From gary at zope.com  Fri Dec  1 02:49:51 2006
From: gary at zope.com (Gary Poster)
Date: Thu, 30 Nov 2006 20:49:51 -0500
Subject: [Pythonmac-SIG] Does anyone on the planet know how to build
	python-2.5 (WITH READLINE)?
In-Reply-To: <455B6B5B.5020208@lanl.gov>
References: 20060627143827.45945.qmail@web52303.mail.yahoo.com
	<455B6B5B.5020208@lanl.gov>
Message-ID: <EA666BD1-0799-49C9-AFCB-B147EB60153B@zope.com>


On Nov 15, 2006, at 2:32 PM, Todd Thal wrote:

> I am having one "hades" of a heckuva time getting readline support for
> Python-2.5 on a Mac OS 10.4.8.8.0

Thanks to Ronald and Skip for replying to this thread.  It took me a  
little bit to connect the dots from their replies, so I thought I'd  
share my experiences.  Maybe this can help someone, and maybe also  
lead to Ronald or someone else clarifying the right way to use his  
CFLAGS and LDFLAGS patch.

I'm using darwinports, so first I had to

sudo port install readline

Once that was done, I wanted to build a local, non-framework Python  
with readline.  darwinports uses /opt/local, so Ronald's approach led  
me to this best guess:

$ CFLAGS=-I/opt/local/include LDFLAGS=-L/opt/local/lib ./configure -- 
prefix=/Users/gary/py
$ make

However, that was unsuccessful.

$ ./python.exe
Python 2.4.4 (#1, Nov 30 2006, 20:36:29)
[GCC 4.0.1 (Apple Computer, Inc. build 5367)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
 >>> import readline
Traceback (most recent call last):
   File "<stdin>", line 1, in ?
ImportError: No module named readline

Maybe someone could let me know what I did wrong?

Skip's approach was a little more unusual, at least to me, since it  
was a step in-between `make` and `make install`.  However, it  
worked. :-)  If, after a configure and a make (but, again, before a  
make install) I run this command...

$ ./python.exe setup.py build_ext --include-dirs=/opt/local/include -- 
library-dirs=/opt/local/lib

...then readline appears in glory.

$ ./python.exe Python 2.4.4 (#1, Nov 30 2006, 20:36:29)
[GCC 4.0.1 (Apple Computer, Inc. build 5367)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
 >>> import readline
 >>>

Yay!

Of course, I could have copied readline.so from MacPython but that  
would have been cheating. :-)

HTH someone

Gary

From ronaldoussoren at mac.com  Fri Dec  1 17:04:03 2006
From: ronaldoussoren at mac.com (Ronald Oussoren)
Date: Fri, 1 Dec 2006 17:04:03 +0100
Subject: [Pythonmac-SIG] Does anyone on the planet know how to build
	python-2.5 (WITH READLINE)?
In-Reply-To: <EA666BD1-0799-49C9-AFCB-B147EB60153B@zope.com>
References: "20060627143827.45945.qmail@web52303.mail.yahoo.com"
	<455B6B5B.5020208@lanl.gov>
	<EA666BD1-0799-49C9-AFCB-B147EB60153B@zope.com>
Message-ID: <DBF30ED3-E5F3-46A4-AF5D-17F299D8B553@mac.com>


On  1 Dec 2006, at 2:49 AM, Gary Poster wrote:

>
> On Nov 15, 2006, at 2:32 PM, Todd Thal wrote:
>
>> I am having one "hades" of a heckuva time getting readline support  
>> for
>> Python-2.5 on a Mac OS 10.4.8.8.0
>
> Thanks to Ronald and Skip for replying to this thread.  It took me a
> little bit to connect the dots from their replies, so I thought I'd
> share my experiences.  Maybe this can help someone, and maybe also
> lead to Ronald or someone else clarifying the right way to use his
> CFLAGS and LDFLAGS patch.
>
> I'm using darwinports, so first I had to
>
> sudo port install readline
>
> Once that was done, I wanted to build a local, non-framework Python
> with readline.  darwinports uses /opt/local, so Ronald's approach led
> me to this best guess:
>
> $ CFLAGS=-I/opt/local/include LDFLAGS=-L/opt/local/lib ./configure --
> prefix=/Users/gary/py

This should do the trick, more or less. The build script for the  
python.org installer uses:

./configure --prefix=/Users/gary/py LDFLAGS=-L/opt/local/lib OPT=-I/ 
opt/local/include

> $ make

If you're using macports you might as well do 'port install python2.4'.

Ronald

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3562 bytes
Desc: not available
Url : http://mail.python.org/pipermail/pythonmac-sig/attachments/20061201/fd6fbff3/attachment.bin 

From Chris.Barker at noaa.gov  Fri Dec  1 18:18:29 2006
From: Chris.Barker at noaa.gov (Christopher Barker)
Date: Fri, 01 Dec 2006 09:18:29 -0800
Subject: [Pythonmac-SIG] [Fwd: [wxPython-users] wxPython 2.8 RC - Call for
	testers! :-)]
Message-ID: <457063E5.1040801@noaa.gov>

HI all,

This went out over the wxPython list, but I thought I'd post it here as 
generally, the Mac gets the least testing. Also, given that it looks 
like Apple is going to put 2.8 in their next release, and they don't 
have a history of updating things like that, it's all the more important 
to get as much testing as possible on the Mac.

-Chris

-------- Original Message --------
Subject: [wxPython-users] wxPython 2.8 RC - Call for testers! :-)
Date: Thu, 30 Nov 2006 16:38:59 -0800
From: Kevin Ollivier <kevino at tulane.edu>
Reply-To: wxPython-users at lists.wxwidgets.org
To: wxPython-users at lists.wxwidgets.org

Hi all,

Robin made an announcement about the upcoming release of wxWidgets
2.8 next week on wxpython-dev, but I thought that 2.8 Release
Candidate (RC) testing merited an extra call for help on wxPython-
users as well.

So, if you're good at making programs crash, implode, or otherwise
freak out, we need your help! :-) We're on the last stretch for 2.8,
and particularly as it is going to be included in the default install
of OS X 10.5, Leopard, we'd like it to be as solid and slick as
possible. So if you have apps, frameworks, scripts, or one-liners
written using wx, we would appreciate any time you could spend
testing them with the 2.8 RC. Even if it's just 15 minutes, or with
one script, or testing the demo... every little bit helps. We
probably won't get to everything reported before the release, but
it's better to find all the issues we can, so we can at least
prioritize and fix the really nasty ones.

Oh, and if you don't mind fixing bugs too, well, I promise we won't
complain! :-)

The wxPython 2.8 RC can be found here:

http://wxpython.wxcommunity.com/preview/20061129/

Thanks!

Kevin

---------------------------------------------------------------------
To unsubscribe, e-mail: wxPython-users-unsubscribe at lists.wxwidgets.org
For additional commands, e-mail: wxPython-users-help at lists.wxwidgets.org


-- 
Christopher Barker, Ph.D.
Oceanographer

Emergency Response Division
NOAA/NOS/OR&R            (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception

Chris.Barker at noaa.gov

From cdamundsen at gmail.com  Sat Dec  2 04:45:38 2006
From: cdamundsen at gmail.com (Craig Amundsen)
Date: Fri, 1 Dec 2006 19:45:38 -0800
Subject: [Pythonmac-SIG] controlling iTunes with appscript
In-Reply-To: <20061128193021.GA48180@uiuc.edu>
References: <456C66F8.4080306@eurotalk.com>
	<8c7f10c60611280918l786afbecvbfc73b9734381935@mail.gmail.com>
	<456C757C.6040603@eurotalk.com>
	<8dc0c28f0611281109k366efe79lcff010f0ee173f5@mail.gmail.com>
	<20061128193021.GA48180@uiuc.edu>
Message-ID: <8dc0c28f0612011945n988e007x1cf0e7bc4a9b6d3e@mail.gmail.com>

Hi -

I finally had time to play around with appscript, and can now get the
list of tracks. For reasons I don't understand, if I try to get all
the file_tracks in the library, iTunes shoots up to about 80% CPU
utilization and never comes back, even if I set the timeout to 15
minutes. But if I put all the songs in a playlist and get the songs
with

app('iTunes').sources[1].user_playlists['PlaylistOfAllSongs'].file_tracks.get()

it takes a while, but eventually comes back with list of tracks.

> Just FYI if you don't know already, iTunes uses aliases to keep track
> of file locations, so if you're not moving across volumes, you
> shouldn't have to do anything at all.

> You may end up needing to use HFS rather than POSIX paths, or even
> aliases, depending on what iTunes expects.

If I get the location of a track eg,

>>> t.location.get()
macfile.Alias(u'/Volumes/FirewireHD/iTunes/Air/Talkie Walkie/02 Cherry
Blossom Girl.mp3')

the location is definitely a macfile.Alias

I am moving my songs between Volumes, so I do have to do something. If I try:

>>> f = macfile.Alias(u'/Volumes/RAID/iTunes/Air/Talkie Walkie/02
Cherry Blossom Girl.mp3')
>>> t.location.set(f)

I get this error:

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/aeosa/appscript/reference.py",
line 281, in __call__
    raise CommandError(self, (args, kargs), e)
appscript.reference.CommandError: 'NoneType' object has no attribute '__name__'
        Failed command:
app(u'/Applications/iTunes.app').sources.ID(41).user_playlists.ID(388392).file_tracks.ID(388445).location.set(macfile.Alias(u'/Volumes/RAID500/iTunes/Air/Talkie
Walkie/02 Cherry Blossom Girl.mp3'))

Any hints would be greatly appreciated.

Thanks,
- Craig

From bthom at cs.hmc.edu  Mon Dec  4 08:14:49 2006
From: bthom at cs.hmc.edu (belinda thom)
Date: Sun, 03 Dec 2006 23:14:49 -0800
Subject: [Pythonmac-SIG] seeking matplotlib install advice
Message-ID: <400392E2-CA02-44A0-AEF5-AA5AC8DD7109@cs.hmc.edu>

Hi,

I've installed matplotlib (v 0.87.6_0) via macports (os x 10.4.8,  
python 2.4.3). I've tried following the "interactive" instructions at  
matplotlib.sourceforge.net/interactive.html, which tells me to modify  
matplotlibrc as follows:

interactive : True
backend : TkAgg

My general usage is via ipython -pylab, which should automatically  
setup the necessary extra thread for plotting, but I've seen similar  
behavior in vanilla python shells as well.

I can run matplotlib commands (e.g. plot([1,2,3])) but plot windows  
are never shown (even after I explicitly call show()). I can save the  
figure and view it as a pgn, so much is working.

When I configure matplotlibrc via the rec'd TkAgg backend, I get the  
following error:

Traceback (most recent call last):
   File "/opt/local/bin/ipython", line 28, in ?
     IPython.Shell.start().mainloop()
   [snip]
   File "/opt/local/lib/python2.4/site-packages/matplotlib/backends/ 
backend_tkagg.py", line 8, in ?
     import tkagg                 # Paint image to Tk photo blitter  
extension
   File "/opt/local/lib/python2.4/site-packages/matplotlib/backends/ 
tkagg.py", line 1, in ?
     import _tkagg
ImportError: No module named _tkagg

A google search led me to believe I didn't have "the proper tcl-dev/ 
tk-dev" directories. I then noticed
there was a +gt2k macports matplotlib variant, and found another  
search snippet that led me to believe
installing this new variant might relieve my woes.

No such luck.

Has anyone gotten matplotlib to plot figures interactively in the  
macports/python24 setting?

Advice greatly appreciated.

--b 

From njriley at uiuc.edu  Mon Dec  4 08:37:57 2006
From: njriley at uiuc.edu (Nicholas Riley)
Date: Mon, 4 Dec 2006 01:37:57 -0600
Subject: [Pythonmac-SIG] controlling iTunes with appscript
In-Reply-To: <8dc0c28f0612011945n988e007x1cf0e7bc4a9b6d3e@mail.gmail.com>
References: <456C66F8.4080306@eurotalk.com>
	<8c7f10c60611280918l786afbecvbfc73b9734381935@mail.gmail.com>
	<456C757C.6040603@eurotalk.com>
	<8dc0c28f0611281109k366efe79lcff010f0ee173f5@mail.gmail.com>
	<20061128193021.GA48180@uiuc.edu>
	<8dc0c28f0612011945n988e007x1cf0e7bc4a9b6d3e@mail.gmail.com>
Message-ID: <20061204073757.GA14509@uiuc.edu>

On Fri, Dec 01, 2006 at 07:45:38PM -0800, Craig Amundsen wrote:
> >>> t.location.set(f)

Yeah, you can't do this.  If you look in the dictionary (for example,
opening it in Script Editor), you'll see:

location (alias, r/o) : the location of the file represented by this track

The location property is read only.

The only way I can think to move files across disks in iTunes and have
the references stay intact is to use the "Consolidate Library"
command.  To do that, reset the iTunes Music Library location
(Preferences > Advanced > General) to where you want the files to go,
then choose "Consolidate Library" from the Advanced menu, and all your
music will move there.

Sorry for giving you advice before checking what you wanted to do was
possible.  That's the second time today I've made that mistake
recently (the other one cost me about 2 hours of trying to compile
Kerberos on AIX today.)

In the strange coincidences category, "Cherry Blossom Girl" just came
on the streaming station I'm listening to.  Scary.

-- 
Nicholas Riley <njriley at uiuc.edu> | <http://www.uiuc.edu/ph/www/njriley>

From cdamundsen at gmail.com  Mon Dec  4 17:00:53 2006
From: cdamundsen at gmail.com (Craig Amundsen)
Date: Mon, 4 Dec 2006 08:00:53 -0800
Subject: [Pythonmac-SIG] controlling iTunes with appscript
In-Reply-To: <20061204073757.GA14509@uiuc.edu>
References: <456C66F8.4080306@eurotalk.com>
	<8c7f10c60611280918l786afbecvbfc73b9734381935@mail.gmail.com>
	<456C757C.6040603@eurotalk.com>
	<8dc0c28f0611281109k366efe79lcff010f0ee173f5@mail.gmail.com>
	<20061128193021.GA48180@uiuc.edu>
	<8dc0c28f0612011945n988e007x1cf0e7bc4a9b6d3e@mail.gmail.com>
	<20061204073757.GA14509@uiuc.edu>
Message-ID: <8dc0c28f0612040800s331ec0e7tcee42b792e3c668f@mail.gmail.com>

Hi -

On Fri, Dec 01, 2006 at 07:45:38PM -0800, Craig Amundsen wrote:
> > >>> t.location.set(f)
>
> Yeah, you can't do this.  If you look in the dictionary (for example,
> opening it in Script Editor), you'll see:
>
> location (alias, r/o) : the location of the file represented by this track
>
> The location property is read only.


I was afraid of that. Thanks for letting me know.

The only way I can think to move files across disks in iTunes and have
> the references stay intact is to use the "Consolidate Library"
> command.  To do that, reset the iTunes Music Library location
> (Preferences > Advanced > General) to where you want the files to go,
> then choose "Consolidate Library" from the Advanced menu, and all your
> music will move there.


I've tried that. I'm trying to move my library from one external Firewire
drive to an external RAID1 setup. Doing the consolidate moved about 10% of
the files and then crapped out with an error that said something about being
able to read/write a drive. Repeated attempts to consolidate the library now
die with the read/write error immediately. I don't think it's a problem with
either drive since I can read and write from both of them. It may be that
RAID is too slow to keep up with whatever iTunes is trying to do. I was
hoping I could do a manual copy and then update all the locations via
appscript. This having the advantage of allowing me to preserve metadata. I
guess it's time to forget about the metadata since I'm not going to do the
one-at-a-time location update from within iTunes.

- Craig
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/pythonmac-sig/attachments/20061204/7f8f7f99/attachment.htm 

From delza at livingcode.org  Mon Dec  4 19:23:17 2006
From: delza at livingcode.org (Dethe Elza)
Date: Mon, 4 Dec 2006 10:23:17 -0800
Subject: [Pythonmac-SIG] controlling iTunes with appscript
In-Reply-To: <8dc0c28f0612040800s331ec0e7tcee42b792e3c668f@mail.gmail.com>
References: <456C66F8.4080306@eurotalk.com>
	<8c7f10c60611280918l786afbecvbfc73b9734381935@mail.gmail.com>
	<456C757C.6040603@eurotalk.com>
	<8dc0c28f0611281109k366efe79lcff010f0ee173f5@mail.gmail.com>
	<20061128193021.GA48180@uiuc.edu>
	<8dc0c28f0612011945n988e007x1cf0e7bc4a9b6d3e@mail.gmail.com>
	<20061204073757.GA14509@uiuc.edu>
	<8dc0c28f0612040800s331ec0e7tcee42b792e3c668f@mail.gmail.com>
Message-ID: <6B54E4DF-3D89-4402-973A-FC19B12D76CB@livingcode.org>

On 4-Dec-06, at 8:00 AM, Craig Amundsen wrote:

> I've tried that. I'm trying to move my library from one external  
> Firewire
> drive to an external RAID1 setup. Doing the consolidate moved about  
> 10% of
> the files and then crapped out with an error that said something  
> about being
> able to read/write a drive. Repeated attempts to consolidate the  
> library now
> die with the read/write error immediately. I don't think it's a  
> problem with
> either drive since I can read and write from both of them. It may  
> be that
> RAID is too slow to keep up with whatever iTunes is trying to do. I  
> was
> hoping I could do a manual copy and then update all the locations via
> appscript. This having the advantage of allowing me to preserve  
> metadata. I
> guess it's time to forget about the metadata since I'm not going to  
> do the
> one-at-a-time location update from within iTunes.

Can you set the metadata from AppScript? I'm assuming you can, but  
haven't been following this thread that closely.  You could try  
backing up your iTunes/iTunes Music Library.xml file and write a  
script to parse it and set the metadata on your files once you've  
moved them.  It looks like a standard plist file, so you can use the  
python plist library to do the parsing.  I'm sure the unique Track ID  
will be different, but you could key off artist + title + size and  
reset the other information (play counts, rating, etc.).  I'm pretty  
sure iTunes reads this file on startup to set up its internal  
database, but I don't have time to test this theory right now.

If that doesn't work, there still may be value in thinking outside  
the AppScript box.

--Dethe

Windows has detected the mouse has moved. Please restart your system  
for changes to take effect.



