[Numpy-discussion] numpy, Py_ssize_t, cython and 64 bits python 2.4

Dag Sverre Seljebotn dagss@student.matnat.uio...
Thu Nov 13 18:25:32 CST 2008


For mailing list archival purposes, I'm posting the conclusion of this 
story: An update to Cython which also works with NumPy/Python2.4/64-bit 
can now be retrieved from http://hg.cython.org/cython

That Mercurial repo contains previous release + bugfixes deemed to be 
very safe, so I recommend using this instead for SciPy etc. until the 
next Cython release.

Dag Sverre

David Cournapeau wrote:
> On Mon, Nov 10, 2008 at 12:38 PM, David Cournapeau <cournape@gmail.com> wrote:
> 
>> This sounds like the right solution, but OTOH, since I am not familiar
>> with the cython codebase at all, I will prepare a patch following the
>> first solution, and I guess you can always improve it later.
> 
> Ok, I made a patch which makes strides/shape to be Py_ssize_t arrays
> instead of npy_intp, and create temporary buffers when
> sizeof(Py_ssize_t) != sizeof(npy_intp_t). I have not tested it
> thouroughly, but it made the tests which failed work on python 2.4 on
> RHEL 64 bits.
> 
> Can I have an account to put the mercurial bundle somewhere ?
> 
> cheers,
> 
> David
> _______________________________________________
> Numpy-discussion mailing list
> Numpy-discussion@scipy.org
> http://projects.scipy.org/mailman/listinfo/numpy-discussion


-- 
Dag Sverre


More information about the Numpy-discussion mailing list