Compare commits

...

276 Commits

Author SHA1 Message Date
alexei.dolgolyov 14822fb6a0 Merge feat/roadmap-2026-06-19: pre-release review fixes (2026-06-23) 2026-06-23 14:21:42 +03:00
alexei.dolgolyov 0c096db639 fix: address pre-release review findings (2026-06-23)
Hardening from the pre-release review of the 06-19/06-23 roadmap batches:
solar timezone crash, webhook header CRLF, MQTT topic-prefix injection,
get_stats thread-safe copy, MQTT discovery lock, reactive_mode Literal,
and calibration modal accessibility. Adds regression coverage in
test_release_review_2026_06_23.py.
2026-06-23 14:21:25 +03:00
alexei.dolgolyov c1eeefcf06 Merge feat/roadmap-2026-06-19: roadmap batches (solar/image-quality/per-pixel + integrations)
Round one (6745e25): SolarRule, arm64 Docker, LoL poller, color-harmony
gradients, reactive palette, linear-light, dithering, Nanoleaf extControl.
Round two (39b0554): LIFX multizone/Tile, Hue segment mapping, outbound
webhook action, HA MQTT auto-discovery.
2026-06-23 00:55:49 +03:00
alexei.dolgolyov 39b0554444 feat: roadmap round two (2026-06-23) — per-pixel smart-lights + integrations
A) Per-pixel smart-lights
- LIFX multizone (SetExtendedColorZones msg 510, <=82 zones) + Tile
  (SetTileState64 715), auto-detected on connect with single-colour
  fallback; lifx_per_zone threaded like nanoleaf_per_panel
- Hue gradient-lightstrip mapping: Entertainment v2 frame now keyed by
  channel id (was 1 light=1 LED), channels discovered on connect;
  hue_gradient_mode toggle (default on)

B) Integrations bundle
- Outbound webhook automation action (Discord/IFTTT/Zapier/Node-RED),
  SSRF-gated via validate_polling_url at both save and fire time; fires
  on activate/deactivate, best-effort, audited
- Home Assistant MQTT auto-discovery: read-only binary_sensors per
  automation + connectivity, availability via birth/will, cleanup on
  disable/delete, live state from the engine

Shared: pixel_reduce.resample_to_n nearest-neighbour helper.
57 new tests (lifx_multizone, hue_segment, webhook_action, ha_discovery).
Gate: ruff + tsc + build clean, pytest 2719 passed / 2 skipped.
2026-06-23 00:50:22 +03:00
alexei.dolgolyov 6745e25b20 feat: roadmap batch (2026-06-19) — solar/linear-light/dither/nanoleaf + integrations
Eight roadmap features from the 2026-06-19 review, each a full vertical
(backend + tests + frontend + i18n en/ru/zh); ~67 new unit tests:

- automations: SolarRule sunrise/sunset trigger (new utils/solar.py, shared
  with the daylight cycle; window logic mirrors TimeOfDayRule)
- ci: best-effort arm64 multi-arch Docker manifest via QEMU + docker manifest
  (release.yml; amd64 path untouched, continue-on-error)
- game-integration: wire the orphaned LoLPoller via a LoLPollManager + a shared
  runtime_state module (poll lifecycle on enable/CRUD/startup/shutdown)
- ui: color-harmony gradient generator (complementary/analogous/triadic/...)
- effects: audio-reactive palette modulation (new audio_energy_tap; brightness/
  saturation modulation across all 12 procedural effects)
- capture: linear-light blending + spatio-temporal dithering, opt-in per
  calibration (new utils/linear_light.py, utils/dither.py)
- devices: Nanoleaf extControl v2 per-panel UDP streaming (per_panel mode)

Also bundles the pending 2026-06-18 production-review fixes and other
in-progress work already in the working tree (manual-trigger rule, etc.),
since they share files and could not be cleanly separated.

Gate: ruff + tsc clean; pytest 2654 passed / 2 skipped. The single failing
test (automation manual_trigger handler coverage) is a separate in-progress
item owned elsewhere, intentionally left as-is.
2026-06-22 23:21:24 +03:00
alexei.dolgolyov 126d8f2449 feat(auth): add auth.expose_docs flag to view API docs without a token
The /docs, /redoc and /openapi.json routes are gated by AuthRequired, so a
browser can't open them on plain navigation (no way to attach a Bearer token).
Add an opt-in auth.expose_docs flag (default off) that relaxes ONLY those three
routes to anonymous access (loopback + LAN) via a new verify_docs_access
dependency. Every real endpoint stays protected, and a startup WARNING fires
when the flag is on.

- config: AuthConfig.expose_docs: bool = False
- auth: verify_docs_access / DocsAccess dependency
- main: docs routes use DocsAccess; startup warning
- default_config.yaml: documented flag
- tests: docs anonymous when exposed; real endpoints still 401
2026-06-11 00:14:48 +03:00
alexei.dolgolyov e584235676 chore(activity-log): post-merge cleanup + graduate context to CLAUDE.md
- remove plans/activity-log/ (feature merged; learnings in CLAUDE.md + git history)
- server/CLAUDE.md: Activity/Audit Log architecture + extension points (recorder, fire_entity_event hook, sanitize_display, events allowlist, retention, API auth posture)
2026-06-10 18:42:15 +03:00
alexei.dolgolyov b43f821046 Merge feature/activity-log: persistent activity/audit log
Adds a persistent, queryable activity/audit log surfaced in the WebUI:

- storage: dedicated indexed activity_log table + repository (keyset pagination, prune, streaming export); migration 002

- recorder + actor ContextVar + retention engine + lifecycle wiring; activity_logged realtime event

- instrumentation across auth / device / entity-CRUD / capture / system (best-effort, no secrets logged)

- REST API: list/filter, CSV/JSON export, retention settings, clear (auth-gated)

- WebUI: Activity tab (smart filtering, live updates, localized descriptions, export), Dashboard widget, Settings retention panel; en/ru/zh

Reviewed via independent per-phase + final + security agents; full suite 2563 passed.

Includes 17dd2e0 (review-findings fixes) that the feature branch was based on.
2026-06-10 15:31:12 +03:00
alexei.dolgolyov 077c99c7d1 fix(activity-log): no spinner flash on instant filtering
- re-query keeps current rows visible instead of clearing + showing the full 'Loading' spinner
- loading affordance is delayed ~180ms: instant responses show nothing; slow ones get a subtle dim (aria-busy)
- full spinner reserved for the genuine first load; append (load-more) shows no indicator
2026-06-10 15:30:48 +03:00
alexei.dolgolyov ae74cca132 fix(activity-log): UI polish - accessible export menu, i18n placeholders, zero-result spinner fix
- export button is now a click-toggle menu (aria-haspopup/expanded, role=menu, caret, Escape + click-outside close)
- filter placeholders moved to i18n (actor/entity_type .placeholder keys, en/ru/zh)
- _fetchPage clears _loading before render so a zero-result page doesn't spin forever
- toolbar/entry use elevated card surface (--lux-bg-1); light-theme device badge contrast; mobile message grid
2026-06-10 15:24:45 +03:00
alexei.dolgolyov 77284e8e7b fix(activity-log): dashboard section reconciliation + activity column alignment
- dashboard full re-render now reconciles sections (only replaces changed ones) instead of wholesale .dashboard-dynamic innerHTML swap -> editing an entity no longer jumps the whole dashboard
- Recent Activity widget live DOM + perf strip preserved across re-renders; widget skips re-fetch when already populated (no flash)
- sweep stray non-section nodes so empty->populated doesn't leave an orphan 'no targets' banner (review-caught regression)
- Activity list rows use a CSS grid (fixed badge/actor columns) so message column aligns consistently across rows
2026-06-10 12:28:13 +03:00
alexei.dolgolyov ff1ff06cb5 fix(activity-log): post-test polish - localize descriptions, dashboard widget, ticking time
- localize entry descriptions client-side via localizeMessage (activity_log.msg.* + entity_type.* templates x3 locales); server message kept as fallback/export/search
- remove redundant Activity header banner from tab
- Recent Activity widget is now a first-class dashboard section (Customize Dashboard show/hide/reorder; pre-existing layouts preserved)
- live activity event updates the widget surgically (no full dashboard rebuild); single listener with teardown
- relative-time labels tick via shared ensureRelativeTimeTicker (single 30s interval, visibility-aware)
2026-06-10 12:03:18 +03:00
alexei.dolgolyov 3dd1ac3f0d fix(activity-log): final-review fixes - crosslink keys + sanitize parity
- _ENTITY_NAV map keys corrected to match backend entity_type (device, color_strip_source, audio_source data-id) + scene_playlist crosslink added
- sanitize_display applied uniformly to owner-authored names at remaining record sites (dependencies entity_name, device_health, automation_engine, output_targets_control, scene_presets, scene_playlists)
2026-06-09 21:23:22 +03:00
alexei.dolgolyov 6e1dd2111d feat(activity-log): phase 6 - dashboard widget + settings panel + docs
- Dashboard 'Recent Activity' widget: latest 5 entries, live prepend, 'View all' -> Activity tab
- Settings 'Activity Log' panel: retention (enabled/max_days/max_entries) GET/PUT, clear (confirm + auth-required toast), CSV/JSON export
- audit-log vs ephemeral debug Log Viewer distinction note + cross-links
- public helpers fetchRecentEntries/renderCompactEntry on activity-log.ts (reused, no dup markup)
- README Activity Log section; i18n across en/ru/zh
- review fixes: clear 401 surfaces toast; empty widget transitions on first live event
2026-06-09 21:05:40 +03:00
alexei.dolgolyov 9a0137fa4c feat(activity-log): phase 5 - Activity tab (smart filtering, live updates, export)
- new top-level Activity tab: filter toolbar (category/severity chips, presets, debounced search, actor/entity/date), keyset load-more, expandable detail
- live prepend via server:activity_logged; authed CSV/JSON blob export
- formatTimestamp/formatRelativeTime in core/ui.ts; history+severity SVG icons; Ctrl+7 shortcut
- i18n activity_log.* across en/ru/zh; getting-started tutorial step; activity-log.css (themed)
- review fixes: newest-first ordering, attribute-context XSS hardening (_escapeAttr + event delegation)
2026-06-09 20:42:44 +03:00
alexei.dolgolyov 4a0927521a feat(activity-log): phase 4 - REST API (list/export/settings/clear)
- GET /activity-log: filtered, keyset-paginated list (categories/severities/actor/entity/date/q)
- GET /activity-log/export: streaming CSV/JSON, chunked keyset (releases DB lock per batch), CSV formula-injection guard
- GET/PUT /activity-log/settings: retention config (PUT require_authenticated)
- DELETE /activity-log: clear (require_authenticated, self-audited)
- security: export DoS fix, settings-PUT auth gate, CSV \t/\r guard, metadata-as-JSON
- 122 API tests (auth posture, CSV injection, pagination integrity, filters, settings bounds, clear-audited)
2026-06-09 20:09:46 +03:00
alexei.dolgolyov 25c613c5cb feat(activity-log): phase 3 - event instrumentation (4 categories)
- entity CRUD via fire_entity_event choke point (name resolved/sanitized; deletes pass name explicitly)
- auth: failures + WS session establishment (no tokens logged); per-IP audit-record throttle
- device: online/offline (health), discovered/lost (zeroconf), ADB connect/disconnect
- capture/system: target start-stop, scenes, playlists, automations, backup/restore, update, restart, calibration, settings
- security hardening: sanitize_display strips control/NUL/ANSI/newlines from untrusted strings; malformed-IPv6 origin guard
- 129 instrumentation tests (incl. secret-leak, log-injection, throttle, best-effort) + autouse throttle-reset fixture
2026-06-09 19:20:57 +03:00
alexei.dolgolyov 726f39e2ba feat(activity-log): phase 2 - recorder, actor context, retention, lifecycle
- ActivityRecorder: thread-safe record() (inline on loop, call_soon_threadsafe off-loop), best-effort, fires activity_logged event
- current_actor ContextVar set in verify_api_key (both branches), default system
- ActivityLogRetentionEngine: prune loop (max_days+max_entries), settings persistence, rehydrates recorder.enabled on startup
- lifespan wiring: server.shutting_down recorded first on shutdown, retention stop before db.close
- events-ws.ts allowlist + parity; DI getters + module accessor; 62 new tests
2026-06-09 18:10:27 +03:00
alexei.dolgolyov 1ac4a0f66d feat(activity-log): phase 1 - storage model, migration, repository
- ActivityLogEntry dataclass + ActivityCategory/ActivitySeverity + ActivityLogFilters
- additive idempotent migration 002_add_activity_log (indexed activity_log table, seq keyset tiebreaker)
- ActivityLogRepository (record/query/count/prune/clear/iter_export), keyset pagination, parameterized SQL
- 102 unit + adversarial tests (SQL-injection, pagination, prune, codec, migration idempotency)
2026-06-09 17:40:37 +03:00
alexei.dolgolyov 1afe7d6fcc chore(activity-log): scaffold feature plan and phase subplans 2026-06-09 17:14:50 +03:00
alexei.dolgolyov 17dd2e02ba fix: resolve comprehensive review findings (security, concurrency, perf, Android, UI)
Multi-dimension review of v0.8.2. Excludes the deliberately deferred
default_config.yaml weak-default-key item (C1).

Backend:
- calibration: create_default_calibration no longer exceeds led_count for small
  odd counts (bounded trim + regression test)
- game-integration: generic webhook now requires auth_token; constant-time
  compare_digest in all adapters; per-IP failed-auth rate limit on the ingest
  route; auth_token encrypted at rest via secret_box (migration-safe)
- playlist engine: serialize _state/_task under the lifecycle lock to close a
  delete-mid-play race (+ concurrency tests)
- main: stop the calibration session on shutdown (restore prior target)
- home_assistant: validate HA host via the LAN classifier on create/update
- perf: drop slow preview-WS clients instead of blocking the send loop; cache
  composite full-strip resize linspaces; effect_stream lava reuses scratch

Frontend:
- setup/auto-calibration wizard: guard _state after awaits (cancel-safe), await
  session teardown before output start, busy-gate skip-calibration, manual
  display input keeps focus, move focus on step change
- calibration: destroy EntitySelect on modal close
- color-strips test: dirty-flag-gated render + cached ctx/ImageData
- a11y/TV: focus-visible for new wizard/auto-cal/corner controls, aria-labels on
  the spatial corner/edge picker; theme-aware syntax tokens; dead/undefined CSS
  tokens removed; .modal-error styled; i18n titles (en/ru/zh)

Android:
- ApiKeyManager: EncryptedSharedPreferences with verified, data-safe legacy
  migration that never rotates an existing key
- CaptureService: validate MediaProjection token before promoting; satisfy the
  startForeground 5s contract on the bail path
- NotificationListener: connection-scoped executor with lazy fallback
- BLE: request BLUETOOTH_SCAN/CONNECT at runtime + guard handler-thread
  SecurityExceptions
- Root: cancellation-aware su grant probe

Adds 14 tests. Gate: ruff + tsc 5.9.3 + esbuild + pytest (2185 passed) +
compileDebugKotlin all green.
2026-06-09 16:35:08 +03:00
alexei.dolgolyov 7a12f39f49 chore: release v0.8.2
Build Release / create-release (push) Successful in 3s
Build Android APK / build-android (push) Failing after 11s
Build Release / build-docker (push) Successful in 3m14s
Build Release / build-linux (push) Successful in 4m3s
Build Release / build-windows (push) Successful in 4m25s
Build Release / publish-release (push) Successful in 1s
2026-06-08 20:18:28 +03:00
alexei.dolgolyov dd43f3836d fix(calibration-wizard): all-provider discovery + spatial corner picker
- Setup wizard discovery now scans every device provider, not just WLED:
  drop the hardcoded device_type=wled filter so the backend's parallel
  all-provider scan runs (Adalight, DDP, OpenRGB, BLE, DMX, etc). The list
  UI was already multi-type aware (per-type icon + badge).
- Auto-calibration corner picker: finish the unfinished step. The empty
  .autocal-corner-glyph + dead --top-left/etc modifier classes now render a
  mini-screen frame with the matching corner lit, so users map the physical
  lit LED to a button at a glance. Hover/focus lights the corner dot.
- Remove dead empty CSS rule for #wizard-rerun-btn.
2026-06-08 17:59:56 +03:00
alexei.dolgolyov d32961085d Merge feature/edge-calibration-wizard: auto edge-calibration + first-run wizard
Auto edge-calibration via on-screen chase (#4) and a guided first-run setup
wizard (#3); spatial model (#11) intentionally excluded. Also adds scene
playlists + cycling state to the /api/v1/snapshot poll.

- calibration solver + chase session (lock, idle-timeout, stop/restore) + /api/v1/calibration/* (phase 1)
- POST /api/v1/setup/scaffold (rollback, registers target with manager) + onboarding flag (phase 2)
- reusable browser-driven auto-calibration flow + calibration-modal entry (phase 3)
- guided first-run wizard with first-run trigger + tour suppression (phase 4)
- snapshot endpoint returns scene_playlists + playlist_state

Full suite 2149 passed / 2 skipped; tsc clean; build passes; ruff clean.
2026-06-08 17:00:41 +03:00
alexei.dolgolyov 6cd5e057da fix(setup): register scaffolded target with ProcessorManager + final-review hardening
Final-review blocker: the setup scaffold created the LED output target in the
store but never registered it with the ProcessorManager, so the wizard's
"Start" step 404'd on a fresh setup (target not found) — the lights never
started despite a success screen. Now the scaffold calls
target.register_with_manager(manager) right after create (mirroring the
canonical POST /output-targets route, same ValueError guard), so
start_processing finds the target. Rollback unregisters via
manager.remove_target before deleting the store entity, so a post-registration
failure leaves no half-registered target.

Also from the final review:
- solve corner_indices elements now bounded ge=0 (clear 422 instead of silent
  modulo-wrap).
- setup-wizard.ts: reuse tutorials' suppressGettingStartedTour()/TOUR_KEY
  instead of a duplicated 'tour_completed' literal; drop a duplicate manual-form
  submit listener.

Tests: + adversarial pass over the whole feature (solver/session/scaffold edge
cases) and a scaffold->register->startable regression test. Full suite
2149 passed / 2 skipped; tsc clean; build passes; ruff clean.
2026-06-08 16:55:36 +03:00
alexei.dolgolyov 81b18089e1 feat(onboarding): guided first-run setup wizard (phase 4, final)
A multi-step first-run wizard that takes a brand-new user from install to a
running, calibrated ambient light in ~2 minutes, orchestrating the existing
primitives (no node graph required).

- features/setup-wizard.ts + modals/setup-wizard.html: welcome -> find device
  (discovery list + manual add via the canonical, URL-validated POST /devices)
  -> pick screen (GET /config/displays) -> scaffold (POST /setup/scaffold) ->
  calibrate (embeds the phase-3 auto-calibration flow via mountAutoCalibration
  on the scaffolded CSS + device) -> start -> done, with a progress indicator.
- First-run trigger in app.ts (checkAndOpenWizardIfNeeded): on load, if the
  onboarding flag is unset AND no output targets exist, the wizard takes over
  and the tooltip tour is suppressed; on finish/skip it PUTs the onboarding
  flag and sets localStorage tour_completed so neither re-fires. Re-runnable.
- tutorials.ts exposes TOUR_KEY + a takeover hook so the getting-started tour
  and the wizard never double-fire.
- Calibrate step always calls unmountAutoCalibration() on exit so the device
  is restored. i18n in en/ru/zh (wizard.* keys + common.back).

Final phase of the edge-calibration + first-run-wizard feature. Big Bang final
gate green: tsc --noEmit clean, npm run build passes, full pytest suite
2064 passed / 2 skipped, ruff clean.
2026-06-08 16:27:55 +03:00
alexei.dolgolyov abc204c04e feat(snapshot): include scene playlists + cycling state in snapshot
The aggregated /api/v1/snapshot poll now emits a `scene_playlists` section
(each playlist with its `is_running` flag) plus a companion `playlist_state`
key carrying the single global cycling state (running playlist, current index/
preset, dwell) — so the HA-coordinator and other low-overhead pollers get
playlist state in the same round trip as scenes/targets, matching the other
entity sections. Gated by the `scene_playlists` include-section like the rest.

Reuses the existing list_scene_playlists handler; snapshot route tests updated.
2026-06-08 16:22:47 +03:00
alexei.dolgolyov 9550688c1e feat(calibration): browser-driven auto edge-calibration UI (phase 3)
Reusable, chase-driven calibration flow that solves + saves the linear
CalibrationConfig with a few taps — no LED counting — and works in the
browser on desktop and Android (no Tkinter dependency).

- features/auto-calibration.ts: 5-step flow (start corner -> direction ->
  tap-to-mark-corners -> solved preview -> save). Drives the phase-1 session
  endpoints (session/position/solve) and persists via PUT /color-strip-
  sources/{id}. cornerIndices[0] is anchored to strip index 0 per the solver
  contract. unmountAutoCalibration() is the single cleanup gate — the
  calibration session is always stopped (device restored) on cancel, modal
  close, after save, AND on a mid-flow error, so the strip is never left dark
  or stuck.
- Public API mountAutoCalibration({container, cssId, deviceId, onComplete,
  onCancel}) for the phase-4 wizard to embed; showAutoCalibration() standalone.
- "Auto-calibrate" entry added to the existing calibration modal; standalone
  modal template; app.ts/global.d.ts exports; .autocal-* CSS matching the
  ds-section vocabulary; 43 autocal.* i18n keys in en/ru/zh; docs/CALIBRATION.md.

Part of the edge-calibration + first-run-wizard feature (Big Bang; intermediate
phase — full build/suite + test-writer gated at the final phase).
2026-06-08 15:52:45 +03:00
alexei.dolgolyov 9dcd76d264 feat(setup): one-call setup scaffold + onboarding flag (phase 2)
Backend for the first-run wizard (phase 4).

- POST /api/v1/setup/scaffold: given an existing device_id + display_index
  (+ optional calibration), wires a working chain via the real validated
  store create paths — create-or-reuse capture template -> raw picture
  source -> picture color-strip source (calibration or default) -> LED
  output target -> returns the ids. Does NOT auto-start. Rolls back every
  entity it created (reverse order) on any partial failure, leaving no
  orphans; "created" events are deferred until the whole chain succeeds so
  a rolled-back scaffold never leaves ghost cards in the UI.
- Requires an existing device_id (no inline device creation) — the wizard
  creates the device first via the canonical, URL-validated POST /devices,
  so the scaffold can't bypass device validation. display_index is bounded.
- GET/PUT /api/v1/preferences/onboarding: persistent first-run flag
  ({onboarded, completed_at}) via db.set_setting; server stamps completed_at.
- Both routes AuthRequired. Tests: 25 (scaffold happy/reuse/rollback/
  validation + onboarding + calibration round-trip integration). docs/API.md.

Part of the edge-calibration + first-run-wizard feature (Big Bang; intermediate
phase — full build/suite gated at the final phase).
2026-06-08 15:22:04 +03:00
alexei.dolgolyov 0409cd8b66 feat(calibration): auto edge-calibration backend core (phase 1)
Backend engine for guided LED-chase calibration, driven by the upcoming
auto-calibration UI (phase 3) and first-run wizard (phase 4).

- solve_calibration(): pure function mapping start corner + direction + 4
  corner-tap indices to per-edge LED counts, consistent with EDGE_ORDER/
  EDGE_REVERSE so it round-trips through build_segments().
- CalibrationChaseMixin.set_calibration_pixel(): light a specific LED index
  (+ optional window) on a device, reusing the device_test_mode idle-client
  send path.
- CalibrationSession: single-active session with start/position/stop/cancel,
  a 60s idle-timeout watchdog, and a concurrency lock so interleaved calls
  can't corrupt the stop/restore bookkeeping — start() stops + remembers any
  running target on the device and stop/cancel/timeout always restore it
  (never leaves the device dark or stuck in chase).
- Routes /api/v1/calibration/{session,session/position,session/stop,
  session/cancel,session/state,solve} (all AuthRequired, bounds-validated);
  calibration is persisted by reusing the existing PUT /color-strip-sources/
  {id} (hot-reloads running streams) rather than a duplicate endpoint.
- Tests: 19 solver pure-logic + 19 route/bounds. docs/API.md updated.

Part of the edge-calibration + first-run-wizard feature (Big Bang; intermediate
phase — full build/suite gated at the final phase).
2026-06-08 14:59:58 +03:00
alexei.dolgolyov 6180569b10 wip(dashboard): in-progress dashboard customization changes
Snapshot of uncommitted dashboard-customization work (dashboard, customize,
layout, component styles, config defaults, and en/ru/zh locales) committed
as-is to clear the working tree before branching the edge-calibration +
first-run-wizard feature. Not independently verified to build.
2026-06-08 14:33:33 +03:00
alexei.dolgolyov f71e10ee06 feat(scenes): scene playlists with timed auto-cycling
Add ordered, timed sequences of scene presets that auto-cycle — activating
each preset and holding it for its dwell duration before advancing.

Backend:
- ScenePlaylist / PlaylistItem models + SQLite store (new scene_playlists table)
- PlaylistEngine: cycles ONE playlist at a time (starting one stops any other),
  loop/shuffle, re-reads the playlist each cycle so edits/deletes apply at the
  boundary, skips missing presets, guards against busy-loops; reuses the shared
  apply_scene_state path used by scene presets and automations
- REST API: CRUD + /start, /stop, /state with scene-preset reference validation
- Constructed in the app lifespan with a bounded stop on shutdown

Frontend:
- New "Playlists" sub-tab in the Automations tab with start/stop controls and a
  running indicator; editor modal with ordered scene rows (reorder + per-item
  duration), loop/shuffle toggles, and tags
- Live refresh via the playlist_state_changed WebSocket event
- i18n in en/ru/zh

Tests: new unit + API coverage for the store/model, engine (cycling,
single-active exclusivity, missing-preset skip, shuffle, and the
playlist_state_changed event contract), and routes. Full suite green;
ruff and tsc clean.
2026-06-08 13:48:43 +03:00
alexei.dolgolyov ca59546711 feat(capture): region-of-interest (ROI) crop for screen sampling
Sample only a sub-rectangle of the captured frame instead of the whole display,
so a taskbar, game HUD, or letterbox bars don't pollute the border colours — the
first functional gap a reviewer hits (capture was full-display only).

- New pure crop_screen_capture() returns a numpy view (no copy), fast-paths the
  full-frame case, and clamps degenerate/out-of-range ROIs to >=1px.
- ROI lives on CalibrationConfig (simple mode) as fractions 0..1 with a has_roi
  helper; applied in the picture color-strip stream just before border
  extraction, clamping border_width to the cropped size. Additive + backward
  compatible (full-frame default, omitted from serialization when unset -> no
  migration).
- Round-trips through the calibration schema automatically; frontend adds an
  X/Y/Width/Height (%) 'Capture region' group to the calibration editor with
  i18n (en/ru/zh).

10 unit tests (crop geometry, view-not-copy, clamping, ROI round-trip, legacy
default); full suite green (1946 passed).
2026-06-05 11:58:26 +03:00
alexei.dolgolyov 4a82595f26 Merge feat/roadmap-quick-wins: WLED realtime UDP, look presets, weekday/timezone scheduling 2026-06-05 11:44:38 +03:00
alexei.dolgolyov 1ada5ac334 feat(automations): weekday + timezone scheduling for time-of-day rule
Extend the time-of-day condition from a bare server-local HH:MM window to a real
schedule: pick which weekdays it is active (0=Mon..6=Sun, empty = every day) and
an optional IANA timezone (empty = server local). Closes the parity gap where
even a $5 WLED chip has weekday timers.

- Overnight windows (start > end) count toward the day they START on, so the
  after-midnight tail is matched against the previous weekday.
- Timezones are resolved via zoneinfo, cached, and fall back to server-local
  with a one-time warning on an invalid name (the ~1Hz tick never log-spams).
- Backward compatible: new fields default to all-days / server-local, so
  existing automations are unchanged (no migration).
- Frontend: weekday chips + timezone input on the rule editor, day/timezone in
  the rule summary, styles + i18n (en/ru/zh).

10 unit tests (weekday filter, overnight start-day semantics, tz fallback,
round-trip, invalid-day filtering); full suite green (1936 passed).
(Geographic sunrise/sunset triggers are a natural follow-up — the daylight
value source already has the solar math to reuse.)
2026-06-04 23:54:03 +03:00
alexei.dolgolyov e18d56c838 feat(processing): built-in 'look' presets (Cinematic/Vivid/Cozy/Soft/Cool)
Seed five curated, read-only post-processing templates so a non-expert gets
instant good-looking output before discovering the filter pipeline. Each is an
opinionated chain of existing filters (auto-crop/saturation/contrast/colour-
temperature/temporal-blur) tuned for a use case (films, games, evening ambience,
low-flicker, crisp cool-white).

Mirrors the built-in-gradient pattern: adds is_builtin to PostprocessingTemplate,
seeds missing looks on store init (idempotent, additive — no migration), and
makes built-ins read-only (update/delete raise -> 400; clone to customise).
Surfaced via the existing template picker + is_builtin in the response/type.

7 unit tests (seeding, idempotency, read-only protection, round-trip); full
suite green (1926 passed). (A runtime intensity slider is a follow-up — it needs
a filter-chain parameterisation layer.)
2026-06-04 23:43:11 +03:00
alexei.dolgolyov 7728aecb4f feat(wled): native realtime UDP output (DRGB/DRGBW/DNRGB) with auto-revert
Add WLED's native realtime UDP protocol (port 21324) as a third output mode for
LED targets, alongside DDP and HTTP. For the device LedGrab drives most, this
brings three user-visible wins DDP lacks:

- Auto-revert: every packet carries a timeout byte, so if the stream stops
  (host hiccup/sleep/crash) WLED returns to its preset instead of freezing on
  the last frame.
- Correct RGBW whites: the DRGBW variant carries an explicit white channel.
- Lighter on weak Wi-Fi: raw RGB with a 2-byte header.

New WledRealtimeClient auto-selects DRGB (<=490), DRGBW (<=367), or chunked
DNRGB (>490). WLED applies its own per-bus colour order in realtime mode, so we
send plain RGB and the user's colour-order config just works. Protocol 'udp' is
threaded through WLEDConfig/provider/processor and the schema pattern; the
target editor gains a protocol option + badge + i18n (en/ru/zh).

8 unit tests for the packet builder; full suite green (1919 passed).
2026-06-04 23:34:26 +03:00
alexei.dolgolyov e28ab5a956 Merge feat/power-budget-abl: automatic brightness limiting (ABL) / power budget 2026-06-04 23:22:18 +03:00
alexei.dolgolyov 1e395fd09e Merge fix/verified-bugs: weak default key, broken MQTT route, scene brightness sync 2026-06-04 23:22:18 +03:00
alexei.dolgolyov ffee156c17 feat(targets): automatic brightness limiting (ABL) / per-LED power budget
Cap an addressable strip's estimated current draw to a PSU budget so bright/
white scenes can't brown out an under-spec'd supply (voltage sag -> red/orange
shift, flicker, controller resets) — a classic 'it's broken' first impression.

- New core/processing/power_limit.py: pure current estimate (full white over N
  LEDs draws N * mA_per_led) + a (0,1] scale to land a frame on budget.
- Applied in WledTargetProcessor._send_to_device (single choke point, every send
  path; scales into a reusable scratch buffer, never mutates shared frames).
- Two per-target fields on LED targets: max_milliamps (0 = unlimited) and
  milliamps_per_led (default 55), threaded through model/store/manager/processor/
  schema/route with hot-update via update_target_settings. Additive with safe
  defaults (no data migration needed; legacy targets read as unlimited).
- Frontend: editor fields + i18n (en/ru/zh) + LedOutputTarget type.
- Tests: 10 unit tests for the estimator/scale; full suite green (1911 passed).
2026-06-04 22:56:50 +03:00
alexei.dolgolyov 02e2ea37f3 fix(scenes): sync brightness value-source change to live processor
apply_scene_state computed brightness_changed = "brightness" in changed, but
the change dict only ever uses the key "brightness_value_source_id", so the
branch was dead and a running target's brightness source was never live-synced
on scene activation (it only took effect after a restart). Check the correct
key.
2026-06-04 20:46:26 +03:00
alexei.dolgolyov fdc9201660 fix(api): remove broken legacy /system/mqtt/settings route
The GET/PUT /api/v1/system/mqtt/settings handlers read cfg.mqtt.*, but the
single-broker MQTTConfig block was removed in the multi-broker refactor, so any
call raised AttributeError. Brokers are now first-class MQTTSource entities
managed via the mqtt.py router, and the frontend no longer calls this endpoint.
Remove the dead handlers, the _load_mqtt_settings helper, the now-unused
get_config import, and the orphaned MQTTSettings{Request,Response} schemas.
2026-06-04 20:46:24 +03:00
alexei.dolgolyov 5686ae5468 fix(security): remove active weak default API key from shipped config
default_config.yaml shipped api_keys.dev: "development-key-change-in-production"
uncommitted/active, while the surrounding comment claimed it had been removed.
On a non-loopback bind this is a publicly-known credential granting full LAN
access. Restore the documented secure default (empty api_keys -> loopback-only
anonymous, LAN rejected) and leave a commented example instead.
2026-06-04 20:46:13 +03:00
alexei.dolgolyov 9960f15a1b docs(android): remove ANDROID-REVIEW planning/review docs
The Android feature-gap assessment and per-feature design docs have served
their purpose — notification, audio, webcam, and the foreground-app automation
condition are all implemented and merged, so no gaps remain to track. The
implementation is documented in the code, commit messages, and git history; the
review docs are now obsolete. No committed files referenced them (only the
local-only plans/ archives, left as point-in-time records).
2026-06-02 15:05:11 +03:00
alexei.dolgolyov 397a53ed1c Merge feature/android-foreground-app-automation: Android foreground-app automation condition
Foreground-app -> scene automation on the Android-TV build via a Kotlin
ForegroundAppBridge (UsageStatsManager) bridged into PlatformDetector ahead of the
Windows-only ctypes path; LauncherApps-backed app picker (/system/installed-apps) +
platform signal (/system/info); PACKAGE_USAGE_STATS special-access UX (on-device
button + web-UI banner, graceful degradation). Reuses the existing automation engine
unchanged; zero new deps. assembleDebug + 1897 pytest + ruff + tsc + build green;
independent final + security reviews pass.
2026-06-02 14:57:45 +03:00
alexei.dolgolyov 1c1bbe2551 feat(android): foreground-app automation condition
Make the existing Application automation rule (foreground app -> activate
scene) work on the Android-TV build. A Kotlin ForegroundAppBridge reads the
foreground app via UsageStatsManager and lists launchable apps via LauncherApps;
PlatformDetector bridges it in (ahead of the Windows-only ctypes guard) so the
existing AutomationEngine / ApplicationRule / storage / deactivation modes are
unchanged. New /system/installed-apps + /system/info endpoints feed an app picker
that stores package names (vs process names on desktop); on Android the editor
hides the match-type selector since the foreground app is the only obtainable
signal. PACKAGE_USAGE_STATS is granted via an on-device button + a web-UI banner
(no blanket prompt at capture start); detection degrades gracefully until granted.

Zero new Python/Gradle deps (UsageStatsManager + LauncherApps are in-platform;
matching only string-compares the package name, so no QUERY_ALL_PACKAGES).
assembleDebug + 1897 pytest + ruff + tsc + npm build all green; independent final
review (0 blockers) + security review (no critical issues).
2026-06-02 14:57:29 +03:00
alexei.dolgolyov 68040173c6 Merge feature/android-webcam-capture: Android on-device webcam capture
Camera2 + ImageReader (CameraBridge) -> push RGB frames -> AndroidCameraEngine,
reusing the MediaProjection capture-engine selection path so webcams surface as
selectable displays on Android TV. On-demand camera lifecycle, conditional camera
FGS type (granted-only), single-camera ownership lock. Per-phase + final + security
reviews pass; 14 files; new isolated tests; zero new Python/Gradle deps.
assembleDebug + 1883 pytest + ruff all green.
2026-06-02 13:46:59 +03:00
alexei.dolgolyov 4bf3fe65db feat(android): on-device webcam capture via Camera2 (AndroidCameraEngine)
Add on-device webcam capture to the experimental Android-TV build. Desktop
captures webcams via OpenCV (no Chaquopy/Android wheel); this adds a push-based
AndroidCameraEngine that plugs into the same selection path desktop uses
(capture template engine_type="android_camera" + display_index, HAS_OWN_DISPLAYS).

A Kotlin CameraBridge (Camera2) enumerates cameras and opens them on demand —
only while a capture source is active, driven Python->Kotlin via a guarded jclass
singleton (BleBridge pattern) — converts each frame YUV_420_888->RGB, and pushes
RGB bytes into a module-level queue mirroring mediaprojection_engine.py. Cameras
surface as selectable displays like the desktop OpenCV engine; the data-driven
capture-template UI is unchanged. No new Python deps; no new Gradle deps
(Camera2 is in-platform).

Engine: ENGINE_PRIORITY=0 (never auto-selected over MediaProjection=100; explicit
engine_type only). Single-camera ownership is serialized with a lock + ref-count
(same-camera streams attach, different-camera refused, last release stops),
mirroring the desktop CameraEngine guard.

Permission: CAMERA requested at capture-start, gated on FEATURE_CAMERA_ANY so
camera-less TV boxes never prompt; graceful degradation when denied. The service
is promoted with the camera FGS type (+ FOREGROUND_SERVICE_CAMERA) only when
CAMERA is already granted, so backgrounded capture keeps working without risking
a failed startForeground on camera-less boxes (camera can't ride the
MediaProjection token the way audio playback capture does).

Reviewed via multi-agent adversarial pass (13 findings -> 4 fixed: device leak on
session-failure, multi-stream collision, camera FGS type, i18n key; 9 refuted).

Tests: 18 new desktop-CI tests (no device needed); full suite 1883 passed.
Verified: assembleDebug BUILD SUCCESSFUL, ruff clean.

Docs: ANDROID-REVIEW/android-webcam-capture-plan.md (design), updated
android-missing-functionality.md + README feature table + en/ru/zh locales.
2026-06-02 13:36:23 +03:00
alexei.dolgolyov 34db5de8c3 Merge feature/android-notification-capture: Android on-device notification capture
NotificationListenerService -> push_notification() -> _AndroidBackend path so OS
notifications drive the existing color-strip LED effects on Android TV at app-name
parity. Per-phase + final + security reviews all pass; 11 new isolated tests; zero
new Python deps. Android assembleDebug pending device/CI verification.
2026-06-02 11:47:29 +03:00
alexei.dolgolyov 0be3f833df feat(android): on-device OS notification capture (NotificationListenerService)
Add an Android backend to os_notification_listener.py so notifications on the
experimental Android-TV build drive the existing NotificationColorStripSource
LED effects (flash/pulse/sweep, per-app colors + sounds) at app-name parity
with the Windows/Linux backends.

A Kotlin NotificationListenerService forwards the posting app's display label
across the Chaquopy JNI boundary into a new push-based _AndroidBackend +
module-level push_notification() receiver; the existing color-strip pipeline,
per-app colors/filters, and history endpoint are reused unchanged.

- Python: _AndroidBackend (probed first), push_notification() receiver,
  _LinuxBackend.probe() hardened with is_linux() to exclude Android (which
  also reports platform.system() == "Linux").
- Android: LedGrabNotificationListener NLS — serial single-thread executor,
  full crash isolation around Python.getInstance(), label-only forwarding
  (never notification title/body), ongoing/group-summary/self-package noise
  filtering. Manifest service exported + gated by
  BIND_NOTIFICATION_LISTENER_SERVICE (no new uses-permission).
- UX: prompt-once notification-access + manual "Grant notification access"
  button wired into the D-pad focus chain (computed from visible controls);
  en/ru/zh strings.
- Tests: 11 isolated unit tests — module-global + tmp_path history isolation,
  push routing contract, callback-exception swallowing, None app-name, and a
  desktop-regression lock on backend selection order.
- Docs: README OS-support Android column (notification + audio cells),
  ANDROID-REVIEW status flipped to Implemented.

Zero new Python deps; no build.gradle.kts / Chaquopy pip changes.
2026-06-02 11:47:13 +03:00
alexei.dolgolyov 4b2e8fc5ec docs(android): add audio-capture design + missing-functionality review
- android-audio-capture-plan.md — design behind the merged on-device audio
  capture feature (487259a).
- android-missing-functionality.md — Android missing-feature review notes.
2026-06-02 03:30:43 +03:00
alexei.dolgolyov 487259a96d Merge feature/android-audio-capture: Android on-device audio capture
Push-based AndroidAudioEngine (AudioPlaybackCapture, API 29+) reusing the
MediaProjection token, feeding the unchanged AudioAnalyzer. Build green
(assembleDebug), 1854+13 tests pass. Reviewed: READY/SECURE, 0 blockers.
2026-06-02 03:28:37 +03:00
alexei.dolgolyov fd62db1720 feat(audio): Android on-device system playback capture
Enable audio-reactive lighting on the Android-TV build. A push-based
AndroidAudioEngine captures system playback audio via AudioPlaybackCapture
(API 29+), reusing the existing MediaProjection token, and feeds PCM into
the unchanged AudioAnalyzer pipeline. No new Python deps; no Chaquopy/pip
changes (numpy already bundled).

- Python: android_audio_engine.py — module-level queue + configure/
  push_samples/shutdown mirroring mediaprojection_engine; AndroidAudioEngine
  (priority 100) registered behind a guarded import. push_samples copies and
  defensively trims/clamps each block so the analyzer can't crash on
  variable-length or non-frame-divisible PCM.
- Kotlin: AudioCapture.kt — AudioRecord + AudioPlaybackCaptureConfiguration,
  fixed chunk-size block framing, little-endian float32, mic fallback;
  reads back the actual negotiated channel/sample rate. PythonBridge gains
  configureAudio/pushAudio/shutdownAudio with a cached module handle.
- Wiring: CaptureService starts/stops AudioCapture in the MediaProjection
  path (gated on API>=29 + RECORD_AUDIO + live projection); MainActivity
  requests RECORD_AUDIO; manifest declares it. Degrades gracefully when
  denied; root path stays audio-less by design.
- Tests: 13 desktop-CI tests incl. an over-length/non-divisible regression
  guard that exercises the full read_chunk -> AudioAnalyzer.analyze path.
2026-06-02 03:28:22 +03:00
alexei.dolgolyov 669ae20824 feat(value-sources): optional normalization for magnitude sources
Add an opt-out `normalize` flag to the four magnitude value sources
(ha_entity, http, system_metrics, game_event). get_value() stays in
[0,1] for every source (the normalized scalar-bus invariant), so all
existing consumers — brightness sinks, gradient_map, template `name`
bindings, and color-strip bindable floats — are unaffected.

- normalize=False is a clamp-passthrough: skip the min/max rescale and
  clamp the raw reading into [0,1] (for sources already reporting a
  0..1 fraction). The un-normalized magnitude stays available via
  get_raw_value() / template raw[name] / automations.
- Add get_raw_value() + a raw channel to GameEventValueStream (it had
  none). game_event's flag is model/stream-level only (no CRUD schema).
- Finite-safe clamp01() util; harden the composite-layer brightness
  multiply (latent negative-wrap / >=1 skip) with it.
- Preview WebSocket: tolerate non-numeric raw, generalize raw-range.
- Frontend: settings-toggle slider per HA/system_metrics/http editor
  with min/max grey-out; toggle hidden for fixed-mapping percent
  metrics. en/ru/zh locale keys.
- Additive optional field (default True) — JSON round-trip, no migration.

Tests: store create/update round-trip, clamp-passthrough, live
normalize flip, game_event raw channel + build_stream forwarding,
and finite-safe clamp01.
2026-06-02 02:24:40 +03:00
alexei.dolgolyov 6de61b965e feat(value-sources): add sandboxed-Jinja template combinator
A new `template` value source evaluates a hardened, sandboxed Jinja
expression over the live values of other value sources — the system's
first float combinator.

Backend:
- Shared engine (utils/template_expr.py): ImmutableSandboxedEnvironment with
  filters/tests and auto-injected globals stripped; only min/max/abs/round/
  clamp exposed; rejects **, string/collection-literal repetition, attribute
  access and non-global calls; NaN/inf-safe result coercion.
- TemplateValueSource model + TemplateValueStream runtime: compile-once,
  primitives-only eval context, raw[name] exposure, eval_interval throttle,
  ref-counted input acquire/release, rename-safe hot-update.
- Validation: unbound-variable + reserved-name rejection, reference
  cycle/depth guards (depth-only at create, full cycle at update), runtime
  acquire() depth backstop, and delete referential-integrity.
- API: Create/Update/Response schemas + discriminated unions, _RESPONSE_MAP,
  and an advisory POST /value-sources/validate-template endpoint.
- Demo seed: a static source plus a template combinator example.

Frontend:
- Editor modal section: repeatable inputs list (EntitySelect rows), a
  zero-dependency Jinja syntax highlighter, a hints/reference panel, and a
  debounced live validator that gates Save (stale-response-safe).
- Graph editor: read-only template node with one edge per input.
- i18n (en/ru/zh), icon, and card rendering.

Tests: engine, stream, factory/cycle, validate endpoint, and demo seed.
2026-06-01 18:53:56 +03:00
alexei.dolgolyov 12b40e6071 docs: actualize README and API reference, embed screenshots
README: add free & open-source (MIT) framing; add a Platforms table (Windows/Linux/macOS/Docker supported, Android-TV build marked experimental, scrcpy phone-capture retained); expand the LED device list to ~20 protocols; correct storage to SQLite, the auth model, and resource names (Output Targets, Automations); add a prebuilt-download section; remove the Architecture and Acknowledgments sections; embed dashboard, channels, live-preview, and device-picker screenshots.

docs/API.md: full rewrite from the live route modules — all 253 endpoints across 34 modules, grouped by resource with REST + WebSocket tables, accurate auth model and WS handshake, worked examples, and sensitive-endpoint markers. Replaces the stale v0.1.0 stub.

server/CLAUDE.md: storage is now SQLite (BaseSqliteStore / ledgrab.db / LEDGRAB_DATA_DIR / data_migrations.py); fix the auth description (loopback anonymous, LAN rejected with 401 — not all endpoints open); router registration happens in api/__init__.py.
2026-05-29 14:35:45 +03:00
alexei.dolgolyov 498854f04d refactor(storage): gate clone() behind an opt-in allowlist; expand duplicate tests
Follow-up polish from the duplicate-subgraph review:

- BaseSqliteStore.clone() now requires an opt-in `_cloneable` flag (default
  off), so a new or secret-bearing store can never be cloned by accident —
  defence-in-depth on top of the endpoint's `_DUPLICABLE_KINDS` restriction.
  Only value-source and colour-strip stores are flagged.
- Fix `_check_name_unique` annotation (`str | None = None`).
- Add tests: `layers[].brightness_source_id` remap, the warnings safety net
  firing, the clone allowlist invariant, and clone() refusing a non-cloneable
  store.
2026-05-29 11:55:58 +03:00
alexei.dolgolyov 15cfb821d3 feat(graph): duplicate a selected subgraph server-side
A secret-safe equivalent of blueprint import: the graph editor's overflow menu
gains "Duplicate selection", which deep-clones the selected value and
colour-strip sources server-side (full config preserved, never crossing the
wire) and rewires references that point within the selection — shared
dependencies (devices, HA sources, …) stay shared.

- graph_schema.remap_refs: write-twin of extract_refs (same dot/list/bindable
  grammar) that rewrites only in-selection ids; 8 unit tests.
- BaseSqliteStore.clone(): faithful deep-copy clone (no schema round-trip, so no
  field is lost), prefix-preserving fresh id; reusable by any store.
- POST /api/v1/graph/duplicate: two-pass clone-then-rewire restricted to value /
  colour-strip sources (no inline secrets), with a safety net flagging any
  unremapped reference; 7 integration tests vs real stores.
- Frontend: duplicateSubgraph (+cache invalidation), graphDuplicateSelection
  (reload + reselect the new cluster), overflow-menu item, i18n (en/ru/zh).
2026-05-29 11:45:55 +03:00
alexei.dolgolyov 2e51f46dfd feat(graph): make the visual editor a full wiring control surface
Lets users wire the system end-to-end from the graph, and fixes the core
bug that made drag-to-wire silently fail.

- Fix drag-to-wire 422s across 5 entity kinds: updateConnection() now echoes
  the target's discriminator (source_type/stream_type/target_type) into the
  partial PUT, so value/colour-strip/audio/picture sources and output targets
  all wire correctly. New contract test (54 cases) in test_graph_wiring_contract.py.
- Re-wire composite layers / mapped zones from the graph (right-click a
  layer/zone source edge -> Re-wire). Whole-list write preserves every sibling
  layer/zone setting, with an optimistic-concurrency guard and undo.
- Secret-safe /graph topology: project entities to id/name/subtype + reference
  roots so the endpoint cannot leak webhook tokens or other credentials.
- Carry slot indices on list edges; node custom-icon + schema-drift refinements;
  rewire i18n keys (en/ru/zh); wiring-control roadmap (TODO.md).
2026-05-29 02:29:19 +03:00
alexei.dolgolyov 05cf121666 fix(installer): open WebUI once after "Launch LedGrab"
The finish-page launch action started the app and also opened
http://localhost:8080/ itself, but the app already auto-opens the
browser on a manual (non-autostart) launch, so the page appeared
twice. Drop the installer's redundant open and let the app handle it
(it waits on /health, which is more reliable than a fixed sleep).
2026-05-28 23:52:52 +03:00
alexei.dolgolyov d505388f0e docs: graph-editor wiring-control roadmap (review findings A1-D6) 2026-05-28 23:38:04 +03:00
alexei.dolgolyov 6aeda935f1 chore: release v0.8.1
Build Android APK / build-android (push) Failing after 11s
Build Release / create-release (push) Successful in 6s
Build Release / build-docker (push) Successful in 2m57s
Build Release / build-linux (push) Successful in 1m49s
Build Release / build-windows (push) Successful in 3m35s
Build Release / publish-release (push) Successful in 2s
2026-05-28 23:35:35 +03:00
alexei.dolgolyov a5effba553 feat: aggregated snapshot + wiring-graph APIs, MQTT device brokers
Backend
- snapshot: GET /api/v1/snapshot aggregates targets, devices, sources,
  presets and system into one payload for the HA coordinator, collapsing
  the prior ~2N+M request fan-out; per-section ?include= gating.
- graph: GET /api/v1/graph{,/schema,/dependents} backed by a pure,
  unit-tested graph_schema engine — one authoritative connectable-field
  registry so the editor no longer hard-codes topology in two places.
- devices: thread mqtt_source_id through DeviceCreate/Update/Response and
  the routes for multi-broker MQTT; shared validate_mqtt_source_exists
  (_mqtt_validation.py) reused by device + output-target routes; stop
  update_device masking intentional 4xx as 500.
- shutdown: bound uvicorn graceful-shutdown via GRACEFUL_SHUTDOWN_TIMEOUT
  (shared by __main__, android_entry, demo) so a lingering events WebSocket
  can't strand LED targets or block process exit.
- access log: structured _access_log middleware attributing each request to
  its authenticated token label (never the secret); uvicorn access_log off.

Frontend
- graph editor: generic schema-driven port/edge rendering, layout and
  connection handling; service-worker refresh.
- device modals: MQTT broker EntitySelect for device_type=mqtt in add-device
  and settings, wired into load/save/validate/dirty-check/clone.
- i18n: en/ru/zh keys.

Tests: graph routes + schema, snapshot routes, access log, mqtt_source_id
device regressions, bounded-shutdown entrypoint. 1614 passed.
2026-05-28 22:51:04 +03:00
alexei.dolgolyov b83a72e63f chore: release v0.8.0
Build Release / create-release (push) Successful in 3s
Build Android APK / build-android (push) Failing after 8s
Build Release / build-linux (push) Successful in 3m8s
Build Release / build-docker (push) Successful in 4m11s
Build Release / build-windows (push) Successful in 4m55s
Build Release / publish-release (push) Successful in 1s
2026-05-28 17:48:06 +03:00
alexei.dolgolyov 0d840adfca fix(ctypes): share wintypes.MSG with platform_detector to avoid argtype races
WindowsShutdownGuard was binding user32.GetMessageW.argtypes with
POINTER(_MSG) (project-local struct), while PlatformDetector's display-
power monitor binds it with POINTER(wintypes.MSG). argtypes is a
mutable global on the cached WinDLL handle, so whichever module
imported last won, and the other module's byref() then tripped
Python 3.13's strict argtype check with
"expected LP_MSG instance instead of pointer to _MSG".

The two structs are byte-identical (same field types in the same
order, just pt vs pt_x/pt_y naming) and we never touch the pt field,
so aliasing _MSG to wintypes.MSG eliminates the conflict — both
modules now bind the same POINTER class, the writes become idempotent,
and the full test suite passes regardless of import order.

CI runs on Linux so this never fired in release builds, but it broke
the local Windows test run.
2026-05-28 17:36:19 +03:00
alexei.dolgolyov 1f959932c1 fix(notification): allow clearing the sound on per-app overrides and main row
Previously, both the per-app override sound dropdown and the main
notification sound row only attached an EntitySelect when at least
one sound asset was registered — and never with allowNone. That left
users with no way to pick "no sound" once an entry existed, and made
the override dropdown silently inert before any assets were added.

Always construct the EntitySelect (so an empty assets list still
renders a usable, searchable input) and pass allowNone with the
localized none label so "no sound" is a first-class choice in both
the override list and the main row.
2026-05-28 17:28:34 +03:00
alexei.dolgolyov 10eb24b2ce docs: dashboard innerHTML reconciliation review notes
Design doc capturing the architectural pattern behind the perf-card
flicker (every innerHTML rewrite tears down the subtree, restarting
CSS animations / dropping focus / wasting layout) and the two
drivers — poll timers and server:* push events. Inventories every
remaining latent site in perf-charts.ts / dashboard.ts /
entity-events.ts / game-integration.ts, walks the decision ladder
from a setInnerHtmlIfChanged helper through hand-rolled cell
components to a Lit migration, and lands on Lit for polling-heavy
modules with entity-events.ts tab reconciliation sequenced ahead of
the dashboard cards because of its higher blast radius.

Planning artifact only — no implementation here.
2026-05-28 17:26:56 +03:00
alexei.dolgolyov 66b85b0175 fix(css-editor): persist notification_sound + notification_volume
The CSS editor modal collected every other notification field on
save but silently dropped notification_sound and notification_volume,
so toggling them in the modal had no effect on the saved strip.
Include both in the save payload alongside the existing notification
fields.
2026-05-28 17:26:44 +03:00
alexei.dolgolyov bc42604045 ci(release): publish release only after every build job uploads assets
create-release now creates the release as a draft so users never see
a release page that's missing artifacts (or, worse, missing the
sha256 sidecars that the in-app updater requires). A new
publish-release job runs after create-release, build-windows,
build-linux, and build-docker all succeed, and PATCHes the release
to draft=false in one step. If any build fails, the draft stays
hidden and can be deleted manually.
2026-05-28 17:26:28 +03:00
alexei.dolgolyov 3645216669 feat(icons): spectrum aperture icon set + dedicated tray variant
Regenerate the LedGrab icon family from a single Pillow script
(build/generate_icon.py): a rounded-square aperture traced by a
continuous RGB color-wheel stroke over a vignette canvas with a soft
chromatic bloom. 4x supersampled then downsampled per output for
crispness. Outputs 192/512 standard, 512 maskable (safe-area padded
for PWA round-crops), and a new 256 transparent-background tray
variant so the taskbar icon reads cleanly against light themes
instead of showing a dark tile.

icon.ico now embeds 16/24/32/48/64/128/256 frames sourced from the
transparent tray master, fixing the dark-square halo around the
file/taskbar icon on light Windows themes.

__main__ picks icon-tray.png for the tray and falls back to
icon-192.png when the tray asset isn't present (older bundles /
forks).
2026-05-28 17:26:18 +03:00
alexei.dolgolyov 85da2e538d feat(backup): bundle assets in ZIP + partial-write hardening + restart log
Auto-backups now produce a ZIP containing ledgrab.db plus every file
in the assets dir under assets/ — matching the manual
GET /api/v1/system/backup format, so restore accepts either output
interchangeably. Legacy .db backups remain listable, restorable, and
prunable; both extensions count toward max_backups.

Writes stage to <name>.partial then os.replace into place — a crash
mid-ZIP never leaves a half-written backup that masquerades as valid.
Stale .partials from prior crashes are swept on the next run.
Symlinks inside the assets dir are skipped so a hostile link can't
slurp a target outside the dir into every backup. Backups larger than
500 MB log a warning so operators notice unbounded asset growth before
disk fills up.

restart.py: redirect the spawned restart script's stdout/stderr to
restart.log and bail out early if the script is missing — silent
failures (PowerShell off PATH, restart.ps1 erroring) used to vanish
into a detached child with no diagnostic trail.

Tests cover happy path, asset bytes round-trip, partial cleanup,
None/missing assets_dir, failure rollback, stale-partial sweep,
symlink rejection, mixed legacy+new listing, and cross-format prune.
2026-05-28 17:25:55 +03:00
alexei.dolgolyov e4d24a02da fix(ctypes): pin LPMSG across MSG-pump prototypes for Python 3.13
Python 3.13's ctypes tightened argtypes checks and now rejects
mismatched POINTER(MSG) cache entries — each call to POINTER(MSG)
can return a class identity that doesn't match what byref() of an
instance produces, raising "expected LP_MSG instance instead of
pointer to _MSG" inside GetMessageW/TranslateMessage/DispatchMessageW.

Capture POINTER(MSG) once into LPMSG and reuse the same class object
across all three prototypes in both the WindowsShutdownGuard pump and
the PlatformDetector display-power monitor. Restores the 4 failing
win_shutdown tests.
2026-05-28 17:25:37 +03:00
alexei.dolgolyov bb3a316e35 refactor(frontend): shared API client + automations registry (audit M7, H8)
H8 — automations.ts rule-type registry
  Convert the two hand-rolled RuleType dispatch ladders into per-type
  registries (RULE_FIELD_RENDERERS + RULE_COLLECTORS) keyed by RuleType,
  joining the existing RULE_CHIP_RENDERERS. All three are typed
  Record<RuleType, ...> for compile-time exhaustiveness; an import-time
  _assertRuleHandlerCoverage() check logs loudly if any registry drifts
  from RULE_TYPE_KEYS — mirrors the backend's _RULE_HANDLERS shape, the
  one intentional divergence being that the frontend logs rather than
  throws (a thrown error at module import would brick the whole bundle,
  not just the editor).

M7 — shared API client + 35 file migrations
  New core/api-client.ts wrapping fetchWithAuth with typed apiGet /
  apiPost / apiPut / apiPatch / apiDelete. Auth, 401-relogin, retry,
  timeout, and the offline toast all stay owned by fetchWithAuth; the
  client just collapses the
  if (!resp.ok) { detail || HTTP <status> } ... resp.json()
  dance into one typed call. The detail unwrap is hardened to join
  FastAPI validation arrays instead of stringifying to [object Object].

  35 feature/core files migrated to it across many batches, reviewer-
  approved for behaviour parity in three passes covering the riskier
  divergences (bulk Promise.allSettled deletes, inline-error saves,
  array-detail joins, silent-failure GETs, immutable clones).

  9 files remain on fetchWithAuth — the big god-modules tied to the
  pending C8/C9/C10 splits (streams, settings, targets, dashboard,
  color-strips/index, graph-editor, assets, value-sources) plus
  pairing-flow which by design stays on raw fetch (branches on raw
  Response.status codes).

i18n — 14 new locale keys (en / ru / zh)
  Added save/load/delete error keys across automations, pattern,
  audio_processing, audio_template, templates, gradient, target,
  device namespaces, plus backfilled gradient.error.delete_failed into
  ru/zh. Scan confirms no hardcoded English errorMessage strings
  remain in the migrated diff.

AUDIT_REMAINING.md updated to reflect H6, H8, and M7 status.

Verified: tsc --noEmit clean + npm run build clean after every batch.
2026-05-28 14:58:08 +03:00
alexei.dolgolyov 49c35a2ea0 refactor(frontend): split types.ts into 18 per-entity files (audit H6)
Convert the 1140-LOC types.ts into a pure re-export barrel backed by
focused per-entity files under types/, joining the existing
bindable.ts. Every import { ... } from '../types.ts' resolves
unchanged; reviewer-confirmed all 102 type exports preserved.
2026-05-28 14:57:25 +03:00
alexei.dolgolyov ef1f9eade2 feat(android): production-readiness pass — security, perf, compat, UI/UX
Multi-axis lift to ship-quality after a full review:

Security
- ApiKeyManager: per-install random API key, persisted via SharedPreferences
  with synchronous first-write; threaded into uvicorn via the
  LEDGRAB_AUTH__API_KEYS env var; embedded in QR as a URL fragment (#k=)
  so it never appears in HTTP requests or server logs; frontend reads
  location.hash on first visit and strips it via history.replaceState
- Root.runAsRoot(argv: Array<String>) overload with POSIX shell-quoting to
  eliminate the shell-injection footgun (= excluded from unquoted-safe set)
- UsbSerialBridge: ContextCompat.RECEIVER_NOT_EXPORTED + intent.package
  check in the broadcast receiver for defence-in-depth across API levels
- Release builds refuse to silently fall back to debug keystore; require
  ANDROID_KEYSTORE_* env vars or explicit
  ANDROID_ALLOW_DEBUG_SIGNED_RELEASE=1
- Crash log retention capped at 10 entries
- Fatal-error stack trace hidden behind a toggle on the error screen

Performance
- ScreenCapture / RootScreenrecord reuse a single RGBA ByteArray per
  pipeline instead of allocating per frame — eliminates ~15 MB/s GC churn
  at 30 fps on low-end TV boxes
- Frame pacer switched from System.currentTimeMillis() + integer division
  (~30.3 fps drift) to SystemClock.elapsedRealtimeNanos with a catch-up
  accumulator
- ScreenCapture computes capture dimensions from source aspect ratio so
  non-16:9 displays don't get squashed
- RootScreenrecord input pump backs off 5 ms when MediaCodec is starved,
  ending a tight spin that burned a CPU core on decoder stalls
- QR cached by URL — onResume from background no longer rebuilds the
  560×560 bitmap each time
- ApiKey commit() pre-warmed off Main on app startup

Compatibility
- compileSdk / targetSdk bumped to 35 (Play Store requirement)
- armeabi-v7a build path added to build script + conditionally included
  in gradle splits when the matching wheel is present in android/wheels/
- Foreground service type declared as mediaProjection|specialUse with
  PROPERTY_SPECIAL_USE_FGS_SUBTYPE rationale; promotion via
  ServiceCompat.startForeground with the correct type per mode
- NetworkUtils picks Ethernet > Wi-Fi > VPN > cellular instead of just
  activeNetwork — fixes wrong-URL on TV boxes with both Ethernet + Wi-Fi
- enableOnBackInvokedCallback=true for Android 15 predictive-back
- Splash screen API via androidx.core:core-splashscreen — hides Chaquopy
  stdlib unpack delay on cold first launch

UI / UX
- All previously hardcoded English strings (root prompt, permission
  denial, fatal-error screen, notification text) now localised across
  en/ru/zh
- Monochrome notification icon (was a colored launcher → gray blob in
  status bar)
- 320×180 TV banner (was the square launcher → squashed on Leanback row)
- ViewStub-based running panel (deferred inflation)
- ObjectAnimator pulse on the Running status dot for liveness feedback
- "Starting…" button state while root is being probed
- Autostart checkbox hidden entirely on unrooted devices
- "No network" status when getLocalIpAddress returns null
- QR fallback hint text
- Animator cancelled in onStop to avoid leaking view hierarchy

Lifecycle hardening (from review)
- RootScreenrecord: processLock serialises EOF respawn vs concurrent
  stop() to prevent orphaned screenrecord processes
- CaptureService.restartRootPipeline: publish-before-start under
  @Synchronized to close the orphan window during watchdog restarts
- ScreenCapture.MediaProjection.Callback.onStop just flips
  running=false instead of calling stop() (which self-joined
  captureThread and hung 500 ms)
- updateUI early-returns when lateinit not initialised (fatal-error path)
- Watchdog give-up bound fixed (>= instead of >, was allowing 4 attempts)

server/android_entry.py accepts an optional api_key, sets
LEDGRAB_AUTH__API_KEYS={"android":<key>} as JSON before any LedGrab
import, logs a clear error if pydantic-settings parsing doesn't land
the value back in config (defensive guard against future settings
behaviour drift).

server/static/js/app.ts: bootstrap reads #k= from location.hash,
persists to localStorage, then strips via history.replaceState.

Two independent code-review passes; 147 relevant server tests still
pass; TypeScript and ruff clean.
2026-05-26 12:52:14 +03:00
alexei.dolgolyov 8bdcc17799 chore: release v0.7.0
Build Release / create-release (push) Successful in 3s
Build Android APK / build-android (push) Failing after 11s
Build Release / build-linux (push) Successful in 2m54s
Build Release / build-docker (push) Successful in 3m50s
Build Release / build-windows (push) Successful in 4m36s
2026-05-26 00:35:38 +03:00
alexei.dolgolyov f591e258f7 fix(storage/database): reopen connection on lifespan restart
Database opened its sqlite3 connection eagerly in __init__ and closed it
in close(); the lifespan called close() on shutdown. In production this
is fine — the lifespan runs once per process. Under pytest the module-
level ``db`` singleton survives across every TestClient session, so the
second test file's lifespan startup hit
``sqlite3.ProgrammingError: Cannot operate on a closed database`` at
fixture-setup time (AutoBackupEngine.__init__ → db.get_setting("…")
was the first reader). 65 spurious "errors" on a full Windows pytest run.

- Database: extract _open() from __init__, add ensure_open() that
  reopens iff _conn is None, and have close() null _conn after the
  TRUNCATE checkpoint so re-close is idempotent.
- main.py lifespan startup: call db.ensure_open() before any setting
  read, so subsequent TestClient sessions get a live connection.
- tests/storage/test_database_reopen.py: pin the four invariants —
  close→ensure_open round-trips data, ensure_open is a no-op when
  open, close is idempotent, and using the DB after close without
  ensure_open raises (callers must opt in).

Full backend suite: 1551 pass / 1 skip / 0 errors. Ruff clean.
2026-05-26 00:26:36 +03:00
alexei.dolgolyov f6486f9b34 perf(dashboard): diff FPS charts + cache spark SVG nodes; i18n perf strings
- dashboard: only destroy/recreate FPS charts for added/removed/detached
  targets; skip the history fetch when local samples already exist.
  Drops sync-clock `is_running` from the structure signature so toggles
  don't trigger a full rebuild; route clock/automation refresh through
  the in-place path.
- perf-charts: cache SVG skeleton per spark host and mutate node
  attributes instead of rewriting `innerHTML` every poll. Memoize
  patches/devices rendering by content signature so unchanged ticks
  no longer restart CSS animations. Skip render for env-hidden cards.
- perf-charts: switch `/system/performance` poll to `fetchWithAuth`,
  re-read `dashboardPollInterval` per tooltip move, and route the
  remaining hardcoded English strings ("no captures", "{n} total",
  "{rate} skipped/s", tooltip age, metric labels) through `t()`.
- locales: add `perf.no_captures`, `perf.captures_count`,
  `perf.ratio_of_requested`, `perf.total_count`, `perf.skipped_per_sec`,
  `perf.tip.now`, `perf.tip.ago` in en/ru/zh.
2026-05-26 00:12:29 +03:00
alexei.dolgolyov 48dbdb90e9 docs(review-todo): check off items addressed in 2026-05-23 autonomous pass
Mark devices.py PATCH fix, WLED route-level test, IPv6 regression
test, IconSelect XSS audit, PEP-604 sweep, magic-number constants,
api/auth except specificity, and the (window as any) static-access
cleanup as done. Defer items are unchanged: performance items keep
their "profile first" caveat, Hue cert pinning + CSP keep the design-
sensitive note, architecture refactors keep the multi-day banner,
and i18n parity is now annotated with the exact missing-key counts
(328 ru / 325 zh) so the next translator pass has a clear scope.
2026-05-23 01:22:41 +03:00
alexei.dolgolyov 003517247f refactor(types): migrate (window as any) statics to typed window globals
59 sites across 19 feature modules switched from
`(window as any).foo` to the typed `window.foo` form against
global-types.d.ts. The 7 remaining sites use dynamic string indexing
(`window[fnName]`) where a typed access is impossible — those keep
the narrow cast and are documented as the legitimate exception in
the typedef file's header.

global-types.d.ts grows entries for `loadIntegrations`,
`loadUpdateSettings`, `loadUpdateStatus`, `initUpdateSettingsPanel`,
`onTestDisplaySelected`, `openSettingsModal`, `renderAboutPanel`,
`switchSettingsTab`. The `applyAccentColor` signature is widened to
accept the (accent, persist) call shape observed at the appearance
preset call site so tsc validates the real contract.
2026-05-23 01:22:29 +03:00
alexei.dolgolyov 888f8fd16e refactor(types): PEP-604 union sweep + UP007/UP045 enforcement
ruff --select UP007,UP045 --fix converted ~1760 sites across the
backend: `Optional[T]` → `T | None`, `Union[X, Y]` → `X | Y`. The
remaining module-level alias targets that ruff conservatively skips
(BindableFloatInput, ColorList, DeviceConfig) were converted by hand
earlier in the pass. black -formatted the result so the wider unions
fit cleanly under the 100-char line budget.

pyproject.toml now sets [tool.ruff.lint] extend-select = ["UP007",
"UP045"] so future legacy imports fire CI on every push. The
pre-commit ruff hook was bumped from v0.8.0 -> v0.15.12 to recognise
UP045 (split off from UP007 in v0.13).
2026-05-23 01:21:44 +03:00
alexei.dolgolyov ea7ee88490 refactor(api/auth): narrow WS exception catches + observability log
The 11 except Exception sites around websocket.send_json and
websocket.close are now except _WS_SEND_BENIGN_EXC — a narrow tuple of
WebSocketDisconnect, RuntimeError, ConnectionError, OSError. Real
programming errors (AttributeError, TypeError) no longer silently
disappear inside the handshake path. The receive_text branch grows a
narrow `(RuntimeError, ConnectionError, OSError)` case plus a final
`except Exception: logger.exception(...)` catch-all so genuinely
unexpected error shapes are recorded with a stack trace instead of
being swallowed.
2026-05-23 01:14:43 +03:00
alexei.dolgolyov d38021f061 refactor(processing): hot-path magic numbers -> named module constants
processed_stream lifts the 30-iteration filter recheck cadence to
_FILTER_RECHECK_EVERY_N_FRAMES with a comment explaining the 30 fps
trade-off. wled_target_processor lifts SKIP_REPOLL, the diagnostics
interval, and the CSPT recheck cadence to module-level
_SKIP_REPOLL_SLEEP_SECONDS, _DIAGNOSTICS_REPORT_INTERVAL_SECONDS, and
_CSPT_RECHECK_EVERY_N_ITERATIONS. Tests can monkeypatch them now.
2026-05-23 01:14:31 +03:00
alexei.dolgolyov 507e1385a6 feat(ui/icon-select): defence-in-depth XSS sanitiser on icon channel
Every IconSelect caller was audited: each builds item.icon from a
constant ICON_* literal, a lookup-table getter, or
renderDeviceIcon(stored_id) — none of which embed user input today.
The new sanitiseIcon() helper is the belt-and-braces guard for a
future caller that forgets the trusted-SVG contract: reject icon
strings containing <script>/<iframe>/<embed>/<object>/javascript:/on*=
markers, warn to the console, and fall back to empty so the cell still
renders the (escaped) label + desc.
2026-05-23 01:13:55 +03:00
alexei.dolgolyov 907bdaf043 test(url-scheme): WLED route-level integration + IPv6 regression
TestWLEDSchemeInference in test_devices_routes covers the POST/PUT
create-and-update flow with a stubbed WLED provider so the
infer_http_scheme integration hop has end-to-end coverage instead of
just the unit tests.

test_url_scheme grows public IPv6 (Cloudflare / Google / Quad9 DNS),
bracketed-form, and ULA cases. Adds an explicit pin for the Python
ipaddress documentation-prefix quirk (2001:db8::/32 is is_private,
so it routes to http:// even though some audits colloquially call it
"public").
2026-05-23 01:13:44 +03:00
alexei.dolgolyov 0dd8d430b9 fix(devices): preserve existing URL on PATCH-without-url
When a PATCH omits `url` (rename / icon-only edit), normalized_url
arrived at the processor as None and the manager kept whatever it had
cached — or refused to re-sync if it had nothing. Fall back to
existing.url so the processor is always told the current address.
Surfaced by the production-review backlog.
2026-05-23 01:13:13 +03:00
alexei.dolgolyov fd46c51dba docs: TODO + CLAUDE.md notes + locale keys for new features
TODO.md grows the device-support follow-up roadmap. CLAUDE.md trims a
stale section. en/ru/zh locales add the strings used by the new
HTTP-endpoint editor, MiniSelect labels, automations expansion, and
value-source kinds. Ru/zh parity for the older keys is tracked
separately in REVIEW_TODO.md.
2026-05-23 00:50:31 +03:00
alexei.dolgolyov ddae5719cf chore(frontend-infra): inbound-event allowlist + storage/state touch-ups
events-ws gains an inbound-event allowlist matching the new server-side
allowlist; test_events_ws_parity pins the two lists in sync.
state + storage modules and the streams / integrations /
z2m-light-targets / streams-*-templates editors absorb the
closeIfPristine guard alongside small UX fixes. css-editor template
picks up the new MiniSelect markup for the filter-kind picker.
2026-05-23 00:50:15 +03:00
alexei.dolgolyov 898912f8b1 chore(backend): MQTT/WLED/devices/capture/utils + api routes hardening
Bundle the remaining backend touch-ups that the production review
landed individually as small surgical edits across many modules:
- MQTT runtime: fire-and-forget task tracking + drain resilience.
- mqtt_source + store + storage/color_strip_source: secret_box
  encryption for credentials with auto-migration of plaintext fields.
- devices/discovery_watcher: task tracking on watcher start/stop.
- devices/wled_client + wled_provider: URL scheme inference helper
  applied at the create/update boundary so bare hostnames stay valid.
- core/capture/screen_capture: hardened error paths.
- core/processing (mapped/processed/processor_manager/video/wled_target):
  smaller follow-throughs from the registry refactor that landed
  earlier on the branch.
- utils/safe_source + utils/file_ops + utils/__init__: shared URL +
  IP classification helpers + larger streaming upload size caps.
- api/auth: WebSocket Origin allow-list + /docs auth-gate.
- api/dependencies: register the new HTTP-endpoint store.
- api/routes (assets, backup, webhooks): streaming-upload caps +
  asyncio.gather return_exceptions on broadcast loops.
- tests/test_api + tests/e2e/test_backup_flow: cover the new caps and
  the Origin allow-list.
2026-05-23 00:50:01 +03:00
alexei.dolgolyov 45d12b2811 feat(update-service): SSRF-validated redirects + restart hardening
update_service grows explicit URL validation on the redirect chain so a
hostile mirror can't bounce the updater to a private IP. restart.ps1
gets stricter argument handling and clearer log lines.
default_config.yaml exposes the new toggles. test_system_routes pins
the new behaviour.
2026-05-23 00:49:18 +03:00
alexei.dolgolyov 826e680f37 refactor(color-strip): rename static -> single + frontend follow-through
The "static" source kind always rendered a SINGLE color and the name
confused new code paths. Rename the module + kind to "single". Storage
keeps backward-compatible serialisation. Frontend color-strip cards /
gradient / index / test modules and the affected tests follow the new
name.
2026-05-23 00:49:00 +03:00
alexei.dolgolyov 737fd72b73 feat(value-sources): extend storage + schema + UI alongside new kinds
Storage model + Pydantic schema + route gain fields for the value-source
kinds introduced by the per-type factory refactor. Frontend editor adds
inputs for them and the modal template grows new field rows.
2026-05-23 00:48:48 +03:00
alexei.dolgolyov 3fe66d80cb feat(automations): expand automation rules + UI + engine coverage
Storage model + Pydantic schema + route surface gain the new rule
shapes the engine already supports. Frontend automations editor grows
the matching inputs. New core/test_automation_engine.py pins the
dispatch table rules behind ~285 lines of unit coverage.
2026-05-23 00:48:19 +03:00
alexei.dolgolyov f03cb303c3 feat(modal): closeIfPristine save-guard + per-editor adoption
Modal gains closeIfPristine(entityId): when editing an existing entity
and no tracked field has changed, the helper force-closes the modal
silently and returns true so the caller can skip the PUT and the
misleading "updated" toast. Each editor's save handler now early-returns
on the no-op edit path: advanced-calibration, assets,
audio-processing-templates, audio-sources, calibration, devices,
game-integration, ha-light-targets, home-assistant-sources,
mqtt-sources, pattern-templates, scene-presets, sync-clocks, targets,
weather-sources.
2026-05-23 00:48:00 +03:00
alexei.dolgolyov 9ff83bd6ca feat(ui): MiniSelect primitive + IconSelect XSS hardening + typed globals
MiniSelect replaces the forbidden plain <select> in editors where the
option list isn't large enough to justify the full IconSelect grid. It
shares the IconSelect look-and-feel (chip + dropdown panel, keyboard
nav, search). IconSelect grows an explicit HTML-escape for item labels
and keeps `item.icon` documented as a trusted-SVG sink for callers
that build the icon string from constants. global-types.d.ts gives
existing window.* accesses real types so feature modules can stop
falling back to `(window as any)`. The modal.css additions style the
two selectors and the new dropdown panels.
2026-05-23 00:47:45 +03:00
alexei.dolgolyov d6cc80074d feat(http-endpoints): introduce HTTP endpoint output target stack
New output kind that POSTs the current strip frame to a user-configured
HTTP endpoint, alongside WLED / MQTT / Hue. Stack mirrors the existing
output-target shape end-to-end: storage model + store, FastAPI router +
Pydantic schemas, JS feature module + modal template, router wiring in
api/__init__.py and the modal include in index.html. Tests cover both
the routes and the store.
2026-05-23 00:47:31 +03:00
alexei.dolgolyov 06273ba2bc chore(tooling): vex semantic-search config + REVIEW_TODO backlog
Add .vex.toml so `vex` is the project's primary code-search backend with
auto-update + semantic embeddings enabled. Ignore the .fastembed_cache/
directory that vex creates on first --semantic run. REVIEW_TODO.md
captures items flagged by the multi-agent production review that were
deliberately deferred (multi-day refactors, profile-first perf, and
design-sensitive security work).
2026-05-23 00:46:44 +03:00
alexei.dolgolyov 628c6b2f0d docs: capture architecture-audit remainder for follow-up sessions
10 commits in this branch landed the data-safety bugs (C2, C11), the
worst parallel-change problems (C1/C3/C4/C6/C7), and 5 of the 9 HIGH
audit findings. The remainder splits cleanly into four follow-up
sessions: a frontend sprint (C8/C9/C10/H6/H7/H8/M7-M11/L1), a Device
redesign (H4), a BaseTargetProcessor ABC (C5/H5/M1/M2/L2), and polish
(M3/M6/M12/L3-L5).

This file documents what's left, the recommended ordering, and the
three registry patterns already in the codebase that new contributors
should reach for instead of writing fresh if/elif chains.
2026-05-23 00:36:39 +03:00
alexei.dolgolyov 2f15fbb752 refactor(output-targets): registry + coverage assertion for response builders
``_target_to_response`` in ``api/routes/output_targets.py`` used to be
an isinstance ladder over the three OutputTarget subclasses with a
silent fallback that fabricated a ``LedOutputTargetResponse`` for
unknown types (audit finding H3). The fallback masked exactly the
kind of bug we hit on the CSS side in Phase 1.1: a new target subclass
slipped past the ladder and got mis-shaped on the wire.

Replace the ladder with a ``_TARGET_RESPONSE_BUILDERS`` dict keyed by
the concrete subclass plus an import-time
``_assert_target_response_coverage()`` that requires the registry to
exactly match ``{WledOutputTarget, HALightOutputTarget,
Z2MLightOutputTarget}``. ``_target_to_response`` now raises
``RuntimeError`` instead of silently fabricating a LED response for an
unknown subclass — coverage is asserted at import so this branch is
unreachable in normal operation.

Tests: 5 new regression tests cover bijection between expected classes
and registered builders, callable shape, the rogue-target-raises
contract, and missing/extra entry rejection in the assertion. 24
existing output-target tests stay green; ruff clean.
2026-05-23 00:03:01 +03:00
alexei.dolgolyov c1aa2ebec5 fix(value-source): preserve store contract for game_event + error precedence
Two HIGH issues surfaced by review of 3b8f00e:

1. ``_build_game_event`` was newly succeeding where the old store
   raised ``ValueError("Invalid source type: game_event")``. The
   coverage-assertion-symmetry comment was honest about it being
   a path that didn't exist before, but silent broadening of the
   create contract is a real behaviour delta — any internal caller
   that previously caught the error would now succeed.

   Make ``_build_game_event`` raise NotImplementedError. The
   coverage assertion still passes (the entry exists), but the
   historical "you can't create game_event sources through the
   store" contract is preserved. game_event instances continue to
   be wired up by the game-integration setup path.

2. The new ``create_source`` ran ``_check_name_unique`` BEFORE
   ``build_source``. When both ``source_type`` is invalid AND
   ``name`` collides with an existing source, the old code raised
   ``"Invalid source type: …"`` first; the new code raised the
   name-collision error. Swap the order: build first (which
   validates source_type), then check name uniqueness, then
   persist. Bonus: a uuid is no longer minted for a source we end
   up rejecting on type.

New test pins the game_event NotImplementedError so a future
refactor doesn't accidentally re-open the create path.

38 value-source-store + factory tests stay green; ruff clean.
2026-05-23 00:00:30 +03:00
alexei.dolgolyov 3b8f00e3f9 refactor(value-source): per-type factories for create / update dispatch
ValueSourceStore.create_source used to be a ~260-line if/elif chain
over 14 source_type strings; update_source did the same dance again
with 14 isinstance branches (audit finding C7 store-side). Each
branch duplicated the common-fields scaffold and the per-type
defaulting + validation logic.

Lift each per-type create / update body into a free function in a
new ``storage.value_source_factories`` module:

  * ``CREATE_BUILDERS[source_type]`` — owns defaulting + per-type
    validation (HA needs ha_source_id + entity_id; gradient_map
    needs value_source_id; system_metrics validates against
    VALID_SYSTEM_METRICS; http rejects interval_s < 1; the two
    adaptive_* sub-modes route to the same AdaptiveValueSource
    class with different source_type discriminators).
  * ``UPDATE_APPLIERS[source_type]`` — mirrors the above on the
    update side; ``resolve_ref`` is applied to cross-entity
    references so empty-string clears keep working.
  * ``build_source(...)`` / ``apply_update(source, **kwargs)`` are
    the public entry points the store calls.
  * ``_assert_factory_coverage()`` runs at module import and
    requires BOTH registries to match storage's _VALUE_SOURCE_MAP
    exactly.

The store's ``create_source`` shrinks from ~260 lines to ~25;
``update_source`` from ~200 lines to ~40.

Tests: 14 new tests cover registry coverage in both directions
plus drift assertions, representative builder paths (static /
adaptive_time / adaptive_scene / ha_entity / http / unknown),
the AdaptiveValueSource dual-source-type discriminator, and
several applier paths including ``**_`` swallowing unknown kwargs
and HTTP zero-interval rejection. 47 existing value-source store
tests stay green; 769 storage / core / api tests in aggregate.
Ruff clean.
2026-05-22 23:56:10 +03:00
alexei.dolgolyov 05f73eedf9 refactor(types): extract bindable primitives into types/bindable.ts (H6 partial)
types.ts is 1159 lines of kitchen-sink discriminated-union definitions
(audit finding H6) shared across ~30 frontend modules. Splitting the
whole file in one pass would need careful per-group TypeScript wrangling
and verification across every entity shape; this commit lands the
first slice as a proof of pattern.

What changed
------------

* New ``static/js/types/bindable.ts`` owns ``BindableFloat`` /
  ``BindableColor`` plus their four accessor helpers
  (``bindableValue``, ``bindableSourceId``, ``bindableColor``,
  ``bindableColorSourceId``).
* ``static/js/types.ts`` keeps every interface and union shape that
  references those primitives, but the primitives themselves now come
  from the new file. Re-export keeps every existing ``import { ... }
  from '../types.ts'`` site working unchanged.

Why this slice first
--------------------

Bindable types and helpers are the most heavily-imported piece of
types.ts (~30 modules already use ``bindable*`` helpers) and they have
zero downstream dependencies — they don't reference any other type
group. That makes them the safest extraction and the cleanest
demonstration of the barrel-re-export pattern the remaining groups
(devices, sources, integrations, automations, templates, …) will
follow in a follow-up sprint.

Verification
------------

* ``npx tsc --noEmit`` clean (no compile errors anywhere in the
  frontend tree).
* ``npm run build`` clean (esbuild bundle and CSS bundle produced
  without warnings).

The remaining ~1130 lines of types.ts plus the C8/C9/C10 god-module
splits (value-sources.ts, streams.ts, graph-editor.ts) need a
dedicated frontend session with typescript-reviewer + manual UI
testing — deferring those rather than half-finishing them here.
2026-05-22 23:35:42 +03:00
alexei.dolgolyov 9f3f346543 refactor(value-source): MetricSpec registry for SystemMetricsValueStream
SystemMetricsValueStream used to dispatch on its ``self._metric`` string
across three independent if/elif chains (audit finding M5):

  * priming in ``start()`` (cpu_percent seed, initial network counter);
  * raw reading in ``_read_metric_psutil`` plus ``_read_metric_fallback``;
  * normalisation in ``_normalize`` (percent / min-max range / max-rate).

Adding a new metric meant editing all three chains plus the Android
fallback — and forgetting one branch made the metric silently return 0.

Lift each per-metric concern into a free function and register them as a
``MetricSpec(name, read_psutil, read_fallback, normalize, prime)`` in a
new ``core.processing.metric_readers`` module. Shared normalisers
(``_norm_percent`` / ``_norm_range`` / ``_norm_rate`` / ``_zero``) live
once. The stream's ``start()`` / ``_read_metric()`` / ``_normalize()``
collapse to a single registry lookup + delegation.

The stream still owns its mutable state (``_disk_path``,
``_sensor_label``, ``_gpu_unavailable``, ``_prev_net_bytes``,
``_prev_net_time``, etc.) — readers operate on the stream by
parameter, not by inheritance, so the kitchen-sink class shrinks by
~140 lines without losing the per-stream cadence bookkeeping. Each
spec function's docstring documents which fields it reads or mutates.

Tests: 16 new tests cover the 10-metric coverage set, callable shape
of every spec field, the three normaliser primitives' clamping +
divide-by-zero behaviour, prime-hook presence (only the three metrics
that need a baseline: cpu_load + network_rx + network_tx), and
fallback-path expectations (desktop-only sensors -> _zero, cpu/ram ->
real MetricsProvider).

754 existing core / storage / api tests stay green; ruff clean.
2026-05-22 23:29:33 +03:00
alexei.dolgolyov 98fb61d932 refactor(automations): rule dispatch via class-level handler table
AutomationEngine._evaluate_rule used to rebuild a 9-entry dispatch
dict on EVERY rule evaluation (audit finding H2). Unknown rule types
silently returned False — adding a new Rule subclass without an entry
just made it inert forever.

Refactor:

  * Per-rule-type bodies are now ``_handle_<kind>(self, rule, ctx)``
    methods on AutomationEngine.
  * A ``_RuleEvalContext`` frozen dataclass bundles all the
    cross-cutting state (running_procs, topmost_proc,
    topmost_fullscreen, fullscreen_procs, idle_seconds, display_state)
    so adding a new handler does not require widening
    ``_evaluate_rule``'s parameter list.
  * ``AutomationEngine._RULE_HANDLERS`` is bound once at module-import
    time after the class is defined.
  * ``_assert_rule_handler_coverage()`` runs at import: every Rule
    subclass imported by the module must have an entry, and entries
    keyed by an unknown class are also rejected.

Unknown-type fallback now logs a warning instead of silently returning
False, so a future Rule subclass missing from the registry surfaces in
operator logs rather than just behaving as if the automation were off.

The pure storage layer (storage/automation.py) is untouched — the
handler bodies stay on the engine where the cross-layer dependencies
(MQTT runtime, HA manager, HTTP endpoint store, webhook state) live.

Tests: 4 new tests cover the rule-type/handler bijection, callable
shape, missing-entry rejection, and unknown-class rejection. 44
existing automation engine tests stay green; ruff clean.
2026-05-22 23:07:07 +03:00
alexei.dolgolyov 5fec8db901 refactor(capture): lift duplicated edge-to-LED kernels into shared module
PixelMapper and AdvancedPixelMapper in calibration.py used to carry
byte-for-byte copies of two ~80-line numpy kernels (audit finding M4):

  * the vectorised average-colour-per-LED path with its cumsum + take
    scratch-buffer dance; and
  * the per-LED fallback loop for median / dominant colour modes.

Lift both into a new ``core.capture.edge_interpolation`` module exposing
``average_edge_to_leds(edge_pixels, edge_name, led_count, cache,
cache_key)`` and ``fallback_edge_to_leds(edge_pixels, edge_name,
led_count, calc_color)``. The cache parameter is the caller-owned dict
(``self._edge_cache``) so allocations still happen once per
(edge_len, led_count) signature — the difference is that the
boundary-builder, the buffer set, and the inner numpy ops live in
exactly one place.

PixelMapper keys its cache by edge name (``"top"`` / ``"left"`` etc.);
AdvancedPixelMapper keys by line-index int (same dict, no collision).
Both mappers' ``_map_edge_average`` / ``_map_edge_fallback`` shrink to
single delegating lines.

Tests: 9 new kernel-level tests cover uint8 dtype + shape, the cache
reuse / rebuild contract, independent cache keying, a gradient input
producing a monotonic output, the calc_color callable contract for the
fallback path, and segment-position tracking for both axes. 30
existing calibration tests stay green; ruff clean.
2026-05-22 23:03:44 +03:00
alexei.dolgolyov 97dae2cd62 refactor(processing): replace inline effect dispatch with @_effect_renderer registry
EffectColorStripStream._animate_loop used to rebuild a 12-entry dict
``renderers = {"fire": self._render_fire, ...}`` on every frame, then
look up ``renderers.get(self._effect_type, self._render_fire)``. Two
audit smells (H1) at once: per-frame dict-rebuild churn and a silent
fallback to fire whenever ``self._effect_type`` was a typo or any
``_render_*`` method got renamed without updating the dict.

Fix:

  * ``@_effect_renderer("fire")`` stamps an attribute on the unbound
    method.
  * ``@_collect_effect_renderers`` (applied to the class) walks
    members at class-creation, gathers the marked ones into
    ``cls._RENDERERS``, and raises ``RuntimeError`` on duplicate
    registration.

The loop now reads ``type(self)._RENDERERS`` once and calls the
unbound method with explicit ``self``. An unknown ``_effect_type``
logs a warning and skips the frame (sleep one frame_time) instead of
silently rendering fire — louder failure mode without crashing the
animation thread.

Tests: 5 new tests cover the 12-effect coverage set, callable shape,
class-level (not per-instance) dict identity, duplicate-name
rejection, and the marker stamp contract.

343 existing processing / storage / API tests stay green; ruff clean.
2026-05-22 23:00:00 +03:00
alexei.dolgolyov 29bdacf69a refactor(processing): dedupe HA/Z2M _swap_color_source via shared helper
HALightTargetProcessor and Z2MLightTargetProcessor used to carry
character-for-character identical _swap_color_source method bodies
(audit finding C5) — only the log prefix differed. Extract the body
into a free function ``swap_color_source(processor, new_kind,
new_color_vs_id, *, log_label)`` in a new ``light_target_helpers``
module. Each processor's _swap_color_source now delegates to the helper
and then clears its per-entity history (``_previous_colors`` /
``_previous_on``) — that bit stays on the processor because it's per-
target state, not colour-source state.

Scope deliberately narrower than the full BaseLightTargetProcessor ABC
the audit gestured at: the 76 read sites for the per-processor colour
state across the two files made a full state-composition refactor too
risky for the live LED control loop. The free-function helper is the
minimum-blast-radius way to delete the duplication while leaving WLED
(which has no value-stream-vs-CSS dispatch) untouched.

The helper standardises both warning messages on HA's original wording
("failed to acquire color VS stream" / "failed to re-acquire CSS
stream") so existing log alerts/grep patterns keep working.

A LightTargetSwapState Protocol under TYPE_CHECKING documents the
expected processor surface; no runtime enforcement (acceptable trade-
off vs a 76-site touchpoint).

Tests: 7 new tests cover the release+acquire ordering, the not-running
no-op path, the manager-error-swallowing behaviour, the empty-id
short-circuit, and the missing-manager (TargetContext(None, None))
fallback. 354 existing storage + API + e2e + processing tests stay
green; ruff clean.
2026-05-22 22:54:14 +03:00
alexei.dolgolyov 563cbac88c refactor(storage,processing): kind registries + versioned data migrations
Two CRITICAL data-safety bugs from the architecture audit and the two
worst parallel-change problems are fixed in one coherent pass.

Audit findings addressed:

- C2  silent CSS response fallback. The previous _RESPONSE_MAP fell
      through to a fabricated PictureCSSResponse whenever a source
      class lacked an entry; in particular game_event sources were
      silently mis-shaped. Now: GameEventCSSResponse/Create/Update
      schemas exist, _RESPONSE_MAP is re-keyed by source_type string,
      an import-time _assert_response_map_coverage() requires symmetric
      agreement with storage._SOURCE_TYPE_MAP, and the runtime path
      raises instead of fabricating a response.

- C11 string-replace JSON migration. ColorStripStore used
      blob.replace('"source_type": "static"', '"source_type":
      "single_color"') which can corrupt unrelated substrings (e.g.
      an animation type named "static_wave") and provides no audit,
      no transaction, no idempotency. Replaced with
      storage.data_migrations.MigrationRunner backed by a
      data_migrations audit table. Each migration runs inside one
      db.transaction() that covers the applied-check, the apply(),
      and the audit-INSERT — partial failures roll back atomically.
      StaticToSingleColorMigration parses each row with json.loads
      and mutates only the source_type field. Frozen-write databases
      skip with a warning.

