BUG: avoid deprecation warnings from numpy.char.chararray when checking types at runtime by neutrinoceros · Pull Request #19217 · astropy/astropy

neutrinoceros · 2026-01-26T09:02:36Z

Description

ref #19216
fixes the dominant failure mode, and should be backportable

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 · 2026-01-26T09:02:47Z

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 · 2026-01-26T09:04:57Z

            except Exception:
                try:  # then try to convert it to a strings array
                    itemsize = int(recformat[1:])
-                    array = chararray.array(array, itemsize=itemsize)


this is adjacent and not technically part of the fix, so I can revert this part if prefered. My reasoning is that eliminating chararray as a namespace alias removes noise when grepping for remaining occurences of this class

neutrinoceros · 2026-01-26T09:06:30Z



+def is_chararray(obj, /) -> bool:
+    if not hasattr(np, "char"):


numpy dev's doc currently says that the entire module may go away in the future, even though it's not deprecated yet, so I'm using this guard clause for future-proofing

neutrinoceros · 2026-01-26T09:09:31Z

allowed_failures is expected to pass this time around even though I intentionally didn't fix all of the warnings (see #19216), because the deprecation hasn't landed in nightlies yet.

neutrinoceros · 2026-01-26T09:51:37Z

-    if field.dtype.char in ("S", "U") and not isinstance(field, chararray.chararray):
-        field = field.view(chararray.chararray)
+    if field.dtype.char in ("S", "U") and not is_chararray(field):
+        field = field.view(np.char.chararray)


it seems that this single line is responsible for every remaining failures after this patch. I'm not sure whether it's okay to backport the necessary change, as it could be breaking. TBD (possibly in #19216)

Can we jail this line in a warnings.filterwarnings("ignore", category=DeprecationWarning) via context manager like you did in numpycompat ? That would get rid of the remaining warnings until we are ready for exorcise it for real?

no strong opinions from me. @saimn ?

saimn · 2026-01-26T18:20:06Z

So it seems that currently only np.char.chararray is deprecated but np.char.array (which instantiates a np.char.chararray object) is not ? Hmm, we need to see what are Numpy's plans how the rest of np.char, since masking some warnings does not solve the underlying issue (we rely on that class).

pllim · 2026-01-26T19:05:09Z

Maybe @jorenham could advise?

jorenham · 2026-01-26T19:18:29Z

The plan is to eventually get rid of numpy.char. Importing numpy.chararray directly has been deprecated for a long time now, but numpy.char.chararray (accidentally) wasn't.

So it seems that currently only np.char.chararray is deprecated but np.char.array (which instantiates a np.char.chararray object) is not ?

Good catch; I agree that we should also deprecate the charray constructor methods. I'll try and bring it up at the next community meeting.

neutrinoceros · 2026-01-28T10:30:06Z

@saimn the failures resolved here are now showing up as collection-time errors on allowed-failure jobs, making them blind to anything else. I'll rebase this branch to make it clear that it goes a bit further than main on this front.

pllim · 2026-02-02T17:26:14Z

tl;dr -- devdeps still red here. What else is needed to make it green again?

neutrinoceros · 2026-02-02T18:05:48Z

#3862

pllim · 2026-02-02T19:32:14Z

Do you want to rebase one more time, and then hopefully @saimn can do a final review, if this PR is ready? Thanks!

neutrinoceros · 2026-02-02T21:55:57Z

rebased, and ready

…ng types at runtime

pllim · 2026-02-20T20:04:31Z

Superseded by #19267

neutrinoceros added this to the v7.2.1 milestone Jan 26, 2026

neutrinoceros added Bug no-changelog-entry-needed numpy-dev backport-v7.2.x on-merge: backport to v7.2.x labels Jan 26, 2026

github-actions Bot added io.fits utils labels Jan 26, 2026

neutrinoceros commented Jan 26, 2026

View reviewed changes

neutrinoceros marked this pull request as ready for review January 26, 2026 09:35

neutrinoceros requested a review from saimn as a code owner January 26, 2026 09:35

neutrinoceros mentioned this pull request Jan 26, 2026

BUG: numpy.char.chararray is deprecated #19216

Closed

neutrinoceros commented Jan 26, 2026

View reviewed changes

neutrinoceros force-pushed the compat/is_chararray branch from 4a30d22 to 6ccfdd8 Compare January 28, 2026 10:31

This was referenced Jan 31, 2026

EXP: define our own chararray class #19241

Closed

MAINT: shape setting should be deprecated in astropy container classes. #19168

Open

rosteen mentioned this pull request Feb 2, 2026

Ignore upstream chararray deprecation warning astropy/specutils#1314

Merged

WilliamJamieson mentioned this pull request Feb 2, 2026

Set warnings as errors in tests to catch when new warnings arise spacetelescope/gwcs#691

Merged

4 tasks

pllim reviewed Feb 2, 2026

View reviewed changes

Comment thread astropy/utils/compat/numpycompat.py Resolved

neutrinoceros force-pushed the compat/is_chararray branch from 6ccfdd8 to 16f6db2 Compare February 2, 2026 21:55

BUG: avoid deprecation warnings from numpy.char.chararray when checki…

909cc7d

…ng types at runtime

neutrinoceros force-pushed the compat/is_chararray branch from 16f6db2 to 909cc7d Compare February 3, 2026 10:29

pllim mentioned this pull request Feb 3, 2026

TST: Temporarily ignore DeprecationWarning from numpy chararray #19258

Closed

1 task

jorenham mentioned this pull request Feb 5, 2026

DEP: Deprecating numpy.char.array numpy/numpy#30787

Closed

pllim closed this Feb 20, 2026

neutrinoceros deleted the compat/is_chararray branch February 20, 2026 21:57



		def is_chararray(obj, /) -> bool:
		if not hasattr(np, "char"):

Conversation

neutrinoceros commented Jan 26, 2026

Description

Uh oh!

github-actions Bot commented Jan 26, 2026

Uh oh!

neutrinoceros Jan 26, 2026

Choose a reason for hiding this comment

Uh oh!

neutrinoceros Jan 26, 2026

Choose a reason for hiding this comment

Uh oh!

neutrinoceros commented Jan 26, 2026

Uh oh!

neutrinoceros Jan 26, 2026

Choose a reason for hiding this comment

Uh oh!

pllim Feb 3, 2026

Choose a reason for hiding this comment

Uh oh!

neutrinoceros Feb 3, 2026

Choose a reason for hiding this comment

Uh oh!

saimn commented Jan 26, 2026

Uh oh!

pllim commented Jan 26, 2026

Uh oh!

jorenham commented Jan 26, 2026

Uh oh!

neutrinoceros commented Jan 28, 2026

Uh oh!

Uh oh!

pllim commented Feb 2, 2026

Uh oh!

neutrinoceros commented Feb 2, 2026

Uh oh!

pllim commented Feb 2, 2026

Uh oh!

neutrinoceros commented Feb 2, 2026

Uh oh!

pllim commented Feb 20, 2026

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants