# Pending Features & Issues

Priority: `P1` quick win · `P2` moderate · `P3` large effort

## Processing Pipeline

- [x] `P1` **Noise gate** — Suppress small color changes below threshold, preventing shimmer on static content
- [x] `P1` **Color temperature filter** — Already covered by existing Color Correction filter (2000-10000K)
- [ ] `P1` **Zone grouping** — Merge adjacent LEDs into logical groups sharing one averaged color
  - Complexity: medium — doesn't fit the PP filter model (operates on extracted LED colors, not images); needs a new param on calibration/color-strip-source config + PixelMapper changes
  - Impact: high — smooths out single-LED noise, visually cleaner ambilight on sparse strips
- [x] `P2` **Palette quantization** — Force output to match a user-defined palette (preset or custom hex)
- [x] `P2` **Drag-and-drop filter ordering** — Reorder postprocessing filter chains visually
- [ ] `P3` **Transition effects** — Crossfade, wipe, or dissolve between sources/profiles instead of instant cut
  - Complexity: large — requires a new transition layer concept in ProcessorManager; must blend two live streams simultaneously during switch, coordinating start/stop timing
  - Impact: medium — polishes profile switching UX but ambient lighting rarely switches sources frequently

## Output Targets

- [x] `P1` **Rename `picture-targets` to `output-targets`** — Rename API endpoints and internal references for clarity
- [x] `P2` **OpenRGB** — Control PC peripherals (keyboard, mouse, RAM, fans) as ambient targets
- [ ] `P2` **Art-Net / sACN (E1.31)** — Stage/theatrical lighting protocols, DMX controllers
  - Complexity: medium — UDP-based protocols with well-documented specs; similar architecture to DDP client; needs DMX universe/channel mapping UI
  - Impact: medium — opens stage/theatrical use case, niche but differentiating

## Automation & Integration

- [ ] `P2` **Webhook/MQTT trigger** — Let external systems activate profiles without HA integration
  - Complexity: low-medium — webhook: simple FastAPI endpoint calling SceneActivator; MQTT: add `asyncio-mqtt` dependency + subscription loop
  - Impact: high — key integration point for home automation users without Home Assistant
- [ ] `P2` **WebSocket event bus** — Broadcast all state changes over a single WS channel
  - Complexity: low-medium — ProcessorManager already emits events; add a WS endpoint that fans out JSON events to connected clients
  - Impact: medium — enables real-time dashboards, mobile apps, and third-party integrations
- [x] `P3` **Notification reactive** — Flash/pulse on OS notifications (optional app filter)
  - Complexity: large — OS-level notification listener (platform-specific: Win32 `WinToast`/`pystray`, macOS `pyobjc`); needs a new "effect source" type that triggers color pulses
  - Impact: low-medium — fun but niche; platform-dependent maintenance burden

## Multi-Display

- [ ] `P2` **Investigate multimonitor support** — Research and plan support for multi-monitor setups
  - Complexity: research — audit DXGI/MSS capture engine's display enumeration; test with 2+ monitors; identify gaps in calibration UI (per-display config)
  - Impact: high — many users have multi-monitor setups; prerequisite for multi-display unification
- [ ] `P3` **Multi-display unification** — Treat 2-3 monitors as single virtual display for seamless ambilight
  - Complexity: large — virtual display abstraction stitching multiple captures; edge-matching calibration between monitors; significant UI changes
  - Impact: high — flagship feature for multi-monitor users, but depends on investigation results

## Capture Engines

- [ ] `P3` **SCRCPY capture engine** — Implement SCRCPY-based screen capture for Android devices
  - Complexity: large — external dependency on scrcpy binary; need to manage subprocess lifecycle, parse video stream (ffmpeg/AV pipe), handle device connect/disconnect
  - Impact: medium — enables phone screen mirroring to ambient lighting; appeals to mobile gaming use case
- [x] `P3` **Camera / webcam** — Border-sampling from camera feed for video calls or room-reactive lighting