- C3+C4 color-strip stream dispatch. The 7-branch elif in
      ColorStripStreamManager.acquire() and the duplicate one in
      ws_stream._create_stream() now share a single STREAM_BUILDERS
      registry in core.processing.color_strip_kinds, keyed by
      source.source_type. Both call sites populate a StreamDeps bag
      and delegate to build_stream(). _assert_stream_kind_coverage()
      asserts at import that STREAM_BUILDERS plus SHARABLE_KINDS
      partitions storage._SOURCE_TYPE_MAP. ws_stream's preview path
      wraps each FastAPI-DI getter in _safe() so non-audio previews
      no longer crash when audio/CSPT stores are not wired.

- C6+C7 value stream dispatch. The 14-branch isinstance ladder in
      ValueStreamManager._create_stream and its silent
      StaticValueStream(value=1.0) fallback are replaced by
      core.processing.value_kinds.STREAM_BUILDERS, keyed by
      source_type string (so AdaptiveValueSource's adaptive_time and
      adaptive_scene route to different builders correctly). The
      manager retains only the SyncClockRuntime pre-acquisition step
      for animated_color (kinds needing this are listed explicitly
      in NEEDS_CLOCK_RUNTIME). Symmetric coverage assertion plus a
      separate assertion that NEEDS_CLOCK_RUNTIME is a subset of the
      registry.

Bundled in: the static->single_color rename plus the HTTPValueStream
/ http_endpoint introduction that were already in flight on this
branch share these files; the registry refactor naturally absorbs
both via the new "single_color" / "static" alias entries and the
_build_http builder.

Tests: 26 new tests cover response-map coverage drift, migration
runner audit-table mechanics + transactional rollback +
frozen-write skip, and the two stream-builder registries. 343
existing storage / API / e2e tests stay green. Ruff clean.
2026-05-22 22:45:28 +03:00
alexei.dolgolyov e24f9d33cc fix(shutdown): survive PC restart with WAL fsync + Win32 session-end guard
Two bugs caused user data ('G502' target's color-strip ref, etc.) to
revert after PC restart while persisting fine across normal app
restarts:

1. SQLite was in WAL mode with synchronous=NORMAL and Database.close()
   was never called. On graceful Python exit the sqlite3 finalizer
   checkpoints the WAL, but on an unclean PC shutdown (power loss,
   forced reboot, or Windows force-terminating pythonw.exe) the WAL
   stayed in OS cache, never reached disk, and the next boot rolled the
   DB back to the last checkpoint -- losing recent edits.

2. Nothing handled WM_QUERYENDSESSION / WM_ENDSESSION, so on PC
   shutdown Windows force-killed pythonw.exe after ~5s and the FastAPI
   lifespan never ran. The 'stop_targets' setting was silently ignored
   and devices were left at their last frame.

Changes:
- Database: PRAGMA synchronous=FULL + wal_autocheckpoint=100, plus an
  explicit wal_checkpoint(TRUNCATE) inside Database.close().
- New utils/win_shutdown.py: hidden top-level window in a daemon thread
  with a ctypes WindowProc that catches WM_QUERYENDSESSION (calls
  ShutdownBlockReasonCreate to extend Windows' 5s hung-app timeout up
  to the ~20s GUI ceiling), fires the shutdown callback, then waits in
  WM_ENDSESSION on a completion event before returning. Also raises
  the process shutdown priority via SetProcessShutdownParameters. All
  Win32 argtypes/restypes are bound once at import to avoid LPARAM
  overflow on x64.
- New shutdown_state.py: leaf module owning the cross-thread Event so
  __main__ does not import the heavy ledgrab.main at startup.
- main.py lifespan: per-step asyncio.wait_for budgets (8s for
  processor_manager.stop_all, 1.5s each for HA/MQTT, etc.) so a hung
  device cannot starve the DB checkpoint, then db.close() and
  shutdown_complete.set() always run.
- __main__.py: install the Windows shutdown guard before tray start;
  install SIGINT/SIGTERM/SIGBREAK handlers only on the tray path
  (uvicorn overwrites them on no-tray); raise server_thread.join to 20s.
- Tests cover WM_QUERYENDSESSION (fires callback, returns TRUE,
  idempotent), WM_ENDSESSION (waits on event, times out cleanly,
  cancel-path returns instantly), signal handler installation, and
  that main and shutdown_state share the same Event instance.
2026-05-22 21:43:41 +03:00
alexei.dolgolyov e4bf58da19 fix(dashboard): stop showing perpetual MODIFIED for un-edited legacy layouts
The MODIFIED hint in the Customize Dashboard panel was driven by
`presetActive`, recomputed on every save/load via strict deep-equal
against each preset. Any drift between a saved layout and the current
defaults — older app versions that hadn't yet had some new perf cells
added, prior buggy merges that appended new registry keys to the end
of perfCells, or stale `visible` values from intermediate dev builds —
left `presetActive` undefined forever and pinned the panel in MODIFIED
state for users who had not actually edited anything.

Split the two concerns:

- `presetActive` keeps driving the chip highlight (recomputed). When
  the layout happens to match a preset exactly the chip lights up.
- New `userModified` boolean drives the MODIFIED indicator. Set to true
  only on actual edits through the panel (visibility / density /
  ordering / select changes) and on JSON import; cleared by applying a
  preset and by Reset.

Legacy saves without the field load as `userModified: false` so the
indicator no longer fires retroactively on data the user never
touched. Also tighten `_mergeWithDefaults` so newly-added registry
keys land at their canonical positions (subsequence detection) when
the saved order is consistent with defaults, which keeps the chip
highlight stable across upgrades.
2026-05-16 17:05:12 +03:00
alexei.dolgolyov f1b0f0eab2 fix(ui): repaint transport-bar uptime as soon as /health responds
The inline transport-uptime ticker only repainted on its 1 s setInterval,
so the field could sit on - for up to ~1 s after page load (and much
longer if init's first /health response landed between ticks - the next
seed then had to wait for the 10 s connection-monitor poll). Dispatch a
serverUptimeChanged DOM event when window.__serverUptime is seeded and
let the inline IIFE re-render on receipt, so the value appears as soon
as the response arrives.
2026-05-16 12:28:57 +03:00
alexei.dolgolyov 17684afba1 docs: record review-fix pass in TODO.md
Marks the pre-merge code review as complete in TODO.md and lists each
finding alongside the commit that closed it. Branch is in shape to merge.
2026-05-16 11:06:51 +03:00
alexei.dolgolyov 0e3ae78de7 fix(devices): address pre-merge review findings
Closes the issues surfaced by the pre-merge code review of the
expand-device-support branch.

CRITICAL #2 -- update_device double-encrypts secrets in memory.
storage/device_store.py round-tripped through device.to_dict() which
encrypts hue_username / hue_client_key / ble_govee_key / nanoleaf_token
via _enc(), but Device.__init__ does not decrypt. The cached
self._items[device_id] thus held ciphertext where plaintext belonged,
breaking runtime auth for paired devices on any update -- even an
innocuous rename. Sourcing kwargs from vars(device) directly avoids
the round-trip. Regression tests cover Nanoleaf and Hue.

HIGH #3 -- secrets leaked in GET /api/v1/devices response.
DeviceResponse previously returned nanoleaf_token / hue_username /
hue_client_key in plaintext (decrypted server-side from storage),
defeating the encryption-at-rest. Replaced with nanoleaf_paired and
hue_paired booleans. ble_govee_key intentionally stays -- it's a
user-managed value pasted from a third-party tool, must remain visible
for edit. Frontend types.ts + the one nanoleaf_token reader updated to
the boolean.

HIGH #4 -- SSRF surface. validate_lan_host() added to net_classify.py;
called from each new driver's validate_device (DDP / Yeelight / WiZ /
LIFX / Govee / OPC / Nanoleaf) and from pair_device. Rejects literal
public IPs with a descriptive ValueError; non-IP hostnames pass
through (mDNS labels, bare hostnames). RFC6890 ranges (documentation,
former class E) are accepted as LAN-like since Python's
ipaddress.is_private treats them so -- correct policy for LedGrab.

HIGH #5 -- decrypt failure deletes the device row. _dec() now catches
the exception, logs an error, and returns "" instead of propagating.
Without the fix, a regenerated data/.secret_key would silently make
every Hue / Nanoleaf / BLE-Govee device disappear from the device list
on next startup. Regression test asserts a corrupt envelope leaves the
device hydratable.

HIGH #6 -- update_device route does not rstrip("/") for non-WLED.
Moved the trim before the WLED-specific scheme inference so every
device type gets consistent URL normalization between create and
update.

MEDIUM #7 -- Govee discovery port 4002 collision. Added a lazily-
initialized module-level asyncio.Lock that serializes concurrent
discover_govee_devices() calls; the previous behavior had the second
parallel scan silently return [] when the first still held port 4002.
Error message also clarified to mention another Govee tool.

MEDIUM #8 -- Nanoleaf discover() leaked browser tasks on cancellation.
Moved the browser cancel loop into the finally block so an interrupted
mDNS scan still tears them down.

MEDIUM #9 -- pair endpoint logged user-supplied URL with exc_info=True.
Added _sanitize_url_for_log() that strips userinfo + fragment, and
demoted the log from exc_info to type(exc).__name__ + str(exc) so a
hostile receiver's response body can't end up in the log file.

LOW -- Nanoleaf was the only client without a .port property. Added
one (returns NANOLEAF_PORT, fixed) for cross-driver symmetry.

LOW -- no end-to-end pair-then-create coverage. Added
TestPairThenCreateFlow.test_pair_then_create_persists_encrypted_token
which exercises the full path: POST /api/v1/devices/pair returned
fields, store.create_device, then asserts (a) in-memory plaintext,
(b) to_config() plaintext, (c) persisted ciphertext, (d) API response
strip + paired-boolean.

Tests: 1379 pass (was 1358 -- 21 new regression tests added).
ruff clean. TypeScript clean.
2026-05-16 11:06:10 +03:00
alexei.dolgolyov 7736bc6f58 fix(utils): commit url_scheme + net_classify dependencies
The DDP commit (8f1140a) added imports of infer_http_scheme into
api/routes/devices.py but missed bringing in the module itself --
url_scheme.py and its net_classify.py dependency were in the working
tree as untracked files only. On a clean checkout the FastAPI app
fails to start with ModuleNotFoundError.

Caught by the pre-merge code review. The 1358 passing tests only
worked because the local working tree happens to have the files.

This commit adds:
- ledgrab.utils.url_scheme: infer_http_scheme() for LAN-vs-public WLED
  URL scheme inference
- ledgrab.utils.net_classify: HostCategory enum + classify_ip() +
  is_blocked_for_ssrf() + is_local_for_http_default() + is_loopback().
  Single source of truth for IP categorisation used by safe_source
  (SSRF), url_scheme (LAN), and auth (loopback exemption).
- 107 unit tests (test_url_scheme.py + test_net_classify.py).

net_classify.is_blocked_for_ssrf is the primitive the device-driver
validate_device methods will use in the next commit to close HIGH #4
from the review.
2026-05-16 10:46:45 +03:00
alexei.dolgolyov 390d2b472c docs: mark expand-device-support branch ready for merge
Updates TODO.md to reflect the verification pass outcome:
- pixel_reduce extraction marked done (commit cc87fba)
- pre-merge verification pass marked complete: 1358 pytest tests pass,
  ruff and black clean (against pre-commit-pinned 24.10.0),
  npx tsc --noEmit clean, bundle compiles
- Mi-Light / MiBoxer explicitly marked WONTDO with rationale
  (esp8266_milight_hub firmware -> existing MQTT target is the better
  path for modern Mi-Light deployments)
- duplicate Nanoleaf bullet removed; Twinkly noted as deferred

No code changes -- this is purely a checkpoint for the merge review.
2026-05-16 04:18:33 +03:00
alexei.dolgolyov cc87fba0dd refactor(devices): extract _average_color to pixel_reduce
Six single-pixel LED clients (Yeelight, WiZ, LIFX, Govee, Nanoleaf, BLE)
had byte-for-byte identical local copies of the strip-averaging helper.
Consolidates into core/devices/pixel_reduce.average_color so the next
single-pixel driver can drop the local copy and so behavior changes
land in one place.

Hue is intentionally left out -- its Entertainment API addresses up to
seven lights individually rather than averaging.

Behavior is byte-identical (each call site re-imports under the same
underscore-prefixed local name). 1358 tests still pass.
2026-05-16 04:14:36 +03:00
alexei.dolgolyov 426484adf8 feat(devices): Nanoleaf OpenAPI target type + first pair-flow user
Adds support for Nanoleaf controllers (Light Panels / Canvas / Shapes /
Lines / Elements) via the documented HTTP REST API on port 16021.
First concrete consumer of the pair-UX scaffold from commit 2f31680 --
the abstraction is no longer speculative.

Backend:
- NanoleafClient is a single-pixel HTTP adapter: averages the strip to
  one RGB triple, converts to Nanoleaf's HSB scale (H 0-360 / S 0-100 /
  B 0-100), and PUTs to /api/v1/<token>/state with duration:0 so
  transitions are instant for ambilight. Brightness is clamped to >=1
  because Nanoleaf rejects brightness=0.
- pair_nanoleaf(host) implements the two-step handshake: POST
  /api/v1/new during the 30-second pairing window the controller opens
  after the user holds the power button for 5 s.
    200 -> {auth_token: "..."}
    403 -> raises PairingNotReady ("Hold the power button...")
    other / transport error -> RuntimeError wrapping the cause
- NanoleafDeviceProvider.pair_device returns {nanoleaf_token: ...}
  forwarded by POST /api/v1/devices/pair to the frontend for inclusion
  in the subsequent create payload.
- mDNS discovery via _nanoleafapi._tcp (and the v1 variant); failures
  yield [] rather than raising.
- Health check probes /api/v1 without a token (401/403 still proves
  the host is alive).
- NanoleafConfig has nanoleaf_token + nanoleaf_min_interval_ms
  (default 100 ms = ~10 Hz; HTTP overhead caps practical max ~20 Hz).
- Auth token encrypted at rest via _enc/_dec, matching Hue / BLE-Govee.
- 42 unit tests cover URL parsing, RGB->HSB conversion, pairing
  handshake (200 / 403 / 500 / missing-token / transport-error),
  state mutations, brightness clamp, set_power / set_brightness /
  set_color, connection lifecycle, provider validate / pair /
  discover / capabilities, and Device.to_config round-trip including
  the encrypted-token roundtrip via to_dict + from_dict.

Frontend:
- 'nanoleaf' in DEVICE_TYPE_KEYS (next to 'govee'), HEXAGON icon
  (deliberate departure from the smart-bulb lightbulb family --
  Nanoleaf is panels, not bulbs, and the brand identity is hexagonal).
- isNanoleafDevice predicate + per-type field show/hide.
- Pair flow integration: when the device type is Nanoleaf, the add-
  device modal retitles its submit button to "Pair Device" and
  intercepts the submit. handleAddDevice awaits
  runPairingFlow({deviceType: 'nanoleaf', url}), merges result.fields
  ({nanoleaf_token}) into the create body, then POSTs. On
  PairingCancelled the user stays on the modal silently.
- Settings modal exposes the rate-limit field and a read-only
  "Paired" indicator reusing the pair-modal success badge. The token
  itself is never rendered to the DOM and never sent on update --
  re-pairing requires delete + re-add.
- Per-type pairing instructions in en/ru/zh
  (device.nanoleaf.pair.instructions) that the scaffold's i18n lookup
  resolves automatically.
- Bundle: +6.4 KiB (pairing-flow.ts was tree-shaken before this
  commit; now both it and the Nanoleaf branches are baked in).

The pair-UX scaffold is now proven, not speculative. Tuya and Twinkly
can follow the same shape when their phases arrive.
2026-05-16 03:59:38 +03:00
alexei.dolgolyov 2f31680823 feat(devices): pairing-UX scaffold (Phase 2)
Lays the groundwork for device families that require a one-time
physical pairing action (Nanoleaf hold-power-button, Tuya local-key
extraction, Twinkly network-setup mode, Hue link-button). No driver
uses it yet -- Nanoleaf will be the first concrete consumer.

Phase 2 as originally written had three bullets; only this one was
genuinely missing work. The other two (generic NetworkDiscoveryService
fan-out, unified scan-network UI) were already solved at the route
level by the existing /api/v1/devices/discover handler running all
providers in parallel via asyncio.gather(return_exceptions=True).
Marked WONTDO in TODO.md with rationale.

Backend:
- LEDDeviceProvider gains an async pair_device(url) -> dict method.
  Default raises NotImplementedError so missing implementations on a
  requires_pairing provider fail loud at request time.
- New PairingNotReady exception, distinct from generic errors so the
  route handler can return 409 (user must perform the physical action,
  retry possible) instead of 500.
- POST /api/v1/devices/pair endpoint with PairDeviceRequest /
  PairDeviceResponse schemas. Status-code mapping:
    200 -> paired, fields returned for the subsequent create payload
    400 -> unknown device type, or type doesn't support pairing
    409 -> PairingNotReady (retryable from the UI)
    422 -> invalid URL / device configuration (ValueError)
    502 -> transport / network failure (other exceptions)
    500 -> provider returned a non-dict (defensive)
- 8 route tests register a stub provider and exercise every
  status-code path.

Frontend:
- New modals/pair-device.html with five state blocks (idle / pairing
  / not_ready / success / failed) toggled via data-pair-state, plus
  a 30-second SVG progress ring with monospace countdown.
- New features/pairing-flow.ts exposing
  runPairingFlow({deviceType, url, instructionsKey?}) ->
  Promise<{fields: Record<string, unknown>>. Wires the modal to the
  pair endpoint, maps response codes to UI states, AbortControllers
  in-flight fetches on cancel. Exports a PairingCancelled sentinel
  error class.
- Generic pairing.* i18n keys in en/ru/zh. Drivers will add their own
  device.<type>.pair.instructions key that overrides the default.

Design decisions (per frontend-design skill):
- Single SVG ring + centered countdown (HomeKit-style)
- Instructions stay visible during pairing, dimmed to 60% via :has()
- Success state held 450 ms before auto-dismiss
- Cancel-X in the footer; primary action lives in the state block
- prefers-reduced-motion disables pulse/fade/ring transitions

Note: the components.css diff includes a pre-existing MiniSelect block
from the user's parallel work; pairing-specific styles are the second
hunk (lines ~1628+).
2026-05-16 03:26:53 +03:00
alexei.dolgolyov 31c6c3abb2 feat(devices): Open Pixel Control (OPC) target type
Adds support for Open Pixel Control receivers (Fadecandy boards,
xLights/Falcon endpoints, OPC bridges, art-installation controllers,
hobbyist LED driver software). OPC is a tiny TCP protocol on port
7890 with a 4-byte header [channel][cmd][len_hi][len_lo] + RGB body.

Backend:
- OPCClient opens one persistent TCP connection and streams frames as
  header+body byte pairs. Channel 0 broadcasts to every output on the
  OPC server; channels 1-255 address a specific channel on multi-output
  servers (Fadecandy with multiple Open Pixel chains).
- supports_fast_send=True with a synchronous send_pixels_fast hot path.
  The fast path skips the async drain so the OS write-buffer flushes
  on its own schedule -- exactly what ambilight streaming wants.
- Brightness applies client-side before the frame is sent (OPC has no
  reply channel for hardware-side brightness).
- Health check opens a TCP connection and closes it.
- OPCConfig joins the typed config union; storage gains an opc_channel
  field; full to_dict/from_dict/to_config wiring.
- 36 unit tests cover URL parsing, header construction, send_pixels
  emitting header+body in order, brightness application, list and
  flat-array input shapes, drain behavior, connection lifecycle,
  provider validate/discover/capabilities, Device.to_config round-trip.

Frontend:
- 'opc' in DEVICE_TYPE_KEYS (next to 'ddp'), paper-plane icon -- same
  as DDP since both are open pixel-streaming protocols.
- isOpcDevice predicate + per-type field show/hide.
- Optional channel number input (default 0 = broadcast) with hint copy
  explaining the channel semantics.
- Locale strings in en/ru/zh.

No native discovery (OPC has no discovery protocol); users supply
the receiver IP manually.
2026-05-16 03:02:41 +03:00
alexei.dolgolyov 887131d4af feat(devices): Govee LAN target type
Adds support for Govee Wi-Fi smart bulbs and ambient-lighting kits via
their LAN API (opened in 2023). Discovery is multicast UDP on
239.255.255.250:4001; control commands go unicast to the device's port
4003; responses arrive on port 4002.

Each device requires "LAN Control" toggled ON in the Govee Home app
(Device -> settings -> LAN Control). Devices with LAN Control disabled
silently fail to appear in discovery and won't respond to commands; the
UI hint copy reminds users.

Backend:
- GoveeClient is a single-pixel UDP adapter: averages the strip to one
  RGB triple and pushes a 'colorwc' command with colorTemInKelvin=0 to
  select pure RGB mode (non-zero kelvin would switch the bulb to CCT
  mode and ignore the RGB values).
- Brightness folds into the RGB scaling so we burn one packet per
  frame instead of two.
- supports_fast_send=True with a synchronous send_pixels_fast hot path.
  Default rate gate 50 ms (~20 Hz); UDP fire-and-forget tolerates it.
- Multicast discovery: scan request to 239.255.255.250:4001, listen on
  port 4002, parse the inner data dict for IP + device-id + SKU +
  firmware version. Degrades to [] when port 4002 is already bound or
  network is unavailable.
- Health check sends devStatus and waits 1.5s for any reply; the error
  message points at the LAN-Control toggle since that's the #1 root
  cause of silent failures.
- GoveeConfig joins the typed config union; storage gains
  govee_min_interval_ms; full to_dict/from_dict/to_config wiring.
- 40 unit tests cover URL parsing, scan-reply parsing (rejecting
  non-scan commands and malformed JSON), payload builders (colorwc
  with colorTemInKelvin=0, brightness clamping, power as 1/0 not
  true/false), strip averaging, rate limiting, fast-send hot path,
  provider validate/discover/health, Device.to_config round-trip.

Frontend:
- 'govee' in DEVICE_TYPE_KEYS (next to 'lifx'), lightbulb icon
  (deliberate smart-bulb family grouping).
- isGoveeDevice predicate + per-type field show/hide.
- Rate-limit number input (default 50 ms).
- URL hint copy explicitly instructs users to enable LAN Control in
  the Govee Home app -- the #1 source of "why isn't my Govee
  responding?" support churn.
- Locale strings in en/ru/zh.
2026-05-16 02:47:15 +03:00
alexei.dolgolyov 8f9d490063 feat(devices): LIFX LAN target type
Adds support for LIFX smart bulbs and lightstrips that speak the LIFX
binary UDP protocol on port 56700, with broadcast LAN discovery via the
standard GetService/StateService probe.

Backend:
- LIFXClient is a single-pixel UDP adapter: averages the strip to one
  RGB triple, converts to LIFX HSBK (16-bit hue/saturation/brightness +
  kelvin), and pushes a tagged SetColor packet so all bulbs on the
  subnet act on it. Brightness folds into the HSBK brightness channel.
- Hand-rolled packet builder: 36-byte LIFX header (frame +
  frame-address + protocol-header) + variable-length payload. Source
  ID 'LGGR' identifies LedGrab in protocol logs.
- supports_fast_send=True with a synchronous send_pixels_fast hot path
  -- UDP costs nothing, so the default rate gate is 50 ms (~20 Hz) to
  match LIFX's documented <=20 cmd/sec recommendation.
- Broadcast discovery sends GetService and parses StateService replies
  back into IP + MAC + service-port triples. Broadcast failures yield
  [] rather than raising.
- Health check sends GetService and waits 1.5s for any reply on a
  one-shot UDP socket.
- LIFXConfig joins the typed config union; Device storage gains a
  lifx_min_interval_ms field; full to_dict/from_dict/to_config wiring.
- 47 unit tests cover URL parsing, RGB->HSBK conversion (red/green/
  blue/white/black/clamping), packet construction (size, msg type,
  tagged flag, target MAC, sequence byte), SetColor and SetPower
  payload layouts, StateService reply parsing (including rejection
  of wrong msg types and runt payloads), strip averaging, rate
  limiting, fast-send hot path, provider validate/discover/health,
  and Device.to_config round-trip.

Frontend:
- 'lifx' in DEVICE_TYPE_KEYS (next to 'wiz'), lightbulb icon
  (deliberate smart-bulb family grouping with Hue + Yeelight + WiZ).
- isLifxDevice predicate + per-type field show/hide in create and
  settings modals.
- Rate-limit number input (default 50 ms) in both modals with hint
  text referencing LIFX's documented <=20 cmd/sec ceiling.
- Locale strings in en/ru/zh.

LIFX bulbs are reachable from the existing "Scan network" button -- no
new discovery UI affordance was needed. No brightness_control capability
exposed; LIFX brightness is folded into the HSBK on the wire.
2026-05-16 02:30:30 +03:00
alexei.dolgolyov ede627b4ac feat(devices): WiZ Connected LAN target type
Adds support for WiZ Connected (Philips' budget-tier) smart bulbs that
accept JSON commands as UDP datagrams on port 38899 with broadcast LAN
discovery on 255.255.255.255:38899.

Backend:
- WiZClient is a single-pixel UDP adapter: averages the incoming strip
  to one RGB triple and pushes it via setPilot with r/g/b params.
  Brightness folds into the RGB scaling so we burn one packet per frame
  instead of two.
- UDP fire-and-forget tolerates high update rates with no ack overhead,
  so the default rate gate is 50 ms (~20 Hz) -- 10x faster than Yeelight.
- supports_fast_send=True with a synchronous send_pixels_fast hot path.
- Broadcast discovery sends the standard registration envelope; bulb
  replies are parsed for IP+MAC and surfaced as DiscoveredDevice
  entries. Broadcast failures (no network, firewall) yield [] rather
  than raising.
- Health check sends getPilot and waits 1.5s for any reply on a
  one-shot UDP socket.
- WiZConfig joins the typed config union; Device storage gains a
  wiz_min_interval_ms field; full to_dict/from_dict/to_config wiring.
- 36 unit tests cover URL parsing, MAC extraction, strip averaging,
  rate limiting, fast-send hot path, provider validate/discover/health,
  and Device.to_config round-trip.

Frontend:
- 'wiz' in DEVICE_TYPE_KEYS (next to 'yeelight'), lightbulb icon
  (deliberate smart-bulb family grouping with Hue + Yeelight).
- isWizDevice predicate + per-type field show/hide in create and
  settings modals.
- Rate-limit number input (default 50 ms) in both modals with hint
  text noting the UDP fire-and-forget characteristic.
- Locale strings in en/ru/zh.

WiZ bulbs are reachable from the existing "Scan network" button -- no
new discovery UI affordance was needed.
2026-05-16 02:12:01 +03:00
alexei.dolgolyov 4b65005823 feat(devices): Yeelight LAN target type
Adds support for Xiaomi/Yeelight smart bulbs and lightstrips that speak
the bulb-vendor's JSON-RPC protocol over TCP port 55443 with SSDP-style
LAN discovery on 239.255.255.250:1982.

Backend:
- YeelightClient is a single-pixel adapter: it averages the incoming
  strip down to one RGB triple, packs it into the 24-bit color int the
  bulb expects, and pushes it via set_rgb with sudden+0ms effect.
- Brightness folds into the RGB scaling on the wire so we burn one
  command per frame instead of two.
- A configurable client-side rate gate (yeelight_min_interval_ms, default
  500) keeps us under the bulb's ~1 cmd/sec cap. Frames that arrive
  inside the gate no-op without TX. Music mode (~60 Hz via reverse-TCP)
  is deferred -- the MVP caps at ~2 Hz and that's fine for a strip-to-
  single-pixel averaging device.
- SSDP discovery scans 239.255.255.250:1982 with the bulb-specific
  ST: wifi_bulb header; replies are parsed into DiscoveredDevice
  entries. Multicast failures (no network, firewall) yield [] rather
  than raising -- discovery is best-effort.
- Health check opens a TCP socket to the bulb and closes it.
- YeelightConfig joins the typed config union; Device storage gains a
  yeelight_min_interval_ms field; full to_dict/from_dict/to_config wiring.
- 34 unit tests cover URL parsing, RGB packing, strip averaging, rate
  limiting, SSDP response parsing, provider validate/discover/health,
  and Device.to_config round-trip.

Frontend:
- 'yeelight' in DEVICE_TYPE_KEYS (next to 'hue'), lightbulb icon
  (intentional family-grouping signal with Hue).
- isYeelightDevice predicate + per-type field show/hide in create and
  settings modals.
- Rate-limit number input (default 500 ms) in both modals with hint
  text explaining the trade-off.
- Locale strings in en/ru/zh.
- Drive-by: types.ts DeviceType union backfilled with 'ddp' and 'ble'
  for type-safety consistency.

Yeelight bulbs are now reachable from the existing "Scan network"
button -- no new discovery UI affordance was needed.
2026-05-16 01:44:13 +03:00
alexei.dolgolyov 8f1140abad feat(devices): standalone DDP target type
Promotes the existing DDP packet layer (previously WLED-internal) to a
first-class device type so any DDP-speaking receiver (Pixelblaze,
ESPixelStick, xLights/Falcon endpoints, generic firmware) can be driven
directly without WLED in the path.

Backend:
- New DDPLEDClient wraps the DDPClient transport as a proper LEDClient
  with supports_fast_send=True (synchronous UDP push on the hot loop).
- New DDPDeviceProvider — no native discovery, manual LED count,
  capabilities = {manual_led_count, health_check}.
- DDPConfig joins the typed config union; Device storage gains
  ddp_port / ddp_destination_id / ddp_color_order fields with safe
  defaults (0/1/1 -> port 4048, destination 1=display, RGB byte order).
- URL scheme: ddp://host[:port] or bare host[:port] (default 4048).
- Health check resolves the host via async DNS; UDP has no reply
  channel so reachability is best-effort by design.
- 29 new tests in test_ddp_led_client.py cover URL parsing, packet
  hot path (brightness, list/numpy input shapes, fast vs async send),
  provider validate/discover/capabilities, config round-trip via
  Device.to_config() and to_dict/from_dict.

Frontend:
- 'ddp' in DEVICE_TYPE_KEYS (next to 'dmx'), paper-plane icon.
- isDdpDevice predicate + per-type field show/hide in the create &
  settings modals.
- Color-order picker uses IconSelect (project rule bans plain select).
- Locale strings added in en/ru/zh.

Note: this commit also carries two pre-existing in-flight hunks that
were intermixed in the same files and could not be split out
non-interactively:
- api/routes/devices.py: URL-scheme inference for bare WLED hosts,
  safer error messages, exception-isolated parallel discovery.
- storage/device_store.py: secret_box helpers + at-rest encryption of
  Hue / BLE-Govee / MQTT credentials.
Both are independent of DDP and intentional per the user.
2026-05-16 01:26:45 +03:00
alexei.dolgolyov 337984c618 feat(color-strips): in-editor live preview for all viable source types
Extend the editor's Preview button to render unsaved form values for every
CSS source type that can be previewed without external calibration. New
types now supported transiently: audio, math_wave, weather, game_event,
api_input, mapped, composite, processed.

Backend (preview WebSocket):
- Dispatch in _create_stream by source_type, injecting the dependencies each
  stream needs (audio managers, weather manager, value stream manager, CSPT
  store via public get_cspt_store, color strip stream manager).
- Roll back clock + stream resources if start() fails so failed previews
  don't leak refs.
- On source_type change mid-preview, drop the rebuilt-stream reference if
  rebuild fails and close the WS rather than poll a stopped stream.

Stream lifecycle fixes flushed out by the new preview paths:
- MappedColorStripStream and ProcessedColorStripStream now stamp a per-
  instance UUID into the sub-stream consumer_id so concurrent consumers
  (multiple preview WS connections) don't collide in the CSM registry.
- ProcessedColorStripStream.update_source now re-acquires the input stream
  when input_source_id changes (previously silently kept the old input).

Frontend:
- Expand _PREVIEW_TYPES; route non-quirky types through a new exported
  getCSSEditorPreviewPayload helper that reuses the existing per-type
  handler registry.
- For picture / picture_advanced / key_colors (which depend on calibration
  or rectangles edited elsewhere), show a clearer "save the source first"
  message instead of the generic "unsupported" toast.
2026-05-16 00:40:26 +03:00
alexei.dolgolyov 530316c2c3 feat(mqtt): multi-broker MQTT + Zigbee2MQTT light target
- New Z2MLightOutputTarget storage, processor, editor and routes for
  Zigbee2MQTT light entities (shares the HA-Light editor UI via the new
  light-target-editor module)
- Replace global MQTTService/MQTTConfig with per-source MQTTManager +
  MQTTRuntime; thread mqtt_source_id through Z2M targets, DIY MQTT
  devices, and the automation engine
- Migrate legacy single-broker YAML/env config to a "Default Broker"
  MQTTSource on startup (core/mqtt/legacy_migration.py) and drop the
  obsolete core/mqtt/mqtt_service.py
- Refresh /api/v1/system integration status to surface every MQTT source
- Extract shared light-target editor and refactor OutputTargetStore +
  output_targets routes around typed factories / auto-registry
- Modal CSS polish, locale strings, and storage/bindable test coverage
2026-05-12 18:06:09 +03:00
alexei.dolgolyov 6e4c1b6642 perf(wled): cache per-frame max-pixel for brightness threshold
Cache the np.max(frame) result keyed on frame identity. The
min_brightness_threshold check ran a full reduction over the LED
array on every loop iteration even when the frame reference was
unchanged — at 60 fps with multiple targets this added up to
hundreds of redundant reductions per second.
2026-05-12 15:06:34 +03:00
alexei.dolgolyov ee4fa81376 perf(processing): event-driven frame hand-off and scheduling fixes
- LiveStream: add frame_id counter + Condition with wait_for_new_frame()
  helper. Producers (ScreenCaptureLiveStream, ProcessedLiveStream,
  StaticImageLiveStream, VideoCaptureLiveStream) now signal_new_frame()
  on each new frame; consumers (PictureColorStripStream, ProcessedLive
  Stream) wait on the event with frame_time as a safety timeout
  instead of polling + sleeping. Cuts glass-to-LED latency at matched
  FPS by up to one frame_time.
- ProcessedLiveStream ring buffer: 3 -> 5 slots. The previous "max 2
  frames in flight" assumption ignored the multi-consumer case where
  several PictureColorStripStream/HA-target threads can hold the same
  _latest_frame reference while we wrap. 5 slots gives ~83 ms of
  consumer-read margin at 60 FPS.
- PictureColorStripStream advanced mode: reuse the already-fetched
  primary frame instead of re-acquiring its lock from _live_streams.
- _blend_u16: use cv2.addWeighted (single SIMD-fused pass) when cv2
  is available; numpy fallback unchanged. Output verified bit-equal
  to the existing 6-pass implementation.
- FrameLimiter.wait: drop the 1 ms minimum-sleep floor. Over-budget
  loops no longer add an extra ms per iteration; the cap on achievable
  rate (~750 fps) is removed.
2026-05-12 15:06:11 +03:00
alexei.dolgolyov f184ef0afb perf(capture): vectorize hot paths and fix engine bugs
- WGC: replace per-frame ~30 MB BGRA->RGB fancy-index allocation with
  cv2.cvtColor into a 3-slot pre-allocated RGB pool. Use gc.collect(0)
  on cleanup instead of full GC to avoid multi-hundred-ms stalls.
- MSS: switch from screenshot.rgb (pure-Python BGRA->RGB rebuild) to
  screenshot.raw + cv2.cvtColor into a pooled buffer. Add cheap 256-byte
  hash-based change detection so idle frames return None — matches
  DXcam/BetterCam semantics.
- DXcam/BetterCam: fix silent factory leak — Python name-mangling
  rewrote self._dxcam.__factory to _DXcamCaptureStream__factory inside
  the class body, so cleanup never reached the real attribute. Use
  getattr with string literal to bypass mangling.
- calculate_dominant_color: replace np.random.choice(replace=False)
  (full sort) with np.random.randint, and np.unique(axis=0) (lexsort)
  with packed-RGB np.bincount. ~10x faster on dominant mode.
- calibration._map_edge_average: switch cached scratch buffers from
  float64 to float32. Halves memory bandwidth on the dominant reduction
  path; range-safe up to 8K screens.
- All engines: per-frame DEBUG logs use structlog kwarg style instead
  of f-strings to avoid per-frame string allocation.
2026-05-12 15:05:52 +03:00
alexei.dolgolyov ad84b60ae4 fix(ha-light): apply brightness_scale once and respect boost multipliers
`_send_entity_color` was multiplying the per-mapping `brightness_scale`
into the brightness payload twice when the effective scale was below 1,
yielding a quartered output for a configured half-scale. Conversely,
when the value-stream multiplier exceeded 1.0 with a default scale,
the entire scaling step was skipped and the boost was lost.

Compute brightness as `clamp(max(r,g,b) * bs * vs, 0, 255)` once and
ship it directly, with regression tests pinning the half-scale, boost,
and 255-clamp cases.
2026-05-11 01:42:02 +03:00
alexei.dolgolyov cdf7d94652 feat(ui): expand card icon picker (44 -> 120 icons, +5 categories)
Add 76 new icons to the custom card-icon picker and introduce five new
categories: weather, nature, controls, status, office. Existing icon ids
are unchanged so persisted card icons keep resolving.

- icon-paths.ts: +36 Lucide path constants (weather, nature, room,
  office, media, hardware, lighting variants)
- device-icons.ts: extend IconCategory union and CATEGORIES; add
  registry entries with labels + search aliases
- en/ru/zh locales: 5 new category labels + 76 per-icon labels each
  (126 device.icon keys per locale, fully aligned)

Tabs scroll horizontally via existing overflow-x; no migration needed
(picker reads/writes ids by value, missing ids fall back to inheritance).
2026-05-11 01:38:40 +03:00
alexei.dolgolyov 09792a9a05 chore: release v0.6.1
Build Release / create-release (push) Successful in 4s
Build Android APK / build-android (push) Failing after 9s
Build Release / build-linux (push) Successful in 2m13s
Build Release / build-docker (push) Successful in 3m9s
Build Release / build-windows (push) Successful in 4m6s
2026-05-10 23:57:47 +03:00
alexei.dolgolyov 75ca487be1 feat(ui): per-surface card presentation modes (C/M/D/R)
Adds a comfortable/compact/dense/row toggle to every card grid in the
app. Each surface (LED devices, targets, automations, scenes, sources,
streams, dashboard subsections, etc.) remembers its mode independently.

Persistence mirrors dashboard-layout: localStorage cache for first paint,
debounced PUT to /api/v1/preferences/card-modes (new endpoint) for
cross-browser sync. Surface registry is open — any non-empty key
accepted server-side; modes validated against {comfortable, compact,
dense, row}.

CSS is token-driven: grid min-width and gap come from --card-grid-min /
--card-grid-gap / --card-grid-min-narrow / --card-grid-gap-narrow /
--templates-grid-min / --templates-grid-gap defined on :root, overridden
per [data-card-mode]. Dense/row also hide .mod-leds, collapse secondary
button labels, and tighten .mod-metrics; row collapses the grid to one
full-width column. Coexists with the existing per-section [data-density]
on the dashboard tab — different attribute, additive concern.

Toggle UI auto-mounts into every CardSection header (18+ surfaces) plus
the six dashboard subsections via post-render mount; teardown tracking
keeps the listener Set bounded across re-renders.

i18n: card_mode.{tooltip,comfortable,compact,dense,row} in en/ru/zh.
Tests: 9 new cases in tests/test_preferences_card_modes_api.py covering
defaults, round-trip, validation, open-registry keys, row mode, delete.
2026-05-10 23:49:14 +03:00
alexei.dolgolyov e65dcb41f4 chore: clean up cfg abbreviation and stale TODO link
Rename `cfg` parameter/local in resolve_mqtt_password to `config`
for PEP 8 compliance. Drop the broken reference to the long-removed
docs/plans/device-typed-configs.md from TODO.md.
2026-05-10 23:19:15 +03:00
alexei.dolgolyov 6a07a6b1a2 fix(shutdown): apply target stop actions before tearing down HA/MQTT
Reorder the lifespan shutdown so processor_manager.stop_all() runs before
ha_manager.shutdown(), mqtt_manager.shutdown(), and mqtt_service.stop().
HA-light targets check `_ha_runtime.is_connected` before applying their
`stop_action` (turn_off / restore) and silently skip when HA is already
disconnected; MQTT-output devices need the broker connection alive to
send restore frames. The previous order tore those down first, turning
"stop_targets" into a no-op for those targets — most visible when
closing via the tray Shutdown button.

Also moves automation_engine.stop(), discovery_watcher.stop(), and the
OS notification listener stop ahead of processor stop so they can no
longer fire events into a shutting-down processor manager. Independent
services (weather, update checker, auto-backup) now run last, where
their order does not matter.

Bonus: if the daemon-thread join times out (10 s) and the rest of
shutdown is cut short, the user-visible part — targets stopping — has
already run.
2026-05-10 22:39:18 +03:00
alexei.dolgolyov 0f5850ef80 feat(ui): customisable card icon for all entity types
Extends the icon-plate work from devices and output targets to every
remaining card type — 18 new entities, 20 in total. Users can now pick
a curated icon (with optional colour override) for any card on any tab,
and the picker reuses the same modal, recent-strip, search, and
category tabs introduced for the device picker.

Foundation:
- icon-picker.ts — replace the hardcoded 2-entry adapter record with a
  Map<EntityType, EntityTypeAdapter> and expose
  registerIconEntityType() + makeSimpleIconAdapter() so each feature
  module owns its own adapter (~6 lines per type).
- bodyExtras hook on adapters, keyed off id, lets discriminated routes
  (output-targets target_type, picture-sources stream_type, audio /
  value / color-strip-sources source_type) accept icon-only PUTs.
- core/card-icon.ts — new makeCardIconFields(type, id, entity) helper
  spreads iconHtml / iconColor / iconAttrs into a mod-card head in one
  line.
- _onDocumentClick now accepts any registered type instead of a
  hardcoded device/target check.

Backend (purely additive — no migrations needed thanks to JSON-blob
storage):
- 18 dataclasses gained icon: str = "" + icon_color: str = "" with
  emit-when-truthy serialisation and "" defaults on load.
- All matching Create / Update / Response Pydantic schemas gained the
  fields with the standard Optional[str] + max_length=64/32 +
  description set.
- All routes' response builders use
  getattr(entity, "icon", "") or "" so existing rows render unchanged.
- ValueSource and CSS handle icon/icon_color on the base class so all
  source-type subclasses inherit them automatically.

Frontend wiring (12 modules):
- streams.ts — picture sources, capture templates, PP templates,
  CSPT, audio sources, audio templates, gradients (built-in
  gradients keep no plate).
- automations, scene-presets, sync-clocks, weather-sources,
  value-sources, mqtt-sources, home-assistant-sources,
  game-integration, audio-processing-templates, assets,
  color-strips/cards.
- pattern-templates skipped — uses the legacy wrapCard({content,
  actions}) string API, separate migration.

Dashboard cards now also display the chosen icon:
- Targets already had it (with device inheritance for LED targets).
- Sync clocks, automations, and scene presets gained the same plate
  via a shared _dashboardIconPlate helper that mirrors the mod-card
  layout (mod-head--with-icon class flips on when present).

i18n: 20 new device.icon.entity.<type> labels in en/ru/zh.

Verification:
- ruff check src/ tests/ — clean.
- npx tsc --noEmit — clean.
- npm run build — 2.6 MB bundle.
- pytest tests/ --no-cov — 949 passed (no regressions).

Pending: manual smoke test on each card type — open picker, save, and
confirm the channel-color preview matches the live card.
2026-05-09 16:19:20 +03:00
alexei.dolgolyov a79f4bf73c feat(ha-light): broadcast a single Color Value Source to all entities
HALightOutputTarget gains a `source_kind` field with two modes:
- `css` (existing): per-mapping LED segments averaged from a ColorStripSource.
- `color_vs` (new): one colour from a colour-returning ValueSource pushed to
  every mapped entity (mapping LED ranges are ignored in this mode).

Backend wiring:
- Schema/route: add `source_kind` + `color_value_source_id` to create/update/
  response payloads, with VS existence + return_type=color validation.
- Storage: persist new fields, with defensive `or ""` coalesce so legacy rows
  written via resolve_ref with None survive the str-typed response schema.
- Processor: ha_light_target_processor reworked to drive both source kinds
  (incl. update_target_settings hot-swap of source mode). New unit tests in
  tests/core/test_ha_light_target_processor.py and extended store tests.

Frontend:
- ha-light editor modal: collapsed Color Strip + Color VS into one
  "Color Source" picker with grouped headers; mappings list shows a
  mode-aware hint when broadcasting a single colour.
- EntityPalette: support non-selectable header rows (with keyboard / filter
  handling) for grouped source pickers.

Bundled UI polish (icon inheritance + cleanup):
- Custom card icons now flow into more surfaces: command palette, dashboard
  target cards, scene-preset target picker, calibration test-device picker,
  and the LED-target device picker. LED targets inherit their device's icon
  when none is set on the target itself.
- Empty mod-card icon plates render as a dashed "+" placeholder when an
  icon-picker hook is wired, so the action stays discoverable.
- Icon picker: distinct "HA light target" eyebrow label and supports
  HA-light cards (data-ha-target-id) for channel-colour resolution.
- Update banner: "View release" now opens the in-app Update settings tab
  instead of an external link; uses the sparkles icon.
- Color-strip delete: cleaner toast on 409 conflict.
2026-05-04 14:27:22 +03:00
alexei.dolgolyov ced72fc864 feat(targets): customisable card icon + HA-light stop action
Extends the icon-plate work from the device cards to LED and HA-light
output targets, and adds finalization behaviour for HA-light targets.

Targets:
- Add `icon` and `icon_color` fields to OutputTarget (LED + HA-light).
- LED target cards inherit the icon from their referenced device when
  no override is set; the icon picker shows an "inherited" indicator.
- Promote the device link from the meta line to a chip with the
  device's custom icon, leaving the head row free for the icon plate.

HA-light:
- New `stop_action` field with three modes: `none` / `turn_off` /
  `restore`. Processor snapshots mapped-entity states at start and
  applies the chosen action on stop (rgb / hs / color_temp / brightness
  restored where present).
- Editor modal exposes the choice via an IconSelect of three modes.

Adjacent fixes:
- Fader slider hit-zone now overlays the visible track exactly,
  regardless of label/value column widths.
- Dashboard customise drag-drop indicator splits into before/after
  rather than highlighting the whole row.
- Picture-source EntitySelect resyncs its visible value on load.
2026-05-04 00:43:55 +03:00
alexei.dolgolyov 49ddabbc36 feat(ui): customisable card icon plate for devices
A user-chosen icon ("mouse", "motherboard", "keyboard"…) renders as a
44x44 instrument-panel face plate at the leading edge of .mod-head on
device cards. Optional per-card; null hides the plate and reverts to
the existing badge-only head.

- Storage/schema: new icon, icon_color fields on Device + DeviceUpdate /
  DeviceResponse. SQLite stores entities as a JSON blob, so no migration
  is needed; from_dict defaults handle existing rows.
- Curated 47-icon library across six categories (Hardware / Lighting /
  Rooms / Media / Signal / Ambience), reusing the existing Lucide path
  module; adds circuit-board, bed, armchair, leaf paths.
- mod-card.ts: ModHeadOpts gains iconHtml / iconColor / iconAttrs;
  ModMenuItemOpts gains optional dataAttrs. The plate is rendered when
  iconHtml is supplied; otherwise no layout change.
- Picker modal (icon-picker.html + features/icon-picker.ts): live
  preview, search, six category tabs, recent strip, channel-color
  override toggle. Wired through document-level click delegation on
  [data-icon-picker-trigger="<deviceId>"] — no window globals, no
  inline onclick string. Sets the precedent for migrating other card
  actions off window in a follow-up.
- en/ru/zh locales for picker UI + categories.

Includes a docs/ mockup that's the source-of-truth for the design.
2026-05-03 15:08:17 +03:00
alexei.dolgolyov a026f0b349 ci(android): fail-fast on missing release keystore before SDK setup
Move the keystore guard from after the Decode step (step 9) to right
after Resolve build label (step 3). A release tag pushed without
ANDROID_KEYSTORE_BASE64 configured now fails in seconds instead of
after JDK + Python + Android SDK + NDK install (~3-5 min of wasted
runner time). Switched the condition from steps.keystore.outputs.present
to env.ANDROID_KEYSTORE_BASE64 since the env var is set at job level
and the keystore decode step has not yet run at the new position.
2026-05-01 19:18:46 +03:00
alexei.dolgolyov 5ef6ac1317 chore: release v0.6.0
Build Release / create-release (push) Successful in 3s
Build Android APK / build-android (push) Failing after 3m52s
Build Release / build-linux (push) Successful in 5m20s
Build Release / build-docker (push) Successful in 6m20s
Build Release / build-windows (push) Successful in 7m7s
2026-05-01 19:11:15 +03:00
alexei.dolgolyov 0980cf4dde fix(ui): audio-source modal — preserve device on refresh, relocate refresh action
- Move the device refresh button into the label row next to "Audio Device:"
  so it can no longer overflow the Source panel edge; introduces a small
  .label-row-action style alongside .hint-toggle.
- Restore device selection after refresh by matching on (index, loopback)
  value first, with a trimmed name fallback for OS-side reindexing.
- _selectAudioDevice now syncs the EntitySelect trigger so the visible
  label matches the underlying <select> when the modal opens in edit mode.
- Drop unused min-width/overflow on .transport-status.
2026-05-01 19:04:36 +03:00
alexei.dolgolyov fdac26b9d9 feat: daylight tz, camera engine, value stream + modal/UI polish
- daylight: new daylight_settings module + daylight-tz frontend helper; expanded daylight_stream behavior
- camera engine: capture path additions plus new test_camera_engine suite
- value stream: schema + processing updates (~178 lines)
- color strip: drop cycle effect (cycle.py / color-cycle.ts removed), tighten static path
- modal CSS: large refactor (+883), components.css polish (+110)
- templates: settings, css-editor, value-source-editor, test-template, display-picker, image-lightbox
- frontend core: state, modal, icons, graph-nodes, app
- frontend features: displays, streams, streams-capture-templates, value-sources, settings, color-strips/cards
- locales: en/ru/zh
- storage: color_strip, picture_source, value_source loaders touched
- preferences/sync_clocks/picture_sources routes; home_assistant + templates schemas
2026-05-01 18:42:43 +03:00
alexei.dolgolyov 816a27db73 refactor(ui): drop app footer, move author info to About panel
The "Created by Alexei Dolgolyov..." line lived in a global app
footer that took up vertical space on every page. Move the author
+ contact details into the About tab of the global settings modal
(rendered by renderAboutPanel), where they sit next to the version
pill and license. Adds a localized "donation.about_author" key
(en/ru/zh) and matching .about-hero .about-author styles. Removes
the now-unused .app-footer / .footer-content rules.
2026-05-01 10:55:31 +03:00
alexei.dolgolyov 797b806972 feat: LED hot-path perf, tutorials expansion, modal markup polish
Performance (LED hot path, allocation-free per-frame):
- Adalight: dedicated single-worker tx executor (avoids asyncio.to_thread
  overhead), pre-allocated wire buffer + uint8 scratch, header struct
  precomputed. New tests cover header format, buffer reuse, non-contiguous
  input, and brightness scaling.
- DDP: pre-built struct.Struct for the 10-byte header, allocation-free
  send buffer + memoryview emit path. New tests cover RGB/RGBW packets,
  sequence/PUSH semantics, and multi-packet fragmentation.
- Calibration: precomputed Phase 3 skip-LED resampling (floor/ceil indices,
  fractional weights, take/blend scratch buffers) — per-frame work is now
  np.take + in-place blend, no allocations.
- WLED target processor: matching hot-path tightening.

Tutorials:
- Sub-tab switching, breadcrumb header, and prepare/switchSubTab hooks
  so a tour can open/close the dashboard customize panel and resolve
  targets behind sub-tabs.
- New steps for integrations tab, dashboard customize panel (presets,
  global, sections, perf cells), targets, scenes, sync-clocks.
- en/ru/zh locales updated with the new tour strings.

Dashboard layout:
- Structural deep-equal so the "modified" indicator reflects truth after
  a user edits then reverts, instead of a stale flag.

UI polish:
- Mod-card / modal markup pass across ~33 modal templates and the
  tutorial overlay partial.
- appearance.css, modal.css, tutorials.css refresh.

Tooling:
- Add .mcp.json with code-review-graph MCP server config so the graph
  tools are available to the team out of the box.
2026-05-01 03:02:13 +03:00
alexei.dolgolyov 9d4a534ec6 feat(ui): release notes overlay v2 + settings/streams/dashboard polish
Release Notes overlay redesign (scoped via .release-notes-shell)
- Backend exposes release.assets (name/size/download_url) through
  UpdateReleaseInfo so the frontend can render real download links.
- New masthead: eyebrow + display-font title + tag/published/pre-release
  chip strip + close/external action buttons; opts out of layout.css's
  global `header { height: 60px }` and `header::before` accent bar that
  were leaking into the overlay's <header>.
- Markdown body: <code> filenames are wrapped in clickable <a> via fuzzy
  asset match (exact basename, then same-extension token-overlap), with
  per-asset description tooltip and a small download glyph.
- Per-asset description derived from filename pattern (Windows installer
  /portable/msi, Linux tarball/AppImage/deb/rpm, macOS dmg/pkg, Android
  apk/aab, iOS ipa, generic archives) with i18n keys in en/ru/zh.
- Hide checksum / signature side-files (.sha256/.sha512/.sig/.asc/...).

Settings modal & dashboard polish
- ds-section refresh, rail-channel routing, notif matrix updates.
- Dashboard customize panel + per-account layout updates.
- New docs/settings-modal-redesign.html design reference.

Streams / targets / color-strip
- Stream cards rewrite (cards.css, streams.css, streams.ts).
- Composite stream + metrics history adjustments.
- WLED target processor + color-strip pipeline refinements.
- Color-strip WS source streamer touch-ups.

Misc
- Perf charts overhaul; tabular game-integration / HA / MQTT / weather
  source cards; donation/sync-clocks/scene-presets minor polish.
- New i18n keys across en/ru/zh.

Test infrastructure
- conftest pre-creates the test DB so main.py's legacy-data migration
  doesn't shovel the user's production DB into the test temp dir.
- test_preferences_notifications wipes its own setting at the start of
  the defaults test (was relying on isolation it never enforced).

Pre-commit gates: ruff clean, tsc clean, npm run build clean,
pytest 899/899 passing.
2026-04-29 17:14:05 +03:00
alexei.dolgolyov 51eebf21d5 feat(ui): redesign target pipeline as compact strip + chip row
Drops the legacy "Pipeline details" collapsible block on running
LED target cards. Instead:

- Always-visible 4px segmented timing bar (extract / map / smooth /
  send for video, read / fft / render / send for audio) — same
  stage colors as before, scaled by per-segment ms cost
- One chip row beneath it: total ms / frames count / keepalive
  count, using a new .chip--inline variant (display-weighted number
  + tiny mono-caps unit)
- _patchTargetMetrics now writes only the bar's segments and the
  data-tm spans — bar wrapper survives across polls so the
  flex-transition animates smoothly between samples
- _buildLedTimingHTML replaced by _buildLedTimingSegments (no more
  header / total / legend wrappers — those live in the chip row)

Cleanup
- Drop .target-metrics-collapse / -toggle / -animate / -expanded
  CSS — no callers remain
- Drop targets.metrics.pipeline from en/ru/zh locales — toggle
  label is gone
2026-04-27 01:52:24 +03:00
alexei.dolgolyov 9067db2639 feat(ui): align Targets metric cells with dashboard pattern
mod-card.ts
- ModMetricOpts.extra: raw HTML appended after the .v cell — used
  to embed a sparkline canvas inside the FPS metric block
- ModMetricOpts.valueDataAttrs: data-attrs on the .v element so
  live-update patchers can target the value directly

LED target card
- FPS sparkline (mod-metric-spark-canvas) is embedded INSIDE the
  FPS cell as a sibling of .v — was a separate target-fps-row
  block before, which floated under the metrics grid
- Label hardcoded to "FPS" (the i18n value "Target FPS:" was
  meant for the editor field, not the readout)
- Uptime cell gets ICON_CLOCK; Errors cell gets ICON_OK / ICON_WARNING
  based on count — matches dashboard cell decorations
- Drops the leading FPS icon (display-font number is the focal
  element; no icon needed)
- _patchTargetMetrics now emits the dashboard FPS shape:
  current<span.dashboard-fps-target>/target</span>
  <span.dashboard-fps-avg>avg N.N</span> — picks up dashboard.css
  styling for free

HA Light target card
- Same icon treatment (Uptime → clock; HA → ok/warning by
  ha_connected); FPS icon dropped

Grid sizing
- .devices-grid bumped from minmax(300px, 1fr) / gap 20px to
  minmax(min(380px, 100%), 1fr) / gap 14px — matches the
  dashboard's section grid so metric values like "1m 43s" stop
  truncating at the typical desktop width
2026-04-27 01:42:26 +03:00
alexei.dolgolyov 233b463ac3 feat(ui): migrate Targets cards to mod-card system
LED targets and HA Light targets adopt the dashboard's instrument-
readout vocabulary (mod-head, mod-leds, mod-metrics, mod-foot,
mod-patch, mod-btn, kebab menu) — same classes and tokens already
used by Dashboard and device cards.

mod-card.ts
- ModBtnOpts.dataAttrs for arbitrary data-attrs (used by the LED
  preview toggle's data-led-preview-btn binding)
- ModBodyOpts.extraHtml escape-hatch for live-update widgets that
  don't fit the predefined slots (FPS sparkline canvas, entity
  swatch grid, collapsible pipeline metrics)

LED target card (targets.ts)
- Badge "LED · TGT" pairs with device "WLED · OUT"-style badges
- Meta row: device link → protocol badge → fps → pixel count
- LED bezel: 1-3 dots reflecting checking / streaming / online /
  offline / unreachable
- Headline metrics on running cards (FPS / ERR / UPTIME) preserve
  data-tm selectors so _patchTargetMetrics still patches in place
- Chips for CSS source link, brightness/value-source, threshold
- Patch indicator: STREAMING / UNREACHABLE / STANDBY / OFFLINE /
  CHECKING
- Foot: START/STOP go/stop variant + LED preview + Edit
- Kebab menu: Duplicate / Hide / Delete (replaces top-right trash)
- FPS sparkline + collapsible pipeline preserved via extraHtml
- Tag chips and LED preview panel appended after wrap (mirrors
  devices.ts pattern)

HA Light target card (ha-light-targets.ts)
- Badge "HA · LIGHT"
- Meta: HA source link → light count → update rate
- LEDs: blink running, fault when ha_connected === false, off idle
- Running metrics: RATE / UPTIME / HA status
- Patch: STREAMING / DISCONNECTED / STANDBY / NOT CONFIGURED
- Buttons keep [data-action] for initHALightTargetDelegation
- Live entity color swatches preserved via extraHtml

Misc
- Chip border-radius dropped from 999px (pill) to var(--lux-r-sm,
  3px) — sharp corners match badges/metrics/buttons elsewhere
- _patchTargetMetrics FPS readout uses <small> for the target
  fraction instead of the legacy target-fps-target span
2026-04-27 01:33:13 +03:00
alexei.dolgolyov de13f44f24 feat(autostart): suppress browser auto-open on Windows login
When the user enables "Start with Windows" in the installer, the app
launches on every PC login. Previously each login popped a fresh WebUI
tab, which is noisy for a tray-resident background service.

The autostart shortcut now passes --autostart to start-hidden.vbs, which
sets LEDGRAB_AUTOSTART=1 in the child env. __main__ checks this flag
alongside LEDGRAB_RESTART when deciding whether to open the browser.

Manual launches (desktop/start-menu shortcuts) and the installer's
post-install "Launch LedGrab" finish-page action are unchanged — they
don't pass the arg, so they still open the WebUI tab.
2026-04-26 23:41:03 +03:00
alexei.dolgolyov 1c9acc5afb feat(api-input): make SegmentPayload start/length optional
start defaults to 0, length defaults to led_count - start (the rest of
the strip from start). A single segment with only mode + color now
fills the entire strip — no more length: 9999 magic value clients have
to pass.

Buffer auto-grow only fires for segments with an explicit length past
the current end; implicit "to the end" segments adapt to the current
strip size.
2026-04-26 23:34:42 +03:00
alexei.dolgolyov a56569b02f feat(ui): cards redesign + settings, modal, toolbar polish
Dashboard cards (mod-card system)
- New mod-card / mod-menu modules backing dashboard cards
- Reworked card colors, sections, dashboard layout, perf charts
- Channel-stripe styling, hairline borders, signal-flow animation
  on running cards, refined metric grid

Multiselect bulk toolbar
- Replaced tri-state checkbox with explicit Select-all / Deselect-all
  icon buttons; both disable when not applicable
- Dim + slight blur on non-selected siblings during selection mode so
  the active picks pop; selected card gains a subtle lift + primary-color
  glow halo
- Bulk tick uses ICON_CHECK from the icon registry (was U+2713) and
  scale-pops in via a cubic-bezier overshoot keyframe
- Toolbar restyled with luxury gradient bg, top accent stripe, glass
  blur, neon hover glows on each button group

Settings modal
- Tab bar converted to icon-only (cog / hard-drive / bell / palette /
  refresh / help) so labels never overflow at any locale; title and
  aria-label preserve translated names. Tabs distribute evenly via
  flex: 1 1 0 + space-around — no overflow possible
- IconSelect auto-populates <option> elements when the underlying select
  is empty, fixing the blank notification triggers (root cause: setting
  .value on an empty select is a no-op)
- Tab activation calls scrollIntoView on the active button as a safety
  net for narrow viewports

Modal exit animation
- Added symmetric fadeOut + slideDown keyframes; .modal.closing applies
  them with animation-fill-mode: forwards
- Modal.forceClose() defers display:none until animationend (with timer
  fallback). State cleanup (focus, body lock, stack) runs immediately so
  callers querying state get correct values
- isOpen returns false during the close animation; open() cancels any
  in-flight close so re-open works during the animation
- prefers-reduced-motion disables all modal animations

Locale picker
- Dropped redundant English/Русский/中文 long-form labels — picker now
  shows only EN / RU / ZH
- IconSelect trigger/cell hides empty icon/label slots via :empty so the
  layout collapses cleanly for minimal items

Filter input (cards section)
- Embedded magnifier icon via data URI (no HTML change); monospace
  uppercase placeholder, lux-bg-0 background, neon focus ring with inset
  shadow + outer glow
- Reset button only shows when the input has content (CSS-only via
  :placeholder-shown sibling selector — JS-resilient)

Snack toast
- Glass background (gradient + backdrop-blur) with top channel-color
  accent stripe matching the modal/toolbar language
- Per-type --toast-ch drives border/glow/timer color (success → primary,
  error → danger, info → info)
- Undo button gets a tinted hover with channel-color halo

Top header toolbar
- Removed hairline border from .header-btn for a flatter look; hover
  keeps the subtle background tint and primary-color glow

Device URL hyperlink
- Styled .mod-meta__link to pick up the card's --ch accent (instead of
  inheriting browser-blue underline). Dotted underline at rest solidifies
  on hover; soft text-shadow glow; web icon dims at rest, brightens on
  hover

Misc
- ICON_CHECK and ICON_HARD_DRIVE added to the icon registry
- Existing card-redesign demos checked in under docs/
- Removed obsolete docs/plans/device-typed-configs.md
2026-04-26 03:10:16 +03:00
alexei.dolgolyov ccf4406349 Merge branch 'feat/device-event-notifications'
Configurable device-event notifications: snackbar + Web Notifications
for online/offline (configured targets) and discovery (new WLED/serial
on the LAN/USB) events, with per-event channel matrix and background
discovery toggle.
2026-04-25 17:49:30 +03:00
alexei.dolgolyov 8aa3a323d6 feat(notifications): device event notifications (snack + Web Notifications)
Surface device connection state changes (configured target online/offline)
and discovery events (new WLED on LAN, new serial port, devices that
disappear) through a configurable per-event channel matrix:
none / snack / OS / both.

- Backend: long-running mDNS browser + 10 s serial poller in
  core/devices/discovery_watcher.py, gated by user pref. Reuses the
  existing device_health_changed event for online/offline transitions.
  New GET/PUT /api/v1/preferences/notifications endpoint with Pydantic v2
  schema (channel matrix + background-discovery flag + grace/debounce).
  13 new tests, full suite still 899 passing.
- Frontend: features/notifications-watcher.ts with startup-grace +
  flap-debounce + bulk-coalesce pipeline. Web Notifications API for the
  OS channel (no platform-specific code, works in PWA shell).
  New "Notifications" tab in Settings with 4 IconSelect rows + bg toggle
  + permission row + test button. en/ru/zh translations.

Defaults: device_offline=both (urgent), online/discovered=snack, lost=none,
background discovery on. Already-configured devices are filtered from
discovery events to avoid double-notifications.
2026-04-25 17:49:20 +03:00
alexei.dolgolyov 8e109f32b9 fix(pwa): add mobile-web-app-capable meta tag
Chrome deprecated apple-mobile-web-app-capable in favor of the
standard mobile-web-app-capable. Add the new tag while keeping the
Apple variant for iOS Safari compatibility.
2026-04-25 15:36:59 +03:00
alexei.dolgolyov 033c1f6a92 ci: add workflow_dispatch and skip lint/test on release commits
Release-bump commits don't change code that affects lint/tests, and
release.yml already runs in parallel. Manual dispatch lets us re-run
on demand if needed.
2026-04-25 15:36:51 +03:00
alexei.dolgolyov 0804f54537 chore: release v0.5.0
Build Release / create-release (push) Successful in 3s
Build Android APK / build-android (push) Failing after 2m16s
Build Release / build-linux (push) Successful in 3m54s
Build Release / build-docker (push) Successful in 7m30s
Build Release / build-windows (push) Successful in 8m37s
Lint & Test / test (push) Successful in 8m45s
2026-04-25 15:21:02 +03:00
alexei.dolgolyov 66f921c07f Merge branch 'feat/lumenworks-ui-redesign'
Lint & Test / test (push) Successful in 2m36s
Lumenworks studio-console redesign + per-account dashboard customization
+ Inputs/Integrations/Graph treatment + transport-bar uptime + server
shutdown action.

Sub-features (in order on the branch):
- feat(ui): Lumenworks tokens, fonts, transport bar, channel-strip sidebar
- feat(ui): dashboard polish, perf strip, transport-bar controls
- feat(dashboard): per-account customizable dashboard with slide-in panel
- feat(ui): item-card restyle, perf hover tooltips, FPS ceiling
- feat(ui): Lumenworks treatment for Inputs / Integrations / Graph tabs
- fix(ui): cards on pure black/white, decoupled from bg-anim
- fix(ui): single-row header + readable sidebar labels at narrow widths
- feat: server shutdown action with public cancel_task lifecycle method
- feat(ui): live card-color picker, monotonic uptime ticker, default
  preset uses base palette
- fix(ui): channel stripe paints only on custom-color or running cards
- chore: harden test isolation, gitignore stale src/data, mark TODO done

Pre-merge audit:
- 886/886 pytest passed twice in a row
- ruff + tsc clean
- frontend bundle rebuilt at static/dist
- python package reinstalled in editable mode (dev WebUI now reports
  0.4.2 instead of stale 0.3.0 dist-info)
2026-04-25 15:12:27 +03:00
alexei.dolgolyov 80f01d4813 chore: harden test isolation, gitignore stale src/data, mark shutdown action done
- ``tests/test_preferences_api.py`` no longer captures the auth API
  key at module-import time. The new ``client`` fixture resolves it
  inside its body and bakes the Bearer header into ``TestClient.headers``,
  so the e2e conftest swapping the global config singleton during
  collection cannot leave the test holding a stale 401-bound header.
  Same proven pattern as ``test_audio_processing_templates_api.py``.
- ``.gitignore`` now anchors ``/server/src/data/`` defensively. If the
  server is launched from ``server/src/`` (uncommon but possible during
  ad-hoc debugging), its relative ``data/`` resolves there. Templates
  now live in SQLite (``capture_templates`` / ``pattern_templates`` /
  ``postprocessing_templates`` tables); any stale ``*.json`` that
  lands in that directory is a runtime export and must not be
  committed.
- Three such stale exports were untracked at the start of the
  pre-merge audit and have been deleted from the working tree.
- ``TODO.md`` flips the shutdown-action checklist to done and notes
  that real-hardware verification (WLED + serial after Ctrl+C) is
  still pending.
2026-04-25 15:11:39 +03:00
alexei.dolgolyov b1ee3c3942 fix(ui): channel stripe paints only on custom-color or running cards
Reported during pre-merge review of the Lumenworks redesign:
non-dashboard cards (Inputs, Integrations, Targets) all showed
aggressive cyan / green left stripes "regardless of custom color set
or not", and even the ``+`` Add card carried a stripe.

Root cause: ``.template-card`` defaulted to ``--ch: cyan`` and
``::before`` painted it unconditionally; ``.add-template-card``
inherits ``.template-card`` so it picked the stripe up too.

Fix: gate the ``::before`` channel stripe behind opt-in selectors.
It now paints only when the card carries ``data-has-color="1"``
(the user picked a personal colour via the picker, set by ``wrapCard``)
or has the ``card-running`` class (the "patched and live" indicator).
Dashboard module rows are unchanged — their ``::before`` already
runs at ``opacity: 0.6`` and was approved as the visual benchmark.

``add-template-card::before`` is hidden unconditionally with
``!important`` since the Add card is not an entity and should never
carry a channel hue.
2026-04-25 15:11:24 +03:00
alexei.dolgolyov e0ff40f4f5 feat(ui): live card-color picker, monotonic uptime ticker tweaks, default preset uses base palette
Three adjacent UI fixes that surfaced while soaking the Lumenworks
redesign:

- ``card-colors.ts`` now writes the user's picked color through to
  *every* card representing the same entity (e.g. the targets-tab card
  AND its dashboard mirror), not just the one that owns the picker.
  Sets the ``--ch`` custom property on each match instead of a literal
  ``border-left``, which avoided the double-stripe (custom border +
  Lumenworks ``::before`` channel stripe) the old approach produced
  and reaches mirrors the picker callback's ``.closest()`` lookup
  couldn't.
- ``appearance.ts`` "default" preset now *clears* its colour overrides
  instead of stamping the historic muted greys (#1a1a1a / #2d2d2d /
  #f5f5f5 / #ffffff). With the redesign's pure-black / pure-white base
  palette in ``base.css``, "default" should mean "use the base" — the
  preset swatch in Appearance now matches what ships out of the box.
  Existing users with "default" selected will see a one-time visual
  shift to the new neutrals on next reload; this is intentional.
- ``dashboard.css`` mod-metric label row gets explicit sizing for the
  small status glyphs (clock / check / warning) so they sit beside the
  mono-caps label without competing with the big value. Errors cell
  picks up the coral channel tint when the count is non-zero.
2026-04-25 15:11:09 +03:00
alexei.dolgolyov 3f80ef2101 feat: server shutdown action with public cancel_task lifecycle method
Lets users choose what happens to LED targets when the server shuts
down. Default ("stop_targets") runs the existing per-device stop
sequence, so devices with auto-restore replay their prior state.
"Nothing" cancels the capture tasks without sending restore frames,
so the LEDs keep displaying their last frame on shutdown.

Backend:
- New setting ``shutdown_action`` persisted in db.settings
  (``stop_targets`` default | ``nothing``) with GET/PUT
  ``/api/v1/system/shutdown-action`` endpoints
- ``ProcessorManager.stop_all(restore_devices: bool = True)`` now
  picks the path based on the flag — ``proc.stop()`` for the normal
  branch, public ``proc.cancel_task()`` for the "nothing" branch.
- ``TargetProcessor.cancel_task()`` (new, on the abstract base) cancels
  the loop task and *awaits* its termination so no half-written frame
  is in flight when the process exits. Replaces an earlier draft that
  reached into the private ``_task`` attribute via ``getattr``.
- Lifespan in ``main.py`` reads the setting at shutdown and forwards
  the flag; falls back to ``stop_targets`` on any read error.
- ``/health`` exposes ``uptime_seconds`` (process-wide monotonic clock
  captured at first import of ``api.routes.system``) so the WebUI can
  show the *server's* uptime instead of the browser session's.

Browser launch:
- ``__main__._open_browser`` now polls ``/health`` for up to 30 s
  instead of sleeping a flat 2 s, so the tab opens once the server
  actually accepts requests.

Frontend:
- New "Shutdown action" picker in Settings → General, rendered via
  IconSelect with ICON_SQUARE / ICON_CIRCLE (added to ``core/icons.ts``
  + ``circle`` path to ``icon-paths.ts``).
- Transport-bar uptime ticker reads ``window.__serverUptime`` (typed
  in ``global.d.ts``); shows "—" until the first /health response
  lands so refresh doesn't briefly flash 00:00:00. After 99 h the
  format widens to "Dd HH:MM:SS".
- New i18n keys for the action picker (label, hint, opt.stop /
  opt.nothing + descriptions, saved / save_error toasts) in en/ru/zh.

No data migration needed — the setting is additive and defaults to
the existing behavior.
2026-04-25 15:10:48 +03:00
alexei.dolgolyov 2bae304107 fix(ui): single-row header + readable sidebar labels at narrow widths
At ≤1100px the header grid only declared 3 tracks for 4 children, so
the toolbar wrapped to a second row, doubling the header height. Add a
4th track, tighten the meta cluster, and hide non-essential toolbar
items (API link, tour-restart) so everything fits in one row. At
≤900px drop CPU/Mem cells (Uptime + Poll remain) so the toolbar still
fits beside the meta cluster.

Sidebar tab captions on the 56 px icon rail were ellipsis-truncated to
"DASHBO…" / "AUTOMA…" / "INTEGR…". Switch to a 2-line clamp with
tighter font/tracking so each label renders in full.
2026-04-25 13:54:18 +03:00
alexei.dolgolyov dd415e2813 fix(ui): cards on pure black/white, decoupled from bg-anim
Three related fixes after the Phase-4 migration landed:

- `--card-bg` flipped from `#101216` / `#f5f6f8` to pure `#000000` /
  `#ffffff` in base.css. Off-pure greys read as muddy when sitting on a
  pure-black/white page background; pure values keep card surfaces flush
  with the rest of the chrome and let the channel stripe + corner
  bracket carry all the visual differentiation.

- Removed the `[data-bg-anim="on"] .card { background: rgba(...) }`
  block that turned every entity card translucent whenever the WebGL
  background was enabled. Card backgrounds are now stable across the
  toggle — the shader bleeds through `body { background: transparent }`
  only, not through cards. The same card now reads identically with the
  shader on or off.

- WebGL shader base colour (`_bgColor` in bg-anim.ts and bg-shaders.ts)
  was using the legacy mid-grey `#1a1a1a` / `#f5f5f5`. That added a
  constant grey haze under the additive accent glow that didn't exist
  on the surrounding pure-black/white page. Switched to `[0,0,0]` /
  `[1,1,1]` so the shader composes against the same base as the page.

- Reverted two leftovers from the Phase-4 commit where I had migrated
  `.template-card` and `.graph-node-body` away from `var(--card-bg)`
  toward `var(--lux-bg-1, …)`. Those backgrounds now live on
  `var(--card-bg)` again, matching every other migrated card.
2026-04-25 02:42:57 +03:00
alexei.dolgolyov b43e1cf375 feat(ui): Lumenworks treatment for Inputs / Integrations / Graph tabs
Brings the remaining tabs in line with the Channels-tab visual language:

- .template-card now mirrors .card and .dashboard-target — channel stripe
  on the left edge with glow, silkscreened corner bracket top-right,
  hairline border on --lux-bg-1, hover lift + stripe widen-and-glow.
  Covers streams, capture / pp / cspt / pattern / audio templates and
  every Integrations card (HA / MQTT / weather / value / sync clocks /
  game integrations).

- Channel mapping extended in cards.css. Direct attribute hooks for the
  per-domain ids; section-scoped hooks via [data-card-section="…"] for
  the cards that share a generic data-id (HA / MQTT / weather / value
  → cyan, game-integrations → amber, sync-clocks → violet,
  HA-light-targets → signal). No JS changes — uses the section markup
  CardSection.render already emits.

- Graph editor nodes pick up the studio-console palette: --lux-bg-1
  fill with hairline stroke, hover bold-line, selected/running stroke
  --ch-signal with drop-shadow glow. Title font moved off Big Shoulders
  Display (which read as "stretched" at 12 px) onto --font-body
  (Manrope); subtitle keeps the mono-uppercase caption treatment with a
  conservative letter-spacing. Running gradient now rides the channel
  palette (signal → cyan → signal) rather than the legacy primary /
  success colours. Port labels and grid dots adopt --lux-line tokens.

- Graph node titles get real text-overflow:ellipsis behaviour. SVG
  <text> can't do that natively, so renderNodes runs a post-mount fit
  pass that binary-searches the longest character prefix that fits
  inside the clip rect (with 2 px slack), suffixed with "…". Trailing
  whitespace is stripped before the ellipsis so we never get "Foo …".
  Full text is stashed on data-full-text so the fit can be re-run on
  re-renders.

Also bundles two perf-charts fixes from the same session:

- Hover regression — listener was bound to .perf-charts-grid, which
  rerenderPerfGrid() replaces. Moved to document.body with a guard, and
  the cursor → sample math now uses the same sliceN as the spark
  rendering so the tooltip stays accurate when the user changes the
  window setting.

- Color picker on every perf cell. Patches / Total FPS / Devices now
  expose the same color picker as the spark cells; defaults added to
  METRIC_CSS_VARS. Each card gets an inline --perf-accent on render so
  saved colours apply immediately, including across rerenderPerfGrid.
2026-04-25 02:27:38 +03:00
alexei.dolgolyov 56853b7123 feat(dashboard): per-account customizable dashboard with slide-in panel
Open-registry section/perf-cell schema persisted server-side under
db.get_setting('dashboard_layout'); localStorage cache for instant
first-paint, server sync after auth. 5 built-in presets
(Studio/Operator/Showrunner/Diagnostics/TV); JSON export/import.

Slide-in Customize panel toggles section + perf-cell visibility,
reorders via hand-rolled HTML5 drag (with up/down buttons for
keyboard/TV-remote use), changes density per section, and exposes
global Width / Animations / Perf-mode / Window with per-cell Inherit
overrides.

Window setting now drives the actual sparkline slice (30s/1m/2m/5m at
configurable poll interval) instead of always rendering 120 fixed
samples. Perf-grid edits re-render in place — sparklines repaint from
persistent module-level history, value labels replay from cached
last-fetch payload, so there is no flicker frame and no zero-data
window between layout change and next poll. initPerfCharts now fires
an immediate fetch on init so reload no longer shows "—" until the
first interval tick.

Reset confirmation uses the project's themed showConfirm modal
instead of the browser dialog. Reserved registry keys (audio-meters,
alerts, led-preview, source-thumbs, pinned, flow) are forward-
compatible so v1.1 cards slot in without a schema bump.

Backend exposes GET/PUT/DELETE /api/v1/preferences/dashboard-layout
treating the body as opaque JSON with a numeric version gate; covered
by 6 round-trip / validation / unknown-field tests.
2026-04-25 01:43:14 +03:00
alexei.dolgolyov 70c95d1c09 feat(ui): item-card restyle, perf hover tooltips, FPS ceiling
Item cards (Automations, Channels, Inputs, Integrations):
- `.card-title` — bumped to weight 700, -0.01em tracking, solid --lux-ink
  for better presence against the flat card bg.
- `.card-subtitle` / `.card-meta` — mono font, 0.04em tracking, tighter
  gap so rule chips pack in a readable row.
- `.stream-card-prop` rule chips — rectangular 2px radius + hairline
  border + flat dark bg (was rounded 10px grey pill). Channel-signal
  icon tint; hover fades in a channel-green wash with matching border.
- `.badge` generic — rectangular 2px radius, mono 0.62rem, 0.12em
  tracking, hairline border slot for variants.
  - `.badge-automation-active` — channel-signal tinted bg + border +
    soft outer glow so the "ACTIVE" state reads at a glance.
  - `.badge-automation-inactive` / `-disabled` — transparent with a
    hairline outline so they sit quietly alongside the active variant.
- `.device-url-badge` — switched from rounded pill to rectangular
  hairline mono chip; hover shifts to filled bg + bolder border +
  brighter ink.
- `.card-actions` — 1px hairline top divider, 6px gap.
- `.btn-icon` — 7/10px padding, 1rem icon, hairline border, channel-
  signal glow on hover (replaces the old scale(1.1) jiggle).
  - `.btn-icon.btn-warning` — amber ink + hairline + amber hover glow
    (drives the "disable" action in the automation card).
  - `.btn-icon.btn-success` — signal-green ink + hairline + green hover
    glow ("enable" action).

Cross-link navigation highlight:
- `cardHighlight` keyframes were using an undefined `--primary-rgb` var,
  so the outer glow fell back to 59/130/246 (the Tailwind blue default).
  Rewritten with `var(--ch-signal)` + color-mix so the highlight tracks
  the accent picker and reads as signal-green. Added double-layer
  box-shadow (ring + 32px/10px bloom) so the highlight is obvious on
  the flat dark/light card surfaces. Added .dashboard-target to the
  selector + `isolation: isolate` so the glow isn't clipped inside
  overflow: hidden containers (perf strip cells, tree-nav panels).

Perf strip (follow-up polish):
- Total FPS cell shows `/<N>` ceiling suffix next to the live value —
  sum of fps_target across running targets, styled like the Patches
  "/12". A dashed horizontal reference line at that ceiling is rendered
  on the sparkline so the live value reads as "percentage of max
  achievable throughput." Y-axis ceiling grows to targetSum * 1.1 so
  the dashed line never clips.
- Removed the empty `.perf-chart-app` pill in the FPS cell (no app
  variant). Added `:empty { display: none }` as a safety so any other
  unpopulated cell doesn't render a ghost pill.
- Hover tooltips on all sparks — single floating `.perf-chart-tooltip`
  in <body> with fixed positioning; event-delegated from the perf
  grid so re-renders don't need rebinding. Shows metric label + sys
  value + app value (in both-mode) + "−Ns ago" age line derived from
  the poll interval. Vertical marker line follows the cursor over the
  spark; `cursor: crosshair` on the spark container signals interact-
  ability. `pointer-events: none` shifted from the spark container
  down to the inner SVG so hover events land on the container.

Grid:
- Perf strip capped at 4 cols even on widescreen; wraps to 2 rows ×
  4 when the full 7 cells are present. Responsive breakpoints at
  1100 / 760 / 480 px.
- Big value font uses `clamp(1.8rem, 2.8vw, 2.8rem)` so readouts
  like "18.9/31.8 GB" fit a 1fr cell at desktop while still scaling
  down on narrow viewports. `white-space: nowrap; flex-wrap: nowrap;
  overflow: hidden; text-overflow: clip` prevents mid-text wrapping.
- `.perf-chart-spark` uses `margin-top: auto` so sparkline baselines
  align across cells regardless of whether a subtitle is present
  (CPU/GPU model name, FPS min/max).

Dashboard target meta:
- Integrations card stripe reverted to the default signal color so it
  matches the overall accent picker; the health-dot inside the card
  carries the connection state. Removed the per-integration channel
  override in both cards.css and dashboard.css.

Section headers:
- `.dashboard-section-header` / `.subtab-section-header` underline
  switched from dashed to solid; channel-green 40px accent rule on
  the left remains.
- Section count badge (`.dashboard-section-count`) restyled to match
  the rest of the badge family (mono tabular-nums, 2px radius, hairline
  border, --lux-bg-3 fill).

Build: tsc --noEmit clean; CSS bundle stable at ~216 KB.
2026-04-24 21:59:30 +03:00
alexei.dolgolyov e5a2af9821 feat(ui): dashboard polish, richer perf strip, transport-bar controls
Dashboard perf strip:
- Unified rack-module shell with hairline-divided cells (mockup parity)
  replacing 3 separate perf cards. Cells auto-wrap to 2 rows of 4 on
  widescreen; responsive breakpoints at 1100 / 760 / 480 px.
- Active Patches cell (first) shows running/total channel count plus up
  to 4 live FPS readouts with channel-colored stripes; bottom-right
  radial glow anchors the "live channel bank" corner.
- Total FPS cell — aggregate throughput across running targets, mono
  "fps" unit suffix, session-peak-scaled sparkline with a 60 FPS floor.
- Devices cell — online/total count + per-device dot strip (green when
  online with signal-glow, coral when offline, tooltip with name +
  latency), fed from /devices/batch/states (added to the dashboard
  batch poll).
- Value font uses clamp(1.8rem, 2.8vw, 2.8rem) + white-space: nowrap so
  long readouts (RAM "18.9/31.8 GB", GPU "50% · 37°C") scale down
  instead of wrapping.
- Sparklines anchor to the cell bottom via margin-top: auto so baselines
  align across cells regardless of subtitle presence.
- App-load tag ("APP 3.1%") moved to a pinned top-right position per
  card, accent-colored pill; replaces the subdued inline badge.
- Perf mode toggle (System / App / Both) triggers an immediate poll so
  positioning updates without waiting for the next tick.
- Chart.js removed from perf-charts — inline SVG sparklines with
  drop-shadow filter for the "lit instrument" feel. Chart.js still used
  for per-target FPS charts via chart-utils (now owns the registration).
- Fixed history seed bug: app_ram is MB in the server history payload,
  not percent — convert to percent using sample's ram_total before
  pushing into _appHistory.ram. Skip seeding app_gpu_mem since the
  history schema has no gpu_memory_total.
- Temperature card reveals with an explanatory hint when the backend
  reports cpu_temp_hint_key (e.g. Windows without LibreHardwareMonitor)
  instead of silently hiding; .perf-chart-card-hint neutralizes the big
  display font so the message reads as plain body copy.

Transport bar:
- LED brand mark — 28 px, double-layer signal glow (0 22px + 0 8px),
  brandPulse animation. Brand-stack wraps the title + version so
  "LED GRAB" sits above "V0.3.0" on a single line each.
- Transport status chip — bigger (9/18 padding), mono uppercase,
  inner+outer signal glow when .is-armed.
- Transport meta cells — Uptime (JS-local session ticker), CPU (app
  CPU share), Mem (app RAM, G/M format) as stacked KEY/VALUE mono
  readouts with hairline separators.
- New interactive Poll cell cycles through 1/2/5/10s presets on click;
  replaces the range slider that used to live in the Dashboard toolbar
  (it controlled the whole app, not just the Dashboard).
- Header icon buttons — hairline-bordered 30 px squares with channel-
  glow on hover, replacing the pill container.
- Perf poll moved to global bootstrap so transport CPU / Mem stay live
  across all tabs (was paused when leaving the Dashboard).
- Connection pip (#server-status) hidden; the brand mark itself turns
  coral when offline via :has() selector on .header-title.

Dashboard cards:
- renderDashboardTarget now emits full rack-module markup with CH badge,
  name, meta, LED cluster, 3-cell metric grid (FPS / Uptime / Errors),
  and patch-label + stop button. Running cards get the signal-flow
  strip at the bottom. data-fps-text / data-uptime-text / data-errors-
  text hooks preserved so _updateRunningMetrics updates in place.
- LED count surfaced in the target card meta line (e.g. "LED · WLED ·
  144 LED · GRADIENT") when the linked device reports led_count > 0.
- Integrations (HA + MQTT) picked up .mod-head markup — compact module
  layout with online/offline patch indicator. Integration card stripe
  uses the default signal color (not cyan or amber).
- Scene presets, sync clocks, automations gain the same compact module
  treatment. Automations/scenes dropped into a dashboard-autostart-grid
  so they share the visual language.
- Perf mode toggle, stream sub-tabs, cs-count / tree-count /
  tab-badge / dashboard-section-count badges all use the mono
  rectangular style with tabular-nums.

Command palette:
- Flat background (no gradient), channel-accent rule across the top,
  mono placeholder / group headers / footer, active result gets a
  channel-green left stripe.

Modals:
- Popover + backdrop get a stronger radial dim + 6 px blur.
- Per-modal-ID channel lanes (target→green, source→cyan, audio→magenta,
  automation/scene→violet, settings→amber, confirm→coral) via --modal-ch
  override.
- Modal header picks up a vertical channel stripe + hairline divider;
  footer gets hairline top + subtle wash.

Components:
- Inputs use hairline borders + tabular-nums mono for number fields;
  focus state has channel-green ring + soft glow.
- Buttons switch to mono-uppercase with signal-glow on primary,
  coral-glow on danger, hairline border on secondary.
- Card background flattened — removed gradient wash in favor of solid
  --lux-bg-1 for both dark (#0e1014) and light (#f6f8fb).
- Page background: pure black for dark, pure white for light.

Color-picker:
- Always detaches to <body> with fixed positioning when its swatch sits
  inside an overflow: hidden / auto / clip ancestor (perf strip, modal
  bodies, tree-dd panels). Prevents the popover getting clipped.

Settings modal:
- Remembers the last-opened tab via localStorage key
  settings_active_tab; falls back to 'general' if the tab id no longer
  exists. Explicit overrides (donation → about, update badge →
  updates) still work because callers invoke switchSettingsTab after
  openSettingsModal.

Microcopy:
- Sidebar / transport localization for en/ru/zh:
  sidebar.workspaces · transport.meta.{uptime,cpu,mem,poll,poll_hint}
  · transport.status.{ready,armed} · dashboard.perf.{active_patches,
  total_fps,devices}

Backend (coordinated with frontend):
- /system/performance now returns cpu_temp_hint_key when no live CPU
  temperature is available, so the Temperature card can render an
  actionable explainer instead of being hidden. Frontend respects the
  key via t() lookup.

Section headers:
- Underline switched from dashed to solid; channel-green accent rule
  (40 px) on the left remains.

Build / tests:
- ruff clean on touched Python files.
- tsc --noEmit clean.
- Python metrics-provider tests: 18 passed.
- CSS bundle ~214 KB.
2026-04-24 20:28:44 +03:00
alexei.dolgolyov 539e43195f feat(ui): Lumenworks studio-console WebUI redesign
Full-app UI/UX refresh committing to a tech-instrument / studio-console
aesthetic inspired by hardware synths, Eurorack panels, and DAW layouts.

Design tokens and fonts:
- Embed Manrope (body), JetBrains Mono (labels/metrics), Big Shoulders
  Display (numeric readouts) as local .woff2 variable fonts with
  latin + latin-ext + cyrillic + cyrillic-ext subsets via unicode-range.
- New Lumenworks token layer in base.css: --lux-bg-0..3, --lux-line(-bold),
  --lux-ink(-dim/-mute/-faint), --ch-signal/-cyan/-magenta/-amber/-coral/
  -violet channel palette, --lux-signal-glow, --lux-shadow-rack, all
  theme-aware for dark + light. Existing tokens untouched for compat.

Shell (header + sidebar):
- Header rebuilt as a 3-column CSS-grid transport bar (brand | center |
  toolbar) with a glowing LED brand mark rendered via pseudo-elements on
  .header-title. Gradient channel-color rule under the bottom border.
- New sidebar.css introduces a vertical channel-strip nav. Active tab
  gets a glowing left stripe + radial tint + LED pip. .sidebar-foot
  contains a live CPU/FPS meter plate.
- Sidebar collapses to a 56 px icon rail at <=1100 px and hides via
  display:contents at <=600 px so mobile.css's fixed bottom tab-bar
  flows through unchanged.

Cards and dashboard:
- .card gets channel stripe (data-card-type + .ch-* utilities auto-map
  from data-target-id / data-stream-id / data-automation-id etc.), corner
  bracket, gradient background, subtle rack shadow.
- .card-running replaces the old @property --border-angle conic-gradient
  rotating border with a lightweight signalFlow linear-gradient strip on
  the bottom edge (cheaper paint, no GPU layer compositing per card).
- Skeleton loaders rewritten: left hairline + corner bracket + gradient
  shimmer instead of the old text-color opacity pulse.
- .dashboard-target rows pick up the same channel-stripe + signalFlow
  treatment. Section headers use mono micro-caps with a channel-green
  underline accent consistent across the app.
- .perf-chart-card: channel stripe replaces old border-top; per-metric
  accents moved to the channel palette (CPU=coral, RAM=violet, GPU=green,
  temp=amber). Metric values use tabular-nums + a soft glow.

Live bindings (no new endpoints):
- _updateSidebarMeter: binds the sidebar Load + FPS bars to the existing
  /system/performance poll.
- _updateTransportStatus: toggles the transport chip between "Ready" and
  "Armed - N live" whenever the dashboard's running-target set is
  recomputed.

Tree-nav + sub-tabs:
- tree-nav.css trigger pill gets a channel-stripe left edge that glows
  when open; panel has a gradient channel-accent rule across the top;
  group headers use silkscreened micro-caps; active leaf has a pulsing
  LED pip + channel tint.
- .stream-tab-btn / .subtab-section-header adopt the same mono-caps +
  channel-underline language for consistency.
- Graph editor toolbar gets gradient + hairline + rack shadow + backdrop
  blur. Canvas and nodes untouched.

Modals (40+ modals share modal.css):
- Radial-dim + 6 px blur backdrop. Content gets a gradient background,
  hairline border, deep rack shadow, top channel-accent rule driven by
  --modal-ch, bottom-right corner bracket (hidden on mobile fullscreen).
- Per-modal-ID channel lanes: target editors = green, source/input
  editors = cyan, audio = magenta, automation/scene/game = violet,
  settings/auth = amber, confirm = coral.
- Modal headers: vertical channel stripe left of the title + hairline
  divider. Modal footers: hairline top border + subtle gradient wash.

Forms:
- Inputs use hairline borders; number inputs switch to mono + tabular-nums
  for column alignment. Focus state: channel-green ring + soft glow.
- Buttons use mono-uppercase type with signal-glow on primary and coral-
  glow on danger.

Mobile (<=600 px):
- Fixed bottom .tab-bar gets the full Lumenworks treatment: gradient fill,
  top channel-accent rule matching the transport bar, backdrop blur.
  Active tab has an LED pip above the icon + channel tint + icon recolor.
- Fullscreen modals: corner bracket hidden, header stripe slimmed.

