BUG: avoid deprecated interface for mutating shape attibutes (io.fits) - #19128

Thank you for your contribution to Astropy! 🌌 This checklist is meant to remind the package maintainers who will review this pull request of some common things to look for.

• Do the proposed changes actually accomplish desired goals?
• Do the proposed changes follow the Astropy coding guidelines?
• Are tests added/updated as required? If so, do they follow the Astropy testing guidelines?
• Are docs added/updated as required? If so, do they follow the Astropy documentation guidelines?
• Is rebase and/or squash necessary? If so, please provide the author with appropriate instructions. Also see instructions for rebase and squash.
• Did the CI pass? If no, are the failures related? If you need to run daily and weekly cron jobs as part of the PR, please apply the "Extra CI" label. Codestyle issues can be fixed by the bot.
• Is a change log needed? If yes, did the change log check pass? If no, add the "no-changelog-entry-needed" label. If this is a manual backport, use the "skip-changelog-checks" label unless special changelog handling is necessary.
• Is this a big PR that makes a "What's new?" entry worthwhile and if so, is (1) a "what's new" entry included in this PR and (2) the "whatsnew-needed" label applied?
• At the time of adding the milestone, if the milestone set requires a backport to release branch(es), apply the appropriate "backport-X.Y.x" label(s) before merge.

I dropped part of the patch that turned out to break backward compat. This PR is just aimed at getting trivial changes out of the way so we can focus discussions where needed, so the remaining patch will get a dedicated issue.

Yeah, as Marten says it changes a bit what happens internally. So if we want to keep the test unchanged we need to use the new _set_data method but then this restricts this part of the test to the new Numpy version. It is probably also interesting to test with .data = ... since this is what users will use and I'm not sur if this is tested elsewhere.

Does it make sense to update:

we need to use the new _set_data method but then this restricts this part of the test to the new Numpy version. It is probably also interesting to test with .data = ... since this is what users will use and I'm not sur if this is tested elsewhere.

Maybe there's a bit of confusion going on here, at least in the form of typing mishaps, because _set_data isn't a thing (the private method we can use is ndarray._set_shape). As a result, I'm not entirely sure what you meant.
I'll push an update with my best guess but let me know if I got it wrong !

@braingram absolutely. I initially left it out intentionally because I was worried the discussion there would be thorny, especially because I wasn't sure others would be okay with resorting to _set_shape, but I think now that it makes total sense there so I just included it here. Thanks for pointing it out again !

@neutrinoceros - Yes _set_shape, sorry. We should test inplace shape modification (with _set_shape for numpy ≥ 1.25) so that the test is unchanged, and could also test (with a new block) that setting data works the same way since I'm not sure if this is tested elsewhere (and this is what users would do).

how about now ?

how about now ?

Looks good, thanks (just a minor typo).

thank you. Fixed

You wanna rebase one more time, just in case?

done

Thanks! No need for wait for OSX.

Down to 46 warnings/failures and looks like coordinates is the next big one. 🤪