🌐 Mailman 3 January 2002 - Pythonmac-SIG - python.org

Fwd: [Pythonmac-SIG] sys.exit() functionality
by Jack Jansen Jan. 31, 2002

Jan. 31, 2002

This discussion started on pythonmac-SIG, but someone suggested that it isn't really a MacPython-specific issue (even though the implementation will be different for MacPython from unix-Python). Any opinions? Begin forwarded message: > From: Martin Miller <mmiller(a)adobe.com> > Date: Wed Jan 30, 2002 08:14:13 PM Europe/Amsterdam > To: pythonmac-sig(a)python.org > Subject: Re: [Pythonmac-SIG] sys.exit() functionality > > On Wed, 30 Jan 2002 15:29:21 +0100, Jack Jansen wrote: >> >> On Tuesday, January 29, 2002, at 08:54 , Jon Bradley wrote: >> >>> hey all, >>> >>> In embedded Python - why does sys.exit() quit out of the application >>> that's >>> embedding the interpreter? Is there any way to trap or >>> disregard this? >>> >>> If a user creates an application with Python and runs it through the >>> embedded interpreter, calling quit or exit on the Python application >>> itself >>> is more than ok, but allowing it to force out of the parent >>> application >>> isn't. >> >> Sounds reasonable. How about a routine PyMac_SetExitFunc() that you >> could call to set your own exit function, (similar to >> PyMac_SetConsoleHandler())? MacPython would then do all it's normal >> cleanup, but at the very end call your routine in stead of exit(). > > With an approach like the above, wouldn't it be better to have a > platform-independent way of defining a custom exit function, > rather than > calling a Mac-only system function -- or is this whole thing only an > issue with MacPython embedding? > > Martin > > _______________________________________________ > Pythonmac-SIG maillist - Pythonmac-SIG(a)python.org > http://mail.python.org/mailman/listinfo/pythonmac-sig > -- - Jack Jansen <Jack.Jansen(a)oratrix.com> http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman -

2 1

modules
by Rob Managan Jan. 31, 2002

Jan. 31, 2002

I have been out of touch with Python changes since about 1.5. Where is the best place and way to build modules I am going to write for my own use? Should I learn to use distutils or just dive back into doing it directly with Codewarrior? Historically I did all this in the Extensions folder and just added what I needed to PYTHON_PATH. My first thought is that this would be good for the development cycle.

2 1

[Pyobjc-dev] pyobjc 0.6.1 release (fwd)
by Steven D. Majewski Jan. 30, 2002

Jan. 30, 2002

---------- Forwarded message ---------- Date: Thu, 31 Jan 2002 02:27:57 -0500 (EST) From: Steven D. Majewski <sdm7g(a)minsky.med.virginia.edu> To: pyobjc-dev(a)lists.sourceforge.net Subject: [Pyobjc-dev] pyobjc 0.6.1 release There is a new release pyobjc-0.6.1 on SourceForge: http://prdownloads.sourceforge.net/pyobjc/pyobjc-0.6.1.tar.gz which should build and work with Python-2.2 (It includes a redefinition to correct a bug in Python-2.2/Include/abstract.h, so you don't need a patched version of Python-2.2 ) It includes a Cocoa package with Foundation.py and AppKit.py wrappers. ( You can now do "from pyobjc.Foundation import *" ) and a HelloWorld.py program in the Examples folder. It has one major bugfix and several smaller ones. Since there is still at least one major bug in this version, it has some additional support for debugging: It builds with -g for gdb debugging by default. If you run Python with '-v' switch, pyobjc will dump a log of message sends to a /tmp file. This is not yet the 'New Improved Future Version that fixes the Type/Class barrier' -- I've run into a few design questions and I thought it worth exploring some of those issues with the current model a bit longer. Is anyone else going to be at Python10 next week? Maybe we can have a BOF for pyobjc or the state of MacOSX Python? -- Steve Majewski _______________________________________________ Pyobjc-dev mailing list Pyobjc-dev(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/pyobjc-dev

1 0

Python 10: MacPython or Mac OSX BOF ?
by Steven Majewski Jan. 30, 2002

Jan. 30, 2002

Who is going to Python 10 ? Do we have enough interest for a MacPython or state-of-Python-on-OSX BOF ? -- Steve Majewski (Who only today found out that he is available to go to at least part of it! Gotta go register now. )

1 0

[Python-Dev] Apple PublicSource license [was: next vs darwin] (fwd)
by Steven Majewski Jan. 30, 2002