Microcopy (en / ru / zh):
- "Targets" -> "Channels" / "Каналы" / "通道"
- "Sources" -> "Inputs"    / "Входы"   / "输入"
- Internal tab keys (dashboard/automations/targets/streams/integrations/
  graph) kept stable so no JS or localStorage migration is needed.
- Added: sidebar.workspaces, sidebar.load, sidebar.fps,
  transport.status.ready, transport.status.armed.

Compatibility:
- All existing class hooks preserved (.tab-bar, .tab-btn, .card,
  .card-running, .tree-dd-*, .cs-*, .perf-chart-card, .modal-content,
  .dashboard-target, etc.). No JS or API changes required for the new
  look to take effect.
- Tour selectors survive (header .header-title, #tab-btn-*, onclick
  markers on theme/settings/search, #cp-wrap-accent, etc.).
- Mobile <=600 px bottom tab-bar keeps working via display:contents
  fall-through in the new sidebar.

Build: tsc --noEmit clean; npm run build clean. CSS bundle grew from
~177 KB to ~201 KB for the full new visual system. Fonts loaded lazily
per unicode-range subset (~98 KB critical path for English).

Phased plan + deferred follow-ups (dashboard hero strip, legacy-token
cleanup) recorded at the top of TODO.md.

Reference mockup: server/docs/ui-redesign-mockup.html.
2026-04-24 15:46:47 +03:00
alexei.dolgolyov c44bb38c43 docs(release): refresh v0.4.2 notes with fix(release) and refactor commits
Build Release / create-release (push) Successful in 3s
Build Android APK / build-android (push) Failing after 2m5s
Build Release / build-linux (push) Successful in 4m53s
Build Release / build-docker (push) Successful in 5m39s
Build Release / build-windows (push) Successful in 6m55s
Lint & Test / test (push) Successful in 7m13s
2026-04-22 20:20:30 +03:00
alexei.dolgolyov be2d5e1670 refactor(color-strips): move Key Colors test from lightbox into test-css-source modal
Lint & Test / test (push) Successful in 6m37s
Removes the inlined FPS select and auto-refresh button from the shared
image lightbox and rehosts the Key Colors live preview inside the
dedicated test-css-source modal alongside the other CSS test views.

- Drop initLightbox() / lightbox-fps-select IconSelect — the lightbox no
  longer owns streaming controls.
- Add #css-test-kc-view (canvas + meta) and .css-test-kc-* styles.
- Reroute _testKeyColorsSource() through the existing modal session
  lifecycle so KC, CSPT, and standard CSS tests share teardown paths.
2026-04-22 20:18:46 +03:00
alexei.dolgolyov 5db6eddcf8 fix(release): ship prebuilt assets and bump fallback version
Two release-blocking bugs traced to the same root cause: the unanchored
`data/` rule in .gitignore matched server/src/ledgrab/data/, which is
where shipped package assets live (prebuilt sounds, game adapters).
The files were never `git add`-able without -f, so they never reached
the v0.4.2 tag and CI builds couldn't include them.

- .gitignore: anchor /data/ and /server/data/ so nested package data
  dirs are not ignored.
- Track previously-excluded shipped assets:
  - server/src/ledgrab/data/prebuilt_sounds/{alert,bell,chime,ping,pop}.wav
  - server/src/ledgrab/data/game_adapters/{minecraft,rocket_league,valorant}.yaml
- Bump _FALLBACK_VERSION 0.3.0 -> 0.4.2 to match pyproject.toml.
  The Windows installer strips ledgrab-*.dist-info, so
  importlib.metadata falls back to this literal — which is why
  v0.4.2 reports v0.3.0 in the WebUI.
- Patch _FALLBACK_VERSION at bundle time in build-common.sh and
  build-dist.ps1 so future drift is auto-corrected by the build.
2026-04-22 20:17:10 +03:00
alexei.dolgolyov a8a4296a56 chore: release v0.4.2
Build Android APK / build-android (push) Failing after 1m48s
Build Release / create-release (push) Successful in 3s
Build Release / build-linux (push) Successful in 3m58s
Build Release / build-docker (push) Successful in 5m6s
Build Release / build-windows (push) Successful in 5m54s
Lint & Test / test (push) Successful in 6m14s
2026-04-22 19:48:37 +03:00
alexei.dolgolyov 9ce1dc33bf feat(ui): restyle enhanced header locale picker as LED-accent badge 2026-04-22 19:48:08 +03:00
alexei.dolgolyov 03d2e6b1f2 ci(release): publish .sha256 sidecars alongside release assets
Lint & Test / test (push) Successful in 2m4s
The in-app update service (`ledgrab.core.update.update_service`) refuses
to install any downloaded artifact that has no published sha256 — either
as a sibling `<asset>.sha256` asset on the Gitea release, or embedded in
the release body. The release workflow uploaded the ZIP, setup.exe, and
Linux tarball but never published checksums, so every auto-update 500'd
with "Update checksum unavailable; install aborted".

Generate sha256sum sidecars for the Windows ZIP, Windows setup.exe, and
Linux tar.gz and upload them next to the primary asset on each tagged
release. Existing v0.4.x releases stay broken — ship v0.4.2 (or manually
upload sidecars to v0.4.1) to unblock in-app updates.
2026-04-22 19:40:46 +03:00
alexei.dolgolyov c2c9af3c60 chore: release v0.4.1
Build Release / create-release (push) Successful in 4s
Build Android APK / build-android (push) Failing after 1m41s
Build Release / build-linux (push) Successful in 3m3s
Build Release / build-docker (push) Successful in 4m13s
Lint & Test / test (push) Successful in 5m24s
Build Release / build-windows (push) Successful in 5m33s
2026-04-22 19:21:27 +03:00
alexei.dolgolyov 4f7794ccd4 fix(installer): bundle cryptography + just-playback, set TCL env, clean stale debug.bat
Lint & Test / test (push) Successful in 2m20s
Windows installer silently failed to launch because build-dist-windows.sh
maintained its own DEPS list that drifted from server/pyproject.toml and
was missing `cryptography` — ledgrab.utils.secret_box imports AESGCM at
module load, so pythonw.exe crashed before the tray icon appeared. Also
missing: just-playback (lazy import, silent until a sound triggers).

- Add cryptography + just-playback to DEPS with a sync-with-pyproject
  warning comment
- Extend the post-cleanup on-disk check to abort the build if
  cryptography / cffi / just_playback go missing again
- Launcher now exports TCL_LIBRARY / TK_LIBRARY so the screen-overlay
  tkinter thread stops logging "Can't find init.tcl" at startup
- Installer wipes stale debug.bat / debug.log on install and uninstall
  (leftovers from the pre-rename wled_controller era produced a
  misleading ModuleNotFoundError when users tried to diagnose launch
  failures)
2026-04-22 19:19:07 +03:00
alexei.dolgolyov a0d63a3663 docs(release): drop stale WLED-rename task, document android signing secrets
Lint & Test / test (push) Successful in 2m1s
- Remove the top-of-file "IMPORTANT: Remove WLED naming throughout the
  app" checklist. The effort was absorbed by the multi-backend refactor
  (BLE / USB-serial / ESP-NOW / MQTT / OpenRGB providers all shipped),
  and the remaining user-facing copy has been swept in separate commits.
- Add an "Android Signing Secrets (Gitea)" section covering the four
  secrets the release APK CI expects, the one-off `keytool` command to
  generate `release.jks`, the consequences of losing the keystore, and
  a checklist of the remaining setup steps before tagging v0.4.1.
2026-04-21 20:01:26 +03:00
alexei.dolgolyov 35b75a2ed8 ci(android): fix keystore env scoping, fail loudly on release without key
Lint & Test / test (push) Successful in 3m17s
Primary bug — step-level env is not visible in that same step's `if:`
expression. `Decode signing keystore` had
  if: env.ANDROID_KEYSTORE_BASE64 != ''
  env:
    ANDROID_KEYSTORE_BASE64: ${{ secrets.ANDROID_KEYSTORE_BASE64 }}
so the env context seen by the `if:` evaluator was empty regardless of
whether the secret was configured. The step was skipped, keystore.present
never became 'true', and every release tag silently fell back to
assembleDebug. Result: APKs named `LedGrab-0.4.0-android-debug.apk` that
can't upgrade a previously-release-signed install (signature mismatch).

Fix — move ANDROID_KEYSTORE_BASE64 to the job-level env block. It's now
resolvable in the if-expression of any step in the job, and the shell
inherits it exactly the same way as before.

Secondary — add a "Guard release tag against missing keystore" step that
fires between the decode attempt and the gradle build. If is_release=true
but keystore.present!='true', the job fails with a clear error directing
the operator to configure the four signing secrets. Previously a
misconfigured Gitea silently shipped debug APKs labeled as releases.
2026-04-21 19:55:55 +03:00
alexei.dolgolyov 4ed099d564 docs(release): drop WLED-specific language from auto-generated release notes
The "discover your WLED devices" line predates BLE / USB-serial / ESP-NOW /
MQTT / OpenRGB support and misrepresents what the app does. Replaced with
a generic "add your LED devices" — the device-add UI lists what's actually
supported, and INSTALLATION.md carries the long-form detail.
2026-04-21 19:55:55 +03:00
alexei.dolgolyov d467eb5dae chore: release v0.4.0
Build Release / create-release (push) Successful in 3s
Build Android APK / build-android (push) Successful in 5m57s
Build Release / build-linux (push) Successful in 5m44s
Build Release / build-docker (push) Successful in 7m51s
Lint & Test / test (push) Successful in 8m59s
Build Release / build-windows (push) Successful in 8m51s
2026-04-21 19:41:40 +03:00
alexei.dolgolyov 524e422517 ci: decouple android release attach, add workflow_dispatch to release.yml
Lint & Test / test (push) Successful in 2m16s
build-android.yml
- Attach step upserts the Gitea release: GET /releases/tags/<TAG>, and
  POST to create it on 404 instead of warning-and-skipping. Removes the
  ordering dependency on release.yml's create-release job — the Android
  workflow can now own its own release attachment end-to-end.
- Fail loudly on broken DEPLOY_TOKEN: curl -f on every asset call so
  403/422 surface as job failures instead of "Uploaded" lies, and an
  explicit check that the token is non-empty before starting.
- Preserve the pre-existing replace-on-re-run behavior for idempotent
  asset uploads.

release.yml
- Add workflow_dispatch trigger with optional `version` input so the
  Windows/Linux/Docker builds can be exercised on demand between real
  releases (was tag-push only).
- Gate create-release on github.event_name == 'push' so a manual
  dispatch doesn't create a stray Gitea release.
- Each build job gets `if: !cancelled() && (needs.create-release.result
  in (success, skipped))` so dispatch runs still produce artifacts even
  though create-release was skipped.
- Gate each "Attach * to release" step on github.event_name == 'push'.
- Docker: login + push are push-only; build runs on both triggers so
  dispatch validates the Dockerfile without needing registry creds.
2026-04-21 19:10:14 +03:00
alexei.dolgolyov 5d6310f28c fix(android): make wheels find-links URL work on Linux CI
Lint & Test / test (push) Successful in 3m31s
Chaquopy's pip --find-links argument was built with a hard-coded
"file:///" prefix plus rootDir.absolutePath. That works on Windows
(paths start with "C:/...", so prefix + path produces three slashes
after "file:") but breaks on Linux (paths start with "/workspace/...",
so prefix + path produces four slashes — pip then parses "workspace"
as the URL's hostname and aborts with
  "ValueError: non-local file URIs are not supported on this platform".

Pick the prefix based on whether the absolute path already starts with
"/", so we always end up with exactly three slashes between "file:" and
the drive letter or root.
2026-04-21 18:58:17 +03:00
alexei.dolgolyov 7ef17c1595 ci(android): fix missing python symlink parent, restrict to release tags
Lint & Test / test (push) Successful in 3m53s
- Create android/app/src/main/python before `ln -sfn` — the parent dir
  isn't committed (android/.gitignore:17 ignores /ledgrab inside it) so
  fresh CI checkouts had nothing to link into, failing with
  "No such file or directory".
- Also wrap the step with `set -euo pipefail` and a `test -d` check so a
  broken link aborts the job instead of producing a silently-empty APK.
- Drop the `branches: [master]` push trigger and the `paths:` filter —
  every master commit touching android/** or server/src/ledgrab/** was
  queueing a ~100 MB APK build that nobody used. Only tag pushes (v*)
  and workflow_dispatch remain.
2026-04-21 18:53:43 +03:00
alexei.dolgolyov b3775b2f98 feat(android): boot-time autostart, capture watchdog, versionCode from git
Boot-time startup so LedGrab has display capture and control without user
interaction on rooted TV boxes. Also folds in a batch of review findings
from the Android package audit.

Autostart
- BootReceiver fires on BOOT_COMPLETED / LOCKED_BOOT_COMPLETED /
  MY_PACKAGE_REPLACED, gated by AutostartPrefs and Root.looksRooted().
  Dispatches CaptureService.createRootIntent via
  ContextCompat.startForegroundService. Unrooted devices are a no-op
  because MediaProjection consent cannot be bypassed silently.
- AutostartPrefs: thin SharedPreferences wrapper, defaults to enabled.
  Exposed as a CheckBox on the stopped panel; greyed out when not rooted.
- Manifest: RECEIVE_BOOT_COMPLETED, REQUEST_IGNORE_BATTERY_OPTIMIZATIONS,
  WAKE_LOCK permissions + the new BootReceiver.
- MainActivity prompts for battery-optimization exemption on first opt-in
  so Doze/App Standby doesn't kill the FG service on phones.

Service stability
- onStartCommand now flips isRunning only after startForeground succeeds
  (was stuck=true forever if the FG transition threw) and resets on
  exception. Returns START_REDELIVER_INTENT for root mode so the OS can
  restart the service with the original intent (no consent token to
  invalidate); MediaProjection mode keeps START_NOT_STICKY.
- Watchdog coroutine monitors RootScreenrecord.framesDelivered. Respawns
  the pipeline on stall (reusing the existing Python bridge — no server
  restart), caps at 3 consecutive restarts before giving up.
- RootScreenrecord.framesDelivered is now an AtomicInteger, exposed as a
  public property for the watchdog.
- ScreenCapture takes an onProjectionStopped lambda; when the user taps
  the system Cast/Screen-capture stop banner, the whole service is torn
  down instead of leaving a stale FG notification.
- MainActivity's two startForegroundService calls switch to
  ContextCompat.startForegroundService, clearing pre-existing NewApi lint
  errors (minSdk=24 < API 26 native method).

Build
- versionCode derived from git rev-list --count HEAD (or the
  ANDROID_VERSION_CODE env var for CI). Was pinned to 1 — sideload
  upgrades were silently refusing to install.
- New i18n strings (autostart_label, autostart_unavailable, version_prefix)
  in en/ru/zh; version_text now uses the resource instead of string
  concat.

TODO.md: new "Android Autostart on Boot" section tracking done/pending
items; real-hardware verification on a Magisk'd TV box is the remaining
checkbox.
2026-04-21 18:53:27 +03:00
alexei.dolgolyov 45f93fd30e fix(devices): SP110E vendor handshake + Windows/bleak robustness
Build Android APK / build-android (push) Failing after 1m38s
Lint & Test / test (push) Successful in 4m32s
SP110E peripherals silently tear down the GATT link ~1s after connect
unless a two-write vendor handshake (01 00 → FFE2, 01 B7 E3 D5 → FFE1)
arrives immediately. Without it the first real write hangs 30s then
reconnect-loops forever. Adds optional BLEProtocol.init_writes executed
on connect, plumbs a per-write char_uuid through both transports, and
fixes the SP110E color/power frames from an incorrect 5 bytes to the
documented 4 bytes.

Windows/WinRT robustness:
- asyncio.wait_for hangs on bleak because WinRT IAsyncOperations refuse
  to cancel. _bounded_await() uses asyncio.wait() instead so timeouts
  actually return control even when the inner task is uncancellable.
- BleakClient connect by raw MAC string times out when WinRT guesses
  address type wrong; switched to pre-scanning with BleakScanner and
  passing the resolved BLEDevice, which carries the address type.
- Target-start fetch timeout bumped to 30s with retry disabled so the
  UI doesn't abort during the BLE pre-scan + connect + handshake path.

UI:
- Settings modal exposes Protocol Family (IconSelect grid, shared with
  add-device via parameterized ensureBleFamilyIconSelect) so users can
  fix a wrong family pick without recreating the device. Govee AES key
  row toggles on/off with family selection.

Also turns LAN auth back on in default_config.yaml, logs start_processing
requests on entry for easier diagnosis, and captures the full debug trail
in docs/BLE_LED_CONTROLLERS.md for future BLE work.

Refs the mbullington SP110E protocol gist for the handshake bytes.
2026-04-21 17:45:21 +03:00
alexei.dolgolyov 2b5dac2c42 feat(devices): BLE LED controller support (SP110E/Triones/Zengge/Govee)
Build Android APK / build-android (push) Failing after 1m44s
Lint & Test / test (push) Successful in 4m22s
End-to-end BLE streaming: provider + client + per-protocol wire encoders
with whole-strip averaging, desktop (bleak) and Android (Kotlin BleBridge
via Chaquopy) transports, discovery with protocol-family detection that
auto-fills the UI, throttled not-connected warning + 10 s reconnect
cooldown so a dropped link no longer stalls the pipeline at ~30 s/frame,
and an explicit asyncio.wait_for wrapper around bleak connect() since
the WinRT backend doesn't always honor the timeout kwarg.

Also rewrites server/restart.ps1 to be parameterized (-Port / -Module /
-PythonVersion / timeouts / -Quiet), pick the right interpreter via the
py launcher, pre-flight the target module, poll port readiness on both
shutdown and startup, redirect child stdout/stderr so Start-Process
doesn't hang on inherited Git-Bash handles, and return proper exit codes.

Rolls in concurrent work: Android BLE permissions + launcher icons + ru/zh
resources, Chaquopy-safe value_stream psutil fallback, setup-required
modal, asset-store test coverage, and misc system/config touch-ups.
2026-04-21 14:58:35 +03:00
alexei.dolgolyov d3a6416a1d refactor(devices): per-provider typed configs (phases 1-4)
Phase 1 — DeviceConfig hierarchy (device_config.py):
- 17 @dataclass(frozen=True) subclasses (WLEDConfig, AdalightConfig, …) sharing
  BaseDeviceConfig; DeviceConfig = Union[all 17]
- Device.to_config() in device_store.py: single flat→typed dispatch point

Phase 2+3 — Typed provider signatures + call-site migration:
- ProviderDeps(device_store) frozen dataclass in led_client.py
- LEDDeviceProvider.create_client(config, *, deps) abstract signature
- create_led_client(config, *, deps) factory dispatches via config.device_type
- All 17 providers narrowed to their specific config type; drop kwargs.get()
- GroupLEDClient.connect() uses device.to_config() + create_led_client()
- wled_target_processor: replaced 21-field DeviceInfo unpacking with to_config()
  + dataclasses.replace(config, use_ddp=…) for DDP override
- device_test_mode: build typed config via to_config() + ProviderDeps
- Deleted DeviceInfo dataclass, _get_device_info(), _DEVICE_FIELD_DEFAULTS
- TargetContext: replaced get_device_info callback with is_test_mode_active

Phase 4 — Test migration:
- 47-case test suite in tests/core/devices/test_device_config.py (100% coverage)
- test_group_device.py TestGroupLEDClient migrated to GroupConfig + ProviderDeps
- Removed legacy keyword-arg init path from GroupLEDClient
2026-04-18 01:24:27 +03:00
alexei.dolgolyov 123da1b5c4 fix: comprehensive security, stability, and code quality audit
Build Android APK / build-android (push) Failing after 1m45s
Lint & Test / test (push) Successful in 4m54s
Security:
- Force API key auth for LAN (non-loopback) requests; remove shipped dev key
- Block path-traversal in backup restore; require auth on backup endpoints
- SSRF protection: DNS resolve + private/loopback/link-local IP rejection
- AES-256-GCM encryption for HA tokens and MQTT passwords with auto-migration
- WebSocket auth migrated from query-string to first-message protocol
- Asset upload: extension allowlist, server-side mime, Content-Disposition
- Update installer: SHA256 verification, tar/zip member validation
- Tightened CORS (explicit methods/headers, no credentials)
- ADB serial regex allowlist, webhook rate-limit key fix, log scrubbing

Android:
- Root-capture: ordered teardown, screenrecord respawn watchdog, child reaping
- USB permission blocking API via CompletableDeferred
- Python init crash guard with fatal-error screen
- Moved root grant + QR generation off Main thread
- Cached PyObject engine for per-frame bridge calls
- Ordered ScreenCapture resource cleanup, allowBackup=false

Python:
- Replaced all asyncio.get_event_loop() with get_running_loop/to_thread
- Split color_strip_sources.py (1683->5 files) and color_strip_stream.py
  (1324->7 files) into packages
- Extracted FrameLimiter utility, migrated 9 stream loops
- Provider base-class reuse, WLED state caching + URL normalization
- Narrowed broad except-pass in WS routes, threading fixes in BaseStore

Frontend:
- XSS fix: escapeHtml on dynamic option labels, reconcile-based list renders
- Typed DOM helpers, safe localStorage access, AbortController listener hygiene
- openAuthedWs helper for first-message WS auth protocol
- Migrated remaining plain <select>s to IconSelect/EntitySelect

Design:
- WCAG AA primary color on light theme (#2e7d32, 5.4:1 contrast)
- Android TV 10-foot breakpoint (tv.css)
- Consolidated z-index tokens, unified easing, card-running GPU hints
2026-04-16 04:56:04 +03:00
alexei.dolgolyov 5fcb9f82bd feat(android): root-based screen capture bypassing MediaProjection
Build Android APK / build-android (push) Failing after 1m40s
Lint & Test / test (push) Successful in 3m40s
On rooted TV boxes, spawn `su -c screenrecord ... -` and feed the
H.264 stdout through MediaCodec into an ImageReader, surfacing RGBA
frames via PythonBridge. RootScreenrecordEngine (priority 110) is
picked automatically when root is available; falls back to
MediaProjection when Root.requestGrant() returns false.
2026-04-14 19:30:26 +03:00
alexei.dolgolyov 928d626620 refactor(devices): route ESP-NOW client through SerialTransport
Build Android APK / build-android (push) Failing after 1m39s
Lint & Test / test (push) Successful in 4m54s
Drops the direct pyserial imports from espnow_client/espnow_provider
in favor of open_transport/list_serial_ports/port_exists. The gateway
protocol is write-only, so no read() extension was needed. ESP-NOW
gateways are now reachable via usb:VID:PID URLs on Android.
2026-04-14 19:15:08 +03:00
alexei.dolgolyov 580bd692e6 fix(scenes): coerce BindableFloat fps to int when snapshotting
Build Android APK / build-android (push) Failing after 1m41s
Lint & Test / test (push) Successful in 4m32s
OutputTarget.fps is a BindableFloat but TargetSnapshot.fps is a plain
int — capture_current_snapshot was stuffing the BindableFloat directly
into the snapshot, causing json.dumps to fail on recapture (500) and
polluting the in-memory cache so subsequent list calls also 500'd with
pydantic validation errors.
2026-04-14 19:03:58 +03:00
alexei.dolgolyov 7fcb8dd346 feat(devices): Android USB-serial support for Adalight/AmbiLED controllers
Build Android APK / build-android (push) Failing after 1m41s
Lint & Test / test (push) Successful in 4m51s
Adds end-to-end support for driving USB-connected Adalight / AmbiLED
LED controllers from Android TV boxes. Android's security model blocks
direct USB access from Python, so writes route through a Kotlin
UsbSerialBridge singleton via Chaquopy.

Python side:
- New SerialTransport Protocol (serial_transport.py) with open / write /
  flush / close. Desktop uses PySerialTransport (wraps pyserial),
  Android uses AndroidSerialTransport (wraps the Kotlin bridge).
- list_serial_ports() factory returns desktop COM ports on desktop,
  USB devices on Android — callers don't branch.
- URL scheme extended: existing COM3[:baud] and /dev/ttyUSB0[:baud]
  unchanged; new usb:VID:PID[:serial][@baud] for Android (@ is the
  baud separator since : is already used between VID and PID).
- AdalightClient and SerialDeviceProvider refactored to go through
  the transport — no more direct pyserial imports in hot paths.
- 17 new unit tests cover URL parsing, PySerial transport, factory
  selection, platform-branching discovery. Full suite 750 passing.

Kotlin side:
- UsbSerialBridge.kt singleton uses com.hoho.android.usbserial (mik3y)
  which ships drivers for CH340, CP2102, FTDI, Prolific, and CDC-ACM
  (Arduino). Exposes listDevices, open, write, close via @JvmStatic
  for Chaquopy. First open() attempt without permission triggers the
  system USB permission dialog; next call succeeds once user grants.
- usb-serial-for-android is distributed via JitPack — added that repo
  in settings.gradle.kts and the dependency in app/build.gradle.kts.
- AndroidManifest declares uses-feature android.hardware.usb.host
  (required=false so non-USB-host phones still install).
- LedGrabApp.onCreate calls UsbSerialBridge.init(this) so the bridge
  resolves the UsbManager without needing an Activity ref.

Verified: ./gradlew compileDebugKotlin succeeds; off-Android import
of android_serial_transport works. Real-hardware smoke test on a
TV box with a CH340/CP2102/FTDI adapter still pending.

ESP-NOW (espnow_client / espnow_provider) still imports pyserial
directly because it needs bidirectional reads — separate refactor
to extend the transport with read() if that path ever needs Android
USB support.
2026-04-14 16:34:09 +03:00
alexei.dolgolyov ecae05d00b feat(metrics): battery + thermal-zone readings with dashboard temp chart
Build Android APK / build-android (push) Failing after 1m40s
Lint & Test / test (push) Successful in 4m18s
Extends MetricsProvider with thermals() returning a ThermalSnapshot
(battery_percent, battery_temp_c, cpu_temp_c — all optional). Each
provider implements it independently:

- AndroidMetricsProvider reads /sys/class/power_supply/battery/{capacity,
  temp} (battery temp is tenths of degC) and walks
  /sys/class/thermal/thermal_zone*, filtering by zone type
  (cpu/soc/tsens/core) so battery and skin sensors don't dominate the
  reading. Rejects nonsense values like INT_MAX from buggy zones.
- PsutilMetricsProvider uses sensors_battery() and
  sensors_temperatures() when present (Linux+laptops); no-ops on
  Windows/macOS where psutil doesn't expose them.
- NullMetricsProvider returns the empty snapshot.

PerformanceResponse gains battery_percent / battery_temp_c / cpu_temp_c.
The metrics-history ring buffer also carries cpu_temp / battery_pct /
battery_temp per sample so the dashboard can graph them over time.

Frontend dashboard (perf-charts.ts) gets a new Temperature chart card,
hidden by default and revealed only after seed/poll confirms the
backend reports cpu_temp_c. Battery temperature shows inline as a
secondary badge. The GPU card now also hides entirely when the backend
reports gpu=null instead of showing an "unavailable" placeholder.
HOST_ONLY_KEYS prevents the System/App/Both toggle from flipping a
non-existent app dataset for temp.

Tests: 6 new for thermals (battery tenths-of-degC parsing, CPU zone
filtering, fallback when sensors absent, INT_MAX rejection); 18 metrics
tests total; full suite 733 passing.
2026-04-14 13:48:01 +03:00
alexei.dolgolyov 546b24d015 refactor(metrics): MetricsProvider abstraction with Android /proc backend
Build Android APK / build-android (push) Failing after 1m39s
Lint & Test / test (push) Successful in 4m20s
Moves direct psutil.* calls behind a MetricsProvider Protocol so the
codebase no longer needs ad-hoc `if psutil is not None` guards at every
call site. Each provider lives in its own module under
utils/metrics/: PsutilMetricsProvider for desktop, NullMetricsProvider
as a zeroed fallback, AndroidMetricsProvider that reads /proc/stat,
/proc/meminfo, /proc/self/stat, and /proc/self/status directly (psutil
isn't available under Chaquopy). The Android provider tracks the
previous CPU sample so cpu_percent() returns delta-based percentages
matching psutil's interval=None semantics, and degrades to zeros when
any /proc file is unreadable instead of crashing the dashboard.

Factory get_metrics_provider() in utils/metrics/__init__.py picks
Android > psutil > Null. api/routes/system.py and
core/processing/metrics_history.py now go through the factory; psutil
import is confined to one place. 12 new unit tests cover paren-in-comm
parsing of /proc/self/stat, delta CPU%, missing-file resilience, and
factory selection order. Full suite: 727 passing.
2026-04-14 13:34:32 +03:00
alexei.dolgolyov 488df98996 fix(frontend): add autocomplete attrs to credential inputs
Build Android APK / build-android (push) Failing after 1m40s
Lint & Test / test (push) Successful in 3m35s
Silences browser accessibility warnings on the HA token, MQTT username,
and MQTT password fields. Uses new-password for the secret inputs to
discourage Chrome's site-password autofill from leaking into broker /
HA-token configuration.
2026-04-14 12:50:50 +03:00
alexei.dolgolyov 2477e00fae ci: add Android APK row to release downloads table
Lint & Test / test (push) Successful in 1m37s
2026-04-14 12:47:27 +03:00
alexei.dolgolyov 151cea3ecb ci: Android multi-ABI APK pipeline + pydantic-core wheel rebuild
Build Android APK / build-android (push) Failing after 2m31s
Lint & Test / test (push) Successful in 6m15s
Adds .gitea/workflows/build-android.yml — Linux runner installs JDK 17,
Python 3.11, Android SDK/NDK, symlinks server/src/ledgrab into the
Chaquopy python source dir, and runs assembleDebug on master pushes /
assembleRelease on v* tags. APK is uploaded as an artifact and attached
to the Gitea release on tag push. Conditional signing config in
build.gradle.kts reads keystore from env vars (CI secrets) and falls
back to debug signing locally. Gradle wrapper (gradlew/gradlew.bat/
gradle-wrapper.jar) committed so CI can drive the build.

Rebuilds pydantic-core wheels for arm64-v8a and x86_64 — both were
missing libpython3.11.so in NEEDED, which would have crashed at import
on real devices. build-pydantic-core.sh rewritten as a multi-ABI builder:
selects targets via args, sets RUSTFLAGS=-C link-arg=-Wl,--no-as-needed
-C link-arg=-lpython3.11 to force the symbol-resolution dependency,
uses the per-ABI sysconfigdata + libpython staged in
android/.build-cache/, prefers `py -3.11` on Windows (Git Bash's
python3.11 is an MSStore stub), uses the .cmd clang wrapper on Windows
(fixes os error 193), and verifies NEEDED via llvm-readelf after each
build. abiFilters restored to the full triple in build.gradle.kts;
multi-ABI debug APK builds cleanly (~99 MB).
2026-04-14 12:36:13 +03:00
alexei.dolgolyov 8574424fb7 feat: Android TV app embedding Python server via Chaquopy
Lint & Test / test (push) Successful in 2m10s
Adds a native Android TV application that runs the full LedGrab Python
server in-process via Chaquopy. Captures the TV box screen using the
MediaProjection API and exposes the existing web UI on the device's
local network — users configure via phone/tablet browser.

Android (new /android/ module):
- Kotlin shell: MainActivity, CaptureService (foreground service),
  ScreenCapture (MediaProjection + ImageReader), PythonBridge (Chaquopy).
- Polished Leanback-themed UI with QR code for easy web UI access.
- AGP 8.9 + Chaquopy 17 + Gradle 8.11 (avoids the AGP 8.7 thread-lock bug).
- Pre-built pydantic-core wheels for arm64-v8a, x86_64, x86 cross-compiled
  with maturin + Android NDK, linked against Chaquopy's libpython3.11.so.

Python server platform guards:
- New utils/platform.py with is_android()/is_windows()/is_linux() helpers.
- Guard every top-level import of desktop-only packages (mss, psutil,
  sounddevice, pyserial, PyAudioWPatch, etc.) with try/except ImportError.
- Android-incompatible calls gated with None-checks so the server runs on
  reduced capabilities on Android (no CPU/RAM metrics, no mss displays).
- utils/image_codec.py gains a Pillow fallback for resize + JPEG encode
  when cv2 is unavailable; all internal cv2.resize callers migrated.
- New android_entry.py start_server/stop_server invoked from Kotlin.
- get_displays API falls back to best available engine when mss fails.

New capture engines:
- MediaProjectionEngine: receives RGBA frames pushed from Kotlin through
  a thread-safe queue; caches last frame for static-screen previews.
- ScrcpyClientEngine: optional H.264 streaming via scrcpy-client library
  (priority 10, overrides the ADB-screencap engine when installed).

Frontend:
- Tab loaders previously required an apiKey; now correctly treat
  "auth disabled" as authenticated (Android has no auth by default).
- Re-trigger the active tab's loader after loadServerInfo resolves
  authRequired, since initTabs runs earlier.
- Add i18n keys for the demo / mediaprojection / scrcpy_client engines.

Docs:
- TODO.md: follow-ups for multi-ABI wheel rebuilds, CI pipeline, USB
  serial LED controllers, root-only capture, perf metrics abstraction.
- CLAUDE.md: Android dependency sync policy (pip --exclude doesn't exist).

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-14 03:11:43 +03:00
alexei.dolgolyov a0b65e3fcb refactor: move build scripts to build/ directory
Lint & Test / test (push) Successful in 2m21s
Declutters the repo root by consolidating build-common.sh,
build-dist.sh, build-dist-windows.sh, build-dist.ps1, and installer.nsi
into build/. Updates all path references in CI workflows, NSIS installer,
and documentation.
2026-04-12 23:16:40 +03:00
alexei.dolgolyov 02cd9d519c refactor: rename project to LedGrab, split HA integration into separate repo
Lint & Test / test (push) Successful in 1m56s
- Rename Python package: wled_controller -> ledgrab
- Rename env var prefix: WLED_ -> LEDGRAB_ (with auto-migration for old vars)
- Rename localStorage key: wled_api_key -> ledgrab_api_key (with migration)
- Rename HA integration domain: wled_screen_controller -> ledgrab
- Update all imports, build scripts, Docker, installer, config, docs
- Remove HA integration (moved to ledgrab-haos-integration repo)
- Remove hacs.json (belongs in HA repo now)
- Add startup warning for users with old WLED_ env vars
- All tests pass (715/715), ruff clean, tsc clean, frontend builds
2026-04-12 22:45:28 +03:00
alexei.dolgolyov 38f73badbf fix: register pattern-templates API route; add responsive toolbar overflow menu
Lint & Test / test (push) Successful in 2m19s
Pattern templates route existed but was never wired into the router,
dependencies, or database table allowlist — causing 404 on graph tab load.

Graph toolbar now collapses secondary actions into a "more" overflow menu
on viewports narrower than 700px. Primary controls (fit, zoom, add) stay
visible; search, filter, panels, undo/redo, relayout, fullscreen, and
help move into the dropdown.
2026-04-12 21:22:50 +03:00
alexei.dolgolyov e678e5590a docs: update TODO and frontend context docs
Lint & Test / test (push) Successful in 1m54s
Replace TODO-css-improvements.md with TODO-release.md. Add EntityPalette
and IconSelect patterns for dynamic entity lists to frontend context.
2026-04-12 20:55:58 +03:00
alexei.dolgolyov 83ceaeda9d fix: HA Light Target cards flickering on every poll cycle
Lint & Test / test (push) Has been cancelled
Use stable placeholder values in card HTML for volatile metrics (fps,
uptime, HA status, entity swatches) so CardSection.reconcile() skips
unchanged cards. Actual values are patched in-place via
patchHALightTargetMetrics() — same pattern LED target cards already use.
2026-04-12 20:55:30 +03:00
alexei.dolgolyov d3cd48e7a7 fix: EntitySelect not showing selected value in weather/processed CSS editors
Lint & Test / test (push) Successful in 1m46s
After _populateWeatherSourceDropdown() and _populateProcessedSelectors()
create new EntitySelect instances, the subsequent .value = assignment on
the native <select> doesn't trigger _syncTrigger(), so the trigger
button still shows "—". Call .refresh() after setting the value.
2026-04-12 20:47:33 +03:00
alexei.dolgolyov cc9900d801 feat: support nesting for composite color strip sources
Lint & Test / test (push) Successful in 2m17s
Allow composite sources to reference other composite/mapped sources as
layers. Adds cycle detection (via transitive dependency graph walk),
depth limiting (MAX_COMPOSITE_DEPTH=4), and a runtime safety net in the
stream manager. Frontend layer dropdown now shows all source types
except the source being edited.

17 new tests covering cycles, depth limits, and valid nesting — all
715 tests passing.
2026-04-12 20:41:15 +03:00
alexei.dolgolyov 4940007e54 feat: add Group device type for combining multiple devices
Lint & Test / test (push) Successful in 2m19s
Introduces a new "group" device type that aggregates multiple physical
(or nested group) devices into one virtual device. Supports two modes:
- Sequence: LEDs concatenated end-to-end (led_count = sum of children)
- Independent: full pixel array resampled to each child independently

Includes cycle detection (DFS) to prevent circular group references,
delete protection for devices referenced by groups, recursive LED count
resolution for nested groups, and reorder controls (move up/down) for
child devices in the UI.

Backend: Device model, API schemas, GroupLEDClient, GroupDeviceProvider,
route validation, processing pipeline integration.
Frontend: type picker, child device picker with reorder, mode selector,
i18n (en/ru/zh), layers icon, CSS for group child rows.
Tests: 20 unit tests for cycle detection, LED count resolution, and
GroupLEDClient (sequence slicing, independent resampling, cleanup).
2026-04-11 02:26:56 +03:00
alexei.dolgolyov 92585e7c19 fix(build): bundle bettercam/dxcam/windows-capture in installer
Lint & Test / test (push) Successful in 2m5s
The Windows installer was only shipping mss as a screen-capture
backend, so EngineRegistry.get_available_engines() reported just
['mss', 'camera', 'demo'] on installed builds. Picture sources
configured to use bettercam/dxcam/wgc were rejected at the
test/ws handshake with HTTP 403 (close-before-accept on
"Engine '<x>' not available").

Add the three Windows screen-capture wheels to WIN_DEPS so the
installer build matches a 'pip install -e .' dev environment.
2026-04-08 23:15:10 +03:00
alexei.dolgolyov 0e09eaf43b fix(launcher): set TCL_LIBRARY/TK_LIBRARY for embedded Python
Embedded Python ships with tcl8.6/ and tk8.6/ next to python.exe, but
Tcl's auto-detection searches <exe>/../lib/tcl8.6 — a path that doesn't
exist in our layout. Without these env vars, tkinter.Tk() raises
"Can't find a usable init.tcl", breaking the screen overlay (and any
other tk-based UI) on installed builds.
2026-04-08 23:14:58 +03:00
alexei.dolgolyov adfc39f9d1 chore: release v0.3.0
Build Release / create-release (push) Successful in 3s
Build Release / build-linux (push) Successful in 1m23s
Build Release / build-docker (push) Successful in 2m19s
Lint & Test / test (push) Successful in 3m10s
Build Release / build-windows (push) Successful in 3m29s
2026-04-08 12:41:28 +03:00
alexei.dolgolyov d037a2e929 fix(tray): replace tkinter messagebox with Win32 MessageBoxW
Lint & Test / test (push) Successful in 2m3s
The packaged embedded Python distribution does not ship the tcl/tk
runtime, so tkinter.messagebox.askyesno crashed with 'Can't find a
usable init.tcl' when the user clicked Shutdown or Restart in the
tray menu. Use ctypes + user32.MessageBoxW instead — no tcl/tk,
no extra dependencies.
2026-04-08 12:16:32 +03:00
alexei.dolgolyov fc8ee34369 fix(launcher): start-hidden.vbs must be ASCII + CRLF, use python.exe
Lint & Test / test (push) Successful in 1m40s
Three separate bugs in the VBS launcher wedged together:

1. The previous fix added a UTF-8 em-dash in a comment. wscript.exe
   on Windows refused to execute the file with "Execution of the
   Windows Script Host failed. (Not enough memory resources are
   available to complete this operation.)" — a misleading error that
   actually means "I could not parse this file as ANSI VBScript".
   Fix: keep the file pure ASCII, convert to CRLF.

2. The launcher was invoking pythonw.exe. WshShell.Run spawning
   pythonw.exe inside the wscript host exited immediately (no process,
   no log). python.exe with WindowStyle=0 works reliably and matches
   the pattern used by the Media Server sibling app's VBS launcher,
   which has been running on this machine without issue.

3. The env vars (PYTHONPATH, WLED_CONFIG_PATH) must be set before the
   child process spawns, otherwise config.py falls back to the CWD
   default path that does not exist at install time.
2026-04-08 00:15:49 +03:00
alexei.dolgolyov e262a8b004 fix(launcher): set PYTHONPATH and WLED_CONFIG_PATH in start-hidden.vbs
Lint & Test / test (push) Successful in 2m1s
The VBS launcher (used by Start Menu, desktop, and autostart shortcuts
created by the NSIS installer) ran pythonw.exe without setting any env
vars. LedGrab.bat sets PYTHONPATH and WLED_CONFIG_PATH; the VBS did not.

With CWD set to the install root, config.py fell through to its default
lookup (./config/default_config.yaml), which does not exist there — the
real file is at app/config/default_config.yaml. The server silently ran
with built-in defaults on every shortcut launch: no devices, wrong data
dir, nothing persisted where the user expected.

The fix uses WshShell.Environment("Process") to set env vars on the
current VBS process, which child processes spawned via .Run inherit.
Kept CurrentDirectory = appRoot to preserve prior behavior for anyone
depending on CWD-relative paths inside the app.
2026-04-08 00:02:56 +03:00
alexei.dolgolyov d4ffe2e985 refactor: drop packaging dependency, inline version parsing
Lint & Test / test (push) Successful in 3m9s
The only user of 'packaging' was version_check.py — two small functions
(normalize_version, is_newer) that just need to parse "1.2.3-alpha.1"
and compare PEP 440-style versions. That's well within stdlib reach.

- Inline a NamedTuple-based Version with kind/pre_num ordering
  (dev < alpha < beta < rc < release), same regex-normalized format
- Define a local InvalidVersion exception
- Remove packaging>=23.0 from pyproject.toml dependencies

Why now: the Windows cross-build uses a hard-coded DEPS array in
build-dist-windows.sh, which was never updated when 'packaging' was
added on March 25. Result: importable from pip-installed dev envs,
missing from the portable installer — tray icon appeared but uvicorn
died with ModuleNotFoundError: No module named 'packaging'.

Removing the dep entirely is cleaner than adding one more hard-coded
entry to the Windows DEPS list. Tests (678 passing) and a manual test
matrix covering dev/alpha/beta/rc/release ordering all pass.
2026-04-07 23:54:27 +03:00
alexei.dolgolyov feb91ad281 fix(build): fix shell syntax error in smoke_test_imports heredoc
Lint & Test / test (push) Successful in 2m6s
The 'cmd <<EOF || { ... }' pattern confuses bash's parser — the closing
brace of the inline error block collides with the function's closing
brace. Rewrote to capture the python script into a local var via
$(cat <<EOF), then run it with -c and a plain 'if !' guard.
2026-04-07 23:41:42 +03:00
alexei.dolgolyov 17c5c02993 fix(build): keep .py sources + make smoke test skip uninstalled modules
Lint & Test / test (push) Successful in 1m36s
- compile_and_strip_sources: stop deleting .py files after compileall.
  OpenCV's loader does literal file I/O on cv2/config.py (not a Python
  import), so stripping it breaks `import cv2` with "missing
  configuration file: ['config.py']". Other packages may do similar
  file-based introspection tricks — the ~30% size win isn't worth
  playing whack-a-mole with broken installers. We already hit this
  with numpy.linalg and zeroconf._services; enough incidents.
- smoke_test_imports: only assert importability for modules whose
  top-level dir actually exists in site-packages. Pillow for example
  is a Windows-only dep, and was failing the Linux build spuriously.
  Rewrote as a heredoc for readability.
2026-04-07 23:37:54 +03:00
alexei.dolgolyov fd6776aeac fix(build): stop stripping zeroconf/_services + add import smoke test
Lint & Test / test (push) Successful in 2m39s
- build-common.sh: remove zeroconf/_services from the strip list.
  zeroconf's compiled Cython _listener.pyd imports from _services
  internally, so stripping it broke `import zeroconf` at runtime with
  ModuleNotFoundError — same class of bug as the numpy.linalg strip.
- build-common.sh: add smoke_test_imports() that imports every top-level
  dependency against the stripped site-packages. Catches "we stripped
  something that was actually needed" regressions at build time instead
  of on a user's machine after install.
- build-dist.sh: wire smoke test into the Linux flow (runs real imports).
- build-dist-windows.sh: cross-build can't load win_amd64 .pyd files with
  the host python, so instead verify that the known-required submodule
  dirs (numpy.linalg/lib/matrixlib/ma, zeroconf._services) exist after
  cleanup. Fails loud if any future strip-rule removes them.
2026-04-07 23:32:50 +03:00
alexei.dolgolyov 9f34ffb0a0 fix(build): stop stripping numpy.lib/linalg from site-packages
Lint & Test / test (push) Successful in 2m57s
numpy's own __init__.py imports lib and matrixlib (which in turn imports
numpy.linalg via defmatrix.py). Removing any of these submodules to save
dist size makes `import numpy` raise ModuleNotFoundError on the target,
which cascades into every wled_controller import.

Symptom: v0.0.0.dev0 Windows installer showed a tray icon but uvicorn
died silently in its background thread and port 8080 never listened —
browser got ERR_CONNECTION_REFUSED.

Keep stripping: polynomial, distutils, f2py, typing, _pyinstaller.
These are genuinely unused by numpy's own import chain.
2026-04-07 23:17:12 +03:00
alexei.dolgolyov b5842e6424 fix(build): normalize non-PEP440 versions, fix .py/compileall ordering, wipe NSIS payload dirs
Lint & Test / test (push) Successful in 3m18s
- build-common.sh: detect_version() normalizes non-PEP440 labels (e.g. 'dev',
  'nightly') to 0.0.0.dev0 so stamping pyproject.toml doesn't break pip install.
- build-common.sh: split .py deletion out of cleanup_site_packages into a new
  compile_and_strip_sources() that runs 'compileall -b' FIRST, then removes
  sources. compileall now fails loud instead of silently no-op'ing.
- build-dist.sh: add missing compile_and_strip_sources call on Linux site-packages
  (previous tarballs shipped empty packages with no .py and no .pyc).
- build-dist-windows.sh: reorder so compile_and_strip_sources runs right after
  cleanup_site_packages, not after .py files have already been deleted.
- installer.nsi: RMDir /r payload dirs (python/, app/, scripts/) and delete
  LedGrab.bat at the top of SecCore before File /r. NSIS File /r MERGES into
  existing dirs, so upgrades left half-old/half-new state that surfaced as
  'version mismatch' or duplicate-package ImportErrors. data/ and logs/ remain
  untouched to preserve user config.
2026-04-07 23:04:38 +03:00
alexei.dolgolyov 7a9c368448 refactor: split color-strips.ts into focused modules under color-strips/ folder
Lint & Test / test (push) Successful in 2m6s
Monolithic 3060-line color-strips.ts split into 11 modules:
- index.ts (core orchestrator, modal, type switching, editor, save, CRUD)
- cards.ts (card rendering for all source types)
- game-event.ts (game event mappings, presets, UI)
- gradient.ts (gradient entity modal + CRUD)
- audio.ts (audio viz widgets, load/reset state)
- math-wave.ts (wave layers + waveform selects)
- mapped.ts (mapped zone helpers)
- color-cycle.ts (color cycle add/remove/render)
- composite.ts, notification.ts, test.ts (previously extracted, moved into folder)
2026-04-05 12:54:15 +03:00
alexei.dolgolyov ce53ca6872 feat: add card glare effect to dashboard and perf chart cards
Lint & Test / test (push) Has been cancelled
Extend cursor-tracking spotlight to .dashboard-target and
.perf-chart-card elements with a smaller 100px radius suited
for compact cards.
2026-04-05 12:23:39 +03:00
alexei.dolgolyov b04978af58 feat: add music sync viz modes and auto_gain audio filter
Lint & Test / test (push) Has been cancelled
Add 4 new audio visualization modes powered by MusicAnalyzer:
- pulse_on_beat: BPM-synced pulsing with smooth beat phase
- energy_gradient: bass/mid/treble mapped to scrolling gradient
- spectrum_bands: three VU zones for frequency bands
- strobe_on_drop: state-driven strobe on detected musical drops

MusicAnalyzer provides BPM estimation (median IBI), beat phase tracking,
asymmetric energy envelope, 3-band frequency splitting, and drop
detection state machine (idle/buildup/drop/recovery).

Add auto_gain audio filter for automatic level normalization via rolling
peak tracking with configurable target level and response time.

Deprecate auto_gain on Audio Value Source (use the filter instead).
2026-04-05 01:40:34 +03:00
alexei.dolgolyov 6e8b159126 fix: weather CSS card shows empty source name after hard refresh
Lint & Test / test (push) Has been cancelled
Weather sources cache was not fetched during streams tab load, so the
card renderer could not resolve weather source names. Also widened
lat/lon inputs from 80px to 100px to fit longer coordinates.
2026-04-05 00:49:28 +03:00
alexei.dolgolyov ace24715c8 feat: add math_wave color strip source type
Lint & Test / test (push) Has been cancelled
Mathematical wave generator that produces per-LED colors from
configurable waveform layers (sine, triangle, sawtooth, square) with
superposition, mapped through a gradient palette. Supports sync clocks,
bindable speed, and up to 8 wave layers.

- Storage model with wave validation and apply_update
- Numpy-vectorized stream with gradient LUT color mapping
- API schemas (create/update/response) and route registration
- Frontend editor with dynamic wave layer rows, gradient picker,
  speed widget, and IconSelect waveform selectors
- i18n in en/ru/zh
2026-04-05 00:41:07 +03:00
alexei.dolgolyov edc6d27e2e fix: replace HA test icon with refresh, make automation rules collapsible
Lint & Test / test (push) Has been cancelled
Use refresh icon instead of flask for HA test connection buttons.
Add collapse/expand chevron to automation rule rows (collapsed by default).
2026-04-04 21:28:51 +03:00
alexei.dolgolyov b7da4ab6b5 feat: add Integrations tab and responsive icon-only tabs
Lint & Test / test (push) Successful in 1m48s
Move HA sources, weather sources, game integration, and MQTT settings
into a dedicated Integrations top-level tab with tab-registry pattern.
Collapse tab labels to icon-only at narrow desktop widths (<=1100px)
to prevent toolbar overflow.
2026-04-02 15:29:38 +03:00
alexei.dolgolyov 99460a8043 fix: make pystray a core dependency on Windows instead of optional extra
Lint & Test / test (push) Successful in 2m5s
The tray icon should always appear on Windows, not only when installed
with the [tray] extra.
2026-04-02 14:22:18 +03:00
alexei.dolgolyov 89990f8d63 chore: remove processed-audio-sources plan files 2026-04-02 13:42:37 +03:00
alexei.dolgolyov 0cc0aaa411 feat: processed audio sources with composable filter pipeline
Replace hardcoded MonoAudioSource/BandExtractAudioSource with a
composable ProcessedAudioSource + AudioProcessingTemplate + AudioFilter
system. 11 audio filters: channel extract, band extract, peak hold,
gain, noise gate, envelope follower, spectral smoothing, compressor,
inverter, beat gate, delay. Full frontend UI with filter editor,
tree navigation, and i18n support.
2026-04-02 13:42:18 +03:00
alexei.dolgolyov af2c89c8df fix: audio tree structure, filter i18n, and IconSelect for filter options
Restructure audio tree nav into Capture (Sources + Engine Templates)
and Processed (Sources + Filter Templates) subgroups.
Add missing i18n description keys for all 11 audio filters.
Wrap plain select filter options with IconSelect grids.
2026-04-02 13:37:50 +03:00
alexei.dolgolyov d04192ffb7 fix: add reference check before deleting audio processing template
Prevent deleting templates that are still referenced by
ProcessedAudioSource entities. Returns 400 with source names.
2026-04-01 23:28:58 +03:00
alexei.dolgolyov 992495e2e4 fix: isolate tests from production database
Tests that imported wled_controller.main at module level caused the real
production database (data/ledgrab.db) to be opened before test fixtures
could patch the config. This led to silent data loss.

Patch the global config singleton at conftest module level (before any
test imports main.py) to redirect all DB access to a temp directory.
2026-04-01 19:01:56 +03:00
alexei.dolgolyov 6b0e4e5539 feat(processed-audio-sources): phase 8 - frontend design consistency review
Fix audio source modal error class (modal-error), use Modal.showError(),
reorder audio source card description, remove redundant APT filter count
badge, clean up unused imports in audio-sources.ts.
2026-03-31 23:11:17 +03:00
alexei.dolgolyov ce1f4847f3 feat(processed-audio-sources): phase 7 - testing and polish
Fix test_list_filters test (filter_id field name mismatch).
Add tests for audio filters, template store, and source store.
All 678 tests pass, ruff clean, tsc clean, esbuild clean.
No dead code remaining from old source types.
2026-03-31 22:50:02 +03:00
alexei.dolgolyov 1ce0dc6c61 feat(processed-audio-sources): phase 6 - frontend source type cleanup
Rewrite audio source editor modal for capture/processed types only.
Remove old multichannel/mono/band_extract HTML sections and i18n keys.
Clean up legacy DOM section null-checks in audio-sources.ts.
2026-03-31 19:40:37 +03:00
alexei.dolgolyov 553463935e feat(processed-audio-sources): phase 5 - frontend audio processing templates
Add Audio Processing Templates management UI to Streams tab:
- Template editor modal with filter list via FilterListManager
- CardSection with reconciliation for template cards
- DataCache instances for templates and audio filter defs
- Audio filter icon mappings in filter-list.ts
- i18n keys in en/ru/zh locales
2026-03-31 19:32:17 +03:00
alexei.dolgolyov ab43578049 feat(processed-audio-sources): phase 4 - runtime filter integration
Add AudioFilterPipeline for chained filter execution on AudioAnalysis.
Wire filter pipelines into AudioColorStripStream, AudioValueStream,
and WebSocket test endpoint. Add hot-update support via
ProcessorManager.refresh_audio_filter_pipelines(). Thread
AudioProcessingTemplateStore through dependency injection hierarchy.
2026-03-31 19:15:29 +03:00
alexei.dolgolyov 353c090b42 feat(processed-audio-sources): phase 3 - processed audio source model
Replace MultichannelAudioSource with CaptureAudioSource, add
ProcessedAudioSource (audio_source_id + audio_processing_template_id),
remove MonoAudioSource and BandExtractAudioSource entirely.
Update store resolution to walk processed chains collecting template IDs.
Update all API schemas, routes, and frontend references.
2026-03-31 19:01:46 +03:00
alexei.dolgolyov eb94066386 feat(processed-audio-sources): phase 2 - implement 11 audio filters
Add all audio filters that transform AudioAnalysis data:
- Channel Extract, Band Extract (migration from old source types)
- Peak Hold, Gain, Noise Gate, Envelope Follower
- Spectral Smoothing, Compressor, Inverter, Beat Gate, Delay
All registered via AudioFilterRegistry with option schemas.
2026-03-31 18:43:36 +03:00
alexei.dolgolyov 86a9d344e6 feat(processed-audio-sources): phase 1 - audio filter framework
Add the foundation for audio processing filters, mirroring the existing
picture filter/postprocessing template system:
- AudioFilter base class, AudioFilterRegistry, AudioFilterOptionDef
- AudioProcessingTemplate dataclass + SQLite-backed store
- audio_filter_template meta-filter with recursive resolution
- Full REST API: CRUD templates + filter registry discovery
- Dependency injection wired in dependencies.py and main.py
2026-03-31 17:35:39 +02:00
alexei.dolgolyov c59107c7c7 feat: refactor MQTT from global config to multi-instance entity model
Lint & Test / test (push) Successful in 1m32s
MQTT broker connections are now managed as entities (like HA sources)
instead of a single global config. Each MQTTSource gets its own
runtime with auto-reconnect, ref-counted via MQTTManager.

Backend:
- MQTTSource dataclass + MQTTSourceStore (SQLite)
- MQTTRuntime (per-broker connection, refactored from MQTTService)
- MQTTManager (ref-counted pool, same pattern as HAManager)
- CRUD API at /api/v1/mqtt/sources + test + status endpoints
- MQTTRule gains mqtt_source_id field for source selection
- Automation engine acquires/releases MQTT runtimes automatically
- Legacy MQTTService kept for backward compat during transition

Frontend:
- MQTT source cards in Streams > Integrations tab
- Create/edit modal with test button
- Dashboard integration cards with health-dot indicators
- Removed MQTT tab from settings modal
2026-03-31 18:02:19 +03:00
alexei.dolgolyov e7c9a568dc feat: HA source cards use health-dot indicators
Lint & Test / test (push) Successful in 1m39s
Replace unstyled status-dot classes with the existing health-dot
pattern (green/red glow) matching LED device cards. Tooltip shows
connection status and entity count.
2026-03-31 16:05:01 +03:00
alexei.dolgolyov b36ddfd395 Merge branch 'feature/game-integration'
Lint & Test / test (push) Successful in 1m49s
Game integration system: receive real-time events from games
(CS2, Dota 2, LoL, etc.) and drive LED effects via color strip
streams and value source bindings.
2026-03-31 14:23:38 +03:00
alexei.dolgolyov 492bdb95e3 feat: game integration system
Receive real-time events from games (CS2, Dota 2, LoL, etc.) and drive
LED effects through the existing color strip and value source pipelines.

Core:
- GameEventBus (thread-safe pub/sub) with standardized 23-type event vocabulary
- GameAdapter ABC + AdapterRegistry + MappingAdapter (YAML-driven)
- Built-in adapters: CS2 GSI, Dota 2 GSI, LoL Live Client, Generic Webhook
- Community YAML adapters: Minecraft, Valorant, Rocket League
- GameEventColorStripStream with 5 effects (flash/pulse/sweep/color_shift/breathing)
- GameEventValueSource with EMA smoothing and timeout
- 4 built-in effect presets (FPS Combat, MOBA Health, Racing, Generic Alert)
- Auto-setup for Valve GSI games (Steam path detection, cfg file writing)
- Demo capture engine exposed to non-demo mode

Frontend:
- Game tab in Streams tree navigation with integration cards
- Game integration editor modal with adapter picker, config fields, event mappings
- game_event source type in CSS and ValueSource editors
- Setup instructions overlay (markdown rendered)
- Live event monitor and connection test

API:
- Full CRUD for game integrations
- Event ingestion endpoint (adapter-level auth)
- Adapter metadata, presets, auto-setup, status/diagnostics endpoints
2026-03-31 13:17:52 +03:00
alexei.dolgolyov b6713be390 feat: system_metrics value source type
Lint & Test / test (push) Successful in 1m32s
New value source that monitors host hardware via psutil/pynvml:
cpu_load, cpu_temp, gpu_load, gpu_temp, ram_usage, disk_usage,
network_rx, network_tx, battery_level, fan_speed.

Each metric normalizes to 0.0-1.0 with configurable ranges, poll
interval, EMA smoothing, and sensor_label for multi-sensor systems.
Conditional editor fields show/hide based on selected metric.

Also fixes: WS test crash when raw_value streams lack _min_ha attr,
toast timer overlap on rapid calls, SW cache bump to v34.
2026-03-30 18:22:58 +03:00
alexei.dolgolyov db5008aaeb feat: system theme option + fix toast timer overlap
Lint & Test / test (push) Successful in 1m27s
Add third theme mode (system) that follows OS prefers-color-scheme.
Theme button cycles dark → light → system with monitor icon.
Listens for OS preference changes in real time when in system mode.

Fix showToast clearing previous timer so rapid calls don't cause
the toast to disappear early.
2026-03-30 13:55:38 +03:00
alexei.dolgolyov 4b7a8d75f4 feat: value source card crosslinks + gradient_map test shows input value
Lint & Test / test (push) Successful in 1m23s
Add navigateToCard crosslinks for ha_entity (→ HA source), gradient_map
(→ input value source + gradient entity), and css_extract (→ color strip).
Gradient map test now charts the input interpolation factor instead of
output luminance, making the 0–1 chart meaningful.
2026-03-30 03:23:05 +03:00
alexei.dolgolyov f6c25cd15f feat: color value source test visualization
Lint & Test / test (push) Successful in 1m6s
WS now sends color RGB data for color-type streams. Test modal renders
a color history swatch strip below the chart, colors the chart line and
fill area with the current output color, and shows rgb() in the stats.
Works for static_color, animated_color, and adaptive_time_color sources.
2026-03-30 03:12:57 +03:00
alexei.dolgolyov 0a8737157c feat: HA value source test — raw value axis + behavior IconSelect
Lint & Test / test (push) Successful in 1m9s
- WS sends raw_value and raw_range for HA entity streams
- Canvas draws right-side Y-axis labels: configured min/max range
  and current raw HA value (green, positioned at correct Y)
- Replace plain <select> for scene behavior with IconSelect (moon/sun)
2026-03-30 03:06:44 +03:00
alexei.dolgolyov 11d5d6b5e1 fix: device card header layout — URL badge overflow and hide button gap
Lint & Test / test (push) Successful in 1m24s
Move URL badge from card-title to card-subtitle row to prevent
overlap with top-right action buttons. Widen card-header padding-right
to 60px for 2-button clearance. Reorder hide button to first position
in top-actions so power and trash stay visually adjacent.
2026-03-30 02:05:45 +03:00
alexei.dolgolyov 384362ccf1 feat: new value source types (HA entity, gradient map, strip extract) + UI fixes
Lint & Test / test (push) Successful in 1m27s
New value source types:
- ha_entity: reads numeric values from HA entity state/attribute, normalizes
  via min/max range, applies EMA smoothing. EntitySelect for HA connection
  and entity selection with live entity list fetching.
- gradient_map: maps a float value source (0-1) through a gradient entity.
  EntitySelect for both input source and gradient with inline previews.
- css_extract: extracts single color by averaging LED range from a color
  strip source. EntitySelect for source selection.

Value source type picker:
- Filter tabs (All / Numeric / Color) above the icon grid
- showTypePicker extended with filterTabs + onFilterChange support

Palette selectors converted to EntitySelect:
- Effect palette, gradient preset, and audio palette selectors now use
  command-palette style EntitySelect with gradient strip previews

Tab indicator fixes:
- Icon now updates on tab switch (was passing no args to updateTabIndicator)
- Visible with any background effect active, not just Noise Field
- Noise Field is the default background effect for new users

Dashboard section collapse fix:
- Split header into clickable toggle (chevron+label) and non-clickable
  actions area — buttons no longer trigger collapse/expand

Discriminated union fix (422 errors):
- source_type/target_type now always included in update payloads for:
  CSS editor, LED target, HA light target, simple calibration,
  advanced calibration
2026-03-29 20:38:22 +03:00
alexei.dolgolyov ea812bb4d5 feat: check if port is busy before starting the server
Lint & Test / test (push) Successful in 1m16s
2026-03-29 14:21:35 +03:00
alexei.dolgolyov a9e6e8cb82 fix: KC color strip test preview — use LiveStreamManager instead of raw engine
Lint & Test / test (push) Successful in 1m28s
BetterCam (DXGI) only supports one capture session per display, so creating
a second engine for the KC test produced no frames when a target was already
running. Now uses LiveStreamManager.acquire() which ref-counts and reuses
the running capture.

Also removed double postprocessing — ProcessedLiveStream already applies
filters, so re-applying them in the KC test halved the resolution (960x400
→ 240x100).
2026-03-29 14:11:01 +03:00
alexei.dolgolyov 78ce6c84d7 fix: composite layer opacity/brightness widgets + CSS layout
Lint & Test / test (push) Successful in 1m7s
- Fix opacity widget empty space (CSS selector .composite-layer-opacity → .composite-layer-opacity-container)
- Replace brightness select dropdown with BindableScalarWidget (slider + VS toggle)
- Legacy brightness_source_id auto-converted to BindableFloat on load
- Add .composite-layer-brightness-container CSS rule
2026-03-29 00:42:42 +03:00
alexei.dolgolyov 8a17bb5caa feat: BindableFloat — universal value source binding for all scalar properties
Lint & Test / test (push) Successful in 1m20s
Introduce BindableFloat abstraction that allows any numeric property to be
either a static value or dynamically driven by a ValueSource. Backward-compatible
serialization: plain float when unbound, {value, source_id} dict when bound.

Backend:
- storage/bindable.py — BindableFloat dataclass + bfloat() helper
- 25+ scalar properties converted across all entity types
- Runtime VS acquisition in ColorStripStreamManager for CSS bindings
- All stream hot loops use self.resolve() for live values
- KeyColorsColorStripStream now inherits ColorStripStream

Frontend:
- BindableScalarWidget (slider + VS picker toggle) for all editors
- TypeScript BindableFloat type + helpers
- Graph editor edges for all bindable properties
- Audio source channel IconSelect grid

Fixes: daylight longitude, candlelight wind_strength/candle_type from_dict
2026-03-29 00:33:24 +03:00
alexei.dolgolyov 5f70302263 feat: use custom app icon for shortcuts and installer
Lint & Test / test (push) Successful in 1m15s
- Generate icon.ico from icon-512.png (16-256px sizes)
- Set MUI_ICON/MUI_UNICON for installer wizard
- Point all shortcuts to icon.ico instead of pythonw.exe
- Add DisplayIcon registry entry for Add/Remove Programs
2026-03-28 18:41:12 +03:00
alexei.dolgolyov 40751fecb7 feat: HA light target live color preview — per-entity swatches via WebSocket
Lint & Test / test (push) Successful in 1m24s
- Cache per-entity colors in HALightTargetProcessor._update_lights()
- Broadcast colors_update to WS clients at target's update_rate
- WS endpoint: /api/v1/output-targets/{target_id}/ha-light/ws
- Frontend: connect WS when target runs, update swatch colors live
- Card shows colored boxes per mapped entity with entity name labels
2026-03-28 18:28:16 +03:00
alexei.dolgolyov 381ee75371 fix: HA light target — brightness source, transition=0, dashboard type label
Lint & Test / test (push) Successful in 1m13s
- Add brightness_value_source_id to HALightOutputTarget model, to_dict,
  from_dict, update_fields, register_with_manager, API response
- Wire value stream in HALightTargetProcessor: acquire/release on
  start/stop, multiply brightness in _update_lights loop
- Fix transition=0 not saving (parseFloat("0") || 0.5 was falsy)
- Fix dashboard showing "Key Colors" for HA targets — now "Home Assistant"
- Fix dashboard FPS showing 0/2 — HA targets show target/target
- Add CSS source subtitle to HA target dashboard cards
2026-03-28 16:03:06 +03:00
alexei.dolgolyov 3e6760f726 refactor: key colors targets → CSS source type, HA target improvements
Lint & Test / test (push) Successful in 1m26s
Key Colors refactor:
- New `key_colors` CSS source type with inline rectangles
- KeyColorsColorStripStream: extracts N colors from screen regions
- CSS editor: EntitySelect for picture source, IconSelect for color mode
- Configure Regions button on card opens pattern canvas editor
- Live WS preview at 5 FPS with rectangle overlay + color swatches
- Removed KC target type, pattern template entity, and related API routes
- Removed KC/pattern template sections from Targets tab

HA light target improvements:
- Update rate, transition, mappings, brightness VS now editable via PUT
- Card crosslinks for HA source, CSS source, brightness VS
- HA connection status icon, text metrics (Hz, uptime)
- Brightness value source selector in editor
2026-03-28 15:28:22 +03:00
alexei.dolgolyov 89d1b13854 fix: rename HA Lights → Home Assistant, HA Light Targets → Light Targets
Lint & Test / test (push) Successful in 1m38s
2026-03-28 11:30:40 +03:00
alexei.dolgolyov 324a308805 feat: entity picker for HA light mapping — searchable EntitySelect for light entities
Lint & Test / test (push) Failing after 11m19s
Replaces plain text input with EntitySelect dropdown that fetches
available light.* entities from the selected HA source. Changing the
HA source refreshes the entity list across all mapping rows.
2026-03-28 00:35:42 +03:00
alexei.dolgolyov cb9289f01f feat: HA light output targets — cast LED colors to Home Assistant lights
Lint & Test / test (push) Has been cancelled
New output target type `ha_light` that sends averaged LED colors to HA
light entities via WebSocket service calls (light.turn_on/turn_off):

Backend:
- HARuntime.call_service(): fire-and-forget WS service calls
- HALightOutputTarget: data model with light mappings, update rate, transition
- HALightTargetProcessor: processing loop with delta detection, rate limiting
- ProcessorManager.add_ha_light_target(): registration
- API schemas/routes updated for ha_light target type

Frontend:
- HA Light Targets section in Targets tab tree nav
- Modal editor: HA source picker, CSS source picker, light entity mappings
- Target cards with start/stop/clone/edit actions
- i18n keys for all new UI strings
2026-03-28 00:08:49 +03:00
alexei.dolgolyov fb98e6e2b8 ci: add manual build workflow for testing artifacts
Lint & Test / test (push) Has been cancelled
workflow_dispatch-triggered build.yml that produces Windows
installer/portable and Linux tarball as CI artifacts without
creating a release. Trigger from Gitea UI → Actions → Run.
2026-03-27 23:41:22 +03:00
alexei.dolgolyov 3c2efd5e4a refactor: move Weather and Home Assistant sources to Integrations tree group
Separates external service connections from utility entities in the
Streams tree navigation for clearer organization.
2026-03-27 23:21:40 +03:00
alexei.dolgolyov 2153dde4b7 feat: Home Assistant integration — WebSocket connection, automation conditions, UI
Add full Home Assistant integration via WebSocket API:
- HARuntime: persistent WebSocket client with auth, auto-reconnect, entity state cache
- HAManager: ref-counted runtime pool (like WeatherManager)
- HomeAssistantCondition: new automation trigger type matching entity states
- REST API: CRUD for HA sources + /test, /entities, /status endpoints
- /api/v1/system/integrations-status: combined MQTT + HA dashboard indicators
- Frontend: HA Sources tab in Streams, condition type in automation editor
- Modal editor with host, token, SSL, entity filters
- websockets>=13.0 dependency added
2026-03-27 22:42:48 +03:00
alexei.dolgolyov f3d07fc47f feat: donation banner, About tab, settings UI improvements
Lint & Test / test (push) Has been cancelled
- Dismissible donation/open-source banner after 3+ sessions (30-day snooze)
- New About tab in Settings: version, repo link, license info
- Centralize project URLs (REPO_URL, DONATE_URL) in __init__.py, served via /health
- Center settings tab bar, reduce tab padding for 6-tab fit
- External URL save button: icon button instead of full-width text button
- Remove redundant settings footer close button
- Footer "Source Code" link replaced with "About" opening settings
- i18n keys for en/ru/zh
2026-03-27 21:09:34 +03:00
alexei.dolgolyov f61a0206d4 feat: custom file drop zone for asset upload modal; fix review issues
Lint & Test / test (push) Successful in 1m29s
Replace plain <input type="file"> with a styled drag-and-drop zone
featuring hover/drag states, file info display, and remove button.

Fix 10 review issues: add 401 handling to upload, remove non-null
assertions, add missing i18n keys (common.remove, error.play_failed,
error.download_failed), normalize close button glyphs to &#x2715;,
i18n the dropzone aria-label, replace silent error catches with toast
notifications, use DataTransfer for cross-browser file assignment.
2026-03-26 21:43:08 +03:00
alexei.dolgolyov f345687600 chore: remove python3.11 version pin from pre-commit config
Lint & Test / test (push) Successful in 1m6s
2026-03-26 20:41:34 +03:00
alexei.dolgolyov e2e1107df7 feat: asset-based image/video sources, notification sounds, UI improvements
Lint & Test / test (push) Has been cancelled
- Replace URL-based image_source/url fields with image_asset_id/video_asset_id
  on StaticImagePictureSource and VideoCaptureSource (clean break, no migration)
- Resolve asset IDs to file paths at runtime via AssetStore.get_file_path()
- Add EntitySelect asset pickers for image/video in stream editor modal
- Add notification sound configuration (global sound + per-app overrides)
- Unify per-app color and sound overrides into single "Per-App Overrides" section
- Persist notification history between server restarts
- Add asset management system (upload, edit, delete, soft-delete)
- Replace emoji buttons with SVG icons throughout UI
- Various backend improvements: SQLite stores, auth, backup, MQTT, webhooks
2026-03-26 20:40:25 +03:00
alexei.dolgolyov c0853ce184 fix: improve command palette actions and automation condition button
Lint & Test / test (push) Successful in 1m16s
- Action items (start/stop, enable/disable) no longer close the palette
- Action items toggle state after success (Start→Stop, Enable→Disable)
- Toast z-index raised above command palette backdrop (3000→3500)
- Automation condition remove button uses ICON_TRASH SVG instead of ✕
2026-03-26 02:21:52 +03:00
alexei.dolgolyov 3e0bf8538c feat: add api_input LED interpolation; fix LED preview, FPS charts, dashboard layout
Lint & Test / test (push) Successful in 1m26s
API Input:
- Add interpolation mode (linear/nearest/none) for LED count mismatch
  between incoming data and device LED count
- New IconSelect in editor, i18n for en/ru/zh
- Mark crossfade as won't-do (client owns temporal transitions)
- Mark last-write-wins as already implemented

LED Preview:
- Fix zone-mode preview parsing composite wire format (0xFE header
  bytes were rendered as color data, garbling multi-zone previews)
- Fix _restoreLedPreviewState to handle zone-mode panels

FPS Charts:
- Seed target card charts from server metrics-history on first load
- Add fetchMetricsHistory() with 5s TTL cache shared across
  dashboard, targets, perf-charts, and graph-editor
- Fix chart padding: pass maxSamples per caller (120 for dashboard,
  30 for target cards) instead of hardcoded 120
- Fix dashboard chart empty on tab switch (always fetch server history)
- Left-pad with nulls for consistent chart width across targets