From altern2 at gmail.com  Tue Dec  5 11:07:27 2006
From: altern2 at gmail.com (altern)
Date: Tue, 05 Dec 2006 11:07:27 +0100
Subject: [Pythonmac-SIG] weird pygame problem on 10.4
Message-ID: <457544DF.30607@gmail.com>

hi all

I just upgraded my old mac laptop to 10.4.8 from 10.3.9. I am running 
pygame + pyopengl scripts and I find this weird error that did not 
happen on 10.3.9. The 'canvas', the surface were the drawing happens it 
is misplaced within the window by 100 px to the right and 100 px to the 
top. So I get a 100 px wide empty band on the left bottom of the window.

If I run the same script using wxPython instead of pygame to handle the 
window events there is no problem at all.

The funny thing is that neither the pygame or the pyOpenGL examples have 
this problem, so it must be my code. But I dont really know where to 
start debugging from.

I am running python 2.4 and i installed all the stuff from
http://www.pythonmac.org/packages/py24-fat/index.html
including :
python-2.4.4-macosx2006-10-18.dmg
pygame-1.8.0pre-py2.4-macosx10.4.zip
PyOpenGL-2.0.2.01-py2.4-macosx10.4.zip

any hints would be very much appreciated

thanks!

enrike

From ronaldoussoren at mac.com  Wed Dec  6 07:39:11 2006
From: ronaldoussoren at mac.com (Ronald Oussoren)
Date: Wed, 6 Dec 2006 07:39:11 +0100
Subject: [Pythonmac-SIG] py2app and haxies
Message-ID: <6BAC4E30-C17B-47B1-906E-1983531231AD@mac.com>

Hi,

The current release of py2app doesn't mix well[1] with haxies and  
other backdoor mechanism's to insert code into applications (such as  
rogue input managers). I've checked in a workaround in the py2app  
repository, updating to 'py2app==dev' should be safe as there isn't  
much development at the moment.

Ronald

[1] Aka "my application crashes when a haxie is present"
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3562 bytes
Desc: not available
Url : http://mail.python.org/pipermail/pythonmac-sig/attachments/20061206/20da8449/attachment.bin 

From bthom at cs.hmc.edu  Thu Dec  7 08:27:41 2006
From: bthom at cs.hmc.edu (belinda thom)
Date: Wed, 06 Dec 2006 23:27:41 -0800
Subject: [Pythonmac-SIG] using matplotlib, ipython, etc. w/MacPython
Message-ID: <50CA97C7-2B53-4CB2-924B-FD9E965C8214@cs.hmc.edu>

Hi,

I've spent the last 2 weeks wrestling w/macports python/packages.  
Today, in advice to a question, Robert Kern advised that I use the  
MacPython "official" distribution.

I now need to quickly get MacPython up to the (hopefully less buggy)  
state than where I had gotten w/the macports setup. Specifically, I  
need to:

1) use matplotlib interactively, which requires something like TkAgg  
or WxAgg. (W/macports WxAgg is the only backend that works)
2) have python access to Tk (aka tkinter)
3) have ipython
4) have Numeric access

I'd also be interested in numpy/scipy if it is an easy install.

I'm entirely new to Tk, so there could use more detailed advice.

I did find a useful blurb on http://matplotlib.sourceforge.net/ 
installing.html (appended), but alas:

1) pythonmac.org/packages server is not responding
2) the MacEnthon link leads to a page that has yet to be created
3) the "step-by-step" link says "under revision; come back soon"

Perhaps the Moad bundle will work?

Advice gratefully appreciated.

--b

--------------
All of the backends run on OS X. Chris Barker has built a binary  
package (fink users see below) for matplotlib which is hosted on  
pythonmac, and works with Agg, Wx and Tk; see the step-by-step  
instructions kindly provided by Michael Tobis. It does not included  
the GTK* backends, because pygtk is usually provided through fink and/ 
or darwinport. This version should work on an stock OS-X 10.3.*, and  
you need Numeric or numarray (available on the same site), and TK or  
wxPython, if you want to use those back ends.

There are other binary packages to look at which each have different  
components included. Robery Kern has built an all-in-one installer  
which includes scipy, Numeric, numarray, matplotlib, ipython, VTK,  
MayaVi, PIL, the enthought tool suite and much more; see >MacEnthon.  
Charlie Moad has bundled matplotlib with freetype, png, wx, numarray,  
numeric and doesn't rely on fink or darwinports; available in the  
Tools Section. Another bundle is scisoft which packages many  
scientific packages for python on OS X, including matplotlib,  
although it is designed for astronomical analysis.

If you want to compile yourself on OS X, make sure you read the  
compiling instructions above. You will need to install freetype2,  
libpng and zlib via fink or from src, or from a package manager such  
as fink or i-Installer. You will also need the base libraries for a  
given backend. Eg, if you want to run TkAgg, you will need a python  
with Tkinter; if you want to use WxAgg, install wxpython. See  
backends for a more comprehensive discussion of the various backend  
requirements. Edit setup.py to configure the backends you want to  
build as described above.

Note when running a GUI backend in OSX, you should launch your  
programs with pythonw rather than python, or you may get  
nonresponsive GUIs.

-----

I do not want to compile code myself unless absolutely necessary. I  
was wondering what was up with the "MacEnton" suite; clicking on the  
link described at the above web page informs me that the "MacEnthon"  
page does not exist.

  

From robert.kern at gmail.com  Thu Dec  7 08:38:39 2006
From: robert.kern at gmail.com (Robert Kern)
Date: Thu, 07 Dec 2006 01:38:39 -0600
Subject: [Pythonmac-SIG] using matplotlib, ipython, etc. w/MacPython
In-Reply-To: <50CA97C7-2B53-4CB2-924B-FD9E965C8214@cs.hmc.edu>
References: <50CA97C7-2B53-4CB2-924B-FD9E965C8214@cs.hmc.edu>
Message-ID: <el8gen$q7f$1@sea.gmane.org>

belinda thom wrote:
> I do not want to compile code myself unless absolutely necessary. I  
> was wondering what was up with the "MacEnton" suite; clicking on the  
> link described at the above web page informs me that the "MacEnthon"  
> page does not exist.

Umm, ignore it. It targetted a now-old Python distribution, and I don't have
time to update it anymore. References recommending it should be removed.

-- 
Robert Kern

"I have come to believe that the whole world is an enigma, a harmless enigma
 that is made terrible by our own mad attempt to interpret it as though it had
 an underlying truth."
  -- Umberto Eco


From bob at redivi.com  Thu Dec  7 08:58:21 2006
From: bob at redivi.com (Bob Ippolito)
Date: Thu, 7 Dec 2006 15:58:21 +0800
Subject: [Pythonmac-SIG] using matplotlib, ipython, etc. w/MacPython
In-Reply-To: <50CA97C7-2B53-4CB2-924B-FD9E965C8214@cs.hmc.edu>
References: <50CA97C7-2B53-4CB2-924B-FD9E965C8214@cs.hmc.edu>
Message-ID: <6a36e7290612062358g5c5c50ddia658ef37ffa0e7b7@mail.gmail.com>

On 12/7/06, belinda thom <bthom at cs.hmc.edu> wrote:
>
> 1) pythonmac.org/packages server is not responding

Works fine here.

-bob

From bthom at cs.hmc.edu  Thu Dec  7 09:29:58 2006
From: bthom at cs.hmc.edu (belinda thom)
Date: Thu, 07 Dec 2006 00:29:58 -0800
Subject: [Pythonmac-SIG] pythonmac packages question
Message-ID: <B94FADCE-B655-49E3-B08D-51776FC35AB2@cs.hmc.edu>

Hi,

As of my prior post, I was able to get into pythonmac.org.

There I found many of the packages it seems I'll need. I was able to  
install Numeric and WxPython (I chose the later dmg, v 2.7.1.3) and  
matplotlib.

This site has no Tkinter and no IPython packages. Where do MacPython  
people go for such things?

Meanwhile...

If I run python alone and try out pylab, I get the following errors:
----------------
85 % python
Python 2.4.4 (#1, Oct 18 2006, 10:34:39)
[GCC 4.0.1 (Apple Computer, Inc. build 5341)] on darwin
[snip]
 >>> import pylab
 >>> pylab.plot([1,2,3],[10,20,30])
/Library/Frameworks/Python.framework/Versions/2.4/lib/python2.4/site- 
packages/matplotlib/axes.py:1223: UserWarning: xmax too close to  
xmin; adjusting
   warnings.warn("xmax too close to xmin; adjusting")
Traceback (most recent call last):  File "/Library/Frameworks/ 
Python.framework/Versions/2.4/lib/python2.4/site-packages/matplotlib/ 
axes.py", line 1227, in set_xlim
     self.viewLim.intervalx().set_bounds(xmin, xmax)
TypeError: only length-1 arrays can be converted to Python scalars.
------------------

I've read elsewhere IPython is the rec'd way to use pylab b/c it sets  
up some needed things like an extra thread. As a temporary hack, I  
added a symlink to /Library/Frameworks/Python.framework/Versions/2.4/ 
lib/python2.4/site-packages that points to the macports IPython  
install I'd done previously. This seems to work (ipython runs).

However, when I do:
-------------------
4 % ipython -pylab
/Library/Frameworks/Python.framework/Versions/2.4/lib/python2.4/site- 
packages/IPython/Shell.py:709: DeprecationWarning: The wxPython  
compatibility package is no longer automatically generated or activly  
maintained.  Please switch to the wx package as soon as possible.
   import wxPython.wx as wx
Python 2.4.4 (#1, Oct 18 2006, 10:34:39)
Type "copyright", "credits" or "license" for more information.
[snikp]
IPython 0.7.2 -- An enhanced Interactive Python.
   Welcome to pylab, a matplotlib-based Python environment.
   For more information, type 'help(pylab)'.

In [1]: from pylab import *
In [2]: plot([1,2,3])
[snip]
exceptions.MemoryError                               Traceback (most  
recent call last)
[snip]
/Library/Frameworks/Python.framework/Versions/2.4/lib/python2.4/site- 
packages/matplotlib/backends/backend_wx.py in _onPaint(self, evt)
MemoryError: _wxagg.convert_agg_to_wx_bitmap(): could not create the  
wx.Bitmap
 > /Library/Frameworks/Python.framework/Versions/2.4/lib/python2.4/ 
site-packages/matplotlib/backends/backend_wxagg.py(63)draw()
      62
---> 63         self.bitmap = _convert_agg_to_wx_bitmap 
(self.get_renderer(), None)
      64         if repaint:
----------

The initial warning implies the MacPython stuff isn't happy with  
macport's IPython's.

I'm confused as to how to proceed. Advice greatly appreciated.

--b



From njriley at uiuc.edu  Thu Dec  7 09:35:10 2006
From: njriley at uiuc.edu (Nicholas Riley)
Date: Thu, 7 Dec 2006 02:35:10 -0600
Subject: [Pythonmac-SIG] pythonmac packages question
In-Reply-To: <B94FADCE-B655-49E3-B08D-51776FC35AB2@cs.hmc.edu>
References: <B94FADCE-B655-49E3-B08D-51776FC35AB2@cs.hmc.edu>
Message-ID: <20061207083510.GB52065@uiuc.edu>

On Thu, Dec 07, 2006 at 12:29:58AM -0800, belinda thom wrote:
> This site has no Tkinter and no IPython packages. Where do MacPython  
> people go for such things?

Tkinter comes with MacPython; you don't need to install it separately.

For IPython, just use easy_install.  You may need to add Python's bin
directory to your path, depending on how you configure it.

-- 
Nicholas Riley <njriley at uiuc.edu> | <http://www.uiuc.edu/ph/www/njriley>

From ronaldoussoren at mac.com  Thu Dec  7 09:38:44 2006
From: ronaldoussoren at mac.com (Ronald Oussoren)
Date: Thu, 7 Dec 2006 09:38:44 +0100
Subject: [Pythonmac-SIG] pythonmac packages question
In-Reply-To: <B94FADCE-B655-49E3-B08D-51776FC35AB2@cs.hmc.edu>
References: <B94FADCE-B655-49E3-B08D-51776FC35AB2@cs.hmc.edu>
Message-ID: <75A0405E-EE3E-4E6F-8A65-B48025011AD3@mac.com>


On Dec 7, 2006, at 9:29 AM, belinda thom wrote:

>
> This site has no Tkinter and no IPython packages. Where do MacPython
> people go for such things?

tkinter is part of the standard installation for macpython.

Ronald

From bthom at cs.hmc.edu  Thu Dec  7 09:38:36 2006
From: bthom at cs.hmc.edu (belinda thom)
Date: Thu, 07 Dec 2006 00:38:36 -0800
Subject: [Pythonmac-SIG] pythonmac packages question
In-Reply-To: <A484A536-370B-4891-B728-14D0893F65EB@cs.hmc.edu>
References: <B94FADCE-B655-49E3-B08D-51776FC35AB2@cs.hmc.edu>
	<20061207083510.GB52065@uiuc.edu>
	<A484A536-370B-4891-B728-14D0893F65EB@cs.hmc.edu>
Message-ID: <07B00F4F-0ECE-48D9-886D-5A9DAE290FF9@cs.hmc.edu>

Whoops, forgot to reply to the forum...

On Dec 7, 2006, at 12:36 AM, belinda thom wrote:

> Thanks Nicholas,
>
> So how do I use easy install :-)?
>
> On Dec 7, 2006, at 12:35 AM, Nicholas Riley wrote:
>
>> For IPython, just use easy_install.  You may need to add Python's bin
>> directory to your path, depending on how you configure it.
>


From njriley at uiuc.edu  Thu Dec  7 09:48:16 2006
From: njriley at uiuc.edu (Nicholas Riley)
Date: Thu, 7 Dec 2006 02:48:16 -0600
Subject: [Pythonmac-SIG] pythonmac packages question
In-Reply-To: <07B00F4F-0ECE-48D9-886D-5A9DAE290FF9@cs.hmc.edu>
References: <B94FADCE-B655-49E3-B08D-51776FC35AB2@cs.hmc.edu>
	<20061207083510.GB52065@uiuc.edu>
	<A484A536-370B-4891-B728-14D0893F65EB@cs.hmc.edu>
	<07B00F4F-0ECE-48D9-886D-5A9DAE290FF9@cs.hmc.edu>
Message-ID: <20061207084816.GA52585@uiuc.edu>

On Thu, Dec 07, 2006 at 12:38:36AM -0800, belinda thom wrote:
> On Dec 7, 2006, at 12:36 AM, belinda thom wrote:
> 
> >Thanks Nicholas,
> >
> >So how do I use easy install :-)?

<http://peak.telecommunity.com/DevCenter/EasyInstall>

Of particular importance is the section on "custom installation
locations":

<http://peak.telecommunity.com/DevCenter/EasyInstall#custom-installation-locations>

Before starting you probably want to edit the ~/.pydistutils.cfg file
to specify a different location for installing scripts (for example,
/usr/local/bin).  If you just want to install the packages for
yourself, then check out the Mac OS X "User" installation section.

Then download and run ez_setup.py, which bootstraps setuptools, and
run "easy_install IPython".  This will work with many simple packages,
and some complex ones too.

-- 
Nicholas Riley <njriley at uiuc.edu> | <http://www.uiuc.edu/ph/www/njriley>

From ronaldoussoren at mac.com  Thu Dec  7 10:25:35 2006
From: ronaldoussoren at mac.com (Ronald Oussoren)
Date: Thu, 7 Dec 2006 10:25:35 +0100
Subject: [Pythonmac-SIG] pythonmac packages question
In-Reply-To: <20061207084816.GA52585@uiuc.edu>
References: <B94FADCE-B655-49E3-B08D-51776FC35AB2@cs.hmc.edu>
	<20061207083510.GB52065@uiuc.edu>
	<A484A536-370B-4891-B728-14D0893F65EB@cs.hmc.edu>
	<07B00F4F-0ECE-48D9-886D-5A9DAE290FF9@cs.hmc.edu>
	<20061207084816.GA52585@uiuc.edu>
Message-ID: <C4CB26F4-9D0F-44B7-9E90-BF607F56B11E@mac.com>


On Dec 7, 2006, at 9:48 AM, Nicholas Riley wrote:

> On Thu, Dec 07, 2006 at 12:38:36AM -0800, belinda thom wrote:
>> On Dec 7, 2006, at 12:36 AM, belinda thom wrote:
>>
>>> Thanks Nicholas,
>>>
>>> So how do I use easy install :-)?
>
> <http://peak.telecommunity.com/DevCenter/EasyInstall>
>
> Of particular importance is the section on "custom installation
> locations":
>
> <http://peak.telecommunity.com/DevCenter/EasyInstall#custom- 
> installation-locations>
>
> Before starting you probably want to edit the ~/.pydistutils.cfg file
> to specify a different location for installing scripts (for example,
> /usr/local/bin).

No you don't want that. Add /Library/Frameworks/Python.framework/ 
Versions/Current/bin to your shell's PATH, which the macpython  
installer should do automaticly.

The major advantages of that are that all python stuff is nicely self- 
contained in the python framework and it is possible to do side-by- 
side installations of multiple versions of python (such as Python 2.4  
and 2.5).

Ronald


From bthom at cs.hmc.edu  Thu Dec  7 11:21:31 2006
From: bthom at cs.hmc.edu (belinda thom)
Date: Thu, 07 Dec 2006 02:21:31 -0800
Subject: [Pythonmac-SIG] pythonmac packages question
In-Reply-To: <C4CB26F4-9D0F-44B7-9E90-BF607F56B11E@mac.com>
References: <B94FADCE-B655-49E3-B08D-51776FC35AB2@cs.hmc.edu>
	<20061207083510.GB52065@uiuc.edu>
	<A484A536-370B-4891-B728-14D0893F65EB@cs.hmc.edu>
	<07B00F4F-0ECE-48D9-886D-5A9DAE290FF9@cs.hmc.edu>
	<20061207084816.GA52585@uiuc.edu>
	<C4CB26F4-9D0F-44B7-9E90-BF607F56B11E@mac.com>
Message-ID: <B8FBA9C9-14F2-4511-8C9D-BA75D0499B26@cs.hmc.edu>

Thanks folks....the saga continues...

