René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"[MNT]: deprecate auto hatch fallback to patch.edgecolor when edgecolor='None'","articleBody":"### Summary\r\n\r\nSince #28104 now separates out hatchcolor, users should not be allowed to specifically ask to fallback to edgecolor and also explicitly set that edgecolor to none because silently falling back on ```edgecolor=\"None\"``` introduces the problem that:\r\n\r\n1. because of eager color resolution and frequently setting \"none\" by knocking out the alpha, the way to check for none is checking the alpha, which leads to fallback being dependent on the interplay between alpha and edgecolor:\r\n![Figure_1](https://github.com/user-attachments/assets/49f44a68-1d4c-47d8-bdd8-f05dcbfe076f)\r\n\r\n3. fallback to edgecolor rcParams doesn't check if the rcParam is also none so could sink the problem a layer deeper\r\n\r\n\r\n\r\n### Proposed fix\r\n\r\nDeprecate this fallback behavior and raise a warning on \r\n\r\n``` Rectangle( (0,0), .5, .5, hatchcolor='edge', edgecolor='None')```\r\n\r\nThe alternatives are:\r\n* set a hatchcolor \r\n* don't set edgecolor at all, .i.e. ``` Rectangle( (0,0), .5, .5, hatchcolor='edge')``` falls back to the rcParam in `get_edgecolor`","author":{"url":"https://github.com/story645","@type":"Person","name":"story645"},"datePublished":"2024-12-25T00:19:58.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":2},"url":"https://github.com/29380/matplotlib/issues/29380"}