fix(ingestion): lazy-load optional grammars so analyze never crashes when one is missing (#2091, #2093) by magyargergo · Pull Request #2101 · abhigyanpatwari/GitNexus

magyargergo · 2026-06-08T21:34:02Z

Problem

gitnexus analyze aborts at module-load with ERR_MODULE_NOT_FOUND for an optional tree-sitter grammar (e.g. tree-sitter-swift) on any install where that grammar's native binding is absent — regardless of the repo's actual languages. This blocks all indexing (incl. the .githooks auto-reindex), the MCP server, and doctor.

Closes #2091. Closes #2093.

Root cause

The optional grammars are swift / dart / kotlin (cobol is regex-based with no grammar; vue reuses the TypeScript grammar; tree-sitter-proto is a guarded gRPC-extractor concern, not an ingestion language). Each languages/<lang>/query.ts did a top-level static import X from 'tree-sitter-Y'. The scope-resolution registry (scope-resolution/pipeline/registry.ts) and the language-provider index (languages/index.ts) statically import every provider, so those query.ts modules are pulled onto the main thread at module-load — and the static grammar import throws ERR_MODULE_NOT_FOUND before any runtime gate can run.

parse-worker.ts, parser-loader.ts, and the optional-grammars warning helper were all already guarded; only these three query.ts modules defeated the design. GITNEXUS_SKIP_OPTIONAL_GRAMMARS didn't help because the static import fails before any runtime check, and that env was install-time-only.

Changes

Lazy grammar load (the fix): the three query.ts getters (getXParser/getXScopeQuery, consumed only by the sibling captures.ts) now resolve the grammar through the existing lazy, guarded parser-loader.getLanguageGrammar(). No grammar is required at module-load — only at first use, inside the worker, for a file of that language.
Runtime opt-out: parser-loader now honors GITNEXUS_SKIP_OPTIONAL_GRAMMARS at runtime (1/true/all/* or a comma list of lang-ids / tree-sitter-* package names). Required deps routed through the optional machinery for ABI safety (C, severity:'error') are never skippable this way.
Pipeline exclusion: the scope-resolution phase now also filters files by isLanguageAvailable (mirroring the parse phase), so an unavailable/opted-out language's files are cleanly excluded instead of falling through to the main-thread re-extract (which logged a noisy "Unsupported language" and needlessly loaded the grammar on the main thread).
Accurate diagnostics: when a grammar is skipped via the runtime opt-out, the parse-phase warning says so instead of suggesting npm rebuild a grammar that built fine. Added kotlin (.kt/.kts) to the optional-grammar warning list and widened the analyze precheck glob to .swift/.kt/.kts (was .dart/.proto only).

Testing

New test/integration/optional-grammars/registry-import-closure.test.ts: importing the built scope-resolution registry loads no optional grammar binding (with a non-vacuity guard that a required binding still loads, proving the import chain is still walked). Characterization-first — fails on the pre-fix static import.
New test/integration/optional-grammars/skip-optional-pipeline.test.ts: drives the real pipeline over a Python+Swift repo with the opt-out set — Python indexes, Swift is cleanly excluded, no "scope extraction failed" noise.
New test/unit/parser-loader-skip-optional.test.ts: runtime skip-gate semantics (install-state-robust).
Verified locally: tsc + eslint clean; targeted + full scope-resolution suites pass (1189 tests); end-to-end analyze works both with the grammar present (Swift indexed) and with it skipped (Swift excluded, exit 0, no crash).