> No you don't want that. Add /Library/Frameworks/Python.framework/ 
> Versions/Current/bin to your shell's PATH, which the macpython  
> installer should do automaticly.
>
> The major advantages of that are that all python stuff is nicely  
> self-contained in the python framework and it is possible to do  
> side-by-side installations of multiple versions of python (such as  
> Python 2.4 and 2.5).
>
> Ronald

I've gotten easy-install running, which allowed me to download a  
brand-spanking new version of ipython (0.7.2). Recall I've also  
installed (via http://pythonmac.org/packages/py24-fat/index.html)  
matplotlib, Numeric, and wxPython (the latest version: 2.7.1.3; there  
was also a 2.6.3.3 version).

I run my new ipython with the -pylab option (using WXAgg backend and  
interactive True) and still get the following warning:
---------
/Library/Frameworks/Python.framework/Versions/2.4/lib/python2.4/site- 
packages/ipython-0.7.2-py2.4.egg/IPython/Shell.py:709:  
DeprecationWarning: The wxPython compatibility package is no longer  
automatically generated or activly maintained.  Please switch to the  
wx package as soon as possible.
   import wxPython.wx as wx
----------

followed by the same ability to not be able to use matplotlib  
commands, e.g.

     from pylab import *
     plot([1,2,3])

gives the same Memory / wx error:
-------
In [3]:
exceptions.MemoryError                               Traceback (most  
recent call last)
/Library/Frameworks/Python.framework/Versions/2.4/lib/python2.4/site- 
packages/matplotlib/backends/backend_wx.py in _onPaint(self, evt)
    1046             self.realize()
    1047         # Render to the bitmap
-> 1048         self.draw(repaint=False)
    1049         # Update the display using a PaintDC
    1050         self.gui_repaint(drawDC=wx.PaintDC(self))
/Library/Frameworks/Python.framework/Versions/2.4/lib/python2.4/site- 
packages/matplotlib/backends/backend_wxagg.py in draw(self, repaint)
      61         FigureCanvasAgg.draw(self)
      62
---> 63         self.bitmap = _convert_agg_to_wx_bitmap 
(self.get_renderer(), None)
      64         if repaint:
      65             self.gui_repaint()
MemoryError: _wxagg.convert_agg_to_wx_bitmap(): could not create the  
wx.Bitmap
--------

I also tried using the TkAgg backend, which produces very different  
complaints, e.g:
----------------
In [16]: rcParamsDefault['backend']
Out[16]: 'TkAgg'
<snip>
In [18]: plot([1,2,3])
<snip>
exceptions.TypeError                                 Traceback (most  
recent call last)
<snip>
/Library/Frameworks/Python.framework/Versions/2.4/lib/python2.4/site- 
packages/matplotlib/axes.py in set_xlim(self, xmin, xmax, emit)
    1225             xmax += 1e-38
    1226
-> 1227         self.viewLim.intervalx().set_bounds(xmin, xmax)
    1228         if emit: self._send_xlim_event()
    1229         return xmin, xmax
<snip>
TypeError: only length-1 arrays can be converted to Python scalars.
-----------------

Have any of you been able to use matplotlib w/ipython in the MacPorts  
setting? What am I doing incorrectly?

Thanks again,

--b

  

From kw at codebykevin.com  Thu Dec  7 13:50:27 2006
From: kw at codebykevin.com (Kevin Walzer)
Date: Thu, 07 Dec 2006 07:50:27 -0500
Subject: [Pythonmac-SIG] pythonmac packages question
In-Reply-To: <75A0405E-EE3E-4E6F-8A65-B48025011AD3@mac.com>
References: <B94FADCE-B655-49E3-B08D-51776FC35AB2@cs.hmc.edu>
	<75A0405E-EE3E-4E6F-8A65-B48025011AD3@mac.com>
Message-ID: <45780E13.6070900@codebykevin.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


> tkinter is part of the standard installation for macpython.
> 


What does this mean? Does MacPython 2.5 now install a universal build of
the Tcl/Tk frameworks in /Library/Frameworks?

- --
Kevin Walzer
Code by Kevin
http://www.codebykevin.com
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFeA4TEsLm8HXyq4sRAr6AAJsErqcP7BcUEYlH3LpzBtnj/rXktACgh2dp
m17UbFdUdXOk7vfeR5PHh0w=
=SRIb
-----END PGP SIGNATURE-----

From ronaldoussoren at mac.com  Thu Dec  7 14:24:11 2006
From: ronaldoussoren at mac.com (Ronald Oussoren)
Date: Thu, 7 Dec 2006 14:24:11 +0100
Subject: [Pythonmac-SIG] pythonmac packages question
In-Reply-To: <45780E13.6070900@codebykevin.com>
References: <B94FADCE-B655-49E3-B08D-51776FC35AB2@cs.hmc.edu>
	<75A0405E-EE3E-4E6F-8A65-B48025011AD3@mac.com>
	<45780E13.6070900@codebykevin.com>
Message-ID: <033E2123-B805-4CBC-9C98-9A1975A10877@mac.com>


On Dec 7, 2006, at 1:50 PM, Kevin Walzer wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
>
>> tkinter is part of the standard installation for macpython.
>>
>
>
> What does this mean? Does MacPython 2.5 now install a universal  
> build of
> the Tcl/Tk frameworks in /Library/Frameworks?

No it means that MacPython ships with _tkinter.so. If you're running  
on a 10.3 system you still have to install Tcl/Tk by hand.

Either way you're hosed because the Aqua version of Tk sucks.

Ronald


From rowen at cesmail.net  Thu Dec  7 18:40:14 2006
From: rowen at cesmail.net (Russell E. Owen)
Date: Thu, 07 Dec 2006 09:40:14 -0800
Subject: [Pythonmac-SIG] pythonmac packages question
References: <B94FADCE-B655-49E3-B08D-51776FC35AB2@cs.hmc.edu>
Message-ID: <rowen-30BBA4.09401307122006@sea.gmane.org>

In article <B94FADCE-B655-49E3-B08D-51776FC35AB2 at cs.hmc.edu>,
 belinda thom <bthom at cs.hmc.edu> wrote:

> Hi,
> 
> As of my prior post, I was able to get into pythonmac.org.
> 
> There I found many of the packages it seems I'll need. I was able to  
> install Numeric and WxPython (I chose the later dmg, v 2.7.1.3) and  
> matplotlib.
> 
> This site has no Tkinter and no IPython packages. Where do MacPython  
> people go for such things?

Tkinter is already included with MacPython. It will use the built-in 
(crummy) Aqua Tcl/Tk or you can install a newer better version.

ActiveState recently released 8.4.14 as a universal binary (the first 
universal binary version), which is handy if you want to redistribute 
your code. So far it seems pretty robust, though it does have a few 
known bugs.

-- Russell


From lou_boog2000 at yahoo.com  Fri Dec  8 16:33:47 2006
From: lou_boog2000 at yahoo.com (Lou Pecora)
Date: Fri, 8 Dec 2006 07:33:47 -0800 (PST)
Subject: [Pythonmac-SIG] [SciPy-user] C-extensions for NumPy code available.
Message-ID: <364838.73688.qm@web34407.mail.mud.yahoo.com>

For those of you who do numerical calculations, need
to speed up your code, and use NumPy arrays:

I've written several C extensions that handle NumPy
arrays.  They are simple, but they seem to work well.
They will show you how to pass Python variables and
NumPy arrays to your C code. Once you learn how to do
it, it's pretty straight-forward. I suspect they will
suffice for most numerical code.  I've written it up
as a draft and have made the code and document file
available.  If you want a copy, just email me or post
a followup here to this message.  I will send it to
you as a tar.gzip file (only 192 KB).

Remember it is only a draft (no guarantees or
warantees) and you should test the code for your own
uses.  I would really love people who are much more
knowledgable about these things than I to look it over
and set me straight on any mistakes (like not
INREF'ing something).

In addition, is there a place I could put this up on
the web for others?  I don't have any way to do that
at my lab.  I would like to share with the Python
community since I've been helped so much on these
mailing lists.

By the way I did this after a long time of searching
for ways to speed up Python and connect with my C and
C++ code.  There are a lot of solutions out there, but
I decided that they all provided so much more than I
needed and required so much more learning time that it
wasn't worth it. I am not knocking them. I found for
my numerical needs I really only need to pass a
limited set of things (integers, floats, strings, and
NumPy arrays).  If that's your category, this code
might help you.

-- Lou Pecora
    Naval Research Lab
    Washington, DC


 
____________________________________________________________________________________
Need a quick answer? Get one in minutes from people who know.
Ask your question on www.Answers.yahoo.com

From lou_boog2000 at yahoo.com  Fri Dec  8 17:44:00 2006
From: lou_boog2000 at yahoo.com (Lou Pecora)
Date: Fri, 8 Dec 2006 08:44:00 -0800 (PST)
Subject: [Pythonmac-SIG] [SciPy-user] C-extensions for NumPy code
	available.
In-Reply-To: <45798E9A.9050600@gmail.com>
Message-ID: <195291.33451.qm@web34415.mail.mud.yahoo.com>

This is a good idea, but I'm afraid I have no idea how
to even start this process.  I've never done anything
on a Wiki.  Is there some tutorial for newbies like
me?  Maybe I missed a link to one there. 

Thanks.

-- Lou Pecora

--- Robert Kern <robert.kern at gmail.com> wrote:

> Temporarily, you can write up a wiki page about it
> on www.scipy.org and attach
> the tarball to the page. Once we can look at the
> contents a bit more, we might
> have better ideas where they can go. Possibly, they
> can go into the numpy
> distribution as examples.
> 
> -- 
> Robert Kern



 
____________________________________________________________________________________
Any questions? Get answers on any topic at www.Answers.yahoo.com.  Try it now.

From Larry.A.Meyn at nasa.gov  Fri Dec  8 18:49:30 2006
From: Larry.A.Meyn at nasa.gov (Larry Meyn)
Date: Fri, 8 Dec 2006 09:49:30 -0800
Subject: [Pythonmac-SIG] [SciPy-user] C-extensions for NumPy code
	available.
In-Reply-To: <195291.33451.qm@web34415.mail.mud.yahoo.com>
References: <195291.33451.qm@web34415.mail.mud.yahoo.com>
Message-ID: <CC31A22D-F62A-451A-A185-B3FFE0A4FD81@nasa.gov>

Lou,

I've never used the SciPy wiki, but it's similar to the one I use.  
Hopefully this will give you a start.

First you need an account.  For that go to http://www.scipy.org/ 
UserPreferences and enter your information.

To edit pages click on the little balloon icon in the upper right.   
Help on editing can be found at
http://www.scipy.org/HelpOnEditing

Good luck  and thanks for donating you code.

Larry

On Dec 8, 2006, at 8:44 AM, Lou Pecora wrote:

> This is a good idea, but I'm afraid I have no idea how
> to even start this process.  I've never done anything
> on a Wiki.  Is there some tutorial for newbies like
> me?  Maybe I missed a link to one there.
>
> Thanks.
>
> -- Lou Pecora
>
> --- Robert Kern <robert.kern at gmail.com> wrote:
>
>> Temporarily, you can write up a wiki page about it
>> on www.scipy.org and attach
>> the tarball to the page. Once we can look at the
>> contents a bit more, we might
>> have better ideas where they can go. Possibly, they
>> can go into the numpy
>> distribution as examples.
>>
>> -- 
>> Robert Kern
>
>
>
>
> ______________________________________________________________________ 
> ______________
> Any questions? Get answers on any topic at www.Answers.yahoo.com.   
> Try it now.
> _______________________________________________
> Pythonmac-SIG maillist  -  Pythonmac-SIG at python.org
> http://mail.python.org/mailman/listinfo/pythonmac-sig

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/pythonmac-sig/attachments/20061208/3c6080fc/attachment.htm 

From bthom at cs.hmc.edu  Sun Dec 10 02:37:39 2006
From: bthom at cs.hmc.edu (belinda thom)
Date: Sat, 09 Dec 2006 17:37:39 -0800
Subject: [Pythonmac-SIG] eggs and extraction
Message-ID: <E23933F1-923B-4256-898E-D97AB0356E1C@cs.hmc.edu>

Hi,

I've installed matplotlib from http://pythonmac.org/packages/py24- 
fat/. Things are running finally. [Details: I ran into a problem  
where backend TkAgg and numerix Numeric didn't get along; to the best  
of my knowledge this problem is undocumented; my stuff worked w/TkAgg  
only after I changed to numerix numpy; I couldn't get wxPython w/ 
matplotlib and WXAgg to work at all].

Now I can't figure out how to access the documentation. Part of the  
problem is that I'm pretty confused about what an egg is (I've  
scanned easy_install and distutil related stuff, but learned little;  
I'm not even sure where to look; Google often brings up content that  
deals w/chicken eggs :-)

In particular, I wanted to look at matplotlib's two_scales.py. In  
site-packages, I've got a matplotlib-0.87.7-py2.4.egg-info/ 
SOURCES.txt, which contains the line:

   examples/two_scales.py

I had thought perhaps:

   easy_install -editable matplotlib-0.87.7-py2.4.egg-info/

would do the trick, but it complains:

   error: No urls, filenames, or requirements specified (see --help)

What do I have to do to install all the examples onto my machine?

Thanks,

--b





From robert.kern at gmail.com  Sun Dec 10 02:54:37 2006
From: robert.kern at gmail.com (Robert Kern)
Date: Sat, 09 Dec 2006 19:54:37 -0600
Subject: [Pythonmac-SIG] eggs and extraction
In-Reply-To: <E23933F1-923B-4256-898E-D97AB0356E1C@cs.hmc.edu>
References: <E23933F1-923B-4256-898E-D97AB0356E1C@cs.hmc.edu>
Message-ID: <elfpdn$1fu$1@sea.gmane.org>

belinda thom wrote:
> Now I can't figure out how to access the documentation. Part of the  
> problem is that I'm pretty confused about what an egg is (I've  
> scanned easy_install and distutil related stuff, but learned little;  
> I'm not even sure where to look; Google often brings up content that  
> deals w/chicken eggs :-)

The documentation and examples were not included in the binary distribution that
you installed. You can get them from the source distribution.

  http://sourceforge.net/project/showfiles.php?group_id=80706

The first 7 or so hits on a Google search for "python eggs" are all on-topic and
tell you what you want to know about eggs (although they won't tell you where to
get matplotlib documentation and examples).

-- 
Robert Kern

"I have come to believe that the whole world is an enigma, a harmless enigma
 that is made terrible by our own mad attempt to interpret it as though it had
 an underlying truth."
  -- Umberto Eco


From robert.kern at gmail.com  Sun Dec 10 02:56:07 2006
From: robert.kern at gmail.com (Robert Kern)
Date: Sat, 09 Dec 2006 19:56:07 -0600
Subject: [Pythonmac-SIG] eggs and extraction
In-Reply-To: <E23933F1-923B-4256-898E-D97AB0356E1C@cs.hmc.edu>
References: <E23933F1-923B-4256-898E-D97AB0356E1C@cs.hmc.edu>
Message-ID: <elfpgg$1fu$2@sea.gmane.org>

belinda thom wrote:
> I had thought perhaps:
> 
>    easy_install -editable matplotlib-0.87.7-py2.4.egg-info/
> 
> would do the trick, but it complains:
> 
>    error: No urls, filenames, or requirements specified (see --help)

And the appropriate form for this command is the following:


[src]$ easy_install --editable -b . matplotlib
Searching for matplotlib
Reading http://www.python.org/pypi/matplotlib/
Reading http://matplotlib.sourceforge.net
Reading
http://sourceforge.net/project/showfiles.php?group_id=80706&amp;package_id=82474
Reading http://www.python.org/pypi/matplotlib/0.87.7
Best match: matplotlib 0.87.7
Downloading
http://downloads.sourceforge.net/matplotlib/matplotlib-0.87.7.tar.gz?modtime=1161866270&amp;big_mirror=0
Processing matplotlib-0.87.7.tar.gz

Extracted editable version of matplotlib to ./matplotlib

If it uses setuptools in its setup script, you can activate it in
"development" mode by going to that directory and running::


/Library/Frameworks/Python.framework/Versions/2.5/Resources/Python.app/Contents/MacOS/Python
setup.py develop

See the setuptools documentation for the "develop" command for more info.


-- 
Robert Kern

"I have come to believe that the whole world is an enigma, a harmless enigma
 that is made terrible by our own mad attempt to interpret it as though it had
 an underlying truth."
  -- Umberto Eco


From niko at alum.mit.edu  Sun Dec 10 08:25:48 2006
From: niko at alum.mit.edu (Niko Matsakis)
Date: Sun, 10 Dec 2006 08:25:48 +0100
Subject: [Pythonmac-SIG] py2app plugin problem: "Error loading code"
Message-ID: <0FE46EBB-032F-4157-BD8D-318EDB94A47F@alum.mit.edu>

All of a sudden, for reasons I completely fail to understand, py2app- 
created plug-ins will no longer load in my application.  I am not  
100% sure if this is the right list for this kind of question: feel  
free to set me straight.

The error I get is as follows:

2006-12-09 17:10:59.988 drachma[25673] *** -[NSBundle load]: Error  
loading code /Users/niko/hg/proj/drachma/src/build/Debug/drachma.app/ 
Contents/Resources/PyPlugin.plugin/Contents/MacOS/PyPlugin for  
bundle /Users/niko/hg/proj/drachma/src/build/Debug/drachma.app/ 
Contents/Resources/PyPlugin.plugin, error code 2 (link edit error  
code 0, error number 0 ())

The only slightly strange thing I noticed is that when I use "file"  
on the content of the plugin I get:

/Users/niko/hg/proj/drachma/src/build/Debug/drachma.app/Contents/ 
Resources/PyPlugin.plugin/Contents/MacOS/PyPlugin: Mach-O bundle i386

Which says "i386", but this is running on a PPC.  I tried to use the  
"--prefer-ppc" option, but I get the same result either way.

Any ideas on what's wrong, or where to find more information?  It  
doesn't seem to matter whether I use -A mode or not.

I am using a version of Python 2.4 from DarwinPorts; PyObjC was  
installed the same way.

