René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"[MNT]: Use more-inclusive language in _trirefine.py","articleBody":"### Summary\n\nCurrently, both 'agg_scanline_boolean_algebra.h' and '_trirefine.py' use \"master\"/\"slave\" terminology:\n\n**agg_scanline_boolean_algebra.h** (Comments only)\n\n```c\n// Prepare the scanline generators.\n// Here \"sg1\" is master, \"sg2\" is `slave`.\n\n// Synchronize \"slave\" with \"master\"\n\n// Advance the \"master\"\n```\n\n**_trirefine.py** (Comments and variable names)\n\n```python\n# We first build 2 * ntri arrays of edge starting nodes (edge_elems,\n# edge_apexes); we then extract only the masters to avoid overlaps.\n# The so-called 'master' is the triangle with biggest index\n# The 'slave' is the triangle with lower index\n# (can be -1 if border edge)\n# For slave and master we will identify the apex pointing to the edge\n# start\n…\n```\n  \nI recognize that #21371 mentioned a desire to not touch agg (although this may have changed - there have been recent changes). I'm not sure if this applies to '_trirefine.py' or not.\n\n\"child\" already has meaning in '_trirefine.py'. Perhaps 'major'/'minor' could work here? Or maybe there is a mathematical term to mean \"biggest index\" \"lower index\".\n\n(Tagging @story645 as we chatted about this)\n\n### Proposed fix\n\n_No response_","author":{"url":"https://github.com/iccir","@type":"Person","name":"iccir"},"datePublished":"2026-07-13T03:25:15.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":4},"url":"https://github.com/32039/matplotlib/issues/32039"}