BUG: avoid deprecated interface for mutating shape attibutes (uncertainty) by neutrinoceros · Pull Request #19131 · astropy/astropy

neutrinoceros · 2026-01-08T12:12:21Z

Description

ref #19126

By checking this box, the PR author has requested that maintainers do NOT use the "Squash and Merge" button. Maintainers should respect this when possible; however, the final decision is at the discretion of the maintainer that merges the PR.

github-actions · 2026-01-08T12:12:35Z

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.

neutrinoceros · 2026-01-08T13:56:48Z

docs failure are unrelated (#19134)
@pllim, would you be able to force merge this ?

pllim · 2026-01-08T15:47:48Z

I could after uncertainty maintainer approves here. :)

mhvk

This one is a bit trickier -- see inline comment. (Which also means it was useful to split them up, thanks!).

mhvk · 2026-01-08T16:15:14Z

        structured.dtype = new_dtype
        # Get rid of trailing dimension of 1.
-        structured.shape = samples.shape[:-1]
+        structured = np.reshape(structured, samples.shape[:-1])


I think this is one case where it is not unreasonable to just set the shape, since we really just created the array and are full owners of it. If we keep the reshape, we certainly should add copy=False (which I think is available in all versions, not sure), since it is a bug if a copy would need to get made.

Personally, I'd go for using the private _set_shape method here -- we'll have to do the same with the dtype later...

copy=False (which I think is available in all versions, not sure)

the copy kwarg was only introduced in numpy 2.1, so I can't use it without branching.

I'd go for using the private _set_shape method

I meant to ask: is it a method on the ndarray class ? when was it introduced ? my attempt to use it with numpy dev was unsuccesful (got an AttributeError), so I think I'm missing some context.

From the changelog entry: https://github.com/numpy/numpy/pull/29536/files#diff-fdbbec0927064b4f5de50467962cad77f4225ba28c9bfbcb2ef16e73e3acda8e

Directly setting the shape on an array is discouraged, but for cases where it is difficult to work
around, e.g., in __array_finalize__ possible with the private method np.ndarray._set_shape.

So, this would be version dependent...

hum okay maybe I just got confused. I'll try this again with a branch (tomorrow)

neutrinoceros · 2026-01-12T11:36:48Z

trying this again using the semi-private ndarray._set_shape method (conditionally).

…inty)

neutrinoceros · 2026-01-14T10:10:50Z

@mhvk @pllim I think this one should be prioritized now, it's by far the most impactful change needed in this area: this one line is responsible for about 150 tests failing and erroring.

mhvk

Thanks for the reminder: the current change is great. I think we eventually want to move away from setting the shape directly this way, but now we can do it calmly, in due time!

mhvk

This certainly seems right even if it doesn't fix all errors yet (see below), so I think we should get it in.

Note: the logs are now clipped by default; I think we should not use --verbose for pytest, so that we can be sure we always get the final summary and don't have to go to the raw logs to see it.