What's very odd is that all of this USED to work --- I had to do a  
bit of a hack to make the -A mode work, introducing a .pth file into  
~/Library/Python/2.4/site-packages, but now nothing works, whether or  
not I add or remove that file.  The only thing I can think of that I  
did  in between is to attempt to build a "Release" version of the  
program instead of a "Debug" version, but now neither Release nor  
Debug works.


thanks in advance for any clues,

Niko


From robert.kern at gmail.com  Fri Dec  8 17:11:06 2006
From: robert.kern at gmail.com (Robert Kern)
Date: Fri, 08 Dec 2006 10:11:06 -0600
Subject: [Pythonmac-SIG] [SciPy-user] C-extensions for NumPy code
	available.
In-Reply-To: <364838.73688.qm@web34407.mail.mud.yahoo.com>
References: <364838.73688.qm@web34407.mail.mud.yahoo.com>
Message-ID: <45798E9A.9050600@gmail.com>

Lou Pecora wrote:

> In addition, is there a place I could put this up on
> the web for others?  I don't have any way to do that
> at my lab.  I would like to share with the Python
> community since I've been helped so much on these
> mailing lists.

Temporarily, you can write up a wiki page about it on www.scipy.org and attach
the tarball to the page. Once we can look at the contents a bit more, we might
have better ideas where they can go. Possibly, they can go into the numpy
distribution as examples.

-- 
Robert Kern

"I have come to believe that the whole world is an enigma, a harmless enigma
 that is made terrible by our own mad attempt to interpret it as though it had
 an underlying truth."
  -- Umberto Eco

From cmiller at securityevaluators.com  Sat Dec  9 03:30:33 2006
From: cmiller at securityevaluators.com (Charlie Miller)
Date: Fri, 08 Dec 2006 20:30:33 -0600
Subject: [Pythonmac-SIG] Python on mac - can't extend interpreter with
 "catch_exception_raise" wrapper.
Message-ID: <457A1FC9.9040803@securityevaluators.com>

Hi.  I've run into a very strange problem with Python on the mac.  I'm
trying to write a python script which will monitor an application for
exceptions.  It turns out on a mac, to do this you want to use the Mach
IPC interface, see:

http://web.mit.edu/darwin/src/modules/xnu/osfmk/man/
http://www.wodeveloper.com/omniLists/macosx-dev/2000/June/msg00137.html

Once set up, you call the mach function exc_server which then calls
"catch_raise_exception" which you provide.  This works great in C, so I
thought I'd wrap it for python.  But, it turns out the python
interpreter already has a symbol named "catch_raise_exception", (in my
case at 0x2030 under gdb).  The problem is when exc_server calls
catch_raise_exception it calls the one from the python binary and not
the one I provided and dies.  Does anyone have any ideas on how to deal
with this perplexing problem?!?

Thanks!!!

Charlie

From niko at alum.mit.edu  Sun Dec 10 22:19:15 2006
From: niko at alum.mit.edu (Niko Matsakis)
Date: Sun, 10 Dec 2006 22:19:15 +0100
Subject: [Pythonmac-SIG] py2app plugin problem: "Error loading code"
In-Reply-To: <0FE46EBB-032F-4157-BD8D-318EDB94A47F@alum.mit.edu>
References: <0FE46EBB-032F-4157-BD8D-318EDB94A47F@alum.mit.edu>
Message-ID: <7C1340EA-9112-4728-A693-E8BD9296574A@alum.mit.edu>

As a follow up, I decided to see if the problem was related to my  
project in any way by followiing the tutorial at http:// 
pyobjc.sourceforge.net/doc/extending_objc_with_python.php.  I still  
get the same error, so it must be something in my machine.

I can't seem to find any kind of reference to what "error code 2"  
might mean....


Niko

On Dec 10, 2006, at 8:25 AM, Niko Matsakis wrote:

> All of a sudden, for reasons I completely fail to understand, py2app-
> created plug-ins will no longer load in my application.  I am not
> 100% sure if this is the right list for this kind of question: feel
> free to set me straight.
>
> The error I get is as follows:
>
> 2006-12-09 17:10:59.988 drachma[25673] *** -[NSBundle load]: Error
> loading code /Users/niko/hg/proj/drachma/src/build/Debug/drachma.app/
> Contents/Resources/PyPlugin.plugin/Contents/MacOS/PyPlugin for
> bundle /Users/niko/hg/proj/drachma/src/build/Debug/drachma.app/
> Contents/Resources/PyPlugin.plugin, error code 2 (link edit error
> code 0, error number 0 ())
>
> The only slightly strange thing I noticed is that when I use "file"
> on the content of the plugin I get:
>
> /Users/niko/hg/proj/drachma/src/build/Debug/drachma.app/Contents/
> Resources/PyPlugin.plugin/Contents/MacOS/PyPlugin: Mach-O bundle i386
>
> Which says "i386", but this is running on a PPC.  I tried to use the
> "--prefer-ppc" option, but I get the same result either way.
>
> Any ideas on what's wrong, or where to find more information?  It
> doesn't seem to matter whether I use -A mode or not.
>
> I am using a version of Python 2.4 from DarwinPorts; PyObjC was
> installed the same way.
>
> What's very odd is that all of this USED to work --- I had to do a
> bit of a hack to make the -A mode work, introducing a .pth file into
> ~/Library/Python/2.4/site-packages, but now nothing works, whether or
> not I add or remove that file.  The only thing I can think of that I
> did  in between is to attempt to build a "Release" version of the
> program instead of a "Debug" version, but now neither Release nor
> Debug works.
>
>
> thanks in advance for any clues,
>
> Niko
>
> _______________________________________________
> Pythonmac-SIG maillist  -  Pythonmac-SIG at python.org
> http://mail.python.org/mailman/listinfo/pythonmac-sig


From ronaldoussoren at mac.com  Sun Dec 10 22:37:21 2006
From: ronaldoussoren at mac.com (Ronald Oussoren)
Date: Sun, 10 Dec 2006 22:37:21 +0100
Subject: [Pythonmac-SIG] py2app plugin problem: "Error loading code"
In-Reply-To: <7C1340EA-9112-4728-A693-E8BD9296574A@alum.mit.edu>
References: <0FE46EBB-032F-4157-BD8D-318EDB94A47F@alum.mit.edu>
	<7C1340EA-9112-4728-A693-E8BD9296574A@alum.mit.edu>
Message-ID: <50193F5C-BDEF-4C33-BF27-E553CD76304C@mac.com>


On Dec 10, 2006, at 10:19 PM, Niko Matsakis wrote:

> As a follow up, I decided to see if the problem was related to my
> project in any way by followiing the tutorial at http://
> pyobjc.sourceforge.net/doc/extending_objc_with_python.php.  I still
> get the same error, so it must be something in my machine.
>
> I can't seem to find any kind of reference to what "error code 2"
> might mean....

The problem is that py2app/bundletemplate/prebuild/main in the py2app  
distribution is an intel-only binary (at least in the current trunk)  
instead of a universal binary.

Removing the file and running 'setup.py' in the bundletemplate  
directory should rebuild it with the proper architecture for your  
machine.

Ronald
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3562 bytes
Desc: not available
Url : http://mail.python.org/pipermail/pythonmac-sig/attachments/20061210/ec3240ce/attachment.bin 

From niko at alum.mit.edu  Sun Dec 10 23:38:21 2006
From: niko at alum.mit.edu (Niko Matsakis)
Date: Sun, 10 Dec 2006 23:38:21 +0100
Subject: [Pythonmac-SIG] py2app plugin problem: "Error loading code"
In-Reply-To: <50193F5C-BDEF-4C33-BF27-E553CD76304C@mac.com>
References: <0FE46EBB-032F-4157-BD8D-318EDB94A47F@alum.mit.edu>
	<7C1340EA-9112-4728-A693-E8BD9296574A@alum.mit.edu>
	<50193F5C-BDEF-4C33-BF27-E553CD76304C@mac.com>
Message-ID: <01ED529F-AF16-464F-8B8B-A5014195EEF5@alum.mit.edu>

>
> The problem is that py2app/bundletemplate/prebuild/main in the  
> py2app distribution is an intel-only binary (at least in the  
> current trunk) instead of a universal binary.

Okay, I must say, I would *never* have guessed that!  I wonder how  
that file changed?  As I said, this used to work... perhaps  
DarwinPorts took it upon itself to upgrade py2app in some way?  In  
any case, after doing your setup.py trick, everything seemed better.   
Thanks a lot!

Now, next question:

to get python imports to work and the python bundle to start, I found  
I have to add a symbolic link from "Contents/Resources/Python" to  
"Contents/Resources/lib/python2.4".  While I can do this if  
necessary, do you have any idea why things don't just end up in the  
right places on their own?

Is DarwinPorts not the best way to go about using things, anyhow?  It  
seems that most things don't really work too smoothly.  If so, what  
is the recommended configuration?


Niko

From bthom at cs.hmc.edu  Mon Dec 11 07:09:16 2006
From: bthom at cs.hmc.edu (belinda thom)
Date: Sun, 10 Dec 2006 22:09:16 -0800
Subject: [Pythonmac-SIG] permissions on packages
Message-ID: <21EC38A2-D0AB-4A66-B5E3-79075BAE2DC2@cs.hmc.edu>

Hi,

I'm writing to seek a better understanding of how permissions should  
work in python on the Mac.

Out-of-the-MacPython-2.4-install box, files in /Library/Frameworks/ 
Python.framework/Versions/2.4/lib/python2.4 have permissions like these:

-rw-rw-r--     1 root  admin   26255 Oct 18 01:39 zipfile.py
-rw-rw-r--     1 root  admin   18931 Dec  6 22:14 zipfile.pyc
-rw-rw-r--     1 root  admin   18931 Dec  6 22:14 zipfile.pyo

In the site-packages directory, packages I installed (via  
macpython.com, e.g. numpy, Numeric, matplotlib, etc) have permissions  
like:

drwxrwxr-x    41 501    admin   1394 Jul  5 20:11 Numeric/
drwxrwxr-x   157 501    admin   5338 Dec  9 20:57 matplotlib/
drwxr-xr-x   130 root   admin   4420 Dec  7 04:39 numarray/
drwxrwxr-x    47 501    admin   1598 Oct 27 16:54 numpy/
drwxrwxr-x     6 501    admin    204 Oct 27 16:54 numpy-1.0-py2.4.egg- 
info/
-rw-rw-r--     1 501    admin     31 Oct 26 06:26 pylab.py
-rw-rw-r--     1 501    admin    205 Oct 30 10:15 pylab.pyc
-rw-rw-r--     1 501    admin    205 Oct 30 10:15 pylab.pyo
drwxrwxr-x    43 501    admin   1462 Sep 17 18:04 scipy/
drwxrwxr-x     6 501    admin    204 Sep 17 18:04 scipy-0.5.1- 
py2.4.egg-info/
drwxr-xr-x    10 bthom  admin    340 Dec  7 01:33 setuptools-0.6c3- 
py2.4.egg/
-rw-r--r--     1 bthom  admin     29 Dec  7 01:33 setuptools.pth
drwxrwxr-x     4 root   admin    136 Jul 15 16:03 wx-2.6-mac-unicode/
-rw-r--r--     1 root   admin     18 Jul 15 16:03 wx.pth
drwxr-xr-x     4 root   admin    136 Jul 15 16:03 wxaddons/
-rw-r--r--     1 root   admin  17813 Jan 20  2006 wxversion.py

Is there a reason why these packages have owner 501? And is there a  
reason why group have read access? I'd have thought the best way to  
install something that are supposed to work as-is would be to have  
root as owner, giving sole write permissions to owner. (Then to  
modify something, you'd have to sudo).

I'm beginning to realize its not that simple. For instance, the first  
time I (w/admin privs) try and edit a file (e.g. zipfile.py),  
TextEdit asks if I'd like to overwrite the permissions on save.  
Saying yes overwrites the file. The disturbing thing is that on  
future opens (even after TextEdit's been closed) I am no longer  
asked; overwriting saves occur automatically. Fortunately, when I  
created another user w/o admin permissions, overwriting when logged  
in as this less priv'd user didn't seem to be an option.

...but I'd like to protect against myself (this is esp. true b/c I  
use ipython w/in xemacs w/%pdb on, so I'm often dumped into the  
editor w/a file like bpd.py).

Some more poking in a package directory (e.g. matplotlib) shows  
permissions like:

-rw-rw-r--     1 root   admin    40789 Oct 26 09:30 __init__.py
-rw-r--r--     1 bthom  admin    39389 Dec 10 21:02 __init__.pyc
-rw-rw-r--     1 root   admin    30134 Oct 30 10:15 __init__.pyo

What's interesting here is that pyc is owned by me (perhaps b/c it  
was compiled via my first call to __init__.py). However, I'm pretty  
sure what I installed off macpython.com was binarys---so there should  
be no need to compile the files, right? Again, I (perhaps naively)  
would feel better if all of these files were owned by root (and that  
only root could override them).

Security issues become even more important as students will be using  
python/ipython from the same machine.

Perhaps I'm missing something really basic (would a Finder view to  
permissions provide better control?). Or perhaps the ability to not  
protect as tightly against myself as admin as I'd have liked is a  
"feature".

What have others done wrt this issue?

Thanks,
--b

From bob at redivi.com  Mon Dec 11 07:21:17 2006
From: bob at redivi.com (Bob Ippolito)
Date: Mon, 11 Dec 2006 14:21:17 +0800
Subject: [Pythonmac-SIG] permissions on packages
In-Reply-To: <21EC38A2-D0AB-4A66-B5E3-79075BAE2DC2@cs.hmc.edu>
References: <21EC38A2-D0AB-4A66-B5E3-79075BAE2DC2@cs.hmc.edu>
Message-ID: <6a36e7290612102221x49533c17jc8a648fcddd23870@mail.gmail.com>

On 12/11/06, belinda thom <bthom at cs.hmc.edu> wrote:
> Hi,
>
> I'm writing to seek a better understanding of how permissions should
> work in python on the Mac.
>
> Out-of-the-MacPython-2.4-install box, files in /Library/Frameworks/
> Python.framework/Versions/2.4/lib/python2.4 have permissions like these:
>
> -rw-rw-r--     1 root  admin   26255 Oct 18 01:39 zipfile.py
> -rw-rw-r--     1 root  admin   18931 Dec  6 22:14 zipfile.pyc
> -rw-rw-r--     1 root  admin   18931 Dec  6 22:14 zipfile.pyo
>
> In the site-packages directory, packages I installed (via
> macpython.com, e.g. numpy, Numeric, matplotlib, etc) have permissions
> like:
>
> drwxrwxr-x    41 501    admin   1394 Jul  5 20:11 Numeric/
> drwxrwxr-x   157 501    admin   5338 Dec  9 20:57 matplotlib/
> drwxr-xr-x   130 root   admin   4420 Dec  7 04:39 numarray/
> drwxrwxr-x    47 501    admin   1598 Oct 27 16:54 numpy/
> drwxrwxr-x     6 501    admin    204 Oct 27 16:54 numpy-1.0-py2.4.egg-
> info/
> -rw-rw-r--     1 501    admin     31 Oct 26 06:26 pylab.py
> -rw-rw-r--     1 501    admin    205 Oct 30 10:15 pylab.pyc
> -rw-rw-r--     1 501    admin    205 Oct 30 10:15 pylab.pyo
> drwxrwxr-x    43 501    admin   1462 Sep 17 18:04 scipy/
> drwxrwxr-x     6 501    admin    204 Sep 17 18:04 scipy-0.5.1-
> py2.4.egg-info/
> drwxr-xr-x    10 bthom  admin    340 Dec  7 01:33 setuptools-0.6c3-
> py2.4.egg/
> -rw-r--r--     1 bthom  admin     29 Dec  7 01:33 setuptools.pth
> drwxrwxr-x     4 root   admin    136 Jul 15 16:03 wx-2.6-mac-unicode/
> -rw-r--r--     1 root   admin     18 Jul 15 16:03 wx.pth
> drwxr-xr-x     4 root   admin    136 Jul 15 16:03 wxaddons/
> -rw-r--r--     1 root   admin  17813 Jan 20  2006 wxversion.py
>
> Is there a reason why these packages have owner 501? And is there a
> reason why group have read access? I'd have thought the best way to
> install something that are supposed to work as-is would be to have
> root as owner, giving sole write permissions to owner. (Then to
> modify something, you'd have to sudo).
>
> I'm beginning to realize its not that simple. For instance, the first
> time I (w/admin privs) try and edit a file (e.g. zipfile.py),
> TextEdit asks if I'd like to overwrite the permissions on save.
> Saying yes overwrites the file. The disturbing thing is that on
> future opens (even after TextEdit's been closed) I am no longer
> asked; overwriting saves occur automatically. Fortunately, when I
> created another user w/o admin permissions, overwriting when logged
> in as this less priv'd user didn't seem to be an option.
>
> ...but I'd like to protect against myself (this is esp. true b/c I
> use ipython w/in xemacs w/%pdb on, so I'm often dumped into the
> editor w/a file like bpd.py).
>
> Some more poking in a package directory (e.g. matplotlib) shows
> permissions like:
>
> -rw-rw-r--     1 root   admin    40789 Oct 26 09:30 __init__.py
> -rw-r--r--     1 bthom  admin    39389 Dec 10 21:02 __init__.pyc
> -rw-rw-r--     1 root   admin    30134 Oct 30 10:15 __init__.pyo
>
> What's interesting here is that pyc is owned by me (perhaps b/c it
> was compiled via my first call to __init__.py). However, I'm pretty
> sure what I installed off macpython.com was binarys---so there should
> be no need to compile the files, right? Again, I (perhaps naively)
> would feel better if all of these files were owned by root (and that
> only root could override them).
>
> Security issues become even more important as students will be using
> python/ipython from the same machine.
>
> Perhaps I'm missing something really basic (would a Finder view to
> permissions provide better control?). Or perhaps the ability to not
> protect as tightly against myself as admin as I'd have liked is a
> "feature".
>
> What have others done wrt this issue?

That's not really intentional. If you want to fix it, feel free. The
code that makes these packages is bdist_mpkg and the source is here:

http://svn.pythonmac.org/bdist_mpkg/bdist_mpkg/trunk/

I don't have the time or the need to fix it myself any time soon.

