[SciPy-dev] scipy.test() segfaults

[Nils Wagner]

On Thu, 8 Oct 2009 13:18:49 -0600
  Charles R Harris <charlesr.harris at gmail.com> wrote:
> On Thu, Oct 8, 2009 at 1:04 PM, Nils Wagner 
><nwagner at iam.uni-stuttgart.de>wrote:
> 
>> On Thu, 8 Oct 2009 13:03:50 -0600
>>   Charles R Harris <charlesr.harris at gmail.com> wrote:
>> > On Thu, Oct 8, 2009 at 12:48 PM, Nils Wagner
>> > <nwagner at iam.uni-stuttgart.de>wrote:
>> >
>> >> On Thu, 8 Oct 2009 12:29:48 -0600
>> >>  Charles R Harris <charlesr.harris at gmail.com> wrote:
>> >> > On Thu, Oct 8, 2009 at 12:10 PM, Nils Wagner
>> >> > <nwagner at iam.uni-stuttgart.de>wrote:
>> >> >
>> >> >> Hi,
>> >> >>
>> >> >> Is this a known issue ?
>> >> >>
>> >> >>
>> >> > Maybe. What release of numpy was scipy compiled
>> >>against?
>> >>
>> >> I am using the latest svn version of numpy and scipy.
>> >> >>> import numpy
>> >> >>> numpy.__version__
>> >> '1.4.0.dev7507'
>> >> >>> import scipy
>> >> >>> scipy.__version__
>> >> '0.8.0.dev5964'
>> >>
>> >>
>> > Did you delete the build directories? Did you compile
>> >numpy first, then
>> > scipy?
>> >
>> Yes and yes.
>>
>>
> OK, I'm seeing this also. It may be related to David's 
>change to
> NeighborhoodIter in numpy r7452.
> 
> Chuck

I have also removed everything under

/home/nwagner/local/lib64/python2.6/site-packages/numpy
/home/nwagner/local/lib64/python2.6/site-packages/scipy

Now scipy.test() passed.

Ran 3866 tests in 85.541s

FAILED (KNOWNFAIL=6, SKIP=17, failures=7)
<nose.result.TextTestResult run=3866 errors=0 failures=7>


Nils