BUG: fix support for gufuncs with more than one core dimension - #16120
Conversation
In particular, `erfa.rxp` and `erfa.rxr`.
|
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.
|
| tuple(range(-1, -1 - len(sig), -1)) | ||
| if axis is None | ||
| else axis |
neutrinoceros
left a comment
There was a problem hiding this comment.
I find it pretty elegant for a hack :-)
More importantly, it does fix the issue in a back portable way, so that's more than enough for me. Thank you very much for going at it so quickly !
neutrinoceros
left a comment
There was a problem hiding this comment.
Actually, now I'm seeing a new error in MaskedArray.__array_ufunc__ in #16070 (which this PR is supposed to unblock) https://github.com/astropy/astropy/actions/runs/8076412008/job/22064849015?pr=16070
I don't know yet if that's a new bug in this PR or if it's also present on main (will inspect). I just wanted to hold off this one until we know more. I'll update my review status as soon as possible !
|
@pllim - this was reviewed by @neutrinoceros, which I think should suffice. Could you merge for us? |
pllim
left a comment
There was a problem hiding this comment.
Approving by proxy. Thanks!
(I don't grok this but I trust you...)
…n one core dimension
…120-on-v6.0.x Backport PR #16120 on branch v6.0.x (BUG: fix support for gufuncs with more than one core dimension)
In particular,
erfa.rxpanderfa.rxr.This just fixes the bug, and is slightly weird in its treatment of
axis. I have a follow-up PR in the works which just adds proper support foraxes- but that would be less suitable for backporting.Fixes #16116