-bob

From Chris.Barker at noaa.gov  Mon Dec 11 19:29:33 2006
From: Chris.Barker at noaa.gov (Christopher Barker)
Date: Mon, 11 Dec 2006 10:29:33 -0800
Subject: [Pythonmac-SIG] eggs and extraction
In-Reply-To: <E23933F1-923B-4256-898E-D97AB0356E1C@cs.hmc.edu>
References: <E23933F1-923B-4256-898E-D97AB0356E1C@cs.hmc.edu>
Message-ID: <457DA38D.30506@noaa.gov>

belinda thom wrote:
> Details: I ran into a problem  
> where backend TkAgg and numerix Numeric didn't get along; to the best  
> of my knowledge this problem is undocumented;

I don't think I've heard of it -- odd.

 > my stuff worked w/TkAgg
> only after I changed to numerix numpy;

numpy is a better option anyway, unless you need Numeric for legacy code.

> I couldn't get wxPython w/ 
> matplotlib and WXAgg to work at all].

Darn. Unfortunately, MPL needs to be built with exactly the same version 
of wx that you run it with -- we've tried to keep the MPL and wx 
packages on pythonmac in sync, but they've gotten messed up. I hope 
someone will fix this, I'm not sure when I can, I still haven't set 
myself up with universal builds of the dependencies -- Charlie?

-Chris

-- 
Christopher Barker, Ph.D.
Oceanographer

Emergency Response Division
NOAA/NOS/OR&R            (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception

Chris.Barker at noaa.gov

From rowen at cesmail.net  Mon Dec 11 21:20:23 2006
From: rowen at cesmail.net (Russell E. Owen)
Date: Mon, 11 Dec 2006 12:20:23 -0800
Subject: [Pythonmac-SIG] cyptes problem on MacOS X
Message-ID: <rowen-C597F3.12202311122006@sea.gmane.org>

I've been trying out ctypes (from MacPython 2.5) on MacOS X on a PPC Mac 
and have run into a problem. I hope I'm just doing something stupid, but 
it seems to match the tutorial so I'm really puzzled:

import ctypes
libc = ctypes.CDLL("libc.dylib")
libc.printf("int=%d float=%f double=%f\n", ctypes.c_int(1), 
ctypes.c_float(2.0), ctypes.c_double(3.0));

results in:
int=1 float=-1.996124 double=0.000000

I realize the docs doesn't actually claim to handle dylib files, but 
I've seen at least some info on the web that claims it works. One site 
suggested the need for ctypes.RTLD_GLOBAL to CDLL but adding it made no 
difference.

On a related subject...assuming I can get this to work...does anyone 
have any idea how the speed of ctypes relates to programming a python C 
extension or using SWIG? This is mostly for use with NumPy and 
interfacing python to C/C++-based heavy duty astronomical image 
processing code.

Other suggestions are welcome. We'll mostly be interfacing to C++ so 
I've considered trying out Boost, but it hasn't had an update in years 
which worries me. Robustness, ease of use and speed are all major 
requirements. Linux and MacOS X are the main platforms.

-- Russell


From pedz at easesoftware.com  Mon Dec 11 21:39:24 2006
From: pedz at easesoftware.com (Perry Smith)
Date: Mon, 11 Dec 2006 14:39:24 -0600
Subject: [Pythonmac-SIG] cyptes problem on MacOS X
In-Reply-To: <rowen-C597F3.12202311122006@sea.gmane.org>
References: <rowen-C597F3.12202311122006@sea.gmane.org>
Message-ID: <18702D1C-0D2E-4B4B-9616-7F1CCD23BE30@easesoftware.com>

I don't use floats and double much.  But the printf man page implies  
that all
floats are actually passed as doubles.  I vaguely recall that to be  
the case.

But I doubt if the python thingy knows that.  Try not passing  
floats.  You might
also have problems trying to pass char's and short's (since they are  
promoted
to int's).

On Dec 11, 2006, at 2:20 PM, Russell E. Owen wrote:

> I've been trying out ctypes (from MacPython 2.5) on MacOS X on a  
> PPC Mac
> and have run into a problem. I hope I'm just doing something  
> stupid, but
> it seems to match the tutorial so I'm really puzzled:
>
> import ctypes
> libc = ctypes.CDLL("libc.dylib")
> libc.printf("int=%d float=%f double=%f\n", ctypes.c_int(1),
> ctypes.c_float(2.0), ctypes.c_double(3.0));
>
> results in:
> int=1 float=-1.996124 double=0.000000
>
> I realize the docs doesn't actually claim to handle dylib files, but
> I've seen at least some info on the web that claims it works. One site
> suggested the need for ctypes.RTLD_GLOBAL to CDLL but adding it  
> made no
> difference.
>
> On a related subject...assuming I can get this to work...does anyone
> have any idea how the speed of ctypes relates to programming a  
> python C
> extension or using SWIG? This is mostly for use with NumPy and
> interfacing python to C/C++-based heavy duty astronomical image
> processing code.
>
> Other suggestions are welcome. We'll mostly be interfacing to C++ so
> I've considered trying out Boost, but it hasn't had an update in years
> which worries me. Robustness, ease of use and speed are all major
> requirements. Linux and MacOS X are the main platforms.
>
> -- Russell
>
> _______________________________________________
> Pythonmac-SIG maillist  -  Pythonmac-SIG at python.org
> http://mail.python.org/mailman/listinfo/pythonmac-sig
>

Perry Smith ( pedz at easesoftware.com )
Ease Software, Inc. ( http://www.easesoftware.com )

Low cost SATA Disk Systems for IBMs p5, pSeries, and RS/6000 AIX systems


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/pythonmac-sig/attachments/20061211/273a9885/attachment.htm 

From rowen at cesmail.net  Mon Dec 11 22:17:29 2006
From: rowen at cesmail.net (Russell E Owen)
Date: Mon, 11 Dec 2006 13:17:29 -0800
Subject: [Pythonmac-SIG] cyptes problem on MacOS X
In-Reply-To: <18702D1C-0D2E-4B4B-9616-7F1CCD23BE30@easesoftware.com>
References: <rowen-C597F3.12202311122006@sea.gmane.org>
	<18702D1C-0D2E-4B4B-9616-7F1CCD23BE30@easesoftware.com>
Message-ID: <p06240802c1a379573a0d@[128.95.99.44]>

At 2:39 PM -0600 2006-12-11, Perry Smith wrote:
>I don't use floats and double much.  But the printf man page implies that all
>floats are actually passed as doubles.  I vaguely recall that to be the case.
>
>But I doubt if the python thingy knows that.  Try not passing 
>floats.  You might
>also have problems trying to pass char's and short's (since they are promoted
>to int's).

You're right. I should not have passed floats to printf. But the 
example also showed doubles being mishandled. Here is a more 
extensive example:

import ctypes as ct
libc = ct.CDLL("libc.dylib")
libc.printf("int=%d  double=%f\n", ct.c_int(1), ct.c_double(3.0))
libc.printf("int=%d  double=%f  double=%f\n", ct.c_int(1), 
ct.c_double(3.0), ct.c_double(3.0))

On MacOS X this prints:
int=1  double=-1.996155
int=1  double=-1.996140  double=0.000000

On a linux box, with a suitably modified ct.CDLL statement, this 
produces the more reasonable:
int=1  double=3.000000
int=1  double=3.000000  double=3.000000

I am totally mystified by the MacOS X results.

-- Russell

From pedz at easesoftware.com  Mon Dec 11 22:36:28 2006
From: pedz at easesoftware.com (Perry Smith)
Date: Mon, 11 Dec 2006 15:36:28 -0600
Subject: [Pythonmac-SIG] cyptes problem on MacOS X
In-Reply-To: <p06240802c1a379573a0d@[128.95.99.44]>
References: <rowen-C597F3.12202311122006@sea.gmane.org>
	<18702D1C-0D2E-4B4B-9616-7F1CCD23BE30@easesoftware.com>
	<p06240802c1a379573a0d@[128.95.99.44]>
Message-ID: <E99F78B0-6094-4692-A520-6BC7F52122E4@easesoftware.com>

Does this set up (python calling C code) have a way to introduce a C  
function prototype?

I *think* that on the PPC, to call a function that has a variable  
number of arguments of
mixed types, the compiler HAS to see the prototype so it knows to  
push the arguments on
the stack and not pass them in registers.  I'm getting old and I  
can't remember how the PPC
guys solved this problem.  The MIPs guys blew it.

I think, the best I can recall, a function like:

void f(double d);

passes d in floating point register 0.  A function like:

void g(int i);

passes i in general purpose register 3.  (The next argument takes  
register 4, etc).

But when the compile sees:

int printf(char *f, ...);

It says "oh I gotta pass everything on the stack".

Try this: right a tiny C program.  Do not include stdio.h.  Do not  
include anything.

Just call printf like you are calling it and see what happens.  I bet  
it doesn't work.

Then include stdio.h, and I bet it will magically start working.


On Dec 11, 2006, at 3:17 PM, Russell E Owen wrote:

> At 2:39 PM -0600 2006-12-11, Perry Smith wrote:
>> I don't use floats and double much.  But the printf man page  
>> implies that all
>> floats are actually passed as doubles.  I vaguely recall that to  
>> be the case.
>>
>> But I doubt if the python thingy knows that.  Try not passing  
>> floats.  You might
>> also have problems trying to pass char's and short's (since they  
>> are promoted
>> to int's).
>
> You're right. I should not have passed floats to printf. But the  
> example also showed doubles being mishandled. Here is a more  
> extensive example:
>
> import ctypes as ct
> libc = ct.CDLL("libc.dylib")
> libc.printf("int=%d  double=%f\n", ct.c_int(1), ct.c_double(3.0))
> libc.printf("int=%d  double=%f  double=%f\n", ct.c_int(1),  
> ct.c_double(3.0), ct.c_double(3.0))
>
> On MacOS X this prints:
> int=1  double=-1.996155
> int=1  double=-1.996140  double=0.000000
>
> On a linux box, with a suitably modified ct.CDLL statement, this  
> produces the more reasonable:
> int=1  double=3.000000
> int=1  double=3.000000  double=3.000000
>
> I am totally mystified by the MacOS X results.
>
> -- Russell
>

Perry Smith ( pedz at easesoftware.com )
Ease Software, Inc. ( http://www.easesoftware.com )

Low cost SATA Disk Systems for IBMs p5, pSeries, and RS/6000 AIX systems



From cmiller at securityevaluators.com  Mon Dec 11 23:43:06 2006
From: cmiller at securityevaluators.com (Charlie Miller)
Date: Mon, 11 Dec 2006 16:43:06 -0600
Subject: [Pythonmac-SIG] Catching exceptions inside python with
	catch_exception_raise() doesn't work
Message-ID: <506DCEAC-1E98-4084-B378-58537E3A4148@securityevaluators.com>

Is it possible to monitor for exceptions using the Mac OS X  
catch_exception_raise function inside python?  As you may know, after  
some setup, this function gets called whenever an exception occurs.   
I have a program which works great in C but when I wrap it in python  
it fails.  This is due to the fact that it calls exc_server, which  
uses _dyld_lookup_and_bind, to try to find the location of the  
function catch_exception_raise.  When wrapped in python,  
dyld_lookup_and_bind returns 0x2030 which is NOT the correct  
address.  It then jumps there and tries executing which causes a BUS  
ERROR.

In C at least, dyld_lookup_and_bind works if the  
catch_exception_raise function is compiled into the binary or linked  
dynamically at runtime via a shared library (dylib).  However, it  
doesn't work if it is brought into the executable space via dlopen -  
which presumably is how python does it.

How can I set it up so that I can monitor applications for exceptions  
in a python script?  I could fork and exec a program with its own  
address space, but that seems ridiculous.  I could recompile the  
python interpreter and include catch_exception_raise, but that also  
seems terrible.  Or is it impossible to wrap such a program in python  
- which seems silly?

Any help is appreciated.  Thanks,

Charlie



From Chris.Barker at noaa.gov  Tue Dec 12 01:21:15 2006
From: Chris.Barker at noaa.gov (Christopher Barker)
Date: Mon, 11 Dec 2006 16:21:15 -0800
Subject: [Pythonmac-SIG] cyptes problem on MacOS X
In-Reply-To: <rowen-C597F3.12202311122006@sea.gmane.org>
References: <rowen-C597F3.12202311122006@sea.gmane.org>
Message-ID: <457DF5FB.9040105@noaa.gov>

Russell E. Owen wrote:
> I've been trying out ctypes (from MacPython 2.5) on MacOS X on a PPC Mac 
> and have run into a problem.

I'm no help here.

> On a related subject...assuming I can get this to work...does anyone 
> have any idea how the speed of ctypes relates to programming a python C 
> extension or using SWIG?

probably a bit better performance for ctypes -- swig does put a fair bit 
  between C and Python.

> This is mostly for use with NumPy and 
> interfacing python to C/C++-based heavy duty astronomical image 
> processing code.

In this case, you'll mostly  be passing large arrays back and forth -- I 
doubt the overhead of any wrapping technique will be noticeable.

That being the case, what you want is whatever is most comfortable for 
you.

> We'll mostly be interfacing to C++ so 
> I've considered trying out Boost, but it hasn't had an update in years 
> which worries me.

I'm pretty sure that someone has updated the boost array code for numpy, 
but I'm not sure where to find that out. I'd ask on the numpy list - if 
no one there is using it, I'd call it risky. You might also look at the 
C++-sig -- there's been discussion of this there recently.

pyrex is another candidate worth looking at. There should be examples on 
the scipy Wiki.

This has been my way of thinking:

If you have a substantial existing C/C++ library, particularly one that 
is undergoing continued development and needs to be used with both c/C++ 
and Python code: use SWIG (or maybe SIP, but I've never heard of that 
being used with numpy)

If you have a single dll (or *.so, etc) or small lib, use ctypes.

If you have ONLY the dll, and not source code, definitely use ctypes!

If you're developing extensions just for use with Python, use pyrex or 
boost. Prefer pyrex is it's C, boost if it's C++. Boost gets even more 
compelling if you are using other boost libs for your C++ stuff.

Cxx maybe worth a look-see. I think it's being used for matplotlib, so 
there must be a way to so numpy with it.

http://sourceforge.net/projects/cxx/

It would be great it we could put a note like this, with more detail, on 
the Scipy Wiki. Maybe you could do that when you learn more.

(oh, and be sure to check the Scipy Wiki. there is some stuff there already)

-Chris





-- 
Christopher Barker, Ph.D.
Oceanographer

Emergency Response Division
NOAA/NOS/OR&R            (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception

Chris.Barker at noaa.gov

From ronaldoussoren at mac.com  Tue Dec 12 10:44:26 2006
From: ronaldoussoren at mac.com (Ronald Oussoren)
Date: Tue, 12 Dec 2006 10:44:26 +0100
Subject: [Pythonmac-SIG] Catching exceptions inside python with
	catch_exception_raise() doesn't work
In-Reply-To: <506DCEAC-1E98-4084-B378-58537E3A4148@securityevaluators.com>
References: <506DCEAC-1E98-4084-B378-58537E3A4148@securityevaluators.com>
Message-ID: <97B5399F-EF49-4E4E-B1C3-5ABC8622272A@mac.com>

Hi,

I haven't used the raw mach API's yet, so can't help you here. I must  
say that this looks like a pretty evil API. I wonder why they didn't  
use a manually passed in callback function like sane APIs.

Ronald

On Dec 11, 2006, at 11:43 PM, Charlie Miller wrote:

> Is it possible to monitor for exceptions using the Mac OS X
> catch_exception_raise function inside python?  As you may know, after
> some setup, this function gets called whenever an exception occurs.
> I have a program which works great in C but when I wrap it in python
> it fails.  This is due to the fact that it calls exc_server, which
> uses _dyld_lookup_and_bind, to try to find the location of the
> function catch_exception_raise.  When wrapped in python,
> dyld_lookup_and_bind returns 0x2030 which is NOT the correct
> address.  It then jumps there and tries executing which causes a BUS
> ERROR.
>
> In C at least, dyld_lookup_and_bind works if the
> catch_exception_raise function is compiled into the binary or linked
> dynamically at runtime via a shared library (dylib).  However, it
> doesn't work if it is brought into the executable space via dlopen -
> which presumably is how python does it.
>
> How can I set it up so that I can monitor applications for exceptions
> in a python script?  I could fork and exec a program with its own
> address space, but that seems ridiculous.  I could recompile the
> python interpreter and include catch_exception_raise, but that also
> seems terrible.  Or is it impossible to wrap such a program in python
> - which seems silly?
>
> Any help is appreciated.  Thanks,
>
> Charlie
>
>
> _______________________________________________
> Pythonmac-SIG maillist  -  Pythonmac-SIG at python.org
> http://mail.python.org/mailman/listinfo/pythonmac-sig

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3562 bytes
Desc: not available
Url : http://mail.python.org/pipermail/pythonmac-sig/attachments/20061212/7788113d/attachment.bin 

From ronaldoussoren at mac.com  Tue Dec 12 11:11:58 2006
From: ronaldoussoren at mac.com (Ronald Oussoren)
Date: Tue, 12 Dec 2006 11:11:58 +0100
Subject: [Pythonmac-SIG] cyptes problem on MacOS X
In-Reply-To: <rowen-C597F3.12202311122006@sea.gmane.org>
References: <rowen-C597F3.12202311122006@sea.gmane.org>
Message-ID: <CCE7FA09-7F6C-450F-BF06-8EBECC766E8C@mac.com>


On Dec 11, 2006, at 9:20 PM, Russell E. Owen wrote:

> I've been trying out ctypes (from MacPython 2.5) on MacOS X on a  
> PPC Mac
> and have run into a problem. I hope I'm just doing something  
> stupid, but
> it seems to match the tutorial so I'm really puzzled:

I can reproduce this on an intel system as well.  You might want to  
ask about this on the ctypes list (https://lists.sourceforge.net/ 
lists/listinfo/ctypes-users), maybe you're using ctypes incorrectly.

Ronald

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3562 bytes
Desc: not available
Url : http://mail.python.org/pipermail/pythonmac-sig/attachments/20061212/2f94b70c/attachment.bin 

From Hubert.Holin at lmd.polytechnique.fr  Tue Dec 12 11:02:12 2006
From: Hubert.Holin at lmd.polytechnique.fr (Hubert Holin)
Date: Tue, 12 Dec 2006 11:02:12 +0100
Subject: [Pythonmac-SIG] cyptes problem on MacOS X
In-Reply-To: <mailman.4329.1165876993.32030.pythonmac-sig@python.org>
References: <mailman.4329.1165876993.32030.pythonmac-sig@python.org>
Message-ID: <9748969D-8E22-4548-80B0-672B2186DC05@lmd.polytechnique.fr>

Paris (U.E.), le 12/12/2006

	Bonjour

On Mon Dec 11 21:20:23 CET 2006, Russell E. Owen wrote:

> Other suggestions are welcome. We'll mostly be interfacing to C++ so
> I've considered trying out Boost, but it hasn't had an update in years
> which worries me. Robustness, ease of use and speed are all major
> requirements. Linux and MacOS X are the main platforms.

		Admittedly, the latest Boost release is taking longer than what we  
would have hoped (it is an overwhelmingly volunteer effort, and as  
far as I am concerned, I know I was far less available to help than I  
would have liked, so I can understand that others are in the same  
boat...), but I can assure you that Boost is alive and well, and last  
I heard the next release is indeed now at hand.

		As regards the Python bindings in particular, a lot has been done,  
as can be seen from messages on the c++-sig list. At any rate, it is  
always possible to join Boost and download the WIP library rather  
than the official release if there is a particular, urgent, need.

	Bon courage

			Hubert Holin


From lou_boog2000 at yahoo.com  Tue Dec 12 17:24:13 2006
From: lou_boog2000 at yahoo.com (Lou Pecora)
Date: Tue, 12 Dec 2006 08:24:13 -0800 (PST)
Subject: [Pythonmac-SIG] Best/Safest way to parse NumPy array in C extension?
Message-ID: <579946.48484.qm@web34408.mail.mud.yahoo.com>

I've done this two ways.  Which is better or safer?

Say I'm passing a NumPy array (float) to a C
extension.  One way to "parse" it is (ignoring checks
on return values, etc. for now),

  PyArrayObject *mat;
  PyArg_ParseTuple(args, "O!", &PyArray_Type, &mat);

or is this better,

  PyObject *Pymat;
  PyArrayObject *mat;
  PyArg_ParseTuple(args, "O", &Pymat);
  mat=PyArray_ContiguousFromObject(objin,NPY_DOUBLE,
2,2);

The latter appears to have the constraint that the
array be contiguous.  Or is that an illusion and it's
saying it _expects_ a Python object that has
contiguous data?

I've done both.  Pointing C arrays to the resulting
PyArrays' data works fine, but I fear one way or the
other might be courting disaster.

BTW, I do other checks on dimension, type, etc., but I
left those out here for clarity.  (I hope.)





-- Lou Pecora
      My views are my own.


 
____________________________________________________________________________________
Do you Yahoo!?
Everyone is raving about the all-new Yahoo! Mail beta.
http://new.mail.yahoo.com

From lou_boog2000 at yahoo.com  Tue Dec 12 17:28:31 2006
From: lou_boog2000 at yahoo.com (Lou Pecora)
Date: Tue, 12 Dec 2006 08:28:31 -0800 (PST)
Subject: [Pythonmac-SIG] [SciPy-user] Best/Safest way to parse NumPy
	array in C extension?
In-Reply-To: <579946.48484.qm@web34408.mail.mud.yahoo.com>
Message-ID: <117751.42356.qm@web34405.mail.mud.yahoo.com>

YIKES.  That one line should read

mat=PyArray_ContiguousFromObject(Pymat,NPY_DOUBLE,2,2);

Very sorry.

--- Lou Pecora <lou_boog2000 at yahoo.com> wrote:

> I've done this two ways.  Which is better or safer?
> 
> Say I'm passing a NumPy array (float) to a C
> extension.  One way to "parse" it is (ignoring
> checks
> on return values, etc. for now),
> 
>   PyArrayObject *mat;
>   PyArg_ParseTuple(args, "O!", &PyArray_Type, &mat);
> 
> or is this better,
> 
>   PyObject *Pymat;
>   PyArrayObject *mat;
>   PyArg_ParseTuple(args, "O", &Pymat);
>   mat=PyArray_ContiguousFromObject(objin,NPY_DOUBLE,
> 2,2);
> 
> The latter appears to have the constraint that the
> array be contiguous.  Or is that an illusion and
> it's
> saying it _expects_ a Python object that has
> contiguous data?
> 
> I've done both.  Pointing C arrays to the resulting
> PyArrays' data works fine, but I fear one way or the
> other might be courting disaster.
> 
> BTW, I do other checks on dimension, type, etc., but
> I
> left those out here for clarity.  (I hope.)
> 
> 
> 
> 
> 
> -- Lou Pecora
>       My views are my own.
> 
> 
>  
>
____________________________________________________________________________________
> Do you Yahoo!?
> Everyone is raving about the all-new Yahoo! Mail
> beta.
> http://new.mail.yahoo.com
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.org
>
http://projects.scipy.org/mailman/listinfo/scipy-user
> 


-- Lou Pecora
      My views are my own.


 
____________________________________________________________________________________
Have a burning question?  
Go to www.Answers.yahoo.com and get answers from real people who know.

From Chris.Barker at noaa.gov  Tue Dec 12 18:33:17 2006
From: Chris.Barker at noaa.gov (Christopher Barker)
Date: Tue, 12 Dec 2006 09:33:17 -0800
Subject: [Pythonmac-SIG] [SciPy-user] Best/Safest way to parse
 NumPy	array in C extension?
In-Reply-To: <117751.42356.qm@web34405.mail.mud.yahoo.com>
References: <117751.42356.qm@web34405.mail.mud.yahoo.com>
Message-ID: <457EE7DD.9050304@noaa.gov>

Lou,

This type of question is best posed on the numpy list:

http://projects.scipy.org/mailman/listinfo/numpy-discussion

Lou Pecora wrote:

>>   PyArrayObject *mat;
>>   PyArg_ParseTuple(args, "O!", &PyArray_Type, &mat);

If you do this, you then need to check and see if the array that got 
passed in has the properties you want: size, shape, type.

>>   PyObject *Pymat;
>>   PyArrayObject *mat;
>>   PyArg_ParseTuple(args, "O", &Pymat);
>>   mat=PyArray_ContiguousFromObject(objin,NPY_DOUBLE,
>> 2,2);

I've always used PyArray_ContiguousFromObject, because then it does all 
the checking for me, and it allows my code to accept anything that can 
by turned into a numpy array.

>> The latter appears to have the constraint that the
>> array be contiguous.  Or is that an illusion and
>> it's
>> saying it _expects_ a Python object that has
>> contiguous data?

What it does is convert whatever you pass in to a Contiguous array. If 
it already is a contiguous array, then it returns itself (I think -- 
check the docs). There is probably something like:

PyArray_FromObject

That does not guarantee contiguous, if you don't need that.

By the way, I think these are the old Numeric names, which numpy has a 
compatibility layer for, but you may want to use the new, numpy ones. 
There are a lot, but most are special cases of:

Py_Array_FromAny

This is well documented in the Numpy book, I'm not sure about what's online.


>> I've done both.  Pointing C arrays to the resulting
>> PyArrays' data works fine, but I fear one way or the
>> other might be courting disaster.
>>
>> BTW, I do other checks on dimension, type, etc., but
>> I left those out here for clarity.  (I hope.)

If you do all those checks, you're OK, but do make sure you are not 
assuming contiguous, if you haven't forced it -- that can slip through 
testing.

-Chris
-- 
Christopher Barker, Ph.D.
Oceanographer

Emergency Response Division
NOAA/NOS/OR&R            (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception

Chris.Barker at noaa.gov

From lou_boog2000 at yahoo.com  Tue Dec 12 18:53:07 2006
From: lou_boog2000 at yahoo.com (Lou Pecora)
Date: Tue, 12 Dec 2006 09:53:07 -0800 (PST)
Subject: [Pythonmac-SIG] [SciPy-user] Best/Safest way to parse NumPy
	array in C extension?
In-Reply-To: <457EE7DD.9050304@noaa.gov>
Message-ID: <20061212175307.96350.qmail@web34414.mail.mud.yahoo.com>

Thanks for the answers, Chris.  A few comments and
questions beyond what I asked and you answered:

--- Christopher Barker <Chris.Barker at noaa.gov> wrote:

> Lou,
> 
> This type of question is best posed on the numpy
> list:
> 
>
http://projects.scipy.org/mailman/listinfo/numpy-discussion

Ah, yet another mailing list to join.  Sigh.  :-) 

> Lou Pecora wrote:
> 
> >>   PyArrayObject *mat;
> >>   PyArg_ParseTuple(args, "O!", &PyArray_Type,
> &mat);
> 
> If you do this, you then need to check and see if
> the array that got 
> passed in has the properties you want: size, shape,
> type.

I do, but I don't know how to check in Python for
contiguous.  I will search for that.  Should be a way.
> 
> >>   PyObject *Pymat;
> >>   PyArrayObject *mat;
> >>   PyArg_ParseTuple(args, "O", &Pymat);
> >>  
> mat=PyArray_ContiguousFromObject(objin,NPY_DOUBLE,
> >> 2,2);
> 
> I've always used PyArray_ContiguousFromObject,
> because then it does all 
> the checking for me, and it allows my code to accept
> anything that can 
> by turned into a numpy array.

Got it.  That's what I suspected.  

> >> The latter appears to have the constraint that
> the
> >> array be contiguous.  Or is that an illusion and
> >> it's
> >> saying it _expects_ a Python object that has
> >> contiguous data?
> 
> What it does is convert whatever you pass in to a
> Contiguous array. If 
> it already is a contiguous array, then it returns
> itself (I think -- 
> check the docs). There is probably something like:
> 
> PyArray_FromObject
> 
> That does not guarantee contiguous, if you don't
> need that.

That can lead to a problem.  If I have a C extension
in which I expect to operate on the array data in
place, then passing a non-contiguous array will cause
a new data set to be developed (new memory, I would
assume) and the original array would not be changed in
place by my C operations.  Then I have a problem.  The
original array is unchanged, but I assume it has been
changed.  In this case I would want a check on whether
the input array is contiguous and an exception raised
if not ... or something to say there's a
data-structure problem.




-- Lou Pecora
      My views are my own.


 
____________________________________________________________________________________
Yahoo! Music Unlimited
Access over 1 million songs.
http://music.yahoo.com/unlimited

From delza at livingcode.org  Wed Dec 13 01:28:22 2006
From: delza at livingcode.org (Dethe Elza)
Date: Tue, 12 Dec 2006 16:28:22 -0800
Subject: [Pythonmac-SIG] py2app plugin problem: "Error loading code"
In-Reply-To: <50193F5C-BDEF-4C33-BF27-E553CD76304C@mac.com>
References: <0FE46EBB-032F-4157-BD8D-318EDB94A47F@alum.mit.edu>
	<7C1340EA-9112-4728-A693-E8BD9296574A@alum.mit.edu>
	<50193F5C-BDEF-4C33-BF27-E553CD76304C@mac.com>
Message-ID: <98BD2CB2-3EC4-457C-8196-932F731D6063@livingcode.org>

On 10-Dec-06, at 1:37 PM, Ronald Oussoren wrote:

> The problem is that py2app/bundletemplate/prebuild/main in the  
> py2app distribution is an intel-only binary (at least in the  
> current trunk) instead of a universal binary.
>
> Removing the file and running 'setup.py' in the bundletemplate  
> directory should rebuild it with the proper architecture for your  
> machine.

Ytrewq1 and I were working on this and he(?) came up with a solution  
that seems to work for universal binaries:

On 10-Nov-2006 ytrewq1 wrote:

> I think I got something working.
>
> The steps I followed were:
>
> 1. Move the prebuilt main binary out of the way
> (py2app/bundletmplate/prebuilt/main)
> 2. Modify py2app/bundletemplate/setup.py so that it builds  
> universal binaries
>
> In step 2, I changed the CFLAGS line to be:
>
> CFLAGS = '-Os -Wall -isysroot /Developer/SDKs/MacOSX10.4u.sdk -arch
> i386 -arch ppc'
>
> That seemed to do it for me.

This worked for me, but I had other issues, which turned out to be  
that when I'd installed Stackless it had taken over as my main  
python, and since it mostly worked, it took me awhile to see the  
obvious.

So far using the above recipe is working well for me.

--Dethe

When laws are outlawed, only outlaws will have laws.



From generalkernel at gmail.com  Tue Dec 12 22:19:20 2006
From: generalkernel at gmail.com (Gen Kazama)
Date: Tue, 12 Dec 2006 16:19:20 -0500
Subject: [Pythonmac-SIG] CPython on Mac OSX
Message-ID: <AF064B05-E293-471A-BB78-E5C5402B9436@gmail.com>

Hi I'm running python 2.3 on 10.4.  i made an application using a C  
backend and a python frontend.

This is a school project and although it works perfectly on the  
school computers, i have trouble compiling the C code correctly on  
the mac to make it into a python module (maybe this is because the  
school computers used python 2.4-ie, an updated version of Python.h).

Anyway here is how I compile it at school:

         gcc -I /usr/include/python2.4 -c mandel.c
         gcc -shared mandel.o -o mandel.so

which works with no flaws.  then , I try compiling it on my mac:

         gcc -I /usr/include/python2.3 -c mandel.c
         gcc mandel.o -o mandel.so

(note: -shared wasn't found on my mac for some reason). I get the  
following errors:

/usr/bin/ld: Undefined symbols:
_main
_PyArg_ParseTuple
_PyInt_FromLong
_PyString_FromString
_PyTuple_New
_PyTuple_SetItem
_Py_InitModule4
collect2: ld returned 1 exit status


If anyone could tell me how to compile this C code into a python  
module on my mac, i would greatly appreciate it.  thanks!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/pythonmac-sig/attachments/20061212/b36856da/attachment.htm 

From bob at redivi.com  Thu Dec 14 11:28:29 2006
From: bob at redivi.com (Bob Ippolito)
Date: Thu, 14 Dec 2006 18:28:29 +0800
Subject: [Pythonmac-SIG] CPython on Mac OSX
In-Reply-To: <AF064B05-E293-471A-BB78-E5C5402B9436@gmail.com>
References: <AF064B05-E293-471A-BB78-E5C5402B9436@gmail.com>
Message-ID: <6a36e7290612140228w72627897l988b7f647a881093@mail.gmail.com>

On 12/13/06, Gen Kazama <generalkernel at gmail.com> wrote:
> Hi I'm running python 2.3 on 10.4.  i made an application using a C backend
> and a python frontend.
>
> This is a school project and although it works perfectly on the school
> computers, i have trouble compiling the C code correctly on the mac to make
> it into a python module (maybe this is because the school computers used
> python 2.4-ie, an updated version of Python.h).
>
> Anyway here is how I compile it at school:
>
>         gcc -I /usr/include/python2.4 -c mandel.c
>         gcc -shared mandel.o -o mandel.so
>
> which works with no flaws.  then , I try compiling it on my mac:
>
>         gcc -I /usr/include/python2.3 -c mandel.c
>         gcc mandel.o -o mandel.so
>
> (note: -shared wasn't found on my mac for some reason). I get the following
> errors:
>
> /usr/bin/ld: Undefined symbols:
> _main
> _PyArg_ParseTuple
> _PyInt_FromLong
> _PyString_FromString
> _PyTuple_New
> _PyTuple_SetItem
> _Py_InitModule4
> collect2: ld returned 1 exit status
>
>
> If anyone could tell me how to compile this C code into a python module on
> my mac, i would greatly appreciate it.  thanks!

The best way is to use distutils (or setuptools). Create a setup.py
that compiles it for you. This works on every platform, even windows.

-bob

From ronaldoussoren at mac.com  Thu Dec 14 11:31:17 2006
From: ronaldoussoren at mac.com (Ronald Oussoren)
Date: Thu, 14 Dec 2006 11:31:17 +0100
Subject: [Pythonmac-SIG] CPython on Mac OSX
In-Reply-To: <AF064B05-E293-471A-BB78-E5C5402B9436@gmail.com>
References: <AF064B05-E293-471A-BB78-E5C5402B9436@gmail.com>
Message-ID: <AD3863E0-71AF-40D7-B0FE-0CEBE4D8F9F4@mac.com>

The best way to do this is using distutils, which is the standard  
python package for building and installing python packages.

The document at <http://docs.python.org/ext/building.html> should get  
you going.

To quote from that, create a setup.py containing:

#BEGIN OF FILE
from distutils.core import setup, Extension

module1 = Extension('demo',
                     sources = ['mandel.c'])

setup (name = 'MandelPackage',
        version = '1.0',
        description = 'This is a demo package',
        ext_modules = [module1])
#END OF FILE

If you want to install the extension: 'python setup.py install'. This  
will install it into the site-packages directory.

If you create a setup.cfg containing the text below you can use  
'python setup.py build' to build the extension in the current directory

# setup.cfg
[build_ext]
inplace = 1
# end setup.cfg

Ronald


On Dec 12, 2006, at 10:19 PM, Gen Kazama wrote:

> Hi I'm running python 2.3 on 10.4.  i made an application using a C  
> backend and a python frontend.
>
> This is a school project and although it works perfectly on the  
> school computers, i have trouble compiling the C code correctly on  
> the mac to make it into a python module (maybe this is because the  
> school computers used python 2.4-ie, an updated version of Python.h).
>
> Anyway here is how I compile it at school:
>
>         gcc -I /usr/include/python2.4 -c mandel.c
>         gcc -shared mandel.o -o mandel.so
>
> which works with no flaws.  then , I try compiling it on my mac:
>
>         gcc -I /usr/include/python2.3 -c mandel.c
>         gcc mandel.o -o mandel.so
>
> (note: -shared wasn't found on my mac for some reason). I get the  
> following errors:
>
> /usr/bin/ld: Undefined symbols:
> _main
> _PyArg_ParseTuple
> _PyInt_FromLong
> _PyString_FromString
> _PyTuple_New
> _PyTuple_SetItem
> _Py_InitModule4
> collect2: ld returned 1 exit status
>
>
> If anyone could tell me how to compile this C code into a python  
> module on my mac, i would greatly appreciate it.  thanks! 
> _______________________________________________
> Pythonmac-SIG maillist  -  Pythonmac-SIG at python.org
> http://mail.python.org/mailman/listinfo/pythonmac-sig

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3562 bytes
Desc: not available
Url : http://mail.python.org/pipermail/pythonmac-sig/attachments/20061214/4b59889d/attachment.bin 

From dan at chalkie.org.uk  Thu Dec 14 15:03:12 2006
From: dan at chalkie.org.uk (Dan White)
Date: Thu, 14 Dec 2006 16:03:12 +0200
Subject: [Pythonmac-SIG] undefined symbols when making python wrapped c
	libs for itk
In-Reply-To: <mailman.8.1166094003.711.pythonmac-sig@python.org>
References: <mailman.8.1166094003.711.pythonmac-sig@python.org>
Message-ID: <F570DAFD-0544-4303-BA07-306BD63A8F82@chalkie.org.uk>

Hi,

Actually i see the same error message using python2.5 when building  
ITK python wrapping.

I was wondering what i am doing wrong that  ld is not seeing pytohn  
properly to find those symbols?

Any one know what is going wrong to give error message like

>> /usr/bin/ld: Undefined symbols:
>> _main
>> _PyArg_ParseTuple
>> _PyInt_FromLong
>> _PyString_FromString
>> _PyTuple_New
>> _PyTuple_SetItem
>> _Py_InitModule4
>> collect2: ld returned 1 exit status

which lib or headers or something is ld not seeing?

cheers

Dan


On 14 Dec 2006, at 13:00, pythonmac-sig-request at python.org wrote:

> From: "Bob Ippolito" <bob at redivi.com>
> Date: 14 December 2006 12:28:29 GMT+02:00
> To: "Gen Kazama" <generalkernel at gmail.com>
> Cc: pythonmac-sig at python.org
> Subject: Re: [Pythonmac-SIG] CPython on Mac OSX
>
>
> On 12/13/06, Gen Kazama <generalkernel at gmail.com> wrote:
>> Hi I'm running python 2.3 on 10.4.  i made an application using a  
>> C backend
>> and a python frontend.
>>
>> This is a school project and although it works perfectly on the  
>> school
>> computers, i have trouble compiling the C code correctly on the  
>> mac to make
>> it into a python module (maybe this is because the school  
>> computers used
>> python 2.4-ie, an updated version of Python.h).
>>
>> Anyway here is how I compile it at school:
>>
>>         gcc -I /usr/include/python2.4 -c mandel.c
>>         gcc -shared mandel.o -o mandel.so
>>
>> which works with no flaws.  then , I try compiling it on my mac:
>>
>>         gcc -I /usr/include/python2.3 -c mandel.c
>>         gcc mandel.o -o mandel.so
>>
>> (note: -shared wasn't found on my mac for some reason). I get the  
>> following
>> errors:
>>
>> /usr/bin/ld: Undefined symbols:
>> _main
>> _PyArg_ParseTuple
>> _PyInt_FromLong
>> _PyString_FromString
>> _PyTuple_New
>> _PyTuple_SetItem
>> _Py_InitModule4
>> collect2: ld returned 1 exit status
>>
>>
>> If anyone could tell me how to compile this C code into a python  
>> module on
>> my mac, i would greatly appreciate it.  thanks!
>
> The best way is to use distutils (or setuptools). Create a setup.py
> that compiles it for you. This works on every platform, even windows.
>
> -bob

Dr. Daniel James White BSc. (Hons.) PhD
Bioimaging Coordinator
Nanoscience Centre and Department of Biological and Environmental  
Sciences
Division of Molecular Recognition
Ambiotica C242
PO Box 35
University of Jyv?skyl?
Jyv?skyl?
FIN 40014
Finland

+358 14 260 4183 (work)
+358 468102840 (mobile)
http://www.bioimagexd.org
http://www.chalkie.org.uk
dan at chalkie.org.uk
white at cc.jyu.fi




From bob at redivi.com  Thu Dec 14 18:20:22 2006
From: bob at redivi.com (Bob Ippolito)
Date: Fri, 15 Dec 2006 01:20:22 +0800
Subject: [Pythonmac-SIG] undefined symbols when making python wrapped c
	libs for itk
In-Reply-To: <F570DAFD-0544-4303-BA07-306BD63A8F82@chalkie.org.uk>
References: <mailman.8.1166094003.711.pythonmac-sig@python.org>
	<F570DAFD-0544-4303-BA07-306BD63A8F82@chalkie.org.uk>
Message-ID: <6a36e7290612140920h751020b5pfe96b0c4154ebc90@mail.gmail.com>

The correct way to link Python extensions is to use distutils, end of
story. The compiler/linker flags are different for every platform and
sometimes differ based upon the way it's built. Specifically in this
case all that should be necessary is -undefined dynamic_lookup --
however, the only good fix is to use distutils.

-bob

On 12/14/06, Dan White <dan at chalkie.org.uk> wrote:
> Hi,
>
> Actually i see the same error message using python2.5 when building
> ITK python wrapping.
>
> I was wondering what i am doing wrong that  ld is not seeing pytohn
> properly to find those symbols?
>
> Any one know what is going wrong to give error message like
>
> >> /usr/bin/ld: Undefined symbols:
> >> _main
> >> _PyArg_ParseTuple
> >> _PyInt_FromLong
> >> _PyString_FromString
> >> _PyTuple_New
> >> _PyTuple_SetItem
> >> _Py_InitModule4
> >> collect2: ld returned 1 exit status
>
> which lib or headers or something is ld not seeing?
>
> cheers
>
> Dan
>
>
> On 14 Dec 2006, at 13:00, pythonmac-sig-request at python.org wrote:
>
> > From: "Bob Ippolito" <bob at redivi.com>
> > Date: 14 December 2006 12:28:29 GMT+02:00
> > To: "Gen Kazama" <generalkernel at gmail.com>
> > Cc: pythonmac-sig at python.org
> > Subject: Re: [Pythonmac-SIG] CPython on Mac OSX
> >
> >
> > On 12/13/06, Gen Kazama <generalkernel at gmail.com> wrote:
> >> Hi I'm running python 2.3 on 10.4.  i made an application using a
> >> C backend
> >> and a python frontend.
> >>
> >> This is a school project and although it works perfectly on the
> >> school
> >> computers, i have trouble compiling the C code correctly on the
> >> mac to make
> >> it into a python module (maybe this is because the school
> >> computers used
> >> python 2.4-ie, an updated version of Python.h).
> >>
> >> Anyway here is how I compile it at school:
> >>
> >>         gcc -I /usr/include/python2.4 -c mandel.c
> >>         gcc -shared mandel.o -o mandel.so
> >>
> >> which works with no flaws.  then , I try compiling it on my mac:
> >>
> >>         gcc -I /usr/include/python2.3 -c mandel.c
> >>         gcc mandel.o -o mandel.so
> >>
> >> (note: -shared wasn't found on my mac for some reason). I get the
> >> following
> >> errors:
> >>
> >> /usr/bin/ld: Undefined symbols:
> >> _main
> >> _PyArg_ParseTuple
> >> _PyInt_FromLong
> >> _PyString_FromString
> >> _PyTuple_New
> >> _PyTuple_SetItem
> >> _Py_InitModule4
> >> collect2: ld returned 1 exit status
> >>
> >>
> >> If anyone could tell me how to compile this C code into a python
> >> module on
> >> my mac, i would greatly appreciate it.  thanks!
> >
> > The best way is to use distutils (or setuptools). Create a setup.py
> > that compiles it for you. This works on every platform, even windows.
> >
> > -bob
>
> Dr. Daniel James White BSc. (Hons.) PhD
> Bioimaging Coordinator
> Nanoscience Centre and Department of Biological and Environmental
> Sciences
> Division of Molecular Recognition
> Ambiotica C242
> PO Box 35
> University of Jyv?skyl?
> Jyv?skyl?
> FIN 40014
> Finland
>
> +358 14 260 4183 (work)
> +358 468102840 (mobile)
> http://www.bioimagexd.org
> http://www.chalkie.org.uk
> dan at chalkie.org.uk
> white at cc.jyu.fi
>
>
>
> _______________________________________________
> Pythonmac-SIG maillist  -  Pythonmac-SIG at python.org
> http://mail.python.org/mailman/listinfo/pythonmac-sig
>

From josh.p.marshall at gmail.com  Fri Dec 15 04:49:06 2006
From: josh.p.marshall at gmail.com (Josh Marshall)
Date: Fri, 15 Dec 2006 14:49:06 +1100
Subject: [Pythonmac-SIG] Universal binary SciPy on OS X
Message-ID: <3539EE2B-72B8-4BF7-8431-A3F8AE8302E2@gmail.com>

I have been trying to build a universal SciPy that works, so I can  
include it in a redistributable application I build with py2app.

The one at http://pythonmac.org/packages/py24-fat/index.html is  
*incomplete*, it contains some dynamic libraries which are ppc only,  
which is what also happens when I try to build a universal binary  
myself.Some libs build fine, but some are only a single arch (the  
native one).

The results of
      for FILE in `find . | grep '\.so'` ; do lipo -info $FILE ; done
are:
------------------------------------------------------------------------ 
-----------
Architectures in the fat file: ./scipy/cluster/_vq.so are: i386 ppc
Non-fat file: ./scipy/fftpack/_fftpack.so is architecture: ppc
Non-fat file: ./scipy/fftpack/convolve.so is architecture: ppc
Non-fat file: ./scipy/integrate/_odepack.so is architecture: ppc
Non-fat file: ./scipy/integrate/_quadpack.so is architecture: ppc
Non-fat file: ./scipy/integrate/vode.so is architecture: ppc
Non-fat file: ./scipy/interpolate/_fitpack.so is architecture: ppc
Non-fat file: ./scipy/interpolate/dfitpack.so is architecture: ppc
Architectures in the fat file: ./scipy/io/numpyio.so are: i386 ppc
Architectures in the fat file: ./scipy/lib/blas/cblas.so are: i386 ppc
Non-fat file: ./scipy/lib/blas/fblas.so is architecture: ppc
Architectures in the fat file: ./scipy/lib/lapack/atlas_version.so  
are: i386 ppc
Non-fat file: ./scipy/lib/lapack/calc_lwork.so is architecture: ppc
Architectures in the fat file: ./scipy/lib/lapack/clapack.so are:  
i386 ppc
Architectures in the fat file: ./scipy/lib/lapack/flapack.so are:  
i386 ppc
Non-fat file: ./scipy/linalg/_flinalg.so is architecture: ppc
Non-fat file: ./scipy/linalg/_iterative.so is architecture: ppc
Architectures in the fat file: ./scipy/linalg/atlas_version.so are:  
i386 ppc
Non-fat file: ./scipy/linalg/calc_lwork.so is architecture: ppc
Architectures in the fat file: ./scipy/linalg/cblas.so are: i386 ppc
Architectures in the fat file: ./scipy/linalg/clapack.so are: i386 ppc
Non-fat file: ./scipy/linalg/fblas.so is architecture: ppc
Non-fat file: ./scipy/linalg/flapack.so is architecture: ppc
Architectures in the fat file: ./scipy/linsolve/_csuperlu.so are:  
i386 ppc
Architectures in the fat file: ./scipy/linsolve/_dsuperlu.so are:  
i386 ppc
Architectures in the fat file: ./scipy/linsolve/_ssuperlu.so are:  
i386 ppc
Architectures in the fat file: ./scipy/linsolve/_zsuperlu.so are:  
i386 ppc
Architectures in the fat file: ./scipy/ndimage/_nd_image.so are: i386  
ppc
Non-fat file: ./scipy/odr/__odrpack.so is architecture: ppc
Non-fat file: ./scipy/optimize/_cobyla.so is architecture: ppc
Non-fat file: ./scipy/optimize/_lbfgsb.so is architecture: ppc
Non-fat file: ./scipy/optimize/_minpack.so is architecture: ppc
Architectures in the fat file: ./scipy/optimize/_zeros.so are: i386 ppc
Non-fat file: ./scipy/optimize/minpack2.so is architecture: ppc
Architectures in the fat file: ./scipy/optimize/moduleTNC.so are:  
i386 ppc
Architectures in the fat file: ./scipy/signal/sigtools.so are: i386 ppc
Architectures in the fat file: ./scipy/signal/spline.so are: i386 ppc
Non-fat file: ./scipy/sparse/sparsetools.so is architecture: ppc
Non-fat file: ./scipy/special/_cephes.so is architecture: ppc
Non-fat file: ./scipy/special/specfun.so is architecture: ppc
Non-fat file: ./scipy/stats/futil.so is architecture: ppc
Non-fat file: ./scipy/stats/mvn.so is architecture: ppc
Non-fat file: ./scipy/stats/statlib.so is architecture: ppc
Architectures in the fat file: ./scipy/stsci/convolve/_correlate.so  
are: i386 ppc
Architectures in the fat file: ./scipy/stsci/image/_combine.so are:  
i386 ppc
------------------------------------------------------------------------ 
-----------

So I didn't pursue this any further for the time being. Since I am  
not an expert on the SciPy build process, I'm trying to work around  
this.

Instead I took your (Chris') versions from http://trichech.us/? 
page_id=4 , both ppc and intel SuperPacks. I yanked the scipy dirs  
out of the packages, and called them scipy-ppc and scipy-i386. Then  
what I did was lipo the shared libs together, as described on Apple's  
site:
http://developer.apple.com/opensource/buildingopensourceuniversal.html

Like so:

mv scipy-ppc scipy-universal
cd scipy-universal
for FILE in `find . | grep '\.so'` ; do file $FILE ;  
done               # shows all ppc libs
for LIB in `find . | grep '\.so'` ; do lipo -create $LIB ../scipy- 
i386/$LIB -output $LIB ; done
for FILE in `find . | grep '\.so'` ; do file $FILE ;  
done              # shows all universal libs, yay!
cd ..
sudo cp -r scipy-universal $PYTHON_SITEPACKAGES/scipy

Now this works fine on my G4. I won't have a chance to test on an  
Intel Mac until later in the week, but this looks promising.

Chris, since the package seem to be identical apart from the shared  
libs, you could probably do this on the SuperPack and only need to  
distribute one version.

Could I also have some advice on how to deal with this sort of stuff.  
I imagine that much of it is tied up with not having an Apple-blessed  
Fortran compiler. (Fingers crossed for Leopard).

Thanks for all your help everyone,
Josh

From vivacarlie at gmail.com  Sat Dec 16 23:18:47 2006
From: vivacarlie at gmail.com (Nehemiah Dacres)
Date: Sat, 16 Dec 2006 16:18:47 -0600
Subject: [Pythonmac-SIG] About Obj-C 2.0 language bridge
Message-ID: <65fadfc30612161418q11fe1572l5376cb791fa55217@mail.gmail.com>

I just wanted to bring the attention to the community that there will
be a python/Objective
C <http://theocacao.com/document.page/373>
<http://theocacao.com/document.page/373>language bridge implemented come the
release of Leopard, Mac OS
10.5<http://developer.apple.com/leopard/overview/apptech.html>
.
These are a few questions I wanted to ask the MacPython Dev team
I was wondering how are we going to tailor the  MacPython programming
environment for those who want to use the bridge in programming for python?
Will Py2App take advantage of this new ability and if so, how?
How does this contrast with using PyObjC?


-- 

"lalalalala! it's not broken because I can use it"

http://linux.slashdot.org/comments.pl?sid=194281&threshold=1&commentsort=0&mode=thread&cid=15927703
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/pythonmac-sig/attachments/20061216/e7271873/attachment.html 

From ronaldoussoren at mac.com  Sun Dec 17 11:17:55 2006
From: ronaldoussoren at mac.com (Ronald Oussoren)
Date: Sun, 17 Dec 2006 02:17:55 -0800
Subject: [Pythonmac-SIG] About Obj-C 2.0 language bridge
In-Reply-To: <65fadfc30612161418q11fe1572l5376cb791fa55217@mail.gmail.com>
References: <65fadfc30612161418q11fe1572l5376cb791fa55217@mail.gmail.com>
Message-ID: <32B7418C-010F-1000-BA3F-2BA499D39836-Webmail-10009@mac.com>

 
On Saturday, December 16, 2006, at 11:19PM, "Nehemiah Dacres" <vivacarlie at gmail.com> wrote:
>I just wanted to bring the attention to the community that there will
>be a python/Objective
>C <http://theocacao.com/document.page/373>
><http://theocacao.com/document.page/373>language bridge implemented come the
>release of Leopard, Mac OS
>10.5<http://developer.apple.com/leopard/overview/apptech.html>
>.
>These are a few questions I wanted to ask the MacPython Dev team
>I was wondering how are we going to tailor the  MacPython programming
>environment for those who want to use the bridge in programming for python?
>Will Py2App take advantage of this new ability and if so, how?
>How does this contrast with using PyObjC?

As Leopard is still under NDA please don't expect answers with any kind of substance. 

Ronald


From jrus at hcs.harvard.edu  Mon Dec 18 20:06:10 2006
From: jrus at hcs.harvard.edu (Jacob Rus)
Date: Mon, 18 Dec 2006 14:06:10 -0500
Subject: [Pythonmac-SIG] About Obj-C 2.0 language bridge
In-Reply-To: <65fadfc30612161418q11fe1572l5376cb791fa55217@mail.gmail.com>
References: <65fadfc30612161418q11fe1572l5376cb791fa55217@mail.gmail.com>
Message-ID: <em6or5$ria$1@sea.gmane.org>

Nehemiah Dacres wrote:
> I just wanted to bring the attention to the community that there will
> be a python/Objective C language bridge implemented come the
> release of Leopard, Mac OS 10.5
> 
> These are a few questions I wanted to ask the MacPython Dev team
> I was wondering how are we going to tailor the  MacPython programming
> environment for those who want to use the bridge in programming for python?
> Will Py2App take advantage of this new ability and if so, how?
> How does this contrast with using PyObjC?

I don't actually have any inside knowledge (I don't have access to 
Leopard developer builds, etc.), but I rather doubt that Apple is 
completely re-implementing PyObjC.  At least a few of the people who 
work on PyObjC are Apple employees, and there was a tutorial showing how 
to make a Cocoa app using PyObjC at apple.com a few months ago.  It was 
my impression that the Cocoa-python bridge discussed in the leopard 
preview docs *is* PyObjC, though of course I could be way off the mark, etc.

-Jacob


From hamish at gmail.com  Tue Dec 19 00:11:27 2006
From: hamish at gmail.com (Hamish Allan)
Date: Mon, 18 Dec 2006 23:11:27 +0000
Subject: [Pythonmac-SIG] How to fathom appscript
Message-ID: <597e7edb0612181511n689164f6o3bb6f462f21536a4@mail.gmail.com>

Hi,

I'm new to appscript (and relatively new to python) and I'm having
trouble fathoming out how to determine what can be done with each type
of object.

