[SciPy-Dev] ANN: SciPy 0.9.0 beta 1

Pauli Virtanen pav@iki...
Mon Dec 13 16:43:22 CST 2010


On Mon, 13 Dec 2010 14:21:13 -0800, Christoph Gohlke wrote:
[clip]
> Yes, see SWIG bug #3057804 "[Python] wrapper does not compile with
> python 3.2"
> <http://sourceforge.net/tracker/?
func=detail&aid=3057804&group_id=1645&atid=101645>

Possibly addressed by this:

	https://github.com/pv/scipy-work/tree/bug/sparsetools-reswig

If it fixes it, then it's probably the correct solution.
It's also possible that the issue (4) is related to this on Python 
2.7/3.1, since I can't think of much else than the 
PyCObject/PyCapsule issue that would have changed in between.

>>> 3) 32-bit Python 2.6 is the only platform where scipy.test() finishes;
[clip]
>> sc.airye(-1)
> 
> __main__:1: SpecialFunctionWarning: airye:: unknown error
> __main__:1: SpecialFunctionWarning: airye:: unknown error 
> (nan, nan, 0.10399738949694459, 0.59237562642279229)

Then I don't understand why in the test the function gives
(nan, nan, nan, nan)

OTOH, there are some other bugs, where when linked with MKL, some 
non-local effects (indicating memory corruption / fp state stuff) on 
function return values appear.

-- 
Pauli Virtanen



More information about the SciPy-Dev mailing list