2026-01-12T11:59:22.2662642Z �[31mFAILED�[0m ../../.tox/py313-test-devdeps/lib/python3.13/site-packages/astropy/utils/masked/tests/test_function_helpers.py::�[1mTestArraySetOps::test_isin[int]�[0m - DeprecationWarning: Setting the shape on a NumPy array has been deprecated in NumPy 2.5.
2026-01-12T11:59:22.2663697Z As an alternative, you can create a new view using np.reshape (with copy=False if needed).
2026-01-12T11:59:22.2665154Z �[31mFAILED�[0m ../../.tox/py313-test-devdeps/lib/python3.13/site-packages/astropy/utils/masked/tests/test_function_helpers.py::�[1mTestArraySetOps::test_isin[float]�[0m - DeprecationWarning: Setting the shape on a NumPy array has been deprecated in NumPy 2.5.
2026-01-12T11:59:25.4150085Z As an alternative, you can create a new view using np.reshape (with copy=False if needed).
2026-01-12T11:59:25.4153645Z �[31mFAILED�[0m ../../.tox/py313-test-devdeps/lib/python3.13/site-packages/astropy/utils/masked/tests/test_function_helpers.py::�[1mTestArraySetOps::test_isin[object]�[0m - DeprecationWarning: Setting the shape on a NumPy array has been deprecated in NumPy 2.5.
2026-01-12T11:59:25.4156000Z As an alternative, you can create a new view using np.reshape (with copy=False if needed).
2026-01-12T11:59:25.4158470Z �[31mFAILED�[0m ../../.tox/py313-test-devdeps/lib/python3.13/site-packages/astropy/utils/masked/tests/test_function_helpers.py::�[1mTestArraySetOps::test_setdiff1d�[0m - DeprecationWarning: Setting the shape on a NumPy array has been deprecated in NumPy 2.5.
2026-01-12T11:59:25.4160369Z As an alternative, you can create a new view using np.reshape (with copy=False if needed).
2026-01-12T11:59:25.4161961Z �[31mFAILED�[0m ../../.tox/py313-test-devdeps/lib/python3.13/site-packages/astropy/utils/masked/tests/test_masked.py::�[1mTestMaskedNDArraySubclassCreation::test_viewing_independent_shape�[0m - DeprecationWarning: Setting the shape on a NumPy array has been deprecated in NumPy 2.5.
2026-01-12T11:59:25.4163169Z As an alternative, you can create a new view using np.reshape (with copy=False if needed).
2026-01-12T11:59:25.4164505Z �[31mFAILED�[0m ../../.tox/py313-test-devdeps/lib/python3.13/site-packages/astropy/utils/masked/tests/test_masked.py::�[1mTestMaskedArrayShaping::test_shape_setting�[0m - DeprecationWarning: Setting the shape on a NumPy array has been deprecated in NumPy 2.5.
2026-01-12T11:59:25.4165852Z As an alternative, you can create a new view using np.reshape (with copy=False if needed).
2026-01-12T11:59:25.4167197Z �[31mFAILED�[0m ../../.tox/py313-test-devdeps/lib/python3.13/site-packages/astropy/utils/masked/tests/test_masked.py::�[1mTestMaskedArrayShaping::test_shape_setting_failure�[0m - DeprecationWarning: Setting the shape on a NumPy array has been deprecated in NumPy 2.5.
2026-01-12T11:59:25.4168507Z As an alternative, you can create a new view using np.reshape (with copy=False if needed).

mhvk · 2026-01-14T14:36:14Z

Feel free to either merge or also fix the remaining masked errors associated with shape setting.

neutrinoceros · 2026-01-14T14:40:39Z

I do prefer merging what we have now. Will open other targeted PRs. Hopefully the number of failures should become manageable very soon !

lumberbot-app · 2026-01-14T14:40:59Z

Owee, I'm MrMeeseeks, Look at me.

There seem to be a conflict, please backport manually. Here are approximate instructions:

Checkout backport branch and update it.

git checkout v7.2.x
git pull

Cherry pick the first parent branch of the this PR on top of the older branch:

git cherry-pick -x -m1 a99d1985cdb8b64810e0700d4af0b38d1955083d

You will likely have some merge/cherry-pick conflict here, fix them and commit:

git commit -am 'Backport PR #19131: BUG: avoid deprecated interface for mutating shape attibutes (uncertainty)'

Push to a named branch:

git push YOURFORK v7.2.x:auto-backport-of-pr-19131-on-v7.2.x

Create a PR against branch v7.2.x, I would have named this PR:

"Backport PR #19131 on branch v7.2.x (BUG: avoid deprecated interface for mutating shape attibutes (uncertainty))"

And apply the correct labels and milestones.

Congratulations — you did some good work! Hopefully your backport PR will be tested by the continuous integration and merged soon!

Remember to remove the Still Needs Manual Backport label once the PR gets merged.

