🌐 AI搜索 & 代理 主页
Skip to content

Conversation

@QuLogic
Copy link
Member

@QuLogic QuLogic commented Sep 18, 2025

PR summary

This should fix macOS CI and the disjoint bases issue.

Also, add an ignore for GlyphIndexType in stubtest, which may or may not be an upstream problem, but is the simplest fix for now.

PR checklist

ksunden and others added 30 commits May 9, 2025 12:23
…t-of-pr-30029-on-v3.10.3-doc

Backport PR matplotlib#30029 on branch v3.10.3-doc (Update diagram in subplots_adjust documentation to clarify parameters)
…t-of-pr-30119-on-v3.10.3-doc

Backport PR matplotlib#30119 on branch v3.10.3-doc (Add some types to _mathtext.py)
…t-of-pr-30118-on-v3.10.3-doc

Backport PR matplotlib#30118 on branch v3.10.3-doc (CI: Skip jobs on forks)
…t-of-pr-30212-on-v3.10.3-doc

Backport PR matplotlib#30212 on branch v3.10.3-doc ([Doc]: fix bug in release notes for matplotlib v3.5.0 and v3.7.0)
…t-of-pr-30244-on-v3.10.3-doc

Backport PR matplotlib#30244 on branch v3.10.3-doc (DOC: Recommend to use bare Figure instances for saving to file)
…t-of-pr-30289-on-v3.10.3-doc

Backport PR matplotlib#30289 on branch v3.10.3-doc (DOC: Fix build with pybind11 3)
It may be fundamentally nice not to have to create the figure
though pyplot to be able to use it in pyplot afterwards. You can now do

```
from matplotlib.figure import Figure
import matplotlib.pyplot as plt

fig = Figure()
fig.subplots().plot([1, 3, 2])

plt.figure(fig)  # fig is now tracked in pyplot
plt.show()
```

This also opens up the possibility to more dynamically track
and untrack figures in pyplot, which opens up the road to
optimized figure tracking in pyplot (matplotlib#29849)
When destroying a manager, replace the figure's canvas by a
figure canvas base.
There in now machinery for all of the public API that takes a
renderer is input to get it from the current canvas on the root
figure.  Use that machinery instead.
Flush the wx canvas, not the base canvas that is installed on close.
This goes up the chain to the GUI classes that do not have this method.
This prevent the figures from growing on hi-dpi screens.
Co-authored-by: Ruth Comer <10599679+rcomer@users.noreply.github.com>
@github-actions github-actions bot added topic: canvas and figure manager Documentation: examples files in galleries/examples Documentation: devdocs files in doc/devel Documentation: user guide files in galleries/users_explain or doc/users labels Sep 18, 2025
timhoffm and others added 2 commits September 18, 2025 07:20
On some CI, the Qt backend seems to be picked, but something is confused
and it fails due to an invalid `DISPLAY` variable. But this test doesn't
need an interactive backend, so don't use one.
@QuLogic
Copy link
Member Author

QuLogic commented Sep 18, 2025

May want to wait to also add in #30581 here.

anntzer and others added 3 commits September 18, 2025 09:10
Copy-edit the "fonts in pdf and postscript" table.
TST: Force Agg backend in test_openin_any_paranoid
@QuLogic
Copy link
Member Author

QuLogic commented Sep 18, 2025

I redid the merge now that #30581 is in, and now somehow GitHub has decided to show all the commits from main...

@QuLogic
Copy link
Member Author

QuLogic commented Sep 18, 2025

The last merge into the text-overhaul branch was 9766cbd, which came from 2eccf17, so this PR should be equal to 2eccf17...5cdd827 + the last commit.

@tacaswell
Copy link
Member

Some of the failures look real but one of them is the c-coverage filter issue that we have fixed at least twice...

Copy link
Member

@tacaswell tacaswell left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Leave to @QuLogic discretion to fix the failing tests in this PR or to do it in a follow on.

@QuLogic
Copy link
Member Author

QuLogic commented Sep 18, 2025

Some of the failures look real but one of them is the c-coverage filter issue that we have fixed at least twice...

Ah, I think we have a specific cutout for ubuntu-24.04, but it should also catch ubuntu-22.04-arm as well. It has lcov 2:
https://github.com/matplotlib/matplotlib/actions/runs/17839138498/job/50723813450?pr=30579#step:5:876
(not sure why non-arm is still 1.15 though).

Let me push that and see if it works.

@QuLogic
Copy link
Member Author

QuLogic commented Sep 18, 2025

This is actually affecting the main branch as well; I've opened #30586 to fix it there.

QuLogic and others added 4 commits September 18, 2025 17:58
Since earlier today, the `ubuntu-22.04-arm` image has actually been
Ubuntu 24.04. Instead of working out how to fix that, switch to the
newer image explicitly.
This may be an upstream bug [1], but until that is determined, ignore
the error to get CI working.

[1] python/mypy#19877
@QuLogic QuLogic force-pushed the fix-text-overhaul-ci branch from e4078a2 to b236488 Compare September 19, 2025 03:38
@QuLogic
Copy link
Member Author

QuLogic commented Sep 19, 2025

Since everything is passing now, I'll merge, as normally I've been doing main-to-text-overhaul merges without PRs.

@QuLogic QuLogic merged commit 0da372b into matplotlib:text-overhaul Sep 19, 2025
41 of 42 checks passed
@QuLogic QuLogic deleted the fix-text-overhaul-ci branch September 19, 2025 05:18
@github-project-automation github-project-automation bot moved this to Waiting for other PR in Font and text overhaul Sep 19, 2025
@QuLogic QuLogic moved this from Waiting for other PR to Done in Font and text overhaul Sep 19, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment