Compare commits

...

204 Commits

Author SHA1 Message Date
bvandeusen 725ddca950 Merge pull request 'Milestone 127: recommendation quality — provenance, tiered mixes, For You v2, tuning lab' (#107) from dev into main
release / Build signed APK (tag releases only) (push) Has been skipped
test-go / test (push) Successful in 34s
release / Build + push container image (push) Successful in 36s
test-web / test (push) Successful in 41s
test-go / integration (push) Successful in 4m41s
2026-07-03 10:13:49 -04:00
bvandeusen 29e1e7c64c test(web/tuning): marker summary appears in tooltips too — use getAllByText
test-web / test (push) Successful in 32s
The knob-turn summary renders both as the sparkline tick tooltips (one
per series) and in the list under the chart; the single-element query
tripped on the duplicates in CI run 1906.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TsF3cNoKrqCYsU78cXC8U6
2026-07-03 09:35:51 -04:00
bvandeusen 9ad4343c76 feat(tuning): weekly trend view — per-surface series + knob-turn markers
test-go / test (push) Successful in 33s
test-web / test (push) Failing after 38s
test-go / integration (push) Successful in 4m44s
The verify half of the tune→verify loop (#1251), on the same admin
Tuning page as the knobs:

- RecommendationWeeklyTrends: weekly per-source outcomes aggregated
  across all users (the knobs are global, so judging a turn needs
  global outcomes — rows carry rates only, no track/user identity),
  with a taste-hit count per bucket: plays whose track's artist has a
  positive weight in the player's current taste profile. That's the
  "cheap recompute" reading — retroactive over the whole window, at
  the cost of profile drift.
- GET /api/admin/recommendation-trends?weeks=N (default 12, cap 52):
  per-family weekly series (skip rate, sample-weighted completion,
  taste-hit rate) plus the tuning-audit markers inside the window.
- Web: sparkline table under the tuning cards — skip rate per week on
  a shared axis with dashed ticks at knob turns, latest-week columns,
  window taste-hit rate, low-volume rows dimmed as anecdote, and a
  plain-text list of the window's tuning changes.

Also fixes the revive unused-parameter lint on the tuning GET handler
that failed CI run 1903 on the previous commit.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TsF3cNoKrqCYsU78cXC8U6
2026-07-03 09:29:33 -04:00
bvandeusen 0d0a8f46b1 feat(tuning): scoring weights → DB-backed admin tuning lab
test-go / test (push) Failing after 14s
test-web / test (push) Successful in 34s
test-go / integration (push) Successful in 4m42s
The recommendation scoring knobs move out of YAML (radio profile) and
out of the systemMixWeights hard-code (daily_mix profile) into
DB-backed settings with live effect (#1250) — the defaults-discovery
lab per decision #1247: the operator turns knobs to find good values,
which then get baked back into shipped defaults; end users and other
operators should never need the card.

- Migration 0040: recommendation_weight_profiles (radio / daily_mix,
  8 weight columns), taste_tuning singleton (engagement half-life +
  completion-curve points), recommendation_tuning_audit (one row per
  change with a {field, old, new} diff — the trend view's markers,
  #1251).
- internal/recsettings: boot reconcile seeds shipped defaults without
  clobbering tuned rows (coverart SettingsService pattern), validates
  patches (bounds, curve ordering), writes audit rows, and pushes
  daily_mix weights + taste config into package playlists. No-op
  patches write no audit row.
- playlists gains SetSystemMixWeights / SetTasteConfig swap points
  under a RWMutex — no signature threading through the producers; the
  scheduler's taste rebuild reads the pushed config.
- Radio reads its weight profile from the service per request; the 8
  weight fields leave config.RecommendationConfig (YAML keeps only
  RecentlyPlayedHours / RadioSize / RadioSizeMax).
- Admin API: GET/PATCH/reset under /api/admin/recommendation-tuning,
  echoing current + shipped values.
- Web: new admin Tuning tab — two weight profiles side by side, taste
  card, per-scope save (changed fields only) + reset, deviation dots
  against shipped defaults.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TsF3cNoKrqCYsU78cXC8U6
2026-07-03 09:22:03 -04:00
bvandeusen 9e02878b61 feat(playlists): For You composition v2 — multi-seed blend + weighted fresh tail
test-go / test (push) Successful in 29s
test-go / integration (push) Successful in 4m37s
Two approved composition changes (#1269), mechanism only — the
taste/fresh share stays data-decided (#1252) and pick_kind
attribution is unchanged.

Multi-seed blending: each day's build now seeds from up to 3 of the
user's top-5 tracks (pickDailySeeds, the generalized daily shuffle)
instead of one rotating anchor, so the mix spans neighborhoods within
a day and stops feeling bipolar as the rotation swings between
dissimilar seeds. Per-seed pools merge first-seen-deduped; the head
is filled best-first under 50/30/20 per-seed quotas (60/40 for two
seeds) so one neighborhood can't monopolize it, with thin-seed quota
spilling best-first.

Score-weighted fresh tail: the tail sample (rank 2*headN onward) was
uniform — the 380th-best candidate as likely as the 101st. It now
uses deterministic Efraimidis-Spirakis keys with weight halving every
50 ranks, so freshness keeps its "you'll probably enjoy this" half
while still rotating daily.

The retired single-seed picker's one other caller, You-might-like,
moves to pickDailySeeds(n=1) — a single neighborhood per day is right
for a short shelf, and the behavior note is inline.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TsF3cNoKrqCYsU78cXC8U6
2026-07-03 09:02:35 -04:00
bvandeusen 48f288e2e5 feat(mixes): tiered rebuilds for New for you + First listens (rule #131)
test-go / test (push) Successful in 29s
test-go / integration (push) Successful in 4m39s
Both mixes move from a single hard eligibility rule to the tiered
ladder, with their tier stamped onto playlist_tracks.pick_kind via the
#1270 provenance pipeline.

New for you (#1267) — consume on play, degrade by stepping back:
- "Consumed" = any track attempted >=30s; played albums leave the mix
  at the next build instead of crowding it until the calendar window
  expires.
- Tier 1: unconsumed albums added <30d by direct-affinity artists.
  Tier 2: unconsumed affinity albums from the wider 30-90d window —
  added while you weren't looking. Tier 3: any unconsumed album added
  <90d, newest first.

First listens (#1268) — track-level "attempted" threshold:
- A 2-second accidental brush no longer disqualifies a whole album;
  "attempted" is duration_played_ms >= 30000 per track.
- Tier 1: albums with zero attempted tracks. Tier 2: barely-attempted
  albums (<=25% of tracks reached 30s), minus the attempted tracks
  themselves. The artist-affinity ordering signal also moves to the
  >=30s definition so skip-only contact doesn't read as trust.

Producer plumbing: fetch adapters map the tier column onto pick kinds,
finishMix propagates PickKind into the persisted candidates, and
rotateForDay now rotates within contiguous same-pick-kind blocks so
daily rotation can't hoist tier-3 filler above tier-1's exact fits
(untiered pools are one block — original behavior).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TsF3cNoKrqCYsU78cXC8U6
2026-07-03 08:54:20 -04:00
bvandeusen 2be07ef271 fix(mixes): close three intent gaps found in the system-playlists audit
test-go / test (push) Successful in 29s
test-go / integration (push) Successful in 4m36s
Three discovery-mix defects from the intent audit (Scribe note #1254),
all sharing the same root pattern — skips treated as non-events:

- Deep Cuts (#1257): eligibility counted only unskipped plays, so a
  track skipped twice with zero completed listens read as "barely
  heard" and kept being re-offered. Tracks with >=2 skips no longer
  qualify; a single accidental skip doesn't banish.

- Rediscover (#1258): a skip on a rediscover-sourced play — the user
  explicitly declining the resurfacing invitation — changed nothing,
  so declined tracks re-qualified the next day forever. Such tracks
  now sit out 90 days.

- On This Day (#1256): day-of-year distance used plain ABS, so
  Dec 28 vs Jan 3 read as 359 days apart and the window silently
  gutted itself for ~3 weeks around every New Year. Now circular
  (LEAST(d, 365-d)), anchored on the build-date parameter instead of
  now() so it's testable and consistent with the mix's daily
  determinism.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TsF3cNoKrqCYsU78cXC8U6
2026-07-03 08:47:01 -04:00
bvandeusen a670840114 fix(playlists): Songs-like mixes no longer vanish after a quiet week
test-go / test (push) Successful in 29s
test-go / integration (push) Successful in 4m30s
PickSeedArtists had a hard 7-day window with no fallback: a week
without listening emptied the seed pool, produceSeedMixes returned
zero playlists, and the daily atomic-replace build deleted every
existing "Songs like X" mix until the user played something again
(#1255).

The query now falls back through widening engagement windows — 7d →
30d → all-time → liked artists — the same tiered shape that fixed the
identical vanish for For You's seeds (PickTopPlayedTracksForUser).
Like-boost scoring is preserved in every tier.

All returned rows share the winning tier, and produceSeedMixes maps it
onto the rule-#131 pick-kind ladder (7d = tier1 exact, 30d = tier2,
all-time/liked = tier3) and stamps the built tracks — the #1270
provenance pipeline then attributes plays and skips to seed freshness,
so the metrics card can say whether stale-seeded mixes actually
perform worse.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TsF3cNoKrqCYsU78cXC8U6
2026-07-03 08:38:54 -04:00
bvandeusen 5faa57634b feat(metrics): provenance as standard — pick_kind for all system mixes
test-go / test (push) Successful in 31s
test-web / test (push) Successful in 38s
test-go / integration (push) Successful in 4m36s
The #1249 mechanism (stamp WHY a track is in the snapshot at build
time, freeze it onto the play at ingestion, break it down in metrics)
generalizes from a For You one-off to the standard for every system
mix (#1270):

- Migration 0039 widens both pick_kind CHECKs (drop + re-add in the
  same change) to taste/fresh + Discover's dormant/cross_user/random
  + tier1-3 for the rule-#131 eligibility ladders.
- GetForYouPickKindForTrack becomes GetSystemPickKindForTrack
  (user, variant, track); ingestion stamps any systemPlaylistSources
  play from its own variant's live snapshot, live + offline paths.
- Discover stamps its candidate bucket on discoverTrack before the
  interleave, making the 40/30/30 allocation measurable; dedup keeps
  the taking bucket's stamp.
- Metrics replace the for_you special-case with one pick-kind
  vocabulary — any family with attributed plays gets a breakdown,
  future stamping mixes need no metrics change.
- Web: breakdown sub-rows are now toggled per surface (collapsed by
  default) so eight stamping mixes don't swamp the card.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TsF3cNoKrqCYsU78cXC8U6
2026-07-02 23:26:52 -04:00
bvandeusen fb4431207d feat(recommendation): For You exploration attribution — taste vs fresh picks
test-go / test (push) Successful in 31s
test-web / test (push) Successful in 37s
test-go / integration (push) Successful in 4m37s
Milestone #127 step 2 (#1249). For You deliberately blends two
populations — a head of top-scored taste picks and a tail sampled from
deeper ranking (the freshness injection) — but the metrics judged it as
one blob, so its skip rate couldn't distinguish "the taste engine is
missing" from "the freshness tax is too high". That number decides the
exploration share before we tune it.

- Migration 0038: nullable pick_kind ('taste'|'fresh') on both
  playlist_tracks (stamped at snapshot build) and play_events (frozen at
  play-ingestion — the snapshot rebuilds daily, so attribution cannot be
  reconstructed at read time).
- Builder: pickHeadAndTail marks head=taste / tail=fresh; the small-pool
  fallback is all taste (top-N-by-score IS the taste mechanism). Other
  variants persist NULL.
- Ingestion: for_you plays (live + offline replay) look the track up in
  the user's current snapshot; not found → unattributed, never guessed.
- Metrics: For You's row gains a breakdown (taste / fresh / earlier
  unattributed plays), parent row stays the sum; web card renders the
  sub-rows indented with the same baseline deltas + low-data dimming.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TsF3cNoKrqCYsU78cXC8U6
2026-07-02 20:39:41 -04:00
bvandeusen 60533073ad feat(metrics): bucketed surface families + manual-plays baseline (#1248, milestone 127)
test-go / test (push) Successful in 31s
test-web / test (push) Successful in 37s
test-go / integration (push) Successful in 4m30s
The recommendation metrics table was observable but not actionable: raw
source strings (album:<uuid> one-offs) drowned the stable surfaces, and
manual plays were excluded so skip rates had no control group.

- SQL: include NULL-source rows (the baseline) and carry completion_n
  so family merges can weight avg_completion correctly.
- Handler buckets raw sources into stable families (radio:<uuid> →
  Radio, album:/artist: → direct plays, etc.) grouped by surface
  intent: go-to / discovery / direct — each band judged against its
  job, since discovery mixes are expected to skip hotter. Families
  under 20 plays are flagged low-confidence, not hidden.
- Settings card renders the baseline row and per-surface deltas in
  percentage points vs baseline (worse-than-baseline deltas in danger
  color), intent hint copy per group, low-data rows dimmed.
- Pure-unit test for the bucketing/merge; DB test updated to the new
  contract (baseline included, radio:<uuid> collapse).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TsF3cNoKrqCYsU78cXC8U6
2026-07-02 18:00:40 -04:00
bvandeusen 4b150a277e fix(recommendation): Rediscover no longer ships a one-song playlist (#1246)
test-go / test (push) Successful in 29s
test-go / integration (push) Successful in 4m35s
Confirmed against prod: exactly one track (17 plays, cold since May 21)
met the c>=5 + 30d-cold bar, and three process defects turned that into
a 1-track playlist instead of the locked placeholder.

- ListRediscoverTracks: collapse the two-tier UNION into one blended
  pool. The old shallow-tier gate (WHERE NOT EXISTS deep) was
  all-or-nothing — one 6-month row suppressed the entire 30-day tier —
  and deep was a strict subset of shallow anyway. Eligibility drops to
  >=3 non-skip plays (on a weeks-old history the >=5-play tracks are
  precisely the ones still in rotation); ordering prefers >=6mo cold,
  then >=5 plays, then raw count.
- Minimum viable mix floor for all five discovery mixes: below
  minLen (15; 5 for the album-coherent NewForYou/FirstListens) the
  variant is withheld so Home renders the 'listen more to unlock'
  placeholder instead of a mix that reads as built-wrong.
- /api/events: clamp client-supplied 'at' to [user.created_at,
  now+5m]. Unbounded client clocks could write arbitrarily old plays
  and poison the 6-month ordering (prod data verified clean — no
  scrub needed).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TsF3cNoKrqCYsU78cXC8U6
2026-07-02 17:29:41 -04:00
bvandeusen d145fee35d Merge pull request 'fix(android/resilience): auto-recover failed loads on reconnect' (#106) from dev into main
android / Build + lint + test (push) Successful in 4m44s
release / Build signed APK (tag releases only) (push) Successful in 4m31s
release / Build + push container image (push) Successful in 16s
2026-07-02 16:45:13 -04:00
bvandeusen 7628330f72 test(android/library): stub SyncController.lastSyncError with a real StateFlow
android / Build + lint + test (push) Successful in 3m45s
The VM now combines lastSyncError into uiState; a relaxed-mock flow
never emits, so the combine never fired and the state sat at Loading.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TsF3cNoKrqCYsU78cXC8U6
2026-07-02 16:10:45 -04:00
bvandeusen 09102dc615 fix(android/resilience): auto-recover failed loads on reconnect (#1245)
android / Build + lint + test (push) Failing after 3m3s
Failed loads used to stay failed forever: no screen ViewModel listened
to server-health recovery, and the cache-first screens swallowed refresh
errors so a cold load against a down server looked like an empty account.

- connectivity/Recovery.kt: recoveries() — per-collector flow of
  down→Healthy transitions; the screen-level half of the idiom
  SyncController/MutationReplayer/DiagnosticsUploader already use.
- Every screen VM now re-runs its load on recovery (cache-first screens
  unconditionally; direct-load screens when sitting in Error).
- Cache-first error surfacing: Home / Playlists / Liked track refresh
  failure; Library reads SyncController.lastSyncError (new) — empty
  cache + failed refresh now renders Error-with-Retry, not welcome copy.
- Requests: 12s poll also retries from Error (was structurally unable
  to escape it — the in-flight predicate required a Success state).
- Search: retry() bypasses the distinctUntilChanged query pipeline so a
  same-text resubmit after a transient failure actually re-runs.
- ArtistDetail: secondary sections (similar artists / top tracks)
  re-fetch on recovery instead of staying silently absent.
- ErrorRetry: LazyColumn wrapper (pull-to-refresh works on error states,
  same rationale as EmptyState) + optional title; adopted on every error
  branch; PlaylistDetail's one-shot ErrorBlock removed in its favor.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TsF3cNoKrqCYsU78cXC8U6
2026-07-02 16:02:14 -04:00
bvandeusen 611715154b Merge pull request 'M9 diagnostics follow-ups: playback relabel, sort, connected fix, per-skip + track-identity' (#105) from dev into main
test-go / test (push) Successful in 38s
test-web / test (push) Successful in 47s
android / Build + lint + test (push) Successful in 4m18s
test-go / integration (push) Successful in 4m39s
release / Build signed APK (tag releases only) (push) Successful in 3m46s
release / Build + push container image (push) Successful in 16s
2026-06-30 19:19:15 -04:00
bvandeusen 392454b249 feat(android/diagnostics): track-identity enrichment + zero stale Sonos state
android / Build + lint + test (push) Successful in 3m27s
Numeric indices wobble across re-casts (offset +1↔0 seen during output
toggling), making "same track?" ambiguous. Enrich both the track_change
event and the heartbeat with local_track_id (TrackRef.id) and sonos_uri
(RemotePlayerState.currentTrackUri — the URL the speaker is actually
streaming), so a desync is unambiguous.

Also fixes the cast→phone stale-state pollution (#1211): sonos_* is now
zeroed unless a remote route is active, via a shared putSonos() helper —
so a just-ended cast's RemotePlayerState can't masquerade as live Sonos
data in the diagnostics.

Refs Scribe M9 (#119), tasks #1210 #1211.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01K55iTxn95BtshocgdE1shW
2026-06-30 19:15:12 -04:00
bvandeusen cdfc79e6ab feat(android/diagnostics): per-skip track-change event
android / Build + lint + test (push) Successful in 3m35s
Heartbeats are 45s apart and missed a rapid skip burst (local_index
16→22 in one gap). Add a 'playback' track_change event emitted on each
queue-index / current-track change, snapshotting local vs Sonos
index+position + server_health + upnp_loading + route — so a transient
skip-induced desync is captured at the instant it happens. (uiState is a
conflated StateFlow, so a very rapid burst may coalesce intermediate
indices; we still get the boundaries + the snapshot.)

Refs Scribe M9 (#119), task #1210.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01K55iTxn95BtshocgdE1shW
2026-06-30 18:48:35 -04:00
bvandeusen 79f2d79a2e feat(diagnostics): 'playback' kind, newest-first sort, fix active-route subtitle
test-go / test (push) Successful in 32s
test-web / test (push) Successful in 41s
android / Build + lint + test (push) Successful in 3m40s
test-go / integration (push) Successful in 4m30s
Relabel (#1204): route + player_state events fire for every output route,
not just UPnP — split them into a new 'playback' kind; 'upnp_sync' now
means genuinely UPnP/Sonos signal (drops, resync). Migration 0037 adds
'playback' to the kind CHECK; server whitelist, Android reporter labels,
and the web kind filter updated.

Web sort: the diagnostics list gains a Newest/Oldest-first sort (default
newest at top); export follows the displayed order.

Fix (#1205): OutputRoute.isConnected was derived from RouteInfo.connectionState,
which stays DISCONNECTED for local SYSTEM routes even when active — so a
connected Bluetooth device showed "Available" and reported connected:false.
The picker subtitle now uses isSelected (route == selected route); the dead
isConnected field is removed and the misleading `connected` field dropped
from the diagnostics route event (it only ever logs the active route).

Refs Scribe M9 (#119), tasks #1204 #1205.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01K55iTxn95BtshocgdE1shW
2026-06-30 16:33:01 -04:00
bvandeusen 96b15b75e6 feat(web/diagnostics): default to recent 500, move time window to Advanced
test-web / test (push) Successful in 39s
The diagnostics view already defaulted to the most recent 500 events (no
window); make that the obvious path. Device/Kind stay primary; the
start/end window + row cap move into a collapsed "Advanced filters"
disclosure (auto-opens when a window is active) with a "Reset to recent
500" action. Caption now states whether you're seeing the recent default
or a windowed slice.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01K55iTxn95BtshocgdE1shW
2026-06-30 12:42:25 -04:00
bvandeusen 23a82fb38d Merge pull request 'M9 — Device diagnostics & debug reporting (connectivity + UPnP desync)' (#104) from dev into main
test-go / test (push) Successful in 35s
test-web / test (push) Successful in 47s
android / Build + lint + test (push) Successful in 4m6s
test-go / integration (push) Successful in 4m37s
release / Build signed APK (tag releases only) (push) Successful in 3m56s
release / Build + push container image (push) Successful in 1m38s
2026-06-29 19:24:16 -04:00
bvandeusen 782f152d37 test(web/admin): AdminTabs now has seven tabs (Diagnostics added)
test-web / test (push) Successful in 31s
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01K55iTxn95BtshocgdE1shW
2026-06-29 19:11:39 -04:00
bvandeusen bffa5b28bd fix(diagnostics): StateFlow distinctUntilChanged build error + AdminUser test fixtures
test-web / test (push) Failing after 33s
android / Build + lint + test (push) Successful in 3m29s
- DiagnosticsReporter.collectServerHealth: drop distinctUntilChanged() on
  networkStatus.state (StateFlow is already distinct; the deprecation
  warning is a hard error under allWarningsAsErrors).
- web users.test.ts: add debug_mode_enabled to the alice/bob AdminUser
  fixtures now that the field is required on the type.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01K55iTxn95BtshocgdE1shW
2026-06-29 19:05:50 -04:00
bvandeusen 8a58f07237 fix(android/diagnostics): keep uploader drain within ReturnCount gate
android / Build + lint + test (push) Failing after 2m28s
drainSafe/drain each had 3 returns (detekt ReturnCount ≤ 2). Collapse the
guard clauses and convert drain's loop to a `more` flag — same behavior,
zero/two returns.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01K55iTxn95BtshocgdE1shW
2026-06-29 18:59:05 -04:00
bvandeusen 4d42e298dd feat(android+web/diagnostics): on-device debug reporter + admin timeline (M9)
test-web / test (push) Failing after 11s
android / Build + lint + test (push) Failing after 1m19s
Android: a gated DiagnosticsReporter taps connectivity, server-health,
UPnP drops/player-state/route, power (Doze/battery-opt/screen), and
app fg/bg, plus a heartbeat snapshotting Sonos-vs-local position — the
locked-phone desync signal. Events buffer in a Room ring buffer
(deliberately NOT the MutationQueue: high-volume best-effort telemetry
that must survive the dead zone being debugged) and DiagnosticsUploader
drains them on a tick / health-recovery / sign-in.

Gating: the account flag (users.debug_mode_enabled) reaches the device
via a new /api/me refresh in AuthController; a per-device local OFF
switch lives in Settings. Reporter runs only when enabled && !optOut;
disabling drops the unsent buffer.

Web admin: /admin/diagnostics — pick account+device+kind+time-window,
see a chronological timeline, flip an account's debug mode remotely, and
Copy-JSON / Download-NDJSON the slice for analysis.

Room schema 6→7 (new diagnostic_events table + auth_session.diagnosticsOptOut;
pre-v1 destructive fallback).

Refs Scribe M9 (#119), tasks #1174 #1175 #1176 #1177.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01K55iTxn95BtshocgdE1shW
2026-06-29 18:56:36 -04:00
bvandeusen 4ed831d9c3 feat(server/diagnostics): device debug-reporting ingest + admin timeline + retention (M9)
test-go / test (push) Successful in 37s
test-go / integration (push) Successful in 4m44s
New diagnostic_events table + per-account users.debug_mode_enabled flag.
When an account's flag is on, its client(s) POST a batch timeseries of
connectivity / UPnP-sync / power / lifecycle events to /api/diagnostics
(no-op 204 when off, kind whitelist mirrors the CHECK constraint).

Admin surface: GET /api/admin/diagnostics (optional account/device/kind/
time-window filters, RFC3339-or-epoch-ms, export-sized paging) + a
/diagnostics/devices overview + PUT /api/admin/users/{id}/debug-mode to
flip an account remotely while a bug is live. debug_mode_enabled is now
exposed on /api/me (client gate) and the admin user views.

Retention: a 30-day gc-worker sweep (GcPruneDiagnostics), keyed on the
server clock so a skewed device clock can't keep rows alive.

Refs Scribe M9 (#119), tasks #1172 #1173.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01K55iTxn95BtshocgdE1shW
2026-06-29 18:38:56 -04:00
bvandeusen 0de2437689 Merge pull request 'Image rendering + player resilience (#968, #980)' (#103) from dev into main
test-go / test (push) Successful in 46s
test-web / test (push) Successful in 55s
android / Build + lint + test (push) Successful in 4m11s
test-go / integration (push) Successful in 4m44s
release / Build signed APK (tag releases only) (push) Successful in 3m53s
release / Build + push container image (push) Successful in 1m41s
2026-06-20 20:57:39 -04:00
bvandeusen f4f4df7708 feat(android/playlists): stale-view snackbar + Refresh on open system playlist after rebuild
android / Build + lint + test (push) Successful in 3m38s
#980, parity with web e932ab43. When playlist.system_rebuilt arrives (SSE)
while a system-playlist detail screen is open, the ViewModel marks it stale and
the screen shows an indefinite "This mix was refreshed · Refresh" snackbar.
Refresh re-resolves the rotated variant via PlaylistsRepository.systemShuffle
and reuses the existing regenerated navigate-replace flow to land on the fresh
playlist id — without triggering another server rebuild (unlike the manual
regenerate button). Dismiss clears the flag. Functional behaviors were already
correct; this closes the cosmetic stale-list gap.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-20 19:52:13 -04:00
bvandeusen e932ab438c feat(web/playlists): stale-view banner + Refresh on an open system-playlist after rebuild
test-web / test (push) Successful in 39s
#980. When the daily rebuild fires while a system-playlist detail page is
open, its cached data goes stale and can't be refetched in place — the
playlist id rotated, so the old id 404s. serverEvents now exposes a monotonic
rebuild counter; the detail page shows a "this mix was refreshed" banner with
a Refresh that re-resolves the variant (systemShuffle) to the new playlist id
and navigates there. No forced redirect, no auto-reload — the user refreshes
on their terms. Functional behaviors were already correct (tapping a song
plays it; tiles load the current mix); this closes the cosmetic list-staleness.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-20 19:48:52 -04:00
bvandeusen d05264ff80 fix(web/playlists): attribute source when playing a system playlist from its detail page
test-web / test (push) Successful in 38s
Playing a system playlist from /playlists/<id> previously sent no source, so
it never advanced that playlist's rotation — inconsistent with the home tile
(and the Android detail screen, which already tags the variant). Pass
source: variant alongside the existing self-heal closure so a play is
attributed regardless of the surface it started from. Issue #968.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-20 17:35:45 -04:00
bvandeusen a23e2e36ca feat(android/home): refresh Home on playlist.system_rebuilt
android / Build + lint + test (push) Successful in 3m46s
Parity with the web SSE consumer (5a80a1e4). HomeViewModel now subscribes to
EventsStream and re-pulls Home (refreshIndex + system-playlist status) when
the server emits playlist.system_rebuilt — the daily 03:00 rebuild or a
manual refresh — so the system-playlist tiles and You-might-like rows reflect
the new snapshot without a manual reload. Browse-only: the active playback
queue is left to self-heal on the failure path. Issue #968.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-20 15:46:35 -04:00
bvandeusen 5a80a1e460 feat(web): subscribe to SSE and refresh home/playlists on playlist.system_rebuilt
test-web / test (push) Successful in 39s
The web client only ever SENT events; it had no inbound SSE listener, so a
tab left open across the daily system-playlist rebuild kept showing
yesterday's home + playlist snapshots until a manual reload (the stale-
browse-view bug behind #968). Add useServerEvents(): opens /api/events/stream
while authenticated and, on playlist.system_rebuilt, invalidates the home,
playlists, and system-playlist-status query caches. Deliberately does not
disturb the active playback queue — that self-heals on the failure path.
Issue #968.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-20 15:44:05 -04:00
bvandeusen 16f76ea707 feat(server): emit playlist.system_rebuilt on daily + manual system-playlist rebuild
test-go / test (push) Successful in 28s
test-go / integration (push) Successful in 4m27s
The daily 03:00 scheduler rebuild (and the manual refresh endpoint) replace
a user's system playlists + You-might-like rows but published no event, so a
client left open across the rebuild served yesterday's snapshot until a
manual reload — the stale-tab case behind #968. Add a user-scoped
playlist.system_rebuilt event (envelope {kind,user_id,data:{}}) from both the
scheduler (bus threaded into NewScheduler) and handleSystemPlaylistRefresh.
Clients consume it to invalidate home / system-playlist views and proactively
re-pull a stale active queue. Issue #968.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-20 12:59:27 -04:00
bvandeusen d4cc177db4 feat(android/player): self-heal a stale system-playlist / radio queue on total failure
android / Build + lint + test (push) Successful in 3m56s
Web parity with 27766ae0. When a load error exhausts a fully-unplayable
queue, re-pull the source instead of stopping: a bare-variant source is a
refreshable system playlist (re-pull via PlaylistsRepository.systemShuffle),
"radio:<seed>" re-seeds via RadioController. Reads the source from the
current MediaItem extra; bounded to one re-pull per exhaustion (reset when
a track next loads with real audio) so a still-stale refresh can't loop.
Album / artist / user-playlist / offline sources have nothing to refresh
and still stop. Issue #968.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-20 12:56:01 -04:00
bvandeusen 27766ae063 feat(web/player): self-heal a stale system-playlist / radio queue on total failure
test-web / test (push) Successful in 39s
When the whole queue proves unplayable (e.g. a tab left open across the
daily system-playlist rebuild — the exact stale-snapshot case), the player
now re-pulls the fresh snapshot and resumes instead of dead-ending on
"Try again". The seeder hands the store an opaque refetch closure so the
store stays decoupled from the playlist API and the per-artist
(songs_like_artist) identity problem: single-instance variants re-pull via
systemShuffle, per-artist mixes via getPlaylist(id), radio re-seeds from
its track. Bounded to one self-heal per exhaustion (reset on the next
successful play) so a still-broken refresh can't loop; "Try again" stays
the genuine last resort. Wired from PlaylistCard, the playlist detail page,
and playRadio. Issue #968.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-20 12:51:47 -04:00
bvandeusen 335d782215 fix(android/player): auto-skip a failed track on load error, not just zero-duration
android / Build + lint + test (push) Successful in 3m47s
onPlayerError fired a `load_failed` event and the snackbar reporter coalesced
it into "Skipped N unplayable tracks" — but nothing actually skipped, so a
bad/stale track stranded playback while the toast claimed otherwise. Mirror
the zero_duration path: advance to the next item and re-prepare (a load error
leaves the player IDLE), or stop at the end. Forward-only bounds a fully-
unplayable queue. Web parity with 2a8de82a. Issue #968.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-20 12:41:30 -04:00
bvandeusen 2a8de82a17 fix(web/player): auto-skip a failed track instead of dead-ending on "Try again"
test-web / test (push) Successful in 39s
A track that fails to load (e.g. a stale system-playlist snapshot pointing
at a rebuilt/removed file) hard-set the player to the 'error' state and
stranded the user on a "Try again" button that just re-queued the same
failing track. Now a load error advances to the next track; the error
state only surfaces once the whole queue has proven unplayable — every
track failed, or we reached the end. A failure streak capped at queue
length stops a fully-broken queue from cycling, and resets on the next
successful play.

Next (Track B cont.): self-heal a stale system-playlist / radio queue by
re-pulling the fresh snapshot on total failure, plus the Android
equivalent. Issue #968.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-20 12:40:08 -04:00
bvandeusen 096a3c0b15 fix(clients): never leave cover tiles blank — shared web <Cover> + Android Coil placeholder/error
test-web / test (push) Successful in 48s
android / Build + lint + test (push) Successful in 4m10s
Cover tiles (worst in the "You might like" home row, which surfaces
unplayed items whose art is often not yet backfilled) sat empty while
loading and stayed blank on a 404. The server returns a fast 404; the
gap was missing client-side loading/fallback states.

Web: new shared Cover.svelte owns the loading placeholder + onerror
fallback (static cover, or Disc3 for artists). AlbumCard, ArtistCard and
CompactTrackCard now reuse it instead of three hand-rolled <img> tags
that disagreed on fallback handling — notably ArtistCard had no onerror.

Android: ServerImage tracks Coil's load state so the per-caller fallback
doubles as a placeholder (loading) and an error state (404 / unreachable),
instead of only guarding the null-URL case. All five call sites pass an
explicit size modifier, so the new Box wrapper is layout-safe.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-20 12:29:05 -04:00
bvandeusen a251dce7e3 Merge pull request 'feat: scan on startup by default + README first-run walkthrough & screenshots' (#102) from dev into main
release / Build signed APK (tag releases only) (push) Has been skipped
test-go / test (push) Successful in 37s
release / Build + push container image (push) Successful in 1m44s
test-go / integration (push) Successful in 4m35s
2026-06-20 11:40:56 -04:00
bvandeusen 97e0e88483 feat(server): scan library on startup by default + README first-run walkthrough
test-go / test (push) Successful in 37s
test-go / integration (push) Successful in 4m38s
Make a fresh install usable out of the box and document the first-run flow
for the public-facing repo.

- Default scan_on_startup to true (Default() + config.example.yaml, which is
  the live config baked into the image). Previously false, so a fresh stack
  came up with an empty library and no hint to scan. Scans are incremental
  (mtime skip), so the per-restart cost is just a directory walk. Re-point
  the env-override test to exercise the override against the new default.
- README: add a "First run" walkthrough (register -> scan -> integrations ->
  install Android app -> invite users), each grounded in a real route.
- Add docs/screenshots/ with six captures, referenced via width-constrained
  <img> wrapped in a link (shrink inline + click to open full size).
  API token and invite token were cropped/redacted out of the captures
  before commit so no live credential lands in the public history.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-20 11:35:22 -04:00
bvandeusen 938dae7163 Merge pull request 'docs: correct README setup/OOBE + fix data-volume mount' (#101) from dev into main 2026-06-20 10:58:07 -04:00
bvandeusen ad37937949 docs: correct README setup/OOBE + fix data-volume mount
Bring the README and client/README in line with the current product and
release pipeline ahead of making the repo public-facing:

- Highlights: replace the retired "Flutter client in flight" bullet with
  the native-Android-APK-bundled-in-the-image story.
- Quickstart: document the purpose of each compose volume mount, and fix
  the minstrel-data mount (/data -> /app/data) so it matches the image's
  MINSTREL_STORAGE_DATA_DIR and generated artefacts actually persist.
- Configuration: tie MINSTREL_STORAGE_DATA_DIR to the /app/data mount.
- Updating: replace the bogus :v1.0.x scheme with the real tag model
  (:latest, immutable :vYYYY.MM.DD, :main) and note the bundled APK.
- client/README: rewrite the Production section to match release.yml
  (android-release + needs: ordering, non-tag :latest bundle path).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-20 10:57:28 -04:00
bvandeusen 93365cb555 Merge pull request 'ci(release): bundle the latest release APK into non-tag :latest builds' (#100) from dev into main
release / Build signed APK (tag releases only) (push) Has been skipped
release / Build + push container image (push) Successful in 17s
2026-06-14 22:55:23 -04:00
bvandeusen 2c5c477a0d ci(release): bundle the latest release APK into non-tag :latest builds
Every main push moves :latest, but main builds don't build an APK — so the
in-app update channel silently vanished from :latest until the next tag.

Now the image-release job, on non-tag builds, pulls the most-recent
release's signed APK from the gitea API and reconstructs its exact
versionName (${TAG#v}.$(git rev-list --count TAG) — the same formula
android-release bakes in) for the version sidecar. No rebuild, just
rebundle; tag builds still bundle their own freshly-built APK. Checkout
gains fetch-depth:0 + fetch-tags so the commit count resolves. Degrades to
an empty client/ (404 update channel) — never a wrong version — if no
release / APK asset / tag count can be resolved.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-14 22:54:54 -04:00
bvandeusen eeabdf1f2c Merge pull request 'Web UI: Most Played hover fix, narrower seek bar, Android-parity track kebab' (#99) from dev into main
release / Build signed APK (tag releases only) (push) Has been skipped
test-web / test (push) Successful in 35s
release / Build + push container image (push) Successful in 1m50s
2026-06-14 22:26:52 -04:00
bvandeusen f7278f2417 feat(web): drop "Remove from library" from the track kebab
test-web / test (push) Successful in 32s
No single-click destructive action belongs in the kebab. Removing the item
orphaned its whole path (RemoveTrackPopover was its only caller, and the
admin/tracks API client was the popover's only caller), so per the repo's
no-dead-code convention the chain is fully removed: the menu item + its
admin/isAdmin plumbing in TrackMenu, RemoveTrackPopover(.svelte/.test),
src/lib/api/admin/tracks(.ts/.test), and the now-needless transitive mocks
in the CompactTrackCard / PlaylistTrackRow / playlist specs.

The kebab is now an 8-item, admin-agnostic menu. The DELETE /api/admin/tracks
server endpoint is untouched — a future safer admin surface can rebind it.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-14 22:17:35 -04:00
bvandeusen 9a166f9032 test(web): update TrackMenu spec for Android-parity kebab
test-web / test (push) Successful in 32s
The kebab gained "Start radio" and dropped the duplicate "Flag this track…"
(its action now lives solely under "Hide", which opens the same FlagPopover).
Net item count is unchanged (9 admin / 8 non-admin), but the named-item and
flag-entry assertions needed updating:
- mock playRadio in the store mock; assert Start radio dispatches playRadio.
- swap the flag-item presence check for start-radio.
- replace the "click Flag" test with "click Hide opens the popover".

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-14 21:48:57 -04:00
bvandeusen b3d6785543 feat(web): match track kebab to Android + add it to full-screen Now Playing
test-web / test (push) Failing after 32s
Bring the web TrackMenu to parity with Android's canonical TrackActionsSheet
so the kebab reads the same on both clients:
- Reorder to Android's groups: queue → like/add-to-playlist/start-radio →
  go-to-album/artist → hide.
- Drop the duplicate "Flag this track…" item — it opened the very same
  FlagPopover as "Hide" (Android folds flag into a single Hide).
- Align icons (ListVideo / ListMusic / ListPlus / Disc3 / User).
- Admin-only "Remove from library" stays as a web superset (Android has no
  surface for it), past its own divider.

Mount the kebab on the full-screen /now-playing route with hideQueueActions,
mirroring Android's NowPlayingScreen — Start radio / Add to playlist / Hide
were previously unreachable there (only like + volume + queue existed).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-14 21:46:54 -04:00
bvandeusen 773275e916 feat(web): surface "Start radio" in the track kebab menu
test-web / test (push) Failing after 31s
Radio was fully wired (playRadio → /api/radio + 80% auto-refresh) but its
only entry point was TrackRow's inline 📻 button, so it was unreachable from
the kebab — i.e. missing on the Most Played compact cards and the mini-player.
Add a "Start radio" item to TrackMenu, shown even under hideQueueActions since
reseeding a station from the current track is meaningful there.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-14 21:40:16 -04:00
bvandeusen 066616e196 fix(web): de-overlap Most Played hover controls + narrow desktop seek bar
test-web / test (push) Successful in 40s
CompactTrackCard is a short one-line row but reused CardActionCluster,
which corner-splits Like+Add (top) and the kebab menu (bottom). That split
is right for the tall square Album/Artist cards but makes the two groups
collide on the compact row's hover state. Give the compact card a single
inline, vertically-centred right cluster (Like + Add + menu in one group)
and widen its right padding reserve to match.

In the desktop PlayerBar, the left info column was a fixed w-72 (title kept
truncating) while the seek column was flex-1 (the scrubber hogged the slack
on wide screens). Let the left column grow up to max-w-md while holding its
288px floor at md, and cap the seek/transport column at max-w-xl centred, so
freed width flows to the title instead of stretching the bar.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-14 21:36:26 -04:00
bvandeusen 3e258507bb Merge pull request 'fix(android): UPnP cast resilience — drop-suppression, session adopt, recovery hardening' (#98) from dev into main
android / Build + lint + test (push) Successful in 4m39s
release / Build signed APK (tag releases only) (push) Successful in 4m31s
release / Build + push container image (push) Successful in 15s
2026-06-12 21:25:25 -04:00
bvandeusen fd7d6dac4c feat(android): adopt a running UPnP session + tighten cast recovery/anti-stickiness
android / Build + lint + test (push) Successful in 3m53s
Three related improvements to UPnP/Sonos session handling, on top of the
WiFi-lock + drop-suppression fixes.

1. Adopt a running session instead of clear+reload (the headline).
   Selecting a renderer always did removeAllTracksFromQueue + full reload --
   a jarring restart if the speaker was already playing our queue (e.g. after
   the phone got disconnected but the autonomous Sonos kept going). selectUpnp
   now probes the renderer first; if it's mid-playback on the same track id at
   the same queue index, we ATTACH in place: sync the local cursor to its
   position, wire ActiveUpnpHolder, start polling -- no clear, no reload, and
   skip/seek immediately drive its live queue. Falls through to clear+reload
   when it isn't our queue. New PlayerController.moveCursorTo aligns the local
   cursor without auto-playing.

2. Discovery expiry + selection revert (anti-stickiness). upsertRoute only
   ever added, so a powered-off renderer lingered in the picker forever and
   could pin a stale selection. Stamp lastSeen per route; after the picker's
   active M-SEARCH scan, prune routes that didn't re-announce. A collector
   reverts the selection to the phone when the selected route leaves discovery
   while we're not actively casting -- so a later play never targets a ghost.
   Pruning is tied to picker-open scans only (no background timer -> no row
   flicker).

3. Reconcile immediately on network recovery. When NetworkStatus flips back to
   Healthy while a route is active, nudge an immediate poll instead of waiting
   up to POLL_INTERVAL_MS -- the held session re-confirms the renderer in one
   round-trip.

Verified (read-only): the tap-play-onto-dead-route fallback still fires when
the phone's network is Healthy (the poll-loop drop path is unchanged for that
case); the drop-suppression gate only holds during phone-side outages, where a
local fallback couldn't play either.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-12 21:20:26 -04:00
bvandeusen d6290a3ef0 fix(android): don't drop Sonos (or blast local audio) when the phone's own network is down
android / Build + lint + test (push) Successful in 3m43s
Observed on device: casting to Sonos on battery + screen locked, a
transient ~67s reachability gap (NetworkStatus -> ServerDown while WiFi
itself stayed associated) starved the 1 Hz poll past DROP_THRESHOLD. The
poll loop then dropped the route and fell back to the local player, which
honored the play-intent -- so the phone suddenly started playing the song
out loud locally while the Sonos was still happily streaming it.

A poll failure during a phone-side network outage means "we can't see the
renderer right now," not "the renderer died": a UPnP renderer streams
autonomously and keeps playing, and the local player we'd fall back to
can't reach the server either. Dropping is strictly worse than waiting.

Gate the drop on NetworkStatusController: only drop when the phone's
network is Healthy (renderer genuinely unreachable on an otherwise-fine
link). While Unstable/ServerDown/Offline, hold the route, keep polling,
and clear the failure streak so recovery re-evaluates from scratch rather
than re-dropping on the first post-recovery hiccup. The poll reconciles to
the renderer's real (advanced) position once the network returns.

Complements the CastNetworkLock fix: the lock reduces how often these gaps
happen; this stops a gap that does happen from punishing the user.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-12 20:48:40 -04:00
bvandeusen 9550d8daaf Merge pull request 'fix(android): hold WiFi+wake lock during UPnP cast (locked-screen poll starvation)' (#97) from dev into main
android / Build + lint + test (push) Successful in 4m5s
release / Build signed APK (tag releases only) (push) Successful in 4m4s
release / Build + push container image (push) Successful in 13s
2026-06-12 19:45:44 -04:00
bvandeusen 4d5af3599e fix(android): hold WiFi+wake lock during UPnP cast so a locked screen doesn't starve the poll
android / Build + lint + test (push) Successful in 3m46s
While casting, the wrapped ExoPlayer is paused, releasing its
WAKE_MODE_NETWORK locks -- so nothing kept the phone's radio awake. On a
locked, on-battery phone the WiFi power-saves within seconds and the CPU
dozes, stalling the 1 Hz liveness poll to the renderer and the
queue-extend calls to the server. The poll then trips DROP_THRESHOLD and
playback falls back to a phone that also has no network: silence, while
the Sonos was streaming fine the whole time.

Diagnosed from logcat: ~12s after screen-off the phone logged
"Unable to resolve host minstrel.fabledsword.com" (its own DNS, not the
server), the extend aborted (1/58 appended), then the drop tripped and
the local fallback came up active=null. USB charging masks it (no Doze
while charging), which is why it only bit on battery.

Add CastNetworkLock: a high-perf/low-latency WifiLock + partial WakeLock
acquired when a UPnP route goes active and released on drop/switch-back
(every teardown path funnels through holder.set(null) -> onActiveChanged).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-12 19:39:42 -04:00
bvandeusen db393bbe65 Merge PR #96: recommendation batch (You-might-like fallback + taste 2b + observability)
test-go / test (push) Successful in 41s
test-web / test (push) Successful in 43s
release / Build signed APK (tag releases only) (push) Successful in 4m2s
test-go / integration (push) Successful in 4m38s
release / Build + push container image (push) Successful in 13s
2026-06-12 01:10:09 -04:00
bvandeusen 1a7515e6ea feat(taste): phase 4 — recommendation observability (#796)
test-go / test (push) Successful in 33s
test-web / test (push) Successful in 41s
test-go / integration (push) Successful in 4m24s
Per-source play outcomes so the operator can see whether each recommendation
surface is landing and tune the now-operator-tunable taste weights.

Server:
- query RecommendationSourceMetricsForUser: groups the user's play_events by
  source (system-playlist surface), reporting plays / skips / avg completion
  over a window; NULL-source (library/radio) plays excluded.
- GET /api/me/recommendation-metrics?days=30 (default 30, capped 365) →
  {window_days, sources:[{source, plays, skips, skip_rate, avg_completion}]}.
- handler test: 401 unauth; per-source aggregation + NULL-source exclusion +
  skip_rate / avg_completion math.

Web:
- lib/api/metrics.ts: query + friendly source labels.
- settings page gains a "Recommendation metrics" card (table of surface / plays
  / skip rate / avg completion), with loading/error/empty states.
- settings tests mock the new query (manual subscribe-store, hoisting-safe).

Note: You-might-like plays aren't source-tagged (it's a Home row, not a system
playlist), so this covers For-You / Discover / the mixes. Tagging YML plays
would be a client follow-up.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-12 00:28:30 -04:00
bvandeusen 6c26ba807e feat(taste): phase 2b — taste_overlap candidate arm (#796)
test-go / test (push) Successful in 37s
test-go / integration (push) Successful in 4m41s
2a re-ranks the existing pool by TasteMatch; this ensures taste-relevant tracks
ARE in the pool. Adds a 6th arm to LoadRadioCandidatesV2: in-library tracks by
the user's top positively-weighted taste-profile artists ($10 K, weight > 0,
deterministic weight-DESC,id order so it doesn't reintroduce same-day
nondeterminism). Pool-inclusion only (sim_score 0) — TasteMatch already scores
the fit. Empty for cold-start users (no profile).

- CandidateSourceLimits.TasteOverlap; default 20 (radio), 80 for For-You via
  systemForYouSourceLimits.
- You-might-like deliberately sets TasteOverlap=0: it surfaces NOT-actively-
  engaged artists, so flooding its pool with top-taste (mostly already-played)
  artists would just feed the read-time dedup.
- Test: positive-weight artist's track enters via the arm; negative-weight one
  is excluded (weight > 0). Existing pool tests unaffected (no profile seeded).

Deferred within 2b: profile-seeded For-You — marginal given the arm + TasteMatch
already inject taste broadly (top-played seed ≈ top-taste artist).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-12 00:05:50 -04:00
bvandeusen c7adf2c87a fix(recommendation): broaden You-might-like fallback to liked album/track artists (#790)
test-go / test (push) Successful in 29s
test-go / integration (push) Successful in 4m36s
The fallback pulled artists only from explicit artist-likes (general_likes_artists),
but most users like albums and tracks far more than artists — so the artists row
still came up thin (a couple of tiles) even with a rich library, while the albums
row filled fine.

Broaden both fallbacks to "entities you've shown affinity for":
- artist fallback = explicit artist-likes ∪ artists of liked albums ∪ artists of
  liked tracks.
- album fallback = explicit album-likes ∪ albums of liked tracks.
New dedicated queries (ListYouMightLike{Artist,Album}FallbackForUser) replace the
narrow Rediscover-fallback reuse; same projection so the Go layer still converts
directly. (Aliased + fully-qualified the UNION arms — sqlc merges UNION scopes,
so unqualified user_id was ambiguous across the three like tables.)

Test: 12 liked TRACKS by distinct artists, no artist-likes → the artist row now
fills from their artists (was empty before).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-11 23:55:08 -04:00
bvandeusen 7b7bd0c3e8 Merge PR #95: You-might-like liked-entity fallback (#790)
test-go / test (push) Successful in 30s
release / Build signed APK (tag releases only) (push) Successful in 3m55s
test-go / integration (push) Successful in 4m38s
release / Build + push container image (push) Successful in 13s
2026-06-11 23:33:25 -04:00
bvandeusen 26c368c35e fix(recommendation): use type conversion for fallback rows (staticcheck S1016)
test-go / test (push) Successful in 28s
test-go / integration (push) Successful in 4m35s
golangci-lint v2 (CI-only; local is v1) flagged the field-by-field struct
literals — the fallback and you-might-like row types are identical, so convert
directly instead.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-11 23:26:35 -04:00
bvandeusen 36786defd1 feat(recommendation): fall back to liked entities when You-might-like is thin (#790)
test-go / test (push) Failing after 14s
test-go / integration (push) Has been cancelled
The taste roll-up surfaces top-similar albums/artists, which for a heavy
listener are mostly ones they already play — so the read-time dedup (vs Most
Played + Rediscover + Last Played) can strip the section down to a single tile
(reported on the artists row). The code was sound; the section was just starved.

Adds a read-time fallback: when a You-might-like row comes up short after dedup,
top it up from the user's LIKED artists/albums — a far larger pool than the
12-entity similarity roll-up, so the same exclusions still leave plenty. Reuses
the existing Rediscover-fallback queries (no new SQL), applies the same
exclusions (already-shown + Rediscover + Most/Last Played) so it never
duplicates a tile or suggests an actively-played entity, and is best-effort
(a query error leaves the section as-is). Takes effect immediately — no rebuild.

A cold-start user with no likes gets nothing from the fallback, so the
new-user-empty behaviour is preserved (test still passes).

Test: 20 liked artists, none played → Rediscover fills 10, You-might-like
fallback fills the other 10, disjoint.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-11 23:24:21 -04:00
bvandeusen 8019537b02 Merge PR #94: "You might like" web client row (#790)
test-web / test (push) Successful in 35s
release / Build signed APK (tag releases only) (push) Successful in 3m50s
release / Build + push container image (push) Successful in 1m56s
2026-06-11 23:01:38 -04:00
bvandeusen 3f240bc777 feat(web): "You might like" Home row (#790 web client slice)
test-web / test (push) Successful in 32s
The web UI rendered a fixed set of Home sections and had no code for the
server's you_might_like_albums / you_might_like_artists (shipped in
v2026.06.11), so the row was absent in the web client. Adds it, mirroring
the Rediscover block, positioned directly under the system-playlists row.

- types.ts HomePayload: two new slices (server always emits them; web ships
  in lockstep with the server).
- +page.svelte: a "You might like" section (albums + artists scrollers) as the
  first section under the playlists row, with a "still learning your taste"
  empty state for the cold-start/gated case. Reuses existing AlbumCard /
  ArtistCard / HorizontalScrollRow.
- home.test.ts / page.test.ts: mock payloads gain the two fields.

Completes the You-might-like row across all three clients (server already
emits it; Android in v2026.06.11; web here).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-11 22:59:58 -04:00
bvandeusen e41d603c12 Merge PR #93: move "You might like" under system-playlists row
android / Build + lint + test (push) Successful in 4m24s
release / Build signed APK (tag releases only) (push) Successful in 4m29s
release / Build + push container image (push) Successful in 14s
2026-06-11 22:55:52 -04:00
bvandeusen 1cc9553d1f fix(android): move "You might like" directly under the system-playlists row
android / Build + lint + test (push) Successful in 3m37s
Operator expected the row immediately beneath the system-generated playlists
section, not below Rediscover. Reorders the section call (presentation only —
no logic/state change) and updates the doc comment.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-11 22:50:02 -04:00
bvandeusen a62f07bd3a Merge PR #92: "You might like" Android client row (#790)
android / Build + lint + test (push) Successful in 4m19s
release / Build signed APK (tag releases only) (push) Successful in 4m12s
release / Build + push container image (push) Successful in 15s
2026-06-11 22:36:22 -04:00
bvandeusen c2f9bbaff4 fix(android): suppress detekt TooManyFunctions on HomeRepository
android / Build + lint + test (push) Successful in 3m32s
The two new you-might-like observe accessors pushed HomeRepository from 11 to
13 functions, tripping detekt's per-class default. It's accessor density (one
observe method per Home row) on a thin pass-through repository, not complexity —
suppressed with a one-line rationale per the project convention. ktlint already
passed; this was the only detekt finding.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-11 22:19:13 -04:00
bvandeusen abc225e12a feat(android): "You might like" Home rows (#790 client slice)
android / Build + lint + test (push) Failing after 1m17s
Surface the server's you_might_like_albums / you_might_like_artists sections
(daily-built, cold-start gated, taste-aware) on the Home screen, mirroring the
Rediscover block.

- HomeIndexWire: two new slices, defaulted to emptyList() so decode is safe
  against older servers that don't emit them (Class-B discipline).
- HomeRepository: two section constants + observeYouMightLikeAlbums/Artists
  (reusing the existing album/artist hydration helpers) + refreshIndex now
  replaces both sections and pre-warms their artists. No Room schema change —
  cached_home_index stores the section string verbatim.
- HomeScreen: HomeSections gains the two fields (+ isAllEmpty); the ViewModel
  combine is split (core 5 → +2 you-might-like → +playlists) to stay within the
  coroutines 5-arity limit; a YouMightLikeBlock + youMightLikeSection render an
  albums-then-artists block below Rediscover, with a "still learning your taste"
  empty state for the cold-start/gated case.

Server side already shipped in v2026.06.11; this makes it visible on device.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-11 22:16:32 -04:00
bvandeusen 3c646c6974 Merge PR #91: "You might like" rows + taste profile (learn + apply)
test-go / test (push) Successful in 29s
test-go / integration (push) Successful in 4m34s
release / Build signed APK (tag releases only) (push) Successful in 4m5s
release / Build + push container image (push) Successful in 15s
2026-06-11 21:41:17 -04:00
bvandeusen aff346c731 feat(taste): phase 2a — apply the taste profile via a TasteMatch scoring term (#796)
test-go / test (push) Successful in 39s
test-go / integration (push) Successful in 4m34s
The profile built in phase 1 now changes what gets surfaced. Adds a TasteMatch
term to the weighted-shuffle score so candidates are re-ranked by their fit to
the user's learned taste (positive draws toward it; negative reflects passive
avoidance; 0 at cold start).

- recommendation/score.go: ScoringInputs.TasteMatchScore ([-1,+1]) +
  ScoringWeights.TasteWeight + the term in Score.
- recommendation/taste.go: LoadTasteProfile reads the taste_profile_* tables;
  TasteProfile.Match blends the candidate's artist weight (0.7) and avg genre-tag
  weight (0.3), each tanh-squashed by a fixed scale so one outlier artist can't
  compress the rest. Unknown artist/tags and empty profiles → 0 (neutral).
- candidates.go: both candidate loaders set TasteMatchScore per candidate, so
  every Score caller (system playlists incl. You-might-like, radio) becomes
  taste-aware automatically.
- weights: systemMixWeights.TasteWeight = 1.5 (daily mixes are the primary
  taste surface); config.RecommendationConfig gains taste_weight (default 1.0,
  lighter — radio is seed-directed) wired into the radio handler.
- tests: pure (Match curve incl. saturation/clamp/empty-neutral, Score term
  add+subtract) + DB round-trip (seed taste rows → Match positive). All green
  vs real Postgres; existing playlist/radio tests unaffected (empty profile →
  zero taste effect).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-11 21:29:42 -04:00
bvandeusen 13b3fca949 fix(taste): drop unused now param + tighten test (golangci-lint revive)
test-go / test (push) Successful in 28s
test-go / integration (push) Successful in 4m25s
golangci-lint v2 (CI-only; local is v1) flagged two unused-parameter issues:
- BuildTasteProfile's `now` was genuinely dead — decay/windowing are computed
  DB-side via now(), so no Go-side timestamp is threaded. Removed it (a
  phase-3 context model that needs a pinned reference time would re-add it);
  updated the scheduler call site.
- the degenerate-params engagement test ignored t; reworked it to assert the
  result stays in [-1,1], which also strengthens the test.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-11 20:58:43 -04:00
bvandeusen 6e0d0e5723 feat(taste): phase 1 — persistent per-user taste profile from graded engagement (#796)
test-go / test (push) Failing after 25s
test-go / integration (push) Has been cancelled
Build a persistent, decaying model of each user's taste, recomputed daily,
that later phases consume across every recommendation surface. Phase 1 only
BUILDS the object — no behaviour change to what's surfaced yet.

Core mechanic — graded engagement (replaces binary was_skipped for learning;
was_skipped stays for History): a play's completion ratio maps to a signal in
[-1,+1] via two linear ramps (instant-skip → -1, ~0.30 neutral, ≥0.90 → +1).
Time-decayed (half-life ~75d) so recent behaviour dominates and the profile
tracks drift.

Per operator constraints:
- No explicit dislike button — negatives come only from passive behaviour
  (early skips). Nothing recorded to regret or opt out of.
- Negatives are track-scoped; artist/tag weight is the decayed SUM of their
  tracks' engagement, so one skip nets out against many good plays (a
  DB test asserts a liked artist stays positive despite an early-skipped
  track). A floor clamp bounds how negative any single entity can get.

- migration 0035: taste_profile_artists / taste_profile_tags (signed weight,
  indexed by (user, weight DESC)).
- internal/taste: engagement.go (pure curve + decay) + profile.go
  (accumulate plays + like bonuses, floor damping, size caps, atomic-replace).
- scheduler: rebuildUserDaily recomputes the profile before the playlist
  build (so phase 2 can read it), best-effort — a taste failure never blocks
  playlist building. Wired into the daily job + startup catch-up only (not
  manual/lazy rebuilds).
- tests: pure (engagement curve, decay, ranking, floor, genre split) +
  DB-backed (positive/negative weights, aggregation-protects-artist, like
  bonus, atomic replace). All green vs real Postgres.

Config knobs live in taste.DefaultConfig() for now; wiring them into the
server RecommendationConfig is a later follow-up.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-11 20:55:14 -04:00
bvandeusen 752906b054 fix(playlists): pin candidate order before jitter to make daily builds deterministic
test-go / test (push) Successful in 29s
test-go / integration (push) Successful in 4m33s
scoreAndSortCandidates drew per-candidate jitter by slice position, but
the candidate query (LoadRadioCandidatesV2) has ORDER BY random() arms and
no stable outer ordering, so DB row order varies call-to-call. When the
recency spread between candidates is smaller than the ±jitter (small or
recency-clustered libraries), two same-day rebuilds assigned jitter to
different tracks and reordered near-ties — so the build was not actually
deterministic-within-a-day as documented.

Pre-existing latent flake in TestBuildSystemPlaylists_DailyNonceDeterminism
(passed in isolation / by luck in CI; deterministically reproduced when the
system-build tests run in sequence). Confirmed independent of the
You-might-like change by neutralizing buildYouMightLike — the flake
persisted.

Fix: sort the candidate slice by track id before assigning jitter, so the
jitter for a track is a function of (track, day) alone, independent of DB
return order. Verified: full playlists package green 4/4 and the build-test
sequence green 5/5 (was 0/4 before).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-11 19:56:13 -04:00
bvandeusen fdd14ef04c feat(server): "You might like" album/artist Home rows (#790)
test-go / test (push) Successful in 39s
test-go / integration (push) Failing after 4m39s
Surface in-library albums/artists the listener doesn't actively spin but
is predicted to enjoy, derived from the same similarity + like-weighted
candidate engine that powers For-You — rolled up from track scores to
album/artist granularity. Built in the daily 3am BuildSystemPlaylists
pass, atomic-replaced alongside the system playlists, and read back by
/api/home (+ /api/home/index).

Cold-start gate: skips generation entirely below 20 distinct unskipped
tracks AND 5 distinct artists, so a thin profile ships empty rows rather
than near-random tiles.

- migration 0034: you_might_like_albums / you_might_like_artists (id+rank,
  CASCADE, per-user rank index).
- playlists/you_might_like.go: cold-start gate + similarity roll-up
  (sum-of-top-3 aggregation, per-artist album cap, daily-rotating via the
  same userIDHash jitter as For-You) + atomic-replace persist in the tx.
- recommendation/home.go: two new HomePayload sections with read-time
  cross-section dedup vs Most Played / Rediscover / Last Played, trimmed
  to 10 each.
- api: you_might_like_albums / you_might_like_artists on /api/home and
  /api/home/index, reusing albumRefFrom / artistRefFromCovered.
- tests: pure roll-up/aggregation/cap unit tests + DB-backed gate,
  sufficiency, and atomic-replace tests (all green vs real Postgres).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-11 19:33:46 -04:00
bvandeusen 9a57dc4bec Merge pull request 'Offline/playback recording robustness (contract-audit follow-ups)' (#90) from dev into main
test-go / test (push) Successful in 29s
android / Build + lint + test (push) Successful in 4m23s
test-go / integration (push) Successful in 4m25s
release / Build signed APK (tag releases only) (push) Successful in 4m20s
release / Build + push container image (push) Successful in 1m53s
2026-06-11 14:19:27 -04:00
bvandeusen 4ecb1680bf fix(android): drop second loop jump in supersededLikeToggleIds (detekt)
android / Build + lint + test (push) Successful in 3m51s
LoopWithTooManyJumpStatements — replace the two continues with a
filtered sequence + a single null guard.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-11 12:11:54 -04:00
bvandeusen d2a22e49e3 fix: harden offline/playback recording (contract-audit follow-ups)
test-go / test (push) Successful in 42s
android / Build + lint + test (push) Failing after 1m27s
test-go / integration (push) Successful in 4m42s
Audit of every Android↔server connection point (2026-06-11) cleared the
silent-contract class that caused the events `type` bug, but surfaced a
cluster of offline/playback-robustness defects. Fixes:

1. Playback double-count on background. PlayEventsReporter no longer
   enqueues a partial play_offline + leaves the live row open on every
   screen-lock. closeCurrent() now routes by whether the server has an
   open row: close-by-id (durable PLAY_ENDED on failure) when it does,
   offline only when no row exists, and a no-op while a play_started is
   in flight (the server auto-closes that orphan). onStop only durably
   closes a *paused* play — a still-playing one is left to the live path
   under the foreground service. Adds the PLAY_ENDED mutation kind.

2. Replayer poison rows. MutationReplayer now classifies each replay as
   SENT / DROP / RETRY: permanent 4xx (and corrupt payloads) are dropped
   instead of retried forever; 408/429/5xx/transport still retry.

3. Offline-play / close-by-id idempotency (server). RecordOfflinePlay
   dedups on (user, track, started_at); RecordPlayEnded skips a second
   skip_events insert when re-closing an already-ended row. Makes the
   at-least-once replay safe against lost-response duplicates.

4. Like-toggle collapse. Replayer drops like-toggles superseded by a
   later toggle for the same entity, so partial-failure + differential
   retry can't invert the final like state.

5. Connectivity-return trigger. MutationReplayer + SyncController now
   also drain/sync when NetworkStatusController recovers to Healthy, so
   an offline→online transition mid-session doesn't wait for a cold
   start. SyncController.syncSafe gains a single-in-flight mutex.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-11 12:05:35 -04:00
bvandeusen f167ddfbfb Merge pull request 'fix: restore native Android play-event recording (History was empty)' (#89) from dev into main
test-go / test (push) Successful in 38s
test-go / integration (push) Successful in 4m29s
android / Build + lint + test (push) Successful in 4m42s
release / Build signed APK (tag releases only) (push) Successful in 3m59s
release / Build + push container image (push) Successful in 14s
2026-06-11 09:08:20 -04:00
bvandeusen 653ed95f14 fix(android): always serialize event type so plays aren't dropped
android / Build + lint + test (push) Successful in 5m21s
The native client's /api/events requests omitted the `type`
discriminator entirely. The app's Json is configured with
encodeDefaults=false (AppModule), so a `type` left at its data-class
default ("play_started" etc.) is never written to the wire. The server
multiplexes on `type` and returns 400 "unknown event type" for an
empty one, which PlayEventsReporter's catch swallows — and the
play_started path has no offline fallback, so the play is lost with no
trace.

Net effect: EVERY native Android play event (started/ended/skipped/
offline) has 400'd since this code was written. Listening History only
ever populated from the Flutter/web clients; as usage moved to the
native app, History went sparse. Confirmed live in the server access
log: POST /api/events -> 400 on every play, while reads 200.

Force the discriminator onto the wire with
@EncodeDefault(Mode.ALWAYS) on each request type's `type` field.
Surgical (vs flipping encodeDefaults globally), and idiomatic for a
constant-valued discriminator.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-11 00:01:13 -04:00
bvandeusen b2c6f6f0e9 fix(server): apply skip rule on auto-close instead of force-hiding orphans
test-go / test (push) Successful in 35s
test-go / integration (push) Successful in 4m34s
autoClosePriorOpen hardcoded was_skipped=true for every orphaned
play_event (a play_started whose play_ended never arrived, e.g. the
client backgrounded mid-track). That hid fully-listened tracks from
History — a play that sat open past its own length was capped to the
track duration (ratio ~1) yet still flagged skipped. Observed live:
History showed 3 plays for a day of listening because most rows were
auto-closed orphans marked skipped.

Now the auto-close applies the same skip rule as RecordPlayEnded to the
duration-capped elapsed estimate: ratio >= threshold OR elapsed >= the
duration floor -> a real play that lands in History; a genuine
quick-abandon still classifies as a skip. Still writes no skip_events
row, so the ambiguous auto-close never feeds the skip-ratio /
recommendation signal.

This is the server half. The client-side root cause (backgrounded
track transitions never closed, orphaning the rows in the first place)
is tracked separately.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-10 23:15:22 -04:00
bvandeusen 962b4dbc8c Merge pull request 'v2026.06.07 — Sonos: one SOAP failure no longer drops to local' (#88) from dev into main
android / Build + lint + test (push) Successful in 4m37s
release / Build signed APK (tag releases only) (push) Successful in 4m21s
release / Build + push container image (push) Successful in 15s
2026-06-07 19:14:45 -04:00
bvandeusen aa4089118e chore: configure Renovate (tuned)
release / Build signed APK (tag releases only) (push) Has been skipped
release / Build + push container image (push) Successful in 1m34s
Activate Renovate with a tuned config: target dev, ignore retired
flutter_client/**, auto-merge GREEN patch/minor bumps, hold majors behind
dependency-dashboard approval, and group go/CI/docker/gradle/npm updates.
Throttled to a weekend schedule with prHourlyLimit 2.
2026-06-07 12:00:44 -04:00
bvandeusen dabca3ad24 fix(android): stop single SOAP failure from dropping Sonos to local
android / Build + lint + test (push) Successful in 3m44s
When the phone was locked, a transport command (lock-screen/Bluetooth/
watch media button, or queue-resync play) issued during a WiFi power-save
stall would hit the 2s connect timeout and throw. handleSoapFailure then
cancelled the poll loop and fired onDrop on that single failure -- showing
"Disconnected from <Sonos>" and reverting to local playback, even though
the renderer was perfectly reachable. This bypassed the poll loop's
deliberate 30-consecutive-failure tolerance (DROP_THRESHOLD, bumped from 3
precisely for screen-off WiFi sleep / Doze).

Make the 1 Hz poll loop the sole drop arbiter:
- Transport SOAP commands retry transient IO failures (retryTransientIo:
  3 attempts, 400ms backoff) so a brief WiFi stall lands the command once
  WiFi wakes instead of abandoning it. A SoapFaultException (renderer
  answered, rejected the action) is not retried -- the device is alive.
- handleTransportFailure no longer cancels the poll loop or fires onDrop;
  it logs and nudges an immediate poll so the UI reconciles to Sonos's
  actual state. If the renderer is truly gone, the poll loop trips the
  drop on its own via DROP_THRESHOLD.

Extract retryTransientIo as an internal top-level fn + unit test covering
first-success, retry-then-succeed, exhaust-and-rethrow, and no-retry-on-
SoapFault. Refresh now-stale drop-heuristic comments.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-07 10:54:14 -04:00
bvandeusen 7c791dc8e4 v2026.06.06 — UPnP recovery, library watcher, artist discovery, request auto-poll (#87)
test-go / test (push) Successful in 35s
test-web / test (push) Successful in 45s
android / Build + lint + test (push) Successful in 4m7s
test-go / integration (push) Successful in 4m30s
release / Build signed APK (tag releases only) (push) Successful in 3m44s
release / Build + push container image (push) Successful in 14s
2026-06-06 23:31:27 -04:00
bvandeusen 8f29cc7414 feat: poll in-flight requests on native + refresh after submit (#369)
test-web / test (push) Successful in 33s
android / Build + lint + test (push) Successful in 3m37s
Native RequestsViewModel gains poll-while-approved (silent reloads, paused
when nothing in-flight) for parity with the web auto-poll, and the SSE
collector now reloads silently instead of flashing the loading spinner.
Web discover submit invalidates qk.myRequests() so a new request appears
on /requests immediately.
2026-06-06 23:18:44 -04:00
bvandeusen c8b21aa76e fix(android): grid items() resolution on artist detail (drop bad alias)
android / Build + lint + test (push) Successful in 3m35s
2026-06-06 22:56:58 -04:00
bvandeusen 483804fc9e fix(android): extract ArtistSuccessBody to satisfy detekt LongMethod
android / Build + lint + test (push) Failing after 2m56s
2026-06-06 22:48:45 -04:00
bvandeusen 12a8cfccb5 feat(android): similar-artists strip + top-tracks on artist detail
test-go / test (push) Successful in 32s
test-web / test (push) Successful in 40s
android / Build + lint + test (push) Failing after 1m16s
test-go / integration (push) Successful in 4m29s
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-06 22:39:54 -04:00
bvandeusen e358a92cf4 feat(web): similar-artists strip + top-tracks panel on artist detail 2026-06-06 22:34:46 -04:00
bvandeusen 63b25e65ad feat(server): similar-artists + per-user artist top-tracks endpoints
GET /api/artists/{id}/similar — in-library artists ranked by similarity
score (deduped across sources), ArtistRef list with cover + album count.
GET /api/artists/{id}/top-tracks — current user's most-played tracks for
the artist (skips excluded, quarantine filtered).
2026-06-06 22:30:41 -04:00
bvandeusen a766b7193f test(server): drop removed scheduler arg from New() callers
test-go / test (push) Successful in 28s
test-go / integration (push) Successful in 4m27s
2026-06-06 22:00:21 -04:00
bvandeusen e95138d412 style(server): gofmt watcher_test map alignment
test-go / test (push) Failing after 10s
test-web / test (push) Successful in 33s
test-go / integration (push) Failing after 4m27s
2026-06-06 21:54:14 -04:00
bvandeusen 7426f6c718 feat(web): remove scan-schedule admin UI (scheduler retired)
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-06 21:52:20 -04:00
bvandeusen e994aae613 feat(server): retire scan scheduler for watcher + safety-net walk
Wire the fsnotify watcher and a fixed 12h safety-net delta walk in main;
remove the configurable scan scheduler (scheduler.go, scan_schedule table via
migration 0033, GET/PATCH /api/admin/scan/schedule, and the server/api
plumbing). Manual scan + scan status are unchanged.
2026-06-06 21:50:08 -04:00
bvandeusen c39a9ca18f feat(server): targeted ScanFiles + fsnotify library watcher
Add Scanner.ScanFiles (watcher-driven targeted scan returning changed album
IDs) and a recursive fsnotify Watcher that debounces filesystem events and
enriches just the affected albums inline. Pure classifyEvent/drainPending
seams unit-tested; ScanFiles covered in the scanner integration test.
2026-06-06 21:43:27 -04:00
bvandeusen 06e155abb6 fix(android): rename IdleRevertDecision.kt to satisfy detekt MatchingDeclarationName
android / Build + lint + test (push) Successful in 4m11s
2026-06-06 17:36:07 -04:00
bvandeusen 6b11208e5a perf(android): fail-fast connect timeout for UPnP transport SOAP
android / Build + lint + test (push) Failing after 1m8s
2026-06-06 17:30:28 -04:00
bvandeusen 41da012494 fix(android): failed UPnP play reverts to phone and resumes locally 2026-06-06 17:29:58 -04:00
bvandeusen 8a3104443c feat(android): record play-intent on remote play/pause 2026-06-06 17:29:36 -04:00
bvandeusen 5f5ab69da6 feat(android): track UPnP play-intent surviving SOAP errors 2026-06-06 17:29:19 -04:00
bvandeusen 5386ae870f feat(android): revert UPnP output to phone after 5-min idle 2026-06-06 17:28:32 -04:00
bvandeusen d7b011e52f feat(android): pure idle-revert decision for UPnP output 2026-06-06 17:27:52 -04:00
bvandeusen 11466e1525 Merge pull request 'Unify offline detection + offline playlist UX (NetworkStatusController)' (#86) from dev into main
android / Build + lint + test (push) Successful in 3m56s
release / Build signed APK (tag releases only) (push) Successful in 3m50s
release / Build + push container image (push) Successful in 13s
2026-06-05 13:27:36 -04:00
bvandeusen e185b36138 fix(android): arbitrate op-failure probe off the reducer thread
android / Build + lint + test (push) Successful in 3m33s
Awaiting probeOnce() inline in the OpFailure branch blocked the single-consumer
reducer for the /healthz timeout, delaying a concurrent self-proving success
from snapping back to Healthy. Launch the probe instead so recovery stays fast.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-05 12:47:46 -04:00
bvandeusen 31d8c30dfe feat(android): grey offline-unavailable playlists, lead with cached pools
android / Build + lint + test (push) Successful in 3m28s
Home + Playlists list derive offline from NetworkStatusController (Offline or
ServerDown, not the raw device link), fixing the consistency gap. PlaylistRef
gains unavailableOffline (refreshable || !fullyCached); PlaylistCard dims the
whole tile when greyed but stays tappable. buildPlaylistsRow offline: pools
lead, real playlists partitioned available-first, placeholders dropped.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-05 12:40:30 -04:00
bvandeusen 69ccd7b25d feat(android): per-playlist fullyCached via cache-index join
android / Build + lint + test (push) Successful in 3m27s
CachedPlaylistDao.observeCachedCounts LEFT-JOINs cached_playlist_tracks ×
audio_cache_index per playlist; PlaylistsRepository.observeAll combines it in
and stamps PlaylistRef.fullyCached (trackCount>0 && cached>=trackCount). Merge
extracted to a pure mergePlaylistsWithCache for Android-free unit tests.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-05 12:29:59 -04:00
bvandeusen 8b77c6be97 feat(android): 4-state connection banner with unstable + back-online flash
android / Build + lint + test (push) Successful in 3m27s
Banner VM collects NetworkStatusController.state directly (drops the redundant
WhileSubscribed re-wrap). Adds a mild 'Reconnecting…' treatment for the
non-gating Unstable state and a transient 'Back online' confirmation on
down→Healthy recovery (try/finally guards against a stuck flash).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-05 12:23:55 -04:00
bvandeusen c33ef18a50 fix(android): extract OfflineGatedDataSource gate to satisfy detekt ThrowsCount
android / Build + lint + test (push) Successful in 3m24s
The added IOException rethrow pushed open() to 3 throws (max 2). Move the
two gating throws into a private gateOnHealth() helper.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-05 12:19:29 -04:00
bvandeusen 4c9450c117 feat(android): feed API/playback/pull-refresh signals into NetworkStatusController
android / Build + lint + test (push) Failing after 1m10s
OkHttp ReachabilityReportingInterceptor (Lazy to break the Hilt cycle) runs
first in the chain and reports only PLACEHOLDER_HOST (Minstrel-bound) 2xx/IO
outcomes so external artwork fetches don't read as server reachability.
OfflineGatedDataSource reports stream open success/failure; PlaybackErrorReporter
arbitrates on track failures; PullToRefreshScaffold re-probes on every pull.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-05 12:14:50 -04:00
bvandeusen b467cb7532 refactor(android): unify offline detection into NetworkStatusController
android / Build + lint + test (push) Successful in 3m25s
Absorbs VersionCheckController (/healthz poll + version parse) and
ServerHealthController (tri-state derive) into one signal-driven authority.
Adds the non-gating Unstable state across all ServerHealth branch sites
(OfflineGatedDataSource, SearchRepository, TrackRow, banner). Repoints
MinstrelApplication, MainActivity, PlayerFactory, VersionTooOldViewModel.
Drops the now-unused nowMs params the detekt UnusedParameter rule flagged.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-05 12:06:22 -04:00
bvandeusen c78bbb7ba5 feat(android): pure ReachabilityMachine + 4-state ServerHealth enum
android / Build + lint + test (push) Failing after 1m16s
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-05 11:35:22 -04:00
bvandeusen 301c3bfb86 Merge pull request 'fix(android): don't flip offline on WAN-validation flicker — trust /healthz' (#85) from dev into main
android / Build + lint + test (push) Successful in 5m6s
release / Build signed APK (tag releases only) (push) Successful in 4m42s
release / Build + push container image (push) Successful in 14s
2026-06-04 23:02:28 -04:00
bvandeusen 5c0db429b3 fix(android): don't flip offline on WAN-validation flicker — trust /healthz
android / Build + lint + test (push) Successful in 3m38s
A self-hosted Minstrel server is usually on the LAN, but ConnectivityObserver
gated 'online' on NET_CAPABILITY_VALIDATED — which tracks whether Android
reached its own WAN internet-validation probe, not whether Minstrel is
reachable. A transient WAN/DNS blip (or Android's periodic re-validation)
momentarily drops VALIDATED while the LAN server stays reachable. That flipped
ServerHealth -> Offline with NO debounce (only the /healthz path got hysteresis),
and OfflineGatedDataSource fast-failed the in-flight stream read with
OfflineException -> ExoPlayer SOURCE error -> the load_failed 'Source error'
event. On-device: 'app said server offline while it wasn't', one track failed,
then recovered when VALIDATED returned.

- ConnectivityObserver: require INTERNET only, not VALIDATED. The /healthz poll
  (VersionCheckController, with its own failure hysteresis) is the authority on
  whether Minstrel is reachable; the device-link signal only answers 'is there a
  network at all' (airplane mode).
- ServerHealthController: add a WARN-tier transition log. The signal had zero
  instrumentation, which is why this was hard to diagnose from logcat.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-04 22:57:57 -04:00
bvandeusen 4d8c7d6566 Merge pull request 'fix(android): notification art uses album cover, not embedded stream tags' (#84) from dev into main
android / Build + lint + test (push) Successful in 4m1s
release / Build signed APK (tag releases only) (push) Successful in 6m12s
release / Build + push container image (push) Successful in 15s
2026-06-04 22:35:21 -04:00
bvandeusen 58810a860b fix(android): notification art uses album cover, not embedded stream tags
android / Build + lint + test (push) Successful in 3m38s
The MediaController notification / lock-screen background pulled artwork
from the stream's embedded ID3/FLAC tags (artworkData) because the
MediaItem never set artworkUri — a different source than the in-app
album cover (/api/albums/{id}/cover). For tracks whose embedded tag art
differs from the server album cover, the two surfaces disagreed.

- PlayerController.toMediaItem: set artworkUri to TrackRef.coverUrl.
  MediaMetadata.populate() overwrites artworkUri+artworkData as a pair,
  so the MediaItem URI clears the embedded bytes ExoPlayer extracts from
  the stream — the album cover now wins on both surfaces.
- PlayerFactory.buildBitmapLoader: OkHttp-backed CacheBitmapLoader so the
  authed placeholder cover URL resolves (the default DefaultHttpDataSource
  loader can't rewrite placeholder.invalid or attach the auth cookie).
- MinstrelPlayerService: attach it via MediaSession.setBitmapLoader.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-04 22:24:20 -04:00
bvandeusen d6e6caa223 Merge pull request 'Sonos queue resync + cold-start prefetcher gate' (#83) from dev into main
android / Build + lint + test (push) Successful in 4m19s
release / Build signed APK (tag releases only) (push) Successful in 4m1s
release / Build + push container image (push) Successful in 13s
2026-06-04 17:36:31 -04:00
bvandeusen 9a31955fa4 perf(android): gate audio prefetch on isPlaying
android / Build + lint + test (push) Successful in 3m41s
Cold-start playback on a fresh install was taking ~25 s before any
audio played. Logcat showed AudioPrefetcher was kicking off N
concurrent CacheWriter jobs the instant setQueue updated uiState --
each prefetch a full upcoming-track download over the same OkHttp
client as the current-track DataSource. Five-way bandwidth split
plus parallel Coil cover fetches starved the current track until
its full file body had streamed through (~12 MB at ~1 MB/s under
contention).

Now reconcile() observes uiState.isPlaying and starts upcoming-track
prefetches only when the current track is actually playing.
Cancellation of out-of-window jobs always runs so a queue switch or
skip still frees the pipe immediately, even while paused. Cold start
should drop from ~25 s -> 5-7 s on the user's network: just the
single-stream throughput plus the one-time TLS/DNS tax.

Refactored the inline reconcile body into computeTargets /
cancelOutOfWindowLocked / startInWindowLocked helpers to keep
ReturnCount under the detekt cap.
2026-06-04 17:29:19 -04:00
bvandeusen e7d7cb2471 fix(android): incremental Sonos queue sync for playNext + radio-append
android / Build + lint + test (push) Successful in 3m55s
The previous fix re-loaded Sonos's full queue on every uiState.queue
identity change -- correct for playlist-switch (full replacement) but
disruptive for in-queue mutations: playNext and radio-append would
restart the currently-playing track on Sonos because removeAllTracks
+ AddURIToQueue x N + SetAVTransportURI re-anchors the transport.

Now the resync runs a longest-common-prefix / common-suffix diff first.
When the current Sonos track lies in the preserved prefix, applies the
minimum-incremental SOAP operations -- RemoveTrackRangeFromQueue on the
removed middle, AddURIToQueue at the same insertion point -- so Sonos
keeps playing the current track and the new entries land in place
without interrupting playback. Falls back to the full removeAllTracks
reload when the current track is in the removed slice (playlist
switch).

Adds AVTransportClient.removeTrackRangeFromQueue (Sonos-specific,
UpdateID=0 skips the queue-version check).

Cases now covered:
  - Playlist switch -> full reload (current track replaced, prefix=0)
  - playNext insert  -> 1 AddURIToQueue at the right slot
  - Radio-append     -> RemoveTrackRangeFromQueue for old tail + N
                        AddURIToQueue for new tracks at the end
2026-06-04 17:07:26 -04:00
bvandeusen 8017934334 fix(android): re-prime Sonos queue when user plays a different playlist
android / Build + lint + test (push) Successful in 3m56s
Before: tapping a different playlist while Sonos was the active route
updated the player view but Sonos kept the old queue and played those
tracks (or whatever was last there). PlayerController.setQueue replaced
the local ExoPlayer queue and called play(), which forwarded SOAP Play
to Sonos -- but Sonos's native queue (loaded once at route selection
via removeAllTracks + AddURIToQueue + SetAVTransportURI) was never
touched on subsequent setQueue calls.

Now: MinstrelForwardingPlayer.setMediaItems (all 3 overloads) clears
holder.active + sets target synchronously so the immediately-following
play() drops via isLoadingUpnp(). OutputPickerController observes
uiState.queue identity changes; when target or active is non-null and
the queue key shifted, it re-runs loadQueueOnSonos under the existing
selectUpnpMutex and restores active when done. Sonos resync failures
drop cleanly to local (selectedUpnpRouteIdInternal nulled).

Doesn't touch addMediaItem / radio-append paths -- those leave Sonos's
queue stale and need a separate AddURIToQueue extension hook; out of
scope for this fix.
2026-06-04 17:00:24 -04:00
bvandeusen 8b08482d13 Merge pull request 'fix(android): hysteresis on /healthz reachable signal' (#82) from dev into main
android / Build + lint + test (push) Successful in 4m46s
release / Build signed APK (tag releases only) (push) Successful in 4m26s
release / Build + push container image (push) Successful in 13s
2026-06-04 14:23:49 -04:00
bvandeusen faa0c7024b fix(android): hysteresis on /healthz reachable signal — 3 consecutive failures
android / Build + lint + test (push) Successful in 3m55s
The single-failure flip-to-false produced two false-positive permanent
banner cases:

1. Startup race — AuthStore.baseUrl loads from Room asynchronously, so the
   first runOnce() can fire against AuthStore.DEFAULT_BASE_URL
   ("http://localhost:8080") before the real server URL has hydrated. One
   failure was enough to lock the banner on for the next 5 minutes.

2. Deployments whose reverse proxy routes only /api/* to the Go server —
   /healthz never reaches the handler, so /healthz polls fail forever even
   though every real /api/* call succeeds. User sees "Server unreachable"
   permanently and OfflineGatedDataSource starts throwing OfflineException
   on every audio cache miss, silently breaking playback of uncached
   tracks.

Now we require 3 consecutive failures (~15 min at the 5-min poll cadence)
before flipping reachable=false, and any single success resets the
counter. Adds Timber.w/i at the flip transitions so operator logcat can
diagnose genuine outages.
2026-06-04 13:40:49 -04:00
bvandeusen 7cf04fe24b Merge pull request 'Android #618 offline-mode UX + Sonos polish + server DRY' (#81) from dev into main
android / Build + lint + test (push) Successful in 4m35s
release / Build signed APK (tag releases only) (push) Successful in 4m23s
release / Build + push container image (push) Successful in 13s
2026-06-04 12:53:59 -04:00
bvandeusen 1e17eeda72 feat(android): #618 Phase 5 — confirm offline writes via shell snackbar
android / Build + lint + test (push) Successful in 3m58s
MutationQueue now emits "Saved — will sync when online" on a SharedFlow
whenever a user-driven enqueue lands (like toggle, playlist append,
request create/cancel, quarantine flag/unflag). Background enqueues
(play-offline events, playback-error reports) do not emit — those fire
from non-foreground paths where a snackbar would be either dropped
(no shell mounted) or jarring (lock-screen toggle).

ShellScaffold subscribes via OfflineWriteHintViewModel and routes the
hint through its existing snackbar host. Replaces the prior silent-
queue UX where a tap on a like / playlist add looked successful but
the user couldn't tell whether the server had been hit or the call
was deferred for replay.
2026-06-04 12:40:59 -04:00
bvandeusen 1daea79f64 feat(android): #618 Phase 4 — row-level offline-unavailable affordance
android / Build + lint + test (push) Has been cancelled
TrackRow now consumes the LocalServerHealth CompositionLocal (provided
once at MainActivity from ServerHealthController.state). When the server
is Offline or ServerDown and the track id isn't in LocalCachedTrackIds,
the row dims to 0.4 alpha and a tap fires a Toast instead of attempting
playback. Replaces the silent "tap-and-fail-to-OfflineException" UX with
explicit at-a-glance signaling of which rows in a long list will work.

Trailing slot (kebab / like / playlist-add) stays interactive so write
affordances can route through MutationQueue — Phase 5 gates those at
the action level.
2026-06-04 12:37:42 -04:00
bvandeusen 48de720514 feat(android): #618 Phase 2 — search falls back to cached entities when offline
android / Build + lint + test (push) Has been cancelled
When ServerHealthController reports Offline or ServerDown, SearchRepository
runs Room LIKE queries against cached_artists / cached_albums / cached_tracks
instead of hitting /api/search. The screen draws a one-line hint above the
results so the user can tell server matches from on-device-only matches.

Adds searchByName / searchByTitle DAO methods; LOCAL_SEARCH_LIMIT=20 matches
the server's default page size.
2026-06-04 12:35:01 -04:00
bvandeusen fced6b681e feat(android): cache-only audio path when ServerHealth != Healthy
android / Build + lint + test (push) Successful in 4m58s
Phase 3 of #618. Wraps the OkHttpDataSource upstream of CacheDataSource with OfflineGatedDataSource. CacheDataSource only consults the upstream factory on a cache miss, so playback of cached audio is unaffected. Offline tap on a non-cached track now throws OfflineException immediately (subclass of IOException for ExoPlayer's PlaybackException to wrap) instead of waiting on a multi-second OkHttp timeout. AudioPrefetcher keeps its own ungated upstream -- writes fail silently when offline, no user-visible impact.
2026-06-04 11:27:11 -04:00
bvandeusen 80a6be25aa feat(android): ServerHealth tri-state composite + banner distinguishes offline vs server-down
android / Build + lint + test (push) Has been cancelled
Phase 1 of #618. VersionCheckController gains reachable: StateFlow<Boolean> from the same /healthz poll (no double polling). ServerHealthController combines connectivity.online + versionCheck.reachable into ServerHealth { Healthy, Offline, ServerDown }. ConnectionErrorBanner now branches on the tri-state, distinguishing 'no Wi-Fi' from 'server unreachable.' Phases 2-5 (local search, cache-only audio source, row-level not-cached affordance, write-affordance gray-out) ship separately as independent slices.
2026-06-04 11:25:22 -04:00
bvandeusen 4d0a0b8e09 fix(android): throttle UPnP extend with 50ms delay per successful AddURIToQueue
android / Build + lint + test (push) Successful in 4m10s
Closes Scribe #611. The 2026-06-04 logcat showed 33 consecutive AddURIToQueue failures clustered at ~10ms intervals once the burst hit offset 39 -- characteristic of Sonos's burst-add rate-limit. 50ms between successful adds adds ~5s to the 100-track background extension but eliminates the burst rejection. Next reproduction with the SOAP fault detail logging (audit commit c5b326c6) will confirm the fault code if any tracks still fail.
2026-06-04 11:03:55 -04:00
bvandeusen 6184c62721 feat(android): MediaSession picks up UPnP state via direct listener invocation
android / Build + lint + test (push) Has been cancelled
Closes Scribe #606. Two pieces: MediaMetadata gets durationMs (lock-screen scrubber gets a known total even when wrapped ExoPlayer is paused under UPnP); MinstrelForwardingPlayer keeps an externalListeners registry that mirrors super.addListener so we can directly invoke onIsPlayingChanged / onPlaybackStateChanged / onMediaItemTransition when remoteState mutates. Fires from pollOnce + play()/pause() onSuccess. dedup via lastNotified guards so we don't spam events at 1Hz when nothing changed.
2026-06-04 11:02:51 -04:00
bvandeusen 222a0ff636 Merge pull request 'dev → main: collage center-crop, server DRY, CI durability-off' (#80) from dev into main
test-go / test (push) Successful in 29s
test-go / integration (push) Successful in 4m27s
release / Build signed APK (tag releases only) (push) Successful in 4m1s
release / Build + push container image (push) Successful in 12s
2026-06-04 08:42:36 -04:00
bvandeusen 28300e19fd perf(ci): turn off Postgres durability in integration tests
test-go / test (push) Successful in 28s
test-go / integration (push) Successful in 4m27s
TRUNCATE-everything ResetDB before every test forces a commit fsync; the CI DB is rebuilt each run so durability buys nothing. ALTER SYSTEM via docker exec (the services: block can't override the postgres command line). Non-fatal so a perms surprise degrades to slow, never red.

Per the playbook the operator shared from another project (~17x speedup observed there). Measure before/after in the next two CI runs.
2026-06-04 08:31:06 -04:00
bvandeusen 024493f2a7 refactor(server): unify stream URL builders + MIME tables + cover-path helper
test-go / test (push) Successful in 28s
test-go / integration (push) Has been cancelled
Closes Scribe #614, #615, server half of #616 surfaced by the 2026-06-04 divergent-provider audit.

- streamURL helper now used everywhere /api/tracks/{id}/stream is built (was inline concat in playlists.go and cast_token.go); add streamURLWithExt for the .ext cast variant.

- audioContentType in media.go is the canonical file_format -> MIME lookup; mimeForFormat in cast_token.go is now a thin wrapper that overrides the unknown-format fallback to audio/mpeg (Sonos rejects octet-stream). Adds mpeg/vorbis/wave aliases. Subsonic's contentTypeForFormat stays frozen per docs.

- coverart.ResolveAlbumPath extracted; api and subsonic both delegate to it.
2026-06-04 08:29:51 -04:00
bvandeusen edd198cdf5 fix(server): collage drawScaled uses center-crop instead of stretch
test-go / test (push) Successful in 27s
test-go / integration (push) Has been cancelled
2026-06-04 08:22:59 -04:00
bvandeusen d75c1ae37f Merge pull request 'dev → main: Android UPnP/Sonos transport parity + server stream URL extension' (#79) from dev into main
release / Build signed APK (tag releases only) (push) Has been skipped
test-go / test (push) Successful in 30s
release / Build + push container image (push) Successful in 1m24s
android / Build + lint + test (push) Successful in 4m12s
test-go / integration (push) Successful in 9m16s
2026-06-04 08:15:15 -04:00
bvandeusen 8cd2383a42 test(server): seed track for cast-token tests + assert file-ext in URL
test-go / test (push) Successful in 29s
test-go / integration (push) Successful in 9m34s
2026-06-04 07:44:33 -04:00
bvandeusen 27bd38e005 feat(server): stream URL gets file extension so Sonos can probe duration
test-go / test (push) Successful in 37s
test-go / integration (push) Failing after 10m34s
2026-06-04 07:29:28 -04:00
bvandeusen aa23a72693 fix(android): effectiveDuration uses desiredIdx so duration tracks the displayed title
android / Build + lint + test (push) Successful in 4m2s
2026-06-04 07:17:17 -04:00
bvandeusen 4021938046 fix(android): polling tick no longer force-syncs controller -- avoids Sonos seek-to-0
android / Build + lint + test (push) Successful in 3m26s
2026-06-04 07:06:07 -04:00
bvandeusen 7486bc2444 fix(android): split tickPositionPoll into helpers for detekt complexity
android / Build + lint + test (push) Has been cancelled
2026-06-04 07:04:01 -04:00
bvandeusen ee8a1fdc93 fix(android): event-driven pending-transport clear (Sonos ack OR 5s safety)
android / Build + lint + test (push) Failing after 1m12s
2026-06-04 07:00:22 -04:00
bvandeusen 8e578d2068 fix(android): 2s transport-ack lockout so Prev/Next isn't undone by stale poll
android / Build + lint + test (push) Successful in 3m25s
2026-06-04 06:55:39 -04:00
bvandeusen cacb280832 fix(android): polling tick track update is forward-only so user Next isn't undone
android / Build + lint + test (push) Successful in 3m27s
2026-06-04 06:50:35 -04:00
bvandeusen 36054506c2 fix(android): polling tick owns track-change updates when delegate.seekTo silent
android / Build + lint + test (push) Successful in 3m26s
2026-06-04 06:46:38 -04:00
bvandeusen 5db90844cb fix(android): DIDL-Lite includes Rincon namespace + cdudn desc for Sonos
android / Build + lint + test (push) Successful in 3m39s
2026-06-04 06:34:00 -04:00
bvandeusen d5437d517e fix(android): single break in extend loop for detekt LoopWithTooManyJumpStatements
android / Build + lint + test (push) Successful in 3m52s
2026-06-04 01:07:57 -04:00
bvandeusen 3085d6f409 fix(android): DIDL-Lite restricted=true / id=-1 so Sonos accepts metadata
android / Build + lint + test (push) Failing after 1m33s
2026-06-04 00:59:31 -04:00
bvandeusen c5b326c620 fix(android): UPnP extend captures SOAP fault, aborts after 3 failures
android / Build + lint + test (push) Has been cancelled
2026-06-04 00:56:48 -04:00
bvandeusen 389c896d65 fix(android): force poll on ON_RESUME + cap interpolation drift at 5s
android / Build + lint + test (push) Has been cancelled
2026-06-04 00:55:10 -04:00
bvandeusen 41230b5afb fix(android): single jump per polling loop for detekt LoopWithTooManyJumpStatements
android / Build + lint + test (push) Successful in 4m0s
2026-06-04 00:42:35 -04:00
bvandeusen c245b1ef0b fix(android): hold UI patch until cursor sync lands; reanchor on track flip
android / Build + lint + test (push) Failing after 1m21s
2026-06-04 00:37:34 -04:00
bvandeusen 2425a305eb fix(android): interpolate UPnP position between polls + suppress post-seek race
android / Build + lint + test (push) Successful in 4m5s
2026-06-04 00:16:57 -04:00
bvandeusen 88b161193d fix(android): TrackRef.durationSec is final fallback so duration is never 0
android / Build + lint + test (push) Successful in 5m54s
2026-06-04 00:03:48 -04:00
bvandeusen 9628ed1749 fix(android): duration falls back to local while Sonos hasn't reported one
android / Build + lint + test (push) Has been cancelled
2026-06-04 00:00:45 -04:00
bvandeusen 85926f4ec0 fix(android): keep service alive when UPnP is playing -- override playWhenReady
android / Build + lint + test (push) Successful in 3m53s
2026-06-03 23:51:54 -04:00
bvandeusen 47b0894ad6 test(android): update RemotePlayerState threshold expectation to 30
android / Build + lint + test (push) Has been cancelled
2026-06-03 23:50:37 -04:00
bvandeusen e62fac3a0e fix(android): onEvents reads UPnP state so resume keeps duration
android / Build + lint + test (push) Has been cancelled
2026-06-03 23:48:52 -04:00
bvandeusen eae5dcad23 fix(android): correct ErrorCopy import path in PlaylistPlayback
android / Build + lint + test (push) Failing after 6m2s
2026-06-03 23:42:18 -04:00
bvandeusen 3576e241c0 fix(android): split playPlaylistShuffled to satisfy detekt ReturnCount
android / Build + lint + test (push) Failing after 3m42s
2026-06-03 23:37:30 -04:00
bvandeusen 8f89279fa4 fix(android): UPnP survives screen-off + cursor catches up to Sonos
android / Build + lint + test (push) Has been cancelled
2026-06-03 23:36:43 -04:00
bvandeusen b1a66f18bd feat(android): shuffle on system playlist tile play -- Home + Playlists list
android / Build + lint + test (push) Failing after 1m30s
2026-06-03 23:29:42 -04:00
bvandeusen 6a7958c921 fix(android): debounce non-PLAYING poll to suppress UPnP track-change flicker
android / Build + lint + test (push) Successful in 4m13s
2026-06-03 23:19:29 -04:00
bvandeusen 33285b53c6 fix(android): refresh uiState.isPlaying from remoteState during UPnP
android / Build + lint + test (push) Successful in 4m1s
2026-06-03 23:08:32 -04:00
bvandeusen 87ad7f4dc2 feat(android): lazy queue activation + loading spinner during UPnP load
android / Build + lint + test (push) Successful in 3m55s
Part A: split loadQueueOnSonos into an initial phase (tracks[0..currentIndex]
only, then SetAV+Seek+Play) plus a background extendQueueOnSonos coroutine
that appends the remainder after activation. Reduces the UPnP activation
block from ~17s (100 tracks serial) to ~200ms (1 track at currentIndex=0).
Background extension cancels cleanly when activeUpnpHolder.active changes.

Part B: add PlayerUiState.isUpnpLoading (target set, active null). Projected
inline in onEvents so it stays consistent with the rest of the snapshot, plus
a separate combine(target, active) collector that updates uiState between
player-event fires. NowPlayingScreen.TransportRow and MiniPlayer.MiniRow
replace the play/pause icon with a CircularProgressIndicator while loading
and disable the button tap to prevent premature commands to the Sonos queue.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 22:56:06 -04:00
bvandeusen 9c0013f4b6 fix(android): defer holder.active until SOAP wired -- drop transport during load
android / Build + lint + test (push) Successful in 4m34s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 22:46:57 -04:00
bvandeusen 6129536153 diag(android): log ForwardingPlayer transport overrides on entry
android / Build + lint + test (push) Successful in 3m50s
2026-06-03 22:32:33 -04:00
bvandeusen e6c3c959fa fix(android): handleZeroDurationIfNeeded ReturnCount within cap
android / Build + lint + test (push) Successful in 3m45s
2026-06-03 22:25:53 -04:00
bvandeusen e011b04e04 fix(android): remove pollLoop cursor sync -- races with queue load and SOAP
android / Build + lint + test (push) Has been cancelled
2026-06-03 22:24:34 -04:00
bvandeusen e2866795ef fix(android): skip zero-duration auto-advance during UPnP playback
android / Build + lint + test (push) Failing after 1m26s
2026-06-03 22:18:11 -04:00
bvandeusen e20d7b1438 fix(android): correct SOAPACTION assertions in next/previous tests
android / Build + lint + test (push) Successful in 3m58s
2026-06-03 22:06:54 -04:00
bvandeusen 2a098a78fe feat(android): Sonos queue mode -- ClearQueue + AddURIToQueue + x-rincon-queue
android / Build + lint + test (push) Failing after 9m9s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 21:55:03 -04:00
bvandeusen ece37e9a92 fix(android): remove pollLoop natural-advance -- races with selectUpnp and skip
android / Build + lint + test (push) Successful in 4m1s
Polling alone cannot distinguish Sonos auto-advancing via SetNextAVTransportURI
from URI changes we made ourselves via syncCurrentItemToRemote.  This produced
two races: (1) activation race -- first poll returns stale URI from prior
session, second returns new URI, false-positive fires and double-advances the
cursor; (2) user-skip race -- skip's syncCurrentItemToRemote changes the URI,
next poll sees the change and fires again.  Remove the detection block and
previousTrackUri capture from pollOnce entirely.  pollLoop is now a pure
state-tracker (position + transport state) plus the one-shot initial pre-queue
gate.  GENA event subscriptions to AVTransport LastChange are the correct fix;
deferred to its own slice (see parity-map).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 21:42:01 -04:00
bvandeusen 8a1203c4a1 fix(android): revert x-rincon-mp3radio Sonos URI -- plain https for music tracks
android / Build + lint + test (push) Successful in 4m13s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 21:31:31 -04:00
bvandeusen 487d1bd430 fix(android): SOAP/UPnP parsers enable processNamespaces -- correct name extraction
android / Build + lint + test (push) Successful in 4m4s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 20:58:59 -04:00
bvandeusen 5c99341b34 fix(android): test assertions use JUnit Jupiter for lazy-message Supplier<String>
android / Build + lint + test (push) Failing after 3m31s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 20:51:28 -04:00
bvandeusen 8fe3308afd fix(android): detekt -- parseHhMmSs ReturnCount/MagicNumber + readUntilEndTag jumps
android / Build + lint + test (push) Failing after 3m5s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 20:45:00 -04:00
bvandeusen 96594ba52b fix(android): Sonos ZGT robust extraction -- escaped + nested element fallback + diag
android / Build + lint + test (push) Failing after 1m29s
2026-06-03 20:40:52 -04:00
bvandeusen 2c61d7a333 fix(android): Sonos UDN comparison strips _MR/_MS suffix -- coordinator routing
android / Build + lint + test (push) Failing after 2m41s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 20:29:08 -04:00
bvandeusen 8652b86f40 fix(android): Sonos x-rincon-mp3radio URI transform for SetAVTransportURI
android / Build + lint + test (push) Failing after 1m20s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 20:20:25 -04:00
bvandeusen 75132a2afe diag(android): log raw SOAP response body for first 3 UPnP polls
android / Build + lint + test (push) Failing after 1m22s
Add optional onRawResponse callback to SoapClient; loggingSoapClient
factory emits the first 6 GetPositionInfo/GetTransportInfo bodies
(3 poll cycles) at WARN so release logs capture them. Wire into
transportFor so every AVTransportClient for a new UPnP session logs.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 20:12:15 -04:00
bvandeusen 673f98487f fix(android): UPnP local pause via delegate -- no SOAP race
android / Build + lint + test (push) Failing after 1m30s
Move local ExoPlayer pause from OutputPickerController.selectUpnp
into MinstrelForwardingPlayer.onActiveChanged (handler.post { delegate.pause() }).
This guarantees the pause hits ExoPlayer before the holder is live, eliminating
the async race that caused SOAP fault 701 on Sonos when pause() was dispatched
via playerController after holder.active was already set.
Also adds per-poll Timber.w before initialPreQueueDone for diagnostics.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 19:53:41 -04:00
bvandeusen c556388a6b fix(android): UPnP activation order -- pause local before holder.set; sync+preQueue sequential; diagnostic Timber.w
android / Build + lint + test (push) Failing after 1m28s
- OutputPickerController.selectUpnp: pause ExoPlayer BEFORE setting
  activeUpnpHolder so ForwardingPlayer.pause() routes to ExoPlayer,
  not SOAP; remove now-redundant playerController.pause() from inside
  runCatching; bump activation Timber.i -> Timber.w for release logcat
- MinstrelForwardingPlayer: remove Player.Listener onMediaItemTransition
  that raced with seekToNext/Prev override's syncCurrentItemToRemote;
  seekToNext/Prev now launch sync -> preQueueNext sequentially in one
  coroutine; remove early preQueueNext from onActiveChanged (raced with
  selectUpnp SOAP); move initial pre-queue to pollLoop, fires once
  trackUri lands confirming Sonos accepted SetAV+Play
- Extract pollOnce from pollLoop to stay within detekt LongMethod=60;
  natural-advance branch now calls preQueueNext explicitly (no listener)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 19:47:19 -04:00
bvandeusen edffdec2b2 fix(android): UPnP drop falls back to local; pollLoop detects natural advance
android / Build + lint + test (push) Failing after 1m10s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 18:07:20 -04:00
bvandeusen 1ab21d81ca feat(android): SetNextAVTransportURI pre-queue for gap-free UPnP advance
android / Build + lint + test (push) Failing after 1m35s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 17:53:10 -04:00
bvandeusen 81794e2475 fix(android): UPnP volume cache keyed to active route id
android / Build + lint + test (push) Has been cancelled
2026-06-03 17:52:00 -04:00
bvandeusen 29309d9bfb ui(android): drop snackbar + hardware volume keys when UPnP route active
android / Build + lint + test (push) Failing after 1m24s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 17:43:55 -04:00
bvandeusen 70b29567fb refactor(android): service holds Player not ExoPlayer for ForwardingPlayer wrap
android / Build + lint + test (push) Failing after 1m25s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 17:36:35 -04:00
bvandeusen 2f4d67d3c8 feat(android): PlayerFactory wraps ExoPlayer in ForwardingPlayer + drop events
android / Build + lint + test (push) Failing after 1m35s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 17:34:35 -04:00
bvandeusen b2bfe96559 ui(android): LazyColumn stable keys for in-place output picker row continuity
android / Build + lint + test (push) Failing after 1m21s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 17:33:01 -04:00
bvandeusen e9dd3e4d2a fix(android): OutputPicker -- local capture, selectUpnp mutex, shared dedup helper, suppress
android / Build + lint + test (push) Has been cancelled
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 17:31:40 -04:00
bvandeusen b29875fd30 feat(android): UPnP selection state, disconnect, BuiltIn-pinned alphabetical sort
android / Build + lint + test (push) Failing after 1m21s
OutputPickerController now owns selection state for the UPnP leg and
runs the disconnect flow when the user picks a system route while a
renderer is active.

- Inject OkHttpClient + RemotePlayerState so we can build a
  RenderingControlClient at selection time and capture the last-known
  remote position on disconnect.
- selectUpnp publishes ActiveUpnp(routeId, routeName, avTransport,
  rendering) to ActiveUpnpHolder, marks the route id in
  selectedUpnpRouteIdInternal, and honors Sonos topology by routing
  through coordinatorRouteFor before SOAP.
- selectSystem now does the disconnect: AVTransport.Stop -> clear the
  holder -> seek local ExoPlayer to the remembered position -> resume
  if the remote was playing.
- routesState combines 4 sources (system, UPnP, Sonos topology,
  upnp-selected id). Non-coordinator Sonos members are filtered out
  of the visible list. current resolves from the merged list when a
  UPnP route is selected; otherwise from the system snapshot.
- sortRoutes drops the current-first rule -- BuiltIn "Phone speaker"
  pins to the top, everything else lowercase-alphabetical. Selection
  state moves to the radio-button indicator in the picker row.
- RemotePlayerState gets @Singleton + @Inject constructor() so Hilt
  can provide the shared instance to both the picker and the
  forthcoming MinstrelForwardingPlayer.
2026-06-03 17:24:08 -04:00
bvandeusen 85cea8d559 fix(android): UPnP forwarding -- route mint failures through drop, no double drop
android / Build + lint + test (push) Failing after 1m25s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 17:18:53 -04:00
bvandeusen ab6c3a1354 feat(android): MinstrelForwardingPlayer wraps ExoPlayer with UPnP branching
android / Build + lint + test (push) Failing after 1m25s
Task 7 of UPnP transport-parity slice. Introduces the central
ForwardingPlayer that branches between local ExoPlayer and the active
UPnP renderer:

- MinstrelForwardingPlayer wraps the delegate Player; play/pause/seek
  and the next/previous transport calls translate to AVTransport SOAP
  when an ActiveUpnp is set, otherwise forward to super. Position +
  isPlaying + duration + playbackState reads pull from
  RemotePlayerState while remote.
- 1Hz poll loop drives GetPositionInfo + GetTransportInfo, feeding
  RemotePlayerState; the rolling-3 failure heuristic fires onDrop on
  the looper for the factory to surface as a snackbar.
- StreamTokenProvider extracts the CastApi.create() Retrofit wiring
  into a Hilt singleton so the service-side player and the
  controller-side picker share one CastApi instance.
- OutputPickerController constructor swaps Retrofit for
  StreamTokenProvider + ActiveUpnpHolder (the holder is wired now for
  Task 8). selectUpnp now mints via streamTokens.mint(trackId).

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-06-03 17:12:11 -04:00
bvandeusen 3aee2276bc feat(android): RemotePlayerState container for UPnP-synthesized player state
android / Build + lint + test (push) Failing after 1m28s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 17:08:04 -04:00
bvandeusen 9a7d3b2d30 feat(android): ActiveUpnpHolder singleton for picker -> player handoff
android / Build + lint + test (push) Failing after 1m24s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 17:06:33 -04:00
bvandeusen 9002cf5559 refactor(android): UPnP discovery cleanup -- bareUdn helper, Timber, suppress, kdoc
android / Build + lint + test (push) Has been cancelled
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 17:05:31 -04:00
bvandeusen a5e4570f01 feat(android): UPnP in-place updates + Sonos topology aggregation
android / Build + lint + test (push) Failing after 1m14s
2026-06-03 16:59:31 -04:00
bvandeusen bfcb9c42a0 feat(android): Sonos ZoneGroupTopology client + parser
android / Build + lint + test (push) Failing after 1m27s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 16:53:29 -04:00
bvandeusen 799d50024c test(android): RenderingControl lower clamp + GetVolume request body checks
android / Build + lint + test (push) Failing after 1m17s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 16:51:07 -04:00
bvandeusen 8c0c4c8600 feat(android): RenderingControl get/set volume
android / Build + lint + test (push) Failing after 1m33s
2026-06-03 16:48:40 -04:00
bvandeusen 3cdb416f94 fix(android): AVTransport test assertions escape DIDL; consolidate xmlEscape
android / Build + lint + test (push) Failing after 1m18s
DIDL assertion now checks for XML-escaped form (&lt;dc:title&gt;) since
SoapClient.buildEnvelope escapes all arg values. Lifts xmlEscape to a
top-level internal fun in SoapClient.kt, removing the duplicate private
copy from AVTransportClient. Fixes @Suppress rationale (not Compose).
Renames seek test to reflect colon-separated format; adds unknown-state
getTransportInfo test.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 16:47:18 -04:00
bvandeusen d62a3b8134 feat(android): AVTransport pause/seek/nextURI/positionInfo/transportInfo
android / Build + lint + test (push) Failing after 1m35s
- Add pause(), seek(positionMs), setNextAVTransportURI(uri, mime, title)
- Add getPositionInfo() -> PositionInfo, getTransportInfo() -> TransportInfo
- Extract buildDidlLite() helper; add formatHhMmSs / parseHhMmSs helpers
- Add PositionInfo, TransportState, TransportInfo top-level types
- Add AVTransportClientTest covering all four new call shapes

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-03 16:42:04 -04:00
bvandeusen 574bf29a7e fix(android): PlayerController transport methods dispatch to controller thread
android / Build + lint + test (push) Successful in 4m17s
After DIDL fix, Sonos accepted SetAVTransportURI + Play, but
playerController.pause() threw IllegalStateException 'method is
called from a wrong thread' because selectUpnp runs the whole
UPnP-selection flow on Dispatchers.Default. UI tap handlers were
fine - they're already on Main - but the cross-thread background
call from OutputPickerController.selectUpnp hit the MediaController's
application-thread guard.

Same fix as the cold-boot resume one earlier today (commit e69a5204
wrapped setQueue): pause / play / seekTo / skipToNext / skipToPrevious
now route through runOnControllerThread, which is a no-op when
already on the application looper and Handler.post otherwise.

Logcat from on-device confirmed Sonos plays after this fix lands -
SetAVTransportURI -> 200, Play -> 200, then the IllegalStateException
was the last failure path.
2026-06-03 16:00:53 -04:00
bvandeusen 24b7c92abd fix(server+android): DIDL-Lite metadata for Sonos UPnP (error 1023)
test-go / test (push) Successful in 29s
android / Build + lint + test (push) Successful in 4m14s
test-go / integration (push) Failing after 9m12s
After the X-Forwarded-Proto fix Sonos now gets a clean https:// URL
but returns vendor error 1023 - empty CurrentURIMetaData. Sonos
requires DIDL-Lite metadata with at minimum <res protocolInfo>
carrying the audio MIME type so it can validate the source before
playback. The original spec said 'Sonos accepts empty DIDL; recoverable
if a device rejects' - that was wrong for Sonos.

Server (cast_token.go):
- Look up the track and return mime (from tracks.file_format) +
  title in the cast-token response. mimeForFormat covers the common
  formats - mp3, flac, m4a/aac, ogg, opus, wav - falling through to
  audio/mpeg for unknowns.
- Missing track returns 404 (apierror.NotFound) instead of letting the
  caller mint a token for nothing.

Client (CastApi.kt, AVTransportClient.kt, OutputPickerController.kt):
- StreamTokenResponse gains mime + title (defaulted so old contracts
  stay parseable).
- AVTransportClient.setAVTransportURIWithMetadata builds minimal Sonos-
  acceptable DIDL-Lite around the URL + MIME + title. xml-escaped.
- selectUpnp calls the new overload; Timber.i now logs the MIME so the
  next on-device test shows it.

