)]}'
{
  "commit": "1159cb541c8679af8aeca885d2c38d2c321e6c7e",
  "tree": "2e4bcc52d8f0bb0991767dfa5bfa33ca8e446c4f",
  "parents": [
    "77a4fb62f70c6ea05e1820216d903938e331d42b",
    "cf3249c0ccc2ef99adab21cb7b31a7a005826bb8"
  ],
  "author": {
    "name": "Guillaume Gomez",
    "email": "contact@guillaume-gomez.fr",
    "time": "Wed May 27 20:45:09 2026 +0200"
  },
  "committer": {
    "name": "GitHub",
    "email": "noreply@github.com",
    "time": "Wed May 27 20:45:09 2026 +0200"
  },
  "message": "Rollup merge of #156970 - qaijuang:async_closure_coverage, r\u003dZalathar\n\ncoverage: Use original HIR info for synthetic by-move coroutine bodies\n\n\u003e This is a copy of #156952(because my git was acting weird)\n\nSynthetic by-move coroutine bodies created for async closures don\u0027t have useful HIR of their own. Coverage was falling back to the parent async closure for HIR info, which means the executed `AsyncFnOnce` body could inherit hole spans from the wrong body and report the user-written closure body as uncovered.\n\nThis PR uses the synthetic body\u0027s coroutine type to recover the original coroutine body def-id, then extracts HIR info from that body instead. That keeps the coverage spans tied to the body the synthetic MIR was cloned from.\n\nFixes rust-lang/rust#151135.\n\nr? Zalathar\n",
  "tree_diff": []
}