diff options
| author | Max Bernstein <max.bernstein@shopify.com> | 2025-03-25 12:26:04 -0400 |
|---|---|---|
| committer | Takashi Kokubun <takashikkbn@gmail.com> | 2025-04-18 21:53:01 +0900 |
| commit | 556fda08640d69bf18f704f8983cf9bfe2bd9f7c (patch) | |
| tree | 76481c8951a2c265bf5df410a0790e388acedccc /doc | |
| parent | d9720eaf10ed1dda3c2a9b9b4f4b8254b33a0c3c (diff) | |
Use expect-test for HIR tests
This makes it easier to update cascading test failures while still
making output reviewable in the PR.
Notes
Notes:
Merged: https://github.com/ruby/ruby/pull/13131
Diffstat (limited to 'doc')
| -rw-r--r-- | doc/zjit.md | 10 |
1 files changed, 10 insertions, 0 deletions
diff --git a/doc/zjit.md b/doc/zjit.md index a6e2f745d9..e71a0df539 100644 --- a/doc/zjit.md +++ b/doc/zjit.md @@ -50,6 +50,16 @@ You can also run a single test case by specifying the function name: make zjit-test ZJIT_TESTS=test_putobject ``` +If you expect that your changes cause tests to fail and they do, you can have +`expect-test` fix the expected value for you by putting `UPDATE_EXPECT=1` +before your test command, like so: + +``` +UPDATE_EXPECT=1 make zjit-test ZJIT_TESTS=test_putobject +``` + +Test changes will be reviewed alongside code changes. + <details> <summary>Setting up zjit-test</summary> |
