### Bug summary I have taken the results from https://github.com/matplotlib/matplotlib/actions/runs/16533707714/job/46764218849?pr=30334, extracted the diff, then put all the common and `-` lines in one file and the common and `+` files in another, and wrote them to separate files. This produces a much simpler diff with just the lines that changed instead of a whole byte string diff: ``` --- a.eps 2025-07-25 20:29:08.047303419 -0400 +++ b.eps 2025-07-25 20:30:15.458646631 -0400 @@ -10771,27 +10771,7 @@ 0 setlinecap 0 0 1 setrgbcolor gsave -/c6 { -279.447273 58.176 m -387.665455 58.176 l -387.665455 138.816 l -279.447273 138.816 l -cl -333.556364 78.336 m -333.556364 78.336 326.381414 78.336 319.499366 80.460185 c -314.425911 84.240727 309.352456 88.02127 306.501818 93.149506 c -306.501818 98.496 306.501818 103.842494 309.352456 108.97073 c -314.425911 112.751273 319.499366 116.531815 326.381414 118.656 c -333.556364 118.656 340.731313 118.656 347.613361 116.531815 c -352.686816 112.751273 357.760272 108.97073 360.610909 103.842494 c -360.610909 98.496 360.610909 93.149506 357.760272 88.02127 c -352.686816 84.240727 347.613361 80.460185 340.731313 78.336 c -cl - -clip -newpath -} bind def -c6 +c5 333.556364 150.912 m 315.970909 121.192794 l 272.638562 124.704 l ``` ### Code for reproduction ```Python pytest ``` ### Actual outcome <img width="1155" height="902" alt="Image" src="https://github.com/user-attachments/assets/0c8e0094-de2f-4a5e-bafe-1a2ad3151508" /> ### Expected outcome <img width="1155" height="902" alt="Image" src="https://github.com/user-attachments/assets/e43c2ed7-5d64-457b-ac9e-2c7f8e4ad127" /> ### Additional information _No response_ ### Operating system _No response_ ### Matplotlib Version tip of #30334 ### Matplotlib Backend _No response_ ### Python version _No response_ ### Jupyter version _No response_ ### Installation None
Activity
QuLogic commentedon Jul 26, 2025
Looking at https://github.com/matplotlib/matplotlib/actions/runs/16535584290/job/46769299546, the diff is different this time. The star in the bottom left Axes is missing instead (and a few path names after it are renumbered).
apteryks commentedon Dec 4, 2025
I also hit this on Guix.
rcomer commentedon Dec 4, 2025
Also recently saw this fail in the minver tests. Is it always python 3.11?
apteryks commentedon Dec 4, 2025
Yes, testing with current Python in Guix, which is 3.11.11 in the build container.
QuLogic commentedon Apr 16, 2026
This continues to happen even with the re-arrangement of the subplots. And curiously, it is still the star that disappears.