🌐 Mailman 3 March 2006 - Pythonmac-SIG - python.org

Installing numpy 0.9.6 problems
by James Boyle March 27, 2006

March 27, 2006

I'm not sure whether to post this to this list or the numpy list but I'll start here. I am running OX 10.3.9. I installed python 2.4.2 using Universal MacPython 2.4.2.dmg. When I try to install numpy 0.9.6 it fails with the message. . . gcc: _configtest.c gcc: cannot specify -o with -c or -S and multiple compilations gcc: cannot specify -o with -c or -S and multiple compilations failure. . . I am running gcc 3.3. I note that the python 2.4.2 was built using gcc 4.0.1. Could this difference in gcc versions be the problem? At present I am at a loss as to what to try next. --Jim gcc: _configtest.c gcc: cannot specify -o with -c or -S and multiple compilations gcc: cannot specify -o with -c or -S and multiple compilations failure.

6 8

Python 2.4.3c1 universal binaries
by Ronald Oussoren March 27, 2006

March 27, 2006

I have a universal build of the Python 2.4.3c1 release in the public folder of my idisk (at http://homepage.mac.com/ronaldoussoren). This passes the unittests on OSX 10.4 (both intel and ppc). I haven't tested it with 10.3.9 yet. This also solves the problems Nicolas Cole was having w.r.t. resizing terminals and curses applications. This is because this build includes it's own version of the curses library that don't suffer from the bug in Apple's build. Hopefully this also means that the curses module works on 10.3.9. Ronald

1 0

Problems installing Numpy using Universal MacPython
by James Boyle March 27, 2006

March 27, 2006

I posted this to the numpy mailing list - I am re-posting here since it is a more complete description of the situation than my previous post to this list. I am running OX 10.3.9. I installed python 2.4.2 using Universal MacPython 2.4.2.dmg. I am on a PowerPC G4. The Universal build is designed to install on the Mac PPC and Intel machines. When I run the numpy 0.9.6 install it apparently fails on the configure. The whole trace back is at the end of this posting. One thing that I noted was the line: gcc options: '-arch ppc -arch i386 -isysroot /Developer/SDKs/MacOSX10.4u.sdk -fno-strict-aliasing -Wno-long-double -no-cpp-precomp -mno-fused-madd -fno-common -dynamic -DNDEBUG -g' It appears that the configure routines think that I have two architectures - ppc and i386. At my low level of understanding, all I can do is speculate that the Universal install generates some confusion as to the actual cpu. Somewhere numpy is keying on information that is ambiguous in the universal install. I have installed Numeric, matplotlib and other packages without a problem. --Jim The trace back from >python setup.py build: Running from numpy source directory. Warning: not existing path in numpy/distutils: site.cfg F2PY Version 2_2236 blas_opt_info: FOUND: extra_link_args = ['-Wl,-framework', '-Wl,Accelerate'] define_macros = [('NO_ATLAS_INFO', 3)] extra_compile_args = ['-faltivec', '-I/System/Library/Frameworks/vecLib.framework/Headers'] lapack_opt_info: FOUND: extra_link_args = ['-Wl,-framework', '-Wl,Accelerate'] define_macros = [('NO_ATLAS_INFO', 3)] extra_compile_args = ['-faltivec'] running build running config_fc running build_src building py_modules sources creating build creating build/src creating build/src/numpy creating build/src/numpy/distutils building extension "numpy.core.multiarray" sources creating build/src/numpy/core Generating build/src/numpy/core/config.h customize NAGFCompiler customize AbsoftFCompiler customize IbmFCompiler customize GnuFCompiler customize GnuFCompiler customize GnuFCompiler using config gcc options: '-arch ppc -arch i386 -isysroot /Developer/SDKs/MacOSX10.4u.sdk -fno-strict-aliasing -Wno-long-double -no-cpp-precomp -mno-fused-madd -fno-common -dynamic -DNDEBUG -g' compile options: '-I/Library/Frameworks/Python.framework/Versions/2.4/include/python2.4 -Inumpy/core/src -Inumpy/core/include -I/Library/Frameworks/Python.framework/Versions/2.4/include/python2.4 -c' gcc: _configtest.c gcc: cannot specify -o with -c or -S and multiple compilations gcc: cannot specify -o with -c or -S and multiple compilations failure. removing: _configtest.c _configtest.o Traceback (most recent call last): File "setup.py", line 76, in ? setup_package() File "setup.py", line 69, in setup_package setup( **config.todict() ) File "/Volumes/IBM1/macPython/numpy-0.9.6/numpy/distutils/core.py", line 85, in setup return old_setup(**new_attr) File "/Library/Frameworks/Python.framework/Versions/2.4//lib/python2.4/ distutils/core.py", line 149, in setup dist.run_commands() File "/Library/Frameworks/Python.framework/Versions/2.4//lib/python2.4/ distutils/dist.py", line 946, in run_commands self.run_command(cmd) File "/Library/Frameworks/Python.framework/Versions/2.4//lib/python2.4/ distutils/dist.py", line 966, in run_command cmd_obj.run() File "/Library/Frameworks/Python.framework/Versions/2.4//lib/python2.4/ distutils/command/build.py", line 112, in run self.run_command(cmd_name) File "/Library/Frameworks/Python.framework/Versions/2.4//lib/python2.4/ distutils/cmd.py", line 333, in run_command self.distribution.run_command(command) File "/Library/Frameworks/Python.framework/Versions/2.4//lib/python2.4/ distutils/dist.py", line 966, in run_command cmd_obj.run() File "/Volumes/IBM1/macPython/numpy-0.9.6/numpy/distutils/command/ build_src.py", line 84, in run self.build_sources() File "/Volumes/IBM1/macPython/numpy-0.9.6/numpy/distutils/command/ build_src.py", line 99, in build_sources self.build_extension_sources(ext) File "/Volumes/IBM1/macPython/numpy-0.9.6/numpy/distutils/command/ build_src.py", line 209, in build_extension_sources sources = self.generate_sources(sources, ext) File "/Volumes/IBM1/macPython/numpy-0.9.6/numpy/distutils/command/ build_src.py", line 267, in generate_sources source = func(extension, build_dir) File "numpy/core/setup.py", line 37, in generate_config_h raise "ERROR: Failed to test configuration" ERROR: Failed to test configuration

