fix(parser): skip interpolated templates in disabled-feature recovery#889
Merged
Conversation
When the parser rejects a disabled traditional `for (;;)` loop, it records
the warning and skips the loop's header and body via the error-recovery
routines (SkipBalancedParens, SkipBlock, SkipUntilSemicolon). Those routines
counted braces token by token, but the `}` that closes a `${ ... }` template
substitution is lexed as a plain RightBrace under the default goal — the
TemplateMiddle/TemplateTail span is only produced when the lexer is invoked
with the template-tail goal *after* that `}`. So the skip miscounted the
substitution's `}` as the structural brace ending the skipped region, stopped
early, and the trailing backtick was re-scanned as a fresh, unterminated
template literal running to EOF. The result was a misleading
"Unterminated template literal" error at EOF instead of the real
"enable --compat-traditional-for-loop" diagnostic at the `for` keyword.
Add a shared SkipTemplateLiteral helper that skips a complete template literal
as a unit — recursing through nested substitutions and templates, balancing
inner object/block braces, and re-lexing each continuation span with the
template-tail goal (as ParseTemplateLiteral does). Route SkipBlock,
SkipBalancedParens, and SkipUntilSemicolon through it, which fixes the whole
class of disabled-feature recovery skips (for-loop body/header/non-block body,
and the sibling while / do-while constructs), not just the reported case.
Tests (both verified red without the parser change):
- tests/language/statements/unsupported-features.js: a disabled for-loop with
an interpolated template (block body, non-block body, header, nested) is
skipped as a clean no-op and subsequent code runs.
- scripts/test-cli-parser.ts: recovery succeeds in interpreter and bytecode
mode, never regresses to "Unterminated template literal", and the
human-readable diagnostic references --compat-traditional-for-loop.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
|
The latest updates on your projects. Learn more about Vercel for GitHub. 1 Skipped Deployment
|
Contributor
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (1)
📝 WalkthroughWalkthroughAdds template-literal-aware parser recovery and expands regression tests for disabled-feature parsing with interpolated template literals in Template-literal error recovery for disabled loops
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes Possibly related PRs
Suggested labels
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
Comment |
coderabbitai
Bot
added
bug
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
🧹 Nitpick comments (1)
scripts/test-cli-parser.ts (1)
191-255: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick winCover the shared
while/do...whilerecovery path too.
SkipTemplateLiteralis now exercised here only through disabledforrecovery, but the sameSkipBalancedParenspath also drives disabledwhileanddo...whilerecovery insource/units/Goccia.Parser.pas. Adding one regression per construct would lock down the broader contract this parser change now depends on.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@scripts/test-cli-parser.ts` around lines 191 - 255, The recovery coverage here only exercises the disabled traditional for-loop path, but the same SkipBalancedParens/SkipTemplateLiteral flow is also used by disabled while and do...while recovery in Goccia.Parser. Extend the recoveryCases in test-cli-parser.ts with one regression each for while and do...while template-literal recovery, and keep the same assertions around json.ok, exitCode, expected output, and the compat-flag diagnostic so the shared parser contract stays covered.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Nitpick comments:
In `@scripts/test-cli-parser.ts`:
- Around line 191-255: The recovery coverage here only exercises the disabled
traditional for-loop path, but the same SkipBalancedParens/SkipTemplateLiteral
flow is also used by disabled while and do...while recovery in Goccia.Parser.
Extend the recoveryCases in test-cli-parser.ts with one regression each for
while and do...while template-literal recovery, and keep the same assertions
around json.ok, exitCode, expected output, and the compat-flag diagnostic so the
shared parser contract stays covered.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro Plus
Run ID: a4db99c8-bb3b-4d37-9f50-010b63fae0de
📒 Files selected for processing (3)
scripts/test-cli-parser.tssource/units/Goccia.Parser.pastests/language/statements/unsupported-features.js
Contributor
Suite TimingTest Runner (interpreted: 10,962 passed; bytecode: 10,962 passed)
MemoryGC rows aggregate the main thread plus all worker thread-local GCs. Test runner worker shutdown frees thread-local heaps in bulk; that shutdown reclamation is not counted as GC collections or collected objects.