🤖 Generated with Claude Code

…when one is missing (abhigyanpatwari#2091, abhigyanpatwari#2093) The scope-resolution registry and the language-provider index statically import every language provider; each optional grammar's query.ts did a top-level `import X from 'tree-sitter-Y'` (swift/dart/kotlin). On a default install where the vendored/optional native binding is absent, that import resolution threw ERR_MODULE_NOT_FOUND at module-load on the main thread — before any runtime gate — crashing `gitnexus analyze` regardless of the repo's actual languages. The worker, parser-loader, and optional-grammars warning helper were all already guarded; only these three query.ts modules defeated the design. - Route the three query.ts getters through the existing lazy, guarded parser-loader.getLanguageGrammar(), so a grammar binding is only required at first use (inside the worker, for a file of that language) — never at module-load. This removes the crash for analyze, the MCP server, and doctor. - Honor GITNEXUS_SKIP_OPTIONAL_GRAMMARS at runtime in parser-loader (it was an install-time-only env): a runtime opt-out for swift/dart/kotlin. Required deps routed through the optional machinery for ABI safety (C, severity:'error') are never skippable this way. - Exclude unavailable-grammar files at the scope-resolution phase too, mirroring the parse phase, so a skipped-language file never falls through to the main-thread re-extract (which would log a noisy "Unsupported language" and needlessly load the grammar on the main thread). - When a grammar is skipped via the runtime opt-out (not a genuinely-missing binding), the parse-phase warning says so instead of suggesting `npm rebuild`. - Close the warning-coverage gaps: add kotlin (.kt/.kts) to the optional-grammar warning list and widen the analyze precheck glob to .swift/.kt/.kts. - Tests: registry static-import-closure regression (no optional grammar binding loads — with a non-vacuity guard that a required binding still does — when importing the scope-resolution registry); a pipeline-level regression that the scope-resolution phase cleanly excludes an opted-out grammar (no "scope extraction failed" noise) while still indexing available languages; and parser-loader runtime skip-gate unit tests. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

vercel · 2026-06-08T21:34:06Z

@magyargergo is attempting to deploy a commit to the NexusCore Team on Vercel.

A member of the Team first needs to authorize it.

github-actions · 2026-06-08T21:52:31Z

CI Report

✅ All checks passed

Pipeline Status

Stage	Status	Details
✅ Typecheck	`success`	tsc --noEmit
✅ Tests	`success`	unit tests, 3 platforms
✅ E2E	`success`	gitnexus-web changes only

Test Results

Tests	Passed	Failed	Skipped	Duration
10754	10738	0	16	520s

✅ All 10738 tests passed

16 test(s) skipped — expand for details

COBOL pipeline benchmark > scales with file count
C++ ADL emit benchmark > emit phase scales sub-quadratically with co-scaled files and sites
C++ pipeline benchmark > scales with file count
C# pipeline benchmark > scales with file count — namespaces spread across the solution
C# pipeline benchmark > scales with file count — all types in one (global) namespace bucket
C# pipeline benchmark > scales with file count — all types in one (named) namespace bucket
Go pipeline benchmark > scales with file count (workers enabled)
Go pipeline benchmark — worker pool (issue Worker idle timeout kills long Go scope extraction and surfaces as Napi::Error during analyze #1848) > does not quarantine the large generated Go file on sub-batch idle timeout
Go structural interface detection benchmark > scales linearly with interface × struct count
Go structural interface detection split-phase benchmark > separates index-build and detection time
PHP pipeline benchmark > scales with file count (workers enabled)
Ruby pipeline benchmark > scales with file count (workers enabled)
Rust pipeline benchmark > scales with file count (workers enabled)
Vue pipeline benchmark > scales with component count
run.cjs direct-exec entrypoint (fix(cli): steer docs, skills, and hooks through a CLI-neutral project-local runner (#1939) #1945) > resolves a .cmd shim via the Windows shell branch, passing args and exit code
buildTypeEnv > known limitations (documented skip tests) > Ruby block parameter: users.each { |user| } — closure param inference, different feature

Code Coverage

Tests

Metric	Coverage	Covered	Base	Delta	Status
Statements	74.92%	34892/46569	N/A%	—	🟢 ██████████████░░░░░░
Branches	62.7%	21599/34448	N/A%	—	🟢 ████████████░░░░░░░░
Functions	80.7%	3773/4675	N/A%	—	🟢 ████████████████░░░░
Lines	78.74%	31561/40082	N/A%	—	🟢 ███████████████░░░░░

_{📋 View full run · Generated by CI}

magyargergo

Tri-review — optional-grammar lazy-load fix

Methods (3): GitNexus swarm (risk, test/CI) + Compound-Engineering personas (correctness, adversarial, performance, maintainability, testing) + Codex (live). Engine breakdown: 7 Claude lanes + 1 Codex. Only Codex is a different engine; agreement among the Claude lanes is "consistent across personas," not independent confirmation.