Dashboard:
- Fix metrics row alignment (grid layout with fixed column widths)
- Fix FPS label overflow into uptime column
2026-03-26 02:06:49 +03:00
alexei.dolgolyov be4c98b543 fix: show template name instead of ID in filter list and card badges
Lint & Test / test (push) Successful in 1m7s
Collapsed filter cards in the modal showed raw template IDs (e.g.
pp_cb72e227) instead of resolving select options to their labels.
Card filter chain badges now include the referenced template name.
2026-03-25 23:56:40 +03:00
alexei.dolgolyov dca2d212b1 fix: clip graph node title and subtitle to prevent overflow
Long entity names overflowed past the icon area on graph cards.
Added SVG clipPath to constrain text within the node bounds.
2026-03-25 23:56:30 +03:00
alexei.dolgolyov 53986f8d95 fix: replace emoji with SVG icons on weather and daylight cards
Weather card used  and 🌡 emoji, daylight card used 🕐 and .
Replaced with ICON_FAST_FORWARD, ICON_THERMOMETER, and ICON_CLOCK.
Added thermometer icon path.
2026-03-25 23:56:21 +03:00
alexei.dolgolyov a4a9f6f77f fix: send gradient_id instead of palette in effect transient preview
Lint & Test / test (push) Successful in 1m19s
The preview config was sending `palette` which defaults to "fire" on
the server, ignoring the user's selected gradient. Also removed the
dead fallback notification block and stale custom_palette check.
2026-03-25 23:43:33 +03:00
alexei.dolgolyov 9fcfdb8570 ci: use sparse checkout for release notes in release workflow
Only fetch RELEASE_NOTES.md instead of full repo checkout, and
simplify the file detection to a direct path check.
2026-03-25 23:43:31 +03:00
1056 changed files with 192241 additions and 49464 deletions
+248
View File
@@ -0,0 +1,248 @@
name: Build Android APK
on:
# Release tags only — building the ~100 MB APK on every master push
# burned Gitea runner minutes without producing a useful artifact.
# Use workflow_dispatch for on-demand dev builds.
push:
tags: ['v*']
workflow_dispatch:
inputs:
version:
description: 'Version label (e.g. dev, 0.3.0-test)'
required: false
default: 'dev'
jobs:
build-android:
runs-on: ubuntu-latest
env:
JAVA_VERSION: '17'
PYTHON_VERSION: '3.11'
ANDROID_CMDLINE_TOOLS_VERSION: '11076708'
ANDROID_SDK_PLATFORM: 'android-34'
ANDROID_BUILD_TOOLS: '34.0.0'
ANDROID_NDK_VERSION: '26.1.10909125'
# Surfaced at job level (not step level) so the `if: env.X != ''`
# check on the Decode step actually sees it — step-level env is
# NOT available in that step's own `if:` expression, which
# silently skipped the decode and produced debug-signed release
# APKs until it was noticed.
ANDROID_KEYSTORE_BASE64: ${{ secrets.ANDROID_KEYSTORE_BASE64 }}
steps:
- name: Checkout
uses: actions/checkout@v4
with:
fetch-depth: 0
- name: Resolve build label
id: label
run: |
REF="${{ gitea.ref_name }}"
if echo "$REF" | grep -qE '^v[0-9]'; then
LABEL="${REF#v}"
IS_RELEASE="true"
elif [ -n "${{ inputs.version }}" ]; then
LABEL="${{ inputs.version }}"
IS_RELEASE="false"
else
LABEL="dev-${{ gitea.sha }}"
IS_RELEASE="false"
fi
LABEL="${LABEL:0:40}"
echo "label=$LABEL" >> "$GITHUB_OUTPUT"
echo "is_release=$IS_RELEASE" >> "$GITHUB_OUTPUT"
echo "Build label: $LABEL (release=$IS_RELEASE)"
- name: Guard release tag against missing keystore
# Release tags MUST produce a release-signed APK, otherwise existing
# installs can't upgrade (signature mismatch). Fail loudly instead
# of silently falling back to the debug signing config.
# Runs before JDK/Python/SDK/NDK setup so a misconfigured release
# tag fails in seconds instead of after several minutes of setup.
if: ${{ steps.label.outputs.is_release == 'true' && env.ANDROID_KEYSTORE_BASE64 == '' }}
run: |
echo "::error::Release tag ${{ gitea.ref_name }} requires ANDROID_KEYSTORE_BASE64 (plus KEYSTORE_PASSWORD, KEY_ALIAS, KEY_PASSWORD) to be configured in Gitea → Settings → Secrets."
exit 1
- name: Setup JDK ${{ env.JAVA_VERSION }}
uses: actions/setup-java@v4
with:
distribution: temurin
java-version: ${{ env.JAVA_VERSION }}
- name: Setup Python ${{ env.PYTHON_VERSION }}
uses: actions/setup-python@v5
with:
python-version: ${{ env.PYTHON_VERSION }}
- name: Setup Android SDK + NDK
run: |
set -euo pipefail
SDK_ROOT="$HOME/android-sdk"
mkdir -p "$SDK_ROOT/cmdline-tools"
cd "$SDK_ROOT/cmdline-tools"
curl -sSL --retry 3 \
-o cmdline-tools.zip \
"https://dl.google.com/android/repository/commandlinetools-linux-${ANDROID_CMDLINE_TOOLS_VERSION}_latest.zip"
unzip -q cmdline-tools.zip
rm cmdline-tools.zip
mv cmdline-tools latest
export ANDROID_SDK_ROOT="$SDK_ROOT"
export ANDROID_HOME="$SDK_ROOT"
SDKMANAGER="$SDK_ROOT/cmdline-tools/latest/bin/sdkmanager"
yes | "$SDKMANAGER" --licenses > /dev/null 2>&1 || true
"$SDKMANAGER" --install \
"platform-tools" \
"platforms;${ANDROID_SDK_PLATFORM}" \
"build-tools;${ANDROID_BUILD_TOOLS}" \
"ndk;${ANDROID_NDK_VERSION}" > /dev/null
echo "ANDROID_SDK_ROOT=$SDK_ROOT" >> "$GITHUB_ENV"
echo "ANDROID_HOME=$SDK_ROOT" >> "$GITHUB_ENV"
echo "ANDROID_NDK_HOME=$SDK_ROOT/ndk/${ANDROID_NDK_VERSION}" >> "$GITHUB_ENV"
echo "$SDK_ROOT/platform-tools" >> "$GITHUB_PATH"
echo "$SDK_ROOT/cmdline-tools/latest/bin" >> "$GITHUB_PATH"
- name: Create local.properties
run: |
echo "sdk.dir=$ANDROID_SDK_ROOT" > android/local.properties
- name: Link Python source (junction equivalent)
run: |
set -euo pipefail
# Chaquopy reads Python modules from android/app/src/main/python/
# On Windows dev machines this is a directory junction; on Linux
# CI use a symlink. The parent dir is .gitignore'd (the junction
# target is the real content), so we have to create it first.
mkdir -p android/app/src/main/python
ln -sfn "$(pwd)/server/src/ledgrab" android/app/src/main/python/ledgrab
ls -la android/app/src/main/python/
# Sanity check — readlink resolves the link and the directory exists.
test -d android/app/src/main/python/ledgrab
- name: Decode signing keystore
id: keystore
if: env.ANDROID_KEYSTORE_BASE64 != ''
run: |
set -euo pipefail
mkdir -p android/keystore
echo "$ANDROID_KEYSTORE_BASE64" | base64 -d > android/keystore/release.jks
echo "path=$(pwd)/android/keystore/release.jks" >> "$GITHUB_OUTPUT"
echo "present=true" >> "$GITHUB_OUTPUT"
- name: Build APK
working-directory: android
env:
ANDROID_KEYSTORE_PATH: ${{ steps.keystore.outputs.path }}
ANDROID_KEYSTORE_PASSWORD: ${{ secrets.ANDROID_KEYSTORE_PASSWORD }}
ANDROID_KEY_ALIAS: ${{ secrets.ANDROID_KEY_ALIAS }}
ANDROID_KEY_PASSWORD: ${{ secrets.ANDROID_KEY_PASSWORD }}
run: |
chmod +x gradlew
if [ "${{ steps.keystore.outputs.present }}" = "true" ] && [ "${{ steps.label.outputs.is_release }}" = "true" ]; then
echo "Building signed release APK"
./gradlew --no-daemon assembleRelease
else
echo "Building debug APK (no signing keystore available or not a release tag)"
./gradlew --no-daemon assembleDebug
fi
- name: Locate and rename APK
id: apk
run: |
set -euo pipefail
SRC=$(ls android/app/build/outputs/apk/release/*.apk 2>/dev/null | head -1 || true)
if [ -z "$SRC" ]; then
SRC=$(ls android/app/build/outputs/apk/debug/*.apk | head -1)
VARIANT="debug"
else
VARIANT="release"
fi
DEST="build/LedGrab-${{ steps.label.outputs.label }}-android-${VARIANT}.apk"
mkdir -p build
cp "$SRC" "$DEST"
echo "path=$DEST" >> "$GITHUB_OUTPUT"
echo "name=$(basename "$DEST")" >> "$GITHUB_OUTPUT"
ls -lh "$DEST"
- name: Upload APK artifact
uses: actions/upload-artifact@v3
with:
name: LedGrab-${{ steps.label.outputs.label }}-android
path: ${{ steps.apk.outputs.path }}
retention-days: 90
- name: Attach APK to Gitea release (upsert)
if: ${{ steps.label.outputs.is_release == 'true' }}
env:
GITEA_TOKEN: ${{ secrets.DEPLOY_TOKEN }}
run: |
set -euo pipefail
TAG="${{ gitea.ref_name }}"
BASE_URL="${{ gitea.server_url }}/api/v1/repos/${{ gitea.repository }}"
APK_PATH="${{ steps.apk.outputs.path }}"
APK_NAME="${{ steps.apk.outputs.name }}"
if [ -z "${GITEA_TOKEN:-}" ]; then
echo "::error::DEPLOY_TOKEN secret not configured — cannot attach APK"
exit 1
fi
# Upsert: look up release by tag. If it exists, reuse it; if 404,
# create one. Makes the Android workflow self-sufficient — no
# ordering dependency on release.yml's create-release job.
HTTP=$(curl -s -o /tmp/release.json -w "%{http_code}" \
"$BASE_URL/releases/tags/$TAG" \
-H "Authorization: token $GITEA_TOKEN")
case "$HTTP" in
200)
RELEASE_ID=$(python3 -c "import json; print(json.load(open('/tmp/release.json'))['id'])")
echo "Found existing release id=$RELEASE_ID"
;;
404)
echo "No release for tag $TAG — creating one"
IS_PRE="false"
if echo "$TAG" | grep -qE '(alpha|beta|rc)'; then
IS_PRE="true"
fi
CREATE_HTTP=$(curl -s -o /tmp/created.json -w "%{http_code}" \
-X POST "$BASE_URL/releases" \
-H "Authorization: token $GITEA_TOKEN" \
-H "Content-Type: application/json" \
-d "{\"tag_name\":\"$TAG\",\"name\":\"LedGrab $TAG\",\"draft\":false,\"prerelease\":$IS_PRE}")
if [ "$CREATE_HTTP" != "201" ] && [ "$CREATE_HTTP" != "200" ]; then
echo "::error::Failed to create release (HTTP $CREATE_HTTP)"
cat /tmp/created.json
exit 1
fi
RELEASE_ID=$(python3 -c "import json; print(json.load(open('/tmp/created.json'))['id'])")
echo "Created release id=$RELEASE_ID"
;;
*)
echo "::error::Unexpected HTTP $HTTP when looking up release for tag $TAG"
cat /tmp/release.json
exit 1
;;
esac
# Replace existing asset if present (re-run safety).
EXISTING_ID=$(curl -fsS "$BASE_URL/releases/$RELEASE_ID/assets" \
-H "Authorization: token $GITEA_TOKEN" \
| python3 -c "import sys,json; assets=json.load(sys.stdin); print(next((str(a['id']) for a in assets if a['name']=='$APK_NAME'),''))")
if [ -n "$EXISTING_ID" ]; then
curl -fsS -X DELETE "$BASE_URL/releases/$RELEASE_ID/assets/$EXISTING_ID" \
-H "Authorization: token $GITEA_TOKEN"
echo "Replaced existing asset: $APK_NAME"
fi
# -f: exit non-zero on 4xx/5xx so a broken token fails the job
# loudly instead of the previous silent "Uploaded" lie.
curl -fsS -X POST \
"$BASE_URL/releases/$RELEASE_ID/assets?name=$APK_NAME" \
-H "Authorization: token $GITEA_TOKEN" \
-H "Content-Type: application/octet-stream" \
--data-binary "@$APK_PATH"
echo "Uploaded: $APK_NAME"
+80
View File
@@ -0,0 +1,80 @@
name: Build Artifacts
on:
workflow_dispatch:
inputs:
version:
description: 'Version label (e.g. dev, 0.3.0-test)'
required: false
default: 'dev'
jobs:
build-windows:
runs-on: ubuntu-latest
steps:
- name: Checkout
uses: actions/checkout@v4
with:
fetch-depth: 0
- name: Setup Python
uses: actions/setup-python@v5
with:
python-version: '3.11'
- name: Setup Node.js
uses: actions/setup-node@v4
with:
node-version: '20'
- name: Install system dependencies
run: |
sudo apt-get update
sudo apt-get install -y --no-install-recommends zip libportaudio2 nsis msitools
- name: Cross-build Windows distribution
run: |
chmod +x build/build-dist-windows.sh
./build/build-dist-windows.sh "v${{ inputs.version }}"
- uses: actions/upload-artifact@v3
with:
name: LedGrab-${{ inputs.version }}-win-x64
path: |
build/LedGrab-*.zip
build/LedGrab-*-setup.exe
retention-days: 90
build-linux:
runs-on: ubuntu-latest
steps:
- name: Checkout
uses: actions/checkout@v4
with:
fetch-depth: 0
- name: Setup Python
uses: actions/setup-python@v5
with:
python-version: '3.11'
- name: Setup Node.js
uses: actions/setup-node@v4
with:
node-version: '20'
- name: Install system dependencies
run: |
sudo apt-get update
sudo apt-get install -y --no-install-recommends libportaudio2
- name: Build Linux distribution
run: |
chmod +x build/build-dist.sh
./build/build-dist.sh "v${{ inputs.version }}"
- uses: actions/upload-artifact@v3
with:
name: LedGrab-${{ inputs.version }}-linux-x64
path: build/LedGrab-*.tar.gz
retention-days: 90
+157 -35
View File
@@ -4,16 +4,30 @@ on:
push:
tags:
- 'v*'
# Manual dispatch builds Windows/Linux/Docker artifacts without creating
# a Gitea release — for validating build scripts between real releases.
# Attach/push steps are gated on github.event_name == 'push'.
workflow_dispatch:
inputs:
version:
description: 'Version label for dispatch builds (artifacts only, no release)'
required: false
default: 'dev'
jobs:
# ── Create the release first (shared by all build jobs) ────
create-release:
# Skipped on workflow_dispatch — dispatch is for build validation only.
if: github.event_name == 'push'
runs-on: ubuntu-latest
outputs:
release_id: ${{ steps.create.outputs.release_id }}
steps:
- name: Checkout
- name: Fetch RELEASE_NOTES.md only
uses: actions/checkout@v4
with:
sparse-checkout: RELEASE_NOTES.md
sparse-checkout-cone-mode: false
- name: Create Gitea release
id: create
@@ -33,11 +47,9 @@ jobs:
REPO=$(echo "${{ gitea.repository }}" | tr '[:upper:]' '[:lower:]')
DOCKER_IMAGE="${SERVER_HOST}/${REPO}"
# Scan for RELEASE_NOTES.md (check repo root first, then recursively)
NOTES_FILE=$(find . -maxdepth 3 -name "RELEASE_NOTES.md" -type f | head -1)
if [ -n "$NOTES_FILE" ]; then
export RELEASE_NOTES=$(cat "$NOTES_FILE")
echo "Found release notes: $NOTES_FILE"
if [ -f RELEASE_NOTES.md ]; then
export RELEASE_NOTES=$(cat RELEASE_NOTES.md)
echo "Found RELEASE_NOTES.md"
else
export RELEASE_NOTES=""
echo "No RELEASE_NOTES.md found"
@@ -64,6 +76,7 @@ jobs:
| Windows (installer) | \`LedGrab-{tag}-setup.exe\` | Install with Start Menu shortcut, optional autostart, uninstaller |
| Windows (portable) | \`LedGrab-{tag}-win-x64.zip\` | Unzip anywhere, run LedGrab.bat |
| Linux | \`LedGrab-{tag}-linux-x64.tar.gz\` | Extract, run ./run.sh |
| Android | \`LedGrab-{tag}-android-release.apk\` | Sideload on Android 7.0+ (API 24+) — TV boxes, Fire TV, phones, tablets. arm64-v8a / x86_64 / x86 |
| Docker | See below | docker pull + docker run |
After starting, open **http://localhost:8080** in your browser.
@@ -77,14 +90,17 @@ jobs:
### First-time setup
1. Change the default API key in config/default_config.yaml
2. Open http://localhost:8080 and discover your WLED devices
3. See INSTALLATION.md for detailed configuration
1. Change the default API key in `config/default_config.yaml`.
2. Open http://localhost:8080 and add your LED devices.
3. See `INSTALLATION.md` for detailed configuration.
''').strip())
print(json.dumps('\n\n'.join(sections)))
")
# Created as draft so the release isn't user-visible until every
# build job has attached its assets. The publish-release job at
# the end of the workflow flips draft=false once all builds pass.
RELEASE=$(curl -s -X POST "$BASE_URL/releases" \
-H "Authorization: token $GITEA_TOKEN" \
-H "Content-Type: application/json" \
@@ -92,7 +108,7 @@ jobs:
\"tag_name\": \"$TAG\",
\"name\": \"LedGrab $TAG\",
\"body\": $BODY_JSON,
\"draft\": false,
\"draft\": true,
\"prerelease\": $IS_PRE
}")
@@ -110,6 +126,9 @@ jobs:
# ── Windows portable ZIP (cross-built from Linux) ─────────
build-windows:
needs: create-release
# `!cancelled()` lets this job run even when create-release was skipped
# (dispatch) or failed. The attach step itself is still push-gated.
if: ${{ !cancelled() && (needs.create-release.result == 'success' || needs.create-release.result == 'skipped') }}
runs-on: ubuntu-latest
steps:
- name: Checkout
@@ -134,8 +153,8 @@ jobs:
- name: Cross-build Windows distribution
run: |
chmod +x build-dist-windows.sh
./build-dist-windows.sh "${{ gitea.ref_name }}"
chmod +x build/build-dist-windows.sh
./build/build-dist-windows.sh "${{ gitea.ref_name }}"
- name: Upload build artifacts
uses: actions/upload-artifact@v3
@@ -147,6 +166,8 @@ jobs:
retention-days: 90
- name: Attach assets to release
# Push (tag) only — dispatch runs produce artifacts but no release.
if: github.event_name == 'push'
env:
GITEA_TOKEN: ${{ secrets.DEPLOY_TOKEN }}
run: |
@@ -173,15 +194,26 @@ jobs:
echo "Uploaded: $NAME"
}
# Publish an asset plus its .sha256 sidecar. The in-app update
# service refuses to install without a published checksum, so
# every artifact needs its hash uploaded alongside.
upload_with_sha256() {
local FILE="$1"
upload_asset "$FILE"
(cd "$(dirname "$FILE")" && sha256sum "$(basename "$FILE")" > "$(basename "$FILE").sha256")
upload_asset "$FILE.sha256"
}
ZIP_FILE=$(ls build/LedGrab-*.zip | head -1)
[ -f "$ZIP_FILE" ] && upload_asset "$ZIP_FILE"
[ -f "$ZIP_FILE" ] && upload_with_sha256 "$ZIP_FILE"
SETUP_FILE=$(ls build/LedGrab-*-setup.exe 2>/dev/null | head -1)
[ -f "$SETUP_FILE" ] && upload_asset "$SETUP_FILE"
[ -f "$SETUP_FILE" ] && upload_with_sha256 "$SETUP_FILE"
# ── Linux tarball ──────────────────────────────────────────
build-linux:
needs: create-release
if: ${{ !cancelled() && (needs.create-release.result == 'success' || needs.create-release.result == 'skipped') }}
runs-on: ubuntu-latest
steps:
- name: Checkout
@@ -206,8 +238,8 @@ jobs:
- name: Build Linux distribution
run: |
chmod +x build-dist.sh
./build-dist.sh "${{ gitea.ref_name }}"
chmod +x build/build-dist.sh
./build/build-dist.sh "${{ gitea.ref_name }}"
- name: Upload build artifact
uses: actions/upload-artifact@v3
@@ -217,34 +249,44 @@ jobs:
retention-days: 90
- name: Attach tarball to release
if: github.event_name == 'push'
env:
GITEA_TOKEN: ${{ secrets.DEPLOY_TOKEN }}
run: |
RELEASE_ID="${{ needs.create-release.outputs.release_id }}"
BASE_URL="${{ gitea.server_url }}/api/v1/repos/${{ gitea.repository }}"
upload_asset() {
local FILE="$1"
local NAME
NAME=$(basename "$FILE")
EXISTING_ID=$(curl -s "$BASE_URL/releases/$RELEASE_ID/assets" \
-H "Authorization: token $GITEA_TOKEN" \
| python3 -c "import sys,json; assets=json.load(sys.stdin); print(next((str(a['id']) for a in assets if a['name']=='$NAME'),''))" 2>/dev/null)
if [ -n "$EXISTING_ID" ]; then
curl -s -X DELETE "$BASE_URL/releases/$RELEASE_ID/assets/$EXISTING_ID" \
-H "Authorization: token $GITEA_TOKEN"
echo "Replaced existing asset: $NAME"
fi
curl -s -X POST \
"$BASE_URL/releases/$RELEASE_ID/assets?name=$NAME" \
-H "Authorization: token $GITEA_TOKEN" \
-H "Content-Type: application/octet-stream" \
--data-binary "@$FILE"
echo "Uploaded: $NAME"
}
TAR_FILE=$(ls build/LedGrab-*.tar.gz | head -1)
TAR_NAME=$(basename "$TAR_FILE")
# Delete existing asset with same name to prevent duplicates on re-run
EXISTING_ID=$(curl -s "$BASE_URL/releases/$RELEASE_ID/assets" \
-H "Authorization: token $GITEA_TOKEN" \
| python3 -c "import sys,json; assets=json.load(sys.stdin); print(next((str(a['id']) for a in assets if a['name']=='$TAR_NAME'),''))" 2>/dev/null)
if [ -n "$EXISTING_ID" ]; then
curl -s -X DELETE "$BASE_URL/releases/$RELEASE_ID/assets/$EXISTING_ID" \
-H "Authorization: token $GITEA_TOKEN"
echo "Replaced existing asset: $TAR_NAME"
if [ -f "$TAR_FILE" ]; then
upload_asset "$TAR_FILE"
(cd "$(dirname "$TAR_FILE")" && sha256sum "$(basename "$TAR_FILE")" > "$(basename "$TAR_FILE").sha256")
upload_asset "$TAR_FILE.sha256"
fi
curl -s -X POST \
"$BASE_URL/releases/$RELEASE_ID/assets?name=$TAR_NAME" \
-H "Authorization: token $GITEA_TOKEN" \
-H "Content-Type: application/octet-stream" \
--data-binary "@$TAR_FILE"
echo "Uploaded: $TAR_NAME"
# ── Docker image ───────────────────────────────────────────
build-docker:
needs: create-release
if: ${{ !cancelled() && (needs.create-release.result == 'success' || needs.create-release.result == 'skipped') }}
runs-on: ubuntu-latest
steps:
- name: Checkout
@@ -268,6 +310,9 @@ jobs:
- name: Login to Gitea Container Registry
id: docker-login
# Dispatch runs don't need registry credentials — the build step
# verifies the Dockerfile locally and push is skipped.
if: github.event_name == 'push'
continue-on-error: true
run: |
echo "${{ secrets.DEPLOY_TOKEN }}" | docker login \
@@ -275,7 +320,10 @@ jobs:
-u "${{ gitea.actor }}" --password-stdin
- name: Build Docker image
if: steps.docker-login.outcome == 'success'
# Always build — dispatch uses this to validate the Dockerfile.
# On push, still gate on successful login so we don't build a
# tagged image that can't be pushed.
if: github.event_name != 'push' || steps.docker-login.outcome == 'success'
run: |
TAG="${{ gitea.ref_name }}"
REGISTRY="${{ steps.meta.outputs.registry }}"
@@ -294,7 +342,7 @@ jobs:
fi
- name: Push Docker image
if: steps.docker-login.outcome == 'success'
if: github.event_name == 'push' && steps.docker-login.outcome == 'success'
run: |
TAG="${{ gitea.ref_name }}"
REGISTRY="${{ steps.meta.outputs.registry }}"
@@ -305,3 +353,77 @@ jobs:
if ! echo "$TAG" | grep -qE '(alpha|beta|rc)'; then
docker push "$REGISTRY:latest"
fi
# Best-effort arm64 (Raspberry Pi / arm64 HAOS hosts). Runs AFTER the
# amd64 push so the amd64 image always ships even if this fails.
# Deliberately avoids `docker buildx` (its docker-container driver needs
# nested networking the TrueNAS runners lack — see contexts/ci-cd.md):
# instead it cross-builds a single arm64 image via QEMU binfmt and folds
# amd64 + arm64 into multi-arch manifest lists under the existing tags.
# `continue-on-error` keeps a runner that can't emulate arm64 from
# failing the release; the plain amd64 tags pushed above remain valid.
- name: Build + publish arm64 (multi-arch manifest, best-effort)
if: github.event_name == 'push' && steps.docker-login.outcome == 'success'
continue-on-error: true
run: |
set -e
export DOCKER_CLI_EXPERIMENTAL=enabled
TAG="${{ gitea.ref_name }}"
REGISTRY="${{ steps.meta.outputs.registry }}"
VERSION="${{ steps.meta.outputs.version }}"
# Register arm64 emulation. If the runner forbids privileged
# containers this fails and the whole step is skipped.
docker run --privileged --rm tonistiigi/binfmt --install arm64
# Cross-build the arm64 image (QEMU-emulated — slow but uses arm64
# manylinux wheels, so no source compilation). Stays in the local
# daemon alongside the amd64 image from the previous build step.
DOCKER_BUILDKIT=1 docker build \
--platform linux/arm64 \
--build-arg APP_VERSION="$VERSION" \
--label "org.opencontainers.image.version=$VERSION" \
--label "org.opencontainers.image.revision=${{ gitea.sha }}" \
-t "$REGISTRY:$VERSION-arm64" \
./server
# Fold amd64 + arm64 into a multi-arch manifest list under each
# user-facing tag. The arch-suffixed tags remain pullable directly.
publish_manifest() {
local t="$1"
docker tag "$REGISTRY:$t" "$REGISTRY:$t-amd64"
docker push "$REGISTRY:$t-amd64"
docker tag "$REGISTRY:$VERSION-arm64" "$REGISTRY:$t-arm64"
docker push "$REGISTRY:$t-arm64"
docker manifest create --amend "$REGISTRY:$t" \
"$REGISTRY:$t-amd64" "$REGISTRY:$t-arm64"
docker manifest push "$REGISTRY:$t"
}
publish_manifest "$TAG"
publish_manifest "$VERSION"
if ! echo "$TAG" | grep -qE '(alpha|beta|rc)'; then
publish_manifest "latest"
fi
# ── Publish the release (flip draft=false) ─────────────────
# Runs only after every build job succeeded so users never see a
# release that's missing artifacts or sha256 sidecars (the in-app
# updater refuses to install without them).
publish-release:
needs: [create-release, build-windows, build-linux, build-docker]
if: github.event_name == 'push' && success()
runs-on: ubuntu-latest
steps:
- name: Promote draft release to published
env:
GITEA_TOKEN: ${{ secrets.DEPLOY_TOKEN }}
run: |
RELEASE_ID="${{ needs.create-release.outputs.release_id }}"
BASE_URL="${{ gitea.server_url }}/api/v1/repos/${{ gitea.repository }}"
curl -s -X PATCH "$BASE_URL/releases/$RELEASE_ID" \
-H "Authorization: token $GITEA_TOKEN" \
-H "Content-Type: application/json" \
-d '{"draft": false}'
echo "Published release $RELEASE_ID"
+6
View File
@@ -5,9 +5,15 @@ on:
branches: [master]
pull_request:
branches: [master]
# Allow manual runs (e.g. to validate after a release commit was skipped).
workflow_dispatch:
jobs:
test:
# Skip release-publishing commits — version bumps don't affect lint/tests
# and the release.yml pipeline is already running. PRs and manual dispatch
# always run.
if: ${{ github.event_name != 'push' || !startsWith(github.event.head_commit.message, 'chore: release') }}
runs-on: ubuntu-latest
steps:
- name: Checkout
+30 -3
View File
@@ -4,7 +4,16 @@ __pycache__/
*$py.class
*.so
.Python
build/
# Build output artifacts (LedGrab/, *.zip, *.exe, *.tar.gz, cached downloads)
build/LedGrab/
build/*.zip
build/*.exe
build/*.tar.gz
build/*.msi
build/python-embed-*.zip
build/pip-wheels/
build/win-wheels/
build/tk-extract/
develop-eggs/
dist/
downloads/
@@ -16,6 +25,10 @@ parts/
sdist/
var/
wheels/
# …but keep pre-built Android wheels (pydantic-core cross-compiled for
# arm64-v8a / x86_64 / x86, required by the Chaquopy build)
!android/wheels/
!android/wheels/*
*.egg-info/
.installed.cfg
*.egg
@@ -49,8 +62,17 @@ htmlcov/
logs/
*.log.*
# Runtime data
data/
# Runtime data — anchor to repo root so nested package data dirs
# (server/src/ledgrab/data/prebuilt_sounds, game_adapters) are NOT ignored.
# An unanchored `data/` rule silently broke the v0.4.2 release by keeping
# shipped sound assets out of the CI tag checkout.
/data/
/server/data/
# Defensive: if the server is launched from server/src/ (uncommon path),
# its relative `data/` dir resolves to server/src/data/. Templates now
# live in SQLite, so any *.json that lands here is stale runtime export
# and must not be committed.
/server/src/data/
*.db
*.sqlite
*.json.bak
@@ -73,3 +95,8 @@ tmp/
# OS
Thumbs.db
.DS_Store
# Added by code-review-graph
.code-review-graph/
# vex semantic-search embedding cache (auto-downloaded on first --semantic run)
.fastembed_cache/
+12
View File
@@ -0,0 +1,12 @@
{
"mcpServers": {
"code-review-graph": {
"command": "uvx",
"args": [
"code-review-graph",
"serve"
],
"type": "stdio"
}
}
}
+4 -2
View File
@@ -4,10 +4,12 @@ repos:
hooks:
- id: black
args: [--line-length=100, --target-version=py311]
language_version: python3.11
- repo: https://github.com/astral-sh/ruff-pre-commit
rev: v0.8.0
# Bumped from v0.8.0 so the hook recognises UP045
# (non-pep604-annotation-optional), which the v0.13+ ruff split off
# from UP007. Pyproject.toml extend-selects both rules.
rev: v0.15.12
hooks:
- id: ruff
args: [--line-length=100, --target-version=py311]
+25
View File
@@ -0,0 +1,25 @@
# vex configuration — https://github.com/tenatarika/vex
#
# Place this file in your project root as .vex.toml
# Glob patterns to exclude from indexing (gitignore syntax, on top of .gitignore)
# exclude = [
# "vendor/**",
# "node_modules/**",
# "*.generated.go",
# "dist/**",
# ]
# Default output format: "text", "json", or "compact"
# format = "text"
# Enable semantic embeddings by default (slower indexing, enables meaning-based search)
semantic = true
# Automatically run `vex update` before search if the index is stale
auto_update = true
# Embedder used for semantic indexing. IDs: minilm-l6-v2 (default, CPU-fast),
# jina-code (code-specialized, GPU-worthy), bge-base-en-v1.5, bge-large-en-v1.5.
# Changing the embedder requires a full reindex.
embedder = "jina-code"
+428
View File
@@ -0,0 +1,428 @@
# LedGrab Architecture Audit — Remaining Items
Roadmap for the architecture-audit refactor sprint that started 2026-05-22.
This file lists every audit finding that is **not yet addressed**; the ones
already landed in commits `563cbac..2f15fbb` are summarised below for
context.
## Already done (10 commits)
| Commit | Findings addressed |
|---|---|
| `563cbac` | C2, C11, C1 (parallel-change only), C3, C4, C6, C7-streams |
| `29bdacf` | C5 (HA/Z2M swap helper; full ABC deferred) |
| `97dae2c` | H1 |
| `5fec8db` | M4 |
| `98fb61d` | H2 |
| `9f3f346` | M5 |
| `05f73ee` | H6 (bindable extraction only) |
| `3b8f00e` + `c1aa2eb` | C7 store-side |
| `2f15fbb` | H3 |
| _uncommitted (2026-05-27 autonomous pass)_ | H6-rest, H8, M7 (foundation + 3 reference files) |
All commits have ≥1 code-review subagent pass with HIGH findings fixed
before commit. Tests pass on each commit; ruff clean; tsc + bundle build
clean for the frontend commit.
The two CRITICAL **data-safety** items (C2 silent CSS fallback, C11
string-replace JSON migration) are fixed. The two CRITICAL
**parallel-change** problems for color-strip + value-source dispatch are
fixed. The two HIGH dispatch problems (H1 effects, H2 rules) are fixed.
---
## Remaining backend items
### HIGH
#### H4 — `Device.__init__` 40+ params mixing per-type fields
**File:** `server/src/ledgrab/storage/device_store.py:46-150`
The `Device` dataclass constructor accepts ~40 parameters that mix common
fields with DMX-only / DDP-only / Hue-only / Yeelight-only / Wiz-only /
LIFX-only / Govee-only / Nanoleaf-only / SPI-only / Chroma-only /
GameSense-only fields. Setting `hue_username` on a WLED device is
silently ignored.
**Approach:** introduce per-device-type config dataclasses
(`DmxConfig`, `HueConfig`, `DdpConfig`, …) and make `Device.config` a
discriminated union. Per-type validation moves to the config classes.
Wire migration: every existing device row needs to be re-parsed; use the
versioned `MigrationRunner` introduced in Phase 1.2.
**Risk:** medium-high. Touches:
- `storage/device_store.py` — Device dataclass, `from_dict`, `to_dict`,
`create_device`, `update_device`
- `api/schemas/devices.py` — Pydantic schemas
- `api/routes/devices.py` — request validation
- `core/devices/*` — every provider reads device fields
- A new migration to translate flat fields → nested `config`
**Estimated scope:** ~1500 LOC diff, 1-2 dedicated sessions.
#### H5 — `WledTargetProcessor` god class (32 methods, 5 responsibilities)
**File:** `server/src/ledgrab/core/processing/wled_target_processor.py` (1238 LOC)
Conflates:
1. Device connectivity (probe, liveness, reconnect)
2. FPS negotiation (adaptive_fps, keepalive_interval, state_check_interval)
3. LED resampling (`_fit_to_device` — 60 lines of numpy)
4. Preview WebSocket fanout (`_preview_clients`, `_broadcast_led_preview`)
5. Metrics emission (`get_state`, `get_metrics`)
**Approach:** extract `WledDeviceConnector`, `WledPixelSender`,
`TargetFitProcessor`, `TargetPreviewBroadcaster`, `TargetMetricsCollector`.
`WledTargetProcessor` becomes an orchestrator that composes them.
**Risk:** HIGHEST in the audit. This class drives physical LED hardware
in production. A regression caught at runtime (in the user's living
room) is the expensive failure mode. Needs manual verification with at
least one real WLED device after the refactor.
**Coupled with:** C5 (HA/Z2M shared the same shape; should extract a
common `BaseTargetProcessor` ABC at the same time so all three
processors share lifecycle / preview / metrics code).
**Estimated scope:** ~2000 LOC diff, 2-3 dedicated sessions, with manual
device testing after each.
#### H7 — `device-discovery.ts` 1745 LOC
Frontend mirror of H4. The `onDeviceTypeChanged` handler has a giant
switch with 15+ device kinds and 15+ `_showXxxFields` / `_buildXxxItems`
helpers. Adding a device type requires editing 5 separate frontend hooks.
**Approach:** mirror the H4 backend redesign — once the storage layer
has per-type config objects, the frontend can have a per-type field-set
registry. Best done **after** H4 lands so the schemas drive the
registry.
**Estimated scope:** 1-2 sessions; coupled to H4.
#### H8 — `automations.ts` 1410 LOC — ✅ DONE (uncommitted, 2026-05-27)
Frontend mirror of H2 (rule polymorphism). Already addressed on the
backend in `98fb61d`; the frontend dispatch on `RuleType` was
hand-rolled.
**Done:** the two remaining hand-rolled dispatch ladders were converted
to registries keyed by `RuleType`, alongside the pre-existing
`RULE_CHIP_RENDERERS`:
- `RULE_FIELD_RENDERERS` — the `renderFields` if/elif ladder was
extracted into module-level `_renderXxxFields(container, data)`
functions (they only ever closed over `container`); the in-row
`renderFields` is now a 3-line dispatcher.
- `RULE_COLLECTORS` — the `getAutomationEditorRules` if/elif ladder
became per-type collectors; the loop is now a registry lookup.
- All three registries are typed `Record<RuleType, …>` (compile-time
exhaustiveness) and an import-time `_assertRuleHandlerCoverage()`
logs loudly if any registry drifts from `RULE_TYPE_KEYS`. (Frontend
logs rather than throws — a thrown error at import would brick the
whole bundle, not just the editor — the one intentional divergence
from the backend's raising `_assert_rule_handler_coverage`.)
Adding a new rule type now means: one entry in `RULE_TYPE_KEYS`,
`RULE_TYPE_ICONS`, and each of the three registries — and tsc + the
coverage check flag any omission.
Verified: tsc + bundle build clean; typescript-reviewer APPROVE (the
extracted renderer bodies are byte-identical to the originals; no stray
closure captures; http_poll widget-stash + HA entity loading preserved).
### MEDIUM
#### M1 — `ProcessorManager.add_target` shotgun (11 args, WLED-leak)
**File:** `server/src/ledgrab/core/processing/processor_manager.py:396`
Method is named generically (`add_target`) but accepts `protocol="ddp"`
and `keepalive_interval` — WLED-only fields. HA and Z2M have sibling
methods with their own bespoke params.
**Approach:** extract a `TargetFactory` (per-kind builders, similar to
`value_source_factories.py` from Phase 7). Couple with H5/C5 work.
#### M2 — `TargetContext` god-bag
**File:** `server/src/ledgrab/core/processing/processor_manager.py`
`@dataclass TargetContext` exposes ~8 attributes (device_store,
color_strip_stream_manager, value_stream_manager, metrics_history,
mqtt_manager, ha_manager, …). Processors silently depend on whichever
fields they read. Tests have to construct a huge mock context.
**Approach:** make per-processor explicit dependency injection. Couple
with H5 work.
#### M3 — Validation duplicated across layers
Field-level constraints (composite nesting depth, name uniqueness, span
ranges) are enforced in route + schema + store. Adding a new constraint
means editing 3 places.
**Approach:** move all validation to the model/schema layer (Pydantic
validators + dataclass `__post_init__`). Routes trust the schema; store
trusts the model.
**Risk:** moderate — cross-cutting; needs careful review of which layer
currently owns which constraint.
#### M6 — `ws_stream.py` mixed concerns (699 LOC)
**File:** `server/src/ledgrab/api/routes/color_strip_sources/ws_stream.py`
The worst part (stream-creation dispatch) was fixed in Phase 2.1 — it
now calls `color_strip_kinds.build_stream(source, deps)`. The remaining
699 lines mix config parsing + WebSocket lifecycle + frame loop. Could
extract the frame loop into a separate `PreviewFrameLoop` class.
**Estimated scope:** half a session. Low impact since the parallel-change
problem is already fixed.
#### M7 — No shared frontend API client — 🟡 FOUNDATION DONE (uncommitted, 2026-05-27)
**File:** every `static/js/features/*.ts`
`fetchWithAuth(...)` + bespoke error-unwrapping is copy-pasted in every
feature's save / load function. ~45 files, ~243 call sites.
**Done:** `static/js/core/api-client.ts` now provides typed
`apiGet` / `apiPost` / `apiPut` / `apiPatch` / `apiDelete` that wrap
`fetchWithAuth` (so auth, 401-relogin, retry, timeout, and the offline
toast are unchanged) and collapse the repeated
`if (!resp.ok) { detail || HTTP <status> } … resp.json()` dance into one
call returning a typed body and throwing `ApiError` on failure. The
`detail` unwrap is hardened to join FastAPI validation arrays instead of
stringifying to `[object Object]`. **35 feature/core files migrated**
(covers GET/POST/PUT/DELETE, typed response bodies, custom i18n error
messages, silent-failure GETs, bulk `Promise.allSettled` deletes,
inline-error saves, array-`detail` joins, fire-and-forget POSTs, and
local catch handling) — reviewer-approved for behaviour parity across
the riskier divergences. Migrated files include the integration sources
(weather / HA / MQTT / HTTP), the template families (capture / audio /
audio-processing / pattern), the scene-preset CRUD, the simple-CRUD
entity files (sync-clocks / audio-sources / game-integration /
gradient / displays / device-discovery), the light-target editors
(z2m / ha), the preferences modules (dashboard-layout / card-modes /
notifications-watcher), the calibration editors (simple + advanced),
the entire `automations.ts` and `devices.ts` CRUD surfaces, and several
core utilities (`api-client.ts` itself, `cache.ts`, `command-palette.ts`,
`graph-connections.ts`, `tag-input.ts`, `process-picker.ts`,
`perf-charts.ts`, `icon-picker.ts`, `update.ts`, `integrations.ts`).
Also added **14 new locale keys** (en / ru / zh) so the fallback
messages the migration surfaces — `pattern.error.save_failed`,
`audio_processing.error.save_failed`, `audio_template.error.save_failed`,
`audio_template.error.load_failed`, `templates.error.save_failed`,
`templates.error.load_failed`, `gradient.error.save_failed`,
`target.error.load_failed`, `device.error.load_failed`,
`automations.error.{load,save,delete,toggle}_failed`, plus
`gradient.error.delete_failed` for ru/zh — are translated instead of
hardcoded English. A scan confirms **no `errorMessage: '<English>'`
strings remain** in the migrated diff.
**Remaining:** 9 feature files (~94 call sites). All but one are the
big god-modules whose migration is best done as part of their C8/C9/C10
splits: `streams.ts` (18), `settings.ts` (18), `targets.ts` (16),
`dashboard.ts` (15), `color-strips/index.ts` (8), `graph-editor.ts` (7),
`assets.ts` (6 — also blocked by multipart upload + blob download paths
that legitimately bypass the JSON client), and `value-sources.ts` (5).
The lone leaf file still on `fetchWithAuth` is `pairing-flow.ts` (1) —
its branching on raw `Response.status` codes (200 / 409 / 4xx) doesn't
fit the api-client contract, so it stays on raw fetch by design.
Migration is mechanical but **not** a blind find/replace — each site
carries its own localised error key that must be preserved as the
`errorMessage` option, and binary/multipart endpoints (e.g.
`assets.ts` file upload / blob download) must stay on raw
`fetchWithAuth` (the client is JSON-only). Each migrated file ideally
gets manual UI smoke-testing. **Behaviour note:** migrated GET sites now
prefer the server's `detail` over the generic localised fallback when
present — matching what the write paths already did; intended, but
user-visible.
#### M8 — Global `_cached*` `let` vars
Mutable module-level state mutated from multiple feature modules. No
subscription model — features manually `invalidate()` after CRUD.
**Approach:** introduce a reactive cache (EventEmitter pattern or a tiny
store like Nano Stores). Couple with M7 (the API client can drive cache
invalidation on write).
#### M9 — `dashboard.ts` 1421 LOC
Frontend god-module orchestrating + rendering device / target / CSS
cards. Couple with C8/C9/C10 frontend split work.
#### M10 — Duplicate frontend modal classes
`ValueSourceModal`, `StreamEditorModal`, `TargetEditorModal`,
`AddDeviceModal`, etc. each reimplement pristine-check / undo / focus
management.
**Approach:** introduce a `FormModal<T>` base class.
#### M11 — Hardcoded `_getSectionForSource` / `_getTabForSource`
Routing tables duplicated across multiple feature files (streams.ts,
value-sources.ts). Adding a new stream type requires hunting strings.
**Approach:** single routing registry keyed by source_type.
#### M12 — Late imports masking cycles
Partially addressed by the kind registries (Phase 2.1, 2.2). Some
late-imports still exist in `value_stream.py`, `audio_stream.py`, the
target processors. Resolving them requires restructuring module layout
to break the circular dependencies.
**Estimated scope:** small follow-up after H5.
### LOW
#### L1 — `(src as any).field` casts in `value-sources.ts`
Discriminated unions aren't narrowed properly. Couple with C8 frontend
split.
#### L2 — Mutable state without locks
`_preview_clients`, `_last_preview_data`, `_color_stream`,
`_css_stream` are mutated from multiple async tasks without explicit
locks. Production has not exhibited issues but the contract is fragile.
**Approach:** add explicit `asyncio.Lock` per processor. Couple with H5.
#### L3 — `Calibration.validate()` raises instead of returning result
**File:** `server/src/ledgrab/core/capture/calibration.py:164`
All 4 call sites currently rely on the raise; converting to
`ValidationResult` would force every caller to check a return value
without adding safety. **Recommendation:** skip — current design is
appropriate.
#### L4 — `_SOURCE_TYPE_MAP` is module-private
No public `GET /api/v1/source-types` discovery endpoint. Frontend
hardcodes the list of source types in `types.ts`.
**Approach:** add a discovery route + matching frontend fetch. Couple
with H6 frontend split (since `types.ts` is involved).
#### L5 — `AudioValueStream` implicit state machine
**File:** `server/src/ledgrab/core/processing/value_stream.py:169-383`
`get_value()` can be called before `start()`; transitions are implicit.
**Approach:** explicit State pattern. Low value (production callers
always start before reading).
---
## Remaining frontend items (all)
### CRITICAL
- **C8** — `value-sources.ts` 1972 LOC (4 god-functions, type-dispatch ladders)
- **C9** — `graph-editor.ts` 2707 LOC (layout + interaction + state + WS sync + …)
- **C10** — `streams.ts` 2341 LOC (picture / audio / template kitchen-sink)
### Other frontend (severity in main list above)
- **H6 rest** — ✅ DONE (uncommitted, 2026-05-27): `types.ts` (1140 LOC)
split into 18 per-entity files under `types/` (joining the existing
`bindable.ts`); `types.ts` is now a ~200-line pure re-export barrel, so
every `import { … } from '../types.ts'` still resolves. Reviewer
confirmed all 102 exported symbols preserved, none renamed.
- **H7** — `device-discovery.ts` 1745 LOC (couple with H4)
- **H8** — `automations.ts` 1410 LOC (mirror H2)
- **M7** — shared API client
- **M8** — reactive cache
- **M9** — `dashboard.ts` 1421 LOC
- **M10** — `FormModal<T>` base
- **M11** — routing registry
- **L1** — narrowing the discriminated unions
The frontend remainder is **multi-day work** even when broken up by
finding. Recommended approach: a dedicated frontend sprint with the
typescript-reviewer agent + manual UI testing for each god-module
split. Order:
1. Finish `types.ts` split (H6) — pure organisation, low risk, unblocks
the rest
2. Introduce API client (M7) — every feature file gains a cleaner shape
3. Split `value-sources.ts` (C8) — uses the API client + per-type
registry pattern
4. Split `streams.ts` (C10)
5. Split `graph-editor.ts` (C9) — needs the most care; the file owns
the entire visual editor
6. Polish: `dashboard.ts` (M9), `device-discovery.ts` (H7),
`automations.ts` (H8), `FormModal` (M10), routing registry (M11),
reactive cache (M8), narrowing (L1)
---
## Recommended ordering for future sessions
### Session A — Frontend sprint (multi-day)
Address H6-rest, C8, C9, C10, H7, H8, M7-M11, L1. See order above.
Critical to have typescript-reviewer feedback + manual UI testing after
each split.
> **Progress (2026-05-27, uncommitted):** steps 1 & 2 of the order above
> are done — H6-rest (`types.ts` split) and M7-foundation (`api-client.ts`
> + 3 reference migrations). H8 (automations registry) also landed. Still
> open: C8, C9, C10, H7, the remaining ~40 M7 file migrations, M8-M11, L1.
> Next per the order: introduce the API client everywhere (finish M7),
> then split `value-sources.ts` (C8).
### Session B — Device redesign (1-2 sessions)
Address H4 alone. Touches device storage + provider classes; needs a
data migration. Once H4 lands, H7 frontend mirror can follow.
### Session C — BaseTargetProcessor ABC (2-3 sessions)
Address C5 (full) + H5 + M1 + M2 + L2 together. Highest risk in the
audit because it drives physical LED hardware. Each step needs manual
verification with a real device.
### Session D — Polish (half a session)
Address M3, M6 (remainder), M12 (remainder), L3 (decision: skip), L4,
L5.
---
## Pattern reference for new contributors
Three registry-pattern templates that already exist in the codebase and
should be the model for the remaining dispatch ladders:
1. **Class-level handler dict + import-time coverage assertion**
- `core/processing/effect_stream.py::_RENDERERS`
(`@_effect_renderer` decorator + `@_collect_effect_renderers`
class decorator)
- `core/automations/automation_engine.py::AutomationEngine._RULE_HANDLERS`
(module-level binding after class definition)
- `api/routes/output_targets.py::_TARGET_RESPONSE_BUILDERS`
(response-shape dispatch keyed by storage class)
2. **Per-type free functions + dependency-bag dataclass**
- `core/processing/color_strip_kinds.py` (`StreamDeps` + `STREAM_BUILDERS`)
- `core/processing/value_kinds.py` (`ValueStreamDeps` + `STREAM_BUILDERS`)
- `storage/value_source_factories.py` (`CREATE_BUILDERS` + `UPDATE_APPLIERS`)
3. **Versioned migration runner**
- `storage/data_migrations.py` (`MigrationRunner` + `DataMigration` ABC)
- Used for any storage rename / field-shape change in the future.
- Audit-table contract: atomic transaction covers
applied-check + apply + record, so partial-failure cannot leave
data rewritten but unrecorded.
Adding a new feature that touches dispatch should reach for one of
these three patterns before writing a fresh if/elif chain.
-118
View File
@@ -1,118 +0,0 @@
# Feature Brainstorm — LED Grab
## New Automation Conditions (Profiles)
Right now profiles only trigger on **app detection**. High-value additions:
- **Time-of-day / Schedule** — "warm tones after sunset, off at midnight." Schedule-based value sources pattern already exists
- **Display state** — detect monitor on/off/sleep, auto-stop targets when display is off
- **System idle** — dim or switch to ambient effect after N minutes of no input
- **Sunrise/sunset** — fetch local solar times, drive circadian color temperature shifts
- **Webhook/MQTT trigger** — let external systems activate profiles without HA integration
## New Output Targets
Currently: WLED, Adalight, AmbileD, DDP. Potential:
- **MQTT publish** — generic IoT output, any MQTT subscriber becomes a target
- **Art-Net / sACN (E1.31)** — stage/theatrical lighting protocols, DMX controllers
- **OpenRGB** — control PC peripherals (keyboard, mouse, RAM, fans) as ambient targets
- **HTTP webhook** — POST color data to arbitrary endpoints
- **Recording target** — save color streams to file for playback later
## New Color Strip Sources
- **Spotify / media player** — album art color extraction or tempo-synced effects
- **Weather** — pull conditions from API, map to palettes (blue=rain, orange=sun, white=snow)
- **Camera / webcam** — border-sampling from camera feed for video calls or room-reactive lighting
- **Script source** — user-written JS/Python snippets producing color arrays per frame
- **Notification reactive** — flash/pulse on OS notifications (optional app filter)
## Processing Pipeline Extensions
- **Palette quantization** — force output to match a user-defined palette
- **Zone grouping** — merge adjacent LEDs into logical groups sharing one averaged color
- **Color temperature filter** — warm/cool shift separate from hue shift (circadian/mood)
- **Noise gate** — suppress small color changes below threshold, preventing shimmer on static content
## Multi-Instance & Sync
- **Multi-room sync** — multiple instances with shared clock for synchronized effects
- **Multi-display unification** — treat 2-3 monitors as single virtual display for seamless ambilight
- **Leader/follower mode** — one target's output drives others with optional delay (cascade)
## UX & Dashboard
- **PWA / mobile layout** — mobile-first layout + "Add to Home Screen" manifest
- **Scene presets** — bundled source + filters + brightness as one-click presets ("Movie night", "Gaming")
- **Live preview on dashboard** — miniature screen with LED colors rendered around its border
- **Undo/redo for calibration** — reduce frustration in the fiddly calibration editor
- **Drag-and-drop filter ordering** — reorder postprocessing filter chains visually
## API & Integration
- **WebSocket event bus** — broadcast all state changes over a single WS channel
- **OBS integration** — detect active scene, switch profiles; or use OBS virtual camera as source
- **Plugin system** — formalize extension points into documented plugin API with hot-reload
## Creative / Fun
- **Effect sequencer** — timeline-based choreography of effects, colors, and transitions
- **Music BPM sync** — lock effect speed to detected BPM (beat detection already exists)
- **Color extraction from image** — upload photo, extract palette, use as gradient/cycle source
- **Transition effects** — crossfade, wipe, or dissolve between sources/profiles instead of instant cut
---
## Deep Dive: Notification Reactive Source
**Type:** New `ColorStripSource` (`source_type: "notification"`) — normally outputs transparent RGBA, flashes on notification events. Designed to be used as a layer in a **composite source** so it overlays on top of a persistent base (gradient, effect, screen capture, etc.).
### Trigger modes (both active simultaneously)
1. **OS listener (Windows)**`pywinrt` + `Windows.UI.Notifications.Management.UserNotificationListener`. Runs in background thread, pushes events to source via queue. Windows-only for now; macOS (`pyobjc` + `NSUserNotificationCenter`) and Linux (`dbus` + `org.freedesktop.Notifications`) deferred to future.
2. **Webhook**`POST /api/v1/notifications/{source_id}/fire` with optional body `{ "app": "MyApp", "color": "#FF0000" }`. Always available, cross-platform by nature.
### Source config
```yaml
os_listener: true # enable Windows notification listener
app_filter:
mode: whitelist|blacklist # which apps to react to
apps: [Discord, Slack, Telegram]
app_colors: # user-configured app → color mapping
Discord: "#5865F2"
Slack: "#4A154B"
Telegram: "#26A5E4"
default_color: "#FFFFFF" # fallback when app has no mapping
effect: flash|pulse|sweep # visual effect type
duration_ms: 1500 # effect duration
```
### Effect rendering
Source outputs RGBA color array per frame:
- **Idle**: all pixels `(0,0,0,0)` — composite passes through base layer
- **Flash**: instant full-color, linear fade to transparent over `duration_ms`
- **Pulse**: sine fade in/out over `duration_ms`
- **Sweep**: color travels across the strip like a wave
Each notification starts its own mini-timeline from trigger timestamp (not sync clock).
### Overlap handling
New notification while previous effect is active → restart timer with new color. No queuing.
### App color resolution
1. Webhook body `color` field (explicit override) → highest priority
2. `app_colors` mapping by app name
3. `default_color` fallback
---
## Top Picks (impact vs effort)
1. **Time-of-day + idle profile conditions** — builds on existing profile/condition architecture
2. **MQTT output target** — opens the door to an enormous IoT ecosystem
3. **Scene presets** — purely frontend, bundles existing features into one-click UX
+86 -7
View File
@@ -1,10 +1,41 @@
# Claude Instructions for WLED Screen Controller
# Claude Instructions for LedGrab
## Code Search
**If `ast-index` is available, use it as the PRIMARY code search tool.** It is significantly faster than grep and returns structured, accurate results. Fall back to grep/Glob only when ast-index is not installed, returns empty results, or when searching regex patterns/string literals/comments.
**Priority order: `vex` (PRIMARY) → `ast-index` (fallback) → Grep/Glob (last resort).** This repo has a fully-featured `.vex.toml` index. Use vex first for any symbol/definition/usage/call-graph lookup. Fall back to ast-index only when vex legitimately can't help, and to Grep/Glob only for regex patterns, string literals, comments, config files, or unparsed languages.
**IMPORTANT for subagents:** When spawning Agent subagents (Plan, Explore, general-purpose, etc.), always instruct them to use `ast-index` via Bash for code search instead of grep/Glob. Example: include "Use `ast-index search`, `ast-index class`, `ast-index usages` etc. via Bash for code search" in the agent prompt.
**IMPORTANT — use ALL vex indexing features.** The index is built with every capability enabled, and queries must take advantage of them. Keep them ON and exploit them:
| Capability | Status | Powers |
| ---------- | ------ | ------ |
| Semantic embeddings (`jina-code`, 768-dim) | ON | `vex search` (semantic channel), `similar`, `find_similar`, `duplicates` |
| Call graph | ON | `vex callers`, `callees`, `paths`, `reachable`, `bundle --mode pr-impact` |
| BM25 | ON | hybrid RRF text channel in `vex search` |
| Pattern index | ON | `vex pattern` AST-shape matching |
| C++ includes | ON | include-graph resolution |
| Body tokens (incremental HNSW) | ON | fast incremental reindex |
| History | ON | `vex history`, `vex diff <rev>` blame/evolution queries |
**In-session, use the `mcp__vex__*` MCP tools** (`search`, `show`, `usages`, `callers`, `callees`, `bundle`, `outline`, `implementations`, `similar`, `grep`, `status`, etc.) — MCP output is far cheaper in tokens than `Bash("vex …")`. Drop to Bash `vex` only for CLI-only features (`pattern`, `diff`, `paths`, `reachable`, `bundle`, `history`, `--strict`/`--why` flags), for subagent prompts, or for shell composition.
```bash
vex search "query" --semantic # Hybrid semantic + BM25 search
vex show <Symbol> # Definition body (prefer over Read)
vex usages <Symbol> --strict # Reference sites (AST-precise on T1 langs)
vex callers <Function> # Call sites (function-scoped)
vex callees <Function> # Outgoing calls
vex paths --from <A> --to <B> # Multi-hop call-graph path
vex bundle --mode pr-impact --base master # Changed symbols + callers + reachable tests
vex pattern '$X async fn returning Response' # AST-shape (metavariables)
vex diff master # Symbol-level branch diff
vex history <Symbol> # Commit evolution of a symbol
```
**Maintenance:** the index has `auto_update = true`, so it refreshes on stale queries. After a `vex self-update`, rerun `vex index --history --semantic --embedder jina-code --device cuda` so newly-added extractors populate and all features stay enabled. Verify with `vex status` — every capability line should read `yes`.
**IMPORTANT for subagents:** Subagents don't inherit MCP. When spawning Agent subagents (Plan, Explore, general-purpose, etc.), instruct them to use `vex` via Bash for code search (e.g. include "Use `vex search`, `vex show`, `vex usages`, `vex callers` via Bash for code search; ast-index is the fallback"). Don't tell them to default to grep/Glob.
**Fallback — `ast-index`** (use only when vex is unavailable):
```bash
ast-index search "Query" # Universal search
@@ -28,8 +59,21 @@ ast-index changed --base master # Show symbols changed in current bran
## Project Structure
- `/server` — Python FastAPI backend (see [server/CLAUDE.md](server/CLAUDE.md))
- `/android` — Android TV app (Kotlin shell + embedded Python via Chaquopy)
- `/contexts` — Context files for Claude (frontend conventions, graph editor, Chrome tools, server ops, demo mode)
## Android Dependency Sync (CRITICAL)
The Android app (`android/app/build.gradle.kts`) installs the server package with `--no-deps` and lists Android-compatible dependencies **explicitly** in the Chaquopy `pip {}` block. This is because `server/pyproject.toml` includes desktop-only packages (mss, psutil, sounddevice, etc.) that have no Android wheels.
**When adding a new dependency to `server/pyproject.toml`:**
1. If the package is **pure Python or has Chaquopy wheels** (check [Chaquopy PyPI](https://chaquo.com/pypi-13.1/)), also add it to `android/app/build.gradle.kts` in the `pip { install(...) }` block
2. If the package is **desktop-only** (native C/Rust extension without Android support), do NOT add it to `build.gradle.kts` — and guard its import with `try/except ImportError` in Python code
3. If unsure, check Chaquopy's package index first
**Incident context:** Chaquopy's pip runs on the build machine (Windows), not on Android. Platform markers like `sys_platform != 'linux'` evaluate against the BUILD host, not the target device. `pip install --exclude` does not exist. The only reliable way to exclude packages is to not list them.
## Context Files
| File | When to read |
@@ -42,10 +86,6 @@ ast-index changed --base master # Show symbols changed in current bran
| [Gitea Python CI/CD Guide](https://git.dolgolyov-family.by/alexei.dolgolyov/claude-code-facts/src/branch/main/gitea-python-ci-cd.md) | Reusable CI/CD patterns: Gitea Actions, cross-build, NSIS, Docker |
| [server/CLAUDE.md](server/CLAUDE.md) | Backend architecture, API patterns, common tasks |
## Task Tracking via TODO.md
Use `TODO.md` in the project root as the primary task tracker. **Do NOT use the TodoWrite tool** — all progress tracking goes through `TODO.md`.
## Documentation Lookup
**Use context7 MCP tools for library/framework documentation lookups** (FastAPI, OpenCV, Pydantic, yt-dlp, etc.) instead of relying on potentially outdated training data.
@@ -91,3 +131,42 @@ Do NOT commit code that fails linting or tests. Fix the issues first.
- Follow existing code style and patterns
- Update documentation when changing behavior
- Never make commits or pushes without explicit user approval
<!-- code-review-graph MCP tools -->
## MCP Tools: code-review-graph
**IMPORTANT: This project has a knowledge graph. ALWAYS use the
code-review-graph MCP tools BEFORE using Grep/Glob/Read to explore
the codebase.** The graph is faster, cheaper (fewer tokens), and gives
you structural context (callers, dependents, test coverage) that file
scanning cannot.
### When to use graph tools FIRST
- **Exploring code**: `semantic_search_nodes` or `query_graph` instead of Grep
- **Understanding impact**: `get_impact_radius` instead of manually tracing imports
- **Code review**: `detect_changes` + `get_review_context` instead of reading entire files
- **Finding relationships**: `query_graph` with callers_of/callees_of/imports_of/tests_for
- **Architecture questions**: `get_architecture_overview` + `list_communities`
Fall back to Grep/Glob/Read **only** when the graph doesn't cover what you need.
### Key Tools
| Tool | Use when |
|------|----------|
| `detect_changes` | Reviewing code changes — gives risk-scored analysis |
| `get_review_context` | Need source snippets for review — token-efficient |
| `get_impact_radius` | Understanding blast radius of a change |
| `get_affected_flows` | Finding which execution paths are impacted |
| `query_graph` | Tracing callers, callees, imports, tests, dependencies |
| `semantic_search_nodes` | Finding functions/classes by name or keyword |
| `get_architecture_overview` | Understanding high-level codebase structure |
| `refactor_tool` | Planning renames, finding dead code |
### Workflow
1. The graph auto-updates on file changes (via hooks).
2. Use `detect_changes` for code review.
3. Use `get_affected_flows` to understand impact.
4. Use `query_graph` pattern="tests_for" to check coverage.
+4 -4
View File
@@ -9,8 +9,8 @@
## Development Setup
```bash
git clone https://git.dolgolyov-family.by/alexei.dolgolyov/wled-screen-controller-mixed.git
cd wled-screen-controller-mixed/server
git clone https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab.git
cd ledgrab/server
# Python environment
python -m venv venv
@@ -29,7 +29,7 @@ npm run build
cd server
export PYTHONPATH=$(pwd)/src # Linux/Mac
# set PYTHONPATH=%CD%\src # Windows
python -m wled_controller.main
python -m ledgrab.main
```
Open http://localhost:8080 to access the dashboard.
@@ -55,7 +55,7 @@ ruff check src/ tests/
## Frontend Changes
After modifying any file under `server/src/wled_controller/static/js/` or `static/css/`, rebuild the bundle:
After modifying any file under `server/src/ledgrab/static/js/` or `static/css/`, rebuild the bundle:
```bash
cd server
+29 -80
View File
@@ -1,15 +1,17 @@
# Installation Guide
Complete installation guide for LED Grab (WLED Screen Controller) server and Home Assistant integration.
Complete installation guide for the LedGrab server.
## Table of Contents
1. [Docker Installation (recommended)](#docker-installation)
2. [Manual Installation](#manual-installation)
3. [First-Time Setup](#first-time-setup)
4. [Home Assistant Integration](#home-assistant-integration)
5. [Configuration Reference](#configuration-reference)
6. [Troubleshooting](#troubleshooting)
4. [Configuration Reference](#configuration-reference)
5. [Troubleshooting](#troubleshooting)
> **Home Assistant integration** has moved to a separate repository:
> [ledgrab-haos-integration](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab-haos-integration)
---
@@ -20,8 +22,8 @@ The fastest way to get running. Requires [Docker](https://docs.docker.com/get-do
1. **Clone and start:**
```bash
git clone https://git.dolgolyov-family.by/alexei.dolgolyov/wled-screen-controller-mixed.git
cd wled-screen-controller/server
git clone https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab.git
cd ledgrab/server
docker compose up -d
```
@@ -54,7 +56,7 @@ cd server
docker build -t ledgrab .
docker run -d \
--name wled-screen-controller \
--name ledgrab \
-p 8080:8080 \
-v $(pwd)/data:/app/data \
-v $(pwd)/logs:/app/logs \
@@ -84,8 +86,8 @@ Screen capture from inside a container requires X11 access. Uncomment `network_m
1. **Clone the repository:**
```bash
git clone https://git.dolgolyov-family.by/alexei.dolgolyov/wled-screen-controller-mixed.git
cd wled-screen-controller/server
git clone https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab.git
cd ledgrab/server
```
2. **Build the frontend bundle:**
@@ -95,7 +97,7 @@ Screen capture from inside a container requires X11 access. Uncomment `network_m
npm run build
```
This compiles TypeScript and bundles JS/CSS into `src/wled_controller/static/dist/`.
This compiles TypeScript and bundles JS/CSS into `src/ledgrab/static/dist/`.
3. **Create a virtual environment:**
@@ -123,7 +125,6 @@ Screen capture from inside a container requires X11 access. Uncomment `network_m
```bash
pip install ".[perf]" # DXCam, BetterCam, WGC (Windows only)
pip install ".[notifications]" # OS notification capture
pip install ".[tray]" # System tray icon (Windows only)
pip install ".[dev]" # pytest, black, ruff (development)
```
@@ -132,11 +133,11 @@ Screen capture from inside a container requires X11 access. Uncomment `network_m
```bash
# Linux / macOS
export PYTHONPATH=$(pwd)/src
uvicorn wled_controller.main:app --host 0.0.0.0 --port 8080
uvicorn ledgrab.main:app --host 0.0.0.0 --port 8080
# Windows (cmd)
set PYTHONPATH=%CD%\src
uvicorn wled_controller.main:app --host 0.0.0.0 --port 8080
uvicorn ledgrab.main:app --host 0.0.0.0 --port 8080
```
6. **Verify:** open <http://localhost:8080> in your browser.
@@ -161,7 +162,7 @@ auth:
Option B -- set an environment variable:
```bash
export WLED_AUTH__API_KEYS__dev="your-secure-key-here"
export LEDGRAB_AUTH__API_KEYS__dev="your-secure-key-here"
```
Generate a random key:
@@ -185,7 +186,7 @@ server:
Or via environment variable:
```bash
WLED_SERVER__CORS_ORIGINS='["http://localhost:8080","http://192.168.1.100:8080"]'
LEDGRAB_SERVER__CORS_ORIGINS='["http://localhost:8080","http://192.168.1.100:8080"]'
```
### Discover devices
@@ -194,57 +195,12 @@ Open the dashboard and go to the **Devices** tab. Click **Discover** to find WLE
---
## Home Assistant Integration
### Option 1: HACS (recommended)
1. Install [HACS](https://hacs.xyz/docs/setup/download) if you have not already.
2. Open HACS in Home Assistant.
3. Click the three-dot menu, then **Custom repositories**.
4. Add URL: `https://git.dolgolyov-family.by/alexei.dolgolyov/wled-screen-controller-mixed`
5. Set category to **Integration** and click **Add**.
6. Search for "WLED Screen Controller" in HACS and click **Download**.
7. Restart Home Assistant.
8. Go to **Settings > Devices & Services > Add Integration** and search for "WLED Screen Controller".
9. Enter your server URL (e.g., `http://192.168.1.100:8080`) and API key.
### Option 2: Manual
Copy the `custom_components/wled_screen_controller/` folder from this repository into your Home Assistant `config/custom_components/` directory, then restart Home Assistant and add the integration as above.
### Automation example
```yaml
automation:
- alias: "Start ambient lighting when TV turns on"
trigger:
- platform: state
entity_id: media_player.living_room_tv
to: "on"
action:
- service: switch.turn_on
target:
entity_id: switch.living_room_tv_processing
- alias: "Stop ambient lighting when TV turns off"
trigger:
- platform: state
entity_id: media_player.living_room_tv
to: "off"
action:
- service: switch.turn_off
target:
entity_id: switch.living_room_tv_processing
```
---
## Configuration Reference
The server reads configuration from three sources (in order of priority):
1. **Environment variables** -- prefix `WLED_`, double underscore as nesting delimiter (e.g., `WLED_SERVER__PORT=9090`)
2. **YAML config file** -- `server/config/default_config.yaml` (or set `WLED_CONFIG_PATH` to override)
1. **Environment variables** -- prefix `LEDGRAB_`, double underscore as nesting delimiter (e.g., `LEDGRAB_SERVER__PORT=9090`)
2. **YAML config file** -- `server/config/default_config.yaml` (or set `LEDGRAB_CONFIG_PATH` to override)
3. **Built-in defaults**
See [`server/.env.example`](server/.env.example) for every available variable with descriptions.
@@ -253,14 +209,14 @@ See [`server/.env.example`](server/.env.example) for every available variable wi
| Variable | Default | Description |
| -------- | ------- | ----------- |
| `WLED_SERVER__PORT` | `8080` | HTTP listen port |
| `WLED_SERVER__LOG_LEVEL` | `INFO` | `DEBUG`, `INFO`, `WARNING`, `ERROR` |
| `WLED_SERVER__CORS_ORIGINS` | `["http://localhost:8080"]` | Allowed CORS origins (JSON array) |
| `WLED_AUTH__API_KEYS` | `{"dev":"development-key..."}` | API keys (JSON object) |
| `WLED_STORAGE__DATABASE_FILE` | `data/ledgrab.db` | SQLite database path |
| `WLED_MQTT__ENABLED` | `false` | Enable MQTT for HA auto-discovery |
| `WLED_MQTT__BROKER_HOST` | `localhost` | MQTT broker address |
| `WLED_DEMO` | `false` | Enable demo mode (sandbox with virtual devices) |
| `LEDGRAB_SERVER__PORT` | `8080` | HTTP listen port |
| `LEDGRAB_SERVER__LOG_LEVEL` | `INFO` | `DEBUG`, `INFO`, `WARNING`, `ERROR` |
| `LEDGRAB_SERVER__CORS_ORIGINS` | `["http://localhost:8080"]` | Allowed CORS origins (JSON array) |
| `LEDGRAB_AUTH__API_KEYS` | `{"dev":"development-key..."}` | API keys (JSON object) |
| `LEDGRAB_STORAGE__DATABASE_FILE` | `data/ledgrab.db` | SQLite database path |
| `LEDGRAB_MQTT__ENABLED` | `false` | Enable MQTT for HA auto-discovery |
| `LEDGRAB_MQTT__BROKER_HOST` | `localhost` | MQTT broker address |
| `LEDGRAB_DEMO` | `false` | Enable demo mode (sandbox with virtual devices) |
---
@@ -277,7 +233,7 @@ python --version # must be 3.11+
**Check the frontend bundle exists:**
```bash
ls server/src/wled_controller/static/dist/app.bundle.js
ls server/src/ledgrab/static/dist/app.bundle.js
```
If missing, run `cd server && npm ci && npm run build`.
@@ -289,7 +245,7 @@ If missing, run `cd server && npm ci && npm run build`.
docker compose logs -f
# Manual install
tail -f logs/wled_controller.log
tail -f logs/ledgrab.log
```
### Cannot access the dashboard from another machine
@@ -298,13 +254,6 @@ tail -f logs/wled_controller.log
2. Check your firewall allows inbound traffic on port 8080.
3. Add your server's LAN IP to `cors_origins` (see [Configure CORS](#configure-cors-for-lan-access) above).
### Home Assistant integration not appearing
1. Verify HACS installed the component: check that `config/custom_components/wled_screen_controller/` exists.
2. Clear your browser cache.
3. Restart Home Assistant.
4. Check logs at **Settings > System > Logs** and search for `wled_screen_controller`.
### WLED device not responding
1. Confirm the device is powered on and connected to Wi-Fi.
@@ -325,4 +274,4 @@ tail -f logs/wled_controller.log
- [API Documentation](docs/API.md)
- [Calibration Guide](docs/CALIBRATION.md)
- [Repository Issues](https://git.dolgolyov-family.by/alexei.dolgolyov/wled-screen-controller-mixed/issues)
- [Repository Issues](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/issues)
+152 -112
View File
@@ -1,36 +1,58 @@
# LED Grab
Ambient lighting system that captures screen content and drives LED strips in real time. Supports WLED, Adalight, AmbileD, and DDP devices with audio-reactive effects, pattern generation, and automated profile switching.
Ambient lighting system that captures screen content and drives LED strips and smart lights in real time. Supports a wide range of devices — WLED, DDP, Adalight, smart bulbs, PC peripherals, Bluetooth strips, and more — with audio-reactive effects, pattern generation, and condition-based automation.
**Free and open source.** LedGrab is released under the [MIT license](LICENSE) — free to use, modify, and self-host, with no accounts, telemetry, or cloud dependency. Everything runs locally on your own machine and network.
## What It Does
The server captures pixels from a screen (or Android device via ADB), extracts border colors, applies post-processing filters, and streams the result to LED strips at up to 60 fps. A built-in web dashboard provides device management, calibration, live LED preview, and real-time metrics — no external UI required.
The server captures pixels from a screen (or from a connected Android phone via ADB), extracts border colors, applies a post-processing filter pipeline, and streams the result to your LED devices at up to 60 fps. A built-in web dashboard provides device management, calibration, a visual wiring editor, live LED preview, and real-time metrics — no external UI required.
A Home Assistant integration exposes devices as entities for smart home automation.
A separate Home Assistant integration exposes devices as entities for smart-home automation.
## Screenshots
![LedGrab dashboard](docs/screenshots/dashboard.PNG)
*Dashboard — live system performance, integrations, automations, and scene presets at a glance.*
![Channels view](docs/screenshots/dashboard-targets.PNG)
*Channels — start, stop, and monitor each source-to-device pipeline with live FPS.*
![Live capture preview](docs/screenshots/test-preview-capture.PNG)
*Live preview — inspect the processed capture output in real time before it reaches the LEDs.*
## Features
### Screen Capture
- Multi-monitor support with per-target display selection
- 6 capture engine backends MSS (cross-platform), DXCam, BetterCam, Windows Graphics Capture (Windows), Scrcpy (Android via ADB), Camera/Webcam (OpenCV)
- Capture engine backends: MSS (cross-platform), DXCam, BetterCam, Windows Graphics Capture (Windows only), and Camera/Webcam (OpenCV)
- Capture from a connected Android phone's screen via scrcpy (ADB) — the device is a *source*; LedGrab itself runs on your desktop
- Configurable capture regions, FPS, and border width
- Capture templates for reusable configurations
- Reusable capture templates
### LED Device Support
- WLED (HTTP/UDP) with mDNS auto-discovery
- Adalight (serial) — Arduino-compatible LED controllers
- AmbileD (serial)
- DDP (Distributed Display Protocol, UDP)
- OpenRGB — PC peripherals (keyboard, mouse, RAM, fans, LED strips)
- Serial port auto-detection and baud rate configuration
LedGrab speaks many protocols, so a single setup can drive everything from a DIY strip to off-the-shelf smart bulbs:
![Device type picker](docs/screenshots/devices.PNG)
- **Network LED controllers** — WLED (HTTP/UDP, with mDNS auto-discovery), DDP (Pixelblaze, ESPixelStick, Falcon), Open Pixel Control (OPC), Art-Net / sACN (E1.31), ESP-NOW, and generic WebSocket streaming
- **Serial / direct hardware** — Adalight (Arduino-compatible), AmbiLED, SPI-attached strips (e.g. WS2812B), and USB HID controllers
- **Smart bulbs & panels** — Philips Hue (Entertainment API), Nanoleaf, Yeelight, WiZ, LIFX, and Govee (Wi-Fi LAN)
- **Bluetooth LE strips** — SP110E, Triones / HappyLighting, Zengge, and Govee BLE
- **PC peripherals** — OpenRGB, Razer Chroma, and SteelSeries GameSense (keyboards, mice, RAM, fans, etc.)
- **Device groups** — combine multiple devices into one logical target
- Serial port auto-detection and baud-rate configuration
### Color Processing
- Post-processing filter pipeline: brightness, gamma, saturation, color correction, auto-crop, frame interpolation, pixelation, flip
- Post-processing filter pipeline: brightness, gamma, saturation, color correction, auto-crop, frame interpolation, pixelation, flip, and more
- Reusable post-processing templates
- Color strip sources: audio-reactive, pattern generator, composite layering, audio-to-color mapping
- Color strip sources: audio-reactive, pattern generator, gradients, composite layering, and audio-to-color mapping
- Pattern templates with customizable effects
### Audio Integration
@@ -38,17 +60,20 @@ A Home Assistant integration exposes devices as entities for smart home automati
- Multichannel audio capture from any system device (input or loopback)
- WASAPI engine on Windows, Sounddevice (PortAudio) engine on Linux/macOS
- Per-channel mono extraction
- Audio-reactive color strip sources driven by frequency analysis
- Audio filter / processing pipeline feeding audio-reactive color sources driven by frequency analysis
### Automation
- Profile engine with condition-based switching (time of day, active window, etc.)
- Dynamic brightness value sources (schedule-based, scene-aware)
- Key Colors (KC) targets with live WebSocket color streaming
- Automations engine with condition-based rules — switch targets, scenes, or brightness by time of day, active window/process, MQTT, webhooks, or game events
- Scene presets for one-click lighting changes
- Dynamic value sources for brightness and other parameters (schedule-based, weather-based, scene-aware)
- Weather sources, clock sync, webhooks, and inbound/outbound HTTP endpoints
- Game integration adapters (e.g. League of Legends)
### Dashboard
- Web UI at `http://localhost:8080` — no installation needed on the client side
- Web UI at `http://localhost:8080` — nothing to install on the client side
- Visual node-graph editor for wiring sources → processing → targets
- Progressive Web App (PWA) — installable on phones and tablets with offline caching
- Responsive mobile layout with bottom tab navigation
- Device management with auto-discovery wizard
@@ -57,38 +82,76 @@ A Home Assistant integration exposes devices as entities for smart home automati
- Real-time FPS, latency, and uptime charts
- Localized in English, Russian, and Chinese
### Activity Log
The **Activity** tab is a persistent, queryable audit log of everything LedGrab has done — entity changes, auth events, device connections, and system actions.
- Filter by category (auth, device, entity, capture, system), severity, actor, entity type, date range, or free text
- Live-append of new events as they happen
- Export as CSV or JSON (authentication required)
- Entity crosslinks navigate directly to the relevant card
- **Retention settings** (Settings → Activity Log): configure max age, max entry count, and toggle recording on/off
- **Clear log** (Settings → Activity Log, requires authentication) — audited: a system entry records who cleared the log and when
> **Note:** The Activity Log is distinct from the **debug Log Viewer** (Settings → General → Open Log Viewer). The Log Viewer is an ephemeral real-time tail of the server's Python log stream (WARNING/ERROR lines, resets on disconnect). The Activity Log is a structured, persistent SQLite-backed record of semantic application events.
### Home Assistant Integration
- HACS-compatible custom component
- HACS-compatible custom component (separate repository)
- Light, switch, sensor, and number entities per device
- Real-time metrics via data coordinator
- Real-time metrics via a data coordinator
- WebSocket-based live LED preview in HA
## Platforms
LedGrab runs as a desktop / server application:
| Platform | Status | Notes |
| -------- | ------ | ----- |
| Windows | ✅ Supported | Installer (`.exe`) and portable ZIP; all capture/audio backends |
| Linux | ✅ Supported | Tarball and Docker image; X11 capture (Wayland in-container capture not supported) |
| macOS | ✅ Supported | Runs from source / Docker; MSS capture |
| Docker | ✅ Supported | Multi-arch container image |
| Android (TV) | ⚠️ Experimental | An on-device Android-TV build exists (APK attached to releases) but is emulator-verified only and **not officially supported** |
> **There is no production Android app.** Android phones are only supported as a *capture source* (via scrcpy/ADB) from a desktop host. The on-device Android-TV build is experimental.
### Feature support by OS
| Feature | Windows | Linux / macOS | Android TV (experimental) |
| ------- | ------- | ------------- | ------------------------- |
| Screen capture | DXCam, BetterCam, WGC, MSS | MSS | MediaProjection; root `screenrecord` (rooted devices) |
| Webcam capture | OpenCV (DirectShow) | OpenCV (V4L2) | Camera2 (on-demand, while capture is running) |
| Audio capture | WASAPI, Sounddevice | Sounddevice (PulseAudio/PipeWire) | AudioPlaybackCapture (API 29+) |
| GPU monitoring | NVIDIA (nvidia-ml-py) | NVIDIA (nvidia-ml-py) | — (CPU/RAM/battery/thermal via `/proc`) |
| Capture from Android phone | scrcpy (ADB) | scrcpy (ADB) | — (captures its own screen instead) |
| Notification capture | WinRT | dbus (Linux) | NotificationListenerService |
| Monitor names | Friendly names (WMI) | Generic ("Display 0") | Single built-in display |
| LED transports | Network, USB-serial, BLE | Network, USB-serial, BLE | Network, USB-serial (Android driver), BLE (Android bridge) |
| Automation: window/process conditions | Supported | Partial | Foreground-app condition (UsageStatsManager) |
## Requirements
- Python 3.11+ (or Docker)
- A supported LED device on the local network or connected via USB
- Windows, Linux, or macOS — all core features work cross-platform
### Platform Notes
| Feature | Windows | Linux / macOS |
| ------- | ------- | ------------- |
| Screen capture | DXCam, BetterCam, WGC, MSS | MSS |
| Webcam capture | OpenCV (DirectShow) | OpenCV (V4L2) |
| Audio capture | WASAPI, Sounddevice | Sounddevice (PulseAudio/PipeWire) |
| GPU monitoring | NVIDIA (pynvml) | NVIDIA (pynvml) |
| Android capture | Scrcpy (ADB) | Scrcpy (ADB) |
| Monitor names | Friendly names (WMI) | Generic ("Display 0") |
| Profile conditions | Process/window detection | Not yet implemented |
- A supported LED device on the local network, connected via USB/serial, or reachable over Bluetooth
- Windows, Linux, or macOS
## Quick Start
### Docker (recommended)
### Prebuilt downloads
Grab a ready-to-run build from the [Releases page](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/releases):
- **Windows** — `LedGrab-<version>-setup.exe` (installer, no admin required) or `LedGrab-<version>-win-x64.zip` (portable)
- **Linux** — `LedGrab-<version>-linux-x64.tar.gz`
- **Docker** — see below
- **Android TV** — `.apk` (experimental, see [Platforms](#platforms))
### Docker (recommended for servers)
```bash
git clone https://git.dolgolyov-family.by/alexei.dolgolyov/wled-screen-controller-mixed.git
cd wled-screen-controller/server
git clone https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab.git
cd ledgrab/server
docker compose up -d
```
@@ -97,8 +160,8 @@ docker compose up -d
Requires Python 3.11+ and Node.js 18+.
```bash
git clone https://git.dolgolyov-family.by/alexei.dolgolyov/wled-screen-controller-mixed.git
cd wled-screen-controller/server
git clone https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab.git
cd ledgrab/server
# Build the frontend bundle
npm ci && npm run build
@@ -112,97 +175,76 @@ pip install .
# Start the server
export PYTHONPATH=$(pwd)/src # Linux/Mac
# set PYTHONPATH=%CD%\src # Windows
uvicorn wled_controller.main:app --host 0.0.0.0 --port 8080
uvicorn ledgrab.main:app --host 0.0.0.0 --port 8080
```
Open **http://localhost:8080** to access the dashboard.
Open <http://localhost:8080> to access the dashboard.
> **Important:** The default API key is `development-key-change-in-production`. Change it before exposing the server outside localhost. See [INSTALLATION.md](INSTALLATION.md) for details.
> **Network access:** By default, LedGrab allows anonymous access only from `localhost`. Any request from another machine on your LAN is rejected unless you configure an API key (`auth.api_keys`). Set a key before exposing the server on your network — see [INSTALLATION.md](INSTALLATION.md).
See [INSTALLATION.md](INSTALLATION.md) for the full installation guide, including configuration, Docker manual builds, and Home Assistant setup.
See [INSTALLATION.md](INSTALLATION.md) for the full installation guide, including configuration, Docker manual builds, and CORS setup.
## Architecture
## Demo Mode
```text
wled-screen-controller/
├── server/ # Python FastAPI backend
│ ├── src/wled_controller/
│ │ ├── main.py # Application entry point
│ │ ├── config.py # YAML + env var configuration
│ │ ├── api/
│ │ │ ├── routes/ # REST + WebSocket endpoints
│ │ │ └── schemas/ # Pydantic request/response models
│ │ ├── core/
│ │ │ ├── capture/ # Screen capture, calibration, pixel processing
│ │ │ ├── capture_engines/ # MSS, DXCam, BetterCam, WGC, Scrcpy, Camera backends
│ │ │ ├── devices/ # WLED, Adalight, AmbileD, DDP, OpenRGB clients
│ │ │ ├── audio/ # Audio capture engines
│ │ │ ├── filters/ # Post-processing filter pipeline
│ │ │ ├── processing/ # Stream orchestration and target processors
│ │ │ └── profiles/ # Condition-based profile automation
│ │ ├── storage/ # JSON-based persistence layer
│ │ ├── static/ # Web dashboard (vanilla JS, CSS, HTML)
│ │ │ ├── js/core/ # API client, state, i18n, modals, events
│ │ │ ├── js/features/ # Feature modules (devices, streams, targets, etc.)
│ │ │ ├── css/ # Stylesheets
│ │ │ └── locales/ # en.json, ru.json, zh.json
│ │ └── utils/ # Logging, monitor detection
│ ├── config/ # default_config.yaml
│ ├── tests/ # pytest suite
│ ├── Dockerfile
│ └── docker-compose.yml
├── custom_components/ # Home Assistant integration (HACS)
│ └── wled_screen_controller/
├── docs/
│ ├── API.md # REST API reference
│ └── CALIBRATION.md # LED calibration guide
├── INSTALLATION.md
└── LICENSE # MIT
Demo mode runs the server with virtual devices, sample data, and isolated storage — useful for exploring the UI without real hardware.
Set the `LEDGRAB_DEMO` environment variable to `true`, `1`, or `yes`:
```bash
# Docker
docker compose run -e LEDGRAB_DEMO=true server
# Python
LEDGRAB_DEMO=true uvicorn ledgrab.main:app --host 0.0.0.0 --port 8081
```
Demo mode uses port **8081**, config file `config/demo_config.yaml`, and stores data under `data/demo/` (separate from production data). It can run alongside the main server.
## Configuration
Edit `server/config/default_config.yaml` or use environment variables with the `WLED_` prefix:
Edit `server/config/default_config.yaml` or use environment variables with the `LEDGRAB_` prefix:
```yaml
server:
host: "0.0.0.0"
port: 8080
log_level: "INFO"
cors_origins:
- "http://localhost:8080"
auth:
api_keys:
dev: "development-key-change-in-production"
storage:
devices_file: "data/devices.json"
templates_file: "data/capture_templates.json"
# Empty (default) → loopback-only anonymous access; LAN requests are rejected.
# Add a key to enable LAN/remote access (generate one with: openssl rand -hex 32).
api_keys: {}
# api_keys:
# dev: "your-secret-key-here"
logging:
format: "json"
file: "logs/wled_controller.log"
file: "logs/ledgrab.log"
max_size_mb: 100
```
Environment variable override example: `WLED_SERVER__PORT=9090`.
- Application data is stored in a SQLite database (`data/ledgrab.db` by default). Set `LEDGRAB_DATA_DIR` to relocate the data root (database + assets).
- Environment variable override example: `LEDGRAB_SERVER__PORT=9090`.
See [INSTALLATION.md](INSTALLATION.md) and [`server/.env.example`](server/.env.example) for the full configuration reference.
## API
The server exposes a REST API (with Swagger docs at `/docs`) covering:
The server exposes a REST API (with interactive Swagger docs at `/docs`) plus WebSocket endpoints. Resources include:
- **Devices** — CRUD, discovery, validation, state, metrics
- **Capture Templates** — Screen capture configurations
- **Picture Sources** — Screen capture stream definitions
- **Picture Targets** — LED target management, start/stop processing
- **Post-Processing Templates** — Filter pipeline configurations
- **Color Strip Sources** — Audio, pattern, composite, mapped sources
- **Audio Sources** — Multichannel and mono audio device configuration
- **Pattern Templates** — Effect pattern definitions
- **Value Sources** — Dynamic brightness/value providers
- **Key Colors Targets** — KC targets with WebSocket live color stream
- **Profiles** — Condition-based automation profiles
- **Capture Templates** & **Picture Sources** — screen capture configuration and stream definitions
- **Output Targets** — LED target management, start/stop processing, live color stream
- **Post-Processing Templates** — filter pipeline configurations
- **Color Strip Sources**, **Pattern Templates**, **Gradients** — color generation
- **Audio Sources / Templates / Filters** — audio capture and reactive processing
- **Value Sources**, **Weather Sources**, **Scene Presets** — dynamic parameters and presets
- **Automations**, **Webhooks**, **HTTP Endpoints**, **Game Integration** — triggers and rules
- **MQTT** & **Home Assistant**broker sources and HA integration
All endpoints require API key authentication via `X-API-Key` header or `?token=` query parameter.
Authentication uses a Bearer token (`Authorization: Bearer <api-key>`) when API keys are configured; loopback requests are anonymous by default. WebSocket connections authenticate via a first-message handshake.
See [docs/API.md](docs/API.md) for the full reference.
@@ -214,9 +256,7 @@ See [docs/CALIBRATION.md](docs/CALIBRATION.md) for a step-by-step guide.
## Home Assistant
Install via HACS (add as a custom repository) or manually copy `custom_components/wled_screen_controller/` into your HA config directory. The integration creates light, switch, sensor, and number entities for each configured device.
See [INSTALLATION.md](INSTALLATION.md) for detailed setup instructions.
For Home Assistant integration, see the separate [ledgrab-haos-integration](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab-haos-integration) repository.
## Development
@@ -237,16 +277,16 @@ ruff check src/ tests/
Optional extras:
```bash
pip install -e ".[perf]" # High-performance capture engines (Windows)
pip install -e ".[camera]" # Webcam capture via OpenCV
pip install -e ".[perf]" # High-performance capture engines (Windows: DXCam, BetterCam, WGC)
pip install -e ".[notifications]" # OS notification capture (WinRT / dbus)
pip install -e ".[scrcpy]" # Capture from an Android phone via scrcpy
pip install -e ".[ble]" # Bluetooth LE LED controllers (desktop only)
```
## Contributing
Contributions are welcome. LedGrab is MIT-licensed, so you're free to fork, modify, and self-host. Please open an issue or pull request on the [repository](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab).
## License
MIT — see [LICENSE](LICENSE).
## Acknowledgments
- [WLED](https://github.com/Aircoookie/WLED) — LED control firmware
- [FastAPI](https://fastapi.tiangolo.com/) — Python web framework
- [MSS](https://python-mss.readthedocs.io/) — Cross-platform screen capture
MIT — see [LICENSE](LICENSE). Free and open source.
+91 -6
View File
@@ -1,15 +1,100 @@
## v0.2.2 (2025-03-25)
## v0.8.2 (2026-06-08)
### Features
- Add 4 built-in gradients, searchable gradient picker, cleaner modal titles ([a5e7a4e](https://git.dolgolyov-family.by/alexei.dolgolyov/wled-screen-controller-mixed/commit/a5e7a4e))
### User-facing changes
#### Features
##### WLED native realtime UDP output
- New realtime UDP sink speaking WLED's **DRGB / DRGBW / DNRGB** protocols, with automatic revert to the device's prior state when streaming stops ([7728aec](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/7728aec))
##### Automatic brightness limiting (ABL) / power budget
- Per-LED power budgeting that caps total draw by scaling brightness to a configurable current/PSU limit, preventing brownouts on long strips ([ffee156](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/ffee156))
##### Scene playlists
- Scenes can be grouped into **playlists with timed auto-cycling**, so a target can rotate through looks on a schedule ([f71e10e](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/f71e10e))
- Playlist + cycling state is included in the aggregated `/snapshot` response ([abc204c](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/abc204c))
##### Auto edge-calibration + guided first-run setup wizard
- Backend core for **automatic screen-edge calibration** ([0409cd8](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/0409cd8)), a one-call setup scaffold with an onboarding flag ([9dcd76d](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/9dcd76d)), and a browser-driven calibration UI ([9550688](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/9550688))
- A **guided first-run setup wizard** ties it together for new installs ([81b1808](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/81b1808)), with all-provider source discovery and a spatial corner picker ([dd43f38](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/dd43f38))
##### Region-of-interest (ROI) screen capture
- Screen sampling can now be cropped to a **region of interest** instead of the whole display ([ca59546](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/ca59546))
##### Built-in "look" presets
- One-click looks: **Cinematic / Vivid / Cozy / Soft / Cool** ([e18d56c](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/e18d56c))
##### Weekday + timezone scheduling
- The time-of-day automation rule now supports **weekday selection and explicit timezones** ([1ada5ac](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/1ada5ac))
##### Value sources
- New **sandboxed-Jinja template combinator** for composing value sources ([6de61b9](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/6de61b9)) and optional normalization for magnitude sources ([669ae20](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/669ae20))
##### Visual graph editor
- The editor is now a **full wiring control surface** ([2e51f46](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/2e51f46)), and you can **duplicate a selected subgraph** server-side ([15cfb82](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/15cfb82))
##### Android on-device capture
- **System audio playback capture** ([fd62db1](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/fd62db1)), **OS notification capture** via NotificationListenerService ([0be3f83](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/0be3f83)), **webcam capture** via Camera2 ([4bf3fe6](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/4bf3fe6)), and a **foreground-app automation condition** ([1c1bbe2](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/1c1bbe2))
#### Bug Fixes
- **Security:** removed an active **weak default API key** from the shipped config — fresh installs no longer ship with a guessable key. Set your own key on first run ([5686ae5](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/5686ae5))
- Removed a broken legacy `/system/mqtt/settings` route ([fdc9201](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/fdc9201))
- Scene brightness value-source changes now sync to the live processor immediately ([02e2ea3](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/02e2ea3))
- Wizard hardening: scaffolded targets are registered with the ProcessorManager and the final review step is more robust ([6cd5e05](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/6cd5e05))
- Installer opens the WebUI only once after "Launch LedGrab" ([05cf121](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/05cf121))
---
### Development / Internal
#### Backend / Storage
- `clone()` is now gated behind an **opt-in allowlist**, with expanded duplicate-handling tests ([498854f](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/498854f))
#### Frontend
- In-progress dashboard customization groundwork ([6180569](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/6180569))
#### Docs
- Actualized README + API reference with embedded screenshots ([12b40e6](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/12b40e6)), graph-editor wiring-control roadmap ([d505388](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/d505388)), Android audio-capture design notes ([4b2e8fc](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/4b2e8fc)); removed stale ANDROID-REVIEW planning docs ([9960f15](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/9960f15))
#### Tests
- Large new suites for calibration solver/session (incl. adversarial), setup & scene-playlist routes, playlist engine, and ROI capture. Full suite: **2149 passing, 2 skipped**
---
<details>
<summary>All Commits</summary>
<summary>All Commits (31)</summary>
| Hash | Message | Author |
|------|---------|--------|
| [a5e7a4e](https://git.dolgolyov-family.by/alexei.dolgolyov/wled-screen-controller-mixed/commit/a5e7a4e) | feat: add 4 built-in gradients, searchable gradient picker, cleaner modal titles | alexei.dolgolyov |
| ---- | ------- | ------ |
| [dd43f38](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/dd43f38) | fix(calibration-wizard): all-provider discovery + spatial corner picker | alexei.dolgolyov |
| [6cd5e05](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/6cd5e05) | fix(setup): register scaffolded target with ProcessorManager + final-review hardening | alexei.dolgolyov |
