NotificationsYou must be signed in to change notification settings
Fork7.9k
Star21.4k

See if pinning numpy to < 1.22 fixes tests#22076

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 ourterms of service andprivacy statement. We’ll occasionally send you account related emails.

Already on GitHub?Sign in to your account

Jump to bottom

Closed

dstansby wants to merge1 commit intomatplotlib:mainfromdstansby:numpy-pin

Closed

See if pinning numpy to < 1.22 fixes tests#22076

dstansby wants to merge1 commit intomatplotlib:mainfromdstansby:numpy-pin

Conversation

Copy link

Member

dstansby commentedJan 1, 2022

PR Summary

In the last couple of days a few of the tests have been failing on some builds. The test fails are:

FAILED lib/matplotlib/tests/test_axes.py::test_errorbar[png] - matplotlib.tes...FAILED lib/matplotlib/tests/test_axes.py::test_errorbar[svg] - matplotlib.tes...FAILED lib/matplotlib/tests/test_axes.py::test_errorbar[pdf] - matplotlib.tes...FAILED lib/matplotlib/tests/test_streamplot.py::test_direction[png] - matplot...FAILED lib/mpl_toolkits/tests/test_axisartist_grid_helper_curvelinear.py::test_axis_direction[png]FAILED lib/mpl_toolkits/tests/test_mplot3d.py::test_trisurf3d[png] - matplotl...FAILED lib/mpl_toolkits/tests/test_mplot3d.py::test_stem3d[png] - matplotlib....

Running a diff between installed packages on a succesful and failed run shows that the only packages that have changed are numpy-1.22.0, pikepdf-4.3.1, and pygments-2.11.1. I don't think it's a MPL commit causing the new test failures, so try pinning numpy to see if that's the cause.

PR Checklist

Tests and Styling

Has pytest style unit tests (andpytest passes).
IsFlake 8 compliant (installflake8-docstrings and runflake8 --docstring-convention=all).

Documentation

New features are documented, with examples if plot related.
New features have an entry indoc/users/next_whats_new/ (follow instructions in README.rst there).
API changes documented indoc/api/next_api_changes/ (follow instructions in README.rst there).
Documentation is sphinx and numpydoc compliant (the docs shouldbuild without error).

Copy link

Contributor

greglucas commentedJan 2, 2022

CI pip installs MPL with--no-deps, so you'll have to pin it in the GitHub Actions file I think.

See here for some other discussion of checks I was trying before the release:#21634
I still can't reproduce these failures locally using the same numpy manylinux wheels that CI is using, so it is going to be tough to debug this.

Copy link

Contributor

StefRe commentedJan 2, 2022

I'm not sure it's just thenumpy version alone, on#22063 (comment) I get the following results (I think it's the same for all the issues since yesterday):

Python  OS      numpy  result3.7 min Ubuntu  1.17.0 pass3.7     Ubuntu  1.21.5 pass3.8     Ubuntu  1.22.0 fail3.9     Ubuntu  1.22.0 fail3.10    Ubuntu  1.22.0 pass3.7     Windows 1.21.5 pass3.8     Windows 1.22.0 pass3.9     Windows 1.22.0 pass3.7     mac     1.21.5 pass3.8     mac1.22.0 pass3.9     mac1.22.0 pass