Requre photutils 3+

[larrybradley]

Resolves JP-nnnn

This PR uses a context manager for locally-scoped changes instead of relying on or potentially mutating a global photutils variable.

CC: @braingram

Tasks

• If you have a specific reviewer in mind, tag them.
• add a build milestone, i.e. Build 12.0 (use the latest build if not sure)
• Does this PR change user-facing code / API? (if not, label with no-changelog-entry-needed)
  • write news fragment(s) in changes/: echo "changed something" > changes/<PR#>.<changetype>.rst (see changelog readme for instructions)
    • if your change breaks step-level or public API (as defined in the docs), also add a changes/<PR#>.breaking.rst news fragment
  • update or add relevant tests
  • update relevant docstrings and / or docs/ page
  • start a regression test and include a link to the running job (click here for instructions)
    • Do truth files need to be updated ("okified")?
      • after the reviewer has approved these changes, run okify_regtests to update the truth files
• if a JIRA ticket exists, make sure it is resolved properly

[braingram]

Thanks! @tapastro mentioned bumping the pin to >=3. Since it's now out, would that be easier (perhaps using the same approach as you added in romancal)?

[larrybradley]

Sure, I can do that. I wasn't sure if jwst was ready for that (romancal needs photutils 3+). I'll update this PR.

[codecov]

Codecov Report

❌ Patch coverage is 90.00000% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 86.48%. Comparing base (c9bd647) to head (cb9f2cb).
⚠️ Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
jwst/extract_1d/ifu.py	0.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #10471      +/-   ##
==========================================
+ Coverage   86.43%   86.48%   +0.05%     
==========================================
  Files         374      374              
  Lines       40276    40217      -59     
==========================================
- Hits        34811    34782      -29     
+ Misses       5465     5435      -30     

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[larrybradley]

I also bumped numpy >= 2.0 and astropy >= 6.1.4. Those are required for photutils >= 3.0.

[larrybradley]

Regression tests:
https://github.com/spacetelescope/RegressionTests/actions/runs/24690043180

[larrybradley]

The oldestdeps CI failures are unrelated -- due to PyparsingDeprecationWarning and ValueError: numpy.dtype size changed.

[melanieclarke]

It looks like there are some significant differences in the regression tests, for the MIRI and NIRCam image3 results, looks like because the source catalog found some extra sources in both cases. Is that expected with the changes here?

Also, we'll need to update the oldest dependencies allowed to get the oldestdeps tests passing.

[larrybradley]

Thanks, @melanieclarke . I updated the min versions of packaging and scikit-image and now the oldestdeps tests pass.

For the regression tests, it appears that the FGS, MIRI, and NIRCam i2d images have differences, which could lead to a slightly different number of sources in the downstream catalogs. Are the differences in the i2d images expected from other recent pipeline changes or dependency updates? For example, there are many differences in the MIRI image (https://github.com/spacetelescope/RegressionTests/actions/runs/24690043180/job/72209316131#step:42:1371), including the WCS (https://github.com/spacetelescope/RegressionTests/actions/runs/24690043180/job/72209316131#step:42:1270).

[melanieclarke]

Thanks, @melanieclarke . I updated the min versions of packaging and scikit-image and now the oldestdeps tests pass.

Thank you!

For the regression tests, it appears that the FGS, MIRI, and NIRCam i2d images have differences, which could lead to a slightly different number of sources in the downstream catalogs. Are the differences in the i2d images expected from other recent pipeline changes or dependency updates? For example, there are many differences in the MIRI image (https://github.com/spacetelescope/RegressionTests/actions/runs/24690043180/job/72209316131#step:42:1371), including the WCS (https://github.com/spacetelescope/RegressionTests/actions/runs/24690043180/job/72209316131#step:42:1270).

Our regression tests are clean with current dependencies, so it's not likely other pipeline changes. I expect a few pixels different from upgrading numpy and scipy, but nothing on the scale of the differences for the regtests on the branch here.

I think the i2d differences are because the tweakreg solution is slightly different in the image3 run, so all the pixels are shifted a little compared to the regtest truth files. So, not directly because of the source catalog differences (sorry I misspoke!), but presumably because the sources chosen in tweakreg are a little different with the changes here.

[larrybradley]

Thanks. I'll need to investigate the tweakreg differences.

[emolter]

Looks good, thanks for handling this! I had just a few small comments.

[larrybradley]

Regtests: https://github.com/spacetelescope/RegressionTests/actions/runs/26308022406

[larrybradley]

@melanieclarke My last update fixed the image differences.

The failing regtests are due to the orientation angles now being wrapped to [0, 360) degrees. They will need to be okified.

[melanieclarke]

Looks like all comments are addressed and tests are passing. Thanks for getting this cleaned up!