| [81b1808](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/81b1808) | feat(onboarding): guided first-run setup wizard (phase 4, final) | alexei.dolgolyov |
| [abc204c](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/abc204c) | feat(snapshot): include scene playlists + cycling state in snapshot | alexei.dolgolyov |
| [9550688](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/9550688) | feat(calibration): browser-driven auto edge-calibration UI (phase 3) | alexei.dolgolyov |
| [9dcd76d](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/9dcd76d) | feat(setup): one-call setup scaffold + onboarding flag (phase 2) | alexei.dolgolyov |
| [0409cd8](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/0409cd8) | feat(calibration): auto edge-calibration backend core (phase 1) | alexei.dolgolyov |
| [6180569](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/6180569) | wip(dashboard): in-progress dashboard customization changes | alexei.dolgolyov |
| [f71e10e](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/f71e10e) | feat(scenes): scene playlists with timed auto-cycling | alexei.dolgolyov |
| [ca59546](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/ca59546) | feat(capture): region-of-interest (ROI) crop for screen sampling | alexei.dolgolyov |
| [1ada5ac](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/1ada5ac) | feat(automations): weekday + timezone scheduling for time-of-day rule | alexei.dolgolyov |
| [e18d56c](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/e18d56c) | feat(processing): built-in 'look' presets (Cinematic/Vivid/Cozy/Soft/Cool) | alexei.dolgolyov |
| [7728aec](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/7728aec) | feat(wled): native realtime UDP output (DRGB/DRGBW/DNRGB) with auto-revert | alexei.dolgolyov |
| [ffee156](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/ffee156) | feat(targets): automatic brightness limiting (ABL) / per-LED power budget | alexei.dolgolyov |
| [02e2ea3](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/02e2ea3) | fix(scenes): sync brightness value-source change to live processor | alexei.dolgolyov |
| [fdc9201](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/fdc9201) | fix(api): remove broken legacy /system/mqtt/settings route | alexei.dolgolyov |
| [5686ae5](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/5686ae5) | fix(security): remove active weak default API key from shipped config | alexei.dolgolyov |
| [9960f15](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/9960f15) | docs(android): remove ANDROID-REVIEW planning/review docs | alexei.dolgolyov |
| [1c1bbe2](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/1c1bbe2) | feat(android): foreground-app automation condition | alexei.dolgolyov |
| [4bf3fe6](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/4bf3fe6) | feat(android): on-device webcam capture via Camera2 (AndroidCameraEngine) | alexei.dolgolyov |
| [0be3f83](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/0be3f83) | feat(android): on-device OS notification capture (NotificationListenerService) | alexei.dolgolyov |
| [4b2e8fc](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/4b2e8fc) | docs(android): add audio-capture design + missing-functionality review | alexei.dolgolyov |
| [fd62db1](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/fd62db1) | feat(audio): Android on-device system playback capture | alexei.dolgolyov |
| [669ae20](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/669ae20) | feat(value-sources): optional normalization for magnitude sources | alexei.dolgolyov |
| [6de61b9](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/6de61b9) | feat(value-sources): add sandboxed-Jinja template combinator | alexei.dolgolyov |
| [12b40e6](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/12b40e6) | docs: actualize README and API reference, embed screenshots | alexei.dolgolyov |
| [498854f](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/498854f) | refactor(storage): gate clone() behind an opt-in allowlist; expand duplicate tests | alexei.dolgolyov |
| [15cfb82](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/15cfb82) | feat(graph): duplicate a selected subgraph server-side | alexei.dolgolyov |
| [2e51f46](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/2e51f46) | feat(graph): make the visual editor a full wiring control surface | alexei.dolgolyov |
| [05cf121](https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab/commit/05cf121) | fix(installer): open WebUI once after "Launch LedGrab" | alexei.dolgolyov |
</details>
+249
View File
@@ -0,0 +1,249 @@
# Dashboard Reconciliation — Review Notes
*Captured 2026-05-26. Session focused on dashboard + perf-card flicker and per-poll re-rendering.*
*Updated 2026-05-27 — widened the audit beyond the two poll timers and found a **second driver** (server push) plus the **highest-blast-radius site** (`entity-events.ts`). Added §3.5, corrected the "out of scope" reasoning in §5, and confirmed the decision: **commit to the Lit migration**. Implementation deferred — this is still a planning doc, not a spec.*
This is a thinking-aloud document for whoever picks up reconciliation work next (likely me). It captures the bug class, what's already shipped, what's still latent, the decision ladder we walked through, and the recommendation we landed on. It is **not** a spec — treat any code shown as illustrative.
---
## 1. The bug class in one sentence
> Every place a data-driven render — a poll timer **or** a server-pushed `server:*` event — writes `el.innerHTML = ...`, the existing DOM is torn down — even when the new HTML equals the old — which restarts CSS animations, drops focus, skips transitions, and burns wasted DOM mutation cycles.
The symptom only becomes visually loud when the destroyed subtree contains a CSS keyframe animation (e.g. the pulsing `.perf-patches-empty-dot`). Everywhere else the cost is silent: lost transitions, broken focus, wasted layout work. The bug is **load-bearing in the architecture**, not in any single call site — that's why we keep coming back to it.
---
## 2. What landed in commit `f6486f9` (this session)
Tactical work — solves the worst cases, does not change the architecture.
### `server/src/ledgrab/static/js/features/dashboard.ts`
- Collapsed the two fast-path branches into one. Fast path runs when `structureUnchanged && !forceFullRender` regardless of `running.length`. Previously, **zero running targets meant every poll rebuilt the entire dashboard** even when nothing changed.
- `_lastSyncClockIds` no longer fingerprints `is_running` — pausing/resuming a clock no longer tears down every card. `_updateSyncClocksInPlace` already handles the toggle.
- `_updateAutomationsInPlace` now called from the unified fast path. Automation badges were silently going stale on the fast path.
- `_initFpsCharts` rewritten diff-based: only destroy charts for ids that left or whose canvas was detached by a DOM swap; only create for new ids; only fetch `/api/metrics/history` when there are genuinely new ids needing seed data.
- Sync-clock pause/resume/reset callers + `server:automation_state_changed` SSE handler now use `loadDashboard()` (no force) — `forceFullRender` is now actually load-bearing, meaning "settings changed, full rebuild required."
### `server/src/ledgrab/static/js/features/perf-charts.ts`
- `_renderChartSvg` no longer rewrites `innerHTML` per poll. The SVG skeleton (ref line + sys area/line + app line) is built once via `_ensureSparkNodes` and mutated thereafter. WeakMap cache (`_sparkNodeCache`) keyed by host element avoids the per-tick `querySelector` cost.
- Hidden cards (env-disabled GPU/Temp) skip render entirely.
- `_fetchPerformance` switched to `fetchWithAuth`.
- Hardcoded English strings replaced with `t()` calls. New keys: `perf.no_captures`, `perf.captures_count.{one,few,many,other}`, `perf.ratio_of_requested`, `perf.total_count`, `perf.skipped_per_sec`, `perf.tip.now`, `perf.tip.ago` (en/ru/zh).
- Tooltip reads `dashboardPollInterval` per mousemove tick (was captured at bind time).
- Dead `<defs><linearGradient>` block removed.
- `updateTotalCaptureFpsActual` now delegates to `_paintCaptureFpsActualValue` — single code path.
- `updateActivePatches` / `updateDevices` skip the `innerHTML` write when content signature hasn't changed. This is the direct fix for the "READY TO LAUNCH flickers every update" report — the empty-state dot's CSS pulse no longer resets.
- Two missing semicolons in `_seedAggregateHistories` (ASI was saving us).
### Reviewer findings addressed (typescript-reviewer pass)
- **HIGH:** `_metricLabel` was looking up `dashboard.perf.${key}` but the FPS family uses `dashboard.perf.total_fps`, `total_capture_fps`, `total_capture_fps_actual`. Tooltip would have shouted `FPS` / `CAPTURE_FPS` / `CAPTURE_FPS_ACTUAL`. Fixed via explicit `METRIC_LABEL_KEYS` map.
- **HIGH:** `_ensureSparkNodes` silently coerced `null` children to non-null when the SVG existed but a child was missing. Hardened to validate all four children and rebuild if any are missing.
---
## 3. Hot spots still latent
These are the call sites where `innerHTML` is still written every poll. None are flickering today (no CSS animations on their inner elements), but every one is the same bug shape and will bite the next time someone adds a keyframe / transition / focus target inside.
### `perf-charts.ts`
| Line | Site | Fires per poll? | Notes |
|------|------|-----------------|-------|
| 462 | `updateActivePatches``listEl.innerHTML` | yes | guarded by signature compare (✓) |
| 493 | `updateTotalFps``valEl.innerHTML` | yes | FPS value, no inner animation |
| 526 | `updateTotalCaptureFps``valEl.innerHTML` | yes | same |
| 638 | `_paintNetworkValue``valEl.innerHTML` | yes | bytes/s value |
| 655 | `_paintDeviceLatencyValue``valEl.innerHTML` (no-devices hint) | yes | hint span |
| 657 | `_paintDeviceLatencyValue``valEl.innerHTML` (offline hint) | yes | hint span |
| 660 | `_paintDeviceLatencyValue``valEl.innerHTML` (ms value) | yes | value |
| 676 | `_paintSendTimingValue``valEl.innerHTML` (idle hint) | yes | hint span |
| 679 | `_paintSendTimingValue``valEl.innerHTML` (ms value) | yes | value |
| 738 | `_paintErrorsValue``valEl.innerHTML` | yes | rate value |
| 806 | `updateDevices``dotsEl.innerHTML` | yes | guarded by signature compare (✓) |
| 1086 | `_renderValuePair``mainEl.innerHTML = appVal` | yes | dual sys/app value |
| 1088 | `_renderValuePair``mainEl.innerHTML = sysVal` | yes | dual sys/app value |
| 1094 | `_renderValuePair``tagEl.innerHTML` (App tag) | mode='both' only | App tag in `both` mode |
| 1181 | `_applyPerfDataToDom` temp hint | only when cpu_temp_hint_key changes | rare |
| 1449 | `_paintFpsValue` | seed only | once per init |
| 1456 | `_paintCaptureFpsValue` | seed only | once per init |
| 1463 | `_paintCaptureFpsActualValue` (no-captures hint) | yes via live updater | now goes through painter |
| 1469 | `_paintCaptureFpsActualValue` (value) | yes via live updater | same |
| 1499 | `_paintErrorsValue` (duplicate of 738) | seed only | once per init |
| 1823 | tooltip `tip.innerHTML` | per mousemove | rate-limited by hover only |
### `dashboard.ts`
| Line | Site | Fires per poll? | Notes |
|------|------|-----------------|-------|
| 275 | `_updateRunningMetrics``fpsEl.innerHTML` | per running target | live FPS pill — visible churn |
| 293 | `_updateRunningMetrics``labelEl.innerHTML` (errors label) | per running target | rebuilt each poll |
| 340 | `_updateAutomationsInPlace``btn.innerHTML` | only on enable/disable change | low frequency |
| 366 | `_updateSyncClocksInPlace``btn.innerHTML` | per poll for every clock | wasteful |
| 975 | `loadDashboard` first-load → `container.innerHTML` | once per init | fine |
| 989 | `loadDashboard` slow path → `dynamic.innerHTML = dynamicHtml` | only when slow path fires | the **big** swap, scoped already |
| 1010 | `loadDashboard` error path | rare | fine |
| 1416 | `subscribeDashboardLayout` clear | rare | fine |
### What this list tells us
- The remaining innerHTML writes are **per-cell value updates** that paint formatted spans (`{value}<span class="perf-fps-unit">fps</span>`). Each rewrite destroys two text nodes + a span every poll across ~10 cells. Not flickering today; will flicker the moment anyone adds an animation to `.perf-fps-unit` or `.perf-fps-ceiling`.
- The pattern can be killed without architectural change by splitting these into a stable structure (number text node + static unit span) and only updating `textContent` of the number. That's what L3 / Lit would force naturally.
---
## 3.5 Beyond dashboard/perf — push-driven reconciliation
*Added 2026-05-27. The §3 audit was scoped to the two poll timers we were debugging. Widening the `\.innerHTML\s*=` search showed the bug class has a **second driver** and lives outside dashboard/perf too.*
### Two drivers, not one
The teardown is triggered by anything that re-renders **without user intent**:
- **Poll timers** (`setInterval`) — what §2/§3 covered (`dashboard.ts` `_uptimeTimer` + main refresh, `perf-charts.ts` `_pollTimer`).
- **Server-pushed `server:*` events** — `core/events-ws.ts` turns each WS message into a `server:*` CustomEvent; feature modules listen and re-render through the *same* `innerHTML` paths.
So the one-line bug class in §1 reads "poll- **or** push-driven," not just poll.
### Genuinely-affected sites outside dashboard/perf
| Site | Driver | Shape | Notes |
| ---- | ------ | ----- | ----- |
| `core/entity-events.ts` `_invalidateAndReload` | push (`server:entity_changed`, `server:device_health_changed`) | full-**tab** rebuild via `loadTargetsTab` / `loadPictureSources` / `loadAutomations` / `loadIntegrations` | **highest blast radius.** A single pushed entity change tears down and rebuilds an entire tab — losing scroll, focus, open inline editors, restarting card-enter animations. |
| `features/game-integration.ts` event feed (`_eventMonitorTimer`) | poll (2 s) | `feed.innerHTML = events.slice(0,20).map(...)` | full 20-item list rebuild every 2 s while the panel is open. |
| `features/game-integration.ts` connection test (`_connectionTestTimer`) | poll | `panel.innerHTML = …` per tick | transient, low frequency. |
`entity-events.ts` already has the **L1 floor applied by hand**: a 600 ms debounce plus a diff check (`oldData === newData`, then length + `id` + `updated_at` compare) that skips the reload when nothing changed. That kills the *no-op* case — but a **real** change still does the full-tab teardown. This is exactly the §4-L1 limitation ("still tears down when content *does* differ"), live across the whole app.
### Counter-examples that already do it right
Two poll loops never flicker because they mutate `textContent` on a **stable structure** instead of rewriting `innerHTML`:
- `core/api.ts` `loadServerInfo` (connection-check poll) — `versionEl.textContent` / `statusEl.textContent`.
- `features/color-strips/test.ts` FPS sampler (1 s) — `valueEl.textContent` / `avgEl.textContent`.
These are live proof that "stable structure + mutate text node" is the fix — i.e. what L3 / Lit force by construction.
### What this changes about the plan
The §4 ladder was reasoned entirely around **per-cell** rendering, because that was the visible flicker. The push-driven finding surfaces a second, qualitatively different problem:
- **Problem A — cell value churn:** every poll, one value span. Loud only with animations. *Mostly fixed in `f6486f9`.* → wants `setText` / skip-if-unchanged.
- **Problem B — list/tab teardown:** on change/push, an entire list or tab. Loses scroll/focus/open editors. *Unaddressed.* `entity-events.ts` and the game feed are Problem B. → wants **keyed list reconciliation**.
Problem B is a **list-level** concern, not a cell-level one. In Lit terms it maps to a keyed `repeat()` directive over the tab/list body — the dashboard-card work in Phase 2 already needs this, but `entity-events.ts` needs it for tabs that §5 used to list as "out of scope." This does **not** change the chosen direction (Lit); it adds `entity-events.ts` as a first-class, high-priority target.
---
## 4. Decision ladder
Walked through with the user 2026-05-26. Captured here so we don't re-litigate.
### L1 — drop-in `setInnerHtmlIfChanged` helper
- **Shape:** `WeakMap<Element, string>` cache; replace every `el.innerHTML = x` with `setInnerHtmlIfChanged(el, x)`.
- **Wins:** stops the no-change rewrites globally; zero behavior risk; ~30 call-site changes.
- **Misses:** still tears down DOM when content *does* differ (e.g. FPS row values change every tick); doesn't preserve focus/transition state inside a list.
- **Verdict:** floor, not ceiling. Worth doing for cells that don't get migrated to L3/Lit.
### L2 — lint guard
- **Shape:** pre-commit script greps `\.innerHTML\s*=` in `static/js/` outside an allowlist, fails the commit.
- **Wins:** keeps the discipline; cheap.
- **Misses:** only useful as a pair with L1+; bare guard with no helper makes contributors angry.
- **Verdict:** pair with whatever helper we land on.
### L3 — hand-rolled cell-component pattern
- **Shape:** `defineCell({ html, refs, mount, update, unmount })` + `reconcileList(host, items, binding)` + `setText/setClass/setAttr` mutators. ~150300 lines of runtime.
- **Wins:** correct by construction; no dependencies; explicit about what mutates; composes with existing customize panel / color picker.
- **Misses:** we own the abstraction — it grows over time as we need transitions, async data, focus, devtools, error boundaries. Death by a thousand features.
- **Verdict:** second-best. Strong contender if zero-deps is a hard constraint.
### Lit migration of polling modules — **recommended**
- **Shape:** convert each perf cell + each dashboard card cell to a Lit web component. Use `html\`<span>${value}</span>\`` tagged-template + targeted diff. ~5KB gzip added to bundle, no new build step (esbuild handles it).
- **Wins:** solves the bug class by design; maintained by Google + community; web-components-based so no framework lock-in; composes with vanilla DOM trivially; mental model is close to current template-string idiom; non-polling code can stay vanilla forever.
- **Misses:** introduces a dependency; contributors learn one more thing; rare edge cases (`@html`-equivalent exists and reintroduces the bug if misused).
- **Verdict:** best ceiling-to-cost ratio for a small team. Recommended.
### Full framework rewrite (React / Vue / Solid)
- **Verdict:** overkill. The bug class lives in polling paths; the rest of the app is fine. Spending the migration budget on rebuilding IconSelect / EntitySelect / modals / customize panel / graph editor — none of which are broken — is a bad trade.
---
## 5. Recommendation
**Lit for the polling-heavy modules.**
Migration plan:
### Phase 0 — spike (2-hour time-box)
- Convert `patches` cell to a Lit component, end to end.
- Verify it plays nicely with: color picker integration, customize panel layout reorder, `rerenderPerfGrid` reconciliation, `setPerfMode` toggle, hidden-by-env state, the spark tooltip handler.
- If any of those break in an unfixable way → pivot to L3.
- If they work → commit to the migration.
### Phase 1 — perf-charts cells
1. `patches` (already spiked)
2. `devices`
3. `fps` / `capture_fps` / `capture_fps_actual` (share a sparkline base class)
4. `cpu` / `ram` / `gpu` / `temp` (share `_sparkCardHtml` template family)
5. `network` / `device_latency` / `send_timing` / `errors`
Each is its own PR, dashboard stays working at every step. `renderPerfSection` becomes a registry of Lit components; `rerenderPerfGrid` becomes "reorder existing elements in the grid" (which it mostly already does).
### Phase 2 — dashboard card cells
6. Output target cards (running variant — biggest payoff, has live FPS + uptime + errors)
7. Output target cards (stopped variant)
8. Sync clock cards
9. Automation cards
10. Integration (HA / MQTT) cards
These get bigger wins from the migration because they have nested mutable state (FPS pill, errors cell, health dot, action button) that's currently rebuilt per poll via the `_updateRunningMetrics` path.
### Highest-impact: `entity-events.ts` tab reconciliation (sequence early)
`entity-events.ts` (§3.5) is the single highest-blast-radius site and is **not** on the dashboard — it re-renders the Targets / Integrations / Automations tabs on server push. Whether or not those tabs' cells become Lit components, the loader path (`loadTargetsTab` / `loadIntegrations` / `loadAutomations`) should switch from a full `innerHTML` rebuild to a **keyed list reconcile** (a Lit `repeat()` over the tab body). This preserves scroll / focus / open inline editors across pushes. If the goal is "biggest UX win first" rather than "lowest-risk first," sequence this ahead of Phase 2.
### Phase 3 — stopgap helper for the rest
Add `setInnerHtmlIfChanged` and apply to any remaining vanilla polling sites we don't plan to migrate. Add the L2 lint guard at this point — by now everything that polls is either Lit-managed or uses the helper.
### Out of scope (deliberately) — with one correction (2026-05-27)
- Targets tab, automations editor, integrations, scene presets — these render on-demand, **but they are ALSO re-rendered on server push** via `entity-events.ts` (see §3.5). The original claim that "the bug class doesn't bite them" was **wrong**: a pushed `server:entity_changed` does a full-tab `innerHTML` teardown. The *editor / on-demand views* can stay vanilla, but the **list/tab render that entity-events triggers needs reconciliation** (a keyed list diff) regardless of whether those cells become Lit components. Treat the entity-events reload path as **in-scope** — it is the highest-blast-radius Problem B site.
- Color strips editor, graph editor, settings — genuinely on-demand, no push re-render path, stay vanilla.
- Transport bar cells (CPU/Mem chip in the top bar) — read from the same perf payload, can be migrated opportunistically but not urgent.
---
## 6. Open questions to settle before committing
These came up during the discussion and weren't resolved:
1. **Bundle-size budget.** Is +5KB acceptable? Current bundle is 2.7MB so this is noise — but worth confirming there isn't a strict cap (e.g. for slow networks / Android Chaquopy embed).
2. **Contributor model.** If the project will grow to multiple contributors, Lit's smaller community vs React's is a recruiting tradeoff. Currently solo-ish, so probably moot.
3. **Android TV target.** Chaquopy embed serves the same bundle. Lit works fine in any modern browser — Android TV WebView is Chromium-based. Should be a no-op but verify in Phase 0 spike.
4. **Long-term framework intent.** If there's a chance we ever migrate to React/Vue/Solid for the rest of the app, doing Lit now is *not* lock-in (web components are standard), but it does add a second mental model. Probably fine; just naming the tradeoff.
5. **Customize panel.** The drag-reorder code in `dashboard-customize.ts` mutates `.dashboard-section` DOM directly. Lit components reorder cleanly via `moveBefore` / `insertBefore` since they're just elements, but the dnd library needs to treat them as opaque drag handles. Phase 0 spike should confirm.
---
## 7. Pointers
- Source files most relevant:
- `server/src/ledgrab/static/js/features/dashboard.ts`
- `server/src/ledgrab/static/js/features/perf-charts.ts`
- `server/src/ledgrab/static/js/features/dashboard-layout.ts` (cell ordering + visibility)
- `server/src/ledgrab/static/js/features/dashboard-customize.ts` (drag-reorder UI)
- `server/src/ledgrab/static/js/core/card-modes.ts` (mode toggle that hangs off section headers)
- `server/src/ledgrab/static/js/core/entity-events.ts` (push-driven tab reloads — §3.5, highest blast radius)
- `server/src/ledgrab/static/js/core/events-ws.ts` (WS → `server:*` CustomEvent dispatch)
- `server/src/ledgrab/static/js/features/game-integration.ts` (2 s event-feed list rebuild — §3.5)
- Most recent reconciliation commit: `f6486f9`.
- Related skill files in `~/.claude/skills/`: `frontend-patterns`, `documentation-lookup` (for Lit docs via Context7).
- Locale convention: `perf.*` for cross-card primitives, `dashboard.perf.*` for cell titles.
---
## 8. If this doc gets stale
If you read this and the perf cells are already Lit components — delete this file. If you read this and there's a new flicker / focus / transition bug nobody can explain — search for `\.innerHTML\s*=` in `static/js/features/` **and `static/js/core/`** (`entity-events` lives in core) and you've probably found it. For *state loss on a server event* (scroll jump, focus drop, an inline editor closing itself), look at the `server:*` listeners in `core/entity-events.ts` first.
+180
View File
@@ -0,0 +1,180 @@
# Production Review — Remaining Items
Output of the multi-agent production review (security / Python / TypeScript /
performance / architecture / code-quality). Each entry below is something
the original audit flagged and the autonomous hardening pass deliberately
did **not** address — either because it needs design input, profiling
validation, or a multi-day refactor that should land in its own session.
The hardening pass landed everything else: see git log between `master` and
the head of the review branch for the applied changes (URL-scheme +
malicious-input rejection, IconSelect XSS escape, MiniSelect for forbidden
plain `<select>`s, WebSocket Origin allow-list, /docs auth-gate, security
headers middleware, streaming upload size caps, fire-and-forget task
tracking + drain resilience in MQTT runtime, discovery_watcher task
tracking, asyncio.gather return_exceptions, secret_box encryption for MQTT
/ Hue / Govee credentials with auto-migration, SSRF-validated update
redirects, single source of truth for IP classification in
`utils/net_classify.py`, allowlist + parity test for inbound WS events,
typed `Window` globals, and more).
## Items completed in the follow-up autonomous pass (2026-05-23)
- [x] **devices.py PATCH-without-url processor desync**`update_device`
now falls back to `existing.url` so a rename / icon-only edit
always tells the processor the current address.
- [x] **WLED scheme integration test** on `/api/v1/devices` — covers
bare IPv4 (`http://`), public hostname (`https://`), and trailing-slash
normalisation; lives in `tests/api/routes/test_devices_routes.py`.
- [x] **IPv6 regression test**`tests/test_url_scheme.py` now pins
public IPv6 → `https://`, ULA → `http://`, and documents the
Python-`ipaddress` documentation-prefix classification quirk.
- [x] **IconSelect XSS audit + defence-in-depth** — every caller
audited (all feed `icon` from constants or lookup tables); added
`sanitiseIcon` that rejects `<script>`, `javascript:`, `on*=`,
`<iframe>`, `<embed>`, `<object>` and warns to the console.
- [x] **`Optional[T]``T | None` (PEP 604)** — 55 sites cleaned via
`ruff --fix UP007`. The remaining `Union[…]` aliases for
pixel/colour/device-config typing converted by hand. `UP007` now
lives in `pyproject.toml` so the rule fires on new code.
- [x] **Hot-path magic numbers → named constants**`processed_stream`
gains `_FILTER_RECHECK_EVERY_N_FRAMES`; `wled_target_processor`
gains `_SKIP_REPOLL_SLEEP_SECONDS`, `_DIAGNOSTICS_REPORT_INTERVAL_SECONDS`,
`_CSPT_RECHECK_EVERY_N_ITERATIONS`.
- [x] **`api/auth.py` `except Exception` tightening** — every WS send /
close site is now `except _WS_SEND_BENIGN_EXC` (a narrow tuple of
WebSocketDisconnect / RuntimeError / ConnectionError / OSError).
The auth-receive path catches the same set plus a final
`logger.exception` catch-all for observability on truly unexpected
shapes.
- [x] **`(window as any)` cleanup** — 59 static-property accesses
migrated to typed `window.<name>` against `global-types.d.ts`. The
remaining 7 sites use dynamic string indexing (`window[fnName]`)
and intentionally keep the cast (documented in the typedef file).
---
## Architecture refactors (multi-day — own session)
- [ ] **Split `core/processing/value_stream.py`** (1856 LOC, 14 stream classes)
into a `value_streams/` package. Each value-stream type gets its own
file ≤300 LOC; `manager.py` holds `ValueStreamManager`.
- [ ] **Split `storage/color_strip_source.py`** (1841 LOC, 18 source kinds)
into a `color_strip_sources/` package mirroring `value_streams/`.
- [ ] **Frontend file splits**`graph-editor.ts` (2707), `streams.ts`
(2335), `value-sources.ts` (1889), `types.ts` (1062). Highest-churn
modules; mixed UI / state / network responsibilities.
- [ ] **Layering reversal**: introduce a neutral `domain/` package and move
shared DTOs (`FilterInstance`, `CalibrationConfig`, etc.) into it so
`storage/` no longer imports `core/`. Eliminates 7+ layering
violations and the lazy-import hacks used to break the resulting
circulars.
- [ ] **`main.py` boot refactor** — extract import-time side effects into
`bootstrap.py` + `create_app()` factory. `lifespan()` becomes the
single place that wires stores and managers.
- [ ] **DI consolidation** — replace `api/dependencies.py` getter sprawl
(30+ `get_*()` functions reading a process-global `_deps` dict) with
a single typed `get_container()` dependency. Makes test-overrides
trivial; ban direct getter calls in handler bodies.
- [ ] **Exception hierarchy** — define `ledgrab/errors.py` (`LedGrabError`,
`NotFoundError`, `ValidationError`, `RemoteUnavailableError`,
`SSRFBlockedError`). Move HTTP translation into a FastAPI exception
handler. Stop raising `HTTPException` from `utils/safe_source.py`.
- [ ] **Lazy-import audit** — 289 in-function `from ledgrab.*` imports.
Specifically `core/processing/daylight_settings.py` imports
`api.dependencies` (core → api inversion). Pass the database in via
the constructor instead of service-locator lookup.
## Performance (profile before applying)
- [ ] **`composite_stream.py` blend modes** — pre-allocate scratch buffers
in `_blend_override / overlay / hard_light / soft_light / difference
/ exclusion`. Each currently allocates per frame (`mul`, `scr`,
`blended`, `np.where(...)`). At 100 LEDs × 30 fps × N layers this
adds up.
- [ ] **`mapped_stream` / `composite_stream` zone resize** — replace the
per-channel `np.interp` calls with a cached `floor/ceil/frac` LUT
(same trick as `wled_target_processor._fit_to_device`) or a single
`cv2.resize` call on the (N,3) array. `np.interp` allocates a new
`float64` array per channel per frame even on cache-hit.
- [ ] **`processed_stream._processing_loop`** — add ping-pong output
buffers and pass them as `out=` to filter `process_strip()` calls.
Today every filter that returns a fresh allocation costs us a copy
per frame. Also: the loop uses `time.sleep` instead of an
event-driven wait on the input stream — input updates faster than
30 fps see up to `frame_time` of latency.
- [ ] **`mqtt_client.py` `send_pixels`** — add a binary publish path (or
at minimum cache the outer dict skeleton). Today every frame
`pixels.tolist()` + `json.dumps` for ~300 LEDs × 30 fps × N devices.
- [ ] **Frontend `static/js/features/color-strips/test.ts`** — cache
`ImageData` per canvas (`canvas._imageData`); only re-create on
dimension change; use a `Uint32Array` view to copy pixels in one
loop instead of the per-pixel JS loop. Border-overlay rebuild on
every frame should also be debounced to dimension changes only.
- [ ] **`ws_stream.py` composite branch** — pre-allocate a `bytearray`
sized to the largest frame and write into slices instead of
`b"".join(tobytes()) per layer` every iteration. Same anti-pattern
in `wled_target_processor._broadcast_led_preview`.
- [ ] **Preview broadcast slow-client guard**`asyncio.gather` over
preview clients waits for the slowest. Move to `asyncio.wait` with a
timeout and drop slow clients, or fire-and-forget with a
`ws.application_state` filter.
## Security (deferred — non-trivial or design-sensitive)
- [ ] **Content-Security-Policy header** — would need careful tuning
because the UI uses inline event handlers / Jinja templates.
Mis-set CSP would break the app silently. Defer until templates can
move to event-delegated handlers, then add a strict policy.
- [x] **`api/auth.py` exception specificity** — done in the 2026-05-23
pass; see top of file.
- [ ] **Hue bridge cert pinning**`httpx.AsyncClient(verify=False)` for
Hue bridge (self-signed cert by design). Should record the
certificate fingerprint at pairing time and pin it on subsequent
requests; otherwise an on-path attacker can MITM the bridge.
## Mechanical / code-quality (low risk, high line-count)
- [ ] **i18n parity** — confirmed **328** keys missing in `ru.json` and
**325** missing in `zh.json` against the canonical English file.
Translation work — needs a native speaker, not a machine-translation
pass. Run `py scripts/diff_locale_keys.py` (or copy the diff block
out of the 2026-05-23 pass log) to get the exact key list.
- [x] **`Optional[T]``T | None`** — done; `UP007` now enforced via
`pyproject.toml` so the rule prevents regressions.
- [ ] **Hot-path `logger.error(f"...")` → `logger.error("... %s", e)`**
lazy-eval — 658 sites flagged by `ruff --select G004`. Deferred
because it is genuinely cosmetic at ERROR level (always emitted)
and the cumulative cost is negligible. Worth doing if/when ruff
gains a safe autofix, or as a Codemod in a dedicated session.
- [x] **Remaining `(window as any)` sites** — 59 migrated to typed
`window.<name>` access; the 7 surviving sites use dynamic string
indexing and are documented as the legitimate exception.
- [x] **Magic numbers → named constants** — done; see `processed_stream`
and `wled_target_processor` constants at the top of each module.
- [ ] **Standardise `from __future__ import annotations`** — partially
mooted by the UP007 cleanup. Files that previously relied on
`Optional`/`Union` no longer need the future import; the few that
already use `__future__` keep it for forward-reference convenience.
A blanket policy would still help — leave as a stylistic followup.
## Test gaps
- [x] **Route-level integration test** for the WLED scheme inference —
done; covers create + update in `tests/api/routes/test_devices_routes.py::TestWLEDSchemeInference`.
- [x] **IPv6 public address regression** — done; pinned in
`tests/test_url_scheme.py` for both bracketless and bracketed forms.
## Pre-existing issues surfaced during the audit (not in our diff)
These were flagged by the auditors but predate the review session — kept
here as a future-work backlog:
- [x] **`icon-select.ts:_buildGrid` `item.icon` is interpolated raw** —
audited; all callers pass project-owned literals or table-lookup
results. Added a runtime sanitiser as defence-in-depth.
- [x] **`devices.py` `manager.update_device_info(device_url=update_data.url)`**
None-on-PATCH path — fixed; now falls back to `existing.url`.
- [ ] **`asyncio.gather` over uncapped client lists** in preview broadcasts
— slow clients block the loop. Already noted under Performance
above; pre-existing.
-114
View File
@@ -1,114 +0,0 @@
# TODO
## IMPORTANT: Remove WLED naming throughout the app
- [ ] Rename all references to "WLED" in user-facing strings, class names, module names, config keys, file paths, and documentation
- [ ] The app is **LedGrab** — not tied to WLED specifically. WLED is just one of many supported output protocols
- [ ] Audit: i18n keys, page titles, tray labels, installer text, pyproject.toml description, README, CLAUDE.md, context files, API docs
- [ ] Rename `wled_controller` package → decide on new package name (e.g. `ledgrab`)
- [ ] Update import paths, entry points, config references, build scripts, Docker, CI/CD
- [ ] **Migration required** if renaming storage paths or config keys (see data migration policy in CLAUDE.md)
---
## Donation / Open-Source Banner
- [ ] Add a persistent but dismissible banner or notification in the dashboard UI informing users that the project is open-source and under active development, and that donations are highly appreciated
- [ ] Include a link to the donation page (GitHub Sponsors, Ko-fi, or similar — decide on platform)
- [ ] Remember dismissal in localStorage so it doesn't reappear every session
- [ ] Add i18n keys for the banner text (`en.json`, `ru.json`, `zh.json`)
---
# Color Strip Source Improvements
## New Source Types
- [x] **`weather`** — Weather-reactive ambient: maps weather conditions (rain, snow, clear, storm) to colors/animations via API
- [ ] **`music_sync`** — Beat-synced patterns: BPM detection, energy envelope, drop detection (higher-level than raw `audio`)
- [ ] **`math_wave`** — Mathematical wave generator: user-defined sine/triangle/sawtooth expressions, superposition
- [ ] **`text_scroll`** — Scrolling text marquee: bitmap font rendering, static text or RSS/API data source *(delayed)*
### Discuss: `home_assistant`
Need to research HAOS communication options first (WebSocket API, REST API, MQTT, etc.) before deciding scope.
### Deferred
- `image` — Static image sampler *(not now)*
- `clock` — Time display *(not now)*
## Improvements to Existing Sources
### `effect` (now 12 types)
- [x] Add effects: rain, comet, bouncing ball, fireworks, sparkle rain, lava lamp, wave interference
- [x] Custom palette support: user-defined [[pos,R,G,B],...] stops via JSON textarea
### `gradient`
- [x] Noise-perturbed gradient: value noise displacement on stop positions (`noise_perturb` animation type)
- [x] Gradient hue rotation: `hue_rotate` animation type — preserves S/V, rotates H
- [x] Easing functions between stops: linear, ease_in_out (smoothstep), step, cubic
### `audio`
- [x] New audio source type: band extractor (bass/mid/treble split) — responsibility of audio source layer, not CSS
- [ ] Peak hold indicator: global option on audio source (not per-mode), configurable decay time
### `daylight`
- [x] Longitude support for accurate solar position (NOAA solar equations)
- [x] Season awareness (day-of-year drives sunrise/sunset via solar declination)
### `candlelight`
- [x] Wind simulation: correlated flicker bursts across all candles (wind_strength 0.0-2.0)
- [x] Candle type presets: taper (steady), votive (flickery), bonfire (chaotic) — applied at render time
- [x] Wax drip effect: localized brightness dips with fade-in/fade-out recovery
### `composite`
- [ ] Allow nested composites (with cycle detection)
- [x] More blend modes: overlay, soft light, hard light, difference, exclusion
- [x] Per-layer LED range masks (optional start/end/reverse on each composite layer)
### `notification`
- [x] Chase effect (light bounces across strip with glowing tail)
- [x] Gradient flash (bright center fades to edges, exponential decay)
- [x] Queue priority levels (color_override = high priority, interrupts current)
### `api_input`
- [ ] Crossfade transition when new data arrives
- [ ] Interpolation when incoming LED count differs from strip count
- [ ] Last-write-wins from any client (no multi-source blending)
## Architectural / Pipeline
### Processing Templates (CSPT)
- [x] HSL shift filter (hue rotation + lightness adjustment)
- [x] ~~Color temperature filter~~ — already exists as `color_correction`
- [x] Contrast filter
- [x] ~~Saturation filter~~ — already exists
- [x] ~~Pixelation filter~~ — already exists as `pixelate`
- [x] Temporal blur filter (blend frames over time)
### Transition Engine
Needs deeper design discussion. Likely a new entity type `ColorStripSourceTransition` that defines how source switches happen (crossfade, wipe, etc.). Interacts with automations when they switch a target's active source.
### Deferred
- Global BPM sync *(not sure)*
- Recording/playback *(not now)*
- Source preview in editor modal *(not needed — overlay preview on devices is sufficient)*
---
## Remaining Open Discussion
1. **`home_assistant` source** — Need to research HAOS communication protocols first
2. **Transition engine** — Design as `ColorStripSourceTransition` entity: what transition types? (crossfade, wipe, dissolve?) How does a target reference its transition config? How do automations trigger it?
+54
View File
@@ -0,0 +1,54 @@
# TODO
## Donation / Open-Source Banner
- [x] Add a persistent but dismissible banner or notification in the dashboard UI informing users that the project is open-source and under active development, and that donations are highly appreciated
- [x] Include a link to the donation page (GitHub Sponsors, Ko-fi, or similar — decide on platform)
- [x] Remember dismissal in localStorage so it doesn't reappear every session
- [x] Add i18n keys for the banner text (`en.json`, `ru.json`, `zh.json`)
- [ ] Configure `DONATE_URL` and `REPO_URL` constants in `donation.ts` once platform is chosen
## Android Signing Secrets (Gitea)
The CI workflow `build-android.yml` produces a signed release APK **only** when all four secrets below are configured in Gitea → Settings → Secrets. When any one is missing, the "Guard release tag against missing keystore" step hard-fails a `v*` tag build — previously we silently shipped a debug-signed APK labeled as release.
| Secret | Contents |
| --- | --- |
| `ANDROID_KEYSTORE_BASE64` | Output of `base64 -w0 release.jks` — the whole keystore as one line |
| `ANDROID_KEYSTORE_PASSWORD` | Keystore password (the `-storepass` passed to `keytool`) |
| `ANDROID_KEY_ALIAS` | Key alias (e.g. `ledgrab-release`) |
| `ANDROID_KEY_PASSWORD` | Key password (can be the same as keystore password) |
### Generate the keystore (one-time, ~2 min)
```bash
keytool -genkeypair -v \
-storetype JKS \
-keystore release.jks \
-alias ledgrab-release \
-keyalg RSA -keysize 4096 \
-validity 9125 \
-dname "CN=LedGrab, O=Dolgolyov, C=BY"
base64 -w0 release.jks > release.jks.b64 # Linux / Git Bash
# Windows alternative:
# certutil -encode release.jks release.jks.b64
# (strip the -----BEGIN/END CERTIFICATE----- header/footer lines)
```
### Critical — back up `release.jks` outside the repo
- 1Password attachment, encrypted USB stick, or printed hex + password written down somewhere physical.
- Losing the keystore = every existing sideloaded install is permanently unable to upgrade. The only workaround is uninstall-then-reinstall, which wipes user data.
- The `release.jks` file itself must **never** be committed to git. Only the base64 string lives in Gitea secrets.
### Why it matters even without Play Store
Android's package manager refuses to install an upgrade whose signature differs from the currently-installed APK's signature — enforced by the OS, not Play. So once users install a build signed by key X, every future build they can upgrade to must also be signed by key X.
### Current state
- [ ] Generate `release.jks` with `keytool` (above) and back it up
- [ ] Upload the four secrets to Gitea
- [ ] Tag a throwaway `v0.4.1-test` to verify signed release APK is produced (then delete the tag + release)
- [ ] Note: any existing `v0.4.0` debug-signed install cannot upgrade to a release-signed v0.4.1 — users must uninstall first
+1107 -29
View File
File diff suppressed because it is too large Load Diff
+20
View File
@@ -0,0 +1,20 @@
*.iml
.gradle
/local.properties
/.idea
.DS_Store
/build
/app/build
/captures
.externalNativeBuild
.cxx
local.properties
# Chaquopy build cache
.build-cache/
# Python source junction (points at ../server/src/ledgrab — do not commit)
/app/src/main/python/ledgrab
# Signing keystore decoded from CI secrets
/keystore/
+220
View File
@@ -0,0 +1,220 @@
import java.util.concurrent.TimeUnit
plugins {
id("com.android.application")
id("org.jetbrains.kotlin.android")
id("com.chaquo.python")
}
// Derive versionCode from git commit count so sideload/Play upgrades
// always see a strictly-greater value without anyone remembering to
// bump by hand. ANDROID_VERSION_CODE env var wins for CI pipelines
// that prefer explicit values; fallback is `1` when neither is
// available (e.g. building from a tarball with no .git).
val ledgrabVersionCode: Int = run {
System.getenv("ANDROID_VERSION_CODE")?.toIntOrNull()?.let { return@run it }
try {
val process = ProcessBuilder("git", "rev-list", "--count", "HEAD")
.directory(rootDir)
.redirectErrorStream(true)
.start()
if (process.waitFor(5, TimeUnit.SECONDS) && process.exitValue() == 0) {
process.inputStream.bufferedReader().readText().trim().toIntOrNull() ?: 1
} else {
1
}
} catch (_: Exception) {
1
}
}
android {
namespace = "com.ledgrab.android"
// SDK 35 (Android 15) — required for Play Store from Aug 2025 onward.
compileSdk = 35
defaultConfig {
applicationId = "com.ledgrab.android"
minSdk = 24 // Android 7.0 — covers nearly all TV boxes
targetSdk = 35
// Derived from git commit count (or ANDROID_VERSION_CODE env var
// in CI). See ledgrabVersionCode above. Was stuck at 1 before —
// sideload updates silently refused to install.
versionCode = ledgrabVersionCode
versionName = "0.8.2"
// ABI selection. Detect armeabi-v7a wheel presence and opt the
// ABI in only when the matching pydantic-core wheel is on disk —
// otherwise Chaquopy would fail the build searching for it. The
// build script (build-scripts/build-pydantic-core.sh) is the
// source of truth for which ABIs we *can* ship.
val v7Wheel = file("$rootDir/wheels").listFiles().orEmpty()
.any { it.name.startsWith("pydantic_core-") && it.name.contains("armeabi_v7a") }
val ledgrabAbis = buildList {
add("arm64-v8a")
add("x86_64")
add("x86")
if (v7Wheel) add("armeabi-v7a")
}
ndk {
// arm64-v8a is the primary target (real TV hardware).
// x86_64/x86 cover emulators.
// armeabi-v7a is opt-in: many pre-2018 Mecool/X96/H96 TV boxes
// still ship 32-bit ARMv7 — when a wheel exists in wheels/ we
// automatically include the ABI in builds.
abiFilters += ledgrabAbis
}
}
// Per-ABI APK splits — reduces download size by ~60% vs universal APK.
// Each split contains only one native ABI's shared libraries + wheels.
splits {
abi {
val v7Wheel = file("$rootDir/wheels").listFiles().orEmpty()
.any { it.name.startsWith("pydantic_core-") && it.name.contains("armeabi_v7a") }
isEnable = true
reset()
include("arm64-v8a", "x86_64", "x86")
if (v7Wheel) include("armeabi-v7a")
isUniversalApk = true // also produce a fat APK for sideloading
}
}
// Signing config from env vars (CI) — only registered when all four are set.
// Local release builds fall back to the debug signing config.
val ciKeystorePath = System.getenv("ANDROID_KEYSTORE_PATH")
val ciKeystorePassword = System.getenv("ANDROID_KEYSTORE_PASSWORD")
val ciKeyAlias = System.getenv("ANDROID_KEY_ALIAS")
val ciKeyPassword = System.getenv("ANDROID_KEY_PASSWORD")
val hasCiSigning = listOf(ciKeystorePath, ciKeystorePassword, ciKeyAlias, ciKeyPassword)
.all { !it.isNullOrBlank() } && file(ciKeystorePath!!).exists()
signingConfigs {
if (hasCiSigning) {
create("release") {
storeFile = file(ciKeystorePath!!)
storePassword = ciKeystorePassword
keyAlias = ciKeyAlias
keyPassword = ciKeyPassword
}
}
}
buildTypes {
release {
// TODO(minify): keep R8 disabled until Chaquopy reflection is
// verified end-to-end. Chaquopy resolves Kotlin classes & static
// methods (PythonBridge, UsbSerialBridge, Root) by name from
// Python via PyObject — silent stripping breaks the app at
// runtime, after release. proguard-rules.pro contains keep
// rules covering the known entry points, but until we have
// a release smoke test that exercises every PyObject path we
// do NOT ship a minified release.
isMinifyEnabled = false
proguardFiles(
getDefaultProguardFile("proguard-android-optimize.txt"),
"proguard-rules.pro",
)
// Refuse to silently sign release APKs with the debug
// keystore — that's how a debug-signed release accidentally
// ships. CI must provide all four signing env vars. If a
// local "release" build is genuinely intended for testing,
// set ANDROID_ALLOW_DEBUG_SIGNED_RELEASE=1 to opt out.
val allowDebugSigned =
System.getenv("ANDROID_ALLOW_DEBUG_SIGNED_RELEASE") == "1"
signingConfig = when {
hasCiSigning -> signingConfigs.getByName("release")
allowDebugSigned -> signingConfigs.getByName("debug")
else -> throw GradleException(
"Release builds require signing env vars " +
"(ANDROID_KEYSTORE_PATH/PASSWORD, ANDROID_KEY_ALIAS/PASSWORD). " +
"Set ANDROID_ALLOW_DEBUG_SIGNED_RELEASE=1 to force a debug-signed release."
)
}
}
}
compileOptions {
sourceCompatibility = JavaVersion.VERSION_17
targetCompatibility = JavaVersion.VERSION_17
}
kotlinOptions {
jvmTarget = "17"
}
}
chaquopy {
defaultConfig {
version = "3.11"
pip {
// Pre-built wheels directory (for pydantic-core etc.) as an
// absolute file:// URI with forward slashes. Pip requires
// exactly three slashes after `file:` for an empty-host local
// path; the path-normalization differs by platform:
// Windows: absolute path is "C:/..." → prefix "file:///"
// Linux: absolute path is "/..." → prefix "file://"
// Using a fixed "file:///" prefix on both made CI produce
// "file:////workspace/…", which pip rejects as non-local.
val wheelsPath = rootDir.absolutePath.replace("\\", "/")
val wheelsUrlPrefix = if (wheelsPath.startsWith("/")) "file://" else "file:///"
options("--find-links", "$wheelsUrlPrefix$wheelsPath/wheels/")
// ── Android-compatible dependencies ─────────────────
// Listed explicitly because pyproject.toml includes
// desktop-only packages with no Android wheels.
// See CLAUDE.md "Android Dependency Sync" for policy.
install("fastapi")
install("uvicorn") // without [standard] — no uvloop/httptools
install("httpx")
install("numpy")
install("pydantic") // needs pydantic-core wheel in wheels/
install("pydantic-settings")
install("PyYAML")
install("structlog")
install("python-json-logger")
install("python-dateutil")
install("python-multipart")
install("jinja2")
install("zeroconf")
install("aiomqtt")
install("openrgb-python")
// opencv-python-headless: no cp311 Android wheel on Chaquopy.
// LedGrab's cv2 usage is guarded with try/except ImportError
// and falls back to numpy/Pillow alternatives on Android.
install("Pillow")
install("websockets")
install("cryptography") // AES-GCM secret-box for HA/MQTT credentials
}
}
}
// LedGrab Python source is included via a directory junction:
// android/app/src/main/python/ledgrab -> server/src/ledgrab
// This is the standard Chaquopy way to include local Python packages.
// Create the junction (run from repo root, no admin needed):
// cmd /c "mklink /J android\app\src\main\python\ledgrab server\src\ledgrab"
dependencies {
implementation("androidx.core:core-ktx:1.13.1")
implementation("androidx.appcompat:appcompat:1.7.0")
implementation("androidx.leanback:leanback:1.0.0")
implementation("com.google.android.material:material:1.12.0")
implementation("androidx.lifecycle:lifecycle-service:2.8.7")
implementation("androidx.lifecycle:lifecycle-runtime-ktx:2.8.7")
implementation("org.jetbrains.kotlinx:kotlinx-coroutines-android:1.8.1")
// SplashScreen API — keeps a friendly logo on screen while Chaquopy
// unpacks the Python stdlib on first launch (can take 1-3s).
implementation("androidx.core:core-splashscreen:1.0.1")
// QR code generation for displaying server URL on TV
implementation("com.google.zxing:core:3.5.3")
// EncryptedSharedPreferences (Android Keystore-backed) for the per-install
// server API key (see ApiKeyManager). Falls back to plain SharedPreferences
// when the keystore is unavailable.
implementation("androidx.security:security-crypto:1.1.0-alpha06")
// USB-serial drivers (CH340, CP2102, FTDI, Prolific, CDC-ACM) for
// driving Adalight/AmbiLED controllers plugged into Android TV boxes.
implementation("com.github.mik3y:usb-serial-for-android:3.8.1")
}
+27
View File
@@ -0,0 +1,27 @@
# LedGrab ProGuard / R8 rules.
#
# IMPORTANT: Chaquopy resolves Java/Kotlin classes and static methods by
# name from Python (e.g. UsbSerialBridge.INSTANCE.listDevices()) via
# reflection. Anything reachable through PyObject must be kept by name
# or the release build will throw NoSuchMethod / ClassNotFound at
# runtime silently, only on the user's device.
#
# Keep ALL of com.ledgrab.android.* members for safety. The app is
# small enough that the size win from stripping these isn't worth the
# fragility.
-keep class com.ledgrab.android.** { *; }
# Chaquopy runtime itself.
-keep class com.chaquo.python.** { *; }
-dontwarn com.chaquo.python.**
# usb-serial-for-android driver classes are loaded via the prober's
# default device-id list, which uses reflection in some chip drivers.
-keep class com.hoho.android.usbserial.driver.** { *; }
-dontwarn com.hoho.android.usbserial.**
# Kotlin coroutines keep the debug agent off and the metadata intact.
-dontwarn kotlinx.coroutines.**
# Standard Android best-practice keeps.
-keepattributes Signature, InnerClasses, EnclosingMethod, *Annotation*
+186
View File
@@ -0,0 +1,186 @@
<?xml version="1.0" encoding="utf-8"?>
<manifest xmlns:android="http://schemas.android.com/apk/res/android"
xmlns:tools="http://schemas.android.com/tools">
<!-- BLE scanning and connecting — API ≥31 uses granular permissions;
older releases need BLUETOOTH + ACCESS_FINE_LOCATION for scanning.
neverForLocation avoids the location permission dialog on API 31+. -->
<uses-permission android:name="android.permission.BLUETOOTH"
android:maxSdkVersion="30" />
<uses-permission android:name="android.permission.BLUETOOTH_ADMIN"
android:maxSdkVersion="30" />
<uses-permission android:name="android.permission.ACCESS_FINE_LOCATION"
android:maxSdkVersion="30" />
<uses-permission android:name="android.permission.BLUETOOTH_SCAN"
android:usesPermissionFlags="neverForLocation"
tools:targetApi="s" />
<uses-permission android:name="android.permission.BLUETOOTH_CONNECT"
tools:targetApi="s" />
<!-- BLE hardware — required=false so non-BT boxes still install. -->
<uses-feature
android:name="android.hardware.bluetooth_le"
android:required="false" />
<!-- Network access for WLED HTTP/UDP, web UI, MQTT -->
<uses-permission android:name="android.permission.INTERNET" />
<uses-permission android:name="android.permission.ACCESS_NETWORK_STATE" />
<uses-permission android:name="android.permission.ACCESS_WIFI_STATE" />
<!-- Foreground service permissions.
FOREGROUND_SERVICE_MEDIA_PROJECTION: required on API 34+ for the
MediaProjection capture path.
FOREGROUND_SERVICE_SPECIAL_USE: required on API 34+ for the root
screenrecord capture path (it doesn't use MediaProjection).
Both are declared because the service may run in either mode. -->
<uses-permission android:name="android.permission.FOREGROUND_SERVICE" />
<uses-permission android:name="android.permission.FOREGROUND_SERVICE_MEDIA_PROJECTION" />
<uses-permission android:name="android.permission.FOREGROUND_SERVICE_SPECIAL_USE" />
<!-- FOREGROUND_SERVICE_CAMERA (API 34+): required to keep camera access while
the app is backgrounded during on-device webcam capture. The service is
promoted with the `camera` FGS type ONLY when CAMERA is already granted
(see CaptureService.onStartCommand) — unlike audio playback capture (which
rides the MediaProjection token under the mediaProjection type), the camera
has no such coupling and needs its own FGS type to survive backgrounding. -->
<uses-permission android:name="android.permission.FOREGROUND_SERVICE_CAMERA" />
<!-- POST_NOTIFICATIONS for Android 13+ foreground service notification -->
<uses-permission android:name="android.permission.POST_NOTIFICATIONS" />
<!-- RECORD_AUDIO for on-device system-playback capture (AudioPlaybackCapture,
API 29+) feeding audio-reactive lighting. Runtime "dangerous" permission,
requested in MainActivity; capture degrades gracefully when denied.
Playback capture runs under the existing mediaProjection FGS type, so no
FOREGROUND_SERVICE_MICROPHONE / microphone FGS type is needed (that would
only be required if the mic-fallback path ran inside the service). -->
<uses-permission android:name="android.permission.RECORD_AUDIO" />
<!-- CAMERA for on-device webcam capture (Camera2). Runtime "dangerous"
permission, requested in MainActivity gated on FEATURE_CAMERA_ANY so
camera-less TV boxes never see the prompt; capture degrades gracefully
when denied. The camera is opened ON DEMAND (only while a camera
capture source is active). To keep capturing after the app is
backgrounded, the service is promoted with the `camera` FGS type
(FOREGROUND_SERVICE_CAMERA above) — but only when CAMERA is already
granted, so a camera-less / not-yet-granted box never risks a failed
service start. -->
<uses-permission android:name="android.permission.CAMERA" />
<!-- PACKAGE_USAGE_STATS — read the foreground app for the "Application"
automation rule (foreground app -> activate scene) via UsageStatsManager.
A special-access permission: it can't be granted at runtime; the user
toggles it under Settings > Usage access (opened from MainActivity).
tools:ignore="ProtectedPermissions" silences the build warning that this
is a system/signature-level permission — it is honoured as a user-grantable
special access. NO QUERY_ALL_PACKAGES is needed: matching only compares the
foreground package NAME, and the app picker uses LauncherApps. -->
<uses-permission
android:name="android.permission.PACKAGE_USAGE_STATS"
tools:ignore="ProtectedPermissions" />
<!-- Autostart on boot — BootReceiver spawns CaptureService in root
mode so capture resumes without the user touching the remote. -->
<uses-permission android:name="android.permission.RECEIVE_BOOT_COMPLETED" />
<!-- Exempt from Doze/App Standby so the FG service isn't killed
overnight on phones; essentially a no-op on TV boxes. -->
<uses-permission android:name="android.permission.REQUEST_IGNORE_BATTERY_OPTIMIZATIONS" />
<!-- Optional wake lock for sustained-performance boxes that aggressively
sleep the CPU with the display off. -->
<uses-permission android:name="android.permission.WAKE_LOCK" />
<!-- Android TV declarations -->
<uses-feature
android:name="android.software.leanback"
android:required="true" />
<uses-feature
android:name="android.hardware.touchscreen"
android:required="false" />
<!-- USB host — for USB-to-TTL adapters driving Adalight/AmbiLED
controllers. required=false so phones without USB host still install. -->
<uses-feature
android:name="android.hardware.usb.host"
android:required="false" />
<!-- Camera hardware — for on-device webcam capture. required=false so
camera-less TV boxes (the common case) still install; the camera
engine simply reports no displays on such devices. camera.any covers
built-in (front/back) and external/USB-UVC cameras the platform
routes through Camera2. -->
<uses-feature
android:name="android.hardware.camera.any"
android:required="false" />
<application
android:name=".LedGrabApp"
android:allowBackup="false"
android:enableOnBackInvokedCallback="true"
android:icon="@mipmap/ic_launcher"
android:label="@string/app_name"
android:banner="@drawable/banner_tv"
android:networkSecurityConfig="@xml/network_security_config"
android:theme="@style/Theme.LedGrab">
<!-- TV launcher activity. Boots through the SplashScreen theme so
the (sometimes multi-second) Chaquopy stdlib unpack doesn't
show as a black screen on first launch. -->
<activity
android:name=".MainActivity"
android:exported="true"
android:theme="@style/Theme.LedGrab.Splash">
<intent-filter>
<action android:name="android.intent.action.MAIN" />
<category android:name="android.intent.category.LEANBACK_LAUNCHER" />
</intent-filter>
</activity>
<!-- Foreground service for screen capture + Python server.
Declares BOTH mediaProjection AND specialUse: only one is
active at a time but Android needs to see the union of
possible types up-front so it doesn't kill the service when
we promote it with a different type at runtime.
FOREGROUND_SERVICE_TYPE_SPECIAL_USE on API 34+ requires the
PROPERTY_SPECIAL_USE_FGS_SUBTYPE rationale below. -->
<service
android:name=".CaptureService"
android:foregroundServiceType="mediaProjection|specialUse|camera"
android:exported="false">
<property
android:name="android.app.PROPERTY_SPECIAL_USE_FGS_SUBTYPE"
android:value="Root-mode screen capture for ambient LED sync. Uses /system/bin/screenrecord on rooted devices to avoid MediaProjection's persistent capture indicator overlay, which is required for the always-on ambient-lighting use case." />
</service>
<!-- Notification capture — a NotificationListenerService bound by
system_server. exported="true" is REQUIRED here (the system binds
it cross-process) and intentionally diverges from CaptureService
(exported="false"); access is gated by the system-held
BIND_NOTIFICATION_LISTENER_SERVICE permission, so no new
<uses-permission> is needed. The user grants access via
Settings > Notification access (opened from MainActivity). -->
<service
android:name=".LedGrabNotificationListener"
android:label="@string/notification_listener_label"
android:exported="true"
android:permission="android.permission.BIND_NOTIFICATION_LISTENER_SERVICE">
<intent-filter>
<action android:name="android.service.notification.NotificationListenerService" />
</intent-filter>
</service>
<!-- Autostart — fires on device boot (and package replace).
On rooted devices, launches CaptureService directly so capture
resumes without the user tapping Start. Unrooted devices are
no-op because MediaProjection consent cannot be bypassed. -->
<receiver
android:name=".BootReceiver"
android:exported="true"
android:enabled="true">
<intent-filter>
<action android:name="android.intent.action.BOOT_COMPLETED" />
<action android:name="android.intent.action.LOCKED_BOOT_COMPLETED" />
<action android:name="android.intent.action.MY_PACKAGE_REPLACED" />
</intent-filter>
</receiver>
</application>
</manifest>
@@ -0,0 +1,243 @@
package com.ledgrab.android
import android.content.Context
import android.content.SharedPreferences
import android.util.Log
import androidx.security.crypto.EncryptedSharedPreferences
import androidx.security.crypto.MasterKey
import java.security.SecureRandom
/**
* Persists the per-install API key for the embedded FastAPI server.
*
* The server's auth gate ([ledgrab.api.auth]) requires a Bearer token
* for any non-loopback request when ``auth.api_keys`` is configured.
* Without a key, LAN clients (phone, laptop) get 401 — which is the
* server's secure default but breaks the QR-scan workflow.
*
* This class generates one key per install (random 32-byte → 64-char
* hex), persists it to SharedPreferences, and exposes it to:
* - [PythonBridge] which sets ``LEDGRAB_AUTH__API_KEYS=android:<key>``
* before uvicorn starts.
* - [MainActivity] which embeds the key as a URL fragment
* (``http://ip:port/#k=<key>``) in the QR. Fragments are never sent
* to the server in HTTP requests, so the key doesn't appear in
* access logs.
*/
class ApiKeyManager(context: Context) {
private val appContext = context.applicationContext
// Prefer Android-Keystore-backed EncryptedSharedPreferences for the API
// key. If the keystore is unavailable (some OEM TV-box ROMs ship a broken
// or absent keystore, or a key got corrupted), creation throws — fall back
// to plain SharedPreferences so a keystore failure NEVER bricks the local
// API key (which would 401 every LAN client).
private val prefs: SharedPreferences
init {
val (store, isEncrypted) = buildPrefs(appContext)
prefs = store
// Only run the plain→encrypted migration when the encrypted store is
// actually available; on the degraded plain path there is nothing to
// migrate INTO (and recoverLegacyKey reads the backup directly).
if (isEncrypted) migrateLegacyKeyIfPresent()
}
// Once we've materialised a key in this process, cache it so
// subsequent reads don't hit prefs and don't risk re-checking
// length under contention.
@Volatile private var cached: String? = null
private val lock = Any()
/** Persistent random API key, generated lazily on first access. */
val apiKey: String
get() = getOrCreateKey()
/** Force a new key. Useful if a user thinks the QR was photographed. */
fun rotate(): String {
synchronized(lock) {
val next = generateKey()
// apply() is fine for rotation — by definition the user
// initiated this and will see the new QR; the worst case
// on crash is they need to re-rotate.
prefs.edit().putString(KEY_API_KEY, next).apply()
cached = next
Log.i(TAG, "Rotated API key")
return next
}
}
private fun getOrCreateKey(): String {
cached?.let { return it }
synchronized(lock) {
// Double-checked under the lock.
cached?.let { return it }
val existing = prefs.getString(KEY_API_KEY, null)
if (existing != null && existing.length >= MIN_KEY_LENGTH) {
cached = existing
return existing
}
// Before minting a fresh key, fall back to any key still in the
// legacy plain store (covers a failed/partial encrypted migration:
// commit() can return false WITHOUT throwing, so migration may have
// left the live key only in the legacy file). Rotating the
// per-install key would 401 every already-paired client, so we
// generate a brand-new key ONLY when no key exists anywhere.
recoverLegacyKey()?.let { recovered ->
// Best-effort persist into the encrypted store; cache regardless
// so we still return the recovered key if the write keeps failing.
runCatching { prefs.edit().putString(KEY_API_KEY, recovered).commit() }
cached = recovered
Log.i(TAG, "Recovered existing API key from legacy storage")
return recovered
}
val generated = generateKey()
// commit() (synchronous disk write) on the FIRST write so
// the key is durable before MainActivity encodes it into a
// QR. If the process is killed between QR display and the
// async write landing, the user's phone would scan a key
// the server never learned about. Subsequent rotates can
// safely use apply().
prefs.edit().putString(KEY_API_KEY, generated).commit()
cached = generated
Log.i(TAG, "Generated new API key (length=${generated.length})")
return generated
}
}
/**
* Build the backing store, preferring EncryptedSharedPreferences. Returns
* (store, isEncrypted). Any keystore failure falls back to the plain prefs
* file so the local API key is never lost on a broken-keystore device.
*/
private fun buildPrefs(context: Context): Pair<SharedPreferences, Boolean> {
return try {
createEncrypted(context) to true
} catch (e: Exception) {
// The keystore can become invalidated (OS upgrade, device restore,
// OEM keystore bug), after which create() throws on EVERY launch and
// the corrupt encrypted file is never cleaned up — degrading to plain
// prefs forever and (because the live key was only in the encrypted
// store) rotating the per-install key on the next mint, 401-ing every
// paired client. Self-heal once: delete the corrupt store + master key
// alias and retry create() before degrading.
Log.w(TAG, "EncryptedSharedPreferences unavailable, attempting one-time reset: ${e.message}")
runCatching {
context.deleteSharedPreferences(ENCRYPTED_PREFS_NAME)
runCatching {
val ks = java.security.KeyStore.getInstance("AndroidKeyStore").apply { load(null) }
if (ks.containsAlias(MasterKey.DEFAULT_MASTER_KEY_ALIAS)) {
ks.deleteEntry(MasterKey.DEFAULT_MASTER_KEY_ALIAS)
}
}.onFailure { Log.w(TAG, "Master-key alias cleanup failed: ${it.message}") }
createEncrypted(context) to true
}.getOrElse {
// Still failing after reset — degrade to plain prefs rather than
// crashing. Worst case the key is stored unencrypted on a
// single-user TV box, which is the pre-existing behaviour.
Log.w(TAG, "EncryptedSharedPreferences still unavailable after reset, using plain prefs: ${it.message}")
context.getSharedPreferences(PREFS_NAME, Context.MODE_PRIVATE) to false
}
}
}
private fun createEncrypted(context: Context): SharedPreferences {
val masterKey = MasterKey.Builder(context)
.setKeyScheme(MasterKey.KeyScheme.AES256_GCM)
.build()
return EncryptedSharedPreferences.create(
context,
ENCRYPTED_PREFS_NAME,
masterKey,
EncryptedSharedPreferences.PrefKeyEncryptionScheme.AES256_SIV,
EncryptedSharedPreferences.PrefValueEncryptionScheme.AES256_GCM,
)
}
/**
* One-time migration: if a key exists in the legacy plain-text prefs file
* (from before encrypted storage), copy it into the encrypted store and
* remove the plain copy. Preserves the existing key so already-scanned QR
* clients keep working — generating a fresh key here would silently 401
* every LAN client (see the Data Migration Policy in CLAUDE.md).
*/
private fun migrateLegacyKeyIfPresent() {
// Don't migrate if the encrypted store already holds a key.
if (!prefs.getString(KEY_API_KEY, null).isNullOrEmpty()) return
runCatching {
val legacy = appContext.getSharedPreferences(PREFS_NAME, Context.MODE_PRIVATE)
val legacyKey = legacy.getString(KEY_API_KEY, null)
if (legacyKey != null && legacyKey.length >= MIN_KEY_LENGTH) {
// commit() returns false on write failure WITHOUT throwing, so the
// runCatching wrapper alone does NOT protect this path. Verify the
// encrypted store both committed AND reads back the identical value
// before touching the legacy copy — otherwise a silent write
// failure could delete the only surviving copy of the key and
// rotate it on next launch (401s every paired client — the exact
// silent-data-loss the Data Migration Policy forbids).
val ok = prefs.edit().putString(KEY_API_KEY, legacyKey).commit()
if (ok && prefs.getString(KEY_API_KEY, null) == legacyKey) {
// Keep the value as a .migrated backup (don't hard-delete) per
// the migration policy; remove only the live legacy key so the
// plaintext copy no longer answers reads.
legacy.edit()
.putString(KEY_API_KEY_MIGRATED, legacyKey)
.remove(KEY_API_KEY)
.apply()
Log.i(TAG, "Migrated API key from plain to encrypted storage")
} else {
// Leave the legacy key untouched; getOrCreateKey() will recover
// it via recoverLegacyKey() rather than minting a fresh one.
Log.w(TAG, "Encrypted key write unverified — keeping legacy key, not migrating")
}
}
}.onFailure { Log.w(TAG, "Legacy API key migration failed: ${it.message}") }
}
/**
* Recover a still-present key from the legacy plain store — either the live
* key (failed/never-run migration) or the `.migrated` backup. Returns null
* only when no valid key survives.
*
* This MUST run on the degraded plain-prefs path too (not just the encrypted
* path): after a successful migration the live key is moved to the
* `.migrated` backup in this same plain file, so when the keystore later
* fails and we degrade to plain prefs, the backup is the only surviving
* copy. Returning null here (the previous `if (!encrypted) return null`
* guard) would mint a fresh key and rotate the per-install key, 401-ing every
* paired client.
*/
private fun recoverLegacyKey(): String? {
val legacy = appContext.getSharedPreferences(PREFS_NAME, Context.MODE_PRIVATE)
val candidate = legacy.getString(KEY_API_KEY, null)
?: legacy.getString(KEY_API_KEY_MIGRATED, null)
return candidate?.takeIf { it.length >= MIN_KEY_LENGTH }
}
private fun generateKey(): String {
val bytes = ByteArray(KEY_BYTES)
SecureRandom().nextBytes(bytes)
// Hex-encode so the key survives copy/paste, URL fragments, env
// vars, and YAML config without escaping concerns. Mask to 0xff
// first — Kotlin's Byte is signed, and `%02x` on a negative
// Byte sign-extends to an 8-char hex string ("ffffffff" instead
// of "ff"), which would produce an invalid key.
return bytes.joinToString("") { "%02x".format(it.toInt() and 0xff) }
}
companion object {
private const val TAG = "ApiKeyManager"
private const val PREFS_NAME = "ledgrab_auth"
private const val ENCRYPTED_PREFS_NAME = "ledgrab_auth_enc"
private const val KEY_API_KEY = "api_key"
// Backup of a migrated legacy key, kept in the plain store per the
// Data Migration Policy (never hard-delete user data on rename/move).
private const val KEY_API_KEY_MIGRATED = "api_key_migrated"
private const val KEY_BYTES = 32
private const val MIN_KEY_LENGTH = 32
/** Label used as the LEDGRAB_AUTH__API_KEYS map key. */
const val LABEL = "android"
}
}
@@ -0,0 +1,234 @@
package com.ledgrab.android
import android.annotation.SuppressLint
import android.media.AudioAttributes
import android.media.AudioFormat
import android.media.AudioPlaybackCaptureConfiguration
import android.media.AudioRecord
import android.media.MediaRecorder
import android.media.projection.MediaProjection
import android.os.Build
import android.util.Log
import java.nio.ByteBuffer
import java.nio.ByteOrder
/**
* Captures audio with [AudioRecord] and pushes interleaved float32 PCM to
* the LedGrab Python server via [PythonBridge], where the
* `android_audio_engine` feeds it into the unchanged audio-analysis
* pipeline.
*
* Two sources:
* - [start] — system playback capture via `AudioPlaybackCapture` (API 29+),
* reusing the same [MediaProjection] token the app already holds for
* screen capture. This is the primary path on the consent flow.
* - [startMic] — microphone fallback (`AudioSource.MIC`) for paths with no
* MediaProjection (root mode) or API < 29.
*
* Mirrors [ScreenCapture]'s shape: a dedicated capture thread, a single
* reusable cross-JNI buffer (no per-block allocation → no GC churn on
* low-end TV boxes), and graceful teardown in [stop].
*
* The capture format is negotiated by [AudioRecord]; the **actual**
* channel count and sample rate are read back and forwarded to
* `configureAudio` so the Python analyzer's interleaving matches the bytes
* we push (e.g. a stereo request that the device satisfies as mono).
*/
class AudioCapture(
private val projection: MediaProjection?,
private val bridge: PythonBridge,
private val sampleRate: Int = 48000,
private val channels: Int = 2,
private val chunkFrames: Int = 1024,
) {
companion object {
private const val TAG = "AudioCapture"
private const val BYTES_PER_FLOAT = 4
}
private var audioRecord: AudioRecord? = null
private var captureThread: Thread? = null
@Volatile private var running = false
/**
* Start system playback capture (API 29+). Requires the app to hold
* RECORD_AUDIO and a valid [projection]. Returns true if capture began.
*/
@SuppressLint("MissingPermission")
fun start(): Boolean {
if (running) return true
if (Build.VERSION.SDK_INT < Build.VERSION_CODES.Q) {
Log.i(TAG, "Playback capture needs API 29+; skipping (have ${Build.VERSION.SDK_INT})")
return false
}
val proj = projection
if (proj == null) {
Log.i(TAG, "No MediaProjection; playback capture unavailable")
return false
}
val config = AudioPlaybackCaptureConfiguration.Builder(proj)
.addMatchingUsage(AudioAttributes.USAGE_MEDIA)
.addMatchingUsage(AudioAttributes.USAGE_GAME)
.addMatchingUsage(AudioAttributes.USAGE_UNKNOWN)
.build()
val record = try {
AudioRecord.Builder()
.setAudioFormat(audioFormat())
.setBufferSizeInBytes(bufferBytes())
.setAudioPlaybackCaptureConfig(config)
.build()
} catch (e: Exception) {
Log.e(TAG, "Failed to build playback AudioRecord: ${e.message}")
return false
}
return begin(record, "playback")
}
/**
* Start microphone capture (fallback). Works on API 24+ and needs no
* MediaProjection. Requires RECORD_AUDIO. Returns true if capture began.
*
* ⚠️ SECURITY/POLICY: currently UNWIRED (no caller). Microphone capture is
* a materially different posture than playback capture — it records real
* room audio (bystander voices). Before wiring this into [CaptureService]:
* - add FOREGROUND_SERVICE_MICROPHONE permission + the `microphone` FGS
* type (on API 34+ the service is killed without it), and
* - add the Play Store privacy disclosure for microphone use,
* - re-trigger a security review.
* Do NOT call this from inside the foreground service without the above.
*/
@SuppressLint("MissingPermission")
fun startMic(): Boolean {
if (running) return true
val record = try {
AudioRecord.Builder()
.setAudioSource(MediaRecorder.AudioSource.MIC)
.setAudioFormat(audioFormat())
.setBufferSizeInBytes(bufferBytes())
.build()
} catch (e: Exception) {
Log.e(TAG, "Failed to build mic AudioRecord: ${e.message}")
return false
}
return begin(record, "mic")
}
/** Stop capturing and release all resources. Idempotent. */
fun stop() {
running = false
// AudioRecord.stop() unblocks a pending READ_BLOCKING read within
// milliseconds, so the loop sees running=false and returns well inside
// the 500ms join window — release() below won't race a live read.
// (Mirrors ScreenCapture's bounded join.)
runCatching { audioRecord?.stop() }
captureThread?.let { runCatching { it.join(500) } }
captureThread = null
runCatching { audioRecord?.release() }
audioRecord = null
runCatching { bridge.shutdownAudio() }
Log.i(TAG, "Audio capture stopped")
}
// ── internals ──────────────────────────────────────────────────────
private fun begin(record: AudioRecord, mode: String): Boolean {
if (record.state != AudioRecord.STATE_INITIALIZED) {
Log.e(TAG, "AudioRecord ($mode) failed to initialize")
runCatching { record.release() }
return false
}
val actualChannels = record.channelCount.coerceAtLeast(1)
val actualRate = record.sampleRate
// Confirm recording actually started before reporting success —
// startRecording() can throw (exclusive-capture contention) or
// leave the record in a non-recording state, in which case read()
// would only ever return errors.
val started = runCatching { record.startRecording() }.isSuccess &&
record.recordingState == AudioRecord.RECORDSTATE_RECORDING
if (!started) {
Log.e(TAG, "AudioRecord ($mode) failed to start recording")
runCatching { record.release() }
return false
}
// Recording confirmed — tell Python the real negotiated format
// before frames flow, so the analyzer's channel/sample-rate match
// the interleaving we push.
bridge.configureAudio(actualRate, actualChannels, chunkFrames)
audioRecord = record
running = true
captureThread = Thread(
{ captureLoop(record, actualChannels) },
"LedGrab-AudioCapture",
).also { it.start() }
Log.i(TAG, "Audio capture started ($mode, sr=$actualRate ch=$actualChannels chunk=$chunkFrames)")
return true
}
/**
* Blocking read loop. Accumulates into fixed `chunkFrames * channels`
* float blocks and pushes only COMPLETE blocks — [AudioRecord.read]
* returns a variable count, so partial reads are stitched here rather
* than handed to Python as ragged chunks (the analyzer requires
* whole-frame, ≤ chunk-size blocks).
*/
private fun captureLoop(record: AudioRecord, actualChannels: Int) {
val blockFloats = chunkFrames * actualChannels
val floatBuf = FloatArray(blockFloats)
// Reusable little-endian byte buffer — Python copies on push, so the
// same backing array is safe to overwrite next block. Default
// ByteBuffer order is BIG_ENDIAN, which would corrupt every sample;
// LITTLE_ENDIAN matches numpy's native float32 on all Android ABIs.
val byteBuf = ByteArray(blockFloats * BYTES_PER_FLOAT)
val floatView = ByteBuffer.wrap(byteBuf).order(ByteOrder.LITTLE_ENDIAN).asFloatBuffer()
var filled = 0
while (running) {
val n = record.read(floatBuf, filled, blockFloats - filled, AudioRecord.READ_BLOCKING)
if (n < 0) {
if (running) {
// A negative read (e.g. ERROR_DEAD_OBJECT after an audio-route
// change, ERROR_INVALID_OPERATION) means this AudioRecord is
// finished. Deactivate the Python engine so is_available() stops
// advertising a dead stream and the audio-reactive consumer isn't
// left polling an empty queue forever. We're on the capture thread,
// so we can't call stop() (it would self-join) — just flip running
// and shut the engine down; onDestroy's stop() releases the record.
Log.w(TAG, "AudioRecord.read error: $n — stopping audio capture")
running = false
runCatching { bridge.shutdownAudio() }
}
break
}
filled += n
if (filled < blockFloats) continue
floatView.clear()
floatView.put(floatBuf, 0, blockFloats)
bridge.pushAudio(byteBuf)
filled = 0
}
}
private fun channelMask(): Int =
if (channels >= 2) AudioFormat.CHANNEL_IN_STEREO else AudioFormat.CHANNEL_IN_MONO
private fun audioFormat(): AudioFormat =
AudioFormat.Builder()
.setEncoding(AudioFormat.ENCODING_PCM_FLOAT)
.setSampleRate(sampleRate)
.setChannelMask(channelMask())
.build()
private fun bufferBytes(): Int {
val minBuf = AudioRecord.getMinBufferSize(sampleRate, channelMask(), AudioFormat.ENCODING_PCM_FLOAT)
// A few blocks of headroom so a slow consumer doesn't overrun the
// hardware buffer between reads.
val want = chunkFrames * channels * BYTES_PER_FLOAT * 4
return if (minBuf > 0) maxOf(minBuf, want) else want
}
}
@@ -0,0 +1,26 @@
package com.ledgrab.android
import android.content.Context
/**
* Thin SharedPreferences wrapper for the boot-autostart toggle.
*
* Default is `true`: [BootReceiver] still bails out when the device
* isn't rooted, so a true default is safe — it only takes effect on
* boxes where we can actually honor it silently.
*/
class AutostartPrefs(context: Context) {
private val prefs = context.applicationContext
.getSharedPreferences(PREFS_NAME, Context.MODE_PRIVATE)
var isEnabled: Boolean
get() = prefs.getBoolean(KEY_ENABLED, DEFAULT_ENABLED)
set(value) { prefs.edit().putBoolean(KEY_ENABLED, value).apply() }
companion object {
private const val PREFS_NAME = "ledgrab_autostart"
private const val KEY_ENABLED = "autostart_enabled"
private const val DEFAULT_ENABLED = true
}
}
@@ -0,0 +1,319 @@
package com.ledgrab.android
import android.bluetooth.BluetoothGatt
import android.bluetooth.BluetoothGattCallback
import android.bluetooth.BluetoothGattCharacteristic
import android.bluetooth.BluetoothManager
import android.bluetooth.BluetoothProfile
import android.bluetooth.le.ScanCallback
import android.bluetooth.le.ScanResult
import android.content.Context
import android.os.Build
import android.os.Handler
import android.os.HandlerThread
import android.util.Log
import java.util.Collections
import java.util.UUID
import java.util.concurrent.ConcurrentHashMap
import java.util.concurrent.atomic.AtomicInteger
import kotlinx.coroutines.CompletableDeferred
import kotlinx.coroutines.runBlocking
import kotlinx.coroutines.withTimeout
import kotlinx.coroutines.TimeoutCancellationException
/**
* Android BLE bridge exposed to the Python server via Chaquopy.
*
* Wraps the Android BluetoothGatt / BluetoothLeScanner APIs into
* synchronous, blocking calls that can be safely invoked from
* a Python thread (Chaquopy proxy threads are real OS threads).
*
* All GATT callbacks run on a private [HandlerThread] so they don't
* block the main looper. [runBlocking] is used to bridge callback
* completions back to the calling Python thread.
*
* Python callers access the singleton via
* `BleBridge.INSTANCE.scan()` etc. — see
* `server/src/ledgrab/core/devices/android_ble_transport.py`.
*/
object BleBridge {
private const val TAG = "BleBridge"
private const val CONNECT_TIMEOUT_MS = 18_000L // connect + service discovery
private const val WRITE_TIMEOUT_MS = 5_000L
@Volatile private var appContext: Context? = null
private val handleSeq = AtomicInteger(1)
// Dedicated looper thread so BLE callbacks don't land on the main thread.
private val bleHandlerThread = HandlerThread("LedGrab-BLE").also { it.start() }
private val bleHandler = Handler(bleHandlerThread.looper)
private data class GattHandle(
val gatt: BluetoothGatt,
val writeChar: BluetoothGattCharacteristic,
)
private val handles = ConcurrentHashMap<Int, GattHandle>()
// Write completion futures, keyed by handle. Only populated for
// WRITE_TYPE_DEFAULT (with-response) writes.
private val pendingWrites = ConcurrentHashMap<Int, CompletableDeferred<Boolean>>()
/** Called once from [LedGrabApp.onCreate] to bind the application context. */
@JvmStatic
fun init(context: Context) {
appContext = context.applicationContext
}
private fun ctx(): Context =
appContext ?: error("BleBridge.init() not called — app context unavailable")
private fun adapter() =
ctx().getSystemService(BluetoothManager::class.java)?.adapter
// ─── Public API ──────────────────────────────────────────────────────────
/**
* Scan for BLE peripherals for [timeoutMs] milliseconds.
*
* Returns a list of `"address|name|rssi"` strings. Addresses are
* deduplicated — only the last-seen RSSI for each address is kept.
* Returns an empty list if Bluetooth is off or the permission is denied.
*/
@JvmStatic
@JvmOverloads
fun scan(timeoutMs: Long = 4_000L): List<String> {
val adapter = adapter() ?: return emptyList()
if (!adapter.isEnabled) return emptyList()
val scanner = try { adapter.bluetoothLeScanner } catch (_: SecurityException) { null }
?: return emptyList()
val seen = Collections.synchronizedMap(LinkedHashMap<String, String>())
val callback = object : ScanCallback() {
override fun onScanResult(callbackType: Int, result: ScanResult) {
val address = result.device.address ?: return
val name = result.scanRecord?.deviceName ?: result.device.name ?: ""
seen[address] = "$address|$name|${result.rssi}"
}
override fun onScanFailed(errorCode: Int) {
Log.w(TAG, "BLE scan failed with error $errorCode")
}
}
try {
// startScan runs on the BLE handler thread; a denied
// BLUETOOTH_SCAN throws SecurityException there, which would
// crash the whole process (an uncaught exception on a handler
// thread is fatal). Catch it inside the posted body and report.
bleHandler.post {
try {
scanner.startScan(callback)
} catch (e: SecurityException) {
Log.w(TAG, "BLUETOOTH_SCAN permission denied — scan skipped", e)
} catch (e: Exception) {
Log.w(TAG, "BLE startScan failed: ${e.message}")
}
}
Thread.sleep(timeoutMs)
} catch (_: InterruptedException) {
Thread.currentThread().interrupt()
} finally {
bleHandler.post {
try {
scanner.stopScan(callback)
} catch (e: SecurityException) {
Log.w(TAG, "BLUETOOTH_SCAN permission denied — stopScan skipped", e)
} catch (e: Exception) {
Log.w(TAG, "BLE stopScan failed: ${e.message}")
}
}
}
return seen.values.toList()
}
/**
* Connect to the BLE peripheral at [address] and locate the GATT
* characteristic identified by [writeCharUuid] across all services.
*
* Blocks until connected + services discovered, or returns -1 on failure.
* The returned integer is an opaque handle passed to [write]/[disconnect].
*/
@JvmStatic
fun connect(address: String, writeCharUuid: String): Int {
val adapter = adapter() ?: return -1
val device = try { adapter.getRemoteDevice(address) } catch (e: Exception) {
Log.e(TAG, "Invalid BLE address '$address': ${e.message}")
return -1
}
val readyDeferred = CompletableDeferred<Boolean>()
val callback = object : BluetoothGattCallback() {
override fun onConnectionStateChange(gatt: BluetoothGatt, status: Int, newState: Int) {
when {
newState == BluetoothProfile.STATE_CONNECTED
&& status == BluetoothGatt.GATT_SUCCESS -> {
Log.d(TAG, "GATT connected to $address, discovering services")
// Runs on the BLE handler thread; a denied
// BLUETOOTH_CONNECT throws SecurityException here, which
// would crash the process. Catch and fail the connect.
try {
gatt.discoverServices()
} catch (e: SecurityException) {
Log.e(TAG, "BLUETOOTH_CONNECT denied during discoverServices", e)
readyDeferred.complete(false)
} catch (e: Exception) {
Log.w(TAG, "discoverServices failed: ${e.message}")
readyDeferred.complete(false)
}
}
newState == BluetoothProfile.STATE_DISCONNECTED -> {
Log.w(TAG, "GATT disconnected from $address (status=$status)")
readyDeferred.complete(false)
}
}
}
override fun onServicesDiscovered(gatt: BluetoothGatt, status: Int) {
readyDeferred.complete(status == BluetoothGatt.GATT_SUCCESS)
}
override fun onCharacteristicWrite(
gatt: BluetoothGatt,
characteristic: BluetoothGattCharacteristic,
status: Int,
) {
val h = handles.entries.firstOrNull { it.value.gatt === gatt }?.key ?: return
pendingWrites.remove(h)?.complete(status == BluetoothGatt.GATT_SUCCESS)
}
}
val gatt: BluetoothGatt = try {
if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.O) {
device.connectGatt(
ctx(), false, callback,
android.bluetooth.BluetoothDevice.TRANSPORT_LE,
android.bluetooth.BluetoothDevice.PHY_LE_1M_MASK,
bleHandler,
)
} else {
@Suppress("DEPRECATION")
device.connectGatt(
ctx(), false, callback,
android.bluetooth.BluetoothDevice.TRANSPORT_LE,
)
}
} catch (e: SecurityException) {
Log.e(TAG, "BLUETOOTH_CONNECT permission denied for $address", e)
return -1
} catch (e: Exception) {
Log.e(TAG, "connectGatt failed for $address", e)
return -1
}
val ready = try {
runBlocking { withTimeout(CONNECT_TIMEOUT_MS) { readyDeferred.await() } }
} catch (_: TimeoutCancellationException) {
Log.e(TAG, "BLE connect+discovery timed out for $address")
runCatching { gatt.close() }
return -1
}
if (!ready) {
runCatching { gatt.close() }
return -1
}
val charUuid = try { UUID.fromString(writeCharUuid) } catch (e: Exception) {
Log.e(TAG, "Invalid characteristic UUID '$writeCharUuid'")
gatt.disconnect(); gatt.close()
return -1
}
val writeChar = gatt.services.flatMap { it.characteristics }
.firstOrNull { it.uuid == charUuid }
if (writeChar == null) {
Log.e(TAG, "Characteristic $writeCharUuid not found on $address")
gatt.disconnect(); gatt.close()
return -1
}
val handle = handleSeq.getAndIncrement()
handles[handle] = GattHandle(gatt, writeChar)
Log.i(TAG, "BLE connected: address=$address char=$writeCharUuid handle=$handle")
return handle
}
/**
* Write [data] to the characteristic associated with [handle].
*
* [withResponse] controls the GATT write type:
* - `true` → Write Request (waits for device ACK, slower but reliable)
* - `false` → Write Command (fire-and-forget, faster, used by SP110E/Triones/Zengge)
*
* Returns `true` on success, `false` on any error.
*/
@JvmStatic
fun write(handle: Int, data: ByteArray, withResponse: Boolean): Boolean {
val entry = handles[handle] ?: return false
val gatt = entry.gatt
val char = entry.writeChar
val writeType = if (withResponse)
BluetoothGattCharacteristic.WRITE_TYPE_DEFAULT
else
BluetoothGattCharacteristic.WRITE_TYPE_NO_RESPONSE
return if (withResponse) {
val deferred = CompletableDeferred<Boolean>()
pendingWrites[handle] = deferred
val initiated = gattWrite(gatt, char, data, writeType)
if (!initiated) {
pendingWrites.remove(handle)
return false
}
try {
runBlocking { withTimeout(WRITE_TIMEOUT_MS) { deferred.await() } }
} catch (_: TimeoutCancellationException) {
pendingWrites.remove(handle)
Log.w(TAG, "BLE write-with-response timed out on handle $handle")
false
}
} else {
gattWrite(gatt, char, data, writeType)
}
}
/** Disconnect and close the GATT connection for [handle]. */
@JvmStatic
fun disconnect(handle: Int) {
val entry = handles.remove(handle) ?: return
pendingWrites.remove(handle)?.complete(false)
runCatching {
entry.gatt.disconnect()
entry.gatt.close()
}.onFailure { Log.w(TAG, "BLE disconnect error for handle $handle: ${it.message}") }
Log.i(TAG, "BLE disconnected handle=$handle")
}
// ─── Internal helpers ─────────────────────────────────────────────────────
private fun gattWrite(
gatt: BluetoothGatt,
char: BluetoothGattCharacteristic,
data: ByteArray,
writeType: Int,
): Boolean = if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.TIRAMISU) {
gatt.writeCharacteristic(char, data, writeType) ==
android.bluetooth.BluetoothStatusCodes.SUCCESS
} else {
@Suppress("DEPRECATION")
char.writeType = writeType
@Suppress("DEPRECATION")
char.value = data
@Suppress("DEPRECATION")
gatt.writeCharacteristic(char)
}
}
@@ -0,0 +1,62 @@
package com.ledgrab.android
import android.content.BroadcastReceiver
import android.content.Context
import android.content.Intent
import android.util.Log
import androidx.core.content.ContextCompat
/**
* Starts the LedGrab capture service automatically on device boot and
* after the app is updated.
*
* Autostart is best-effort and deliberately limited:
* - Requires root. MediaProjection consent cannot be granted silently,
* so we can't resume capture on unrooted boxes without the user
* walking up to the TV and tapping Start.
* - Gated behind [AutostartPrefs] so the user can opt out.
*
* Receivers declared with BOOT_COMPLETED must be fast — we hand off to
* a foreground service within milliseconds and let the service do the
* heavy lifting (Chaquopy bootstrap, pipeline setup).
*/
class BootReceiver : BroadcastReceiver() {
override fun onReceive(context: Context, intent: Intent) {
val action = intent.action
Log.i(TAG, "Boot event: $action")
if (action != Intent.ACTION_BOOT_COMPLETED
&& action != Intent.ACTION_LOCKED_BOOT_COMPLETED
&& action != Intent.ACTION_MY_PACKAGE_REPLACED
) {
return
}
if (!AutostartPrefs(context).isEnabled) {
Log.i(TAG, "Autostart disabled — skipping")
return
}
// Cheap check first (no process spawn). The real `su -c id` probe
// would block and may not complete before the OS reclaims us.
if (!Root.looksRooted()) {
Log.i(TAG, "Device not rooted — cannot autostart without MediaProjection consent")
return
}
try {
ContextCompat.startForegroundService(
context,
CaptureService.createRootIntent(context),
)
Log.i(TAG, "LedGrab capture service dispatched on boot")
} catch (e: Exception) {
Log.e(TAG, "Failed to start service on boot", e)
}
}
companion object {
private const val TAG = "BootReceiver"
}
}
@@ -0,0 +1,411 @@
package com.ledgrab.android
import android.Manifest
import android.annotation.SuppressLint
import android.content.Context
import android.content.pm.PackageManager
import android.graphics.ImageFormat
import android.hardware.camera2.CameraCaptureSession
import android.hardware.camera2.CameraCharacteristics
import android.hardware.camera2.CameraDevice
import android.hardware.camera2.CameraManager
import android.media.Image
import android.media.ImageReader
import android.os.Handler
import android.os.HandlerThread
import android.os.SystemClock
import android.util.Log
import android.util.Size
import android.view.Surface
import com.chaquo.python.PyObject
import com.chaquo.python.Python
import kotlin.coroutines.resume
import kotlin.coroutines.resumeWithException
