.icnet.uk> <14166.51558.850416.153999@bitdiddle.cnri.reston.va.us> <8790a1njeu.fsf@pc-hrvoje.srce.hr> <14166.60091.353043.829598@amarok.cnri.reston.va.us>
Message-ID: <87k8tks2y5.fsf@pc-hrvoje.srce.hr>

"Andrew M. Kuchling" <akuchlin at mems-exchange.org> writes:

> Hrvoje Niksic writes:
> >Jeremy Hylton <jeremy at cnri.reston.va.us> writes:
> >> The short answer is no.  The Python interpreter has a lock that
> >> prevents allows only a single thread to execute at a time.
> >
> >Eek.  That's depressing.  Isn't supporting threads and not allowing
> >simultaneous execution a bit weird?
> 
> I don't think so, because the common case is probably not the one
> where lots of work is being done in Python code.

Well, I do think so.  The "common case" may be that lot of work is not 
done in Python code, but maybe such common cases emerge from lousy
execution to begin with.

Dividing threaded programs to IO-heavy and computation-heavy misses
the point because many realistic cases have some of both -- for
instance, an internet server under heavy load.  Python's
characteristics make it quite suitable for such a purpose, and having
threads lock each other off is a _definite_ minus on a multi-CPU
machine.

Maybe it would be possible to make the actual interpreter thread-safe,
and only lock when an extension of unknown thread-safety is called.



From news at dorb.com  Wed Jun 23 14:45:06 1999
From: news at dorb.com (Darrell)
Date: Wed, 23 Jun 1999 14:45:06 -0400
Subject: Newbie: How do you "clear" PythonWin between runs?
References: <37700E61.5388E101@alliedsignal.com> <cUVb3.1952$U6.6930@newsr2.twcny.rr.com> <3771006D.22A9EC2D@alliedsignal.com>
Message-ID: <hZ9c3.2164$U6.7813@newsr2.twcny.rr.com>

I went to the python faq and used the "Python FAQ Wizard" and searched on
"reload"
http://grail.cnri.reston.va.us/cgi-bin/faqw.py


--
--Darrell
H. P. Friedrichs <" HPeter.Friedrichs"@alliedsignal.com> wrote in message
news:3771006D.22A9EC2D at alliedsignal.com...
> Darrell:
>
> What exactly is this URL? What document were you referencing? Is there
another
> way to get it?
>
> I can't seem to connect to this URL...I keep getting "server busy or
offline"
> type errors.
>
> Thanks,
> Pete
>
>
> Darrell wrote:
>
> >
http://grail.cnri.reston.va.us/cgi-bin/faqw.py?query=reload&querytype=simple
> > &casefold=yes&req=search
> >
> > --
> > --Darrell
> > H. P. Friedrichs <" HPeter.Friedrichs"@alliedsignal.com> wrote in
message
> > news:37700E61.5388E101 at alliedsignal.com...
> > > I've just started playing with python, so all of this is very new to
me.
> > > I have been writing some scripts, saving them to *.py files, and
running
> > > them under PythonWin 1.5.2 . The process is iterative; I code, save,
> > > run, code, save, run, etc.
> > >
> > > What I have noticed is that python doesn't "reset" between runs.
> > > Sometimes error  messages reappear even though I have fixed the
problem,
> > > because some artifact of a previous run lingers.
> > >
> > > So, my question is this: How can I "flush" or "clear" the python
> > > interpreter between successive runs? Is there a command I can add to
my
> > > script to initialize python before running the script?
> > >
> > > Thanks,
> > >
> > > Pete
> > > hpeter.friedrichs at alliedsignal.com
> > >
>
> -----------------------------------------------
> | The opinions expressed here are my own,     |
> | and not necessarily those of my employer.   |
> -----------------------------------------------
>
>





From paul at prescod.net  Sun Jun 27 11:52:07 1999
From: paul at prescod.net (Paul Prescod)
Date: Sun, 27 Jun 1999 15:52:07 GMT
Subject: Pythonwin and Office 2000
Message-ID: <377648A7.9B9BE59F@prescod.net>

I believe that Pythonwin is not compatible with Office 2000. Here are some
symptoms:

D:\Program Files\PYTHON\WIN32COM\TEST>python testMSOffice.py

Starting Word 8 for dynamic test
Starting Word 8 for non-lazy dynamic test
Starting MSWord for generated test
Traceback (innermost last):
  File "testMSOffice.py", line 143, in TestAll
    TestWord()
  File "testMSOffice.py", line 46, in TestWord
    if gencache.EnsureModule("{00020905-0000-0000-C000-000000000046}",
1033, 8,
0) is None:
  File "D:\Program Files\Python\win32com\client\gencache.py", line 226, in
Ensur
eModule
    tlbAttr = pythoncom.LoadRegTypeLib(typelibCLSID, major, minor,
lcid).GetLibA
ttr()
com_error: (-2147319779, 'Library not registered.', None, None)


>>> a=win32com.client.Dispatch("MSWord.Application")
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File "D:\Program Files\Python\win32com\client\__init__.py", line 13, in
Dispatch
    dispatch, userName =
dynamic._GetGoodDispatchAndUserName(dispatch,userName,clsctx)
  File "D:\Program Files\Python\win32com\client\dynamic.py", line 67, in
_GetGoodDispatchAndUserName
    return (_GetGoodDispatch(IDispatch, clsctx), userName)
  File "D:\Program Files\Python\win32com\client\dynamic.py", line 58, in
_GetGoodDispatch
  IDispatch = pythoncom.CoCreateInstance(IDispatch, None, clsctx,
pythoncom.IID_IDispatch)
com_error: (-2147221005, 'Invalid class string', None, None)




  File "..\..\slides\slidesppvisitor.py", line 4, in ?
    ppmod = win32com.client.gencache.GetModuleForProgID(
"PowerPoint.Application" )
  File "D:\Program Files\Python\win32com\client\gencache.py", line 134, in
GetModuleForProgID
    return GetModuleForCLSID(iid)
  File "D:\Program Files\Python\win32com\client\gencache.py", line 152, in
GetModuleForCLSID
    return GetModuleForTypelib(typelibCLSID, lcid, major, minor)
  File "D:\Program Files\Python\win32com\client\gencache.py", line 167, in
GetModuleForTypelib
    return _GetModule(modName)
  File "D:\Program Files\Python\win32com\client\gencache.py", line 328, in
_GetModule
    mod = __import__("win32com.gen_py.%s" % fname)
  File "D:\Program
Files\Python\win32com\gen_py\91493440-5A91-11CF-8700-00AA0060263Bx0x2x14.py",
line 4549
    def AddOLEObject(self, Left=0.0, Top=0.0, Width=-1.0, Height=-1.0,
ClassName
=L'', FileName=L'', DisplayAsIcon=0, IconFileName=L'', IconIndex=0,
IconLabel=L'
', Link=0):

Is this "L" syntax a future syntax for Unicode that Python 1.5.2 doesn't
support? Perhaps PythonWin should test for support for that syntax before
it generates the Python code?

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for only himself
 http://itrc.uwaterloo.ca/~papresco

Perhaps the war in Kosovo would get more press if it were directed by
George Lucas.



From Fredrik.Lundh at p98.f112.n480.z2.fidonet.org  Wed Jun 30 04:09:30 1999
From: Fredrik.Lundh at p98.f112.n480.z2.fidonet.org (Fredrik Lundh)
Date: Wed, 30 Jun 1999 09:09:30 +0100
Subject: PySequence_Check
Message-ID: <000015a9@bossar.com.pl>

From: "Fredrik Lundh" <fredrik at pythonware.com>

Reuben Sumner wrote:
> How can I do the equivalent of a PySequence_Check in Python?
> GvR told me that the question was illdefined (if I explained it correctly)
> and to ask here.  It seems that some of the functions in abstract.h
> are available from python (len(), callable()) and others are not
> (PySequence_Check, PyNumber_Check).

import operator
operator.isSequenceType
operator.isNumberType
operator.isMappingType

not documented, and should probably be considered as
deprecated (see replies from Paul and Tim for some back-
ground).

> What am I missing?  I have a nested structure of lists and tuple and I
> just want to know when I have bottomed out.  I can check for tuple or
> list indivually but that is ugly and inflexible.

> I could just try indexing and see if there was an exception but
> that is not much better.

given that Python has no support for declared inter-
faces, that's a reasonable approach.  works just fine,
and adds no extra overhead.

on the other hand, it doesn't work if you have strings
in there.  but that's true for most other approaches
as well:

>>> operator.isSequenceType("spam")
1
>>> operator.isSequenceType("spam"[0])
1
>>> operator.isSequenceType("spam"[0][0])
1
>>> operator.isSequenceType("spam"[0][0][0])
1
>>> operator.isSequenceType("spam"[0][0][0][0])
1
>>> operator.isSequenceType("spam"[0][0][0][0][0])
1

</F>



From moshez at math.huji.ac.il  Fri Jun 18 07:37:07 1999
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 18 Jun 1999 14:37:07 +0300
Subject: When the code and the comments disagree
Message-ID: <Pine.SUN.3.95-heb-2.07.990618143540.12733B-100000@sunset.ma.huji.ac.il>

Or maybe I have misunderstood something. From dospath.py

---------------
# Is a path a mount point?
# XXX This degenerates in: 'is this the root?' on DOS

def ismount(path):
        return isabs(splitdrive(path)[1])
----------------

Why checking isabs, instead of checking =='\\'??????
Beyond me. Anybody?
--
Moshe Zadka <mzadka at geocities.com>. 
#!/usr/bin/tail -1
Just another tail hacker.




From arcege at shore.net  Thu Jun 24 13:43:56 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Thu, 24 Jun 1999 17:43:56 GMT
Subject: Default printing behavior for classes
References: <7ktoe5$bbl$1@nnrp1.deja.com>
Message-ID: <w7uc3.98$Of3.18161@news.shore.net>

Ben Glazer <glazer at scicomp.com> wrote:
: I'm looking for a way to set default printing behavior for a class I've
: defined.  Lists are printed in a certain format by default.  I'd like to
: be able to define this for any type of class.  It'd be really nice if
: there were something I could include in my class definition, e.g.

:  class Board:
:      def __init__(self, width=2, height=4, length=20):
:          self.width = width
:          self.height = height
:          self.length = length

:      def __print__:
:          print "[%s,%s,%s]" % (width,height,length)

: such that

:  b = Board(8,8,2)
:  print b

: would print "[8,8,2]".

: Is there anything in Python that encapsulates this functionality, or
: should I resign this desire to a feature req for Python 2.0?

There are two such things: the result from repr(object) and from
str(object).  You can create your own versions by creating the methods
"__repr__" and "__str__" respectively.  Both are to return a string.

The more general is the output of repr(), if there is no method to
output str(), then repr() is used.  The "print" statement uses str(),
containers (lists, dictionaries, etc.) and string conversion (`...`)
both use repr().

Why are there two such functions?  Taking just one simple example, look
at the output of a Python string:

  >>> a = 'Hi\tBen\n'
  >>> print a
  Hi      Ben

  >>> print str(a)
  Hi      Ben

  >>> print repr(a)
  'Hi\tBen\n'
  >>>

With the str() function it is difficult to see the special characters.
The repr() is used to show the internal representation of the object,
not the human-readable "pretty" representation.

Preferably, the result of repr() is something that can be called from
eval to recreate the object.

Since printing tuples, lists and dictionaries isn't generally "human
friendly" and more for use with eval(), repr() is used to print the
contained objects.

So for your class, you could write something like:

  class Board:
    def __init__(self, width=2, height=4, length=20):
      self.width = width
      self.height = height
      self.length = length
    def __str__(self):
      return '[%d, %d, %d]' % (self.width, self,height, self.length)
    def __repr__(self):
      return '%s(%d, %d, %d)' % (
        self.__class__.__name__,  # the instance class name
        self.width, self.height, self.length
      )

>>> print b
[8, 8, 2]
>>> print `b`
Board(8, 8, 2)
>>> b
Board(8, 8, 2)
>>> print 'The board is %s.' % b
The board is [8, 8, 2].
>>>

For more information, read the Python Language Reference manual.
(http://www.python.org/doc/current/ref/customization.html#l2h-288)

  -Arcege




From adjih at technologist.com  Wed Jun  9 18:10:20 1999
From: adjih at technologist.com (adjih at technologist.com)
Date: Wed, 09 Jun 1999 22:10:20 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com> <7j450j$908$2@cronkite.cc.uga.edu> <7j5sr7$eb4$1@nnrp1.deja.com> <7j6j8d$58f$2@cronkite.cc.uga.edu> <7j902g$hob$1@nnrp1.deja.com> <87n1yc4k2e.fsf@ev.netlab.co.jp>
Message-ID: <7jmoo5$f6$1@nnrp1.deja.com>

In article <87n1yc4k2e.fsf at ev.netlab.co.jp>,
  Yukihiro Matsumoto <matz at netlab.co.jp> wrote:
>
> adjih at technologist.com writes:
>
> |  It is true that it shouldn't be hard to make a mark-sweep collector
> |that is working in practice on many platforms.
> |
> |  Pedantically I doubt that ANSI-C tells you the current stack top
> |address, or guarantees that the whole address space between the
current
> |stack pointer and this top stack address is stack space. But I agree
> |that actual C compilers wouldn't do something too stupid, we
> |shouldn't care about this :-)
>
> You don't have to know stack address to implement mark-sweep register,
> unless you want to make it conservative.  The mark-sweep GC can be
> written in ANSI-C portable.

But then this can break Python C modules, because references are held
from the C stack. Just the snippet code:
[...]
PyObject* myint=PyInt_FromLong(19991999);
[...]
would break with a mark-sweep GC ignoring the stack.
(almost all the C/Python code creating objects is in danger).

Of course, the solution is to call GC only at some places, but isn't
there an relation between the Python stack and the C stack ?

For instance if a module creates an object and then calls a function
(which can call "eval"), isn't the big interpreter switch entered
a second time ? Can one GC at all before the function/eval return ?

-- Cedric


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From doughellmann at mindspring.com  Tue Jun  1 10:20:17 1999
From: doughellmann at mindspring.com (Doug Hellmann)
Date: Tue, 01 Jun 1999 14:20:17 GMT
Subject: intractable(?) Tkinter scrollbar problem
References: <19990601001054.36247@nms.otc.com.au> <jonathan-3105992257320001@borg.onegoodidea.net> <19990601124302.48567@nms.otc.com.au>
Message-ID: <3753ECEE.58AE682A@mindspring.com>

Part of the problem may be related to comparing floating point values
for exact equality to one another.  Specifically, the comparison of
float(hi) >= 1.0.
 
I was able to make the code not show the scrollbars at all, with no
flicker, by changing:
 
if float(lo) <= 0.0 and float(hi) >= 1.0:
 
to

if lowtest <= 0.0 and (float(hi) - 1.0) <= 0.0:

However, when I resize the window smaller set() isn't called again, so
there isn't an opportunity to have them come back.  That seems to be an
issue of catching the resize event, though, but I'm a bit unclear on how
that works.  I tried registering a function for '<Configure>', and it
was called once on startup.  But when I resize the window it isn't
called.

Have I headed off in the wrong direction here?

Doug



From dkuhlman at netcom.com  Mon Jun 28 14:46:53 1999
From: dkuhlman at netcom.com (G. David Kuhlman)
Date: 28 Jun 1999 18:46:53 GMT
Subject: ANN: Stackless Python 0.2
References: <000c01bec11f$8dfbf800$e19e2299@tim>
Message-ID: <7l8fut$es6@dfw-ixnews6.ix.netcom.com>

Tim Peters (tim_one at email.msn.com) wrote:

> 
> That's mostly true today too:  Python threads run serially now, one at a
> time, and if a thread calling out to C doesn't release the global lock no
> other thread will run until it returns.
> 

I'll bite. How *does* a thread that calls out to C release the global
lock?

Would it have to do this thingy that is at the beginning of the
interpreter loop in python/ceval.c?

    if (PyThreadState_Swap(NULL) != tstate)
        Py_FatalError("ceval: tstate mix-up");
    PyThread_release_lock(interpreter_lock);

Would this work? or would it screw up the interpreter?

Is there a way to do this in a Python script *before* it calls out
to C.

Is there an approved solution to this problem?


  - (mystified) Dave



From da at ski.org  Tue Jun 22 00:04:32 1999
From: da at ski.org (David Ascher)
Date: Mon, 21 Jun 1999 21:04:32 -0700 (Pacific Daylight Time)
Subject: need some fast balanced-tree datastructure
In-Reply-To: <376ED372.FA73994F@cs.ucsd.edu>
Message-ID: <Pine.WNT.4.05.9906212103320.147-100000@david.ski.org>

On Mon, 21 Jun 1999, Pavel Velikho wrote:

>     I am looking for a fast implementation of incremental container
> that maintains an ordering on the keys... (i.e. I want an AVL,
> Red-Black, or anything similar). I have tried the b+tree in Python but
> its not fast enough. Is there anything like that available as a C
> extension module?

I've used Sam Rushing's AVL module with great success.  It is available
at:

  ftp://squirl.nightmare.com/pub/python/python-ext/avl/

Cheers,

--david ascher




From gregm at iname.com  Mon Jun 21 05:56:35 1999
From: gregm at iname.com (Greg McFarlane)
Date: Mon, 21 Jun 1999 09:56:35 GMT
Subject: Tkinter missing methods
Message-ID: <19990621195635.02550@nms.otc.com.au>

Tkinter.py revision 1.127, (python1.5.2) is missing these Text
methods:

    def mark_next(self, index):
	return self.tk.call(self._w, 'mark', 'next', index) or None

    def mark_previous(self, index):
	return self.tk.call(self._w, 'mark', 'previous', index) or None

-- 
Greg McFarlane:    INMS Telstra Australia (gregm at iname.com)
Today's forecast:  Sunny, with occasional cloudy periods and a chance
		   of precipitation in some areas.



From dalke at bioreason.com  Fri Jun 25 17:09:11 1999
From: dalke at bioreason.com (Andrew Dalke)
Date: Fri, 25 Jun 1999 15:09:11 -0600
Subject: GC In Python: YAS (Yet Another Summary)
References: <Pine.SUN.3.95-heb-2.07.990616155228.15761A-100000@sunset.ma.huji.ac.il> <p53dzrfcea.fsf@bidra241.bbn.hp.com> <930267199.139037@clint.waikato.ac.nz> <37730BAA.28A7C738@bioreason.com> <930340397.465868@clint.waikato.ac.nz>
Message-ID: <3773EFF7.27D29899@bioreason.com>

Stuart Yeates <syeates at cs.waikato.ac.nz> said to me:
> I'm not quite sure what you mean by 'integer handle,' but if you
> do a straight cast then it shouldn't be a problem (since the 
> collector operates on the binaries representation). If you mean
> that it uses an index to an array of pointers (or some such), 
> then that shouldn't be a problem either, since the pointer still
> exists somewhere (i.e. the array). 

The latter.   Example object creation code might look like:

dt_Handle mol = dt_smilin("CCC", 3);

where dt_Handle is a typedef to unsigned long.  In the SWIGged
wrapped version, this is mol = dt_smilin("CCC").

Yes, the library has a pointer table to the object.  The problem
I see is, how does the Boehm collector (or other GC) know what to
collect from this vendor library?  I can't see how, given that
my Python code stores it as an integer.

That's why I don't see how you say "It fully supports C and C++"
when this is a package which doesn't do "pointer XORing or other
crimes" but cannot be usable with a GC w/o modifications.

But then, I'm a computational biophysicist by training and don't
know much about GC other than the general concepts.  One of the
references on the SGI page says Boehm can be used in "uncooperative
environments" so I'll end by saying that I don't know enough.

						Andrew Dalke
						dalke at bioreason.com



From stig at pvv.ntnu.no  Mon Jun 14 15:31:47 1999
From: stig at pvv.ntnu.no (Stig Hemmer)
Date: 14 Jun 1999 21:31:47 +0200
Subject: OO (was: Why is tcl broken?)
References: <1283099377-77267484@hypernet.com> <Pine.SGI.4.05.9906101104460.3354-100000@lemur.magnet.com> <375FDCA7.62FF@mailserver.hursley.ibm.com> <7josm1$bgc$1@Starbase.NeoSoft.COM> <37628EBF.207F7A8F@iname.com> <3762988E.C94AA252@iname.com> <phXineXly-1206991437590001@usrns72.dialup.hawaii.edu> <3763C746.5CA8FE78@iname.com> <m3u2sby4gf.fsf@soma.andreas.org> <87pv2zjhhl.fsf@home.ivm.de>
Message-ID: <ekvn1y2efz0.fsf@gnoll.pvv.ntnu.no>

> Prefix rocks, infix is crap.

(think I (meant you (and (rocks prefix) (is-crap infix))))

Stig Hemmer,
Jack of a Few Trades.




From mwh21 at cam.ac.uk  Tue Jun  1 05:54:32 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 01 Jun 1999 10:54:32 +0100
Subject: Python 2.0
References: <11A17AA2B9EAD111BCEA00A0C9B4179301E10ED6@forest.origin.ea.com>
Message-ID: <m3r9nwi6uf.fsf@atrus.jesus.cam.ac.uk>

"Stidolph, David" <stidolph at origin.ea.com> writes:

> >OK, I'll try to propose something constructive. Maybe we need an object
> >protocol, that would enumerate
> >all references held by an object ?  Writing a portable GC would be then
> much
> >easier.
> 
> Sounds good to me for debugging.  A call that could return a list of
> everything that holds a reference to an object - that would be cool!
> 
> list = GetReferences(object)
> print 'List of referencest to',object
> for item in list:
>   print 'item:',item
> 
> Anybody know of a current way to do this?

If you recompile Python with Py_TRACE_REFS defined, then the sys
module sprouts a "getobjects" function that returns a list of all
objects in existence. This could probably be used to implement
something like this. It would be veeeeery slow, I suspect.

I've never resorted to this approach, tending always to make cyclic
references go away be staring at them very, very hard.

I'm not sure this is at all what is being asked for, but it seems kind
of relavent.

Yours,
Michael



From tre17 at cosc.canterbury.ac.nz  Tue Jun  8 20:30:07 1999
From: tre17 at cosc.canterbury.ac.nz (Timothy R Evans)
Date: 09 Jun 1999 12:30:07 +1200
Subject: In what order does Python read dictionary keys?
References: <7jjnt5$snk$1@nnrp1.deja.com>
Message-ID: <c754skitdv4.fsf@pc142.cosc.canterbury.ac.nz>

tcpeter at my-deja.com writes:

> Hello, all.  I've checked the docs and queried DejaNews for this but I
> couldn't come up with an answer.  I wrote a script to convert rgb
> values to hex for when I'm doing colors in HTML pages and don't have
> Photoshop handy.  The code appears below.  My question is, how does
> Python decide in which order to read the keys?  Notice that I have the
> dictionary set up in this order: red, blue, green.  (I mistyped it while
> writing the code).  But when I ran the code, the options were presented
> in the desired order. (A trifle creepy, if you ask me.  Turing would be
> delighted.)  Anyway, how does Python decide the order. Thanks.
> 
> 
> #! /usr/bin/python
> 
> # The first thing to do is create a dictionary containing the
> # questions to be asked where the color name is the dictionary key.
> 
> entries = {'red':'Type the red value, please: ',
> 	'blue':'Type the blue value, please: ',
> 	'green':'Type the green value, please: '}
> 
> for key in entries.keys():		# Now get a loop going
> 					# through the dictionaries
> 
> 	foo = input(entries[key])	# Create a variable to hold
> 					# the return from the prompt
> 					# for each key
> 
> 	if (foo) < 16:			# Since hex numbers don't
> 					# have leading zeroes below
> 					# F, I added them.
> 
> 		print 'The hex equivalent for',foo,'is 0%X' % foo
> 	else:
> 		print 'The hex equivalent for',foo,'is %X' % foo
> 
> 
> Tim Peter
> (Not him, a different guy)
> 
> 
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.

The order that keys are retrieved is random, from the tutorial section
5.4 on Dictionaries:

"""
The keys() method of a dictionary object returns a list of all the
keys used in the dictionary, in random order (if you want it sorted,
just apply the sort() method to the list of keys). To check whether a
single key is in the dictionary, use the has_key()method of the
dictionary.
"""

To get the keys sorted alter your loop to be:

for key in entries.keys().sort():

>From the look of this example, you probably would be better off using
a nested tuple like this:

"""
entries = (('red', 'Type the red value, please: '),
           ('green', 'Type the green value, please: '),
           ('blue', 'Type the blue value, please: '))

for key,value in entries:
...
"""

or even:

"""
entries = ('red', 'green', 'blue')

for key in entries:
    value = 'Type the ' + key + ' value, please: '
"""

This way you can control the order exactly and it might be faster as
you avoid the cost of a dictionary lookup.

--
Tim Evans




From hniksic at srce.hr  Tue Jun  8 03:11:19 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 08 Jun 1999 09:11:19 +0200
Subject: waste of resources ?
In-Reply-To: "Barry A. Warsaw"'s message of "Mon, 7 Jun 1999 22:51:13 -0400 (EDT)"
References: <375C0F91.1C73E4E3@icrf.icnet.uk> <375C16A9.5ADC4990@icrf.icnet.uk> <375C2003.B01BEC3@arlut.utexas.edu> <87yahvzqz3.fsf@pc-hrvoje.srce.hr> <14172.34081.978798.911000@anthem.cnri.reston.va.us>
Message-ID: <87n1ybxj3c.fsf@pc-hrvoje.srce.hr>

"Barry A. Warsaw" <bwarsaw at cnri.reston.va.us> writes:

> >>>>> "HN" == Hrvoje Niksic <hniksic at srce.hr> writes:
> 
>     |     except (IOError, os.error), detail:
> 
> Or you could just catch EnvironmentError. :)

It probably didn't exist in Python 1.5.1, under which I was writing
the code in question.



From hniksic at srce.hr  Thu Jun  3 05:56:36 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 03 Jun 1999 11:56:36 +0200
Subject: Use of StringIO vs cStringIO in standard modules
Message-ID: <8790a1620b.fsf@pc-hrvoje.srce.hr>

I noticed that many standard modules use StringIO and not cStringIO,
although they don't need subclassing.  Is this intentional?

For example, base64.py uses StringIO to implement encodestring() and
decodestring().  Since both functions write to output line by line, I
imagine the performance hit of StringIO vs cStringIO might be
non-negligible.

Furthermore, is there a particular reason for maintaining two parallel
StringIO implementations?  If subclassing is the reason, I assume it
would be trivial to rewrite StringIO to encapsulate cStringIO the same 
way that UserDict encapsulates dictionary objects.



From tim_one at email.msn.com  Wed Jun  2 00:53:40 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 2 Jun 1999 00:53:40 -0400
Subject: Python 2.0
In-Reply-To: <37533165.C6CF92C7@hex.net>
Message-ID: <001601beacb3$e299b440$d89e2299@tim>

[Raymond G. Beausoleil]
> Although I have not followed the recent evolution of Java very
> carefully, I have strong misgivings about the compromises made in
> the numerical core of Java. For  details, I recommend a quick
> review of William Kahan's presentation on this topic:

Make that Welvel Kahan, or W. Kahan as he seems to prefer.

> http://www.cs.berkeley.edu/~wkahan/JAVAhurt.pdf
>
> I'm new to Python, but I'm very excited about NumPy and PIL,
> and I wonder what compromises would have to made to these extensions
> if Python were rewritten in Java?

None, actually -- Python has no IEEE-754 story at all to tell today,
inheriting whatever random stuff the platform's C delivers, and having a
design in some areas that precludes 754 conformance (e.g., x == x is always
true in Python, and by design).  Java's rules are actually closer to 754
conformance than most popular languages'.

[Peter Koren]
> Has Sun done anything about Kahan's objections? This is indeed a serious
> matter.

If you read to the end of the paper, you'll find

    Java?s floating-point suffers from serious oversights. The same could
    be said of several other programming languages,

Ahem -- try replacing "several" with "virtually all" <0.1 wink>.

    some of them venerable, but Java lacks their historical excuses.

Recognize the polemic for what it is.  Java is likely much closer to
mandating substantial 754 conformance that whatever language you're using
today!

> A huge percentage of the statistically based studies prior to
> the works of Wilkinson and later, Lawson and Hanson, was garbage,
> because of poor numerics. This is a show stopper issue for Java in
> science and engineering.

Java is a whale of a lot closer than Fortran or C++ -- or are the sci/eng
folks all programming Borneo now <wink>?

754 provides some wonderful tools for true experts that Java doesn't yet
expose, but no tweaking of finite floating-point arithmetic rules is going
to save anyone from a numerically naive algorithm.  Directed rounding modes
and signed zeroes et alia don't do you any good unless you know how to use
them and apply them with extreme care -- and very few people do or can.  For
the rest, Java mandates IEEE nearest/even results, which is likely more than
what they're guaranteed today.

> If Java does not change, it should be rejected. Too bad. I am a Linux
> junkie and would love to see a reformed Java win and Microsoft lose.

MS has seemed to move away from the 754 spirit in its language
implementations rather than closer to it.  So if full access to all 754
features is what drives the market, MS will surely die within a few short
years <breath-holding wink>.

> I too am thrilled by NumPy, PIL and the other scientific packages that
> Hugunin, Hinsen, and Andrich have cooked up and/or packaged. The user
> community is as important as the language and Python is rich in both
> respects.

Yes it is!  But if Java's 754 story is a disaster, we'll need a new word for
Python's.

754-is-a-hw-std-that-almost-all-sw-still-ignores-ly y'rs  - tim





From gmcm at hypernet.com  Wed Jun  9 10:27:31 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 9 Jun 1999 09:27:31 -0500
Subject: Python/C API and setattr
In-Reply-To: <375E5257.8579541B@serop.abb.se>
Message-ID: <1283183791-72189810@hypernet.com>

Thomas S. Strinnhed writes:
> 
> I'm building an extention type for Python in C using NT, MSVC++.

> I've looked pretty much at Mark Lutz's example StackType  from
> Programming Python and I think I'm doing the same, but I've added
> setattrfunc to be able to directly manipulate value as c.value = 23
> and when I use it something goes wrong. (inc() and dec() work fine)
> 
> Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on
> win32 Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> from CounterType import *
> >>> c = Counter()
> >>> c.value = 23
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> SystemError: error return without exception set

You returned NULL (or 0, or the compiler generated a return of NULL 
because you didn't state anything) from your C code. That signifies 
an error to Python, which expects you to have set an error string.

Return (an increfed) None and Python will be happy.

> >>> c.value
> 23
> >>>
> The value is set properly but I can't figure out the system error,
> it seems to be something wrong with my setattrfunc. 
> 
> static void counter_setattr
>  (counterobject *self, char *name, PyObject *v)
> {
>  if(strcmp(name, "value") == 0)
>  {
>   self->value = (int)PyInt_AsLong(v);
>  }
> }
> 
> But what would that be??
> I have a hard time finding examples of how to do setattr so
> any pointers would be great. Also if someone understands this error,
> please post an answer.
> 
> My counterobject is this:
> typedef struct
> {
>  PyObject_HEAD
>  int value;
> } counterobject;
> 
> Thanks 
>  -- Thomas S. Strinnhed, thstr at serop.abb.se
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list

- Gordon



From skip at mojam.com  Tue Jun  8 08:59:55 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 08 Jun 1999 12:59:55 GMT
Subject: whichdb module fails to recognize Berkeley DB v2 hash files
Message-ID: <375D1457.D97D365B@mojam.com>

I guess in 1.5.2 a new module, whichdb, was added that attempts to
divine the nature of a database file.  This module doesn't know anything
about Berkeley DB v2 files.  In v2, Sleepycat added a 12-byte null pad
in front of the old magic numbers (at least for hash and btree files). 
I've been using v2 for awhile and upgrading to 1.5.2 broke all my
anydbm.open calls. I believe the following patch corrects the problem.

*** whichdb.py.orig	Wed Jun  2 13:27:54 1999
--- whichdb.py	Tue Jun  8 08:59:59 1999
***************
*** 32,38 ****
          return None
  
!     # Read the first 4 bytes of the file -- the magic number
!     s = f.read(4)
      f.close()
  
      # Return "" if not at least 4 bytes
--- 32,39 ----
          return None
  
!     # Read the start of the file -- the magic number
!     s16 = f.read(16)
      f.close()
+     s = s16[0:4]
  
      # Return "" if not at least 4 bytes
***************
*** 49,52 ****
--- 50,63 ----
      if magic == 0x13579ace:
          return "gdbm"
+ 
+     # Check for BSD hash
+     if magic in (0x00061561, 0x61150600):
+         return "dbhash"
+ 
+     # BSD hash v2 has a 12-byte NULL pad in front of the file type
+     try:
+ 	(magic,) = struct.unpack("=l", s16[-4:])
+     except struct.error:
+         return ""
  
      # Check for BSD hash

-- 
Skip Montanaro	| Mojam: "Uniting the World of Music"
http://www.mojam.com/
skip at mojam.com  | Musi-Cal: http://www.musi-cal.com/
518-372-5583



From mnot at pobox.com  Tue Jun 22 06:29:23 1999
From: mnot at pobox.com (Mark Nottingham)
Date: Tue, 22 Jun 1999 10:29:23 GMT
Subject: htmllib: CR in CDATA
References: <00d201bebc5a$adb129a0$0301a8c0@mnot.net> <012601bebc91$f3ded760$0301a8c0@mnot.net>
Message-ID: <014401bebc9a$1963a9e0$0301a8c0@mnot.net>

OK, I'm starting to have a really nice conversation with myself now ;-)

htmllib DOESN'T change the newline to a single space - it leaves it in.

CDATA is a sequence of characters from the document character set and may
include character entities. User agents should interpret attribute values as
follows:
          Replace character entities with characters,
          Ignore line feeds,
          Replace each carriage return or tab with a single space.
     User agents may ignore leading and trailing white space in CDATA
attribute values (e.g., "   myval   " may be interpreted as "myval").
Authors should not declare attribute values with leading or trailing white
space.

If I'm wrong this time, someone please save us both the trouble and shoot
me.

Example:

#!/opt/local/bin/python
body = '''\
<HTML>
<A HREF="image.
jpg">
</HTML>
'''
import formatter, htmllib, sys, string
parser = htmllib.HTMLParser(formatter.NullFormatter())
parser.feed(body)
print parser.anchorlist



----- Original Message -----
From: Mark Nottingham <mnot at pobox.com>
To: Python <python-list at cwi.nl>
Sent: Tuesday, June 22, 1999 7:31
Subject: Re: htmllib: CR in CDATA


> Whooops, nevermind, I misread the spec -- carriage returns are turned into
> spaces (which is what htmllib does) - *line feeds* should be ignored...
>
> --
> "Get me the phone book."
>   "Which one?"
> "Doesn't matter."
>
>
> ----- Original Message -----
> From: Mark Nottingham <mnot at pobox.com>
> To: Python <python-list at cwi.nl>
> Sent: Tuesday, June 22, 1999 12:55
> Subject: htmllib: CR in CDATA
>
>
> > It appears that htmllib doesn't ignore returns in CDATA fields, as HTML
> 4.0
> > says it should:
> > http://www.w3.org/TR/REC-html40/types.html#type-cdata
> > http://www.w3.org/TR/REC-html40/sgml/dtd.html
> >
> > As a result, htmllib improperly parses any CDATA element that wraps
across
> a
> > line; this affects elements like
> >
> > <A href="foo.
> > gif">
> >
> > I'm happy to work up a patch, but I thought I'd ask around first. It may
> be
> > a bit involved to fix it properly; every CDATA should be handled this
way,
> > which practically means almost every tag attribute.
> >
> > Regards,
> >
> >
> > Mark Nottingham, Melbourne Australia
> > mnot at pobox.com  http://www.mnot.net/
> >
> >
> >
>
>




From matomira at iname.com  Thu Jun 10 12:47:13 1999
From: matomira at iname.com (Fernando Mato Mira)
Date: Thu, 10 Jun 1999 18:47:13 +0200
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <375FD703.C6D@mailserver.hursley.ibm.com>
Message-ID: <375FEC11.FD20EDE6@iname.com>


Paul Duffin wrote:

> Whether the choice of language is a good or bad idea depends entirely on
> the context. i.e. environment and job to be done.
>
> You cannot avoid taking into account the context even if you only look
> at the language from a semantic / syntactic view point.

OK. Let's drop the word `choice'. Why do people think tcl is a bad
language then, strictly from a _design_ viewpoint? Let's assume
imperative languages are OK (responses like "it's not declarative"
are obvious and uninteresting and do not help comparing it to other
imperative languages).






From gmcm at hypernet.com  Wed Jun  2 18:05:01 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 2 Jun 1999 17:05:01 -0500
Subject: Defining VCL-like framework for Python
In-Reply-To: <375344AA.322F50E1@compaq.com>
Message-ID: <1283761146-37459666@hypernet.com>

Greg Ewing wrote:

> Gordon McMillan wrote:
> > 
> > And where your ideas
> > don't fit the framework, you're in a prison.
> 
> I think it's unfair to characterise the general concept
> of a framework as being inflexible. Some particular
> frameworks might be, but I don't see why a framework
> can't be designed to help with the most common things
> while allowing the freedom to do others.
> 
> In fact, when the framework is implemented in a user-
> friendly OO language like Python, it would be hard *not* 
> to make it flexible!

I think my statement was fair. I believe I said some frameworks are 
more flexible than others. But no framework covers all possible 
domains (except the null framework) and no framework is useful 
outside its domain.

FWIW, I like frameworks, but I'm not going to bother learning one 
unless I'll be spending some time inside its domain.

- Gordon



From amackay at starvision.com  Thu Jun 17 17:33:00 1999
From: amackay at starvision.com (Angus MacKay)
Date: Thu, 17 Jun 1999 14:33:00 -0700
Subject: sleep()ing
Message-ID: <3769698C.FD19B200@starvision.com>

is there anyway to sleep in a python program asside from:
import os
os.system("sleep 1")

or

import threading
threading._sleep(1)

I mean it is part of unistd.h so you would think that it would
appear in os or posix or somewhere.

cheers, Angus.



From emile at fenx.com  Mon Jun  7 10:24:18 1999
From: emile at fenx.com (Emile van Sebille)
Date: Mon, 7 Jun 1999 07:24:18 -0700
Subject: Indentation and grouping proposal
References: <000001beb0a1$bc540a00$5b9e2299@tim> <375BCAD1.BE6E391B@erols.com>
Message-ID: <iCQ63.1184$VL2.175421@news.direcpc.com>

Re: One Liners:

>>> def myexec(b):
...  for stmnt in string.split(b,'\n'):
...   exec(stmnt)
...
>>> b="if a: print 'Spam!';print 'Spam2!';\nprint'Yum!'"
>>> myexec(b)
Yum!
>>> a=1
>>> myexec(b)
Spam!
Spam2!
Yum!
>>>
--

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







From fredrik at pythonware.com  Mon Jun 28 09:51:41 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 28 Jun 1999 13:51:41 GMT
Subject: strange behaviour at termination time
References: <377765F0.38230773@steria.fr>
Message-ID: <00e601bec16d$5a65f560$f29b12c2@secret.pythonware.com>

Fabien COUTANT <fabien.coutant at steria.fr> wrote:
> Basically there is a base class name that I found bound to None,
> which prevents me to call base methods.

also see:
http://www.python.org/doc/essays/cleanup.html
(this *was* implemented in 1.5.1)

</F>




From hniksic at srce.hr  Fri Jun  4 06:01:04 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 04 Jun 1999 12:01:04 +0200
Subject: "The Python Way"
References: <001101beae49$39af7bc0$949e2299@tim>
Message-ID: <87hfoos2sf.fsf@pc-hrvoje.srce.hr>

"Tim Peters" <tim_one at email.msn.com> writes:

> It's just old.  python.org has archives of the pre-c.l.py days, and
> you'll find many of the old-timers arguing vigorously about all the
> "current issues" 8 years ago already.  There have been few notable
> advances in the art of GC-- let alone whitespace --since then, and
> repetition is tiring.

Actually, the GC discussion is way older than eight years, and it's
very sad that Python chose the pessimal strategy.  :-(



From arcege at shore.net  Tue Jun 15 14:16:55 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 15 Jun 1999 18:16:55 GMT
Subject: Python and C, arrays
References: <7jrqs5$otb$1@nnrp1.deja.com>
Message-ID: <rMw93.5785$nn.1055711@news.shore.net>

John Fisher <jfisher at are.berkeley.edu> wrote:
: Hey folks,

: I've run into a problem with a module I'm writing in C.  Basically, I
: need to be able to convert Python tuples or lists (it's unimportant
: which) into C arrays.  I know at compile-time the type of the array (and
: that the tuple or list will be homogeneous, and of compatible type), but
: I don't know the size.

: I then need to find a way to take a C array and stuff it back into a
: Python tuple or list.  I have some ideas involving using "stack-like"
: lists or tuples in Python, with a "pop" function which I could call from
: C, repeated in a loop to get everything.  But this seems overly complex,
: and indeed like it has the potential to become truly hideous.

: I would imagine these two tasks are actually quite common.  Is there
: some simpler way to do either?  Dare I hope that PyArg_ParseTuple and
: Py_BuildValue can do it?

Use the Abstract Layer's PySequence_*() functions in the C API
 (http://www.python.org/doc/current/api/sequence.html)

PyObject_Length() will give you the number of elements in a tuple or
list.  And from the other direction (if you do not know the size of the
array), use (sizeof(array)/sizeof(array[0])).

Here are some specific converters for you (from a Python sequence to a
C array of PyObject* and back).

  int convert_PySeq2C(seq, array)
    PyObject *seq;
    PyObject ***array;
    { int i, n;
      PyObject *tmp;

      n = PyObject_Length(seq);
      *array = (PyObject *)malloc(n * sizeof(PyObject *));
      for (i = 0; i < n; i++) {
        tmp = PySequence_GetItem(seq, i);  /* not a 'borrowed' reference */
        if (tmp == NULL) {
          free(*array);
          *array = NULL;
          return -1; /* exception set already */
        }
        Py_INCREF(tmp);
        (*array)[i] = tmp;
      }
      return n;
    }

Converting back to Python is a little simplier:

  PyObject *convert_PySeq2C(array, n)
    PyObject *array[];
    int n;
    { int i;
      PyObject *result;

      result = PyTuple_New(n);
      if (result != NULL) {
        for (i=0; i < n; i++)
          PyTuple_SetItem(result, i, array[i]); /* increments the ref count */
      }
      return result;
    }

There will be other, more efficient, means that would be tailored to
your app.

  -Arcege



From holger at phoenix-edv.netzservice.de  Mon Jun 14 12:46:11 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Mon, 14 Jun 1999 16:46:11 GMT
Subject: Newbie: easy filecopy?
Message-ID: <376531D3.7E42AC42@phoenix-edv.netzservice.de>

Sorry,
it's me again.;-)

I couldn't believe it. I think there must be an easy copy-function
in python (I only want to copy files under ext2- or FAT-FileSystems).
Call me "4 blind mice"-Joey, but I couldn't find it. 
For sure, I could write a thing like open-read-write-close, but I
think, someone else has done already, hasn't he?

Thanx again for patience with me...[snuff]

Ciao,
Holger



From chad at vision.arc.nasa.gov  Sat Jun 19 08:25:53 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Sat, 19 Jun 1999 05:25:53 -0700
Subject: An Un-optimization anecdote with text-processing
References: <376B1AAC.19FE8BCE@mysolution.com> <d%Ga3.498$dR.35932@newsfeed.avtel.net>
Message-ID: <376B8C51.5D59C774@vision.arc.nasa.gov>

Phil Mayes wrote:

> A raw email always has a blank line between the header and the body.
> So you can read it in and find the gap by looking for 2 EOLs:
>
> [Code snipped]
> --
> Phil Mayes    pmayes AT olivebr DOT com

Well, I was in the process of starting to write a Perl program
for a simple text processing task I have to do, when this post
steered me back to my language of choice.  As I wrote my script,
which is called by the Unix "find" command, I got to thinking
about how it was anti-optimized...  Here is what I wrote:

import string
import sys

# Way stupid algorithm, but I'm running it overnight :)
filename = sys.argv[1]
f = open(filename, 'r')
text = f.read()
while 1:
    i = string.find(text, r'/"')
    if i == -1: break
    text = text[:i+1] + "index.html" + text[i+1:]

f.close()
f = open(filename, 'w')
f.write(text)
f.close()


As you can see, the core loop scans the entire file for a string
match, builds a new string by concatenation, then starts scanning from
the beginning yet again.  It almost doesn't seem like it could be made
much worse, unless one deliberately tried.

Which brings me to my new game...  Rather than finding a nice hack
that would speed things up and use less lines of code, how about
finding the hack that uses the least amount of code to make this as
slow as possible?  Adding pointless operations, or other obfuscations
don't count.  Try to make this operation as elegantly short, and as
painfully slow as possible...

For example:

I had originally coded this with string.index() using a try/except
pair to break out of the loop, which took more lines of code but is
probably faster than my loop (it avoids the 'if' statement in the
loop, for the overhead of one exception).  So I changed it to be
shorter and (possibly) slower.  Can anyone think of other elegant
unoptimizations?

Chad Netzer

PS.  I'm not suggesting running elaborate benchmarks or anything, just
hoping to hear some thoughts from the real Python masters.  It's
Saturday morning, 5:18 AM, and I'm STILL at work, so that might
explain the loopy idea.





From gmcm at hypernet.com  Wed Jun 16 00:38:32 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 15 Jun 1999 23:38:32 -0500
Subject: Newbie: easy filecopy?
In-Reply-To: <37665D08.267A697C@phoenix-edv.netzservice.de>
Message-ID: <1282614322-106445528@hypernet.com>

Holger Jannsen writes:

> Another ascpect to my question of yesterday:
> 
> The problem with 
> f=open(aFile)
> x=f.read()
> f.close
> f.open(toFile)
> f.write(x)
> f.close
> is, that some special ASCII-codes are not copied as usual 
> under Windows-systems, so e.g. CarriageReturn (tested with normal
> Asccii-text).

You're doing something funny, but not sure what from your pseudo 
code.

open(aFile) is equivalent to open(aFile, 'r'), which is text mode 
- it will translate \r\n to \n on input. open(toFile,'w') would also 
open text mode, and translate \n to \r\n on ouput. Your symptoms 
match reading from a file in text mode, but writing in binary mode 
(eg, open(toFile, 'wb')).

Don't worry - you're in good company. Netscape gets them confused 
frequently too.

- Gordon



From dellaq2 at pdq.net  Tue Jun  8 00:20:14 1999
From: dellaq2 at pdq.net (John Dell'Aquila)
Date: Mon, 07 Jun 1999 23:20:14 -0500
Subject: Extreme Python Testing
Message-ID: <375C99FE.E61E511@pdq.net>

I admire the Smalltalk "Extreme Programming" methodology enough that
I'm applying it in my Python development. XP is impossible without an
automated test framework, so the first thing I did was write a Python
version of Kent Beck's Testing Framework.

Shortly thereafter, I saw Tim Peter's brilliant doctest.py and
realized there was no need to write separate TestCase classes. With
doctest, the test case coulde *be* the documentation, inseparably
linked to the object being exercised. After a lot of cutting and
twisting, a very simple framework emerged.

I like the result well enough that I sometimes catch myself doing the
Right Thing - writing test cases *before* starting to code. I attach
xp.py in the hope that others may find it useful and I offer apologies
in advance if my abuse of docStrings treads on *your* abuse.

John

# Module xp
# John Dell'Aquila <jbd at alum.mit.edu>
# 6 June 1999

# Public domain

# Usual disclaimers apply: "as is", "use at your own risk",
# "no warranty of fitness or merchantability", . . .

""" Module xp - Extreme Python Test Framework.
Inspired by Kent Beck's Testing Framework (see Extreme Programming web
pages at http://c2.com/cgi/wiki?ExtremeProgrammingRoadmap) and derived
from Tim Peter's doctest.py (comp.lang.python archives for March,
1999).

xp.Test examines ALL docStrings in specified modules, classes, methods
or functions and *executes* any lines having the Python interpreter
prompt (>>>) as their first non-whitespace.  DocStrings are classified
as passed (no errors), failed (exception raised) or noTest (no
docString or no code within docString).

The general idea is to create test objects and then verify their
correct behavior with 'assert' statements. Multi-line Python
statements are explicitly NOT supported; complex tests must be
implemented as helper functions/methods in the code being
tested. Exception testing IS supported (see TestResult.runDocString)
and so are unqualified imports ('from foo import *' will NOT cause
foo's objects to be tested).

See the docStrings in this file for examples.

Usage:
    import xp
    tst = xp.Test()           # create Test harness
    tst.check(myModule)       # check some stuff
    tst.check(moduleX.Class1).check(moduleX.Class2)
     . . .
    print tst.result          # print summary
    print tst.result.errors() # display formatted error messages
    print tst.result.passed   # list objects that passed tests
    print tst.result.noTest   # list objects that had no tests
    print tst.result.failed   # list raw error tuples

Sample test summary:
<TestResult: 10 passed, 1 failed, 0 noTest, 0.03 sec>

Sample error message:
-------------------------------------------------
DocString:   FUNCTION:__main__.foo
Statement:   assert 1==2
Exception:   AssertionError:

>>> None  # stops noTest snivels, i.e this space intentionally blank
"""

__version__ = '1.0'

import sys, types, time, re, string, traceback

_PS1 = re.compile(r'\s*>>>\s*').match
_summaryFmt = (
    "<TestResult: %(passed)d passed, %(failed)d failed, "
                 "%(noTest)d noTest, %(elapsed)g sec>"
    )
_errorFmt = (
    "-------------------------------------------------\n"
    "DocString:   %s\n"
    "Statement:   %s\n"
    "Exception:   %s\n"
    )

_Module = types.ModuleType
_Class = types.ClassType
_Method = types.MethodType
_Func = types.FunctionType
_Dict = types.DictionaryType
_None = types.NoneType
_Long = types.LongType
_BuiltIn = types.BuiltinFunctionType
_String = types.StringType

_typeAbbrev = {
    _Method: 'METHOD',
    _Long: 'LONG',
    _BuiltIn: 'BUILTIN'
    }

_expect_Fn = """
def _expect_(stmt, exc):
    try:
        exec stmt
    except exc:
        return
    else:
        raise '_expect_', '%s not raised' % exc.__name__
"""

def moduleName(obj):
    """ Return object's module name, if known, otherwise None.

    >>> assert moduleName(re) == 're'
    >>> assert moduleName(re.compile) == 're'
    >>> assert moduleName(re.RegexObject) == 're'
    >>> assert moduleName(re.RegexObject.match) == 're'
    >>> assert moduleName('foo') == None
    """
    if type(obj) == _Module:
        return obj.__name__
    if type(obj) == _Class:
        return obj.__module__
    if type(obj) == _Method:
        return  obj.im_class.__module__
    if type(obj) == _Func:
        return obj.func_globals['__name__']
    return None

def objRepr(obj):
    """ Return printable representation of object.

    >>> assert objRepr(re) == 'MODULE:re'
    >>> assert objRepr(re.RegexObject) == 'CLASS:re.RegexObject'
    >>> assert objRepr(Test.check) == 'METHOD:%s.Test.check' % __name__
    >>> assert objRepr(objRepr) == 'FUNCTION:%s.objRepr' % __name__
    >>> assert objRepr('foo') == 'STRING:'
    >>> assert objRepr(3L) == 'LONG:'
    >>> assert objRepr(string.lower) == 'BUILTIN:lower'
    """
    typ = _typeAbbrev.get( type(obj),
                           string.upper(type(obj).__name__) )
    if type(obj) == _Module:
        return '%s:%s' % (typ, obj.__name__)
    if type(obj) == _Class:
        return '%s:%s.%s' % (typ, obj.__module__, obj.__name__)
    if type(obj) == _Method:
        return '%s:%s.%s' % (typ, str(obj.im_class), obj.__name__)
    if type(obj) == _Func:
        return '%s:%s.%s' % (typ, obj.func_globals['__name__'],
obj.__name__)
    if hasattr(obj, '__name__'):
        return '%s:%s' % (typ, obj.__name__)
    return '%s:' % typ


class TestResult:
    """ Accumulate test results. Helper class to Test.
    >>> None
    """

    def __init__(self):
        """ >>> None """
        self.start = self.stop = 0
        self.passed = []
        self.failed = []
        self.noTest = []

    def runDocString(self, obj, env):
        """ Execute obj.__doc__ in a *shallow* copy of the
        environment dictionary, env. If possible, insert
        _expect_ function into environment to allow exception
        testing. Tally results and timestamp.

        >>> tr = TestResult()  # check for *expected* exception
        >>> _expect_('tr.runDocString(3, {})', AssertionError)
        """
        assert type(obj) in (_Module, _Class, _Method, _Func)

        if not self.start:
            self.start = time.time()
        name = objRepr(obj)
        globs = env.copy()
        if not globs.has_key('_expect_'):
            exec _expect_Fn in globs

        status = None
        for line in string.split(obj.__doc__ or '', '\n'):
            matchPS1 = _PS1(line)
            stmt = matchPS1 and line[matchPS1.end(0):]
            if not stmt:
                continue
            try:
                exec stmt in globs
                status = 1
            except:
                etype, evalue = sys.exc_info()[:2]
                edesc = traceback.format_exception_only(etype, evalue)
                edesc = string.strip(string.join(edesc, ''))
                status = 0
                break

        if status == None:
            self.noTest.append(name)
        elif status == 0:
            self.failed.append((name, stmt, edesc))
        elif status == 1:
            self.passed.append(name)
        self.stop = time.time()

    def __repr__(self):
        """ Display summary results.
        >>> None
        """
        vars = { 'passed': len(self.passed),
                 'failed': len(self.failed),
                 'noTest': len(self.noTest),
                 'start': time.ctime(self.start),
                 'stop': time.ctime(self.stop),
                 'elapsed' : self.stop-self.start }
        return _summaryFmt % vars

    def errors(self):
        """ Return error details as printable string.
        >>> None
        """
        return string.join(map(lambda x, f=_errorFmt: f%x, self.failed),
'')


class Test:
    """ DocString tester.
    >>> None
    """

    def __init__(self):
        """ >>> None """
        self.result = TestResult()

    def check(self, obj, env=None):
        """ Recursively test docStrings of object and any subobjects
        (module or class members). Import safe - does not cross module
        boundaries (use successive calls to check other modules).
        For convenience, Strings are taken as module names, i.e.
        check(__name__) tests the current module.

        DocStrings execute in env dictionary, if supplied, otherwise
        in the appropriate module namespace. A *shallow* copy
        is made for each docString, so temporary variables may be
        used freely, but changing mutable objects can 'pollute'
        environment of succeeding tests.

        >>> None
        """
        assert type(obj) in (_Module, _Class, _Method, _Func, _String)

        if type(obj) == _String:
            thisModuleName, obj = obj, sys.modules[obj]
        else:
            thisModuleName = moduleName(obj)
        if not env:
            env = sys.modules[thisModuleName].__dict__

        self.result.runDocString(obj, env)

        if type(obj) not in (_Module, _Class):
            return self
        for name in obj.__dict__.keys():
            nxtObj = getattr(obj, name) # NOT obj.__dict__[name] !!!
            if type(nxtObj) not in (_Class, _Method, _Func):
                continue
            if moduleName(nxtObj) == thisModuleName:
                self.check(nxtObj, env)
        return self

if __name__ == '__main__':
    tst = Test().check('__main__')
    print tst.result
    print tst.result.errors()





From cvlcareer at yahoo.com  Fri Jun 11 13:08:36 1999
From: cvlcareer at yahoo.com (lokesh ceeba)
Date: Fri, 11 Jun 1999 10:08:36 -0700 (PDT)
Subject: mxODBC and MySQL
Message-ID: <19990611170837.18906.rocketmail@web1006.mail.yahoo.com>

Hello Everybody,
     I've installed Mysql on Windows-NT 4.0.
Used those commands to NET START MYSQL.
But i donot see any Mysql Prompt.
Anybody would be able to help me out in running Mysql successfuly. I
thank you in advance for your time.
At the sql prompt i should be able to run Sql queries.

Sincerely
Thx,
Loki

--- Joachim Schmitz <perry at ac-copy.net> wrote:
> 
> Hi,
> 
> I compiled and installed  mxODBC for MySQL on SuSE
> 6.1 without problems.
> I get this:
> Python 1.5.2b2 (#6, Feb 28 1999, 21:41:19)  [GCC
> egcs-2.91.60 19981201
> (egcs-1.1
> .1  on linux2
> Copyright 1991-1995 Stichting Mathematisch Centrum,
> Amsterdam
> >>> import ODBC.MySQL
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File
> "/usr/local/lib/python1.5/ODBC/MySQL/__init__.py",
> line 7, in ?
>     from mxODBC import *
> ImportError:
> /usr/local/lib/python1.5/ODBC/MySQL/mxODBC.so:
> undefined
> symbol: PyBuffer_Type
> >>
> 
> regards 
> 
> Joachim Schmitz
> 
>
------------------------------------------------------------------------------
>  P R O F I . N E T  A G    *    Joachim Schmitz   *
> Senior Network Consultant
>
------------------------------------------------------------------------------
>   profi.net                |    Aindlinger Str. 7  
> | Voice  +49 821 27062-0
>   Hassler & Mair AG        |    86167 Augsburg     
> | Fax    +49 821 27062-99
>                            |    Germany            
> | Email  js at profi.net  
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 

_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




From nobody at newsfeeds.com  Tue Jun 22 17:40:20 1999
From: nobody at newsfeeds.com (Anonymous)
Date: Tue, 22 Jun 1999 17:40:20 -0400
Subject: Thanks for Python...
References: <376ead64@anonymous.newsfeeds.com> <7knb7p$kas$2@news.udel.edu>
Message-ID: <377000f1@anonymous.newsfeeds.com>

Terry Reedy <tjreedy at udel.edu> wrote in message
news:7knb7p$kas$2 at news.udel.edu...
> In article <376ead64 at anonymous.newsfeeds.com>,
>
> John Brainard posting anonymously as nobody at newsfeeds.com says...
>
> >write a mach-server that emulated the real server. I was successful at
>
> >mach server that I had originally written in Perl and it works much
>
> and >I hope that my boss is impressed.
>
> To impress your boss more, use 'mock' (imitation) instead of 'mach'
> (speed number) when you write your report.
>
> TJR (With a real email address, I would have sent this privately.)

