Codecov Report

Attention: Patch coverage is 62.16216% with 14 lines in your changes missing coverage. Please review.

Project coverage is 73.77%. Comparing base (d0d1a67) to head (8191748).
Report is 847 commits behind head on main.

@@            Coverage Diff             @@
##             main    #9126      +/-   ##
==========================================
- Coverage   73.77%   73.77%   -0.01%     
==========================================
  Files         372      372              
  Lines       37265    37272       +7     
==========================================
+ Hits        27492    27497       +5     
- Misses       9773     9775       +2     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

I started regression tests here:
https://github.com/spacetelescope/RegressionTests/actions/runs/13145149831

It would be nice to have a unit test or two to cover the new code. Do you have bandwidth to give that a try?

Edit: regtest results look fine. There are a few unrelated errors, but no changes as a result of this PR, as expected. I'll run again when code changes are done.

Good call; I've added in a simple test.

Tagging @jemorrison for a second pair of eyes, if you have a little time to review.

Looks good to me. One tiny comment on doc. The clean_showers module is surprising simple considering what it is trying to remove.

I was surprised at the large number for shower_x_stddev (18.0). Just for my understanding why is this set so large? The bad pixels have been naned out and the 3rd plane of the regions file is use to mark pixels that receive light from the sky and as I understand the shower_low and high reject are also used to further screen pixels.

Conceptually, it's large to enable the routine to reach across the masked-out IFU slices that can be 30+ pixels wide. Anecdotally, I tried a few different sizes and x=18 y=5 looked like it gave the best results visually for my test cases. It's entirely possible that there may be some better set of parameters though.