Generic UPnP renderers tolerate the DIDL shape too - no downside to
sending it everywhere.
2026-06-03 15:54:56 -04:00
300 changed files with 19445 additions and 2828 deletions
+54 -5
View File
@@ -2,8 +2,8 @@ name: release
# Builds and pushes the minstrel container image to the Gitea registry.
#
# push to main → :main and :latest (no APK bundled)
# push tag vYYYY.MM.DD → :vYYYY.MM.DD and :latest (APK bundled)
# push to main → :main and :latest (latest-release APK bundled)
# push tag vYYYY.MM.DD → :vYYYY.MM.DD and :latest (freshly-built APK bundled)
# workflow_dispatch → manual trigger (same rules based on the ref)
#
# Release model: per-day CalVer tags (no trailing patch digit). The day's
@@ -20,6 +20,14 @@ name: release
# happens in the same android-release job, so the Release-page download
# link and the in-image bundled APK are both populated atomically.
#
# :latest always carries an APK. Because every main push also moves
# :latest (not just tags), a main build with no APK would silently strip
# the in-app update channel off :latest until the next release. So on
# non-tag builds image-release pulls the MOST RECENT release's signed APK
# and reconstructs its exact versionName (tag + commit-count, the same
# formula android-release bakes in) for the version sidecar — no rebuild,
# just rebundle. Tag builds keep bundling their own freshly-built APK.
#
# Android testing (lint + detekt + unit tests, debug APK upload on main)
# lives in android.yml and runs independently on every push.
@@ -181,6 +189,12 @@ jobs:
steps:
- name: Checkout
uses: actions/checkout@v4
with:
# Full history + tags so non-tag :latest builds can resolve the
# latest release tag's commit count and reconstruct the bundled
# APK's exact versionName (see "Bundle latest release APK" below).
fetch-depth: 0
fetch-tags: true
- name: Detect buildable project
id: guard
@@ -221,9 +235,8 @@ jobs:
| docker login git.fabledsword.com -u "${{ github.actor }}" --password-stdin
- name: Download signed APK artifact
# Tag pushes only — android-release just produced this. Main
# pushes skip and the image ships with empty client/ (the
# /api/client/version endpoint then returns 404 by design).
# Tag pushes only — android-release just produced this. Non-tag
# builds take the "Bundle latest release APK" path below instead.
if: steps.guard.outputs.ready == 'true' && startsWith(github.ref, 'refs/tags/v')
uses: actions/download-artifact@v3
with:
@@ -247,6 +260,42 @@ jobs:
echo "${APK_VERSION_NAME}" > client/minstrel.apk.version
ls -lh client/
- name: Bundle latest release APK (non-tag :latest builds)
# Main pushes don't build an APK, but they DO move :latest — so
# without this the in-app update channel would vanish from :latest
# until the next tag. Pull the most-recent release's signed APK and
# reconstruct its exact versionName (${TAG#v}.$(git rev-list --count
# TAG) — identical to android-release's formula) so the version
# sidecar the server hands clients matches the installed build.
# Degrades to an empty client/ (404 update channel) — never a wrong
# version — if no release / APK asset / tag-count can be resolved.
if: steps.guard.outputs.ready == 'true' && !startsWith(github.ref, 'refs/tags/v')
shell: bash
env:
CI_TOKEN: ${{ secrets.CI_TOKEN }}
run: |
set -eu
REPO="${GITHUB_REPOSITORY}"
REL_JSON="$(curl -fsSL -H "Authorization: token ${CI_TOKEN}" \
"https://git.fabledsword.com/api/v1/repos/${REPO}/releases/latest" || true)"
if [ -z "${REL_JSON}" ]; then
echo "::notice::no published release — image ships without bundled APK"; exit 0
fi
TAG="$(printf '%s' "${REL_JSON}" | grep -oP '"tag_name":\s*"\K[^"]+' | head -1)"
APK_URL="$(printf '%s' "${REL_JSON}" | grep -oP '"browser_download_url":\s*"\K[^"]+' | grep -E '\.apk$' | head -1)"
if [ -z "${TAG}" ] || [ -z "${APK_URL}" ]; then
echo "::notice::latest release '${TAG:-?}' has no APK asset — image ships without bundled APK"; exit 0
fi
COUNT="$(git rev-list --count "${TAG}" 2>/dev/null || true)"
if [ -z "${COUNT}" ]; then
echo "::notice::could not resolve commit count for ${TAG} (tag not fetched?) — skipping APK bundle"; exit 0
fi
VERSION_NAME="${TAG#v}.${COUNT}"
curl -fsSL -H "Authorization: token ${CI_TOKEN}" -o client/minstrel.apk "${APK_URL}"
echo "${VERSION_NAME}" > client/minstrel.apk.version
echo "::notice::bundled release APK ${TAG} as version ${VERSION_NAME}"
ls -lh client/
- name: Build and push
if: steps.guard.outputs.ready == 'true'
run: |
+21
View File
@@ -116,6 +116,27 @@ jobs:
# Wait for Postgres to accept TCP (no health-check dependency).
for i in $(seq 1 60); do (echo > "/dev/tcp/${PG_IP}/5432") 2>/dev/null && break; sleep 2; done
# Relax durability on the throwaway CI Postgres. Our test pattern
# is dbtest.ResetDB → TRUNCATE … RESTART IDENTITY CASCADE before
# every test, and the per-TRUNCATE commit fsync is the dominant
# cost of the integration suite. The CI DB is rebuilt every run so
# fsync / full_page_writes / synchronous_commit buy nothing. Apply
# via docker exec because:
# - The act_runner `services:` block can't override the container
# command, so `postgres -c fsync=off` at boot isn't an option.
# - ALTER SYSTEM cannot run inside a transaction; psql -c
# auto-commits each statement, which is what we need.
# - fsync / full_page_writes are sighup GUCs and
# synchronous_commit is user-context, so pg_reload_conf() picks
# all three up with no restart.
# Non-fatal: a perms surprise degrades to "slower", never red CI.
docker exec "$PG_ID" psql -U minstrel -d minstrel_test \
-c "ALTER SYSTEM SET fsync = off" \
-c "ALTER SYSTEM SET synchronous_commit = off" \
-c "ALTER SYSTEM SET full_page_writes = off" \
-c "SELECT pg_reload_conf()" \
|| echo "WARN: durability relax failed; continuing"
# Apply embedded migrations to the fresh test DB, then run the
# full suite (no -short → integration tests execute). -p 1:
# every integration package TRUNCATEs the one shared test DB;
+44 -7
View File
@@ -4,7 +4,7 @@ A self-hosted music server that thinks for you. Smart shuffle, contextual likes,
> State and intelligence belong on the server, not the client.
<!-- TODO: screenshot of the home page -->
<a href="docs/screenshots/home.png"><img src="docs/screenshots/home.png" width="820" alt="Minstrel home — your library at a glance"></a>
## Highlights
@@ -13,7 +13,7 @@ A self-hosted music server that thinks for you. Smart shuffle, contextual likes,
- **ListenBrainz radio.** Session-aware "more like this" pulls from ListenBrainz similarity data, not a static genre tag.
- **Lidarr integration.** Triggered scans, request-driven album imports, and a quarantine flow when something doesn't fit.
- **Built-in web SPA.** Full-feature library, search, queue, playlists, and admin — no separate frontend container to deploy.
- **Flutter mobile client in flight.** Tracking issue [#356](https://git.fabledsword.com/bvandeusen/minstrel/issues/356).
- **Native Android client, shipped with the server.** The signed APK is bundled into every image and attached to each [release](https://git.fabledsword.com/bvandeusen/minstrel/releases) — sideload it once, then the app self-updates straight from your own server (no app store, no separate download to track).
## Quickstart
@@ -24,10 +24,18 @@ services:
image: git.fabledsword.com/bvandeusen/minstrel:latest
ports: ['4533:4533']
volumes:
# Your music library. Point ./music at wherever your audio files
# live. Mounted read-only — Minstrel never writes to your library.
- ./music:/music:ro
- minstrel-data:/data
# Generated data: playlist cover collages, artist art, caches.
# The path must match MINSTREL_STORAGE_DATA_DIR, which the image
# sets to /app/data — keep this mount on /app/data or your cache
# won't survive a container recreate.
- minstrel-data:/app/data
environment:
MINSTREL_DATABASE_URL: postgres://minstrel:minstrel@db:5432/minstrel?sslmode=disable
# Colon-separated library roots to scan; must match the container
# path of the read-only music mount above (/music here).
MINSTREL_LIBRARY_SCAN_PATHS: /music
depends_on: [db]
@@ -37,6 +45,8 @@ services:
POSTGRES_USER: minstrel
POSTGRES_PASSWORD: minstrel
POSTGRES_DB: minstrel
# Postgres data dir — users, likes, play history, sessions, settings.
# The one volume you must never lose; back it up with pg_dump.
volumes: [pgdata:/var/lib/postgresql/data]
volumes:
@@ -48,7 +58,29 @@ volumes:
docker compose up -d
```
After the stack is up, visit `http://localhost:4533/register` and create your admin account. The first user to register on a fresh instance is automatically marked as the administrator; subsequent users can register through the same form (or via invite tokens generated from the admin Users panel, depending on how you configure registration).
## First run
With the stack up, a handful of in-app steps get you to a working library. Use your own host in place of `localhost` if you're reaching the server over a LAN/VPN address (plain `http://` is fine — no TLS required).
**1. Create your admin account.** Visit `http://localhost:4533/register`. The first account on a fresh instance is automatically the administrator; later users join through the same form or an invite token (step 5).
<a href="docs/screenshots/register.png"><img src="docs/screenshots/register.png" width="320" alt="Creating the first (admin) account on a fresh instance"></a>
**2. Let the first library scan finish.** `scan_on_startup` is on by default, so Minstrel walks your mounted library on boot and imports artists, albums, and tracks — no button to press. Watch progress (and re-scan any time) on the **Admin** page (`/admin`); the scan runs in stages and is incremental, so later restarts only pick up what changed.
<a href="docs/screenshots/library-scan.png"><img src="docs/screenshots/library-scan.png" width="820" alt="The Admin page, where the library scan runs and reports progress"></a>
**3. (Optional) Name the instance and wire up integrations.** In admin **Settings → Integrations** (`/admin/integrations`), add a ListenBrainz token (scrobbling + similarity radio) and/or a Lidarr URL + API key (the request flow). These live in the UI and apply without a restart; the display name can also be set via `MINSTREL_BRANDING_APP_NAME`.
<a href="docs/screenshots/integrations.png"><img src="docs/screenshots/integrations.png" width="820" alt="ListenBrainz and Lidarr integration cards in admin Settings"></a>
**4. Install the Android app.** Open **Settings** (`/settings`) and use the *Install the Android app* card to download the APK that ships inside this server image, then sign in with the same account. From then on the app self-updates straight from your server.
<a href="docs/screenshots/android-download.png"><img src="docs/screenshots/android-download.png" width="820" alt="The &quot;Install the Android app&quot; download card in Settings"></a>
**5. Invite the rest of the household.** From admin **Users** (`/admin/users`), generate an invite token (or enable open registration). Each person gets their own account, so likes, play history, and recommendations stay per-user.
<a href="docs/screenshots/invite-users.png"><img src="docs/screenshots/invite-users.png" width="820" alt="Generating an invite token in admin Users"></a>
For the full configuration surface, see [`config.example.yaml`](./config.example.yaml).
@@ -57,7 +89,7 @@ For the full configuration surface, see [`config.example.yaml`](./config.example
Most operators only need the env vars in the quickstart above. A few extras worth knowing:
- `MINSTREL_BRANDING_APP_NAME` — rename the instance ("Family Jukebox", "Office Music"). Surfaces in the header, browser tab, and OG share previews.
- `MINSTREL_STORAGE_DATA_DIR`defaults to `./data`. Holds playlist cover collages and other generated artefacts.
- `MINSTREL_STORAGE_DATA_DIR`where generated artefacts (playlist cover collages, artist art, caches) are written. The container image sets this to `/app/data`, which is why the quickstart mounts the `minstrel-data` volume there.
- `MINSTREL_LIBRARY_SCAN_PATHS` — colon-separated list of music library roots to scan. Supports multiple roots (`/music:/podcasts`).
ListenBrainz integration (per-user scrobble + similarity tokens) and Lidarr integration (URL + API key) are configured through the admin Settings UI rather than env vars or yaml — per Minstrel's "config in UI" rule, integration settings live where operators can edit them without restarting.
@@ -66,8 +98,13 @@ Most operational keys have a `MINSTREL_<SECTION>_<FIELD>` env override. Recommen
## Updating
- `:main` — rolling, follows the dev branch's tested tip. Recommended only for the operator who's running an upstream-watching deployment.
- `:v1.0.x` — pinned releases. Recommended default. Database migrations run automatically at startup; rollbacks require restoring a Postgres dump.
Image tags (`git.fabledsword.com/bvandeusen/minstrel:<tag>`):
- `:latest` — the newest blessed image. Moves on every `main` push **and** every release. Recommended for most operators.
- `:vYYYY.MM.DD` — immutable per-day release tags. Pin one of these for a deployment you don't want moving under you. (Per-day CalVer — no trailing patch digit; a same-day re-cut moves the tag forward.)
- `:main` — the rolling post-merge tip. Same image as `:latest` at push time; choose it if you want to track `main` explicitly rather than the release line.
Every `:latest` and every `:vYYYY.MM.DD` bundles the current signed Android APK, so the in-app update channel is always live. Database migrations run automatically at startup; rollbacks require restoring a Postgres dump.
## Specs
+13
View File
@@ -1754,6 +1754,19 @@
<option name="screenX" value="1600" />
<option name="screenY" value="2560" />
</PersistentDeviceSelectionData>
<PersistentDeviceSelectionData>
<option name="api" value="36" />
<option name="brand" value="google" />
<option name="codename" value="tangorpro" />
<option name="formFactor" value="Tablet" />
<option name="id" value="tangorpro" />
<option name="labId" value="google" />
<option name="manufacturer" value="Google" />
<option name="name" value="Pixel Tablet" />
<option name="screenDensity" value="320" />
<option name="screenX" value="1600" />
<option name="screenY" value="2560" />
</PersistentDeviceSelectionData>
<PersistentDeviceSelectionData>
<option name="api" value="35" />
<option name="brand" value="google" />
-1
View File
@@ -1,4 +1,3 @@
<?xml version="1.0" encoding="UTF-8"?>
<project version="4">
<component name="ExternalStorageConfigurationManager" enabled="true" />
<component name="ProjectRootManager" version="2" languageLevel="JDK_21" default="true" project-jdk-name="jbr-21" project-jdk-type="JavaSDK">
@@ -0,0 +1,690 @@
{
"formatVersion": 1,
"database": {
"version": 6,
"identityHash": "fb73ed8674efb1d82a586551baba5ef0",
"entities": [
{
"tableName": "sync_metadata",
"createSql": "CREATE TABLE IF NOT EXISTS `${TABLE_NAME}` (`id` INTEGER NOT NULL, `cursor` INTEGER NOT NULL, `lastSyncAt` INTEGER, PRIMARY KEY(`id`))",
"fields": [
{
"fieldPath": "id",
"columnName": "id",
"affinity": "INTEGER",
"notNull": true
},
{
"fieldPath": "cursor",
"columnName": "cursor",
"affinity": "INTEGER",
"notNull": true
},
{
"fieldPath": "lastSyncAt",
"columnName": "lastSyncAt",
"affinity": "INTEGER"
}
],
"primaryKey": {
"autoGenerate": false,
"columnNames": [
"id"
]
}
},
{
"tableName": "cached_artists",
"createSql": "CREATE TABLE IF NOT EXISTS `${TABLE_NAME}` (`id` TEXT NOT NULL, `name` TEXT NOT NULL, `sortName` TEXT NOT NULL, `mbid` TEXT, `artistThumbPath` TEXT, `artistFanartPath` TEXT, `fetchedAt` INTEGER NOT NULL, PRIMARY KEY(`id`))",
"fields": [
{
"fieldPath": "id",
"columnName": "id",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "name",
"columnName": "name",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "sortName",
"columnName": "sortName",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "mbid",
"columnName": "mbid",
"affinity": "TEXT"
},
{
"fieldPath": "artistThumbPath",
"columnName": "artistThumbPath",
"affinity": "TEXT"
},
{
"fieldPath": "artistFanartPath",
"columnName": "artistFanartPath",
"affinity": "TEXT"
},
{
"fieldPath": "fetchedAt",
"columnName": "fetchedAt",
"affinity": "INTEGER",
"notNull": true
}
],
"primaryKey": {
"autoGenerate": false,
"columnNames": [
"id"
]
}
},
{
"tableName": "cached_albums",
"createSql": "CREATE TABLE IF NOT EXISTS `${TABLE_NAME}` (`id` TEXT NOT NULL, `artistId` TEXT NOT NULL, `title` TEXT NOT NULL, `sortTitle` TEXT NOT NULL, `releaseDate` TEXT, `coverPath` TEXT, `mbid` TEXT, `fetchedAt` INTEGER NOT NULL, PRIMARY KEY(`id`))",
"fields": [
{
"fieldPath": "id",
"columnName": "id",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "artistId",
"columnName": "artistId",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "title",
"columnName": "title",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "sortTitle",
"columnName": "sortTitle",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "releaseDate",
"columnName": "releaseDate",
"affinity": "TEXT"
},
{
"fieldPath": "coverPath",
"columnName": "coverPath",
"affinity": "TEXT"
},
{
"fieldPath": "mbid",
"columnName": "mbid",
"affinity": "TEXT"
},
{
"fieldPath": "fetchedAt",
"columnName": "fetchedAt",
"affinity": "INTEGER",
"notNull": true
}
],
"primaryKey": {
"autoGenerate": false,
"columnNames": [
"id"
]
}
},
{
"tableName": "cached_tracks",
"createSql": "CREATE TABLE IF NOT EXISTS `${TABLE_NAME}` (`id` TEXT NOT NULL, `albumId` TEXT NOT NULL, `artistId` TEXT NOT NULL, `title` TEXT NOT NULL, `durationMs` INTEGER NOT NULL, `trackNumber` INTEGER, `discNumber` INTEGER, `filePath` TEXT, `fileFormat` TEXT, `genre` TEXT, `fetchedAt` INTEGER NOT NULL, PRIMARY KEY(`id`))",
"fields": [
{
"fieldPath": "id",
"columnName": "id",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "albumId",
"columnName": "albumId",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "artistId",
"columnName": "artistId",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "title",
"columnName": "title",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "durationMs",
"columnName": "durationMs",
"affinity": "INTEGER",
"notNull": true
},
{
"fieldPath": "trackNumber",
"columnName": "trackNumber",
"affinity": "INTEGER"
},
{
"fieldPath": "discNumber",
"columnName": "discNumber",
"affinity": "INTEGER"
},
{
"fieldPath": "filePath",
"columnName": "filePath",
"affinity": "TEXT"
},
{
"fieldPath": "fileFormat",
"columnName": "fileFormat",
"affinity": "TEXT"
},
{
"fieldPath": "genre",
"columnName": "genre",
"affinity": "TEXT"
},
{
"fieldPath": "fetchedAt",
"columnName": "fetchedAt",
"affinity": "INTEGER",
"notNull": true
}
],
"primaryKey": {
"autoGenerate": false,
"columnNames": [
"id"
]
}
},
{
"tableName": "cached_likes",
"createSql": "CREATE TABLE IF NOT EXISTS `${TABLE_NAME}` (`userId` TEXT NOT NULL, `entityType` TEXT NOT NULL, `entityId` TEXT NOT NULL, `likedAt` INTEGER NOT NULL, PRIMARY KEY(`userId`, `entityType`, `entityId`))",
"fields": [
{
"fieldPath": "userId",
"columnName": "userId",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "entityType",
"columnName": "entityType",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "entityId",
"columnName": "entityId",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "likedAt",
"columnName": "likedAt",
"affinity": "INTEGER",
"notNull": true
}
],
"primaryKey": {
"autoGenerate": false,
"columnNames": [
"userId",
"entityType",
"entityId"
]
}
},
{
"tableName": "cached_playlists",
"createSql": "CREATE TABLE IF NOT EXISTS `${TABLE_NAME}` (`id` TEXT NOT NULL, `userId` TEXT NOT NULL, `name` TEXT NOT NULL, `description` TEXT NOT NULL, `isPublic` INTEGER NOT NULL, `coverPath` TEXT, `trackCount` INTEGER NOT NULL, `durationSec` INTEGER NOT NULL, `systemVariant` TEXT, `fetchedAt` INTEGER NOT NULL, PRIMARY KEY(`id`))",
"fields": [
{
"fieldPath": "id",
"columnName": "id",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "userId",
"columnName": "userId",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "name",
"columnName": "name",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "description",
"columnName": "description",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "isPublic",
"columnName": "isPublic",
"affinity": "INTEGER",
"notNull": true
},
{
"fieldPath": "coverPath",
"columnName": "coverPath",
"affinity": "TEXT"
},
{
"fieldPath": "trackCount",
"columnName": "trackCount",
"affinity": "INTEGER",
"notNull": true
},
{
"fieldPath": "durationSec",
"columnName": "durationSec",
"affinity": "INTEGER",
"notNull": true
},
{
"fieldPath": "systemVariant",
"columnName": "systemVariant",
"affinity": "TEXT"
},
{
"fieldPath": "fetchedAt",
"columnName": "fetchedAt",
"affinity": "INTEGER",
"notNull": true
}
],
"primaryKey": {
"autoGenerate": false,
"columnNames": [
"id"
]
}
},
{
"tableName": "cached_playlist_tracks",
"createSql": "CREATE TABLE IF NOT EXISTS `${TABLE_NAME}` (`playlistId` TEXT NOT NULL, `trackId` TEXT NOT NULL, `position` INTEGER NOT NULL, PRIMARY KEY(`playlistId`, `trackId`))",
"fields": [
{
"fieldPath": "playlistId",
"columnName": "playlistId",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "trackId",
"columnName": "trackId",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "position",
"columnName": "position",
"affinity": "INTEGER",
"notNull": true
}
],
"primaryKey": {
"autoGenerate": false,
"columnNames": [
"playlistId",
"trackId"
]
}
},
{
"tableName": "cached_quarantine_mine",
"createSql": "CREATE TABLE IF NOT EXISTS `${TABLE_NAME}` (`trackId` TEXT NOT NULL, `reason` TEXT NOT NULL, `notes` TEXT, `createdAt` TEXT NOT NULL, `trackTitle` TEXT NOT NULL, `trackDurationMs` INTEGER NOT NULL, `albumId` TEXT NOT NULL, `albumTitle` TEXT NOT NULL, `albumCoverArtPath` TEXT, `artistId` TEXT NOT NULL, `artistName` TEXT NOT NULL, `fetchedAt` INTEGER NOT NULL, PRIMARY KEY(`trackId`))",
"fields": [
{
"fieldPath": "trackId",
"columnName": "trackId",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "reason",
"columnName": "reason",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "notes",
"columnName": "notes",
"affinity": "TEXT"
},
{
"fieldPath": "createdAt",
"columnName": "createdAt",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "trackTitle",
"columnName": "trackTitle",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "trackDurationMs",
"columnName": "trackDurationMs",
"affinity": "INTEGER",
"notNull": true
},
{
"fieldPath": "albumId",
"columnName": "albumId",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "albumTitle",
"columnName": "albumTitle",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "albumCoverArtPath",
"columnName": "albumCoverArtPath",
"affinity": "TEXT"
},
{
"fieldPath": "artistId",
"columnName": "artistId",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "artistName",
"columnName": "artistName",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "fetchedAt",
"columnName": "fetchedAt",
"affinity": "INTEGER",
"notNull": true
}
],
"primaryKey": {
"autoGenerate": false,
"columnNames": [
"trackId"
]
}
},
{
"tableName": "audio_cache_index",
"createSql": "CREATE TABLE IF NOT EXISTS `${TABLE_NAME}` (`trackId` TEXT NOT NULL, `path` TEXT NOT NULL, `sizeBytes` INTEGER NOT NULL, `cachedAt` INTEGER NOT NULL, `lastPlayedAt` INTEGER, `source` TEXT NOT NULL, PRIMARY KEY(`trackId`))",
"fields": [
{
"fieldPath": "trackId",
"columnName": "trackId",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "path",
"columnName": "path",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "sizeBytes",
"columnName": "sizeBytes",
"affinity": "INTEGER",
"notNull": true
},
{
"fieldPath": "cachedAt",
"columnName": "cachedAt",
"affinity": "INTEGER",
"notNull": true
},
{
"fieldPath": "lastPlayedAt",
"columnName": "lastPlayedAt",
"affinity": "INTEGER"
},
{
"fieldPath": "source",
"columnName": "source",
"affinity": "TEXT",
"notNull": true
}
],
"primaryKey": {
"autoGenerate": false,
"columnNames": [
"trackId"
]
}
},
{
"tableName": "cached_mutations",
"createSql": "CREATE TABLE IF NOT EXISTS `${TABLE_NAME}` (`id` INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL, `kind` TEXT NOT NULL, `payload` TEXT NOT NULL, `createdAt` INTEGER NOT NULL, `lastAttemptAt` INTEGER, `attempts` INTEGER NOT NULL)",
"fields": [
{
"fieldPath": "id",
"columnName": "id",
"affinity": "INTEGER",
"notNull": true
},
{
"fieldPath": "kind",
"columnName": "kind",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "payload",
"columnName": "payload",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "createdAt",
"columnName": "createdAt",
"affinity": "INTEGER",
"notNull": true
},
{
"fieldPath": "lastAttemptAt",
"columnName": "lastAttemptAt",
"affinity": "INTEGER"
},
{
"fieldPath": "attempts",
"columnName": "attempts",
"affinity": "INTEGER",
"notNull": true
}
],
"primaryKey": {
"autoGenerate": true,
"columnNames": [
"id"
]
}
},
{
"tableName": "cached_resume_state",
"createSql": "CREATE TABLE IF NOT EXISTS `${TABLE_NAME}` (`id` INTEGER NOT NULL, `json` TEXT NOT NULL, `updatedAt` INTEGER NOT NULL, PRIMARY KEY(`id`))",
"fields": [
{
"fieldPath": "id",
"columnName": "id",
"affinity": "INTEGER",
"notNull": true
},
{
"fieldPath": "json",
"columnName": "json",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "updatedAt",
"columnName": "updatedAt",
"affinity": "INTEGER",
"notNull": true
}
],
"primaryKey": {
"autoGenerate": false,
"columnNames": [
"id"
]
}
},
{
"tableName": "cached_home_index",
"createSql": "CREATE TABLE IF NOT EXISTS `${TABLE_NAME}` (`section` TEXT NOT NULL, `position` INTEGER NOT NULL, `entityType` TEXT NOT NULL, `entityId` TEXT NOT NULL, `fetchedAt` INTEGER NOT NULL, PRIMARY KEY(`section`, `position`))",
"fields": [
{
"fieldPath": "section",
"columnName": "section",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "position",
"columnName": "position",
"affinity": "INTEGER",
"notNull": true
},
{
"fieldPath": "entityType",
"columnName": "entityType",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "entityId",
"columnName": "entityId",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "fetchedAt",
"columnName": "fetchedAt",
"affinity": "INTEGER",
"notNull": true
}
],
"primaryKey": {
"autoGenerate": false,
"columnNames": [
"section",
"position"
]
}
},
{
"tableName": "cached_history_snapshot",
"createSql": "CREATE TABLE IF NOT EXISTS `${TABLE_NAME}` (`id` INTEGER NOT NULL, `json` TEXT NOT NULL, `updatedAt` INTEGER NOT NULL, PRIMARY KEY(`id`))",
"fields": [
{
"fieldPath": "id",
"columnName": "id",
"affinity": "INTEGER",
"notNull": true
},
{
"fieldPath": "json",
"columnName": "json",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "updatedAt",
"columnName": "updatedAt",
"affinity": "INTEGER",
"notNull": true
}
],
"primaryKey": {
"autoGenerate": false,
"columnNames": [
"id"
]
}
},
{
"tableName": "auth_session",
"createSql": "CREATE TABLE IF NOT EXISTS `${TABLE_NAME}` (`id` INTEGER NOT NULL, `sessionCookie` TEXT, `baseUrl` TEXT NOT NULL, `userJson` TEXT, `themeMode` TEXT, `clientId` TEXT, `cacheSettingsJson` TEXT, PRIMARY KEY(`id`))",
"fields": [
{
"fieldPath": "id",
"columnName": "id",
"affinity": "INTEGER",
"notNull": true
},
{
"fieldPath": "sessionCookie",
"columnName": "sessionCookie",
"affinity": "TEXT"
},
{
"fieldPath": "baseUrl",
"columnName": "baseUrl",
"affinity": "TEXT",
"notNull": true
},
{
"fieldPath": "userJson",
"columnName": "userJson",
"affinity": "TEXT"
},
{
"fieldPath": "themeMode",
"columnName": "themeMode",
"affinity": "TEXT"
},
{
"fieldPath": "clientId",
"columnName": "clientId",
"affinity": "TEXT"
},
{
"fieldPath": "cacheSettingsJson",
"columnName": "cacheSettingsJson",
"affinity": "TEXT"
}
],
"primaryKey": {
"autoGenerate": false,
"columnNames": [
"id"
]
}
}
],
"setupQueries": [
"CREATE TABLE IF NOT EXISTS room_master_table (id INTEGER PRIMARY KEY,identity_hash TEXT)",
"INSERT OR REPLACE INTO room_master_table (id,identity_hash) VALUES(42, 'fb73ed8674efb1d82a586551baba5ef0')"
]
}
}
@@ -21,6 +21,9 @@ import androidx.lifecycle.compose.collectAsStateWithLifecycle
import androidx.navigation.compose.rememberNavController
import com.fabledsword.minstrel.auth.ui.AuthGateViewModel
import com.fabledsword.minstrel.cache.CachedTrackIds
import com.fabledsword.minstrel.connectivity.LocalServerHealth
import com.fabledsword.minstrel.connectivity.ServerHealth
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.nav.DetailSeedCache
import com.fabledsword.minstrel.nav.LocalDetailSeedCache
import com.fabledsword.minstrel.nav.MinstrelNavGraph
@@ -38,6 +41,7 @@ import javax.inject.Inject
class MainActivity : ComponentActivity() {
@Inject lateinit var seedCache: DetailSeedCache
@Inject lateinit var cachedTrackIds: CachedTrackIds
@Inject lateinit var serverHealth: NetworkStatusController
// Flipped to true when the user taps the media notification (or
// any other entry point that asks for the full player). The App
@@ -54,6 +58,7 @@ class MainActivity : ComponentActivity() {
App(
seedCache = seedCache,
cachedTrackIds = cachedTrackIds,
serverHealth = serverHealth,
pendingOpenNowPlaying = pendingOpenNowPlaying.asStateFlow(),
onOpenedNowPlaying = { pendingOpenNowPlaying.value = false },
)
@@ -86,6 +91,7 @@ class MainActivity : ComponentActivity() {
private fun App(
seedCache: DetailSeedCache,
cachedTrackIds: CachedTrackIds,
serverHealth: NetworkStatusController,
pendingOpenNowPlaying: StateFlow<Boolean>,
onOpenedNowPlaying: () -> Unit,
themeVm: ThemePreferenceViewModel = hiltViewModel(),
@@ -93,11 +99,13 @@ private fun App(
) {
val theme by themeVm.themeMode.collectAsStateWithLifecycle()
val cached by cachedTrackIds.ids.collectAsStateWithLifecycle()
val health: ServerHealth by serverHealth.state.collectAsStateWithLifecycle()
val pending by pendingOpenNowPlaying.collectAsStateWithLifecycle()
MinstrelTheme(darkOverride = theme.toDarkOverride()) {
CompositionLocalProvider(
LocalDetailSeedCache provides seedCache,
LocalCachedTrackIds provides cached,
LocalServerHealth provides health,
) {
val startDestination by gate.startDestination.collectAsStateWithLifecycle()
val resolved = startDestination
@@ -10,6 +10,8 @@ import com.fabledsword.minstrel.cache.CacheIndexer
import com.fabledsword.minstrel.cache.mutations.MutationReplayer
import com.fabledsword.minstrel.cache.sync.SyncController
import com.fabledsword.minstrel.di.ApplicationScope
import com.fabledsword.minstrel.diagnostics.DiagnosticsReporter
import com.fabledsword.minstrel.diagnostics.DiagnosticsUploader
import com.fabledsword.minstrel.events.EventsStream
import com.fabledsword.minstrel.events.LiveEventsDispatcher
import com.fabledsword.minstrel.metadata.FreshnessSweeper
@@ -19,7 +21,7 @@ import com.fabledsword.minstrel.player.PlayEventsReporter
import com.fabledsword.minstrel.player.PlaybackErrorReporter
import com.fabledsword.minstrel.player.ResumeController
import com.fabledsword.minstrel.update.data.UpdateBannerController
import com.fabledsword.minstrel.update.data.VersionCheckController
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import dagger.hilt.android.HiltAndroidApp
import kotlinx.coroutines.CoroutineScope
import kotlinx.coroutines.launch
@@ -114,12 +116,13 @@ class MinstrelApplication :
@Suppress("unused") @Inject lateinit var audioPrefetcher: AudioPrefetcher
/**
* Same construct-the-singleton trick — VersionCheckController's
* init block starts a 5-min poll loop against /healthz so the
* shell-level VersionTooOldBanner can surface min_client_version
* mismatches without waiting for the next user-driven request.
* Same construct-the-singleton trick — NetworkStatusController owns the
* /healthz poll loop + the device-link collector + the reachability state
* machine, and is the single authority on the tri-state ServerHealth
* signal (plus the VersionTooOld byproduct). It must exist from launch so
* the poll loop runs and the StateFlow stays warm for every consumer.
*/
@Suppress("unused") @Inject lateinit var versionCheckController: VersionCheckController
@Suppress("unused") @Inject lateinit var networkStatusController: NetworkStatusController
/**
* Same construct-the-singleton trick — UpdateBannerController polls
@@ -154,6 +157,11 @@ class MinstrelApplication :
*/
@Suppress("unused") @Inject lateinit var liveEventsDispatcher: LiveEventsDispatcher
// Device diagnostics (M9). The reporter gates itself on the account's
// debug flag; the uploader drains its buffer on a tick / recovery.
@Suppress("unused") @Inject lateinit var diagnosticsReporter: DiagnosticsReporter
@Suppress("unused") @Inject lateinit var diagnosticsUploader: DiagnosticsUploader
@Inject @ApplicationScope lateinit var appScope: CoroutineScope
override fun onCreate() {
@@ -4,6 +4,8 @@ import androidx.lifecycle.ViewModel
import androidx.lifecycle.viewModelScope
import com.fabledsword.minstrel.admin.data.AdminInvitesRepository
import com.fabledsword.minstrel.api.ErrorCopy
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.recoveries
import com.fabledsword.minstrel.models.Invite
import dagger.hilt.android.lifecycle.HiltViewModel
import kotlinx.coroutines.channels.Channel
@@ -25,6 +27,7 @@ data class AdminInvitesUiState(
@HiltViewModel
class AdminInvitesViewModel @Inject constructor(
private val repository: AdminInvitesRepository,
networkStatus: NetworkStatusController,
) : ViewModel() {
private val internal = MutableStateFlow(AdminInvitesUiState())
@@ -36,6 +39,13 @@ class AdminInvitesViewModel @Inject constructor(
init {
refresh()
// Screen-level auto-recovery (issue #1245): reload a failed list
// when server health returns instead of waiting for a manual pull.
viewModelScope.launch {
networkStatus.recoveries().collect {
if (internal.value.message != null) refresh()
}
}
}
fun refresh() {
@@ -29,6 +29,7 @@ import androidx.navigation.NavHostController
import com.fabledsword.minstrel.models.AdminQuarantineItemRef
import com.fabledsword.minstrel.nav.AdminQuarantine
import com.fabledsword.minstrel.shared.widgets.EmptyState
import com.fabledsword.minstrel.shared.widgets.ErrorRetry
import com.fabledsword.minstrel.shared.widgets.LoadingCentered
import com.fabledsword.minstrel.shared.widgets.MinstrelTopAppBar
import com.fabledsword.minstrel.shared.widgets.PullToRefreshScaffold
@@ -62,9 +63,10 @@ fun AdminQuarantineScreen(
body = "When users flag tracks as bad rips, wrong tags, or " +
"duplicates, their reports get aggregated and surfaced here.",
)
is AdminQuarantineUiState.Error -> EmptyState(
is AdminQuarantineUiState.Error -> ErrorRetry(
title = "Couldn't load queue",
body = s.message,
message = s.message,
onRetry = { viewModel.refresh() },
)
is AdminQuarantineUiState.Success -> QueueList(
rows = s.rows,
@@ -4,6 +4,8 @@ import androidx.lifecycle.ViewModel
import androidx.lifecycle.viewModelScope
import com.fabledsword.minstrel.admin.data.AdminQuarantineRepository
import com.fabledsword.minstrel.api.ErrorCopy
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.recoveries
import com.fabledsword.minstrel.events.EventsStream
import com.fabledsword.minstrel.models.AdminQuarantineItemRef
import dagger.hilt.android.lifecycle.HiltViewModel
@@ -26,6 +28,7 @@ sealed interface AdminQuarantineUiState {
class AdminQuarantineViewModel @Inject constructor(
private val repository: AdminQuarantineRepository,
private val eventsStream: EventsStream,
networkStatus: NetworkStatusController,
) : ViewModel() {
private val internal = MutableStateFlow<AdminQuarantineUiState>(AdminQuarantineUiState.Loading)
@@ -38,6 +41,13 @@ class AdminQuarantineViewModel @Inject constructor(
.filter { it.kind.startsWith("quarantine.") }
.collect { refresh() }
}
// Screen-level auto-recovery (issue #1245): reload a failed list
// when server health returns instead of waiting for a manual pull.
viewModelScope.launch {
networkStatus.recoveries().collect {
if (internal.value is AdminQuarantineUiState.Error) refresh()
}
}
}
fun refresh(): Job = viewModelScope.launch {
@@ -28,6 +28,7 @@ import androidx.navigation.NavHostController
import com.fabledsword.minstrel.models.RequestRef
import com.fabledsword.minstrel.nav.AdminRequests
import com.fabledsword.minstrel.shared.widgets.EmptyState
import com.fabledsword.minstrel.shared.widgets.ErrorRetry
import com.fabledsword.minstrel.shared.widgets.LoadingCentered
import com.fabledsword.minstrel.shared.widgets.MinstrelTopAppBar
import com.fabledsword.minstrel.shared.widgets.PullToRefreshScaffold
@@ -61,9 +62,10 @@ fun AdminRequestsScreen(
body = "When users ask Lidarr for new music, their pending " +
"requests show up here for approval.",
)
is AdminRequestsUiState.Error -> EmptyState(
is AdminRequestsUiState.Error -> ErrorRetry(
title = "Couldn't load requests",
body = s.message,
message = s.message,
onRetry = { viewModel.refresh() },
)
is AdminRequestsUiState.Success -> RequestList(
rows = s.rows,
@@ -5,6 +5,8 @@ import androidx.lifecycle.viewModelScope
import com.fabledsword.minstrel.admin.data.AdminRequestsRepository
import com.fabledsword.minstrel.admin.data.AdminUsersRepository
import com.fabledsword.minstrel.api.ErrorCopy
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.recoveries
import com.fabledsword.minstrel.events.EventsStream
import com.fabledsword.minstrel.models.RequestRef
import dagger.hilt.android.lifecycle.HiltViewModel
@@ -35,6 +37,7 @@ class AdminRequestsViewModel @Inject constructor(
private val repository: AdminRequestsRepository,
private val usersRepository: AdminUsersRepository,
private val eventsStream: EventsStream,
networkStatus: NetworkStatusController,
) : ViewModel() {
private val internal = MutableStateFlow<AdminRequestsUiState>(AdminRequestsUiState.Loading)
@@ -47,6 +50,13 @@ class AdminRequestsViewModel @Inject constructor(
.filter { it.kind in RELEVANT_EVENT_KINDS }
.collect { refresh() }
}
// Screen-level auto-recovery (issue #1245): reload a failed list
// when server health returns instead of waiting for a manual pull.
viewModelScope.launch {
networkStatus.recoveries().collect {
if (internal.value is AdminRequestsUiState.Error) refresh()
}
}
}
fun refresh(): Job = viewModelScope.launch {
@@ -4,6 +4,8 @@ import androidx.lifecycle.ViewModel
import androidx.lifecycle.viewModelScope
import com.fabledsword.minstrel.admin.data.AdminUsersRepository
import com.fabledsword.minstrel.api.ErrorCopy
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.recoveries
import com.fabledsword.minstrel.models.AdminUserRef
import dagger.hilt.android.lifecycle.HiltViewModel
import kotlinx.coroutines.Job
@@ -23,6 +25,7 @@ sealed interface AdminUsersUiState {
@HiltViewModel
class AdminUsersViewModel @Inject constructor(
private val repository: AdminUsersRepository,
networkStatus: NetworkStatusController,
) : ViewModel() {
private val internal = MutableStateFlow<AdminUsersUiState>(AdminUsersUiState.Loading)
@@ -30,6 +33,13 @@ class AdminUsersViewModel @Inject constructor(
init {
refresh()
// Screen-level auto-recovery (issue #1245): reload a failed list
// when server health returns instead of waiting for a manual pull.
viewModelScope.launch {
networkStatus.recoveries().collect {
if (internal.value is AdminUsersUiState.Error) refresh()
}
}
}
fun refresh(): Job = viewModelScope.launch {
@@ -1,6 +1,7 @@
package com.fabledsword.minstrel.api
import com.fabledsword.minstrel.BuildConfig
import com.fabledsword.minstrel.connectivity.ReachabilityReportingInterceptor
import com.jakewharton.retrofit2.converter.kotlinx.serialization.asConverterFactory
import dagger.Module
import dagger.Provides
@@ -45,9 +46,15 @@ object NetworkModule {
fun provideOkHttp(
baseUrl: BaseUrlInterceptor,
auth: AuthCookieInterceptor,
reachability: ReachabilityReportingInterceptor,
logging: HttpLoggingInterceptor,
): OkHttpClient =
OkHttpClient.Builder()
// ReachabilityReportingInterceptor MUST run first: it identifies
// Minstrel-bound requests by the still-unrewritten PLACEHOLDER_HOST
// (so external artwork fetches don't read as server reachability)
// and observes the final transport outcome by wrapping the chain.
.addInterceptor(reachability)
// AuthCookieInterceptor MUST run before BaseUrlInterceptor.
// Both scope on `host == PLACEHOLDER_HOST` to distinguish
// Minstrel-server requests from external image fetches
@@ -39,11 +39,18 @@ data class StreamTokenRequest(
/**
* Response body. [url] is a fully-formed stream URL with [token] and
* [exp] already embedded as query params — callers pass it verbatim
* to `AVTransport.SetAVTransportURI`.
* to `AVTransport.SetAVTransportURI`. [mime] + [title] are the bits
* the client needs to build DIDL-Lite metadata for that call: Sonos
* rejects empty DIDL with vendor error 1023, so the server hands back
* the track's MIME (from `tracks.file_format`) and title so the
* client can populate `<res protocolInfo>` and `<dc:title>` without
* a follow-up round trip.
*/
@Serializable
data class StreamTokenResponse(
val token: String,
val exp: Long,
val url: String,
val mime: String = "audio/mpeg",
val title: String = "",
)
@@ -0,0 +1,36 @@
package com.fabledsword.minstrel.api.endpoints
import kotlinx.serialization.SerialName
import kotlinx.serialization.Serializable
import kotlinx.serialization.json.JsonElement
import retrofit2.http.Body
import retrofit2.http.POST
/**
* Retrofit interface for device diagnostics ingest (M9). One call
* uploads a batch of buffered events. The server stores them only when
* the account's debug_mode_enabled flag is on (it returns 204 otherwise,
* which the uploader treats as success so it can drop the batch).
*/
interface DiagnosticsApi {
@POST("api/diagnostics")
suspend fun report(@Body body: DiagnosticsReportRequest)
}
@Serializable
data class DiagnosticsReportRequest(
@SerialName("client_id") val clientId: String,
@SerialName("app_version") val appVersion: String? = null,
@SerialName("os_version") val osVersion: String? = null,
val events: List<DiagnosticEventWire>,
)
@Serializable
data class DiagnosticEventWire(
val kind: String,
// Device-clock epoch milliseconds. The server stamps its own
// received_at; both are stored so a skewed device clock is visible.
@SerialName("occurred_at") val occurredAt: Long,
// Opaque structured payload carrying the event sub-type + fields.
val payload: JsonElement,
)
@@ -2,6 +2,7 @@ package com.fabledsword.minstrel.api.endpoints
import com.fabledsword.minstrel.models.wire.AlbumDetailWire
import com.fabledsword.minstrel.models.wire.ArtistDetailWire
import com.fabledsword.minstrel.models.wire.ArtistWire
import com.fabledsword.minstrel.models.wire.TrackWire
import retrofit2.http.GET
import retrofit2.http.Path
@@ -35,9 +36,26 @@ interface LibraryApi {
@GET("api/artists/{id}/tracks")
suspend fun getArtistTracks(@Path("id") id: String): List<TrackWire>
@GET("api/artists/{id}/similar")
suspend fun getSimilarArtists(
@Path("id") id: String,
@Query("limit") limit: Int = SIMILAR_ARTISTS_LIMIT,
): List<ArtistWire>
@GET("api/artists/{id}/top-tracks")
suspend fun getArtistTopTracks(
@Path("id") id: String,
@Query("limit") limit: Int = TOP_TRACKS_LIMIT,
): List<TrackWire>
@GET("api/albums/{id}")
suspend fun getAlbumDetail(@Path("id") id: String): AlbumDetailWire
@GET("api/library/shuffle")
suspend fun shuffleLibrary(@Query("limit") limit: Int = 100): List<TrackWire>
private companion object {
const val SIMILAR_ARTISTS_LIMIT = 12
const val TOP_TRACKS_LIMIT = 5
}
}
@@ -1,6 +1,7 @@
package com.fabledsword.minstrel.auth
import com.fabledsword.minstrel.api.endpoints.AuthApi
import com.fabledsword.minstrel.api.endpoints.MeApi
import com.fabledsword.minstrel.di.ApplicationScope
import com.fabledsword.minstrel.models.UserRef
import com.fabledsword.minstrel.models.wire.LoginRequestBody
@@ -35,6 +36,7 @@ class AuthController @Inject constructor(
retrofit: Retrofit,
) {
private val api: AuthApi = retrofit.create()
private val meApi: MeApi = retrofit.create()
private val currentUserState = MutableStateFlow<UserRef?>(null)
val currentUser: StateFlow<UserRef?> = currentUserState.asStateFlow()
@@ -55,6 +57,10 @@ class AuthController @Inject constructor(
currentUserState.value = raw?.let { decodeUser(it) }
}
}
// Refresh from /api/me on startup so a remotely-changed account
// flag (e.g. admin enabling debug mode, M9) reaches the device
// without a re-login. Best-effort; failures keep the cached value.
scope.launch { refreshProfile() }
}
/**
@@ -72,9 +78,32 @@ class AuthController @Inject constructor(
)
currentUserState.value = user
authStore.setUserJson(json.encodeToString(UserRef.serializer(), user))
// Pull the fuller /me shape (carries debug_mode_enabled) right
// after login so the diagnostics gate is correct without waiting
// for the next startup refresh.
scope.launch { refreshProfile() }
return user
}
/**
* Re-fetch the caller's profile from /api/me and update currentUser
* + persisted userJson. Carries account-level flags (debug mode)
* that aren't in the login response. Best-effort: a network failure
* leaves the cached identity untouched. No-op when signed out.
*/
suspend fun refreshProfile() {
if (!isSignedIn) return
val p = runCatching { meApi.getProfile() }.getOrNull() ?: return
val user = UserRef(
id = p.id,
username = p.username,
isAdmin = p.isAdmin,
debugModeEnabled = p.debugModeEnabled,
)
currentUserState.value = user
authStore.setUserJson(json.encodeToString(UserRef.serializer(), user))
}
/**
* Clears the local session immediately and best-effort hits
* `/api/auth/logout` so the server can drop its session row.
@@ -59,6 +59,9 @@ class AuthStore @Inject constructor(
private val cacheSettingsState = MutableStateFlow(CacheSettings.DEFAULT)
val cacheSettings: StateFlow<CacheSettings> = cacheSettingsState.asStateFlow()
private val diagnosticsOptOutState = MutableStateFlow(false)
val diagnosticsOptOut: StateFlow<Boolean> = diagnosticsOptOutState.asStateFlow()
private val json = Json { ignoreUnknownKeys = true }
init {
@@ -70,6 +73,7 @@ class AuthStore @Inject constructor(
themeModeState.value = row?.themeMode
clientIdState.value = row?.clientId
cacheSettingsState.value = decodeCacheSettings(row?.cacheSettingsJson)
diagnosticsOptOutState.value = row?.diagnosticsOptOut ?: false
}
}
}
@@ -112,6 +116,11 @@ class AuthStore @Inject constructor(
scope.launch { persistCacheSettings(encoded) }
}
fun setDiagnosticsOptOut(value: Boolean) {
diagnosticsOptOutState.value = value
scope.launch { persistDiagnosticsOptOut(value) }
}
private suspend fun persistCookie(value: String?) {
if (dao.get() == null) {
dao.upsert(currentEntity().copy(sessionCookie = value))
@@ -160,6 +169,14 @@ class AuthStore @Inject constructor(
}
}
private suspend fun persistDiagnosticsOptOut(value: Boolean) {
if (dao.get() == null) {
dao.upsert(currentEntity().copy(diagnosticsOptOut = value))
} else {
dao.setDiagnosticsOptOut(value)
}
}
private fun currentEntity(): AuthSessionEntity = AuthSessionEntity(
id = ROW_ID,
sessionCookie = sessionCookieState.value,
@@ -171,6 +188,7 @@ class AuthStore @Inject constructor(
CacheSettings.serializer(),
cacheSettingsState.value,
),
diagnosticsOptOut = diagnosticsOptOutState.value,
)
companion object {
@@ -16,6 +16,7 @@ import com.fabledsword.minstrel.cache.db.dao.CachedResumeStateDao
import com.fabledsword.minstrel.cache.db.dao.CachedPlaylistTrackDao
import com.fabledsword.minstrel.cache.db.dao.CachedQuarantineDao
import com.fabledsword.minstrel.cache.db.dao.CachedTrackDao
import com.fabledsword.minstrel.cache.db.dao.DiagnosticEventDao
import com.fabledsword.minstrel.cache.db.dao.SyncMetadataDao
import com.fabledsword.minstrel.cache.db.entities.AudioCacheIndexEntity
import com.fabledsword.minstrel.cache.db.entities.AuthSessionEntity
@@ -30,6 +31,7 @@ import com.fabledsword.minstrel.cache.db.entities.CachedResumeStateEntity
import com.fabledsword.minstrel.cache.db.entities.CachedPlaylistTrackEntity
import com.fabledsword.minstrel.cache.db.entities.CachedQuarantineEntity
import com.fabledsword.minstrel.cache.db.entities.CachedTrackEntity
import com.fabledsword.minstrel.cache.db.entities.DiagnosticEventEntity
import com.fabledsword.minstrel.cache.db.entities.SyncMetadataEntity
/**
@@ -61,8 +63,11 @@ import com.fabledsword.minstrel.cache.db.entities.SyncMetadataEntity
CachedHomeIndexEntity::class,
CachedHistorySnapshotEntity::class,
AuthSessionEntity::class,
DiagnosticEventEntity::class,
],
version = 6,
// v7: + diagnostic_events table (M9) and the diagnosticsOptOut column
// on auth_session. Pre-v1 destructive fallback rebuilds on mismatch.
version = 7,
exportSchema = true,
)
@TypeConverters(MinstrelTypeConverters::class)
@@ -81,4 +86,5 @@ abstract class AppDatabase : RoomDatabase() {
abstract fun cachedHomeIndexDao(): CachedHomeIndexDao
abstract fun cachedHistorySnapshotDao(): CachedHistorySnapshotDao
abstract fun authSessionDao(): AuthSessionDao
abstract fun diagnosticEventDao(): DiagnosticEventDao
}
@@ -10,6 +10,7 @@ import com.fabledsword.minstrel.cache.db.dao.CachedHistorySnapshotDao
import com.fabledsword.minstrel.cache.db.dao.CachedHomeIndexDao
import com.fabledsword.minstrel.cache.db.dao.CachedLikeDao
import com.fabledsword.minstrel.cache.db.dao.CachedMutationDao
import com.fabledsword.minstrel.cache.db.dao.DiagnosticEventDao
import com.fabledsword.minstrel.cache.db.dao.CachedPlaylistDao
import com.fabledsword.minstrel.cache.db.dao.CachedPlaylistTrackDao
import com.fabledsword.minstrel.cache.db.dao.CachedQuarantineDao
@@ -105,5 +106,10 @@ object DatabaseModule {
fun provideAudioCacheIndexDao(db: AppDatabase): AudioCacheIndexDao =
db.audioCacheIndexDao()
@Provides
@Singleton
fun provideDiagnosticEventDao(db: AppDatabase): DiagnosticEventDao =
db.diagnosticEventDao()
private const val DATABASE_NAME = "minstrel.db"
}
@@ -42,4 +42,8 @@ interface AuthSessionDao {
/** Partial update: change only the serialized cache settings. */
@Query("UPDATE auth_session SET cacheSettingsJson = :json WHERE id = 0")
suspend fun setCacheSettingsJson(json: String?)
/** Partial update: change only the per-device diagnostics opt-out. */
@Query("UPDATE auth_session SET diagnosticsOptOut = :optOut WHERE id = 0")
suspend fun setDiagnosticsOptOut(optOut: Boolean)
}
@@ -24,6 +24,14 @@ interface CachedAlbumDao {
@Query("SELECT * FROM cached_albums WHERE id = :id")
fun observeById(id: String): Flow<CachedAlbumEntity?>
@Query(
"SELECT * FROM cached_albums " +
"WHERE title LIKE '%' || :q || '%' COLLATE NOCASE " +
"ORDER BY sortTitle COLLATE NOCASE ASC " +
"LIMIT :limit",
)
suspend fun searchByTitle(q: String, limit: Int): List<CachedAlbumEntity>
@Query("SELECT id FROM cached_albums WHERE fetchedAt < :before LIMIT :limit")
suspend fun idsStaleBefore(before: Long, limit: Int): List<String>
@@ -18,6 +18,14 @@ interface CachedArtistDao {
@Query("SELECT * FROM cached_artists WHERE id = :id")
fun observeById(id: String): Flow<CachedArtistEntity?>
@Query(
"SELECT * FROM cached_artists " +
"WHERE name LIKE '%' || :q || '%' COLLATE NOCASE " +
"ORDER BY sortName COLLATE NOCASE ASC " +
"LIMIT :limit",
)
suspend fun searchByName(q: String, limit: Int): List<CachedArtistEntity>
@Query("SELECT id FROM cached_artists WHERE fetchedAt < :before LIMIT :limit")
suspend fun idsStaleBefore(before: Long, limit: Int): List<String>
@@ -31,6 +31,20 @@ interface CachedPlaylistDao {
@Query("SELECT * FROM cached_playlists WHERE id = :id")
suspend fun getById(id: String): CachedPlaylistEntity?
/**
* Per-playlist count of member tracks resident in the audio cache index.
* LEFT JOINs so playlists with zero cached tracks still appear
* (cachedCount = 0). Drives the offline "fully cached" greying.
*/
@Query(
"SELECT p.id AS playlistId, COUNT(a.trackId) AS cachedCount " +
"FROM cached_playlists p " +
"LEFT JOIN cached_playlist_tracks t ON t.playlistId = p.id " +
"LEFT JOIN audio_cache_index a ON a.trackId = t.trackId " +
"GROUP BY p.id",
)
fun observeCachedCounts(): Flow<List<PlaylistCachedCount>>
@Insert(onConflict = OnConflictStrategy.REPLACE)
suspend fun upsertAll(rows: List<CachedPlaylistEntity>)
@@ -24,6 +24,14 @@ interface CachedTrackDao {
@Query("SELECT * FROM cached_tracks WHERE id IN (:ids)")
suspend fun getByIds(ids: List<String>): List<CachedTrackEntity>
@Query(
"SELECT * FROM cached_tracks " +
"WHERE title LIKE '%' || :q || '%' COLLATE NOCASE " +
"ORDER BY title COLLATE NOCASE ASC " +
"LIMIT :limit",
)
suspend fun searchByTitle(q: String, limit: Int): List<CachedTrackEntity>
@Query("SELECT id FROM cached_tracks WHERE fetchedAt < :before LIMIT :limit")
suspend fun idsStaleBefore(before: Long, limit: Int): List<String>
@@ -0,0 +1,35 @@
package com.fabledsword.minstrel.cache.db.dao
import androidx.room.Dao
import androidx.room.Insert
import androidx.room.Query
import com.fabledsword.minstrel.cache.db.entities.DiagnosticEventEntity
@Dao
interface DiagnosticEventDao {
@Insert
suspend fun insert(row: DiagnosticEventEntity): Long
/** FIFO drain order so the uploader sends oldest-first. */
@Query("SELECT * FROM diagnostic_events ORDER BY id ASC LIMIT :limit")
suspend fun takeBatch(limit: Int): List<DiagnosticEventEntity>
@Query("SELECT COUNT(*) FROM diagnostic_events")
suspend fun count(): Int
@Query("DELETE FROM diagnostic_events WHERE id IN (:ids)")
suspend fun deleteByIds(ids: List<Long>)
/**
* Ring-buffer trim: drop the oldest rows beyond [keep]. Called after
* insert so a long offline stretch can't grow the buffer unbounded.
*/
@Query(
"DELETE FROM diagnostic_events WHERE id NOT IN " +
"(SELECT id FROM diagnostic_events ORDER BY id DESC LIMIT :keep)",
)
suspend fun trimToNewest(keep: Int)
@Query("DELETE FROM diagnostic_events")
suspend fun clear()
}
@@ -0,0 +1,11 @@
package com.fabledsword.minstrel.cache.db.dao
/**
* Projection: how many of a playlist's member tracks are resident in the audio
* cache index. Backs the offline "fully cached" greying — a playlist is fully
* available offline when [cachedCount] reaches its `trackCount`.
*/
data class PlaylistCachedCount(
val playlistId: String,
val cachedCount: Int,
)
@@ -36,4 +36,11 @@ data class AuthSessionEntity(
* CacheSettings shape evolves.
*/
val cacheSettingsJson: String? = null,
/**
* Per-device opt-out of diagnostics reporting (M9). When the
* account's debug mode is enabled by an admin, the user can still
* turn reporting OFF on this device (battery/privacy) — that local
* choice lives here. Default false = honor the account flag.
*/
val diagnosticsOptOut: Boolean = false,
)
@@ -14,6 +14,11 @@ import kotlinx.datetime.Instant
* - "rediscover_artists"
* - "most_played_tracks"
* - "last_played_artists"
* - "you_might_like_albums"
* - "you_might_like_artists"
*
* New `section` values need no schema change — the column stores the
* string verbatim and the DAO keys on it.
*
* `entityType` is "album" | "artist" | "track" — dispatches hydration
* to the right per-entity endpoint when a tile is rendered.
@@ -0,0 +1,28 @@
package com.fabledsword.minstrel.cache.db.entities
import androidx.room.Entity
import androidx.room.PrimaryKey
/**
* One buffered device-diagnostics event (M9). The DiagnosticsReporter
* writes rows here when the account's debug mode is on; the
* DiagnosticsUploader drains them to POST /api/diagnostics and deletes
* on success.
*
* This is a deliberate ring buffer that is NOT routed through the offline
* MutationQueue: diagnostics are high-volume, best-effort telemetry, and
* the bug we're chasing (roaming / dead-zone recovery) happens WHILE
* offline — so events must persist locally through the dead zone and
* upload on recovery, without clogging the user-data mutation replay.
*
* `occurredAtMillis` is the device-clock epoch-ms when the event happened
* (the server also stamps its own received_at). `payloadJson` is an
* opaque JSON object carrying the event sub-type + fields.
*/
@Entity(tableName = "diagnostic_events")
data class DiagnosticEventEntity(
@PrimaryKey(autoGenerate = true) val id: Long = 0,
val kind: String,
val occurredAtMillis: Long,
val payloadJson: String,
)
@@ -2,12 +2,18 @@ package com.fabledsword.minstrel.cache.mutations
import com.fabledsword.minstrel.cache.db.dao.CachedMutationDao
import com.fabledsword.minstrel.cache.db.entities.CachedMutationEntity
import kotlinx.coroutines.channels.BufferOverflow
import kotlinx.coroutines.flow.MutableSharedFlow
import kotlinx.coroutines.flow.SharedFlow
import kotlinx.coroutines.flow.asSharedFlow
import kotlinx.serialization.Serializable
import kotlinx.serialization.encodeToString
import kotlinx.serialization.json.Json
import javax.inject.Inject
import javax.inject.Singleton
private const val QUEUED_MESSAGE = "Saved — will sync when online"
/**
* Stable mutation kinds the queue knows how to replay. Strings are
* persisted in `cached_mutations.kind` so renaming a variant breaks
@@ -22,6 +28,13 @@ object MutationKind {
const val PLAY_OFFLINE: String = "play_offline"
const val REQUEST_CANCEL: String = "request_cancel"
const val PLAYBACK_ERROR_REPORT: String = "playback_error_report"
// Durable close-by-id for a play whose server row is already open
// (play_started succeeded). Distinct from PLAY_OFFLINE, which records
// a *whole* play when no server row exists. Using close-by-id on
// background avoids the duplicate + orphan row the old offline-on-stop
// path produced (see 2026-06-11 contract audit).
const val PLAY_ENDED: String = "play_ended"
}
/**
@@ -71,61 +84,79 @@ class MutationQueue @Inject constructor(
private val dao: CachedMutationDao,