I apologise for posting anonymously to this group. I had no idea that
newsfeeds would hide my e-mail address when I posted. I have nothing to hide
from this news group, or any other that I read from and will change my sig
file to include that and my icq number.

As for the misspelling, I am truly embarassed by it. I am normally quite
about checking it.



  --------== Posted Anonymously via Newsfeeds.Com ==-------
     Featuring the worlds only Anonymous Usenet Server
    -----------== http://www.newsfeeds.com ==----------



From amitp at Xenon.Stanford.EDU  Fri Jun 18 22:32:54 1999
From: amitp at Xenon.Stanford.EDU (Amit Patel)
Date: 19 Jun 1999 02:32:54 GMT
Subject: NameError Problem
References: <7kels4$9tv$1@eskinews.eskimo.com>
Message-ID: <7kevgm$8f5$1@nntp.Stanford.EDU>

 Jonathon <jblake at eskimo.com> wrote:
| 
| 	I'm writing a long << 25K line >> script to verify data from a
|  	database.
| 		
| 	The following bit of script is giving me a NameError, and
| 	I can't figure out why.
| 
| 	datafile = open ( the_data_file_is, "r" )
| 	new_datafile = open ( the_temp_data_file_is, "w" )
| 	line_kounter = 0
| 	while line_kounter < 50000
| 	    line_kounter = line_kounter + 1
| 	    templine = datafile.readline(line_kounter)
|             validate_line = templine
| 	    if ( the_file_is == "john_www.db" ) :
| 	         temp_record_id = int(validate_line[2:18])
| 		 temp_country  = strip(validate_line[20:50])
|             elif ( the_file_is == "jan_www.db " ):
|                  temp_record_id = int(validate_line[2:10])
|                  temp_country = strip(validate_line[12:20])
|             print temp_record_id

You have an 'if' and then some 'elif's.  However, if the_file_is does
not match any of the strings, then temp_record_id does not get set,
and you get a NameError.  That's probably what's happening.

One thing to check:                            |
                                               v
|             elif ( the_file_is == "jan_www.db " ):
                                               ^
                                               |
                                  Did you mean to put this space here?


    - Amit



From mgutierrez at dss.es  Thu Jun 17 11:12:14 1999
From: mgutierrez at dss.es (mgutierrez at dss.es)
Date: Thu, 17 Jun 1999 15:12:14 GMT
Subject: Starting a project: any hints?
References: <19990617154643.A1680@optiplex.palga.uucp>
Message-ID: <7kb384$lgs$1@nnrp1.deja.com>

In article <19990617154643.A1680 at optiplex.palga.uucp>,
  Gerrit Holl <gerrit.holl at pobox.com> wrote:
> Hello python-aholics,
>
> I'm a 13-years-old-boy and I've about half a year programming
experience. And
>
[snip]
I started a similar project called Lritaunas Peki , currently this
project is quite ''frozen'' ( or dead) : www.ctv.es/USERS/irmina

I have started nearly a dozen of small (freeware) projects and I
can give you some pieces of advice:
- Make your program works from the very first day , it doesn't matter
if it barely works.
- Don't be ambitious, be humble and try easier/simpler things always.
- Involve other people in the project, keeping an open mind and a lot
of patience!
- Post a lot about it in newsgroups and freshmeat.
- Work a lot in your project, if you don't nobody will it do for you!
- Use python, it's by far the best thing you can use: objects and
readibility.

I wish you luck, boy.
And remember if you don't succeed at the first time,... next will be
Manolo



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From genius at acme.ca  Thu Jun 10 11:13:36 1999
From: genius at acme.ca (Snoopy :-)))
Date: Thu, 10 Jun 1999 11:13:36 -0400
Subject: Free Sex Links  7546
References: <3CN73.93097$pl3.57568@newreader.ukcore.bt.net>
Message-ID: <375fd775@nemo.idirect.com>

F#@*^ OFF ;-((

kvrekp at hotmail.com wrote in message
<3CN73.93097$pl3.57568 at newreader.ukcore.bt.net>...
>For Nasty Sex links visit:
>
>http://website.lineone.net/~simonsi/sexlinks.htm
>ztjorzgfvo
>





From kiki at pixar.com  Thu Jun 24 00:33:50 1999
From: kiki at pixar.com (kiki)
Date: Wed, 23 Jun 1999 21:33:50 -0700
Subject: Tkinter: radiobutton
Message-ID: <3771B52E.80F3FA46@pixar.com>

Newbie speaks: I'm sure there's an elegant solution to this, but I just
can't find it, and my head hurts from reading man pages...

I have a color menu that has "Add Color" and "Change Color", and an
appended list of user-created colors as radiobuttons.

What I would like to do is when the user chooses "Change Color", the new
color overwrites the current color selected in the radiobutton list.

The problem is I don't know how to find out which button is selected.
self.color_menu.entryconfig("active"...) configures the "Change Color"
button!!  Arg!  The index is otherwise implicit...

A group of radiobuttons are associated by the fact that they share a
variable.  When that variable changes, the states of the radio buttons
change.  Certainly there must be a way to find out the index of the
selected radiobutton in a group of radiobuttons and modify the contents
[eg "value", "activebackground", etc.] of it...

Thanx, and sorry if this was an obvious question.

Kiki




From donn at u.washington.edu  Wed Jun  2 11:28:15 1999
From: donn at u.washington.edu (Donn Cave)
Date: 2 Jun 1999 15:28:15 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>  <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>  <87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net>  <87hfosquc0.fsf@ev.netlab.co.jp> <3753F121.E4A9BE8B@prescod.net> <87yai3ntdt.fsf@ev.netlab.co.jp>
Message-ID: <7j3iif$134e$1@nntp6.u.washington.edu>

Yukihiro Matsumoto <matz at netlab.co.jp> writes:
...
| I don't mean real GC does everything ref counting does.  For
| exapmle, with real GC:
|
|   dialog = GUItoolkit.dialog(params)
|   ... do something with the dialog ...
|   dialog = None
|
| will close the dialog at unpredictable timing, which is not
| desirable for everyone.  But I prefer explicit termination of the
| dialog for this case.
|
|   dialog = GUItoolkit.dialog(params)
|   ... do something with the dialog ...
|   dialog.close

You're in good company here, I think the leading lights of the Python
world all prefer explicit termination in these cases.  Ref counting sure
offers a great convenience here, though.  Files close and flush their
buffers, dialogues disappear, all in accordance with their programmatic
lifetime.  I personally will miss this finalization more than any resource
management if reference counting is taken away.

I wonder if reference counts could be maintained at a higher level,
still internal but optional as a complement to GC or whatever.  The
C extension writer could forget about reference counts, unless the
extension's own objects use the option or its functions handle external
objects that might have reference counts - or perhaps a C extension
could choose to use reference counts when its external library routines
have memory allocation issues that can't be integrated with the GC.

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



From a.mueller at icrf.icnet.uk  Thu Jun  3 13:50:48 1999
From: a.mueller at icrf.icnet.uk (Arne Mueller)
Date: Thu, 03 Jun 1999 18:50:48 +0100
Subject: threads
Message-ID: <3756C078.5D998D3B@icrf.icnet.uk>

Hi All,

I'm using the threading module in python 1.5.2b2. I'm wondering wheather
the different threads are all executed in parallel on a multiprocessor
machine.

I'm working on an SGI Power Challange with 12 processors running irix
6.5 and the pthreads library. I call 4 Thread objects and start each
immedatelty after creation. In the 'top' list of running processes my
program is reported to occupy 200 % cpu and I wonder how I can manage to
make it using ~400 % cpu.

I'd be greatful for any suggestions, maybe some general hints about
threading etc. Maybe it's a matter of the os or the appropiate pthreads
implementation. 

	Greetings,

	Arne
-- 
Arne Mueller
Biomolecular Modelling Laboratory
Imperial Cancer Research Fund
44 Lincoln's Inn Fields
London WC2A 3PX, U.K.
phone : +44-(0)171 2693405      | Fax : +44-(0)171 269 3258
email : a.mueller at icrf.icnet.uk | http://www.icnet.uk/bmm/



From fredrik at pythonware.com  Wed Jun 16 07:02:07 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 16 Jun 1999 11:02:07 GMT
Subject: Tk Scrollbar
References: <37663C34.2957622F@dlr.de>
Message-ID: <01ae01beb7e7$ae8e43c0$f29b12c2@pythonware.com>

> I have only an easy question. I think it is possible to bind a scrollbar
> and an other widget e.g. a frame with buttons. But I have not found any
> literature or programming examples. I want to bind the scrollbar to an
> frame and I don't know how. Please help me.

you cannot.  period.

but you can put the frame in a canvas, and
add scrollbars to the canvas.    the following
example does this (and more).

</F>

#
# tk045 -- scrolled frame with autohiding scrollbars
#
# fredrik lundh, august 1998
#
# fredrik at pythonware.com
# http://www.pythonware.com
#

from Tkinter import *

class AutoScrollbar(Scrollbar):
    # a scrollbar that hides itself if it's not needed.  only
    # works if you use the grid geometry manager.
    def set(self, lo, hi):
        if float(lo) <= 0.0 and float(hi) >= 1.0:
            # grid_remove is currently missing from Tkinter!
            self.tk.call("grid", "remove", self)
        else:
            self.grid()
        Scrollbar.set(self, lo, hi)
    def pack(self, **kw):
        raise TclError, "cannot use pack with this widget"
    def place(self, **kw):
        raise TclError, "cannot use place with this widget"

#
# create scrolled canvas

root = Tk()

vscrollbar = AutoScrollbar(root)
vscrollbar.grid(row=0, column=1, sticky=N+S)
hscrollbar = AutoScrollbar(root, orient=HORIZONTAL)
hscrollbar.grid(row=1, column=0, sticky=E+W)

canvas = Canvas(root, 
                yscrollcommand=vscrollbar.set, 
                xscrollcommand=hscrollbar.set)
canvas.grid(row=0, column=0, sticky=N+S+E+W)

vscrollbar.config(command=canvas.yview)
hscrollbar.config(command=canvas.xview)

# make the canvas expandable
root.grid_rowconfigure(0, weight=1)
root.grid_columnconfigure(0, weight=1)

#
# create canvas contents

frame = Frame(canvas)
frame.rowconfigure(1, weight=1)
frame.columnconfigure(1, weight=1)

rows = 5 
for i in range(1,rows):
    for j in range(1,10):
        button = Button(frame, padx=7, pady=7, text="[%d,%d]" % (i,j))
        button.grid(row=i, column=j, sticky='news')

canvas.create_window(0, 0, anchor=NW, window=frame)

frame.update_idletasks()

canvas.config(scrollregion=canvas.bbox("all"))

root.mainloop()




From Fred.L..Drake at p98.f112.n480.z2.fidonet.org  Tue Jun 29 15:10:33 1999
From: Fred.L..Drake at p98.f112.n480.z2.fidonet.org (Fred L. Drake)
Date: Tue, 29 Jun 1999 20:10:33 +0100
Subject: Binary pickle portability
Message-ID: <00001589@bossar.com.pl>

From: "Fred L. Drake" <fdrake at cnri.reston.va.us>


Olaf Delgado writes:
 > After looking a bit closer, I found that the problem is probably related
 > to the Numeric module (i.e., pickling matrices) and doesn't depend on
 > whether binary or ordinary mode is used. So please don't bother now!
 > Still, I am interested in whether people use binary pickles.

  The last release of Grail uses binary pickles to store caches of
parsed bookmark files.  There are no floating point or NumPy objects
in the pickles, so that wouldn't test cases you may be interested in.
My choice to use binary pickles was entirely based on the performance
improvement; I doubt there are any substantial cross-platform issues
for bookmark files, especially when only used as caches!


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From rozzin at geekspace.com  Fri Jun 25 19:34:17 1999
From: rozzin at geekspace.com (Joshua Rosen)
Date: Fri, 25 Jun 1999 19:34:17 -0400
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <m3aetr2yk9.fsf@atrus.jesus.cam.ac.uk> <7ks38p$l4m$1@brick.cswv.com> <slrn7f68br.2bp.thantos@brimstone.mecha> <7kuocf$1an$1@brick.cswv.com>
Message-ID: <377411F9.DAF42195@geekspace.com>

Neel Krishnaswami wrote:
> Maybe we could re-use the class keyword and let people write class.foo
> inside class definitions, by analogy to writing self.foo to refer to
> the current instance? And the class.foo syntax would get translated to
> the equivalent of self.__class__?

	How about defining __getattr__ and company?

		-Rozzin.




From velzev at freemail.nl  Fri Jun 11 10:36:36 1999
From: velzev at freemail.nl (Vivienne)
Date: Fri, 11 Jun 1999 16:36:36 +0200
Subject: Trouble with ListBoxes in wxPyhton
Message-ID: <7jr72m$s72$1@zonnetje.NL.net>

Hi there,

I'm new to wxPython and I'm having some difficulties with Listboxes. I
create a panel with two listboxes that are filled with strings. When I
select an item from one listbox, the other should go empty (eventually there
should appear a list of objects that are linked to the chosen item). At the
moment I try it like this :

def UpdateCList(self, panel):
    x = CList(panel)
    x.EmptyList()

class PList:
    def __init__(self, panel):
        self.lijst = ['bla','bla','bla']
        self.panel = panel
        self.box = wxListBox(panel, 10, wxPoint(100,100), wxDefaultSize,
self.lijst, wxLB_SINGLE|wxLB_SORT)
        EVT_LISTBOX(panel, 10, self.OnLeftClick)

    def OnLeftClick(self, event):
        #some code
        UpdateCList(self, self.panel)

#class CList has a similar implementation, with the following method
included:
class CList:
    def __init__#etc.

    def EmptyList(self):
        self.box.Clear()

class MyApp(wxApp):
    def OnInit(self):
        frame = wxFrame(NULL,-1,'Test')
        panel = wxPanel(frame, -1)
        x = CList(panel)
        y = PList(panel)

if __name__=='__main__':
    app = MyApp(0)
    app.MainLoop()

But 1. This isn't working, and 2. It seems to be a very roundabout way.
Could anyone please help me?
Thanks,

Vivienne






From yopen at my-deja.com  Thu Jun 24 22:52:05 1999
From: yopen at my-deja.com (Bill Wilkinson)
Date: Fri, 25 Jun 1999 02:52:05 GMT
Subject: ODBC connectivity to Access!
References: <37723F25.31033307@wipsys.soft.net>
Message-ID: <7kuqsi$o7c$1@nnrp1.deja.com>

In article <37723F25.31033307 at wipsys.soft.net>,
  kvbalaji at wipsys.soft.net wrote:

> Hi ,
>   Can any body tell me how to connect and retrieve data stored in a
> MS Access database using ODBC in python?
> I am new to this language, Hints, pointers welcome!
> bala

Have you tried using the pythonwin  com extensions to get at the Access
data through the jet engine?  This works very well and is quick enough
for most uses.  If your client software is running on Windows, I think
this is the best way.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From tim_one at email.msn.com  Sun Jun 13 03:22:36 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 13 Jun 1999 03:22:36 -0400
Subject: fork()
In-Reply-To: <df983.24$KR2.60812@news.corridex.com>
Message-ID: <000601beb56d$829f6720$319e2299@tim>

[Evan Simpson]
> ...
> Perhaps I'm being naive, but I have the impression that when cycles
> arise it's usually a simple accident and easy to fix once you know
> you've done it.

Exactly so:  you're being naive <wink>.  Note that there are even cycles in
Python's implementation, most notably that a module namespace points to the
top-level functions defined in the module, while each of the latter points
back to the module namespace (which serves as the function's global
namespace, so must be reachable from the function).  There's no reasonable
way known to break that cycle, and indeed it causes Guido quite a bit of
pain at system shutdown time (when these cycles get  artifically snapped).
OTOH, nobody *notices* these cycles because they *expect* module functions
to stick around as long as the module does, and vice versa.  It's only at
shutdown time that surprises occur.

Stuff "like that" pops up in all sorts of apps.  With practice you learn
ways around it most often.  I've noted more than once that the resulting
cycle-free design is often more robust and comprehensible than its cyclic
ancestor, so I think it's a fruitful exercise to do this; doesn't always
succeed, though, and when you're in a rush it's a pain in the ass to have to
bother.

> ...
> Would it be possible (and not too expensive) to distinguish actual cycle
> members from dangly bits (maximal non-cyclic subgraphs)?  If so, it might
> be worth releasing the dangly bits normally through their root object
> after flushing the cycle.  That way you could have cycles in a program
> yet still have a place to put resource finalizers.

It doesn't fall out naturally in this approach, and suspect it's a lot more
expensive than simply marking the live objects.  There *are* elegant schemes
that work like this:  keep a refcount not on individual objects, but on the
strongly connected components (SCCs).  The SCCs of any graph necessarily
form a DAG, so the "cycle problem" can't arise *at* the SCC level; e.g.,

class C:
    def __init__(self, id):
        self.id = id
    def __del__(self):
        print self.id, "going away"

a, b, c = C('a'), C('b'), C('c')
a.partner = b
b.partner = a
a.extra = c
del a, b, c

a and b embrace each other, c is a "dangly bit" referenced by a, and at this
point all have a refcount of 1 and all are unreachable.  In the SCC scheme,
the SCCs are {a, b} and {c}, the refcount on the former is *zero* now and
the refcount on the latter is 1 (because {c} is referenced by {a, b}).  Does
a perfect job of separating the cycles from the dangly bits, and gives a
destruction order with the least possible number of headaches (because it
still respects dependencies *between* cycles and among the dangly bits).
Also nukes cycles the instant they become unreachable!  Alas, best I know
it's still thought impractically expensive to update the SCC info under
mutation, and so the scheme only shows up in some pure functional languages.

What you suggest is weaker, effectively lumping the SCCs into two gross
classes (singletons and non-singletons), and nuking the latter first.  I
think that just pushes the problems back one level; e.g., in the example
above, create another embracing pair {d, e} referenced from {c}.  You'll
invoke c's finalizer after {d, e} gets nuked, but since {c} did reference
{d, e} that probably means c's finalizer will blow up.  SCC would nuke {a,
b} first, *then* (c}, and *then* {d, e}.

Anyway, that's all by way of hinting <wink> that the scheme you suggest has
merit, but won't really work unless extended to the SCC approach (which
doesn't *need* to be incremental -- it could be done from scratch, at
massive expense <wink>, along with M&S).

too-bad-"if-the-implementation-is-hard-to-explain-it's-a-bad-idea"-is-
    one-of-our-core-principles<wink>-ly y'rs  - tim





From mlh at swl.msd.ray.com  Fri Jun 25 15:08:53 1999
From: mlh at swl.msd.ray.com (Milton L. Hankins)
Date: Fri, 25 Jun 1999 15:08:53 -0400
Subject: Environment inheritance under Windows NT 4, Attempt 2
In-Reply-To: <Pine.SGI.3.95q.990624162529.153620D-100000@masox202>
References: <Pine.SGI.3.95q.990624162529.153620D-100000@masox202>
Message-ID: <Pine.SGI.3.95q.990624200821.155434C-100000@masox202>

The mystery has been solved.  It turns out that my environment has a
duplicate name problem -- that is, two entries for the environment
variable exist in the data structure, each with two different values. This
can be easily be verified by writing a C program to print the contents of
_environ.  (I have such a program if anyone is interested.) 

This is not entirely the fault of Windows NT.  On my system, the
environment is being tainted by another appliacation.  However, NT is
permitting this to take place.  Feh.

Thus, an environment variable lookup becomes dependent on the
implementation's ordering.  In Python's implementation, the last entry
wins because it slurps the data from (C) _environ and stores it in
posix.environ.  The command prompt and getenv() seem to always get the
"right" value, somehow.

--
              Milton L. Hankins              \\  ><> Ezekiel 18:21 ><>
      http://www.snurgle.org/~mhankins       //  <mlh at swl.msd.ray.com>
    These are my opinions, not Raytheon's.   \\      W. W. J. D. ?




From mlh at swl.msd.ray.com  Thu Jun 24 16:34:36 1999
From: mlh at swl.msd.ray.com (Milton L. Hankins)
Date: Thu, 24 Jun 1999 16:34:36 -0400
Subject: Environment inheritance under Windows NT 4, Attempt 2
In-Reply-To: <092b01bebe6c$ef6b88e0$69c35f18@rochester.rr.com>
Message-ID: <Pine.SGI.3.95q.990624162529.153620D-100000@masox202>

I suppose this could be a side-effect of the way our IT department sets up
NT.  But I shudder to think how or what really causes it. 

On Thu, 24 Jun 1999, Darrell wrote:

> Works fine for me.
> Use the same case for the python version.
> > C:\>python -c "import os; print os.environ['WINDIR']"

Thanks for the response.  Unfortunately, this doesn't help.  :(  I tried
exact, all upper, and all lower case to the same result.

I've developed a workaround for posixmodule which uses getenv() under
Windows instead of direct access to environ.  It works great.  I had
really hoped others experienced this problem.  Sigh.  I will send my patch
to Guido anyway.  Maybe it will be deemed worthy. 


Fingers crossed...

--
              Milton L. Hankins              \\  ><> Ezekiel 18:21 ><>
Software Engineer, Raytheon Systems Company  //  <mlh at swl.msd.ray.com>
      http://amasts.msd.ray.com/~mlh         \\  RayComNet  7-225-4728




From doctorwes at mindspring.com  Sun Jun 27 00:30:33 1999
From: doctorwes at mindspring.com (Wesley Phoa)
Date: Sat, 26 Jun 1999 21:30:33 -0700
Subject: Advice for Newbie
References: <377588E9.82E8F567@dsuper.net> <7l4255$c22@dfw-ixnews13.ix.netcom.com>
Message-ID: <7l49q5$p5u$1@nntp2.atl.mindspring.net>

_Learning Python_ is much better than _Programming Python_, which is a bit
disorganized and hard to follow, and also a bit dated. And if you're
interested in the win32 extensions, or in JPython, then _Learning Python_ is
definitely the place to start. It also has some useful sections on 'gotchas'
that beginners (like me) often run into.

But, I have to say, I found that after a glance through the books, it was
pretty easy just to start writing Python code with the occasional glance at
the online documentation. And the code just works. Totally different from my
experience learning C.

Wesley.

Aahz Maruch wrote in message <7l4255$c22 at dfw-ixnews13.ix.netcom.com>...
>In article <377588E9.82E8F567 at dsuper.net>, Dave  <redsand at dsuper.net>
wrote:
>>
>>I read the Python homepage. I sounds promising. What books do you
>>commend for a beginner?
>
>If you already have experience with other programming languages, some
>combination of the following: _Programming Python_, _Learning Python_,
>and the on-line tutorial/documentation.  If you're not a programmer and
>you're not in a hurry, wait for _Teach Yourself Python in 24 Hours_.
>--
>                      --- Aahz (@netcom.com)
>
>Hugs and backrubs -- I break Rule 6       <*>
http://www.rahul.net/aahz/
>Androgynous poly kinky vanilla queer het





From guido at CNRI.Reston.VA.US  Thu Jun 24 07:45:14 1999
From: guido at CNRI.Reston.VA.US (Guido van Rossum)
Date: Thu, 24 Jun 1999 07:45:14 -0400
Subject: List <-> news hosed again?
In-Reply-To: Your message of "Wed, 23 Jun 1999 23:58:53 EDT."
             <000601bebdf5$e0572fc0$549e2299@tim> 
References: <000601bebdf5$e0572fc0$549e2299@tim> 
Message-ID: <199906241145.HAA26877@eric.cnri.reston.va.us>

> [Guido]
> > We're looking into this; uunet changed their configuration and we
> > neglected to make a matching change.  It should be fixed now
> > (presuming this message shows up on the uunet side of things).
> 
> Alas, don't think it worked -- DejaNews doesn't have your msg yet, nor my
> ISP's (admittedly flaky) c.l.py feed.  Appears to have stopped working very
> early AM (EDT) on 18-June-99.

Indeed, it didn't work.  Inspection of the logs indicates that the
uunet news software raises a "500 command not understood" error
whenever our news server tries to upload anything; this suggests a
serious version skew.  I think we may have to install a different
version of the news software (unless it is caused by bogus data sent
in a corrupted article that somehow upsets the command parser).

> in-the-last-hideous-stage-of-ego-surfing-withdrawal<wink>-ly y'rs  - tim

You could always post to python-list at cwi.nl, that gateway still works...

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



From proteus at cloud9.net  Thu Jun  3 10:27:05 1999
From: proteus at cloud9.net (Michael Muller)
Date: Thu, 03 Jun 1999 10:27:05 -0400
Subject: Threaded Tkinter
References: <m3btex4vxu.fsf@tiana.kli.prv>
Message-ID: <375690B9.37C1B8B6@cloud9.net>

Gerald Gutierrez wrote:
[snip]
> My first one -- I'm starting out with Tkinter. I wrote the code below,
> which will supposedly create a button with an image, that image
> changing every second. If I set the button to call swapIcons() when
> pressed, the image changes fine, but using the SpinThread, the icons
> don't change. Is this because Tkinter is not thread-safe and I cannot
> safely change the button's image property from within a thread other
> than the main one?

Tkinter is thread safe.  Tk is not.  To deal with this, the tkinter code
uses a global lock, serializing access to the Tk API.

Because of this, your background thread will block on Tkinter calls
while
the foreground thread is sitting in the mainloop.

You can deal with this by using a queue to communicate from your
background
thread and polling (arghhh!) the queue from the Tk thread like so:

  from Tkinter import Tk
  tk = Tk()

  # ...

  def poll():
     # clean out the queue
     while not myqueue.empty():
        processMessage(myqueue.get())

     # poll again in 1 second
     tk.after(1000, poll)

  poll()

Does anyone know of a better way to do this?

Alternately, in the case of your example, you could just put the image
swapping code in the poll() function.

[snip] 
> My second question -- I'd like to use Tkinter an Fnorb (CORBA)
> together as the UI and the network agent, respectively. Incoming CORBA
> calls will change appliation state and cause a change in the UI.
> 
> Now, since both Tkinter and Fnorb have their own event loops, I
> probably have to run them each in a separate thread. This is fine with
> me. The problem is that, if my first problem above is caused by the
> fact that I cannot use Tkinter in mutiple threads, I will need to send
> UI change requests for processing from within the main thread.
> 
> One way I know to do this is to have some sort of queue to which I can
> send messages. Tkinter's event loop will have to process the
> queue. How do I go about accomplishing this?
> 
> If this is not the correct way, what is?
> 

Fnorb is Tkinter aware: see examples/tkinter.  I think it deals with Tk
events and ORB requests in the same thread using select().  In my brief
experiments with this, I've had mixed results.

If the native mechanisms fail you, you can probably use the queue
approach
as identified above.

> Thanks for any help.
> 
> Please forward replies to gutier at intergate.bc.ca.


=============================================================================
michaelMuller = proteus at cloud9.net | http://www.cloud9.net/~proteus
-----------------------------------------------------------------------------
Mantra for the 60's: Tune in, turn on, drop out
Mantra for the 90's: Turn on, jack in, jerk off
=============================================================================



From matz at netlab.co.jp  Tue Jun  1 00:36:26 1999
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 01 Jun 1999 13:36:26 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
 <87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>
Message-ID: <87n1ykr0z9.fsf@ev.netlab.co.jp>

Paul Prescod <paul at prescod.net> writes:

|First you say:
|> I don't get it.  There's no relation between destructors and real
|> garbage collection.  For example, Java uses destructors called
|> finalizers, even though most Java VM use real GC.
|
|But then you say:
|> I know, by ref counting, object is
|> recycled as soon as it looses last reference to it, but relying on
|> that behavior is error prone, I think.
|
|So you do get it. There is a relation, but you just don't think that the
|"Python way" is right, You consider the guarantees that Java gives to be
|more important than the guarantees that Python does.

Hmm, I said ``there's no relation between destructors and real GC'',
then said ``deterministic destructor invocation is not needed''.

Even though I believe Java garantees is far more important than Python
garantees as you said, that's nothing do with ``relation between
destructors and real GC''.

Remember the original statement is like this:

>As many people have noted before, you would not be able to use
>destructors to free up resources like memory or file descriptors.

which, as I understand, means destructors can not be used with real
GC.  That's just not true.

|But I think that the bigger problem with "real GC" is that C programs
|often hang on to references to Python objects and any kind of compacting
|GC would invalidate those references.

It's the common legend especially here in comp.lang.python.