import kotlinx.coroutines.runBlocking
import kotlinx.coroutines.suspendCancellableCoroutine
import kotlinx.coroutines.withTimeout
import org.json.JSONArray
import org.json.JSONObject
/**
* Android camera bridge exposed to the Python server via Chaquopy.
*
* Wraps the Camera2 API into synchronous, blocking calls that can be
* invoked from a Python thread (Chaquopy proxy threads are real OS
* threads). The physical camera is opened **on demand** — Python's
* `android_camera_engine` calls [startCamera] when a capture stream
* initializes and [stopCamera] when it cleans up, so the camera-in-use
* indicator and battery cost are limited to actual use.
*
* Each captured frame is converted YUV_420_888 → RGB and pushed to the
* Python engine's `push_frame`, mirroring how [ScreenCapture] feeds
* `mediaprojection_engine`. Camera2 callbacks run on a private
* [HandlerThread] so they never touch the main looper.
*
* Python callers access the singleton via
* `jclass("com.ledgrab.android.CameraBridge").INSTANCE` — see
* `server/src/ledgrab/core/capture_engines/android_camera_engine.py`.
*/
object CameraBridge {
private const val TAG = "CameraBridge"
private const val ENGINE_MODULE = "ledgrab.core.capture_engines.android_camera_engine"
private const val OPEN_TIMEOUT_MS = 8_000L
private const val MAX_IMAGES = 2
private const val TARGET_FPS = 20
// "auto" capture size — balanced for ambient LED sampling (the LED
// pipeline downscales anyway), kept modest so the per-frame YUV→RGB
// conversion stays cheap on low-end TV boxes.
private const val DEFAULT_W = 1280
private const val DEFAULT_H = 720
private const val BYTES_PER_RGB = 3
@Volatile private var appContext: Context? = null
// Dedicated looper thread so Camera2 callbacks don't land on main.
private val camThread = HandlerThread("LedGrab-Camera").also { it.start() }
private val camHandler = Handler(camThread.looper)
// Active session state — guarded by [lock]. One camera at a time.
private val lock = Any()
private var cameraDevice: CameraDevice? = null
private var captureSession: CameraCaptureSession? = null
private var imageReader: ImageReader? = null
@Volatile private var running = false
private var activeIndex = -1
// Cached Python engine module handle for the per-frame push fast path.
@Volatile private var engineModule: PyObject? = null
// Reusable conversion buffers — sized once per session (output size is
// fixed for the session), reused to avoid per-frame GC churn on TV boxes.
private var rgbBuffer: ByteArray? = null
private var yBuf: ByteArray? = null
private var uBuf: ByteArray? = null
private var vBuf: ByteArray? = null
// Monotonic frame pacing (mirrors ScreenCapture's accumulator).
private val frameIntervalNanos = 1_000_000_000L / TARGET_FPS.coerceAtLeast(1)
private var nextFrameNanos = 0L
/** Called once from [LedGrabApp.onCreate] to bind the application context. */
@JvmStatic
fun init(context: Context) {
appContext = context.applicationContext
}
/**
* Enumerate cameras as a JSON array string the Python engine parses:
* `[{"index":0,"name":"Back camera","facing":"back","cameraId":"0"}, ...]`
*
* Indices are stable (positional in [CameraManager.cameraIdList]) so
* Python's `display_index` maps 1:1 to [startCamera]'s `index`.
* Enumeration needs no CAMERA permission. Returns `[]` on any error.
*/
@JvmStatic
fun listCameras(): String {
val arr = JSONArray()
val ctx = appContext
if (ctx == null) {
Log.w(TAG, "listCameras: context not bound (init not called)")
return arr.toString()
}
try {
val mgr = ctx.getSystemService(Context.CAMERA_SERVICE) as CameraManager
mgr.cameraIdList.forEachIndexed { idx, id ->
val facing = facingOf(mgr, id)
val name = when (facing) {
"front" -> "Front camera"
"back" -> "Back camera"
"external" -> "External camera $idx"
else -> "Camera $idx"
}
arr.put(
JSONObject()
.put("index", idx)
.put("name", name)
.put("facing", facing)
.put("cameraId", id),
)
}
} catch (e: Exception) {
Log.w(TAG, "listCameras failed: ${e.message}")
}
return arr.toString()
}
/**
* Open camera [index] and start streaming RGB frames to Python.
* Blocks until the capture session is configured (or fails/times out).
*
* Returns false — without throwing across the JNI boundary — when the
* CAMERA permission is missing, the index is out of range, or the
* device/session fails to configure. Closes any previously-open camera
* first (one active at a time).
*/
@SuppressLint("MissingPermission")
@JvmStatic
fun startCamera(index: Int, width: Int, height: Int): Boolean {
synchronized(lock) {
closeLocked()
val ctx = appContext ?: run {
Log.w(TAG, "startCamera: context not bound")
return false
}
if (ctx.checkSelfPermission(Manifest.permission.CAMERA)
!= PackageManager.PERMISSION_GRANTED
) {
Log.w(TAG, "startCamera: CAMERA permission not granted")
return false
}
val mgr = ctx.getSystemService(Context.CAMERA_SERVICE) as CameraManager
val ids = try {
mgr.cameraIdList
} catch (e: Exception) {
Log.w(TAG, "startCamera: cameraIdList failed: ${e.message}")
return false
}
if (index < 0 || index >= ids.size) {
Log.w(TAG, "startCamera: index $index out of range (${ids.size} cameras)")
return false
}
val cameraId = ids[index]
val size = chooseSize(mgr, cameraId, width, height) ?: run {
Log.w(TAG, "startCamera: no YUV output sizes for camera $index")
return false
}
val reader = ImageReader.newInstance(
size.width, size.height, ImageFormat.YUV_420_888, MAX_IMAGES,
)
// Size the conversion buffers once for this session.
rgbBuffer = ByteArray(size.width * size.height * BYTES_PER_RGB)
yBuf = null; uBuf = null; vBuf = null
nextFrameNanos = SystemClock.elapsedRealtimeNanos()
reader.setOnImageAvailableListener({ r -> onFrame(r) }, camHandler)
return try {
runBlocking {
withTimeout(OPEN_TIMEOUT_MS) {
// Publish each resource to its field as soon as it exists so
// closeLocked() (in the catch) can release it if a LATER step
// throws. Assigning only after setRepeatingRequest succeeds
// would orphan the opened CameraDevice on a createSession /
// setRepeatingRequest failure (camera stuck on; subsequent
// opens fail with CAMERA_IN_USE).
imageReader = reader
val device = openCamera(mgr, cameraId)
cameraDevice = device
val session = createSession(device, reader.surface)
captureSession = session
val request = device.createCaptureRequest(CameraDevice.TEMPLATE_PREVIEW)
.apply { addTarget(reader.surface) }
.build()
session.setRepeatingRequest(request, null, camHandler)
activeIndex = index
running = true
Log.i(TAG, "Camera $index opened (${size.width}x${size.height} @ ${TARGET_FPS}fps)")
true
}
}
} catch (e: Exception) {
Log.e(TAG, "startCamera($index) failed: ${e.message}")
// imageReader/cameraDevice/captureSession are now whatever got
// assigned before the failure — closeLocked releases each exactly
// once (idempotent, runCatching-wrapped).
closeLocked()
false
}
}
}
/** Stop streaming and release the camera. Idempotent; safe if not started. */
@JvmStatic
fun stopCamera() {
synchronized(lock) { closeLocked() }
Log.i(TAG, "Camera stopped")
}
// ── internals ────────────────────────────────────────────────────────
private fun facingOf(mgr: CameraManager, id: String): String =
when (mgr.getCameraCharacteristics(id).get(CameraCharacteristics.LENS_FACING)) {
CameraCharacteristics.LENS_FACING_FRONT -> "front"
CameraCharacteristics.LENS_FACING_BACK -> "back"
CameraCharacteristics.LENS_FACING_EXTERNAL -> "external"
else -> "unknown"
}
/** Pick the supported YUV size closest in area to the request (or the
* balanced default for `auto`/0). */
private fun chooseSize(mgr: CameraManager, cameraId: String, reqW: Int, reqH: Int): Size? {
val map = mgr.getCameraCharacteristics(cameraId)
.get(CameraCharacteristics.SCALER_STREAM_CONFIGURATION_MAP) ?: return null
val sizes = map.getOutputSizes(ImageFormat.YUV_420_888)
if (sizes == null || sizes.isEmpty()) return null
val targetArea = (if (reqW > 0) reqW else DEFAULT_W).toLong() *
(if (reqH > 0) reqH else DEFAULT_H)
return sizes.minByOrNull { kotlin.math.abs(it.width.toLong() * it.height - targetArea) }
}
@SuppressLint("MissingPermission")
private suspend fun openCamera(mgr: CameraManager, cameraId: String): CameraDevice =
suspendCancellableCoroutine { cont ->
mgr.openCamera(cameraId, object : CameraDevice.StateCallback() {
override fun onOpened(device: CameraDevice) {
if (cont.isActive) cont.resume(device) else device.close()
}
override fun onDisconnected(device: CameraDevice) {
device.close()
if (cont.isActive) cont.resumeWithException(IllegalStateException("camera disconnected"))
}
override fun onError(device: CameraDevice, error: Int) {
device.close()
if (cont.isActive) cont.resumeWithException(IllegalStateException("camera error $error"))
}
}, camHandler)
}
@Suppress("DEPRECATION")
private suspend fun createSession(device: CameraDevice, surface: Surface): CameraCaptureSession =
suspendCancellableCoroutine { cont ->
// createCaptureSession(List, callback, handler) is deprecated at
// API 30 but is the correct API down to minSdk 24 (the
// SessionConfiguration overload is API 28+).
device.createCaptureSession(
listOf(surface),
object : CameraCaptureSession.StateCallback() {
override fun onConfigured(session: CameraCaptureSession) {
if (cont.isActive) cont.resume(session)
}
override fun onConfigureFailed(session: CameraCaptureSession) {
if (cont.isActive) cont.resumeWithException(IllegalStateException("session configure failed"))
}
},
camHandler,
)
}
/** ImageReader callback — paced, converts YUV→RGB, pushes to Python. */
private fun onFrame(reader: ImageReader) {
if (!running) {
runCatching { reader.acquireLatestImage()?.close() }
return
}
val now = SystemClock.elapsedRealtimeNanos()
if (now < nextFrameNanos) {
runCatching { reader.acquireLatestImage()?.close() }
return
}
val image = runCatching { reader.acquireLatestImage() }.getOrNull() ?: return
try {
val w = image.width
val h = image.height
val out = ensureRgbBuffer(w * h * BYTES_PER_RGB)
yuv420ToRgb(image, out, w, h)
pushFrame(out, w, h)
nextFrameNanos += frameIntervalNanos
if (now - nextFrameNanos > frameIntervalNanos * 4) {
nextFrameNanos = now + frameIntervalNanos
}
} catch (e: Exception) {
Log.w(TAG, "frame processing error: ${e.message}")
} finally {
runCatching { image.close() }
}
}
private fun ensureRgbBuffer(size: Int): ByteArray {
val buf = rgbBuffer
if (buf != null && buf.size == size) return buf
return ByteArray(size).also { rgbBuffer = it }
}
/**
* Stride-aware YUV_420_888 → packed RGB (3 bytes/px) using BT.601
* fixed-point coefficients. Handles both planar and semi-planar
* (NV21-like, pixelStride 2) chroma layouts via the plane strides.
*/
private fun yuv420ToRgb(image: Image, out: ByteArray, width: Int, height: Int) {
val planes = image.planes
val yPlane = planes[0]
val uPlane = planes[1]
val vPlane = planes[2]
val yRowStride = yPlane.rowStride
val yPixStride = yPlane.pixelStride
val uRowStride = uPlane.rowStride
val uPixStride = uPlane.pixelStride
val vRowStride = vPlane.rowStride
val vPixStride = vPlane.pixelStride
// Copy each plane to a reusable array for fast indexed access
// (ByteBuffer absolute-get per pixel is far slower).
val yByteBuf = yPlane.buffer
val uByteBuf = uPlane.buffer
val vByteBuf = vPlane.buffer
val yArr = ensurePlane(yBuf, yByteBuf.remaining()).also { yBuf = it }
val uArr = ensurePlane(uBuf, uByteBuf.remaining()).also { uBuf = it }
val vArr = ensurePlane(vBuf, vByteBuf.remaining()).also { vBuf = it }
yByteBuf.get(yArr, 0, yArr.size)
uByteBuf.get(uArr, 0, uArr.size)
vByteBuf.get(vArr, 0, vArr.size)
var o = 0
for (row in 0 until height) {
val yRowBase = row * yRowStride
val uvRow = row shr 1
val uRowBase = uvRow * uRowStride
val vRowBase = uvRow * vRowStride
for (col in 0 until width) {
val y = (yArr[yRowBase + col * yPixStride].toInt() and 0xFF)
val uvCol = col shr 1
val u = (uArr[uRowBase + uvCol * uPixStride].toInt() and 0xFF) - 128
val v = (vArr[vRowBase + uvCol * vPixStride].toInt() and 0xFF) - 128
// BT.601 full-range, fixed-point (<<16).
var r = y + ((91881 * v) shr 16)
var g = y - ((22554 * u + 46802 * v) shr 16)
var b = y + ((116130 * u) shr 16)
if (r < 0) r = 0 else if (r > 255) r = 255
if (g < 0) g = 0 else if (g > 255) g = 255
if (b < 0) b = 0 else if (b > 255) b = 255
out[o++] = r.toByte()
out[o++] = g.toByte()
out[o++] = b.toByte()
}
}
}
/** Return [cached] if it already fits [n] bytes, else a fresh array. */
private fun ensurePlane(cached: ByteArray?, n: Int): ByteArray =
if (cached != null && cached.size == n) cached else ByteArray(n)
private fun pushFrame(rgb: ByteArray, width: Int, height: Int) {
val module = engineModule ?: runCatching {
Python.getInstance().getModule(ENGINE_MODULE)
}.getOrNull()?.also { engineModule = it } ?: return
try {
module.callAttr("push_frame", rgb, width, height)
} catch (e: Exception) {
Log.w(TAG, "push_frame failed: ${e.message}")
}
}
/** Tear down the active session. Caller holds [lock]. */
private fun closeLocked() {
running = false
activeIndex = -1
runCatching { imageReader?.setOnImageAvailableListener(null, null) }
runCatching { captureSession?.stopRepeating() }
runCatching { captureSession?.close() }
captureSession = null
runCatching { cameraDevice?.close() }
cameraDevice = null
runCatching { imageReader?.close() }
imageReader = null
}
}
@@ -0,0 +1,476 @@
package com.ledgrab.android
import android.app.Notification
import android.app.NotificationChannel
import android.app.NotificationManager
import android.app.PendingIntent
import android.Manifest
import android.app.Service
import android.content.Context
import android.content.Intent
import android.content.pm.PackageManager
import android.content.pm.ServiceInfo
import android.media.projection.MediaProjection
import android.media.projection.MediaProjectionManager
import android.os.Build
import android.os.IBinder
import android.util.DisplayMetrics
import android.util.Log
import android.view.WindowManager
import androidx.core.app.NotificationCompat
import androidx.core.app.ServiceCompat
import kotlinx.coroutines.CoroutineScope
import kotlinx.coroutines.Dispatchers
import kotlinx.coroutines.Job
import kotlinx.coroutines.SupervisorJob
import kotlinx.coroutines.cancel
import kotlinx.coroutines.delay
import kotlinx.coroutines.isActive
import kotlinx.coroutines.launch
/**
* Foreground service that runs the Python LedGrab server and captures
* the screen via MediaProjection or root screenrecord.
*
* On Android 14+ the foreground-service "type" must match the work
* being done. We promote the service with the correct type (mediaProjection
* for the consent path, specialUse for the root path) instead of
* declaring a single fixed type in the manifest — the manifest now
* declares the *union* so promotion at runtime is permitted.
*/
class CaptureService : Service() {
companion object {
private const val TAG = "CaptureService"
private const val CHANNEL_ID = "ledgrab_capture"
private const val NOTIFICATION_ID = 1
private const val EXTRA_RESULT_CODE = "result_code"
private const val EXTRA_RESULT_DATA = "result_data"
private const val EXTRA_USE_ROOT = "use_root"
private const val SERVER_PORT = 8080
private const val CAPTURE_WIDTH = 480
private const val CAPTURE_HEIGHT = 270
private const val CAPTURE_FPS = 30
// Watchdog: if no new root-capture frames arrive inside this
// window the pipeline is considered stalled and respawned.
// Grace gives screenrecord time to produce its first I-frame.
private const val WATCHDOG_GRACE_MS = 5_000L
private const val WATCHDOG_CHECK_MS = 5_000L
private const val WATCHDOG_MAX_RESTARTS = 3
/** True while the service is alive. Survives activity recreation. */
@Volatile
@JvmStatic
var isRunning: Boolean = false
private set
fun createIntent(
context: Context,
resultCode: Int,
resultData: Intent,
): Intent {
return Intent(context, CaptureService::class.java).apply {
putExtra(EXTRA_RESULT_CODE, resultCode)
putExtra(EXTRA_RESULT_DATA, resultData)
}
}
/** Root-mode intent — no MediaProjection consent data required. */
fun createRootIntent(context: Context): Intent {
return Intent(context, CaptureService::class.java).apply {
putExtra(EXTRA_USE_ROOT, true)
}
}
}
private var bridge: PythonBridge? = null
private var screenCapture: ScreenCapture? = null
private var rootCapture: RootScreenrecord? = null
private var audioCapture: AudioCapture? = null
private var mediaProjection: MediaProjection? = null
// Service-scoped coroutine scope for the root-capture watchdog.
// SupervisorJob so a watchdog crash doesn't tear down other children.
private val serviceScope = CoroutineScope(SupervisorJob() + Dispatchers.Default)
private var watchdogJob: Job? = null
override fun onBind(intent: Intent?): IBinder? = null
override fun onCreate() {
super.onCreate()
createNotificationChannel()
}
override fun onStartCommand(intent: Intent?, flags: Int, startId: Int): Int {
val useRoot = intent?.getBooleanExtra(EXTRA_USE_ROOT, false) ?: false
// CRITICAL (Android 14+): for the MediaProjection path, validate the
// projection token BEFORE promoting to a foreground service with the
// mediaProjection FGS type. On service recreation (system redelivery
// or a stale relaunch) the consent token is gone — promoting first and
// then discovering the dead token causes a spurious foreground-service
// start + immediate stop, which on strict OEMs flickers the
// notification or trips a stopSelf loop. Bail out cleanly here, before
// startForeground, when the MediaProjection consent data is missing.
val localIp = NetworkUtils.getLocalIpAddress(this) ?: ""
val url = "http://$localIp:$SERVER_PORT"
val mediaProjectionResultData: Intent? =
if (!useRoot) extractProjectionResultData(intent) else null
if (!useRoot && (intent == null || mediaProjectionResultData == null)) {
// MediaProjection mode can't recover from a redelivery —
// the consent token in the original intent is single-use.
//
// We were launched via startForegroundService(), so the OS REQUIRES
// a startForeground() within ~5s even on this immediate-stop path,
// or it raises the fatal ForegroundServiceDidNotStartInTimeException.
// Promote with a benign SPECIAL_USE type (NOT mediaProjection — we
// have no valid consent token, and requesting that type without an
// active projection is exactly what we're avoiding) just long enough
// to satisfy the contract, then stop.
Log.w(TAG, "MediaProjection start without a valid consent token — stopping")
runCatching {
val bailType = if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.UPSIDE_DOWN_CAKE) {
ServiceInfo.FOREGROUND_SERVICE_TYPE_SPECIAL_USE
} else {
0
}
ServiceCompat.startForeground(this, NOTIFICATION_ID, buildNotification(url), bailType)
}.onFailure { Log.w(TAG, "Bail-path startForeground failed: ${it.message}") }
stopSelf()
return START_NOT_STICKY
}
// startForeground must be called IMMEDIATELY after the token check —
// before any heavier work like getMediaProjection(). The service type
// must match the work; pass it explicitly via ServiceCompat so we stay
// compatible back to API 24. The MEDIA_PROJECTION type is only used
// here once resultData is confirmed non-null (checked above).
try {
val type = if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.UPSIDE_DOWN_CAKE) {
var t = if (useRoot) {
ServiceInfo.FOREGROUND_SERVICE_TYPE_SPECIAL_USE
} else {
ServiceInfo.FOREGROUND_SERVICE_TYPE_MEDIA_PROJECTION
}
// On-demand webcam capture opens the camera from this service.
// To retain camera access once the app is backgrounded (the
// always-on ambient-lighting case), API 34+ requires the camera
// FGS type. Add it ONLY when CAMERA is already granted — promoting
// with the camera type without the runtime permission throws and
// would kill the whole service on the (common) camera-less or
// not-yet-granted box. If CAMERA is granted later, it takes effect
// on the next Start (matches the audio/permission UX).
if (checkSelfPermission(Manifest.permission.CAMERA) ==
PackageManager.PERMISSION_GRANTED
) {
t = t or ServiceInfo.FOREGROUND_SERVICE_TYPE_CAMERA
}
t
} else {
0
}
ServiceCompat.startForeground(
this,
NOTIFICATION_ID,
buildNotification(url),
type,
)
} catch (e: Exception) {
// Most common cause: missing foregroundServiceType permission
// or denied POST_NOTIFICATIONS on API 33+.
Log.e(TAG, "startForeground failed — service cannot run", e)
stopSelf()
return START_NOT_STICKY
}
// Only flip the public flag once the FG transition has succeeded,
// otherwise `isRunning=true` sticks forever when startForeground throws.
isRunning = true
try {
if (useRoot) {
startRootCapture(url)
} else {
// mediaProjectionResultData is guaranteed non-null here — the
// token was validated before startForeground above.
startMediaProjectionCapture(intent!!, mediaProjectionResultData!!, url)
}
} catch (e: Exception) {
Log.e(TAG, "Failed to start capture", e)
isRunning = false
stopSelf()
}
// Root mode can be cleanly restarted from the original intent
// because it carries no perishable consent token. MediaProjection
// mode cannot — a restart there would silently start the service
// with a dead projection. START_NOT_STICKY there, the user has
// to tap Start again.
return if (useRoot) START_REDELIVER_INTENT else START_NOT_STICKY
}
private fun apiKey(): String? =
(application as? LedGrabApp)?.apiKeyManager?.apiKey
private fun startRootCapture(url: String) {
val newBridge = PythonBridge(this).also { b ->
b.configureRootCapture(CAPTURE_WIDTH, CAPTURE_HEIGHT)
b.startServer(SERVER_PORT, apiKey())
}
bridge = newBridge
val pipeline = RootScreenrecord(
bridge = newBridge,
width = CAPTURE_WIDTH,
height = CAPTURE_HEIGHT,
fps = CAPTURE_FPS,
)
if (!pipeline.start()) {
Log.w(TAG, "Root capture failed to launch — stopping service")
stopSelf()
return
}
rootCapture = pipeline
startWatchdog()
Log.i(TAG, "LedGrab service started (root mode) — web UI at $url")
}
/**
* Replace the active root pipeline with a fresh instance, reusing
* the existing Python bridge (no server restart). Returns true if
* the new pipeline launched, false otherwise.
*
* Synchronized so a concurrent onDestroy() either (a) sees the old
* instance and stops it then null-out, or (b) sees the new instance
* and stops it. There is no window where a fresh instance can be
* orphaned with no one holding a reference to it.
*/
@Synchronized
private fun restartRootPipeline(): Boolean {
val currentBridge = bridge ?: return false
// Tear down the old instance first so we don't run two
// screenrecord processes simultaneously fighting for the GPU.
rootCapture?.let { old ->
rootCapture = null
runCatching { old.stop() }
}
val next = RootScreenrecord(
bridge = currentBridge,
width = CAPTURE_WIDTH,
height = CAPTURE_HEIGHT,
fps = CAPTURE_FPS,
)
// Publish BEFORE start() — if onDestroy fires after this
// assignment but before start() completes, the field is non-null
// and onDestroy will stop() it properly. start() is idempotent
// enough (running=true, then resource construction) that being
// raced by stop() at most produces a brief partial-init that
// the next stop() call cleans up.
rootCapture = next
if (!next.start()) {
Log.e(TAG, "Root capture failed to restart")
// start() already called stop() on itself on the failure
// path — but null out the field so the watchdog/onDestroy
// don't try to stop it again.
rootCapture = null
return false
}
return true
}
/**
* Monitor the root pipeline's frame counter. If no new frames
* arrive within one check window, respawn the pipeline. Caps at
* [WATCHDOG_MAX_RESTARTS] consecutive failures before giving up so
* we don't burn battery forever on a device that genuinely can't
* produce frames.
*/
private fun startWatchdog() {
watchdogJob?.cancel()
watchdogJob = serviceScope.launch {
delay(WATCHDOG_GRACE_MS)
var last = rootCapture?.framesDelivered ?: 0
var restartAttempts = 0
while (isActive) {
delay(WATCHDOG_CHECK_MS)
val pipe = rootCapture ?: break
val current = pipe.framesDelivered
if (current == last) {
restartAttempts += 1
Log.w(
TAG,
"Root capture stalled (no new frames in ${WATCHDOG_CHECK_MS}ms); " +
"restart attempt $restartAttempts/$WATCHDOG_MAX_RESTARTS",
)
if (restartAttempts >= WATCHDOG_MAX_RESTARTS) {
Log.e(TAG, "Watchdog gave up after $WATCHDOG_MAX_RESTARTS restarts")
stopSelf()
return@launch
}
if (!restartRootPipeline()) {
stopSelf()
return@launch
}
last = rootCapture?.framesDelivered ?: 0
} else {
// Forward progress — forgive earlier glitches.
restartAttempts = 0
last = current
}
}
}
}
/**
* Extract the single-use MediaProjection consent token from the start
* intent, or null if the intent is missing/redelivered without it.
* Called BEFORE startForeground so the mediaProjection FGS type is only
* ever requested when a valid token is present (see onStartCommand).
*/
private fun extractProjectionResultData(intent: Intent?): Intent? {
if (intent == null) return null
@Suppress("DEPRECATION")
return if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.TIRAMISU) {
intent.getParcelableExtra(EXTRA_RESULT_DATA, Intent::class.java)
} else {
intent.getParcelableExtra(EXTRA_RESULT_DATA)
}
}
private fun startMediaProjectionCapture(intent: Intent, resultData: Intent, url: String) {
val resultCode = intent.getIntExtra(EXTRA_RESULT_CODE, 0)
val projectionManager =
getSystemService(Context.MEDIA_PROJECTION_SERVICE) as MediaProjectionManager
val projection = projectionManager.getMediaProjection(resultCode, resultData)
if (projection == null) {
Log.e(TAG, "Failed to create MediaProjection")
stopSelf()
return
}
mediaProjection = projection
val metrics = if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.R) {
val windowMetrics = (getSystemService(Context.WINDOW_SERVICE) as WindowManager)
.currentWindowMetrics
DisplayMetrics().apply {
val bounds = windowMetrics.bounds
widthPixels = bounds.width()
heightPixels = bounds.height()
densityDpi = resources.displayMetrics.densityDpi
}
} else {
val windowManager = getSystemService(Context.WINDOW_SERVICE) as WindowManager
DisplayMetrics().also { m ->
@Suppress("DEPRECATION")
windowManager.defaultDisplay.getRealMetrics(m)
}
}
val newBridge = PythonBridge(this).also { b ->
b.configureCapture(CAPTURE_WIDTH, CAPTURE_HEIGHT)
b.startServer(SERVER_PORT, apiKey())
}
bridge = newBridge
screenCapture = ScreenCapture(
projection = projection,
metrics = metrics,
bridge = newBridge,
targetWidth = CAPTURE_WIDTH,
targetHeight = CAPTURE_HEIGHT,
targetFps = CAPTURE_FPS,
// If the user taps the system Cast/Screen-capture stop banner,
// MediaProjection.Callback.onStop fires — tear the whole
// service down so the notification/Python server don't linger.
onProjectionStopped = { stopSelf() },
).also { it.start() }
// Reuse the same projection to capture system playback audio so
// audio-reactive lighting works on-device (API 29+, RECORD_AUDIO
// granted). Best-effort: screen capture and the server keep running
// if audio is unavailable. Started AFTER ScreenCapture so the
// projection's callback is already registered.
if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.Q &&
checkSelfPermission(Manifest.permission.RECORD_AUDIO) ==
PackageManager.PERMISSION_GRANTED
) {
audioCapture = AudioCapture(projection, newBridge).also { ac ->
if (!ac.start()) {
Log.i(TAG, "Playback audio capture unavailable — continuing without audio")
audioCapture = null
}
}
} else {
Log.i(TAG, "RECORD_AUDIO not granted or API < 29 — audio-reactive capture disabled")
}
Log.i(TAG, "LedGrab service started (MediaProjection) — web UI at $url")
}
override fun onDestroy() {
isRunning = false
watchdogJob?.cancel()
watchdogJob = null
serviceScope.cancel()
screenCapture?.stop()
screenCapture = null
// Stop audio before the server: stop() calls bridge.shutdownAudio().
audioCapture?.stop()
audioCapture = null
rootCapture?.stop()
rootCapture = null
bridge?.stopServer()
bridge = null
mediaProjection?.stop()
mediaProjection = null
Log.i(TAG, "LedGrab service destroyed")
super.onDestroy()
}
private fun createNotificationChannel() {
if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.O) {
val channel = NotificationChannel(
CHANNEL_ID,
getString(R.string.notification_channel_name),
NotificationManager.IMPORTANCE_LOW,
).apply {
description = getString(R.string.notification_channel_description)
}
val manager = getSystemService(NotificationManager::class.java)
manager.createNotificationChannel(channel)
}
}
private fun buildNotification(url: String): Notification {
val tapIntent = PendingIntent.getActivity(
this,
0,
Intent(this, MainActivity::class.java).apply {
flags = Intent.FLAG_ACTIVITY_SINGLE_TOP
},
PendingIntent.FLAG_IMMUTABLE,
)
return NotificationCompat.Builder(this, CHANNEL_ID)
.setContentTitle(getString(R.string.notification_title))
.setContentText(getString(R.string.notification_text, url))
// ic_notification is a monochrome 24dp vector — status-bar
// icons must be white-on-transparent or they render as a
// gray blob on Android 5+.
.setSmallIcon(R.drawable.ic_notification)
.setColor(0xFF64FFDA.toInt())
.setColorized(true)
.setContentIntent(tapIntent)
.setOngoing(true)
.build()
}
}
@@ -0,0 +1,154 @@
package com.ledgrab.android
import android.app.AppOpsManager
import android.app.usage.UsageEvents
import android.app.usage.UsageStatsManager
import android.content.Context
import android.content.pm.LauncherApps
import android.os.Build
import android.os.Process
import android.util.Log
import org.json.JSONArray
import org.json.JSONObject
/**
* Foreground-app + installed-app bridge exposed to the Python server via Chaquopy.
*
* Backs the Android implementation of the "Application" automation rule
* (foreground app -> activate scene). Desktop detects the foreground process via
* Win32 ctypes in ``platform_detector.py``; Android has no such API, so this
* bridge wraps two in-platform services into synchronous calls a Python thread
* can invoke (Chaquopy proxy threads are real OS threads):
*
* - [getForegroundPackage] via [UsageStatsManager] (needs PACKAGE_USAGE_STATS,
* a special-access permission granted from Settings — see MainActivity).
* - [listLaunchableApps] via [LauncherApps] for the automation editor's app
* picker (no QUERY_ALL_PACKAGES needed — getActivityList is the sanctioned
* launchable-app enumeration API).
* - [hasUsageAccess] so the server / UI can detect the missing grant.
*
* Detection only ever string-compares the foreground *package name*, so no label
* resolution / package visibility is required at match time.
*
* Python callers access the singleton via
* `jclass("com.ledgrab.android.ForegroundAppBridge").INSTANCE` — see
* `server/src/ledgrab/core/automations/platform_detector.py`.
*/
object ForegroundAppBridge {
private const val TAG = "ForegroundAppBridge"
// Trailing window for queryEvents. queryEvents reports discrete foreground
// transitions (not "current app"), and events can lag a few seconds, so we
// look back far enough to reliably catch the latest MOVE_TO_FOREGROUND while
// staying recent enough not to report a stale app on the ~1s automation tick.
private const val WINDOW_MS = 10_000L
@Volatile private var appContext: Context? = null
/** Called once from [LedGrabApp.onCreate] to bind the application context. */
@JvmStatic
fun init(context: Context) {
appContext = context.applicationContext
}
/**
* Package name of the most recently foregrounded app, or null when none is
* found in the trailing window, Usage Access is not granted, or on any error.
* Never throws across the JNI boundary.
*/
@JvmStatic
fun getForegroundPackage(): String? {
val ctx = appContext ?: run {
Log.w(TAG, "getForegroundPackage: context not bound (init not called)")
return null
}
return try {
val usm = ctx.getSystemService(Context.USAGE_STATS_SERVICE) as? UsageStatsManager
?: return null
val end = System.currentTimeMillis()
val events = usm.queryEvents(end - WINDOW_MS, end)
val event = UsageEvents.Event()
var latestPkg: String? = null
var latestTs = Long.MIN_VALUE
while (events.hasNextEvent()) {
events.getNextEvent(event)
// ACTIVITY_RESUMED (API 29+) shares the value of the legacy
// MOVE_TO_FOREGROUND constant, so the single check covers both.
// >= (not >) so that on an exact-timestamp tie the later-iterated
// event wins — events arrive chronologically, so that is the most
// recent foreground transition.
if (event.eventType == UsageEvents.Event.MOVE_TO_FOREGROUND &&
event.timeStamp >= latestTs
) {
latestTs = event.timeStamp
latestPkg = event.packageName
}
}
latestPkg
} catch (e: Exception) {
// SecurityException when access is missing, plus any service error.
Log.w(TAG, "getForegroundPackage failed: ${e.message}")
null
}
}
/** Whether the user has granted Usage Access (PACKAGE_USAGE_STATS) to this app. */
@JvmStatic
fun hasUsageAccess(): Boolean {
val ctx = appContext ?: return false
return try {
val appOps = ctx.getSystemService(Context.APP_OPS_SERVICE) as? AppOpsManager
?: return false
val mode = if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.Q) {
appOps.unsafeCheckOpNoThrow(
AppOpsManager.OPSTR_GET_USAGE_STATS, Process.myUid(), ctx.packageName,
)
} else {
@Suppress("DEPRECATION")
appOps.checkOpNoThrow(
AppOpsManager.OPSTR_GET_USAGE_STATS, Process.myUid(), ctx.packageName,
)
}
mode == AppOpsManager.MODE_ALLOWED
} catch (e: Exception) {
Log.w(TAG, "hasUsageAccess failed: ${e.message}")
false
}
}
/**
* Launchable apps as a JSON array string the Python server parses:
* `[{"package":"com.netflix.mediaclient","label":"Netflix"}, ...]`
*
* Uses [LauncherApps.getActivityList] (launcher + leanback launchables) —
* no QUERY_ALL_PACKAGES. De-duplicated by package, sorted by label.
* Returns `[]` on any error.
*/
@JvmStatic
fun listLaunchableApps(): String {
val arr = JSONArray()
val ctx = appContext ?: run {
Log.w(TAG, "listLaunchableApps: context not bound (init not called)")
return arr.toString()
}
try {
val launcher = ctx.getSystemService(Context.LAUNCHER_APPS_SERVICE) as? LauncherApps
?: return arr.toString()
val seen = HashSet<String>()
val items = ArrayList<Pair<String, String>>()
for (info in launcher.getActivityList(null, Process.myUserHandle())) {
val pkg = info.applicationInfo?.packageName ?: continue
if (!seen.add(pkg)) continue
val label = info.label?.toString().takeUnless { it.isNullOrBlank() } ?: pkg
items.add(pkg to label)
}
items.sortBy { it.second.lowercase() }
for ((pkg, label) in items) {
arr.put(JSONObject().put("package", pkg).put("label", label))
}
} catch (e: Exception) {
Log.w(TAG, "listLaunchableApps failed: ${e.message}")
}
return arr.toString()
}
}
@@ -0,0 +1,120 @@
package com.ledgrab.android
import android.app.Application
import android.util.Log
import com.chaquo.python.Python
import com.chaquo.python.android.AndroidPlatform
import java.io.File
import java.io.PrintWriter
import java.text.SimpleDateFormat
import java.util.Date
import java.util.Locale
/**
* Application class — initializes the Chaquopy Python runtime and
* installs a global uncaught exception handler that persists crash
* logs to app-private storage.
*
* `Python.start()` must be called once before any Python code runs.
* It loads libpython, extracts stdlib + pip packages from APK assets
* (first launch only), and sets up `sys.path`.
*/
class LedGrabApp : Application() {
/** Set if [Python.start] threw — surfaced by MainActivity. */
@Volatile
var initError: Throwable? = null
private set
/** Lazily-initialized API-key manager (see [ApiKeyManager]). */
val apiKeyManager: ApiKeyManager by lazy { ApiKeyManager(this) }
override fun onCreate() {
super.onCreate()
installCrashLogger()
pruneOldCrashLogs()
try {
if (!Python.isStarted()) {
Python.start(AndroidPlatform(this))
}
} catch (t: Throwable) {
// Don't crash here — MainActivity will render a failure
// screen with a Copy log button so the user can report it.
Log.e(TAG, "Python.start() failed", t)
initError = t
return
}
// Bind application context for the USB-serial bridge so Python
// can enumerate and open USB-to-TTL adapters without needing
// an Activity reference.
UsbSerialBridge.init(this)
// Bind application context for the BLE bridge so Python can
// scan and connect to BLE LED controllers.
BleBridge.init(this)
// Bind application context for the camera bridge so Python can
// enumerate cameras and open them on demand (webcam capture).
CameraBridge.init(this)
// Bind application context for the foreground-app bridge so Python can
// detect the foreground app (Application automation rule) and list
// launchable apps for the editor's picker.
ForegroundAppBridge.init(this)
// Pre-warm the API key on a background thread. First-launch
// generation does a SharedPreferences.commit() (synchronous
// disk write — 10-50 ms on slow TV-box flash), which would
// hit the Main thread otherwise when MainActivity / CaptureService
// reads it. Doing it here makes subsequent reads memory-only.
Thread({
runCatching { apiKeyManager.apiKey }
}, "ledgrab-apikey-warmup").apply { isDaemon = true }.start()
}
/**
* Install a global uncaught exception handler that writes the
* stack trace to `files/crash-<timestamp>.log` before letting
* the default handler terminate the process. Logs survive app
* restarts and can be pulled via `adb pull` for diagnostics.
*/
private fun installCrashLogger() {
val defaultHandler = Thread.getDefaultUncaughtExceptionHandler()
Thread.setDefaultUncaughtExceptionHandler { thread, throwable ->
try {
val ts = SimpleDateFormat("yyyyMMdd-HHmmss", Locale.US).format(Date())
val logFile = File(filesDir, "crash-$ts.log")
PrintWriter(logFile).use { pw ->
pw.println("LedGrab crash at $ts")
pw.println("Thread: ${thread.name}")
pw.println()
throwable.printStackTrace(pw)
}
Log.e(TAG, "Crash log written to ${logFile.absolutePath}")
} catch (_: Exception) {
// Best effort — don't crash inside the crash handler.
}
// Chain to the default handler so Android shows the crash dialog
// and terminates the process.
defaultHandler?.uncaughtException(thread, throwable)
}
}
/**
* Keep only the most recent [MAX_CRASH_LOGS] crash files so a
* long-lived install doesn't slowly fill its private storage with
* historical traces. Cheap on every launch — listFiles is O(n)
* but n is tiny by construction.
*/
private fun pruneOldCrashLogs() {
val logs = filesDir.listFiles { f ->
f.isFile && f.name.startsWith("crash-") && f.name.endsWith(".log")
} ?: return
if (logs.size <= MAX_CRASH_LOGS) return
logs.sortedByDescending { it.lastModified() }
.drop(MAX_CRASH_LOGS)
.forEach { runCatching { it.delete() } }
}
companion object {
private const val TAG = "LedGrabApp"
private const val MAX_CRASH_LOGS = 10
}
}
@@ -0,0 +1,178 @@
package com.ledgrab.android
import android.app.Notification
import android.service.notification.NotificationListenerService
import android.service.notification.StatusBarNotification
import android.util.Log
import com.chaquo.python.Python
import java.util.concurrent.ConcurrentHashMap
import java.util.concurrent.ExecutorService
import java.util.concurrent.Executors
import java.util.concurrent.RejectedExecutionException
/**
* Captures posted OS notifications and forwards the posting app's display
* label to the Python notification pipeline, where the existing
* `NotificationColorStripSource` fires its one-shot LED effect.
*
* Direction is Kotlin -> Python via the process-global Chaquopy instance
* (NOT a per-[CaptureService] [PythonBridge]): `system_server` binds this
* service independently of [CaptureService], so it resolves Python itself.
* The Python receiver (`os_notification_listener.push_notification`) is a
* no-op whenever the server/listener isn't running, so a notification
* arriving before — or after — a capture session is safely ignored.
*/
class LedGrabNotificationListener : NotificationListenerService() {
// Serial executor: the Python receiver does a (non-concurrency-safe) history
// disk write and may play a sound, so pushes must not overlap. Off the main
// looper to keep the system service responsive.
//
// Tied to the listener-connection lifecycle (onListenerConnected /
// onListenerDisconnected), NOT onDestroy: this is a system-rebindable
// service, so it can be connected/disconnected multiple times across a
// single onCreate..onDestroy span. Managing the executor here — combined
// with the runCatching guard at the submit site — keeps a notification
// that races teardown from triggering RejectedExecutionException on a
// shut-down executor. @Volatile so the connect/disconnect callbacks (which
// may run on a different thread than onNotificationPosted) publish safely.
@Volatile private var pushExecutor: ExecutorService? = null
// Guards executor creation so the lazy submit-site fallback and
// onListenerConnected can't race two executors into existence.
private val executorLock = Any()
// Tracks whether the listener is currently connected. ensureExecutor() only
// CREATES a new executor while connected — otherwise a notification racing
// onListenerDisconnected (which nulls pushExecutor) would spin up a fresh
// executor that nothing reaps until the next disconnect cycle (a thread leak).
@Volatile private var connected: Boolean = false
// packageName -> resolved human-readable label. Matches the app_name the
// Windows/Linux backends pass, so per-app colors/filters keep working.
// Naturally bounded by the number of notification-posting apps (tens) and
// cleared with the process — no eviction needed.
private val labelCache = ConcurrentHashMap<String, String>()
override fun onNotificationPosted(sbn: StatusBarNotification?) {
val notification = sbn ?: return
// The Python server (and thus the listener) only exists during a capture
// session. isRunning is a coarse early-out — the authoritative gate is the
// Python receiver's None-check — but it avoids needless JNI churn here.
if (!CaptureService.isRunning) return
// Filter notifications that should never drive an effect:
// - ongoing (media transport, downloads): not user-facing "alerts"
// - group summaries: duplicate their child notifications
// - our own foreground-service notification: would self-trigger
if (notification.isOngoing) return
if ((notification.notification.flags and Notification.FLAG_GROUP_SUMMARY) != 0) return
if (notification.packageName == packageName) return
val label = resolveAppLabel(notification.packageName)
// Obtain (creating if needed) the executor. onListenerConnected normally
// creates it, but that callback is not reliably invoked on every
// OEM/version (re)bind, and a notification can arrive before it fires —
// lazily creating here keeps a missing/late onListenerConnected from
// permanently disabling notification forwarding. A late submit onto an
// executor that onListenerDisconnected is shutting down throws
// RejectedExecutionException — guard with runCatching so a notification
// racing teardown can never crash this system-bound service.
val executor = ensureExecutor() ?: run {
Log.d(TAG, "no executor (listener disconnected) — skipping push")
return
}
runCatching {
executor.execute {
try {
Python.getInstance()
.getModule(PY_MODULE)
.callAttr("push_notification", label)
} catch (t: Throwable) {
// Never crash a system-bound service. Python.getInstance() throws
// IllegalStateException if Python.start() hasn't run (e.g. the
// service was bound at boot before the app process initialized).
// Log at debug — the label is potentially sensitive on a shared TV.
Log.d(TAG, "push_notification failed: ${t.message}")
}
}
}.onFailure { e ->
if (e is RejectedExecutionException) {
Log.d(TAG, "push rejected — listener disconnecting")
} else {
throw e
}
}
}
/** Resolve (and cache) a package's human-readable label; fall back to the package name. */
private fun resolveAppLabel(pkg: String): String {
labelCache[pkg]?.let { return it }
// Only cache SUCCESSFUL resolutions. Caching the package-name fallback
// would permanently pin a wrong label if the PackageManager lookup
// failed transiently (e.g. the app was mid-install / still updating).
val resolved = runCatching {
val info = packageManager.getApplicationInfo(pkg, 0)
packageManager.getApplicationLabel(info).toString()
}.getOrNull()
if (resolved != null) {
labelCache[pkg] = resolved
return resolved
}
return pkg
}
/**
* Return the push executor, creating it under [executorLock] if absent AND
* the listener is connected. Returns null when disconnected so a notification
* racing teardown neither submits onto a shutting-down executor nor spins up
* a stray one. Safe against a concurrent onListenerConnected/onNotificationPosted
* race (single executor) and against a missing onListenerConnected callback.
*/
private fun ensureExecutor(): ExecutorService? {
pushExecutor?.let { return it }
synchronized(executorLock) {
if (!connected) return null
return pushExecutor ?: Executors.newSingleThreadExecutor().also { pushExecutor = it }
}
}
override fun onListenerConnected() {
Log.i(TAG, "Notification listener connected")
// Spin up the push executor on connect. The system can disconnect and
// later reconnect this service without destroying it, so own the
// executor here rather than in onCreate/onDestroy. onNotificationPosted
// also lazily creates it (via ensureExecutor) in case this callback is
// late or skipped on some ROMs.
connected = true
ensureExecutor()
}
override fun onListenerDisconnected() {
Log.i(TAG, "Notification listener disconnected")
// Mark disconnected BEFORE nulling the executor so a racing ensureExecutor
// sees !connected and skips creating a replacement. Tear the executor
// down; a fresh one is created on the next onListenerConnected.
connected = false
pushExecutor?.let { exec ->
pushExecutor = null
exec.shutdown()
}
}
override fun onDestroy() {
// Defensive: onListenerDisconnected normally clears this first, but
// shut down here too in case onDestroy fires without a prior disconnect.
connected = false
pushExecutor?.shutdown()
pushExecutor = null
super.onDestroy()
}
companion object {
private const val TAG = "LedGrabNotifListener"
private const val PY_MODULE = "ledgrab.core.processing.os_notification_listener"
}
}
@@ -0,0 +1,653 @@
package com.ledgrab.android
import android.Manifest
import android.animation.ObjectAnimator
import android.animation.ValueAnimator
import android.annotation.SuppressLint
import android.app.Activity
import android.content.Intent
import android.content.pm.PackageManager
import android.graphics.Bitmap
import android.media.projection.MediaProjectionManager
import android.net.Uri
import android.os.Build
import android.os.Bundle
import android.os.PowerManager
import android.provider.Settings
import android.util.Log
import android.view.View
import android.view.ViewStub
import android.view.animation.AccelerateDecelerateInterpolator
import android.widget.Button
import android.widget.CheckBox
import android.widget.ImageView
import android.widget.LinearLayout
import android.widget.ScrollView
import android.widget.TextView
import androidx.core.app.NotificationManagerCompat
import androidx.core.content.ContextCompat
import androidx.core.splashscreen.SplashScreen.Companion.installSplashScreen
import com.google.zxing.BarcodeFormat
import com.google.zxing.qrcode.QRCodeWriter
import kotlinx.coroutines.CoroutineScope
import kotlinx.coroutines.Dispatchers
import kotlinx.coroutines.MainScope
import kotlinx.coroutines.cancel
import kotlinx.coroutines.launch
import kotlinx.coroutines.runInterruptible
import kotlinx.coroutines.withContext
/**
* Main (and only) Activity for the Android TV app.
*
* Two-state UI: stopped (Start button) and running (URL + QR + Stop).
* Navigable with D-pad / USB controller.
*/
class MainActivity : Activity() {
// Activity-scoped coroutine scope. We don't depend on AppCompat /
// androidx.lifecycle's lifecycleScope here because the TV launcher
// theme inherits from Leanback (non-AppCompat).
private val uiScope: CoroutineScope = MainScope()
companion object {
private const val TAG = "MainActivity"
private const val SERVER_PORT = 8080
private const val REQUEST_MEDIA_PROJECTION = 1001
private const val REQUEST_POST_NOTIFICATIONS = 1002
private const val REQUEST_RECORD_AUDIO = 1003
private const val REQUEST_CAMERA = 1004
private const val REQUEST_BLUETOOTH = 1005
private const val QR_SIZE_PX = 560
private const val NOTIF_PREFS = "ledgrab_notif"
private const val KEY_NOTIF_ACCESS_PROMPTED = "notif_access_prompted"
}
// Stopped-state views (always inflated).
private lateinit var stoppedPanel: View
private lateinit var statusText: TextView
private lateinit var toggleButton: Button
private lateinit var versionText: TextView
private lateinit var autostartCheck: CheckBox
private lateinit var autostartPrefs: AutostartPrefs
private lateinit var grantNotificationButton: Button
private lateinit var grantUsageAccessButton: Button
// Running-state views (lazy-inflated via ViewStub).
private lateinit var runningPanelStub: ViewStub
private var runningPanel: View? = null
private var urlText: TextView? = null
private var qrImage: ImageView? = null
private var stopButtonRunning: Button? = null
private var statusDot: View? = null
private var statusDotAnimator: ObjectAnimator? = null
// Cache of the most recently rendered QR (and the URL it encodes).
// updateUI() runs on every onResume (HDMI-CEC wakes, app switches,
// overlay dismissal, etc.). Rebuilding the 560×560 bitmap each time
// is wasteful — usually the IP and key are unchanged. Cache and
// short-circuit when the URL matches.
private var cachedQrUrl: String? = null
private var cachedQrBitmap: Bitmap? = null
override fun onCreate(savedInstanceState: Bundle?) {
// Install the splash screen BEFORE super.onCreate so the system
// keeps it on screen until our first frame is ready. This hides
// the Chaquopy stdlib unpack delay on cold first launch.
val splashScreen = installSplashScreen()
super.onCreate(savedInstanceState)
// Surface fatal Python init errors instead of crashing.
val initError = (application as? LedGrabApp)?.initError
if (initError != null) {
// Tell the splash screen to dismiss immediately — we're
// about to render an error screen, not the main UI.
splashScreen.setKeepOnScreenCondition { false }
showFatalErrorScreen(initError)
return
}
setContentView(R.layout.activity_main)
stoppedPanel = findViewById(R.id.stopped_panel)
runningPanelStub = findViewById(R.id.running_panel_stub)
statusText = findViewById(R.id.status_text)
toggleButton = findViewById(R.id.toggle_button)
versionText = findViewById(R.id.version_text)
autostartCheck = findViewById(R.id.autostart_check)
grantNotificationButton = findViewById(R.id.grant_notification_button)
grantUsageAccessButton = findViewById(R.id.grant_usage_access_button)
val versionName = packageManager.getPackageInfo(packageName, 0).versionName
versionText.text = getString(R.string.version_prefix, versionName ?: "?")
autostartPrefs = AutostartPrefs(this)
// Autostart only takes effect on rooted devices. Hide the
// checkbox entirely on unrooted hardware instead of showing a
// disabled-but-visible control, which reads as broken UI from
// across the room.
if (Root.looksRooted()) {
autostartCheck.visibility = View.VISIBLE
autostartCheck.isChecked = autostartPrefs.isEnabled
autostartCheck.setOnCheckedChangeListener { _, isChecked ->
autostartPrefs.isEnabled = isChecked
if (isChecked) ensureIgnoringBatteryOptimizations()
}
} else {
autostartCheck.visibility = View.GONE
}
grantNotificationButton.setOnClickListener { openNotificationListenerSettings() }
grantUsageAccessButton.setOnClickListener { openUsageAccessSettings() }
toggleButton.setOnClickListener { startCapture() }
updateStoppedPermissionButtons()
updateUI()
}
override fun onDestroy() {
stopStatusDotPulse()
uiScope.cancel()
super.onDestroy()
}
override fun onStop() {
// ObjectAnimator retains a hard reference to the dot View. On
// backgrounded TV apps onDestroy may never fire, so cancel here
// to avoid leaking the entire view hierarchy through an
// INFINITE-repeat animator.
stopStatusDotPulse()
super.onStop()
}
override fun onResume() {
super.onResume()
if (!::stoppedPanel.isInitialized) return
// Restart the pulse if we returned to the foreground while the
// service is still running. The running panel's view may have been
// recreated; ensureRunningPanelInflated already keys off the field
// reference. When stopped, refresh the notification-access button —
// the user may have just granted/revoked access in Settings.
if (CaptureService.isRunning) {
updateUI()
} else {
updateStoppedPermissionButtons()
}
}
/**
* Decide whether to go through the MediaProjection consent flow or
* jump straight into root capture. Root check is fast but may block
* briefly the first time Magisk shows its grant dialog — running it
* on the UI thread is acceptable because we're responding to a
* button press and we want to block until the user answers.
*/
private fun startCapture() {
// `su -c id` can block for seconds while Magisk shows its grant
// dialog; running it on the Main thread caused ANRs. Render an
// explicit "starting" state so the button doesn't look frozen.
val originalText = toggleButton.text
toggleButton.isEnabled = false
toggleButton.text = getString(R.string.btn_starting)
statusText.text = getString(R.string.status_checking_root)
uiScope.launch(Dispatchers.IO) {
// runInterruptible so a config change (rotation) during the
// up-to-10s `su` probe cancels the coroutine AND interrupts the
// blocking probe thread — Root.requestGrant honours the interrupt,
// destroys the su child, and rethrows, so we don't leak the
// process + drain thread. Without this, IO-dispatcher cancellation
// would not interrupt the blocking waitFor().
val rooted = runInterruptible { Root.requestGrant() }
withContext(Dispatchers.Main) {
toggleButton.isEnabled = true
toggleButton.text = originalText
statusText.text = ""
if (rooted) {
Log.i(TAG, "Root available — skipping MediaProjection consent")
startRootCaptureService()
} else {
requestMediaProjection()
}
}
}
}
private fun requestMediaProjection() {
val manager = getSystemService(MEDIA_PROJECTION_SERVICE) as MediaProjectionManager
@Suppress("DEPRECATION")
startActivityForResult(manager.createScreenCaptureIntent(), REQUEST_MEDIA_PROJECTION)
}
private fun startRootCaptureService() {
ensureNotificationPermission()
ensureNotificationListenerAccess()
ensureCameraPermission()
ensureBluetoothPermissions()
ContextCompat.startForegroundService(this, CaptureService.createRootIntent(this))
updateUI()
}
@Deprecated("Using deprecated API for plain Activity compatibility")
override fun onActivityResult(requestCode: Int, resultCode: Int, data: Intent?) {
super.onActivityResult(requestCode, resultCode, data)
if (requestCode == REQUEST_MEDIA_PROJECTION) {
if (resultCode == RESULT_OK && data != null) {
startCaptureService(resultCode, data)
} else {
statusText.text = getString(R.string.status_permission_denied)
Log.w(TAG, "MediaProjection permission denied")
}
}
}
private fun startCaptureService(resultCode: Int, resultData: Intent) {
ensureNotificationPermission()
ensureNotificationListenerAccess()
ensureAudioPermission()
ensureCameraPermission()
ensureBluetoothPermissions()
val intent = CaptureService.createIntent(this, resultCode, resultData)
ContextCompat.startForegroundService(this, intent)
updateUI()
}
private fun stopCaptureService() {
stopService(Intent(this, CaptureService::class.java))
updateUI()
}
private fun ensureRunningPanelInflated(): View {
runningPanel?.let { return it }
val view = runningPanelStub.inflate()
urlText = view.findViewById(R.id.url_text)
qrImage = view.findViewById(R.id.qr_image)
stopButtonRunning = view.findViewById(R.id.stop_button_running)
statusDot = view.findViewById(R.id.status_dot)
stopButtonRunning?.setOnClickListener { stopCaptureService() }
runningPanel = view
return view
}
private fun updateUI() {
// Fatal-init-error path took over setContentView and the
// lateinit view fields are unassigned. Guard so any future
// caller (Resume, broadcast receiver, etc.) doesn't NPE.
if (!::stoppedPanel.isInitialized) return
if (CaptureService.isRunning) {
val running = ensureRunningPanelInflated()
val localIp = NetworkUtils.getLocalIpAddress(this)
if (localIp == null) {
// No network — show the no-network state inside the
// stopped panel and keep capture stopped. The service
// is alive (capture works on loopback) but the URL/QR
// are useless without a routable address.
statusText.text = getString(R.string.status_no_network)
stoppedPanel.visibility = View.VISIBLE
versionText.visibility = View.VISIBLE
running.visibility = View.GONE
toggleButton.requestFocus()
return
}
val displayUrl = "http://$localIp:$SERVER_PORT"
val qrUrl = qrUrlFor(displayUrl)
urlText?.text = displayUrl
val cachedForUrl = cachedQrBitmap?.takeIf { cachedQrUrl == qrUrl }
if (cachedForUrl != null) {
qrImage?.setImageBitmap(cachedForUrl)
} else {
qrImage?.setImageBitmap(null)
// Build the bitmap pixels off the Main thread — encode + 313k
// setPixel calls were noticeably janky on slow TV boxes.
uiScope.launch(Dispatchers.Default) {
val bitmap = generateQrCode(qrUrl)
withContext(Dispatchers.Main) {
if (CaptureService.isRunning && urlText?.text == displayUrl) {
cachedQrUrl = qrUrl
cachedQrBitmap = bitmap
qrImage?.setImageBitmap(bitmap)
}
}
}
}
stoppedPanel.visibility = View.GONE
versionText.visibility = View.GONE
running.visibility = View.VISIBLE
stopButtonRunning?.requestFocus()
startStatusDotPulse()
} else {
stopStatusDotPulse()
urlText?.text = ""
qrImage?.setImageBitmap(null)
// Drop the cached bitmap so a Start → IP change → Start
// sequence rebuilds the QR for the new address.
cachedQrUrl = null
cachedQrBitmap = null
runningPanel?.visibility = View.GONE
stoppedPanel.visibility = View.VISIBLE
versionText.visibility = View.VISIBLE
toggleButton.requestFocus()
}
}
/**
* Build the URL we encode into the QR. Embeds the API key as a
* URL fragment (``#k=<token>``) so:
* - The token never appears in HTTP requests (fragments aren't
* sent over the wire) — no access-log leak.
* - The frontend can read [location.hash] on first visit and
* persist the key to localStorage (see static/js/app.ts).
* - The visible URL chip stays short and human-readable.
*
* The chip text in [updateUI] intentionally uses the *base* URL
* (without the fragment) so a human reading the URL out loud
* doesn't have to dictate 64 hex chars; only the QR carries the
* key. Do not collapse these into a single string — that would
* leak the key onto the screen.
*/
private fun qrUrlFor(base: String): String {
val key = (application as? LedGrabApp)?.apiKeyManager?.apiKey
return if (key.isNullOrBlank()) base else "$base/#k=$key"
}
private fun startStatusDotPulse() {
val dot = statusDot ?: return
if (statusDotAnimator?.isStarted == true) return
val animator = ObjectAnimator.ofFloat(dot, "alpha", 1f, 0.35f).apply {
duration = 900
repeatCount = ValueAnimator.INFINITE
repeatMode = ValueAnimator.REVERSE
interpolator = AccelerateDecelerateInterpolator()
}
animator.start()
statusDotAnimator = animator
}
private fun stopStatusDotPulse() {
statusDotAnimator?.cancel()
statusDotAnimator = null
statusDot?.alpha = 1f
}
private fun generateQrCode(text: String): Bitmap {
val size = QR_SIZE_PX
val bitMatrix = QRCodeWriter().encode(text, BarcodeFormat.QR_CODE, size, size)
// ALPHA_8 = 1 byte/px instead of 2 (RGB_565) or 4 (ARGB_8888).
// The ImageView gets tinted white via the matrix — for a pure
// black-and-white QR that's all we need and it halves heap usage
// compared to the previous RGB_565 path.
val bitmap = Bitmap.createBitmap(size, size, Bitmap.Config.ARGB_8888)
val pixels = IntArray(size * size)
for (y in 0 until size) {
val rowOffset = y * size
for (x in 0 until size) {
pixels[rowOffset + x] =
if (bitMatrix[x, y]) 0xFF000000.toInt() else 0xFFFFFFFF.toInt()
}
}
bitmap.setPixels(pixels, 0, size, 0, 0, size, size)
return bitmap
}
/**
* Minimal failure UI shown when Python.start() (Chaquopy) blew up.
* Stack trace is hidden behind a "Show details" toggle so we don't
* print user-path data on shared TV screens by default.
*/
private fun showFatalErrorScreen(error: Throwable) {
Log.e(TAG, "Fatal init error — showing error screen", error)
val stackText = Log.getStackTraceString(error)
val container = LinearLayout(this).apply {
orientation = LinearLayout.VERTICAL
setPadding(48, 48, 48, 48)
}
val title = TextView(this).apply {
text = getString(R.string.fatal_title)
textSize = 22f
}
val description = TextView(this).apply {
text = getString(R.string.fatal_body_prefix)
textSize = 14f
setPadding(0, 24, 0, 12)
}
val body = TextView(this).apply {
text = stackText
textSize = 12f
setTextIsSelectable(true)
visibility = View.GONE
}
val scroll = ScrollView(this).apply {
addView(body)
visibility = View.GONE
}
val toggleBtn = Button(this).apply {
text = getString(R.string.fatal_show_details)
setOnClickListener {
val showing = scroll.visibility == View.VISIBLE
scroll.visibility = if (showing) View.GONE else View.VISIBLE
body.visibility = scroll.visibility
text = getString(
if (showing) R.string.fatal_show_details else R.string.fatal_hide_details,
)
}
}
val copyBtn = Button(this).apply {
text = getString(R.string.fatal_copy_log)
setOnClickListener {
val cm = getSystemService(CLIPBOARD_SERVICE)
as android.content.ClipboardManager
cm.setPrimaryClip(
android.content.ClipData.newPlainText("LedGrab error", stackText)
)
}
}
container.addView(title)
container.addView(description)
container.addView(toggleBtn)
container.addView(copyBtn)
container.addView(scroll)
setContentView(container)
}
/**
* Prompt the user to exempt LedGrab from battery optimization.
* Strictly a phone-side concern (Doze/App Standby kill the FG
* service after hours of sleep); essentially a no-op on TV boxes.
* Only asked when autostart is turned on, which is itself only
* available on rooted devices.
*
* Play Store flags REQUEST_IGNORE_BATTERY_OPTIMIZATIONS by default
* — LedGrab's ambient-capture use case falls under the documented
* acceptable-use exceptions (a foreground service that must not be
* killed to fulfill its primary function).
*/
@SuppressLint("BatteryLife")
private fun ensureIgnoringBatteryOptimizations() {
if (Build.VERSION.SDK_INT < Build.VERSION_CODES.M) return
val pm = getSystemService(POWER_SERVICE) as PowerManager
if (pm.isIgnoringBatteryOptimizations(packageName)) return
try {
val intent = Intent(Settings.ACTION_REQUEST_IGNORE_BATTERY_OPTIMIZATIONS).apply {
data = Uri.parse("package:$packageName")
}
startActivity(intent)
} catch (e: Exception) {
// Some TV-box OEM builds strip this intent. Fall back to the
// generic settings screen so the user can find it manually.
Log.w(TAG, "Direct exemption intent unavailable: ${e.message}")
runCatching {
startActivity(Intent(Settings.ACTION_IGNORE_BATTERY_OPTIMIZATION_SETTINGS))
}
}
}
/**
* Request POST_NOTIFICATIONS permission on Android 13+ so the
* foreground service notification is visible. On older API levels
* this is a no-op.
*/
private fun ensureNotificationPermission() {
if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.TIRAMISU) {
if (checkSelfPermission(Manifest.permission.POST_NOTIFICATIONS)
!= PackageManager.PERMISSION_GRANTED
) {
@Suppress("DEPRECATION")
requestPermissions(
arrayOf(Manifest.permission.POST_NOTIFICATIONS),
REQUEST_POST_NOTIFICATIONS,
)
}
}
}
/**
* Request RECORD_AUDIO (API 29+) so the capture service can capture
* system playback audio for audio-reactive lighting. Fire-and-forget,
* like [ensureNotificationPermission]: capture still works without it
* (just no audio), so we don't block on the result. If first granted
* here, audio becomes available on the next Start.
*/
private fun ensureAudioPermission() {
if (Build.VERSION.SDK_INT < Build.VERSION_CODES.Q) return
if (checkSelfPermission(Manifest.permission.RECORD_AUDIO)
!= PackageManager.PERMISSION_GRANTED
) {
@Suppress("DEPRECATION")
requestPermissions(
arrayOf(Manifest.permission.RECORD_AUDIO),
REQUEST_RECORD_AUDIO,
)
}
}
/**
* Request CAMERA so the capture service can open the device camera for
* on-device webcam capture. Fire-and-forget, like [ensureAudioPermission]:
* capture still works without it (just no camera engine), so we don't block
* on the result. Gated on actual camera hardware via FEATURE_CAMERA_ANY so
* camera-less TV boxes (the common case) never see the prompt. The camera
* is opened on demand only while a camera source is active — granting this
* does not keep the camera on. If first granted here, the camera engine
* becomes available on the next Start.
*/
private fun ensureCameraPermission() {
if (!packageManager.hasSystemFeature(PackageManager.FEATURE_CAMERA_ANY)) return
if (checkSelfPermission(Manifest.permission.CAMERA)
!= PackageManager.PERMISSION_GRANTED
) {
@Suppress("DEPRECATION")
requestPermissions(
arrayOf(Manifest.permission.CAMERA),
REQUEST_CAMERA,
)
}
}
/**
* Request BLUETOOTH_SCAN + BLUETOOTH_CONNECT (API 31+) so the embedded
* server can discover and drive BLE LED controllers (SP110E / Triones /
* Zengge). On API < 31 these are install-time legacy permissions
* (BLUETOOTH / BLUETOOTH_ADMIN / ACCESS_FINE_LOCATION, maxSdk=30) and
* need no runtime grant — so this is a no-op there. Fire-and-forget,
* like [ensureAudioPermission]: screen capture works without BLE, and
* BleBridge degrades gracefully (empty scan / failed connect) when the
* grant is denied, so we don't block on the result. If first granted
* here, BLE devices become reachable on the next scan/connect.
*/
private fun ensureBluetoothPermissions() {
if (Build.VERSION.SDK_INT < Build.VERSION_CODES.S) return
val needed = listOf(
Manifest.permission.BLUETOOTH_SCAN,
Manifest.permission.BLUETOOTH_CONNECT,
).filter {
checkSelfPermission(it) != PackageManager.PERMISSION_GRANTED
}
if (needed.isEmpty()) return
@Suppress("DEPRECATION")
requestPermissions(needed.toTypedArray(), REQUEST_BLUETOOTH)
}
/** Whether the user has granted notification-listener access to this app. */
private fun isNotificationAccessGranted(): Boolean =
NotificationManagerCompat.getEnabledListenerPackages(this).contains(packageName)
/** Open the system Notification-access screen (manual affordance / re-grant). */
private fun openNotificationListenerSettings() {
runCatching {
startActivity(Intent(Settings.ACTION_NOTIFICATION_LISTENER_SETTINGS))
}.onFailure { Log.w(TAG, "Notification-access settings unavailable: ${it.message}") }
}
/**
* Whether Usage Access (PACKAGE_USAGE_STATS) is granted — needed by the
* foreground-app automation rule. Delegates to the bridge's AppOps check.
*/
private fun isUsageAccessGranted(): Boolean = ForegroundAppBridge.hasUsageAccess()
/**
* Open the system Usage-Access screen so the user can grant LedGrab access
* for the foreground-app automation rule. Falls back to the generic Settings
* screen on TV-box OEM builds that strip the dedicated intent.
*/
private fun openUsageAccessSettings() {
runCatching {
startActivity(Intent(Settings.ACTION_USAGE_ACCESS_SETTINGS))
}.onFailure {
Log.w(TAG, "Usage-access settings unavailable: ${it.message}")
runCatching { startActivity(Intent(Settings.ACTION_SETTINGS)) }
}
}
/**
* Prompt-once-then-remember: the first time capture starts without
* notification-listener access, open the settings screen so the user can
* grant it — then never nag again (the manual "Grant notification access"
* button stays available). Fire-and-forget like [ensureNotificationPermission].
*/
private fun ensureNotificationListenerAccess() {
if (isNotificationAccessGranted()) return
val prefs = getSharedPreferences(NOTIF_PREFS, MODE_PRIVATE)
if (prefs.getBoolean(KEY_NOTIF_ACCESS_PROMPTED, false)) return
prefs.edit().putBoolean(KEY_NOTIF_ACCESS_PROMPTED, true).apply()
openNotificationListenerSettings()
}
/**
* Show each "Grant <permission> access" button only while that access is
* missing, then re-wire the D-pad focus chain. Called on create and on resume
* (access can change in Settings while we're backgrounded). The usage-access
* button is a passive affordance (no auto-prompt at capture start) — the
* primary guidance is the web-UI banner when an Android app rule needs it.
*/
private fun updateStoppedPermissionButtons() {
if (!::grantNotificationButton.isInitialized) return
grantNotificationButton.visibility =
if (isNotificationAccessGranted()) View.GONE else View.VISIBLE
grantUsageAccessButton.visibility =
if (isUsageAccessGranted()) View.GONE else View.VISIBLE
wireStoppedFocusChain()
}
/**
* Link the visible stopped-panel controls into a single up/down D-pad chain.
* The optional controls (the grant-access buttons and the root-only autostart
* checkbox) may be GONE, so the chain is computed from whatever is visible —
* a static nextFocus pointing at a GONE view would strand the focus on a TV
* remote.
*/
private fun wireStoppedFocusChain() {
val chain = listOfNotNull(
toggleButton,
grantNotificationButton.takeIf { it.visibility == View.VISIBLE },
grantUsageAccessButton.takeIf { it.visibility == View.VISIBLE },
autostartCheck.takeIf { it.visibility == View.VISIBLE },
)
chain.forEachIndexed { i, view ->
view.nextFocusUpId = (chain.getOrNull(i - 1) ?: view).id
view.nextFocusDownId = (chain.getOrNull(i + 1) ?: view).id
}
}
}
@@ -0,0 +1,70 @@
package com.ledgrab.android
import android.content.Context
import android.net.ConnectivityManager
import android.net.LinkProperties
import android.net.Network
import android.net.NetworkCapabilities
import java.net.Inet4Address
/**
* Network utilities for discovering the device's local IP address.
*/
object NetworkUtils {
/**
* Return the device's local IPv4 address, preferring (in order):
* - Ethernet (wired TV-box link)
* - Wi-Fi
* - any other transport
* - whatever the active network reports
*
* Returns ``null`` only when no IPv4 link addresses exist at all.
*
* Why not just ``activeNetwork``: on TV boxes with both Ethernet
* AND Wi-Fi connected, Android's active-network heuristic can
* pick Wi-Fi while the user's phone is on the Ethernet subnet —
* leading to a URL/QR that the phone can't reach.
*/
fun getLocalIpAddress(context: Context): String? {
val cm = context.getSystemService(Context.CONNECTIVITY_SERVICE) as ConnectivityManager
// TODO(AP-mode): On TV boxes acting as a Wi-Fi tether/hotspot,
// TRANSPORT_WIFI here will resolve to the AP-side interface
// (typically 192.168.43.x) which clients on the user's actual
// home LAN can't reach. Detecting AP mode requires the @SystemApi
// WifiManager.getWifiApState reflection trick — defer until a
// user reports needing it.
val networks = cm.allNetworks
if (networks.isEmpty()) return ipv4Of(cm, cm.activeNetwork ?: return null)
val ranked = networks
.mapNotNull { n ->
val caps = cm.getNetworkCapabilities(n) ?: return@mapNotNull null
val rank = when {
caps.hasTransport(NetworkCapabilities.TRANSPORT_ETHERNET) -> 0
caps.hasTransport(NetworkCapabilities.TRANSPORT_WIFI) -> 1
caps.hasTransport(NetworkCapabilities.TRANSPORT_VPN) -> 3
caps.hasTransport(NetworkCapabilities.TRANSPORT_CELLULAR) -> 4
else -> 2
}
Triple(rank, n, caps)
}
.sortedBy { it.first }
for ((_, network, _) in ranked) {
val ip = ipv4Of(cm, network)
if (ip != null) return ip
}
return null
}
private fun ipv4Of(cm: ConnectivityManager, network: Network): String? {
val props: LinkProperties = cm.getLinkProperties(network) ?: return null
return props.linkAddresses
.asSequence()
.map { it.address }
.filterIsInstance<Inet4Address>()
.firstOrNull { !it.isLoopbackAddress && !it.isLinkLocalAddress }
?.hostAddress
}
}
@@ -0,0 +1,196 @@
package com.ledgrab.android
import android.content.Context
import android.util.Log
import com.chaquo.python.PyObject
import com.chaquo.python.Python
/**
* Bridge between Kotlin and the LedGrab Python server.
*
* All Python calls go through Chaquopy's `Python.getInstance()`.
* Frame data crosses the JNI boundary as a `ByteArray` (reused across
* frames — see ScreenCapture / RootScreenrecord for buffer pools).
*/
class PythonBridge(private val context: Context) {
companion object {
private const val TAG = "PythonBridge"
}
private var serverThread: Thread? = null
@Volatile private var running = false
// Cached PyObject handles for the per-frame fast path. Looking these
// up via Python.getInstance().getModule(...) every frame was a real
// measurable cost (~1ms/frame on TV boxes). Cached once at configure
// time and read on the capture thread — @Volatile is enough for the
// single-writer/single-reader pattern we have here.
@Volatile private var mediaProjectionEngine: PyObject? = null
@Volatile private var rootEngine: PyObject? = null
@Volatile private var androidAudioEngine: PyObject? = null
/**
* Configure the MediaProjection engine with screen dimensions.
* Must be called before [startServer].
*/
fun configureCapture(width: Int, height: Int) {
val py = Python.getInstance()
val engine = py.getModule("ledgrab.core.capture_engines.mediaprojection_engine")
engine.callAttr("configure", width, height)
mediaProjectionEngine = engine
Log.i(TAG, "MediaProjection engine configured: ${width}x${height}")
}
/**
* Configure the root screenrecord engine with screen dimensions.
* Used instead of [configureCapture] when root is available.
*/
fun configureRootCapture(width: Int, height: Int) {
val py = Python.getInstance()
val engine = py.getModule("ledgrab.core.capture_engines.root_screenrecord_engine")
engine.callAttr("configure", width, height)
rootEngine = engine
Log.i(TAG, "Root screenrecord engine configured: ${width}x${height}")
}
/**
* Configure the Android playback-capture audio engine with the format
* actually negotiated by [AudioCapture]'s `AudioRecord`. Must be called
* before [pushAudio]. Caches the module handle for the per-block fast
* path (same pattern as [configureCapture]).
*/
fun configureAudio(sampleRate: Int, channels: Int, chunkFrames: Int) {
val py = Python.getInstance()
val engine = py.getModule("ledgrab.core.audio.android_audio_engine")
engine.callAttr("configure", sampleRate, channels, chunkFrames)
androidAudioEngine = engine
Log.i(TAG, "Android audio engine configured: sr=$sampleRate ch=$channels chunk=$chunkFrames")
}
/**
* Push one interleaved little-endian float32 PCM block to the Python
* audio engine. Called from [AudioCapture]'s capture thread. The byte
* array crosses the JNI boundary; Python copies it on receipt, so the
* caller may reuse the same buffer for the next block.
*/
fun pushAudio(pcmFloat32: ByteArray) {
if (!running) return
val engine = androidAudioEngine ?: return
try {
engine.callAttr("push_samples", pcmFloat32)
} catch (e: Exception) {
Log.w(TAG, "Failed to push audio: ${e.message}")
}
}
/**
* Deactivate the Python audio engine. Called from [AudioCapture.stop].
*/
fun shutdownAudio() {
val engine = androidAudioEngine ?: return
try {
engine.callAttr("shutdown")
} catch (e: Exception) {
Log.w(TAG, "Failed to shut down audio engine: ${e.message}")
}
androidAudioEngine = null
}
/**
* Start the LedGrab FastAPI server on a background thread.
*
* Passes [apiKey] through so the Python server's auth gate accepts
* Bearer-authenticated LAN requests; null disables auth (loopback
* only — see [ApiKeyManager]).
*
* This blocks until [stopServer] is called, so it runs in its own
* thread.
*/
fun startServer(port: Int = 8080, apiKey: String? = null) {
if (running) {
Log.w(TAG, "Server already running")
return
}
running = true
val dataDir = context.filesDir.absolutePath
serverThread = Thread({
try {
Log.i(TAG, "Starting Python server (dataDir=$dataDir, port=$port)")
val py = Python.getInstance()
val entry = py.getModule("ledgrab.android_entry")
if (apiKey != null) {
entry.callAttr("start_server", dataDir, port, apiKey)
} else {
entry.callAttr("start_server", dataDir, port)
}
} catch (e: Exception) {
Log.e(TAG, "Python server error", e)
} finally {
running = false
}
}, "ledgrab-python-server")
serverThread?.start()
}
/**
* Signal the Python server to shut down gracefully.
*/
fun stopServer() {
if (!running) return
try {
val py = Python.getInstance()
val entry = py.getModule("ledgrab.android_entry")
entry.callAttr("stop_server")
} catch (e: Exception) {
Log.e(TAG, "Error stopping server", e)
}
serverThread?.join(10_000)
serverThread = null
running = false
Log.i(TAG, "Server stopped")
}
/**
* Push a captured RGBA frame to the Python MediaProjection engine.
*
* Called from [ScreenCapture] on the capture thread. The byte array
* crosses the JNI boundary — keep frames small (downscale to 480p
* before calling) and pass reusable buffers (see ScreenCapture's
* buffer pool).
*/
fun pushFrame(rgbaBytes: ByteArray, width: Int, height: Int) {
if (!running) return
val engine = mediaProjectionEngine ?: return
try {
engine.callAttr("push_frame", rgbaBytes, width, height)
} catch (e: Exception) {
Log.w(TAG, "Failed to push frame: ${e.message}")
}
}
/**
* Push a frame produced by the root `screenrecord` pipeline.
*
* Routed to a separate Python module so the two capture paths stay
* independently introspectable (engine priority, availability flags,
* dashboard diagnostics).
*/
fun pushRootFrame(rgbaBytes: ByteArray, width: Int, height: Int) {
if (!running) return
val engine = rootEngine ?: return
try {
engine.callAttr("push_frame", rgbaBytes, width, height)
} catch (e: Exception) {
Log.w(TAG, "Failed to push root frame: ${e.message}")
}
}
val isRunning: Boolean get() = running
}
@@ -0,0 +1,219 @@
package com.ledgrab.android
import android.util.Log
import java.io.BufferedReader
import java.io.InputStreamReader
import java.util.concurrent.TimeUnit
/**
* Lightweight root-access utilities.
*
* Detection is two-phase:
* 1. Cheap check for a `su` binary in common paths (no process spawn).
* 2. On demand, a real `su -c id` execution that returns true only when
* the shell actually runs as UID 0. This triggers Magisk's grant
* dialog the first time, exactly like any other root request.
*
* The heavier check is cached after it succeeds so we don't ask Magisk
* repeatedly mid-session.
*/
object Root {
private const val TAG = "Root"
// Slice length for the cancellation-aware su probe wait loop. Short
// enough that coroutine cancellation is honoured promptly, long enough
// to avoid busy-spinning while Magisk's grant dialog is up.
private const val POLL_SLICE_MS = 100L
private val SU_PATHS = listOf(
"/system/bin/su",
"/system/xbin/su",
"/sbin/su",
"/su/bin/su",
"/debug_ramdisk/su",
"/vendor/bin/su",
)
@Volatile private var cachedGranted: Boolean? = null
/** Cheap probe: true if a known `su` binary exists. Doesn't run anything. */
@JvmStatic
fun looksRooted(): Boolean = SU_PATHS.any { java.io.File(it).exists() }
/**
* Actually exercise `su`. Triggers Magisk's grant dialog on first call.
* Returns true if the shell reports UID 0 within the timeout; false
* otherwise (no root, user denied, or timeout).
*/
@JvmStatic
fun requestGrant(timeoutSeconds: Long = 10): Boolean {
cachedGranted?.let { return it }
if (!looksRooted()) {
cachedGranted = false
return false
}
var process: Process? = null
val granted = try {
// redirectErrorStream merges stderr into stdout so a single
// drain thread is enough — avoids the classic pipe-buffer
// deadlock where waitFor() blocks because stderr filled up.
val proc = ProcessBuilder("su", "-c", "id")
.redirectErrorStream(true)
.start()
process = proc
val outputBuilder = StringBuilder()
val drain = Thread({
try {
BufferedReader(InputStreamReader(proc.inputStream)).use { r ->
val buf = CharArray(512)
while (true) {
val n = r.read(buf)
if (n < 0) break
synchronized(outputBuilder) { outputBuilder.append(buf, 0, n) }
}
}
} catch (_: Exception) {
// Process gone — drain ends.
}
}, "Root-su-drain").apply { isDaemon = true; start() }
// Cancellation-aware wait: callers run this on a coroutine
// (MainActivity wraps it in runInterruptible), so a config change
// mid-probe cancels the coroutine and interrupts this thread.
// Poll waitFor() in short slices and honour interruption so we
// don't leak the `su` child + its drain thread for up to 10s.
// The catch(InterruptedException) below destroys the process; we
// re-arm the interrupt and rethrow so coroutine cancellation
// propagates cleanly.
val deadlineNanos = System.nanoTime() + TimeUnit.SECONDS.toNanos(timeoutSeconds)
var finished = false
while (System.nanoTime() < deadlineNanos) {
if (proc.waitFor(POLL_SLICE_MS, TimeUnit.MILLISECONDS)) {
finished = true
break
}
// Throws InterruptedException if the thread was interrupted
// by coroutine cancellation — handled below to tear down.
if (Thread.interrupted()) throw InterruptedException("su probe cancelled")
}
if (!finished) {
proc.destroyForcibly()
drain.join(500)
Log.w(TAG, "su -c id timed out after ${timeoutSeconds}s")
false
} else {
drain.join(500)
val output = synchronized(outputBuilder) { outputBuilder.toString() }
if (proc.exitValue() != 0) {
Log.w(TAG, "su -c id exited with ${proc.exitValue()} output='${output.trim()}'")
false
} else {
val rooted = output.contains("uid=0")
Log.i(TAG, "su -c id → '${output.trim()}' → rooted=$rooted")
rooted
}
}
} catch (e: InterruptedException) {
// Coroutine cancelled mid-probe (e.g. config change). Kill the
// su child so it doesn't outlive the cancelled work, re-arm the
// interrupt flag, and rethrow so the coroutine cancels cleanly.
// Do NOT cache a result — the probe never completed.
runCatching { process?.destroyForcibly() }
Thread.currentThread().interrupt()
throw e
} catch (e: Exception) {
Log.w(TAG, "su invocation failed: ${e.message}")
runCatching { process?.destroyForcibly() }
false
}
cachedGranted = granted
return granted
}
/**
* Run a command as root.
*
* The [argv] array is passed to `su -c` as **a single string** built by
* shell-quoting each element. This prevents the shell-injection class
* of bug where a caller passes user-influenced data containing
* spaces, semicolons, or backticks: each element is treated as a
* single shell token regardless of contents.
*
* Returns true on exit-zero. Failure invalidates the cached grant so
* the next [requestGrant] re-checks (covers cases like Magisk grant
* being revoked mid-session).
*/
@JvmStatic
@JvmOverloads
fun runAsRoot(argv: Array<String>, timeoutSeconds: Long = 5): Boolean {
require(argv.isNotEmpty()) { "runAsRoot called with empty argv" }
val quoted = argv.joinToString(" ") { shellQuote(it) }
return execSu(quoted, timeoutSeconds)
}
/**
* Convenience for fully-trusted constant commands (e.g.
* ``runAsRoot("pkill -TERM screenrecord")``). DO NOT pass anything
* derived from user input through this overload — use [runAsRoot]
* with an argv array instead so each token is quoted individually.
*/
@JvmStatic
@JvmOverloads
fun runAsRoot(command: String, timeoutSeconds: Long = 5): Boolean {
return execSu(command, timeoutSeconds)
}
private fun execSu(shellLine: String, timeoutSeconds: Long): Boolean {
return try {
val process = ProcessBuilder("su", "-c", shellLine)
.redirectErrorStream(true)
.start()
val finished = process.waitFor(timeoutSeconds, TimeUnit.SECONDS)
if (!finished) {
process.destroyForcibly()
cachedGranted = null
false
} else if (process.exitValue() != 0) {
cachedGranted = null
false
} else {
true
}
} catch (e: Exception) {
Log.w(TAG, "runAsRoot('$shellLine') failed: ${e.message}")
cachedGranted = null
false
}
}
/**
* POSIX-shell-style single-quote escape. Wraps in single quotes and
* escapes embedded single quotes as ``'\''`` so shell metacharacters
* inside [s] are inert.
*/
private fun shellQuote(s: String): String {
if (s.isEmpty()) return "''"
// Optimisation: if the string contains only safe characters,
// skip the quoting overhead. The set is intentionally narrow —
// notably `=` is excluded because an unquoted "FOO=bar" at the
// start of a command would be parsed as a shell variable
// assignment, not a literal arg. Quoting it forces literal use.
if (s.all { it.isLetterOrDigit() || it in "_-./" }) return s
val sb = StringBuilder(s.length + 2)
sb.append('\'')
for (ch in s) {
if (ch == '\'') sb.append("'\\''") else sb.append(ch)
}
sb.append('\'')
return sb.toString()
}
/** Forget the cached grant result — useful if Magisk permission was revoked. */
@JvmStatic
fun invalidateCache() {
cachedGranted = null
}
}
@@ -0,0 +1,352 @@
package com.ledgrab.android
import android.graphics.PixelFormat
import android.media.ImageReader
import android.media.MediaCodec
import android.media.MediaFormat
import android.os.Handler
import android.os.HandlerThread
import android.util.Log
import java.io.InputStream
import java.util.concurrent.atomic.AtomicInteger
/**
* Root-only screen capture backed by `/system/bin/screenrecord`.
*
* When the device is rooted (Magisk), we spawn ``su -c screenrecord
* --output-format=h264 …`` and read the encoder's H.264 bitstream from
* stdout. A MediaCodec decoder is configured to render into an
* ImageReader's Surface, and the ImageReader's `onImageAvailable`
* callback feeds RGBA bytes back to the Python pipeline — same sink as
* the MediaProjection path, just a different source.
*
* Why bother: MediaProjection forces Android to draw a persistent
* capture indicator overlay (unavoidable on stock Android 14+). The
* root path sidesteps that entirely because `screenrecord` runs as
* system UID through `su`. It also skips the one-time MediaProjection
* consent dialog, which matters on TV-only setups where keyboard input
* is awkward.
*/
class RootScreenrecord(
private val bridge: PythonBridge,
private val width: Int = 480,
private val height: Int = 270,
private val bitRate: Int = 4_000_000,
private val fps: Int = 30,
) {
companion object {
private const val TAG = "RootScreenrecord"
private const val MIME_TYPE = MediaFormat.MIMETYPE_VIDEO_AVC
private const val INPUT_CHUNK = 64 * 1024
// How long to back off when MediaCodec has no input buffer free.
// 50 ms keeps the input pump from busy-spinning if the decoder
// is stalled (codec init, severe stall, etc.).
private const val NO_BUFFER_BACKOFF_MS = 5L
}
// Instance is single-use: stop() permanently disposes it. Callers
// wanting to restart the pipeline must construct a new instance —
// see CaptureService.restartRootPipeline().
@Volatile private var process: Process? = null
private var decoder: MediaCodec? = null
private var imageReader: ImageReader? = null
private var readerThread: HandlerThread? = null
private var inputThread: Thread? = null
private var outputThread: Thread? = null
@Volatile private var running = false
private val framesDeliveredCounter = AtomicInteger(0)
// disposed gates duplicate-stop calls only — not start() after
// stop() (which is unsupported, see note above). Set at the START
// of cleanup so a second concurrent stop() (rare under @Synchronized
// but possible if a future caller drops it) doesn't re-run runCatching
// blocks against already-released resources.
@Volatile private var disposed = false
// Guards process respawn vs. concurrent disposal. The input pump
// can spawn a fresh `su -c screenrecord` after EOF; without this
// lock, stop() could destroy the OLD process between spawn and
// assignment, leaving the new one orphaned (GPU encoder leak).
private val processLock = Any()
// Reusable RGBA buffer for ImageReader callbacks (single-threaded
// reader callback). See ScreenCapture for the rationale: avoids
// ~15 MB/s of per-frame garbage at 30 fps × 480×270×4 B.
private val frameBuffer: ByteArray = ByteArray(width * height * 4)
/** Monotonic count of frames pushed to the Python bridge. */
val framesDelivered: Int get() = framesDeliveredCounter.get()
/** True once at least one frame has reached the Python bridge. */
val hasProducedFrame: Boolean get() = framesDelivered > 0
/**
* Start the capture pipeline. Returns false if `su`/`screenrecord`
* couldn't be launched at all; true doesn't guarantee frames will
* actually flow (the caller should monitor [hasProducedFrame] and
* be ready to fall back to MediaProjection on timeout).
*/
fun start(): Boolean {
if (running) return true
running = true
try {
val reader = buildImageReader().also { imageReader = it }
val codec = buildDecoder(reader).also { decoder = it }
val proc = spawnScreenrecord() ?: run {
stop()
return false
}
process = proc
startInputPump(proc.inputStream, codec)
startOutputDrain(codec)
Log.i(TAG, "Root capture pipeline started (${width}x$height @ ${fps}fps)")
return true
} catch (e: Exception) {
Log.e(TAG, "Failed to start root capture", e)
stop()
return false
}
}
/** Stop everything and release resources. Idempotent. Single-use: do not call start() again. */
@Synchronized
fun stop() {
if (disposed) return
disposed = true
// Order matters: signal first so worker loops drop out, then
// stop the codec on the thread that created it (this one), then
// join workers BEFORE releasing the codec/ImageReader they may
// still be touching, then kill the external screenrecord process.
running = false
runCatching { decoder?.stop() }
inputThread?.interrupt()
outputThread?.interrupt()
runCatching { inputThread?.join(500) }
runCatching { outputThread?.join(500) }
inputThread = null
outputThread = null
// Best-effort: kill the screenrecord child before reaping `su`,
// otherwise screenrecord can outlive su as an orphan and keep
// the GPU encoder busy. Fire-and-forget; ignore failures.
runCatching {
Root.runAsRoot(arrayOf("pkill", "-TERM", "screenrecord"), timeoutSeconds = 2)
}
runCatching { decoder?.release() }
decoder = null
runCatching { imageReader?.setOnImageAvailableListener(null, null) }
runCatching { imageReader?.close() }
imageReader = null
readerThread?.quitSafely()
runCatching { readerThread?.join(500) }
readerThread = null
// Use the same lock as the respawn path so we don't destroy a
// not-yet-published process or leak one that was spawned after
// we already destroyed the old reference.
synchronized(processLock) {
runCatching { process?.destroy() }
process = null
}
Log.i(TAG, "Root capture pipeline stopped (frames delivered: ${framesDelivered})")
}
private fun buildImageReader(): ImageReader {
val thread = HandlerThread("RootReader").apply { start() }
readerThread = thread
val handler = Handler(thread.looper)
val reader = ImageReader.newInstance(width, height, PixelFormat.RGBA_8888, 3)
reader.setOnImageAvailableListener({ r ->
val image = r.acquireLatestImage() ?: return@setOnImageAvailableListener
try {
val plane = image.planes[0]
val buffer = plane.buffer
val rowStride = plane.rowStride
val pixelStride = plane.pixelStride
val rowBytes = width * pixelStride
val expected = rowBytes * height
if (rowStride == rowBytes && buffer.remaining() >= expected) {
buffer.get(frameBuffer, 0, expected)
} else {
for (row in 0 until height) {
buffer.position(row * rowStride)
buffer.get(frameBuffer, row * rowBytes, rowBytes)
}
}
// CONTRACT: frameBuffer is REUSED across frames (single-threaded
// reader callback — no copy here). Safety depends on the Python
// receiver copying the bytes before this callback returns and
// overwrites the buffer for the next frame. It does:
// PythonBridge.pushRootFrame → root_screenrecord_engine.push_frame
// (server/src/ledgrab/core/capture_engines/root_screenrecord_engine.py)
// does `rgba[:, :, :3].copy()`, so the queued frame owns its
// pixels independently of this buffer. Do NOT remove that copy.
bridge.pushRootFrame(frameBuffer, width, height)
framesDeliveredCounter.incrementAndGet()
} catch (e: Exception) {
Log.w(TAG, "Root frame delivery failed: ${e.message}")
} finally {
image.close()
}
}, handler)
return reader
}
private fun buildDecoder(reader: ImageReader): MediaCodec {
val format = MediaFormat.createVideoFormat(MIME_TYPE, width, height).apply {
setInteger(MediaFormat.KEY_FRAME_RATE, fps)
}
val codec = MediaCodec.createDecoderByType(MIME_TYPE)
codec.configure(format, reader.surface, null, 0)
codec.start()
return codec
}
private fun spawnScreenrecord(): Process? {
// argv form — passes safely through Root.runAsRoot's shell-quote
// logic so future changes to flag values can't introduce injection.
val args = arrayOf(
"screenrecord",
"--output-format=h264",
"--size=${width}x$height",
"--bit-rate=$bitRate",
// Time limit 0 isn't supported; the largest accepted is 180s.
// We restart the process ourselves if it exits early.
"--time-limit=180",
"-",
)
// Inline ProcessBuilder so we have direct access to the child's
// stdout (Root.runAsRoot returns Boolean). We still pass args
// unquoted because the entire array is a fixed program+flags
// with no user-controlled content.
return try {
ProcessBuilder("su", "-c", args.joinToString(" "))
.redirectErrorStream(false)
.start()
} catch (e: Exception) {
Log.e(TAG, "Failed to spawn `su -c screenrecord`: ${e.message}")
null
}
}
private fun startInputPump(initialStream: InputStream, codec: MediaCodec) {
inputThread = Thread({
val buf = ByteArray(INPUT_CHUNK)
var stream: InputStream = initialStream
try {
outer@ while (running) {
val n = try {
stream.read(buf)
} catch (e: Exception) {
if (!running) break
Log.w(TAG, "screenrecord read error: ${e.message}")
-1
}
if (n <= 0) {
if (!running) break
// screenrecord caps at --time-limit=180s. When it
// exits cleanly we respawn so capture survives
// long sessions instead of freezing after ~3min.
Log.i(TAG, "screenrecord EOF — respawning")
synchronized(processLock) {
runCatching { process?.destroy() }
process = null
}
val next = spawnScreenrecord()
if (next == null) {
// Avoid a tight loop if `su` is suddenly unhappy.
try { Thread.sleep(500) } catch (_: InterruptedException) { break }
continue@outer
}
// Publish the new process under the lock so a
// concurrent stop() either (a) sees no process,
// tears down later, and lets us assign it for
// the destroy on the NEXT stop call — or (b) sees
// !running and we destroy the new process ourselves.