private val json: Json,
) {
// capacity=1 DROP_OLDEST so a burst of user enqueues (e.g. liking N
// tracks while offline) surfaces as one snackbar rather than queueing
// N. replay=0 because a hint observed at enqueue time isn't useful
// to a screen that mounts later.
private val _userEnqueueHints = MutableSharedFlow<String>(
replay = 0,
extraBufferCapacity = 1,
onBufferOverflow = BufferOverflow.DROP_OLDEST,
)
/**
* Hint stream consumed by [com.fabledsword.minstrel.shared.widgets.ShellScaffold]
* to surface "Saved — will sync when online" as a snackbar whenever a
* user-driven write hits the offline-fallback path. Background-only
* enqueues (play-events, playback-error reports) do not emit — those
* fire from non-foreground paths where a snackbar would be either
* dropped (no shell mounted) or jarring (lock-screen toggle).
*/
val userEnqueueHints: SharedFlow<String> = _userEnqueueHints.asSharedFlow()
suspend fun enqueueLikeToggle(
entityType: String,
entityId: String,
desiredState: Boolean,
): Long = dao.insert(
CachedMutationEntity(
kind = MutationKind.LIKE_TOGGLE,
payload = json.encodeToString(
): Long = insertUserDriven(
MutationKind.LIKE_TOGGLE,
json.encodeToString(
LikeTogglePayload.serializer(),
LikeTogglePayload(entityType, entityId, desiredState),
),
),
)
suspend fun enqueueRequestCreate(payload: RequestCreatePayload): Long = dao.insert(
CachedMutationEntity(
kind = MutationKind.REQUEST_CREATE,
payload = json.encodeToString(RequestCreatePayload.serializer(), payload),
),
suspend fun enqueueRequestCreate(payload: RequestCreatePayload): Long = insertUserDriven(
MutationKind.REQUEST_CREATE,
json.encodeToString(RequestCreatePayload.serializer(), payload),
)
suspend fun enqueueQuarantineUnflag(trackId: String): Long = dao.insert(
CachedMutationEntity(
kind = MutationKind.QUARANTINE_UNFLAG,
payload = json.encodeToString(
suspend fun enqueueQuarantineUnflag(trackId: String): Long = insertUserDriven(
MutationKind.QUARANTINE_UNFLAG,
json.encodeToString(
QuarantineUnflagPayload.serializer(),
QuarantineUnflagPayload(trackId),
),
),
)
suspend fun enqueuePlaylistAppend(
playlistId: String,
trackIds: List<String>,
): Long = dao.insert(
CachedMutationEntity(
kind = MutationKind.PLAYLIST_APPEND,
payload = json.encodeToString(
): Long = insertUserDriven(
MutationKind.PLAYLIST_APPEND,
json.encodeToString(
PlaylistAppendPayload.serializer(),
PlaylistAppendPayload(playlistId, trackIds),
),
),
)
suspend fun enqueueQuarantineFlag(
trackId: String,
reason: String,
notes: String,
): Long = dao.insert(
CachedMutationEntity(
kind = MutationKind.QUARANTINE_FLAG,
payload = json.encodeToString(
): Long = insertUserDriven(
MutationKind.QUARANTINE_FLAG,
json.encodeToString(
QuarantineFlagPayload.serializer(),
QuarantineFlagPayload(trackId, reason, notes),
),
)
suspend fun enqueueRequestCancel(requestId: String): Long = insertUserDriven(
MutationKind.REQUEST_CANCEL,
json.encodeToString(
RequestCancelPayload.serializer(),
RequestCancelPayload(requestId),
),
)
@@ -136,22 +167,29 @@ class MutationQueue @Inject constructor(
),
)
suspend fun enqueueRequestCancel(requestId: String): Long = dao.insert(
CachedMutationEntity(
kind = MutationKind.REQUEST_CANCEL,
payload = json.encodeToString(
RequestCancelPayload.serializer(),
RequestCancelPayload(requestId),
),
),
)
suspend fun enqueuePlaybackErrorReport(payload: PlaybackErrorReportPayload): Long = dao.insert(
CachedMutationEntity(
kind = MutationKind.PLAYBACK_ERROR_REPORT,
payload = json.encodeToString(PlaybackErrorReportPayload.serializer(), payload),
),
)
/**
* Durable close-by-id for an already-open server play row. Background
* path only (no user hint) — same as [enqueuePlayOffline].
*/
suspend fun enqueuePlayEnded(payload: PlayEndedPayload): Long = dao.insert(
CachedMutationEntity(
kind = MutationKind.PLAY_ENDED,
payload = json.encodeToString(PlayEndedPayload.serializer(), payload),
),
)
private suspend fun insertUserDriven(kind: String, payload: String): Long {
val id = dao.insert(CachedMutationEntity(kind = kind, payload = payload))
_userEnqueueHints.tryEmit(QUEUED_MESSAGE)
return id
}
}
/**
@@ -203,6 +241,19 @@ data class PlayOfflinePayload(
val source: String? = null,
)
/**
* Persisted payload for `MutationKind.PLAY_ENDED`. The replayer re-fires
* `POST /api/events` with type=play_ended, closing the already-open server
* row [playEventId] at [durationPlayedMs]. Server-side close-by-id is
* idempotent (re-ending updates the same row and skips a duplicate
* skip_events insert), so a lost-response retry can't duplicate history.
*/
@Serializable
data class PlayEndedPayload(
val playEventId: String,
val durationPlayedMs: Long,
)
/**
* Persisted payload for `MutationKind.REQUEST_CANCEL` — the
* `POST /api/requests/{id}/cancel` call lost during a connectivity
@@ -1,3 +1,5 @@
@file:Suppress("TooManyFunctions") // one dispatcher per mutation kind + replay helpers
package com.fabledsword.minstrel.cache.mutations
import com.fabledsword.minstrel.api.endpoints.AppendTracksRequest
@@ -10,6 +12,9 @@ import com.fabledsword.minstrel.api.endpoints.PlaybackErrorsApi
import com.fabledsword.minstrel.api.endpoints.PlaylistsApi
import com.fabledsword.minstrel.api.endpoints.QuarantineApi
import com.fabledsword.minstrel.api.endpoints.RequestsApi
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.ServerHealth
import com.fabledsword.minstrel.models.wire.PlayEndedRequest
import com.fabledsword.minstrel.models.wire.PlayOfflineRequest
import com.fabledsword.minstrel.auth.AuthStore
import com.fabledsword.minstrel.cache.db.dao.CachedMutationDao
@@ -19,12 +24,16 @@ import com.fabledsword.minstrel.likes.data.LikesRepository
import com.fabledsword.minstrel.models.wire.CreateRequestBody
import kotlinx.coroutines.CoroutineScope
import kotlinx.coroutines.flow.distinctUntilChanged
import kotlinx.coroutines.flow.filter
import kotlinx.coroutines.flow.filterNotNull
import kotlinx.coroutines.flow.map
import kotlinx.coroutines.launch
import kotlinx.coroutines.sync.Mutex
import kotlinx.coroutines.sync.withLock
import kotlinx.datetime.Clock
import kotlinx.serialization.SerializationException
import kotlinx.serialization.json.Json
import retrofit2.HttpException
import retrofit2.Retrofit
import retrofit2.create
import javax.inject.Inject
@@ -34,25 +43,24 @@ import javax.inject.Singleton
* Drains the offline MutationQueue. Reads pending rows in FIFO order,
* re-fires each call against the raw Retrofit API (NOT the high-level
* Repository, which would re-enqueue on failure → infinite loop), and
* deletes the row on 2xx success.
* resolves each to an [Outcome]:
* - SENT — server accepted (2xx) → delete the row.
* - DROP — permanently un-sendable (4xx, unknown kind/variant, corrupt
* payload) → delete the row so it can't wedge the queue forever.
* - RETRY — transient (transport, 5xx, 408, 429) → leave queued, bump
* attempts, retry on the next trigger.
*
* Triggered by:
* - app start (MinstrelApplication @Inject forces the singleton's
* init block to subscribe)
* - every time AuthStore.sessionCookie transitions to a non-null
* value (sign-in, app launch with persisted cookie)
* - app start (MinstrelApplication @Inject forces the singleton's init)
* - AuthStore.sessionCookie transitioning to non-null (sign-in / cold
* start with a persisted cookie)
* - NetworkStatusController recovering to Healthy (server reachable again
* mid-session — covers offline→online without needing a cold start)
*
* Single in-flight via the `mutex` — back-to-back triggers collapse
* into one drain pass. A proper connectivity-listener / WorkManager
* job that re-tries on Wi-Fi-came-back is a follow-up; for MVP the
* sign-in + app-open triggers cover the common offline → online
* transition (operator unlocks phone, opens app).
*
* Failed rows stay in the queue with their `attempts` and
* `lastAttemptAt` updated. They get another chance on the next
* trigger; no exponential backoff or max-attempts yet — the rows are
* small (low single-digit count for typical use) so retry-forever is
* cheap.
* Single in-flight via the [mutex]; back-to-back triggers collapse into one
* drain pass. No exponential backoff / max-attempts: the rows are few for
* typical use, and permanent failures are now dropped rather than retried
* forever.
*/
@Singleton
class MutationReplayer @Inject constructor(
@@ -60,6 +68,7 @@ class MutationReplayer @Inject constructor(
private val authStore: AuthStore,
private val json: Json,
@ApplicationScope private val scope: CoroutineScope,
networkStatus: NetworkStatusController,
retrofit: Retrofit,
) {
private val likesApi: LikesApi = retrofit.create()
@@ -72,6 +81,8 @@ class MutationReplayer @Inject constructor(
private val mutex = Mutex()
private enum class Outcome { SENT, DROP, RETRY }
init {
scope.launch {
authStore.sessionCookie
@@ -79,6 +90,16 @@ class MutationReplayer @Inject constructor(
.distinctUntilChanged()
.collect { drainSafe() }
}
// Server-reachable-again: replay queued writes the moment the
// health signal recovers, so an offline→online transition mid-
// session doesn't wait for the next cold start / re-auth.
scope.launch {
networkStatus.state
.map { it == ServerHealth.Healthy }
.distinctUntilChanged()
.filter { it }
.collect { drainSafe() }
}
}
/** Public entry-point; coalesces concurrent triggers via a Mutex. */
@@ -93,69 +114,98 @@ class MutationReplayer @Inject constructor(
private suspend fun drain() {
val rows = dao.getAll()
// Collapse superseded like-toggles: only the latest desired state per
// (entity) is replayed; older toggles for the same entity are dropped
// unsent. Without this, partial-failure + differential retry could
// replay an older toggle last and invert the final like state.
val superseded = supersededLikeToggleIds(rows)
for (row in rows) {
val sent = runCatching { dispatch(row) }.getOrDefault(false)
if (sent) {
if (row.id in superseded) {
dao.delete(row.id)
} else {
dao.recordAttempt(row.id, Clock.System.now())
continue
}
when (outcomeFor(row)) {
Outcome.SENT, Outcome.DROP -> dao.delete(row.id)
Outcome.RETRY -> dao.recordAttempt(row.id, Clock.System.now())
}
}
}
/** Row ids of like-toggles superseded by a later toggle for the same entity. */
private fun supersededLikeToggleIds(rows: List<CachedMutationEntity>): Set<Long> {
val latestByEntity = HashMap<String, Long>()
val superseded = HashSet<Long>()
rows.asSequence()
.filter { it.kind == MutationKind.LIKE_TOGGLE }
.forEach { row ->
val decoded = runCatching {
json.decodeFromString(LikeTogglePayload.serializer(), row.payload)
}.getOrNull()
if (decoded != null) {
val key = "${decoded.entityType}:${decoded.entityId}"
// `rows` is ascending by id, so a prior entry is always older.
latestByEntity.put(key, row.id)?.let(superseded::add)
}
}
return superseded
}
private suspend fun outcomeFor(row: CachedMutationEntity): Outcome = try {
dispatch(row)
} catch (e: HttpException) {
if (isPermanent(e.code())) Outcome.DROP else Outcome.RETRY
} catch (@Suppress("SwallowedException") e: SerializationException) {
// Corrupt / schema-incompatible payload — permanent, drop it.
Outcome.DROP
} catch (@Suppress("TooGenericExceptionCaught", "SwallowedException") e: Throwable) {
// Transport / unknown — leave queued for the next pass.
Outcome.RETRY
}
/** 4xx are permanent except the two "retry me" statuses. */
private fun isPermanent(code: Int): Boolean =
code in HTTP_CLIENT_ERR_MIN..HTTP_CLIENT_ERR_MAX &&
code != HTTP_TIMEOUT && code != HTTP_TOO_MANY
/**
* Returns true if the server accepted the call, false on any
* failure (transport, server error, decode). The replayer treats
* false as "leave the row for next pass" — it doesn't try to
* distinguish permanent rejections from transient ones; a row
* that consistently fails will accumulate attempt count and
* lastAttemptAt for diagnostics.
* Returns the outcome of replaying one row. SENT on a 2xx; DROP for an
* unknown kind/variant. Network / HTTP failures THROW and are classified
* by [outcomeFor] — dispatchers no longer swallow.
*/
private suspend fun dispatch(row: CachedMutationEntity): Boolean = when (row.kind) {
private suspend fun dispatch(row: CachedMutationEntity): Outcome = when (row.kind) {
MutationKind.LIKE_TOGGLE -> dispatchLikeToggle(row.payload)
MutationKind.REQUEST_CREATE -> dispatchRequestCreate(row.payload)
MutationKind.QUARANTINE_UNFLAG -> dispatchQuarantineUnflag(row.payload)
MutationKind.PLAYLIST_APPEND -> dispatchPlaylistAppend(row.payload)
MutationKind.QUARANTINE_FLAG -> dispatchQuarantineFlag(row.payload)
MutationKind.PLAY_OFFLINE -> dispatchPlayOffline(row.payload)
MutationKind.PLAY_ENDED -> dispatchPlayEnded(row.payload)
MutationKind.REQUEST_CANCEL -> dispatchRequestCancel(row.payload)
MutationKind.PLAYBACK_ERROR_REPORT -> dispatchPlaybackErrorReport(row.payload)
else -> {
// Unknown kind — drop the row by claiming success so a
// stale schema entry can't wedge the queue forever.
true
}
// Unknown kind — drop so a stale schema entry can't wedge the queue.
else -> Outcome.DROP
}
private suspend fun dispatchLikeToggle(payload: String): Boolean {
private suspend fun dispatchLikeToggle(payload: String): Outcome {
val decoded = json.decodeFromString(LikeTogglePayload.serializer(), payload)
val kindPath = when (decoded.entityType) {
LikesRepository.ENTITY_ARTIST -> "artists"
LikesRepository.ENTITY_ALBUM -> "albums"
LikesRepository.ENTITY_TRACK -> "tracks"
else -> return true // unknown variant — drop.
else -> return Outcome.DROP
}
return try {
if (decoded.desiredState) {
likesApi.like(kindPath, decoded.entityId)
} else {
likesApi.unlike(kindPath, decoded.entityId)
}
true
} catch (
@Suppress("TooGenericExceptionCaught", "SwallowedException") e: Throwable,
) {
// Intentional swallow: the row stays in the queue with
// attempts + lastAttemptAt incremented; next drain pass
// retries. Per-attempt diagnostic logging lands when we
// wire up Timber at the replayer level.
false
}
return Outcome.SENT
}
private suspend fun dispatchRequestCreate(payload: String): Boolean {
private suspend fun dispatchRequestCreate(payload: String): Outcome {
val decoded = json.decodeFromString(RequestCreatePayload.serializer(), payload)
val body = CreateRequestBody(
discoverApi.createRequest(
CreateRequestBody(
kind = decoded.kind,
artistMbid = decoded.artistMbid,
artistName = decoded.artistName,
@@ -163,56 +213,28 @@ class MutationReplayer @Inject constructor(
albumTitle = decoded.albumTitle,
trackMbid = decoded.trackMbid,
trackTitle = decoded.trackTitle,
),
)
return try {
discoverApi.createRequest(body)
true
} catch (
@Suppress("TooGenericExceptionCaught", "SwallowedException") e: Throwable,
) {
// Intentional swallow: the row stays in the queue with
// attempts + lastAttemptAt incremented; next drain pass
// retries. Per-attempt diagnostic logging lands when we
// wire up Timber at the replayer level.
false
}
return Outcome.SENT
}
private suspend fun dispatchQuarantineUnflag(payload: String): Boolean {
private suspend fun dispatchQuarantineUnflag(payload: String): Outcome {
val decoded = json.decodeFromString(QuarantineUnflagPayload.serializer(), payload)
return try {
quarantineApi.unflag(decoded.trackId)
true
} catch (
@Suppress("TooGenericExceptionCaught", "SwallowedException") e: Throwable,
) {
// Intentional swallow: the row stays in the queue with
// attempts + lastAttemptAt incremented; next drain pass
// retries. Per-attempt diagnostic logging lands when we
// wire up Timber at the replayer level.
false
}
return Outcome.SENT
}
private suspend fun dispatchPlaylistAppend(payload: String): Boolean {
private suspend fun dispatchPlaylistAppend(payload: String): Outcome {
val decoded = json.decodeFromString(PlaylistAppendPayload.serializer(), payload)
return try {
playlistsApi.appendTracks(
playlistId = decoded.playlistId,
body = AppendTracksRequest(trackIds = decoded.trackIds),
)
true
} catch (
@Suppress("TooGenericExceptionCaught", "SwallowedException") e: Throwable,
) {
// Intentional swallow: row stays queued; next drain pass retries.
false
}
return Outcome.SENT
}
private suspend fun dispatchQuarantineFlag(payload: String): Boolean {
private suspend fun dispatchQuarantineFlag(payload: String): Outcome {
val decoded = json.decodeFromString(QuarantineFlagPayload.serializer(), payload)
return try {
quarantineApi.flag(
FlagRequest(
trackId = decoded.trackId,
@@ -220,18 +242,11 @@ class MutationReplayer @Inject constructor(
notes = decoded.notes,
),
)
true
} catch (
@Suppress("TooGenericExceptionCaught", "SwallowedException") e: Throwable,
) {
// Intentional swallow: row stays queued; next drain pass retries.
false
}
return Outcome.SENT
}
private suspend fun dispatchPlayOffline(payload: String): Boolean {
private suspend fun dispatchPlayOffline(payload: String): Outcome {
val decoded = json.decodeFromString(PlayOfflinePayload.serializer(), payload)
return try {
eventsApi.playOffline(
PlayOfflineRequest(
trackId = decoded.trackId,
@@ -241,31 +256,28 @@ class MutationReplayer @Inject constructor(
source = decoded.source,
),
)
true
} catch (
@Suppress("TooGenericExceptionCaught", "SwallowedException") e: Throwable,
) {
// Intentional swallow: row stays queued; next drain pass retries.
false
}
return Outcome.SENT
}
private suspend fun dispatchRequestCancel(payload: String): Boolean {
private suspend fun dispatchPlayEnded(payload: String): Outcome {
val decoded = json.decodeFromString(PlayEndedPayload.serializer(), payload)
eventsApi.playEnded(
PlayEndedRequest(
playEventId = decoded.playEventId,
durationPlayedMs = decoded.durationPlayedMs,
),
)
return Outcome.SENT
}
private suspend fun dispatchRequestCancel(payload: String): Outcome {
val decoded = json.decodeFromString(RequestCancelPayload.serializer(), payload)
return try {
requestsApi.cancel(decoded.requestId)
true
} catch (
@Suppress("TooGenericExceptionCaught", "SwallowedException") e: Throwable,
) {
// Intentional swallow: row stays queued; next drain pass retries.
false
}
return Outcome.SENT
}
private suspend fun dispatchPlaybackErrorReport(payload: String): Boolean {
private suspend fun dispatchPlaybackErrorReport(payload: String): Outcome {
val decoded = json.decodeFromString(PlaybackErrorReportPayload.serializer(), payload)
return try {
playbackErrorsApi.report(
PlaybackErrorReportRequest(
trackId = decoded.trackId,
@@ -274,12 +286,13 @@ class MutationReplayer @Inject constructor(
clientId = decoded.clientId,
),
)
true
} catch (
@Suppress("TooGenericExceptionCaught", "SwallowedException") e: Throwable,
) {
// Intentional swallow: row stays queued; next drain pass retries.
false
return Outcome.SENT
}
private companion object {
const val HTTP_CLIENT_ERR_MIN = 400
const val HTTP_CLIENT_ERR_MAX = 499
const val HTTP_TIMEOUT = 408
const val HTTP_TOO_MANY = 429
}
}
@@ -0,0 +1,20 @@
package com.fabledsword.minstrel.cache.mutations
import androidx.lifecycle.ViewModel
import dagger.hilt.android.lifecycle.HiltViewModel
import kotlinx.coroutines.flow.Flow
import javax.inject.Inject
/**
* Hilt-injectable wrapper exposing [MutationQueue.userEnqueueHints] to
* ShellScaffold. The queue itself is an app-scoped singleton; this VM
* just bridges its SharedFlow into a `hiltViewModel()`-resolvable
* surface so ShellScaffold can collect it without an EntryPoint
* accessor. Mirrors PlaybackErrorViewModel.
*/
@HiltViewModel
class OfflineWriteHintViewModel @Inject constructor(
mutationQueue: MutationQueue,
) : ViewModel() {
val messages: Flow<String> = mutationQueue.userEnqueueHints
}
@@ -1,7 +1,10 @@
package com.fabledsword.minstrel.cache.sync
import com.fabledsword.minstrel.api.ErrorCopy
import com.fabledsword.minstrel.api.endpoints.SyncApi
import com.fabledsword.minstrel.auth.AuthStore
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.ServerHealth
import com.fabledsword.minstrel.cache.db.dao.CachedAlbumDao
import com.fabledsword.minstrel.cache.db.dao.CachedArtistDao
import com.fabledsword.minstrel.cache.db.dao.CachedTrackDao
@@ -15,9 +18,15 @@ import com.fabledsword.minstrel.models.wire.SyncAlbumWire
import com.fabledsword.minstrel.models.wire.SyncArtistWire
import com.fabledsword.minstrel.models.wire.SyncTrackWire
import kotlinx.coroutines.CoroutineScope
import kotlinx.coroutines.flow.MutableStateFlow
import kotlinx.coroutines.flow.StateFlow
import kotlinx.coroutines.flow.asStateFlow
import kotlinx.coroutines.flow.distinctUntilChanged
import kotlinx.coroutines.flow.filter
import kotlinx.coroutines.flow.filterNotNull
import kotlinx.coroutines.flow.map
import kotlinx.coroutines.launch
import kotlinx.coroutines.sync.Mutex
import kotlinx.datetime.Clock
import retrofit2.Retrofit
import retrofit2.create
@@ -50,10 +59,26 @@ class SyncController @Inject constructor(
private val trackDao: CachedTrackDao,
private val authStore: AuthStore,
@ApplicationScope private val scope: CoroutineScope,
networkStatus: NetworkStatusController,
retrofit: Retrofit,
) {
private val api: SyncApi = retrofit.create()
// Single in-flight guard: the cookie + health-recovery triggers (and a
// pull-to-refresh) can fire concurrently; coalesce them into one pass.
private val mutex = Mutex()
private val lastSyncErrorInternal = MutableStateFlow<String?>(null)
/**
* Human copy for the most recent sync failure; null after any clean
* pass. Lets the Library screen distinguish "empty because the first
* sync failed" (show error + retry) from "genuinely empty library"
* (show welcome copy) — a failed sync over a populated cache stays
* silent, since the cached content is still the better surface.
*/
val lastSyncError: StateFlow<String?> = lastSyncErrorInternal.asStateFlow()
init {
scope.launch {
authStore.sessionCookie
@@ -61,11 +86,27 @@ class SyncController @Inject constructor(
.distinctUntilChanged()
.collect { syncSafe() }
}
// Re-sync when the server becomes reachable again mid-session, so a
// delta missed while offline lands without waiting for a cold start.
scope.launch {
networkStatus.state
.map { it == ServerHealth.Healthy }
.distinctUntilChanged()
.filter { it }
.collect { syncSafe() }
}
}
/** Public entry-point for "Sync now" affordances. Swallows errors. */
/** Public entry-point for "Sync now" affordances. Swallows errors into [lastSyncError]. */
suspend fun syncSafe() {
if (!mutex.tryLock()) return
try {
runCatching { sync() }
.onSuccess { lastSyncErrorInternal.value = null }
.onFailure { lastSyncErrorInternal.value = ErrorCopy.fromThrowable(it) }
} finally {
mutex.unlock()
}
}
private suspend fun sync() {
@@ -14,11 +14,23 @@ import javax.inject.Inject
import javax.inject.Singleton
/**
* Single source of truth for the device's "is the internet usable
* right now" signal — wraps [ConnectivityManager] and exposes a hot
* cold-startable Flow that emits `false` while the active network
* lacks INTERNET + VALIDATED capabilities (airplane mode, no carrier,
* captive portal, etc.) and `true` once a usable network appears.
* Single source of truth for "does the device have a network link at
* all" — wraps [ConnectivityManager] and exposes a hot cold-startable
* Flow that emits `false` only when there is no active INTERNET-capable
* network (airplane mode, no carrier/Wi-Fi) and `true` once any network
* link appears.
*
* Deliberately does NOT require `NET_CAPABILITY_VALIDATED`. VALIDATED
* tracks whether Android reached its own WAN internet-validation probe
* (Google's `generate_204`) — which is the wrong question for a
* self-hosted server that is usually on the LAN. A transient WAN/DNS
* blip (or Android's periodic re-validation) momentarily drops VALIDATED
* while the Minstrel box stays perfectly reachable; gating on it flipped
* the app to Offline with no debounce and fast-failed in-flight playback
* via [com.fabledsword.minstrel.player.OfflineGatedDataSource]. The
* authority on whether *Minstrel* is reachable is the `/healthz` poll
* ([com.fabledsword.minstrel.connectivity.NetworkStatusController], which
* has its own failure hysteresis), not this coarse device-link signal.
*
* Used by the shell-level ConnectionErrorBanner; downstream
* repositories can also collect this to gate retry loops.
@@ -35,21 +47,22 @@ class ConnectivityObserver @Inject constructor(
.build()
val callback = object : ConnectivityManager.NetworkCallback() {
override fun onAvailable(network: Network) {
trySend(hasUsableInternet())
trySend(hasActiveNetwork())
}
override fun onLost(network: Network) {
trySend(hasUsableInternet())
trySend(hasActiveNetwork())
}
override fun onCapabilitiesChanged(
network: Network,
capabilities: NetworkCapabilities,
) {
// INTERNET only -- NOT VALIDATED. A WAN/validation flicker
// must not read as "device offline" when the LAN (and the
// Minstrel server on it) is still reachable. /healthz is the
// authority on server reachability.
trySend(
capabilities.hasCapability(
NetworkCapabilities.NET_CAPABILITY_INTERNET,
) &&
capabilities.hasCapability(
NetworkCapabilities.NET_CAPABILITY_VALIDATED,
),
)
}
@@ -57,14 +70,13 @@ class ConnectivityObserver @Inject constructor(
cm.registerNetworkCallback(request, callback)
// Seed the initial value so the banner doesn't flash before the
// first capability callback fires.
trySend(hasUsableInternet())
trySend(hasActiveNetwork())
awaitClose { cm.unregisterNetworkCallback(callback) }
}.distinctUntilChanged()
private fun hasUsableInternet(): Boolean {
private fun hasActiveNetwork(): Boolean {
val caps = cm.activeNetwork?.let { cm.getNetworkCapabilities(it) }
return caps != null &&
caps.hasCapability(NetworkCapabilities.NET_CAPABILITY_INTERNET) &&
caps.hasCapability(NetworkCapabilities.NET_CAPABILITY_VALIDATED)
caps.hasCapability(NetworkCapabilities.NET_CAPABILITY_INTERNET)
}
}
@@ -0,0 +1,176 @@
package com.fabledsword.minstrel.connectivity
import androidx.compose.runtime.staticCompositionLocalOf
import com.fabledsword.minstrel.BuildConfig
import com.fabledsword.minstrel.auth.AuthStore
import com.fabledsword.minstrel.di.ApplicationScope
import com.fabledsword.minstrel.update.api.HealthzApi
import com.fabledsword.minstrel.update.api.HealthzResponse
import com.fabledsword.minstrel.update.data.VersionResult
import com.fabledsword.minstrel.update.data.isVersionNewer
import kotlinx.coroutines.CoroutineScope
import kotlinx.coroutines.channels.Channel
import kotlinx.coroutines.delay
import kotlinx.coroutines.flow.MutableStateFlow
import kotlinx.coroutines.flow.StateFlow
import kotlinx.coroutines.flow.asStateFlow
import kotlinx.coroutines.launch
import retrofit2.Retrofit
import timber.log.Timber
import java.util.concurrent.atomic.AtomicLong
import javax.inject.Inject
import javax.inject.Singleton
private const val POLL_HEALTHY_MS = 5 * 60 * 1000L
private const val POLL_DEGRADED_MS = 20_000L
private const val ARBITRATE_MIN_GAP_MS = 2_000L
/**
* THE single authority on network/server reachability. Absorbs the former
* VersionCheckController (the /healthz poll + version parsing) and
* ServerHealthController (the tri-state derive) into one signal-driven unit so
* the app has one place that answers "can we reach Minstrel?" — not three
* half-systems reporting differently.
*
* Inputs (all funnel through a single-consumer intent channel for thread
* safety — [reportSuccess]/[reportFailure] are called from the audio read path
* and OkHttp threads):
* - device link transitions ([ConnectivityObserver]); link-return triggers an
* immediate probe so recovery is near-instant (fixes the sticky banner).
* - periodic /healthz probe, adaptive cadence (calm when Healthy, fast when down).
* - reportSuccess / reportFailure from the API interceptor, the audio data
* source, and the playback-error reporter.
* - recheck() from pull-to-refresh and the banner.
*
* Version compatibility is a byproduct of the same /healthz response.
*
* Constructed at launch via the construct-the-singleton trick in
* [com.fabledsword.minstrel.MinstrelApplication].
*/
@Singleton
class NetworkStatusController @Inject constructor(
@ApplicationScope private val scope: CoroutineScope,
connectivity: ConnectivityObserver,
private val authStore: AuthStore,
retrofit: Retrofit,
) {
private val api: HealthzApi = retrofit.create(HealthzApi::class.java)
private val machine = ReachabilityMachine()
private val lastProbeAtMs = AtomicLong(0)
private val stateInternal = MutableStateFlow(ServerHealth.Healthy)
val state: StateFlow<ServerHealth> = stateInternal.asStateFlow()
private val versionInternal = MutableStateFlow(VersionResult.SKIPPED)
val versionResult: StateFlow<VersionResult> = versionInternal.asStateFlow()
private sealed interface Intent {
data class Link(val up: Boolean) : Intent
data class Probe(val resp: HealthzResponse?) : Intent
object OpSuccess : Intent
object OpFailure : Intent
}
private val intents = Channel<Intent>(Channel.UNLIMITED)
init {
scope.launch { reduceLoop() }
scope.launch {
connectivity.online.collect { up ->
intents.trySend(Intent.Link(up))
if (up) probeOnce()
}
}
scope.launch { pollLoop() }
}
/** A real server op verifiably succeeded — self-proving recovery. Cheap; safe on hot paths. */
fun reportSuccess() {
if (stateInternal.value != ServerHealth.Healthy) intents.trySend(Intent.OpSuccess)
}
/** A real network op failed — triggers /healthz arbitration. No-op when already Offline. */
fun reportFailure() {
if (stateInternal.value == ServerHealth.Offline) return
intents.trySend(Intent.OpFailure)
}
/** One-shot recheck for pull-to-refresh and the banner. */
fun recheck() {
scope.launch { probeOnce(force = true) }
}
private suspend fun reduceLoop() {
for (intent in intents) {
val now = System.currentTimeMillis()
when (intent) {
is Intent.Link -> machine.onLinkChange(intent.up)
is Intent.Probe -> applyProbe(intent.resp, now)
Intent.OpSuccess -> machine.onSuccess()
Intent.OpFailure -> {
machine.onOpFailure(now)
// Arbitrate off the reducer thread: awaiting a stalled
// /healthz here would block a concurrent self-proving
// success from snapping us straight back to Healthy.
scope.launch { probeOnce() }
}
}
emit(machine.health())
}
}
private fun applyProbe(resp: HealthzResponse?, now: Long) {
if (resp == null) {
machine.onProbeFailure(now)
} else {
machine.onSuccess()
versionInternal.value = versionResultFor(resp)
}
}
private fun emit(next: ServerHealth) {
if (stateInternal.value != next) {
Timber.w("NetworkStatus -> %s", next)
stateInternal.value = next
}
}
private suspend fun pollLoop() {
probeOnce()
while (true) {
val interval =
if (stateInternal.value == ServerHealth.Healthy) POLL_HEALTHY_MS
else POLL_DEGRADED_MS
delay(interval)
probeOnce()
}
}
private suspend fun probeOnce(force: Boolean = false) {
// Startup guard: don't poll the localhost placeholder before AuthStore
// hydrates the real base URL — that false failure used to flash the banner.
if (authStore.baseUrl.value == AuthStore.DEFAULT_BASE_URL) return
val now = System.currentTimeMillis()
if (!force && now - lastProbeAtMs.get() < ARBITRATE_MIN_GAP_MS) return
lastProbeAtMs.set(now)
val resp = runCatching { api.check() }.getOrNull()
intents.trySend(Intent.Probe(resp))
}
private fun versionResultFor(resp: HealthzResponse): VersionResult {
val min = resp.minClientVersion
return when {
min.isEmpty() -> VersionResult.SKIPPED
isVersionNewer(min, BuildConfig.VERSION_NAME) -> VersionResult.TOO_OLD
else -> VersionResult.OK
}
}
}
/**
* Reactive [ServerHealth] snapshot provided once at the app root. Lets leaf
* composables (TrackRow gating, write-affordance disabling) branch on health
* without re-injecting the controller. Defaults to Healthy so previews/tests
* don't crash.
*/
val LocalServerHealth = staticCompositionLocalOf { ServerHealth.Healthy }
@@ -0,0 +1,88 @@
package com.fabledsword.minstrel.connectivity
internal const val ESCALATE_AFTER_MS = 120_000L
internal const val CORROBORATION_WINDOW_MS = 30_000L
internal const val CORROBORATION_OP_THRESHOLD = 2
/**
* Pure reachability state machine. No Android, no coroutines, no real clock —
* every entry point takes `nowMs`, so it is fully deterministic and unit-
* testable. [NetworkStatusController] wires real time + signals around it.
*
* Reachability (independent of the device link):
* - Reachable last evidence says the server answered.
* - Unstable a probe failed; arbitration/escalation pending.
* - Unreachable corroborated or sustained failure.
*
* [health] folds the device link over that: no link → Offline; otherwise the
* reachability maps Reachable→Healthy, Unstable→Unstable, Unreachable→ServerDown.
*
* Principle: **success is self-proving, failure is ambiguous.** [onSuccess]
* (a real server byte-read or API 2xx, or a successful /healthz) snaps straight
* back to Reachable. A failure only escalates when a /healthz probe corroborates
* it ([onProbeFailure]) — either via fresh op-failure corroboration or the
* sustained-time backstop.
*/
class ReachabilityMachine {
private enum class Reachability { Reachable, Unstable, Unreachable }
private var linkUp = true
private var reachability = Reachability.Reachable
private var failureStreakStartMs: Long? = null
private val recentOpFailures = ArrayDeque<Long>()
fun onLinkChange(up: Boolean) {
linkUp = up
// Link transitions don't reset reachability — a restored link keeps the
// last-known server reachability until a fresh probe/op result arrives.
if (!up) recentOpFailures.clear()
}
/** A real successful server op (stream read, API 2xx) or a successful /healthz. */
fun onSuccess() {
reachability = Reachability.Reachable
failureStreakStartMs = null
recentOpFailures.clear()
}
/** A real network op failed. Ambiguous on its own — records corroboration. */
fun onOpFailure(nowMs: Long) {
pruneOpFailures(nowMs)
recentOpFailures.addLast(nowMs)
}
/** A /healthz probe failed — the arbiter. Escalates per corroboration/backstop. */
fun onProbeFailure(nowMs: Long) {
pruneOpFailures(nowMs)
if (reachability == Reachability.Reachable) {
reachability = Reachability.Unstable
failureStreakStartMs = nowMs
}
if (reachability == Reachability.Unstable && shouldEscalate(nowMs)) {
reachability = Reachability.Unreachable
}
}
fun health(): ServerHealth = when {
!linkUp -> ServerHealth.Offline
reachability == Reachability.Reachable -> ServerHealth.Healthy
reachability == Reachability.Unstable -> ServerHealth.Unstable
else -> ServerHealth.ServerDown
}
private fun shouldEscalate(nowMs: Long): Boolean {
val corroborated = recentOpFailures.size >= CORROBORATION_OP_THRESHOLD
val sustained =
failureStreakStartMs?.let { nowMs - it >= ESCALATE_AFTER_MS } ?: false
return corroborated || sustained
}
private fun pruneOpFailures(nowMs: Long) {
while (recentOpFailures.isNotEmpty() &&
nowMs - recentOpFailures.first() > CORROBORATION_WINDOW_MS
) {
recentOpFailures.removeFirst()
}
}
}
@@ -0,0 +1,49 @@
package com.fabledsword.minstrel.connectivity
import com.fabledsword.minstrel.api.BaseUrlInterceptor.Companion.PLACEHOLDER_HOST
import dagger.Lazy
import okhttp3.Interceptor
import okhttp3.Response
import java.io.IOException
import javax.inject.Inject
import javax.inject.Singleton
private const val HEALTHZ_PATH = "/healthz"
/**
* Feeds real Minstrel API outcomes into [NetworkStatusController]. A 2xx is
* self-proving proof the server is reachable → reportSuccess(); a transport
* [IOException] (no response at all) → reportFailure(), which triggers /healthz
* arbitration.
*
* MUST run first in the OkHttp chain (before [BaseUrlInterceptor]) so the host
* is still the [PLACEHOLDER_HOST] sentinel: this shared client also fetches
* EXTERNAL artwork (musicbrainz / coverartarchive), and an external image
* loading must NOT be read as "our server is reachable" — only sentinel-host
* requests are Minstrel-bound. 5xx is deliberately NOT a failure (the server
* answered), and /healthz is skipped to avoid a feedback loop with the poll.
*
* [NetworkStatusController] is injected as a [Lazy] to break the Hilt cycle:
* the controller needs `Retrofit`, which needs `OkHttpClient`, which needs this
* interceptor. By the time a request flows through, the controller singleton is
* already constructed (construct-the-singleton trick in MinstrelApplication).
*/
@Singleton
class ReachabilityReportingInterceptor @Inject constructor(
private val networkStatus: Lazy<NetworkStatusController>,
) : Interceptor {
override fun intercept(chain: Interceptor.Chain): Response {
val request = chain.request()
val isMinstrel = request.url.host == PLACEHOLDER_HOST
val isHealthz = request.url.encodedPath.endsWith(HEALTHZ_PATH)
if (!isMinstrel || isHealthz) return chain.proceed(request)
return try {
val response = chain.proceed(request)
if (response.isSuccessful) networkStatus.get().reportSuccess()
response
} catch (e: IOException) {
networkStatus.get().reportFailure()
throw e
}
}
}
@@ -0,0 +1,27 @@
package com.fabledsword.minstrel.connectivity
import kotlinx.coroutines.flow.Flow
import kotlinx.coroutines.flow.distinctUntilChanged
import kotlinx.coroutines.flow.drop
import kotlinx.coroutines.flow.filter
import kotlinx.coroutines.flow.map
/**
* Emits once each time server health RETURNS to [ServerHealth.Healthy]
* after the collector subscribed — the StateFlow's replayed current value
* is dropped, so only genuine down→up transitions fire (a screen that
* subscribes while already Healthy doesn't double-load).
*
* This is the screen-level half of the recovery idiom the app-lifetime
* singletons (SyncController / MutationReplayer / DiagnosticsUploader)
* already use: a ViewModel collects this in its viewModelScope and re-runs
* its load, so a surface that failed while the server was unreachable
* heals itself the moment connectivity returns instead of sitting in the
* failed state until a manual pull-to-refresh.
*/
fun NetworkStatusController.recoveries(): Flow<Unit> = state
.map { it == ServerHealth.Healthy }
.distinctUntilChanged()
.drop(1)
.filter { it }
.map { }
@@ -0,0 +1,15 @@
package com.fabledsword.minstrel.connectivity
/**
* The single reachability signal every consumer branches on.
*
* - [Healthy] link up, /healthz ok — normal network behavior.
* - [Unstable] link up, a recent failure with arbitration pending —
* INFORMATIONAL ONLY. Does NOT gate playback; preserves the
* anti-flicker intent of commit 5c0db429 while still warning
* the user that something is flaky.
* - [ServerDown] link up but /healthz failing, corroborated or sustained —
* gate to cache-only.
* - [Offline] no device link at all — gate to cache-only.
*/
enum class ServerHealth { Healthy, Unstable, ServerDown, Offline }
@@ -14,77 +14,121 @@ import androidx.compose.material3.Icon
import androidx.compose.material3.MaterialTheme
import androidx.compose.material3.Text
import androidx.compose.runtime.Composable
import androidx.compose.runtime.LaunchedEffect
import androidx.compose.runtime.getValue
import androidx.compose.runtime.mutableStateOf
import androidx.compose.runtime.remember
import androidx.compose.runtime.setValue
import androidx.compose.ui.Alignment
import androidx.compose.ui.Modifier
import androidx.compose.ui.graphics.Color
import androidx.compose.ui.unit.dp
import androidx.hilt.navigation.compose.hiltViewModel
import androidx.lifecycle.SavedStateHandle
import androidx.lifecycle.ViewModel
import androidx.lifecycle.compose.collectAsStateWithLifecycle
import androidx.lifecycle.viewModelScope
import com.composables.icons.lucide.CloudOff
import com.composables.icons.lucide.Lucide
import com.fabledsword.minstrel.connectivity.ConnectivityObserver
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.ServerHealth
import dagger.hilt.android.lifecycle.HiltViewModel
import kotlinx.coroutines.flow.SharingStarted
import kotlinx.coroutines.delay
import kotlinx.coroutines.flow.StateFlow
import kotlinx.coroutines.flow.stateIn
import javax.inject.Inject
private const val ONLINE_SHARE_STOP_TIMEOUT_MS = 5_000L
private const val BACK_ONLINE_FLASH_MS = 2_000L
/**
* Tiny VM that just lifts the [ConnectivityObserver] singleton's
* Flow into a StateFlow with the standard sharing strategy. Keeps
* the banner composable pure-presentation.
* Exposes [NetworkStatusController]'s tri-state directly to the banner. No
* re-wrapping StateFlow — the controller's is already app-scoped and warm.
*/
@HiltViewModel
class ConnectivityBannerViewModel @Inject constructor(
observer: ConnectivityObserver,
networkStatus: NetworkStatusController,
@Suppress("UnusedPrivateProperty") savedStateHandle: SavedStateHandle,
) : ViewModel() {
val online: StateFlow<Boolean> = observer.online.stateIn(
scope = viewModelScope,
started = SharingStarted.WhileSubscribed(ONLINE_SHARE_STOP_TIMEOUT_MS),
initialValue = true,
)
val health: StateFlow<ServerHealth> = networkStatus.state
}
/**
* Banner shown at the top of the shell when the device has no usable
* internet. Mirrors Flutter's ConnectionErrorBanner: red-tinted error
* surface, CloudOff icon, "No connection — check Wi-Fi or mobile
* data" copy. Auto-hides via slide+fade when connectivity returns.
* Shell banner. Tells the user *why* they're degraded — no link vs server-down
* — plus a non-alarming "Reconnecting…" for the transient [ServerHealth.Unstable]
* window, and a brief "Back online" confirmation when health recovers so
* recovery is unmistakable. Sentence case, understated voice (design system).
*/
@Composable
fun ConnectionErrorBanner(
viewModel: ConnectivityBannerViewModel = hiltViewModel(),
) {
val online by viewModel.online.collectAsStateWithLifecycle()
val health by viewModel.health.collectAsStateWithLifecycle()
var showBackOnline by remember { mutableStateOf(false) }
var wasDown by remember { mutableStateOf(false) }
LaunchedEffect(health) {
val down = health == ServerHealth.Offline || health == ServerHealth.ServerDown
if (health == ServerHealth.Healthy && wasDown) {
// try/finally so a mid-delay cancellation (health flips again) can't
// orphan the flag and leave "Back online" stuck on screen.
try {
showBackOnline = true
delay(BACK_ONLINE_FLASH_MS)
} finally {
showBackOnline = false
}
}
wasDown = down
}
AnimatedVisibility(
visible = !online,
visible = health != ServerHealth.Healthy || showBackOnline,
enter = expandVertically() + fadeIn(),
exit = shrinkVertically() + fadeOut(),
) {
BannerContent(health = health, backOnline = showBackOnline)
}
}
@Composable
private fun BannerContent(health: ServerHealth, backOnline: Boolean) {
val scheme = MaterialTheme.colorScheme
val background: Color
val foreground: Color
when {
backOnline -> {
background = scheme.secondaryContainer
foreground = scheme.onSecondaryContainer
}
health == ServerHealth.Unstable -> {
background = scheme.surfaceVariant
foreground = scheme.onSurfaceVariant
}
else -> {
background = scheme.errorContainer
foreground = scheme.onErrorContainer
}
}
Row(
modifier = Modifier
.fillMaxWidth()
.background(MaterialTheme.colorScheme.errorContainer)
.background(background)
.padding(horizontal = 16.dp, vertical = 10.dp),
verticalAlignment = Alignment.CenterVertically,
horizontalArrangement = Arrangement.spacedBy(12.dp),
) {
Icon(
imageVector = Lucide.CloudOff,
contentDescription = null,
tint = MaterialTheme.colorScheme.onErrorContainer,
)
Icon(imageVector = Lucide.CloudOff, contentDescription = null, tint = foreground)
Text(
text = "No connection — check Wi-Fi or mobile data.",
text = bannerText(health, backOnline),
style = MaterialTheme.typography.bodyMedium,
color = MaterialTheme.colorScheme.onErrorContainer,
color = foreground,
)
}
}
}
private fun bannerText(health: ServerHealth, backOnline: Boolean): String = when {
backOnline -> "Back online."
health == ServerHealth.Offline -> "No connection — check Wi-Fi or mobile data."
health == ServerHealth.ServerDown ->
"Server unreachable — your cached content is still available."
health == ServerHealth.Unstable -> "Reconnecting…"
else -> ""
}
@@ -0,0 +1,328 @@
package com.fabledsword.minstrel.diagnostics
import android.content.BroadcastReceiver
import android.content.Context
import android.content.Intent
import android.content.IntentFilter
import android.os.PowerManager
import androidx.lifecycle.DefaultLifecycleObserver
import androidx.lifecycle.LifecycleOwner
import androidx.core.content.ContextCompat
import androidx.lifecycle.ProcessLifecycleOwner
import com.fabledsword.minstrel.auth.AuthController
import com.fabledsword.minstrel.auth.AuthStore
import com.fabledsword.minstrel.cache.db.dao.DiagnosticEventDao
import com.fabledsword.minstrel.cache.db.entities.DiagnosticEventEntity
import com.fabledsword.minstrel.connectivity.ConnectivityObserver
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.di.ApplicationScope
import com.fabledsword.minstrel.player.PlayerController
import com.fabledsword.minstrel.player.RemotePlayerState
import com.fabledsword.minstrel.player.output.OutputPickerController
import com.fabledsword.minstrel.player.output.OutputRoute
import dagger.hilt.android.qualifiers.ApplicationContext
import kotlinx.coroutines.CoroutineScope
import kotlinx.coroutines.Dispatchers
import kotlinx.coroutines.Job
import kotlinx.coroutines.delay
import kotlinx.coroutines.flow.combine
import kotlinx.coroutines.flow.distinctUntilChanged
import kotlinx.coroutines.flow.map
import kotlinx.coroutines.launch
import kotlinx.serialization.json.Json
import kotlinx.serialization.json.JsonObject
import kotlinx.serialization.json.buildJsonObject
import kotlinx.serialization.json.put
import javax.inject.Inject
import javax.inject.Singleton
/**
* On-device diagnostics collector (M9). Active only while the account's
* debug mode is enabled (server flag, read via AuthController.currentUser)
* AND the user hasn't opted out locally. When active it taps existing
* connectivity / player / UPnP / power seams READ-ONLY and writes a
* timeseries of events into the [DiagnosticEventDao] ring buffer; the
* [DiagnosticsUploader] drains that buffer to the server.
*
* The aim is to explain two failure classes after the fact: roaming /
* poor-data recovery (connectivity + server-health transitions) and
* locked-phone UPnP desync (power/Doze transitions correlated with the
* Sonos-vs-local position deltas captured in the heartbeat).
*/
@Suppress("TooManyFunctions") // one collector + one (un)register pair per signal; cohesive
@Singleton
class DiagnosticsReporter @Inject constructor(
@ApplicationContext private val context: Context,
@ApplicationScope private val scope: CoroutineScope,
private val dao: DiagnosticEventDao,
private val authController: AuthController,
private val authStore: AuthStore,
private val connectivity: ConnectivityObserver,
private val networkStatus: NetworkStatusController,
private val playerController: PlayerController,
private val outputPicker: OutputPickerController,
private val remoteState: RemotePlayerState,
private val json: Json,
) {
@Volatile private var enabled = false
private var collectorJob: Job? = null
private var powerReceiver: BroadcastReceiver? = null
private var lifecycleObserver: DefaultLifecycleObserver? = null
init {
// Gate = account flag (currentUser.debugModeEnabled) AND not the
// per-device local opt-out.
scope.launch {
combine(authController.currentUser, authStore.diagnosticsOptOut) { user, optOut ->
(user?.debugModeEnabled == true) && !optOut
}.distinctUntilChanged().collect { setEnabled(it) }
}
// Periodically re-pull /me so a remote admin flip propagates
// without a re-login or app restart.
scope.launch {
while (true) {
delay(GATE_REFRESH_MS)
authController.refreshProfile()
}
}
}
private fun setEnabled(value: Boolean) {
if (value == enabled) return
enabled = value
if (value) start() else stop()
}
private fun start() {
registerPowerReceiver()
registerLifecycleObserver()
collectorJob = scope.launch {
record("lifecycle", buildJsonObject { put("event", "reporting_started") })
launch { collectConnectivity() }
launch { collectServerHealth() }
launch { collectUpnpDrops() }
launch { collectPlayerState() }
launch { collectTrackChanges() }
launch { collectRoutes() }
launch { heartbeatLoop() }
}
}
private fun stop() {
collectorJob?.cancel()
collectorJob = null
unregisterPowerReceiver()
unregisterLifecycleObserver()
// Off means off: drop any unsent buffer (honors the user OFF
// intent and avoids leaking data captured before disable).
scope.launch { runCatching { dao.clear() } }
}
private suspend fun collectConnectivity() {
connectivity.online.distinctUntilChanged().collect { online ->
record("connectivity", buildJsonObject {
put("event", "online_changed")
put("online", online)
})
}
}
private suspend fun collectServerHealth() {
// state is a StateFlow — already conflated/distinct, so no
// distinctUntilChanged (it's a deprecation warning = build error).
networkStatus.state.collect { s ->
record("connectivity", buildJsonObject {
put("event", "server_health")
put("state", s.name)
})
}
}
private suspend fun collectUpnpDrops() {
playerController.dropEvents.collect { msg ->
record("upnp_sync", buildJsonObject {
put("event", "drop")
put("message", msg)
})
}
}
private suspend fun collectPlayerState() {
// 'playback', not 'upnp_sync' — player state applies to every output
// route (phone speaker, Bluetooth, UPnP), not just casting.
playerController.uiState
.map { Triple(it.currentSource, it.isUpnpLoading, it.playbackError) }
.distinctUntilChanged()
.collect { (source, upnpLoading, err) ->
record("playback", buildJsonObject {
put("event", "player_state")
put("source", source ?: "")
put("upnp_loading", upnpLoading)
if (err != null) put("error", err)
})
}
}
// Emit a snapshot at each track/queue-index change — captures a
// skip-induced local↔Sonos desync at the INSTANT it happens, which the
// 45s heartbeat misses. Note: uiState is a conflated StateFlow, so a
// very rapid skip burst may coalesce intermediate indices (we still get
// the boundaries + the local-vs-Sonos snapshot).
private suspend fun collectTrackChanges() {
playerController.uiState
.map { it.queueIndex to it.currentTrack?.id }
.distinctUntilChanged()
.collect { (index, _) ->
val ui = playerController.uiState.value
val casting = outputPicker.routesState.value.current.protocol !=
OutputRoute.Protocol.SYSTEM
record("playback", buildJsonObject {
put("event", "track_change")
put("local_index", index)
// Track IDENTITY, not just index — indices wobble across
// re-casts, so the id + Sonos URI make a desync unambiguous.
put("local_track_id", ui.currentTrack?.id ?: "")
put("local_pos_ms", ui.positionMs)
putSonos(this, casting)
put("upnp_loading", ui.isUpnpLoading)
put("server_health", networkStatus.state.value.name)
put("route", outputPicker.routesState.value.current.name)
})
}
}
private suspend fun collectRoutes() {
// 'playback' — route changes happen for all outputs. This only ever
// logs the ACTIVE route (routesState.current), so no "connected" flag.
outputPicker.routesState.map { it.current }.distinctUntilChanged().collect { r ->
record("playback", buildJsonObject {
put("event", "route")
put("id", r.id)
put("name", r.name)
put("kind", r.kind.name)
put("protocol", r.protocol.name)
})
}
}
private suspend fun heartbeatLoop() {
while (true) {
delay(HEARTBEAT_MS)
val ui = playerController.uiState.value
val route = outputPicker.routesState.value.current
val routeActive = route.protocol != OutputRoute.Protocol.SYSTEM
// Idle = nothing worth sampling. Skip to keep the buffer lean.
if (!ui.isPlaying && !routeActive) continue
record("heartbeat", buildJsonObject {
put("server_health", networkStatus.state.value.name)
put("is_playing", ui.isPlaying)
put("source", ui.currentSource ?: "")
put("local_index", ui.queueIndex)
put("local_track_id", ui.currentTrack?.id ?: "")
put("local_pos_ms", ui.positionMs)
put("route", route.name)
put("route_protocol", route.protocol.name)
putSonos(this, routeActive)
addPowerFields(this)
})
}
}
// --- power + lifecycle taps -------------------------------------------
private fun registerPowerReceiver() {
if (powerReceiver != null) return
val receiver = object : BroadcastReceiver() {
override fun onReceive(c: Context?, intent: Intent?) {
val action = intent?.action ?: return
if (!enabled) return
scope.launch { record("power", powerEvent(action)) }
}
}
val filter = IntentFilter().apply {
addAction(Intent.ACTION_SCREEN_ON)
addAction(Intent.ACTION_SCREEN_OFF)
addAction(PowerManager.ACTION_DEVICE_IDLE_MODE_CHANGED)
}
// System-protected broadcasts, but pass NOT_EXPORTED explicitly so
// the API 34+ registerReceiver flag requirement is satisfied.
ContextCompat.registerReceiver(
context, receiver, filter, ContextCompat.RECEIVER_NOT_EXPORTED,
)
powerReceiver = receiver
}
private fun unregisterPowerReceiver() {
powerReceiver?.let { runCatching { context.unregisterReceiver(it) } }
powerReceiver = null
}
private fun registerLifecycleObserver() {
if (lifecycleObserver != null) return
val obs = object : DefaultLifecycleObserver {
override fun onStart(owner: LifecycleOwner) = logLifecycle("app_foreground")
override fun onStop(owner: LifecycleOwner) = logLifecycle("app_background")
}
lifecycleObserver = obs
scope.launch(Dispatchers.Main) {
ProcessLifecycleOwner.get().lifecycle.addObserver(obs)
}
}
private fun unregisterLifecycleObserver() {
val obs = lifecycleObserver ?: return
lifecycleObserver = null
scope.launch(Dispatchers.Main) {
ProcessLifecycleOwner.get().lifecycle.removeObserver(obs)
}
}
private fun logLifecycle(event: String) {
if (!enabled) return
scope.launch { record("lifecycle", buildJsonObject { put("event", event) }) }
}
private fun powerEvent(action: String): JsonObject = buildJsonObject {
put("event", action.substringAfterLast('.'))
addPowerFields(this)
}
// Sonos/UPnP remote-vs-local desync fields. Only meaningful while a
// remote route is active; zeroed otherwise so a stale RemotePlayerState
// from a just-ended cast can't masquerade as live Sonos data (the
// cast→phone handoff artifact). currentTrackUri is the desync ground
// truth — it carries the track the speaker is actually streaming.
private fun putSonos(builder: kotlinx.serialization.json.JsonObjectBuilder, casting: Boolean) {
builder.put("sonos_track", if (casting) remoteState.trackNumber else 0)
builder.put("sonos_pos_ms", if (casting) remoteState.positionMs else 0)
builder.put("sonos_playing", casting && remoteState.isPlaying)
if (casting) builder.put("sonos_uri", remoteState.currentTrackUri)
}
private fun addPowerFields(builder: kotlinx.serialization.json.JsonObjectBuilder) {
val pm = context.getSystemService(Context.POWER_SERVICE) as PowerManager
builder.put("doze", pm.isDeviceIdleMode)
builder.put("screen_on", pm.isInteractive)
builder.put("battery_opt_ignored", pm.isIgnoringBatteryOptimizations(context.packageName))
}
private suspend fun record(kind: String, payload: JsonObject) {
runCatching {
val id = dao.insert(
DiagnosticEventEntity(
kind = kind,
occurredAtMillis = System.currentTimeMillis(),
payloadJson = json.encodeToString(JsonObject.serializer(), payload),
),
)
if (id % TRIM_INTERVAL == 0L) dao.trimToNewest(MAX_BUFFER)
}
}
private companion object {
const val HEARTBEAT_MS = 45_000L
const val GATE_REFRESH_MS = 300_000L // re-pull /me every 5 min
const val MAX_BUFFER = 5_000 // ring-buffer ceiling, oldest dropped
const val TRIM_INTERVAL = 50L // trim every N inserts, not every one
}
}
@@ -0,0 +1,118 @@
package com.fabledsword.minstrel.diagnostics
import android.os.Build
import com.fabledsword.minstrel.BuildConfig
import com.fabledsword.minstrel.api.endpoints.DiagnosticEventWire
import com.fabledsword.minstrel.api.endpoints.DiagnosticsApi
import com.fabledsword.minstrel.api.endpoints.DiagnosticsReportRequest
import com.fabledsword.minstrel.auth.AuthStore
import com.fabledsword.minstrel.cache.db.dao.DiagnosticEventDao
import com.fabledsword.minstrel.cache.db.entities.DiagnosticEventEntity
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.ServerHealth
import com.fabledsword.minstrel.di.ApplicationScope
import kotlinx.coroutines.CoroutineScope
import kotlinx.coroutines.delay
import kotlinx.coroutines.flow.distinctUntilChanged
import kotlinx.coroutines.flow.filter
import kotlinx.coroutines.flow.filterNotNull
import kotlinx.coroutines.flow.map
import kotlinx.coroutines.launch
import kotlinx.coroutines.sync.Mutex
import kotlinx.serialization.json.Json
import retrofit2.Retrofit
import retrofit2.create
import javax.inject.Inject
import javax.inject.Singleton
/**
* Drains the [DiagnosticEventDao] ring buffer to POST /api/diagnostics
* (M9). Triggers mirror [com.fabledsword.minstrel.cache.mutations.MutationReplayer]:
* a periodic tick, a sign-in (sessionCookie non-null), and a server-health
* recovery to [ServerHealth.Healthy] — so a dead-zone's buffered events
* upload the moment the link comes back.
*
* Deliberately NOT the MutationQueue: diagnostics are high-volume,
* best-effort telemetry. A failed upload just retries on the next tick;
* the server no-ops (204) when the account's debug flag is off, which the
* uploader treats as success so the buffer drains either way.
*/
@Singleton
class DiagnosticsUploader @Inject constructor(
private val dao: DiagnosticEventDao,
private val authStore: AuthStore,
private val json: Json,
@ApplicationScope private val scope: CoroutineScope,
networkStatus: NetworkStatusController,
retrofit: Retrofit,
) {
private val api: DiagnosticsApi = retrofit.create()
private val mutex = Mutex()
init {
scope.launch {
while (true) {
delay(UPLOAD_INTERVAL_MS)
drainSafe()
}
}
scope.launch {
authStore.sessionCookie.filterNotNull().distinctUntilChanged().collect { drainSafe() }