From my experience with Scheme and Ruby, both with real GC, the real
GC is mostly very fast.  Fast enough no one notices its "hang" on most
cases.  The legend was formed up by old GC implementation and slower
machines of the past, I guess.

|But note that Python's guarantees are more important in small, simple
|programs and Java's guarantees are better for large, complex programs with
|complicated data structures.

Do you really mean Python the object-oriented programming language
rarely treat complicated data structures?  I think Python is good
language, good enough to make people write complex program with
complicated data structures.

I admit ref-counting is far easier to implement.  But it's too simple
way for the real language like Python.

|http://www.quoininc.com/quoininc/Design_Java0197.html

This document says ``the real GC with the finalizers do the most of
the jobs'', right?

						matz.



From roger.burnham at gte.net  Wed Jun 23 22:06:45 1999
From: roger.burnham at gte.net (Roger Burnham)
Date: Thu, 24 Jun 1999 02:06:45 GMT
Subject: Windows - Drag & drop on Python script?
References: <37717600.D076BF10@compaq.com>
Message-ID: <3771908b.1913288@news.gte.net>

On Thu, 24 Jun 1999 12:04:16 +1200, Greg Ewing <greg.ewing at compaq.com> wrote:

>How can I create an icon onto which I can drop a file,
>and have it run a Python script with the dropped file
>as an argument?
>
>I've tried myriad combinations of .bat files, shortcuts,
>etc., but nothing seems to work.
>
>The obvious thing would seem to be to use a .bat
>file, but the dropped file seems to end up as the
>standard input or something, rather than getting
>passed as an argument.
>
>There *must* be a way of doing this, surely?
>

Greg,

One choice would be to hack up Pythonwin\pywin\framework\app.py and modify the
OnDropFiles method to do your thing rather than opening the dropped file in the
editor.  Then you could drop files onto the PythonWin icon.

I played a bit with this, the best I could come up with is the following dialog
app, which must be running, and will accept files dropped onto it and print its
output into an edit window.


# adapted from Pythonwin\pywin\Demos\app\dlgappdemo.py

from pywin.framework import dlgappcore, app
import win32ui, win32con, win32api
import sys
import regsub

class DropScriptDialogApp(dlgappcore.DialogApp):
    def CreateDialog(self):
        return DropScriptAppDialog()

class DropScriptAppDialog(dlgappcore.AppDialog):
    def __init__(self):
        self.edit = None
        dlgappcore.AppDialog.__init__(self, win32ui.IDD_LARGE_EDIT)
        
    def OnInitDialog(self):
        self.SetWindowText('Drop script application')
        self.edit = self.GetDlgItem(win32ui.IDC_EDIT1)
        self.DragAcceptFiles()
        self.HookMessage(self.OnDropFiles, win32con.WM_DROPFILES)
        return 1

    def OnDropFiles(self, msg):
        hDropInfo = msg[2]
        nFiles = win32api.DragQueryFile(hDropInfo)
        try:
            for iFile in range(0, nFiles):
                fileName = win32api.DragQueryFile(hDropInfo, iFile)
                print '%s dropped...' % fileName
        finally:
            win32api.DragFinish(hDropInfo);
        return 0
                
    def PreDoModal(self):
        sys.stdout = sys.stderr = self

    def write(self, str):
        if self.edit:
            self.edit.SetSel(-2)
            # translate \n to \n\r
            self.edit.ReplaceSel(regsub.gsub('\n','\r\n',str))
        else:
            win32ui.OutputDebug(
                'dlgapp - no edit control! >>\n%s\n<<\n' % str )

app.AppBuilder = DropScriptDialogApp

if __name__=='__main__':
    import demoutils
    demoutils.NeedApp()
        

Roger Burnham   
Cambridge Research & Instrumentation   
rburnham at cri-inc.com   
http://www.cri-inc.com/   
http://starship.python.net/crew/roger/   
PGP Key: http://www.nai.com/default_pgp.asp   
PGP Fingerprint: 5372 729A 9557 5F36 177F  084A 6C64 BE27 0BC4 CF2D   



From meowing at banet.net  Wed Jun 23 09:51:01 1999
From: meowing at banet.net (meow)
Date: 23 Jun 1999 09:51:01 -0400
Subject: interactive-logging
References: <376FF09A.8170CB@icrf.icnet.uk> <y0jso7jghez.fsf@vier.idi.ntnu.no>
Message-ID: <87yahbc9fe.fsf@banet.net>

Magnus L Hetland <mlh at idt.ntnu.no> writes:

> Arne Mueller <a.mueller at icrf.icnet.uk> writes:
>> I'd like to add some logging to the interpreter's interactive modus so
>> that everything I type in and everything what's printed to the screen
>> (errors and 'normal' output) is copied to a log-file.

> Hm. Isn't there a UN*X command that does that? (That logs everything
> going on in a terminal...)

'python | tee log.file' would be a simple way to get that kind of
thing.



From rdev at my-deja.com  Wed Jun 30 15:18:38 1999
From: rdev at my-deja.com (rdev at my-deja.com)
Date: Wed, 30 Jun 1999 19:18:38 GMT
Subject: theads & global namespaces
References: <377A5C88.32A1FAB6@notes.cba.ufl.edu>
Message-ID: <7ldqi2$nc1$1@nnrp1.deja.com>

Two suggestions:
- If you're running on Solaris, the thread won't get to run
unless you give up some time.  So try putting a time.sleep(.001)
into your while loop.  On NT, this is not necessary (don't know
why.)
- Your thread will only increment once and then exit, so you'll
never see x>1.  If you want to keep the thread running, put it in
a while loop.  Make sure you give up the time slice there as well.

Roger Dev
dev at opticominc.com

In article <377A5C88.32A1FAB6 at notes.cba.ufl.edu>,
  andersdt at notes.cba.ufl.edu wrote:
> Greetings all,
>
> I am trying to share globals between threads.  Is is possible for me
to
> share globals between threads that call functions from different
> modules?  Thus far I have had little success.  I ask this question
> because  I would prefer not to put all my code in one module.
>
> For example, I have 2 .py files: a.py and b.py.  b.py contains the
> function ('foo()') to be executed  under the new thread, as well as
the
> global variable 'x'.  a.py contains the call to
> start_new_thread(foo,()).  a.py also tries to use the global variable
x.
>
> =========================================
> b.py
>
> x = 0
>
> def foo():
>    global x
>    x  = x + 1
>
> =========================================
> a.py
>
> from b.py import *
>
> if __name__ == '__main__':
>
>     start_new_thread(foo, ())
>     while 1:
>         if x != 0:
>             print x
>
> =========================================
>
> So far, my code in a.py does not detect any change in x.  Any
> suggestions for a workaround?
>
> Many thanks,
>
> Taylor
>
>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From davecook at home.com  Wed Jun 16 01:54:02 1999
From: davecook at home.com (David M. Cook)
Date: Wed, 16 Jun 1999 05:54:02 GMT
Subject: urllib: an alternative URLOpener
References: <Pine.SUN.3.95-heb-2.07.990611155013.10151A-100000@sunset.ma.huji.ac.il> <7jt7uc$sh4@news.rhrz.uni-bonn.de> <14178.38855.323759.698523@bitdiddle.cnri.reston.va.us> <14181.8918.956963.762261@bitdiddle.cnri.reston.va.us>
Message-ID: <slrn7mef1g.1b0.davecook@sputnik.escnd1.sdca.home.com>

On Mon, 14 Jun 1999 11:42:14 -0400 (EDT), Jeremy Hylton
<jeremy at cnri.reston.va.us> wrote:

>Ok.  There is now an alpha version of urllib2 available from my Web
>page: http://the-tech.mit.edu/~jeremy/python/

Please forgive the naive question, but how does one get the "realm" info
from a site?

I want to connect to a page that requires logging in with username and
password and then retrieve the data there, but it seems I have to know the
realm info as well?

Dave Cook



From tony at lsl.co.uk  Mon Jun 28 08:48:11 1999
From: tony at lsl.co.uk (Tony J. Ibbs (Tibs))
Date: Mon, 28 Jun 1999 12:48:11 GMT
Subject: [comp.lang.python.*] Posting guidelines -- bi-weekly posting
Message-ID: <04e501bec164$7b238b90$f0c809c0@lslp7o.lsl.co.uk>

Would it be worth pointing out, near the top, that, since comp.lang.python
is
gatewayed to a mailing list, it is considered good etiquette not
to assume that one is *only* writing to News - i.e., many people
do not have the option of ignoring messages based upon the
message header, but get the whole thing automatically delivered.

I know that we still get people who don't read the document contributing
to the mailing list, but perhaps this might stop *some* of the people
who (non-maliciously) start long, winding, off-topic discussions which
they believe "can be ignored by people who aren't interested"...

--
Tony J Ibbs (Tibs)      http://www.tibsnjoan.demon.co.uk/
Feet first with 5 wheels...
My views! Mine! Mine! (Unless Laser-Scan ask nicely to borrow them.)




From aahz at netcom.com  Wed Jun  9 19:18:12 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Wed, 9 Jun 1999 23:18:12 GMT
Subject: How do I defect?
References: <375EEBE2.3F20390F@lucent.com>
Message-ID: <aahzFD322C.Fu1@netcom.com>

In article <375EEBE2.3F20390F at lucent.com>,
Burton Kent  <burton at lucent.com> wrote:
>
>So how do I defect from Perl to Python?  I have the Programming Python
>book on order.  What can I read to get started ASAP?

http://www.python.org/doc/current/tut/tut.html
-- 
                      --- Aahz (@netcom.com)

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

"I'm not tense, i'm just terribly, terribly alert"  -- unknown



From holger at phoenix-edv.netzservice.de  Thu Jun 10 05:43:02 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Thu, 10 Jun 1999 09:43:02 GMT
Subject: MAC-adress
Message-ID: <375F88A6.469BC143@phoenix-edv.netzservice.de>

High,

is there a way to get the MAC-adresses and other properties of network-cards with python?

thanx,
Holger



From greg.ewing at compaq.com  Mon Jun 28 23:18:04 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Tue, 29 Jun 1999 15:18:04 +1200
Subject: Why is tcl broken?
References: <375F82C7.CF0F0477@iname.com> <z%Q93.4399$5E4.36952@news2.randori.com> <m2n1xsmh10.fsf@desk.crynwr.com> <37776F1C.5656@mailserver.hursley.ibm.com>
Message-ID: <37783AEC.931EC3C9@compaq.com>

Paul Duffin wrote:

> Russell Nelson wrote:

> > Because it necessarily parses strings.

> What do you mean by "necessarily parses strings" ?

It's better than unnecessarily parsing
strings, I suppose! :-)

Greg



From mwh21 at cam.ac.uk  Sun Jun 13 12:26:10 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 13 Jun 1999 17:26:10 +0100
Subject: SWIG'in python question
References: <37542E8B.ABEAA531@kw.igs.net>
Message-ID: <m3aeu4t6cd.fsf@atrus.jesus.cam.ac.uk>

JB <jimbag at kw.igs.net> writes:

> I've got a .so that I built with swig for python. Everything builds fine
> but when I import i get this message from python "_mymod.so: undefined
> symbol: __eh_pc". I wondered if anyone had seen this before and could
> give me a clue. I think I'm missing a library somewhere. I'm building on
> RedHat 5.2.

Noone else has replied to this yet have they?

Don't know, but the name __eh_pc suggests exception handling. Not
compiling C as C++ or anything like that are you?

HTH
Michael
 
> thanks
> jb



From suzuki611 at okisoft.co.jp  Fri Jun  4 03:25:46 1999
From: suzuki611 at okisoft.co.jp (Hisao Suzuki)
Date: 04 Jun 1999 16:25:46 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
	<87pv3iq6ss.fsf@ev.netlab.co.jp> <3752C76D.604CD474@prescod.net>
	<87n1ykr0z9.fsf@ev.netlab.co.jp> <375372F6.A874554@prescod.net>
	<87hfosquc0.fsf@ev.netlab.co.jp> <3753F121.E4A9BE8B@prescod.net>
	<87yai3ntdt.fsf@ev.netlab.co.jp>
	<7j3iif$134e$1@nntp6.u.washington.edu>
	<7j45iv$908$4@cronkite.cc.uga.edu>
	<7j4hhj$118u$1@nntp6.u.washington.edu>
	<7j4jbe$dhh$1@cronkite.cc.uga.edu> <37564A02.F115B9C5@infercor.no>
	<045501beadf8$610e62e0$f29b12c2@pythonware.com>
	<14166.61071.201719.405639@bitdiddle.cnri.reston.va.us>
Message-ID: <u31zfszath.fsf@ares.sys1.ngo.okisoft.co.jp>

In article <14166.61071.201719.405639 at bitdiddle.cnri.reston.va.us>,
Jeremy Hylton <jeremy at cnri.reston.va.us> wrote:
| To the contrary, the community is too busy using Python to bother with 
| all these silly discussions about changing Python.  The fact that I'm
| posting in this threads is... ummm... the exception that proves the
| rule.

Yes, we are too busy using Python to bother with some fancy
ideas, in particular, those of some other (superior?) languages.

| Most of the volume has been a vigorous but unenlightening back and
| forth about garbage collection.  I've for one have concluded that
| there have been enough papers written about garbage collection that,
| for any particular argument against GC, there exists at least one
| paper describing a system for which that argument doesn't apply.  
| 
| Of course, none of that has anything to do with Python.  I'm tired of
| hearing what GC can do in theory.  I'd be happy to hear about some
| actual experience implementing a GC system for Python.  Otherwise,
| let's put this on to bed.

Me too.  And I am afraid that some evangelists would do nothing
but _only_ point out Python's some so-called fault in order to
drive us into convert to their favorite or invented language...

Anyway, as you say, I'd think that we need an actual experience.

The following might be served as the very basic, half-baked
demonstration of mark-and-sweep garbage collection.  (Actually
it does not collect any garbages.  It just report them.)

------8<------8<------8<------8<------8<------8<------8<------8<------
import sys
from types import *

def get_referred_instances():
    """Collect all the validly referred instances [PRELIMINARY VERSION].

    Return a dictionary of id_of_instance:name_of_instance, where
    id_of_instance is an integer value returned by id(instance) and
    name_of_instance is a human-readable string generated in some ad
    hoc way.

    Note that the reference counting of the interpreter does not
    affected by this function.  The return value does not have
    instances themselves but id values of them.
    """
    instances = {}
    dicts = {}
    _get_referred_sub(sys.modules, "", dicts, instances)
    return instances

def _get_referred_sub(val, name, dicts, instances):
    if isinstance(val, DictionaryType):
	for key, obj in val.items():
	    _get_referred_sub(key, '%s\'s key' % name,
			      dicts, instances)
	    _get_referred_sub(obj, '%s[%s]' % (name, key),
			      dicts, instances)
    elif isinstance(val, ListType) or isinstance(val, TupleType):
        for i, obj in map(None, range(len(val)), val):
	    _get_referred_sub(obj, '%s[%d]' % (name, i),
			      dicts, instances)
    else:
	# instance, class, module, ...
	if isinstance(val, InstanceType): instances[id(val)] = name
	elif isinstance(val, ModuleType): name = val.__name__
	try: d = val.__dict__
	except: pass
	else:
	    _id = id(d)
	    if not dicts.has_key(_id):
		dicts[_id] = name
		for key, obj in d.items():
		    key = name + '.' + key
		    _get_referred_sub(obj, key, dicts, instances)

if __name__ == '__main__':
    class A:				# sample class
	instances = []
	def __init__(self): A.instances.append(id(self))
	def __del__(self): A.instances.remove(id(self))
    a = A(); b = A(); a.b = b; b.a = a	# construct circular references
    x = [A(), A(), A()]

    print "Referred instances are", get_referred_instances()
    print "Class A instances are", A.instances
    print "Delete variables a, b, and x."
    del a, b, x
    print "Now, class A instances are", A.instances

    # Mark all objects referred.
    marked = get_referred_instances().keys()
    # Then, sweep the garbage of class A.
    garbages = []
    for _id in A.instances:
	if _id not in marked:
	    garbages.append(_id)
    print "Unreferred (= of garbage) A instances are", garbages
------8<------8<------8<------8<------8<------8<------8<------8<------

And this is the result:

Referred instances are {135060000: '__main__.a.b.a', 135060024: '__main__.b', 135060128: '__main__.x[2]', 135060072: '__main__.x[1]', 134996472: 'os.environ', 135060048: '__main__.x[0]'}
Class A instances are [135060000, 135060024, 135060048, 135060072, 135060128]
Delete variables a, b, and x.
Now, class A instances are [135060000, 135060024]
Unreferred (= of garbage) A instances are [135060000, 135060024]

--===-----========------------- Sana esprimo naskas sanan ideon.
SUZUKI Hisao            suzuki611 at okisoft.co.jp, suzuki at acm.org.



From Mathieu.Lecarme at p98.f112.n480.z2.fidonet.org  Wed Jun 30 11:15:19 1999
From: Mathieu.Lecarme at p98.f112.n480.z2.fidonet.org (Mathieu Lecarme)
Date: Wed, 30 Jun 1999 16:15:19 +0100
Subject: looking for cookie.py
Message-ID: <000015b1@bossar.com.pl>

From: "Mathieu Lecarme" <zehp at francemel.com>

the link on python web site doesn't work!
where can i find it?




From alt at artisan.com  Mon Jun 14 15:50:06 1999
From: alt at artisan.com (alt at artisan.com)
Date: 14 Jun 1999 12:50:06 -0700
Subject: is there a Gantt chart toolkit?
Message-ID: <7in1y2imtt.fsf@betty.artisan.com>

subject says it all.

Thanks!
Albert 



From aahz at netcom.com  Tue Jun 15 00:36:13 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Tue, 15 Jun 1999 04:36:13 GMT
Subject: Newbie: easy filecopy?
References: <376531D3.7E42AC42@phoenix-edv.netzservice.de>
Message-ID: <aahzFDCq4D.Do1@netcom.com>

In article <376531D3.7E42AC42 at phoenix-edv.netzservice.de>,
Holger Jannsen  <holger at phoenix-edv.netzservice.de> wrote:
>
>I couldn't believe it. I think there must be an easy copy-function
>in python (I only want to copy files under ext2- or FAT-FileSystems).
>Call me "4 blind mice"-Joey, but I couldn't find it. 
>For sure, I could write a thing like open-read-write-close, but I
>think, someone else has done already, hasn't he?

Look at the shutil module.
-- 
                      --- Aahz (@netcom.com)

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

"That doesn't stop me from wanting to work all three of them over with
the clue stick for while, with no safewords allowed."  --abostick



From paul at prescod.net  Sun Jun 27 21:37:42 1999
From: paul at prescod.net (Paul Prescod)
Date: Mon, 28 Jun 1999 01:37:42 GMT
Subject: Pythonwin and Office 2000
References: <377648A7.9B9BE59F@prescod.net> <7l672r$5k0$1@m2.c2.telstra-mm.net.au>
Message-ID: <3776D1E6.45E1AAF7@prescod.net>

Mark Hammond wrote:
> 
> Paul Prescod wrote in message <377648A7.9B9BE59F at prescod.net>...
> >I believe that Pythonwin is not compatible with Office 2000. Here are some
> >symptoms:
> 
> I havent tried this, although I believe it!

Well, I got it to work eventually. I used the command line makepy instead
of the one in the GUI. It didn't test the result before it registered it
in the .dat file. Then I edited the file by hand to remove the upper-case
"L" characters.

> This is expected (unfortunately).  Word will have a new typelibrary, and
> hence a new IID for the library.  The "fix" will be to work out what the
> library ID of the new one is (by using "makepy -i") and changing TestWord.py
> to gracefully allow either version.

 >>> # Use these commands in Python code to auto generate .py support
 >>> from win32com.client import gencache
 >>> gencache.EnsureModule('{91493440-5A91-11CF-8700-00AA0060263B}', 0, 2,
14)

Is there an equivalent to EnsureModule that works from the textual library
name? I know that the textual library name still changes from version to
version but I don't think it is Pythonish (easy, clean and readable) to
put UUIDs in code at all.

In fact, the most Pythonish way would be to do what Python does with
Pyc's. When Pythonwin does a win32com.client.Dispatch, it could generate
the gencache module if possible. If not possible then no problem, it just
works in the dynamic mode.

Also, is there any chance of having win32com.client.Dispatch renamed
CreateObject() for JavaScript and VBScript compatibility?

> >    def AddOLEObject(self, Left=0.0, Top=0.0, Width=-1.0, Height=-1.0,
> >ClassName
> >=L'', FileName=L'', DisplayAsIcon=0, IconFileName=L'', IconIndex=0,
> >IconLabel=L'
> >', Link=0):

Once I deleted that strange "L" thing, the generated module worked.

In case it helps, here is the code:

def AddOLEObject(self, Left=0.0, Top=0.0, Width=-1.0, Height=-1.0,
ClassName=L'', FileName=L'', DisplayAsIcon=0, IconFileName=L'',
IconIndex=0, IconLabel=L'', Link=0):
		return self._ApplyTypes_(0x7d8, 1, (9, 32), ((4, 49), (4, 49), (4, 49),
(4, 49), (8, 49), (8, 49), (3, 49), (8, 49), (22, 49), (8, 49), (3, 49)),
'AddOLEObject', '{91493479-5A91-11CF-8700-00AA0060263B}', Left, Top,
Width, Height, ClassName, FileName, DisplayAsIcon, IconFileName,
IconIndex, IconLabel, Link)

...
	# Result is of type TextRange
	def InsertAfter(self, NewText=L''):
		return self._ApplyTypes_(0x7e1, 1, (9, 32), ((8, 49),), 'InsertAfter',
'{9149348F-5A91-11CF-8700-00AA0060263B}', NewText)
...


	def Help(self, HelpFile=L'vbappt9.chm', ContextID=0):
		return self._ApplyTypes_(0x7e4, 1, (24, 32), ((8, 49), (22, 49)),
'Help', None, HelpFile, ContextID)


-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for only himself
 http://itrc.uwaterloo.ca/~papresco

Those who profess to favor freedom and yet depreciate agitation, are men 
who want crops without plowing up the ground. They want rain without 
thunder and lightning. They want the ocean without the roar of its many 
waters. - Fredrick Douglass
 http://www.informamerica.com/Articles/Quotes.htm



From Cepl at fpm.cz  Wed Jun  2 09:03:36 1999
From: Cepl at fpm.cz (Matej Cepl)
Date: Wed, 2 Jun 1999 14:03:36 +0100
Subject: HTTPlib.py error
Message-ID: <1318D78C9072D11195C9006094EA98A72C20B8@ocesrv>

What's wrong with me?

Probably, the missing thing is some kind of configuration of httplib.py,
but even in documentation I did not find anything helpfull.

		Thanks for any response

					Matthew
--------------------------------------------------------------
C:\Program Files\Python\lib>python "C:\Program
Files\Python\lib\httplib.py"
Traceback (innermost last):
  File "C:\Program Files\Python\lib\httplib.py", line 193, in ?
    test()
  File "C:\Program Files\Python\lib\httplib.py", line 179, in test
    h.connect(host)
  File "C:\Program Files\Python\lib\httplib.py", line 79, in connect
    self.sock.connect(host, port)
  File "<string>", line 1, in connect
socket.error: host not found 



From mikael at pobox.com  Wed Jun 30 19:51:16 1999
From: mikael at pobox.com (Mikael Lyngvig)
Date: Wed, 30 Jun 1999 23:51:16 GMT
Subject: Python GUIs: Summary and Conclusion
References: <99FA4A1A1280BEBD.0A1ABA62CC989479.4C2C6022F695BEAC@lp.airnews.net> <MTOd3.139$6M6.44310@news.shore.net>
Message-ID: <ADA11DBD29B53BB7.BF5C474E8A0D15AB.A4DD9ACF36310266@lp.airnews.net>

On Mon, 28 Jun 1999 18:09:48 GMT, "Michael P. Reilly"
<arcege at shore.net> wrote:

>I've already rewritten _tkinter to use a new pytcl module, but the
>functionality is still the same, ie. executing Tk calls from within an
>interpreter.  But I'd like to see it using the TK/C API (so much for
>my hard word on rewriting _tkinter *wink*).
>
>The main issue would be to make it extensible for adding the new
>widgets and other extensions (BTL, PIL, etc.).  Also, the interface
>between Python and C would not be as "simple" as the existing module.
>It's doable tho.
>
>Note that using the "native" Tk API would still require linking with
>the Tcl library.
>
>I could scratch my _tkinter rewrite and work on a native solution.
>(I could still publish my pytcl module.)  Any thoughts?

I've downloaded the Tcl 8.1.1 and Tk 8.1.1 sources, rebuilt them, and
studied them for a bit.

I think we're looking at a major task.  As far as I can determine from
the Tk sources, the parser is an integral part of the Tk library.
E.g., the Tk sources appear to be high-level down to the lowest level.

Even the lowest level functions takes argv[] style parameters and,
aassisted by a horde of strncmp()s (350+ in the generic directory),
parse the parameter/option list into a command, which is executed by
code which is interleaved with the parsing code.

I incorrectly assumed the design would be something like:

	[Tcl parser layer]
	[Tk parser layer]
	[Tk implementation layer]

But it appears to be:

	[Tcl parser layer]
	[Tk parser and implementation layer]

My hope was to make a Python interface run on top of the
implementation layer, but that appears to be impossible with the
current Tk design.

So, in order to make a native Tk layer for Python, we'd have to create
the implementation layer ourselves, and convince the Tk maintainers to
adopt this (if we'd want their future to be part of our TkInter
automatically).

I can illustrate what I mean by looking at the Tk bell command's
implementation:

int
Tk_BellObjCmd(clientData, interp, objc, objv)
    ClientData clientData;      /* Main window associated with
interpreter. */
    Tcl_Interp *interp;         /* Current interpreter. */
    int objc;                   /* Number of arguments. */
    Tcl_Obj *CONST objv[];      /* Argument objects. */
{
    static char *bellOptions[] = {"-displayof", (char *) NULL};
    Tk_Window tkwin = (Tk_Window) clientData;
    char *displayName;
    int index;

    if ((objc != 1) && (objc != 3)) {
        Tcl_WrongNumArgs(interp, 1, objv, "?-displayof window?");
        return TCL_ERROR;
    }

    if (objc == 3) {
        if (Tcl_GetIndexFromObj(interp, objv[1], bellOptions,
"option", 0,
                &index) != TCL_OK) {
            return TCL_ERROR;
        }
        displayName = Tcl_GetStringFromObj(objv[2], (int *) NULL);
        
        tkwin = Tk_NameToWindow(interp, displayName, tkwin);
        if (tkwin == NULL) {
            return TCL_ERROR;
        }
    }
    XBell(Tk_Display(tkwin), 0);
    XForceScreenSaver(Tk_Display(tkwin), ScreenSaverReset);
    XFlush(Tk_Display(tkwin));
    return TCL_OK;
}

The way I see it, this should have been something like this:

int Tki_WindowBell(Window)
   Tk_Window Window;
{
    XBell(Tk_Display(Window), 0);
    XForceScreenSaver(Tk_Display(Window), ScreenSaverReset);
    XFlush(Tk_Display(Window));
    return TCL_OK;
}

int
Tk_BellObjCmd(clientData, interp, objc, objv)
    ClientData clientData;      /* Main window associated with
interpreter. */
    Tcl_Interp *interp;         /* Current interpreter. */
    int objc;                   /* Number of arguments. */
    Tcl_Obj *CONST objv[];      /* Argument objects. */
{
    static char *bellOptions[] = {"-displayof", (char *) NULL};
    Tk_Window tkwin = (Tk_Window) clientData;
    char *displayName;
    int index;

    if ((objc != 1) && (objc != 3)) {
        Tcl_WrongNumArgs(interp, 1, objv, "?-displayof window?");
        return TCL_ERROR;
    }

    if (objc == 3) {
        if (Tcl_GetIndexFromObj(interp, objv[1], bellOptions,
"option", 0,
                &index) != TCL_OK) {
            return TCL_ERROR;
        }
        displayName = Tcl_GetStringFromObj(objv[2], (int *) NULL);
        
        tkwin = Tk_NameToWindow(interp, displayName, tkwin);
        if (tkwin == NULL) {
            return TCL_ERROR;
        }
    }

    Tki_WindowBell(tkwin)
    return TCL_OK;
}

That way we could easily add a very fast Python layer on top of the
Tki_Xxx() functions.

So, in order to accomplish the task we'd have to restructure the
entire Tk source distribution - basically modify all functions in all
source files.

>We already know that Python 1.6 won't be until the new year, so any
>changes to Tkinter would be as a patch.

>From what I know now, I suspect that Python 3.0 could be done by the
time we're through :)

Also, if we did separate the parser and the implementation, we'd end
up with something which is directly incompatible with add-on widgets
and so on.  Although we could probably make a TkParser module, which
could interface to the parser layer.

My view, given the latest info, is that the task of making a native
TkInter is so huge that it doesn't pay off, unless your goal is to use
Tk in some form (not to get a fast, portable GUI).  The quickest way
to make TkInter faster is to wait a couple of years till your (then)
computer has quadrubled in speed.

