René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"pypy failures","articleBody":"Going through the logs from @mattip 's logs in #19154 I think the failures we are seeing can be classified as below.\r\n\r\n\r\n## finalizer / weakrefs different in pypy\r\n\r\nWe probably should just detect and skip pypy on these.\r\n\r\n- [x] FAILED lib/matplotlib/tests/test_animation.py::test_animation_delete[anim0]\r\n- [x] FAILED lib/matplotlib/tests/test_animation.py::test_animation_repr_html[anim0-none-None-imagemagick]\r\n- [x] FAILED lib/matplotlib/tests/test_animation.py::test_animation_repr_html[anim0-none-None-ffmpeg]\r\n- [x] FAILED lib/matplotlib/tests/test_animation.py::test_funcanimation_cache_frame_data[False]\r\n- [x] FAILED lib/matplotlib/tests/test_cbook.py::Test_callback_registry::test_callback_complete\r\n\r\n## pypy does not implement a working `sys.getsizeof`\r\n\r\nSkip this test on pypy, we are only using it to make sure that the output of rendering makes sense.\r\n\r\n- [x] FAILED lib/matplotlib/tests/test_backend_pdf.py::test_multipage_properfinalize\r\n\r\n\r\n## stack walking different?\r\n\r\nThe failures due to raising an unexpected warning are a bit more concerning.  We are doing some frame walking / inspection to sort out if we should warn or not in https://github.com/matplotlib/matplotlib/blob/09ab2c280bd7abb07b3e35a39134a9f791781fc9/lib/matplotlib/backend_bases.py#L1624-L1665  I assume this is not working on pypy because we are relying on some cpython implementation detail?  \r\n\r\nIf someone who understands pypy can fix that to work on both that would be best, but if not we should detect if we are on pypy and fall back to never warning.  It is operating in a slightly degraded mode (users may not get warnings they should see), but the status quo is going to warn for things the user can not do anything about (I think every time we use tightlayout of `bbox_inches='tight'`).\r\n\r\nThis is probably the only failures I would hold 3.4 over.\r\n\r\n- [x] FAILED lib/matplotlib/tests/test_bbox_tight.py::test_bbox_inches_tight[png]\r\n- [x] FAILED lib/matplotlib/tests/test_bbox_tight.py::test_bbox_inches_tight[svg]\r\n- [x] FAILED lib/matplotlib/tests/test_bbox_tight.py::test_bbox_inches_tight_suptile_legend[pdf]\r\n- [x] FAILED lib/matplotlib/tests/test_bbox_tight.py::test_bbox_inches_tight_clipping[png]\r\n- [x] FAILED lib/matplotlib/tests/test_bbox_tight.py::test_bbox_inches_tight_clipping[svg]\r\n- [x] FAILED lib/matplotlib/tests/test_bbox_tight.py::test_bbox_inches_tight_raster[pdf]\r\n- [x] FAILED lib/matplotlib/tests/test_backend_pgf.py::test_bbox_inches_tight - mat...\r\n- [x] FAILED lib/matplotlib/tests/test_backend_pgf.py::test_mixedmode[pdf] - matplo...\r\n- [x] FAILED lib/matplotlib/tests/test_bbox_tight.py::test_bbox_inches_tight[pdf]\r\n- [x] FAILED lib/matplotlib/tests/test_bbox_tight.py::test_bbox_inches_tight_suptile_legend[png]\r\n- [x] FAILED lib/matplotlib/tests/test_bbox_tight.py::test_bbox_inches_tight_suptile_legend[svg]\r\n- [x] FAILED lib/matplotlib/tests/test_bbox_tight.py::test_bbox_inches_tight_clipping[pdf]\r\n- [x] FAILED lib/matplotlib/tests/test_bbox_tight.py::test_bbox_inches_tight_raster[png]\r\n- [x] FAILED lib/matplotlib/tests/test_bbox_tight.py::test_bbox_inches_tight_raster[svg]\r\n- [x] FAILED lib/matplotlib/tests/test_bbox_tight.py::test_tight_pcolorfast - matpl...\r\n- [x] FAILED lib/matplotlib/tests/test_figure.py::test_tightbbox_box_aspect[svg] - ...\r\n- [x] FAILED lib/matplotlib/tests/test_bbox_tight.py::test_only_on_non_finite_bbox\r\n- [x] FAILED lib/matplotlib/tests/test_bbox_tight.py::test_noop_tight_bbox - matplo...\r\n- [x] FAILED lib/matplotlib/tests/test_offsetbox.py::test_annotationbbox_extents - ...\r\n- [x] FAILED lib/mpl_toolkits/tests/test_axes_grid1.py::test_image_grid[png] - matp...\r\n\r\n## segfaults in tk\r\n\r\nWe should either track this down or just skip the tests.\r\n\r\n- [x] FAILED lib/matplotlib/tests/test_backends_interactive.py::test_interactive_backend[toolbar2-tkagg]\r\n- [x] FAILED lib/matplotlib/tests/test_backends_interactive.py::test_interactive_backend[toolmanager-gtk3agg]\r\n- [x] FAILED lib/matplotlib/tests/test_backends_interactive.py::test_interactive_backend[toolmanager-tkagg]\r\n\r\n\r\n## subprocess hung\r\n\r\nWe should either track this down or just skip the tests.\r\n\r\n\r\n- [x] FAILED lib/matplotlib/tests/test_backend_tk.py::test_never_update - Failed: S...\r\n- [x] FAILED lib/matplotlib/tests/test_backend_tk.py::test_figuremanager_cleans_own_mainloop\r\n- [x] FAILED lib/matplotlib/tests/test_backends_interactive.py::test_interactive_backend[toolbar2-gtk3agg]\r\n- [x] FAILED lib/matplotlib/tests/test_backends_interactive.py::test_interactive_thread_safety[gtk3agg]\r\n- [x] FAILED lib/matplotlib/tests/test_backends_interactive.py::test_interactive_thread_safety[tkagg]\r\n\r\n## not matching numpy?\r\n\r\nThese both look like they are upstream bugs in pypy?  We should probably just skip these. \r\n\r\n- [x] FAILED lib/matplotlib/tests/test_cbook.py::test_step_fails[args1] - SystemErr...\r\n- [x] FAILED lib/matplotlib/tests/test_matplotlib.py::test_importable_with__OO - As...\r\n\r\n## checking refcounts\r\n\r\nThese tests do not make sense for pypy, we should skip them.\r\n\r\n- [x] FAILED lib/matplotlib/tests/test_quiver.py::test_quiver_memory_leak - Attribu...\r\n- [x] FAILED lib/matplotlib/tests/test_quiver.py::test_quiver_key_memory_leak - Att...\r\n\r\n## sockets not working\r\n\r\nThe comments around this suggest we never got this working on azure (which I assume is what is hosting GH actions) so we should sort out why the code to not run these tests in that case is not being triggered here.\r\n\r\n- [x] FAILED lib/matplotlib/tests/test_backends_interactive.py::test_webagg - Faile...\r\n\r\n","author":{"url":"https://github.com/tacaswell","@type":"Person","name":"tacaswell"},"datePublished":"2020-12-21T22:54:01.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":9},"url":"https://github.com/19160/matplotlib/issues/19160"}