val accepted = synchronized(processLock) {
if (!running) {
false
} else {
process = next
true
}
}
if (!accepted) {
// running flipped false between EOF and now —
// someone called stop(). Drop the new process
// on the floor; the codec and output thread
// are stop()'s responsibility (it's the only
// writer to `running`, so we don't need to
// tear them down here).
runCatching { next.destroy() }
break@outer
}
stream = next.inputStream
continue@outer
}
var offset = 0
while (offset < n && running) {
val index = codec.dequeueInputBuffer(50_000)
if (index < 0) {
// Codec is starved — back off briefly instead
// of spinning. Without this, a stalled codec
// burns 100% of one core hammering dequeue.
try {
Thread.sleep(NO_BUFFER_BACKOFF_MS)
} catch (_: InterruptedException) {
break
}
continue
}
val inputBuffer = codec.getInputBuffer(index) ?: continue
inputBuffer.clear()
val chunk = minOf(n - offset, inputBuffer.capacity())
inputBuffer.put(buf, offset, chunk)
codec.queueInputBuffer(
index,
0,
chunk,
System.nanoTime() / 1_000,
0,
)
offset += chunk
}
}
} catch (_: InterruptedException) {
// Expected on stop()
} catch (e: Exception) {
Log.w(TAG, "Input pump error: ${e.message}")
}
}, "RootDecoderInput").also { it.start() }
}
private fun startOutputDrain(codec: MediaCodec) {
outputThread = Thread({
val info = MediaCodec.BufferInfo()
try {
while (running) {
val index = codec.dequeueOutputBuffer(info, 50_000)
if (index >= 0) {
// `true` = render to the configured surface (ImageReader),
// which triggers onImageAvailable on the reader's handler.
codec.releaseOutputBuffer(index, true)
if (info.flags and MediaCodec.BUFFER_FLAG_END_OF_STREAM != 0) {
Log.i(TAG, "Decoder reported EOS")
break
}
}
}
} catch (_: InterruptedException) {
// Expected on stop()
} catch (e: Exception) {
Log.w(TAG, "Output drain error: ${e.message}")
}
}, "RootDecoderOutput").also { it.start() }
}
}
@@ -0,0 +1,211 @@
package com.ledgrab.android
import android.graphics.PixelFormat
import android.hardware.display.DisplayManager
import android.hardware.display.VirtualDisplay
import android.media.ImageReader
import android.media.projection.MediaProjection
import android.os.Handler
import android.os.HandlerThread
import android.os.SystemClock
import android.util.DisplayMetrics
import android.util.Log
/**
* Captures the Android screen via MediaProjection and feeds frames
* to [PythonBridge].
*
* Frames are downscaled to roughly [targetWidth] x [targetHeight] before
* crossing the JNI boundary to minimize overhead. The actual capture
* dimensions preserve the source screen's aspect ratio (snapped to even
* pixels for codec friendliness) so non-16:9 displays don't get
* squashed.
*/
class ScreenCapture(
private val projection: MediaProjection,
private val metrics: DisplayMetrics,
private val bridge: PythonBridge,
targetWidth: Int = 480,
targetHeight: Int = 270,
private val targetFps: Int = 30,
private val onProjectionStopped: () -> Unit = {},
) {
companion object {
private const val TAG = "ScreenCapture"
private const val VIRTUAL_DISPLAY_NAME = "LedGrabCapture"
}
// Snap to the source aspect ratio so we don't squash 21:9 / portrait
// / rotated screens. Width is the budget; height follows.
private val captureWidth: Int
private val captureHeight: Int
init {
val srcW = metrics.widthPixels.coerceAtLeast(1).toFloat()
val srcH = metrics.heightPixels.coerceAtLeast(1).toFloat()
val budget = targetWidth.coerceAtLeast(16)
val aspect = srcW / srcH
val w = budget
val h = (w / aspect).toInt().coerceAtLeast(16)
// Bias toward even dimensions — some encoders/ImageReaders are
// unhappy with odd sizes when row strides come into play.
captureWidth = (w and 1.inv()).coerceAtLeast(16)
captureHeight = (h and 1.inv()).coerceAtLeast(16)
if (captureWidth != targetWidth || captureHeight != targetHeight) {
Log.i(
TAG,
"Capture size adjusted for ${srcW.toInt()}x${srcH.toInt()} " +
"(${"%.2f".format(aspect)}:1) → ${captureWidth}x$captureHeight",
)
}
}
private var virtualDisplay: VirtualDisplay? = null
private var imageReader: ImageReader? = null
private var captureThread: HandlerThread? = null
private var captureHandler: Handler? = null
@Volatile private var running = false
// Reusable RGBA frame buffer — sized once for the capture dimensions.
// The capture handler is single-threaded so no synchronisation is
// required around this buffer (each callback runs to completion
// before the next is dispatched). Eliminates ~15 MB/s of per-frame
// garbage at 30 fps × 480×270×4 B that previously caused GC pauses
// on low-end TV boxes.
private val frameBuffer: ByteArray = ByteArray(captureWidth * captureHeight * 4)
// Monotonic frame pacing. `nextFrameNanos` is the target render
// time of the next frame; carrying it forward as an accumulator
// avoids the integer-division drift the wall-clock version had
// (e.g. 30 fps → 33 ms produced ~30.3 fps).
private val frameIntervalNanos = (1_000_000_000L / targetFps.coerceAtLeast(1))
private var nextFrameNanos = 0L
/**
* Start capturing the screen.
*/
fun start() {
if (running) return
running = true
nextFrameNanos = SystemClock.elapsedRealtimeNanos()
captureThread = HandlerThread("LedGrab-Capture").also { it.start() }
captureHandler = Handler(captureThread!!.looper)
// Android 14+ requires registering a callback before createVirtualDisplay
projection.registerCallback(object : MediaProjection.Callback() {
override fun onStop() {
Log.i(TAG, "MediaProjection stopped (external)")
// We're on captureHandler's thread here — calling stop()
// directly would self-join captureThread (handler.join()
// from inside the handler thread hangs until the join
// timeout, then closes resources while we're STILL
// inside this callback). Just flip `running` to halt
// frame processing and hand off to the service; its
// onDestroy will call stop() from the main thread,
// which is safe to join captureThread from.
running = false
onProjectionStopped()
}
}, captureHandler)
imageReader = ImageReader.newInstance(
captureWidth,
captureHeight,
PixelFormat.RGBA_8888,
3, // maxImages — small ring buffer; 3 is more forgiving than 2 under jitter
)
imageReader?.setOnImageAvailableListener({ reader ->
if (!running) return@setOnImageAvailableListener
val now = SystemClock.elapsedRealtimeNanos()
if (now < nextFrameNanos) {
// Too early — drop this image to stay on cadence.
reader.acquireLatestImage()?.close()
return@setOnImageAvailableListener
}
val image = reader.acquireLatestImage() ?: return@setOnImageAvailableListener
try {
val plane = image.planes[0]
val buffer = plane.buffer
val rowStride = plane.rowStride
val pixelStride = plane.pixelStride
val rowBytes = captureWidth * pixelStride
val expected = rowBytes * captureHeight
// Fill the reusable buffer. Two paths:
// - rowStride == rowBytes: bulk get into the buffer
// - rowStride > rowBytes: row-by-row copy stripping padding
if (rowStride == rowBytes && buffer.remaining() >= expected) {
buffer.get(frameBuffer, 0, expected)
} else {
for (row in 0 until captureHeight) {
buffer.position(row * rowStride)
buffer.get(frameBuffer, row * rowBytes, rowBytes)
}
}
// CONTRACT: frameBuffer is REUSED across frames (single-threaded
// capture handler — no copy here). Safety depends on the Python
// receiver copying the bytes before this callback returns and
// overwrites the buffer for the next frame. It does:
// PythonBridge.pushFrame → mediaprojection_engine.push_frame
// (server/src/ledgrab/core/capture_engines/mediaprojection_engine.py)
// does `rgba[:, :, :3].copy()`, so the queued frame owns its
// pixels independently of this buffer. Do NOT remove that copy.
bridge.pushFrame(frameBuffer, captureWidth, captureHeight)
// Advance the pacing accumulator. If we fell badly behind
// (long GC, JNI stall), snap forward to "now" instead of
// accumulating a burst of catch-up frames.
nextFrameNanos += frameIntervalNanos
if (now - nextFrameNanos > frameIntervalNanos * 4) {
nextFrameNanos = now + frameIntervalNanos
}
} catch (e: Exception) {
Log.w(TAG, "Frame processing error: ${e.message}")
} finally {
image.close()
}
}, captureHandler)
virtualDisplay = projection.createVirtualDisplay(
VIRTUAL_DISPLAY_NAME,
captureWidth,
captureHeight,
metrics.densityDpi,
DisplayManager.VIRTUAL_DISPLAY_FLAG_AUTO_MIRROR,
imageReader?.surface,
null,
captureHandler,
)
Log.i(TAG, "Screen capture started (${captureWidth}x${captureHeight} @ ${targetFps}fps)")
}
/**
* Stop capturing and release all resources.
*/
fun stop() {
running = false
// Order matters: detach the listener BEFORE releasing the
// VirtualDisplay so the handler can't be re-entered with stale
// resources, then quit & join the handler thread, only then
// close the ImageReader.
runCatching { imageReader?.setOnImageAvailableListener(null, null) }
runCatching { virtualDisplay?.release() }
virtualDisplay = null
captureThread?.quitSafely()
runCatching { captureThread?.join(500) }
captureThread = null
captureHandler = null
runCatching { imageReader?.close() }
imageReader = null
Log.i(TAG, "Screen capture stopped")
}
}
@@ -0,0 +1,307 @@
package com.ledgrab.android
import android.app.PendingIntent
import android.content.BroadcastReceiver
import android.content.Context
import android.content.Intent
import android.content.IntentFilter
import android.hardware.usb.UsbManager
import android.os.Build
import android.util.Log
import androidx.core.content.ContextCompat
import com.hoho.android.usbserial.driver.UsbSerialDriver
import com.hoho.android.usbserial.driver.UsbSerialPort
import com.hoho.android.usbserial.driver.UsbSerialProber
import java.util.concurrent.ConcurrentHashMap
import java.util.concurrent.atomic.AtomicBoolean
import java.util.concurrent.atomic.AtomicInteger
import kotlinx.coroutines.CompletableDeferred
import kotlinx.coroutines.runBlocking
import kotlinx.coroutines.withTimeout
import kotlinx.coroutines.TimeoutCancellationException
/**
* USB-serial bridge exposed to the Python server via Chaquopy.
*
* Uses the `usb-serial-for-android` library (mik3y) which ships drivers
* for the common USB-to-TTL chips (CH340, CP2102, FTDI, Prolific, and
* CDC-ACM) found on Arduino boards and Adalight/AmbiLED controllers.
*
* Python callers access the singleton instance via
* `UsbSerialBridge.INSTANCE.listDevices()` etc. — see
* `server/src/ledgrab/core/devices/android_serial_transport.py`.
*
* The bridge holds no Context of its own; [init] must be called once
* from [LedGrabApp.onCreate] to bind the application context.
*/
object UsbSerialBridge {
private const val TAG = "UsbSerialBridge"
private const val ACTION_USB_PERMISSION = "com.ledgrab.android.USB_PERMISSION"
@Volatile private var appContext: Context? = null
private val handleSeq = AtomicInteger(1)
private val openPorts = HashMap<Int, UsbSerialPort>()
private val initialized = AtomicBoolean(false)
private val pendingPermissions = ConcurrentHashMap<String, CompletableDeferred<Boolean>>()
/** Called once from [LedGrabApp.onCreate] so we can resolve services. */
@JvmStatic
fun init(context: Context) {
val app = context.applicationContext
appContext = app
// Idempotent: re-entrant init() must not double-register the
// receiver (which would leak listeners and double-fire callbacks).
if (!initialized.compareAndSet(false, true)) return
val filter = IntentFilter(ACTION_USB_PERMISSION)
val ourPackage = app.packageName
val receiver = object : BroadcastReceiver() {
override fun onReceive(ctx: Context, intent: Intent) {
// Defence-in-depth: the receiver is registered as
// RECEIVER_NOT_EXPORTED, but on pre-API-33 platforms
// older Android versions historically defaulted to
// exported. Also enforce the package check here so an
// explicit-intent attack from another app on the device
// is rejected even if the OS treats us as exported.
if (intent.`package` != null && intent.`package` != ourPackage) {
Log.w(
TAG,
"Ignoring USB permission broadcast from " +
"package='${intent.`package`}' (not us)",
)
return
}
val granted = intent.getBooleanExtra(
UsbManager.EXTRA_PERMISSION_GRANTED,
false,
)
val device = intent.getParcelableExtra<android.hardware.usb.UsbDevice>(
UsbManager.EXTRA_DEVICE,
)
Log.i(TAG, "USB permission broadcast: granted=$granted device=${device?.deviceName}")
device?.deviceName?.let { name ->
pendingPermissions.remove(name)?.complete(granted)
}
}
}
// ContextCompat handles the RECEIVER_NOT_EXPORTED flag correctly
// across all supported API levels (it's a no-op on platforms
// where the flag doesn't exist, and explicit on API ≥33 where
// Android enforces it).
ContextCompat.registerReceiver(
app,
receiver,
filter,
ContextCompat.RECEIVER_NOT_EXPORTED,
)
}
private fun ctx(): Context =
appContext ?: error("UsbSerialBridge.init() not called — app context unavailable")
private fun safeSerial(driver: UsbSerialDriver): String =
try {
driver.device.serialNumber ?: ""
} catch (_: SecurityException) {
// Reading the serial requires USB permission on API 29+.
""
}
/**
* Enumerate attached USB-serial devices.
*
* Each entry is `"VID|PID|serial|description"` with VID/PID as
* 4-char lowercase hex. Pipe is used as the separator so device
* descriptions containing colons (common on FTDI strings) don't
* confuse the Python parser.
*/
@JvmStatic
fun listDevices(): List<String> {
val manager = ctx().getSystemService(Context.USB_SERVICE) as UsbManager
val drivers = UsbSerialProber.getDefaultProber().findAllDrivers(manager)
return drivers.map { driver ->
val dev = driver.device
val vid = "%04x".format(dev.vendorId)
val pid = "%04x".format(dev.productId)
val serial = safeSerial(driver)
val description = buildString {
append(dev.manufacturerName ?: "USB")
val product = dev.productName
if (!product.isNullOrBlank()) {
append(' ')
append(product)
}
}.trim().ifEmpty { "USB $vid:$pid" }
"$vid|$pid|$serial|$description"
}
}
/**
* Open the first matching USB-serial device. Returns a non-negative
* opaque handle on success, -1 on failure (device not found, user
* denied permission, or driver error). Failures also trigger an
* async permission-request dialog when applicable — subsequent
* open() calls will succeed once the user grants.
*/
@JvmStatic
fun open(vendorId: Int, productId: Int, serial: String, baud: Int): Int {
val context = ctx()
val manager = context.getSystemService(Context.USB_SERVICE) as UsbManager
val drivers = UsbSerialProber.getDefaultProber().findAllDrivers(manager)
val driver = drivers.firstOrNull { d ->
val dev = d.device
dev.vendorId == vendorId &&
dev.productId == productId &&
(serial.isEmpty() || safeSerial(d) == serial)
}
if (driver == null) {
Log.w(TAG, "No matching device for $vendorId:$productId:$serial")
return -1
}
if (!manager.hasPermission(driver.device)) {
Log.w(TAG, "USB permission not yet granted for ${driver.device.deviceName}")
requestPermission(context, manager, driver)
return -1
}
val connection = manager.openDevice(driver.device)
if (connection == null) {
Log.w(TAG, "openDevice returned null for ${driver.device.deviceName}")
return -1
}
val port = driver.ports.firstOrNull()
if (port == null) {
connection.close()
Log.w(TAG, "Driver reports no ports for ${driver.device.deviceName}")
return -1
}
try {
port.open(connection)
port.setParameters(
baud,
8,
UsbSerialPort.STOPBITS_1,
UsbSerialPort.PARITY_NONE,
)
} catch (e: Exception) {
Log.e(TAG, "Failed to configure serial port", e)
runCatching { port.close() }
return -1
}
val handle = handleSeq.getAndIncrement()
synchronized(openPorts) { openPorts[handle] = port }
Log.i(
TAG,
"Opened USB serial ${driver.device.deviceName} baud=$baud handle=$handle",
)
return handle
}
/** Write bytes to the previously-opened handle. Throws if invalid. */
@JvmStatic
fun write(handle: Int, data: ByteArray) {
val port = synchronized(openPorts) { openPorts[handle] }
?: throw IllegalStateException("Invalid handle $handle")
// 1s write timeout matches the old pyserial `timeout=1` behavior.
port.write(data, 1_000)
}
/** Close a previously-opened handle. Silently ignores unknown handles. */
@JvmStatic
fun close(handle: Int) {
val port = synchronized(openPorts) { openPorts.remove(handle) } ?: return
runCatching { port.close() }
.onFailure { Log.w(TAG, "close($handle): ${it.message}") }
}
/**
* Block until the user grants (or denies) USB permission for the
* device with [deviceName] (e.g. "/dev/bus/usb/001/004"). Returns
* true if granted within [timeoutMs], false otherwise. Safe to call
* from a Python thread via Chaquopy.
*/
@JvmStatic
@JvmOverloads
fun requestPermissionBlocking(deviceName: String, timeoutMs: Long = 15_000L): Boolean {
val context = ctx()
val manager = context.getSystemService(Context.USB_SERVICE) as UsbManager
val driver = UsbSerialProber.getDefaultProber().findAllDrivers(manager)
.firstOrNull { it.device.deviceName == deviceName }
?: return false
if (manager.hasPermission(driver.device)) return true
// Coalesce concurrent requests for the same device — only the
// first caller actually fires the system dialog.
val deferred = pendingPermissions.computeIfAbsent(deviceName) {
CompletableDeferred<Boolean>().also {
requestPermission(context, manager, driver)
}
}
return try {
runBlocking {
withTimeout(timeoutMs) { deferred.await() }
}
} catch (_: TimeoutCancellationException) {
pendingPermissions.remove(deviceName)
Log.w(TAG, "Permission request timed out for $deviceName")
false
} catch (e: Exception) {
pendingPermissions.remove(deviceName)
Log.w(TAG, "Permission request failed for $deviceName: ${e.message}")
false
}
}
/**
* Like [open] but blocks for permission first. Use this from Python
* instead of relying on the open()/retry pattern.
*/
@JvmStatic
@JvmOverloads
fun openWithPermission(
vendorId: Int,
productId: Int,
serial: String,
baud: Int,
timeoutMs: Long = 15_000L,
): Int {
val context = ctx()
val manager = context.getSystemService(Context.USB_SERVICE) as UsbManager
val driver = UsbSerialProber.getDefaultProber().findAllDrivers(manager)
.firstOrNull { d ->
val dev = d.device
dev.vendorId == vendorId &&
dev.productId == productId &&
(serial.isEmpty() || safeSerial(d) == serial)
} ?: return -1
if (!manager.hasPermission(driver.device)) {
val granted = requestPermissionBlocking(driver.device.deviceName, timeoutMs)
if (!granted) return -1
}
return open(vendorId, productId, serial, baud)
}
private fun requestPermission(
context: Context,
manager: UsbManager,
driver: UsbSerialDriver,
) {
val flags =
if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.M) {
PendingIntent.FLAG_IMMUTABLE
} else {
0
}
val intent = Intent(ACTION_USB_PERMISSION).apply {
setPackage(context.packageName)
}
val pending = PendingIntent.getBroadcast(context, 0, intent, flags)
manager.requestPermission(driver.device, pending)
}
}
@@ -0,0 +1,5 @@
<?xml version="1.0" encoding="utf-8"?>
<selector xmlns:android="http://schemas.android.com/apk/res/android">
<item android:state_focused="true" android:color="#ffffff" />
<item android:color="@color/purple_accent" />
</selector>
@@ -0,0 +1,58 @@
<?xml version="1.0" encoding="utf-8"?>
<!-- Android TV launcher banner: 320x180 landscape.
Shown on the leanback home row. The previous build reused the square
launcher icon, which letterboxed badly. -->
<vector xmlns:android="http://schemas.android.com/apk/res/android"
android:width="320dp"
android:height="180dp"
android:viewportWidth="320"
android:viewportHeight="180">
<!-- Background -->
<path
android:fillColor="#0d1117"
android:pathData="M0,0 L320,0 L320,180 L0,180 Z" />
<!-- Subtle teal glow top-left -->
<path
android:fillColor="#1A64ffda"
android:pathData="M0,0 L160,0 L160,90 L0,90 Z" />
<!-- Subtle purple glow bottom-right -->
<path
android:fillColor="#15bb86fc"
android:pathData="M160,90 L320,90 L320,180 L160,180 Z" />
<!-- TV body, centered -->
<path
android:fillColor="#1c2333"
android:pathData="M88,56 L196,56 Q204,56 204,64 L204,116 Q204,124 196,124 L88,124 Q80,124 80,116 L80,64 Q80,56 88,56 Z" />
<!-- TV screen -->
<path
android:fillColor="#161b22"
android:pathData="M92,60 L192,60 Q196,60 196,64 L196,116 Q196,120 192,120 L92,120 Q88,120 88,116 L88,64 Q88,60 92,60 Z" />
<!-- LED glow strips -->
<path
android:fillColor="#64ffda"
android:fillAlpha="0.8"
android:pathData="M94,50 L190,50 L190,54 L94,54 Z" />
<path
android:fillColor="#bb86fc"
android:fillAlpha="0.7"
android:pathData="M72,62 L76,62 L76,118 L72,118 Z" />
<path
android:fillColor="#ff6b6b"
android:fillAlpha="0.7"
android:pathData="M208,62 L212,62 L212,118 L208,118 Z" />
<path
android:fillColor="#ffd93d"
android:fillAlpha="0.7"
android:pathData="M94,126 L190,126 L190,130 L94,130 Z" />
<!-- Wordmark "LedGrab" — drawn as paths so we don't depend on the
system font cache being warm at TV launch. -->
<!-- L -->
<path android:fillColor="#64ffda"
android:pathData="M222,72 L228,72 L228,100 L240,100 L240,106 L222,106 Z" />
<!-- e -->
<path android:fillColor="#e6edf3"
android:pathData="M244,82 L260,82 Q264,82 264,86 L264,94 L250,94 L250,100 L262,100 L262,106 L246,106 Q244,106 244,104 Z M250,86 L250,90 L258,90 L258,86 Z" />
<!-- d -->
<path android:fillColor="#e6edf3"
android:pathData="M266,72 L272,72 L272,82 L284,82 Q286,82 286,84 L286,106 L268,106 Q266,106 266,104 Z M272,88 L272,100 L280,100 L280,88 Z" />
</vector>
@@ -0,0 +1,31 @@
<?xml version="1.0" encoding="utf-8"?>
<selector xmlns:android="http://schemas.android.com/apk/res/android">
<item android:state_focused="true">
<layer-list>
<item android:left="-6dp" android:top="-6dp" android:right="-6dp" android:bottom="-6dp">
<shape android:shape="rectangle">
<solid android:color="#4064ffda" />
<corners android:radius="44dp" />
</shape>
</item>
<item>
<shape android:shape="rectangle">
<solid android:color="#64ffda" />
<corners android:radius="36dp" />
</shape>
</item>
</layer-list>
</item>
<item android:state_pressed="true">
<shape android:shape="rectangle">
<solid android:color="#3dccb0" />
<corners android:radius="36dp" />
</shape>
</item>
<item>
<shape android:shape="rectangle">
<solid android:color="@color/teal_accent" />
<corners android:radius="36dp" />
</shape>
</item>
</selector>
@@ -0,0 +1,33 @@
<?xml version="1.0" encoding="utf-8"?>
<selector xmlns:android="http://schemas.android.com/apk/res/android">
<item android:state_focused="true">
<layer-list>
<item android:left="-6dp" android:top="-6dp" android:right="-6dp" android:bottom="-6dp">
<shape android:shape="rectangle">
<solid android:color="#40bb86fc" />
<corners android:radius="44dp" />
</shape>
</item>
<item>
<shape android:shape="rectangle">
<solid android:color="#7c4dff" />
<corners android:radius="36dp" />
</shape>
</item>
</layer-list>
</item>
<item android:state_pressed="true">
<shape android:shape="rectangle">
<solid android:color="#9966d4" />
<corners android:radius="36dp" />
<stroke android:width="2dp" android:color="@color/purple_accent" />
</shape>
</item>
<item>
<shape android:shape="rectangle">
<solid android:color="#1Abb86fc" />
<corners android:radius="36dp" />
<stroke android:width="2dp" android:color="@color/purple_accent" />
</shape>
</item>
</selector>
@@ -0,0 +1,28 @@
<?xml version="1.0" encoding="utf-8"?>
<layer-list xmlns:android="http://schemas.android.com/apk/res/android">
<item>
<color android:color="@color/bg_navy" />
</item>
<item>
<shape android:shape="rectangle">
<gradient
android:type="radial"
android:gradientRadius="900dp"
android:centerX="0.5"
android:centerY="-0.2"
android:startColor="#1A64ffda"
android:endColor="#000d1117" />
</shape>
</item>
<item>
<shape android:shape="rectangle">
<gradient
android:type="radial"
android:gradientRadius="600dp"
android:centerX="1.1"
android:centerY="1.2"
android:startColor="#12bb86fc"
android:endColor="#000d1117" />
</shape>
</item>
</layer-list>
@@ -0,0 +1,16 @@
<?xml version="1.0" encoding="utf-8"?>
<layer-list xmlns:android="http://schemas.android.com/apk/res/android">
<item android:left="-8dp" android:top="-8dp" android:right="-8dp" android:bottom="-8dp">
<shape android:shape="rectangle">
<solid android:color="#2064ffda" />
<corners android:radius="24dp" />
</shape>
</item>
<item>
<shape android:shape="rectangle">
<solid android:color="#ffffff" />
<corners android:radius="16dp" />
<stroke android:width="3dp" android:color="@color/teal_accent" />
</shape>
</item>
</layer-list>
@@ -0,0 +1,9 @@
<?xml version="1.0" encoding="utf-8"?>
<!-- Static fallback for the status dot. The animated version
(animated_status_dot.xml) is used at runtime; this is what
XML rendering tools show in the editor. -->
<shape xmlns:android="http://schemas.android.com/apk/res/android"
android:shape="oval">
<solid android:color="@color/green_status" />
<size android:width="18dp" android:height="18dp" />
</shape>
@@ -0,0 +1,7 @@
<?xml version="1.0" encoding="utf-8"?>
<shape xmlns:android="http://schemas.android.com/apk/res/android"
android:shape="rectangle">
<solid android:color="@color/bg_surface_elevated" />
<corners android:radius="12dp" />
<stroke android:width="1dp" android:color="#2264ffda" />
</shape>
@@ -0,0 +1,52 @@
<?xml version="1.0" encoding="utf-8"?>
<vector xmlns:android="http://schemas.android.com/apk/res/android"
android:width="108dp"
android:height="108dp"
android:viewportWidth="108"
android:viewportHeight="108">
<!-- Background circle -->
<path
android:fillColor="#0d1117"
android:pathData="M54,54m-50,0a50,50 0,1 1,100 0a50,50 0,1 1,-100 0" />
<!-- Border ring -->
<path
android:strokeColor="#2264ffda"
android:strokeWidth="1.5"
android:fillColor="#00000000"
android:pathData="M54,54m-48,0a48,48 0,1 1,96 0a48,48 0,1 1,-96 0" />
<!-- TV body -->
<path
android:fillColor="#1c2333"
android:pathData="M26,32 L82,32 Q86,32 86,36 L86,68 Q86,72 82,72 L26,72 Q22,72 22,68 L22,36 Q22,32 26,32 Z" />
<!-- TV screen -->
<path
android:fillColor="#161b22"
android:pathData="M28,35 L80,35 Q82,35 82,37 L82,66 Q82,68 80,68 L28,68 Q26,68 26,66 L26,37 Q26,35 28,35 Z" />
<!-- LED glow - top (teal) -->
<path
android:fillColor="#64ffda"
android:fillAlpha="0.7"
android:pathData="M30,28 L78,28 L78,30 L30,30 Z" />
<!-- LED glow - left (purple) -->
<path
android:fillColor="#bb86fc"
android:fillAlpha="0.6"
android:pathData="M18,34 L20,34 L20,70 L18,70 Z" />
<!-- LED glow - right (red) -->
<path
android:fillColor="#ff6b6b"
android:fillAlpha="0.6"
android:pathData="M88,34 L90,34 L90,70 L88,70 Z" />
<!-- LED glow - bottom (yellow) -->
<path
android:fillColor="#ffd93d"
android:fillAlpha="0.6"
android:pathData="M30,74 L78,74 L78,76 L30,76 Z" />
<!-- TV stand -->
<path
android:fillColor="#1c2333"
android:pathData="M44,72 L44,78 L64,78 L64,72" />
<path
android:fillColor="#1c2333"
android:pathData="M38,78 L70,78 L70,80 L38,80 Z" />
</vector>
@@ -0,0 +1,44 @@
<?xml version="1.0" encoding="utf-8"?>
<!-- Adaptive icon foreground: TV with LED glow strips.
Centered in the 108dp safe zone (inner 72dp is guaranteed visible). -->
<vector xmlns:android="http://schemas.android.com/apk/res/android"
android:width="108dp"
android:height="108dp"
android:viewportWidth="108"
android:viewportHeight="108">
<!-- TV body -->
<path
android:fillColor="#1c2333"
android:pathData="M26,32 L82,32 Q86,32 86,36 L86,68 Q86,72 82,72 L26,72 Q22,72 22,68 L22,36 Q22,32 26,32 Z" />
<!-- TV screen -->
<path
android:fillColor="#161b22"
android:pathData="M28,35 L80,35 Q82,35 82,37 L82,66 Q82,68 80,68 L28,68 Q26,68 26,66 L26,37 Q26,35 28,35 Z" />
<!-- LED glow - top (teal) -->
<path
android:fillColor="#64ffda"
android:fillAlpha="0.7"
android:pathData="M30,28 L78,28 L78,30 L30,30 Z" />
<!-- LED glow - left (purple) -->
<path
android:fillColor="#bb86fc"
android:fillAlpha="0.6"
android:pathData="M18,34 L20,34 L20,70 L18,70 Z" />
<!-- LED glow - right (red) -->
<path
android:fillColor="#ff6b6b"
android:fillAlpha="0.6"
android:pathData="M88,34 L90,34 L90,70 L88,70 Z" />
<!-- LED glow - bottom (yellow) -->
<path
android:fillColor="#ffd93d"
android:fillAlpha="0.6"
android:pathData="M30,74 L78,74 L78,76 L30,76 Z" />
<!-- TV stand -->
<path
android:fillColor="#1c2333"
android:pathData="M44,72 L44,78 L64,78 L64,72" />
<path
android:fillColor="#1c2333"
android:pathData="M38,78 L70,78 L70,80 L38,80 Z" />
</vector>
@@ -0,0 +1,23 @@
<?xml version="1.0" encoding="utf-8"?>
<!-- Monochrome status-bar icon. Android requires white-on-transparent for
notification icons since API 21 - reusing the colored launcher would
render as a gray blob. -->
<vector xmlns:android="http://schemas.android.com/apk/res/android"
android:width="24dp"
android:height="24dp"
android:viewportWidth="24"
android:viewportHeight="24"
android:tint="#FFFFFFFF">
<!-- TV body -->
<path
android:fillColor="#FFFFFFFF"
android:pathData="M5,7 L19,7 Q20,7 20,8 L20,16 Q20,17 19,17 L5,17 Q4,17 4,16 L4,8 Q4,7 5,7 Z M5.5,8.5 L5.5,15.5 L18.5,15.5 L18.5,8.5 Z" />
<!-- TV stand -->
<path
android:fillColor="#FFFFFFFF"
android:pathData="M10,17 L10,18.5 L14,18.5 L14,17 Z M9,19 L15,19 L15,20 L9,20 Z" />
<!-- LED glow strips around the TV (bright dots) -->
<path
android:fillColor="#FFFFFFFF"
android:pathData="M6,5.5 L18,5.5 L18,6.5 L6,6.5 Z" />
</vector>
@@ -0,0 +1,37 @@
<?xml version="1.0" encoding="utf-8"?>
<!-- Splash screen icon (API 31+ uses a 1:1 vector inside a 240dp circle).
The SplashScreen API masks this with a circle automatically. -->
<vector xmlns:android="http://schemas.android.com/apk/res/android"
android:width="240dp"
android:height="240dp"
android:viewportWidth="108"
android:viewportHeight="108">
<!-- TV body -->
<path
android:fillColor="#1c2333"
android:pathData="M26,32 L82,32 Q86,32 86,36 L86,68 Q86,72 82,72 L26,72 Q22,72 22,68 L22,36 Q22,32 26,32 Z" />
<!-- TV screen -->
<path
android:fillColor="#161b22"
android:pathData="M28,35 L80,35 Q82,35 82,37 L82,66 Q82,68 80,68 L28,68 Q26,68 26,66 L26,37 Q26,35 28,35 Z" />
<!-- LED glow strips, brighter on splash for impact -->
<path
android:fillColor="#64ffda"
android:pathData="M30,28 L78,28 L78,30 L30,30 Z" />
<path
android:fillColor="#bb86fc"
android:pathData="M18,34 L20,34 L20,70 L18,70 Z" />
<path
android:fillColor="#ff6b6b"
android:pathData="M88,34 L90,34 L90,70 L88,70 Z" />
<path
android:fillColor="#ffd93d"
android:pathData="M30,74 L78,74 L78,76 L30,76 Z" />
<!-- TV stand -->
<path
android:fillColor="#1c2333"
android:pathData="M44,72 L44,78 L64,78 L64,72" />
<path
android:fillColor="#1c2333"
android:pathData="M38,78 L70,78 L70,80 L38,80 Z" />
</vector>
@@ -0,0 +1,133 @@
<?xml version="1.0" encoding="utf-8"?>
<FrameLayout
xmlns:android="http://schemas.android.com/apk/res/android"
xmlns:tools="http://schemas.android.com/tools"
android:layout_width="match_parent"
android:layout_height="match_parent"
android:background="@drawable/bg_main">
<!-- STOPPED STATE -->
<LinearLayout
android:id="@+id/stopped_panel"
android:layout_width="match_parent"
android:layout_height="match_parent"
android:orientation="vertical"
android:gravity="center"
android:paddingStart="160dp"
android:paddingEnd="160dp">
<ImageView
android:layout_width="72dp"
android:layout_height="72dp"
android:src="@drawable/ic_launcher"
android:contentDescription="@null"
android:layout_marginBottom="24dp" />
<TextView
android:layout_width="wrap_content"
android:layout_height="wrap_content"
android:text="@string/app_name"
android:textColor="@color/teal_accent"
android:textSize="64sp"
android:textStyle="bold"
android:letterSpacing="0.08"
android:layout_marginBottom="12dp"
android:fontFamily="sans-serif" />
<TextView
android:id="@+id/tagline_text"
android:layout_width="wrap_content"
android:layout_height="wrap_content"
android:text="@string/tagline"
android:textColor="@color/text_secondary"
android:textSize="28sp"
android:layout_marginBottom="24dp" />
<!-- Transient status (root probing / permission denial). Always
present so the layout doesn't reflow when text appears. -->
<TextView
android:id="@+id/status_text"
android:layout_width="wrap_content"
android:layout_height="32dp"
android:gravity="center"
android:textColor="@color/text_secondary"
android:textSize="20sp"
android:layout_marginBottom="32dp"
tools:text="Checking root access…" />
<Button
android:id="@+id/toggle_button"
style="@style/Widget.LedGrab.Button.Primary"
android:layout_width="320dp"
android:layout_height="72dp"
android:text="@string/btn_start"
android:textSize="22sp"
android:focusable="true"
android:focusableInTouchMode="true"
android:nextFocusDown="@+id/autostart_check" />
<!-- Shown only while notification-listener access is missing. The D-pad
focus chain is wired at runtime (wireStoppedFocusChain) because this
button and the autostart checkbox are both conditionally visible. -->
<Button
android:id="@+id/grant_notification_button"
style="@style/Widget.LedGrab.Button.Secondary"
android:layout_width="320dp"
android:layout_height="56dp"
android:layout_marginTop="20dp"
android:text="@string/btn_grant_notification_access"
android:textSize="18sp"
android:focusable="true"
android:focusableInTouchMode="true"
android:visibility="gone" />
<!-- Shown only while Usage Access is missing (needed by the foreground-app
automation rule). Like the grant-notification button, its D-pad focus
chain is wired at runtime (wireStoppedFocusChain). -->
<Button
android:id="@+id/grant_usage_access_button"
style="@style/Widget.LedGrab.Button.Secondary"
android:layout_width="320dp"
android:layout_height="56dp"
android:layout_marginTop="20dp"
android:text="@string/btn_grant_usage_access"
android:textSize="18sp"
android:focusable="true"
android:focusableInTouchMode="true"
android:visibility="gone" />
<CheckBox
android:id="@+id/autostart_check"
android:layout_width="wrap_content"
android:layout_height="wrap_content"
android:layout_marginTop="32dp"
android:text="@string/autostart_label"
android:textColor="@color/text_secondary"
android:textSize="20sp"
android:buttonTint="@color/teal_accent"
android:focusable="true"
android:focusableInTouchMode="true"
android:nextFocusUp="@id/toggle_button" />
</LinearLayout>
<!-- Version at bottom (always visible — looks polished on TV idle). -->
<TextView
android:id="@+id/version_text"
android:layout_width="wrap_content"
android:layout_height="wrap_content"
android:layout_gravity="bottom|center_horizontal"
android:layout_marginBottom="32dp"
android:textColor="@color/text_hint"
android:textSize="18sp"
tools:text="v0.1.0" />
<!-- RUNNING STATE — deferred-inflate via ViewStub so first paint is
cheaper and the inflater doesn't measure two competing layouts. -->
<ViewStub
android:id="@+id/running_panel_stub"
android:layout_width="match_parent"
android:layout_height="match_parent"
android:inflatedId="@+id/running_panel"
android:layout="@layout/panel_running"
android:visibility="gone" />
</FrameLayout>
@@ -0,0 +1,124 @@
<?xml version="1.0" encoding="utf-8"?>
<!-- RUNNING STATE -->
<LinearLayout
xmlns:android="http://schemas.android.com/apk/res/android"
xmlns:tools="http://schemas.android.com/tools"
android:layout_width="match_parent"
android:layout_height="match_parent"
android:orientation="horizontal"
android:gravity="center_vertical"
android:paddingStart="120dp"
android:paddingEnd="120dp"
android:paddingTop="80dp"
android:paddingBottom="80dp">
<!-- Left: status + URL + stop -->
<LinearLayout
android:layout_width="0dp"
android:layout_height="wrap_content"
android:layout_weight="1"
android:orientation="vertical"
android:gravity="start|center_vertical"
android:paddingEnd="64dp">
<LinearLayout
android:layout_width="wrap_content"
android:layout_height="wrap_content"
android:orientation="horizontal"
android:gravity="center_vertical"
android:layout_marginBottom="32dp">
<View
android:id="@+id/status_dot"
android:layout_width="18dp"
android:layout_height="18dp"
android:background="@drawable/bg_status_dot"
android:layout_marginEnd="16dp" />
<TextView
android:layout_width="wrap_content"
android:layout_height="wrap_content"
android:text="@string/status_running"
android:textColor="@color/green_status"
android:textSize="28sp"
android:textStyle="bold"
android:letterSpacing="0.05" />
</LinearLayout>
<TextView
android:layout_width="wrap_content"
android:layout_height="wrap_content"
android:text="@string/label_web_ui"
android:textColor="@color/text_secondary"
android:textSize="22sp"
android:layout_marginBottom="8dp" />
<TextView
android:id="@+id/url_text"
android:layout_width="wrap_content"
android:layout_height="wrap_content"
android:textColor="@color/teal_accent"
android:textSize="30sp"
android:maxLines="1"
android:textStyle="bold"
android:background="@drawable/bg_url_chip"
android:paddingStart="24dp"
android:paddingEnd="24dp"
android:paddingTop="12dp"
android:paddingBottom="12dp"
android:layout_marginBottom="56dp"
tools:text="http://192.168.1.5:8080" />
<Button
android:id="@+id/stop_button_running"
style="@style/Widget.LedGrab.Button.Secondary"
android:layout_width="240dp"
android:layout_height="64dp"
android:text="@string/btn_stop"
android:textSize="20sp"
android:focusable="true"
android:focusableInTouchMode="true"
android:nextFocusUp="@id/stop_button_running"
android:nextFocusDown="@id/stop_button_running" />
</LinearLayout>
<!-- Right: QR code + fallback hint -->
<LinearLayout
android:layout_width="wrap_content"
android:layout_height="wrap_content"
android:orientation="vertical"
android:gravity="center">
<FrameLayout
android:layout_width="wrap_content"
android:layout_height="wrap_content"
android:background="@drawable/bg_qr_container"
android:padding="20dp"
android:layout_marginBottom="20dp">
<ImageView
android:id="@+id/qr_image"
android:layout_width="280dp"
android:layout_height="280dp"
android:contentDescription="@string/qr_description"
android:scaleType="fitXY" />
</FrameLayout>
<TextView
android:layout_width="wrap_content"
android:layout_height="wrap_content"
android:text="@string/scan_to_configure"
android:textColor="@color/text_secondary"
android:textSize="22sp"
android:gravity="center" />
<TextView
android:layout_width="280dp"
android:layout_height="wrap_content"
android:text="@string/scan_fallback_hint"
android:textColor="@color/text_hint"
android:textSize="14sp"
android:gravity="center"
android:layout_marginTop="6dp" />
</LinearLayout>
</LinearLayout>
@@ -0,0 +1,5 @@
<?xml version="1.0" encoding="utf-8"?>
<adaptive-icon xmlns:android="http://schemas.android.com/apk/res/android">
<background android:drawable="@color/bg_navy" />
<foreground android:drawable="@drawable/ic_launcher_foreground" />
</adaptive-icon>
@@ -0,0 +1,31 @@
<?xml version="1.0" encoding="utf-8"?>
<resources>
<string name="app_name">LedGrab</string>
<string name="tagline">Фоновая подсветка для телевизора</string>
<string name="btn_start">Начать захват</string>
<string name="btn_starting">Запуск…</string>
<string name="btn_stop">Стоп</string>
<string name="status_running">Работает</string>
<string name="status_checking_root">Проверка root-доступа…</string>
<string name="status_permission_denied">Доступ запрещён — для захвата экрана требуется разрешение</string>
<string name="status_no_network">Нет сети — подключите Wi-Fi или Ethernet</string>
<string name="label_web_ui">Адрес веб-интерфейса</string>
<string name="scan_to_configure">Сканируйте для настройки</string>
<string name="scan_fallback_hint">или откройте этот адрес с любого устройства в сети</string>
<string name="qr_description">QR-код для веб-интерфейса</string>
<string name="version_prefix">v%1$s</string>
<string name="autostart_label">Запускать при загрузке (только с root)</string>
<string name="autostart_unavailable">Запуск при загрузке — недоступно (нужен root)</string>
<string name="fatal_title">Не удалось запустить LedGrab</string>
<string name="fatal_body_prefix">Ошибка инициализации Python:</string>
<string name="fatal_copy_log">Скопировать журнал</string>
<string name="fatal_show_details">Показать подробности</string>
<string name="fatal_hide_details">Скрыть подробности</string>
<string name="notification_channel_name">Захват LedGrab</string>
<string name="notification_channel_description">Отображается, пока LedGrab захватывает экран.</string>
<string name="notification_title">LedGrab работает</string>
<string name="notification_text">Веб-интерфейс: %1$s</string>
<string name="notification_listener_label">Захват уведомлений LedGrab</string>
<string name="btn_grant_notification_access">Разрешить доступ к уведомлениям</string>
<string name="btn_grant_usage_access">Разрешить доступ к статистике использования</string>
</resources>
@@ -0,0 +1,31 @@
<?xml version="1.0" encoding="utf-8"?>
<resources>
<string name="app_name">LedGrab</string>
<string name="tagline">电视氛围灯光</string>
<string name="btn_start">开始捕获</string>
<string name="btn_starting">正在启动…</string>
<string name="btn_stop">停止</string>
<string name="status_running">运行中</string>
<string name="status_checking_root">正在检查 root 权限…</string>
<string name="status_permission_denied">权限被拒绝 — 屏幕捕获需要授权</string>
<string name="status_no_network">无网络 — 请连接 Wi-Fi 或以太网</string>
<string name="label_web_ui">Web界面地址</string>
<string name="scan_to_configure">扫码配置</string>
<string name="scan_fallback_hint">或在同一网络的任何设备上访问上方网址</string>
<string name="qr_description">Web界面二维码</string>
<string name="version_prefix">v%1$s</string>
<string name="autostart_label">开机自启(仅限 root)</string>
<string name="autostart_unavailable">开机自启 — 不可用(需要 root)</string>
<string name="fatal_title">LedGrab 启动失败</string>
<string name="fatal_body_prefix">Python 运行时初始化失败:</string>
<string name="fatal_copy_log">复制日志</string>
<string name="fatal_show_details">显示详情</string>
<string name="fatal_hide_details">隐藏详情</string>
<string name="notification_channel_name">LedGrab 屏幕捕获</string>
<string name="notification_channel_description">LedGrab 捕获屏幕时显示。</string>
<string name="notification_title">LedGrab 运行中</string>
<string name="notification_text">Web界面:%1$s</string>
<string name="notification_listener_label">LedGrab 通知捕获</string>
<string name="btn_grant_notification_access">授予通知访问权限</string>
<string name="btn_grant_usage_access">授予使用情况访问权限</string>
</resources>
@@ -0,0 +1,20 @@
<?xml version="1.0" encoding="utf-8"?>
<resources>
<color name="bg_navy">#0d1117</color>
<color name="bg_navy_mid">#111827</color>
<color name="bg_surface">#161b22</color>
<color name="bg_surface_elevated">#1c2333</color>
<color name="teal_accent">#64ffda</color>
<color name="teal_accent_dim">#3399aa</color>
<color name="teal_accent_alpha30">#4D64ffda</color>
<color name="teal_accent_alpha15">#2664ffda</color>
<color name="purple_accent">#bb86fc</color>
<color name="purple_accent_dim">#7c4dff</color>
<color name="purple_accent_alpha30">#4Dbb86fc</color>
<color name="purple_accent_alpha15">#26bb86fc</color>
<color name="green_status">#4caf50</color>
<color name="green_status_dim">#1b5e20</color>
<color name="text_primary">#e6edf3</color>
<color name="text_secondary">#8b949e</color>
<color name="text_hint">#484f58</color>
</resources>
@@ -0,0 +1,31 @@
<?xml version="1.0" encoding="utf-8"?>
<resources>
<string name="app_name">LedGrab</string>
<string name="tagline">Ambient lighting for your TV</string>
<string name="btn_start">Start Capture</string>
<string name="btn_starting">Starting…</string>
<string name="btn_stop">Stop</string>
<string name="status_running">Running</string>
<string name="status_checking_root">Checking root access…</string>
<string name="status_permission_denied">Permission denied — screen capture requires authorization</string>
<string name="status_no_network">No network — connect Wi-Fi or Ethernet</string>
<string name="label_web_ui">Web UI address</string>
<string name="scan_to_configure">Scan to configure</string>
<string name="scan_fallback_hint">or visit the URL above on any device on this network</string>
<string name="qr_description">QR code for web UI</string>
<string name="version_prefix">v%1$s</string>
<string name="autostart_label">Start on boot (root only)</string>
<string name="autostart_unavailable">Start on boot — unavailable (root required)</string>
<string name="fatal_title">LedGrab failed to start</string>
<string name="fatal_body_prefix">Python runtime initialization failed:</string>
<string name="fatal_copy_log">Copy log</string>
<string name="fatal_show_details">Show details</string>
<string name="fatal_hide_details">Hide details</string>
<string name="notification_channel_name">LedGrab capture</string>
<string name="notification_channel_description">Shows while LedGrab is capturing the screen.</string>
<string name="notification_title">LedGrab Running</string>
<string name="notification_text">Web UI: %1$s</string>
<string name="notification_listener_label">LedGrab notification capture</string>
<string name="btn_grant_notification_access">Grant notification access</string>
<string name="btn_grant_usage_access">Grant usage access</string>
</resources>
@@ -0,0 +1,40 @@
<?xml version="1.0" encoding="utf-8"?>
<resources>
<style name="Theme.LedGrab" parent="@style/Theme.Leanback">
<item name="android:windowBackground">@color/bg_navy</item>
<item name="android:colorBackground">@color/bg_navy</item>
<item name="android:windowNoTitle">true</item>
<item name="android:textColorPrimary">@color/text_primary</item>
<item name="android:textColorSecondary">@color/text_secondary</item>
<item name="android:textColorHint">@color/text_hint</item>
<item name="android:colorAccent">@color/teal_accent</item>
<item name="android:colorControlHighlight">@color/teal_accent_dim</item>
<item name="android:colorControlActivated">@color/teal_accent</item>
</style>
<!-- Splash screen theme. Compatible across API levels via the
androidx.core:core-splashscreen library. On API 31+ the system
splash uses the foreground icon; on older versions the launch
theme just paints the navy background, which is harmless. -->
<style name="Theme.LedGrab.Splash" parent="Theme.SplashScreen">
<item name="windowSplashScreenBackground">@color/bg_navy</item>
<item name="windowSplashScreenAnimatedIcon">@drawable/ic_splash</item>
<item name="postSplashScreenTheme">@style/Theme.LedGrab</item>
</style>
<style name="Widget.LedGrab.Button.Primary" parent="@android:style/Widget.Button">
<item name="android:background">@drawable/bg_button_primary</item>
<item name="android:textColor">@color/bg_navy</item>
<item name="android:textStyle">bold</item>
<item name="android:focusable">true</item>
<item name="android:stateListAnimator">@null</item>
</style>
<style name="Widget.LedGrab.Button.Secondary" parent="@android:style/Widget.Button">
<item name="android:background">@drawable/bg_button_secondary</item>
<item name="android:textColor">@color/btn_secondary_text</item>
<item name="android:textStyle">bold</item>
<item name="android:focusable">true</item>
<item name="android:stateListAnimator">@null</item>
</style>
</resources>
@@ -0,0 +1,29 @@
<?xml version="1.0" encoding="utf-8"?>
<!--
LedGrab is a LAN-only app:
- Inbound: web UI / API on the device (HTTP, port 8080)
- Outbound: WLED HTTP/UDP, Home Assistant, MQTT brokers, mDNS
All of these are plaintext on the local network. Android's network
security config doesn't support CIDR allowlists, so we cannot
restrict cleartext to RFC1918 ranges declaratively — we have to
permit cleartext base-wide.
Defence-in-depth that ACTUALLY mitigates this:
1. Inbound: the FastAPI server in this app rejects non-loopback
requests when no API key is configured (see ledgrab.api.auth).
The Android launcher auto-generates an API key on first run
(see ApiKeyManager.kt) and injects it via the
LEDGRAB_AUTH__API_KEYS env var before uvicorn starts. The
user's phone receives the key by scanning the QR, which
embeds the key as a URL fragment (never logged server-side).
2. Outbound: targets are validated by net_classify in the Python
layer (LAN-only HTTP, SSRF-safe).
DO NOT remove the cleartext permission without first migrating
every LAN peer to HTTPS — most WLED firmware, mDNS, and the LAN
HTTP server itself rely on this flag.
-->
<network-security-config>
<base-config cleartextTrafficPermitted="true" />
</network-security-config>
@@ -0,0 +1,203 @@
#!/usr/bin/env bash
#
# Cross-compile pydantic-core for Android across all supported ABIs:
# arm64-v8a (primary — modern TV hardware)
# x86_64 (modern emulators)
# x86 (legacy emulators)
# armeabi-v7a (32-bit ARMv7 — older cheap TV boxes like X96 mini, MeCool)
#
# Outputs wheels into android/wheels/. Wheels are linked against the real
# libpython3.11.so shipped by Chaquopy (stub .so does NOT work — see
# memory/project_android_app.md for the incident notes).
#
# Prerequisites (on host):
# - Rust + cargo (rustup) with targets:
# aarch64/x86_64/i686/armv7a-linux-android(eabi)
# - Android NDK (ANDROID_NDK_HOME, or installed at Sdk/ndk/*)
# - Python 3.11 (matches Chaquopy's embedded version)
# - maturin (pip install maturin)
#
# Rebuild cadence: whenever PYDANTIC_CORE_VERSION changes, or pydantic's
# core dependency version changes.
#
# Usage:
# ./build-pydantic-core.sh # build all 4 ABIs
# ./build-pydantic-core.sh arm64 # build a single ABI
# ./build-pydantic-core.sh arm64 x86_64 # build a subset
# ./build-pydantic-core.sh armv7 # 32-bit ARM only
#
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
ANDROID_DIR="$(cd "$SCRIPT_DIR/.." && pwd)"
WHEELS_DIR="$ANDROID_DIR/wheels"
BUILD_DIR="$ANDROID_DIR/.build-cache"
PYDANTIC_CORE_VERSION="2.46.0"
API_LEVEL=24
PY_VERSION="3.11"
# Resolve a concrete python3.11 interpreter path (maturin needs it callable).
# Windows Git Bash doesn't ship `python3.11` on PATH; fall back to `py -3.11`.
if [ -z "${PYTHON_BIN:-}" ]; then
# Prefer `py -3.11` on Windows (Git Bash's `python3.11` resolves to a
# non-functional MSStore stub). Fall back to `python3.11` on Linux/macOS.
if command -v py >/dev/null 2>&1 && py -3.11 -c '' 2>/dev/null; then
PYTHON_BIN="$(py -3.11 -c 'import sys; print(sys.executable)' 2>/dev/null | tr -d '\r')"
elif command -v python3.11 >/dev/null 2>&1 && python3.11 -c '' 2>/dev/null; then
PYTHON_BIN="$(command -v python3.11)"
fi
fi
[ -n "${PYTHON_BIN:-}" ] || { echo "ERROR: python3.11 not found; set PYTHON_BIN"; exit 1; }
echo "Python: $PYTHON_BIN"
mkdir -p "$WHEELS_DIR"
# ── Find Android NDK ────────────────────────────────────────────────
if [ -z "${ANDROID_NDK_HOME:-}" ]; then
for candidate in \
"$HOME/Library/Android/sdk/ndk"/* \
"$HOME/Android/Sdk/ndk"/* \
"${LOCALAPPDATA:-}/Android/Sdk/ndk"/* \
"/c/Users/$(whoami)/AppData/Local/Android/Sdk/ndk"/* \
"/usr/local/lib/android/sdk/ndk"/* \
"${ANDROID_SDK_ROOT:-}/ndk"/*; do
if [ -d "$candidate" ]; then
ANDROID_NDK_HOME="$candidate"
break
fi
done
fi
[ -n "${ANDROID_NDK_HOME:-}" ] || { echo "ERROR: NDK not found; set ANDROID_NDK_HOME"; exit 1; }
echo "NDK: $ANDROID_NDK_HOME"
case "$(uname -s)" in
Linux*) HOST_TAG="linux-x86_64" ;;
Darwin*) HOST_TAG="darwin-x86_64" ;;
MINGW*|MSYS*|CYGWIN*) HOST_TAG="windows-x86_64" ;;
*) echo "Unsupported host OS"; exit 1 ;;
esac
TOOLCHAIN="$ANDROID_NDK_HOME/toolchains/llvm/prebuilt/$HOST_TAG"
READELF="$TOOLCHAIN/bin/llvm-readelf"
# ── Prepare source tree ─────────────────────────────────────────────
SRC_DIR="$BUILD_DIR/pydantic_core-$PYDANTIC_CORE_VERSION"
if [ ! -d "$SRC_DIR" ]; then
echo "Downloading pydantic_core $PYDANTIC_CORE_VERSION sdist..."
mkdir -p "$BUILD_DIR"
SDIST="$BUILD_DIR/pydantic_core-$PYDANTIC_CORE_VERSION.tar.gz"
[ -f "$SDIST" ] || curl -sSL --retry 3 -o "$SDIST" \
"https://files.pythonhosted.org/packages/source/p/pydantic_core/pydantic_core-$PYDANTIC_CORE_VERSION.tar.gz"
tar -xzf "$SDIST" -C "$BUILD_DIR"
fi
# ── ABI table ───────────────────────────────────────────────────────
# Columns: short_name rust_target clang_prefix sysconfig_dir
ABI_TABLE=(
"arm64 aarch64-linux-android aarch64-linux-android${API_LEVEL} cross-sysconfig"
"x86_64 x86_64-linux-android x86_64-linux-android${API_LEVEL} cross-sysconfig-x86_64"
"x86 i686-linux-android i686-linux-android${API_LEVEL} cross-sysconfig-x86"
"armv7 armv7-linux-androideabi armv7a-linux-androideabi${API_LEVEL} cross-sysconfig-armv7"
)
declare -A ABI_TAG_MAP=(
[arm64]="arm64_v8a"
[x86_64]="x86_64"
[x86]="x86"
[armv7]="armeabi_v7a"
)
# ── Select which ABIs to build ──────────────────────────────────────
SELECTED=("$@")
if [ ${#SELECTED[@]} -eq 0 ]; then
SELECTED=(arm64 x86_64 x86 armv7)
fi
# ── Ensure rust targets are installed ───────────────────────────────
for name in "${SELECTED[@]}"; do
for row in "${ABI_TABLE[@]}"; do
read -r sname rtarget _ _ <<<"$row"
if [ "$sname" = "$name" ]; then
rustup target add "$rtarget" >/dev/null 2>&1 || true
fi
done
done
# ── Build loop ──────────────────────────────────────────────────────
cd "$SRC_DIR"
for name in "${SELECTED[@]}"; do
MATCHED=""
for row in "${ABI_TABLE[@]}"; do
read -r sname rtarget cprefix sysdir <<<"$row"
if [ "$sname" = "$name" ]; then
MATCHED="1"
break
fi
done
[ -n "$MATCHED" ] || { echo "Unknown ABI: $name"; exit 1; }
CROSS_LIB_DIR="$BUILD_DIR/$sysdir"
[ -f "$CROSS_LIB_DIR/libpython3.11.so" ] || {
echo "ERROR: missing $CROSS_LIB_DIR/libpython3.11.so (extract from a Chaquopy APK)"
exit 1
}
# On Windows hosts the clang wrapper is a .cmd batch file; cargo/rustc
# can't exec it as a linker directly (os error 193). Use the .cmd path
# explicitly so CreateProcess picks up cmd.exe as interpreter.
CC_BIN="$TOOLCHAIN/bin/${cprefix}-clang"
if [ "$HOST_TAG" = "windows-x86_64" ] && [ -f "${CC_BIN}.cmd" ]; then
CC_BIN="${CC_BIN}.cmd"
fi
AR_BIN="$TOOLCHAIN/bin/llvm-ar"
[ "$HOST_TAG" = "windows-x86_64" ] && [ -f "${AR_BIN}.exe" ] && AR_BIN="${AR_BIN}.exe"
[ -f "$CC_BIN" ] || { echo "ERROR: $CC_BIN not found"; exit 1; }
# Normalize rust target name for env var (cargo wants UPPER_WITH_UNDERSCORES)
TARGET_ENV=$(echo "$rtarget" | tr 'a-z-' 'A-Z_')
echo ""
echo "════════════════════════════════════════════════════════════"
echo " Building pydantic_core $PYDANTIC_CORE_VERSION for $name ($rtarget)"
echo "════════════════════════════════════════════════════════════"
env \
CARGO_TARGET_DIR="$SRC_DIR/target" \
"CC_${rtarget//-/_}=$CC_BIN" \
"AR_${rtarget//-/_}=$AR_BIN" \
"CARGO_TARGET_${TARGET_ENV}_LINKER=$CC_BIN" \
PYO3_CROSS=1 \
PYO3_CROSS_LIB_DIR="$CROSS_LIB_DIR" \
PYO3_CROSS_PYTHON_VERSION="$PY_VERSION" \
RUSTFLAGS="-C link-arg=-Wl,--no-as-needed -C link-arg=-lpython3.11 -C link-arg=-L$CROSS_LIB_DIR" \
maturin build \
--release \
--target "$rtarget" \
--interpreter "$PYTHON_BIN" \
--out "$WHEELS_DIR" \
--compatibility linux
# maturin writes the wheel with the correct android_<api>_<abi> platform
# tag directly (the sysconfigdata provides MULTIARCH). Find + verify.
ABI_TAG="${ABI_TAG_MAP[$name]}"
OUT_WHL="$WHEELS_DIR/pydantic_core-$PYDANTIC_CORE_VERSION-cp311-cp311-android_${API_LEVEL}_${ABI_TAG}.whl"
[ -f "$OUT_WHL" ] || { echo "ERROR: expected wheel not found: $OUT_WHL"; exit 1; }
TMP=$(mktemp -d)
unzip -qo "$OUT_WHL" -d "$TMP" "pydantic_core/*.so"
SO=$(find "$TMP" -name "*.so" | head -1)
if "$READELF" -d "$SO" | grep -q "libpython3.11.so"; then
echo " NEEDED OK: libpython3.11.so present in $(basename "$OUT_WHL")"
else
echo " ERROR: libpython3.11.so NOT in NEEDED — wheel will crash at import"
"$READELF" -d "$SO" | grep NEEDED
rm -rf "$TMP"
exit 1
fi
rm -rf "$TMP"
done
echo ""
echo "=== Build complete ==="
ls -lh "$WHEELS_DIR"/pydantic_core-*.whl
+74
View File
@@ -0,0 +1,74 @@
#!/usr/bin/env bash
#
# Set up the cross-compilation environment for building Python native
# extensions targeting Android ARM64.
#
# This script:
# 1. Verifies Android NDK is installed
# 2. Installs the Rust aarch64-linux-android target
# 3. Installs maturin (Python wheel builder for Rust extensions)
# 4. Runs a quick test compile to verify the toolchain works
#
set -euo pipefail
echo "=== LedGrab Android NDK Setup ==="
# ── Check prerequisites ─────────────────────────────────────────────
if ! command -v rustc &>/dev/null; then
echo "ERROR: Rust is not installed."
echo "Install from: https://rustup.rs/"
exit 1
fi
echo "Rust: $(rustc --version)"
if ! command -v cargo &>/dev/null; then
echo "ERROR: Cargo is not installed."
exit 1
fi
echo "Cargo: $(cargo --version)"
if ! command -v python3.11 &>/dev/null && ! command -v python3 &>/dev/null; then
echo "WARNING: Python 3.11 not found. Needed for maturin builds."
fi
# ── Install Rust target ─────────────────────────────────────────────
echo ""
echo "Installing Rust Android target..."
rustup target add aarch64-linux-android
echo "Installed targets:"
rustup target list --installed | grep android
# ── Install maturin ─────────────────────────────────────────────────
echo ""
echo "Installing maturin..."
pip install maturin 2>/dev/null || pip3 install maturin 2>/dev/null || {
echo "WARNING: Could not install maturin. Install manually: pip install maturin"
}
if command -v maturin &>/dev/null; then
echo "maturin: $(maturin --version)"
fi
# ── Verify NDK ──────────────────────────────────────────────────────
echo ""
if [ -n "${ANDROID_NDK_HOME:-}" ]; then
echo "ANDROID_NDK_HOME: $ANDROID_NDK_HOME"
else
echo "ANDROID_NDK_HOME is not set."
echo "Set it to your NDK installation path, e.g.:"
echo " export ANDROID_NDK_HOME=\$HOME/Android/Sdk/ndk/26.1.10909125"
echo ""
echo "Or install NDK via Android Studio:"
echo " SDK Manager → SDK Tools → NDK (Side by side)"
fi
echo ""
echo "=== Setup complete ==="
echo ""
echo "Next steps:"
echo " 1. Ensure ANDROID_NDK_HOME is set"
echo " 2. Run: ./build-pydantic-core.sh"
+5
View File
@@ -0,0 +1,5 @@
plugins {
id("com.android.application") version "8.9.0" apply false
id("org.jetbrains.kotlin.android") version "2.1.0" apply false
id("com.chaquo.python") version "17.0.0" apply false
}
+7
View File
@@ -0,0 +1,7 @@
org.gradle.jvmargs=-Xmx2048m -Dfile.encoding=UTF-8
android.useAndroidX=true
kotlin.code.style=official
android.nonTransitiveRClass=true
org.gradle.parallel=false
org.gradle.configuration-cache=false
org.gradle.unsafe.isolated-projects=false
Binary file not shown.
+7
View File
@@ -0,0 +1,7 @@
distributionBase=GRADLE_USER_HOME
distributionPath=wrapper/dists
distributionUrl=https\://services.gradle.org/distributions/gradle-8.11.1-bin.zip
networkTimeout=10000
validateDistributionUrl=true
zipStoreBase=GRADLE_USER_HOME
zipStorePath=wrapper/dists
+252
View File
@@ -0,0 +1,252 @@
#!/bin/sh
#
# Copyright © 2015-2021 the original authors.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# https://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
# SPDX-License-Identifier: Apache-2.0
#
##############################################################################
#
# Gradle start up script for POSIX generated by Gradle.
#
# Important for running:
#
# (1) You need a POSIX-compliant shell to run this script. If your /bin/sh is
# noncompliant, but you have some other compliant shell such as ksh or
# bash, then to run this script, type that shell name before the whole
# command line, like:
#
# ksh Gradle
#
# Busybox and similar reduced shells will NOT work, because this script
# requires all of these POSIX shell features:
# * functions;
# * expansions «$var», «${var}», «${var:-default}», «${var+SET}»,
# «${var#prefix}», «${var%suffix}», and «$( cmd )»;
# * compound commands having a testable exit status, especially «case»;
# * various built-in commands including «command», «set», and «ulimit».
#
# Important for patching:
#
# (2) This script targets any POSIX shell, so it avoids extensions provided
# by Bash, Ksh, etc; in particular arrays are avoided.
#
# The "traditional" practice of packing multiple parameters into a
# space-separated string is a well documented source of bugs and security
# problems, so this is (mostly) avoided, by progressively accumulating
# options in "$@", and eventually passing that to Java.
#
# Where the inherited environment variables (DEFAULT_JVM_OPTS, JAVA_OPTS,
# and GRADLE_OPTS) rely on word-splitting, this is performed explicitly;
# see the in-line comments for details.
#
# There are tweaks for specific operating systems such as AIX, CygWin,
# Darwin, MinGW, and NonStop.
#
# (3) This script is generated from the Groovy template
# https://github.com/gradle/gradle/blob/HEAD/platforms/jvm/plugins-application/src/main/resources/org/gradle/api/internal/plugins/unixStartScript.txt
# within the Gradle project.
#
# You can find Gradle at https://github.com/gradle/gradle/.
#
##############################################################################
# Attempt to set APP_HOME
# Resolve links: $0 may be a link
app_path=$0
# Need this for daisy-chained symlinks.
while
APP_HOME=${app_path%"${app_path##*/}"} # leaves a trailing /; empty if no leading path
[ -h "$app_path" ]
do
ls=$( ls -ld "$app_path" )
link=${ls#*' -> '}
case $link in #(
/*) app_path=$link ;; #(
*) app_path=$APP_HOME$link ;;
esac
done
# This is normally unused
# shellcheck disable=SC2034
APP_BASE_NAME=${0##*/}
# Discard cd standard output in case $CDPATH is set (https://github.com/gradle/gradle/issues/25036)
APP_HOME=$( cd -P "${APP_HOME:-./}" > /dev/null && printf '%s
' "$PWD" ) || exit
# Use the maximum available, or set MAX_FD != -1 to use that value.
MAX_FD=maximum
warn () {
echo "$*"
} >&2
die () {
echo
echo "$*"
echo
exit 1
} >&2
# OS specific support (must be 'true' or 'false').
cygwin=false
msys=false
darwin=false
nonstop=false
case "$( uname )" in #(
CYGWIN* ) cygwin=true ;; #(
Darwin* ) darwin=true ;; #(
MSYS* | MINGW* ) msys=true ;; #(
NONSTOP* ) nonstop=true ;;
esac
CLASSPATH=$APP_HOME/gradle/wrapper/gradle-wrapper.jar
# Determine the Java command to use to start the JVM.
if [ -n "$JAVA_HOME" ] ; then
if [ -x "$JAVA_HOME/jre/sh/java" ] ; then
# IBM's JDK on AIX uses strange locations for the executables
JAVACMD=$JAVA_HOME/jre/sh/java
else
JAVACMD=$JAVA_HOME/bin/java
fi
if [ ! -x "$JAVACMD" ] ; then
die "ERROR: JAVA_HOME is set to an invalid directory: $JAVA_HOME
Please set the JAVA_HOME variable in your environment to match the
location of your Java installation."
fi
else
JAVACMD=java
if ! command -v java >/dev/null 2>&1
then
die "ERROR: JAVA_HOME is not set and no 'java' command could be found in your PATH.
Please set the JAVA_HOME variable in your environment to match the
location of your Java installation."
fi
fi
# Increase the maximum file descriptors if we can.
if ! "$cygwin" && ! "$darwin" && ! "$nonstop" ; then
case $MAX_FD in #(
max*)
# In POSIX sh, ulimit -H is undefined. That's why the result is checked to see if it worked.
# shellcheck disable=SC2039,SC3045
MAX_FD=$( ulimit -H -n ) ||
warn "Could not query maximum file descriptor limit"
esac
case $MAX_FD in #(
'' | soft) :;; #(
*)
# In POSIX sh, ulimit -n is undefined. That's why the result is checked to see if it worked.
# shellcheck disable=SC2039,SC3045
ulimit -n "$MAX_FD" ||
warn "Could not set maximum file descriptor limit to $MAX_FD"
esac
fi
# Collect all arguments for the java command, stacking in reverse order:
# * args from the command line
# * the main class name
# * -classpath
# * -D...appname settings
# * --module-path (only if needed)
# * DEFAULT_JVM_OPTS, JAVA_OPTS, and GRADLE_OPTS environment variables.
# For Cygwin or MSYS, switch paths to Windows format before running java
if "$cygwin" || "$msys" ; then
APP_HOME=$( cygpath --path --mixed "$APP_HOME" )
CLASSPATH=$( cygpath --path --mixed "$CLASSPATH" )
JAVACMD=$( cygpath --unix "$JAVACMD" )
# Now convert the arguments - kludge to limit ourselves to /bin/sh
for arg do
if
case $arg in #(
-*) false ;; # don't mess with options #(
/?*) t=${arg#/} t=/${t%%/*} # looks like a POSIX filepath
[ -e "$t" ] ;; #(
*) false ;;
esac
then
arg=$( cygpath --path --ignore --mixed "$arg" )
fi
# Roll the args list around exactly as many times as the number of
# args, so each arg winds up back in the position where it started, but
# possibly modified.
#
# NB: a `for` loop captures its iteration list before it begins, so
# changing the positional parameters here affects neither the number of
# iterations, nor the values presented in `arg`.
shift # remove old arg
set -- "$@" "$arg" # push replacement arg
done
fi
# Add default JVM options here. You can also use JAVA_OPTS and GRADLE_OPTS to pass JVM options to this script.
DEFAULT_JVM_OPTS='"-Xmx64m" "-Xms64m"'
# Collect all arguments for the java command:
# * DEFAULT_JVM_OPTS, JAVA_OPTS, JAVA_OPTS, and optsEnvironmentVar are not allowed to contain shell fragments,
# and any embedded shellness will be escaped.
# * For example: A user cannot expect ${Hostname} to be expanded, as it is an environment variable and will be
# treated as '${Hostname}' itself on the command line.
set -- \
"-Dorg.gradle.appname=$APP_BASE_NAME" \
-classpath "$CLASSPATH" \
org.gradle.wrapper.GradleWrapperMain \
"$@"
# Stop when "xargs" is not available.
if ! command -v xargs >/dev/null 2>&1
then
die "xargs is not available"
fi
# Use "xargs" to parse quoted args.
#
# With -n1 it outputs one arg per line, with the quotes and backslashes removed.
#
# In Bash we could simply go:
#
# readarray ARGS < <( xargs -n1 <<<"$var" ) &&
# set -- "${ARGS[@]}" "$@"
#
# but POSIX shell has neither arrays nor command substitution, so instead we
# post-process each arg (as a line of input to sed) to backslash-escape any
# character that might be a shell metacharacter, then use eval to reverse
# that process (while maintaining the separation between arguments), and wrap
# the whole thing up as a single "set" statement.
#
# This will of course break if any of these variables contains a newline or
# an unmatched quote.
#
eval "set -- $(
printf '%s\n' "$DEFAULT_JVM_OPTS $JAVA_OPTS $GRADLE_OPTS" |
xargs -n1 |
sed ' s~[^-[:alnum:]+,./:=@_]~\\&~g; ' |
tr '\n' ' '
)" '"$@"'
exec "$JAVACMD" "$@"
+94
View File
@@ -0,0 +1,94 @@
@rem
@rem Copyright 2015 the original author or authors.
@rem
@rem Licensed under the Apache License, Version 2.0 (the "License");
@rem you may not use this file except in compliance with the License.
@rem You may obtain a copy of the License at
@rem
@rem https://www.apache.org/licenses/LICENSE-2.0
@rem
@rem Unless required by applicable law or agreed to in writing, software
@rem distributed under the License is distributed on an "AS IS" BASIS,
@rem WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
@rem See the License for the specific language governing permissions and
@rem limitations under the License.
@rem
@rem SPDX-License-Identifier: Apache-2.0
@rem
@if "%DEBUG%"=="" @echo off
@rem ##########################################################################
@rem
@rem Gradle startup script for Windows
@rem
@rem ##########################################################################
@rem Set local scope for the variables with windows NT shell
if "%OS%"=="Windows_NT" setlocal
set DIRNAME=%~dp0
if "%DIRNAME%"=="" set DIRNAME=.
@rem This is normally unused
set APP_BASE_NAME=%~n0
set APP_HOME=%DIRNAME%
@rem Resolve any "." and ".." in APP_HOME to make it shorter.
for %%i in ("%APP_HOME%") do set APP_HOME=%%~fi
@rem Add default JVM options here. You can also use JAVA_OPTS and GRADLE_OPTS to pass JVM options to this script.
set DEFAULT_JVM_OPTS="-Xmx64m" "-Xms64m"
@rem Find java.exe
if defined JAVA_HOME goto findJavaFromJavaHome
set JAVA_EXE=java.exe
%JAVA_EXE% -version >NUL 2>&1
if %ERRORLEVEL% equ 0 goto execute
echo. 1>&2
echo ERROR: JAVA_HOME is not set and no 'java' command could be found in your PATH. 1>&2
echo. 1>&2
echo Please set the JAVA_HOME variable in your environment to match the 1>&2
echo location of your Java installation. 1>&2
goto fail
:findJavaFromJavaHome
set JAVA_HOME=%JAVA_HOME:"=%
set JAVA_EXE=%JAVA_HOME%/bin/java.exe
if exist "%JAVA_EXE%" goto execute
echo. 1>&2
echo ERROR: JAVA_HOME is set to an invalid directory: %JAVA_HOME% 1>&2
echo. 1>&2
echo Please set the JAVA_HOME variable in your environment to match the 1>&2
echo location of your Java installation. 1>&2
goto fail
:execute
@rem Setup the command line
set CLASSPATH=%APP_HOME%\gradle\wrapper\gradle-wrapper.jar
@rem Execute Gradle
"%JAVA_EXE%" %DEFAULT_JVM_OPTS% %JAVA_OPTS% %GRADLE_OPTS% "-Dorg.gradle.appname=%APP_BASE_NAME%" -classpath "%CLASSPATH%" org.gradle.wrapper.GradleWrapperMain %*
:end
@rem End local scope for the variables with windows NT shell
if %ERRORLEVEL% equ 0 goto mainEnd
:fail
rem Set variable GRADLE_EXIT_CONSOLE if you need the _script_ return code instead of
rem the _cmd.exe /c_ return code!
set EXIT_CODE=%ERRORLEVEL%
if %EXIT_CODE% equ 0 set EXIT_CODE=1
if not ""=="%GRADLE_EXIT_CONSOLE%" exit %EXIT_CODE%
exit /b %EXIT_CODE%
:mainEnd
if "%OS%"=="Windows_NT" endlocal
:omega
+20
View File
@@ -0,0 +1,20 @@
pluginManagement {
repositories {
google()
mavenCentral()
gradlePluginPortal()
}
}
dependencyResolutionManagement {
repositoriesMode.set(RepositoriesMode.FAIL_ON_PROJECT_REPOS)
repositories {
google()
mavenCentral()
// usb-serial-for-android (mik3y) is distributed via JitPack
maven { url = uri("https://jitpack.io") }
}
}
rootProject.name = "LedGrab"
include(":app")
-140
View File
@@ -1,140 +0,0 @@
#!/usr/bin/env bash
#
# Shared build functions for LedGrab distribution packaging.
# Sourced by build-dist.sh (Linux) and build-dist-windows.sh (Windows).
#
# Expected variables set by the caller before sourcing:
# SCRIPT_DIR, BUILD_DIR, DIST_DIR, SERVER_DIR, APP_DIR
# ── Version detection ────────────────────────────────────────
detect_version() {
# Usage: detect_version [explicit_version]
local version="${1:-}"
if [ -z "$version" ]; then
version=$(git describe --tags --exact-match 2>/dev/null || true)
fi
if [ -z "$version" ]; then
version="${GITEA_REF_NAME:-${GITHUB_REF_NAME:-}}"
fi
if [ -z "$version" ]; then
version=$(grep -oP '^version\s*=\s*"\K[^"]+' "$SERVER_DIR/pyproject.toml" 2>/dev/null || echo "0.0.0")
fi
VERSION_CLEAN="${version#v}"
# Stamp the resolved version into pyproject.toml so that
# importlib.metadata reads the correct value at runtime.
sed -i "s/^version = .*/version = \"${VERSION_CLEAN}\"/" "$SERVER_DIR/pyproject.toml"
}
# ── Clean previous build ─────────────────────────────────────
clean_dist() {
if [ -d "$DIST_DIR" ]; then
echo " Cleaning previous build..."
rm -rf "$DIST_DIR"
fi
mkdir -p "$DIST_DIR"
}
# ── Build frontend ───────────────────────────────────────────
build_frontend() {
echo " Building frontend bundle..."
(cd "$SERVER_DIR" && npm ci --loglevel error && npm run build) 2>&1 | {
grep -v 'RemoteException' || true
}
}
# ── Copy application files ───────────────────────────────────
copy_app_files() {
echo " Copying application files..."
mkdir -p "$APP_DIR"
cp -r "$SERVER_DIR/src" "$APP_DIR/src"
cp -r "$SERVER_DIR/config" "$APP_DIR/config"
mkdir -p "$DIST_DIR/data" "$DIST_DIR/logs"
# Clean up source maps and __pycache__
find "$APP_DIR" -name "*.map" -delete 2>/dev/null || true
find "$APP_DIR" -type d -name __pycache__ -exec rm -rf {} + 2>/dev/null || true
}
# ── Site-packages cleanup ────────────────────────────────────
#
# Strips tests, type stubs, unused submodules, and debug symbols
# from the installed site-packages directory.
#
# Args:
# $1 — path to site-packages directory
# $2 — native extension suffix: "pyd" (Windows) or "so" (Linux)
# $3 — native lib suffix for OpenCV ffmpeg: "dll" or "so"
cleanup_site_packages() {
local sp_dir="$1"
local ext_suffix="${2:-so}"
local lib_suffix="${3:-so}"
echo " Cleaning up site-packages to reduce size..."
# ── Generic cleanup ──────────────────────────────────────
find "$sp_dir" -type d -name __pycache__ -exec rm -rf {} + 2>/dev/null || true
find "$sp_dir" -type d -name tests -exec rm -rf {} + 2>/dev/null || true
find "$sp_dir" -type d -name test -exec rm -rf {} + 2>/dev/null || true
find "$sp_dir" -type d -name "*.dist-info" -exec rm -rf {} + 2>/dev/null || true
find "$sp_dir" -name "*.pyi" -delete 2>/dev/null || true
# ── pip / setuptools (not needed at runtime) ─────────────
rm -rf "$sp_dir"/pip "$sp_dir"/pip-* 2>/dev/null || true
rm -rf "$sp_dir"/setuptools "$sp_dir"/setuptools-* "$sp_dir"/pkg_resources 2>/dev/null || true
rm -rf "$sp_dir"/_distutils_hack 2>/dev/null || true
# ── OpenCV ───────────────────────────────────────────────
local cv2_dir="$sp_dir/cv2"
if [ -d "$cv2_dir" ]; then
# Remove ffmpeg (28 MB on Windows), Haar cascades, dev files
rm -f "$cv2_dir"/opencv_videoio_ffmpeg*."$lib_suffix" 2>/dev/null || true
rm -rf "$cv2_dir/data" "$cv2_dir/gapi" "$cv2_dir/misc" "$cv2_dir/utils" 2>/dev/null || true
rm -rf "$cv2_dir/typing_stubs" "$cv2_dir/typing" 2>/dev/null || true
fi
# ── NumPy ────────────────────────────────────────────────
# Remove unused submodules (only core, fft, random are used)
for mod in polynomial linalg ma lib distutils f2py typing _pyinstaller; do
rm -rf "$sp_dir/numpy/$mod" 2>/dev/null || true
done
rm -rf "$sp_dir/numpy/tests" "$sp_dir/numpy/*/tests" 2>/dev/null || true
# ── Pillow (only used for system tray icon) ──────────────
rm -rf "$sp_dir/PIL/tests" 2>/dev/null || true
# Remove unused image format plugins (keep JPEG, PNG, ICO, BMP)
for plugin in Eps Gif Tiff Webp Psd Pcx Xbm Xpm Dds Ftex Gbr Grib \
Icns Im Imt Iptc McIrdas Mpo Msp Pcd Pixar Ppm Sgi \
Spider Sun Tga Wal Wmf; do
rm -f "$sp_dir/PIL/${plugin}ImagePlugin.py" 2>/dev/null || true
rm -f "$sp_dir/PIL/${plugin}ImagePlugin.pyc" 2>/dev/null || true
done
# ── zeroconf ─────────────────────────────────────────────
rm -rf "$sp_dir/zeroconf/_services" 2>/dev/null || true
# ── Strip debug symbols ──────────────────────────────────
if command -v strip &>/dev/null; then
echo " Stripping debug symbols from .$ext_suffix files..."
find "$sp_dir" -name "*.$ext_suffix" -exec strip --strip-debug {} \; 2>/dev/null || true
fi
# ── Remove .py source (keep .pyc bytecode) ───────────────
echo " Removing .py source from site-packages (keeping .pyc)..."
find "$sp_dir" -name "*.py" ! -name "__init__.py" -delete 2>/dev/null || true
# ── Remove wled_controller if pip-installed ───────────────
rm -rf "$sp_dir"/wled_controller* "$sp_dir"/wled*.dist-info 2>/dev/null || true
local cleaned_size
cleaned_size=$(du -sh "$sp_dir" | cut -f1)
echo " Site-packages after cleanup: $cleaned_size"
}
+283
View File
@@ -0,0 +1,283 @@
#!/usr/bin/env bash
#
# Shared build functions for LedGrab distribution packaging.
# Sourced by build-dist.sh (Linux) and build-dist-windows.sh (Windows).
#
# Expected variables set by the caller before sourcing:
# SCRIPT_DIR, BUILD_DIR, DIST_DIR, SERVER_DIR, APP_DIR
# ── Version detection ────────────────────────────────────────
detect_version() {
# Usage: detect_version [explicit_version]
local version="${1:-}"
if [ -z "$version" ]; then
version=$(git describe --tags --exact-match 2>/dev/null || true)
fi
if [ -z "$version" ]; then
version="${GITEA_REF_NAME:-${GITHUB_REF_NAME:-}}"
fi
if [ -z "$version" ]; then
version=$(grep -oP '^version\s*=\s*"\K[^"]+' "$SERVER_DIR/pyproject.toml" 2>/dev/null || echo "0.0.0")
fi
VERSION_CLEAN="${version#v}"
# Normalize non-PEP440 version labels (e.g. "dev", "nightly", "snapshot")
# to a valid PEP440 dev release. Without this, pip/setuptools rejects the
# pyproject.toml with: `project.version` must be pep440.
if ! [[ "$VERSION_CLEAN" =~ ^[0-9]+(\.[0-9]+)*((a|b|rc|\.dev|\.post)[0-9]+)*(\+[a-zA-Z0-9.]+)?$ ]]; then
echo " Warning: '$VERSION_CLEAN' is not PEP440-compliant, using 0.0.0.dev0"
VERSION_CLEAN="0.0.0.dev0"
fi
# Stamp the resolved version into pyproject.toml so that
# importlib.metadata reads the correct value at runtime.
sed -i "s/^version = .*/version = \"${VERSION_CLEAN}\"/" "$SERVER_DIR/pyproject.toml"
}
# ── Clean previous build ─────────────────────────────────────
clean_dist() {
if [ -d "$DIST_DIR" ]; then
echo " Cleaning previous build..."
rm -rf "$DIST_DIR"
fi
mkdir -p "$DIST_DIR"
}
# ── Build frontend ───────────────────────────────────────────
build_frontend() {
echo " Building frontend bundle..."
(cd "$SERVER_DIR" && npm ci --loglevel error && npm run build) 2>&1 | {
grep -v 'RemoteException' || true
}
}
# ── Copy application files ───────────────────────────────────
copy_app_files() {
echo " Copying application files..."
mkdir -p "$APP_DIR"
cp -r "$SERVER_DIR/src" "$APP_DIR/src"
cp -r "$SERVER_DIR/config" "$APP_DIR/config"
mkdir -p "$DIST_DIR/data" "$DIST_DIR/logs"
# Clean up source maps and __pycache__
find "$APP_DIR" -name "*.map" -delete 2>/dev/null || true
find "$APP_DIR" -type d -name __pycache__ -exec rm -rf {} + 2>/dev/null || true
# Patch the fallback version in the bundled __init__.py. Bundled installs
# strip ledgrab-*.dist-info from site-packages, so importlib.metadata
# falls back to this literal at runtime — and a stale literal is what
# silently shipped v0.4.2 reporting "0.3.0" in the WebUI.
local bundled_init="$APP_DIR/src/ledgrab/__init__.py"
if [ -f "$bundled_init" ] && [ -n "${VERSION_CLEAN:-}" ]; then
sed -i "s/_FALLBACK_VERSION = \"[^\"]*\"/_FALLBACK_VERSION = \"${VERSION_CLEAN}\"/" "$bundled_init"
echo " Patched _FALLBACK_VERSION -> ${VERSION_CLEAN}"
fi
}
# ── Site-packages cleanup ────────────────────────────────────
#
# Strips tests, type stubs, unused submodules, and debug symbols
# from the installed site-packages directory.
#
# Args:
# $1 — path to site-packages directory
# $2 — native extension suffix: "pyd" (Windows) or "so" (Linux)
# $3 — native lib suffix for OpenCV ffmpeg: "dll" or "so"
cleanup_site_packages() {
local sp_dir="$1"
local ext_suffix="${2:-so}"
local lib_suffix="${3:-so}"
echo " Cleaning up site-packages to reduce size..."
# ── Generic cleanup ──────────────────────────────────────
find "$sp_dir" -type d -name __pycache__ -exec rm -rf {} + 2>/dev/null || true
find "$sp_dir" -type d -name tests -exec rm -rf {} + 2>/dev/null || true
find "$sp_dir" -type d -name test -exec rm -rf {} + 2>/dev/null || true
find "$sp_dir" -type d -name "*.dist-info" -exec rm -rf {} + 2>/dev/null || true
find "$sp_dir" -name "*.pyi" -delete 2>/dev/null || true
# ── pip / setuptools (not needed at runtime) ─────────────
rm -rf "$sp_dir"/pip "$sp_dir"/pip-* 2>/dev/null || true
rm -rf "$sp_dir"/setuptools "$sp_dir"/setuptools-* "$sp_dir"/pkg_resources 2>/dev/null || true
rm -rf "$sp_dir"/_distutils_hack 2>/dev/null || true
# ── OpenCV ───────────────────────────────────────────────
local cv2_dir="$sp_dir/cv2"
if [ -d "$cv2_dir" ]; then
# Remove ffmpeg (28 MB on Windows), Haar cascades, dev files
rm -f "$cv2_dir"/opencv_videoio_ffmpeg*."$lib_suffix" 2>/dev/null || true
rm -rf "$cv2_dir/data" "$cv2_dir/gapi" "$cv2_dir/misc" "$cv2_dir/utils" 2>/dev/null || true
rm -rf "$cv2_dir/typing_stubs" "$cv2_dir/typing" 2>/dev/null || true
fi
# ── NumPy ────────────────────────────────────────────────
# Only strip modules that are safely unused by numpy's own import chain.
# DO NOT strip: lib, linalg, ma, matrixlib — numpy.__init__ imports them
# transitively (e.g. matrixlib → defmatrix → linalg), so removing any of
# these breaks `import numpy` itself, cascading into every downstream
# module. Learned the hard way in the v0.0.0.dev0 Windows build.
for mod in polynomial distutils f2py typing _pyinstaller; do
rm -rf "$sp_dir/numpy/$mod" 2>/dev/null || true
done
rm -rf "$sp_dir/numpy/tests" "$sp_dir/numpy/*/tests" 2>/dev/null || true
# ── Pillow (only used for system tray icon) ──────────────
rm -rf "$sp_dir/PIL/tests" 2>/dev/null || true
# Remove unused image format plugins (keep JPEG, PNG, ICO, BMP)
for plugin in Eps Gif Tiff Webp Psd Pcx Xbm Xpm Dds Ftex Gbr Grib \
Icns Im Imt Iptc McIrdas Mpo Msp Pcd Pixar Ppm Sgi \
Spider Sun Tga Wal Wmf; do
rm -f "$sp_dir/PIL/${plugin}ImagePlugin.py" 2>/dev/null || true
rm -f "$sp_dir/PIL/${plugin}ImagePlugin.pyc" 2>/dev/null || true
done
# ── zeroconf ─────────────────────────────────────────────
# DO NOT strip zeroconf/_services — the compiled Cython _listener.pyd
# imports from it, and the import fails at runtime with:
# ModuleNotFoundError: No module named 'zeroconf._services'
# Same class of bug as numpy — "presumed unused" submodule is actually
# imported internally by the package's own compiled code.
# ── Strip debug symbols ──────────────────────────────────
if command -v strip &>/dev/null; then
echo " Stripping debug symbols from .$ext_suffix files..."
find "$sp_dir" -name "*.$ext_suffix" -exec strip --strip-debug {} \; 2>/dev/null || true
fi
# ── Remove ledgrab if pip-installed ───────────────
rm -rf "$sp_dir"/ledgrab* "$sp_dir"/ledgrab*.dist-info 2>/dev/null || true
local cleaned_size
cleaned_size=$(du -sh "$sp_dir" | cut -f1)
echo " Site-packages after cleanup: $cleaned_size"
}
# ── Pre-compile .py → .pyc (keep sources) ────────────────────
#
# MUST run AFTER cleanup_site_packages. Speeds up first startup by
# producing .pyc alongside .py. We deliberately do NOT delete .py
# sources afterwards:
#
# 1. OpenCV's loader does literal file I/O on cv2/config.py (not
# an import) — stripping it breaks `import cv2` with:
# "OpenCV loader: missing configuration file: ['config.py']".
# 2. Other packages may do similar tricks (inspect.getsource,
# runtime introspection, __file__-relative data loading).
# 3. The size saving (~30%) isn't worth the whack-a-mole of
# shipping broken installers. We already hit this with
# numpy.linalg and zeroconf._services — enough incidents.
#
# Args:
# $1 — directory to compile (site-packages or app/src)
# $2 — python executable to use (default: python3)
compile_and_strip_sources() {
local target_dir="$1"
local py_cmd="${2:-python3}"
if [ ! -d "$target_dir" ]; then
return 0
fi
echo " Pre-compiling Python bytecode in $(basename "$target_dir")..."
"$py_cmd" -m compileall -b -q "$target_dir" 2>/dev/null || {
echo " ERROR: compileall failed for $target_dir — aborting"
return 1
}
# Drop __pycache__ to save the duplicated PEP-3147 copies; the
# `-b` flag above placed legacy .pyc next to each .py, so nothing
# of value is lost here.
find "$target_dir" -type d -name __pycache__ -exec rm -rf {} + 2>/dev/null || true
}
# ── Import smoke test ────────────────────────────────────────
#
# Verifies that every top-level dependency that ledgrab actually
# uses can be imported from the stripped site-packages. Catches regressions
# where cleanup_site_packages removes a submodule that turns out to be
# imported internally by the package (e.g. numpy.linalg, zeroconf._services).
# Failing here is cheap; failing on a user's machine after install is not.
#
# Args:
# $1 — path to site-packages to test against
# $2 — python executable
# $3 — (optional) extra PYTHONPATH entry (e.g. app/src for ledgrab)
smoke_test_imports() {
local sp_dir="$1"
local py_cmd="${2:-python3}"
local extra_path="${3:-}"
echo " Running import smoke test..."
local pypath="$sp_dir"
if [ -n "$extra_path" ]; then
pypath="$extra_path:$sp_dir"
fi
# Modules that MUST import cleanly IF PRESENT. We don't enforce
# installation — Pillow for example is only a Windows dep. But if a
# module's top-level package dir exists in site-packages and we
# can't import it, that's a broken install and we abort.
local smoke_script
smoke_script=$(cat <<'PYEOF'
import importlib
import os
import sys
sp_dir = sys.argv[1]
# (module_name, site-packages path to check for presence)
candidates = [
('numpy', 'numpy'),
('numpy.linalg', 'numpy/linalg'),
('numpy.lib', 'numpy/lib'),
('numpy.matrixlib', 'numpy/matrixlib'),
('cv2', 'cv2'),
('fastapi', 'fastapi'),
('uvicorn', 'uvicorn'),
('starlette', 'starlette'),
('pydantic', 'pydantic'),
('zeroconf', 'zeroconf'),
('zeroconf._services', 'zeroconf/_services'),
('PIL', 'PIL'),
('PIL.Image', 'PIL'),
('yaml', 'yaml'),
]
tested = 0
skipped = 0
failed = []
for mod, path in candidates:
if not os.path.exists(os.path.join(sp_dir, path)):
skipped += 1
continue
try:
importlib.import_module(mod)
tested += 1
except Exception as e:
failed.append(f'{mod}: {type(e).__name__}: {e}')
if failed:
print('SMOKE TEST FAILED:', file=sys.stderr)
for f in failed:
print(f' {f}', file=sys.stderr)
sys.exit(1)
print(f' Smoke test passed ({tested} imported, {skipped} not installed)')
PYEOF
)
if ! PYTHONPATH="$pypath" "$py_cmd" -c "$smoke_script" "$sp_dir"; then
echo " ERROR: smoke test failed — site-packages is broken, aborting build"
return 1
fi
}
@@ -14,10 +14,11 @@
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
BUILD_DIR="$SCRIPT_DIR/build"
REPO_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
BUILD_DIR="$SCRIPT_DIR"
DIST_NAME="LedGrab"
DIST_DIR="$BUILD_DIR/$DIST_NAME"
SERVER_DIR="$SCRIPT_DIR/server"
SERVER_DIR="$REPO_ROOT/server"
PYTHON_DIR="$DIST_DIR/python"
APP_DIR="$DIST_DIR/app"
PYTHON_VERSION="${PYTHON_VERSION:-3.11.9}"
@@ -66,7 +67,7 @@ if ! grep -q 'Lib\\site-packages' "$PTH_FILE"; then
echo 'Lib\site-packages' >> "$PTH_FILE"
fi
# Embedded Python ._pth overrides PYTHONPATH, so we must add the app
# source directory here for wled_controller to be importable
# source directory here for ledgrab to be importable
if ! grep -q '\.\./app/src' "$PTH_FILE"; then
echo '../app/src' >> "$PTH_FILE"
fi
@@ -177,10 +178,16 @@ echo "[6/9] Downloading Windows dependencies..."
WHEEL_DIR="$BUILD_DIR/win-wheels"
mkdir -p "$WHEEL_DIR"
# Core dependencies (cross-platform, should have win_amd64 wheels)
# Core dependencies (cross-platform, should have win_amd64 wheels).
# KEEP IN SYNC with server/pyproject.toml [project.dependencies] — this
# list duplicates it because cross-build on Linux can't invoke `pip install
# <path>` against pyproject.toml with a Windows target. Missing entries
# ship a broken installer that silently fails under pythonw.exe (no
# traceback visible to the user). Audit after every pyproject.toml edit.
DEPS=(
"fastapi>=0.115.0"
"uvicorn[standard]>=0.32.0"
"cryptography>=42.0.0"
"httpx>=0.27.2"
"mss>=9.0.2"
"numpy>=2.1.3"
@@ -200,6 +207,7 @@ DEPS=(
"aiomqtt>=2.0.0"
"openrgb-python>=0.2.15"
"opencv-python-headless>=4.8.0"
"just-playback>=0.1.7"
)
# Windows-only deps
@@ -213,6 +221,10 @@ WIN_DEPS=(
# System tray (Pillow needed by pystray for tray icon)
"pystray>=0.19.0"
"Pillow>=10.4.0"
# Windows screen capture engines (mss is the only fallback without these)
"dxcam>=0.0.5"
"bettercam>=1.0.0"
"windows-capture>=1.5.0"
)
# Download cross-platform deps (prefer binary, allow source for pure Python)
@@ -273,6 +285,28 @@ rm -rf "$SITE_PACKAGES"/pythonwin 2>/dev/null || true
rm -f "$SITE_PACKAGES"/PyWin32.chm 2>/dev/null || true
find "$SITE_PACKAGES/winrt" -name "*.pyi" -delete 2>/dev/null || true
# Pre-compile and strip .py sources. MUST run AFTER cleanup (so we don't
# waste work compiling files about to be deleted). Uses host python —
# PYTHON_VERSION above must match the embedded Python major.minor or
# the generated .pyc will ImportError on the target.
compile_and_strip_sources "$SITE_PACKAGES" "python"
# Windows cross-build: host python can't load win_amd64 .pyd files, so
# we can't `import numpy` for real. Instead, check that the submodules
# known to be imported internally exist on disk — the same landmines that
# cost users a broken v0.0.0.dev0 installer.
echo " Verifying required submodules exist after cleanup..."
for required in \
"numpy/linalg" "numpy/lib" "numpy/matrixlib" "numpy/ma" \
"zeroconf/_services" \
"cryptography" "cffi" "just_playback"; do
if [ ! -d "$SITE_PACKAGES/$required" ] && [ ! -f "$SITE_PACKAGES/$required.py" ] && [ ! -f "$SITE_PACKAGES/$required.pyc" ]; then
echo " ERROR: $required missing from site-packages — either cleanup_site_packages removed something required, or DEPS is out of sync with pyproject.toml. Aborting."
exit 1
fi
done
echo " All required submodules present."
WHEEL_COUNT=$(ls "$WHEEL_DIR"/*.whl 2>/dev/null | wc -l)
echo " Installed $WHEEL_COUNT packages"
@@ -286,10 +320,9 @@ build_frontend
echo "[8/9] Copying application files..."
copy_app_files
# Pre-compile Python bytecode for faster startup
echo " Pre-compiling Python bytecode..."
# Pre-compile app source for faster startup (keep .py too — app source
# is small and easier to debug in-place if a user reports an issue)
python -m compileall -b -q "$APP_DIR/src" 2>/dev/null || true
python -m compileall -b -q "$SITE_PACKAGES" 2>/dev/null || true
# ── Create launcher ──────────────────────────────────────────
@@ -301,14 +334,20 @@ cd /d "%~dp0"
:: Set paths
set PYTHONPATH=%~dp0app\src
set WLED_CONFIG_PATH=%~dp0app\config\default_config.yaml
set LEDGRAB_CONFIG_PATH=%~dp0app\config\default_config.yaml
:: Tcl/Tk ship under python\ but Tk's default search path is
:: <python.exe>\..\lib\tcl8.6. Point it at the right location so
:: the screen-overlay feature (tkinter) can start without errors.
set TCL_LIBRARY=%~dp0python\tcl8.6
set TK_LIBRARY=%~dp0python\tk8.6
:: Create data directory if missing
if not exist "%~dp0data" mkdir "%~dp0data"
if not exist "%~dp0logs" mkdir "%~dp0logs"
:: Start the server (tray icon handles UI and exit)
"%~dp0python\pythonw.exe" -m wled_controller
"%~dp0python\pythonw.exe" -m ledgrab
LAUNCHER
# Convert launcher to Windows line endings
@@ -316,7 +355,7 @@ sed -i 's/$/\r/' "$DIST_DIR/LedGrab.bat"
# Copy hidden launcher VBS
mkdir -p "$DIST_DIR/scripts"
cp server/scripts/start-hidden.vbs "$DIST_DIR/scripts/"
cp "$SERVER_DIR/scripts/start-hidden.vbs" "$DIST_DIR/scripts/"
# ── Create autostart scripts ─────────────────────────────────
+28 -11
View File
@@ -34,11 +34,11 @@ param(
$ErrorActionPreference = 'Stop'
$ProgressPreference = 'SilentlyContinue' # faster downloads
$ScriptRoot = $PSScriptRoot
$BuildDir = Join-Path $ScriptRoot "build"
$RepoRoot = Split-Path $PSScriptRoot -Parent
$BuildDir = $PSScriptRoot
$DistName = "LedGrab"
$DistDir = Join-Path $BuildDir $DistName
$ServerDir = Join-Path $ScriptRoot "server"
$ServerDir = Join-Path $RepoRoot "server"
$PythonDir = Join-Path $DistDir "python"
$AppDir = Join-Path $DistDir "app"
@@ -58,7 +58,7 @@ if (-not $Version) {
}
if (-not $Version) {
# Parse from __init__.py
$initFile = Join-Path $ServerDir "src\wled_controller\__init__.py"
$initFile = Join-Path $ServerDir "src\ledgrab\__init__.py"
$match = Select-String -Path $initFile -Pattern '__version__\s*=\s*"([^"]+)"'
if ($match) { $Version = $match.Matches[0].Groups[1].Value }
}
@@ -107,7 +107,7 @@ if ($pthContent -notmatch 'Lib\\site-packages') {
$pthContent = $pthContent.TrimEnd() + "`nLib\site-packages`n"
}
# Embedded Python ._pth overrides PYTHONPATH, so add the app source path
# directly for wled_controller to be importable
# directly for ledgrab to be importable
if ($pthContent -notmatch '\.\.[/\\]app[/\\]src') {
$pthContent = $pthContent.TrimEnd() + "`n..\app\src`n"
}
@@ -130,7 +130,7 @@ if ($LASTEXITCODE -ne 0) { throw "Failed to install pip" }
# ── Install dependencies ──────────────────────────────────────
Write-Host "[5/8] Installing dependencies..."
$extras = "camera,notifications,tray"
$extras = "camera,notifications"
if (-not $SkipPerf) { $extras += ",perf" }
# Install the project (pulls all deps via pyproject.toml), then remove
@@ -144,10 +144,10 @@ if ($LASTEXITCODE -ne 0) {
Write-Host " Some optional deps may have failed (continuing)..." -ForegroundColor Yellow
}
# Remove the installed wled_controller package to avoid duplication
# Remove the installed ledgrab package to avoid duplication
$sitePackages = Join-Path $PythonDir "Lib\site-packages"
Get-ChildItem -Path $sitePackages -Filter "wled*" -Directory | Remove-Item -Recurse -Force -ErrorAction SilentlyContinue
Get-ChildItem -Path $sitePackages -Filter "wled*.dist-info" -Directory | Remove-Item -Recurse -Force -ErrorAction SilentlyContinue
Get-ChildItem -Path $sitePackages -Filter "ledgrab*" -Directory | Remove-Item -Recurse -Force -ErrorAction SilentlyContinue
Get-ChildItem -Path $sitePackages -Filter "ledgrab*.dist-info" -Directory | Remove-Item -Recurse -Force -ErrorAction SilentlyContinue
# Clean up caches and test files to reduce size
Write-Host " Cleaning up caches..."
@@ -196,6 +196,17 @@ New-Item -ItemType Directory -Path (Join-Path $DistDir "logs") -Force | Out-Null
Get-ChildItem -Path $srcDest -Recurse -Filter "*.map" | Remove-Item -Force -ErrorAction SilentlyContinue
Get-ChildItem -Path $srcDest -Recurse -Directory -Filter "__pycache__" | Remove-Item -Recurse -Force -ErrorAction SilentlyContinue
# Patch the fallback version in the bundled __init__.py so the WebUI always
# reports the release version — the installer strips ledgrab-*.dist-info from
# site-packages (above), so importlib.metadata falls back to this literal.
$bundledInit = Join-Path $srcDest "ledgrab\__init__.py"
if (Test-Path $bundledInit) {
$initContent = Get-Content $bundledInit -Raw
$patched = [regex]::Replace($initContent, '_FALLBACK_VERSION\s*=\s*"[^"]*"', "_FALLBACK_VERSION = `"$VersionClean`"")
Set-Content -Path $bundledInit -Value $patched -NoNewline
Write-Host " Patched _FALLBACK_VERSION -> $VersionClean"
}
# ── Create launcher ────────────────────────────────────────────
Write-Host "[8/8] Creating launcher..."
@@ -206,14 +217,20 @@ cd /d "%~dp0"
:: Set paths
set PYTHONPATH=%~dp0app\src
set WLED_CONFIG_PATH=%~dp0app\config\default_config.yaml
set LEDGRAB_CONFIG_PATH=%~dp0app\config\default_config.yaml
:: Tcl/Tk ship under python\ but Tk's default search path is
:: <python.exe>\..\lib\tcl8.6. Point it at the right location so
:: the screen-overlay feature (tkinter) can start without errors.
set TCL_LIBRARY=%~dp0python\tcl8.6
set TK_LIBRARY=%~dp0python\tk8.6
:: Create data directory if missing
if not exist "%~dp0data" mkdir "%~dp0data"
if not exist "%~dp0logs" mkdir "%~dp0logs"
:: Start the server (tray icon handles UI and exit)
"%~dp0python\pythonw.exe" -m wled_controller
"%~dp0python\pythonw.exe" -m ledgrab
'@
$launcherContent = $launcherContent -replace '%VERSION%', $VersionClean
+15 -4
View File
@@ -10,10 +10,11 @@
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
BUILD_DIR="$SCRIPT_DIR/build"
REPO_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
BUILD_DIR="$SCRIPT_DIR"
DIST_NAME="LedGrab"
DIST_DIR="$BUILD_DIR/$DIST_NAME"
SERVER_DIR="$SCRIPT_DIR/server"
SERVER_DIR="$REPO_ROOT/server"
VENV_DIR="$DIST_DIR/venv"
APP_DIR="$DIST_DIR/app"
@@ -53,6 +54,12 @@ SITE_PACKAGES=$(echo "$VENV_DIR"/lib/python*/site-packages)
# Clean up with shared function
cleanup_site_packages "$SITE_PACKAGES" "so" "so"