What are your thoughts?  

P.S. Can anyone tell me how to ring a bell using the native Tk API?  I
don't see any way, except to do what Tk_BellObjCmd() does yourself.



-- Mikael




From alrice at swcp.com  Thu Jun  3 16:24:22 1999
From: alrice at swcp.com (Alex Rice)
Date: Thu, 03 Jun 1999 14:24:22 -0600
Subject: Alien whitespace eating nanovirus strikes again!
References: <375500C9.742DC539@pop.vet.uu.nl> <3756E3D9.E44D2B39@swcp.com>
Message-ID: <3756E476.ACB50DF1@swcp.com>

Therefore python is more post-modern than perl.

:-)



From cgw at fnal.gov  Tue Jun  1 13:07:39 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Tue, 1 Jun 1999 12:07:39 -0500 (CDT)
Subject: os.access for pre-1.5.2
Message-ID: <14164.4955.804605.200302@buffalo.fnal.gov>

I have some code which uses the os.access function, and need this code 
to be runnable on RedHat5.2/6.0 systems (which have python version
1.5.1).

I wrote the following Python implementation of access, which I am
posting here in case it is useful to anyone else.  Feedback is
welcome.



#!/usr/bin/env python

# $Id: access.py,v 1.1 1999/06/01 16:42:48 cgw Exp $

"""os.access isn't supported in Python versions before 1.5.2, so this module provides a portable implementation"""

import os

F_OK, X_OK, W_OK, R_OK = 0, 1, 2, 4

if hasattr(os,"access"):
    access = os.access
else:
    import stat
    stat_masks={
        X_OK: (stat.S_IXOTH, stat.S_IXGRP, stat.S_IXUSR),
        W_OK: (stat.S_IWOTH, stat.S_IWGRP, stat.S_IWUSR),
        R_OK: (stat.S_IROTH, stat.S_IRGRP, stat.S_IRUSR)
        }
    
    def access(filename, my_mode):
        if not os.path.exists(filename):
            return 0
        try:
            s=os.stat(filename)
        except: # if we can't stat it, all bets are off
            return 0

        if my_mode == F_OK: #check for existence only
            return 1
            
        my_uid=os.getuid() # not geteuid, as per POSIX def'n of "access"
        my_gid=os.getgid()
        file_uid=s[stat.ST_UID]
        file_gid=s[stat.ST_GID]
        file_mode=s[stat.ST_MODE]

        for test in X_OK, W_OK, R_OK:
            if test & my_mode == 0:
                continue
            other_mask, group_mask, user_mask = stat_masks[test]
            other_ok = file_mode & other_mask 
            group_ok = file_mode & group_mask and my_gid==file_gid
            user_ok = file_mode & user_mask and my_uid==file_uid
            if not (other_ok or group_ok or user_ok):
                return 0
        #all tests passwd
        return 1

if __name__ == '__main__':
    testname = {F_OK: 'present',
                X_OK: 'executable',
                W_OK: 'writable',
                R_OK: 'readable'
                }
    for file in '/tmp', '/nonesuch', '/usr', '/bin/sh':
        for test in range(8):
            print "Testing if", file, "is",
            if test==0: print "present",
            for mask in F_OK,X_OK,W_OK,R_OK:
                if mask & test: print testname[mask],
            print "...", access(file, test) and "yes" or "no"

            
            
            



From trigg at parc.xerox.com  Thu Jun 10 21:12:51 1999
From: trigg at parc.xerox.com (Randy Trigg)
Date: Thu, 10 Jun 1999 18:12:51 -0700
Subject: Using os.system and os.popen from pythonwin under Windows NT
Message-ID: <37606293.FA48FB71@parc.xerox.com>

I'm sure I'm missing some really trivial python/windows thing here.  I
have a .exe executable which runs fine when called in the Command Prompt
window (it doesn't return anything - just writes its results in a file).

But when I attempt to invoke the thing from pythonwin using
os.popen(<command>), I get

  Traceback (innermost last):
    File "<interactive input>", line 0, in ?
  OSError: (0, 'Error')

(I thought the "0" value indicated *no error* which just shows how
confused I've become.)

Meanwhile, if I try invoking the same program using
os.system(<command>), it flashes what looks like it might be a shell
window, and then dies returning error code 1.

Thanks in advance for any ideas,

   - Randy




From swartz at erols.com  Sat Jun 19 13:28:27 1999
From: swartz at erols.com (Swartz)
Date: Sat, 19 Jun 1999 13:28:27 -0400
Subject: Curses for MSDOS
Message-ID: <376BD33B.5A6F54BB@erols.com>

Is there anything like curses for MS-DOS?  Thanx.



Paul




From ivnowa at hvision.nl  Tue Jun 22 16:45:22 1999
From: ivnowa at hvision.nl (Hans Nowak)
Date: Tue, 22 Jun 1999 20:45:22 GMT
Subject: YAS to the "Reading line-by-line" Problem
In-Reply-To: <376FDF70.4756AD38@mindspring.com>
References: <376FDF70.4756AD38@mindspring.com>
Message-ID: <199906222044.WAA08518@axil.hvision.nl>

On 22 Jun 99, Jesse D. Sightler wrote:

> Ack is raising an exception for a normally occuring event really a
> _good_ idea?  Generally exceptions are only used for things that
> actually are error conditions, and it would seem to me that using them
> as nothing more than a fancy flow-control construct is a sure path to
> nasty spaghetti code.
> 
> I really do not like this idea.  :)

Hmm, I always wanted file objects to have an eof() method... I never 
quite understood why it isn't there. Assuming file objects are based 
on C's fopen, fclose, fread etc. functions, it puzzles me why feof 
wasn't used as well.

Of course I could write another file object to do this... but it's 
not a big deal, really. I just wondered. :)

--Hans Nowak (ivnowa at hvision.nl)
Homepage: http://fly.to/zephyrfalcon



From a.mueller at icrf.icnet.uk  Wed Jun  9 11:22:41 1999
From: a.mueller at icrf.icnet.uk (Arne Mueller)
Date: Wed, 09 Jun 1999 16:22:41 +0100
Subject: waste of resources ?
References: <375C0F91.1C73E4E3@icrf.icnet.uk> <375C16A9.5ADC4990@icrf.icnet.uk> <375C1FD1.A85565C@aw.sgi.com> <c75so8370by.fsf@pc142.cosc.canterbury.ac.nz> <375D216B.61787B66@icrf.icnet.uk> <87r9nl398o.fsf@psyche.evansnet>
Message-ID: <375E86C1.73B2CF80@icrf.icnet.uk>

Carey Evans wrote:
> 
> Arne Mueller <a.mueller at icrf.icnet.uk> writes:
> 
> [snip]
> 
> > > def sigchild_handler(signum, frame):
> > >     os.wait()
> 
> [snip]
> 
> > Hm, that doesn't work properly  with my program. THe number of zombie
> > processes is steadily at about 20, which means there are 20 unhandled
> > died children.
> 
> Unix doesn't guarantee you'll get exactly one signal delivered per
> event.  To be sure to get all the children, you need to loop over the
> processes with waitpid().
> 
> I think something like this should do it:
> 
> ----------
> def sigchld_handler(signum, frame):
>     print "Checking for processes..."
>     try:
>         while 1:
>             pid, sts = os.waitpid(-1, os.WNOHANG)
>             if pid == 0:
>                 break
>             print "Process %d ended with status %04x." % (pid, sts)
>     except OSError, detail:
>         if detail.args[0] != errno.ECHILD:
>             raise
>         else:
>             print "No more children."
>     else:
>         print "Some children yet to finish."
> ----------

Hm, that's similar what I do in my program but I always have as many
zombies as I fork processes, here's a program fragment:

cpu = 4
used = 0
for fn in files:

        if cpu:
            if used < cpu: 
                r, w = os.pipe()
                pid = os.fork()
                used = used + 1
                if pid: # parent
                    os.close(w)
                    chfds.append(r)
                    chpids[r] = pid # dict. with filedescriptor as key
	            # fork children for available processors
                    if used < cpu and fn != last: continue 
                else:   # child
                    os.close(r)
                    param['files'] = [fn]
		    # some application specific stuff
                    msa = ProcessMSA(param)
                    msa.of = None
                    msa.dbkeys = {}
                    msa.param = {}
	            # pickle objecty and send it to parent
                    p = mpickle.dumps(msa)
                    r, w, o = select.select([], [w], [])
                    os.write(w[0], p)
                    os._exit(0)
	    # parent reads from ready file descriptors 
            while used >= cpu or (fn == last and used):
                r, w, o = select.select(chfds, [], [])
                for i in r:
                    msa = mpickle.loads(readPickleStream(i))
                    os.close(i)
                    stderr.write('number %d\n' % len(stats.ids.keys()))
                    stats.add(msa)  # do something with the received
object
                    used = used - 1 # adjust number of used processors
                    chfds.remove(i)
                    os.waitpid(chpids[i], os.WNOHANG)
                    del chpids[i]

Why does that code produce zombies? Do I've to send a signal to the
child that it has to die? I mean maybe ther child exits when the parent
hasn't reached the os.waitpid?

	Thanks for discussion,

	Arne



From neelk at brick.cswv.com  Wed Jun 23 21:56:41 1999
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 23 Jun 1999 20:56:41 -0500
Subject: total idiot question: +=, .=, etc...
References: <877lov62z1.fsf@pbl.ddns.org> <slrn7n0rr2.2fi.behrends@allegro.cse.msu.edu> <m3aetr2yk9.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <7ks38p$l4m$1@brick.cswv.com>

In article <m3aetr2yk9.fsf at atrus.jesus.cam.ac.uk>,
Michael Hudson  <mwh21 at cam.ac.uk> wrote:
>behrends at cse.msu.edu (Reimer Behrends) writes:
> 
>> Personally, I can easily live without a construct that I can use on
>> average once every three screens of code or so. On the other hand, I
>> _am_ getting pretty sick of having to type self.whatever countless
>> times, where pretty much any other language allows me to discard the
>> "self." part. Not only is it annoying to type, it also reduces the
>> readability of OO code a lot, much like the gratuitious use of
>> punctuation characters as variable prefixes in Perl does.
>
>I sometimes think this when writing Python; but in my summer job which
>I've just started, I'm frequently faced with large screenfulls of C++
>and not being able to tell whether a given variable is local, static,
>class or global at a glance is mightily annoying. 

I don't mind writing self.foo = bar, for the reason you describe, but
I know I hate having to write self.__class__.member = foo in order to
keep inheritance from breaking:

class Foo:
    x = 9
    def munge_x(self, x):
        Foo.x = x

class Bar(Foo):
    x = 12

>>> a = Foo()
>>> a.munge_x(15)
>>> print a.x
15

>>> y = Bar()
>>> y.munge_x(15)
>>> print y.x
12

But if I change Foo to:

class Foo:
    x = 9
    def munge_x(self, x):
        self.__class__.x = x

and try the same thing, I get the more intuitive:

>>> a = Foo()
>>> a.munge_x(15)
>>> print a.x
15

>>> y = Bar()
>>> y.munge_x(15)
>>> print y.x
15

It's just that writing self.__class__ is so *ugly*! Everything else
in Python is beautiful, too, which makes this stand out even more. 


Neel



From greg.ewing at compaq.com  Thu Jun 17 18:37:53 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 18 Jun 1999 10:37:53 +1200
Subject: NEWBIE: opaque error message
References: <14181.50740.800169.537860@lrz.de>
Message-ID: <376978C1.EBAEBFA3@compaq.com>

Eugene Leitl wrote:
> 
>   File "/usr/lib/python1.5/copy.py", line 221, in _deepcopy_inst
>     return x.__deepcopy__(memo)
> TypeError: too many arguments; expected 1, got 2

I suspect that the __deepcopy__ method of one of
your objects is missing a 'self' argument.

Greg



From kas at maps.magnetic-ink.dk  Tue Jun 22 03:58:55 1999
From: kas at maps.magnetic-ink.dk (Klaus Alexander Seistrup)
Date: Tue, 22 Jun 1999 09:58:55 +0200
Subject: newbie idiom question
References: <m3emj4yezq.fsf@monsoon.swcp.com>
Message-ID: <376F423F.E2EB9C4D@maps.magnetic-ink.dk>

Alex Rice wrote:

> I keep trying to do this:
> 
>       >>> eggs = [1,2,3]
>       >>> for spam in eggs:
>       ...       spam = 'cooked'
>       ...
>       >>> eggs
>       [1, 2, 3]
>       >>>
 
How about:

  >>> eggs = [1,2,3]
  >>> for i in range (len (eggs)):
  ...     eggs[i] = 'cooked'
  ... 
  >>> eggs
  ['cooked', 'cooked', 'cooked']
  >>>

?

  //Klaus

-- 
???[ Magnetic Ink ]?????????????????????????????????????????????????????????
???[ http://www.magnetic-ink.dk/ ]??????????????????????????????????????????



From jtravs at debian.org  Wed Jun 16 15:24:17 1999
From: jtravs at debian.org (John Travers)
Date: Wed, 16 Jun 1999 20:24:17 +0100
Subject: extention modules.
Message-ID: <99061620265802.00254@Vegus>

Hi, I am writing an extention module in C. Well I'm writing the low level (math
intensive) stuff in C and want to wrap it up in some native python classes for
a nice api. Should I write it as two seperate modules (the compiled C one and
the native python one) or is there a way to combine them in one. I am using the
SWIG program to wrap my C library.

Help apreciated.

--
John Travers

"Sometimes I think the surest sign that intelligent life exists elsewhere
in the universe is that none of it has tried to contact us!"



From pjc_2 at geocities.com  Thu Jun  3 14:50:52 1999
From: pjc_2 at geocities.com (Philip C.)
Date: Thu, 3 Jun 1999 13:50:52 -0500
Subject: need help with decompress
Message-ID: <66A53.34$a43.449@news.megsinet.net>

using the zlib module o found out how to compress springs but I having
trouble with decompressing them.
is there any way to scramble string sort of encrypt them





From Tim.Peters at p98.f112.n480.z2.fidonet.org  Tue Jun 29 14:57:04 1999
From: Tim.Peters at p98.f112.n480.z2.fidonet.org (Tim Peters)
Date: Tue, 29 Jun 1999 19:57:04 +0100
Subject: PySequence_Check
Message-ID: <000015a8@bossar.com.pl>

From: "Tim Peters" <tim_one at email.msn.com>

[Reuben Sumner]
> How can I do the equivalent of a PySequence_Check in Python?
> GvR told me that the question was illdefined (if I explained it correctly)
> and to ask here.  It seems that some of the functions in abstract.h
> are available from python (len(), callable()) and others are not
> (PySequence_Check, PyNumber_Check).
>
> What am I missing?

For starters, unquestioning faith in Guido's pronouncements <wink>.

> I have a nested structure of lists and tuple and I just want to know
> when I have bottomed out.

Then why not check for that directly and be done with it?

> I can check for tuple or list indivually but that is ugly and
> inflexible.

Not to mention simple, obvious and unsurprising <wink>.

> I could just try indexing and see if there was an exception but that
> is not much better.

OK, you tell me:  what *is* a sequence?  Answer that, and you'll know how to
check for it <0.5 wink>.  Before answering too fast, note that mapping types
can be indexed without raising an exception, and so can strings.  Do you
want your nest of "lists and tuples" *not* to "bottom out" at a string?
Probably not -- but if so, you're not really interested in testing whether
it's a sequence!  Note too that user-defined classes can support the
__getitem__ protocol without supporting the __getslice__ protocol, or vice
versa.  Are either of those sequences, or do they need to support both?
Whichever way you answer, half the world will disagree.

I think that, in most cases, when people say "sequence":  (A) it's
ill-defined <wink>; and, (B) after five rounds of tedious questioning they
end up deciding they mean "responds to slicing but isn't a string".  So
there you go:

def issequence(x):
    try:
        x[0:0]
    except:
        return 0
    return type(x) is not type("")

may-not-match-what-you-mean-by-"sequence"-ly y'rs  - tim






From tim_one at email.msn.com  Thu Jun 10 21:54:50 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 10 Jun 1999 21:54:50 -0400
Subject: Python 2.0
In-Reply-To: <37605213.2ACAE4D3@compaq.com>
Message-ID: <000301beb3ad$64a80680$329e2299@tim>

[Tim, quoting the Java Language Spec]
> "Therefore, we recommend that the design of finalize methods be 
> kept simple and that they be programmed defensively, so that they
> will work in all cases."

[Greg Ewing] 
> In my experience, this advice applies just as much to Python
> __del__ methods, their alleged greater predictability
> notwithstanding. 

I'll repeat my sig, which you cut off <wink>:

good-advice-in-any-language-ly y'rs  - tim





From catlee at my-deja.com  Tue Jun 29 15:15:36 1999
From: catlee at my-deja.com (catlee at my-deja.com)
Date: Tue, 29 Jun 1999 19:15:36 GMT
Subject: Arbitrary precision calculations
Message-ID: <7lb60c$ot1$1@nnrp1.deja.com>

Hi,

I'm looking for a package that will let me do arbitrary precision
arithmatic in python.  I've come across the real.py module written by
Jurjen N.E. Bos, and it seems like a really nice module except for the
fact that you can't compare two numbers for equality,inequality,etc.

Cheers,
Chris

P.S. I found the real.py package by doing a search on www.deja.com for
arbirary precision in comp.land.python


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From schorsch at schorsch.com  Wed Jun 16 08:39:45 1999
From: schorsch at schorsch.com (Georg Mischler)
Date: Wed, 16 Jun 1999 12:39:45 GMT
Subject: do anonymous pipes normally work on NT?
References: <7juatc$fef$1@nnrp1.deja.com> <7k43sk$6vd$1@nnrp1.deja.com> <7k7i1g$87d$1@m2.c2.telstra-mm.net.au>
Message-ID: <7k85ug$j12$1@nnrp1.deja.com>

Mark Hammond wrote:
> Im afraid I can't be of much help here.  I have had a good play,
> and nothing seems obvious.
[...]
> If you are really keen, the next step would probably be to recode
> this small sample in C, and see if that works.  If it does, Python
> (or the win32 extensions) has a problem we should fix as a matter of
> priority.  If not, then we need to find the magic words to chant to
> make it work from C, and move that back to Python...


Thanks for your efforts.

Based on my own further experiments I allmost expected to hear
this, and I'm starting to get mentally prepared to check with
a few lines of C. In the extreme case, I'd actually have to
rewrite it as an extension module since I really need the
functionality.
But of course, finding a mistake of mine, or seeing python resp.
the win32 extensions fixed in case there should be a real problem,
would be highly preferrable.

I'll be back with my further findings.


Thanks

-schorsch


--
Georg Mischler  --  simulation developper  --  schorsch at schorsch.com
+schorsch.com+  --  lighting design tools  --  http://www.schorsch.com/


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From tim_one at email.msn.com  Fri Jun  4 21:42:11 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 4 Jun 1999 21:42:11 -0400
Subject: Python 2.0
In-Reply-To: <199906042125.XAA12398@axil.hvision.nl>
Message-ID: <000901beaef4$a239c1e0$f19e2299@tim>

[Kumar Balachandran]
> ... why not have optional syntactic sugar such as
> if
> ...
> elif
> ...
> else
> ...
> fi
>
> ...

[Hans Nowak]
> If you think it's more elegant, why not try things like this:
>
> if blahblah:
>     do_this()
>     do_that()
> #fi
>
> for x in my_sequence:
>     print 
>     print booze.append(x)
> #rof
>
> Or use #endif, #endfor, etc.
> ...

See pindent.py in the distribution's Tools/Scripts/ directory.

it's-dying-from-loneliness-ly y'rs  - tim





From graham at sloth.math.uga.edu  Wed Jun 16 20:46:30 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 17 Jun 1999 00:46:30 GMT
Subject: Solution to finalisation problem [Re: fork()]
References: <000301beb6c9$a01d2960$b2a22299@tim> <7k4gec$6vg$1@cronkite.cc.uga.edu> <37683DD6.4C42DC3E@compaq.com>
Message-ID: <7k9gh6$ouh$1@cronkite.cc.uga.edu>

Graham Matthews wrote:
: > All
: > objects will be in a state in which they can be finalised in pass 1 since
: > nothing is deleted in pass 1. I don't see the problem here at all.
Greg Ewing (greg.ewing at compaq.com) wrote:
: The only way I can see that the 2-pass collection scheme
: can fail is if the finaliser of object A destroys something
: needed by the finaliser of object B. But if each object
: having a finaliser looks after just one external resource,
: how can that happen?

What do you mean "destroy" here? 

As I am sure Tim will tell you the only way a 2 pass collector can get 
into trouble is if the finaliser for A changes some object O used by the
finaliser for B. In this case the ordering in which you run finalisers
is important. In other words a 2 pass collector solves the "which order
to finalise in" problem as long as finalisers don't change objects
used by other finalisers.

But here is where semantics comes into play. What do we want finalisation
to do? In my opinion finalisation should be *defined* as follows: all
objects collected by the collector are finalised *simultaneously*.
In implementation speak this means that objects are finalised relative to 
the system state at the *beginning* of the finalisation process. This is 
also relatively straighforward to implement I think. Alternatively we
can simply decide that finalisers that play with the state of other
finalisers are "wrong" programs and hence not worry about them. This is
much like the way Python treats threads playing with shared data (it's
up to the programmer to ensure that such code runs in a determinate
manner). Personally I think either solution is fine, since I do believe
that finalisers playing with the state of other finalisers is very rare,
and also "wrong" code.

graham


-- 
           As you grow up and leave the playground
       where you kissed your prince and found your frog
          Remember the jester that showed you tears
                   the script for tears



From pduffin at mailserver.hursley.ibm.com  Mon Jun 28 08:49:18 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Mon, 28 Jun 1999 13:49:18 +0100
Subject: Embedded Tcl for Python
References: <lJT73.2677$nn.808830@news.shore.net>
	 <3760F0A1.1DDC@mailserver.hursley.ibm.com> <B57RNPASzUY3EwGq@jessikat.demon.co.uk>
Message-ID: <37776F4E.6113@mailserver.hursley.ibm.com>

Robin Becker wrote:
> 
> In article <3760F0A1.1DDC at mailserver.hursley.ibm.com>, Paul Duffin
> <pduffin at mailserver.hursley.ibm.com> writes
> >Michael P. Reilly wrote:
> >>
> >> A few weeks ago, someone asked if there was a Tcl module for Python.
> >> People responded (including myself) that Tkinter uses a Tcl/Tk
> >> interpreter, but you couldn't get rid of the annoying window.
> >>
> >> I've since written an extension ("pytcl") that creates Tcl interpreter
> >> objects.  As part of that, I have also rewritten _tkinter to use these
> >> new objects.
> >>
> >> I've had a short discussion with Guido and he's interested in the
> >> idea.  Now I need some alpha testers; I only have access to UNIX boxes,
> >> and very few types at that.
> >>
> >> The _tkinter module has the same functionality as before, including the
> >> same threading issues.  I've include Deiter's Tcl/Tk8.1 patch but have
> >> not had the opportunity to test that yet (right not it is just do-no-
> >> damage).
> >>
> >> Contact me at arcege at shore.net if you are interested in helping me test
> >> these modules.
> >>
> >
> >I was interested in doing something similar in Tcl (e.g. creating a
> >python interpreter object) and I was wondering if someone could give
> >me some pointers as to how to go about doing it.
> >
> Paul I created a socket linked python interpreter and have used it with
> reasonable success from Tcl. Latest code is at
> 
> http://www.jessikat.demon.co.uk in the file pyserver.zip
> 
> this may not be as efficient as havin the in core thing, but gets round
> the problem of threading model problems etc.

Thanks Robin, I will take a look at that, have you got any information
on the threading model problems as I have not yet looked into Python
threading. I must admit I was amazed that Python does not support
being built as a shared library and can think of no practical reason why
not as Python does support dynamically loadable extensions.

Have any Pythoneers out there done this as it is very useful ?
If you have any questions as to how useful it is you just need to look
at Tkinter which could not be dynamically loadable (it is isn't it)
without Tcl and Tk being dynamically loadable.


-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From jimbag at kw.igs.net  Wed Jun  2 12:44:01 1999
From: jimbag at kw.igs.net (JB)
Date: Wed, 02 Jun 1999 16:44:01 +0000
Subject: SWIG'in python question
References: <37542E8B.ABEAA531@kw.igs.net> <19990602123127.B280136@epa.gov>
Message-ID: <37555F51.53A7E84F@kw.igs.net>

Randall Hopper wrote:
> 
> JB:
>  |I've got a .so that I built with swig for python. Everything builds fine
>  |but when I import i get this message from python "_mymod.so: undefined
>  |symbol: __eh_pc". I wondered if anyone had seen this before and could
>  |give me a clue. I think I'm missing a library somewhere. I'm building on
>  |RedHat 5.2.
> 
> Does your code in _mymod.so depend on any user code another .o/.so?
> 
> I'll assume not.  That being the case, it's probably a symbol defined in
> some system library.  Use "nm" to find out which one, then add a "-l"
> reference to it on the link line (the command line where you create
> _mymod.so).
> 
> Check your man page for nm for the right options to print the filename and
> the symbol on one line.  For example:
> 
>     (FreeBSD:)   nm -o  /usr/lib/*.so* | grep <mysymbol>
>     (IRIX:)      nm -Bo /usr/lib/*.so* | grep <mysymbol>
>     (Solaris:)   nm -Rh /usr/lib/*.so* | grep <mysymbol>
> 
> Randall

It turned out to be in libstdc++. I didn't know about nm tho. Thanks for
that tip.

cheers
jim



From news at dorb.com  Mon Jun 28 13:11:54 1999
From: news at dorb.com (Darrell)
Date: Mon, 28 Jun 1999 13:11:54 -0400
Subject: Python for large projects?
References: <3774b5d1.690922@news.force9.net>  <T6bd3.2883$U6.12014@newsr2.twcny.rr.com> <slrn7naup9.jk.thantos@brimstone.mecha> <RGid3.2957$U6.12374@newsr2.twcny.rr.com> <87n1xm85mf.fsf@lisboa.i-did-not-set--mail-host-address--so-shoot-me> <Jhsd3.3090$U6.12410@newsr2.twcny.rr.com> <3776EF65.A07CBEFC@compaq.com> <slrn7ne533.1rmr.marc@oscar.noc.cv.net> <xeLd3.3224$U6.13605@newsr2.twcny.rr.com>
Message-ID: <q6Od3.3254$U6.12368@newsr2.twcny.rr.com>

> I'm glad I brought this up because Michael Hudson suggested using RExec.
> This sounds like a perfect solution to my concerns. Even better than C++
> const.
>
> --Darrell
>

Read up on RExec and don't think it's the solution. Thanks anyway. I had
been meaning to learn about RExec.
--Darrell





From marcoxa at copernico.parades.rm.cnr.it  Tue Jun 29 12:17:55 1999
From: marcoxa at copernico.parades.rm.cnr.it (Marco Antoniotti)
Date: 29 Jun 1999 18:17:55 +0200
Subject: Why is tcl broken?
References: <1283099377-77267484@hypernet.com> <lw67471fyi.fsf@copernico.parades.rm.cnr.it> <7laoq0$bvh$1@Starbase.NeoSoft.COM> <lwso7bdnia.fsf@copernico.parades.rm.cnr.it> <7laqeb$e17$1@Starbase.NeoSoft.COM>
Message-ID: <lwr9mvdlqk.fsf@copernico.parades.rm.cnr.it>

claird at Starbase.NeoSoft.COM (Cameron Laird) writes:

> >Of course I could have defined a much more intricated macro.
> 			.
> 			.
> 			.
> As it happens, there's very good work going on just now
> to beef up Tcl's introspective capabilities.  Is *that*--
> introspection--the real content of your preference?

Nope, you must correct me if I am wrong, but in Tcl, you set up a new
'form' by writing a "proc".  This "proc" takes some Tcl 'list' (with
strings as leaves) and produces another list (with strings as leaves).

When "proc" is used within some code, its code is executed at
runtime.  This is not what happens in Lisp, Scheme and Dylan, where
macros are evaluated at read-time.  This is pure code transformation,
which when fed in the Common Lisp *compiler* produces inlined code.

Introspection is not something I was very interested in when making my
point. I just wanted to point out that with Common Lisp (and Scheme, and
Dylan) you get a native compiler.

If we want to talk about introspection, have you ever seen a Common
Lisp inspector?  It is based on a lot of "introspective" functions,
which were present in Common Lisp in 1984, 15 years ago.

> Incidentally, many of the same ideas and possibilities
> are available to Python, although I don't know of anyone
> actively pursuing them for Python.  While Tim Peters im-
> presses me with, among much else, his ability to code
> clever little methods that tease all sorts of informa-
> tion from a Python interpreter, I occasionally argue that
> 'twould be worth the effort to do introspection for Py-
> thon in a more unified way.