Verdict: no correctness/crash bugs — the core fix is sound

All three methods independently confirmed:

The lazy getLanguageGrammar() returns the same object the old static default-import did (CJS module.exports; no .default interop) — no parse-result regression for repos that have the grammars.
No circular import (query.ts → parser-loader is leaf-direction).
Required-grammar (C) failures are not newly swallowed — still logged at error severity in loadGrammar before the scope-phase skip.
isRuntimeSkippedGrammar env-parsing handles all documented forms.

Adversarial + correctness refuted every break attempt (C silent-drop, circular import, env abuse, worker bypass, test false-pass). Refuted nit: the not.toContain('Foo')/Struct assertion is meaningful — swift structs are labeled Struct.

Worth addressing (all P2/P3 — diagnostics, test-robustness, maintainability)

Inline comments mark the six in-diff items. Plus, in body (unchanged file):

[P2 · Codex+Claude] Worker ignores GITNEXUS_SKIP_OPTIONAL_GRAMMARS — parse-worker.ts:38-56,388-405 keeps its own grammar table. Harmless today (the main-thread parseableScanned filter excludes skipped-grammar files before dispatch), but the enforcement boundary is undocumented; a future change dispatching files without that pre-filter would silently parse a "skipped" grammar. Worth a comment now; full consolidation (route the worker through parser-loader) is a planned follow-up.

Lower-priority P3 (not inlined): memoize the env-parse in isRuntimeSkippedGrammar; add a direct unit test for isGrammarRuntimeSkipped; naming consistency isGrammarRuntimeSkipped ↔ isLanguageAvailable; include result.signal in the registry-test diagnostic; factor the 3 near-identical query.ts comment/getter blocks.

Residual (pre-existing, out of scope)

The worker statically import C from 'tree-sitter-c', so a broken C ABI still crashes all workers — crash-immunity doesn't fully extend to required grammars in the worker (ties into the worker-consolidation follow-up).
COBOL emits a spurious "npm rebuild tree-sitter-cobol" skip warning (regex parser, no grammar).

CI

Green except a non-code Vercel deploy-auth failure and a pending Build&Push. tsc + eslint clean; targeted + scope-resolution suites pass locally.

Automated 3-tool digest (7 Claude lanes + Codex). 11-file diff fully read. Verify before acting.

magyargergo · 2026-06-08T22:11:06Z

+    const result: LoadResult = {
+      ok: false,
+      error: new Error('skipped via GITNEXUS_SKIP_OPTIONAL_GRAMMARS'),
+      note: source.unavailableNote,


[P3 · Codex+Claude · reproduced] Under a deliberate GITNEXUS_SKIP_OPTIONAL_GRAMMARS opt-out, this skip branch reuses source.unavailableNote ("…no prebuilt .node… failed to load") — which contradicts the opt-out; the (skipped via …) suffix is the only hint. Observed live in the skip unit-test log output. Fix: in the skip branch use a dedicated note (e.g. "disabled via GITNEXUS_SKIP_OPTIONAL_GRAMMARS — unset to re-enable") instead of source.unavailableNote.

magyargergo · 2026-06-08T22:11:06Z

+const isRuntimeSkippedGrammar = (key: string, source: GrammarSource): boolean => {
+  // Only genuinely-optional grammars (optionalDependencies / vendored), never
+  // required deps that merely use the optional machinery for ABI safety.
+  if (source.optional !== true || source.severity === 'error') return false;


[P2 · maintainability · code-read] Skippability is inferred from optional !== true || severity === 'error'. A future required grammar added as optional:true without severity:'error' would silently become user-skippable. Consider an explicit userSkippable?: boolean on GrammarSource and gate on that — the C-vs-(swift/dart/kotlin) invariant then reads off the SOURCES table instead of a predicate.

magyargergo · 2026-06-08T22:11:06Z

  // degraded index.
  try {
-    const matches = await glob(['**/*.dart', '**/*.proto'], {
+    const matches = await glob(['**/*.dart', '**/*.proto', '**/*.swift', '**/*.kt', '**/*.kts'], {


[P2 · maintainability · code-read] This hardcoded glob duplicates OPTIONAL_GRAMMARS[].extensions (optional-grammars.ts) — they must be kept in sync by hand. Export e.g. getOptionalGrammarExtensions() and derive the glob so a future grammar addition tracks automatically.

