[SciPy-User] Building Scipy for Mac OS X 10.6

[Scott Stephens]

On Fri, Jun 4, 2010 at 7:31 AM, Ralf Gommers
<ralf.gommers at googlemail.com> wrote:
> On Fri, Jun 4, 2010 at 8:16 PM, Scott Stephens <stephens.js at gmail.com>
> wrote:
>> Looks like all of the created files have the right architectures, and
>> nothing in the unit tests fail because of import errors. I'm still not
>> getting a flawless scipy.test() run though (6 errors and 1 failure).
>> I haven't yet had a chance to do a deep dive and figure out what
>> exactly is failing and if it might be related to build problems.
>>
> Can you post the errors and failure? Currently one test gives an error,
> test_lapack_misaligned. Others that have been failing in recent checkouts
> are one lambertw test and several matlab ones, so no need to investigate
> those.

Here they are:

ERROR: test_complex_nonsymmetric_modes
(test_arpack.TestEigenComplexNonSymmetric)
ERROR: test_nonsymmetric_modes (test_arpack.TestEigenNonSymmetric)
ERROR: test_starting_vector (test_arpack.TestEigenNonSymmetric)
ERROR: test_add_function_ordered (test_catalog.TestCatalog)
ERROR: Test persisting a function in the default catalog
(test_add_function_persistent1 from
scipy/weave/tests/test_catalog.py)
ERROR: Shouldn't get a single file from the temp dir.
(test_get_existing_files2 from scipy/weave/tests/test_catalog.py)
FAIL: test_complex_symmetric_modes (test_arpack.TestEigenComplexSymmetric)

This is scipy-0.7.1 from a tarball.  I'm wondering if if some of the
nonsymmetric failures are related to me not installing UMFPACK.  At
some point in my build adventure I got some messages about UMFPACK not
being found, but nothing about it appears in the installation
instructions, so I haven't done anything with it.


-- Scott