As Common Lisp does? (sorry, I couldn't resist :) )

Cheers

-- 
Marco Antoniotti ===========================================
PARADES, Via San Pantaleo 66, I-00186 Rome, ITALY
tel. +39 - 06 68 10 03 17, fax. +39 - 06 68 80 79 26
http://www.parades.rm.cnr.it/~marcoxa



From tismer at appliedbiometrics.com  Tue Jun  8 12:01:06 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Tue, 8 Jun 1999 16:01:06 GMT
Subject: Using PyArg_VaParse
References: <7jj6hr$l1g$1@nnrp1.deja.com>
Message-ID: <375D3E42.E3264E31@appliedbiometrics.com>


bhosu at my-deja.com wrote:
> 
> Hi All,
>    I want to use PyArg_VaParse to retrieve variable number of arguments.

Why? See below.

> I am not sure if this is the correct function to use. I have embedded
> python in a C++ application and implemented python methods in C++. I
> have a situation where, the method may pass any number of string
> parameters. I would like to get these parameters in my C++ code. Here is
> a scenario of the whole thing.
> 
> (C++ implementation.. for method bar in the module foo.)
> 
> static PyObject *bar(PyObject *self,PyObject *args)
> {
> ....
> if(PyArg_VaParse(args,...) /* What is that I need to use here...? */
> ...
> }

The PyObject *args is already a Tuple, so the only thing
you need to do is to check its length and if
the parameters are stings indeed.

> 
> and a pythod call may be like,
> import foo
> foo.bar('a','b','c')
> foo.bar('a','b','c','d'.....)
> 
> I need to know the Python API call to use to get the arguments. As I
> said before I donot know how many arguments are being passed.

Just use PyTuple_Size(args), in a loop
use PyTuple_GetItem(args, i)
and check them each with PyString_Check.

good luck - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From pvelikho at cs.ucsd.edu  Mon Jun 21 20:06:10 1999
From: pvelikho at cs.ucsd.edu (Pavel Velikho)
Date: Mon, 21 Jun 1999 17:06:10 -0700
Subject: need some fast balanced-tree datastructure
Message-ID: <376ED372.FA73994F@cs.ucsd.edu>

Hi,
    I am looking for a fast implementation of incremental container that

maintains an ordering on the keys... (i.e. I want an AVL, Red-Black, or
anything similar). I have tried the b+tree in Python but its not fast
enough.
Is there anything like that available as a C extension module?

Thank you
Pavel Velikhov
pvelikho at cs.ucsd.edu




From ronERASEjeffries at ERASEacm.org  Sat Jun 19 18:05:54 1999
From: ronERASEjeffries at ERASEacm.org (Ronald E Jeffries)
Date: Sat, 19 Jun 1999 22:05:54 GMT
Subject: [JPYTHON] Installation Problems Help Please
References: <0D1FF10FB191D45F.A19ABABF1E01F5C3.C3A5123AAD9506A3@lp.airnews.net>
Message-ID: <13F9B85A54E3C24C.7F4273C4E5CA0DA3.A6A7F3694B2974E0@lp.airnews.net>

On Sat, 19 Jun 1999 01:24:32 GMT, ronERASEjeffries at ERASEacm.org
(Ronald E Jeffries) wrote:

>Thought I'd try XP in Python, so I just downloaded JPython1.1beta2 and
>when I installed it with JView, got this:
>
>D:\Downloading\jpython>jview JPython11beta2
>InstallShield JavaTM Edition
>Extracting installation code................................done
>Failed to execute MakeScriptAction
>
>The files all appear to be in the directory I chose. There's a
>"jpython", and a "jpython.jar".
>
>However, when I type jpython at the DOS prompt, I get the dreaded
>
>D:\Downloading\jpython>jpython
>Bad command or file name
>
>I'm supposing this is because the MakeScriptAction didn't run.
>
>Suggestions welcome!
>
>Also ... is there an IDE of any kind available for Python yet ... or
>at least something that will get me out of the DOS box?
>
>Thanks!
>
>Ron Jeffries
>Extreme Programmer
>http://www.armaties.com


Ron Jeffries
Extreme Programmer
http://www.armaties.com



From tre17 at cosc.canterbury.ac.nz  Tue Jun 15 23:09:49 1999
From: tre17 at cosc.canterbury.ac.nz (Timothy R Evans)
Date: 16 Jun 1999 15:09:49 +1200
Subject: Tk Scrollbar
References: <37663C34.2957622F@dlr.de>
Message-ID: <c75ogig264i.fsf@pc142.cosc.canterbury.ac.nz>

Sven Drescher <Sven.Drescher at dlr.de> writes:

> Hi!
> I have only an easy question. I think it is possible to bind a scrollbar
> and an other widget e.g. a frame with buttons. But I have not found any
> literature or programming examples. I want to bind the scrollbar to an
> frame and I don't know how. Please help me.
> Thanks for links and Python-code or other help.
> 
> Sven

Find the pmw (python mega-widget) module, it has a Pmw.ScrolledFrame
widget which I think will do what you want.  I haven't used it myself
but the documentations says is works.

Pmw homepage is:

http://www.dscpl.com.au/pmw/

--
Tim Evans




From tim_one at email.msn.com  Fri Jun  4 01:15:11 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 4 Jun 1999 01:15:11 -0400
Subject: threads
In-Reply-To: <8790a1njeu.fsf@pc-hrvoje.srce.hr>
Message-ID: <001001beae49$389b4ca0$949e2299@tim>

[Arne Mueller]
> Hi All, I'm using the threading module in python 1.5.2b2. I'm
> wondering wheather the different threads are all executed in
> parallel on a multiprocessor machine.

[Jeremy Hylton]
> The short answer is no.  The Python interpreter has a lock that
> prevents allows only a single thread to execute at a time.

[Hrvoje Niksic]
> Eek.  That's depressing.  Isn't supporting threads and not allowing
> simultaneous execution a bit weird?

The serialization only applies to those threads trying to interpret Python
bytecodes at a given time; they have to take turns then, but can run off in
parallel while doing anything else (typically doing I/O, or off in an
extension module with its own ideas about threads).

There are a mass of hairy tradeoffs, and for Python 1.4 there exists a large
set of "free threading" patches (by Greg Stein) that removed the global
lock.  Some good news, some bad news, but old news all the same <wink>.

Oh ya -- in answer to your question, yes!  It is a bit weird.

what-does-that-make-us?-ly y'rs  - tim





From aahz at netcom.com  Wed Jun 23 16:21:56 1999
From: aahz at netcom.com (Aahz Maruch)
Date: Wed, 23 Jun 1999 20:21:56 GMT
Subject: help?  strptime ....
References: <rNHvjEdhm5Pp-pn2-dCsXnyPPztJy@d185d18f8.rochester.rr.com>
Message-ID: <aahzFDsr8L.MI5@netcom.com>

In article <rNHvjEdhm5Pp-pn2-dCsXnyPPztJy at d185d18f8.rochester.rr.com>,
Benjamin Schollnick <rfreedm1 at nospam.rochester.rr.com> wrote:
>
>	I'm trying to use strptime (from the time module) and can't....

What version of Python?  I'm getting AttributeError in 1.5.1, so I
suspect it's only in 1.5.2.  I'll bet mxDateTime has what you need,
though, if you don't want to upgrade.
-- 
                      --- Aahz (@netcom.com)

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

Disciple of Uncle Chuck



From news at dorb.com  Wed Jun  2 20:31:19 1999
From: news at dorb.com (Darrell)
Date: Wed, 2 Jun 1999 20:31:19 -0400
Subject: [Off topic] Weird...
References: <l03102809b37b3f90b2a0@[193.78.237.145]> <37559F04.2062C65D@callware.com>
Message-ID: <m4k53.1873$j7.29484@newsr2.twcny.rr.com>

The video is often shown on ZDTV.
Some guy in a cubical gets mad at his computer and trashes it.

--
--Darrell
Ivan Van Laningham <ivanlan at callware.com> wrote in message
news:37559F04.2062C65D at callware.com...
> Sorry, but it's not worth it to me.  ;-)  Maybe you could give us a
> blow-by-blow description, Just?
>






From gmcm at hypernet.com  Thu Jun  3 17:43:30 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 3 Jun 1999 16:43:30 -0500
Subject: COM server error
Message-ID: <1283676036-42579634@hypernet.com>

COM wizards:

 I have a Python class acting as a COM server. It uses CLSCTX_SERVER 
and uses the following code in __init__ to become a "singleton":

self.handle = pythoncom.RegisterActiveObject(
   win32com.server.util.wrap(self), self._reg_clsid_, 1)	

Things work wonderfully. For a while. Then my machine (NT 4 SP4) 
starts acting like a W9x box: the python process that has the COM 
server open won't respond and can't be killed. No sign in Task Mgr of 
the pythonw process running the COM server. Machine won't reboot. Has 
to be powered off (just like an HPUX box with a bad NFS config 
-hee hee!). The event log has the following message:

    The server {7C43A261-ABB7-11D2-93E1-00609736B700} did not 
    register with DCOM within the required timeout. 

I'm afraid I'm still on 1.52b2 and build 123. If you say this is 
fixed in build 124, I'll provisionally believe you <wink>.

The code is complex, but since I started from one of Andy's samples, 
I would imagine I could recreate the problem in something much 
smaller, if needed. (I also searched thru a long thread about 
creating COM singletons but couldn't find a resolution - I chose this 
method based on my understanding of Don Box's comments.)

- Gordon



From befletch at my-deja.com  Sat Jun 26 01:30:16 1999
From: befletch at my-deja.com (Bruce Fletcher)
Date: Sat, 26 Jun 1999 05:30:16 GMT
Subject: [ANN] DejaGrabber 0.1 -- search Dejanews with Python
References: <7kuu73$1ki$1@brick.cswv.com> <7l0520$qe4$1@news.wrc.xerox.com>
Message-ID: <7l1oh9$ol4$1@nnrp1.deja.com>

In article <7l0520$qe4$1 at news.wrc.xerox.com>,
  "Val Bykoski" <vbykovsk at sdsp.mc.xerox.com> wrote:
> Neel:
> This looks really great..
> But I have this:
>
> Traceback (innermost last):
>   File "<interactive input>", line 1, in ?
>   File "C:\Program Files\Py152\Lib\httplib.py", line 51, in __init__
>     if host: self.connect(host, port)
>   File "C:\Program Files\Py152\Lib\httplib.py", line 79, in connect
>     self.sock.connect(host, port)
>   File "<string>", line 1, in connect
> error: (10065, 'winsock error')
>
> when i am trying to reach www.python.org thru a proxy (www.mc.xerox.com).
> How to handle this proxy-related situation?
>
> Thanx -- val

Well, I can't access www.mc.xerox.com; presumably it is Xerox internal
use only.  I was wondering if this machine would be a Windows box by any
chance, possibly running WinProxy?  Your problem looks exactly like one I
had with WinProxy on an NT box.  WinProxy is slightly broken.  The
authors intend to fix it in their next release, but for now you would
need to add a workaround (kindly provided to me by a reader of this
group) to your Python libraries.  Unfortunately I don't have the fix
here, it is on my firewalled machine at work.  You could search for the
fix in this group; just search for WinProxy.

If this isn't your problem, I guess it is something else...

Good luck,
- Bruce


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From Angus.MacKay at p98.f112.n480.z2.fidonet.org  Wed Jun 30 01:43:48 1999
From: Angus.MacKay at p98.f112.n480.z2.fidonet.org (Angus MacKay)
Date: Wed, 30 Jun 1999 06:43:48 +0100
Subject: ioctl in python
Message-ID: <000015d3@bossar.com.pl>

From: Angus MacKay <amackay at starvision.com>

well it seems I have sort of solved my problem. if I use a
string for the 3rd arg then it works properly:
>>> fo=open('/tmp')
>>> fcntl.ioctl(fo.fileno(), 0x80047601, "hi")
'\001\000'
>>> fo=open('/')
>>> fcntl.ioctl(fo.fileno(), 0x80047601, "hi")
'\000\000'
>>> fo=open('/proc')
>>> fcntl.ioctl(fo.fileno(), 0x80047601, "hi")
Traceback (innermost last):
  File "<stdin>", line 1, in ?
IOError: (25, 'Inappropriate ioctl for device')

the question is why? it will still fail with only 2 args or an int:
>>> fcntl.ioctl(fo.fileno(), 0x80047601)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
IOError: (14, 'Bad address')
>>> fcntl.ioctl(fo.fileno(), 0x80047601, 1)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
IOError: (14, 'Bad address')

cheers, Angus.

Angus MacKay wrote:
> 
> is there some magic to the fcntl.ioctl() call?
> 
> I can do these ioctl calls in C no problem (with a small wrapper):
> (amackay at phat)~/tmp$ ./ioctltest /tmp 0x80047601
> 0x80047601: 1
> (amackay at phat)~/tmp$ ./ioctltest / 0x80047601
> 0x80047601: 0
> (amackay at phat)~/tmp$ ./ioctltest /proc 0x80047601
> /proc: Inappropriate ioctl for device
> 
> that was the Linux ioctl for EXT2 version.
> 
> but in python:
> >>> a = 0
> >>> req = 0x80047601
> >>> import fcntl
> >>> fo=open('/tmp')
> >>> fcntl.ioctl(fo.fileno(), req, a)
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> IOError: (14, 'Bad address')
> >>>



From gmcm at hypernet.com  Tue Jun  1 11:01:43 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 1 Jun 1999 10:01:43 -0500
Subject: Python 2.0
In-Reply-To: <87k8toqv70.fsf@ev.netlab.co.jp>
Message-ID: <1283872943-30734874@hypernet.com>

Yukihiro Matsumoto wrote:
> 
> "Vadim Chugunov" <chega_ at yahoo.com> writes:

> |I'm sorry for being unclear, but the key words here were: "to free up
> |resources".
> |And I meant "to free up resources the moment I expect them to", not "some time
> |later".
> 
> Well, OK.  I understand now.  Now only thing I can say is, why do
> you emphasize so much on reclaiming object at the time you exepcted?

As someone used to GC, you would probably never write:

 txt = open('blahbalh.txt','r').read()

This is a common idiom for those of us used to ref counting.
You would certainly take a reference to the file object so you could 
explicitly close it. 

On the other hand, you don't worry about circular references, while 
we train ourselves to recognize them and take corrective action.

> Because of your C++ background?

Ironically, C++ can beat the pants off anything else in this regard. 
With smart pointers, you can handle ref counts automatically, even 
when there are circular references.  And since you have the choice 
between heap and stack, you can be absolutely certain that you 
release resources, even when an exception is thrown.

> |BTW, this is a common expectation from anyone coming from a C++ background.
> |In C++, deallocation in destructors is the only way you can use exceptions
> |without leaking resources.
> 
> Yes, but it's not true for HIGHER languages like Python or Java (or
> Ruby).

In this regard, they're a step backwards.

- Gordon



From larsga at ifi.uio.no  Sun Jun 20 07:02:32 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 20 Jun 1999 13:02:32 +0200
Subject: BiBTeX module?
References: <joe-160619991358444033@chinacat.salk.edu> <7k950a$hsb@cs.vu.nl> <slrn7miqcg.tpi.bernhard@alpha1.csd.uwm.edu> <slrn7mk9ui.dic.gobry@blanc.idiap.ch> <wk1zf9wo5y.fsf@ifi.uio.no> <slrn7moad8.4l5.bernhard@alpha1.csd.uwm.edu>
Message-ID: <wkzp1vt9rr.fsf@ifi.uio.no>

* Bernhard Reiter
|
| What do you mean with XML version?

I mean an XML vocabulary (or DTD, if you like) for marking up BibTeX
entries.
 
| Somebody has to write a good DTD first and then XML doesn't have
| TeX's typesetting capabilities.

That's true, but that's not what's interesting about XML either. The
point about XML is that you can describe the data (the references)
precisely, in a way that makes it easy to write software that works
with them.

As for the good DTD, anyone can write that. :)

| Often I find myself using them to really typeset the References
| precisely.

If that's the only think you ever do, then BibTeX is enough.
 
| If you want something like a .bib format in XML you might have to
| invent the needed typesetting rules, too.

No, all you need is to be able to convert back to .bib. The point is
that XML is easier to work with when you want to do something else
with your data than to typeset them.
 
--Lars M.



From tim_one at email.msn.com  Thu Jun 24 22:32:31 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 24 Jun 1999 22:32:31 -0400
Subject: Windows - Drag & drop on Python script?
In-Reply-To: <3772D23E.10C89EC@compaq.com>
Message-ID: <000101bebeb2$f9c75940$d29e2299@tim>

[Greg Ewing]
> ...
> Another question - what is the recommended way of
> getting the effect of a unix #!? I came up with:
>
> python -x %0 %1
> exit
> <python code here>
>
> which only works by the serendipitous fact that
> there is an 'exit' defined in the python builtins.
> Without the exit, after the python finishes, the
> shell goes on to try and interpret it as msdos
> commands!
>
> Is there a more elegant way to do this?

Dale Nagata has a cute one-liner for this; see

    http://www.deja.com/getdoc.xp?AN=465164252

The bug he reported there (tracebacks are off by a line when using -x)
didn't get fixed in time for 1.5.2, alas; it is fixed in the development
version.  Note too that parts of Dale's method require that you're running
NT and have "cmd extensions" enabled (which is the default, so don't worry
about it if it doesn't make sense).

> Another other question - what is the "Batch File"
> box in the Properties dialog of a .bat file for?
> If you put the name of another .bat file there,
> it gets executed before the main file. But what
> is its purpose? Just curious.

It's more useful for certain kinds of .exe's, where it can be convenient to
run a .bat file to fiddle system properties before the .exe gets run.
Usually see it only in programs with deep Windows backward-compatibility
problems.

but-still-useful-to-e.g.-delete-copies-of-perl.exe-from-your-disk-ly y'rs
    - tim





From mryan at localhost.netaxs.com  Sun Jun  6 11:50:59 1999
From: mryan at localhost.netaxs.com (Michael W. Ryan)
Date: 6 Jun 1999 15:50:59 GMT
Subject: pyscript help
References: <BRS53.233$h14.1175@newsr2.twcny.rr.com> <Tlo63.546$h14.2355@newsr2.twcny.rr.com>
Message-ID: <slrn7ll5vs.14q.mryan@localhost.netaxs.com>

On Sun, 6 Jun 1999 02:12:15 -0400, Darrell <news at dorb.com> wrote:
>
>I just wanted to see an ASP something or other work.
>
>Replaced comcat.dll with an older one.
>NT4 SP5
>IE5
>And I'm afraid to install option pack 4.

Don't be afraid.  Just don't go overboard.  Of course, the recommended
order is:

    1) Install NT
    2) Apply SP3 (Yes, SP3)
    3) Install Option Pack
    4) Apply SP4+ (either SP4 or SP5)

The reason for this is that SP4+ modifies some of the items that the
Option Pack installs/modifies/replaces.

-- 
Michael W. Ryan, MCP          | OTAKON 1999
mryan at netaxs.com              | Convention of Otaku Generation
http://www.netaxs.com/~mryan/ | http://www.otakon.com/

PGP fingerprint: 7B E5 75 7F 24 EE 19 35  A5 DF C3 45 27 B5 DB DF
PGP public key available by fingering mryan at unix.netaxs.com (use -l opt)



From arcege at shore.net  Thu Jun 24 10:40:37 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Thu, 24 Jun 1999 14:40:37 GMT
Subject: Embedding Question
References: <slrn7n4699.g36.tuna@delirious.fsnnet>
Message-ID: <Frrc3.63$Of3.12274@news.shore.net>

Lloyd Weehuizen <tuna at delirious.fsnnet> wrote:
: Hi

: If I use the PyRun_String method, with a string that defines a method called
: "xyz", is there any way to remove this method from the Python interpreters
: "memory" without doing a Finalize() and Initialize(), because I am running
: multiple Python scripts, and I don't want there methods getting all mixed
: up, and I presume Finalize() and Initialize() are not very speedy ;)

: Thanks
: Lloyd 
: Lloyd 

Hi Lloyd,

You can use any number of the "del" based operations.  But from the
high-level (using the PyRun_* functions), you would want:
  
  int delete_function(funcname)
    char *funcname;
    { char buf[500];

      sprintf(buf, "if callable(%s):\n  del %s", funcname, funcname);
      return PyRun_String(buf);
    }

This way you can give it dotted (qualified) function names if they happen
to be in another module.

If you really wanted to get fancy, you could dynamically create modules
and run code within them.

  -Arcege




From landrum at foreman.ac.rwth-aachen.de  Sat Jun 19 06:18:46 1999
From: landrum at foreman.ac.rwth-aachen.de (Gregory A. Landrum)
Date: 19 Jun 1999 12:18:46 +0200
Subject: Tkinter & socket question
References: <19990618215411.23898.qmail@nwcst277.netaddress.usa.net>
Message-ID: <sogic333t.fsf@foreman.ac.rwth-aachen.de>

>2) I need to poll the connection periodically, say, every second. This would
>not have been a problem in console mode... I don't know how to do this in a
>Tkinter environment though. So, how can I do something every second (or every
>n milliseconds, or whatever the time span may be)? Are threads the way to go?
>I haven't used them before.

Wow, a question I can answer.... amazing.
Obligatory-Mandatory-Necessary-Essential Disclaimer:  I bear no
resemblance whatsoever to a python wizard, I just happen to have done
something like what you want to do.  There's almost definitely a
better way of doing this, but my method seems to work.

I assume that you have already opened a nonblocking socket connection,
and that that is what you are polling.  Your main Tkinter window has a
method called "after" which can be used to do the polling.
Documentation can be found at:
http://www.pythonware.com/library/tkinter/introduction/x8550-alarm-handlers-and-other.htm

An example, in my stupid little networked python game, I have the
following polling function:
    def ListenForMessages(self):
        msg = None
        theConn = None
        
	# Check to see if we are the client or the server
        if self.serverActive == 1:
            if self.clientConn is None:
                print "No Client!"
                return
            theConn = self.clientConn
        elif self.clientActive == 1:
            theConn = self.socket

        if theConn is None: return

	# see if there is anything there to read
        try:
            msg = theConn.recv(1024)
	    # Got it!  Dispatch the message.
            self.Dispatch(self.message.Type(msg),msg)
        except error:
	    # nope... no one loves us.
            pass

	# re-register the after callback, with a 200msec delay
        self.listenID = self.Game.rootWin.after(200,self.ListenForMessages)

The last line is important, because the after callback is removed
after being called.  So it needs to be re-registered with Tk.

The callback is first registered immediately upon establishing the
socket connection.


I hope this helps,
-greg

-- 
---------------------
Dr. Greg Landrum  (landrumSPAM at foreman.ac.rwth-aachen.de)
Institute of Inorganic Chemistry
Aachen University of Technology





From Mark.Hammond at p98.f112.n480.z2.fidonet.org  Wed Jun 30 15:00:20 1999
From: Mark.Hammond at p98.f112.n480.z2.fidonet.org (Mark Hammond)
Date: Wed, 30 Jun 1999 20:00:20 +0100
Subject: Python, ASP - Namespace (again)
Message-ID: <0000159b@bossar.com.pl>

From: "Mark Hammond" <MHammond at skippinet.com.au>

Nikolai Kirsebom wrote in message ...
...
>when I run the statements in the script1.asp file in an interactive window,
>I can create the KSdebatt object, however when I run the script on the
>server - I  get the following message:
>
>Python ActiveX Scripting Engine error '80020009'
>Traceback (innermost last): File "<Script Block >", line 4, in ?
>o=KSdebatt() NameError: KSdebatt
>/ksdebatt/pages/Script1.asp, line 6
>o=KSdebatt()
>
>If it is possible I would like to 'hide' the object (o) inside the KSdebatt
>module - that is to have the instanciation inside the module itself - but
>still be able to access it from the ASP-code in the script (o.Attr....).
Is
>this possible ?

There is no obvious reason to me why this wont work.  Note that if you are
not running win32all-125, then I recommend you upgrade - there have been
some fixes (but none that adddress this, as far as I can tell).

The best I can suggest is that you use the win32trace module and print
statements to work out what the hell is going on.

Eg, making your first line "import win32traceutil" will mean that all print
statement go to a debugging collector.  You can view the print statements by
starting Pythonwin on the same machine, and selecting "Tools/Trace Collector
Debugging Output".

Then use lots of print statements - specifically print dir() etc, to see
exactly what the namespace is...

Mark.





From rfreedm1 at nospam.rochester.rr.com  Sat Jun 26 12:34:01 1999
From: rfreedm1 at nospam.rochester.rr.com (Benjamin Schollnick)
Date: Sat, 26 Jun 1999 16:34:01 GMT
Subject: ConfigParser.py?
Message-ID: <rNHvjEdhm5Pp-pn2-ob9OW6dCFSRN@d185d18f8.rochester.rr.com>

Folks,

	I'm trying to use ConfigParser... And I believe I'm doing everything
write, except I don't see any documentation (including in the 
configparser.py file)
telling my what function(s) to use to write the file out to disk???

	Can someone give me a example or syntax on that?

	I've tried the atypical .CLOSE & .WRITE syntax's but had no luck...

			Benjamin



From tjreedy at udel.edu  Tue Jun 22 02:32:54 1999
From: tjreedy at udel.edu (Terry Reedy)
Date: 22 Jun 1999 06:32:54 GMT
Subject: choosing random numbers with weights/probability?
References: <929987267.6227@www.remarq.com> <lbwb3.3$V4.11640@news-dal.corridex.com> <ERwb3.1452$U6.5885@newsr2.twcny.rr.com>
Message-ID: <7knamm$kas$1@news.udel.edu>

In article <ERwb3.1452$U6.5885 at newsr2.twcny.rr.com>, news at dorb.com says...
>
>import time, whrandom
>
>list1 = [('one', 0.25), ('two', 0.25), ('three', 0.5)]
>def wc(list):
>   from whrandom import uniform
>   n = uniform(0, 1)
>   for item, weight in list:
>     if n < weight:
>       break
>     n = n - weight
>   return item

By writing cumulative weights (standard method), the substraction is not 
needed.  Comparison with 1.0 can then be eliminated.  By putting largest 
weights first, the remaining number of comparisons is minimized.  With these 
optimizations, we get:

list1 = [('three', 0.5), ('two', 0.75)] # else 'one'
def wc(list):
   from whrandom import uniform
   n = uniform(0, 1)
   for item, weight in list:
     if n < weight:
       return item
   return 'one'

If there were a enough categories, then binary search of the list (with the 
last item with cumulative weight 1.0 added back) would be faster.

Terry J. Reedy




From luisroco at hotmail.com  Tue Jun  8 16:55:35 1999
From: luisroco at hotmail.com (luisroco at hotmail.com)
Date: Tue, 8 Jun 1999 20:55:35 GMT
Subject: PIL Images in a web site
Message-ID: <7jk007$m8p9@eGroups.com>

Hello all:

How can a CGI script output graphics, such a JPEG image, through Python code using the package Python Imaging Library?

Thanks in advance




From MHammond at skippinet.com.au  Mon Jun 28 04:24:26 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Mon, 28 Jun 1999 18:24:26 +1000
Subject: Pythonwin and Office 2000
References: <377648A7.9B9BE59F@prescod.net> <7l672r$5k0$1@m2.c2.telstra-mm.net.au> <3776D1E6.45E1AAF7@prescod.net>
Message-ID: <7l7bbt$42u$1@m2.c2.telstra-mm.net.au>

[CC'd from the newsgroup to the python-com developers list]

Paul Prescod wrote in message <3776D1E6.45E1AAF7 at prescod.net>...

>Is there an equivalent to EnsureModule that works from the textual library
>name? I know that the textual library name still changes from version to
>version but I don't think it is Pythonish (easy, clean and readable) to
>put UUIDs in code at all.

I agree.  Worse, as you found out, it breaks when upgrading the COM object.
However, the only way to do it textually would be to search the registry for
the name.  Depending on the system, this can be significant.  The IID and
version info give us a direct key into the registry.  On my development
system, searching for an IID name could take a few seconds, and (IMO) we
can't afford this hit just to check.

I can't think of a better solution.  The Python equivilent to the VB system
would be for the developer of a script to nominate the type libraries they
reference, and to generate the necessary makepy support into a single file.
This single file would then be shipped with the script.  (VB puts the
generated code in the same .exe, hiding this extra distribution - in Python
terms, this would be integrating with freeze etc).  To go this route would
really require that makepy no longer generate source code, but instead
generate a tight binary format (which may be desirable anyway).

VBScript, for example, does not have the concept - it only understands (in
Python parlance) "dynamic dispatch".  Delphi, AFAIK, also makes it a
build-time step, rather than a run-time step.

Anyway, I have thought about this, and could not come up with a decent
solution.  The "gencache" implementation was the result - I knew it to be
half-baked, but really couldnt come up with a reasonable solution.

>In fact, the most Pythonish way would be to do what Python does with
>Pyc's. When Pythonwin does a win32com.client.Dispatch, it could generate
>the gencache module if possible. If not possible then no problem, it just
>works in the dynamic mode.

Yes - this could also work.  Indeed, we could change the generation so that
instead of doing the whole type library, it only does the classes of
interest.  In many respects, you could consider makepy itself a hack that
should be removed once Python can do a better job dynamically.

