On Tue, Jul 27, 2010 at 11:20 AM, Nicolas M. Thiery <nicolas.thi...@u-psud.fr> wrote: > On Tue, Jul 27, 2010 at 02:05:21AM -0700, Robert Bradshaw wrote: >> > Option 4 is to fix this defect in Cython. I played around with this >> > the other night and the problem is this: the Python C API has only one >> > slot for both __getattr__ and __getattribute__. If you only overload >> > the former, CPython can't tell, and assumes you overrode the latter. >> > It then creates a pure Python stub in your Python classes which call >> > the __getattribute__ wrapper via Python semantics. I'm thinking if we >> > can surpress or delete that wrapper if only __getattr__ is implemented >> > then it will just inherit the slot. I haven't had a chance to try this >> > out though. >> >> And, catching up on my Cython email, Carl Witty may have essentially >> done this! http://trac.cython.org/cython_trac/ticket/561 > > Carl, Robert, thanks so much for investigating this!
Yes, indeed. This is absolutely fantastic work you guys did! > > Wow a general speed improvement over all non cdef classes will be good :-) > I am looking forward timing results and trying it out! Yes, indeed. And by the way my Cython benchmark was for a non-cdef'd but still Cython class. So everything is consistent. Thanks again Carl, etc.! William > > Cheers, > Nicolas > -- > Nicolas M. Thiéry "Isil" <nthi...@users.sf.net> > http://Nicolas.Thiery.name/ > > -- > You received this message because you are subscribed to the Google Groups > "sage-combinat-devel" group. > To post to this group, send email to sage-combinat-de...@googlegroups.com. > To unsubscribe from this group, send email to > sage-combinat-devel+unsubscr...@googlegroups.com. > For more options, visit this group at > http://groups.google.com/group/sage-combinat-devel?hl=en. > > -- William Stein Professor of Mathematics University of Washington http://wstein.org -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
