[dinwwwh]

Requested by dinwwwh · project thread

What

matchesSegment was a one-line function used in two places: matchesCatchAll, for the segments after a catch-all param, and the closure that createPrefixMatcher returns. Both now do the comparison in place, with a comment that undefined stands for a param, which needs a non-empty segment.

Why

Each top-level function costs V8 about 15k instructions to set up at import, and about as much again to compile when it is first called (see #198). This removes one function from every import. It also removes one compile from the first request that reaches a catch-all, or a prefix with a param.

Measurements

Instruction counts from callgrind, node --single-threaded, averaged over hash seeds 1 and 2:

	before	after
min / min+gzip	5043 / 2100 B	5053 / 2098 B
import	2.167M	2.164M
REST cold+10 (includes a catch-all request)	6.041M	6.017M (−0.4%)
steady, catch-all / tail	4985 / 5256	4979 / 5249
steady, prefix matcher with a param / with %	2453 / 4082	2422 / 4052 (−1%)

Cold+3 on the REST set reaches no catch-all, so it changes only by the import saving.

Checks

• Tests pass with 100% coverage.
• The model fuzzers, including the late-add and prefix ones, find 0 diffs.
• Stryker finds 68 survivors, the same as on main. None are on the new comparisons.

---

Generated by Claude Code

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[codspeed]

Merging this PR will not alter performance

1 warning

⚠️ 1 benchmark is dominated by system calls

System calls cannot be simulated consistently and are left out of the measure. The reported value understates the real cost. Switch to the Walltime instrument to measure them.

Learn more about system calls

See affected benchmarks

✅ 29 untouched benchmarks
⏩ 2 skipped benchmarks¹

---

_{Comparing claude/project-thread-m32ra9 (74625ac) with main (cd598d9)}

Footnotes

1. 2 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

[pullfrog]

✅ No new issues found.

Reviewed changes

• matchesSegment inlined at both call sites — the module-private expected === undefined ? segment !== '' : segment === expected helper is removed and its negation written directly in createPrefixMatcher's closure and in matchesCatchAll's tail loop. The inlined ternaries are the exact De Morgan negation of the helper, so behavior is unchanged; I confirmed both sites and ran the router suite (73 tests pass).
• Comment added documenting that undefined stands for a param requiring a non-empty segment.

The refactor is mechanical and provably behavior-preserving: !matchesSegment expands to expected === undefined ? segment === '' : segment !== expected on both branches. matchesSegment was module-private with only these two callers (impact scan found no tracked references), and no public export changed.

  ^{｜ View workflow run ｜ Using deepseek-v4.1-flash (free via Pullfrog for OSS) ｜ 𝕏}