If these instructions are inaccurate, feel free to suggest an improvement.

pllim · 2026-01-14T16:20:32Z

Are you able to do the manual backport part? Thanks!

neutrinoceros · 2026-01-14T16:24:41Z

yes. On it

(cherry picked from commit a99d198)

Backport PR #19131 to branch v7.2.x (BUG: avoid deprecated interface for mutating shape attibutes (uncertainty))

neutrinoceros added no-changelog-entry-needed numpy-dev backport-v7.2.x on-merge: backport to v7.2.x labels Jan 8, 2026

github-actions Bot added the uncertainty label Jan 8, 2026

neutrinoceros mentioned this pull request Jan 8, 2026

BUG: incoming deprecation warnings from numpy 2.5 (dev): direct mutations to array shape attributes are deprecated #19126

Closed

neutrinoceros added the Bug label Jan 8, 2026

neutrinoceros force-pushed the uncertainty/19126 branch from de70cbd to eb0f387 Compare January 8, 2026 12:43

neutrinoceros added this to the v7.2.1 milestone Jan 8, 2026

neutrinoceros marked this pull request as ready for review January 8, 2026 13:56

mhvk reviewed Jan 8, 2026

View reviewed changes

neutrinoceros force-pushed the uncertainty/19126 branch from eb0f387 to 6274f4f Compare January 12, 2026 11:36

neutrinoceros marked this pull request as draft January 12, 2026 11:36

BUG: avoid deprecated interface for mutating shape attibutes (uncerta…

5c22ea8

…inty)

neutrinoceros force-pushed the uncertainty/19126 branch from 6274f4f to 5c22ea8 Compare January 12, 2026 11:47

neutrinoceros changed the title ~~BUG: avoid mutation of shape attibutes (uncertainty)~~ BUG: avoid deprecated interface for mutating shape attibutes (uncertainty) Jan 12, 2026

neutrinoceros marked this pull request as ready for review January 12, 2026 11:47

mhvk approved these changes Jan 14, 2026

View reviewed changes

neutrinoceros merged commit a99d198 into astropy:main Jan 14, 2026
31 of 33 checks passed

lumberbot-app Bot added the Still Needs Manual Backport label Jan 14, 2026

neutrinoceros deleted the uncertainty/19126 branch January 14, 2026 14:41

neutrinoceros mentioned this pull request Jan 14, 2026

Backport PR #19131 to branch v7.2.x (BUG: avoid deprecated interface for mutating shape attibutes (uncertainty)) #19157

Merged

1 task

neutrinoceros added a commit to neutrinoceros/astropy that referenced this pull request Jan 14, 2026

Merge pull request astropy#19131 from neutrinoceros/uncertainty/19126

19b2a09

(cherry picked from commit a99d198)

pllim added a commit that referenced this pull request Jan 14, 2026

Merge pull request #19157 from neutrinoceros/bp/19131

2eaa495

Backport PR #19131 to branch v7.2.x (BUG: avoid deprecated interface for mutating shape attibutes (uncertainty))

pllim removed the Still Needs Manual Backport label Jan 14, 2026

osasisorae mentioned this pull request Jan 18, 2026

Fix NumPy 2.5 shape mutation deprecation warnings #19179

Closed

7 tasks done

Conversation

neutrinoceros commented Jan 8, 2026

Description

Uh oh!

github-actions Bot commented Jan 8, 2026

Uh oh!

neutrinoceros commented Jan 8, 2026

Uh oh!

pllim commented Jan 8, 2026

Uh oh!

mhvk left a comment

Choose a reason for hiding this comment

Uh oh!

mhvk Jan 8, 2026

Choose a reason for hiding this comment

Uh oh!

neutrinoceros Jan 8, 2026

Choose a reason for hiding this comment

Uh oh!

neutrinoceros Jan 8, 2026

Choose a reason for hiding this comment

Uh oh!

mhvk Jan 8, 2026

Choose a reason for hiding this comment

Uh oh!

neutrinoceros Jan 8, 2026

Choose a reason for hiding this comment

Uh oh!

neutrinoceros commented Jan 12, 2026

Uh oh!

neutrinoceros commented Jan 14, 2026

Uh oh!

mhvk left a comment

Choose a reason for hiding this comment

Uh oh!

mhvk left a comment

Choose a reason for hiding this comment

Uh oh!

mhvk commented Jan 14, 2026

Uh oh!

neutrinoceros commented Jan 14, 2026

Uh oh!

Uh oh!

lumberbot-app Bot commented Jan 14, 2026

Uh oh!

pllim commented Jan 14, 2026

Uh oh!

neutrinoceros commented Jan 14, 2026

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants