Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

TST: NumPy 2 compatibility tracker #4874

Draft
wants to merge 1 commit into
base: main
Choose a base branch
from

Conversation

neutrinoceros
Copy link
Member

@neutrinoceros neutrinoceros commented Apr 12, 2024

PR Summary

This PR isn't meant to be merged

I've been watching numpy/numpy#26191
At the time I'm opening this, all our core dependencies (matplotlib, unyt, ewah-bool-utils) already have compatible releases.
This will unleash full CI with numpy 2 at runtime to reveal which of our optional and transitional dependencies still lack support for it, so I have a chance to reach out to them before the final release drops.

name compatible release on PyPI tracker resource, notes
astropy
cartopy
h5py
netcdf4 Unidata/netcdf4-python#1317
pandas
arm-pyart ARM-DOE/pyart#1550
pykdtree storpipfugl/pykdtree#113
scipy
shapely shapely/shapely#1972
xarray pydata/xarray#8844

This table may be incomplete, but CI doesn't forgive, which is why this is a PR and not an issue.

edit history:
2024/04/12 created
2024/04/13
2024/05/05 new entry: astropy

@neutrinoceros neutrinoceros added the do not merge this change should not be merged until later label Apr 12, 2024
@neutrinoceros neutrinoceros changed the title TST: pin runtime requirement to numpy>=2.0.0rc1 to see which of our optional dependencies still lack compat TST: NumPy 2 compatibility tracker Apr 12, 2024
@neutrinoceros
Copy link
Member Author

neutrinoceros commented Apr 19, 2024

An aspect I hadn't considered earlier is that some packages may have dropped support for Python 3.9 before adding compatibility with numpy 2.0, in which case a brand new env in Python 3.9 may be broken. I'm not going to worry about that right now but it's something to keep in mind if CI breaks specifically on 3.9 even after all our optional dependencies have a numpy2-compatible release.

update 16-05-2024: this was debated on numpy's mailing list so hopefully awareness of this problem is higher now.

@neutrinoceros
Copy link
Member Author

neutrinoceros commented May 16, 2024

My understanding is that netcdf4 maintainers do not wish to build off of a numpy release candidate and are waiting for the final stable release, but the dev branch is already compatible, so I don't expect it to be a blocker for long.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
do not merge this change should not be merged until later
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

1 participant