3 2

python SWIG problem ...
by Darran Edmundson March 26, 2006

March 26, 2006

I've been playing around with using SWIG to generate python bindings on the Mac. Even for the most trivial SWIG/python examples, I keep coming up against the linker not being able to resolve the symbol "PyInstance_Lookup". This is an internal-use function defined in /Library/Frameworks/Python.framework/Headers/ classobject.h: $ grep PyInstance classobject.h } PyInstanceObject; PyAPI_DATA(PyTypeObject) PyClass_Type, PyInstance_Type, PyMethod_Type; #define PyInstance_Check(op) ((op)->ob_type == &PyInstance_Type) PyAPI_FUNC(PyObject *) PyInstance_New(PyObject *, PyObject *, PyAPI_FUNC(PyObject *) PyInstance_NewRaw(PyObject *, PyObject *); PyAPI_FUNC(PyObject *) _PyInstance_Lookup(PyObject *pinst, PyObject *name); If I search the shared libraries (no surprise) it isn't defined ... $ nm /Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/ lib-dynload/*.so | grep PyInstance U PyInstance_New U PyInstance_NewRaw U PyInstance_Type U PyInstance_Type Where else could it be and/or any ideas how to resolve this issue? I've asked on the SWIG mailing list but to no avail. Is there a trick to getting SWIG running on the Mac? FYI, I'm using a G4 Powerbook, OS X 10.4.5, $ which python /sw/bin/python (Fink?) $ python Python 2.3.3 (#1, Jan 6 2005, 15:03:06) [GCC 3.3 20030304 (Apple Computer, Inc. build 1666)] on darwin Type "help", "copyright", "credits" or "license" for more information. >>> As always, any help is much appreciated. Cheers, Darran.

2 1

resizing terminal and curses
by Nicholas Cole March 25, 2006

March 25, 2006

I couldn't find anything in the archives related to this. Using both Apple's python and a build of python2.4.2, there seems to be a bug when using ncurses. curses.newwin() is supposed to give a new window filling the pysical terminal, but if the terminal has been resized this does not seem to happen. Instead, the window seems to be the size that the terminal was when curses was initialised. This is not the case on other platforms, where the window created reflects the new size of the terminal. Is this a known bug? (it seems to affect both xterm and terminal.app) Best wishes, N.

