DEP: expire deprecation for null handling of context and return_scalar in arraywrap - #31119

I am not sure about when to add the release notes, it would be really helpful if anybody could guide me on that. Thank you!

I am not sure about when to add the release notes, it would be really helpful if anybody could guide me on that. Thank you!

Expired deprecations indeed require a release note. See https://github.com/numpy/numpy/blob/main/doc/release/upcoming_changes/README.rst for details.

@jorenham Thank you for the reference, I have added the release notes.

Thanks, this LGTM. But, I have a bit of a feeling that people might have managed to blissfully not notice this, e.g. torch here: https://github.com/pytorch/pytorch/blob/main/torch/_tensor.py#L1259

So that makes me wonder a bit if we should at least only bump it to a VisibleDeprecationWarning, and maybe also open a torch issue and wait til after 2.5 (since that is very soon) to merge.

I.e. as much as I would like to get this done... there seems little rush here.

Thanks, this LGTM. But, I have a bit of a feeling that people might have managed to blissfully not notice this, e.g. torch here: https://github.com/pytorch/pytorch/blob/main/torch/_tensor.py#L1259

So that makes me wonder a bit if we should at least only bump it to a VisibleDeprecationWarning, and maybe also open a torch issue and wait til after 2.5 (since that is very soon) to merge.

I.e. as much as I would like to get this done... there seems little rush here.

That makes sense, thanks for the suggestion.
I will go ahead and open a PR for converting it to VisibleDeprecationWarning and open an issue with PyTorch to check whether this behavior is relied upon there.
Does that sound good?

Yes, that would be great! Could also convert this PR, but either way is good.