For example,

itunes = app('iTunes')
x = itunes.sources.first.playlists[its.name.contains('MyName')]

This code fetches any playlists containing the string 'MyName'. But I
want an exact match. Using 'equals' rather than 'contains' doesn't
work. How do I find out which operations its.name supports? And more
generally, what is possible in other similar situations with different
objects?

I also want to do what the following code suggests:

x = itunes.sources.first.playlist_folders[its.name.equals('MyFolder')]
y = itunes.sources.first.playlists[its.parent.equals(x)]

Is this possible?

Many thanks,
Hamish

From njriley at uiuc.edu  Tue Dec 19 01:20:54 2006
From: njriley at uiuc.edu (Nicholas Riley)
Date: Mon, 18 Dec 2006 18:20:54 -0600
Subject: [Pythonmac-SIG] How to fathom appscript
In-Reply-To: <597e7edb0612181511n689164f6o3bb6f462f21536a4@mail.gmail.com>
References: <597e7edb0612181511n689164f6o3bb6f462f21536a4@mail.gmail.com>
Message-ID: <20061219002054.GA64589@uiuc.edu>

On Mon, Dec 18, 2006 at 11:11:27PM +0000, Hamish Allan wrote:
> itunes = app('iTunes')
> x = itunes.sources.first.playlists[its.name.contains('MyName')]
> 
> This code fetches any playlists containing the string 'MyName'. But I
> want an exact match. Using 'equals' rather than 'contains' doesn't
> work.

You just use '==' instead.

In [4]: itunes.sources.first.playlists[its.name == 'Stations']()
Out[4]: [app(u'/Applications/iTunes.app').sources.ID(41).user_playlists.ID(1936)]

But in this case, you don't need to.

In [6]: itunes.sources.first.playlists['Stations']()
Out[6]: app(u'/Applications/iTunes.app').sources.ID(41).user_playlists.ID(1936)

Most objects support multiple reference forms, as you'll see if you
look at their documentation.

> How do I find out which operations its.name supports? And more
> generally, what is possible in other similar situations with different
> objects?

Use .help().  So, for example, you can see that the playlists can be
referenced by index, name or ID:

In [7]: itunes.sources.first.playlists.help()
==============================================================================
Appscript Help (-t)

Reference: app(u'/Applications/iTunes.app').sources.first.playlists

------------------------------------------------------------------------------
Description of reference

Element: playlists -- by index, name, id
[...]

Note the reference forms above.

It also helps to view the scripting dictionary, either using
appscript's tools for doing so in a Web browser, or just with Script
Editor (as I normally do).

> I also want to do what the following code suggests:
> 
> x = itunes.sources.first.playlist_folders[its.name.equals('MyFolder')]

You want 'folder_playlists' not 'playlist_folders'.  So again it's
pretty simple:

In [15]: itunes.sources.first.folder_playlists['Statistics']()
Out[15]: app(u'/Applications/iTunes.app').sources.ID(41).folder_playlists.ID(122340)

> y = itunes.sources.first.playlists[its.parent.equals(x)]

Unfortunately here you run into a problem, as Apple didn't fully
implement terminology for 'folder playlists', but you can view the
scripting dictionary to figure it out.  'parent', while a property of
playlists, isn't always set.  So the only thing I could figure out was
to iterate through them in Python:

In [45]: stats_id = itunes.folder_playlists['Statistics'].id()
In [46]: [p for p in itunes.user_playlists() if p.parent.exists()
   ....:  and p.parent.id() == stats_id]
Out[46]: 
[app(u'/Applications/iTunes.app').sources.ID(41).user_playlists.ID(1032),
 app(u'/Applications/iTunes.app').sources.ID(41).user_playlists.ID(1158),
 app(u'/Applications/iTunes.app').sources.ID(41).user_playlists.ID(1060)]

-- 
Nicholas Riley <njriley at uiuc.edu> | <http://www.uiuc.edu/ph/www/njriley>

From mithro at mithis.com  Sat Dec 23 11:00:55 2006
From: mithro at mithis.com (Tim Ansell)
Date: Sat, 23 Dec 2006 20:30:55 +1030
Subject: [Pythonmac-SIG] Python and Zeroconf?
Message-ID: <1166868056.6695.3.camel@localhost>

Hello,

My game is using ZeroConf to find local game servers. Under Linux I can
use avahi or pyZeroConf to locate the servers (depending if avahi is
installed or not). 

However as Mac OS X has it's own mDNS server I can't use either of these
options. Is there a way to interface with the Mac's Zeroconf (Bonjour)? 

I have found
http://www-unix.mcs.anl.gov/fl/research/accessgrid/bonjour-py/bonjour-py.html however there doesn't seem to be Universal Python binaries which I need to build binary packages.

Anyway have any ideas?

Thanks for your time.

Tim Ansell


From bob at redivi.com  Sat Dec 23 13:29:36 2006
From: bob at redivi.com (Bob Ippolito)
Date: Sat, 23 Dec 2006 20:29:36 +0800
Subject: [Pythonmac-SIG] Python and Zeroconf?
In-Reply-To: <1166868056.6695.3.camel@localhost>
References: <1166868056.6695.3.camel@localhost>
Message-ID: <6a36e7290612230429g11333c3cx384882cf133c072@mail.gmail.com>

On 12/23/06, Tim Ansell <mithro at mithis.com> wrote:
> Hello,
>
> My game is using ZeroConf to find local game servers. Under Linux I can
> use avahi or pyZeroConf to locate the servers (depending if avahi is
> installed or not).
>
> However as Mac OS X has it's own mDNS server I can't use either of these
> options. Is there a way to interface with the Mac's Zeroconf (Bonjour)?
>
> I have found
> http://www-unix.mcs.anl.gov/fl/research/accessgrid/bonjour-py/bonjour-py.html however there doesn't seem to be Universal Python binaries which I need to build binary packages.

Looks like source code. Compiles here. What's the problem?

-bob

From vivacarlie at gmail.com  Tue Dec 26 19:29:12 2006
From: vivacarlie at gmail.com (Nehemiah Dacres)
Date: Tue, 26 Dec 2006 12:29:12 -0600
Subject: [Pythonmac-SIG] Python and Zeroconf?
In-Reply-To: <6a36e7290612230429g11333c3cx384882cf133c072@mail.gmail.com>
References: <1166868056.6695.3.camel@localhost>
	<6a36e7290612230429g11333c3cx384882cf133c072@mail.gmail.com>
Message-ID: <65fadfc30612261029s28f37b33yb0c69629926c1040@mail.gmail.com>

his problem might be how to make
fat<http://en.wikipedia.org/wiki/Fat_binary>python byte code, or
rather, universal binaries with python.

On 12/23/06, Bob Ippolito <bob at redivi.com> wrote:
>
> On 12/23/06, Tim Ansell <mithro at mithis.com> wrote:
> > Hello,
> >
> > My game is using ZeroConf to find local game servers. Under Linux I can
> > use avahi or pyZeroConf to locate the servers (depending if avahi is
> > installed or not).
> >
> > However as Mac OS X has it's own mDNS server I can't use either of these
> > options. Is there a way to interface with the Mac's Zeroconf (Bonjour)?
> >
> > I have found
> >
> http://www-unix.mcs.anl.gov/fl/research/accessgrid/bonjour-py/bonjour-py.htmlhowever there doesn't seem to be Universal Python binaries which I need to
> build binary packages.
>
> Looks like source code. Compiles here. What's the problem?
>
> -bob
> _______________________________________________
> Pythonmac-SIG maillist  -  Pythonmac-SIG at python.org
> http://mail.python.org/mailman/listinfo/pythonmac-sig
>



-- 

"lalalalala! it's not broken because I can use it"

http://linux.slashdot.org/comments.pl?sid=194281&threshold=1&commentsort=0&mode=thread&cid=15927703
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/pythonmac-sig/attachments/20061226/d33b69ce/attachment.html 

From Jack.Jansen at cwi.nl  Wed Dec 27 00:23:54 2006
From: Jack.Jansen at cwi.nl (Jack Jansen)
Date: Wed, 27 Dec 2006 00:23:54 +0100
Subject: [Pythonmac-SIG] Python and Zeroconf?
In-Reply-To: <65fadfc30612261029s28f37b33yb0c69629926c1040@mail.gmail.com>
References: <1166868056.6695.3.camel@localhost>
	<6a36e7290612230429g11333c3cx384882cf133c072@mail.gmail.com>
	<65fadfc30612261029s28f37b33yb0c69629926c1040@mail.gmail.com>
Message-ID: <CA9BF3D3-40BE-4E25-93E1-6EABDDAB560F@cwi.nl>


On  26-Dec-2006, at 19:29 , Nehemiah Dacres wrote:

> his problem might be how to make fat python byte code, or rather,  
> universal binaries with python.

Python byte code is architecture-independent[*], so that shouldn't be  
a problem.

[*] I think there was a problem with portability between machines  
with 32 and 64 bit integers, but I recall that that has been solved  
also. And at least it isn't a problem for ppc/intel mac portability.
--
Jack Jansen, <Jack.Jansen at cwi.nl>, http://www.cwi.nl/~jack
If I can't dance I don't want to be part of your revolution -- Emma  
Goldman


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/pythonmac-sig/attachments/20061227/c468255c/attachment.htm 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 2255 bytes
Desc: not available
Url : http://mail.python.org/pipermail/pythonmac-sig/attachments/20061227/c468255c/attachment.bin 

From uche at ogbuji.net  Fri Dec 29 20:23:26 2006
From: uche at ogbuji.net (Uche Ogbuji)
Date: Fri, 29 Dec 2006 12:23:26 -0700
Subject: [Pythonmac-SIG] Hello, and Wiki vandalism
Message-ID: <45956B2E.6070108@ogbuji.net>

Hi all,

I just switched to a MacBook Pro for my primary dev box, so I guess I'll 
be hanging out here.  I was trying to get started and went to the WIki, 
the FAQ in particular

http://pythonmac.org/wiki/FAQ

The page was spammed/vandalized on the 18th.  I checked recent changes 
and a lot of pages got this treatment in mid-December,.  I'm surprised 
they haven't been fixed, considering the importance of these pages.  I 
know it's probably just a lack of volunteer time but I also know 
MoinMoin has some serviceable anti-spam features.  I tried to fix the 
FAQ page myself reverting to the pre-spam version, but there seems to be 
some bug where preview shows the fixed version, but when I submit, it 
seems to be re-submitting the spam version.  Really strange.  Leaves me 
wondering what's the status of admin on the wiki.

Thanks, and happy hols, all.

-- 
Uche Ogbuji                               Work: The Kadomo Group, Inc.
http://uche.ogbuji.net                    http://kadomo.com
http://copia.ogbuji.net                   Lead dev at http://4Suite.org
Articles: http://uche.ogbuji.net/tech/publications/


From skip at pobox.com  Fri Dec 29 20:49:25 2006
From: skip at pobox.com (skip at pobox.com)
Date: Fri, 29 Dec 2006 13:49:25 -0600
Subject: [Pythonmac-SIG] Hello, and Wiki vandalism
In-Reply-To: <45956B2E.6070108@ogbuji.net>
References: <45956B2E.6070108@ogbuji.net>
Message-ID: <17813.28997.621882.197196@montanaro.dyndns.org>


    Uche> http://pythonmac.org/wiki/FAQ

    Uche> The page was spammed/vandalized on the 18th.  I checked recent
    Uche> changes and a lot of pages got this treatment in mid-December,.
    Uche> I'm surprised they haven't been fixed, considering the importance
    Uche> of these pages.  I know it's probably just a lack of volunteer
    Uche> time but I also know MoinMoin has some serviceable anti-spam
    Uche> features.  I tried to fix the FAQ page myself reverting to the
    Uche> pre-spam version, but there seems to be some bug where preview
    Uche> shows the fixed version, but when I submit, it seems to be
    Uche> re-submitting the spam version.  Really strange.  Leaves me
    Uche> wondering what's the status of admin on the wiki.

I just set up a login.  I don't even get a revert link with I click the info
link.

Skip

From bob at redivi.com  Sat Dec 30 04:35:26 2006
From: bob at redivi.com (Bob Ippolito)
Date: Fri, 29 Dec 2006 19:35:26 -0800
Subject: [Pythonmac-SIG] Hello, and Wiki vandalism
In-Reply-To: <45956B2E.6070108@ogbuji.net>
References: <45956B2E.6070108@ogbuji.net>
Message-ID: <6a36e7290612291935g3f69e804w19b32b121c3dd372@mail.gmail.com>

On 12/29/06, Uche Ogbuji <uche at ogbuji.net> wrote:
> Hi all,
>
> I just switched to a MacBook Pro for my primary dev box, so I guess I'll
> be hanging out here.  I was trying to get started and went to the WIki,
> the FAQ in particular
>
> http://pythonmac.org/wiki/FAQ
>
> The page was spammed/vandalized on the 18th.  I checked recent changes
> and a lot of pages got this treatment in mid-December,.  I'm surprised
> they haven't been fixed, considering the importance of these pages.  I
> know it's probably just a lack of volunteer time but I also know
> MoinMoin has some serviceable anti-spam features.  I tried to fix the
> FAQ page myself reverting to the pre-spam version, but there seems to be
> some bug where preview shows the fixed version, but when I submit, it
> seems to be re-submitting the spam version.  Really strange.  Leaves me
> wondering what's the status of admin on the wiki.
>
> Thanks, and happy hols, all.
>

The version of MoinMoin that's running on doesn't have anti-spam
features. It's OLD. If someone with MoinMoin experience wants to take
over managing the wiki I'd be all for that, but I can't set that up
until after Jan 3.

I've been away from the country all month (and am still away), and in
general I don't have much time for the wiki. I definitely can't do
anything about it until next year because I'm in Shanghai and the
earthquake in Taipei basically took out the route between China and
the US (~80% packet loss). The ISPs out here aren't smart enough or
just don't want to pay extra peering costs so it doesn't seem like
they're going to route around the problems. The undersea cables won't
be fixed for weeks.

On top of all that my MacBook Pro's power supply melted and they
charge nearly twice what it costs in the US for one out here
(infinitely more considering a replacement would be free at home), so
it's hard to get at my SSH keys right now... even if I could make a
working TCP connection to the states (if only Google had a proxy).

-bob

From uche at ogbuji.net  Sun Dec 31 01:21:20 2006
From: uche at ogbuji.net (Uche Ogbuji)
Date: Sat, 30 Dec 2006 17:21:20 -0700
Subject: [Pythonmac-SIG] Hello, and Wiki vandalism
In-Reply-To: <6a36e7290612291935g3f69e804w19b32b121c3dd372@mail.gmail.com>
References: <45956B2E.6070108@ogbuji.net>
	<6a36e7290612291935g3f69e804w19b32b121c3dd372@mail.gmail.com>
Message-ID: <45970280.50107@ogbuji.net>

Bob Ippolito wrote:
> On 12/29/06, Uche Ogbuji <uche at ogbuji.net> wrote:
>> Hi all,
>>
>> I just switched to a MacBook Pro for my primary dev box, so I guess I'll
>> be hanging out here.  I was trying to get started and went to the WIki,
>> the FAQ in particular
>>
>> http://pythonmac.org/wiki/FAQ
>>
>> The page was spammed/vandalized on the 18th.  I checked recent changes
>> and a lot of pages got this treatment in mid-December,.  I'm surprised
>> they haven't been fixed, considering the importance of these pages.  I
>> know it's probably just a lack of volunteer time but I also know
>> MoinMoin has some serviceable anti-spam features.  I tried to fix the
>> FAQ page myself reverting to the pre-spam version, but there seems to be
>> some bug where preview shows the fixed version, but when I submit, it
>> seems to be re-submitting the spam version.  Really strange.  Leaves me
>> wondering what's the status of admin on the wiki.
>>
>> Thanks, and happy hols, all.
>>
>
> The version of MoinMoin that's running on doesn't have anti-spam
> features. It's OLD. If someone with MoinMoin experience wants to take
> over managing the wiki I'd be all for that, but I can't set that up
> until after Jan 3.
>
> I've been away from the country all month (and am still away), and in
> general I don't have much time for the wiki. I definitely can't do
> anything about it until next year because I'm in Shanghai and the
> earthquake in Taipei basically took out the route between China and
> the US (~80% packet loss). The ISPs out here aren't smart enough or
> just don't want to pay extra peering costs so it doesn't seem like
> they're going to route around the problems. The undersea cables won't
> be fixed for weeks.
>
> On top of all that my MacBook Pro's power supply melted and they
> charge nearly twice what it costs in the US for one out here
> (infinitely more considering a replacement would be free at home), so
> it's hard to get at my SSH keys right now... even if I could make a
> working TCP connection to the states (if only Google had a proxy).

Ouch.  Sorry to hear of all those problems (I had heard of the undersea 
cable damage issue).  I really have no cycles available to admin the 
Wiki, but I hope someone is able to step forward.

I tried to do my little part by at least documenting what I learned and 
what worked for me:

http://copia.ogbuji.net/blog/2006-12-30/First_day_

-- 
Uche Ogbuji                               Work: The Kadomo Group, Inc.
http://uche.ogbuji.net                    http://kadomo.com
http://copia.ogbuji.net                   Lead dev at http://4Suite.org
Articles: http://uche.ogbuji.net/tech/publications/


From skip at pobox.com  Sun Dec 31 06:58:56 2006
From: skip at pobox.com (skip at pobox.com)
Date: Sat, 30 Dec 2006 23:58:56 -0600
Subject: [Pythonmac-SIG] So what can we do to clean up the wiki?
Message-ID: <17815.20896.33046.900341@montanaro.dyndns.org>


I realize Bob's kind of hamstrung at the moment.  I set up my login to the
wiki to monitor all changes.  I went through and fixed what I could.
Without revert or DeletePage actions it's a bit tough.  In addition, when I
went to restore the FAQ and DrawBot content it told me "too many links" and
wouldn't let me replace the crap with the correct content!  I had to do them
in several edits...

If someone knows how to delete pages in that old version of Moin (SystemInfo
barfs with a MemoryError, so I can't tell how old it is), please search for
"junk" or "spam" in the text of all pages.  Many of the pages that are
returned should just be deleted.

Skip