[Pythonmac-SIG] python 2.6 trunk

[Kees, Christopher E]

I updated to r65653 and it seems to be working OK. There are some test
failures. I saved all the output in case it's of any use.

My problem is now with the numpy module, which builds fine but says it can't
install when cross-compiling.  I apparently had an old version of numpy in
site-packages that must not have been a universal but which was running
without any problems as long as I was running in 32 bit. Anybody had any
luck with a 4-way universal numpy build?

Thanks,
Chris

...
317 tests OK.
10 tests failed:
    test_aepack test_applesingle test_array test_distutils
    test_httpservers test_macostools test_scriptpackages test_signal
    test_site test_socket
33 tests skipped:
    test_al test_bsddb test_bsddb3 test_cd test_cl test_codecmaps_cn
    test_codecmaps_hk test_codecmaps_jp test_codecmaps_kr
    test_codecmaps_tw test_curses test_dl test_epoll test_gdbm test_gl
    test_imageop test_imgfile test_largefile test_linuxaudiodev
    test_normalization test_ossaudiodev test_pep277 test_py3kwarn
    test_socketserver test_startfile test_sunaudiodev test_tcl
    test_timeout test_urllib2net test_urllibnet test_winreg
    test_winsound test_zipfile64
2 skips unexpected on darwin:
    test_tcl test_dl
make: *** [test] Error 1



On 8/12/08 7:43 AM, "Ronald Oussoren" <ronaldoussoren at mac.com> wrote:

> 
> On 12 Aug, 2008, at 14:30, Ronald Oussoren wrote:
>>> 
>> 
>> For some reason configure doesn't seem to have detected that the
>> 10.5 SDK. Are you building on OSX 10.5? If so, could you check
>> pyconfig.h to check if HAVE_MACOS105_SDK is defined? (Which it
>> should be when you're building a 4-way universal binary).
> 
> Never mind, there was a real bug here. The actual #define is called
> HAVE_OSX105_SDK.
> 
> I've fixed this in r65653.
> 
> Ronald
> 
>