[SciPy-dev] Different number of tests

Arnd Baecker arnd.baecker at web.de
Wed Dec 21 05:27:02 CST 2005


Hi Nils,

On Wed, 21 Dec 2005, Nils Wagner wrote:
> Arnd Baecker wrote:
> >On Wed, 21 Dec 2005, Nils Wagner wrote:
> >>Hi all,
> >>
> >>I have installed the latest svn versions of core and scipy on two
> >>different machines.
> >>
> >>On the first box
> >>scipy.test(1,10)
> >>Ran 1372 tests in 8.086s
> >>
> >>FAILED (failures=1)
> >><unittest.TextTestRunner object at 0x47fed68c>
> >>
> >>On the second box
> >>Ran 1230 tests in 7.358s
> >>
> >>FAILED (failures=1)
> >><unittest.TextTestRunner object at 0xb2908c8c>
> >>
> >>What is the reason for the difference in the number of tests (1372
> >>versus 1230) ?
> >
> >if you do scipy.test(1) (or 10) you should see a summary like
> >
> >  Found 4 tests for scipy.io.array_import
> >  Found 128 tests for scipy.linalg.fblas
[...]
> >
> >Comparing these two lists should reveal, where less tests
> >are found.

[...]

>  The main difference is here
>  Found 128 tests for scipy.lib.blas.fblas
> which isn't available on the other machine.

Which still leaves a difference of further 14 tests.
So, did your comparison answer your question well enough?

Best, Arnd






More information about the Scipy-dev mailing list