TST: Set tests touched by text overhaul to mpl20 style by QuLogic · Pull Request #31300 · matplotlib/matplotlib

QuLogic · 2026-03-13T17:10:21Z

PR summary

This is based on several other PRs to make it easier for me to track what needs changing. I will rebase those extra commits out once ready.

The first commit changes all related tests to mpl20, while the second commit changes a few so that they better work in the new style (e.g., moves a transparent legend over top something so that you can actually confirm that it's transparent.) I do expect to squash these together, but kept them separate so that it'd be easier to review the secondary changes.

AI Disclosure

None

PR checklist

[n/a] "closes #0000" is in the body of the PR description to link the related issue
new and changed code is tested
[n/a] Plotting related features are demonstrated in an example
[n/a] New Features and API Changes are noted with a directive and release note
[n/a] Documentation complies with general and docstring guidelines

lib/matplotlib/tests/baseline_images/test_axes/o_marker_path_snap.png

QuLogic · 2026-03-13T17:20:09Z

lib/matplotlib/tests/baseline_images/test_axes/rc_spines.png

The test forces a low DPI, but I guess this is okay.

probably an attempt to save space?

QuLogic · 2026-03-13T17:20:39Z

lib/matplotlib/tests/baseline_images/test_axes/rc_grid.png

Again a low DPI, but probably okay?

same as below, probably to save space? Came in like this from 94d044e

lib/matplotlib/tests/baseline_images/test_axes/bar_tick_label_multiple_old_label_alignment.png

QuLogic · 2026-03-13T17:22:30Z

lib/matplotlib/tests/baseline_images/test_legend/rgba_alpha.png

This one seems like it might be wrong, given the red is a bit different, but it could be an illusion since the border is gone.

We need to set the legend.framealpha rcparam to None to get this test back.

Yes, fixed.

lib/matplotlib/tests/baseline_images/test_legend/rcparam_alpha.png

QuLogic · 2026-03-13T17:24:26Z

lib/matplotlib/tests/baseline_images/test_spines/spines_data_positions.png

With the new style, ticks on top/right spines are turned off. I'm not sure if we want to re-enable them here, mostly for aesthetic reasons; it's not really relevant to the test per se.

I'm in favor of turning them back on, but no strong feeling either way.

QuLogic · 2026-03-13T17:26:28Z

lib/matplotlib/tests/baseline_images/test_text/text_contains.png

Bit uncertain why this moved so much; I guess because text is in physical units, while everything else was scaled down for the smaller figure size. I'll try a larger figure to see if it matches again.

the figure got smaller in real units, the text stayed the same size in real units, the limits stayed the same in data units, so the text takes up more space in data units.

Either restore the old figure size in inches or make the range of points we try to hit bigger.

OK, made it bigger and it looks much closer to what it used to be.

lib/matplotlib/tests/baseline_images/test_text/agg_text_clip.png

QuLogic · 2026-03-13T17:30:11Z

lib/matplotlib/tests/test_axes.py


-@image_comparison(['test_stairs_options.png'], remove_text=True)
+@image_comparison(['test_stairs_options.png'], style='mpl20', remove_text=True)
 def test_stairs_options():


I have not yet investigated yet, but this test actually breaks after changing styles:

