Repository navigation
upgrade cibuildwheel action to support Python 3.14 - #1004
Conversation
|
Thanks! It looks like there are some build failures? |
|
Yes, I was unable to judge whether they came because of the new version sinceithe error is on Python 3.11. |
|
The original test function is not deterministic. It isn't guaranteed that 100 random samples will definitely cover all combinations. |
|
Correct, good spot, rerunning failed jobs |
|
Not sure why the rerun could not be scheduled within 24 hours, but can you rerun it again, please? |
|
done is macos-13 still valid? |
|
Actually not, 14 is the oldest https://github.com/actions/runner-images#available-images I'll change that, too, when I get home. |
|
thanks! |
Choose MacOS runner images with native x86 and arm64 architectures, see https://github.com/actions/runner-images#available-images.
5848274 to
d19cf49
Compare
|
Made some changes to also test on Python 3.14. I'm not sure why this problem only pops up now with 3.14, but is there a good reason to install pytest via pip rather than a conda channel? It's also no longer recommended to use the conda defaults channel due to licensing issues. |
|
Good catch, any chance you could pull from conda-forge? The gist is this project does not have direct funding, and as such I can only allocate minimal maintenance effort. We greatly appreciate contributions like yours which help immensely. |
|
Done already in the last commit 😉 |
|
Thanks @Midnighter! |
biom-format has no CPython 3.14 wheels yet (biocore/biom-format#1004 is merged but unreleased), so on 3.14 it builds from source and needs a C compiler. The stage-1 constraints require the docs to say so; the note sits in README's installation section, which docs/index.md includes. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* MAINT: build Python 3.14 wheels #1004 upgraded cibuildwheel to a version that supports 3.14, but the build selector in pyproject.toml still listed cp39-cp313, so no cp314 wheels were produced. Add cp314-* to the selector. Also note the in1d -> isin fix (#1003) in the ChangeLog. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * BUG: to_anndata fails with anndata >= 0.13 anndata 0.13.0 removed the dtype argument to AnnData(), so Table.to_anndata() raised TypeError. Cast the matrix before constructing the AnnData object instead. This also removes the FutureWarning older anndata versions emitted for dtype=. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
3 participants
fix #1002