Set default scalars for filters

[MatthewFlamm]

Overview

Followup to #2433, this time for scalars.

Details

Adding data to DataSets automatically makes them active, but it is still possible that there are no active scalars, e.g. if there is no data at all

[MatthewFlamm]

I will keep this as draft until it is used in all the filters I can find.

[codecov]

Codecov Report

Merging #2474 (d4b7025) into main (87ea7d5) will decrease coverage by 0.02%.
The diff coverage is 92.68%.

@@            Coverage Diff             @@
##             main    #2474      +/-   ##
==========================================
- Coverage   93.72%   93.69%   -0.03%     
==========================================
  Files          75       75              
  Lines       16120    16159      +39     
==========================================
+ Hits        15108    15140      +32     
- Misses       1012     1019       +7     

[adeak]

Since you're already working in this ballpark, I got this warning locally on main:

tests/test_demos.py::test_plot_glyphs
tests/plotting/test_plotting.py::test_plot_compare_four
  /home/adeak/pyvista/pyvista/core/filters/data_set.py:1999: UserWarning: No vector-like data to use for orient. orient will be set to False.
    warnings.warn("No vector-like data to use for orient. orient will be set to False.")

Probably just a matter of a manual False flag, but I haven't been able to investigate yet.

[akaszynski]

tests/test_demos.py::test_plot_glyphs
tests/plotting/test_plotting.py::test_plot_compare_four
/home/adeak/pyvista/pyvista/core/filters/data_set.py:1999: UserWarning: No vector-like data to use for orient. orient will be set to False.
warnings.warn("No vector-like data to use for orient. orient will be set to False.")

Should be fixed in #2460. I'm tempted to merge that so we avoid duplication.

[MatthewFlamm]

It is weird to fix in #2460, I agree we don't need two semi related PRs that fix this.

[akaszynski]

It is weird to fix in #2460, I agree we don't need two semi related PRs that fix this.

Agreed. It was bothering me, but I should have split it up.

[adeak]

OK, so I think what @akaszynski fixed in #2460 was a doc build warning coming from https://github.com/pyvista/pyvista/pull/2460/files#diff-355952f3009b3d15d0d85ebb510556c3db59722b98c0d5c7d57abdb2b6ea5363L454 .

What I said was a test run, somewhere in tests/. So there's still something to fix here.

Ah, no, I missed the changed tests too in https://github.com/pyvista/pyvista/pull/2460/files#diff-14cb3a563d7f5dd0da9f56abbb5a24503d19670735bf67d24036f9524529caaeL1520

Might be all good, sorry for the noise.

[MatthewFlamm]

This is ready for review.

When adding all the filters I could find, I noticed that some filters use SetInputArrayToProcess and some filters leave it to the vtk algorithm to use the active scalars by default. It would be good to eventually align on a preferred usage. But, this is out of scope for this PR.

[akaszynski]

I added one final test, but otherwise this is good to go.