}
scope.launch {
networkStatus.state
.map { it == ServerHealth.Healthy }
.distinctUntilChanged()
.filter { it }
.collect { drainSafe() }
}
}
/** Coalesces concurrent triggers via tryLock — a drain in flight wins. */
private suspend fun drainSafe() {
val clientId = authStore.clientId.value
// Signed out / no client id yet → nothing to do. Single guard keeps
// the return count within the detekt gate.
if (authStore.sessionCookie.value.isNullOrEmpty() || clientId == null) return
if (!mutex.tryLock()) return
try {
drain(clientId)
} finally {
mutex.unlock()
}
}
private suspend fun drain(clientId: String) {
var more = true
while (more) {
val batch = runCatching { dao.takeBatch(BATCH_SIZE) }.getOrNull().orEmpty()
val sent = batch.isNotEmpty() && runCatching { upload(clientId, batch) }.isSuccess
if (sent) runCatching { dao.deleteByIds(batch.map { it.id }) }
// Keep going only while a full batch sent cleanly; otherwise stop
// (empty buffer, or a failure to retry on the next trigger).
more = sent && batch.size >= BATCH_SIZE
}
}
private suspend fun upload(clientId: String, batch: List<DiagnosticEventEntity>) {
api.report(
DiagnosticsReportRequest(
clientId = clientId,
appVersion = BuildConfig.VERSION_NAME,
osVersion = "Android ${Build.VERSION.RELEASE} (${Build.MODEL})",
events = batch.map { row ->
DiagnosticEventWire(
kind = row.kind,
occurredAt = row.occurredAtMillis,
payload = json.parseToJsonElement(row.payloadJson),
)
},
),
)
}
private companion object {
const val UPLOAD_INTERVAL_MS = 60_000L
const val BATCH_SIZE = 100
}
}
@@ -41,6 +41,7 @@ import com.fabledsword.minstrel.models.ArtistSuggestionRef
import com.fabledsword.minstrel.models.LidarrRequestKind
import com.fabledsword.minstrel.models.LidarrSearchResultRef
import com.fabledsword.minstrel.nav.Discover
import com.fabledsword.minstrel.shared.widgets.ErrorRetry
import com.fabledsword.minstrel.shared.widgets.LoadingCentered
import com.fabledsword.minstrel.shared.widgets.MinstrelTopAppBar
import com.fabledsword.minstrel.shared.widgets.PullToRefreshScaffold
@@ -109,6 +110,7 @@ private fun DiscoverBody(
snackbar.showSnackbar(snackbarFor(outcome, s.name))
}
},
onRetry = { viewModel.loadSuggestions() },
)
ResultsState.Loading -> LoadingCentered()
is ResultsState.Loaded -> ResultsList(
@@ -121,7 +123,11 @@ private fun DiscoverBody(
}
},
)
is ResultsState.Error -> CenteredMessage("Search failed: ${r.message}")
is ResultsState.Error -> ErrorRetry(
title = "Search failed",
message = r.message,
onRetry = viewModel::runSearch,
)
}
}
}
@@ -177,10 +183,15 @@ private fun SuggestionsPane(
state: SuggestionState,
locallyRequestedMbids: Set<String>,
onRequest: (ArtistSuggestionRef) -> Unit,
onRetry: () -> Unit,
) {
when (state) {
SuggestionState.Loading -> LoadingCentered()
is SuggestionState.Error -> CenteredMessage("Couldn't load suggestions.")
is SuggestionState.Error -> ErrorRetry(
title = "Couldn't load suggestions",
message = state.message,
onRetry = onRetry,
)
is SuggestionState.Loaded -> SuggestionsList(
items = state.items.filter { it.mbid !in locallyRequestedMbids },
onRequest = onRequest,
@@ -3,6 +3,8 @@ package com.fabledsword.minstrel.discover.ui
import androidx.lifecycle.ViewModel
import androidx.lifecycle.viewModelScope
import com.fabledsword.minstrel.api.ErrorCopy
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.recoveries
import com.fabledsword.minstrel.discover.data.DiscoverRepository
import com.fabledsword.minstrel.discover.data.RequestOutcome
import com.fabledsword.minstrel.models.ArtistSuggestionRef
@@ -46,6 +48,7 @@ sealed interface ResultsState {
@HiltViewModel
class DiscoverViewModel @Inject constructor(
private val repository: DiscoverRepository,
networkStatus: NetworkStatusController,
) : ViewModel() {
private val internal = MutableStateFlow(DiscoverState())
@@ -53,6 +56,15 @@ class DiscoverViewModel @Inject constructor(
init {
loadSuggestions()
// Screen-level auto-recovery (issue #1245): re-run whichever pane
// failed while the server was unreachable once health returns.
viewModelScope.launch {
networkStatus.recoveries().collect {
val s = internal.value
if (s.suggestions is SuggestionState.Error) loadSuggestions()
if (s.results is ResultsState.Error) runSearch()
}
}
}
fun setQuery(value: String) {
@@ -19,6 +19,8 @@ import androidx.lifecycle.ViewModel
import androidx.lifecycle.compose.collectAsStateWithLifecycle
import androidx.lifecycle.viewModelScope
import com.fabledsword.minstrel.api.ErrorCopy
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.recoveries
import com.fabledsword.minstrel.history.data.HistoryEntry
import com.fabledsword.minstrel.history.data.HistoryRepository
import com.fabledsword.minstrel.models.TrackRef
@@ -26,6 +28,7 @@ import com.fabledsword.minstrel.player.PlayerController
import com.fabledsword.minstrel.shared.UiState
import com.fabledsword.minstrel.shared.widgets.TrackRow
import com.fabledsword.minstrel.shared.widgets.EmptyState
import com.fabledsword.minstrel.shared.widgets.ErrorRetry
import com.fabledsword.minstrel.shared.widgets.LoadingCentered
import com.fabledsword.minstrel.shared.widgets.PullToRefreshScaffold
import com.fabledsword.minstrel.shared.widgets.TrackCoverThumb
@@ -56,6 +59,7 @@ private const val SHARE_STOP_TIMEOUT_MS = 5_000L
class HistoryTabViewModel @Inject constructor(
private val repository: HistoryRepository,
private val player: PlayerController,
networkStatus: NetworkStatusController,
) : ViewModel() {
private val refreshError = MutableStateFlow<String?>(null)
@@ -82,6 +86,11 @@ class HistoryTabViewModel @Inject constructor(
init {
refresh()
// Screen-level auto-recovery (issue #1245): re-pull when server
// health returns, so a load that failed offline heals unprompted.
viewModelScope.launch {
networkStatus.recoveries().collect { refresh() }
}
}
fun refresh(): Job = viewModelScope.launch {
@@ -116,9 +125,10 @@ fun HistoryTab(
title = "No listening history yet",
body = "Play something — your recent plays will show up here.",
)
is UiState.Error -> EmptyState(
is UiState.Error -> ErrorRetry(
title = "Couldn't load history",
body = s.message,
message = s.message,
onRetry = { viewModel.refresh() },
)
is UiState.Success -> HistoryList(
entries = s.data,
@@ -39,6 +39,9 @@ import javax.inject.Singleton
* pre-warms the top artists via [HomeArtistPrewarmer].
*/
@Singleton
// Per-section observe accessors (one per Home row) inflate the function
// count past detekt's 11 default — accessor density, not complexity.
@Suppress("TooManyFunctions")
class HomeRepository @Inject constructor(
private val homeIndexDao: CachedHomeIndexDao,
private val metadataProvider: MetadataProvider,
@@ -76,6 +79,12 @@ class HomeRepository @Inject constructor(
fun observeLastPlayedArtists(): Flow<List<HomeTile<ArtistRef>>> =
observeArtistSection(SECTION_LAST_PLAYED_ARTISTS)
fun observeYouMightLikeAlbums(): Flow<List<HomeTile<AlbumRef>>> =
observeAlbumSection(SECTION_YOU_MIGHT_LIKE_ALBUMS)
fun observeYouMightLikeArtists(): Flow<List<HomeTile<ArtistRef>>> =
observeArtistSection(SECTION_YOU_MIGHT_LIKE_ARTISTS)
/**
* Pulls /api/home/index, replaces each cached_home_index section,
* and pre-warms the top artists. The section Flows re-fire on the
@@ -88,7 +97,11 @@ class HomeRepository @Inject constructor(
replaceSection(SECTION_REDISCOVER_ARTISTS, "artist", wire.rediscoverArtists)
replaceSection(SECTION_MOST_PLAYED_TRACKS, "track", wire.mostPlayedTracks)
replaceSection(SECTION_LAST_PLAYED_ARTISTS, "artist", wire.lastPlayedArtists)
prewarmer.warm(wire.rediscoverArtists + wire.lastPlayedArtists)
replaceSection(SECTION_YOU_MIGHT_LIKE_ALBUMS, "album", wire.youMightLikeAlbums)
replaceSection(SECTION_YOU_MIGHT_LIKE_ARTISTS, "artist", wire.youMightLikeArtists)
prewarmer.warm(
wire.rediscoverArtists + wire.lastPlayedArtists + wire.youMightLikeArtists,
)
}
private suspend fun replaceSection(section: String, entityType: String, ids: List<String>) {
@@ -148,5 +161,7 @@ class HomeRepository @Inject constructor(
const val SECTION_REDISCOVER_ARTISTS = "rediscover_artists"
const val SECTION_MOST_PLAYED_TRACKS = "most_played_tracks"
const val SECTION_LAST_PLAYED_ARTISTS = "last_played_artists"
const val SECTION_YOU_MIGHT_LIKE_ALBUMS = "you_might_like_albums"
const val SECTION_YOU_MIGHT_LIKE_ARTISTS = "you_might_like_artists"
}
}
@@ -54,6 +54,8 @@ import com.composables.icons.lucide.History
import com.composables.icons.lucide.Lucide
import com.composables.icons.lucide.Music
import com.fabledsword.minstrel.api.ErrorCopy
import com.fabledsword.minstrel.connectivity.ServerHealth
import com.fabledsword.minstrel.connectivity.recoveries
import com.fabledsword.minstrel.home.data.HomeRepository
import com.fabledsword.minstrel.library.data.LibraryRepository
import com.fabledsword.minstrel.library.widgets.AlbumCard
@@ -69,13 +71,14 @@ import com.fabledsword.minstrel.nav.ArtistDetail
import com.fabledsword.minstrel.nav.Home
import com.fabledsword.minstrel.nav.PlaylistDetail
import com.fabledsword.minstrel.playlists.data.PlaylistsRepository
import com.fabledsword.minstrel.playlists.data.toPlayableTrackRefs
import com.fabledsword.minstrel.playlists.data.playPlaylistShuffled
import com.fabledsword.minstrel.playlists.widgets.OfflinePoolCard
import com.fabledsword.minstrel.playlists.widgets.PlaylistCard
import com.fabledsword.minstrel.playlists.widgets.PlaylistPlaceholderCard
import com.fabledsword.minstrel.shared.UiState
import com.fabledsword.minstrel.shared.asCacheFirstStateFlow
import com.fabledsword.minstrel.shared.widgets.EmptyState
import com.fabledsword.minstrel.shared.widgets.ErrorRetry
import com.fabledsword.minstrel.shared.widgets.HorizontalScrollRow
import com.fabledsword.minstrel.shared.widgets.MinstrelTopAppBar
import com.fabledsword.minstrel.shared.widgets.PullToRefreshScaffold
@@ -94,12 +97,11 @@ import kotlinx.coroutines.flow.combine
import kotlinx.coroutines.flow.map
import kotlinx.coroutines.flow.receiveAsFlow
import kotlinx.coroutines.flow.stateIn
import kotlinx.coroutines.flow.filter
import kotlinx.coroutines.launch
import kotlinx.coroutines.withTimeout
import javax.inject.Inject
private const val SHARE_STOP_TIMEOUT_MS = 5_000L
private const val PLAYLIST_FETCH_TIMEOUT_MS = 8_000L
private const val BOTTOM_PADDING_FOR_MINIPLAYER_DP = 140
// Recently Added is laid out in a multi-row LazyHorizontalGrid that
// scrolls as one panel (same pattern as Most Played). Two rows trades
@@ -117,6 +119,8 @@ data class HomeSections(
val rediscoverArtists: List<HomeTile<ArtistRef>> = emptyList(),
val mostPlayedTracks: List<HomeTile<TrackRef>> = emptyList(),
val lastPlayedArtists: List<HomeTile<ArtistRef>> = emptyList(),
val youMightLikeAlbums: List<HomeTile<AlbumRef>> = emptyList(),
val youMightLikeArtists: List<HomeTile<ArtistRef>> = emptyList(),
) {
val isAllEmpty: Boolean
get() = playlists.isEmpty() &&
@@ -124,7 +128,9 @@ data class HomeSections(
rediscoverAlbums.isEmpty() &&
rediscoverArtists.isEmpty() &&
mostPlayedTracks.isEmpty() &&
lastPlayedArtists.isEmpty()
lastPlayedArtists.isEmpty() &&
youMightLikeAlbums.isEmpty() &&
youMightLikeArtists.isEmpty()
}
// ─── ViewModel ───────────────────────────────────────────────────────
@@ -136,7 +142,8 @@ class HomeViewModel @Inject constructor(
private val libraryRepository: LibraryRepository,
private val player: com.fabledsword.minstrel.player.PlayerController,
private val shuffleSource: com.fabledsword.minstrel.cache.ShuffleSource,
connectivity: com.fabledsword.minstrel.connectivity.ConnectivityObserver,
private val eventsStream: com.fabledsword.minstrel.events.EventsStream,
networkStatus: com.fabledsword.minstrel.connectivity.NetworkStatusController,
) : ViewModel() {
private val systemStatusInternal = MutableStateFlow(SystemPlaylistsStatus())
@@ -144,9 +151,13 @@ class HomeViewModel @Inject constructor(
/** System-playlist build status for the Home placeholder cards. */
val systemStatus: StateFlow<SystemPlaylistsStatus> = systemStatusInternal.asStateFlow()
/** True when the device has no usable internet — gates the offline-pool cards. */
val offline: StateFlow<Boolean> = connectivity.online
.map { !it }
/**
* Cache-only when there's no link OR the server is unreachable. Reads the
* unified [NetworkStatusController] (not the raw device link) so Home reacts
* to ServerDown too; the transient Unstable state stays calm (not offline).
*/
val offline: StateFlow<Boolean> = networkStatus.state
.map { it == ServerHealth.Offline || it == ServerHealth.ServerDown }
.stateIn(
scope = viewModelScope,
started = SharingStarted.WhileSubscribed(SHARE_STOP_TIMEOUT_MS),
@@ -158,8 +169,31 @@ class HomeViewModel @Inject constructor(
/** Transient snackbar messages from offline-pool taps. */
val transientMessages: Flow<String> = poolMessages.receiveAsFlow()
/**
* Copy for the most recent /home/index refresh failure; null once a
* refresh succeeds. Folded into [uiState] so an empty cache + failed
* refresh renders as Error (with retry) instead of masquerading as
* the "Welcome to Minstrel" empty state.
*/
private val refreshError = MutableStateFlow<String?>(null)
init {
refresh()
// Screen-level auto-recovery (issue #1245): a Home that failed to
// load while the server was unreachable re-pulls itself the moment
// health returns — same idiom as SyncController, one layer up.
viewModelScope.launch {
networkStatus.recoveries().collect { refresh() }
}
// #968: the daily 03:00 rebuild (and manual refresh) emit
// playlist.system_rebuilt; re-pull Home so the system-playlist tiles
// and You-might-like rows reflect the new snapshot without a manual
// reload. Mirrors the web SSE consumer.
viewModelScope.launch {
eventsStream.events
.filter { it.kind == "playlist.system_rebuilt" }
.collect { refresh() }
}
}
/**
@@ -259,45 +293,9 @@ class HomeViewModel @Inject constructor(
*/
suspend fun playPlaylist(playlist: PlaylistRef) {
viewModelScope.launch {
val detail = try {
withTimeout(PLAYLIST_FETCH_TIMEOUT_MS) {
if (playlist.refreshable && playlist.systemVariant != null) {
playlistsRepository.systemShuffle(playlist.systemVariant)
} else {
playlistsRepository.refreshDetail(playlist.id)
playPlaylistShuffled(playlist, playlistsRepository, player) {
poolMessages.trySend(it)
}
}
} catch (
@Suppress("SwallowedException") _: kotlinx.coroutines.TimeoutCancellationException,
) {
poolMessages.trySend("Couldn't load playlist - check your connection")
return@launch
} catch (
@Suppress("TooGenericExceptionCaught") e: Throwable,
) {
poolMessages.trySend(
"Couldn't load playlist: ${ErrorCopy.fromThrowable(e)}",
)
return@launch
}
// Shared with PlaylistDetailViewModel.play - filters out
// unplayable rows (missing trackId or empty streamUrl) so the
// queue can't end up with tracks Media3 silently rejects.
val tracks = detail.tracks.toPlayableTrackRefs()
if (tracks.isEmpty()) {
poolMessages.trySend("Mix isn't ready yet - try again in a moment")
return@launch
}
// Drift #564: send the BARE systemVariant string, not
// "playlist:<variant>" — the server's rotation matcher
// (internal/playevents/writer.go systemPlaylistSources)
// keys on the bare variant. Web sends the bare form too
// (web/src/lib/components/PlaylistCard.svelte:83), so this
// brings Android into alignment. Wrong prefix here meant
// system-mix plays from Android Home never advanced the
// rotation.
val source = if (playlist.refreshable) playlist.systemVariant else null
player.setQueue(tracks, initialIndex = 0, source = source)
}.join()
}
@@ -307,7 +305,14 @@ class HomeViewModel @Inject constructor(
* actual completion before hiding the indicator.
*/
fun refresh(): Job = viewModelScope.launch {
val home = launch { runCatching { homeRepository.refreshIndex() } }
refreshError.value = null
val home = launch {
// /home/index is the load-bearing pull: its failure drives the
// empty-cache Error state. A failure over a populated cache
// stays silent — cached sections beat a full-screen error.
runCatching { homeRepository.refreshIndex() }
.onFailure { refreshError.value = ErrorCopy.fromThrowable(it) }
}
val lists = launch { runCatching { playlistsRepository.refreshList() } }
val status = launch {
runCatching { homeRepository.getSystemPlaylistsStatus() }
@@ -322,13 +327,14 @@ class HomeViewModel @Inject constructor(
combineHomeFlows().asCacheFirstStateFlow(viewModelScope)
/**
* Five home-section flows merged into the sections struct without
* playlists; then merged with the playlists flow in [combineHomeFlows].
* Splitting it like this keeps each combine call inside the
* coroutines built-in arity limit (5) and avoids the typed-vararg
* acrobatics needed for a 6-flow combine across heterogeneous types.
* The core five section flows merged into the sections struct. Split
* out so each combine call stays inside the coroutines built-in arity
* limit (5) and avoids the typed-vararg acrobatics a wider combine
* across heterogeneous types would need. [observeHomeSections] then
* folds in the two you-might-like flows, and [combineHomeFlows] the
* playlists flow.
*/
private fun observeHomeSections() = combine(
private fun observeCoreSections() = combine(
homeRepository.observeRecentlyAddedAlbums(),
homeRepository.observeRediscoverAlbums(),
homeRepository.observeRediscoverArtists(),
@@ -344,10 +350,26 @@ class HomeViewModel @Inject constructor(
)
}
private fun observeHomeSections() = combine(
observeCoreSections(),
homeRepository.observeYouMightLikeAlbums(),
homeRepository.observeYouMightLikeArtists(),
) { core, ymlAlbums, ymlArtists ->
core.copy(youMightLikeAlbums = ymlAlbums, youMightLikeArtists = ymlArtists)
}
private fun combineHomeFlows() =
observeHomeSections().combine(playlistsRepository.observeAll()) { sections, playlists ->
combine(
observeHomeSections(),
playlistsRepository.observeAll(),
refreshError,
) { sections, playlists, err ->
val merged = sections.copy(playlists = playlists)
if (merged.isAllEmpty) UiState.Empty else UiState.Success(merged)
when {
!merged.isAllEmpty -> UiState.Success(merged)
err != null -> UiState.Error(err)
else -> UiState.Empty
}
}
}
@@ -398,9 +420,10 @@ fun HomeScreen(
"settings, then come back here for system playlists " +
"and recommendations.",
)
is UiState.Error -> EmptyState(
is UiState.Error -> ErrorRetry(
title = "Couldn't load home",
body = s.message,
message = s.message,
onRetry = { viewModel.refresh() },
)
is UiState.Success -> HomeSuccessContent(
sections = s.data,
@@ -496,6 +519,14 @@ private fun HomeSuccessContent(
onPlayPlaylist = onPlayPlaylist,
)
}
youMightLikeSection(
albums = sections.youMightLikeAlbums,
artists = sections.youMightLikeArtists,
onAlbumClick = onAlbumClick,
onArtistClick = onArtistClick,
onPlayAlbum = onPlayAlbum,
onPlayArtist = onPlayArtist,
)
recentlyAddedSection(sections.recentlyAddedAlbums, onAlbumClick, onPlayAlbum)
rediscoverSection(
albums = sections.rediscoverAlbums,
@@ -700,6 +731,61 @@ private fun RediscoverBlock(
}
}
/**
* "You might like" — in-library albums/artists predicted from the user's
* taste that they don't actively spin (server section, daily-built, cold-
* start gated). Sits directly under the system-playlists row. Same
* albums-then-artists block shape as Rediscover; the artist sub-row drops
* its title when an album sub-row precedes it so the two read as one section.
*/
private fun LazyListScope.youMightLikeSection(
albums: List<HomeTile<AlbumRef>>,
artists: List<HomeTile<ArtistRef>>,
onAlbumClick: (String) -> Unit,
onArtistClick: (String) -> Unit,
onPlayAlbum: suspend (String) -> Unit,
onPlayArtist: suspend (String) -> Unit,
) {
item {
if (albums.isNotEmpty() || artists.isNotEmpty()) {
YouMightLikeBlock(
albums = albums,
artists = artists,
onAlbumClick = onAlbumClick,
onArtistClick = onArtistClick,
onPlayAlbum = onPlayAlbum,
onPlayArtist = onPlayArtist,
)
} else {
EmptySection(
title = "You might like",
body = "We're still learning your taste — keep listening and " +
"picks you might like will show up here.",
)
}
}
}
@Composable
private fun YouMightLikeBlock(
albums: List<HomeTile<AlbumRef>>,
artists: List<HomeTile<ArtistRef>>,
onAlbumClick: (String) -> Unit,
onArtistClick: (String) -> Unit,
onPlayAlbum: suspend (String) -> Unit,
onPlayArtist: suspend (String) -> Unit,
) {
Column {
if (albums.isNotEmpty()) {
AlbumsRow("You might like", albums, onAlbumClick, onPlayAlbum)
}
if (artists.isNotEmpty()) {
val title = if (albums.isEmpty()) "You might like" else ""
ArtistsRow(title, artists, onArtistClick, onPlayArtist)
}
}
}
@Composable
private fun AlbumsRow(
title: String,
@@ -739,18 +825,19 @@ private fun PlaylistsRow(
icon = iconForPool(item.kind),
onClick = { onPlayPool(item.kind) },
)
is PlaylistRowItem.Real -> PlaylistCard(
is PlaylistRowItem.Real -> {
// Greyed offline when the tile needs the live server or
// isn't fully cached — dimmed but still tappable into the
// detail to shuffle whatever subset is cached.
val greyed = offline && item.playlist.unavailableOffline
PlaylistCard(
playlist = item.playlist,
onClick = { onPlaylistClick(item.playlist.id) },
onPlay = { onPlayPlaylist(item.playlist) },
// Match Flutter: refreshable system playlists need
// the live server endpoints, so disable their play
// overlay when offline (user can still tap into the
// detail and shuffle all from cache). User playlists
// play from cache + survive offline.
playEnabled = item.playlist.trackCount > 0 &&
!(offline && item.playlist.refreshable),
playEnabled = item.playlist.trackCount > 0 && !greyed,
greyed = greyed,
)
}
is PlaylistRowItem.Placeholder -> PlaylistPlaceholderCard(
label = item.label,
variant = item.variant,
@@ -766,7 +853,7 @@ private fun iconForPool(kind: OfflinePoolKind) = when (kind) {
}
/** A cache-backed offline pool, a real playlist tile, or a not-yet-generated slot. */
private sealed interface PlaylistRowItem {
internal sealed interface PlaylistRowItem {
data class OfflinePool(val kind: OfflinePoolKind) : PlaylistRowItem
data class Real(val playlist: PlaylistRef) : PlaylistRowItem
data class Placeholder(val label: String, val variant: String) : PlaylistRowItem
@@ -779,55 +866,81 @@ enum class OfflinePoolKind(val label: String) {
}
/**
* Builds the Home Playlists row. When [offline], the two cache-backed
* pool cards (Recently played, Liked) lead the row. Then For You +
* Discover + 3× Songs-like fixed slots (real card when generated,
* placeholder otherwise), then the secondary system kinds (deep cuts /
* rediscover / new for you / on this day / first listens) when they
* exist — no placeholders for these since they're conditional on
* library shape, not guaranteed singletons. Finally user-owned
* playlists.
* Builds the Home Playlists row.
*
* Online: For You + Discover + 3× Songs-like fixed slots (real card when
* generated, placeholder otherwise), then the secondary system kinds (deep cuts
* / rediscover / new for you / on this day / first listens) when they exist —
* no placeholders for these since they're conditional on library shape — then
* user-owned playlists.
*
* Offline: the two cache-backed pools (Recently played, Liked) lead, then the
* same real playlists in curated order but stably partitioned fully-cached
* first / greyed after, and the "building/pending" placeholders are dropped
* (they need the server to generate, so they're meaningless offline).
*
* Diverges from Flutter (`flutter_client/lib/library/home_screen.dart`
* `_buildPlaylistsRow`) which only shows the 5 fixed slots and never
* surfaces the secondary kinds on Home. Operator authorized the
* divergence on 2026-06-01; web UI catch-up tracked as task #53.
*/
private fun buildPlaylistsRow(
internal fun buildPlaylistsRow(
owned: List<PlaylistRef>,
status: SystemPlaylistsStatus,
offline: Boolean,
): List<PlaylistRowItem> {
if (!offline) return buildOnlineRow(owned, status)
val out = mutableListOf<PlaylistRowItem>(
PlaylistRowItem.OfflinePool(OfflinePoolKind.RECENTLY_PLAYED),
PlaylistRowItem.OfflinePool(OfflinePoolKind.LIKED),
)
val (available, greyed) = orderedRealPlaylists(owned).partition { !it.unavailableOffline }
(available + greyed).forEach { out += PlaylistRowItem.Real(it) }
return out
}
/** The online layout: fixed system slots (with placeholders), secondary, user. */
private fun buildOnlineRow(
owned: List<PlaylistRef>,
status: SystemPlaylistsStatus,
): List<PlaylistRowItem> {
val out = mutableListOf<PlaylistRowItem>()
if (offline) {
out += PlaylistRowItem.OfflinePool(OfflinePoolKind.RECENTLY_PLAYED)
out += PlaylistRowItem.OfflinePool(OfflinePoolKind.LIKED)
}
out += owned.firstOrNull { it.systemVariant == "for_you" }
?.let { PlaylistRowItem.Real(it) }
?: PlaylistRowItem.Placeholder("For You", variantFor("for-you", status))
out += owned.firstOrNull { it.systemVariant == "discover" }
?.let { PlaylistRowItem.Real(it) }
?: PlaylistRowItem.Placeholder("Discover", variantFor("discover", status))
val songsLike = owned.filter { it.systemVariant == "songs_like_artist" }.take(3)
val songsLike = owned.filter { it.systemVariant == "songs_like_artist" }.take(SONGS_LIKE_SLOTS)
for (i in 0 until SONGS_LIKE_SLOTS) {
out += songsLike.getOrNull(i)
?.let { PlaylistRowItem.Real(it) }
?: PlaylistRowItem.Placeholder("Songs like…", variantFor("songs-like", status))
}
// Secondary system kinds in server-registry order. Only included
// when actually generated — these depend on library shape (Deep
// cuts needs deep albums, On this day needs prior history, etc.)
// so a missing one means "not enough data" rather than "still
// building".
for (variant in SECONDARY_SYSTEM_VARIANTS) {
owned.firstOrNull { it.systemVariant == variant }
?.let { out += PlaylistRowItem.Real(it) }
owned.firstOrNull { it.systemVariant == variant }?.let { out += PlaylistRowItem.Real(it) }
}
owned.filter { it.systemVariant == null }.forEach { out += PlaylistRowItem.Real(it) }
return out
}
/**
* Curated real-playlist order (system primaries, then secondary, then user).
* Must mirror [buildOnlineRow]'s slot order — the offline row reuses this and
* only differs by dropping placeholders + partitioning available-first.
*/
private fun orderedRealPlaylists(owned: List<PlaylistRef>): List<PlaylistRef> {
val out = mutableListOf<PlaylistRef>()
owned.firstOrNull { it.systemVariant == "for_you" }?.let { out += it }
owned.firstOrNull { it.systemVariant == "discover" }?.let { out += it }
out += owned.filter { it.systemVariant == "songs_like_artist" }.take(SONGS_LIKE_SLOTS)
for (variant in SECONDARY_SYSTEM_VARIANTS) {
owned.firstOrNull { it.systemVariant == variant }?.let { out += it }
}
owned.filter { it.systemVariant == null }.forEach { out += it }
return out
}
private fun variantFor(slot: String, s: SystemPlaylistsStatus): String = when {
s.inFlight -> "building"
s.lastError != null -> "failed"
@@ -105,6 +105,22 @@ class LibraryRepository @Inject constructor(
suspend fun fetchArtistTracks(id: String): List<TrackRef> =
api.getArtistTracks(id).map { it.toDomain() }
/**
* Pulls related artists for the ArtistDetail "Similar artists"
* strip. Network-only — non-critical UI, callers swallow failures
* and render nothing rather than blocking the screen.
*/
suspend fun fetchSimilarArtists(id: String): List<ArtistRef> =
api.getSimilarArtists(id).map { it.toDomain() }
/**
* Pulls the artist's top tracks for the ArtistDetail "Top tracks"
* panel. Network-only; tapping a row plays this exact list as a
* queue, so no persistence is needed.
*/
suspend fun fetchArtistTopTracks(id: String): List<TrackRef> =
api.getArtistTopTracks(id).map { it.toDomain() }
/**
* Pulls the album detail (AlbumRef + tracks), persists both, and
* returns the in-memory snapshot. Lets the AlbumDetail screen
@@ -53,7 +53,7 @@ import com.fabledsword.minstrel.nav.AlbumDetail
import com.fabledsword.minstrel.nav.ArtistDetail
import com.fabledsword.minstrel.shared.formatDuration
import com.fabledsword.minstrel.shared.widgets.TrackRow
import com.fabledsword.minstrel.shared.widgets.EmptyState
import com.fabledsword.minstrel.shared.widgets.ErrorRetry
import com.fabledsword.minstrel.shared.widgets.LikeButton
import com.fabledsword.minstrel.shared.widgets.PullToRefreshScaffold
import com.fabledsword.minstrel.shared.widgets.ServerImage
@@ -116,9 +116,10 @@ private fun AlbumDetailStateContent(
when (val s = state) {
is AlbumDetailUiState.Loading ->
if (s.seed != null) SeededAlbumLoading(s.seed) else SkeletonTrackList()
is AlbumDetailUiState.Error -> EmptyState(
is AlbumDetailUiState.Error -> ErrorRetry(
title = "Couldn't load album",
body = s.message,
message = s.message,
onRetry = { viewModel.refresh() },
)
is AlbumDetailUiState.Success -> {
val albumLiked by viewModel.albumLiked.collectAsStateWithLifecycle()
@@ -5,6 +5,8 @@ import androidx.lifecycle.ViewModel
import androidx.lifecycle.viewModelScope
import androidx.navigation.toRoute
import com.fabledsword.minstrel.api.ErrorCopy
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.recoveries
import com.fabledsword.minstrel.library.data.LibraryRepository
import com.fabledsword.minstrel.likes.data.LikesRepository
import com.fabledsword.minstrel.models.AlbumDetailRef
@@ -36,6 +38,7 @@ class AlbumDetailViewModel @Inject constructor(
private val likes: LikesRepository,
private val player: PlayerController,
private val seedCache: DetailSeedCache,
networkStatus: NetworkStatusController,
savedStateHandle: SavedStateHandle,
) : ViewModel() {
@@ -64,6 +67,14 @@ class AlbumDetailViewModel @Inject constructor(
init {
refresh()
// Screen-level auto-recovery (issue #1245): a detail that failed to
// load while the server was unreachable re-fetches when health
// returns, so coming back to the screen shows content, not an error.
viewModelScope.launch {
networkStatus.recoveries().collect {
if (internal.value is AlbumDetailUiState.Error) refresh()
}
}
}
fun refresh(): Job = viewModelScope.launch {
@@ -1,3 +1,5 @@
@file:Suppress("TooManyFunctions") // Compose screen + private helper composables
package com.fabledsword.minstrel.library.ui
import androidx.compose.animation.Crossfade
@@ -18,6 +20,7 @@ import androidx.compose.foundation.lazy.grid.GridCells
import androidx.compose.foundation.lazy.grid.GridItemSpan
import androidx.compose.foundation.lazy.grid.LazyVerticalGrid
import androidx.compose.foundation.lazy.grid.items
import androidx.compose.foundation.lazy.items as lazyItems
import androidx.compose.foundation.shape.CircleShape
import androidx.compose.material3.Button
import androidx.compose.material3.ExperimentalMaterial3Api
@@ -46,21 +49,27 @@ import com.composables.icons.lucide.Lucide
import com.composables.icons.lucide.Play
import com.composables.icons.lucide.User
import com.fabledsword.minstrel.library.widgets.AlbumCard
import com.fabledsword.minstrel.models.ArtistDetailRef
import com.fabledsword.minstrel.library.widgets.ArtistCard
import com.fabledsword.minstrel.models.ArtistRef
import com.fabledsword.minstrel.models.TrackRef
import com.fabledsword.minstrel.models.albumCoverPath
import com.fabledsword.minstrel.nav.AlbumDetail
import com.fabledsword.minstrel.shared.widgets.EmptyState
import com.fabledsword.minstrel.nav.ArtistDetail
import com.fabledsword.minstrel.shared.formatDuration
import com.fabledsword.minstrel.shared.widgets.ErrorRetry
import com.fabledsword.minstrel.shared.widgets.HorizontalScrollRow
import com.fabledsword.minstrel.shared.widgets.LikeButton
import com.fabledsword.minstrel.shared.widgets.PullToRefreshScaffold
import com.fabledsword.minstrel.shared.widgets.ServerImage
import com.fabledsword.minstrel.shared.widgets.SkeletonAlbumTile
import com.fabledsword.minstrel.shared.widgets.TrackRow
@OptIn(ExperimentalMaterial3Api::class)
@Composable
fun ArtistDetailScreen(
navController: NavHostController,
viewModel: ArtistDetailViewModel = hiltViewModel(),
playerViewModel: com.fabledsword.minstrel.player.ui.PlayerViewModel = hiltViewModel(),
) {
val state by viewModel.uiState.collectAsStateWithLifecycle()
val snackbarHostState = remember { SnackbarHostState() }
@@ -98,25 +107,38 @@ fun ArtistDetailScreen(
} else {
SkeletonArtistAlbumsGrid()
}
is ArtistDetailUiState.Error -> EmptyState(
is ArtistDetailUiState.Error -> ErrorRetry(
title = "Couldn't load artist",
body = s.message,
message = s.message,
onRetry = { viewModel.refresh() },
)
is ArtistDetailUiState.Success -> {
val artistLiked by viewModel.artistLiked
.collectAsStateWithLifecycle()
is ArtistDetailUiState.Success ->
ArtistSuccessBody(s, viewModel, playerViewModel, navController)
}
}
}
}
}
@Composable
private fun ArtistSuccessBody(
state: ArtistDetailUiState.Success,
viewModel: ArtistDetailViewModel,
playerViewModel: com.fabledsword.minstrel.player.ui.PlayerViewModel,
navController: NavHostController,
) {
val artistLiked by viewModel.artistLiked.collectAsStateWithLifecycle()
val playerState by playerViewModel.uiState.collectAsStateWithLifecycle()
ArtistBody(
detail = s.detail,
state = state,
artistLiked = artistLiked,
playingTrackId = playerState.currentTrack?.id,
onPlay = viewModel::playArtist,
onToggleLike = viewModel::toggleLikeArtist,
onAlbumClick = { id -> navController.navigate(AlbumDetail(id)) },
onTopTrackClick = viewModel::playTopTracks,
onSimilarArtistClick = { id -> navController.navigate(ArtistDetail(id)) },
)
}
}
}
}
}
}
private fun titleFor(state: ArtistDetailUiState): String = when (state) {
@@ -127,12 +149,16 @@ private fun titleFor(state: ArtistDetailUiState): String = when (state) {
@Composable
private fun ArtistBody(
detail: ArtistDetailRef,
state: ArtistDetailUiState.Success,
artistLiked: Boolean,
playingTrackId: String?,
onPlay: () -> Unit,
onToggleLike: () -> Unit,
onAlbumClick: (String) -> Unit,
onTopTrackClick: (Int) -> Unit,
onSimilarArtistClick: (String) -> Unit,
) {
val detail = state.detail
LazyVerticalGrid(
columns = GridCells.Adaptive(minSize = 176.dp),
modifier = Modifier.fillMaxSize(),
@@ -149,6 +175,23 @@ private fun ArtistBody(
onToggleLike = onToggleLike,
)
}
if (state.topTracks.isNotEmpty()) {
item(span = { GridItemSpan(maxLineSpan) }) {
TopTracksPanel(
tracks = state.topTracks,
playingTrackId = playingTrackId,
onTrackClick = onTopTrackClick,
)
}
}
if (state.similarArtists.isNotEmpty()) {
item(span = { GridItemSpan(maxLineSpan) }) {
SimilarArtistsStrip(
artists = state.similarArtists,
onArtistClick = onSimilarArtistClick,
)
}
}
if (detail.albums.isEmpty()) {
item(span = { GridItemSpan(maxLineSpan) }) {
EmptyAlbumsHint()
@@ -161,6 +204,67 @@ private fun ArtistBody(
}
}
@Composable
private fun TopTracksPanel(
tracks: List<TrackRef>,
playingTrackId: String?,
onTrackClick: (Int) -> Unit,
) {
Column(modifier = Modifier.fillMaxWidth()) {
Text(
text = "Top tracks",
style = MaterialTheme.typography.titleLarge,
color = MaterialTheme.colorScheme.onBackground,
modifier = Modifier.padding(horizontal = 16.dp, vertical = 8.dp),
)
tracks.forEachIndexed { index, track ->
TopTrackRow(
track = track,
nowPlaying = track.id == playingTrackId,
onClick = { onTrackClick(index) },
)
}
}
}
@Composable
private fun TopTrackRow(
track: TrackRef,
nowPlaying: Boolean,
onClick: () -> Unit,
) {
TrackRow(
title = track.title,
artist = track.artistName,
trackId = track.id,
onClick = onClick,
nowPlaying = nowPlaying,
trailing = {
Text(
text = formatDuration(track.durationSec, zero = "--:--"),
style = MaterialTheme.typography.bodySmall,
color = MaterialTheme.colorScheme.onSurfaceVariant,
modifier = Modifier.padding(start = 12.dp),
)
},
)
}
@Composable
private fun SimilarArtistsStrip(
artists: List<ArtistRef>,
onArtistClick: (String) -> Unit,
) {
HorizontalScrollRow(title = "Similar artists") {
lazyItems(items = artists, key = { it.id }) { artist ->
ArtistCard(
artist = artist,
onClick = { onArtistClick(artist.id) },
)
}
}
}
@Composable
private fun ArtistHeader(
artist: ArtistRef,
@@ -5,10 +5,13 @@ import androidx.lifecycle.ViewModel
import androidx.lifecycle.viewModelScope
import androidx.navigation.toRoute
import com.fabledsword.minstrel.api.ErrorCopy
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.recoveries
import com.fabledsword.minstrel.library.data.LibraryRepository
import com.fabledsword.minstrel.likes.data.LikesRepository
import com.fabledsword.minstrel.models.ArtistDetailRef
import com.fabledsword.minstrel.models.ArtistRef
import com.fabledsword.minstrel.models.TrackRef
import com.fabledsword.minstrel.nav.ArtistDetail
import com.fabledsword.minstrel.nav.DetailSeedCache
import com.fabledsword.minstrel.player.PlayerController
@@ -23,13 +26,18 @@ import kotlinx.coroutines.flow.asStateFlow
import kotlinx.coroutines.flow.receiveAsFlow
import kotlinx.coroutines.flow.stateIn
import kotlinx.coroutines.launch
import timber.log.Timber
import javax.inject.Inject
private const val SHARE_STOP_TIMEOUT_MS = 5_000L
sealed interface ArtistDetailUiState {
data class Loading(val seed: ArtistRef? = null) : ArtistDetailUiState
data class Success(val detail: ArtistDetailRef) : ArtistDetailUiState
data class Success(
val detail: ArtistDetailRef,
val similarArtists: List<ArtistRef> = emptyList(),
val topTracks: List<TrackRef> = emptyList(),
) : ArtistDetailUiState
data class Error(val message: String) : ArtistDetailUiState
}
@@ -39,6 +47,7 @@ class ArtistDetailViewModel @Inject constructor(
private val likes: LikesRepository,
private val player: PlayerController,
private val seedCache: DetailSeedCache,
networkStatus: NetworkStatusController,
savedStateHandle: SavedStateHandle,
) : ViewModel() {
@@ -63,6 +72,19 @@ class ArtistDetailViewModel @Inject constructor(
init {
refresh()
// Screen-level auto-recovery (issue #1245). From Error, redo the
// whole load; from Success, re-pull just the secondary sections —
// their fetch failures are swallowed to emptyList, so a dead zone
// during the first load leaves them permanently absent otherwise.
viewModelScope.launch {
networkStatus.recoveries().collect {
when (internal.value) {
is ArtistDetailUiState.Error -> refresh()
is ArtistDetailUiState.Success -> loadSecondarySections()
is ArtistDetailUiState.Loading -> Unit
}
}
}
}
fun toggleLikeArtist() {
@@ -82,6 +104,7 @@ class ArtistDetailViewModel @Inject constructor(
try {
val detail = repository.refreshArtistDetail(artistId)
internal.value = ArtistDetailUiState.Success(detail)
loadSecondarySections()
} catch (
@Suppress("TooGenericExceptionCaught") e: Throwable,
) {
@@ -89,6 +112,31 @@ class ArtistDetailViewModel @Inject constructor(
}
}
/**
* Pulls the non-critical "Similar artists" + "Top tracks" sections
* and folds them into the live Success state. Each failure is
* swallowed (logged only) — these sections render absent rather than
* blocking, so the core artist view never depends on them.
*/
private suspend fun loadSecondarySections() {
val similar = sectionOrEmpty { repository.fetchSimilarArtists(artistId) }
val top = sectionOrEmpty { repository.fetchArtistTopTracks(artistId) }
val current = internal.value
if (current is ArtistDetailUiState.Success) {
internal.value = current.copy(similarArtists = similar, topTracks = top)
}
}
private inline fun <T> sectionOrEmpty(block: () -> List<T>): List<T> =
try {
block()
} catch (
@Suppress("TooGenericExceptionCaught") e: Throwable,
) {
Timber.w(e, "Artist secondary section fetch failed for %s", artistId)
emptyList()
}
/**
* Pull every track across the artist's albums, shuffle, and start
* playing. Mirrors Flutter's artist Play button — pressing Play on
@@ -118,4 +166,20 @@ class ArtistDetailViewModel @Inject constructor(
}
}
}
/**
* Plays the loaded "Top tracks" list starting at [index] — the top
* tracks are themselves the playable queue (mirrors the web client's
* sectionTracks = top-tracks behaviour). No network round-trip; the
* list is already in the Success state.
*/
fun playTopTracks(index: Int) {
val tracks = (internal.value as? ArtistDetailUiState.Success)?.topTracks.orEmpty()
if (tracks.isEmpty()) return
player.setQueue(
tracks = tracks,
initialIndex = index.coerceIn(0, tracks.lastIndex),
source = "artist-top:$artistId",
)
}
}
@@ -24,11 +24,17 @@ class LibraryViewModel @Inject constructor(
combine(
repository.observeArtists(),
repository.observeAlbums(),
) { artists, albums ->
if (artists.isEmpty() && albums.isEmpty()) {
UiState.Empty
} else {
syncController.lastSyncError,
) { artists, albums, syncError ->
when {
artists.isNotEmpty() || albums.isNotEmpty() ->
UiState.Success(LibraryData(artists, albums))
// Empty cache + failed sync is a load failure, not an empty
// library — surface it so the user gets a Retry instead of
// the welcome copy. Heals automatically: SyncController
// re-syncs on health recovery, clearing lastSyncError.
syncError != null -> UiState.Error(syncError)
else -> UiState.Empty
}
}
.asCacheFirstStateFlow(viewModelScope)
@@ -25,6 +25,9 @@ import androidx.lifecycle.ViewModel
import androidx.lifecycle.compose.collectAsStateWithLifecycle
import androidx.lifecycle.viewModelScope
import androidx.navigation.NavHostController
import com.fabledsword.minstrel.api.ErrorCopy
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.recoveries
import com.fabledsword.minstrel.likes.data.LikesRepository
import com.fabledsword.minstrel.library.widgets.AlbumCard
import com.fabledsword.minstrel.library.widgets.ArtistCard
@@ -38,6 +41,7 @@ import com.fabledsword.minstrel.shared.UiState
import com.fabledsword.minstrel.shared.asCacheFirstStateFlow
import com.fabledsword.minstrel.shared.widgets.TrackRow
import com.fabledsword.minstrel.shared.widgets.EmptyState
import com.fabledsword.minstrel.shared.widgets.ErrorRetry
import com.fabledsword.minstrel.shared.widgets.HorizontalScrollRow
import com.fabledsword.minstrel.shared.widgets.LikeButton
import com.fabledsword.minstrel.shared.widgets.LoadingCentered
@@ -46,6 +50,7 @@ import com.fabledsword.minstrel.shared.widgets.TrackCoverThumb
import com.fabledsword.minstrel.shared.widgets.trackactions.TrackActionsButton
import dagger.hilt.android.lifecycle.HiltViewModel
import kotlinx.coroutines.Job
import kotlinx.coroutines.flow.MutableStateFlow
import kotlinx.coroutines.flow.StateFlow
import kotlinx.coroutines.flow.combine
import kotlinx.coroutines.launch
@@ -68,14 +73,25 @@ data class LikedSections(
class LikedTabViewModel @Inject constructor(
private val repository: LikesRepository,
private val player: PlayerController,
networkStatus: NetworkStatusController,
) : ViewModel() {
/** Copy for the most recent refresh failure; null once one succeeds. */
private val refreshError = MutableStateFlow<String?>(null)
init {
refresh()
// Screen-level auto-recovery (issue #1245): re-pull when server
// health returns, so a load that failed offline heals unprompted.
viewModelScope.launch {
networkStatus.recoveries().collect { refresh() }
}
}
fun refresh(): Job = viewModelScope.launch {
refreshError.value = null
runCatching { repository.refreshIds() }
.onFailure { refreshError.value = ErrorCopy.fromThrowable(it) }
}
/**
@@ -92,9 +108,16 @@ class LikedTabViewModel @Inject constructor(
repository.observeLikedArtists(),
repository.observeLikedAlbums(),
repository.observeLikedTracks(),
) { artists, albums, tracks ->
refreshError,
) { artists, albums, tracks, err ->
val sections = LikedSections(artists, albums, tracks)
if (sections.isAllEmpty) UiState.Empty else UiState.Success(sections)
when {
!sections.isAllEmpty -> UiState.Success(sections)
// Empty cache + failed refresh is a load failure, not "no
// likes yet" — surface Error so the user gets a Retry.
err != null -> UiState.Error(err)
else -> UiState.Empty
}
}.asCacheFirstStateFlow(viewModelScope)
/**
@@ -130,9 +153,10 @@ fun LikedTab(
body = "Tap the heart on an artist, album, or track to start " +
"building your liked collection.",
)
is UiState.Error -> EmptyState(
is UiState.Error -> ErrorRetry(
title = "Couldn't load likes",
body = s.message,
message = s.message,
onRetry = { viewModel.refresh() },
)
is UiState.Success -> LikedContent(
sections = s.data,
@@ -20,6 +20,8 @@ data class PlaylistRef(
val trackCount: Int = 0,
val coverUrl: String = "",
val ownerUsername: String = "",
/** All member tracks resident in the audio cache — playable fully offline. */
val fullyCached: Boolean = false,
) {
val isSystem: Boolean get() = systemVariant != null
@@ -30,6 +32,13 @@ data class PlaylistRef(
* expose a refresh trigger; rebuilds are scheduler-driven).
*/
val refreshable: Boolean get() = isSystem && systemVariant != "songs_like_artist"
/**
* Can't be relied on while offline — either it needs the live server to
* (re)generate (a refreshable system mix) or not all its tracks are cached.
* Callers combine this with the current offline state to grey the tile.
*/
val unavailableOffline: Boolean get() = refreshable || !fullyCached
}
/**
@@ -16,4 +16,11 @@ data class UserRef(
val id: String,
val username: String,
val isAdmin: Boolean = false,
/**
* Account-level diagnostics opt-in (M9). NOT present in the login
* response (that's the narrow UserView) — populated by the /api/me
* refresh in AuthController so a remote admin flip reaches the
* device. Default false until the first refresh.
*/
val debugModeEnabled: Boolean = false,
)
@@ -1,5 +1,9 @@
@file:OptIn(ExperimentalSerializationApi::class)
package com.fabledsword.minstrel.models.wire
import kotlinx.serialization.EncodeDefault
import kotlinx.serialization.ExperimentalSerializationApi
import kotlinx.serialization.SerialName
import kotlinx.serialization.Serializable
@@ -15,7 +19,11 @@ import kotlinx.serialization.Serializable
@Serializable
data class PlayStartedRequest(
val type: String = "play_started",
// Discriminator MUST stay on the wire. The app's Json has
// encodeDefaults=false, so without this a `type` left at its
// default is omitted entirely and the server 400s the event as
// "unknown event type" — silently dropping every native play.
@EncodeDefault(EncodeDefault.Mode.ALWAYS) val type: String = "play_started",
@SerialName("track_id") val trackId: String,
@SerialName("client_id") val clientId: String,
val source: String? = null,
@@ -28,21 +36,21 @@ data class PlayStartedResponse(
@Serializable
data class PlayEndedRequest(
val type: String = "play_ended",
@EncodeDefault(EncodeDefault.Mode.ALWAYS) val type: String = "play_ended",
@SerialName("play_event_id") val playEventId: String,
@SerialName("duration_played_ms") val durationPlayedMs: Long,
)
@Serializable
data class PlaySkippedRequest(
val type: String = "play_skipped",
@EncodeDefault(EncodeDefault.Mode.ALWAYS) val type: String = "play_skipped",
@SerialName("play_event_id") val playEventId: String,
@SerialName("position_ms") val positionMs: Long,
)
@Serializable
data class PlayOfflineRequest(
val type: String = "play_offline",
@EncodeDefault(EncodeDefault.Mode.ALWAYS) val type: String = "play_offline",
@SerialName("track_id") val trackId: String,
@SerialName("client_id") val clientId: String,
val at: String,
@@ -15,8 +15,12 @@ import kotlinx.serialization.Serializable
* - rediscoverArtists → artist
* - mostPlayedTracks → track
* - lastPlayedArtists → artist
* - youMightLikeAlbums → album
* - youMightLikeArtists → artist
*
* All slices default to empty so missing keys don't throw on decode.
* All slices default to empty so missing keys don't throw on decode
* which also keeps the client safe against older servers that don't
* emit the you-might-like sections yet.
*/
@Serializable
data class HomeIndexWire(
@@ -25,4 +29,6 @@ data class HomeIndexWire(
@SerialName("rediscover_artists") val rediscoverArtists: List<String> = emptyList(),
@SerialName("most_played_tracks") val mostPlayedTracks: List<String> = emptyList(),
@SerialName("last_played_artists") val lastPlayedArtists: List<String> = emptyList(),
@SerialName("you_might_like_albums") val youMightLikeAlbums: List<String> = emptyList(),
@SerialName("you_might_like_artists") val youMightLikeArtists: List<String> = emptyList(),
)
@@ -18,4 +18,5 @@ data class MyProfileWire(
@SerialName("display_name") val displayName: String? = null,
val email: String? = null,
@SerialName("is_admin") val isAdmin: Boolean = false,
@SerialName("debug_mode_enabled") val debugModeEnabled: Boolean = false,
)
@@ -58,40 +58,78 @@ class AudioPrefetcher @Inject constructor(
private val activeJobs = mutableMapOf<String, Job>()
private val mutex = Mutex()
private data class ReconcileInput(
val queue: List<Pair<String, String>>,
val index: Int,
val window: Int,
val isPlaying: Boolean,
)
init {
scope.launch {
combine(
playerController.uiState.map { it.queue.map { t -> t.id to t.streamUrl } },
playerController.uiState.map { it.queueIndex },
authStore.cacheSettings.map { it.prefetchWindow },
) { queue, index, window -> Triple(queue, index, window) }
playerController.uiState.map { it.isPlaying },
) { queue, index, window, isPlaying ->
ReconcileInput(queue, index, window, isPlaying)
}
.distinctUntilChanged()
.collect { (queue, index, window) -> reconcile(queue, index, window) }
.collect { input ->
reconcile(input.queue, input.index, input.window, input.isPlaying)
}
}
}
/**
* Apply the prefetch window to the current queue.
*
* Cancellation of out-of-window jobs always runs -- a queue mutation
* or skip should free bandwidth from stale prefetches immediately.
* Starting new prefetches is gated on [isPlaying]: until the current
* track is actually playing, every byte of upstream bandwidth should
* land on it, not on upcoming-track prefetches. Without this gate a
* cold start fanned out 4-6 concurrent CacheWriter jobs against the
* same OkHttp client as the playback DataSource and the user waited
* ~25 s for the first audio to start; with the gate the current
* track gets the full pipe to its first STATE_READY, then the
* prefetcher fills in the next window.
*/
private suspend fun reconcile(
queue: List<Pair<String, String>>,
index: Int,
window: Int,
isPlaying: Boolean,
) {
mutex.withLock {
if (index < 0 || queue.isEmpty() || window <= 0) {
val targets = computeTargets(queue, index, window)
if (targets.isEmpty()) {
cancelAllLocked()
return
}
// Exclude the currently-playing track (it's loaded by the
// player itself) and walk `window` tracks forward.
val firstIdx = (index + 1).coerceAtMost(queue.size)
val lastIdx = (index + window).coerceAtMost(queue.size - 1)
if (firstIdx > lastIdx) {
cancelAllLocked()
return
}
val targets = queue.subList(firstIdx, lastIdx + 1)
val targetIds = targets.mapTo(mutableSetOf()) { it.first }
// Cancellation always runs so a queue mutation or skip frees
// the pipe immediately, even while paused.
cancelOutOfWindowLocked(targetIds)
if (isPlaying) startInWindowLocked(targets)
}
}
// Cancel jobs for tracks that have slid out of the window.
private fun computeTargets(
queue: List<Pair<String, String>>,
index: Int,
window: Int,
): List<Pair<String, String>> {
// Exclude the currently-playing track (it's loaded by the player
// itself) and walk `window` tracks forward.
val firstIdx = index + 1
val lastIdx = (index + window).coerceAtMost(queue.size - 1)
val isValid = index >= 0 && queue.isNotEmpty() && window > 0 && firstIdx <= lastIdx
return if (isValid) queue.subList(firstIdx, lastIdx + 1) else emptyList()
}
private fun cancelOutOfWindowLocked(targetIds: Set<String>) {
activeJobs.entries
.filter { it.key !in targetIds }
.toList()
@@ -99,10 +137,12 @@ class AudioPrefetcher @Inject constructor(
job.cancel()
activeJobs.remove(id)
}
}
// Start prefetches for new arrivals. Skip blank URLs (these
// come from minimal TrackRefs synthesized from playlist rows
// when the upstream track was removed from the library).
private fun startInWindowLocked(targets: List<Pair<String, String>>) {
// Skip blank URLs (these come from minimal TrackRefs synthesized
// from playlist rows when the upstream track was removed from
// the library).
for ((trackId, streamUrl) in targets) {
if (trackId in activeJobs || streamUrl.isBlank()) continue
val job = scope.launch(Dispatchers.IO) {
@@ -112,7 +152,6 @@ class AudioPrefetcher @Inject constructor(
activeJobs[trackId] = job
}
}
}
private fun cancelAllLocked() {
activeJobs.values.forEach { it.cancel() }
@@ -0,0 +1,77 @@
package com.fabledsword.minstrel.player
import android.content.Context
import android.net.wifi.WifiManager
import android.os.Build
import android.os.PowerManager
import timber.log.Timber
/**
* Holds a WiFi + CPU wake lock for the lifetime of an active UPnP cast.
*
* While casting, the wrapped ExoPlayer is paused, so its automatic
* `WAKE_MODE_NETWORK` locks are released and nothing keeps the phone's radio
* awake. On a locked, on-battery phone the WiFi enters power-save within
* seconds and the CPU dozes; that stalls the 1 Hz liveness poll to the
* renderer and the queue-extend calls to the server. The poll then trips the
* drop threshold and playback falls back to a phone that *also* has no
* network — silence, while the renderer was streaming fine the whole time.
*
* A high-performance / low-latency [WifiManager.WifiLock] keeps the radio out
* of power-save, and a partial [PowerManager.WakeLock] keeps the poll
* coroutine scheduled. Both are acquired when a route goes active and released
* the moment it drops or the user switches back to the phone.
*
* All methods are idempotent (`setReferenceCounted(false)` + held-checks) and
* synchronized, so they are safe to call from any thread.
*/
class CastNetworkLock(context: Context) {
private val appContext = context.applicationContext
private val wifiLock: WifiManager.WifiLock? =
(appContext.getSystemService(Context.WIFI_SERVICE) as? WifiManager)
?.createWifiLock(wifiLockMode(), LOCK_TAG)
?.apply { setReferenceCounted(false) }
private val wakeLock: PowerManager.WakeLock? =
(appContext.getSystemService(Context.POWER_SERVICE) as? PowerManager)
?.newWakeLock(PowerManager.PARTIAL_WAKE_LOCK, LOCK_TAG)
?.apply { setReferenceCounted(false) }
@Synchronized
fun acquire() {
runCatching { wifiLock?.takeUnless { it.isHeld }?.acquire() }
.onFailure { Timber.w(it, "CastNetworkLock: wifi acquire failed") }
runCatching { wakeLock?.takeUnless { it.isHeld }?.acquire() }
.onFailure { Timber.w(it, "CastNetworkLock: wake acquire failed") }
Timber.w(
"CastNetworkLock acquired (wifi=%s wake=%s)",
wifiLock?.isHeld, wakeLock?.isHeld,
)
}
@Synchronized
fun release() {
runCatching { wifiLock?.takeIf { it.isHeld }?.release() }
.onFailure { Timber.w(it, "CastNetworkLock: wifi release failed") }
runCatching { wakeLock?.takeIf { it.isHeld }?.release() }
.onFailure { Timber.w(it, "CastNetworkLock: wake release failed") }
Timber.w("CastNetworkLock released")
}
private companion object {
const val LOCK_TAG = "minstrel:upnp-cast"
// WIFI_MODE_FULL_LOW_LATENCY (API 29+) both disables power-save and
// lowers latency; FULL_HIGH_PERF is the pre-Q equivalent for keeping
// the radio fully awake.
fun wifiLockMode(): Int =
if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.Q) {
WifiManager.WIFI_MODE_FULL_LOW_LATENCY
} else {
@Suppress("DEPRECATION")
WifiManager.WIFI_MODE_FULL_HIGH_PERF
}
}
}
@@ -0,0 +1,640 @@
@file:Suppress("TooManyFunctions") // Mirrors Player surface: ~16 methods is the API.
package com.fabledsword.minstrel.player
import android.os.Handler
import android.os.SystemClock
import androidx.lifecycle.DefaultLifecycleObserver
import androidx.lifecycle.LifecycleOwner
import androidx.lifecycle.ProcessLifecycleOwner
import androidx.media3.common.ForwardingPlayer
import androidx.media3.common.MediaItem
import androidx.media3.common.Player
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.ServerHealth
import com.fabledsword.minstrel.player.output.ActiveUpnp
import com.fabledsword.minstrel.player.output.ActiveUpnpHolder
import com.fabledsword.minstrel.player.output.upnp.SoapFaultException
import com.fabledsword.minstrel.player.output.upnp.TransportState
import java.io.IOException
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.channels.Channel
import kotlinx.coroutines.ExperimentalCoroutinesApi
import kotlinx.coroutines.isActive
import kotlinx.coroutines.launch
import kotlinx.coroutines.selects.onTimeout
import kotlinx.coroutines.selects.select
import timber.log.Timber
/**
* Integration point for UPnP transport parity. Wraps the local
* ExoPlayer; every transport method either forwards (local route
* active -- the default) or translates into AVTransport SOAP +
* [RemotePlayerState] updates (UPnP route active).
*
* Created inside [MinstrelPlayerService]; runs on the service's main
* looper. Network SOAP calls fire on [Dispatchers.IO]. While UPnP is
* active, the MediaSession's reads of [Player.isPlaying] and
* [Player.getCurrentPosition] pull from [RemotePlayerState]; the
* wrapped ExoPlayer stays paused at the position it had when the
* route was selected.
*
* Drop heuristic: the 1 Hz poll loop is the *sole* arbiter of route
* liveness -- [RemotePlayerState.recordPollFailure]'s rolling threshold
* (DROP_THRESHOLD consecutive failures) fires [onDrop]. A failed transport
* command (play/pause/seek/next) does NOT drop on its own: a locked phone's
* WiFi power-save can stall a single command's socket I/O for a second or
* two while the renderer is perfectly reachable, so commands retry on
* transient IO failure and otherwise defer to the poll loop. The factory
* wraps the [onDrop] callback into a SharedFlow consumed by the NowPlaying
* surface as a snackbar.
*
* Queue mode: OutputPickerController loads the full queue into Sonos's
* native queue via ClearQueue + AddURIToQueue, then points the