2 2

"mach-o, but wrong architecture"
by Marc Hedlund March 25, 2006

March 25, 2006

Hey, I've seen a few messages about this error, but I'm not clear what the resolution is (if any). I'm using MacOS X 10.4.5, Intel iMac, and the bundled Python 2.3.5. For various reasons I don't want to use Python 2.4 yet. I had been working along just fine using Python and wxPython, until I hit a problem with wxPython today. I realized that I was using an old version of wxPython, 2.5.3.1, and that it didn't support a feature I needed. So I downloaded wxPython 2.6.2.1, along with the TigerPython23Compat.pkg, and installed them both. Now when I try to run python and import wx, I get this error: >>> import wx Traceback (most recent call last): File "<stdin>", line 1, in ? File "//Library/Python/2.3/wx-2.6-mac-unicode/wx/__init__.py", line 42, in ? from wx._core import * File "//Library/Python/2.3/wx-2.6-mac-unicode/wx/_core.py", line 4, in ? import _core_ ImportError: dlopen(/Library/Python/2.3/wx-2.6-mac-unicode/wx/_core_.so, 2): no suitable image found. Did find: /Library/Python/2.3/wx-2.6-mac-unicode/wx/_core_.so: mach-o, but wrong architecture I get that wxPython is built for ppc, and that the version of python on my machine is a Universal binary, and that those aren't playing well together. My question is, can I do anything about it? Is there a way to call python so that it is forced to run under Rosetta? My goal is to run Python 2.3 and wxPython 2.6 on the Intel iMac. Is that possible yet? Thanks much for any help. A more detailed log is below if it helps. -M ---begin--- iMac:~ marc$ python Python 2.3.5 (#1, Dec 25 2005, 07:24:19) [GCC 4.0.1 (Apple Computer, Inc. build 5250)] on darwin Type "help", "copyright", "credits" or "license" for more information. >>> import wxversion >>> print wxversion.getInstalled() ['2.6-mac-unicode', '2.5.3-mac-unicode'] >>> import wx Traceback (most recent call last): File "<stdin>", line 1, in ? File "//Library/Python/2.3/wx-2.6-mac-unicode/wx/__init__.py", line 42, in ? from wx._core import * File "//Library/Python/2.3/wx-2.6-mac-unicode/wx/_core.py", line 4, in ? import _core_ ImportError: dlopen(/Library/Python/2.3/wx-2.6-mac-unicode/wx/_core_.so, 2): no suitable image found. Did find: /Library/Python/2.3/wx-2.6-mac-unicode/wx/_core_.so: mach-o, but wrong architecture >>> ^D iMac:~ marc$ python Python 2.3.5 (#1, Dec 25 2005, 07:24:19) [GCC 4.0.1 (Apple Computer, Inc. build 5250)] on darwin Type "help", "copyright", "credits" or "license" for more information. >>> import wxversion >>> wxversion.select('2.5') >>> import wx >>> print wx.VERSION_STRING 2.5.3.1 >>> ^D iMac:~ marc$ file /Library/Python/2.3/wx-2.6-mac-unicode/wx/_core_.so /Library/Python/2.3/wx-2.6-mac-unicode/wx/_core_.so: Mach-O bundle ppc iMac:~ marc$ file /System/Library/Frameworks/Python.framework/Versions/2.3/bin/python /System/Library/Frameworks/Python.framework/Versions/2.3/bin/python: Mach-O universal binary with 2 architectures /System/Library/Frameworks/Python.framework/Versions/2.3/bin/python (for architecture i386): Mach-O executable i386 /System/Library/Frameworks/Python.framework/Versions/2.3/bin/python (for architecture ppc): Mach-O executable ppc ---end---

2 2

Re: [Pythonmac-SIG] Two appscript problems, plus a stupid question
by has March 25, 2006

March 25, 2006

Andrew Barnert wrote: >In my iTunes scripts, I do this by generating a big >filter: >[...] Come to think of it, you can simplify the OR test a bit, since AND() and OR() can take more than one argument: itunes = app('iTunes') dids = [66, 67, 68, 69, 70] idfilters = [its.database_ID == did for did in dids] idfilter = idfilters[0] if idfilters[1:]: idfilter = idfilter.OR(*idfilters[1:]) print itunes.playlists[1].tracks.filter(idfilter).name() HTH has -- http://freespace.virgin.net/hamish.sanderson/

1 0

Re: [Pythonmac-SIG] Two appscript problems, plus a stupid question
by has March 25, 2006

