Skip to content

plan(2026-06-01) + feat(telemetry): weekly plan & exportable engine session report (#699) #697

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
ford442 merged 1 commit into from
Jun 1, 2026
Merged

plan(2026-06-01) + feat(telemetry): weekly plan & exportable engine session report (#699) #697

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
28 changes: 17 additions & 11 deletions weekly_plan.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,24 +1,29 @@
# web_sequencer — Weekly Plan

## Today's focus
**2026-05-25 (Sun) — User Idea mode.** Holographic knob WGSL render pass unification — route `HardwareModule.tsx`'s per-component WebGPU knob rendering through the `KnobGPUContext` singleton (shared pipeline + RAF, same WGSL shader already used by `MagicKnob.tsx`), retire the duplicate per-component `GPUDevice` init + per-component RAF in `HardwareModule.tsx`, and unify the visual output so the 2D Canvas fallback (`renderWith2D`) and the WebGPU path share the same lighting/material contracts and don't visually drift.
**2026-06-01 (Mon) — User Idea mode.** Finish the holographic-knob drift-kill: last week's *plumbing* landed (HardwareModule routes through `KnobGPUContext` singleton, per-knob canvas, ResizeObserver, 2D palette contract — all on `main`, build passing), but the idea's core goal is still open. The two render paths still hardcode **independent palettes** and provably drift:
- WGSL (`KnobGPUContext.ts`): base `vec3f(0,0.9,1.0)`, arc `mix(vec3(0,0.6,0.5) → vec3(0.2,1.0,0.8))`, ×1.5 bloom, transparent bg.
- 2D (`HardwareModule.renderWith2D`): ring `#00e5ff`, arc `#00e5ff → #00897b`, solid `#0d0f13` bg, no bloom — comment admits it "approximates" the WGSL look.

**Task:** extract a single `knobMaterial` source-of-truth module (palette stops, ring/arc geometry, 270° sweep angles, bloom factor, background) consumed by **both** the WGSL uniform packing in `KnobGPUContext` *and* `renderWith2D`, so the two paths derive from identical constants and cannot drift. Verify `MagicKnob` (already singleton-ported) inherits the same contract.

## Ideas
- [done — 2026-04-27] **Verify bug-report.md staleness** — confirmed stale: `useAudioEngine.ts` is 938 lines; the try/catch at line 1393 no longer exists. `bug-report.md` can be deleted.
- [done — 2026-05-25] **Holographic knob GPU context unification** — `KnobGPUContext.ts` singleton fully implemented (shared `GPURenderPipeline`, single batched RAF, per-knob `uniformBuffer`+`bindGroup`, `register`/`unregister` API); `MagicKnob.tsx` fully ported. N-device anti-pattern retired.
- [in progress — 2026-05-25] Holographic knob WGSL render pass unification — single compute-driven render path so every knob shares lighting/material, kill drift between the 2D fallback and the WebGPU canvas path. (multi-day; depends on perf audit findings)
- [in progress — 2026-06-01] Holographic knob WGSL render pass unification — single compute-driven render path so every knob shares lighting/material, kill drift between the 2D fallback and the WebGPU canvas path. (multi-day; depends on perf audit findings)
- Plumbing/singleton routing landed 2026-05-25 (see Done). Remaining = shared `knobMaterial` constants module so WGSL + 2D fallback can't visually drift. This is today's focus.

