Moose Server unit tests enforce unrelated things about moose

[GiudGiud]

@brandonlangley

[loganharbour]

@brandonlangley can you generalize the test to allow for new params like we've done with some of the other tests?

[GiudGiud]

I ll address the MooseMesh parameter one for now. Thanks!

[lindsayad]

I thought we had an issue for this but I'm too lazy to search rn

[GiudGiud]

ah yes I opened #26033 then we closed it after two fixes. I didnt realize there were more issues at the time

[brandonlangley]

@loganharbour - I just updated the MooseServerTest in #27924 so it will no longer fail when new input syntax is added.

@GiudGiud - Depending on if my #27924 or your #27914 is merged first, one of us will resolve a conflict in that test file.

[GiudGiud]

This is still a problem.
The test still enforces the mesh parameter descriptions
see this failure
https://civet.inl.gov/job/2293872/

[brandonlangley]

@GiudGiud -

But this one is a little different, right? It was not caused by adding any new parameters or blocks.

It was due to the existing add_subdomain_names parameter having its description changed from:

Optional list of subdomain names to be applied to the ids given in add_subdomain_ids...

to now be:

The listed subdomain names will be assumed valid for the mesh...

While still technically unrelated, this will be a much more rare case than adding new input pieces.

I'm not sure how to test the server (that assists MOOSE input) without testing a few input pieces.

But it should rarely happen now and only need a small update when a tested piece gets changed.

[GiudGiud]

But this one is a little different, right? It was not caused by adding any new parameters or blocks.

yes you're right. and in fact as you keep improving the test it will always be something a little different.

I'm not sure how to test the server (that assists MOOSE input) without testing a few input pieces.

maybe it could be using pieces that are not supposed to change? like test objects. Or new objects created within the unit tests.

But it should rarely happen now and only need a small update when a tested piece gets changed.

I did hit this like 1 week after than the second fix so I would not say it's rare yet.
It is a totally small update and takes me like 2 minutes at most. Unfortunately most people are not familiar with our unit tests