)]}'
{
  "commit": "de4e2eebf6156d25b950f12ea9b11667981b7acd",
  "tree": "088f7068676435c93cd5c92dca8909f5abc87937",
  "parents": [
    "0f88cd38f15dce6da7f621c48ec538d1619ecc14",
    "9aa459e742c8b5586142560fd00d8b77d59a83b7"
  ],
  "author": {
    "name": "Jacob Pratt",
    "email": "jacob@jhpratt.dev",
    "time": "Wed Jul 15 06:12:13 2026 -0400"
  },
  "committer": {
    "name": "GitHub",
    "email": "noreply@github.com",
    "time": "Wed Jul 15 06:12:13 2026 -0400"
  },
  "message": "Rollup merge of #159270 - Kobzol:bootstrap-panic, r\u003djieyouxu\n\nOnly force backtraces when a panic occurs in bootstrap\n\nIn https://github.com/rust-lang/rust/pull/159040, I forced a backtrace to be printed in bootstrap when a panic happens. However, since bootstrap also captures backtraces manually during command execution, this meant that backtraces started to be printed during \"normal\" bootstrap failures (e.g. if tidy returned a failure), which is annoying.\n\nThis PR changes it so that the override only happens ona actual panics.\n\nNote: bootstrap still uses a lot of panics for \"normal\" error conditions during a build, rather than for actual bugs or weird bootstrap situations. Those should ideally be replaced with `exit!(1)`.\n",
  "tree_diff": []
}