Jan. 30, 2002

---------- Forwarded message ---------- Date: Wed, 30 Jan 2002 18:23:57 -0500 (EST) From: Steven Majewski <sdm7g(a)virginia.edu> To: python-dev(a)python.org Subject: [Python-Dev] Apple PublicSource license [was: next vs darwin] I did another version of dynload_darwin that took a >10 line function from the dlcompat/dlopen.c code which uses an undocumented (at least in the man pages -- there's probably comments in the Darwin source code) non-public way around the public/private namespace problem we were having with the previous version. I'm waiting for some folks on pythonmac-sig to test it and report back. I'm guessing that this solves the problem without requiring libdl. However it gets into the possible problem of including another license. Could someone who undestands these issues a bit more than I, look at this: Apple Public Source License: http://www.publicsource.apple.com/apsl/ -- Steve BTW: Here's the magic code I added from dlcompat/dlopen.c: ( On the one hand, it's fairly short and trivial. On the other, I wouldn't have had a clue about this without reading the code! ) /* * NSMakePrivateModulePublic() is not part of the public dyld API so we define * it here. The internal dyld function pointer for * __dyld_NSMakePrivateModulePublic is returned so thats all that maters to get * the functionality need to implement the dlopen() interfaces. */ static enum bool NSMakePrivateModulePublic( NSModule module) { static enum bool (*p)(NSModule module) = NULL; if(p == NULL) _dyld_func_lookup("__dyld_NSMakePrivateModulePublic", (unsigned long *)&p); if(p == NULL){ #ifdef DEBUG printf("_dyld_func_lookup of __dyld_NSMakePrivateModulePublic " "failed\n"); #endif return(FALSE); } return(p(module)); } _______________________________________________ Python-Dev mailing list Python-Dev(a)python.org http://mail.python.org/mailman/listinfo/python-dev

1 0

Re: [Pythonmac-SIG] sys.exit() functionality
by Martin Miller Jan. 30, 2002

Jan. 30, 2002

On Wed, 30 Jan 2002 15:29:21 +0100, Jack Jansen wrote: > > On Tuesday, January 29, 2002, at 08:54 , Jon Bradley wrote: > > > hey all, > > > > In embedded Python - why does sys.exit() quit out of the application > > that's > > embedding the interpreter? Is there any way to trap or disregard this? > > > > If a user creates an application with Python and runs it through the > > embedded interpreter, calling quit or exit on the Python application > > itself > > is more than ok, but allowing it to force out of the parent application > > isn't. > > Sounds reasonable. How about a routine PyMac_SetExitFunc() that you > could call to set your own exit function, (similar to > PyMac_SetConsoleHandler())? MacPython would then do all it's normal > cleanup, but at the very end call your routine in stead of exit(). With an approach like the above, wouldn't it be better to have a platform-independent way of defining a custom exit function, rather than calling a Mac-only system function -- or is this whole thing only an issue with MacPython embedding? Martin

2 1

AE support and FSRef
by Alexandre Parenteau Jan. 28, 2002

Jan. 28, 2002

Hello, Trying to use FSRef with MacPython and AppleEvents, I run into all sort of N/A : 1) There is no typeFSRef in AppleEvents.py 2) there is no pack/unpack in aepack.py for FSRef 3) no AEPutKeyDesc support in _AEModule.c 4) no Alias, as_pathname... In macfsmodule.c 5) no RawFSRef in macfsmodule.c (see aepack.py) Did you plan to add these to MacPython ? Are you interested in a patch ? If so, in which CVS branch are you working ? Another related thing : in GenericEventHandler of _AEModule.c it would help a lot if we had a function like : PyObject *AEDesc_New_FromMacOS(AEDesc *itself) This function would create a python AEDesc object which doesn't call AEDisposeDesc when destroyed (by possibly adding a boolean in AEDescObject). This would avoid us from doing the ugly : requestObject->ob_itself.descriptorType = 'null'; requestObject->ob_itself.dataHandle = NULL; replyObject->ob_itself.descriptorType = 'null'; replyObject->ob_itself.dataHandle = NULL; Py_DECREF(args); ... Which took me some time to understand... If we had that, we could pass safely AEDesc coming from MacOS API to Python. I run in this problem because I'm working for fun on a Python context menu plug in for the finder on OSX. Alex. -- Alexandre Parenteau Computer Scientist Core Technologies, AGM Adobe Systems, Inc. 408-536-6166

1 0

Re: Python Reentrant?
by Russell E Owen Jan. 28, 2002

Jan. 28, 2002

Jon Bradley wrote: >Does anyone know forsure whether or not Python 2.2 is reentrant?...We're embedding Python and will want to be able to have more than one thread >evaluating a Python expression at one time. I'm not expert on this, but I hope others will correct any errors. By the way, you'd probably get a better response on a more general Python discussion list or on the news group. I don't believe Python is reentrant, based on: - Only one copy of the Python interpreter can be running at a given time (at least on certain platforms) - Python has a global interpreter lock. Your simplest solution would probably be to write your application using Python's own threads. If you need to have external code running parts of your application, build Python extensions (not that hard to do, though you will have to be careful with the Python interpreter lock). As an alternative, it is conceivable you could get various threads to call Python in a way that obtains the interpreter lock. I have no idea if this is possible or how to go about it. -- Russell

