Get the ACA thermal model with get_xija_model_spec

[jeanconn]

Description

Get the ACA thermal model with get_xija_model_spec

There's a little duplication here but since the proseco code has the convenient limits functionality, I left it alone.

Testing

• [x] Functional testing - Probably needs more testing but doesn't error-out and makes a ccd temperature plot, so looks like it is getting the right model.

Functional testing

I ran AUG3021A. For the flight outputs, I reran flight starcheck so I wouldn't need to play with the run_start_time stuff.

The new top matter looks like

 ------------  Starcheck 13.11.1.dev10+gd4bae3e    -----------------
 Run on Tue Aug 31 11:43:46 EDT 2021 by jeanconn from fido.cfa.harvard.edu
 Configuration:  Using AGASC at /proj/sot/ska3/flight/data/agasc/proseco_agasc_1p7.h5
 chandra_models version: 3.37

in the HTML and txt.
Flight model plot

Test model plot

Diffs flight vs test are expected

jeanconn-fido> diff -u aug3021a.txt /home/jeanconn/git/starcheck/aug3021a.txt
--- aug3021a.txt    2021-08-31 11:53:40.739284000 -0400
+++ /home/jeanconn/git/starcheck/aug3021a.txt   2021-08-31 11:43:48.161616000 -0400
@@ -1,12 +1,13 @@
- ------------  Starcheck 13.11.0    -----------------
- Run on Tue Aug 31 11:53:38 EDT 2021 by jeanconn from fido.cfa.harvard.edu
+ ------------  Starcheck 13.11.1.dev10+gd4bae3e    -----------------
+ Run on Tue Aug 31 11:43:46 EDT 2021 by jeanconn from fido.cfa.harvard.edu
  Configuration:  Using AGASC at /proj/sot/ska3/flight/data/agasc/proseco_agasc_1p7.h5
+ chandra_models version: 3.37

 Short Term Schedule: AUG3021A

 ------------  PROCESSING FILES  -----------------

-DATA = /proj/sot/ska3/flight/lib/python3.8/site-packages/starcheck/data
+DATA = /proj/sot/ska/jeanproj/git/starcheck/starcheck/data
 Using DOT file /data/mpcrit1/mplogs/2021/AUG3021/oflsa/mps/mdAUG3021A.dot
 Using OR file /data/mpcrit1/mplogs/2021/AUG3021/oflsa/mps/or/AUG3021_A.or
 Using TLR file /data/mpcrit1/mplogs/2021/AUG3021/oflsa/CR242_0101.tlr

[taldcroft]

This looks OK now, but can we put the chandra_models version in the starcheck output? Probably a good idea.

[jeanconn]

You mean in the actual output and not the stdout log? Yeah, I was thinking about that too. This PR seems as good a place as any so I'll do that and ask for another review.

[taldcroft]

Yup, up top so it can be seen in the review.

[taldcroft]

The code looks fine now, but I'd like to see a bit more detail with the functional testing:

• The load name that you ran.
• Copy the top bit of text output (or a screen grab) from the HTML showing the chandra models version
• Screen grab of the thermal prediction for the weekly load (1) new and (2) original flight output. This allows a direct visual comparison / sanity check.
• Do a text diff of the starcheck.txt outputs and confirm only expected diffs.

[taldcroft]

@jeanconn - thanks for the test details. As always, functional testing for a PR gets documented for the record in the top Description box, not in the inline comments.

[jeanconn]

Right, I figured I'd move the comment after but thanks for moving it for me.

[taldcroft]

Just start with the functional testing in the right place, and if changes are needed then iterate.