8000 [Bug]: numpy 1.22 masked units test failure · Issue #21632 · matplotlib/matplotlib · GitHub
[go: up one dir, main page]
Skip to content

[Bug]: numpy 1.22 masked units test failure #21632

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.

Sign up for GitHub

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

Jump to bottom
Closed
greglucas opened this issue Nov 14, 2021 · 5 comments
Closed

[Bug]: numpy 1.22 masked units test failure #21632

greglucas opened this issue Nov 14, 2021 · 5 comments

Comments

@greglucas
Copy link
Contributor

Bug summary

Testing with numpy 1.22 produces one test failure in Matplotlib. I'm not sure if it is in the way the test is designed, or something more intrinsic to the units handling.

Code for reproduction

python -m pip install -i https://pypi.anaconda.org/scipy-wheels-nightly/simple numpy
pytest lib/matplotlib/tests/test_units.py::test_plot_masked_units

Actual outcome

Link to CI run logs for this: https://github.com/greglucas/matplotlib/runs/4204821077?check_suite_focus=true

Expected outcome

No failing test

Additional information

No response

Operating system

No response

Matplotlib Version

3.5.0.dev2515+g6c9adc9b33

Matplotlib Backend

No response

Python version

No response

Jupyter version

No response

Installation

No response
@tacaswell
Copy link
Member

I can confirm this, looking into into it...

@tacaswell
Copy link
Member

bisecting numpy changes is not fast 😞

@tacaswell tacaswell mentioned this issue Nov 14, 2021
Closed
@tacaswell
Copy link
Member 
$ git bisect bad
beacb394e94a517d1c84f8a50f9e18f3bf11ced5 is the first bad commit
commit beacb394e94a517d1c84f8a50f9e18f3bf11ced5
Author: default-303 <ajayds2001@gmail.com>
Date:   Wed May 26 18:36:36 2021 +0530

    MAINT: Removed 'if' block which served no purpose in core.py and removed '_linalgRealType(t)' function from linalg.py

 numpy/linalg/linalg.py | 4 ----
 numpy/ma/core.py       | 7 -------
 2 files changed, 11 deletions(-)

Not clear if the issue is a real issue or something about our test.

@dopplershift
Copy link
Contributor

This is a real issue that I can reproduce with pint, which is what the Quantity class in test_units.py is designed to replicate. I'll continue the discussion in numpy/numpy#20372.

@greglucas
Copy link
Contributor Author

This was resolved upstream in numpy by reverting part of that commit: #20386

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@tacaswell @dopplershift @greglucas
0