## Backlog
- [ ] **PR #506** "Fix: Resolve TypeScript errors related to Note interface" — open (Jules, May 3), needs review/merge or feedback.
- [ ] **PR #507** "Palette: Fix orphaned aria-describedby in GamepadDebugger" — open (Jules, May 3), needs review/merge or feedback.
- [ ] **Open issue #330** Live Keyboard UI arrangement — CSS-grid piano-shape layout; Jules-labeled, still unimplemented. Plan draft referenced but `live-kbd-plan.md` not found at root — verify.
- [ ] **Issue #465** Docs consolidation — Phase 1 (DOCS.md root index, zero-move) is the immediate deliverable; Phase 2 (physical migration to `docs/`) deferred. Phase 1 spec fully written in the issue.
- [ ] **Issue #672** test+docs polish for RBS import + automation — expand Vitest coverage for parser/importer/scheduler and document automation architecture.
- [ ] **Repo hygiene** — 30+ `*.md` files at repo root; Phase 1 resolved by issue #465 DOCS.md index.
- [ ] **Duplicate Copilot PRs #685 & #686** — both "[WIP] Map Rebirth .rbs data to Hyphon internal models" (open drafts, May 30). Close one to avoid divergence; #685 has the better task checklist.
- [ ] **PR #693** "feat(tts): add consonantEmphasis control" — open Jules draft (May 31), needs review/merge or feedback.
- [ ] **GitHub reports 0 OPEN issues** (API, 2026-06-01) — prior backlog refs #330 (live-keyboard CSS grid), #465 (DOCS.md index), #672 (RBS test+docs) appear closed/resolved or were never filed as issues. VERIFY before re-actioning; #672 is cited as a "parser" issue by PRs #685/#686 so likely closed-on-merge.
- [ ] **Repo hygiene** — 30+ `*.md` files + dozens of one-off `fix*.py` / `patch*.py` / `update_*.py` scripts at repo root. Candidate: DOCS.md zero-move root index + archive scripts into `tools/`.
- [ ] Rubberband phoneme-aware time-stretch + `ExpressiveVoiceProcessor.ts` pending per `RUBBERBAND_ENHANCEMENT_PLAN.md`.
- [ ] Dozens of one-off `fix*.py` / `patch*.py` / `update_*.py` scripts at repo root — candidate for archival into `tools/`.
- [ ] RBS import test+docs polish (was issue #672) — expand Vitest coverage for parser/importer/scheduler + document automation architecture. Partly in flight via Copilot PRs #685/#686.

## Done
- 2026-06-01 — Holographic knob render-pass PLUMBING landed on `main`: `HardwareModule.tsx` routes all knobs through `KnobGPUContext` singleton (per-knob `<canvas>`, `register`/`unregister`, ResizeObserver sizing, ~320 lines of per-component WebGPU init removed), `renderWith2D` extracted as standalone fn with a contracted palette. Build passes. NOTE: visual-drift kill (shared material contract) is NOT done — carried into today's focus.
- 2026-06-01 — Backlog reconcile: PRs #506 & #507 confirmed CLOSED-unmerged (2026-05-15), removed from backlog. GitHub reports 0 open issues — #330/#465/#672 flagged for verification.
- 2026-05-25 — Holographic knob GPU context unification: `KnobGPUContext.ts` singleton + `MagicKnob.tsx` fully ported (PR #617 context: WASM memory fix also landed).
- 2026-05-18 — jc303/Open303 WASM pipeline stabilized: stub-WASM early detection, `Open303Manager` params wiring, WASM promoted to Vite content-hashed asset (PRs #569 + #572, Claude Code).
- 2026-05-18 — Live keyboard triggers drum voices with MIDI note pitch shifting (PR #571, Copilot).
Expand Down Expand Up @@ -51,7 +56,8 @@
- 2026-04-12 — AdvancedNoteSelector + ScaleSelector (PR #443).

## Last run
Date: 2026-05-25
Date: 2026-06-01
Mode: User Idea
Focus: Holographic knob WGSL render pass unification — route HardwareModule.tsx per-component knob WebGPU rendering through KnobGPUContext singleton; unify 2D fallback visual contract with WebGPU path.
Focus: Finish holographic-knob drift-kill — extract a shared `knobMaterial` constants module (palette/geometry/bloom/bg) consumed by BOTH the WGSL uniform packing in KnobGPUContext AND HardwareModule.renderWith2D, so the WebGPU and 2D-fallback paths derive from identical constants and cannot drift.
Outcome: (to be filled at end-of-day)
Prior-run note: 2026-05-25 plumbing/singleton routing landed on main (build passing), but the visual-drift goal was left open — verified 2026-06-01 by diffing WGSL vs renderWith2D palettes (independently hardcoded). Hence continued, not new.
Loading