Codecov Report

Merging #729 into master will decrease coverage by 0.19%.
The diff coverage is 100%.

@@            Coverage Diff            @@
##           master     #729     +/-   ##
=========================================
- Coverage   51.77%   51.57%   -0.2%     
=========================================
  Files         259      259             
  Lines       23424    23424             
  Branches     3200     3200             
=========================================
- Hits        12127    12081     -46     
- Misses      10509    10558     +49     
+ Partials      788      785      -3

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update b3fc352...e656732. Read the comment docs.

I've spent about half an hour trying to figure out how to get it to the state where there is a VTKNoneArray in there from the Mayavi tests and source with no luck. I'll take some more time trying to remember which use case elicited the error in the first place and try to do some debugging there (but IIRC, it's a pain because it's in a complex interactive Traits-based GUI).

Ahh, sorry to hear that. This is why it is useful to have the test case. :) You can create a VTK dataset perhaps to trigger it without having to go through mlab/mayavi if that helps. I will wait a bit in case the light bulb goes on but if it doesn't, and it seems too hard, I'll see if I can think of a test and add it. Thanks!

You can create a VTK dataset perhaps to trigger it without having to go through mlab/mayavi if that helps.

I've spent another half an hour trying to do this with something like the following in test_core_utils.py:

    def test_get_range_works_for_VTKNoneArray(self):
        id_ = tvtk.ImageData(dimensions=(2, 2, 1), origin=(0, 0, 0),
                             spacing=(1, 1, 1))
        id_.point_data.scalars = dsa.VTKNoneArray()
        dsh = DataSetHelper(id_)
        self.assertEqual(dsh.get_range(), (None, [0, 1]))

But this gives:

E           TypeError: SetScalars argument 1: method requires a VTK object

I cannot figure out how to get the scalars property to be a VTKNoneArray object.

Okay I finally found a test that passes on my system. But it also exposes that there is some updating wackiness in that calling get_range() twice can yield different results. No idea how to fix that one, but at least this does not crash with the ufunc 'isnan' not supported error anymore.

Sorry for the delay, looks good, will merge this and look at your other PR.

Thanks!