Benchmarks (interpreted: 436; bytecode: 436)
MemoryGC rows aggregate the main thread plus all worker thread-local GCs. Benchmark runner performs explicit between-file collections, so collection and collected-object counts can be much higher than the test runner.
Measured on ubuntu-latest x64. |
Contributor
Benchmark Results436 benchmarks · PR vs same-runner Interpreted: 🟢 44 improved · 🔴 50 regressed · 342 unchanged · avg +2.6% Typical per-run noise (median variance): interpreted ±4.3%, bytecode ±2.1%. Deltas within noise overlap and read as unchanged. arraybuffer.js — Interp: 14 unch. · avg -0.1% · Bytecode: 14 unch. · avg -1.3%
arrays.js — Interp: 🔴 6, 13 unch. · avg -1.2% · Bytecode: 🟢 1, 🔴 2, 16 unch. · avg +0.4%
async-await.js — Interp: 6 unch. · avg +1.5% · Bytecode: 🔴 1, 5 unch. · avg -0.1%
async-generators.js — Interp: 2 unch. · avg -2.9% · Bytecode: 2 unch. · avg -0.2%
atomics.js — Interp: 🔴 1, 5 unch. · avg -1.7% · Bytecode: 6 unch. · avg -1.7%
base64.js — Interp: 🟢 3, 7 unch. · avg +2.5% · Bytecode: 🟢 2, 🔴 1, 7 unch. · avg +0.5%
classes.js — Interp: 🟢 1, 🔴 4, 26 unch. · avg -1.9% · Bytecode: 🟢 4, 27 unch. · avg +1.6%
closures.js — Interp: 11 unch. · avg -0.5% · Bytecode: 11 unch. · avg +0.9%
collections.js — Interp: 🔴 1, 11 unch. · avg +1.1% · Bytecode: 12 unch. · avg +1.3%
csv.js — Interp: 🟢 2, 11 unch. · avg +4.8% · Bytecode: 🔴 1, 12 unch. · avg -0.5%
destructuring.js — Interp: 🟢 2, 🔴 4, 16 unch. · avg -2.0% · Bytecode: 🟢 3, 19 unch. · avg +1.0%
fibonacci.js — Interp: 🟢 1, 🔴 1, 6 unch. · avg -1.5% · Bytecode: 8 unch. · avg +0.5%
float16array.js — Interp: 🟢 7, 🔴 3, 22 unch. · avg +0.5% · Bytecode: 🔴 1, 31 unch. · avg +0.7%
for-in/for-in.js — Interp: 🟢 3 · avg +242.3% · Bytecode: 🟢 3 · avg +322.1%
for-of.js — Interp: 🔴 6, 1 unch. · avg -6.4% · Bytecode: 🔴 1, 6 unch. · avg -0.6%
generators.js — Interp: 4 unch. · avg -0.9% · Bytecode: 4 unch. · avg -3.3%
intl.js — Interp: 🟢 1, 🔴 1, 4 unch. · avg -1.3% · Bytecode: 6 unch. · avg +2.4%
iterators.js — Interp: 🟢 1, 🔴 5, 36 unch. · avg -0.3% · Bytecode: 🟢 2, 🔴 2, 38 unch. · avg +1.2%
json.js — Interp: 🟢 3, 20 unch. · avg +0.0% · Bytecode: 🟢 1, 🔴 2, 20 unch. · avg +0.0%
jsx.jsx — Interp: 🟢 1, 🔴 2, 18 unch. · avg +0.4% · Bytecode: 🔴 2, 19 unch. · avg -2.0%
modules.js — Interp: 9 unch. · avg -1.1% · Bytecode: 9 unch. · avg +1.6%
numbers.js — Interp: 🟢 2, 9 unch. · avg +1.8% · Bytecode: 🟢 1, 🔴 2, 8 unch. · avg +1.1%
objects.js — Interp: 🔴 1, 6 unch. · avg -1.2% · Bytecode: 7 unch. · avg -3.8%
promises.js — Interp: 🟢 2, 10 unch. · avg +5.0% · Bytecode: 🔴 1, 11 unch. · avg -0.1%
property-access.js — Interp: 5 unch. · avg -4.5% · Bytecode: 🟢 1, 4 unch. · avg +4.9%
regexp.js — Interp: 🟢 1, 🔴 1, 9 unch. · avg -1.4% · Bytecode: 🟢 1, 🔴 3, 7 unch. · avg -2.3%
strings.js — Interp: 🟢 1, 🔴 3, 15 unch. · avg -0.1% · Bytecode: 🟢 3, 🔴 1, 15 unch. · avg +1.3%
temporal.js — Interp: 🔴 1, 5 unch. · avg -2.8% · Bytecode: 🔴 2, 4 unch. · avg -3.2%
tsv.js — Interp: 🟢 1, 8 unch. · avg +3.8% · Bytecode: 🟢 1, 8 unch. · avg +1.8%
typed-arrays.js — Interp: 🟢 2, 🔴 7, 13 unch. · avg -3.3% · Bytecode: 🟢 1, 🔴 5, 16 unch. · avg -0.0%
uint8array-encoding.js — Interp: 🟢 7, 11 unch. · avg +26.4% · Bytecode: 🟢 4, 🔴 1, 13 unch. · avg -0.2%