However, it is worth noting that speed is no longer a good justification for
using makepy.  Recent changes to dynamic dispatch mean that we are now
comparable speed wise to VB without makepy support.  The reason you often
now need makepy support is to get decent parameter support, but the problem
is worse than that.  In many non-trivial cases, the code you write is simply
not compatible between makepy supported and dynamic dispatch.  So, if you
want to ensure your code will run when installed, you must _insist_ on
makepy support.  (ie, silent failure of makepy is usually not a good option,
as you are simply delaying the inevitable error, but in a less obvious way).

There is plenty of scope to rationalize some of this.  As Python COM support
has grown from its humble beginnings, some of the existing code is no longer
appropriate.  It is also big and scary enough that fiddling will not have
the desired effect - it needs a good, hard think and a revamp.
Compatibility shouldnt be an issue - there are only about 4 truly public
functions in the whole package - 99% of users will only care that
"win32com.client.Dispatch" does what they expect!

>Also, is there any chance of having win32com.client.Dispatch renamed
>CreateObject() for JavaScript and VBScript compatibility?

This sounds reasonable (well, not rename, but alias).  OTOH, maybe we should
reserve this name for the "new world order" - eg, just incase someone _does_
step forward and some of this does get rationalised, there may be
better-but-slightly-different semantics better exposed as CreateObject()
(with Dispatch()" becoming a backwards compatible interface)

Hope this all makes sense...  Just incase people hadnt noticed, COM
development is very quite these days.  Almost no new activity and only
maintenance type work is being done.  This sort of work is unlikely to get
done without a volunteer.

Mark.





From arnold at dstc.edu.au  Tue Jun 29 00:44:59 1999
From: arnold at dstc.edu.au (David Arnold)
Date: Tue, 29 Jun 1999 04:44:59 GMT
Subject: Python GUIs: Summary and Conclusion 
In-Reply-To: Your message of "Tue, 29 Jun 1999 16:04:45 +1200."
             <377845DD.23AD193A@compaq.com> 
References: <377845DD.23AD193A@compaq.com>
Message-ID: <199906290445.OAA02111@piglet.dstc.edu.au>

-->"Greg" == Greg Ewing <greg.ewing at compaq.com> writes:

  Greg> Wasn't there a Tcl-less Tk interface once before, that wasn't
  Greg> kept up to date? Called something beginning with R that I
  Greg> can't remember.

it was called Rivet.

  Greg> Maybe you could use that as a starting point if the Tk C
  Greg> interface hasn't changed too radically.

it used Tk 7.4, and had bits of tcl in it, although it was much more
direct than tkinter.  we still use it (with various hacks that have
accumulated over time) ...

it chief problem (for us) is that it isn't properly thread-safe.



d




From morse at harborcom.net  Fri Jun  4 07:55:46 1999
From: morse at harborcom.net (Kevin Dahlhausen)
Date: Fri, 04 Jun 1999 11:55:46 GMT
Subject: All this self dot is making me...
References: <Vn653.468$HD.99317@typhoon-sf.snfc21.pbi.net> <m33e0a52x2.fsf@chinon.cnrs-orleans.fr> <37551fa6.90445994@news.oh.verio.com> <8Rb53.1622$nn.483094@news.shore.net>
Message-ID: <3757bdfb.262052080@news.oh.verio.com>

"Michael P. Reilly" <arcege at shore.net> wrote:

>I think abbreviations would be better:
>  :abbrev s. self.

Give it a shot.   The abbreviation has to been seen as a word before it is
replaced - that is you must have white space after hitting 's.'  So then you
you have to back up to just after the .

The mapping runs as soon as you hit the '.' and you can continue typing on the
fly.



From fredrik at pythonware.com  Sun Jun 20 12:12:37 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 20 Jun 1999 16:12:37 GMT
Subject: Questions?  Time Module?
References: <rNHvjEdhm5Pp-pn2-5b4dbg1xd7dt@d185d18f8.rochester.rr.com>
Message-ID: <001501bebb37$b78c4050$f29b12c2@pythonware.com>

Benjamin Schollnick wrote:
> I'm trying to figure out where the "time" functions are
> stored, but GREP isn't finding anything in the \lib directory??!?!?
> 
> I'm have some questions on the STRPTIME function, and
> thought I'd examine the code, but can't find it in the 1.51 
> installation?
> Sure, I can use the "IMPORT TIME" and access everything from there
> but heaven help me, if I can't find a TIME.PY or TIME.PYC file 
> anywhere...

http://www.python.org/doc/FAQ.html#4.70

</F>




From MHammond at skippinet.com.au  Mon Jun 21 18:34:10 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Tue, 22 Jun 1999 08:34:10 +1000
Subject: wierdness with pythonwin dialogs, threads & refcounts
References: <1282457297-115891134@hypernet.com> <376e53a0.24292290@nnrp.uk.insnet.net>
Message-ID: <7kmehp$9el$1@m2.c2.telstra-mm.net.au>

Dave Kirby wrote in message <376e53a0.24292290 at nnrp.uk.insnet.net>...

>I am still climbing the learning cliff of Win32 threads so I dont know
>if it is the approved way of doing it, but it seems to work well
>enough for my purposes.

Well, seeing as that is the only way that works, I would say it is
"approved" :-)

The MFC threading stuff is almost identical to a Python thread created with
the thread or threading modules.  Indeed, to Python, they are identical.
The only difference I can see if that we have a real MFC thread backing this
thread.

At some stage I will try and sort this out so standard Python threads also
work correctly...

Mark.





From mlh at idt.ntnu.no  Tue Jun 22 08:40:14 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 22 Jun 1999 14:40:14 +0200
Subject: while(s=infile.readline()):     # reading input
References: <376EA6E2.4E5EE527@starvision.com>
Message-ID: <y0jvhcgh0i9.fsf@vier.idi.ntnu.no>

Angus MacKay <amackay at starvision.com> writes:

> why can't I have syntax like this?

ARG!! Not agaaain! <sob>

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From gvwilson at nevex.com  Fri Jun 25 09:41:53 1999
From: gvwilson at nevex.com (Greg Wilson)
Date: Fri, 25 Jun 1999 09:41:53 -0400
Subject: Python / VTK problem
Message-ID: <37738721.E2B17705@nevex.com>

Hi, folks.  I've just downloaded, compiled, and installed VTK 2.3, and
for the first time, I've
built with Python extensions enabled.  Unfortunately, while I can run
the Tcl examples, I
can't get the Python examples to run.  The command 'python mace.py' (or
similar) produces:

  Traceback (innermost last):
    File "./mace.py", line 2, in ?
      from libVTKGraphicsPython import *
ImportError: /home/gvwilson/lib/libVTKGraphicsPython.so: undefined
symbol: XtStrings

The symbols XtStrings is defined in /usr/X11R6/lib/libXt.so (I'm on Red
Hat Linux).  The directory
/usr/X11R6/lib is in my LD_LIBRARY_PATH and in my PYTHONPATH, as is
/home/gvwilson/lib
(which contains symbolic links to the VTK .so's).  I'd be grateful to
anyone who can tell me what I'm
doing wrong.
Thanks,
Greg
(p.s. I'd also be grateful if you'd reply by direct email to g v w i l s
o n @ n e v e x . c o m)




From fdrake at cnri.reston.va.us  Fri Jun 18 11:01:07 1999
From: fdrake at cnri.reston.va.us (Fred L. Drake)
Date: Fri, 18 Jun 1999 15:01:07 GMT
Subject: HTML bettering?
In-Reply-To: <1318D78C9072D11195C9006094EA98A72C2164@ocesrv>
References: <1318D78C9072D11195C9006094EA98A72C2164@ocesrv>
Message-ID: <14186.24371.249981.437742@weyr.cnri.reston.va.us>

Matej Cepl writes:
 > goodwill, is it? But, to make Pythonians more happy, the story is as

  A worthy goal!  ;-)

 > html2ps which has much better output, but it numbers headings in
 > different way (still unresolved issue, what it the top level -- H2 or

  There are a couple of these out there, including one in Python
included with the Grail sources.  Someone posted a patch to it on the
Grail list, so I'm not sure of the current state of it, but I could
probably answer a few quick questions about it.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From Ovidiu.Predescu at p98.f112.n480.z2.fidonet.org  Wed Jun 30 20:57:23 1999
From: Ovidiu.Predescu at p98.f112.n480.z2.fidonet.org (Ovidiu Predescu)
Date: Thu, 01 Jul 1999 01:57:23 +0100
Subject: strange behaviour at termination time
Message-ID: <000015f0@bossar.com.pl>

From: Ovidiu Predescu <ovidiu at cup.hp.com>

Greg Ewing wrote:

> To avoid problems like this, you have to design
> your __del__ methods so that they can do their
> work without having to refer to any module-level
> names at the time they are called. One way is
> to use the default-argument trick:
> 
> class B (A) :
>     def __del__ (self, A = A) :
>         A.__del__ (self)

Can you explain why this trick works? When are the default arguments
stored in the method instance? I guess this happens when the method
instance for __del__ is created by the interpreter, right?

Thanks,
-- 
Ovidiu Predescu <ovidiu at cup.hp.com>
http://www.geocities.com/SiliconValley/Monitor/7464/



From jo at delorges.in-berlin.de  Wed Jun  2 20:19:27 1999
From: jo at delorges.in-berlin.de (Jo Meder)
Date: 03 Jun 1999 02:19:27 +0200
Subject: Bizarre error while ./configure-ing PIL
References: <m3g14dxduq.fsf@ime.net> <m3n1yii8w9.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <m3k8tmgmpc.fsf@delorges.in-berlin.de>

Michael Hudson <mwh21 at cam.ac.uk> writes:

> Matt Gushee <mgushee at ime.net> writes:

[reasonable permissions for ./configure]
> > bash: ./configure: Permission denied

> Random guess, but check the first line of ./configure. It should
> probably read
> 
> #!/bin/sh
> 
> My guess is either 1) /bin/sh is not executable (very unlikely, all
> sorts of things would barf at that) 2) the line's gotten mangled
> somehows - maybe a misplaced space?

Actually the line reads

#! /bin/sh
  ^note the space

but why this should lead to "Permission denied", I don't know. 

I tried it with tcsh, zsh, bash, ksh, ash and not one of them barfed
at the line.


	Jo.






From xhyuan at oe.fau.edu  Wed Jun 23 09:56:10 1999
From: xhyuan at oe.fau.edu (Xiaohong Yuan)
Date: Wed, 23 Jun 1999 06:56:10 -0700 (PDT)
Subject: How to parse or build unsigned short value?
Message-ID: <Pine.SGI.3.92.990623065051.18876A-100000@sea30>

I am using Py_BuildValue(..), and PyArg_Parse(..) functions. The format
characters are provided for int, short, long, double, float, char type.
However there is no format character for  unsigned int, unsigned short,
unsigned long type. Right now I use pack, or unpack of struct module to
deal with unsigned types. Is there any other convenient way to do it?

Thank you very much.

Xiaohong Yuan




From fredrik at pythonware.com  Wed Jun 16 06:32:45 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 16 Jun 1999 10:32:45 GMT
Subject: <insert favourite flame-bait here> broken
References: <1282792849-95706075@hypernet.com> <3764D7C3.718E125E@iname.com> <002f01beb64d$58e7b5f0$f29b12c2@pythonware.com> <37651951.ED54894E@acm.org> <005501beb674$c5575890$f29b12c2@pythonware.com> <376621CF.88C95B5E@iname.com> <150619990851173299%pereira@research.att.com> <3766653A.D9F6E99B@iname.com>
Message-ID: <014301beb7e3$94707610$f29b12c2@pythonware.com>

> "cult" "kiddie" "flamer" "dogmatic"

interesting.  a lot of new cool things to add
to my business card.  given that fernando has
never met me in person, it's amazing what he
could figure out just by reading a usenet post
or two.  he's a sure genius, that guy ;-)

</F>




From frankn=news at cs.vu.nl  Thu Jun  3 09:34:53 1999
From: frankn=news at cs.vu.nl (Frank Niessink)
Date: Thu, 3 Jun 99 9:34:53
Subject: Help, how to override <= operator
References: <7htsdd$6kd@wapping.ecs.soton.ac.uk> <FBzoJD.2JI@gmx.net> <7i0eao$b4l@cs.vu.nl> <37535A17.C4EE5B44@compaq.com>
Message-ID: <7j5b6t$5qb@cs.vu.nl>

Greg Ewing <greg.ewing at compaq.com> wrote:
> Frank Niessink wrote:
>> 
>> Clemens Hintze <cle at gmx.net> wrote:
>> 
>> > class myint:
>> >     i = 0
>> 
>> This is a bit confusing because you don't need (and don't even use,
>> and probably don't _want_ to use) the class variable i.

> I beg to differ - I think that "declaring" instance variables
> in this way serves a useful documentation purpose, and I do
> so frequently.

But what is `declared' in the snippet above is a _class variable_!
If you are developing an integer class you probably don't want a 
class variable...

This is the original posters' code:
> class myint:
>     i = 0
>     def __init__(self, n):
>         self.i = n
>     def __cmp__(self, b):
>         if self.i < b.i: return -1
>         if self.i > b.i: return 1
>         return 0

The class variable i and the instance variable i are different. Example:
>>> a = myint(1)
>>> a.i
1

Because the instance variable i hides the class variable i we to access 
the class variable indirectly:
>>> a.__class__.i    
0

So my comment still stands: you don't need, don't use, and probably don't
want to use the class variable i (in this particular example of course). 

Ciao, Frank

-- 
An RW6 was the sort of thing you only saw in the sort of magazines that were 
designed to provoke civil unrest.
	-- Douglas Adams, 'Mostly Harmless'



From mgushee at havenrock.com  Sun Jun 20 11:56:07 1999
From: mgushee at havenrock.com (Matt Gushee)
Date: 21 Jun 1999 00:56:07 +0900
Subject: simple text file 'parsing' question
References: <376B1AAC.19FE8BCE@mysolution.com>
Message-ID: <m3ogiaq31k.fsf@havenrock.com>

KP <terocr at mysolution.com> writes:

> Here's my dilema: a directory filled (200+) with small emails. My goal
> is to strip all the headers and combine them into one file. I can read
> all the files just fine and write them all to one file, but I cannot
> discern how to strip the headers.

I have no expertise in this area, but I've been reading the "Internet
Data Handling" section of the Library Reference (Ch. 12 of the 1.5.2
edition), and it seems like there are several modules that might help
you. In particular, check out 'rfc822.'

Hope this helps.

Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com



From jand at easics.be  Wed Jun  2 08:53:34 1999
From: jand at easics.be (Jan Decaluwe)
Date: Wed, 02 Jun 1999 14:53:34 +0200
Subject: Is pcgi usable without Zope?
References: <ltd7zf6wtv.fsf@asfast.com> <3755CA26.BCB7828A@digicool.com>
Message-ID: <3755294E.BBF39060@easics.be>

Paul Everitt wrote:
> 
> Lloyd Zusman wrote:
> >
> > I have to admit that I'm a bit confused about pcgi these days.  The
> > most recent documentation for it was last updated in August, 1998, and
> > it still refers to Bobo and Principia.
> 
> We're certainly not helping -- I agree that stuff is strewn around all
> over the place.
> 
> > Now that we have Zope, is pcgi now intimately bundled with it, never
> > to emerge on its own again?  Or is there a way to run pcgi without
> > Zope?
> 
> I don't think PCGI is intimately bundled with Zope.  Neither is the
> object publisher, object database, the template system, etc.  All can be
> used independently.
> 

We are using pcgi with independent Zope components without problems -
the pcgi documentation is quite good and the new names for the Zope
components don't really matter.

The independent Zope components can also be used as before with Bobo,
at least for the python-only versions. To use the full versions -
including the C extensions and C modules - we do a full installation of
Zope and then point to Zope's lib/python/ directory in the PCGI_INSERT_PATH
variable of the pcgi info file. 

-- 
Jan Decaluwe	       Easics              
Design Manager         SoC design services & web tools 
+32-16-395 600	       Interleuvenlaan 86, B-3001 Leuven, Belgium
mailto:jand at easics.be  http://www.easics.com



From tim_one at email.msn.com  Fri Jun 18 20:01:32 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 18 Jun 1999 20:01:32 -0400
Subject: DATE ARITHMETIC
In-Reply-To: <7kejrv$e57$1@violet.singnet.com.sg>
Message-ID: <000201beb9e6$e3789620$a69e2299@tim>

[Ajith Prasad]
> I need to do some simple date arithmetic - finding the number of days
> between any 2 days, the age of a person on a day given his birthday. The
> mxDateTime module downloadable from the Python website is more
> complex than what I need. Any other suggestions?

Demo/classes/Dates.py, in the std source distribution (the binary Windows
distribution doesn't contain the Demo directory -- you need the source
distribution).  Supplies a very simple date class, that will work for at
least the next trillion years <wink>.

i-laugh-at-y2k-ly y'rs  - tim





From suzuki611 at okisoft.co.jp  Wed Jun  9 20:43:07 1999
From: suzuki611 at okisoft.co.jp (Hisao Suzuki)
Date: 10 Jun 1999 09:43:07 +0900
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu>
	<7j90va$6lk$1@cronkite.cc.uga.edu>
	<u3g1443iht.fsf@ares.sys1.ngo.okisoft.co.jp>
	<87pv384koy.fsf@ev.netlab.co.jp>
Message-ID: <u3909slwbo.fsf@ares.sys1.ngo.okisoft.co.jp>

In article <87pv384koy.fsf at ev.netlab.co.jp>,
Yukihiro Matsumoto <matz at netlab.co.jp> wrote:
| It's because conservative-ness of Ruby's GC, you know.  There should
| be no portability problem with Graham's proposal, mark and sweep GC
| with ref counting, unless implementor try to make it conservative.

and also wrote:

| I claimed in summary:
|   * I don't like mere ref counting; it cannot reclaim cyclic data, and
|     easy to forget INCREF/DECREF.
|   * Many among Python users misunderstood about real GC.
|     (slow, hangs, non portable, etc.)
|   * ref counting with GC may be good for Python 2.0.
| Did I do any advocacy on Ruby?  I was trying constructive discussion.

Well, you have experience of some non-portable conservative
garbage collector on your Ruby, while advocating potential,
portable one for Python, don't you?

Moreover, I am afraid that your experience and argument might
have nothing to do with general, complex finalization.  Once
thinking of better storage management for Python _seriously_,
this will be a real problem.  (I pointed out this matter for
several times recently, you know.)

A very common discussion on GC has been already given.  I hope
you read Doc/ext/ext.tex or `the Extending and Embedding the
Python Interpreter'.  (Judging from your referring to
INCREF/DECREF, you might have read it...)

I'm sorry to say, but your argument is too trivial (or rather
naive) to consider as a constructive discussion.  Having almost
nothing to do with real problems on Python, your referrings to
your Ruby (surely in triumphant tones) sound evangelic.  I'd be
glad if you would present a non-trivial viewpoint or propose a
practical solution _relevant_ to Python actually.

--===-----========------------- Sana esprimo naskas sanan ideon.
SUZUKI Hisao            suzuki611 at okisoft.co.jp, suzuki at acm.org.



From morse at harborcom.net  Tue Jun 15 07:45:34 1999
From: morse at harborcom.net (Kevin Dahlhausen)
Date: Tue, 15 Jun 1999 11:45:34 GMT
Subject: Newbie for Embeding
References: <3766267D.AF89C56B@hotmail.com>
Message-ID: <37663c6c.520743990@news.oh.verio.com>

Denis Lamarche <brainwav at hotmail.com> wrote:

>I was wondering something.  If you can embed python into c++, can you
>then call your c++ fuctions from python?  Can you also modify variable
>in your c++ program using python?
>
>Denis Lamarche

You have to create a module that exposes those elements of your C++ program to
Python.  SWIG (Simplified Wrapper Interface Generator) automates this for simple
cases.   Once you have the module, the answer to both questions is yes.   See
the docs under extending and embedding python.





From dshanbha at clarify.com  Mon Jun 28 21:01:48 1999
From: dshanbha at clarify.com (Dinesh Shanbhag (c))
Date: Mon, 28 Jun 1999 18:01:48 -0700
Subject: newbie question about tkFileDialog.py
Message-ID: <3BE7847CF88DD211893100A0C9D43E1A0259E1ED@hqmail2.clarify.com>

I am using the 'tkFileDialog.py' module to open a file.
 
To allow the user to select a file I can use tkFileDialog.askopenfilename or
tkFileDialog.asksaveasfilename.

I also want the user to be able to select a Directory on the similar
lines...

Did not find anything like tk_getOpenFile in the Tcl/Tk man pages for
selecting Directories...

Am i missing something obvious ?

thanks in advance for any help,
dinesh.



From fatjim at home.com  Tue Jun 22 16:41:26 1999
From: fatjim at home.com (Jim Meier)
Date: Tue, 22 Jun 1999 20:41:26 GMT
Subject: creating dictionnaries
References: <376F82A6.82497D2B@der.edf.fr>
Message-ID: <376FF465.21DA89C5@home.com>


Christian Caremoli wrote:

> Hi,
> Dictionnaries can be created like that :
> d={'a':1,'b':2}
>
> By calling a function you create a dictionnary like that :
> def f(**d):
>    return d
>
> d=f(a=1,b=2)
>
> I would like to be able to create dictionnaries with some similar syntax
> like keyed tuples :
> d=(a=1,b=2)
>
> Is there a way to do that ?

You betcha. Just build a function like this one:

def dict(**kwargs):
    return kwargs

and shazam! You've got your keyword args style dictionary creator.

ex:
>>> dict(a=5, b='jim')
{'b':'jim', 'a':5}


-Jim.




From schorsch at schorsch.com  Tue Jun 15 02:38:34 1999
From: schorsch at schorsch.com (Georg Mischler)
Date: Tue, 15 Jun 1999 06:38:34 GMT
Subject: do anonymous pipes normally work on NT?
References: <7juatc$fef$1@nnrp1.deja.com> <7k43sk$6vd$1@nnrp1.deja.com> <aahzFDCGt4.2Dn@netcom.com>
Message-ID: <7k4sda$e6p$1@nnrp1.deja.com>

Aahz Maruch wrote:
> Georg Mischler wrote:
> >
> >I simplified my class to the absolute minimum I could.  It now does
> >nothing but create three pipes, save the parent ends to self.xxx, and
> >create a process with the child ends of the pipes. After that it
tries
> >to write to the stdin pipe of the child, which goes without error,
and
> >tries to read from stdout, which blocks.
>
> How about sockets?

Because I don't feel like rewriting several dozen existing C programs
as provided by Radiance (radsite.lbl.gov/radiance/) just for this
purpose. It's not that I don't know about sockets. I use them in
situations when I want to run processes on remote hosts (distributed
rendering). But once I'm there, I need a simple way to just run
the existing stuff and get the results back by the simplest
means possible. That's exactly what anonymous pipes are there
for, and I'd *love* them to work as documented (or rather as I
understand the documentation... ;).

But thanks for the suggestion anyway :)


Have fun!

-schorsch

--
Georg Mischler  --  simulation developper  --  schorsch at schorsch.com
+schorsch.com+  --  lighting design tools  --  http://www.schorsch.com/


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From graham at sloth.math.uga.edu  Wed Jun  2 16:42:59 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 2 Jun 1999 20:42:59 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7imr6b$1s0$1@nnrp1.deja.com>
Message-ID: <7j450j$908$2@cronkite.cc.uga.edu>

Martijn Faassen (faassen at pop.vet.uu.nl) wrote:
: This sounds like an interesting idea; I believe I saw references to
: garbage collecting schemes that help clean up circular references caused
: by referencing counting. I forget where, though. :)

Here. I posted comments in this thread and in a similar thread about
such systems. They are generally all variations on mark and sweep
collection. The more efficient ones use colouring.

Martijn Faassen (faassen at pop.vet.uu.nl) wrote:
: * In the complex case, garbage collecting is nice as it cleans up
: circular references. There are unwanted side-effects of its
: unpredictability, though; a possible leaking of resources (files which
: aren't closed), and some difficulties interfacing with C.

I don't buy either of these difficulties since I have worked with and
implemented systems which have full garbage collection and have no
problems interfacing to C, or dealing with resources like files. These
are all arguments against *some* garbage collection schemes, but not
all.

graham

-- 
             Like a bird on a wire
        Like a drunk in a midnight choir
            I have tried in my way
                  To be free



From gmcm at hypernet.com  Tue Jun 15 13:49:36 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 15 Jun 1999 12:49:36 -0500
Subject: easy filecopy? - JPython?
In-Reply-To: <14182.28241.949065.590839@weyr.cnri.reston.va.us>
References: <PBt93.843$B04.2678@newsfeed.slurp.net>
Message-ID: <1282653264-104102639@hypernet.com>

Fred Drake wrote:

>   shutil.copyfile() works just fine (JPython 1.0.3 on java1.2), but
> shutil.copymode() relies on os.chmod(), which isn't there.  I'm sure
> Barry Warsaw (bwarsaw at python.org) would love to receive a patch for
> JPython to support that!  ;-)

Last I recall (a while ago), os.chmod() doesn't work on Win9* and 
acts as a noop on WinNT under CPython. This might explain why the 
functionality isn't there in Java?

- Gordon



From greg.ewing at compaq.com  Tue Jun  1 00:44:13 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Tue, 01 Jun 1999 16:44:13 +1200
Subject: Concrete Proposal: while ... and while ...
References: <7hsv80$g9j$1@news.tamu.edu> <y0j675pmg6c.fsf@vier.idi.ntnu.no> <7hufdb$rrh$1@news.tamu.edu> <_mY03.617$RO1.191275@news.shore.net> <7i1l20$l8p$2@cronkite.cc.uga.edu> <slrn7k8ptl.eqc.quinn@krone.ugcs.caltech.edu> <jliu9nrqq4.fsf@arathorn.bbb.caltech.edu> <7i5bje$vr$1@news.udel.edu>
Message-ID: <3753651D.4ED49CF4@compaq.com>

Terry Reedy wrote:
> 
> Guido recently explained that the interpreter cannot assume that the inner
> call to a recursive function actually calls the function within which it
> sits.

But that's not required in order to do tail recursion.
All you need to do is recognise when you're about to call
a function and then immediately return, and either re-use
the current stack frame or pop the current one before
pushing a new one.

Python could probably do tbis without much difficulty,
as long as the called function is a Python function.
If it's a C function, though, things would get rather
tricky!

I'm not very worried about this personally, since I'm
quite happy to use iteration instead of recursion when
the language encourages it, as Python does with its
lovely for: loop.

Greg



From digitome at iol.ie  Thu Jun 17 13:43:22 1999
From: digitome at iol.ie (Sean Mc Grath)
Date: Thu, 17 Jun 1999 17:43:22 GMT
Subject: XML-toolkit for Windows - binaries available?
References: <376701a3.72288277@newssrv> <wkzp208udt.fsf@ifi.uio.no> <3767fc51.1547455@news.iol.ie> <wkd7yv7t5p.fsf@ifi.uio.no>
Message-ID: <37693395.731772@news.iol.ie>

On 16 Jun 1999 23:05:38 +0200, Lars Marius Garshol <larsga at ifi.uio.no>
wrote:

>
>* Sean Mc Grath
>|
>| The binaries are there but a bit of fiddling is required to get
>| things working. (This is from memory but here goes)
>| 
>| The pyexpat and sgmlop dlls have to be copied into the xml/parsers
>| subdirectory. You also need to edit the __init__.py file to include
>| 'pyexpat' and 'sgmlop' in the __all__ string.
>
>Heh. I didn't know you could do this! :)
>
>I just put them in the python/dlls directory, and that worked for me.
>
Lars, is quite right. python/dlls is a better place to put them than
xml/parsers.

