Proof of Concept: override_aberrations function for WFIRST (Roman)

[robelgeda]

There was an urgent request to make overriding the WFIRST/Roman detector aberrations data possible. This was initially planned to be a part of the WFIRST/Roman prism and grsim implementation. The current plan is to make this functionality available via @robelgeda's override_detector branch until the WFIRST/Roman prism and grsim implementation is complete. This PR gives an overview of what is being provided in that branch which is a temporary means to satisfy the feature request. This PR may be closed when the initial prism and grism PR is opened.

This branch will be closed once the prism and grsim PR is opened.

[mperrin]

Suggestion - let's rename this to override_aberrations rather than override_detector. I think that will make it significantly clearer what is being overridden.

[robelgeda]

Sounds good to me, I have updated the code and PR to reflect this. Only thing I left is the branch name. Thank you for the fast reveiw!

[codecov]

Codecov Report

Merging #382 into develop will decrease coverage by 0.00%.
The diff coverage is 42.85%.

@@             Coverage Diff             @@
##           develop     #382      +/-   ##
===========================================
- Coverage    47.36%   47.35%   -0.01%     
===========================================
  Files           13       13              
  Lines         5559     5566       +7     
===========================================
+ Hits          2633     2636       +3     
- Misses        2926     2930       +4     

Impacted Files	Coverage Δ
webbpsf/wfirst.py	89.03% <42.85%> (-0.74%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 96537c4...a6cd987. Read the comment docs.

[mperrin]

@robelgeda you wrote above

The current plan is to make this functionality available via @robelgeda's override_detector branch until the WFIRST/Roman prism and grsim implementation is complete.

Do you mean/intend that we'll leave this PR open and not merge to develop, so that people who want this functionality should check out the branch from your repo? Or do you intend that we should merge this functionality into develop, at least as a temporary measure while you work on the prism/grism branch? Either approach seems workable, so I'm just asking to make sure I understand this plan. :-)

[robelgeda]

I think both ways work. I created this PR so its easy to compare the changes I made in my branch (and keep track of them) but I personally believe it is better for this to stay in a PR until the other PR is opened. This is because this function may change when we implement the prism and grism. That way we can distinguish what is official and what is temporary. It also doesn't help that we don't have tests for this change.

[mperrin]

Great, sounds good. Then with that agreed, I'll mark this as a 'draft' PR to indicate that we don't plan or intend to merge it now.

[mperrin]

@robelgeda Just to confirm, this PR is no longer needed and is superseded by #416, is that correct?

[robelgeda]

Yes! We can finally close this