[GeoMechanicsApplication] Improved handling of unit tests that need to run conditionally

[avdg81]

📝 Description

Use GTEST_SKIP rather than not compiling unit tests that need to be run conditionally. So far, several unit tests are skipped when non-debug builds are used.

With these changes, the total number of unit tests that is run is identical for all build types. However, the number of skipped tests may differ.

[rfaasse]

Thanks for picking this up. In principle I agree with the changes, but I'm not too sure if it makes it clear e.g. in the pipelines why they are skipped:

Also locally, you'll always get these skipped test cases at the end (like in the first picture) without a reason being printed. Maybe we should add clearly to the names of the test that they should do something only in debug?

[avdg81]

Thanks for picking this up. In principle I agree with the changes, but I'm not too sure if it makes it clear e.g. in the pipelines why they are skipped:

Also locally, you'll always get these skipped test cases at the end (like in the first picture) without a reason being printed. Maybe we should add clearly to the names of the test that they should do something only in debug?

Thanks a lot for your suggestion. I have renamed the test cases that are potentially skipped, to give a hint about the reason why they were skipped. Does this help in your opinion?

[markelov208]

Hi Anne, thank you for the PR. GTEST_SKIP is better and clear approach. Do we a have a style to name unit tests?

[rfaasse]

Now indeed it's clear why these tests would've been skipped from the logging, thanks for this improvement!

[avdg81]

Thanks for picking this up. In principle I agree with the changes, but I'm not too sure if it makes it clear e.g. in the pipelines why they are skipped:

Also locally, you'll always get these skipped test cases at the end (like in the first picture) without a reason being printed. Maybe we should add clearly to the names of the test that they should do something only in debug?

Thanks a lot for your suggestion. I have renamed the test cases that are potentially skipped, to give a hint about the reason why they were skipped. Does this help in your opinion?

Hi Anne, thank you for the PR. GTEST_SKIP is better and clear approach. Do we a have a style to name unit tests?

Good question. I'm not aware of any naming style for the unit tests, but perhaps we should think of one. What do you think?

[markelov208]

Hi Anne, thank you for the PR. GTEST_SKIP is better and clear approach. Do we a have a style to name unit tests?

Good question. I'm not aware of any naming style for the unit tests, but perhaps we should think of one. What do you think?

Sure, it will be nice to have a certain style. In other places a class name is used and underscore to improve readability. It can be more things.