fix(parser): text signal hardening — noise-prefix strip + joined-prefix lookahead

[thewrz]

Summary

• Strips leading ], ]], [_____] noise prefixes from lines before signal cascade — fixes UFGS specifier-note bracket bleed where ] PART 2 PRODUCTS was silently classified as continuation
• Loosens ARTICLE_RE and PR_SIGNALS (pr1, pr2) to accept joined prefixes with no space — fixes CPI export pattern 1.3QUALITY, B.Included, 1.Manufacturers falling to continuation
• Uses (?=\S) lookahead for ARTICLE_RE (digit-letter boundary has no \b) and (?=[^\d\s]) for PR2 to prevent 1.1 bare from matching as pr2

Test plan

pnpm test src/parser/text/signals.test.ts   # 9 new tests + all prior tests (31 total)
pnpm test                                    # full suite, no regressions (432 tests)
pnpm lint                                    # clean

Closes #67

Summary by CodeRabbit

• Tests

  • Added comprehensive test coverage for text line classification with edge cases, including noise-prefix handling and joined-prefix detection.

• Bug Fixes

  • Improved parser robustness for handling unusual formatting patterns, bracketed prefixes, and ambiguous punctuation in document structure detection.

Test plan results

• pnpm test src/parser/text/signals.test.ts — 31 tests (9 new + 22 prior) passed
• pnpm test — 432 tests passed, no regressions
• pnpm lint — ESLint + tsc + prettier clean

[coderabbitai]

📝 Walkthrough

Walkthrough

This PR hardens the plaintext parser's text signal classification with two distinct fixes: (1) stripping noise prefixes (closing brackets, placeholders) that prevent structural keyword matching, and (2) loosening regex patterns and prefix stripping to handle joined prefixes where numbering or letters attach directly to content without whitespace.

Changes

Text signal hardening

Layer / File(s)	Summary
Noise prefix stripping for bracket and placeholder tokens src/parser/text/signals.ts, src/parser/text/signals.test.ts	Introduces NOISE_PREFIX_RE regex and stripNoisePrefixes helper to remove leading ], ]], and [_____] patterns. Integrates stripping into classifyLine before signal detection so structural keywords like PART and SECTION are recognized after noise removal. Tests validate that noise-prefixed lines (e.g., ] PART 2 PRODUCTS, ]] PART 3 EXECUTION) classify with correct type and text.
Joined-prefix regex and stripping updates src/parser/text/signals.ts, src/parser/text/signals.test.ts	Updates ARTICLE_RE and PR_SIGNALS (pr1, pr2) to match article and requirement prefixes immediately adjacent to content using word-boundary and lookahead patterns. Adjusts stripArticlePrefix and stripPrPrefix to remove prefixes without mandatory whitespace suffixes. Tests validate that joined patterns (1.3QUALITY, B.Included, 1.Manufacturers) classify correctly and return text with prefix stripped.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~12 minutes

Poem

A parser once stumbled on brackets that clung,
And prefixes merged into words without space—
Now noise fades away, and the signals ring true,
From ] PART to 1.3QUALITY, each finds its place. 🐰✨

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title directly and clearly describes the main changes: noise-prefix stripping and joined-prefix lookahead handling in the text signal parser.
Linked Issues check	✅ Passed	The PR fully implements both scope items from #67: noise-prefix stripping (NOISE_PREFIX_RE regex and stripNoisePrefixes helper in classifyLine) and joined-prefix lookahead (updated ARTICLE_RE and PR signal regexes), with comprehensive test coverage.
Out of Scope Changes check	✅ Passed	All changes are tightly scoped to the two stated failure modes: noise-prefix handling and joined-prefix regex loosening; test additions directly validate these fixes.
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch feat/issue-67-text-signal-hardening

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}