* transport at x-rincon-queue:<udn>#0. Skip/prev/seekTo delegate to
* AVTransport Next/Previous/SeekToTrack so Sonos manages gap-free
* advance natively. PollLoop syncs the local cursor by comparing the
* 1-based Track index from GetPositionInfo.
*/
class MinstrelForwardingPlayer(
private val delegate: Player,
private val holder: ActiveUpnpHolder,
private val remoteState: RemotePlayerState,
private val castNetworkLock: CastNetworkLock,
private val networkStatus: NetworkStatusController,
private val onDrop: (routeName: String) -> Unit,
) : ForwardingPlayer(delegate) {
private val scope = CoroutineScope(SupervisorJob() + Dispatchers.IO)
private val handler = Handler(delegate.applicationLooper)
private var pollJob: Job? = null
// Tracks consecutive non-PLAYING poll observations so a single transient
// PAUSED_PLAYBACK / STOPPED tick during a Sonos track transition does not
// flip the play/pause button. Manual pause still feels instant because it
// bypasses the poll entirely via applyTransportPaused().
@Volatile private var nonPlayingPollStreak = 0
// Wall-clock of the most-recent within-track seek we issued to Sonos.
// pollOnce uses this to suppress position overwrites for SEEK_ACK_WINDOW_MS
// -- Sonos can take 1-2s to apply a Seek, and a poll landing inside that
// window reports the *old* position. Without the lockout the scrubber
// visibly jumps backwards immediately after a drag, then forwards again.
@Volatile private var lastSeekIssuedAtMs: Long = 0L
// Wake channel for the poll loop. requestImmediatePoll() trySend's a Unit;
// pollLoop's select{} races the delay against this channel so the next
// pollOnce can fire immediately instead of waiting up to POLL_INTERVAL_MS.
// Used on activity resume (ProcessLifecycleOwner.ON_RESUME) so the UI
// catches up to Sonos within RTT rather than the full poll cadence.
// CONFLATED so repeated trySend's between polls don't queue up.
private val pollTrigger = Channel<Unit>(Channel.CONFLATED)
// External Player.Listener registry (separate from super.addListener which
// forwards to the wrapped ExoPlayer). The wrapped player is paused with
// no audio loaded while UPnP is active, so it never fires events for our
// synthesized remote state -- the MediaSession's notification card and
// lock-screen scrubber stay frozen on whatever state was last captured
// before UPnP took over. We dual-register: super.addListener keeps the
// listener attached to the delegate (so local-playback events still
// reach it), AND we hold a ref here so we can directly invoke listener
// callbacks on remote-state changes. The listener's read of isPlaying /
// duration / position then routes through our overrides to remoteState.
private val externalListeners = mutableListOf<Player.Listener>()
@Volatile private var lastNotifiedIsPlaying: Boolean = false
@Volatile private var lastNotifiedTrackIdx: Int = -1
private val lifecycleObserver = object : DefaultLifecycleObserver {
override fun onResume(owner: LifecycleOwner) {
pollTrigger.trySend(Unit)
}
}
init {
scope.launch {
holder.active.collect { active -> onActiveChanged(active) }
}
scope.launch {
// When the phone's own network recovers, reconcile a held cast
// session immediately instead of waiting up to POLL_INTERVAL_MS:
// the renderer kept playing, so one round-trip re-confirms it and
// resyncs the cursor/position. Only a transition *into* Healthy
// while a route is active matters.
var prev = networkStatus.state.value
networkStatus.state.collect { next ->
if (next == ServerHealth.Healthy && prev != ServerHealth.Healthy && isRemote()) {
pollTrigger.trySend(Unit)
}
prev = next
}
}
// Process lifecycle is observed on the main thread; ProcessLifecycleOwner's
// addObserver requires it. The observer just trySend's to the channel.
handler.post {
ProcessLifecycleOwner.get().lifecycle.addObserver(lifecycleObserver)
}
}
private fun isRemote(): Boolean = holder.active.value != null
/**
* Returns true when selectUpnp has marked a UPnP route as the intended
* target but loadQueueOnSonos hasn't yet wired ActiveUpnp. During this
* window we drop transport commands silently -- they would hit Sonos's
* stale state from a prior session and trigger restarts.
*/
private fun isLoadingUpnp(): Boolean =
holder.target.value != null && holder.active.value == null
// ─── setMediaItems intercepts ──────────────────────────────────────
// When PlayerController.setQueue replaces the queue while Sonos is the
// active route, the wrapped delegate's queue gets the new items but
// Sonos's native queue still holds the OLD tracks -- and the play()
// that PlayerController fires immediately after setMediaItems would
// resume the old Sonos queue (user reported on-device: "player view
// updates but Sonos queue does not"). We clear active + set target
// synchronously here so the next play() in the same IPC sequence
// drops via isLoadingUpnp() = true; the OutputPickerController
// observes the uiState.queue change and runs the resync (re-clears
// Sonos's native queue + AddURIToQueue the new tracks + Play).
override fun setMediaItems(mediaItems: List<MediaItem>) {
super.setMediaItems(mediaItems)
markPendingResyncIfRemote()
}
override fun setMediaItems(mediaItems: List<MediaItem>, resetPosition: Boolean) {
super.setMediaItems(mediaItems, resetPosition)
markPendingResyncIfRemote()
}
override fun setMediaItems(mediaItems: List<MediaItem>, startIndex: Int, startPositionMs: Long) {
super.setMediaItems(mediaItems, startIndex, startPositionMs)
markPendingResyncIfRemote()
}
private fun markPendingResyncIfRemote() {
val wasActive = holder.active.value ?: return
Timber.w(
"setMediaItems while UPnP active (%s) -- marking pending resync",
wasActive.routeName,
)
holder.set(null)
holder.setTarget(wasActive.routeId)
}
override fun play() {
if (isLoadingUpnp()) {
Timber.w("ForwardingPlayer.play() dropped -- UPnP loading")
return
}
val active = holder.active.value
Timber.w("ForwardingPlayer.play() active=%s", active?.routeName)
if (active == null) {
super.play()
} else {
remoteState.setPlayIntent(true)
scope.launch {
runCatching { retryTransport { active.avTransport.play() } }
.onSuccess {
remoteState.applyTransportPlaying()
notifyRemoteStateChanged()
}
.onFailure { handleTransportFailure(active, it) }
}
}
}
override fun pause() {
if (isLoadingUpnp()) {
Timber.w("ForwardingPlayer.pause() dropped -- UPnP loading")
return
}
val active = holder.active.value
Timber.w("ForwardingPlayer.pause() active=%s", active?.routeName)
if (active == null) {
super.pause()
} else {
remoteState.setPlayIntent(false)
scope.launch {
runCatching { retryTransport { active.avTransport.pause() } }
.onSuccess {
remoteState.applyTransportPaused()
notifyRemoteStateChanged()
}
.onFailure { handleTransportFailure(active, it) }
}
}
}
override fun seekTo(positionMs: Long) {
if (isLoadingUpnp()) {
Timber.w("ForwardingPlayer.seekTo(positionMs) dropped -- UPnP loading")
return
}
val active = holder.active.value
Timber.w("ForwardingPlayer.seekTo(%dms) active=%s", positionMs, active?.routeName)
if (active == null) {
super.seekTo(positionMs)
} else {
lastSeekIssuedAtMs = SystemClock.elapsedRealtime()
remoteState.applyPositionInfo(
positionMs = positionMs,
durationMs = remoteState.durationMs,
trackUri = remoteState.currentTrackUri,
trackNumber = remoteState.trackNumber,
)
scope.launch {
runCatching { retryTransport { active.avTransport.seek(positionMs) } }
.onFailure { handleTransportFailure(active, it) }
}
}
}
/**
* Widget-driven track change (user taps a track in the queue widget).
* Seeks Sonos to the correct queue slot, then seeks within-track if
* [positionMs] is non-zero.
*/
override fun seekTo(mediaItemIndex: Int, positionMs: Long) {
if (isLoadingUpnp()) {
Timber.w("ForwardingPlayer.seekTo(idx, positionMs) dropped -- UPnP loading")
return
}
val active = holder.active.value
Timber.w(
"ForwardingPlayer.seekTo(idx=%d, %dms) active=%s",
mediaItemIndex, positionMs, active?.routeName,
)
if (active == null) {
super.seekTo(mediaItemIndex, positionMs)
return
}
super.seekTo(mediaItemIndex, positionMs)
remoteState.beginPendingTransport(
SystemClock.elapsedRealtime() + PENDING_TRANSPORT_SAFETY_TIMEOUT_MS,
)
scope.launch {
runCatching {
retryTransport {
active.avTransport.seekToTrack(mediaItemIndex + 1)
if (positionMs > 0L) {
active.avTransport.seek(positionMs)
}
}
}.onFailure { handleTransportFailure(active, it) }
}
}
override fun seekToNextMediaItem() {
if (isLoadingUpnp()) {
Timber.w("ForwardingPlayer.seekToNextMediaItem() dropped -- UPnP loading")
return
}
val active = holder.active.value
Timber.w("ForwardingPlayer.seekToNextMediaItem() active=%s", active?.routeName)
if (active == null) {
super.seekToNextMediaItem()
return
}
// Super first for immediate local cursor advance (UI feedback);
// then delegate to Sonos Next. PollLoop reconciles cursor via
// Track index if they diverge.
super.seekToNextMediaItem()
remoteState.beginPendingTransport(
SystemClock.elapsedRealtime() + PENDING_TRANSPORT_SAFETY_TIMEOUT_MS,
)
scope.launch {
runCatching { retryTransport { active.avTransport.next() } }
.onFailure { handleTransportFailure(active, it) }
}
}
override fun seekToPreviousMediaItem() {
if (isLoadingUpnp()) {
Timber.w("ForwardingPlayer.seekToPreviousMediaItem() dropped -- UPnP loading")
return
}
val active = holder.active.value
Timber.w("ForwardingPlayer.seekToPreviousMediaItem() active=%s", active?.routeName)
if (active == null) {
super.seekToPreviousMediaItem()
return
}
// Super first for immediate local cursor advance (UI feedback);
// then delegate to Sonos Previous. PollLoop reconciles.
super.seekToPreviousMediaItem()
remoteState.beginPendingTransport(
SystemClock.elapsedRealtime() + PENDING_TRANSPORT_SAFETY_TIMEOUT_MS,
)
scope.launch {
runCatching { retryTransport { active.avTransport.previous() } }
.onFailure { handleTransportFailure(active, it) }
}
}
override fun getCurrentPosition(): Long =
if (isRemote()) remoteState.positionMs else super.getCurrentPosition()
override fun getDuration(): Long =
if (isRemote()) remoteState.durationMs else super.getDuration()
override fun isPlaying(): Boolean =
if (isRemote()) remoteState.isPlaying else super.isPlaying()
override fun getPlaybackState(): Int =
if (isRemote()) Player.STATE_READY else super.getPlaybackState()
// Mirror remote state so any consumer that gates on playWhenReady --
// notably MediaSessionService's foreground-keepalive checks and our own
// onTaskRemoved -- sees the remote renderer as the source of truth.
// Without this, swiping the app away with Sonos playing would stop the
// service, kill the poll loop, and leave Sonos orphaned.
override fun getPlayWhenReady(): Boolean =
if (isRemote()) remoteState.isPlaying else super.getPlayWhenReady()
override fun addListener(listener: Player.Listener) {
super.addListener(listener)
synchronized(externalListeners) { externalListeners.add(listener) }
}
override fun removeListener(listener: Player.Listener) {
super.removeListener(listener)
synchronized(externalListeners) { externalListeners.remove(listener) }
}
/**
* Direct-invoke the externally-registered Player.Listeners so the
* MediaSession's PlaybackState publisher (notification card, lock-screen
* scrubber, BT/AVRCP, Auto, Wear OS tile) re-reads our overridden state.
* The listeners then query isPlaying / getDuration / getCurrentPosition,
* all of which route through to remoteState while UPnP is active.
*
* Posted to the player's application looper because Player.Listener
* callbacks contract on the application thread.
*/
private fun notifyRemoteStateChanged() {
if (!isRemote()) return
val playing = remoteState.isPlaying
val trackIdx = (remoteState.trackNumber - 1).coerceAtLeast(0)
val isPlayingChanged = playing != lastNotifiedIsPlaying
val trackChanged = trackIdx != lastNotifiedTrackIdx
if (!isPlayingChanged && !trackChanged) return
lastNotifiedIsPlaying = playing
lastNotifiedTrackIdx = trackIdx
val snapshot = synchronized(externalListeners) { externalListeners.toList() }
handler.post {
for (l in snapshot) {
if (isPlayingChanged) {
l.onIsPlayingChanged(playing)
l.onPlaybackStateChanged(Player.STATE_READY)
}
if (trackChanged) {
val item = if (trackIdx < delegate.mediaItemCount) {
delegate.getMediaItemAt(trackIdx)
} else {
null
}
l.onMediaItemTransition(item, Player.MEDIA_ITEM_TRANSITION_REASON_AUTO)
}
}
}
}
override fun release() {
pollJob?.cancel()
scope.cancel()
handler.post {
ProcessLifecycleOwner.get().lifecycle.removeObserver(lifecycleObserver)
}
super.release()
}
/**
* Run a transport SOAP block, retrying transport-level (IO) failures a
* few times with backoff. A locked phone's WiFi power-save can stall the
* first socket I/O for a second or two; a retry lets the command land once
* WiFi wakes instead of being abandoned. A [SoapFaultException] (the
* renderer answered and rejected the action) is NOT retried -- the device
* is alive and retrying won't change its verdict.
*/
private suspend fun <T> retryTransport(block: suspend () -> T): T =
retryTransientIo(TRANSPORT_RETRY_ATTEMPTS, TRANSPORT_RETRY_BACKOFF_MS, block)
/**
* A transport SOAP command failed even after retries. Deliberately does
* NOT declare the route dropped: the 1 Hz poll loop is the single arbiter
* of liveness (DROP_THRESHOLD consecutive poll failures). A locked phone's
* WiFi power-save can fail one command while the renderer is perfectly
* reachable; dropping on a single command falsely kicked playback back to
* the phone. We log, leave the poll loop running, and nudge an immediate
* poll so the UI reconciles to Sonos's actual state -- if the renderer is
* truly gone, the poll loop trips the drop on its own.
*/
private fun handleTransportFailure(active: ActiveUpnp, t: Throwable) {
if (t is SoapFaultException) {
Timber.w(t, "UPnP transport rejected by %s -- device alive, no drop", active.routeName)
} else {
Timber.w(t, "UPnP transport failed on %s -- poll loop arbitrates", active.routeName)
}
pollTrigger.trySend(Unit)
}
private fun onActiveChanged(active: ActiveUpnp?) {
pollJob?.cancel()
nonPlayingPollStreak = 0
// Reset notify cache so the first poll after a route flip republishes
// playing/track state to the MediaSession even if it happens to match
// the prior session's values numerically.
lastNotifiedIsPlaying = false
lastNotifiedTrackIdx = -1
if (active != null) {
Timber.w("UPnP active: %s -- pollLoop starting", active.routeName)
// Hold the WiFi + CPU awake for the cast: the wrapped ExoPlayer is
// about to be paused (releasing its WAKE_MODE_NETWORK locks), so
// without this the radio power-saves on a locked screen and the
// poll below starves -- see [CastNetworkLock].
castNetworkLock.acquire()
// Pause the wrapped ExoPlayer so we are not playing local audio
// simultaneously with the remote renderer. handler.post targets the
// application looper, so this runs on the same thread that processes
// our override calls -- no race with the pause() override branching
// to SOAP (holder.active is already non-null by the time this post
// fires, but delegate.pause() bypasses the override entirely).
handler.post { delegate.pause() }
pollJob = scope.launch { pollLoop(active) }
} else {
castNetworkLock.release()
remoteState.reset()
}
}
@OptIn(ExperimentalCoroutinesApi::class) // onTimeout / select.onReceive
private suspend fun pollLoop(active: ActiveUpnp) {
var networkDropSuppressed = false
while (scope.isActive && holder.active.value?.routeId == active.routeId) {
val outcome = runCatching { pollOnce(active) }
if (outcome.isSuccess) {
remoteState.recordPollSuccess()
networkDropSuppressed = false
} else if (remoteState.recordPollFailure()) {
if (networkStatus.state.value == ServerHealth.Healthy) {
Timber.w("UPnP drop threshold tripped for %s", active.routeName)
handler.post { onDrop(active.routeName) }
return
}
networkDropSuppressed = suppressDropForNetwork(active, networkDropSuppressed)
}
// Race the normal cadence against any external wake (activity
// resume). Whichever wins continues to the next pollOnce.
select<Unit> {
onTimeout(POLL_INTERVAL_MS) {}
pollTrigger.onReceive {}
}
}
}
/**
* The poll-failure threshold tripped, but the phone's *own* network is down
* (state is not [ServerHealth.Healthy]) -- so this is "we can't see the
* renderer right now," not "the renderer died." A UPnP renderer streams
* autonomously and is almost certainly still playing; meanwhile a local
* fallback couldn't reach the server either, so dropping would only blast
* local audio out of a pocketed phone while the speaker keeps going. Hold
* the route and keep polling -- a successful poll once the network returns
* reconciles to the renderer's real state. Clear the failure streak so
* recovery re-evaluates the renderer from scratch instead of re-dropping on
* the first post-recovery hiccup. Returns the (latched) suppression flag so
* the rationale logs once per outage, not every tick.
*/
private fun suppressDropForNetwork(active: ActiveUpnp, alreadySuppressed: Boolean): Boolean {
if (!alreadySuppressed) {
Timber.w(
"UPnP drop suppressed for %s -- phone network %s, holding route",
active.routeName, networkStatus.state.value,
)
}
remoteState.recordPollSuccess() // clear streak; not a renderer failure
return true
}
/**
* One poll tick: read position + transport state from Sonos, apply to
* [remoteState], and forward-sync the local cursor to Sonos's Track
* index when not in queue load.
*
* Cursor sync is gated on `holder.target == null` (= not loading)
* because during load Sonos reports Track=1 while we're still
* appending, and syncing would race the SetAV+Seek that lands
* after. Outside load, forward sync catches Sonos auto-advances
* (queue end-of-track), Sonos-app driven Next presses, and any
* drift after a brief poll-failure burst that didn't trip the
* drop threshold. Forward-only because a Next override we just
* issued can race with a poll still reporting the prior Track --
* the next poll catches up safely.
*/
private suspend fun pollOnce(active: ActiveUpnp) {
val info = active.avTransport.getPositionInfo()
val now = SystemClock.elapsedRealtime()
val inSeekAckWindow = lastSeekIssuedAtMs > 0L &&
(now - lastSeekIssuedAtMs) < SEEK_ACK_WINDOW_MS
// Inside the seek-ack window, keep the optimistic position we wrote in
// seekTo -- the poll's reported position is stale until Sonos finishes
// processing the Seek SOAP. Other fields still refresh from the poll.
remoteState.applyPositionInfo(
positionMs = if (inSeekAckWindow) remoteState.positionMs else info.relTimeMs,
durationMs = info.trackDurationMs,
trackUri = info.trackUri,
trackNumber = info.track,
)
maybeSyncLocalCursor(info.track)
val transport = active.avTransport.getTransportInfo()
when (transport.state) {
TransportState.PLAYING -> {
nonPlayingPollStreak = 0
remoteState.applyTransportPlaying()
}
TransportState.PAUSED -> {
nonPlayingPollStreak += 1
if (nonPlayingPollStreak >= NON_PLAYING_CONFIRM) {
remoteState.applyTransportPaused()
}
}
TransportState.STOPPED -> {
nonPlayingPollStreak += 1
if (nonPlayingPollStreak >= NON_PLAYING_CONFIRM) {
remoteState.applyTransportStopped()
}
}
TransportState.TRANSITIONING, TransportState.UNKNOWN -> Unit
}
notifyRemoteStateChanged()
}
private fun maybeSyncLocalCursor(sonosTrack: Int) {
if (holder.target.value != null) return
if (sonosTrack <= 0) return
val sonosIdx = sonosTrack - 1
handler.post {
val localIdx = delegate.currentMediaItemIndex
if (sonosIdx > localIdx && sonosIdx < delegate.mediaItemCount) {
Timber.w(
"UPnP cursor catch-up: local=%d -> sonos=%d",
localIdx, sonosIdx,
)
delegate.seekTo(sonosIdx, 0L)
}
}
}
private companion object {
const val POLL_INTERVAL_MS = 1_000L
const val NON_PLAYING_CONFIRM = 2
const val SEEK_ACK_WINDOW_MS = 2_000L
// Safety upper bound on how long the polling tick will wait for
// Sonos to ack a user transport. The common case clears event-driven
// when Sonos's reported Track matches the wrapped player; this only
// kicks in if SOAP fails or Sonos drops the ack entirely.
const val PENDING_TRANSPORT_SAFETY_TIMEOUT_MS = 5_000L
// Transport commands retry transient IO failures so a single WiFi
// power-save stall (locked phone) doesn't abandon the command. 3
// attempts x the SoapClient's 2s connect timeout + backoff bounds the
// worst case at ~7s; a still-failing command then defers to the poll.
const val TRANSPORT_RETRY_ATTEMPTS = 3
const val TRANSPORT_RETRY_BACKOFF_MS = 400L
}
}
/**
* Retry [block] on transient transport-level ([IOException]) failures, with
* [backoffMs] between attempts, up to [attempts] total. Anything that is not
* an [IOException] -- notably [SoapFaultException], where the renderer
* answered and rejected the action -- propagates immediately: the device is
* alive, so retrying won't change its verdict. Extracted from
* [MinstrelForwardingPlayer] so a locked phone's WiFi power-save stall doesn't
* abandon a single transport command (the bug that falsely reverted Sonos
* playback to local audio).
*/
internal suspend fun <T> retryTransientIo(
attempts: Int,
backoffMs: Long,
block: suspend () -> T,
): T {
var attempt = 0
while (true) {
try {
return block()
} catch (io: IOException) {
attempt += 1
if (attempt >= attempts) throw io
Timber.w(io, "transient transport failure (attempt %d) -- retrying", attempt)
delay(backoffMs)
}
}
}
@@ -72,11 +72,12 @@ class MinstrelPlayerService : MediaSessionService() {
override fun onCreate() {
super.onCreate()
val player = playerFactory.build()
val player: Player = playerFactory.build()
val callback = LikeMediaCallback(likesRepository, serviceScope)
val session = MediaSession.Builder(this, player)
.setSessionActivity(buildNowPlayingPendingIntent())
.setCallback(callback)
.setBitmapLoader(playerFactory.buildBitmapLoader())
.setMediaButtonPreferences(ImmutableList.of(buildLikeButton(isLiked = false)))
.build()
mediaSession = session
@@ -176,7 +177,11 @@ class MinstrelPlayerService : MediaSessionService() {
override fun onTaskRemoved(rootIntent: Intent?) {
val player = mediaSession?.player ?: return super.onTaskRemoved(rootIntent)
val activelyPlaying = player.playWhenReady && player.playbackState != Player.STATE_ENDED
// player.isPlaying is overridden on MinstrelForwardingPlayer to return
// remoteState.isPlaying while UPnP is active, so a swipe-away with
// Sonos playing keeps the service (and its UPnP poll loop) alive.
val activelyPlaying = player.isPlaying ||
(player.playWhenReady && player.playbackState != Player.STATE_ENDED)
if (!activelyPlaying) {
stopSelf()
}
@@ -0,0 +1,78 @@
package com.fabledsword.minstrel.player
import androidx.media3.datasource.DataSource
import androidx.media3.datasource.DataSpec
import androidx.media3.datasource.TransferListener
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.ServerHealth
import java.io.IOException
import java.io.InterruptedIOException
/**
* DataSource wrapper that fails the network read immediately when
* [NetworkStatusController] reports a gating state. CacheDataSource only
* calls this upstream factory on cache misses, so playback of cached audio is
* unaffected -- only "tap a non-cached track while offline" hits this branch
* and gets a fast, meaningful error instead of a multi-second network timeout
* (which then surfaced as a silent decode failure to the user).
*
* Wrapping rather than substituting the OkHttp data source lets the cache
* write path remain intact for when health returns and we DO want to fetch:
* we keep the same upstream all the time, just gate `open()`.
*/
class OfflineGatedDataSource(
private val delegate: DataSource,
private val health: NetworkStatusController,
) : DataSource {
override fun open(dataSpec: DataSpec): Long {
gateOnHealth()
return try {
val opened = delegate.open(dataSpec)
health.reportSuccess() // bytes flowing from the server == reachable
opened
} catch (e: IOException) {
health.reportFailure() // real network read failed → arbitrate via /healthz
throw e
}
}
/** Fast-fail before touching the network when the server can't be reached. */
private fun gateOnHealth() {
when (health.state.value) {
ServerHealth.Offline -> throw OfflineException(
"Track not in the on-device cache and the device is offline.",
)
ServerHealth.ServerDown -> throw OfflineException(
"Track not in the on-device cache and the Minstrel server is unreachable.",
)
// Unstable is non-gating: still try the network. Healthy too.
ServerHealth.Unstable, ServerHealth.Healthy -> Unit
}
}
override fun close() = delegate.close()
override fun getUri() = delegate.uri
override fun read(buffer: ByteArray, offset: Int, length: Int): Int =
delegate.read(buffer, offset, length)
override fun addTransferListener(transferListener: TransferListener) =
delegate.addTransferListener(transferListener)
override fun getResponseHeaders() = delegate.responseHeaders
}
class OfflineGatedDataSourceFactory(
private val upstream: DataSource.Factory,
private val health: NetworkStatusController,
) : DataSource.Factory {
override fun createDataSource(): DataSource =
OfflineGatedDataSource(upstream.createDataSource(), health)
}
/**
* Signals the audio-source error path that the request was denied because the
* device is offline / the server is unreachable. ExoPlayer's [androidx.media3
* .common.PlaybackException] catches it via [InterruptedIOException]'s
* `IOException` ancestor and surfaces it as a SOURCE error, which then flows
* through the existing [PlaybackErrorReporter] -> snackbar path.
*/
class OfflineException(message: String) : IOException(message)
@@ -6,6 +6,7 @@ import androidx.lifecycle.ProcessLifecycleOwner
import com.fabledsword.minstrel.api.endpoints.EventsApi
import com.fabledsword.minstrel.auth.AuthStore
import com.fabledsword.minstrel.cache.mutations.MutationQueue
import com.fabledsword.minstrel.cache.mutations.PlayEndedPayload
import com.fabledsword.minstrel.cache.mutations.PlayOfflinePayload
import com.fabledsword.minstrel.di.ApplicationScope
import com.fabledsword.minstrel.models.wire.PlayEndedRequest
@@ -64,11 +65,24 @@ class PlayEventsReporter @Inject constructor(
private var curLastPositionMs: Long = 0
private var curDurationMs: Long = 0
// Whether the currently-tracked play is actively playing (vs paused).
// onStop reads this: a backgrounded play that's still playing is left
// to the live path (the foreground service keeps us alive), while a
// paused one is durably closed so a kill can't orphan its open row.
private var curPlaying: Boolean = false
// Server play_event_id when the live play_started succeeded;
// null if start failed/offline — then the close is captured into
// the offline mutation queue instead of a live ended/skipped.
private var openPlayEventId: String? = null
// True from beginTrack until startLive's play_started resolves. When a
// close races an in-flight start (id still null), the server row is
// being created but we've lost its id — we must NOT also enqueue an
// offline play (that would duplicate); the server auto-closes the
// orphan on the next play_started instead.
private var startInFlight: Boolean = false
private var prevTrackId: String? = null
init {
@@ -95,7 +109,7 @@ class PlayEventsReporter @Inject constructor(
) {
// Track-change → close the prior tracked play.
if (trackId != prevTrackId && curTrackId != null) {
closeCurrent(viaOffline = false)
closeCurrent()
}
// Entered playing for a new track → begin tracking it.
@@ -110,6 +124,7 @@ class PlayEventsReporter @Inject constructor(
if (curTrackId != null && trackId == curTrackId) {
curLastPositionMs = positionMs
if (durationMs > 0) curDurationMs = durationMs
curPlaying = playing
}
prevTrackId = trackId
@@ -121,12 +136,18 @@ class PlayEventsReporter @Inject constructor(
curSource = source
curLastPositionMs = 0
curDurationMs = durationMs
curPlaying = true
openPlayEventId = null
startInFlight = true
startLive(trackId = trackId, source = source)
}
private fun startLive(trackId: String, source: String?) {
val cid = resolveClientId() ?: return
val cid = resolveClientId()
if (cid == null) {
startInFlight = false
return
}
scope.launch {
try {
val response = api.playStarted(
@@ -144,47 +165,61 @@ class PlayEventsReporter @Inject constructor(
) {
// Offline / flaky — openPlayEventId stays null; close
// path will enqueue the completed play for replay.
} finally {
startInFlight = false
}
}
}
private fun closeCurrent(viaOffline: Boolean) {
/**
* Closes the tracked play. Routes by whether the server already has an
* open row, NOT by caller context — this is what keeps background /
* track-change / failure paths from each minting a second row:
* - open server row (id) → close it by id; on failure enqueue a durable
* close-by-id (never an offline play).
* - start in flight (no id yet) → leave it; the server auto-closes the
* orphan on the next play_started. Enqueuing offline here would
* duplicate the in-flight row.
* - no server row at all → record the whole play offline.
*
* Always uses the actual position played; the server's skip rule
* (50%/30s defaults) classifies was_skipped from it.
*/
private fun closeCurrent() {
val trackId = curTrackId
val startedAt = curStartedAt
if (trackId == null || startedAt == null) {
resetCurrent()
return
}
// Always send the actual position played to; the server's
// skip rule (50%/30s defaults, configurable) classifies
// was_skipped from this. If the track ran to completion the
// last observed position is ~= duration, which gives ratio ~1
// and the server records a non-skip. The COMPLETION_TOLERANCE
// rounding the prior shape did is no longer needed — the
// server's threshold is the source of truth.
val durationMs = curLastPositionMs
val source = curSource
val id = openPlayEventId
val inFlight = startInFlight
when {
id != null -> closeById(id, durationMs)
inFlight -> Unit
else -> enqueueOffline(trackId, startedAt, source, durationMs)
}
resetCurrent()
}
if (!viaOffline && id != null) {
private fun closeById(id: String, durationMs: Long) {
scope.launch {
try {
api.playEnded(
PlayEndedRequest(
playEventId = id,
durationPlayedMs = durationMs,
),
PlayEndedRequest(playEventId = id, durationPlayedMs = durationMs),
)
} catch (
@Suppress("TooGenericExceptionCaught", "SwallowedException") e: Throwable,
) {
enqueueOffline(trackId, startedAt, source, durationMs)
// Durable close-by-id (idempotent server-side) — survives a
// process kill without orphaning the open row or duplicating.
mutationQueue.enqueuePlayEnded(
PlayEndedPayload(playEventId = id, durationPlayedMs = durationMs),
)
}
}
} else {
enqueueOffline(trackId, startedAt, source, durationMs)
}
resetCurrent()
}
private fun resetCurrent() {
@@ -193,7 +228,9 @@ class PlayEventsReporter @Inject constructor(
curSource = null
curLastPositionMs = 0
curDurationMs = 0
curPlaying = false
openPlayEventId = null
startInFlight = false
}
private fun enqueueOffline(
@@ -232,8 +269,17 @@ class PlayEventsReporter @Inject constructor(
// ── Lifecycle: durable-close on app background / detach ────────
override fun onStop(owner: LifecycleOwner) {
if (curTrackId != null) {
closeCurrent(viaOffline = true)
// App backgrounded. If the tracked play is still PLAYING, leave it
// alone: the foreground media service keeps the process — and this
// app-scope uiState collector — alive, so the live state machine
// closes the play accurately on the next real track change. Closing
// here (the old always-offline behavior) raced the collector's
// re-begin and minted a duplicate, skip-misclassified row on every
// screen-lock. Only durably close a PAUSED play, where no further
// progress will happen and a process kill would otherwise orphan the
// open server row. See docs/superpowers/2026-06-11 contract audit.
if (curTrackId != null && !curPlaying) {
closeCurrent()
}
}
}
@@ -1,5 +1,6 @@
package com.fabledsword.minstrel.player
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.di.ApplicationScope
import kotlinx.coroutines.CoroutineScope
import kotlinx.coroutines.channels.Channel
@@ -40,6 +41,7 @@ private const val DEBOUNCE_MS = 2_000L
class PlaybackErrorReporter @Inject constructor(
private val playerController: PlayerController,
private val repository: PlaybackErrorRepository,
private val networkStatus: NetworkStatusController,
@ApplicationScope private val scope: CoroutineScope,
) {
private val outChannel = Channel<String>(Channel.BUFFERED)
@@ -52,6 +54,10 @@ class PlaybackErrorReporter @Inject constructor(
val buffer = mutableListOf<String>()
var debounceJob: kotlinx.coroutines.Job? = null
playerController.playbackErrorEvents.collect { event ->
// A track failing to play is ambiguous (dead server vs. one bad
// file) — let the controller arbitrate via /healthz. No-op when
// already Offline; cheap otherwise.
networkStatus.reportFailure()
// Fire-and-forget the server report — repository handles
// success/queue branching so callers don't see throws.
scope.launch { repository.report(event) }
@@ -5,6 +5,8 @@ import android.content.Context
import android.os.Bundle
import android.os.Handler
import android.os.Looper
import android.os.SystemClock
import androidx.core.net.toUri
import androidx.media3.common.MediaItem
import androidx.media3.common.MediaMetadata
import androidx.media3.common.Player
@@ -12,6 +14,8 @@ import androidx.media3.session.MediaController
import androidx.media3.session.SessionToken
import com.fabledsword.minstrel.di.ApplicationScope
import com.fabledsword.minstrel.models.TrackRef
import com.fabledsword.minstrel.playlists.data.PlaylistsRepository
import com.fabledsword.minstrel.playlists.data.toPlayableTrackRefs
import com.fabledsword.minstrel.shared.resolveServerUrl
import dagger.hilt.android.qualifiers.ApplicationContext
import kotlinx.coroutines.CoroutineScope
@@ -20,8 +24,10 @@ import kotlinx.coroutines.channels.Channel
import kotlinx.coroutines.delay
import kotlinx.coroutines.flow.Flow
import kotlinx.coroutines.flow.MutableStateFlow
import kotlinx.coroutines.flow.SharedFlow
import kotlinx.coroutines.flow.StateFlow
import kotlinx.coroutines.flow.asStateFlow
import kotlinx.coroutines.flow.combine
import kotlinx.coroutines.flow.receiveAsFlow
import kotlinx.coroutines.isActive
import kotlinx.coroutines.launch
@@ -59,7 +65,18 @@ class PlayerController @Inject constructor(
@ApplicationContext private val context: Context,
@ApplicationScope private val scope: CoroutineScope,
private val radio: RadioController,
private val playlists: PlaylistsRepository,
private val playerFactory: PlayerFactory,
private val activeUpnpHolder: com.fabledsword.minstrel.player.output.ActiveUpnpHolder,
private val remoteState: RemotePlayerState,
) {
/**
* UPnP drop events surfaced from [PlayerFactory.dropEvents]. NowPlaying
* collects this into its snackbar host so a transport / poll failure
* during UPnP playback shows "Disconnected from <name>" to the user.
*/
val dropEvents: SharedFlow<String> = playerFactory.dropEvents
private val sessionToken =
SessionToken(context, ComponentName(context, MinstrelPlayerService::class.java))
@@ -97,6 +114,13 @@ class PlayerController @Inject constructor(
*/
private var lastEvaluatedItemIndex: Int = -1
/**
* #968: self-heal budget — re-pulls of a stale refreshable queue since
* the last successful play. Capped so a still-broken refresh can't loop;
* reset when any track loads with real audio.
*/
private var selfHealAttempts = 0
/**
* Stable queue snapshot kept in sync with the player's MediaItems —
* the player's own getMediaItem(index) returns Media3 types; we keep
@@ -130,11 +154,24 @@ class PlayerController @Inject constructor(
// ── Transport (no-op until the controller is connected) ──────────────
fun play() { mediaController?.play() }
fun pause() { mediaController?.pause() }
fun seekTo(positionMs: Long) { mediaController?.seekTo(positionMs) }
fun skipToNext() { mediaController?.seekToNextMediaItem() }
fun skipToPrevious() { mediaController?.seekToPreviousMediaItem() }
// Each transport call must run on the MediaController's
// applicationLooper; calling from a background coroutine throws
// IllegalStateException (see PlayerController.setQueue's note).
// UI tap handlers are already on Main so the in-place branch hits;
// the background path only fires for cross-thread callers like
// OutputPickerController.selectUpnp (which calls pause() after
// handing playback off to a UPnP renderer).
fun play() { mediaController?.let { runOnControllerThread(it) { it.play() } } }
fun pause() { mediaController?.let { runOnControllerThread(it) { it.pause() } } }
fun seekTo(positionMs: Long) {
mediaController?.let { runOnControllerThread(it) { it.seekTo(positionMs) } }
}
fun skipToNext() {
mediaController?.let { runOnControllerThread(it) { it.seekToNextMediaItem() } }
}
fun skipToPrevious() {
mediaController?.let { runOnControllerThread(it) { it.seekToPreviousMediaItem() } }
}
/** Flip shuffle on/off. Media3 emits onEvents → uiState reflects. */
fun toggleShuffle() {
@@ -165,6 +202,20 @@ class PlayerController @Inject constructor(
controller.play()
}
/**
* Align the local cursor to [index] + [positionMs] WITHOUT auto-playing.
* Used by the UPnP adopt path: when attaching to a renderer that's already
* playing our queue, we sync the local cursor to the renderer's current
* track/position so skip/seek and the UI map onto its live session — the
* renderer keeps the audio, the local player stays paused. Unlike
* [seekToIndex] this issues no play().
*/
fun moveCursorTo(index: Int, positionMs: Long) {
val controller = mediaController ?: return
if (index !in queueRefs.indices) return
runOnControllerThread(controller) { controller.seekTo(index, positionMs) }
}
/**
* Replace the queue with [tracks] starting at [initialIndex].
* [source] tags the queue with its origin (e.g. "for_you") so the
@@ -299,10 +350,23 @@ class PlayerController @Inject constructor(
* and advance past the dead track. Otherwise no-op.
*/
private fun handleZeroDurationIfNeeded(controller: MediaController, idx: Int) {
val current = queueRefs.getOrNull(idx) ?: return
// Skip during UPnP playback (active) AND during the activation load
// window (target set, active not yet wired). ExoPlayer is intentionally
// paused throughout both windows so its STATE_READY duration is always
// 0 / TIME_UNSET -- without this guard we rapid-advance through the
// entire local queue (and spam /api/playback-errors).
val current = queueRefs.getOrNull(idx)
val upnpEngaged = activeUpnpHolder.active.value != null ||
activeUpnpHolder.target.value != null
if (current == null || upnpEngaged) return
val duration = controller.duration
val isZeroDuration = duration <= 0L || duration == androidx.media3.common.C.TIME_UNSET
if (!isZeroDuration) return
if (!isZeroDuration) {
// A track loaded with real audio — clear the self-heal budget so
// a later stale queue can recover again.
selfHealAttempts = 0
return
}
playbackErrorEventsChannel.trySend(
PlaybackErrorEvent(
trackId = current.id,
@@ -318,6 +382,58 @@ class PlayerController @Inject constructor(
}
}
/**
* #968: the queue is fully unplayable (every track failed to load). If it
* came from a refreshable source — a system playlist (bare-variant source)
* or radio ("radio:<seed>") — the snapshot is probably stale (app/tab left
* open across the daily rebuild); re-pull it and resume instead of stopping.
* Bounded by [selfHealAttempts]. Album / artist / user-playlist / offline
* sources have nothing to refresh — stop.
*/
private fun selfHealOrStop(controller: MediaController) {
val source = controller.currentMediaItem
?.mediaMetadata?.extras?.getString(MINSTREL_SOURCE_KEY)
if (source == null || selfHealAttempts >= MAX_SELF_HEAL_ATTEMPTS) {
controller.stop()
return
}
when {
source.startsWith("radio:") -> {
selfHealAttempts++
scope.launch { selfHealRadio(source.removePrefix("radio:")) }
}
!source.contains(':') -> { // bare variant = refreshable system playlist
selfHealAttempts++
scope.launch { selfHealSystem(source) }
}
else -> controller.stop()
}
}
private suspend fun selfHealSystem(variant: String) {
val refs = runCatching { playlists.systemShuffle(variant).tracks.toPlayableTrackRefs() }
.getOrDefault(emptyList())
if (refs.isEmpty()) {
stopOnControllerThread()
return
}
setQueue(refs, initialIndex = 0, source = variant)
}
private suspend fun selfHealRadio(seedTrackId: String) {
val refs = runCatching { radio.seed(seedTrackId) }.getOrDefault(emptyList())
if (refs.isEmpty()) {
stopOnControllerThread()
return
}
setQueue(refs, initialIndex = 0, source = "radio:$seedTrackId")
}
private fun stopOnControllerThread() {
val controller = mediaController ?: return
runOnControllerThread(controller) { controller.stop() }
}
// ── Internal: async connect + Listener-driven UI state sync ──────────
private suspend fun connectAndObserve() {
@@ -341,6 +457,19 @@ class PlayerController @Inject constructor(
// awaitReady, the Player.Listener is wired too.
if (!readyDeferred.isCompleted) readyDeferred.complete(Unit)
startPositionPolling(controller)
// Keep isUpnpLoading current between player-event fires: holder state
// changes (setTarget / set(active)) are independent of Media3 events, so
// onEvents alone would lag behind by up to one event cycle. This collector
// runs for the process lifetime alongside the position poller.
scope.launch {
combine(
activeUpnpHolder.target,
activeUpnpHolder.active,
) { target, active -> target != null && active == null }
.collect { isLoading ->
uiStateInternal.value = uiStateInternal.value.copy(isUpnpLoading = isLoading)
}
}
controller.addListener(
object : Player.Listener {
override fun onPlayerError(error: androidx.media3.common.PlaybackException) {
@@ -355,6 +484,17 @@ class PlayerController @Inject constructor(
detail = error.message,
),
)
// A failed load leaves the player IDLE; advance past the bad
// track and re-prepare so one unplayable item doesn't strand
// playback (mirrors the zero_duration skip). At the end of a
// fully-unplayable queue, try to self-heal a stale refreshable
// source before stopping.
if (controller.hasNextMediaItem()) {
controller.seekToNextMediaItem()
controller.prepare()
} else {
selfHealOrStop(controller)
}
}
override fun onMediaItemTransition(
@@ -364,6 +504,11 @@ class PlayerController @Inject constructor(
// Reset the per-item evaluation guard so the new
// item's STATE_READY transition gets a fresh check.
lastEvaluatedItemIndex = -1
// The track flipped -- re-anchor the position interpolator
// so the next polling tick treats the new track's
// remoteState.positionMs as fresh rather than carrying the
// old anchor + elapsed forward.
lastSeenRemotePositionMs = -1L
}
override fun onPlaybackStateChanged(playbackState: Int) {
@@ -381,15 +526,38 @@ class PlayerController @Inject constructor(
?.mediaMetadata
?.extras
?.getString(MINSTREL_SOURCE_KEY)
val isUpnpLoading = activeUpnpHolder.target.value != null &&
activeUpnpHolder.active.value == null
// When UPnP is active, the wrapped ExoPlayer is paused with
// no real audio loaded -- player.duration / isPlaying /
// currentPosition all reflect that. Read from remoteState
// instead so an onEvents fire (e.g. activity resume) doesn't
// clobber the UI with zeros. Duration falls back to the
// wrapped player's value when Sonos hasn't reported one yet
// (pre-first-poll window, or Sonos still buffering) -- the
// wrapped ExoPlayer was prepared with the same MediaItem so
// it knows the real duration before any SOAP poll lands.
val upnpActive = activeUpnpHolder.active.value != null
uiStateInternal.value =
PlayerUiState(
currentTrack = current,
queue = queueRefs,
queueIndex = idx,
isPlaying = player.isPlaying,
isBuffering = player.playbackState == Player.STATE_BUFFERING,
positionMs = player.currentPosition.coerceAtLeast(0),
durationMs = player.duration.coerceAtLeast(0),
isPlaying = if (upnpActive) remoteState.isPlaying else player.isPlaying,
isBuffering = !upnpActive &&
player.playbackState == Player.STATE_BUFFERING,
positionMs = if (upnpActive) {
remoteState.positionMs
} else {
player.currentPosition
}.coerceAtLeast(0),
durationMs = effectiveDuration(
upnpActive,
remoteState.durationMs,
player.duration,
desiredIdx = idx,
controllerIdx = idx,
),
bufferedPositionMs = player.bufferedPosition.coerceAtLeast(0),
playbackError = player.playerError?.message,
currentSource = source,
@@ -399,6 +567,7 @@ class PlayerController @Inject constructor(
Player.REPEAT_MODE_ONE -> RepeatMode.ONE
else -> RepeatMode.OFF
},
isUpnpLoading = isUpnpLoading,
)
}
},
@@ -423,15 +592,138 @@ class PlayerController @Inject constructor(
scope.launch(Dispatchers.Main.immediate) {
while (isActive) {
delay(POSITION_POLL_INTERVAL_MS)
if (!controller.isPlaying) continue
uiStateInternal.value = uiStateInternal.value.copy(
positionMs = controller.currentPosition.coerceAtLeast(0),
bufferedPositionMs = controller.bufferedPosition.coerceAtLeast(0),
tickPositionPoll(controller)
}
}
}
/**
* One position-polling tick. Owns track-change detection too: when UPnP
* is active and the wrapped ExoPlayer is paused, `delegate.seekTo` from
* `maybeSyncLocalCursor` may not fire `onMediaItemTransition`, leaving
* uiState.queueIndex stuck on the old track even after Sonos has
* advanced. So the tick reads Sonos's reported Track as the source of
* truth, rebuilds the index/title fields itself, and force-syncs the
* wrapped player as defense in depth.
*/
private fun tickPositionPoll(controller: MediaController) {
val upnpActive = activeUpnpHolder.active.value != null
resolvePendingTransport(controller, upnpActive)
val pendingTransport = upnpActive && remoteState.pendingTransportDeadlineMs > 0L
val effectiveIsPlaying =
if (upnpActive) remoteState.isPlaying else controller.isPlaying
val effectivePosition = if (upnpActive) {
interpolatedRemotePosition(effectiveIsPlaying)
} else {
controller.currentPosition
}
val desiredIdx = desiredQueueIndex(controller, upnpActive)
val current = uiStateInternal.value
val newPos = effectivePosition.coerceAtLeast(0)
val newDur = effectiveDuration(
upnpActive,
remoteState.durationMs,
controller.duration,
desiredIdx = desiredIdx,
controllerIdx = controller.currentMediaItemIndex,
)
val newBuf = controller.bufferedPosition.coerceAtLeast(0)
// Track adjustments are forward-only AND suppressed while a user
// transport press is pending Sonos confirmation. Together those keep
// either direction of user input from being undone by a stale poll.
val trackChanged = !pendingTransport &&
desiredIdx > current.queueIndex &&
desiredIdx in queueRefs.indices
publishTickIfChanged(
current, trackChanged, desiredIdx,
effectiveIsPlaying, newPos, newDur, newBuf,
)
}
/**
* Event-driven primary path: clear pending the moment Sonos's reported
* Track matches the wrapped player's index. Safety fallback: clear on
* deadline so we don't ignore Sonos's actual state forever if SOAP fails.
*/
private fun resolvePendingTransport(controller: MediaController, upnpActive: Boolean) {
if (!upnpActive || remoteState.pendingTransportDeadlineMs <= 0L) return
val sonosIdx = (remoteState.trackNumber - 1).coerceAtLeast(0)
val timedOut = SystemClock.elapsedRealtime() > remoteState.pendingTransportDeadlineMs
if (sonosIdx == controller.currentMediaItemIndex || timedOut) {
remoteState.clearPendingTransport()
}
}
@Suppress("LongParameterList") // assembled at one tick call site; refactor would cost clarity
private fun publishTickIfChanged(
current: PlayerUiState,
trackChanged: Boolean,
desiredIdx: Int,
effectiveIsPlaying: Boolean,
newPos: Long,
newDur: Long,
newBuf: Long,
) {
val somethingChanged = trackChanged ||
current.isPlaying != effectiveIsPlaying ||
current.positionMs != newPos ||
current.durationMs != newDur
if (!somethingChanged) return
val newTrack = if (trackChanged) queueRefs[desiredIdx] else current.currentTrack
val newIdx = if (trackChanged) desiredIdx else current.queueIndex
uiStateInternal.value = current.copy(
currentTrack = newTrack,
queueIndex = newIdx,
isPlaying = effectiveIsPlaying,
positionMs = newPos,
durationMs = newDur,
bufferedPositionMs = newBuf,
)
// Intentionally do NOT call controller.seekTo here. That would route
// through MinstrelForwardingPlayer's seekTo override and re-issue
// AVTransport.SeekToTrack to Sonos -- which seeks Sonos back to the
// start of the same track it's already playing, restarting the song.
// The wrapped player's index is kept in sync by maybeSyncLocalCursor's
// delegate.seekTo (which bypasses the override). If it lags briefly,
// the next pollOnce catches up; the uiState above already reflects
// Sonos's truth for the user.
}
private fun desiredQueueIndex(controller: MediaController, upnpActive: Boolean): Int =
if (upnpActive) {
(remoteState.trackNumber - 1).coerceAtLeast(0)
} else {
controller.currentMediaItemIndex
}
// ── Remote position interpolation state ──────────────────────────────
// remoteState.positionMs is only refreshed by ForwardingPlayer's 1Hz
// SOAP poll (and only when the round-trip completes -- screen-off WiFi
// sleep can stall it for many seconds). To keep the scrubber moving
// smoothly we anchor each fresh reading + an elapsed-realtime stamp;
// between updates we display anchor + elapsed when Sonos is playing.
// A real correction lands as soon as the next poll arrives.
@Volatile private var lastSeenRemotePositionMs: Long = -1L
@Volatile private var positionAnchorMs: Long = 0L
@Volatile private var positionAnchorAtRealtimeMs: Long = 0L
private fun interpolatedRemotePosition(isPlaying: Boolean): Long {
val raw = remoteState.positionMs
val now = SystemClock.elapsedRealtime()
if (raw != lastSeenRemotePositionMs) {
lastSeenRemotePositionMs = raw
positionAnchorMs = raw
positionAnchorAtRealtimeMs = now
}
if (!isPlaying) return positionAnchorMs
// Cap how far past the last anchor we extrapolate. After
// MAX_INTERPOLATION_DRIFT_MS without a poll update, freeze the
// displayed position at anchor + cap rather than projecting wildly.
// The next successful poll re-anchors and motion resumes.
val delta = (now - positionAnchorAtRealtimeMs).coerceAtMost(MAX_INTERPOLATION_DRIFT_MS)
return positionAnchorMs + delta
}
/**
* Bridges Media3's `ListenableFuture<MediaController>.buildAsync()`
* to a suspend function without pulling in `kotlinx-coroutines-guava`
@@ -467,7 +759,23 @@ class PlayerController @Inject constructor(
.setArtist(artistName)
.setAlbumTitle(albumTitle)
.apply {
// Server-known duration -- gives the lock-screen / notification
// scrubber a real total even when the wrapped ExoPlayer is
// paused under UPnP (it never probes a duration in that state).
if (durationSec > 0) setDurationMs(durationSec.toLong() * MS_PER_SECOND)
if (source != null) setExtras(sourceExtras(source))
// Point the notification / lock-screen art at the SAME album
// cover the in-app surfaces use (TrackRef.coverUrl ->
// /api/albums/{id}/cover). Without this, Media3 falls back to
// whatever art is embedded in the stream's tags, which can be a
// different image than the server's album cover. Setting
// artworkUri here is load-bearing: MediaMetadata.populate()
// overwrites artworkUri + artworkData as a pair, so the
// MediaItem's URI clears any embedded artworkData ExoPlayer
// extracts from the stream -- the cover endpoint wins on both
// surfaces. The session's OkHttp-backed BitmapLoader (see
// PlayerFactory) is what makes this authed placeholder URL load.
if (coverUrl.isNotEmpty()) setArtworkUri(coverUrl.toUri())
}
.build()
// Server's stream_url is a relative path (/api/tracks/{id}/stream);
@@ -489,8 +797,40 @@ class PlayerController @Inject constructor(
private fun sourceExtras(source: String): Bundle =
Bundle().apply { putString(MINSTREL_SOURCE_KEY, source) }
/**
* Duration to surface to the UI. Priority: Sonos's reported duration
* (only when UPnP active and non-zero) -> wrapped ExoPlayer's value
* (only valid once it's probed the stream) -> TrackRef.durationSec
* (always known from the server response). The third tier is what
* keeps the scrubber populated when the user taps Sonos before the
* wrapped player has had time to probe its own duration -- without
* it, both top tiers report 0/TIME_UNSET and the field reads empty
* until the first SOAP poll lands.
*/
@Suppress("ReturnCount") // 3-tier fallback reads cleanest as a ladder of early returns
private fun effectiveDuration(
upnpActive: Boolean,
remoteMs: Long,
localMs: Long,
desiredIdx: Int,
controllerIdx: Int,
): Long {
if (upnpActive && remoteMs > 0) return remoteMs
// Tier 2 (wrapped player's probed duration) only valid when the
// wrapped player is on the same track we're trying to show. After a
// Sonos natural advance the polling tick updates desiredIdx from
// Sonos's truth while controllerIdx is briefly stale -- using the
// wrapped player's duration here would surface the old track's
// length under the new track's title.
if (localMs > 0 && controllerIdx == desiredIdx) return localMs
val ref = queueRefs.getOrNull(desiredIdx) ?: return 0
return ref.durationSec.toLong() * MS_PER_SECOND
}
companion object {
const val MINSTREL_SOURCE_KEY: String = "minstrel_source"
private const val MS_PER_SECOND = 1_000L
private const val MAX_INTERPOLATION_DRIFT_MS = 5_000L
}
}
@@ -501,6 +841,10 @@ class PlayerController @Inject constructor(
// they don't need this poll.
private const val POSITION_POLL_INTERVAL_MS = 500L
// #968: at most one stale-queue self-heal re-pull per exhaustion (reset on
// the next successful play) so a still-broken refresh can't loop.
private const val MAX_SELF_HEAL_ATTEMPTS = 1
/**
* Structured playback failure event for [PlaybackErrorReporter]. Drives
* both the user-facing snackbar ("Couldn't play X — skipping") and the
@@ -3,7 +3,10 @@ package com.fabledsword.minstrel.player
import android.content.Context
import androidx.media3.common.AudioAttributes
import androidx.media3.common.C
import androidx.media3.common.Player
import androidx.media3.common.util.BitmapLoader
import androidx.media3.database.StandaloneDatabaseProvider
import androidx.media3.datasource.DataSourceBitmapLoader
import androidx.media3.datasource.cache.CacheDataSink
import androidx.media3.datasource.cache.CacheDataSource
import androidx.media3.datasource.cache.LeastRecentlyUsedCacheEvictor
@@ -11,15 +14,21 @@ import androidx.media3.datasource.cache.SimpleCache
import androidx.media3.datasource.okhttp.OkHttpDataSource
import androidx.media3.exoplayer.ExoPlayer
import androidx.media3.exoplayer.source.DefaultMediaSourceFactory
import androidx.media3.session.CacheBitmapLoader
import com.fabledsword.minstrel.cache.audiocache.CacheConfig
import com.fabledsword.minstrel.player.output.ActiveUpnpHolder
import dagger.hilt.android.qualifiers.ApplicationContext
import kotlinx.coroutines.channels.BufferOverflow
import kotlinx.coroutines.flow.MutableSharedFlow
import kotlinx.coroutines.flow.SharedFlow
import kotlinx.coroutines.flow.asSharedFlow
import okhttp3.OkHttpClient
import java.io.File
import javax.inject.Inject
import javax.inject.Singleton
/**
* Builds the process-singleton ExoPlayer with our shared OkHttp +
* Builds the process-singleton player with our shared OkHttp +
* SimpleCache chain. The MinstrelPlayerService (Phase 6.2) calls
* `build()` once during onCreate.
*
@@ -31,12 +40,22 @@ import javax.inject.Singleton
* (sizeBytes cap = rollingCap); our policy layer in the worker layers
* the 2-bucket protection on top by feeding `removeSpan` only for
* unprotected tracks.
*
* `build()` returns a [MinstrelForwardingPlayer] wrapping the internal
* ExoPlayer. When the UPnP route drops (3 consecutive poll failures or
* a SOAP failure), [dropEvents] emits the route name so the NowPlaying
* surface can show a snackbar. The MutableSharedFlow uses DROP_OLDEST
* with capacity=1 so a burst of failures during a single tear-down
* surfaces as one event rather than queueing N.
*/
@Singleton
class PlayerFactory @Inject constructor(
@ApplicationContext private val context: Context,
private val okHttpClient: OkHttpClient,
private val cacheConfig: CacheConfig,
private val activeUpnpHolder: ActiveUpnpHolder,
private val remoteState: RemotePlayerState,
private val serverHealth: com.fabledsword.minstrel.connectivity.NetworkStatusController,
) {
private val cacheDir: File = File(context.cacheDir, "audio_cache").apply { mkdirs() }
@@ -46,11 +65,38 @@ class PlayerFactory @Inject constructor(
StandaloneDatabaseProvider(context),
)
fun build(): ExoPlayer {
// MutableSharedFlow with extraBufferCapacity=1 + DROP_OLDEST so a burst
// of drop events (rapid SOAP failures during a single tear-down) surfaces
// as one snackbar rather than queueing N.
private val dropEventsInternal = MutableSharedFlow<String>(
replay = 0,
extraBufferCapacity = 1,
onBufferOverflow = BufferOverflow.DROP_OLDEST,
)
val dropEvents: SharedFlow<String> = dropEventsInternal.asSharedFlow()
fun build(): Player {
val exo = buildExoPlayer()
return MinstrelForwardingPlayer(
delegate = exo,
holder = activeUpnpHolder,
remoteState = remoteState,
castNetworkLock = CastNetworkLock(context),
networkStatus = serverHealth,
onDrop = { name -> emitDrop(name) },
)
}
private fun buildExoPlayer(): ExoPlayer {
val httpDataSource = OkHttpDataSource.Factory(okHttpClient)
// Gate network reads on ServerHealth so a cache miss while offline
// fails fast with an OfflineException instead of hitting an OkHttp
// timeout. CacheDataSource only consults the upstream factory on a