Sean






From bruce.adams at rmc-ltd.com  Thu Jun 10 10:46:17 1999
From: bruce.adams at rmc-ltd.com (Bruce S. O. Adams)
Date: Thu, 10 Jun 1999 15:46:17 +0100
Subject: Sharing extensions between scripting languages (was Re: File Lock in 
 TCL)
References: <37555889.CEE8941C@bri.hp.com> <37569190.D3B3D075@equi4.com> <3756AF87.7B7295DA@messagemedia.com> <7j8uht$9c$1@srv38s4u.cas.org> <375CD3ED.66328791@equi4.com> <375FBDA8.8ED5BA14@senteinc.com> <375FCA43.F36A2FD5@equi4.com>
Message-ID: <375FCFB9.C37E46B7@rmc-ltd.com>


Jean-Claude Wippler wrote:

> If you want to lock with a separate lock file ("dotlocking", see below),
> then there is no problem, although there can be a race condition (which,
> on a network, slow link, etc, can be a serious issue).  In this case,
> using the Tcl-only locking approach is perhaps most suitable.
>
> If you want to use fcntl/lockf/flock system calls, then locking may or
> may not work - AFAIK, some platform combinations simply don't lock,
> while some others can cause the underlying faulty lock daemon to hang.
> Here's an excerpt on a note I found through Alta Vista:
>
> ========================================================================
> A brief overview regarding locking methods would be:
> dotlocking (*.lock)         If you want to lock "testfile", try and
>                             create a *new* "testfile.lock" in the same
>                             directory.  Repeat (with ample sleep) until
>                             this succeeds.  To unlock, simply remove the
>                             file.  This method is supported by almost
>                             all configurations.
> fcntl()         (POSIX)     Usually implemented over the lockd program
>                             (thus NFS aware).
> lockf()         (SysV)      Usually implemented over fcntl().
> flock()         (BSD)       On SunOS this only locks locally, on other
>                             systems it might use the lockd.
> ========================================================================
>
> It's by Stephen van den Berg, Procmail's author:
>         http://members.xoom.com/procmail/best-of-procmail-list/3
>
> Another place to look for tips is Mailman (Python, http://www.list.org),
> see the file mailman-1.0rc1/Mailman/flock.py - which says:
>
>     """Portable (?) file locking with timeouts.
>     This code should work with all versions of NFS.
>     The algorithm was suggested by the GNU/Linux open() man page.  Make
>     sure no malicious people have access to link() to the lock file.
>     """
>
> It sure would be nice if there were a package which jumps through all
> known hoops to implement locking, on most Unix'es, Windows, Mac, etc.
> IMHO, this is an example of where the split between Tcl, Python, Perl
> and others is hampering the generation of a general package.
>
> -- Jean-Claude

Tk is already accessible to both Python and Perl, though I don't know how
easy it is to use it?  Perhaps there should be a general discussion on how
to
share extensions between the three platforms.  I only know the TCL mechanism

well at the momment, but I got the general impression that it was the best.
Perhaps, that's just because Perl & Python are less advocated on
comp.lang.tcl
than tcl is :-).  Is anyone going to raise this topic at the forthcoming
scripting conference (which I believe originally centred around perl)?  What
would
we need for a generalised extension architecture?  I'm sure it would involve
more
than just swapping  Tcl_CreateCommand for Perl_CreateCommand or whatever
(especially as we are now encouraged to use Tcl_Obj's in our interfaces.
    A general solution is to have in each language a routine allowing cross
language
calls to be made e.g.  in perl have an InvokeTCLCommand function and in
TCL an InvokePerlCommand function.  Of course some extensions would
require access to the languages interpreter and others wouldn't.  This
solution
would also come at the cost of some efficiency.  Would people be prepared to

pay that cost?  Any thoughts?
                                            Regards,
                                                        Bruce A.











From fredrik at pythonware.com  Tue Jun  1 04:01:57 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 1 Jun 1999 08:01:57 GMT
Subject: Python 2.0
References: <7ik6mi$lbk$1@cronkite.cc.uga.edu> <7iqhkb$d7s@wiscnews.wiscnet.net> <7iu7ma$d33$1@cronkite.cc.uga.edu>
Message-ID: <007901beac05$5c964b00$f29b12c2@pythonware.com>

Graham Matthews <graham at sloth.math.uga.edu> wrote:
> Gang Li (gangli at msn.com) wrote:
> : Python is a real programming language.  We should not turn it in to a Java
> : extension. I hate someone just want to take Python to Java mass and say 'me 
> : too'.  Perl, Tcl/Tk, and Smalltalk have no plan to turn themselves in to a 
> : Java extension.  Why Python?  If you love Jave to death, spent all your 
> : time to help sun conquer the world. Don't mess with Python.
> 
> You are envangelising (a not uncommon response when someone remotely
> criticises Python).

read his message again.  he's saying that Python
shouldn't be tied to Java.  sure looks like a tech-
nical argument to me.

if you want some numbers to support his argument,
take a look at GvR's Python poll from the latest
python conference:
http://www.foretec.com/python/workshops/1998-11/polls.html

10% of the audience used JPython today, for non-web
applications.  about the same amount used COM.

in other words, proposals to make Python dependent on
Java should be treated about as seriously as proposals to
make Python dependent on COM.  they're both available
on all major platforms (at least if you ignore those you
don't happen to use).  and they're both supported by
many other tools (at least if you ignore those I don't
happen to use). etc.

> Stop evangelising and start considering the technical
> issues involve (read (!) other posts for what those issues
> are).

exactly.  and since you seem to imply that you have con-
sidered those issues, maybe you should follow up on some
of the "non-envangelising" posts instead of miscrediting
people who don't happen to agree with you?

</F>




From hniksic at srce.hr  Fri Jun  4 17:45:38 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 04 Jun 1999 23:45:38 +0200
Subject: threads
In-Reply-To: Jeremy Hylton's message of "Fri,  4 Jun 1999 17:41:46 -0400 (EDT)"
References: <1283593620-47537497@hypernet.com> <87emjr67ff.fsf@pc-hrvoje.srce.hr> <14168.18308.709890.544779@bitdiddle.cnri.reston.va.us>
Message-ID: <87u2sn4p31.fsf@pc-hrvoje.srce.hr>

Jeremy Hylton <jeremy at cnri.reston.va.us> writes:

> If I recall correctly, the problem is that taking away one big lock
> usually requires introducing lots of little locks.  (There's some
> kind of conservation of concurrency going on.)  Right now, there is
> no need to have locks on access to builtin types like dictionaries,
> because only a single thread can hold the interpreter lock anyway.
> When that goes away, you need to add lots of fine-grained locking on
> the builtin types.

You are right.

Ironically, when I first saw "tuples" in Python, I assumed they were
invented to speed up threading because you don't need to lock access
to immutable structures.  (The same applies to immutable strings.)
Boy, was I wrong.  :-(



From jason at harlequin.com  Wed Jun  9 07:43:00 1999
From: jason at harlequin.com (Jason Trenouth)
Date: Wed, 09 Jun 1999 11:43:00 GMT
Subject: Humbug (was Re: Indentation and grouping proposal)
References: <000b01beb14d$b502a480$aaa02299@tim> <375D1E23.9367A42C@spirit.gcrc.upenn.edu>
Message-ID: <378852e8.2242010093@newshost>

On Tue, 08 Jun 1999 09:44:03 -0400, Tom Fenn <tom at spirit.gcrc.upenn.edu>
wrote:

> The day Python has {'s, is the day I learn Dylan. :)

format-out("shame...");

__Jason



From gvinther at atex.com  Mon Jun  7 17:56:03 1999
From: gvinther at atex.com (Gordon Vinther)
Date: Mon, 7 Jun 1999 17:56:03 -0400
Subject: Tkinter problem in Python 1.5.2 on Solaris 2.5.1
References: <1283620640-45912067@hypernet.com> <3757CE4B.B8B70983@easics.be>
Message-ID: <928792539.297.78@news.remarQ.com>

I seem to be having the same problem.  Where and how do I find these
patches??

Thanks Thanks



Jan Zegers <janz at easics.be> wrote in message
news:3757CE4B.B8B70983 at easics.be...
> Gordon McMillan wrote:
> >
> > Jan Zegers wrote:
> > >
> > > I have the following problem with Tkinter in Python 1.5.2.
> > /snip/
> > >   This probably means that Tcl wasn't installed properly.
> > >
> > > Note the whitespace between the two last test lines. The
> > > installation is as follows:
> > >   Tcl/Tk 8.1 with --enable-shared so we have libraries
> > >     /usr/local/lib/libtcl8.1.so and /usr/local/lib/libtk8.1.so
> >
> > Tcl / Tk 8.0 is still the official version for Python.
> >
> > On Jun 1, Dieter Maurer posted patches to use 8.1 ("Tkinter-patch for
> > use with Tk8.1...").
> >
> > - Gordon
> Thank you. I applied the patches and it worked!
> Kind regards,
> Jan
> --
> ===================================================================
> Jan Zegers                ===              Easics               ===
> General Manager           ===  VHDL-based ASIC design services  ===
> Tel: +32-16-395 601          ===================================
> Fax: +32-16-395 619      Interleuvenlaan 86, B-3001 Leuven, BELGIUM
> mailto:janz at easics.be              http://www.easics.com





From news at dorb.com  Tue Jun 29 14:43:46 1999
From: news at dorb.com (Darrell)
Date: Tue, 29 Jun 1999 14:43:46 -0400
Subject: run .py not in PATH
Message-ID: <Vv8e3.3676$U6.15962@newsr2.twcny.rr.com>

#!/usr/bin/env python
"""
The .py you want to run is in the python path but not
in your shell path.

I know everyone but me already has this problem solved
in a better way than this. If so please show me the way.

Assuming this is runIt.py then
runIt.py someScript.py arg1 arg2
"""
import sys
import imp

fp, pathname, description = imp.find_module(sys.argv[1])
sys.argv=sys.argv[2:]
md=imp.load_module('__main__', fp, pathname, description)







From sai at dft.chem.cuhk.edu.hk  Mon Jun 28 07:59:40 1999
From: sai at dft.chem.cuhk.edu.hk (SC Zhong)
Date: Mon, 28 Jun 1999 19:59:40 +0800
Subject: math domain error
Message-ID: <Pine.SGI.3.91.990628195446.27381A-100000@dft>

I used the Numerical modules to find the square root of a matrix and
get the "math domain error", what is it?  The matrix I have is 
1096*1096 dimension.  When I tested the program with smaller problem,
it seems to work.  Please advice.

/var/tmp/chk_overlap-> Traceback (innermost last):
  File ".//pop", line 412, in ?
    popMatrix = LdinAnalysis(alphacoefficientsMatrix)
  File ".//pop", line 345, in LdinAnalysis
    M(overlapMatrix, eigvecsT))),eigvecsTInverse)))
ValueError: math domain error


sai





From faassen at pop.vet.uu.nl  Wed Jun  2 06:00:42 1999
From: faassen at pop.vet.uu.nl (Martijn Faassen)
Date: Wed, 02 Jun 1999 12:00:42 +0200
Subject: Alien whitespace eating nanovirus strikes again!
Message-ID: <375500C9.742DC539@pop.vet.uu.nl>

Hi there,

I don't intend to start a flamewar or anything, and I'm sure Larry Wall
is a nice guy, and Perl is nice and all, and the postmodern
rationalisation of the confusion that Perl is to me is *fun* (but does
not work for me), but Larry Wall seems to be infected by the whitespace
eating nanovirus meme. From an interview at:

http://www.linuxjournal.com/issue61/3394.html

Interviewer:  In what way is Perl better than other scripting languages
such as Python and Eiffel? 

[Eiffel a 'scripting language'? hm]

[Larry Wall:]
[Perl is postmodern]
> That is, if something's worth doing, it's worth driving into the ground to the
> exclusion of all other approaches. Look at the use of parentheses in Lisp or 
> the use of white space as syntax in Python. Or the mandatory use of objects in
> many languages, including Java. All of these are ways of taking freedom away
> from the end user ``for their own good''. They're just versions of Orwell's
> Newspeak, in which it's impossible to think bad thoughts. We escaped from the
> fashion police in the 1970s, but many programmers are still slaves of the cyber
> police.
[Perl is postmodern and therefore allows freedom]

One could say the same about Perl and it requiring it curly braces
around blocks? In fact, I heard the claim that its requirement to use
curly braces around single line blocks (unlike C which allows you to
remove them) makes Perl less prone to errors. And Perl excludes the use
of meaningful whitespace indentation! (python does allow #{ and #}
blocks ;)

Okay, I *will* say the same about Perl and its curly braces requirement:

} That is, if something's worth doing, it's worth driving into the
ground to the
} exclusion of all other approaches. Look at the use of curly braces in
} Perl.

*any* language makes it impossible to code in some ways, and encourages
other coding styles, Perl included. Python, for instance, allows
procedural and object oriented coding styles; and people keep trying for
functional styles as well. We've even started using 'assembler style' by
manipulating Python bytecodes! Silly, silly, whitespace meme.

People think bad thoughts in Python all the time (see the 'evil hacks'
threads showing up regularly). It's just that we have the sense to
recognize the bad thoughts for what they are. I suppose that's why
'advisory locking' of member data is good enough for Python. :)

Okay, sorry for all this, it's out of line, I just needed to blow off
steam. :)

Regards,

Martijn



From hniksic at srce.hr  Fri Jun  4 11:54:30 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 04 Jun 1999 17:54:30 +0200
Subject: threads
References: <001001beae49$389b4ca0$949e2299@tim> <3757DDE9.F10CC150@icrf.icnet.uk>
Message-ID: <877lpk7yh5.fsf@pc-hrvoje.srce.hr>

Arne Mueller <a.mueller at icrf.icnet.uk> writes:

> I think I'll try to solve my problem by switching from multiple
> threads to different processes using 'fork'. The only problem I have
> is that all these processes must have one object in common that
> collects all the result data. Maybe a pipe or shared memory can
> help.

A pipe would be useful, yes.  You can use select to wait for several
processes simultaneously.



From stephan at pcrm.win.tue.nl  Tue Jun  8 09:27:08 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 08 Jun 1999 15:27:08 +0200
Subject: Python CPU
References: <14171.17113.39933.904980@lrz.de>
Message-ID: <m3909ubz6b.fsf@pcrm.win.tue.nl>

Eugene Leitl <eugene.leitl at lrz.uni-muenchen.de> writes:

> Have there been any attempts in creating an ASIC capable of
> interpreting Python opcodes directly? 

I don't think so.

A similar attempt has been made Very Long Ago with the Lisp Machine,
the idea being that conventional architectures are optimized for
C-like languages and that Lisp would benefit from a specialised 
architecture. 

However, conventional architectures improved so fast that soon
Lisp interpreters running on them were faster than a hardware
Lisp implementation. 

I think a similar story can be held for Python; it might be
possible to build a Python Machine, but making it as fast
as a Python imterpreter on a i486 will be *very* difficult.

Moreover, what do you do with all those C extension modules?
Anyone volunteering to extend gcc with a Python-opcode backend?

> Of course, with 
> soon-to-arrive Java chips and JPython there might be not 
> that much need for it...

I guess that Sun is relearning the lesson of the Lisp Machine
the hard way...

Greetings,

Stephan



From mark at chem.uwa.edu.au  Thu Jun  3 22:31:20 1999
From: mark at chem.uwa.edu.au (Mark C Favas)
Date: 4 Jun 99 02:31:20 GMT
Subject: intractable(?) Tkinter scrollbar problem
References: <19990601001054.36247@nms.otc.com.au> <jonathan-3105992257320001@borg.onegoodidea.net> <19990601124302.48567@nms.otc.com.au> <3753ECEE.58AE682A@mindspring.com> <19990603122902.18684@nms.otc.com.au>
Message-ID: <mark.928463480@declan>

Greg McFarlane <gregm at iname.com> writes:

>In private mail, someone suggested adding a delay before actually
>mapping/unmapping the scrollbar.  This fix stops the continuous
>mapping and unmapping of the scrollbars.

Unfortunately, this fix doesn't quite work on my system (DEC Alpha,
Digital Unix 4.0D)... At 100ms I get 10-15 redraws, at 200ms 40-50 redraws,
at 1-10ms around 10 redraws, etc.

Mark Favas
--
Email  - mark at chem.uwa.edu.au      ,-_|\                           Mark C Favas
Phone  - +61 9 380 3482           /     \               Department of Chemistry
Fax    - +61 9 380 1005      ---> *_,-._/   The University of Western Australia
                                       v                               Nedlands
Loc    - 31.97 S, 115.81 E                               Western Australia 6009



From moshez at math.huji.ac.il  Wed Jun 16 06:38:04 1999
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 16 Jun 1999 13:38:04 +0300
Subject: GC In Python: YAS (Yet Another Summary)
Message-ID: <Pine.SUN.3.95-heb-2.07.990616133606.13037B-100000@sunset.ma.huji.ac.il>

Regarding the recend discussion here about garbage collection, I wrote 
a small article.

If you plan on getting involved in the discussion, you might want to 
read this: it contains many of the old pros and cons, so you'll know what
you're up against.
--
Moshe Zadka <mzadka at geocities.com>. 
QOTD: My own exit is more likely to be horizontal then perpendicular.






From billtut at microsoft.com  Wed Jun 16 08:15:23 1999
From: billtut at microsoft.com (Bill Tutt)
Date: Wed, 16 Jun 1999 12:15:23 GMT
Subject: do anonymous pipes normally work on NT?
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8100DB90DDC@RED-MSG-50>

Well, I finally noticed this thread, and I'll chime in with what I found out
so far:
(No, I haven't figured it all out yet, its time for sleep.)

Symptoms:
 python -u cat.py < cat.py  is allowed to read from stdin.
 python cat.py < cat.py isn't allowed to read from stdin.
 python -u cat.py > blah is allowed to write to stdout
 python cat.py > blah isn't allowed to wrtie to stdout.
 python runproc.py (with -u) cat.py isn't allowed to write to stdout
 python runproc.py cat.py isn't allowed to write to stdout.

-u does two things:
1) turns stdin/stdout into binary streams
2) turns off stdio buffering

The binary thing shouldn't make any difference. The stdio thing might make a
difference, but how isn't obvious from the read.c source code.

Now the big mystery is: Why does the CRT care? (prolly need to debug read.c
to find out) 
The 2nd question is: Why doesn't -u fix the runproc.py case?

I think the answer to the 2nd question may come from the
	::InitializeSecurityDescriptor(&m_sd, SECURITY_DESCRIPTOR_REVISION);
line in PySECURITY_ATTRIBUTES's constructor.

I'm guessing (not having debug bits atm) that this causes a security
descriptor that doesn't have the default permissions that the current
process has to become filled in. 

(Default permissions are inferred from the current process/thread if
lpSecurityDescriptor is NULL)  

This in turn causes reading/writing to the inherited handles to be denied
because the new process doesn't have the correct permissions. Which in turn
turns into errno == EBADF in read.c.


This seems to be the only meaningful difference between the Python code in
runproc.py, and the win32popen.cpp code I wrote...

So the first thing to try might be to try commenting out the
InitializeSecurityDescriptor line, and rebuilding pywintypes15.dll, and then
using -u when running cat.py.

The next thing to do is to debug _read.c (in the MS CRT) and see why its
complaining if -u isn't specified on the command line.

If somebody could help explore where exactly errno is getting set to EBADF
during the non- "runproc.py" cases that'd be a great help.

Bill
Not part of product support, or PR.




From graham at sloth.math.uga.edu  Fri Jun  4 13:08:39 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 4 Jun 1999 17:08:39 GMT
Subject: "The Python Way"
References: <001101beae49$39af7bc0$949e2299@tim> <7j8pra$55v$4@cronkite.cc.uga.edu> <00f301beaea8$39f419b0$f29b12c2@pythonware.com>
Message-ID: <7j916n$6lk$3@cronkite.cc.uga.edu>

Fredrik Lundh (fredrik at pythonware.com) wrote:
: M> [Fredrik Lundh]
: M> > Would both Guido and TIm Peters be willing to collaborate on a short
: M> > paper
: M> 
: M> Oh thank you Fredrik for posting the above. Does anyone else consider
: M> that the above sounds slightly paranoid?
: 
: I didn't write that.  Sure sounds like you're slightly
: paranoid, Graham.

Not paranoid, just sick of having reasonable well though out suggestions
for Python met with "you are trying to bend Python into uncomfortable
positions ... I suspect that ...". Apologies if you didn't write the
above. I will leave it M to respond then!

graham
-- 
                    One day in a nuclear age 
                  They may understand our rage
              We build machines that we can't control
              And bury the waste in a great big hole



From banderson at boi.hp.com  Thu Jun 24 17:13:13 1999
From: banderson at boi.hp.com (Bill Anderson)
Date: Thu, 24 Jun 1999 15:13:13 -0600
Subject: Python Webserver
References: <3767A69D.50DC7FBF@serop.abb.se>
Message-ID: <37729F69.6F93B78E@boi.hp.com>

"Thomas S. Strinnhed" wrote:
> 
> Hi
> 
> I don't know if I've dreamt, but I think I saw something about
> a webserver written in Python. (It's not the embedded-in-Apache-thing
> I'm thinking of)
> Now I just can't seem to find it anywhere, so does someone know anything
> about this. Hipshots are welcome. I thought it'd be something fun to
> play with.
> 
> Thanks
>  -- Thomas S. Strinnhed, thstr at serop.abb.se

Is Zope the 'embedded-in-apache-thing' you refer to?
If so, you should note that it doesn't *need* apache, and can be used
seperately (I do it this way).

Zope 1.1x uses the Python httpserver mentioned in some of the other
posts, whereas zope 2.0.x(alpha!) uses Zserver, based on Medusa, as
mention in the other posts. :-)

-- 
Bill Anderson 					Linux Administrator
MCS-Boise (ARC)					banderson at boi.hp.com
My opinions are just that; _my_ opinions.



From tjs at longford.cs.monash.edu.au  Mon Jun 28 22:14:41 1999
From: tjs at longford.cs.monash.edu.au (Toby J Sargeant)
Date: Tue, 29 Jun 1999 02:14:41 GMT
Subject: ANN: Stackless Python 0.2
In-Reply-To: <000c01bec11f$8dfbf800$e19e2299@tim>; from Tim Peters on Mon, Jun 28, 1999 at 12:34:48AM -0400
References: <000c01bec11f$8dfbf800$e19e2299@tim>
Message-ID: <19990629121441.A19011@longford.cs.monash.edu.au>

On Mon, Jun 28, 1999 at 12:34:48AM -0400, Tim Peters wrote:
> Nope!  It's a beauty of the implementation already that each code object
> knows exactly how much "Python stack space" it needs, and (just) that much
> is allocated directly into the code object's runtime frame object.  IOW,
> there isn't "a Python stack" as such, so there's nothing to change here --
> the stack is implicit in the way frames link up to each other.

This might be completely irrelevant, but during the course of my masters, I
considered doing this kind of thing to java in order to allow asynchronous
threads to share stack frames (don't ask...). My supervisor complained bitterly
on the grounds that function invocations where orders of magnitude more
common than object creation, and hence associating memory
allocation/deallocation with every call was considered horrendously
inefficient.

It seems that this should affect Stackless Python equally as much. Does anyone
have anything to add on the subject? I would imagine that frames could be
allocated and managed in chunks to alleviate a lot of the memory management
load...

Toby.



From robin at jessikat.demon.co.uk  Tue Jun 29 07:35:31 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Tue, 29 Jun 1999 12:35:31 +0100
Subject: ANN: Stackless Python 0.2
References: <7l8fut$es6@dfw-ixnews6.ix.netcom.com>
 <000801bec207$142b6360$719e2299@tim>
Message-ID: <O3JyuHAD+Ke3Ew2y@jessikat.demon.co.uk>

In article <000801bec207$142b6360$719e2299 at tim>, Tim Peters
<tim_one at email.msn.com> writes
>[G. David Kuhlman]
...
>Py_BEGIN_ALLOW_THREADS
>ThreadedSpam();
>Py_END_ALLOW_THREADS
>
>it's-only-confusing-if-you-think-about-it-too-much<wink>-ly y'rs  - tim
>
>
>
>
so do I then have to poll ThreadedSpam myself to see if it's finished or
is there a python api version of mutexes/locks etc.
-- 
Robin Becker



From alex at mindlube.com  Thu Jun 24 20:07:43 1999
From: alex at mindlube.com (Alex Rice)
Date: 24 Jun 1999 18:07:43 -0600
Subject: newbie idiom question
References: <m3emj4yezq.fsf@monsoon.swcp.com> <m3emj43fnl.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <m37lotdtww.fsf@monsoon.swcp.com>

Thanks everyone. I think I grok it now.

Alex Rice

Michael Hudson <mwh21 at cam.ac.uk> writes:

> 
> If I know I'm not going to be altering the length of the list, I
> generally do it like this:
> 
> for i in range(len(a)):
>     if is_cooked(a[i]):
>         a[i] = "cooked"
> 
> If the list might be changing length, then I generally do
> 
> i = 0
> while i < len(a):
>     ...
> 
> and keep track of i by hand.
> 
> HTH
> Michael




From MHammond at skippinet.com.au  Sat Jun 12 19:25:48 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Sun, 13 Jun 1999 09:25:48 +1000
Subject: do anonymous pipes normally work on NT?
References: <7juatc$fef$1@nnrp1.deja.com>
Message-ID: <7juq84$df3$1@m2.c2.telstra-mm.net.au>

Nothing is particularly obvious, although the code is too large to really be
sure.

* Have you seen win32pipe.popen*?

* I can't see where you keep references to the handles for the sub-process.
Is it possible that the handles are being closed by the parent process,
thereby closing the child process handles?

* Personally I would use win32file.CreateFile instead of
os.open/msvcrt.get_osfhandle.  One advantage is that a layer of indirection
you are not using is removed.  Second advantage is that you deal with
"PyHANDLE" objects, and explicit closing is not necessary (a-la files)

* You may have more luck spotting the problem (or getting someone else to)
if you provide all this in a single, general purpose function.

def CoProc( command_line, infile, outfile ):
  ...

with all command-line parsing and other app specific stuff that doesnt
relate to the problem removed.

Hope this helps...

Mark.






From tim_one at email.msn.com  Thu Jun 10 22:33:02 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 10 Jun 1999 22:33:02 -0400
Subject: Broken 1.5.2 installation? Missing cw3215mt.dll
In-Reply-To: <014501beb3af$48f2e840$a9807018@yec1.on.wave.home.com>
Message-ID: <000901beb3b2$baa6d980$329e2299@tim>

> Problem is encountered when loading command-line Python or IDLE.  IDLE is
> non-functional, command line works fine after acknowledging the
> missing DLLs (as far as I can see).
> ...
> Could be the Cygnus TCL is interfering I suppose.

Aha!  Stop your grinnin' and drop your linen, Mike.  Having more than one
version of tk*.dll or tcl*.dll anywhere in your entire *city* is like
putting a male brain-eater in your left ear and a female brain-eater in your
right ear at the peak of brain-eater mating season, then betting you can
keep them apart by force of moral suasion.  You should reformat your drives
and reinstall NT to be *sure* <wink>, but before that try uninstalling every
Tcl/Tk you have, "dir/s" your drives from the roots to make sure nothing
matching tk*.dll or tcl*.dll survived, uninstall Python, and reinstall
Python with the One True Tcl/Tk that comes with it.

followups-to-cameron-laird<wink>-ly y'rs  - tim





From bwarsaw at cnri.reston.va.us  Thu Jun  3 19:17:15 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Thu, 3 Jun 1999 19:17:15 -0400 (EDT)
Subject: "The Python Way"
References: <045501beadf8$610e62e0$f29b12c2@pythonware.com>
	<9906031407150D.00552@quadra.teleo.net>
	<Pine.A32.3.90.990603172339.1626A-100000@elvis.med.Virginia.EDU>
Message-ID: <14167.3323.707498.300295@anthem.cnri.reston.va.us>

>>>>> "SDM" == Steven D Majewski <sdm7g at Virginia.EDU> writes:

    SDM> What makes you think that Tim and Guido would agree on "The
    SDM> Python Way" ?

What makes you think /Guido/ would agree on "The Python Way"? :)



From tim_one at email.msn.com  Sat Jun 19 22:17:16 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 19 Jun 1999 22:17:16 -0400
Subject: Crawling Python
In-Reply-To: <376b5b32.6312672@news.tpnet.pl>
Message-ID: <000001bebac3$04ea3700$d79e2299@tim>

[MK]
> ...
> Win98. I tried the same several more times in DOS box and got
> different results every time:
>
> >>> for i in range(10000):
> ...     s=s+`range(i,i+20)`
> ...
> >>>
> >>>
> >>> len(s)
> 1645360
> >>>

This didn't show len(s) *before* the loop began, so all that can be
concluded is that the loop added no more than 1645360 bytes to s.

> >>> s=''
> >>>
> >>> for i in range(10000):
> ...     s=s+`range(i,i+20)`
> ...
> >>> len(s)
> 1178515
> >>>

This one shows the loop adding exactly 1178515 bytes -- which it does for me
too.

> >>> for i in range(10000):
> ...     s=s+`range(i,i+20)`
> ...
> >>> len(s)
> 2357030
> >>>

s wasn't reset before this loop, so the loop adds an additional 1178515
bytes on top of the 1178515 s started with, and 1178515 + 1178515 = 2357030.

> Now _that_ is weird.

not-really<wink>-ly y'rs  - tim





From kas at maps.magnetic-ink.dk  Tue Jun 22 13:54:37 1999
From: kas at maps.magnetic-ink.dk (Klaus Alexander Seistrup)
Date: Tue, 22 Jun 1999 19:54:37 +0200
Subject: need some fast balanced-tree datastructure
References: <376ED372.FA73994F@cs.ucsd.edu>
Message-ID: <376FCDDD.1D1BA53E@maps.magnetic-ink.dk>

Pavel Velikho wrote:

> I am looking for a fast implementation of incremental container that
> maintains an ordering on the keys... (i.e. I want an AVL, Red-Black,

The Nightmare Software Collection has an AVL-module:

	http://www.nightmare.com/software.html


  //Klaus

-- 
???[ Magnetic Ink ]?????????????????????????????????????????????????????????
???[ http://www.magnetic-ink.dk/ ]??????????????????????????????????????????