Get ACA penalty and planning limits from chandra_models

[taldcroft]

Description

This gets the ACA penalty and planning limits from chandra_models. It uses the module-level __getattr__ to make these characteristics be lazily defined to prevent import-time issues.

The current code has been tested using a local version of chandra_models in the SKA data that has a tagged "release" from sot/chandra_models#74 as the default model version. Ideally we should get a new version of chandra_models 3.35.1 promoted that includes only the ACA model update.

Testing

• Passes unit tests on MacOS
• Functional testing

[jeanconn]

Regarding "The current code is using a local version of chandra_models that has a tagged "release" from sot/chandra_models#74 as the default model version.", do you just mean that in your testing you have a SKA set with a nonstandard version of chandra_models? Or is there something in the code that I'm missing?

[taldcroft]

do you just mean that in your testing you have a SKA set with a nonstandard version of chandra_models?

No, I changed my local "flight" Ska data version of chandra_models for testing purposes. Since (1) chandra_models is not currently used in any flight tools, and (2) the change is innocuous anyway, I was not concerned about this. My local mod will get wiped out the next time I do a ska_sync.

[jeanconn]

I'm not sure how that is a "no"?

[taldcroft]

I interpreted a "SKA set" to mean changing my SKA env var to a different value where some test Ska data were located.

[jeanconn]

Gotcha. Your SKA is set to a directory that has a nonstandard chandra_models was my point. Though in your strategy you just used the one that you generally use. Either way, there's nothing funny in the current PR/code to pick up a different version, it is just the test environment.

[jeanconn]

I like this implementation. Though now that we have a good use case of get_xija_model_spec I'm wondering about the easiest way to specify a nonstandard location of the chandra_models repo for testing. Env vars are annoying but maybe it is worth one?