On Wed, Jan 2, 2013 at 10:16 AM, Salomon Turgman Cohen
<[email protected]>wrote:

> Dear list,
>     I am having an issue running the tests for fipy. The end of the output
> when issuing fipy.test() is :
>

Which version of FiPy?


> ...
>> ...
>> mayavi version 4.1.0
>> gmsh version 2.6.1
>> To exit: use 'exit', 'quit', or Ctrl-D.An exception has occurred, use %tb
>> to see the full traceback.
>>
>> SystemExit: error: Invalid argument
>>
>
>    If I give %tb in ipython I get:
>
> SystemExit                                Traceback (most recent call last)
> <ipython-input-2-b541bb678baa> in <module>()
> ----> 1 fipy.test()
>
> C:\Enthought\lib\site-packages\fipy\__init__.pyc in test(*args)
>     168         import shutil
>     169         shutil.rmtree(tmpDir)
> --> 170         raise exitErr
>
> SystemExit: error: Invalid argument
>
>  I am running ipython 0.13.1, python 2.7.2, numpy 1.6.2, scipy 0.11.0, and
> matplotlib 1.1.0 in Windows 7. I am starting to suspect that there is
> something wrong with my ipython system but I cannot be sure.
>

I doubt that. It is probably an issue with using "fipy.test()". There are a
few issues with it. For one,  you can't run "fipy.test()" twice in an
ipython session

   http://matforge.org/fipy/ticket/459

What happens when if you do

   $ python -c "import fipy; fipy.test()"

at the command line using the windows command tool rather than using
ipython?


> fipy seems to be working OK when running the 1D diffusion example.
>

"fipy.test()" has only recently been implemented so that the tests can be
run on installed versions of FiPy. Generally we use "python setup.py test"
to run the tests, but you have an installed version of FiPy so that won't
help you.

Cheers

-- 
Daniel Wheeler
_______________________________________________
fipy mailing list
[email protected]
http://www.ctcms.nist.gov/fipy
  [ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]
  • [no subject] Salomon Turgman Cohen
    • Re: Daniel Wheeler
      • Re: Salomon Turgman Cohen
        • Re: Daniel Wheeler

Reply via email to