)]}'
{
  "commit": "6f3aa08983d0fcc4f7e37ff0ad3a86c1030deed8",
  "tree": "fc36fd5c09edaf684c50f8a864257cd7916ba573",
  "parents": [
    "a8ba25588bd21dc5e7bbcdaca3999461ec56c8ae"
  ],
  "author": {
    "name": "Oliver Newman",
    "email": "olivernewman@fuchsia.infra.roller.google.com",
    "time": "Tue Jun 14 01:35:59 2022 +0000"
  },
  "committer": {
    "name": "Copybara-Service",
    "email": "copybara-worker@google.com",
    "time": "Mon Jun 13 18:39:00 2022 -0700"
  },
  "message": "[roll] Roll fuchsia [fint] Fix bug in affected BUILD.gn file analysis\n\nWhen checking whether a test was affected based on whether its BUILD.gn\nfile was changed, we were previously comparing the _absolute_ path of\nthe BUILD.gn file ith the _relative_ path of the test\u0027s GN label, so we\nwere never considering any tests to be affected by their GN file\nchanges.\n\nNow we always use relative paths to compare test labels to changed\nfiles.\n\nAlso simplify the `removeDuplicates()` logic as long as I\u0027m here. It was\noptimized to avoid unnecessary allocations, but that made it\nsignificantly harder to read and made me suspect that it was concealing\nthe bug that was causing this issue. It will never be used to sort more\nthan a few thousand elements at a time, so the optimization was not\nworth the loss of readability.\n\nOriginal-Bug: 101789\nOriginal-Reviewed-on: https://fuchsia-review.googlesource.com/c/fuchsia/+/690149\nOriginal-Revision: 5dd3360ebc12778402016362bd0b2405c1b586da\nGitOrigin-RevId: 6889d554ddb5720ab4eafcbe6dc0e9ba907854c6\nChange-Id: Ia866cf33a28c75d53dc1c9bd19b82ed9798c58e6\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "231de5806490d88f8f5de90dd1e60678b9f1a155",
      "old_mode": 33188,
      "old_path": "stem",
      "new_id": "89245f6763bfa12cc1c99c732dc7f8bd6e7ae4b2",
      "new_mode": 33188,
      "new_path": "stem"
    }
  ]
}