Repository navigation
MNT: Add test for aitoff-projection - #15092
jklymak merged 1 commit into
Conversation
Sorry, something went wrong.
| https://github.com/matplotlib/matplotlib/pull/14451 | ||
| """ | ||
| np.random.seed(42) | ||
| ra_random = np.random.rand(20) * 2 * np.pi |
There was a problem hiding this comment.
Can these be a structured grid (or line of points) instead of random points? That way it's easier to understand changes to the positioning of the points due to any future code changes.
Sorry, something went wrong.
There was a problem hiding this comment.
Ok, changed to use a grid. See updated post above.
Sorry, something went wrong.
|
|
||
| fig, ax = plt.subplots(figsize=(8, 4.2), | ||
| subplot_kw=dict(projection="aitoff")) | ||
| ax.tick_params(left=False, bottom=False, |
There was a problem hiding this comment.
Why remove the ticks?
Sorry, something went wrong.
There was a problem hiding this comment.
Because they aren't relevant for what's being tested. Would you like me to include them?
Sorry, something went wrong.
There was a problem hiding this comment.
If we add them then the image is also testing the tick layout, and will be a good catch of any changes in ticking changes for this projection.
Sorry, something went wrong.
There was a problem hiding this comment.
aitoff projection doesn't have ticks, and the labels are excluded by the remove_text=True option in the @image_comparison decorator (I cannot get the matching freetype to install on my system). But I went to put in a grid now, which would test the tick locations.
Sorry, something went wrong.
PR Summary
Closes #14503
Adds test for fix in #14451
Unfortunately, the initial problem occured at draw time, so one will need to use an image comparison test.
Running the test in matplotlib 3.1 results in
Running the test in current master results in
PR Checklist