# Pre-compile and strip .py sources (must happen AFTER cleanup)
compile_and_strip_sources "$SITE_PACKAGES" "python"
# Fail loud if cleanup broke any required import
smoke_test_imports "$SITE_PACKAGES" "python"
# ── Build frontend ───────────────────────────────────────────
echo "[4/7] Building frontend..."
@@ -63,6 +70,10 @@ build_frontend
echo "[5/7] Copying application files..."
copy_app_files
# Pre-compile app source for faster startup (keep .py too — app source
# is small and easier to debug in-place if a user reports an issue)
python -m compileall -b -q "$APP_DIR/src" 2>/dev/null || true
# ── Create launcher ──────────────────────────────────────────
echo "[6/7] Creating launcher..."
@@ -73,12 +84,12 @@ set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
export PYTHONPATH="$SCRIPT_DIR/app/src"
export WLED_CONFIG_PATH="$SCRIPT_DIR/app/config/default_config.yaml"
export LEDGRAB_CONFIG_PATH="$SCRIPT_DIR/app/config/default_config.yaml"
mkdir -p "$SCRIPT_DIR/data" "$SCRIPT_DIR/logs"
source "$SCRIPT_DIR/venv/bin/activate"
exec python -m wled_controller.main
exec python -m ledgrab.main
LAUNCHER
sed -i "s/VERSION_PLACEHOLDER/${VERSION_CLEAN}/" "$DIST_DIR/run.sh"
+327
View File
@@ -0,0 +1,327 @@
"""Generate LedGrab app icon assets.
Concept: "Spectrum Aperture" — a rounded-square frame (the screen/display)
traced by a continuous RGB color-wheel stroke (the bias-light LED strip),
on a near-black canvas with a soft chromatic bloom behind it.
Outputs:
server/src/ledgrab/static/icons/icon-512.png (standard, opaque vignette bg)
server/src/ledgrab/static/icons/icon-192.png (downscale of 512)
server/src/ledgrab/static/icons/icon-512-maskable.png (safe-area padded, opaque)
server/src/ledgrab/static/icons/icon-tray.png (256, transparent bg, frame + glow)
server/src/ledgrab/static/icons/icon.ico (16/24/32/48/64/128/256)
Run from repo root:
py -3.13 build/generate_icon.py
"""
from __future__ import annotations
import colorsys
import math
from pathlib import Path
from PIL import Image, ImageDraw, ImageFilter
# ── Tunables ────────────────────────────────────────────────────────────
SUPERSAMPLE = 4 # render at 4x and downsample for crispness
BASE = 1024 # logical canvas size
HQ = BASE * SUPERSAMPLE # render canvas
BG_TOP = (12, 14, 22) # near-black, faint cool tint
BG_BOTTOM = (6, 7, 12) # darker at edges (vignette feel)
FRAME_INSET = 0.18 # margin from canvas edge to frame (fraction)
FRAME_RADIUS = 0.22 # corner radius (fraction of frame side)
FRAME_STROKE = 0.085 # stroke width (fraction of canvas)
BLOOM_OPACITY = 0.62 # outer bloom strength (01)
INNER_GLOW_OPACITY = 0.38 # inner chromatic reflection strength
# Hue rotation offset so red sits at the top
HUE_OFFSET = -90.0 # degrees (negative = counter-clockwise shift)
def lerp(a: float, b: float, t: float) -> float:
return a + (b - a) * t
def hue_to_rgb(hue_deg: float) -> tuple[int, int, int]:
"""Bright, slightly desaturated spectral color (LED-like)."""
h = (hue_deg % 360) / 360.0
r, g, b = colorsys.hls_to_rgb(h, 0.58, 0.92)
return int(r * 255), int(g * 255), int(b * 255)
def vignette_background(size: int) -> Image.Image:
"""Dark canvas with a soft radial vignette + faint scanline noise."""
img = Image.new("RGB", (size, size), BG_TOP)
px = img.load()
cx, cy = size / 2, size / 2
max_r = math.hypot(cx, cy)
for y in range(size):
for x in range(size):
d = math.hypot(x - cx, y - cy) / max_r
t = min(1.0, d**1.6)
px[x, y] = (
int(lerp(BG_TOP[0], BG_BOTTOM[0], t)),
int(lerp(BG_TOP[1], BG_BOTTOM[1], t)),
int(lerp(BG_TOP[2], BG_BOTTOM[2], t)),
)
return img
def draw_chromatic_bloom(size: int) -> Image.Image:
"""Soft, large chromatic glow behind the frame — the bias-light effect."""
layer = Image.new("RGBA", (size, size), (0, 0, 0, 0))
draw = ImageDraw.Draw(layer)
cx, cy = size / 2, size / 2
radius = size * 0.36
blob_r = int(size * 0.30)
n_blobs = 24
for i in range(n_blobs):
a = i / n_blobs * 360.0
bx = cx + math.cos(math.radians(a - 90)) * radius
by = cy + math.sin(math.radians(a - 90)) * radius
r, g, b = hue_to_rgb(a + HUE_OFFSET)
alpha = int(255 * BLOOM_OPACITY * 0.55)
draw.ellipse(
(bx - blob_r, by - blob_r, bx + blob_r, by + blob_r),
fill=(r, g, b, alpha),
)
# Heavy blur → continuous, dreamy halo
layer = layer.filter(ImageFilter.GaussianBlur(radius=size * 0.10))
return layer
def rounded_rect_mask(size: int, inset: int, radius: int, stroke: int) -> Image.Image:
"""L-mode mask of a rounded-rect ring (the frame stroke region)."""
mask = Image.new("L", (size, size), 0)
draw = ImageDraw.Draw(mask)
box_outer = (inset, inset, size - inset, size - inset)
box_inner = (
inset + stroke,
inset + stroke,
size - inset - stroke,
size - inset - stroke,
)
r_outer = radius
r_inner = max(0, radius - stroke)
draw.rounded_rectangle(box_outer, radius=r_outer, fill=255)
draw.rounded_rectangle(box_inner, radius=r_inner, fill=0)
return mask
def draw_spectrum_frame(size: int) -> Image.Image:
"""Draw the rounded-square frame stroke filled with a hue-rotation gradient.
Strategy: paint a full-canvas angular hue gradient (centered), then
clip it with the rounded-ring mask. This guarantees a continuous,
seam-free color flow around the entire frame.
"""
cx, cy = size / 2, size / 2
gradient = Image.new("RGB", (size, size), (0, 0, 0))
gpx = gradient.load()
for y in range(size):
dy = y - cy
for x in range(size):
dx = x - cx
ang = math.degrees(math.atan2(dy, dx)) + 90.0 # 0° = top
r, g, b = hue_to_rgb(ang + HUE_OFFSET)
gpx[x, y] = (r, g, b)
inset = int(size * FRAME_INSET)
frame_side = size - 2 * inset
stroke = int(size * FRAME_STROKE)
radius = int(frame_side * FRAME_RADIUS)
mask = rounded_rect_mask(size, inset, radius, stroke)
out = Image.new("RGBA", (size, size), (0, 0, 0, 0))
out.paste(gradient, (0, 0), mask)
return out
def draw_inner_screen(size: int) -> Image.Image:
"""Subtle dark rounded square inside the frame, with faint chromatic
inner reflection along the edges — like a screen catching ambient light."""
inset = int(size * FRAME_INSET)
stroke = int(size * FRAME_STROKE)
frame_side = size - 2 * inset
radius = int(frame_side * FRAME_RADIUS)
pad = int(stroke * 0.35)
box = (
inset + stroke + pad,
inset + stroke + pad,
size - inset - stroke - pad,
size - inset - stroke - pad,
)
r_inner = max(0, radius - stroke - pad)
layer = Image.new("RGBA", (size, size), (0, 0, 0, 0))
draw = ImageDraw.Draw(layer)
# Dark fill, very slight cool tint
draw.rounded_rectangle(box, radius=r_inner, fill=(10, 12, 18, 255))
# Inner chromatic glow: same spectrum, very soft, clipped to the screen
bloom = draw_chromatic_bloom(size)
screen_mask = Image.new("L", (size, size), 0)
ImageDraw.Draw(screen_mask).rounded_rectangle(box, radius=r_inner, fill=255)
bloom_alpha = bloom.split()[-1].point(lambda v: int(v * INNER_GLOW_OPACITY))
bloom.putalpha(bloom_alpha)
masked_bloom = Image.new("RGBA", (size, size), (0, 0, 0, 0))
masked_bloom.paste(bloom, (0, 0), screen_mask)
layer.alpha_composite(masked_bloom)
# Faint highlight glint top-left
glint = Image.new("RGBA", (size, size), (0, 0, 0, 0))
gdraw = ImageDraw.Draw(glint)
glint_box = (
box[0] + int(frame_side * 0.04),
box[1] + int(frame_side * 0.04),
box[0] + int(frame_side * 0.42),
box[1] + int(frame_side * 0.18),
)
gdraw.rounded_rectangle(glint_box, radius=int(frame_side * 0.05), fill=(255, 255, 255, 22))
glint = glint.filter(ImageFilter.GaussianBlur(radius=size * 0.012))
masked_glint = Image.new("RGBA", (size, size), (0, 0, 0, 0))
masked_glint.paste(glint, (0, 0), screen_mask)
layer.alpha_composite(masked_glint)
return layer
def add_outer_frame_glow(frame_rgba: Image.Image) -> Image.Image:
"""Take the spectrum frame and produce a blurred, brightened copy for glow."""
glow = frame_rgba.copy()
# Slightly inflate brightness for glow
r, g, b, a = glow.split()
glow = Image.merge("RGBA", (r, g, b, a.point(lambda v: min(255, int(v * 0.85)))))
glow = glow.filter(ImageFilter.GaussianBlur(radius=glow.width * 0.025))
return glow
def render_tray(size: int) -> Image.Image:
"""Render a tray-optimised icon: transparent background, bolder frame,
tight outer glow. Designed to read clearly at 1632 px on top of any
taskbar color."""
hq = size * SUPERSAMPLE
# Pull the frame inward a touch and beef up the stroke so it reads at 16 px.
global FRAME_INSET, FRAME_STROKE
saved_inset, saved_stroke = FRAME_INSET, FRAME_STROKE
FRAME_INSET = 0.13
FRAME_STROKE = 0.115
try:
frame = draw_spectrum_frame(hq)
finally:
FRAME_INSET, FRAME_STROKE = saved_inset, saved_stroke
# Tight, bright glow that doesn't bleed past the tray cell.
glow = frame.copy()
r, g, b, a = glow.split()
glow = Image.merge("RGBA", (r, g, b, a.point(lambda v: min(255, int(v * 0.95)))))
glow = glow.filter(ImageFilter.GaussianBlur(radius=hq * 0.012))
canvas = Image.new("RGBA", (hq, hq), (0, 0, 0, 0))
canvas.alpha_composite(glow)
canvas.alpha_composite(frame)
return canvas.resize((size, size), Image.LANCZOS)
def render(size: int, *, maskable: bool = False) -> Image.Image:
"""Render the full icon at the given size."""
hq = size * SUPERSAMPLE
if maskable:
# Maskable: pad inward so the entire icon survives a circular crop.
# We render the standard composition at 80% of canvas size, centered.
bg = Image.new("RGB", (hq, hq), BG_BOTTOM).convert("RGBA")
bg.paste(vignette_background(hq), (0, 0))
inner = render(size, maskable=False).resize((int(hq * 0.78), int(hq * 0.78)), Image.LANCZOS)
# Strip the bg from the inner render: composite the spectrum
# parts on top of our maskable background.
ox = (hq - inner.width) // 2
oy = (hq - inner.height) // 2
bg.alpha_composite(inner, (ox, oy))
return bg.resize((size, size), Image.LANCZOS)
bg = vignette_background(hq).convert("RGBA")
bloom = draw_chromatic_bloom(hq)
frame = draw_spectrum_frame(hq)
frame_glow = add_outer_frame_glow(frame)
inner_screen = draw_inner_screen(hq)
# Composite order: bg → bloom → frame_glow → inner_screen → frame
canvas = Image.new("RGBA", (hq, hq), (0, 0, 0, 0))
canvas.alpha_composite(bg)
canvas.alpha_composite(bloom)
canvas.alpha_composite(frame_glow)
canvas.alpha_composite(inner_screen)
canvas.alpha_composite(frame)
return canvas.resize((size, size), Image.LANCZOS)
def main() -> None:
repo_root = Path(__file__).resolve().parent.parent
targets = [
repo_root / "server" / "src" / "ledgrab" / "static" / "icons",
repo_root
/ "android"
/ "app"
/ "build"
/ "python"
/ "sources"
/ "debug"
/ "ledgrab"
/ "static"
/ "icons",
]
print("Rendering 1024 master...")
master = render(1024, maskable=False)
print("Rendering maskable 1024 master...")
maskable_master = render(1024, maskable=True)
print("Rendering tray 512 master (transparent bg)...")
tray_master = render_tray(512)
for icons_dir in targets:
if not icons_dir.exists():
print(f" skip (missing): {icons_dir}")
continue
out_512 = icons_dir / "icon-512.png"
out_192 = icons_dir / "icon-192.png"
out_mask = icons_dir / "icon-512-maskable.png"
out_tray = icons_dir / "icon-tray.png"
out_ico = icons_dir / "icon.ico"
master.resize((512, 512), Image.LANCZOS).save(out_512, "PNG", optimize=True)
master.resize((192, 192), Image.LANCZOS).save(out_192, "PNG", optimize=True)
maskable_master.resize((512, 512), Image.LANCZOS).save(out_mask, "PNG", optimize=True)
tray_master.save(out_tray, "PNG", optimize=True)
# Pre-resize each frame from the 1024 master for maximum crispness.
# Pass them via the `sizes` arg so Pillow embeds every variant.
ico_sizes = [(16, 16), (24, 24), (32, 32), (48, 48), (64, 64), (128, 128), (256, 256)]
# Use the tray (transparent-bg) variant for ICO frames so the file/
# taskbar icon doesn't show a dark tile against light backgrounds.
ico_source = tray_master.resize((256, 256), Image.LANCZOS)
ico_source.save(out_ico, format="ICO", sizes=ico_sizes)
print(f" wrote: {icons_dir}")
if __name__ == "__main__":
main()
+40 -12
View File
@@ -22,7 +22,7 @@
!endif
Name "${APPNAME} v${VERSION}"
OutFile "build\${APPNAME}-v${VERSION}-win-x64-setup.exe"
OutFile "${APPNAME}-v${VERSION}-win-x64-setup.exe"
InstallDir "$LOCALAPPDATA\${APPNAME}"
InstallDirRegKey HKCU "Software\${APPNAME}" "InstallDir"
RequestExecutionLevel user
@@ -30,6 +30,8 @@ SetCompressor /SOLID lzma
; Modern UI Configuration
!define MUI_ICON "..\server\src\ledgrab\static\icons\icon.ico"
!define MUI_UNICON "..\server\src\ledgrab\static\icons\icon.ico"
!define MUI_ABORTWARNING
; Pages
@@ -54,9 +56,10 @@ SetCompressor /SOLID lzma
; Functions
Function LaunchApp
; Only launch the app do NOT open the browser here. A manual launch (no
; --autostart) makes the app open the WebUI itself once /health responds,
; so opening the URL here too made the page appear twice.
ExecShell "open" "wscript.exe" '"$INSTDIR\scripts\${VBSNAME}"'
Sleep 2000
ExecShell "open" "http://localhost:8080/"
FunctionEnd
; Detect running instance before install (file lock check on python.exe)
@@ -85,11 +88,29 @@ Section "!${APPNAME} (required)" SecCore
SetOutPath "$INSTDIR"
; Wipe prior payload dirs before extracting. NSIS File /r MERGES files
; on top of existing ones on an upgrade, stale .pyc/.pyd from the old
; version (and any files removed or renamed since) would survive,
; producing a half-old/half-new install that presents as "version
; mismatch" or "duplicate package" ImportErrors at runtime.
; IMPORTANT: only touch payload dirs never $INSTDIR\data or
; $INSTDIR\logs (user config must be preserved across upgrades).
RMDir /r "$INSTDIR\python"
RMDir /r "$INSTDIR\app"
RMDir /r "$INSTDIR\scripts"
Delete "$INSTDIR\LedGrab.bat"
; Legacy leftovers from the wled_controller-era install. The current
; build does not ship debug.bat, but upgrades from older versions left
; one behind with a stale `-m wled_controller` command that gives a
; misleading ModuleNotFoundError when run. Remove it on upgrade.
Delete "$INSTDIR\debug.bat"
Delete "$INSTDIR\debug.log"
; Copy the entire portable build
File /r "build\LedGrab\python"
File /r "build\LedGrab\app"
File /r "build\LedGrab\scripts"
File "build\LedGrab\LedGrab.bat"
File /r "LedGrab\python"
File /r "LedGrab\app"
File /r "LedGrab\scripts"
File "LedGrab\LedGrab.bat"
; Create data and logs directories
CreateDirectory "$INSTDIR\data"
@@ -102,7 +123,7 @@ Section "!${APPNAME} (required)" SecCore
CreateDirectory "$SMPROGRAMS\${APPNAME}"
CreateShortcut "$SMPROGRAMS\${APPNAME}\${APPNAME}.lnk" \
"wscript.exe" '"$INSTDIR\scripts\${VBSNAME}"' \
"$INSTDIR\python\pythonw.exe" 0
"$INSTDIR\app\src\ledgrab\static\icons\icon.ico" 0
CreateShortcut "$SMPROGRAMS\${APPNAME}\Uninstall.lnk" "$INSTDIR\uninstall.exe"
; Registry: install location + Add/Remove Programs entry
@@ -117,10 +138,12 @@ Section "!${APPNAME} (required)" SecCore
"UninstallString" '"$INSTDIR\uninstall.exe"'
WriteRegStr HKCU "Software\Microsoft\Windows\CurrentVersion\Uninstall\${APPNAME}" \
"InstallLocation" "$INSTDIR"
WriteRegStr HKCU "Software\Microsoft\Windows\CurrentVersion\Uninstall\${APPNAME}" \
"DisplayIcon" "$INSTDIR\app\src\ledgrab\static\icons\icon.ico"
WriteRegStr HKCU "Software\Microsoft\Windows\CurrentVersion\Uninstall\${APPNAME}" \
"Publisher" "Alexei Dolgolyov"
WriteRegStr HKCU "Software\Microsoft\Windows\CurrentVersion\Uninstall\${APPNAME}" \
"URLInfoAbout" "https://git.dolgolyov-family.by/alexei.dolgolyov/wled-screen-controller-mixed"
"URLInfoAbout" "https://git.dolgolyov-family.by/alexei.dolgolyov/ledgrab"
WriteRegDWORD HKCU "Software\Microsoft\Windows\CurrentVersion\Uninstall\${APPNAME}" \
"NoModify" 1
WriteRegDWORD HKCU "Software\Microsoft\Windows\CurrentVersion\Uninstall\${APPNAME}" \
@@ -136,13 +159,16 @@ SectionEnd
Section "Desktop shortcut" SecDesktop
CreateShortcut "$DESKTOP\${APPNAME}.lnk" \
"wscript.exe" '"$INSTDIR\scripts\${VBSNAME}"' \
"$INSTDIR\python\pythonw.exe" 0
"$INSTDIR\app\src\ledgrab\static\icons\icon.ico" 0
SectionEnd
Section "Start with Windows" SecAutostart
; Pass --autostart so the VBS sets LEDGRAB_AUTOSTART=1 and the app suppresses
; the browser auto-open on Windows login. Manual launches (desktop / start
; menu) don't pass the arg, so they keep opening the WebUI tab.
CreateShortcut "$SMSTARTUP\${APPNAME}.lnk" \
"wscript.exe" '"$INSTDIR\scripts\${VBSNAME}"' \
"$INSTDIR\python\pythonw.exe" 0
"wscript.exe" '"$INSTDIR\scripts\${VBSNAME}" --autostart' \
"$INSTDIR\app\src\ledgrab\static\icons\icon.ico" 0
SectionEnd
; Section Descriptions
@@ -171,6 +197,8 @@ Section "Uninstall"
RMDir /r "$INSTDIR\app"
RMDir /r "$INSTDIR\scripts"
Delete "$INSTDIR\LedGrab.bat"
Delete "$INSTDIR\debug.bat"
Delete "$INSTDIR\debug.log"
Delete "$INSTDIR\uninstall.exe"
; Remove logs (but keep data/)
-143
View File
@@ -1,143 +0,0 @@
# Auto-Update Plan — Phase 1: Check & Notify
> Created: 2026-03-25. Status: **planned, not started.**
## Backend Architecture
### Release Provider Abstraction
```
core/update/
release_provider.py — ABC: get_releases(), get_releases_page_url()
gitea_provider.py — Gitea REST API implementation
version_check.py — normalize_version(), is_newer() using packaging.version
update_service.py — Background asyncio task + state machine
```
**`ReleaseProvider` interface** — two methods:
- `get_releases(limit) → list[ReleaseInfo]` — fetch releases (newest first)
- `get_releases_page_url() → str` — link for "view on web"
**`GiteaReleaseProvider`** calls `GET {base_url}/api/v1/repos/{repo}/releases`. Swapping to GitHub later means implementing the same interface against `api.github.com`.
**Data models:**
```python
@dataclass(frozen=True)
class AssetInfo:
name: str # "LedGrab-v0.3.0-win-x64.zip"
size: int # bytes
download_url: str
@dataclass(frozen=True)
class ReleaseInfo:
tag: str # "v0.3.0"
version: str # "0.3.0"
name: str # "LedGrab v0.3.0"
body: str # release notes markdown
prerelease: bool
published_at: str # ISO 8601
assets: tuple[AssetInfo, ...]
```
### Version Comparison
`version_check.py` — normalize Gitea tags to PEP 440:
- `v0.3.0-alpha.1``0.3.0a1`
- `v0.3.0-beta.2``0.3.0b2`
- `v0.3.0-rc.3``0.3.0rc3`
Uses `packaging.version.Version` for comparison.
### Update Service
Follows the **AutoBackupEngine pattern**:
- Settings in `Database.get_setting("auto_update")`
- asyncio.Task for periodic checks
- 30s startup delay (avoid slowing boot)
- 60s debounce on manual checks
**State machine (Phase 1):** `IDLE → CHECKING → UPDATE_AVAILABLE`
No download/apply in Phase 1 — just detection and notification.
**Settings:** `enabled` (bool), `check_interval_hours` (float), `channel` ("stable" | "pre-release")
**Persisted state:** `dismissed_version`, `last_check` (survives restarts)
### API Endpoints
| Method | Path | Purpose |
|--------|------|---------|
| `GET` | `/api/v1/system/update/status` | Current state + available version |
| `POST` | `/api/v1/system/update/check` | Trigger immediate check |
| `POST` | `/api/v1/system/update/dismiss` | Dismiss notification for current version |
| `GET` | `/api/v1/system/update/settings` | Get settings |
| `PUT` | `/api/v1/system/update/settings` | Update settings |
### Wiring
- New `get_update_service()` in `dependencies.py`
- `UpdateService` created in `main.py` lifespan, `start()`/`stop()` alongside other engines
- Router registered in `api/__init__.py`
- WebSocket event: `update_available` fired via `processor_manager.fire_event()`
## Frontend
### Version badge highlight
The existing `#server-version` pill in the header gets a CSS class `has-update` when a newer version exists — changes the background to `var(--warning-color)` with a subtle pulse, making it clickable to open the update panel in settings.
### Notification popup
On `server:update_available` WebSocket event (and on page load if status says `has_update` and not dismissed):
- A **persistent dismissible banner** slides in below the header (not the ephemeral 3s toast)
- Shows: "Version {x.y.z} is available" + [View Release Notes] + [Dismiss]
- Dismiss calls `POST /dismiss` and hides the bar for that version
- Stored in `localStorage` so it doesn't re-show after page refresh for dismissed versions
### Settings tab: "Updates"
New 5th tab in the settings modal:
- Current version display
- "Check for updates" button + spinner
- Channel selector (stable / pre-release) via IconSelect
- Auto-check toggle + interval selector
- When update available: release name, notes preview, link to releases page
### i18n keys
New `update.*` keys in `en.json`, `ru.json`, `zh.json` for all labels.
## Files to Create
| File | Purpose |
|------|---------|
| `core/update/__init__.py` | Package init |
| `core/update/release_provider.py` | Abstract provider interface + data models |
| `core/update/gitea_provider.py` | Gitea API implementation |
| `core/update/version_check.py` | Semver normalization + comparison |
| `core/update/update_service.py` | Background service + state machine |
| `api/routes/update.py` | REST endpoints |
| `api/schemas/update.py` | Pydantic request/response models |
## Files to Modify
| File | Change |
|------|--------|
| `api/__init__.py` | Register update router |
| `api/dependencies.py` | Add `get_update_service()` |
| `main.py` | Create & start/stop UpdateService in lifespan |
| `templates/modals/settings.html` | Add Updates tab |
| `static/js/features/settings.ts` | Update check/settings UI logic |
| `static/js/core/api.ts` | Version badge highlight on health check |
| `static/css/layout.css` | `.has-update` styles for version badge |
| `static/locales/en.json` | i18n keys |
| `static/locales/ru.json` | i18n keys |
| `static/locales/zh.json` | i18n keys |
## Future Phases (not in scope)
- **Phase 2**: Download & stage artifacts
- **Phase 3**: Apply update & restart (external updater script, NSIS silent mode)
- **Phase 4**: Checksums, "What's new" dialog, update history
+17 -10
View File
@@ -7,7 +7,8 @@
| File | Trigger | Purpose |
|------|---------|---------|
| `.gitea/workflows/test.yml` | Push/PR to master | Lint (ruff) + pytest |
| `.gitea/workflows/release.yml` | Tag `v*` | Build artifacts + create Gitea release |
| `.gitea/workflows/release.yml` | Tag `v*` | Build Windows/Linux/Docker artifacts + create Gitea release |
| `.gitea/workflows/build-android.yml` | Push master (android/server paths), tag `v*`, manual | Build Android APK; attach to release on tag |
## Release Pipeline (`release.yml`)
@@ -17,7 +18,7 @@ Four parallel jobs triggered by pushing a `v*` tag:
Creates the Gitea release with a description table listing all artifacts. **The description must stay in sync with actual build outputs** — if you add/remove/rename an artifact, update the body template here.
### 2. `build-windows` (cross-built from Linux)
- Runs `build-dist-windows.sh` on Ubuntu with NSIS + msitools
- Runs `build/build-dist-windows.sh` on Ubuntu with NSIS + msitools
- Downloads Windows embedded Python 3.11 + pip wheels cross-platform
- Bundles tkinter from Python MSI via msiextract
- Builds frontend (`npm run build`)
@@ -25,21 +26,27 @@ Creates the Gitea release with a description table listing all artifacts. **The
- Produces: **`LedGrab-{tag}-win-x64.zip`** (portable) and **`LedGrab-{tag}-setup.exe`** (NSIS installer)
### 3. `build-linux`
- Runs `build-dist.sh` on Ubuntu
- Runs `build/build-dist.sh` on Ubuntu
- Creates a venv, installs deps, builds frontend
- Produces: **`LedGrab-{tag}-linux-x64.tar.gz`**
### 4. `build-docker`
- Plain `docker build` + `docker push` (no Buildx — TrueNAS runners lack nested networking)
- Plain `docker build` + `docker push` for **amd64** (no Buildx — TrueNAS runners lack nested networking)
- Registry: `{gitea_host}/{repo}:{tag}`
- Tags: `v0.x.x`, `0.x.x`, and `latest` (stable only, not alpha/beta/rc)
- **arm64 is best-effort** (Raspberry Pi / arm64 HAOS): a `continue-on-error` step
cross-builds arm64 via **QEMU binfmt** (`tonistiigi/binfmt`) + `docker manifest`
(NOT buildx — sidesteps the docker-container-driver networking limit) and folds
amd64 + arm64 into multi-arch manifest lists under the same tags, plus
`:{tag}-amd64` / `:{tag}-arm64` arch-suffixed tags. If the runner can't run
privileged binfmt the step is skipped and the amd64 tags above remain valid.
## Build Scripts
| Script | Platform | Output |
|--------|----------|--------|
| `build-dist-windows.sh` | Linux → Windows cross-build | ZIP + NSIS installer |
| `build-dist.sh` | Linux native | tarball |
| `build/build-dist-windows.sh` | Linux → Windows cross-build | ZIP + NSIS installer |
| `build/build-dist.sh` | Linux native | tarball |
| `server/Dockerfile` | Docker | Container image |
## Release Versioning
@@ -71,7 +78,7 @@ Build scripts use a fallback chain: CLI argument → exact git tag → CI env va
- **Launch after install**: Use `MUI_FINISHPAGE_RUN_FUNCTION` (not `MUI_FINISHPAGE_RUN_PARAMETERS` — NSIS `Exec` chokes on quoting). Still requires `MUI_FINISHPAGE_RUN ""` defined for checkbox visibility
- **Detect running instance**: `.onInit` checks file lock on `python.exe`, offers to kill process before install
- **Uninstall preserves user data**: Remove `python/`, `app/`, `logs/` but NOT `data/`
- **CI build**: `sudo apt-get install -y nsis msitools zip` then `makensis -DVERSION="${VERSION}" installer.nsi`
- **CI build**: `sudo apt-get install -y nsis msitools zip` then `makensis -DVERSION="${VERSION}" build/installer.nsi`
## Hidden Launcher (VBS)
@@ -135,12 +142,12 @@ The `create-release` job has fallback logic — if the release already exists fo
```bash
npm ci && npm run build # frontend
bash build-dist-windows.sh v1.0.0 # Windows dist
"/c/Program Files (x86)/NSIS/makensis.exe" -DVERSION="1.0.0" installer.nsi # installer
bash build/build-dist-windows.sh v1.0.0 # Windows dist
"/c/Program Files (x86)/NSIS/makensis.exe" -DVERSION="1.0.0" build/installer.nsi # installer
```
### Iterating on installer only
If only `installer.nsi` changed (not app code), skip the full rebuild — just re-run `makensis`. If app code changed, re-run `build-dist-windows.sh` first since `dist/` is a snapshot.
If only `installer.nsi` changed (not app code), skip the full rebuild — just re-run `makensis`. If app code changed, re-run `build/build-dist-windows.sh` first since `dist/` is a snapshot.
### Common issues
+33 -5
View File
@@ -4,7 +4,7 @@
## CSS Custom Properties (Variables)
Defined in `server/src/wled_controller/static/css/base.css`.
Defined in `server/src/ledgrab/static/css/base.css`.
**IMPORTANT:** There is NO `--accent` variable. Always use `--primary-color` for accent/brand color.
@@ -90,10 +90,26 @@ Plain `<select>` dropdowns should be enhanced with visual selectors depending on
Both widgets hide the native `<select>` but keep it in the DOM with its value in sync. **The `<select>` and the visual widget are two separate things — changing one does NOT automatically update the other.** After programmatically changing the `<select>` value, call `.refresh()` (EntitySelect) or `.setValue(val)` (IconSelect) to update the trigger display. Call `.destroy()` when the modal closes.
**CRITICAL pitfall — `<option>` elements required:** `IconSelect` does NOT create `<option>` elements in the native `<select>` — it only builds a visual popup grid. The native `<select>` must already contain matching `<option value="...">` elements (either from the Jinja2 template or added via JS) **before** `.value` is set. Setting `.value` on a `<select>` with no matching `<option>` **silently fails** — the value stays empty, and all downstream logic (section switching, auto-naming, type setup) breaks with no error. **When adding a new type to any IconSelect-enhanced `<select>`, you MUST add the `<option>` in the HTML template too.**
**Common pitfall:** Using a preset/palette selector (e.g. gradient preset dropdown or effect type picker) that changes the underlying `<select>` value but forgets to call `.setValue()` on the IconSelect — the visual grid still shows the old selection.
**IMPORTANT:** For `IconSelect` item icons, use SVG icons from `js/core/icon-paths.ts` (via `_icon(P.iconName)`) or styled `<span>` elements (e.g., `<span style="font-weight:bold">A</span>`). **Never use emoji** — they render inconsistently across platforms and themes.
### Dynamic entity lists (e.g. group child devices)
When a modal needs an **ordered list of entity references** (like child devices in a group), use the `EntityPalette.pick()` pattern instead of plain `<select>` dropdowns per row. Each row should be a styled card with:
- An icon (from `getDeviceTypeIcon()` / `getXxxIcon()`) showing the entity type
- The entity name and metadata (LED count, type, etc.)
- SVG icon buttons for reorder (chevron-up/down) and remove (trash)
- A click handler that opens `EntityPalette.pick()` to change the selection
- `data-*` attributes to store the selected value (not hidden `<select>` elements)
See `_addGroupChildRow()` in `device-discovery.ts` for the reference implementation. This pattern keeps the list visually consistent with the rest of the UI and avoids plain HTML selects.
For **mode/type toggles** with 2-4 fixed options (e.g. group mode: Sequence vs Independent), use `IconSelect` with descriptive icons and labels rather than radio buttons or plain selects. See `ensureGroupModeIconSelect()` for the pattern.
### Modal dirty check (discard unsaved changes)
Every editor modal **must** have a dirty check so closing with unsaved changes shows a "Discard unsaved changes?" confirmation. Use the `Modal` base class pattern from `js/core/modal.ts`:
@@ -175,7 +191,7 @@ When adding **new tabs, sections, or major UI elements**, update the correspondi
When you need a new icon:
1. Find the Lucide icon at https://lucide.dev
2. Copy the inner SVG elements (paths, circles, rects) into `icon-paths.js` as a new export
2. Copy the inner SVG elements (paths, circles, rects) into `icon-paths.ts` as a new export
3. Add a corresponding `ICON_*` constant in `icons.ts` using `_svg(P.myIcon)`
4. Import and use the constant in your feature module
@@ -213,7 +229,19 @@ Static HTML using `data-i18n` attributes is handled automatically by the i18n sy
- `fetchWithAuth('/devices/dev_123', { method: 'DELETE' })` → `DELETE /api/v1/devices/dev_123`
- Passing `/api/v1/gradients` results in **double prefix**: `/api/v1/api/v1/gradients` (404)
For raw `fetch()` without auth (rare), use the full path manually.
**NEVER use raw `fetch()` or `new Audio(url)` / `new Image()` for authenticated endpoints.** These bypass the auth token and will fail with 401. Always use `fetchWithAuth()` and convert to blob URLs when needed (e.g. for `<audio>`, `<img>`, or download links):
```typescript
// WRONG: no auth header — 401
const audio = new Audio(`${API_BASE}/assets/${id}/file`);
// CORRECT: fetch with auth, then create blob URL
const res = await fetchWithAuth(`/assets/${id}/file`);
const blob = await res.blob();
const audio = new Audio(URL.createObjectURL(blob));
```
The only exception is raw `fetch()` for multipart uploads where you must set `Content-Type` to let the browser handle the boundary — but still use `getHeaders()` for the auth token.
## Bundling & Development Workflow
@@ -266,11 +294,11 @@ See [`contexts/chrome-tools.md`](chrome-tools.md) for Chrome MCP tool usage, bro
### Uptime / duration values
Use `formatUptime(seconds)` from `core/ui.js`. Outputs `{s}s`, `{m}m {s}s`, or `{h}h {m}m` via i18n keys `time.seconds`, `time.minutes_seconds`, `time.hours_minutes`.
Use `formatUptime(seconds)` from `core/ui.ts`. Outputs `{s}s`, `{m}m {s}s`, or `{h}h {m}m` via i18n keys `time.seconds`, `time.minutes_seconds`, `time.hours_minutes`.
### Large numbers
Use `formatCompact(n)` from `core/ui.js`. Outputs `1.2K`, `3.5M` etc. Set `element.title` to the exact value for hover detail.
Use `formatCompact(n)` from `core/ui.ts`. Outputs `1.2K`, `3.5M` etc. Set `element.title` to the exact value for hover detail.
### Preventing layout shift
+22 -16
View File
@@ -8,34 +8,38 @@ Two independent server modes with separate configs, ports, and data directories:
| Mode | Command | Config | Port | API Key | Data |
| ---- | ------- | ------ | ---- | ------- | ---- |
| **Real** | `python -m wled_controller` | `config/default_config.yaml` | 8080 | `development-key-change-in-production` | `data/` |
| **Demo** | `python -m wled_controller.demo` | `config/demo_config.yaml` | 8081 | `demo` | `data/demo/` |
| **Real** | `python -m ledgrab` | `config/default_config.yaml` | 8080 | `development-key-change-in-production` | `data/` |
| **Demo** | `python -m ledgrab.demo` | `config/demo_config.yaml` | 8081 | `demo` | `data/demo/` |
Both can run simultaneously on different ports.
Demo mode can also be triggered via the `LEDGRAB_DEMO` environment variable (`true`, `1`, or `yes`). This works with any launch method — Python, Docker (`-e LEDGRAB_DEMO=true`), or the installed app (`set LEDGRAB_DEMO=true` before `LedGrab.bat`).
Both modes can run simultaneously on different ports.
## Restart Procedure
Use the PowerShell restart script — it gracefully shuts the running server down (so stores persist to disk), kills stragglers, launches a detached replacement, and polls the port until it's actually accepting connections. Exit code is 0 on success, 1 if the new server failed to bind the port, 2 on environment errors.
### Real server
Use the PowerShell restart script — it reliably stops only the server process and starts a new detached instance:
```bash
powershell -ExecutionPolicy Bypass -File "c:\Users\Alexei\Documents\wled-screen-controller\server\restart.ps1"
powershell -ExecutionPolicy Bypass -File "c:\Users\Alexei\Documents\led-grab-mixed\led-grab\server\restart.ps1"
```
### Demo server
Find and kill the process on port 8081, then restart:
```bash
# Find PID
powershell -Command "netstat -ano | Select-String ':8081.*LISTEN'"
# Kill it
powershell -Command "Stop-Process -Id <PID> -Force"
# Restart
cd server && python -m wled_controller.demo
powershell -ExecutionPolicy Bypass -File "c:\Users\Alexei\Documents\led-grab-mixed\led-grab\server\restart.ps1" `
-Port 8081 -Module ledgrab.demo -ConfigPath "config\demo_config.yaml"
```
### Useful parameters
- `-Port <int>` / `-Module <name>` — override the target (default: 8080 / `ledgrab`).
- `-StartupTimeoutSec <int>` — how long to wait for the new server to bind the port (default: 30).
- `-ShutdownTimeoutSec <int>` — how long to wait for graceful shutdown before force-killing (default: 15).
- `-Quiet` — suppress progress output.
- `-SkipBrowser:$false` — allow the app to open a browser tab on startup (default: skipped).
**Do NOT use** `Stop-Process -Name python` — it kills unrelated Python processes (VS Code extensions, etc.).
**Do NOT use** bash background `&` jobs — they get killed when the shell session ends.
@@ -43,6 +47,7 @@ cd server && python -m wled_controller.demo
## When to Restart
**Restart required** for changes to:
- API routes (`api/routes/`, `api/schemas/`)
- Core logic (`core/*.py`)
- Configuration (`config.py`)
@@ -50,6 +55,7 @@ cd server && python -m wled_controller.demo
- Data models (`storage/`)
**No restart needed** for:
- Static files (`static/js/`, `static/css/`) — but **must rebuild bundle**: `cd server && npm run build`
- Locale files (`static/locales/*.json`) — loaded by frontend
- Documentation files (`*.md`)
@@ -66,13 +72,13 @@ Auto-reload is disabled (`reload=False` in `main.py`) due to watchfiles causing
2. **New capture engines**: Verify demo mode filtering works — demo engines use `is_demo_mode()` gate in `is_available()`.
3. **New audio engines**: Same as capture engines — `is_available()` must respect `is_demo_mode()`.
4. **New device providers**: Gate discovery with `is_demo_mode()` like `DemoDeviceProvider.discover()`.
5. **New seed data**: Update `server/src/wled_controller/core/demo_seed.py` to include sample entities.
5. **New seed data**: Update `server/src/ledgrab/core/demo_seed.py` to include sample entities.
6. **Frontend indicators**: Demo state exposed via `GET /api/v1/version` -> `demo_mode: bool`. Frontend stores it as `demoMode` in app state and sets `document.body.dataset.demo = 'true'`.
7. **Backup/Restore**: New stores added to `STORE_MAP` in `system.py` automatically work in demo mode since the data directory is already isolated.
### Key files
- Config flag: `server/src/wled_controller/config.py` -> `Config.demo`, `is_demo_mode()`
- Config flag: `server/src/ledgrab/config.py` -> `Config.demo`, `is_demo_mode()`
- Demo engines: `core/capture_engines/demo_engine.py`, `core/audio/demo_engine.py`
- Demo devices: `core/devices/demo_provider.py`
- Seed data: `core/demo_seed.py`
@@ -1,208 +0,0 @@
"""The LED Screen Controller integration."""
from __future__ import annotations
import logging
from datetime import timedelta
import voluptuous as vol
from homeassistant.config_entries import ConfigEntry
from homeassistant.const import Platform
from homeassistant.core import HomeAssistant
from homeassistant.helpers import device_registry as dr
from homeassistant.helpers.aiohttp_client import async_get_clientsession
from .const import (
DOMAIN,
CONF_SERVER_NAME,
CONF_SERVER_URL,
CONF_API_KEY,
DEFAULT_SCAN_INTERVAL,
TARGET_TYPE_KEY_COLORS,
DATA_COORDINATOR,
DATA_WS_MANAGER,
DATA_EVENT_LISTENER,
)
from .coordinator import WLEDScreenControllerCoordinator
from .event_listener import EventStreamListener
from .ws_manager import KeyColorsWebSocketManager
_LOGGER = logging.getLogger(__name__)
PLATFORMS: list[Platform] = [
Platform.BUTTON,
Platform.LIGHT,
Platform.SWITCH,
Platform.SENSOR,
Platform.NUMBER,
Platform.SELECT,
]
async def async_setup_entry(hass: HomeAssistant, entry: ConfigEntry) -> bool:
"""Set up LED Screen Controller from a config entry."""
server_name = entry.data.get(CONF_SERVER_NAME, "LED Screen Controller")
server_url = entry.data[CONF_SERVER_URL]
api_key = entry.data[CONF_API_KEY]
session = async_get_clientsession(hass)
coordinator = WLEDScreenControllerCoordinator(
hass,
session,
server_url,
api_key,
update_interval=timedelta(seconds=DEFAULT_SCAN_INTERVAL),
)
await coordinator.async_config_entry_first_refresh()
ws_manager = KeyColorsWebSocketManager(hass, server_url, api_key)
event_listener = EventStreamListener(hass, server_url, api_key, coordinator)
await event_listener.start()
# Create device entries for each target and remove stale ones
device_registry = dr.async_get(hass)
current_identifiers: set[tuple[str, str]] = set()
if coordinator.data and "targets" in coordinator.data:
for target_id, target_data in coordinator.data["targets"].items():
info = target_data["info"]
target_type = info.get("target_type", "led")
model = (
"Key Colors Target"
if target_type == TARGET_TYPE_KEY_COLORS
else "LED Target"
)
device_registry.async_get_or_create(
config_entry_id=entry.entry_id,
identifiers={(DOMAIN, target_id)},
name=info.get("name", target_id),
manufacturer=server_name,
model=model,
configuration_url=server_url,
)
current_identifiers.add((DOMAIN, target_id))
# Create a single "Scenes" device for scene preset buttons
scenes_identifier = (DOMAIN, f"{entry.entry_id}_scenes")
scene_presets = coordinator.data.get("scene_presets", []) if coordinator.data else []
if scene_presets:
device_registry.async_get_or_create(
config_entry_id=entry.entry_id,
identifiers={scenes_identifier},
name="Scenes",
manufacturer=server_name,
model="Scene Presets",
configuration_url=server_url,
)
current_identifiers.add(scenes_identifier)
# Remove devices for targets that no longer exist
for device_entry in dr.async_entries_for_config_entry(
device_registry, entry.entry_id
):
if not device_entry.identifiers & current_identifiers:
_LOGGER.info("Removing stale device: %s", device_entry.name)
device_registry.async_remove_device(device_entry.id)
# Store data
hass.data.setdefault(DOMAIN, {})
hass.data[DOMAIN][entry.entry_id] = {
DATA_COORDINATOR: coordinator,
DATA_WS_MANAGER: ws_manager,
DATA_EVENT_LISTENER: event_listener,
}
# Track target and scene IDs to detect changes
known_target_ids = set(
coordinator.data.get("targets", {}).keys() if coordinator.data else []
)
known_scene_ids = set(
p["id"] for p in (coordinator.data.get("scene_presets", []) if coordinator.data else [])
)
def _on_coordinator_update() -> None:
"""Manage WS connections and detect target list changes."""
nonlocal known_target_ids, known_scene_ids
if not coordinator.data:
return
targets = coordinator.data.get("targets", {})
# Start/stop WS connections for KC targets based on processing state
for target_id, target_data in targets.items():
info = target_data.get("info", {})
state = target_data.get("state") or {}
if info.get("target_type") == TARGET_TYPE_KEY_COLORS:
if state.get("processing"):
if target_id not in ws_manager._connections:
hass.async_create_task(ws_manager.start_listening(target_id))
else:
if target_id in ws_manager._connections:
hass.async_create_task(ws_manager.stop_listening(target_id))
# Reload if target or scene list changed
current_ids = set(targets.keys())
current_scene_ids = set(
p["id"] for p in coordinator.data.get("scene_presets", [])
)
if current_ids != known_target_ids or current_scene_ids != known_scene_ids:
known_target_ids = current_ids
known_scene_ids = current_scene_ids
_LOGGER.info("Target or scene list changed, reloading integration")
hass.async_create_task(
hass.config_entries.async_reload(entry.entry_id)
)
coordinator.async_add_listener(_on_coordinator_update)
# Register set_leds service (once across all entries)
async def handle_set_leds(call) -> None:
"""Handle the set_leds service call."""
source_id = call.data["source_id"]
segments = call.data["segments"]
# Route to the coordinator that owns this source
for entry_data in hass.data[DOMAIN].values():
coord = entry_data.get(DATA_COORDINATOR)
if not coord or not coord.data:
continue
source_ids = {
s["id"] for s in coord.data.get("css_sources", [])
}
if source_id in source_ids:
await coord.push_segments(source_id, segments)
return
_LOGGER.error("No server found with source_id %s", source_id)
if not hass.services.has_service(DOMAIN, "set_leds"):
hass.services.async_register(
DOMAIN,
"set_leds",
handle_set_leds,
schema=vol.Schema({
vol.Required("source_id"): str,
vol.Required("segments"): list,
}),
)
await hass.config_entries.async_forward_entry_setups(entry, PLATFORMS)
return True
async def async_unload_entry(hass: HomeAssistant, entry: ConfigEntry) -> bool:
"""Unload a config entry."""
entry_data = hass.data[DOMAIN][entry.entry_id]
await entry_data[DATA_WS_MANAGER].shutdown()
await entry_data[DATA_EVENT_LISTENER].shutdown()
unload_ok = await hass.config_entries.async_unload_platforms(entry, PLATFORMS)
if unload_ok:
hass.data[DOMAIN].pop(entry.entry_id)
# Unregister service if no entries remain
if not hass.data[DOMAIN]:
hass.services.async_remove(DOMAIN, "set_leds")
return unload_ok
@@ -1,74 +0,0 @@
"""Button platform for LED Screen Controller — scene preset activation."""
from __future__ import annotations
import logging
from typing import Any
from homeassistant.components.button import ButtonEntity
from homeassistant.config_entries import ConfigEntry
from homeassistant.core import HomeAssistant
from homeassistant.helpers.entity_platform import AddEntitiesCallback
from homeassistant.helpers.update_coordinator import CoordinatorEntity
from .const import DOMAIN, DATA_COORDINATOR
from .coordinator import WLEDScreenControllerCoordinator
_LOGGER = logging.getLogger(__name__)
async def async_setup_entry(
hass: HomeAssistant,
entry: ConfigEntry,
async_add_entities: AddEntitiesCallback,
) -> None:
"""Set up scene preset buttons."""
data = hass.data[DOMAIN][entry.entry_id]
coordinator: WLEDScreenControllerCoordinator = data[DATA_COORDINATOR]
entities = []
if coordinator.data:
for preset in coordinator.data.get("scene_presets", []):
entities.append(
SceneActivateButton(coordinator, preset, entry.entry_id)
)
async_add_entities(entities)
class SceneActivateButton(CoordinatorEntity, ButtonEntity):
"""Button that activates a scene preset."""
_attr_has_entity_name = True
def __init__(
self,
coordinator: WLEDScreenControllerCoordinator,
preset: dict[str, Any],
entry_id: str,
) -> None:
"""Initialize the button."""
super().__init__(coordinator)
self._preset_id = preset["id"]
self._entry_id = entry_id
self._attr_unique_id = f"{entry_id}_scene_{preset['id']}"
self._attr_translation_key = "activate_scene"
self._attr_translation_placeholders = {"scene_name": preset["name"]}
self._attr_icon = "mdi:palette"
@property
def device_info(self) -> dict[str, Any]:
"""Return device information — all scene buttons belong to the Scenes device."""
return {"identifiers": {(DOMAIN, f"{self._entry_id}_scenes")}}
@property
def available(self) -> bool:
"""Return if entity is available."""
if not self.coordinator.data:
return False
return self._preset_id in {
p["id"] for p in self.coordinator.data.get("scene_presets", [])
}
async def async_press(self) -> None:
"""Activate the scene preset."""
await self.coordinator.activate_scene(self._preset_id)
@@ -1,127 +0,0 @@
"""Config flow for LED Screen Controller integration."""
from __future__ import annotations
import logging
from typing import Any
from urllib.parse import urlparse, urlunparse
import aiohttp
import voluptuous as vol
from homeassistant import config_entries
from homeassistant.core import HomeAssistant
from homeassistant.data_entry_flow import FlowResult
from homeassistant.helpers.aiohttp_client import async_get_clientsession
from .const import DOMAIN, CONF_SERVER_NAME, CONF_SERVER_URL, CONF_API_KEY, DEFAULT_TIMEOUT
_LOGGER = logging.getLogger(__name__)
STEP_USER_DATA_SCHEMA = vol.Schema(
{
vol.Optional(CONF_SERVER_NAME, default="LED Screen Controller"): str,
vol.Required(CONF_SERVER_URL, default="http://localhost:8080"): str,
vol.Optional(CONF_API_KEY, default=""): str,
}
)
def normalize_url(url: str) -> str:
"""Normalize URL to ensure port is an integer."""
parsed = urlparse(url)
if parsed.port is not None:
netloc = parsed.hostname or "localhost"
port = int(parsed.port)
if port != (443 if parsed.scheme == "https" else 80):
netloc = f"{netloc}:{port}"
parsed = parsed._replace(netloc=netloc)
return urlunparse(parsed)
async def validate_server(
hass: HomeAssistant, server_url: str, api_key: str
) -> dict[str, Any]:
"""Validate server connectivity and API key."""
session = async_get_clientsession(hass)
timeout = aiohttp.ClientTimeout(total=DEFAULT_TIMEOUT)
# Step 1: Check connectivity via health endpoint (no auth needed)
try:
async with session.get(f"{server_url}/health", timeout=timeout) as resp:
if resp.status != 200:
raise ConnectionError(f"Server returned status {resp.status}")
data = await resp.json()
version = data.get("version", "unknown")
except aiohttp.ClientError as err:
raise ConnectionError(f"Cannot connect to server: {err}") from err
# Step 2: Validate API key via authenticated endpoint (skip if no key and auth not required)
auth_required = data.get("auth_required", True)
if api_key:
headers = {"Authorization": f"Bearer {api_key}"}
try:
async with session.get(
f"{server_url}/api/v1/output-targets",
headers=headers,
timeout=timeout,
) as resp:
if resp.status == 401:
raise PermissionError("Invalid API key")
resp.raise_for_status()
except PermissionError:
raise
except aiohttp.ClientError as err:
raise ConnectionError(f"API request failed: {err}") from err
elif auth_required:
raise PermissionError("Server requires an API key")
return {"version": version}
class WLEDScreenControllerConfigFlow(config_entries.ConfigFlow, domain=DOMAIN):
"""Handle a config flow for LED Screen Controller."""
VERSION = 2
async def async_step_user(
self, user_input: dict[str, Any] | None = None
) -> FlowResult:
"""Handle the initial step."""
errors: dict[str, str] = {}
if user_input is not None:
server_name = user_input.get(CONF_SERVER_NAME, "LED Screen Controller")
server_url = normalize_url(user_input[CONF_SERVER_URL].rstrip("/"))
api_key = user_input[CONF_API_KEY]
try:
await validate_server(self.hass, server_url, api_key)
await self.async_set_unique_id(server_url)
self._abort_if_unique_id_configured()
return self.async_create_entry(
title=server_name,
data={
CONF_SERVER_NAME: server_name,
CONF_SERVER_URL: server_url,
CONF_API_KEY: api_key,
},
)
except ConnectionError as err:
_LOGGER.error("Connection error: %s", err)
errors["base"] = "cannot_connect"
except PermissionError:
errors["base"] = "invalid_api_key"
except Exception as err:
_LOGGER.exception("Unexpected exception: %s", err)
errors["base"] = "unknown"
return self.async_show_form(
step_id="user",
data_schema=STEP_USER_DATA_SCHEMA,
errors=errors,
)
@@ -1,23 +0,0 @@
"""Constants for the LED Screen Controller integration."""
DOMAIN = "wled_screen_controller"
# Configuration
CONF_SERVER_NAME = "server_name"
CONF_SERVER_URL = "server_url"
CONF_API_KEY = "api_key"
# Default values
DEFAULT_SCAN_INTERVAL = 3 # seconds
DEFAULT_TIMEOUT = 10 # seconds
WS_RECONNECT_DELAY = 5 # seconds
WS_MAX_RECONNECT_DELAY = 60 # seconds
# Target types
TARGET_TYPE_LED = "led"
TARGET_TYPE_KEY_COLORS = "key_colors"
# Data keys stored in hass.data[DOMAIN][entry_id]
DATA_COORDINATOR = "coordinator"
DATA_WS_MANAGER = "ws_manager"
DATA_EVENT_LISTENER = "event_listener"
@@ -1,459 +0,0 @@
"""Data update coordinator for LED Screen Controller."""
from __future__ import annotations
import asyncio
from datetime import timedelta
import logging
from typing import Any
import aiohttp
from homeassistant.core import HomeAssistant
from homeassistant.helpers.update_coordinator import DataUpdateCoordinator, UpdateFailed
from .const import (
DOMAIN,
DEFAULT_TIMEOUT,
TARGET_TYPE_KEY_COLORS,
)
_LOGGER = logging.getLogger(__name__)
class WLEDScreenControllerCoordinator(DataUpdateCoordinator):
"""Class to manage fetching LED Screen Controller data."""
def __init__(
self,
hass: HomeAssistant,
session: aiohttp.ClientSession,
server_url: str,
api_key: str,
update_interval: timedelta,
) -> None:
"""Initialize the coordinator."""
self.server_url = server_url
self.session = session
self.api_key = api_key
self.server_version = "unknown"
self._auth_headers = {"Authorization": f"Bearer {api_key}"} if api_key else {}
self._timeout = aiohttp.ClientTimeout(total=DEFAULT_TIMEOUT)
super().__init__(
hass,
_LOGGER,
name=DOMAIN,
update_interval=update_interval,
)
async def _async_update_data(self) -> dict[str, Any]:
"""Fetch data from API."""
try:
async with asyncio.timeout(DEFAULT_TIMEOUT * 3):
if self.server_version == "unknown":
await self._fetch_server_version()
targets_list = await self._fetch_targets()
# Fetch state and metrics for all targets in parallel
targets_data: dict[str, dict[str, Any]] = {}
async def fetch_target_data(target: dict) -> tuple[str, dict]:
target_id = target["id"]
try:
state, metrics = await asyncio.gather(
self._fetch_target_state(target_id),
self._fetch_target_metrics(target_id),
)
except Exception as err:
_LOGGER.warning(
"Failed to fetch data for target %s: %s",
target_id,
err,
)
state = None
metrics = None
result: dict[str, Any] = {
"info": target,
"state": state,
"metrics": metrics,
}
# Fetch rectangles for key_colors targets
if target.get("target_type") == TARGET_TYPE_KEY_COLORS:
kc_settings = target.get("key_colors_settings") or {}
template_id = kc_settings.get("pattern_template_id", "")
if template_id:
result["rectangles"] = await self._fetch_rectangles(
template_id
)
else:
result["rectangles"] = []
else:
result["rectangles"] = []
return target_id, result
results = await asyncio.gather(
*(fetch_target_data(t) for t in targets_list),
return_exceptions=True,
)
for r in results:
if isinstance(r, Exception):
_LOGGER.warning("Target fetch failed: %s", r)
continue
target_id, data = r
targets_data[target_id] = data
# Fetch devices, CSS sources, value sources, and scene presets in parallel
devices_data, css_sources, value_sources, scene_presets = (
await asyncio.gather(
self._fetch_devices(),
self._fetch_css_sources(),
self._fetch_value_sources(),
self._fetch_scene_presets(),
)
)
return {
"targets": targets_data,
"devices": devices_data,
"css_sources": css_sources,
"value_sources": value_sources,
"scene_presets": scene_presets,
"server_version": self.server_version,
}
except asyncio.TimeoutError as err:
raise UpdateFailed(f"Timeout fetching data: {err}") from err
except aiohttp.ClientError as err:
raise UpdateFailed(f"Error communicating with API: {err}") from err
async def _fetch_server_version(self) -> None:
"""Fetch server version from health endpoint."""
try:
async with self.session.get(
f"{self.server_url}/health",
timeout=self._timeout,
) as resp:
resp.raise_for_status()
data = await resp.json()
self.server_version = data.get("version", "unknown")
except Exception as err:
_LOGGER.warning("Failed to fetch server version: %s", err)
self.server_version = "unknown"
async def _fetch_targets(self) -> list[dict[str, Any]]:
"""Fetch all output targets."""
async with self.session.get(
f"{self.server_url}/api/v1/output-targets",
headers=self._auth_headers,
timeout=self._timeout,
) as resp:
resp.raise_for_status()
data = await resp.json()
return data.get("targets", [])
async def _fetch_target_state(self, target_id: str) -> dict[str, Any]:
"""Fetch target processing state."""
async with self.session.get(
f"{self.server_url}/api/v1/output-targets/{target_id}/state",
headers=self._auth_headers,
timeout=self._timeout,
) as resp:
resp.raise_for_status()
return await resp.json()
async def _fetch_target_metrics(self, target_id: str) -> dict[str, Any]:
"""Fetch target metrics."""
async with self.session.get(
f"{self.server_url}/api/v1/output-targets/{target_id}/metrics",
headers=self._auth_headers,
timeout=self._timeout,
) as resp:
resp.raise_for_status()
return await resp.json()
async def _fetch_rectangles(self, template_id: str) -> list[dict]:
"""Fetch rectangles for a pattern template (no cache — always fresh)."""
try:
async with self.session.get(
f"{self.server_url}/api/v1/pattern-templates/{template_id}",
headers=self._auth_headers,
timeout=self._timeout,
) as resp:
resp.raise_for_status()
data = await resp.json()
return data.get("rectangles", [])
except Exception as err:
_LOGGER.warning(
"Failed to fetch pattern template %s: %s", template_id, err
)
return []
async def _fetch_devices(self) -> dict[str, dict[str, Any]]:
"""Fetch all devices with capabilities and brightness."""
try:
async with self.session.get(
f"{self.server_url}/api/v1/devices",
headers=self._auth_headers,
timeout=self._timeout,
) as resp:
resp.raise_for_status()
data = await resp.json()
devices = data.get("devices", [])
except Exception as err:
_LOGGER.warning("Failed to fetch devices: %s", err)
return {}
# Fetch brightness for all capable devices in parallel
async def fetch_device_entry(device: dict) -> tuple[str, dict[str, Any]]:
device_id = device["id"]
entry: dict[str, Any] = {"info": device, "brightness": None}
if "brightness_control" in (device.get("capabilities") or []):
try:
async with self.session.get(
f"{self.server_url}/api/v1/devices/{device_id}/brightness",
headers=self._auth_headers,
timeout=self._timeout,
) as resp:
if resp.status == 200:
bri_data = await resp.json()
entry["brightness"] = bri_data.get("brightness")
except Exception as err:
_LOGGER.warning(
"Failed to fetch brightness for device %s: %s",
device_id, err,
)
return device_id, entry
results = await asyncio.gather(
*(fetch_device_entry(d) for d in devices),
return_exceptions=True,
)
devices_data: dict[str, dict[str, Any]] = {}
for r in results:
if isinstance(r, Exception):
_LOGGER.warning("Device fetch failed: %s", r)
continue
device_id, entry = r
devices_data[device_id] = entry
return devices_data
async def set_brightness(self, device_id: str, brightness: int) -> None:
"""Set brightness for a device."""
async with self.session.put(
f"{self.server_url}/api/v1/devices/{device_id}/brightness",
headers={**self._auth_headers, "Content-Type": "application/json"},
json={"brightness": brightness},
timeout=self._timeout,
) as resp:
if resp.status != 200:
body = await resp.text()
_LOGGER.error(
"Failed to set brightness for device %s: %s %s",
device_id, resp.status, body,
)
resp.raise_for_status()
await self.async_request_refresh()
async def set_color(self, device_id: str, color: list[int] | None) -> None:
"""Set or clear the static color for a device."""
async with self.session.put(
f"{self.server_url}/api/v1/devices/{device_id}/color",
headers={**self._auth_headers, "Content-Type": "application/json"},
json={"color": color},
timeout=self._timeout,
) as resp:
if resp.status != 200:
body = await resp.text()
_LOGGER.error(
"Failed to set color for device %s: %s %s",
device_id, resp.status, body,
)
resp.raise_for_status()
await self.async_request_refresh()
async def set_kc_brightness(self, target_id: str, brightness: int) -> None:
"""Set brightness for a Key Colors target (0-255 mapped to 0.0-1.0)."""
brightness_float = round(brightness / 255, 4)
async with self.session.put(
f"{self.server_url}/api/v1/output-targets/{target_id}",
headers={**self._auth_headers, "Content-Type": "application/json"},
json={"key_colors_settings": {"brightness": brightness_float}},
timeout=self._timeout,
) as resp:
if resp.status != 200:
body = await resp.text()
_LOGGER.error(
"Failed to set KC brightness for target %s: %s %s",
target_id, resp.status, body,
)
resp.raise_for_status()
await self.async_request_refresh()
async def _fetch_css_sources(self) -> list[dict[str, Any]]:
"""Fetch all color strip sources."""
try:
async with self.session.get(
f"{self.server_url}/api/v1/color-strip-sources",
headers=self._auth_headers,
timeout=self._timeout,
) as resp:
resp.raise_for_status()
data = await resp.json()
return data.get("sources", [])
except Exception as err:
_LOGGER.warning("Failed to fetch CSS sources: %s", err)
return []
async def _fetch_value_sources(self) -> list[dict[str, Any]]:
"""Fetch all value sources."""
try:
async with self.session.get(
f"{self.server_url}/api/v1/value-sources",
headers=self._auth_headers,
timeout=self._timeout,
) as resp:
resp.raise_for_status()
data = await resp.json()
return data.get("sources", [])
except Exception as err:
_LOGGER.warning("Failed to fetch value sources: %s", err)
return []
async def _fetch_scene_presets(self) -> list[dict[str, Any]]:
"""Fetch all scene presets."""
try:
async with self.session.get(
f"{self.server_url}/api/v1/scene-presets",
headers=self._auth_headers,
timeout=self._timeout,
) as resp:
resp.raise_for_status()
data = await resp.json()
return data.get("presets", [])
except Exception as err:
_LOGGER.warning("Failed to fetch scene presets: %s", err)
return []
async def push_colors(self, source_id: str, colors: list[list[int]]) -> None:
"""Push flat color array to an api_input CSS source."""
async with self.session.post(
f"{self.server_url}/api/v1/color-strip-sources/{source_id}/colors",
headers={**self._auth_headers, "Content-Type": "application/json"},
json={"colors": colors},
timeout=self._timeout,
) as resp:
if resp.status not in (200, 204):
body = await resp.text()
_LOGGER.error(
"Failed to push colors to source %s: %s %s",
source_id, resp.status, body,
)
resp.raise_for_status()
async def push_segments(self, source_id: str, segments: list[dict]) -> None:
"""Push segment data to an api_input CSS source."""
async with self.session.post(
f"{self.server_url}/api/v1/color-strip-sources/{source_id}/colors",
headers={**self._auth_headers, "Content-Type": "application/json"},
json={"segments": segments},
timeout=self._timeout,
) as resp:
if resp.status not in (200, 204):
body = await resp.text()
_LOGGER.error(
"Failed to push segments to source %s: %s %s",
source_id, resp.status, body,
)
resp.raise_for_status()
async def activate_scene(self, preset_id: str) -> None:
"""Activate a scene preset."""
async with self.session.post(
f"{self.server_url}/api/v1/scene-presets/{preset_id}/activate",
headers=self._auth_headers,
timeout=self._timeout,
) as resp:
if resp.status != 200:
body = await resp.text()
_LOGGER.error(
"Failed to activate scene %s: %s %s",
preset_id, resp.status, body,
)
resp.raise_for_status()
await self.async_request_refresh()
async def update_source(self, source_id: str, **kwargs: Any) -> None:
"""Update a color strip source's fields."""
async with self.session.put(
f"{self.server_url}/api/v1/color-strip-sources/{source_id}",
headers={**self._auth_headers, "Content-Type": "application/json"},
json=kwargs,
timeout=self._timeout,
) as resp:
if resp.status != 200:
body = await resp.text()
_LOGGER.error(
"Failed to update source %s: %s %s",
source_id, resp.status, body,
)
resp.raise_for_status()
async def update_target(self, target_id: str, **kwargs: Any) -> None:
"""Update an output target's fields."""
async with self.session.put(
f"{self.server_url}/api/v1/output-targets/{target_id}",
headers={**self._auth_headers, "Content-Type": "application/json"},
json=kwargs,
timeout=self._timeout,
) as resp:
if resp.status != 200:
body = await resp.text()
_LOGGER.error(
"Failed to update target %s: %s %s",
target_id, resp.status, body,
)
resp.raise_for_status()
await self.async_request_refresh()
async def start_processing(self, target_id: str) -> None:
"""Start processing for a target."""
async with self.session.post(
f"{self.server_url}/api/v1/output-targets/{target_id}/start",
headers=self._auth_headers,
timeout=self._timeout,
) as resp:
if resp.status == 409:
_LOGGER.debug("Target %s already processing", target_id)
elif resp.status != 200:
body = await resp.text()
_LOGGER.error(
"Failed to start target %s: %s %s",
target_id, resp.status, body,
)
resp.raise_for_status()
await self.async_request_refresh()
async def stop_processing(self, target_id: str) -> None:
"""Stop processing for a target."""
async with self.session.post(
f"{self.server_url}/api/v1/output-targets/{target_id}/stop",
headers=self._auth_headers,
timeout=self._timeout,
) as resp:
if resp.status == 409:
_LOGGER.debug("Target %s already stopped", target_id)
elif resp.status != 200:
body = await resp.text()
_LOGGER.error(
"Failed to stop target %s: %s %s",
target_id, resp.status, body,
)
resp.raise_for_status()
await self.async_request_refresh()
@@ -1,95 +0,0 @@
"""WebSocket event listener for server state change notifications."""
from __future__ import annotations
import asyncio
import contextlib
import json
import logging
import aiohttp
from homeassistant.core import HomeAssistant
from homeassistant.helpers.aiohttp_client import async_get_clientsession
from homeassistant.helpers.update_coordinator import DataUpdateCoordinator
from .const import WS_RECONNECT_DELAY, WS_MAX_RECONNECT_DELAY
_LOGGER = logging.getLogger(__name__)
class EventStreamListener:
"""Listens to server WS endpoint for state change events.
Triggers a coordinator refresh whenever a target starts or stops processing,
so HAOS entities react near-instantly to external state changes.
"""
def __init__(
self,
hass: HomeAssistant,
server_url: str,
api_key: str,
coordinator: DataUpdateCoordinator,
) -> None:
self._hass = hass
self._server_url = server_url
self._api_key = api_key
self._coordinator = coordinator
self._task: asyncio.Task | None = None
self._shutting_down = False
async def start(self) -> None:
"""Start listening to the event stream."""
self._task = self._hass.async_create_background_task(
self._ws_loop(),
"wled_screen_controller_events",
)
async def _ws_loop(self) -> None:
"""WebSocket connection loop with reconnection."""
delay = WS_RECONNECT_DELAY
session = async_get_clientsession(self._hass)
ws_base = self._server_url.replace("http://", "ws://").replace(
"https://", "wss://"
)
url = f"{ws_base}/api/v1/events/ws?token={self._api_key}"
while not self._shutting_down:
try:
async with session.ws_connect(url) as ws:
delay = WS_RECONNECT_DELAY # reset on successful connect
_LOGGER.debug("Event stream connected")
async for msg in ws:
if msg.type == aiohttp.WSMsgType.TEXT:
try:
data = json.loads(msg.data)
except json.JSONDecodeError:
continue
if data.get("type") == "state_change":
await self._coordinator.async_request_refresh()
elif msg.type in (
aiohttp.WSMsgType.CLOSED,
aiohttp.WSMsgType.ERROR,
):
break
except asyncio.CancelledError:
raise
except (aiohttp.ClientError, asyncio.TimeoutError, OSError) as err:
_LOGGER.debug("Event stream connection error: %s", err)
except Exception as err:
_LOGGER.error("Unexpected event stream error: %s", err)
if self._shutting_down:
break
await asyncio.sleep(delay)
delay = min(delay * 2, WS_MAX_RECONNECT_DELAY)
async def shutdown(self) -> None:
"""Stop listening."""
self._shutting_down = True
if self._task:
self._task.cancel()
with contextlib.suppress(asyncio.CancelledError):
await self._task
self._task = None
@@ -1,151 +0,0 @@
"""Light platform for LED Screen Controller (api_input CSS sources)."""
from __future__ import annotations
import logging
from typing import Any
from homeassistant.components.light import (
ATTR_BRIGHTNESS,
ATTR_RGB_COLOR,
ColorMode,
LightEntity,
)
from homeassistant.config_entries import ConfigEntry
from homeassistant.core import HomeAssistant
from homeassistant.helpers.entity_platform import AddEntitiesCallback
from homeassistant.helpers.update_coordinator import CoordinatorEntity
from .const import DOMAIN, DATA_COORDINATOR
from .coordinator import WLEDScreenControllerCoordinator
_LOGGER = logging.getLogger(__name__)
async def async_setup_entry(
hass: HomeAssistant,
entry: ConfigEntry,
async_add_entities: AddEntitiesCallback,
) -> None:
"""Set up LED Screen Controller api_input lights."""
data = hass.data[DOMAIN][entry.entry_id]
coordinator: WLEDScreenControllerCoordinator = data[DATA_COORDINATOR]
entities = []
if coordinator.data:
for source in coordinator.data.get("css_sources", []):
if source.get("source_type") == "api_input":
entities.append(
ApiInputLight(coordinator, source, entry.entry_id)
)
async_add_entities(entities)
class ApiInputLight(CoordinatorEntity, LightEntity):
"""Representation of an api_input CSS source as a light entity."""
_attr_has_entity_name = True
_attr_color_mode = ColorMode.RGB
_attr_supported_color_modes = {ColorMode.RGB}
_attr_translation_key = "api_input_light"
_attr_icon = "mdi:led-strip-variant"
def __init__(
self,
coordinator: WLEDScreenControllerCoordinator,
source: dict[str, Any],
entry_id: str,
) -> None:
"""Initialize the light."""
super().__init__(coordinator)
self._source_id: str = source["id"]
self._source_name: str = source.get("name", self._source_id)
self._entry_id = entry_id
self._attr_unique_id = f"{self._source_id}_light"
# Restore state from fallback_color
fallback = self._get_fallback_color()
is_off = fallback == [0, 0, 0]
self._is_on: bool = not is_off
self._rgb_color: tuple[int, int, int] = (
(255, 255, 255) if is_off else tuple(fallback) # type: ignore[arg-type]
)
self._brightness: int = 255
@property
def device_info(self) -> dict[str, Any]:
"""Return device information — one virtual device per api_input source."""
return {
"identifiers": {(DOMAIN, self._source_id)},
"name": self._source_name,
"manufacturer": "WLED Screen Controller",
"model": "API Input CSS Source",
}
@property
def name(self) -> str:
"""Return the entity name."""
return self._source_name
@property
def is_on(self) -> bool:
"""Return true if the light is on."""
return self._is_on
@property
def rgb_color(self) -> tuple[int, int, int]:
"""Return the current RGB color."""
return self._rgb_color
@property
def brightness(self) -> int:
"""Return the current brightness (0-255)."""
return self._brightness
async def async_turn_on(self, **kwargs: Any) -> None:
"""Turn on the light, optionally setting color and brightness."""
if ATTR_RGB_COLOR in kwargs:
self._rgb_color = kwargs[ATTR_RGB_COLOR]
if ATTR_BRIGHTNESS in kwargs:
self._brightness = kwargs[ATTR_BRIGHTNESS]
# Scale RGB by brightness
scale = self._brightness / 255
r, g, b = self._rgb_color
scaled = [round(r * scale), round(g * scale), round(b * scale)]
await self.coordinator.push_segments(
self._source_id,
[{"start": 0, "length": 9999, "mode": "solid", "color": scaled}],
)
# Update fallback_color so the color persists beyond the timeout
await self.coordinator.update_source(
self._source_id, fallback_color=scaled,
)
self._is_on = True
self.async_write_ha_state()
async def async_turn_off(self, **kwargs: Any) -> None:
"""Turn off the light by pushing black and setting fallback to black."""
off_color = [0, 0, 0]
await self.coordinator.push_segments(
self._source_id,
[{"start": 0, "length": 9999, "mode": "solid", "color": off_color}],
)
await self.coordinator.update_source(
self._source_id, fallback_color=off_color,
)
self._is_on = False
self.async_write_ha_state()
def _get_fallback_color(self) -> list[int]:
"""Read fallback_color from the source config in coordinator data."""
if not self.coordinator.data:
return [0, 0, 0]
for source in self.coordinator.data.get("css_sources", []):
if source.get("id") == self._source_id:
fallback = source.get("fallback_color")
if fallback and len(fallback) >= 3:
return list(fallback[:3])
break
return [0, 0, 0]
@@ -1,12 +0,0 @@
{
"domain": "wled_screen_controller",
"name": "LED Screen Controller",
"codeowners": ["@alexeidolgolyov"],
"config_flow": true,
"dependencies": [],
"documentation": "https://git.dolgolyov-family.by/alexei.dolgolyov/wled-screen-controller-mixed",
"iot_class": "local_push",
"issue_tracker": "https://git.dolgolyov-family.by/alexei.dolgolyov/wled-screen-controller-mixed/issues",
"requirements": ["aiohttp>=3.9.0"],
"version": "0.2.0"
}
@@ -1,169 +0,0 @@
"""Number platform for LED Screen Controller (device & KC target brightness)."""
from __future__ import annotations
import logging
from typing import Any
from homeassistant.components.number import NumberEntity, NumberMode
from homeassistant.config_entries import ConfigEntry
from homeassistant.core import HomeAssistant
from homeassistant.helpers.entity_platform import AddEntitiesCallback
from homeassistant.helpers.update_coordinator import CoordinatorEntity
from .const import DOMAIN, DATA_COORDINATOR, TARGET_TYPE_KEY_COLORS
from .coordinator import WLEDScreenControllerCoordinator
_LOGGER = logging.getLogger(__name__)
async def async_setup_entry(
hass: HomeAssistant,
entry: ConfigEntry,
async_add_entities: AddEntitiesCallback,
) -> None:
"""Set up LED Screen Controller brightness numbers."""
data = hass.data[DOMAIN][entry.entry_id]
coordinator: WLEDScreenControllerCoordinator = data[DATA_COORDINATOR]
entities = []
if coordinator.data and "targets" in coordinator.data:
devices = coordinator.data.get("devices") or {}
for target_id, target_data in coordinator.data["targets"].items():
info = target_data["info"]
if info.get("target_type") == TARGET_TYPE_KEY_COLORS:
# KC target — brightness lives in key_colors_settings
entities.append(
WLEDScreenControllerKCBrightness(
coordinator, target_id, entry.entry_id,
)
)
continue
# LED target — brightness lives on the device
device_id = info.get("device_id", "")
if not device_id:
continue
device_data = devices.get(device_id)
if not device_data:
continue
capabilities = device_data.get("info", {}).get("capabilities") or []
if "brightness_control" not in capabilities or "static_color" in capabilities:
continue
entities.append(
WLEDScreenControllerBrightness(
coordinator, target_id, device_id, entry.entry_id,
)
)
async_add_entities(entities)
class WLEDScreenControllerBrightness(CoordinatorEntity, NumberEntity):
"""Brightness control for an LED device associated with a target."""
_attr_has_entity_name = True
_attr_native_min_value = 0
_attr_native_max_value = 255
_attr_native_step = 1
_attr_mode = NumberMode.SLIDER
_attr_icon = "mdi:brightness-6"
def __init__(
self,
coordinator: WLEDScreenControllerCoordinator,
target_id: str,
device_id: str,
entry_id: str,
) -> None:
"""Initialize the brightness number."""
super().__init__(coordinator)
self._target_id = target_id
self._device_id = device_id
self._entry_id = entry_id
self._attr_unique_id = f"{target_id}_brightness"
self._attr_translation_key = "brightness"
@property
def device_info(self) -> dict[str, Any]:
"""Return device information."""
return {"identifiers": {(DOMAIN, self._target_id)}}
@property
def native_value(self) -> float | None:
"""Return the current brightness value."""
if not self.coordinator.data:
return None
device_data = self.coordinator.data.get("devices", {}).get(self._device_id)
if not device_data:
return None
return device_data.get("brightness")
@property
def available(self) -> bool:
"""Return if entity is available."""
if not self.coordinator.data:
return False
targets = self.coordinator.data.get("targets", {})
devices = self.coordinator.data.get("devices", {})
return self._target_id in targets and self._device_id in devices
async def async_set_native_value(self, value: float) -> None:
"""Set brightness value."""
await self.coordinator.set_brightness(self._device_id, int(value))
class WLEDScreenControllerKCBrightness(CoordinatorEntity, NumberEntity):
"""Brightness control for a Key Colors target."""
_attr_has_entity_name = True
_attr_native_min_value = 0
_attr_native_max_value = 255
_attr_native_step = 1
_attr_mode = NumberMode.SLIDER
_attr_icon = "mdi:brightness-6"
def __init__(
self,
coordinator: WLEDScreenControllerCoordinator,
target_id: str,
entry_id: str,
) -> None:
"""Initialize the KC brightness number."""
super().__init__(coordinator)
self._target_id = target_id
self._entry_id = entry_id
self._attr_unique_id = f"{target_id}_brightness"
self._attr_translation_key = "brightness"
@property
def device_info(self) -> dict[str, Any]:
"""Return device information."""
return {"identifiers": {(DOMAIN, self._target_id)}}
@property
def native_value(self) -> float | None:
"""Return the current brightness value (0-255)."""
if not self.coordinator.data:
return None
target_data = self.coordinator.data.get("targets", {}).get(self._target_id)
if not target_data:
return None
kc_settings = target_data.get("info", {}).get("key_colors_settings") or {}
brightness_float = kc_settings.get("brightness", 1.0)
return round(brightness_float * 255)
@property
def available(self) -> bool:
"""Return if entity is available."""
if not self.coordinator.data:
return False
return self._target_id in self.coordinator.data.get("targets", {})
async def async_set_native_value(self, value: float) -> None:
"""Set brightness value."""
await self.coordinator.set_kc_brightness(self._target_id, int(value))
@@ -1,177 +0,0 @@
"""Select platform for LED Screen Controller (CSS source & brightness source)."""
from __future__ import annotations
import logging
from typing import Any
from homeassistant.components.select import SelectEntity
from homeassistant.config_entries import ConfigEntry
from homeassistant.core import HomeAssistant
from homeassistant.helpers.entity_platform import AddEntitiesCallback
from homeassistant.helpers.update_coordinator import CoordinatorEntity
from .const import DOMAIN, DATA_COORDINATOR, TARGET_TYPE_KEY_COLORS
from .coordinator import WLEDScreenControllerCoordinator
_LOGGER = logging.getLogger(__name__)
NONE_OPTION = "— None —"
async def async_setup_entry(
hass: HomeAssistant,
entry: ConfigEntry,
async_add_entities: AddEntitiesCallback,
) -> None:
"""Set up LED Screen Controller select entities."""
data = hass.data[DOMAIN][entry.entry_id]
coordinator: WLEDScreenControllerCoordinator = data[DATA_COORDINATOR]
entities: list[SelectEntity] = []
if coordinator.data and "targets" in coordinator.data:
for target_id, target_data in coordinator.data["targets"].items():
info = target_data["info"]
# Only LED targets
if info.get("target_type") == TARGET_TYPE_KEY_COLORS:
continue
entities.append(
CSSSourceSelect(coordinator, target_id, entry.entry_id)
)
entities.append(
BrightnessSourceSelect(coordinator, target_id, entry.entry_id)
)
async_add_entities(entities)
class CSSSourceSelect(CoordinatorEntity, SelectEntity):
"""Select entity for choosing a color strip source for an LED target."""
_attr_has_entity_name = True
_attr_icon = "mdi:palette"
def __init__(
self,
coordinator: WLEDScreenControllerCoordinator,
target_id: str,
entry_id: str,
) -> None:
super().__init__(coordinator)
self._target_id = target_id
self._entry_id = entry_id
self._attr_unique_id = f"{target_id}_css_source"
self._attr_translation_key = "color_strip_source"
@property
def device_info(self) -> dict[str, Any]:
return {"identifiers": {(DOMAIN, self._target_id)}}
@property
def options(self) -> list[str]:
if not self.coordinator.data:
return []
sources = self.coordinator.data.get("css_sources") or []
return [s["name"] for s in sources]
@property
def current_option(self) -> str | None:
if not self.coordinator.data:
return None
target_data = self.coordinator.data.get("targets", {}).get(self._target_id)
if not target_data:
return None
current_id = target_data["info"].get("color_strip_source_id", "")
sources = self.coordinator.data.get("css_sources") or []
for s in sources:
if s["id"] == current_id:
return s["name"]
return None
@property
def available(self) -> bool:
if not self.coordinator.data:
return False
return self._target_id in self.coordinator.data.get("targets", {})
async def async_select_option(self, option: str) -> None:
source_id = self._name_to_id_map().get(option)
if source_id is None:
_LOGGER.error("CSS source not found: %s", option)
return
await self.coordinator.update_target(
self._target_id, color_strip_source_id=source_id
)
def _name_to_id_map(self) -> dict[str, str]:
sources = (self.coordinator.data or {}).get("css_sources") or []
return {s["name"]: s["id"] for s in sources}
class BrightnessSourceSelect(CoordinatorEntity, SelectEntity):
"""Select entity for choosing a brightness value source for an LED target."""
_attr_has_entity_name = True
_attr_icon = "mdi:brightness-auto"
def __init__(
self,
coordinator: WLEDScreenControllerCoordinator,
target_id: str,
entry_id: str,
) -> None:
super().__init__(coordinator)
self._target_id = target_id
self._entry_id = entry_id
self._attr_unique_id = f"{target_id}_brightness_source"
self._attr_translation_key = "brightness_source"
@property
def device_info(self) -> dict[str, Any]:
return {"identifiers": {(DOMAIN, self._target_id)}}
@property
def options(self) -> list[str]:
if not self.coordinator.data:
return [NONE_OPTION]
sources = self.coordinator.data.get("value_sources") or []
return [NONE_OPTION] + [s["name"] for s in sources]
@property
def current_option(self) -> str | None:
if not self.coordinator.data:
return None
target_data = self.coordinator.data.get("targets", {}).get(self._target_id)
if not target_data:
return None
current_id = target_data["info"].get("brightness_value_source_id", "")
if not current_id:
return NONE_OPTION
sources = self.coordinator.data.get("value_sources") or []
for s in sources:
if s["id"] == current_id:
return s["name"]
return NONE_OPTION
@property
def available(self) -> bool:
if not self.coordinator.data:
return False
return self._target_id in self.coordinator.data.get("targets", {})
async def async_select_option(self, option: str) -> None:
if option == NONE_OPTION:
source_id = ""
else:
name_map = {
s["name"]: s["id"]
for s in (self.coordinator.data or {}).get("value_sources") or []
}
source_id = name_map.get(option)
if source_id is None:
_LOGGER.error("Value source not found: %s", option)
return
await self.coordinator.update_target(
self._target_id, brightness_value_source_id=source_id
)
@@ -1,273 +0,0 @@
"""Sensor platform for LED Screen Controller."""
from __future__ import annotations
import logging
from collections.abc import Callable
from typing import Any
from homeassistant.components.sensor import (
SensorDeviceClass,
SensorEntity,
SensorStateClass,
)
from homeassistant.config_entries import ConfigEntry
from homeassistant.core import HomeAssistant
from homeassistant.helpers.entity_platform import AddEntitiesCallback
from homeassistant.helpers.update_coordinator import CoordinatorEntity
from .const import (
DOMAIN,
TARGET_TYPE_KEY_COLORS,
DATA_COORDINATOR,
DATA_WS_MANAGER,
)
from .coordinator import WLEDScreenControllerCoordinator
from .ws_manager import KeyColorsWebSocketManager
_LOGGER = logging.getLogger(__name__)
async def async_setup_entry(
hass: HomeAssistant,
entry: ConfigEntry,
async_add_entities: AddEntitiesCallback,
) -> None:
"""Set up LED Screen Controller sensors."""
data = hass.data[DOMAIN][entry.entry_id]
coordinator: WLEDScreenControllerCoordinator = data[DATA_COORDINATOR]
ws_manager: KeyColorsWebSocketManager = data[DATA_WS_MANAGER]
entities: list[SensorEntity] = []
if coordinator.data and "targets" in coordinator.data:
for target_id, target_data in coordinator.data["targets"].items():
entities.append(
WLEDScreenControllerFPSSensor(coordinator, target_id, entry.entry_id)
)
entities.append(
WLEDScreenControllerStatusSensor(
coordinator, target_id, entry.entry_id
)
)
# Add color sensors for Key Colors targets
info = target_data["info"]
if info.get("target_type") == TARGET_TYPE_KEY_COLORS:
rectangles = target_data.get("rectangles", [])
for rect in rectangles:
entities.append(
WLEDScreenControllerColorSensor(
coordinator=coordinator,
ws_manager=ws_manager,
target_id=target_id,
rectangle_name=rect["name"],
entry_id=entry.entry_id,
)
)
async_add_entities(entities)
class WLEDScreenControllerFPSSensor(CoordinatorEntity, SensorEntity):
"""FPS sensor for a LED Screen Controller target."""
_attr_has_entity_name = True
_attr_state_class = SensorStateClass.MEASUREMENT
_attr_native_unit_of_measurement = "FPS"
_attr_icon = "mdi:speedometer"
_attr_suggested_display_precision = 1
def __init__(
self,
coordinator: WLEDScreenControllerCoordinator,
target_id: str,
entry_id: str,
) -> None:
"""Initialize the sensor."""
super().__init__(coordinator)
self._target_id = target_id
self._entry_id = entry_id
self._attr_unique_id = f"{target_id}_fps"
self._attr_translation_key = "fps"
@property
def device_info(self) -> dict[str, Any]:
"""Return device information."""
return {"identifiers": {(DOMAIN, self._target_id)}}
@property
def native_value(self) -> float | None:
"""Return the FPS value."""
target_data = self._get_target_data()
if not target_data or not target_data.get("state"):
return None
state = target_data["state"]
if not state.get("processing"):
return None
return state.get("fps_actual")
@property
def extra_state_attributes(self) -> dict[str, Any]:
"""Return additional attributes."""
target_data = self._get_target_data()
if not target_data or not target_data.get("state"):
return {}
return {"fps_target": target_data["state"].get("fps_target")}
@property
def available(self) -> bool:
"""Return if entity is available."""
return self._get_target_data() is not None
def _get_target_data(self) -> dict[str, Any] | None:
if not self.coordinator.data:
return None
return self.coordinator.data.get("targets", {}).get(self._target_id)
class WLEDScreenControllerStatusSensor(CoordinatorEntity, SensorEntity):
"""Status sensor for a LED Screen Controller target."""
_attr_has_entity_name = True
_attr_icon = "mdi:information-outline"
_attr_device_class = SensorDeviceClass.ENUM
_attr_options = ["processing", "idle", "error", "unavailable"]
def __init__(
self,
coordinator: WLEDScreenControllerCoordinator,
target_id: str,
entry_id: str,
) -> None:
"""Initialize the sensor."""
super().__init__(coordinator)
self._target_id = target_id
self._entry_id = entry_id
self._attr_unique_id = f"{target_id}_status"
self._attr_translation_key = "status"
@property
def device_info(self) -> dict[str, Any]:
"""Return device information."""
return {"identifiers": {(DOMAIN, self._target_id)}}
@property
def native_value(self) -> str:
"""Return the status."""
target_data = self._get_target_data()
if not target_data:
return "unavailable"
state = target_data.get("state")
if not state:
return "unavailable"
if state.get("processing"):
errors = state.get("errors", [])
if errors:
return "error"
return "processing"
return "idle"
@property
def available(self) -> bool:
"""Return if entity is available."""
return self._get_target_data() is not None
def _get_target_data(self) -> dict[str, Any] | None:
if not self.coordinator.data:
return None
return self.coordinator.data.get("targets", {}).get(self._target_id)
class WLEDScreenControllerColorSensor(CoordinatorEntity, SensorEntity):
"""Color sensor reporting the extracted screen color for a Key Colors rectangle."""
_attr_has_entity_name = True
_attr_icon = "mdi:palette"
def __init__(
self,
coordinator: WLEDScreenControllerCoordinator,
ws_manager: KeyColorsWebSocketManager,
target_id: str,
rectangle_name: str,
entry_id: str,
) -> None:
"""Initialize the color sensor."""
super().__init__(coordinator)
self._target_id = target_id
self._rectangle_name = rectangle_name
self._ws_manager = ws_manager
self._entry_id = entry_id
self._unregister_ws: Callable[[], None] | None = None
sanitized = rectangle_name.lower().replace(" ", "_").replace("-", "_")
self._attr_unique_id = f"{target_id}_{sanitized}_color"
self._attr_translation_key = "rectangle_color"
self._attr_translation_placeholders = {"rectangle_name": rectangle_name}
@property
def device_info(self) -> dict[str, Any]:
"""Return device information."""
return {"identifiers": {(DOMAIN, self._target_id)}}
async def async_added_to_hass(self) -> None:
"""Register WS callback when entity is added."""
await super().async_added_to_hass()
self._unregister_ws = self._ws_manager.register_callback(
self._target_id, self._handle_color_update
)
async def async_will_remove_from_hass(self) -> None:
"""Unregister WS callback when entity is removed."""
if self._unregister_ws:
self._unregister_ws()
self._unregister_ws = None
await super().async_will_remove_from_hass()
def _handle_color_update(self, colors: dict) -> None:
"""Handle incoming color update from WebSocket."""
if self._rectangle_name in colors:
self.async_write_ha_state()
@property
def native_value(self) -> str | None:
"""Return the hex color string (e.g. #FF8800)."""
color = self._get_color()
if color is None:
return None
return f"#{color['r']:02X}{color['g']:02X}{color['b']:02X}"
@property
def extra_state_attributes(self) -> dict[str, Any]:
"""Return r, g, b, brightness as attributes."""
color = self._get_color()
if color is None:
return {}
r, g, b = color["r"], color["g"], color["b"]
brightness = int(0.299 * r + 0.587 * g + 0.114 * b)
return {
"r": r,
"g": g,
"b": b,
"brightness": brightness,
"rgb_color": [r, g, b],
}
@property
def available(self) -> bool:
"""Return if entity is available."""
return self._get_target_data() is not None
def _get_color(self) -> dict[str, int] | None:
"""Get the current color for this rectangle from WS manager."""
target_data = self._get_target_data()
if not target_data or not target_data.get("state"):
return None
if not target_data["state"].get("processing"):
return None
colors = self._ws_manager.get_latest_colors(self._target_id)
return colors.get(self._rectangle_name)
def _get_target_data(self) -> dict[str, Any] | None:
if not self.coordinator.data:
return None
return self.coordinator.data.get("targets", {}).get(self._target_id)
@@ -1,19 +0,0 @@
set_leds:
name: Set LEDs
description: Push segment data to an api_input color strip source
fields:
source_id:
name: Source ID
description: The api_input CSS source ID (e.g., css_abc12345)
required: true
selector:
text:
segments:
name: Segments
description: >
List of segment objects. Each segment has: start (int), length (int),
mode ("solid"/"per_pixel"/"gradient"), color ([R,G,B] for solid),
colors ([[R,G,B],...] for per_pixel/gradient)
required: true
selector:
object:
@@ -1,91 +0,0 @@
{
"config": {
"step": {
"user": {
"title": "Set up LED Screen Controller",
"description": "Enter the URL and API key for your LED Screen Controller server.",
"data": {
"server_name": "Server Name",
"server_url": "Server URL",
"api_key": "API Key"
},
"data_description": {
"server_name": "Display name for this server in Home Assistant",
"server_url": "URL of your LED Screen Controller server (e.g., http://192.168.1.100:8080)",
"api_key": "API key from your server's configuration file"
}
}
},
"error": {
"cannot_connect": "Failed to connect to server.",
"invalid_api_key": "Invalid API key.",
"unknown": "Unexpected error occurred."
},
"abort": {
"already_configured": "This server is already configured."
}
},
"entity": {
"button": {
"activate_scene": {
"name": "{scene_name}"
}
},
"light": {
"api_input_light": {
"name": "Light"
}
},
"switch": {
"processing": {
"name": "Processing"
}
},
"sensor": {
"fps": {
"name": "FPS"
},
"status": {
"name": "Status",
"state": {
"processing": "Processing",
"idle": "Idle",
"error": "Error",
"unavailable": "Unavailable"
}
},
"rectangle_color": {
"name": "{rectangle_name} Color"
}
},
"number": {
"brightness": {
"name": "Brightness"
}
},
"select": {
"color_strip_source": {
"name": "Color Strip Source"
},
"brightness_source": {
"name": "Brightness Source"
}
}
},
"services": {
"set_leds": {
"name": "Set LEDs",
"description": "Push segment data to an api_input color strip source.",
"fields": {
"source_id": {
"name": "Source ID",
"description": "The api_input CSS source ID (e.g., css_abc12345)."
},
"segments": {
"name": "Segments",
"description": "List of segment objects with start, length, mode, and color/colors fields."
}
}
}
}
}
@@ -1,109 +0,0 @@
"""Switch platform for LED Screen Controller."""
from __future__ import annotations
import logging
from typing import Any
from homeassistant.components.switch import SwitchEntity
from homeassistant.config_entries import ConfigEntry
from homeassistant.core import HomeAssistant
from homeassistant.helpers.entity_platform import AddEntitiesCallback
from homeassistant.helpers.update_coordinator import CoordinatorEntity
from .const import DOMAIN, DATA_COORDINATOR
from .coordinator import WLEDScreenControllerCoordinator
_LOGGER = logging.getLogger(__name__)
async def async_setup_entry(
hass: HomeAssistant,
entry: ConfigEntry,
async_add_entities: AddEntitiesCallback,
) -> None:
"""Set up LED Screen Controller switches."""
data = hass.data[DOMAIN][entry.entry_id]
coordinator: WLEDScreenControllerCoordinator = data[DATA_COORDINATOR]
entities = []
if coordinator.data and "targets" in coordinator.data:
for target_id, target_data in coordinator.data["targets"].items():
entities.append(
WLEDScreenControllerSwitch(coordinator, target_id, entry.entry_id)
)
async_add_entities(entities)
class WLEDScreenControllerSwitch(CoordinatorEntity, SwitchEntity):
"""Representation of a LED Screen Controller target processing switch."""
_attr_has_entity_name = True
def __init__(
self,
coordinator: WLEDScreenControllerCoordinator,
target_id: str,
entry_id: str,
) -> None:
"""Initialize the switch."""
super().__init__(coordinator)
self._target_id = target_id
self._entry_id = entry_id
self._attr_unique_id = f"{target_id}_processing"
self._attr_translation_key = "processing"
self._attr_icon = "mdi:television-ambient-light"
@property
def device_info(self) -> dict[str, Any]:
"""Return device information."""
return {"identifiers": {(DOMAIN, self._target_id)}}
@property
def is_on(self) -> bool:
"""Return true if processing is active."""
target_data = self._get_target_data()
if not target_data or not target_data.get("state"):
return False
return target_data["state"].get("processing", False)
@property
def available(self) -> bool:
"""Return if entity is available."""
return self._get_target_data() is not None
@property
def extra_state_attributes(self) -> dict[str, Any]:
"""Return additional state attributes."""
target_data = self._get_target_data()
if not target_data:
return {}
attrs: dict[str, Any] = {"target_id": self._target_id}
state = target_data.get("state") or {}
metrics = target_data.get("metrics") or {}
if state:
attrs["fps_target"] = state.get("fps_target")
attrs["fps_actual"] = state.get("fps_actual")
if metrics:
attrs["frames_processed"] = metrics.get("frames_processed")
attrs["errors_count"] = metrics.get("errors_count")
attrs["uptime_seconds"] = metrics.get("uptime_seconds")
return attrs
async def async_turn_on(self, **kwargs: Any) -> None:
"""Start processing."""
await self.coordinator.start_processing(self._target_id)
async def async_turn_off(self, **kwargs: Any) -> None:
"""Stop processing."""
await self.coordinator.stop_processing(self._target_id)
def _get_target_data(self) -> dict[str, Any] | None:
"""Get target data from coordinator."""
if not self.coordinator.data:
return None
return self.coordinator.data.get("targets", {}).get(self._target_id)
@@ -1,75 +0,0 @@
{
"config": {
"step": {
"user": {
"title": "Set up LED Screen Controller",
"description": "Enter the URL and API key for your LED Screen Controller server.",
"data": {
"server_name": "Server Name",
"server_url": "Server URL",
"api_key": "API Key"
},
"data_description": {
"server_name": "Display name for this server in Home Assistant",
"server_url": "URL of your LED Screen Controller server (e.g., http://192.168.1.100:8080)",
"api_key": "API key from your server's configuration file"
}
}
},
"error": {
"cannot_connect": "Failed to connect to server.",
"invalid_api_key": "Invalid API key.",
"unknown": "Unexpected error occurred."
},
"abort": {
"already_configured": "This server is already configured."
}
},
"entity": {
"button": {
"activate_scene": {
"name": "{scene_name}"
}
},
"light": {
"api_input_light": {
"name": "Light"
}
},
"switch": {
"processing": {
"name": "Processing"
}
},
"sensor": {
"fps": {
"name": "FPS"
},
"status": {
"name": "Status",
"state": {
"processing": "Processing",
"idle": "Idle",
"error": "Error",
"unavailable": "Unavailable"
}
},
"rectangle_color": {
"name": "{rectangle_name} Color"
}
},
"number": {
"brightness": {
"name": "Brightness"
}
},
"select": {
"color_strip_source": {
"name": "Color Strip Source"
},
"brightness_source": {
"name": "Brightness Source"
}
}
}
}

Some files were not shown because too many files have changed in this diff Show More