Fix KFP component without inputs compilation bug

[ptitzler]

Closes #2644

What changes were proposed in this pull request?

• Fix bug
• Add test case to elyra/tests/pipeline/kfp/test_processor_kfp.py (including resource)

How was this pull request tested?

• Manual testing (add this component to a KFP pipeline and run it)
• pytest -v elyra/tests/pipeline/kfp/test_processor_kfp.py

Developer's Certificate of Origin 1.1

   By making a contribution to this project, I certify that:

   (a) The contribution was created in whole or in part by me and I
       have the right to submit it under the Apache License 2.0; or

   (b) The contribution is based upon previous work that, to the best
       of my knowledge, is covered under an appropriate open source
       license and I have the right under that license to submit that
       work with modifications, whether created in whole or in part
       by me, under the same open source license (unless I am
       permitted to submit under a different license), as indicated
       in the file; or

   (c) The contribution was provided directly to me by some other
       person who certified (a), (b) or (c) and I have not modified
       it.

   (d) I understand and agree that this project and the contribution
       are public and that a record of the contribution (including all
       personal information I submit with it, including my sign-off) is
       maintained indefinitely and may be redistributed consistent with
       this project or the open source license(s) involved.

[elyra-bot]

Thanks for making a pull request to Elyra!

To try out this branch on binder, follow this link:

[ptitzler]

The failing test test_directory_based_component_catalog doesn't take into account that unrelated tests might add other resources (like it is done in this PR) and needs to be reworked to assure its original assumptions remain correct over time.

[kiersten-stokes]

The failing test test_directory_based_component_catalog doesn't take into account that unrelated tests might add other resources (like it is done in this PR) and needs to be reworked to assure its original assumptions remain correct over time.

Each time a test uses the component_cache fixture (which the directory-based catalog test does), the component cache instance should be cleared both before and after the test completes (and hence not side-effect other tests). Obviously, that isn't happening here, so I'll keep looking into a lightweight solution.

[kiersten-stokes]

Thoughts on going back to a 'greater than' comparison for the test_directory_based_component_catalog test? We had it this way as of a few commits ago, and it's probably the simplest solution. Since we go on to check for the presence of certain components, I don't think there would be much of a downside to this.

[ptitzler]

Updates on the CI test failures:

When the entire test is run, the two cases fail because the component cache is not empty at the start

$ pytest -v elyra/tests/pipeline/kfp/test_component_parser_kfp.py 

When only the two test cases are run, they succeed because its setup is not tainted:

$ pytest -v elyra/tests/pipeline/kfp/test_component_parser_kfp.py -k "test_directory_based_component_catalog"

[ptitzler]

Thoughts on going back to a 'greater than' comparison for the test_directory_based_component_catalog test? We had it this way as of a few commits ago, and it's probably the simplest solution. Since we go on to check for the presence of certain components, I don't think there would be much of a downside to this.

It'd be possible. The test would have to take an inventory of the existing components in the cache and eliminate duplicates from its catalog directory list for this to work. However, there's also a good chance that the cache already includes all of the pre-selected components, which would render the test result almost useless because it would only validate that adding components from an empty directory doesn't add any components.

[kiersten-stokes]

It'd be possible. The test would have to take an inventory of the existing components in the cache and eliminate duplicates from its catalog directory list for this to work. However, there's also a good chance that the cache already includes all of the pre-selected components, which would render the test result almost useless because it would only validate that adding components from an empty directory doesn't add any components.

Right. And as of the last commit and my latest testing regarding cache cleanup in certain tests, it seems that we have a larger issue on our hands here.

[kevin-bates]

LGTM

[kiersten-stokes]

LGTM!