__________________ test_stairs_options[png] ___________________ args = (), kwds = {'extension': 'png', 'request': <FixtureRequest for <Function test_stairs_options[png]>>} @wraps(func) def inner(*args, **kwds): with self._recreate_cm(): > return func(*args, **kwds) /usr/lib64/python3.13/contextlib.py:85: _ _ _ _ _ _ _ _ _ _ _ _ lib/matplotlib/figure.py:3508: in savefig self.canvas.print_figure(fname, **kwargs) lib/matplotlib/backend_bases.py:2275: in print_figure result = print_method( lib/matplotlib/backend_bases.py:2132: in <lambda> print_method = functools.wraps(meth)(lambda *args, **kwargs: meth( lib/matplotlib/backends/backend_agg.py:539: in print_png self._print_pil(filename_or_obj, "png", pil_kwargs, metadata) lib/matplotlib/backends/backend_agg.py:487: in _print_pil FigureCanvasAgg.draw(self) lib/matplotlib/backends/backend_agg.py:440: in draw self.figure.draw(self.renderer) lib/matplotlib/artist.py:94: in draw_wrapper result = draw(artist, renderer, *args, **kwargs) lib/matplotlib/artist.py:71: in draw_wrapper return draw(artist, renderer) lib/matplotlib/figure.py:3275: in draw mimage._draw_list_compositing_images( lib/matplotlib/image.py:133: in _draw_list_compositing_images a.draw(renderer) lib/matplotlib/artist.py:71: in draw_wrapper return draw(artist, renderer) lib/matplotlib/axes/_base.py:3282: in draw mimage._draw_list_compositing_images( lib/matplotlib/image.py:133: in _draw_list_compositing_images a.draw(renderer) lib/matplotlib/artist.py:71: in draw_wrapper return draw(artist, renderer) lib/matplotlib/patches.py:740: in draw self._draw_paths_with_artist_properties( lib/matplotlib/patches.py:723: in _draw_paths_with_artist_properties gc.set_dashes(*self._dash_pattern) _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ self = <matplotlib.backend_bases.GraphicsContextBase object at 0x7f3016ba3750>, dash_offset = 0.0, dash_list = [0.0, 0.0] def set_dashes(self, dash_offset, dash_list): """ Set the dash style for the gc. Parameters ---------- dash_offset : float Distance, in points, into the dash pattern at which to start the pattern. It is usually set to 0. dash_list : array-like or None The on-off sequence as points. None specifies a solid line. All values must otherwise be non-negative (:math:`\\ge 0`). Notes ----- See p. 666 of the PostScript `Language Reference <https://www.adobe.com/jp/print/postscript/pdfs/PLRM.pdf>`_ for more info. """ if dash_list is not None: dl = np.asarray(dash_list) if np.any(dl < 0.0): raise ValueError( "All values in the dash list must be non-negative") if dl.size and not np.any(dl > 0.0): > raise ValueError( 'At least one value in the dash list must be positive') E ValueError: At least one value in the dash list must be positive lib/matplotlib/backend_bases.py:897: ValueError

Opened #31302 for this issue.

- `test_backend_ps::test_type3_font` is covered by `test_backend_ps::test_multi_font_type3` - `test_text::test_pdf_chars_beyond_bmp` is covered by `test_backend_pdf::test_multi_font_type3` and `test_backend_pdf::test_multi_font_type42` - `test_text::test_pdf_kerning` is covered by `test_backend_pdf::test_kerning` - `test_text::test_pdf_type42_kerning` is covered by `test_backend_pdf::test_kerning`

These tests had previously kept some code or settings to prevent regenerating the test image, but since they are now going to be regenerated from the text overhaul, we can remove those.

QuLogic added this to the v3.11.0 milestone Mar 13, 2026

github-actions bot added the topic: mplot3d label Mar 13, 2026

QuLogic added the status: waiting for other PR label Mar 13, 2026

QuLogic added this to Font and text overhaul Mar 13, 2026

github-project-automation bot moved this to Waiting for other PR in Font and text overhaul Mar 13, 2026

QuLogic commented Mar 13, 2026

View reviewed changes

QuLogic force-pushed the text-test-style branch 2 times, most recently from 42792b2 to 8109dbb Compare March 13, 2026 22:48

QuLogic force-pushed the text-test-style branch from 8109dbb to 1403060 Compare March 14, 2026 00:17

QuLogic added 6 commits March 13, 2026 21:47

Update test images for font/text overhaul [skip ci]

e66cd55

Merge branch 'text-overhaul-figures' into merges

0d2ca11

Merge branch 'redundant-font-tests' into merges

dc8921c

TST: Cleanup back-compat code in tests touched by text overhaul

4b97b9e

These tests had previously kept some code or settings to prevent regenerating the test image, but since they are now going to be regenerated from the text overhaul, we can remove those.

TST: Set tests touched by text overhaul to mpl20 style

f3030eb

TST: Tweak some tests to work better in mpl20 style

c0ccc2f

QuLogic force-pushed the text-test-style branch from 1403060 to c0ccc2f Compare March 14, 2026 02:23

Uh oh!

Conversation

QuLogic commented Mar 13, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

PR summary

AI Disclosure

PR checklist

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

QuLogic Mar 13, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

QuLogic commented Mar 13, 2026 •

edited

Loading

QuLogic Mar 13, 2026 •

edited

Loading