Dear Nathan, thanks for your feedback! I agree a PR is a better way to discuss code, I had written to the ML and waited for a reply just because the PR template on github recommends "IF IT'S A NEW FEATURE OR API CHANGE, TEST THE WATERS" with a pointer to the ML.
... but anyway, here is the PR: https://github.com/numpy/numpy/pull/26434 And indeed, this version has no API changes! I followed some of your suggestions, in particular the heuristics to decide which code path to take. However I didn't port to C all my code. Let me know if my contribution is interesting anyway. Cheers, Pietro _______________________________________________ NumPy-Discussion mailing list -- numpy-discussion@python.org To unsubscribe send an email to numpy-discussion-le...@python.org https://mail.python.org/mailman3/lists/numpy-discussion.python.org/ Member address: arch...@mail-archive.com