weak-collections.js — Interp: 🟢 3, 🔴 3, 9 unch. · avg +5.1% · Bytecode: 🟢 3, 🔴 1, 11 unch. · avg +8.5%
Deterministic profile diffDeterministic profile diff: no significant changes. Measured on ubuntu-latest x64. Each PR run also builds the |
Contributor
test262 Conformance
Areas closest to 100%
Per-test deltas (+2 / -0)Newly passing (2):
Steady-state failures are non-blocking; regressions vs the cached main baseline (lower total pass count, or any PASS → non-PASS transition) fail the conformance gate. Measured on ubuntu-latest x64, bytecode mode. Areas grouped by the first two test262 path components; minimum 25 attempted tests, areas already at 100% excluded. Δ vs main compares against the most recent cached |
The disabled while and do...while recovery paths share the same SkipBalancedParens / SkipBlock + SkipTemplateLiteral flow as the traditional for-loop, so extend the CLI parser recovery cases with one regression each: a while loop with an interpolated template in its condition and a do...while loop with one in its body. Each case now carries its own compat flag and the human-readable diagnostic assertion runs per case, so json.ok, exit code, expected output, and the compat-flag diagnostic stay covered across the whole shared parser contract. Co-Authored-By: Claude Opus 4.8 <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.
Summary
for (;;)loop containing an interpolated template literal (e.g.obj[`u${level}`] = level;) reported a misleadingSyntaxError: Unterminated template literalat EOF instead of the real "enable--compat-traditional-for-loop" diagnostic at theforkeyword.SkipBalancedParens/SkipBlock/SkipUntilSemicolon. These counted braces naively, but the}that closes a${ … }substitution is lexed as a plainRightBraceunder the default goal (theTemplateMiddle/TemplateTailspan is only produced when the lexer runs with the template-tail goal after that}). The skip miscounted the substitution's}as the structural brace ending the region, stopped early, and the trailing backtick was re-scanned as a fresh, unterminated template literal running to EOF.SkipTemplateLiteralhelper that skips a complete template literal as a unit — recursing through nested substitutions/templates, balancing inner object/block braces, and re-lexing each continuation span with the template-tail goal (mirroringParseTemplateLiteral). Route the three recovery skips through it.${}, and the siblingwhile/do-whileconstructs, while the enabled--compat-traditional-for-looppath still parses and executes templates correctly.Testing
Added tests (both verified red without the parser change):
tests/language/statements/unsupported-features.js— a disabled for-loop with an interpolated template (block body, non-block body, header, nested) is skipped as a clean no-op and subsequent code runs.scripts/test-cli-parser.ts— recovery succeeds in interpreter and bytecode mode, never regresses toUnterminated template literal, and the human-readable diagnostic references--compat-traditional-for-loop.Verification run:
bun scripts/test-cli-parser.ts./format.pas --checkSyntaxError: Unterminated template literal(0 passed / 1 failed) — confirms a genuine regression test🤖 Generated with Claude Code