Coding 2024-04-07
Okay, I fixed stuff up in MOTR:
- I changed the @pytest.fixture() lines to @pytest.fixture.
- I switched the ExceptionGroup checking to using the trio.testing version for now, and got rid of the exceptiongroup package altogether.
At this point, one thing I need to do is really consider whether I want to go through with the idea of decoupling matrix from artifact through parameterization. This is a tough call to make right now, because the purpose of the input field on the axes is purely as part of the API surface, so whatever I do with it has to make sense to someone working with a motrfile. (Unless I somehow come up with yet another layer of API to lather onto this.)
One possibility would be to write a helper function in the artifact module...
It may be best to simply try to note down all of the relevant relations, and just work out how they all interact.
...
Poking at the code, I think the next big improvement I can make is to have a dedicated module for the PathStr alias. The artifact module also defines some type variables that really should get defined in different modules.
Making those changes would do good things, I think. I'll try and get to work on them tomorrow. For now, I want to get to sleep.
Good night.