magyargergo · 2026-06-08T22:11:06Z

+  });
+
+  it('skips the Swift file at the parse phase (non-vacuity: Swift was present)', () => {
+    expect(messages.some((m) => /Skipping 1 swift file\(s\)/.test(m))).toBe(true);


[P2 · testing+adversarial · code-read] /Skipping 1 swift file\(s\)/ is a prefix shared by BOTH the new opt-out message and the old "npm rebuild" message, so the isGrammarRuntimeSkipped routing branch in parse-impl can regress silently. Also assert messages.some(m=>/GITNEXUS_SKIP_OPTIONAL_GRAMMARS/.test(m)) and messages.every(m=>!/npm rebuild/.test(m)).

magyargergo · 2026-06-08T22:11:06Z

+
+  beforeAll(async () => {
+    prevEnv = process.env[ENV];
+    process.env[ENV] = 'swift';


[P2 · Codex+Claude · code-read] Env set in beforeAll without vi.resetModules(); parser-loader's loadCache is a module singleton, so this is safe only via fork-isolation + the fact nothing calls isLanguageAvailable(Swift) before this. Add vi.resetModules() (then dynamic-import the pipeline), or document the constraint, so a later import-order change can't make it pass vacuously.

magyargergo · 2026-06-08T22:11:06Z

+    expect(skipped.isLanguageAvailable(SupportedLanguages.Python)).toBe(pyBase);
+  });
+
+  it('a comma list skips only the named grammars (language-id form)', async () => {


[P2 · testing · code-read] "skips only the named grammars" asserts Swift is skipped but not that Dart/Kotlin are unaffected — a prefix/union bug wouldn't be caught. Add baseline-comparison assertions that the un-named optional grammars are unchanged (same pattern as the C/Python case).

…r handling (abhigyanpatwari#2101) Polish + test-hardening from the PR abhigyanpatwari#2101 tri-review (no correctness bugs were found; the lazy-load crash fix is unchanged). All P2/P3: - parser-loader: a runtime-skipped grammar now logs an accurate "disabled via GITNEXUS_SKIP_OPTIONAL_GRAMMARS (unset to re-enable)" note instead of reusing the install/platform `unavailableNote` ("no prebuilt .node … failed to load"), which misled a deliberate opt-out. - parser-loader: make skippability explicit — add `userSkippable` to GrammarSource (set on swift/dart/kotlin) and gate the runtime opt-out on it, replacing the implicit `optional && severity!=='error'` predicate. C (required dep via the optional machinery, severity:'error') stays non-skippable. - parser-loader: parse GITNEXUS_SKIP_OPTIONAL_GRAMMARS once (module-scope memo) instead of re-reading + re-allocating a Set on every call. - cli: derive the analyze precheck glob from OPTIONAL_GRAMMARS (no hand-synced duplicate extension list), and make the preflight runtime-skip-aware so an installed-but-opted-out grammar gets an accurate upfront note (reason: skipped) rather than silence — `getOptionalGrammarExtensions()` + a `reason` on MissingGrammar. - parse-worker: document the enforcement boundary — the worker intentionally ignores the runtime env; the main-thread `parseableScanned` filter excludes opted-out files before dispatch. (Comment only; the worker→parser-loader consolidation is the deferred Tier-1 follow-up.) - tests: assert the opt-out message routing (env-var named, no "npm rebuild"); harden the pipeline test's module-cache coherence (vi.resetModules + dynamic import); add comma-list isolation, direct isGrammarRuntimeSkipped, and accurate-note assertions; include result.signal in the registry-test diagnostic. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

