fix(res-to-affine): restore the five test fixtures deleted by f766dcb - #741
Merged
Conversation
f766dcb ("chore: remove rescript example configuration") deleted tools/res-to-affine/test/fixtures/*.res as ReScript artefacts. They are not artefacts: they are the *inputs* to the res-to-affine migration tool's own test suite -- synthetic anti-pattern files, each headed "Not a real ReScript program", against which the expected/*.affine goldens are compared. The deletion was silent. test/dune declares the inputs as (glob_files fixtures/*.res); a glob over a directory that does not exist matches zero files and does not fail the build, so the suite kept building and every test died at runtime on Sys_error("fixtures/sample.res: No such file or directory"). Effect: 35 of 35 res-to-affine tests failed, i.e. the tool that performs the ReScript-to-AffineScript migration has had no working correctness check since 2026-08-18. Restoring the six files takes the repo from `dune runtest` rc=1 to rc=0. Refs f766dcb Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: ASSERTIVE Plan: Team Run ID: 📒 Files selected for processing (5)
Included review availability: Your plan provides up to 1 included review per hour; 0 remain after this review. 📜 Recent review details⏰ Context from checks skipped due to timeout. (4)
🔇 Additional comments (3)
📝 SummarySummary by CodeRabbit
WalkthroughThe PR adds five Changesres-to-affine test fixtures
Estimated code review effort: 2 (Simple) | ~10 minutes Merge Risk: ⚪ Minimal · up to The restored fixtures match the migration tool's tested behavior and introduce no actionable merge risk. Suggested reviewers: Poem
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
hyperpolymath
added a commit
that referenced
this pull request
Sep 8, 2026
Implements the owner ruling of 2026-09-07: **no environment may silently run zero tests.** ## The fake green `test_walker.ml:65 skip_unless_ready()` called `Alcotest.skip()` when the tree-sitter CLI or the generated grammar was absent. `tools/vendor/tree-sitter-rescript/src/parser.c` is gitignored (`.gitignore:112`), so **every fresh checkout** took that path: all 32 walker cases skipped and alcotest printed ``` Test Successful in 0.018s. 0 test run. ``` with exit 0. The suite was green precisely when it was testing nothing. This is not hypothetical. It is why the deletion of this suite's own fixtures in `f766dcb` went unnoticed for three weeks (restored in #741): in any environment lacking the grammar, a missing test corpus produced a pass. ## The change Renames the guard to `require_ready()` and replaces both `Alcotest.skip()` calls with `Alcotest.failf` carrying the remedy (`cargo install tree-sitter-cli` / `just install-grammar`). One file, 32 call sites renamed mechanically. ## Verified, both directions | Grammar | Result | |---|---| | absent | `rc=1` — `32 failures! in 0.020s. 32 tests run.` | | present | `rc=0` — `Test Successful in 1.257s. 32 tests run.` | The important column is the **count**: previously the absent case reported `0 test run` and passed. `ci.yml:70-79` installs the CLI and builds the grammar before `dune runtest`, so CI lands in the passing row. ## Accepted cost A fresh clone can no longer run the walker suite without installing the tree-sitter CLI and building the grammar first. That is the deliberate trade, chosen over the milder "fail under CI, skip locally". ## Note `ocamlformat` is not available in my environment, so I could not pre-check `dune build @fmt`. If the formatting gate objects, say so and I will correct it. --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Restores
tools/res-to-affine/test/fixtures/{partial1,phase2c,phase3,phase3b,phase3c}.res,deleted by f766dcb "chore: remove rescript example configuration" (2026-08-18).
sample.reswas restored on main previously; these five were not.Why they are not "rescript example configuration"
They are the inputs to the res-to-affine migration tool's own test suite —
synthetic anti-pattern files, each headed "Not a real ReScript program" — against
which
expected/*.affineis compared. They are the only correctness check theReScript→AffineScript migration tool has.
Why it was silent
Two independent failure-to-signal mechanisms:
test/dunedeclares the inputs as(glob_files fixtures/*.res). A glob over adirectory that does not exist matches zero files and does not fail the build,
so the suite kept compiling and each test died at runtime on
Sys_error("fixtures/<f>.res: No such file or directory").test_walker.ml'sskip_unless_ready ()callsAlcotest.skip ()when thetree-sitter grammar is not built. In an environment without the grammar all 32
walker cases SKIP and alcotest reports
Test Successful … 0 test run, exit 0 —so the suite is green precisely when it is testing nothing.
Evidence
With the grammar built and the fixtures absent: 32 of 32 walker tests fail.
With the fixtures restored: 32 of 32 pass, and repo-wide
dune runtestgoesrc=1 → rc=0.
Not fixed here
Mechanism (2) above — the fail-open skip — is left as-is deliberately; turning a
skip into a hard failure is a policy change for the maintainer, not part of a
restore. Filed as a follow-up.