March 25, 2006

Andrew Barnert wrote: >If this list is the wrong place, I apologize; just let >me know and I'll ask in the right place. Otherwise: Here's fine. >First, whenever I try to get a date through appscript, >at least from iTunes, I get a "CommandError: long int >too large to convert to int". >Second, whenever I ask for a missing value, the result >is something like aem.AEType('gnsm'). Oops, more endian problems. Ta for the heads-up. Fixed and uploaded. >Finally, the stupid question: Is there any easy way to >convert a list of references into a single reference >to the list? (To allow a single IPC call instead of >one per reference.) Nope. There are one or two apps that will accept lists of references (e.g. Finder) as the direct parameter to a command, but that's about it. With stuff like this you're always limited in what you can do by the individual application. Incidentally, the interprocess messaging itself is relative cheap on OS X; the main performance hit tends to be the time taken by applications to evaluate references and perform commands. Lots of opportunities for inefficient algorithms to burn cycles; amalgamating many operations into one simply minimises the amount of waste. >In my iTunes scripts, I do this by generating a big >filter: > > # fast version > dids = [168, 142, 57, 39, 277, ...] > idfilters = [(its.database_ID == did) for did in >dids] > idfilter = idfilters[0] > for f in idfilters[1:]: > idfilter = idfilter.OR(f) > trackrefs = library.file_tracks.filter(idfilter) > names = trackrefs.name() > >Is there an easier way to get all the names in a >single IPC call? Not that I can think of. The obvious thing would be to say: trackrefs = library.file_tracks.filter(its.database_ID.isin(dids)) but iTunes isn't smart enough to understand the isin() in that filter test, so I think you're stuck with your multiple equality tests. Another technique is to grab all the IDs with one command and all names/whatever with another, then filter one against the other in Python; what's quicker will depend on the amount of data involved. >(Also, is there a way I can filter on >the ID instead of having the fetch and keep around the >database_ID for everything?) While iTunes likes to use by-ID specifiers, it makes it damnably hard to get the ID numbers used. Its dictionary doesn't show an 'id' property on any of its classes, although an undocumented one does appear to exist. Feel free to file a bug report on this - it causes problems for AppleScript as well as appscript. To deal with missing terminology in Python, you either use appscript.dump to output the keyword-code translation tables to an external module and patch that by hand, or else drop into aem to build at least part of the reference using raw codes: itunes = app('iTunes') tracksRef = itunes.playlists[1].tracks trackIDsRef = tracksRef.AS_objspec.property('ID ') ids = itunes.get(trackIDsRef) print ids >Here's what I'm doing this for: My other Mac has the >whole library on shuffle, and I hear something and >think, "Wow, what was that song?" So I do this: > > $ ssh emac > $ iwhat `ilast 3` >[...] Nice. HTH has -- http://freespace.virgin.net/hamish.sanderson/

1 0

Two appscript problems, plus a stupid question
by Andrew Barnert March 24, 2006

March 24, 2006

If this list is the wrong place, I apologize; just let me know and I'll ask in the right place. Otherwise: Now that I can use appscript with Python 2.4 on my Intel box, I've been going back over some of my iTunes-controlling scripts, and I've run into two problems and one stupid question. First, whenever I try to get a date through appscript, at least from iTunes, I get a "CommandError: long int too large to convert to int". I'll give the full traceback below. So that my script that should list the last N tracks played in iTunes crashes instead. Second, whenever I ask for a missing value, the result is something like aem.AEType('gnsm'). I have a bunch of scripts that just do things like "if artist:" and now they have to do things like "if isinstance(artist, unicode):" instead. This used to work (on a Mac with ancient versions of appscript and MacPython). Finally, the stupid question: Is there any easy way to convert a list of references into a single reference to the list? (To allow a single IPC call instead of one per reference.) In my iTunes scripts, I do this by generating a big filter: # slow version ids = [281, 1099, 731, 414, 238, ...] tracks = [library.file_tracks.ID(id) for id in ids] names = [track.name() for track in tracks] # fast version dids = [168, 142, 57, 39, 277, ...] idfilters = [(its.database_ID == did) for did in dids] idfilter = idfilters[0] for f in idfilters[1:]: idfilter = idfilter.OR(f) trackrefs = library.file_tracks.filter(idfilter) names = trackrefs.name() Is there an easier way to get all the names in a single IPC call? (Also, is there a way I can filter on the ID instead of having the fetch and keep around the database_ID for everything?) Here's what I'm doing this for: My other Mac has the whole library on shuffle, and I hear something and think, "Wow, what was that song?" So I do this: $ ssh emac $ iwhat `ilast 3` 468: Add N to (X) - Large Number 693: ADULT. - Hairing Impaired 241: T. Raumschmiere - R.Ror $ iwhat -a 693 693: ADULT. - Hairing Impaired (D.U.M.E. 5/6, 2005) 02:36 0/100 $ iplaylist "D.U.M.E." `ibatch 693` $ iplay "D.U.M.E." $ irate 90 ... >From across the room, I figured out which of the recent songs was the one I wanted, made a playlist of all the songs in the same import batch, started playing it, rated the current track (the first one) 4-1/2 stars, etc. (I remembered that I bought D.U.M.E., ripped it, then had to go out and never listened to it....) The ilast and ibatch scripts need dates, the iwhat script needs to understand missing values, and a bunch of the scripts either return a list of IDs (ilast, ibatch, iadd, isearch, ...) or take a list of IDs (iwhat, iplaylist, irate, ...). Hence the questions. Finally, here's the traceback I promised: $ ibatch 142 168 Traceback (most recent call last): File "./icontrol-server", line 117, in ? dates = trackrefs.played_date() File "/Library/Python/2.3/site-packages/appscript/specifier.py", line 356, in __call__ return self.get(*args, **kargs) File "/Library/Python/2.3/site-packages/appscript/specifier.py", line 204, in __call__ raise CommandError(self, (args, kargs), e) appscript.specifier.CommandError: long int too large to convert to int Failed command: app(u'/Applications/iTunes.app').library_playlists[1].file_tracks.filter(((its.database_ID == 168).OR(its.database_ID == 142))).played_date.get()