…olute path (abhigyanpatwari#2111) The recurring Windows `EPERM: operation not permitted, symlink` (errno -4048) when adding the MCP server to Antigravity is NOT the abhigyanpatwari#2101/abhigyanpatwari#2110 module-load crash — it is an install-time arborist failure during the `_npx` reify that the MCP client triggers on every `npx gitnexus` launch. Root cause: the `postinstall` materialize step copied each vendored grammar (`vendor/tree-sitter-{c,dart,proto,swift,kotlin}`) into `node_modules/gitnexus/node_modules/tree-sitter-*` as a real package so runtime `require('tree-sitter-dart')` would resolve. Those packages are in no dependency graph, so every subsequent npm/npx reify treats them as **extraneous** and prunes/relocates them — on Windows the relocation goes through `@npmcli/move-file`'s symlink path and throws EPERM (symlinks need Developer Mode/admin), and on every OS the 2nd run silently deletes the grammars. This is the same class as abhigyanpatwari#1728, which the materialize step itself claimed to have fixed. Fix (the prebuildify + node-gyp-build ecosystem pattern): never copy grammars into node_modules. Load each by absolute path from `vendor/<name>` via the new `requireVendoredGrammar` helper — the grammar's own `bindings/node` runs `node-gyp-build(<dir>)` and loads the committed `vendor/<name>/prebuilds/ <platform>-<arch>/…` directly (all 5 ship all 6 tuples). vendor/ is inside the package but not a node_modules subtree, so arborist never sees the grammars and the reify is idempotent — no EPERM, no silent deletion. - new src/core/tree-sitter/vendored-grammars.ts (requireVendoredGrammar / vendoredGrammarDir / VENDORED_GRAMMAR_PACKAGES; VENDOR_ROOT stable in dev+dist) - route all consumers through it: parser-loader, parse-worker, grpc proto, include-extractor (C), http-patterns kotlin, cli optional-grammars probe - postinstall drops the materialize step; build-tree-sitter-grammars.cjs builds in-place under vendor/ (gitignored) and deletes materialize-vendor-grammars.cjs - tests + grammar-introspection helper load grammars from vendor/ too (single source of truth); new vendored-grammars.test.ts guards against reintroducing a bare `require('tree-sitter-<vendored>')` Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

…olute path (#2111) (#2144) * fix(grammars): load vendored tree-sitter grammars from vendor/ by absolute path (#2111) The recurring Windows `EPERM: operation not permitted, symlink` (errno -4048) when adding the MCP server to Antigravity is NOT the #2101/#2110 module-load crash — it is an install-time arborist failure during the `_npx` reify that the MCP client triggers on every `npx gitnexus` launch. Root cause: the `postinstall` materialize step copied each vendored grammar (`vendor/tree-sitter-{c,dart,proto,swift,kotlin}`) into `node_modules/gitnexus/node_modules/tree-sitter-*` as a real package so runtime `require('tree-sitter-dart')` would resolve. Those packages are in no dependency graph, so every subsequent npm/npx reify treats them as **extraneous** and prunes/relocates them — on Windows the relocation goes through `@npmcli/move-file`'s symlink path and throws EPERM (symlinks need Developer Mode/admin), and on every OS the 2nd run silently deletes the grammars. This is the same class as #1728, which the materialize step itself claimed to have fixed. Fix (the prebuildify + node-gyp-build ecosystem pattern): never copy grammars into node_modules. Load each by absolute path from `vendor/<name>` via the new `requireVendoredGrammar` helper — the grammar's own `bindings/node` runs `node-gyp-build(<dir>)` and loads the committed `vendor/<name>/prebuilds/ <platform>-<arch>/…` directly (all 5 ship all 6 tuples). vendor/ is inside the package but not a node_modules subtree, so arborist never sees the grammars and the reify is idempotent — no EPERM, no silent deletion. - new src/core/tree-sitter/vendored-grammars.ts (requireVendoredGrammar / vendoredGrammarDir / VENDORED_GRAMMAR_PACKAGES; VENDOR_ROOT stable in dev+dist) - route all consumers through it: parser-loader, parse-worker, grpc proto, include-extractor (C), http-patterns kotlin, cli optional-grammars probe - postinstall drops the materialize step; build-tree-sitter-grammars.cjs builds in-place under vendor/ (gitignored) and deletes materialize-vendor-grammars.cjs - tests + grammar-introspection helper load grammars from vendor/ too (single source of truth); new vendored-grammars.test.ts guards against reintroducing a bare `require('tree-sitter-<vendored>')` Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(grammars): throw on a non-vendored name in requireVendoredGrammar Drift guard (PR #2144 review, P3): validate the argument against VENDORED_GRAMMAR_PACKAGES and fail loudly on an unknown name, so the three grammar lists (package set / CLI probe / build registry) drifting out of sync surfaces as a clear error instead of a confusing absolute-path require miss. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(grammars): prepack guard against stray vendor/<g>/build/ shadowing prebuilds Publish hygiene (PR #2144 review, P2). Now that build-tree-sitter-grammars.cjs source-builds into vendor/<name>/build/, a stray build dir would ship in the tarball (files:["vendor"] overrides .gitignore/.npmignore) AND shadow the committed prebuild — node-gyp-build resolves build/Release before prebuilds/. assert-publish-grammar-coverage.cjs (prepack) now fails `npm pack` if any vendor/*/build exists (findStrayBuildArtifacts), with a clear `rm -rf` fix hint. Adds unit coverage for the new pure function. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test(grammars): harden the #2111 no-bare-require regression guard PR #2144 review (P2). The guard regex missed dynamic import(), side-effect `import 'x'`, /subpath, and backtick loads, and only scanned src/. It now covers every node_modules-forcing form (single/double/backtick quotes, optional subpath), scans test/ too (excluding fixtures and the guard file itself), drops the `//`-substring false-negative (leading-comment-only heuristic), and adds a self-test asserting every load form is caught while prose mentions and tree-sitter-cpp are ignored. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * docs(grammars): correct stale vendored-grammar comments PR #2144 review (P3). kotlin/query.ts called tree-sitter-kotlin an "optionalDependency" — it is vendored and loaded from vendor/ by absolute path (#2111). proto.ts now states its remaining `_require` is only for the real `tree-sitter` dependency, not a vendored grammar (which goes through requireVendoredGrammar). Comment-only; no behavior change. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