1 0

Re: [Fwd: Re: [Pythonmac-SIG] OS X import bug?]
by Manoj Plakal Jan. 27, 2002

Jan. 27, 2002

Confirmed a solution to the OS X import bug/feature. The flat namespace used in building extension modules is to blame. I just edited /usr/local/lib/python2.2/config/Makefile and changed both LDSHARED and BLDSHARED to be $(CC) $(LDFLAGS) -bundle -bundle_loader /usr/local/bin/python2.2 -undefined error (disabling the flat namespace and using two-level namespace linking) Now, I can build and import both foo and bar into the same interpreter session without problems. Note that I didn't re-build Python, just edited the Makefile. Not sure if this will have ramifications for other modules ... I think I will submit a bug on Sourceforge asking that flat namespace be disabled. Manoj Manoj Plakal wrote to support(a)wingide.com: > > I got a few responses to my post to > c.l.py and pythonmac-sig about the > OS X import bug. > > OS X 10.1 has two kinds of linking > of loadable modules: flat and > two-level. Two-level is what > is done on Linux (RTLD_LOCAL with > RTLD_GLOBAL not specified) as > well as Windows. However, building > Python releases on OS X enables > the flat namespace option. > > Looks like Python needs to be set > up with the flat namespace option > disabled (when you build your > extension modules). Also, see this > message: > http://mail.python.org/pipermail/pythonmac-sig/2002-January/004707.html > > Manoj > > > -------- Original Message -------- > Subject: Re: [Pythonmac-SIG] OS X import bug? > Date: Fri, 25 Jan 2002 14:46:37 -0500 > From: Marcelinus Prastawa <prastawa(a)cs.unc.edu> > To: Manoj Plakal <terabaap(a)yumpee.org> > > It worked on my machine. > > OS X's dynamic loader is a bit different. The two modules share the same > symbols. If they are built as flat namespace binaries, Python would fail > to build the second one. > > You should remove the "-flat_namespace -undefined suppress" options... If > you are using Mach-O (UNIX) Python you could edit the options in > /usr/local/lib/python2.2/config/Makefile. > > Marcel > > On Thursday, January 24, 2002, at 01:58 , Manoj Plakal wrote: > > > > > Hello, > > > > I've just noticed a weird problem with importing modules > > on MacOS X 10.1 and Python 2.2. > > > > Say I have two modules foo and bar, both of which > > are built with a common file baz.c with some variables > > defined in it. > > > > I make a setup.py which specifies that module foo > > is built from foomodule.c and baz.c, and module bar > > is built from barmodule.c and baz.c. > > > > On Red Hat Linux 7.2 with Python 2.2, this builds fine and I > > can import both foo and bar into the same interpreter session. > > > > On OS X 10.1 with Python 2.2, this builds fine but I can > > only import *ONE* of foo or bar into the same interpreter > > session. E.g., if I import foo first and then bar, > > Python craps out saying "Failure linking new module". > > Similary for bar first and then foo. > > > > This is both with a version built from the 2.2 source > > release, as well as a binary version from Fink > > (fink.sourceforge.net). > > > > I've attached the files which can be used to reproduce > > this: setup.py, foomodule.c, barmodule.c, baz.c. In > > case the attachments don't work, you can grab them > > off the web: http://yumpee.org/python/setup.py > > http://yumpee.org/python/foomodule.c > > http://yumpee.org/python/barmodule.c > > http://yumpee.org/python/baz.c > > > > Is this a bug? Undefined behavior? Feature? > > Is it a fluke that it works on Linux? > > > > If someone else with OS X can reproduce this error, > > (and people on other OSes can confirm that > > it's working), that would be great ... > > > > Manoj > >

4 23

Flat namespace problem
by Jack Jansen Jan. 27, 2002

Jan. 27, 2002

I was going to jump in and look at getting rid of the flat namespace problem when I noticed that if you build a framework based Python (configure --enable-framework) you already get a two-level namespace. Before I go fixing the wrong problem, could someone who has the flat namespace problem please try and build a framework Python, and check that the name collision doesn't happen then? Also, if my understanding is right, this would mean that VTK can't be built with a framework-based Python. At least, not without fiddling LDSHARED in your setup.py. Is that correct? -- - Jack Jansen <Jack.Jansen(a)oratrix.com> http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman -

2 1