1 0

egg question
by Markus March 24, 2006

March 24, 2006

Hi! I've installed 2.4.2 Universal on a machine which has 2.4.1 from fink installed. This worked very well. I had to install some packages with 'python setup.py install' but no problem here. Then I wanted sqlobject and I tried 'easy_install -U sqlobject' which told, that its already installed under /sw/bin/ (the fink path) Hmm, ok. I reinstalled the setuptools, but it still ends in /sw/bin. Now I don't know what to do or where to look. thanks for your help markus Homer:/Users/markus/Documents/Python root# python ez_setup.py Downloading http://cheeseshop.python.org/packages/2.4/s/setuptools/ setuptools-0.6a10-py2.4.egg Processing setuptools-0.6a10-py2.4.egg creating /Library/Frameworks/Python.framework/Versions/2.4/lib/ python2.4/site-packages/setuptools-0.6a10-py2.4.egg Extracting setuptools-0.6a10-py2.4.egg to /Library/Frameworks/ Python.framework/Versions/2.4/lib/python2.4/site-packages Adding setuptools 0.6a10 to easy-install.pth file Installing easy_install script to /Library/Frameworks/ Python.framework/Versions/2.4/bin Installed /Library/Frameworks/Python.framework/Versions/2.4/lib/ python2.4/site-packages/setuptools-0.6a10-py2.4.egg Processing dependencies for setuptools==0.6a10 Setuptools version 0.6a10 or greater has been installed. (Run "ez_setup.py -U setuptools" to reinstall or upgrade.) Homer:/Users/markus/Documents/Python root# python ez_setup.py -U setuptools Searching for setuptools Reading http://www.python.org/pypi/setuptools/ Reading http://peak.telecommunity.com/DevCenter/setuptools Best match: setuptools 0.6a10 Processing setuptools-0.6a10-py2.4.egg setuptools 0.6a10 is already the active version in easy-install.pth Installing easy_install script to /Library/Frameworks/ Python.framework/Versions/2.4/bin Using /Library/Frameworks/Python.framework/Versions/2.4/lib/python2.4/ site-packages/setuptools-0.6a10-py2.4.egg Processing dependencies for setuptools Homer:/Users/markus/Documents/Python root# easy_install -U sqlobject Searching for sqlobject Reading http://www.python.org/pypi/sqlobject/ Couldn't find index page for 'sqlobject' (maybe misspelled?) Scanning index of all packages (this may take a while) Reading http://www.python.org/pypi/ Reading http://www.python.org/pypi/SQLObject/0.7.0 Reading http://sqlobject.org Best match: SQLObject 0.7.0 Processing SQLObject-0.7.0-py2.4.egg SQLObject 0.7.0 is already the active version in easy-install.pth Installing sqlobject-admin script to /sw/bin Using /sw/lib/python2.4/site-packages/SQLObject-0.7.0-py2.4.egg Processing dependencies for sqlobject

1 1