magyargergo commented Jun 8, 2026

View reviewed changes

magyargergo merged commit 4de4d20 into abhigyanpatwari:main Jun 9, 2026
27 of 28 checks passed

magyargergo deleted the fix/optional-grammars-lazy-load branch June 9, 2026 03:58

magyargergo mentioned this pull request Jun 9, 2026

1.6.6 fails to launch #2095

Closed

magyargergo linked an issue Jun 9, 2026 that may be closed by this pull request

1.6.6 fails to launch #2095

Closed

magyargergo mentioned this pull request Jun 9, 2026

npm error code EPERM when add mcp to antigravity, v1.6.6 #2111

Closed

magyargergo linked an issue Jun 9, 2026 that may be closed by this pull request

npm error code EPERM when add mcp to antigravity, v1.6.6 #2111

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

fix(ingestion): lazy-load optional grammars so analyze never crashes when one is missing (#2091, #2093)#2101

fix(ingestion): lazy-load optional grammars so analyze never crashes when one is missing (#2091, #2093)#2101
magyargergo merged 2 commits into
abhigyanpatwari:mainfrom
magyargergo:fix/optional-grammars-lazy-load

magyargergo commented Jun 8, 2026

Uh oh!

vercel Bot commented Jun 8, 2026

Uh oh!

github-actions Bot commented Jun 8, 2026 •

edited

Loading

Uh oh!

magyargergo left a comment

Uh oh!

magyargergo Jun 8, 2026

Uh oh!

magyargergo Jun 8, 2026

Uh oh!

magyargergo Jun 8, 2026

Uh oh!

magyargergo Jun 8, 2026

Uh oh!

magyargergo Jun 8, 2026

Uh oh!

magyargergo Jun 8, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Uh oh!

Conversation

magyargergo commented Jun 8, 2026

Problem

Root cause

Changes

Testing

Uh oh!

vercel Bot commented Jun 8, 2026

Uh oh!

github-actions Bot commented Jun 8, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

CI Report

Pipeline Status

Test Results

Code Coverage

Tests

Uh oh!

magyargergo left a comment

Choose a reason for hiding this comment

Tri-review — optional-grammar lazy-load fix

Verdict: no correctness/crash bugs — the core fix is sound

Worth addressing (all P2/P3 — diagnostics, test-robustness, maintainability)

Residual (pre-existing, out of scope)

CI

Uh oh!

magyargergo Jun 8, 2026

Choose a reason for hiding this comment

Uh oh!

magyargergo Jun 8, 2026

Choose a reason for hiding this comment

Uh oh!

magyargergo Jun 8, 2026

Choose a reason for hiding this comment

Uh oh!

magyargergo Jun 8, 2026

Choose a reason for hiding this comment

Uh oh!

magyargergo Jun 8, 2026

Choose a reason for hiding this comment

Uh oh!

magyargergo Jun 8, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

github-actions Bot commented Jun 8, 2026 •

edited

Loading