feat(editor): grouping-aware click, ruler ticks, and comb (P2-6)

[ChrisBeWithYou]

⚠ STACKED on #276 (P2-5 authored marks) — base is feat/editor-markers-hold-group. After #276 squash-merges: git rebase --onto main feat/editor-markers-hold-group feat/editor-grouping-aware, retarget, ancestry check (the #230 ritual).

What

P2-6 — the grouping field's three consumers, so an authored 7/8 (2+2+3) actually teaches the riff instead of just labeling it:

1. Metronome accents the grouping-cell starts — 2+2+3 clicks strong-weak-strong-weak-strong-weak-weak, so you hear where the pick pattern resets.
2. Ruler group-ticks — accents draw bright at full tick density, and at far zoom (where per-beat ticks are dropped but the bar still has ≥28 px) the few sub-bar ticks are spent on the accents instead of vanishing.
3. The suggest comb corroborates on the FELT pulse — accented positions carry double weight, so a candidate barline whose onsets land on the 2+2+3 accents out-scores one that merely matches an even seven. A wrong grouping (3+2+2 against a 2+2+3 performance) scores below the right one — the detector genuinely prefers the authored feel.

No grouping = bit-identical behavior everywhere (pinned by test): the accent map degrades to downbeat-only, the comb's uniform weights reproduce the exact old mean, the ruler draws what it always drew.

How

• _groupingAccentMapPure([2,2,3], 7) → [1,0,1,0,1,0,0] + a by-measure lookup in tempo-marks.js; the live lookup (_groupingAccentsLive) is memoized on the immutable marks-array identity — no per-frame builds on the draw/scheduler hot paths.
• _metroClicksInWindowPure gains the lookup as plain data (sliced-env safe); a window-edge back-scan keeps mid-bar scheduler windows accent-correct.
• _suggestCombPure(…, accentMap) weighted mean; threaded through the march and the best-candidate probe via opts.accentsByMeasure.

Testing

tests/grouping_aware.test.mjs (5 cases, fails on main): accent-map shapes (incl. bad-sum degradation), lookup coverage, click accents incl. the mid-window back-scan, the comb ordering (right grouping > even > wrong grouping on an accent-only onset fixture) plus exact backwards-compat mean, ruler tick gating. Full suites green: 160 files / 0 fail, lint at baseline.

🤖 Generated with Claude Code

https://claude.ai/code/session_01EBQCHCNA81E9tHmSDHSe2Q

Summary by CodeRabbit

• New Features

  • Metronome accents now follow authored meter groupings, such as 7/8 grouped as 2+2+3.
  • Ruler ticks emphasize grouping boundaries for easier rhythm navigation.
  • Tempo suggestions now give greater weight to onset timing at grouping accents.
  • Bars without groupings retain their previous behavior.

• Tests

  • Added coverage for grouping-aware metronome timing, ruler ticks, and tempo suggestions.

[coderabbitai]

Caution

Review failed

The pull request is closed.

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 12802d0a-235e-41bf-a7e3-822aa2e7bef2

📥 Commits

Reviewing files that changed from the base of the PR and between 0e637b3 and ecfeda8.

📒 Files selected for processing (6)

• CHANGELOG.md
• src/audio.js
• src/ruler.js
• src/tempo-marks.js
• src/tempo-suggest.js
• tests/grouping_aware.test.mjs

---

📝 Walkthrough

Walkthrough

Authored meter groupings now generate accent maps used by metronome clicks, ruler ticks, and tempo suggestion scoring. Ungrouped bars retain downbeat-only behavior, and new tests cover grouping validation and each consumer.

Changes

Grouping-aware timing

Layer / File(s)	Summary
Accent-map generation src/tempo-marks.js, tests/grouping_aware.test.mjs, CHANGELOG.md	Grouping data is converted into memoized measure-scoped accent maps, with tests for valid and invalid groupings and changelog documentation.
Metronome and ruler accents src/audio.js, src/ruler.js, tests/grouping_aware.test.mjs	Metronome clicks and ruler subdivision ticks use grouping accents while preserving ungrouped behavior.
Accent-weighted tempo suggestions src/tempo-suggest.js, tests/grouping_aware.test.mjs	Tempo comb scoring weights accented grouping positions and passes measure-specific accent maps through candidate selection.

Estimated code review effort: 3 (Moderate) | ~20 minutes

Sequence Diagram(s)

sequenceDiagram
  participant TempoMarks
  participant Metronome
  participant Ruler
  participant TempoSuggest
  TempoMarks->>Metronome: provide live grouping accent maps
  TempoMarks->>Ruler: provide live grouping accent maps
  TempoMarks->>TempoSuggest: provide measure accent maps
  Metronome->>Metronome: accent grouped-cell clicks
  Ruler->>Ruler: emphasize grouped subdivision ticks
  TempoSuggest->>TempoSuggest: weight grouped-cell onset scores

Loading

Possibly related PRs

• got-feedBack/feedBack-plugin-editor#215: Shares the assisted tempo-fit and comb-scoring path.
• got-feedBack/feedBack-plugin-editor#231: Also modifies tempo suggestion comb logic.
• got-feedBack/feedBack-plugin-editor#276: Introduces the authored meter-grouping data consumed here.

Suggested reviewers: byrongamatos

✨ 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/editor-grouping-aware

---

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