docs(tests/deepspeech): FIXME cites closed #83, not the open tracking issue #86

[dekobon]

Summary

The FIXME skip-rationale comment in tests/deepspeech_test.rs points readers at the closed issue #83 as the live blocker ("until the issue is resolved"), with no pointer to the open tracking issue #86 that actually governs the five still-excluded files.

Location

• tests/deepspeech_test.rs:8-9

Evidence

// FIXME: Ignoring these files temporarily due to parsing errors (see issue: https://github.com/dekobon/big-code-analysis/issues/83),
// in order to allow CI to pass until the issue is resolved.

Issue states:

• fix(tests,mozcpp): tree-sitter-cpp parse failures skip mozcpp/deepspeech tests #83 — fix(tests,mozcpp): tree-sitter-cpp parse failures skip mozcpp/deepspeech tests — CLOSED.
• track(grammar): tree-sitter-cpp blockers behind mozcpp/deepspeech skips (#83) #86 — track(grammar): tree-sitter-cpp blockers behind mozcpp/deepspeech skips (#83) — OPEN.

The #721 note lower in the same comment (lines 11-17) correctly explains that two files were un-skipped by the upstream-cpp flip and that "the remaining five are shared-base defects," but it never names the live tracking issue. The only clickable issue link in the file is #83, which is closed. A maintainer following that link to find out whether the skip can be removed lands on a resolved issue and gets no signal that #86 is where the remaining blockers are tracked.

The sibling doc tests/README.md:119 gets this right — it writes "(→ [#83]..., tracked in [#86]...)" — so the correct cross-reference already exists elsewhere in the repo; only the test-file comment is stale.

Expected Behavior

The skip-rationale comment should cite the open tracking issue (#86) for the five remaining excluded files, matching tests/README.md. Citing the closed #83 alone (or as "until the issue is resolved") is misleading.

Actual Behavior

The comment cites only the closed #83 and frames the skip as pending "until the issue is resolved," implying an open blocker that no longer exists under that number.

Impact

Documentation drift in a test that deliberately excludes five corpus files. Anyone auditing whether the exclusions are still needed is directed to a closed issue and may either (a) wrongly conclude the skips are obsolete and remove them, re-breaking CI, or (b) waste time reconstructing the real tracking issue. The same "until the issue is resolved" prose also appears in the sibling tests/pdf_js_test.rs:20; fixing both together is advisable, though this issue scopes to deepspeech_test.rs.

---

Resolution

Fixed in commit 4bf9c993 (branch worktree-agent-a0b3a919efbe97cd4).

The FIXME in tests/deepspeech_test.rs:8-9 now reads:

// FIXME: Ignoring these files temporarily due to parsing errors (originally https://github.com/dekobon/big-code-analysis/issues/83,
// now tracked in the open issue https://github.com/dekobon/big-code-analysis/issues/86), to keep CI green until they parse cleanly.

It cites #83 as the origin and the open #86 as the active tracker, mirroring tests/README.md:119. The #721 note block is unchanged. Net-zero SLOC; comment-only, no behavior change.

The sibling tests/pdf_js_test.rs:20 is intentionally out of scope: it cites a different issue (#84), so its correction is deferred to a separate change that can verify #84's state.

[dekobon]

General Plan for Fix

Root Cause

The skip-rationale comment was written when #83 was the active blocker. #83 has since been closed and the long-lived grammar-blocker tracking moved to #86 (and the two newly-parseable files were un-skipped in #723/#494ce4c3), but the FIXME's issue link and "until the issue is resolved" phrasing were never updated to follow that move.

Implementation Steps

1. In tests/deepspeech_test.rs:8-9, update the FIXME to cite the open tracking issue track(grammar): tree-sitter-cpp blockers behind mozcpp/deepspeech skips (#83) #86 for the five remaining excluded files, mirroring the tests/README.md:119 phrasing ("→ fix(tests,mozcpp): tree-sitter-cpp parse failures skip mozcpp/deepspeech tests #83, tracked in track(grammar): tree-sitter-cpp blockers behind mozcpp/deepspeech skips (#83) #86"). Keep the #721 note block as-is.
2. (Optional, same PR) Apply the identical correction to the sibling tests/pdf_js_test.rs:20, which carries the same stale "until the issue is resolved" prose.

Tests to Add/Update

• None. This is a comment-only change; no behavior changes. Verify it is net-zero-line on any file at its .bca-baseline.toml SLOC limit (loc.sloc counts comment lines).

Verification

• cargo test --workspace --all-features still passes (no behavior change)
• The FIXME now links the open tracking issue (track(grammar): tree-sitter-cpp blockers behind mozcpp/deepspeech skips (#83) #86) and no longer frames a closed issue as the live blocker
• Manual verification: clicking the cited issue link in the comment lands on an open issue describing the remaining cpp-grammar blockers

[dekobon]

Fixed in commit 4bf9c993 on branch worktree-agent-a0b3a919efbe97cd4.

The FIXME in tests/deepspeech_test.rs:8-9 now cites #83 as the origin and the open #86 as the active tracker for the five remaining excluded files, matching tests/README.md:119. The misleading "until the issue is resolved" framing (which pointed at the closed #83) is gone.

Confirmed before editing: #83 is CLOSED, #86 is OPEN.

Scope note: the sibling tests/pdf_js_test.rs:20 carries the same "until the issue is resolved" prose but cites a different issue (#84, not #83), so it is left out of this scoped fix — correcting it would require separately verifying #84's tracking state.

Verification:

• rustfmt --check clean
• cargo clippy --workspace --all-targets -- -D warnings clean
• cargo test --test deepspeech_test --all-features passes (comment-only, no behavior change)
• Net-zero SLOC (2 comment lines in, 2 out)