On Wed, May 26, 2010 at 9:34 PM, Jaakko Salli <[email protected]> wrote: > On 26.5.2010 11:15, Sean Gillies wrote: >> I've asked Stefano to check if this fixes the problem he encountered >> in running the tests. If it does, I propose we go immediately to a >> final release. >> > > http://personal.inet.fi/private/jmsalli/Shapely-1.2rc2.win32.exe > > Also, an experimental 64-bit Windows binary: > > http://personal.inet.fi/private/jmsalli/Shapely-1.2rc2.win-amd64.exe > > Tested with 64-bit Python 2.6.5. Only test that fails is array.txt, > and reason for that is the 64-bit numpy, which may generate long > integers for 'shape' member (e.g. a.shape may be (3L,) instead of > 3,)). > > 64-bit GEOS DLL was built with Visual C++ 9.0. It is possible that > any VC++ incompatibilities that existed in previous versions > of libgeos have been fixed in the latest releases. I can migrate > to VC++ DLL for 32-bit binaries as well, but I don't see it as a > priority since MingW ones have been working so well, > > Regards, > Jaakko
Thanks. I've heard only positive reports about 1.2rc2, so can you make a wininst from r1658? I'll tag svn and upload that to the 1.2 release in the morning if no one has reported anything. Cheers, -- Sean _______________________________________________ Community mailing list [email protected] http://lists.gispython.org/mailman/listinfo/community
