Reduce ExtractWorkflowNameFromFile overhead in CLI workflow title parsing

[Copilot]

ExtractWorkflowNameFromFile showed a measurable regression versus historical benchmark levels, with time spent in line-by-line title extraction. This change tightens the hot path in fastParseTitleFromReader to reduce per-line work without altering parsing behavior.

• Hot-path optimization in title scanning

  • Switched scanner processing from string-based operations to byte-slice operations.
  • Replaced scanner.Text() + strings.TrimSpace/HasPrefix with scanner.Bytes() + bytes.TrimSpace/HasPrefix/Equal.
  • Preserves existing semantics for frontmatter detection (---) and H1 extraction (# ), while reducing transient allocations.

• Scope and behavior

  • Changes are isolated to /pkg/cli/workflows.go in fastParseTitleFromReader.
  • No changes to workflow name fallback rules or frontmatter error behavior.

trimmed := bytes.TrimSpace(scanner.Bytes())

if bytes.Equal(trimmed, frontmatterDelimiter) { ... }
if bytes.HasPrefix(trimmed, h1Prefix) {
    return string(bytes.TrimSpace(trimmed[2:])), nil
}

[github-actions]

PR Triage - Run 29041860272. Category: chore | Risk: Low | Score: 28/100 | Action: defer. DRAFT. Micro hot-path optimization in ExtractWorkflowNameFromFile (8+/5-, 1 file). No behavior change. Undraft when ready; low priority.

Generated by 🔧 PR Triage Agent · 78.2 AIC · ⌖ 8.53 AIC · ⊞ 5.4K · ◷

[github-actions]

🤖 PR Triage — Run §29061171191

Field	Value
Category	chore
Risk	🟢 Low
Score	28/100 (Impact 10 + Urgency 8 + Quality 10)
Action	🔃 defer

Reduce ExtractWorkflowNameFromFile overhead. DRAFT, 1 file, 8 additions. Minor perf chore. Low priority.

Generated by 🔧 PR Triage Agent · 155.8 AIC · ⌖ 8.06 AIC · ⊞ 5.4K · ◷

[github-actions]

Thanks for the performance optimization in fastParseTitleFromReader! Switching to byte-slice operations is a solid improvement that reduces transient allocations.

One suggestion before merging:

• Add a benchmark test — since this fixes a measured regression, a BenchmarkExtractWorkflowNameFromFile in the test file would confirm the improvement and guard against future regressions.

If you would like a hand:

Add a benchmark test for ExtractWorkflowNameFromFile in pkg/cli/workflows_test.go (or a new _bench_test.go file).
Cover both frontmatter and H1 title extraction paths to confirm the byte-slice optimization reduces allocations.
Run with: go test -bench=. -benchmem ./pkg/cli/...

Generated by ✅ Contribution Check · 103.9 AIC · ⌖ 11.3 AIC · ⊞ 6.2K · ◷

[pelikhan]

@copilot extract the first header h1, h2, or h3 as the title

[Copilot]

Pull request overview

This PR optimizes CLI workflow title parsing by tightening the hot path in fastParseTitleFromReader, aiming to reduce per-line overhead in ExtractWorkflowNameFromFile without changing parsing semantics.

Changes:

• Switched from scanner.Text() + strings.* operations to scanner.Bytes() + bytes.* operations to reduce transient allocations during line scanning.
• Preserved existing behavior for frontmatter delimiter detection (---) and H1 title extraction (# ), while keeping early-exit scanning.

Show a summary per file

File	Description
pkg/cli/workflows.go	Reworks title scanning to operate on byte slices (bytes.TrimSpace/HasPrefix/Equal) instead of strings to reduce per-line work in the title extraction hot path.

Review details

Tip

Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

• Files reviewed: 1/1 changed files
• Comments generated: 1
• Review effort level: Low

[Copilot]

@copilot extract the first header h1, h2, or h3 as the title

Implemented in bf2f443. Title extraction now returns the first Markdown header among H1, H2, or H3 (while keeping existing frontmatter handling and filename fallback behavior), and tests were updated for H2/H3 and H4 fallback cases.