## Code Health (from review 2026-03-09)

### Bugs

- [x] `P1` **Thread safety: dict mutation during iteration** — composite_stream.py / mapped_stream.py `_sub_streams.clear()` crashes processing loop
- [x] `P1` **Thread safety: SyncClockRuntime.get_time() race** — compound read without lock causes time double-counting
- [x] `P1` **Thread safety: SyncClockManager unprotected dicts** — `_runtimes`/`_ref_counts` mutated from multiple threads without lock
- [x] `P1` **Clock ref-count corruption on hot-swap** — `_release_clock` reads new clock_id from store instead of old one
- [x] `P1` **Path traversal guard** — `auto_backup.py` uses string checks instead of `Path.resolve().is_relative_to()`
- [x] `P2` **Crash doesn't fire state_change event** — fatal exception path in `wled_target_processor.py` doesn't notify dashboard
- [x] `P2` **WS broadcast client mismatch** — `kc_target_processor.py` `zip(clients, results)` can pair wrong clients after concurrent removal

### Performance

- [x] `P1` **Triple FFT for mono audio** — `analysis.py` runs 3 identical FFTs when audio is mono (2x wasted CPU)
- [x] `P2` **Per-frame np.array() from list** — `ddp_client.py:195` allocates new numpy array every frame
- [x] `P2` **frame_time recomputed every loop iteration** — `1.0/fps` in 8 stream files, should be cached
- [x] `P2` **Effect/composite/mapped streams hardcoded to 30 FPS** — ignores target FPS, bottlenecks 60 FPS targets
- [x] `P3` **Spectrum .copy() per audio chunk** — `analysis.py` ~258 array allocations/sec for read-only consumers

### Code Quality

- [x] `P2` **12 store classes with duplicated boilerplate** — no base class; `BaseJsonStore[T]` would eliminate ~60%
- [x] `P2` **DeviceStore.save() uses unsafe temp file** — fixed-path `.tmp` instead of `atomic_write_json`
- [x] `P2` **Route code directly mutates ProcessorManager internals** — `devices.py` accesses `manager._devices` in 13+ places
- [x] `P2` **scene_activator.py accesses ProcessorManager._processors directly** — bypasses public API
- [x] `P3` **datetime.utcnow() deprecated** — 88 call sites in 42 files, should use `datetime.now(timezone.utc)`
- [x] `P3` **color-strips.js 1900+ lines** — should be split into separate modules
- [x] `P3` **No DataCache for color strip sources** — fetched with raw fetchWithAuth in 5+ places

### Features

- [ ] `P1` **Auto-restart crashed processing loops** — add backoff-based restart when `_processing_loop` dies
- [ ] `P1` **"Start All" targets button** — "Stop All" exists but "Start All" is missing
- [ ] `P2` **Manual backup trigger endpoint** — `POST /system/auto-backup/trigger` (~5 lines)
- [ ] `P2` **Scene snapshot should capture device brightness** — `software_brightness` not saved/restored
- [ ] `P2` **Device health WebSocket events** — eliminate 5-30s poll latency for online/offline detection
- [ ] `P2` **Distinguish "crashed" vs "stopped" in dashboard** — `metrics.last_error` is already populated
- [ ] `P3` **Home Assistant MQTT discovery** — publish auto-discovery payloads; MQTT infra already exists
- [ ] `P3` **CSS source import/export** — share individual sources without full config backup
- [ ] `P3` **Exponential backoff on events WS reconnect** — currently fixed 3s retry

## UX

- [x] `P2` **Tags / groups for cards** — Assign tags to devices, targets, and sources; filter and group cards by tag
- [x] `P3` **PWA / mobile layout** — Mobile-first layout + "Add to Home Screen" manifest
- [ ] `P1` **Collapse dashboard running target stats** — Show only FPS chart by default; uptime, errors, and pipeline timings in an expandable section collapsed by default
- [x] `P1` **Review protocol badge on LED target cards** — Review and improve the protocol badge display on LED target cards