// cache miss, so playback of cached audio is unaffected.
val gatedUpstream = OfflineGatedDataSourceFactory(httpDataSource, serverHealth)
val cacheDataSource = CacheDataSource.Factory()
.setCache(simpleCache)
.setUpstreamDataSourceFactory(httpDataSource)
.setUpstreamDataSourceFactory(gatedUpstream)
.setCacheWriteDataSinkFactory(
CacheDataSink.Factory()
.setCache(simpleCache)
@@ -71,4 +117,26 @@ class PlayerFactory @Inject constructor(
.setHandleAudioBecomingNoisy(true)
.build()
}
/**
* BitmapLoader for the MediaSession's notification / lock-screen art.
* Backed by the shared [okHttpClient] so it inherits the same
* BaseUrlInterceptor placeholder rewrite + auth cookie that Coil uses
* for in-app covers — without it the default DefaultHttpDataSource
* loader can't resolve `http://placeholder.invalid/...` and would 401
* on the cover endpoint. Wrapped in CacheBitmapLoader so a cover the
* notification already fetched isn't re-loaded on every metadata
* refresh. Lets the album-cover artworkUri set in
* [PlayerController.toMediaItem] actually render on the media card.
*/
fun buildBitmapLoader(): BitmapLoader =
CacheBitmapLoader(
DataSourceBitmapLoader.Builder(context)
.setDataSourceFactory(OkHttpDataSource.Factory(okHttpClient))
.build(),
)
private fun emitDrop(routeName: String) {
dropEventsInternal.tryEmit(routeName)
}
}
@@ -28,4 +28,6 @@ data class PlayerUiState(
val currentSource: String? = null,
val shuffleEnabled: Boolean = false,
val repeatMode: RepeatMode = RepeatMode.OFF,
/** True while the UPnP initial-batch load is in progress (target set, active not yet wired). */
val isUpnpLoading: Boolean = false,
)
@@ -0,0 +1,110 @@
package com.fabledsword.minstrel.player
import javax.inject.Inject
import javax.inject.Singleton
/**
* Synthesized state for the UPnP route -- what the ForwardingPlayer
* exposes via Player.getCurrentPosition / isPlaying / etc. when the
* remote leg is active. Not a Player; a container.
*
* Updates flow in from:
* - 1Hz GetPositionInfo poll -> applyPositionInfo()
* - Transport SOAP calls landing 200 OK -> applyTransport{Playing,Paused,Stopped}()
* - Poll failures -> recordPollFailure()
*
* The poll-failure counter is the sole drop arbiter: [DROP_THRESHOLD]
* consecutive poll failures = remote considered dropped (returns true
* from recordPollFailure for the caller to surface). Any success resets it.
* Failed transport commands no longer drop the route -- they retry and
* otherwise defer to this counter (see MinstrelForwardingPlayer).
*/
@Singleton
class RemotePlayerState @Inject constructor() {
@Volatile var positionMs: Long = 0L; private set
@Volatile var durationMs: Long = 0L; private set
@Volatile var isPlaying: Boolean = false; private set
@Volatile var currentTrackUri: String = ""; private set
@Volatile var lastError: Throwable? = null; private set
@Volatile var trackNumber: Int = 0; private set
// The operator's last explicit transport intent (play=true, pause=false),
// distinct from the observed remote isPlaying. handleRemoteDrop resumes
// local playback based on THIS, not isPlaying -- so a play() that fails
// its SOAP (stale/dead renderer) still resumes on the phone instead of
// swallowing the tap. Deliberately NOT cleared by applyError.
@Volatile var lastPlayIntent: Boolean = false; private set
// Pending-transport deadline (SystemClock.elapsedRealtime() at which we
// give up waiting). When > 0, a user transport action (next/prev/seekTo
// idx) is in flight: ForwardingPlayer has already moved the wrapped
// player's index, but Sonos's reported Track hasn't refreshed via a
// SOAP poll yet. PlayerController.tickPositionPoll skips track
// adjustments while pending is non-zero. Pending clears when:
// (a) [event-driven, primary] a poll lands and Sonos's reported Track
// matches the wrapped player's currentMediaItemIndex; or
// (b) [safety fallback] the deadline expires (covers SOAP-fail cases
// where Sonos never acks).
@Volatile var pendingTransportDeadlineMs: Long = 0L; private set
fun beginPendingTransport(deadlineMs: Long) {
pendingTransportDeadlineMs = deadlineMs
}
fun clearPendingTransport() {
pendingTransportDeadlineMs = 0L
}
@Volatile private var consecutivePollFailures: Int = 0
fun applyPositionInfo(positionMs: Long, durationMs: Long, trackUri: String, trackNumber: Int) {
this.positionMs = positionMs
this.durationMs = durationMs
this.currentTrackUri = trackUri
this.trackNumber = trackNumber
}
fun applyTransportPlaying() { isPlaying = true }
fun applyTransportPaused() { isPlaying = false }
fun applyTransportStopped() {
isPlaying = false
positionMs = 0L
}
fun setPlayIntent(intended: Boolean) { lastPlayIntent = intended }
fun applyError(t: Throwable) {
isPlaying = false
lastError = t
}
/** Returns true when the rolling threshold trips this call. */
fun recordPollFailure(): Boolean {
consecutivePollFailures += 1
return consecutivePollFailures >= DROP_THRESHOLD
}
fun recordPollSuccess() { consecutivePollFailures = 0 }
fun reset() {
positionMs = 0L
durationMs = 0L
isPlaying = false
lastPlayIntent = false
currentTrackUri = ""
lastError = null
consecutivePollFailures = 0
trackNumber = 0
pendingTransportDeadlineMs = 0L
}
private companion object {
// ~30 seconds of consecutive poll failures before declaring the route
// dropped. Bumped from 3 because screen-off WiFi sleep / brief Doze
// can stall socket I/O for several seconds without the renderer
// actually being unreachable -- a 3-failure drop kicked us back to
// local audio every time the phone went into a pocket.
const val DROP_THRESHOLD = 30
}
}
@@ -0,0 +1,24 @@
package com.fabledsword.minstrel.player
import com.fabledsword.minstrel.api.endpoints.CastApi
import com.fabledsword.minstrel.api.endpoints.StreamTokenRequest
import com.fabledsword.minstrel.api.endpoints.StreamTokenResponse
import retrofit2.Retrofit
import retrofit2.create
import javax.inject.Inject
import javax.inject.Singleton
/**
* Mints stream tokens for a given track id. Pulled into its own Hilt
* singleton so [MinstrelForwardingPlayer] (service-side) and
* [com.fabledsword.minstrel.player.output.OutputPickerController]
* (controller-side) don't each construct their own [CastApi] from
* Retrofit. The shared Retrofit instance is unchanged.
*/
@Singleton
class StreamTokenProvider @Inject constructor(retrofit: Retrofit) {
private val api: CastApi = retrofit.create()
suspend fun mint(trackId: String): StreamTokenResponse =
api.streamToken(StreamTokenRequest(trackId = trackId))
}
@@ -0,0 +1,43 @@
package com.fabledsword.minstrel.player.output
import com.fabledsword.minstrel.player.output.upnp.AVTransportClient
import com.fabledsword.minstrel.player.output.upnp.RenderingControlClient
import kotlinx.coroutines.flow.MutableStateFlow
import kotlinx.coroutines.flow.StateFlow
import kotlinx.coroutines.flow.asStateFlow
import javax.inject.Inject
import javax.inject.Singleton
/**
* Shared singleton handle to the currently-active UPnP route's transport
* + rendering clients. Decouples [com.fabledsword.minstrel.player.MinstrelForwardingPlayer]
* from [OutputPickerController] -- the picker writes; the forwarding
* player reads. Null = no UPnP active (local ExoPlayer path).
*/
data class ActiveUpnp(
val routeId: String,
val routeName: String,
val avTransport: AVTransportClient,
val rendering: RenderingControlClient?,
)
@Singleton
class ActiveUpnpHolder @Inject constructor() {
private val internal = MutableStateFlow<ActiveUpnp?>(null)
val active: StateFlow<ActiveUpnp?> = internal.asStateFlow()
/**
* Pending route id during selectUpnp's queue-load window. Set when
* loadQueueOnSonos starts; cleared on completion (success or failure).
* ForwardingPlayer overrides treat (target != null && active == null)
* as "UPnP intended but SOAP not yet wired" -- drop transport commands
* silently rather than send them to a half-loaded queue.
*/
private val targetInternal = MutableStateFlow<String?>(null)
val target: StateFlow<String?> = targetInternal.asStateFlow()
fun set(active: ActiveUpnp?) { internal.value = active }
fun setTarget(routeId: String?) { targetInternal.value = routeId }
}
@@ -0,0 +1,21 @@
package com.fabledsword.minstrel.player.output
/** Action the idle-revert collector should take on a player-state change. */
internal enum class IdleRevertAction { ARM, CANCEL, IGNORE }
/**
* Pure decision for the UPnP idle-revert timer. [armed] is whether the
* idle timer Job is currently running. We ARM only on the transition INTO
* "engaged + not playing" (so repeated paused emissions don't reset the
* countdown), CANCEL whenever playback resumes or UPnP disengages, and
* otherwise IGNORE.
*/
internal fun idleRevertAction(
upnpEngaged: Boolean,
isPlaying: Boolean,
armed: Boolean,
): IdleRevertAction = when {
upnpEngaged && !isPlaying -> if (armed) IdleRevertAction.IGNORE else IdleRevertAction.ARM
armed -> IdleRevertAction.CANCEL
else -> IdleRevertAction.IGNORE
}
@@ -1,24 +1,37 @@
@file:Suppress("TooManyFunctions") // 5 MediaRouter.Callback overrides inflate the count
package com.fabledsword.minstrel.player.output
import android.content.Context
import androidx.mediarouter.media.MediaControlIntent
import androidx.mediarouter.media.MediaRouteSelector
import androidx.mediarouter.media.MediaRouter
import com.fabledsword.minstrel.api.endpoints.CastApi
import com.fabledsword.minstrel.api.endpoints.StreamTokenRequest
import com.fabledsword.minstrel.di.ApplicationScope
import com.fabledsword.minstrel.models.TrackRef
import com.fabledsword.minstrel.player.PlayerController
import com.fabledsword.minstrel.player.PlayerFactory
import com.fabledsword.minstrel.player.RemotePlayerState
import com.fabledsword.minstrel.player.StreamTokenProvider
import com.fabledsword.minstrel.player.output.upnp.AVTransportClient
import com.fabledsword.minstrel.player.output.upnp.RenderingControlClient
import com.fabledsword.minstrel.player.output.upnp.SoapClient
import com.fabledsword.minstrel.player.output.upnp.SoapFaultException
import com.fabledsword.minstrel.player.output.upnp.TransportState
import com.fabledsword.minstrel.player.output.upnp.UpnpDiscoveryController
import com.fabledsword.minstrel.player.output.upnp.bareUdn
import dagger.hilt.android.qualifiers.ApplicationContext
import kotlinx.coroutines.CoroutineScope
import kotlinx.coroutines.Job
import kotlinx.coroutines.delay
import kotlinx.coroutines.flow.MutableStateFlow
import kotlinx.coroutines.flow.SharingStarted
import kotlinx.coroutines.flow.StateFlow
import kotlinx.coroutines.flow.combine
import kotlinx.coroutines.flow.stateIn
import kotlinx.coroutines.launch
import retrofit2.Retrofit
import retrofit2.create
import kotlinx.coroutines.sync.Mutex
import kotlinx.coroutines.sync.withLock
import okhttp3.HttpUrl.Companion.toHttpUrlOrNull
import okhttp3.OkHttpClient
import timber.log.Timber
import javax.inject.Inject
import javax.inject.Singleton
@@ -26,9 +39,9 @@ import javax.inject.Singleton
/**
* Snapshot of the audio output route state. [current] is the live
* route audio is being delivered to. [available] is every route the
* picker knows about MediaRouter system routes merged with
* UPnP/DLNA renderers discovered on the LAN — sorted current-first
* then by [OutputRoute.Kind] (Bluetooth, Wired, BuiltIn, Other).
* picker knows about -- MediaRouter system routes merged with
* UPnP/DLNA renderers discovered on the LAN -- with the BuiltIn
* "Phone speaker" pinned first, everything else alphabetical.
*/
data class RouteSnapshot(
val current: OutputRoute,
@@ -48,7 +61,7 @@ data class RouteSnapshot(
* - [OutputRoute.Protocol.SYSTEM] — MediaRouter.selectRoute (built-in,
* wired, Bluetooth)
* - [OutputRoute.Protocol.UPNP] — mint a signed stream token via
* [CastApi.streamToken], drive the discovered renderer with
* [StreamTokenProvider.mint], drive the discovered renderer with
* AVTransport.SetAVTransportURI + Play, pause local playback so
* audio yields to the network speaker
* - [OutputRoute.Protocol.CAST] / [OutputRoute.Protocol.SONOS] —
@@ -65,10 +78,12 @@ class OutputPickerController @Inject constructor(
@ApplicationScope private val scope: CoroutineScope,
private val upnpDiscovery: UpnpDiscoveryController,
private val playerController: PlayerController,
retrofit: Retrofit,
private val playerFactory: PlayerFactory,
private val streamTokens: StreamTokenProvider,
private val activeUpnpHolder: ActiveUpnpHolder,
private val remoteState: RemotePlayerState,
private val okHttp: OkHttpClient,
) {
private val castApi: CastApi = retrofit.create()
private val mediaRouter = MediaRouter.getInstance(context)
private val selector = MediaRouteSelector.Builder()
@@ -82,12 +97,26 @@ class OutputPickerController @Inject constructor(
*/
private val systemRoutesInternal = MutableStateFlow(snapshotFromRouter())
private val selectedUpnpRouteIdInternal = MutableStateFlow<String?>(null)
private val selectUpnpMutex = Mutex()
val routesState: StateFlow<RouteSnapshot> = combine(
systemRoutesInternal,
upnpDiscovery.routes,
) { sys, upnp ->
val merged = sys.available + upnp.map { OutputRoute.fromUpnpRoute(it) }
RouteSnapshot(current = sys.current, available = sortRoutes(sys.current, merged))
upnpDiscovery.sonosTopology,
selectedUpnpRouteIdInternal,
) { sys, upnp, _, upnpSelected ->
val suppressed = upnpDiscovery.nonCoordinatorMemberUdns()
val visibleUpnp = upnp
.filter { it.id.bareUdn() !in suppressed } // suppressed set is bare UDNs
.map { OutputRoute.fromUpnpRoute(it) }
val merged = sys.available + visibleUpnp
val current = if (upnpSelected != null) {
merged.firstOrNull { it.id == upnpSelected } ?: sys.current
} else {
sys.current
}
RouteSnapshot(current = current, available = sortRoutes(merged))
}.stateIn(scope, SharingStarted.Eagerly, systemRoutesInternal.value)
private val callback = object : MediaRouter.Callback() {
@@ -113,6 +142,20 @@ class OutputPickerController @Inject constructor(
) = refresh()
}
// Last-synced queue identity. Used by observeQueueChangesForSonosResync
// to detect when the user has mutated the queue (full replacement,
// playNext insert, or radio-append) while UPnP is active and apply the
// minimum-incremental set of Sonos SOAP operations to bring its native
// queue back in sync. Stored as the full id list (not a join-key) so we
// can run the longest-common-prefix / common-suffix diff.
private var lastSyncedQueueIds: List<String>? = null
// Idle-revert timer. Armed when playback is paused/stopped on a UPnP
// route; fires revertToPhoneOnIdle after IDLE_REVERT_MS of continuous
// non-playing. Cancelled on resume, route change, or explicit disconnect.
// See idleRevertAction for the arm/cancel decision.
private var idleRevertJob: Job? = null
init {
// Two-arg addCallback registers with no discovery flag —
// androidx.mediarouter 1.7.0's default passive behavior:
@@ -120,6 +163,309 @@ class OutputPickerController @Inject constructor(
// without forcing Bluetooth scans. (There is no
// CALLBACK_FLAG_PASSIVE_DISCOVERY constant; absent flag = passive.)
mediaRouter.addCallback(selector, callback)
// When MinstrelForwardingPlayer reports the active UPnP route has
// dropped (the poll loop's consecutive-failure threshold tripping --
// the sole drop arbiter; a failed transport command no longer drops),
// clear the UPnP selection state and fall back to local ExoPlayer at
// the last-known remote position. This mirrors selectSystem's disconnect
// path but skips the Stop SOAP since the device is already unreachable.
scope.launch {
playerFactory.dropEvents.collect { handleRemoteDrop() }
}
scope.launch { observeQueueChangesForSonosResync() }
scope.launch { observeIdleRevertWhileUpnp() }
scope.launch { observeSelectedRouteDisappearance() }
}
/**
* Revert the UPnP selection to the phone when the selected renderer leaves
* discovery (powered off / off the LAN) while we are NOT actively casting
* to it. Without this the selection could pin to a ghost route, so a later
* "play" would target a speaker that's gone. An *active* cast is left alone
* -- its own poll loop arbitrates liveness and won't drop on a mere
* discovery gap. Routes leave discovery via [UpnpDiscoveryController]'s
* post-picker-scan prune; this reacts to the resulting routes change.
*/
private suspend fun observeSelectedRouteDisappearance() {
upnpDiscovery.routes.collect { routes ->
val selected = selectedUpnpRouteIdInternal.value
val present = routes.any { it.id == selected }
val casting = activeUpnpHolder.active.value?.routeId == selected
if (selected != null && !present && !casting) {
Timber.w(
"Output: selected UPnP route %s left discovery; reverting to phone",
selected,
)
selectedUpnpRouteIdInternal.value = null
}
}
}
/**
* Arm/cancel the idle-revert timer as playback state changes while a
* UPnP route is engaged. "Engaged" = active OR target set, so the brief
* selectUpnp/resync window (active momentarily null, target set) doesn't
* spuriously fire. The arm/cancel choice is the pure idleRevertAction.
*/
private suspend fun observeIdleRevertWhileUpnp() {
playerController.uiState.collect { state ->
val engaged = activeUpnpHolder.active.value != null ||
activeUpnpHolder.target.value != null
val armed = idleRevertJob?.isActive == true
when (idleRevertAction(engaged, state.isPlaying, armed)) {
IdleRevertAction.ARM ->
idleRevertJob = scope.launch {
delay(IDLE_REVERT_MS)
revertToPhoneOnIdle()
}
IdleRevertAction.CANCEL -> {
idleRevertJob?.cancel()
idleRevertJob = null
}
IdleRevertAction.IGNORE -> Unit
}
}
}
/**
* Timer fired: the operator paused on a UPnP route and never resumed.
* Stop the renderer (best-effort -- it's reachable, unlike a drop),
* clear the UPnP overlay so routesState falls back to the phone, and
* seek local ExoPlayer to the last remote position. No play(): the
* operator was paused, so we revert routing silently. Guarded by the
* selectUpnpMutex so it can't race selectUpnp/resync.
*/
private suspend fun revertToPhoneOnIdle() = selectUpnpMutex.withLock {
if (remoteState.isPlaying) return@withLock // resumed at the boundary
val active = activeUpnpHolder.active.value
val capturedPositionMs = remoteState.positionMs
runCatching { active?.avTransport?.stop() }
.onFailure { Timber.w(it, "UPnP Stop failed during idle revert") }
activeUpnpHolder.set(null)
activeUpnpHolder.setTarget(null)
selectedUpnpRouteIdInternal.value = null
idleRevertJob = null
playerController.seekTo(capturedPositionMs)
}
/**
* When the user plays a different playlist while Sonos is active,
* PlayerController.setQueue replaces the local queue but Sonos's
* native queue still holds the OLD tracks. MinstrelForwardingPlayer's
* setMediaItems override clears holder.active + sets target so the
* imminent play() call drops (drops via isLoadingUpnp() = true). Then
* this collector observes the uiState.queue change and re-runs
* loadQueueOnSonos to push the new tracks to Sonos.
*
* Discrimination: selectUpnp's initial-load path doesn't change
* uiState.queue (the queue was already populated before route
* selection), so this collector doesn't fire during that window. Only
* a fresh setQueue from PlayerController bumps the joined-ids key.
*/
private suspend fun observeQueueChangesForSonosResync() {
playerController.uiState.collect { state ->
val newIds = state.queue.map { it.id }
val oldIds = lastSyncedQueueIds
if (newIds == oldIds) return@collect
lastSyncedQueueIds = newIds
// Route can be in target (setMediaItems-induced clearing already
// ran in ForwardingPlayer) OR in active (queue changed via
// addMediaItem / removeMediaItems etc. which don't hit the
// markPending hook).
val routeId = activeUpnpHolder.target.value
?: activeUpnpHolder.active.value?.routeId
?: return@collect
if (state.queue.isEmpty()) {
Timber.w("Sonos resync skipped: empty queue (clearing target)")
activeUpnpHolder.setTarget(null)
return@collect
}
scope.launch {
resyncSonosQueue(routeId, oldIds.orEmpty(), state.queue, state.queueIndex)
}
}
}
/**
* Bring Sonos's native queue back in sync with the local queue after a
* mutation. Tries an incremental SOAP diff first (RemoveTrackRangeFromQueue
* + AddURIToQueue at the insertion point) so playback continues without
* interruption -- that's what playNext / radio-append want. Falls back to
* the full removeAllTracks + reload path when the diff implies the current
* Sonos track was deleted (e.g. user switched playlists), which is what
* the user-reported "Sonos queue does not update" bug needed.
*/
private suspend fun resyncSonosQueue(
routeId: String,
oldIds: List<String>,
newQueue: List<TrackRef>,
newCurrentIndex: Int,
) = selectUpnpMutex.withLock {
val upnpRoute = upnpDiscovery.routes.value.firstOrNull { it.id == routeId }
val transport = upnpDiscovery.transportFor(routeId)
if (upnpRoute == null || transport == null) {
Timber.w(
"Sonos resync: route or transport gone for %s, dropping to local",
routeId,
)
activeUpnpHolder.setTarget(null)
selectedUpnpRouteIdInternal.value = null
return@withLock
}
val handledIncrementally = runCatching {
tryIncrementalResync(transport, oldIds, newQueue)
}.getOrElse { e ->
Timber.w(e, "Sonos incremental resync errored; falling back to full reload")
false
}
if (handledIncrementally) {
// Active was never cleared on the incremental path; clear any
// target that markPendingResyncIfRemote set (it didn't, for
// incremental cases that don't go through setMediaItems, but
// belt-and-suspenders).
activeUpnpHolder.setTarget(null)
return@withLock
}
// Full rebuild: ensure active is cleared so transport calls drop
// (markPendingResyncIfRemote may already have done this on the
// setMediaItems path).
if (activeUpnpHolder.active.value != null) {
activeUpnpHolder.set(null)
activeUpnpHolder.setTarget(routeId)
}
val outputRoute = OutputRoute.fromUpnpRoute(upnpRoute)
val rendering = renderingClientFor(routeId)
Timber.w("Sonos resync: full reload of %d tracks on %s", newQueue.size, outputRoute.name)
runCatching {
loadQueueOnSonos(transport, outputRoute, newQueue, newCurrentIndex)
activeUpnpHolder.set(
ActiveUpnp(
routeId = routeId,
routeName = outputRoute.name,
avTransport = transport,
rendering = rendering,
),
)
activeUpnpHolder.setTarget(null)
}.onFailure { e ->
Timber.w(e, "Sonos resync (full) failed -- dropping to local")
activeUpnpHolder.setTarget(null)
selectedUpnpRouteIdInternal.value = null
}
}
/**
* Diff-based incremental Sonos queue sync. Returns true when the new
* queue can be produced from the old one with a remove-then-insert at
* the same middle slice -- the common-prefix and common-suffix portions
* stay untouched, and the current Sonos track must lie in the preserved
* prefix (otherwise the diff would orphan playback). Returns false to
* signal the caller to fall back to a full reload.
*/
private suspend fun tryIncrementalResync(
transport: AVTransportClient,
oldIds: List<String>,
newQueue: List<TrackRef>,
): Boolean {
val newIds = newQueue.map { it.id }
if (oldIds == newIds) return true
val prefixLen = commonPrefixLength(oldIds, newIds)
val suffixLen = commonSuffixLength(
oldIds.subList(prefixLen, oldIds.size),
newIds.subList(prefixLen, newIds.size),
)
val removedCount = oldIds.size - prefixLen - suffixLen
val addedCount = newIds.size - prefixLen - suffixLen
// Sonos's current track number is 1-based; compare against the
// preserved-prefix range as 0-based. If the current track is in
// the removed slice, incremental can't preserve playback -- caller
// falls back to full rebuild.
val currentSonosIdx0 = remoteState.trackNumber - 1
val canApply = currentSonosIdx0 in 0 until prefixLen
if (canApply) {
applyQueueDiff(transport, newQueue, prefixLen, removedCount, addedCount)
} else {
Timber.w(
"Sonos incremental: current track %d not in preserved prefix [0,%d); full rebuild",
currentSonosIdx0,
prefixLen,
)
}
return canApply
}
private suspend fun applyQueueDiff(
transport: AVTransportClient,
newQueue: List<TrackRef>,
prefixLen: Int,
removedCount: Int,
addedCount: Int,
) {
if (removedCount > 0) {
Timber.w(
"Sonos incremental: RemoveTrackRangeFromQueue start=%d count=%d",
prefixLen + 1,
removedCount,
)
transport.removeTrackRangeFromQueue(
startingIndex = prefixLen + 1,
numberOfTracks = removedCount,
)
}
if (addedCount == 0) return
Timber.w(
"Sonos incremental: AddURIToQueue x%d starting at position %d",
addedCount,
prefixLen + 1,
)
for (i in 0 until addedCount) {
val ref = newQueue[prefixLen + i]
val token = streamTokens.mint(ref.id)
transport.addURIToQueue(
uri = token.url,
mime = token.mime,
title = token.title,
enqueuedURIPosition = prefixLen + i + 1,
)
if (i > 0) delay(EXTEND_THROTTLE_MS)
}
}
private fun commonPrefixLength(a: List<String>, b: List<String>): Int {
val limit = minOf(a.size, b.size)
for (i in 0 until limit) {
if (a[i] != b[i]) return i
}
return limit
}
private fun commonSuffixLength(a: List<String>, b: List<String>): Int {
val limit = minOf(a.size, b.size)
for (i in 0 until limit) {
if (a[a.size - 1 - i] != b[b.size - 1 - i]) return i
}
return limit
}
/**
* Called when the active UPnP route drops unexpectedly (the poll loop's
* consecutive-failure threshold tripping). Captures the last remote
* position + play state, clears UPnP selection, and resumes local ExoPlayer at
* the same point. Skips the Stop SOAP (device already unreachable).
* The snackbar is handled independently by the NowPlaying surface
* collecting the same [PlayerFactory.dropEvents] via PlayerController.
*/
private fun handleRemoteDrop() {
val capturedPositionMs = remoteState.positionMs
// Resume on the operator's last play-intent, not the observed remote
// state: after a pause, isPlaying is false, but a failed play() that
// triggered this drop means the operator DID ask to play -- honor it.
val wasPlayingRemote = remoteState.lastPlayIntent
activeUpnpHolder.set(null)
activeUpnpHolder.setTarget(null)
selectedUpnpRouteIdInternal.value = null
playerController.seekTo(capturedPositionMs)
if (wasPlayingRemote) playerController.play()
}
/**
@@ -163,46 +509,302 @@ class OutputPickerController @Inject constructor(
}
private fun selectSystem(route: OutputRoute) {
val wasUpnp = selectedUpnpRouteIdInternal.value
if (wasUpnp != null) {
val active = activeUpnpHolder.active.value
val capturedPositionMs = remoteState.positionMs
val wasPlayingRemote = remoteState.isPlaying
scope.launch {
runCatching { active?.avTransport?.stop() }
.onFailure { Timber.w(it, "UPnP Stop failed during disconnect") }
activeUpnpHolder.set(null)
activeUpnpHolder.setTarget(null)
selectedUpnpRouteIdInternal.value = null
playerController.seekTo(capturedPositionMs)
if (wasPlayingRemote) playerController.play()
}
}
val target = mediaRouter.routes.firstOrNull { it.id == route.id } ?: return
mediaRouter.selectRoute(target)
}
/**
* Drive the UPnP renderer: mint a token for the currently playing
* track, set the renderer's URI, play, then pause local playback so
* audio yields to the speaker. Wrapped in `runCatching` at each
* step — token failure, transport-lookup failure, and SOAP failure
* each abandon the selection cleanly rather than crashing. Failures
* log at warn level via Timber so on-device verification can find
* the cause in logcat (OkHttp's logger doesn't cover our own
* deserialize / SOAP-parse code paths).
* Drive the UPnP renderer using Sonos native queue mode:
* clear the device's queue, load every track from our local queue
* via AddURIToQueue, point the transport at the queue URI, seek to
* the current index, and play. Wrapped in `runCatching` — SOAP
* failure abandons the selection cleanly.
*
* Order of operations is deliberate:
* 1. Pause local so the user doesn't keep hearing local audio.
* 2. Set target early so ForwardingPlayer drops transport taps
* while the 17-second queue load is in progress.
* 3. Wire active LAST (after loadQueueOnSonos) so SOAP commands
* are never routed to a half-loaded Sonos queue.
*/
private suspend fun selectUpnp(route: OutputRoute) {
val trackId = playerController.uiState.value.currentTrack?.id
if (trackId == null) {
private suspend fun selectUpnp(route: OutputRoute) = selectUpnpMutex.withLock {
val uiState = playerController.uiState.value
val currentTrack = uiState.currentTrack
if (currentTrack == null) {
Timber.w("UPnP select skipped: no currentTrack (start playback first)")
return
return@withLock
}
val transport = upnpDiscovery.transportFor(route.id)
// Honor Sonos topology: pick the coordinator's route when the user
// tapped a group row. Suppression in routesState already keeps the
// visible row at the coordinator's id, so this is identity in the
// common case -- defensive for follow-up flows.
val effectiveRoute = upnpDiscovery.coordinatorRouteFor(route.id)
?.let { OutputRoute.fromUpnpRoute(it) } ?: route
val transport = upnpDiscovery.transportFor(effectiveRoute.id)
if (transport == null) {
Timber.w(
"UPnP select skipped: no transport for route id=${route.id} " +
"UPnP select skipped: no transport for route id=${effectiveRoute.id} " +
"(route disappeared or id mismatch with discovery list)",
)
return
return@withLock
}
runCatching {
Timber.i("UPnP select: mint token for track=$trackId, route=${route.name}")
val token = castApi.streamToken(StreamTokenRequest(trackId = trackId))
Timber.i("UPnP select: SetAVTransportURI to ${token.url}")
transport.setAVTransportURI(token.url)
Timber.i("UPnP select: Play")
transport.play()
val rendering = renderingClientFor(effectiveRoute.id)
// Pause local before flipping UI state -- user shouldn't keep hearing
// local audio while we queue up Sonos.
playerController.pause()
Timber.i("UPnP select: done")
}.onFailure { e ->
Timber.w(e, "UPnP select failed for route ${route.id}")
selectedUpnpRouteIdInternal.value = effectiveRoute.id
// If the renderer is already playing this exact queue (a "stray"
// session we got disconnected from), attach to it in place instead of
// wiping + reloading -- no jarring restart, and skip/seek land on its
// live queue. Falls through to the clear+reload path when it isn't ours.
if (tryAdoptRunningSession(transport, rendering, effectiveRoute, uiState.queue)) {
return@withLock
}
// Mark UPnP loading. ForwardingPlayer overrides drop transport commands
// silently while target is set but active is null -- the user's premature
// taps don't hit Sonos's stale state from a prior session.
activeUpnpHolder.setTarget(effectiveRoute.id)
runCatching {
loadQueueOnSonos(transport, effectiveRoute, uiState.queue, uiState.queueIndex)
// Wire active LAST -- SOAP path is now safe to use.
activeUpnpHolder.set(
ActiveUpnp(
routeId = effectiveRoute.id,
routeName = effectiveRoute.name,
avTransport = transport,
rendering = rendering,
),
)
// Selecting a route loads the queue and Plays on the renderer, so
// the standing intent is "play" -- a later drop should resume local.
remoteState.setPlayIntent(true)
}.onFailure { e ->
Timber.w(e, "UPnP select failed for route ${effectiveRoute.id}")
activeUpnpHolder.set(null)
activeUpnpHolder.setTarget(null)
selectedUpnpRouteIdInternal.value = null
}
}
/**
* The renderer's current position + transport state, validated as an
* adoptable continuation of our local queue: same track id at the same
* 1-based index. Null means "not ours / not playing" -> caller reloads.
*/
private data class AdoptPlan(
val localIndex: Int,
val positionMs: Long,
val durationMs: Long,
val trackUri: String,
val trackNumber: Int,
val playing: Boolean,
)
/**
* Attach to a renderer that is *already* playing our queue (a session the
* phone got disconnected from but the speaker kept streaming) instead of
* clearing + reloading it. Probes the renderer's live state, and if it's
* mid-playback on the same track id at the same queue index, syncs the
* local cursor to match and wires [ActiveUpnpHolder] -- the speaker never
* skips a beat, and skip/seek now drive its live queue. Returns false (and
* the caller does the full clear+reload) when the renderer isn't playing
* our queue. The local player is already paused by the caller.
*/
private suspend fun tryAdoptRunningSession(
transport: AVTransportClient,
rendering: RenderingControlClient?,
route: OutputRoute,
localQueue: List<TrackRef>,
): Boolean {
val plan = probeAdoptable(transport, route, localQueue) ?: return false
Timber.w(
"UPnP adopt: attaching to live session on %s at track %d (%dms)",
route.name, plan.trackNumber, plan.positionMs,
)
// active is still null here, so this aligns the *local* cursor without
// emitting SOAP; the holder.set below then starts the poll loop, which
// reconciles to the renderer's advancing position from here.
playerController.moveCursorTo(plan.localIndex, plan.positionMs)
remoteState.applyPositionInfo(
positionMs = plan.positionMs,
durationMs = plan.durationMs,
trackUri = plan.trackUri,
trackNumber = plan.trackNumber,
)
remoteState.setPlayIntent(plan.playing)
activeUpnpHolder.set(
ActiveUpnp(
routeId = route.id,
routeName = route.name,
avTransport = transport,
rendering = rendering,
),
)
return true
}
private suspend fun probeAdoptable(
transport: AVTransportClient,
route: OutputRoute,
localQueue: List<TrackRef>,
): AdoptPlan? {
val probe = runCatching {
transport.getTransportInfo() to transport.getPositionInfo()
}.getOrElse {
Timber.w(it, "UPnP adopt: probe failed for %s; will reload", route.name)
return null
}
val (info, pos) = probe
val idx0 = pos.track - 1
val playingId = trackIdFromStreamUrl(pos.trackUri)
val midPlayback =
info.state == TransportState.PLAYING || info.state == TransportState.PAUSED
val aligned = midPlayback && playingId != null &&
idx0 in localQueue.indices && localQueue[idx0].id == playingId
return if (aligned) {
AdoptPlan(
localIndex = idx0,
positionMs = pos.relTimeMs,
durationMs = pos.trackDurationMs,
trackUri = pos.trackUri,
trackNumber = pos.track,
playing = info.state == TransportState.PLAYING,
)
} else {
null
}
}
/** Extract our track id from a stream URL `.../api/tracks/<id>/stream...`. */
private fun trackIdFromStreamUrl(url: String): String? {
val segments = url.toHttpUrlOrNull()?.pathSegments ?: return null
val i = segments.indexOf("tracks")
return if (i >= 0) segments.getOrNull(i + 1) else null
}
private suspend fun loadQueueOnSonos(
transport: AVTransportClient,
route: OutputRoute,
queue: List<TrackRef>,
currentIndex: Int,
) {
Timber.w("UPnP select: clear queue on %s", route.name)
transport.removeAllTracksFromQueue()
val initialEnd = (currentIndex + 1).coerceAtMost(queue.size)
val initialBatch = queue.subList(0, initialEnd)
Timber.w(
"UPnP select: add %d initial tracks (currentIndex=%d, totalQueue=%d)",
initialBatch.size, currentIndex, queue.size,
)
initialBatch.forEachIndexed { idx, ref ->
val token = streamTokens.mint(ref.id)
transport.addURIToQueue(
uri = token.url,
mime = token.mime,
title = token.title,
enqueuedURIPosition = idx + 1,
)
}
val coordinatorUdn = route.id.bareUdn()
val queueUri = "x-rincon-queue:$coordinatorUdn#0"
Timber.w("UPnP select: SetAVTransportURI %s", queueUri)
transport.setAVTransportURI(queueUri, "")
Timber.w("UPnP select: Seek to track %d", currentIndex + 1)
transport.seekToTrack(currentIndex + 1)
Timber.w("UPnP select: Play")
transport.play()
Timber.w("UPnP select: initial done; backgrounding remainder")
val remaining = queue.drop(initialEnd)
if (remaining.isNotEmpty()) {
scope.launch { extendQueueOnSonos(transport, route, remaining, initialEnd) }
}
}
/**
* Background-append tracks after activation. Runs concurrently with
* Sonos playback. Cancels if the user disconnects from this route
* (active.routeId changes or becomes null). Tolerates individual
* AddURIToQueue failures — log and continue so some tracks loaded
* is better than zero tracks loaded.
*/
private suspend fun extendQueueOnSonos(
transport: AVTransportClient,
route: OutputRoute,
tracks: List<TrackRef>,
startPosition: Int,
) {
Timber.w(
"UPnP extend: appending %d tracks starting at position %d",
tracks.size, startPosition + 1,
)
var consecutiveFailures = 0
var succeeded = 0
var aborted = false
for ((i, ref) in tracks.withIndex()) {
if (aborted) break
if (activeUpnpHolder.active.value?.routeId != route.id) {
Timber.w("UPnP extend: cancelled at offset %d (route changed)", i)
aborted = true
} else {
val outcome = runCatching {
val token = streamTokens.mint(ref.id)
transport.addURIToQueue(
uri = token.url,
mime = token.mime,
title = token.title,
enqueuedURIPosition = startPosition + i + 1,
)
}
if (outcome.isSuccess) {
consecutiveFailures = 0
succeeded += 1
// Throttle the burst so we don't tickle Sonos's burst-add
// rejection -- logcat 2026-06-04 showed 33 consecutive
// failures clustered at ~10ms intervals once offset 39 was
// reached, which looks like a rate-limit kicking in. The
// delay is small enough that extending 100 tracks adds
// only ~5s to background work that's already async.
delay(EXTEND_THROTTLE_MS)
} else {
consecutiveFailures += 1
val e = outcome.exceptionOrNull()
val detail = (e as? SoapFaultException)?.let {
"code=${it.code} desc=${it.description}"
} ?: e?.message
Timber.w(e, "UPnP extend: append failed at offset %d -- %s", i, detail)
if (consecutiveFailures >= EXTEND_ABORT_AFTER_FAILURES) {
Timber.w(
"UPnP extend: aborting after %d consecutive failures",
consecutiveFailures,
)
aborted = true
}
}
}
}
Timber.w("UPnP extend: done (%d / %d appended)", succeeded, tracks.size)
}
private fun renderingClientFor(routeId: String): RenderingControlClient? {
val rcUrl = upnpDiscovery.routes.value
.firstOrNull { it.id == routeId }
?.renderingControlUrl ?: return null
return RenderingControlClient(SoapClient(okHttp), rcUrl)
}
private fun refresh() {
@@ -214,24 +816,26 @@ class OutputPickerController @Inject constructor(
.filter { it.matchesSelector(selector) }
.map { OutputRoute.fromRouteInfo(it) }
val current = OutputRoute.fromRouteInfo(mediaRouter.selectedRoute)
return RouteSnapshot(current = current, available = sortRoutes(current, all))
return RouteSnapshot(current = current, available = sortRoutes(all))
}
/**
* Selected first, then Bluetooth, then Wired, then BuiltIn, then
* Other (UPnP renderers fall in Other). Keeps the active output at
* the top + likely-wanted alternatives next + fallback last.
* BuiltIn "Phone speaker" pinned first; everything else
* alphabetical. Selection state is conveyed by the radio button
* indicator in the picker row, not by sort order.
*/
private fun sortRoutes(current: OutputRoute, all: List<OutputRoute>): List<OutputRoute> {
val rank: (OutputRoute) -> Int = { route ->
when {
route.id == current.id -> 0
route.kind == OutputRoute.Kind.Bluetooth -> 1
route.kind == OutputRoute.Kind.Wired -> 2
route.kind == OutputRoute.Kind.BuiltIn -> 3
else -> 4
private fun sortRoutes(all: List<OutputRoute>): List<OutputRoute> {
val (builtIn, rest) = all.partition { it.kind == OutputRoute.Kind.BuiltIn }
return builtIn + rest.sortedBy { it.name.lowercase() }
}
}
return all.sortedBy(rank)
private companion object {
const val EXTEND_ABORT_AFTER_FAILURES = 3
const val EXTEND_THROTTLE_MS = 50L
// 5 minutes of continuous non-playing on a UPnP route before we
// revert to the phone speaker, so a stale Sonos selection can't make
// a later "tap play" do nothing.
const val IDLE_REVERT_MS = 5 * 60 * 1000L
}
}
@@ -4,8 +4,11 @@ import androidx.compose.foundation.layout.Arrangement
import androidx.compose.foundation.layout.Column
import androidx.compose.foundation.layout.Row
import androidx.compose.foundation.layout.fillMaxWidth
import androidx.compose.foundation.layout.heightIn
import androidx.compose.foundation.layout.padding
import androidx.compose.foundation.layout.size
import androidx.compose.foundation.lazy.LazyColumn
import androidx.compose.foundation.lazy.items
import androidx.compose.material3.ExperimentalMaterial3Api
import androidx.compose.material3.Icon
import androidx.compose.material3.MaterialTheme
@@ -69,13 +72,20 @@ fun OutputPickerSheet(
color = MaterialTheme.colorScheme.onSurface,
modifier = Modifier.padding(vertical = 8.dp),
)
snapshot.available.forEach { route ->
LazyColumn(
modifier = Modifier
.fillMaxWidth()
.heightIn(max = ROUTE_LIST_MAX_HEIGHT_DP.dp),
verticalArrangement = Arrangement.spacedBy(4.dp),
) {
items(items = snapshot.available, key = { it.id }) { route ->
RouteRow(
route = route,
isSelected = route.id == snapshot.current.id,
onClick = { onRouteSelected(route) },
)
}
}
if (permissionDenied) {
PermissionHintRow()
}
@@ -123,7 +133,7 @@ private fun RouteRow(
maxLines = 2,
overflow = TextOverflow.Ellipsis,
)
val subtitle = route.description ?: defaultSubtitle(route)
val subtitle = route.description ?: defaultSubtitle(route, isSelected)
Text(
text = subtitle,
style = MaterialTheme.typography.bodySmall,
@@ -188,13 +198,17 @@ private fun MulticastHintRow() {
}
}
private fun defaultSubtitle(route: OutputRoute): String = when (route.kind) {
// isSelected (route == the active/selected route) drives the Bluetooth
// "Connected" subtitle. RouteInfo.connectionState can't — it stays
// DISCONNECTED for local SYSTEM routes even when they're in use.
private fun defaultSubtitle(route: OutputRoute, isSelected: Boolean): String = when (route.kind) {
OutputRoute.Kind.BuiltIn -> "Phone speaker"
OutputRoute.Kind.Wired -> "Wired"
OutputRoute.Kind.Bluetooth -> if (route.isConnected) "Connected" else "Available"
OutputRoute.Kind.Bluetooth -> if (isSelected) "Connected" else "Available"
OutputRoute.Kind.Cast -> "Cast"
OutputRoute.Kind.Other -> "Available"
}
private const val ROW_ICON_DP = 24
private const val HINT_ICON_DP = 20
private const val ROUTE_LIST_MAX_HEIGHT_DP = 400
@@ -24,6 +24,7 @@ import javax.inject.Inject
@HiltViewModel
class OutputPickerViewModel @Inject constructor(
private val controller: OutputPickerController,
private val activeUpnpHolder: ActiveUpnpHolder,
) : ViewModel() {
val routes: StateFlow<RouteSnapshot> = controller.routesState
@@ -33,6 +34,9 @@ class OutputPickerViewModel @Inject constructor(
initialValue = controller.routesState.value,
)
val activeUpnp: StateFlow<ActiveUpnp?> = activeUpnpHolder.active
.stateIn(viewModelScope, SharingStarted.WhileSubscribed(STOP_TIMEOUT_MS), null)
private val sheetVisibleInternal = MutableStateFlow(false)
val sheetVisible: StateFlow<Boolean> = sheetVisibleInternal.asStateFlow()
@@ -21,7 +21,6 @@ data class OutputRoute(
val description: String?,
val kind: Kind,
val protocol: Protocol,
val isConnected: Boolean,
) {
enum class Kind { BuiltIn, Wired, Bluetooth, Cast, Other }
@@ -43,9 +42,11 @@ data class OutputRoute(
/**
* Lift a MediaRouter [route] into the domain model. Kind is
* inferred from [MediaRouter.RouteInfo.getDeviceType]; unknown
* device types fall through to [Kind.Other]. The
* `connectionState` proxy is good enough for the chip's
* "Connected"/"Available" subtitle.
* device types fall through to [Kind.Other]. "Active" is NOT an
* intrinsic of the route — the caller derives it by comparing to
* the selected route (RouteSnapshot.current), because
* RouteInfo.connectionState only reflects remote-route handshakes
* and stays DISCONNECTED for local SYSTEM routes even when in use.
*/
fun fromRouteInfo(route: MediaRouter.RouteInfo): OutputRoute {
val kind = when (route.deviceType) {
@@ -58,15 +59,12 @@ data class OutputRoute(
MediaRouter.RouteInfo.DEVICE_TYPE_SPEAKER -> Kind.Other
else -> Kind.Other
}
val connected =
route.connectionState == MediaRouter.RouteInfo.CONNECTION_STATE_CONNECTED
return OutputRoute(
id = route.id,
name = route.name,
description = route.description,
kind = kind,
protocol = Protocol.SYSTEM,
isConnected = connected,
)
}
@@ -76,10 +74,10 @@ data class OutputRoute(
* network speakers into the same `OutputPickerController`
* routes stream the system routes come through.
*
* `isConnected = false` because UPnP devices have no
* MediaRouter connection-state concept — they're always
* "available" on the LAN, and the picker's selected-route
* rendering handles the "currently playing" indicator.
* UPnP devices have no MediaRouter connection-state concept —
* they're always "available" on the LAN, and the picker's
* selected-route rendering handles the "currently playing"
* indicator.
*
* Subtitle is `manufacturer modelName` joined by a single
* space, falling back to "Network speaker" when both fields
@@ -96,7 +94,6 @@ data class OutputRoute(
description = description,
kind = Kind.Other,
protocol = Protocol.UPNP,
isConnected = false,
)
}
}
@@ -3,12 +3,13 @@ package com.fabledsword.minstrel.player.output.upnp
import okhttp3.HttpUrl
/**
* High-level wrapper for the UPnP AVTransport service. Three calls
* for v1: SetAVTransportURI / Play / Stop. Pause + Seek deferred
* until we have hardware in the loop to verify each device's quirks
* (Sonos and BubbleUPnP accept the standard shape; some smart TVs
* reject Pause without DIDL).
* High-level wrapper for the UPnP AVTransport service.
* Covers SetAVTransportURI / Play / Pause / Stop / Seek /
* GetPositionInfo / GetTransportInfo / queue management
* (RemoveAllTracksFromQueue, AddURIToQueue, Next, Previous, SeekToTrack).
*/
// One method per AVTransport SOAP verb; splitting would obscure the 1:1 protocol mapping.
@Suppress("TooManyFunctions")
class AVTransportClient(
private val soap: SoapClient,
private val controlUrl: HttpUrl,
@@ -26,6 +27,135 @@ class AVTransportClient(
)
}
/**
* Convenience overload that builds DIDL-Lite metadata around [uri]
* + [mime] + [title] and forwards to [setAVTransportURI]. Sonos
* rejects empty DIDL with vendor error 1023; this constructs the
* minimal-but-Sonos-acceptable shape:
* <DIDL-Lite>
* <item id="0" parentID="-1" restricted="1">
* <dc:title>...</dc:title>
* <upnp:class>object.item.audioItem.musicTrack</upnp:class>
* <res protocolInfo="http-get:*:<mime>:*">...</res>
* </item>
* </DIDL-Lite>
* Generic UPnP renderers tolerate this shape too — there's no
* downside to always sending it. Title falls back to "Minstrel"
* when the caller doesn't supply one.
*/
suspend fun setAVTransportURIWithMetadata(uri: String, mime: String, title: String) {
val safeTitle = title.ifBlank { "Minstrel" }
setAVTransportURI(uri, buildDidlLite(uri, mime, safeTitle))
}
/**
* Remove all tracks from the renderer's queue. UPnP action name is
* `RemoveAllTracksFromQueue`. Used at activation time to clear out any
* leftover queue from prior sessions before loading our local queue.
*/
suspend fun removeAllTracksFromQueue() {
soap.call(
controlUrl = controlUrl,
serviceType = SERVICE_TYPE,
action = "RemoveAllTracksFromQueue",
args = mapOf("InstanceID" to "0"),
)
}
/**
* Remove a contiguous range of tracks from the renderer's native queue.
* Sonos-specific extension to AVTransport; `UpdateID=0` skips the queue-
* version check so this works without first calling GetQueue to learn
* the current update id.
*
* [startingIndex] is 1-based per Sonos convention; [numberOfTracks] is
* the count to remove. Used by OutputPickerController's incremental
* queue resync path (radio-append: remove tail, then AddURIToQueue
* the new items).
*/
suspend fun removeTrackRangeFromQueue(startingIndex: Int, numberOfTracks: Int) {
soap.call(
controlUrl = controlUrl,
serviceType = SERVICE_TYPE,
action = "RemoveTrackRangeFromQueue",
args = mapOf(
"InstanceID" to "0",
"UpdateID" to "0",
"StartingIndex" to startingIndex.toString(),
"NumberOfTracks" to numberOfTracks.toString(),
),
)
}
/**
* Append a track to the renderer's queue. Sonos returns the assigned
* track number + new total queue length in the response, but we don't
* read those (we know our intended position). DIDL-Lite metadata is
* required; reuses the same shape as [setAVTransportURIWithMetadata].
*
* [enqueuedURIPosition] is 1-based; 0 means "append to end" per UPnP.
* Our caller passes 1, 2, 3, ... to ensure stable order.
*/
suspend fun addURIToQueue(
uri: String,
mime: String,
title: String,
enqueuedURIPosition: Int = 0,
) {
val safeTitle = title.ifBlank { "Minstrel" }
val didl = buildDidlLite(uri, mime, safeTitle)
soap.call(
controlUrl = controlUrl,
serviceType = SERVICE_TYPE,
action = "AddURIToQueue",
args = mapOf(
"InstanceID" to "0",
"EnqueuedURI" to uri,
"EnqueuedURIMetaData" to didl,
"DesiredFirstTrackNumberEnqueued" to enqueuedURIPosition.toString(),
"EnqueueAsNext" to "0",
),
)
}
/** Skip to the next track in the renderer's queue. */
suspend fun next() {
soap.call(
controlUrl = controlUrl,
serviceType = SERVICE_TYPE,
action = "Next",
args = mapOf("InstanceID" to "0"),
)
}
/** Skip to the previous track in the renderer's queue. */
suspend fun previous() {
soap.call(
controlUrl = controlUrl,
serviceType = SERVICE_TYPE,
action = "Previous",
args = mapOf("InstanceID" to "0"),
)
}
/**
* Seek to a specific track in the queue. UPnP Seek unit "TRACK_NR";
* target is the 1-based track index. Separate from the existing
* [seek] which uses unit REL_TIME for position-within-track.
*/
suspend fun seekToTrack(trackNumber: Int) {
soap.call(
controlUrl = controlUrl,
serviceType = SERVICE_TYPE,
action = "Seek",
args = mapOf(
"InstanceID" to "0",
"Unit" to "TRACK_NR",
"Target" to trackNumber.toString(),
),
)
}
suspend fun play() {
soap.call(
controlUrl = controlUrl,
@@ -35,6 +165,15 @@ class AVTransportClient(
)
}
suspend fun pause() {
soap.call(
controlUrl = controlUrl,
serviceType = SERVICE_TYPE,
action = "Pause",
args = mapOf("InstanceID" to "0"),
)
}
suspend fun stop() {
soap.call(
controlUrl = controlUrl,
@@ -44,7 +183,114 @@ class AVTransportClient(
)
}
suspend fun seek(positionMs: Long) {
soap.call(
controlUrl = controlUrl,
serviceType = SERVICE_TYPE,
action = "Seek",
args = mapOf(
"InstanceID" to "0",
"Unit" to "REL_TIME",
"Target" to formatHhMmSs(positionMs),
),
)
}
suspend fun getPositionInfo(): PositionInfo {
val result = soap.call(
controlUrl = controlUrl,
serviceType = SERVICE_TYPE,
action = "GetPositionInfo",
args = mapOf("InstanceID" to "0"),
)
return PositionInfo(
track = result["Track"]?.toIntOrNull() ?: 0,
trackUri = result["TrackURI"].orEmpty(),
relTimeMs = parseHhMmSs(result["RelTime"].orEmpty()),
trackDurationMs = parseHhMmSs(result["TrackDuration"].orEmpty()),
)
}
suspend fun getTransportInfo(): TransportInfo {
val result = soap.call(
controlUrl = controlUrl,
serviceType = SERVICE_TYPE,
action = "GetTransportInfo",
args = mapOf("InstanceID" to "0"),
)
val state = when (result["CurrentTransportState"]) {
"PLAYING" -> TransportState.PLAYING
"PAUSED_PLAYBACK" -> TransportState.PAUSED
"STOPPED" -> TransportState.STOPPED
"TRANSITIONING" -> TransportState.TRANSITIONING
else -> TransportState.UNKNOWN
}
return TransportInfo(state)
}
private fun buildDidlLite(uri: String, mime: String, title: String): String {
// Sonos requires (a) the rinconnetworks namespace declared on
// <DIDL-Lite> even if we don't use Rincon elements directly, and
// (b) a <desc id="cdudn"> element identifying the URI as an
// external (non-Sonos-library) source. Without those, Sonos
// discards our metadata content and regenerates its own with
// class=object.item and the URL query string as the title
// (logcat 2026-06-04 confirmed). Match SoCo's pattern.
val safeTitle = title.ifBlank { "Minstrel" }
return buildString {
append("<DIDL-Lite ")
append("xmlns=\"urn:schemas-upnp-org:metadata-1-0/DIDL-Lite/\" ")
append("xmlns:dc=\"http://purl.org/dc/elements/1.1/\" ")
append("xmlns:upnp=\"urn:schemas-upnp-org:metadata-1-0/upnp/\" ")
append("xmlns:r=\"urn:schemas-rinconnetworks-com:metadata-1-0/\">")
append("<item id=\"-1\" parentID=\"-1\" restricted=\"true\">")
append("<dc:title>").append(xmlEscape(safeTitle)).append("</dc:title>")
append("<upnp:class>object.item.audioItem.musicTrack</upnp:class>")
append("<res protocolInfo=\"http-get:*:").append(xmlEscape(mime))
append(":*\">").append(xmlEscape(uri)).append("</res>")
append("<desc id=\"cdudn\" ")
append("nameSpace=\"urn:schemas-rinconnetworks-com:metadata-1-0/\">")
append("RINCON_AssociatedZPUDN")
append("</desc>")
append("</item>")
append("</DIDL-Lite>")
}
}
private fun formatHhMmSs(positionMs: Long): String {
val totalSec = (positionMs / MILLIS_PER_SECOND).coerceAtLeast(0)
val h = totalSec / SECONDS_PER_HOUR
val m = (totalSec % SECONDS_PER_HOUR) / SECONDS_PER_MINUTE
val s = totalSec % SECONDS_PER_MINUTE
return "%d:%02d:%02d".format(h, m, s)
}
private fun parseHhMmSs(raw: String): Long {
val parts = raw.split(':').mapNotNull { it.trim().toLongOrNull() }
return if (parts.size == EXPECTED_HMS_PARTS) {
val (h, m, s) = parts
((h * SECONDS_PER_HOUR) + (m * SECONDS_PER_MINUTE) + s) * MILLIS_PER_SECOND
} else {
0L
}
}
private companion object {
const val SERVICE_TYPE = "urn:schemas-upnp-org:service:AVTransport:1"
const val MILLIS_PER_SECOND = 1000L
const val SECONDS_PER_MINUTE = 60L
const val SECONDS_PER_HOUR = 3600L
const val EXPECTED_HMS_PARTS = 3
}
}
data class PositionInfo(
val track: Int,
val trackUri: String,
val relTimeMs: Long,
val trackDurationMs: Long,
)
enum class TransportState { PLAYING, PAUSED, STOPPED, TRANSITIONING, UNKNOWN }
data class TransportInfo(val state: TransportState)
@@ -22,10 +22,12 @@ data class DeviceDescription(
val modelName: String,
val avTransportControlUrl: HttpUrl,
val renderingControlUrl: HttpUrl?,
val zoneGroupTopologyControlUrl: HttpUrl? = null,
) {
companion object {
private const val AVT_SERVICE_TYPE = "urn:schemas-upnp-org:service:AVTransport:1"
private const val RC_SERVICE_TYPE = "urn:schemas-upnp-org:service:RenderingControl:1"
private const val ZGT_SERVICE_TYPE = "urn:schemas-upnp-org:service:ZoneGroupTopology:1"
private const val TAG_SERVICE = "service"
private const val TAG_UDN = "UDN"
@@ -43,6 +45,7 @@ data class DeviceDescription(
*/
fun parse(xml: String, base: HttpUrl): DeviceDescription? {
val parser = XmlPullParserFactory.newInstance().newPullParser().apply {
setFeature(XmlPullParser.FEATURE_PROCESS_NAMESPACES, true)
setInput(xml.reader())
}
val acc = ParseState()
@@ -58,6 +61,7 @@ data class DeviceDescription(
modelName = acc.modelName,
avTransportControlUrl = avt,
renderingControlUrl = acc.rcControlUrl,
zoneGroupTopologyControlUrl = acc.zgtControlUrl,
)
}
@@ -91,6 +95,7 @@ data class DeviceDescription(
when (acc.serviceType) {
AVT_SERVICE_TYPE -> acc.avtControlUrl = resolved
RC_SERVICE_TYPE -> acc.rcControlUrl = resolved
ZGT_SERVICE_TYPE -> acc.zgtControlUrl = resolved
}
acc.inService = false
}
@@ -115,6 +120,7 @@ data class DeviceDescription(
var modelName: String = ""
var avtControlUrl: HttpUrl? = null
var rcControlUrl: HttpUrl? = null
var zgtControlUrl: HttpUrl? = null
var inService: Boolean = false
var serviceType: String = ""
var serviceControlUrl: String = ""
@@ -0,0 +1,44 @@
package com.fabledsword.minstrel.player.output.upnp
import okhttp3.HttpUrl
/**
* RenderingControl service wrapper for hardware-volume routing while a
* UPnP route is active. GetVolume seeds an in-memory cache; SetVolume
* is invoked by NowPlayingScreen's volume-key interceptor. Clamps to
* the UPnP-standard 0..100 range.
*/
class RenderingControlClient(
private val soap: SoapClient,
private val controlUrl: HttpUrl,
) {
suspend fun getVolume(channel: String = "Master"): Int {
val args = soap.call(
controlUrl = controlUrl,
serviceType = SERVICE_TYPE,
action = "GetVolume",
args = mapOf("InstanceID" to "0", "Channel" to channel),
)
return args["CurrentVolume"]?.toIntOrNull() ?: 0
}
suspend fun setVolume(volume: Int, channel: String = "Master") {
val clamped = volume.coerceIn(VOLUME_MIN, VOLUME_MAX)
soap.call(
controlUrl = controlUrl,
serviceType = SERVICE_TYPE,
action = "SetVolume",
args = mapOf(
"InstanceID" to "0",
"Channel" to channel,
"DesiredVolume" to clamped.toString(),
),
)
}
private companion object {
const val SERVICE_TYPE = "urn:schemas-upnp-org:service:RenderingControl:1"
const val VOLUME_MIN = 0
const val VOLUME_MAX = 100
}
}
@@ -1,5 +1,6 @@
package com.fabledsword.minstrel.player.output.upnp
import java.util.concurrent.atomic.AtomicInteger
import kotlinx.coroutines.Dispatchers
import kotlinx.coroutines.withContext
import okhttp3.HttpUrl
@@ -8,6 +9,7 @@ import okhttp3.Request
import okhttp3.RequestBody.Companion.toRequestBody
import org.xmlpull.v1.XmlPullParser
import org.xmlpull.v1.XmlPullParserFactory
import timber.log.Timber
/**
* Minimal SOAP/UPnP envelope builder + POST. Hand-rolled rather than
@@ -25,6 +27,7 @@ import org.xmlpull.v1.XmlPullParserFactory
*/
class SoapClient(
private val okHttp: OkHttpClient,
private val onRawResponse: ((action: String, body: String) -> Unit)? = null,
) {
suspend fun call(
controlUrl: HttpUrl,
@@ -41,6 +44,7 @@ class SoapClient(
.build()
okHttp.newCall(request).execute().use { response ->
val body = response.body?.string().orEmpty()
onRawResponse?.invoke(action, body)
if (!response.isSuccessful) {
throw SoapFaultException(faultCodeOf(body), faultDescriptionOf(body))
}
@@ -69,15 +73,9 @@ class SoapClient(
append("</s:Envelope>")
}
private fun xmlEscape(v: String): String = v
.replace("&", "&amp;")
.replace("<", "&lt;")
.replace(">", "&gt;")
.replace("\"", "&quot;")
.replace("'", "&apos;")
private fun parseResponseArgs(body: String, action: String): Map<String, String> {
val parser = XmlPullParserFactory.newInstance().newPullParser().apply {
setFeature(XmlPullParser.FEATURE_PROCESS_NAMESPACES, true)
setInput(body.reader())
}
val responseTag = "${action}Response"
@@ -102,7 +100,7 @@ class SoapClient(
} else {
if (inResponse) {
val name = parser.name
val text = runCatching { parser.nextText() }.getOrDefault("")
val text = readElementContent(parser, name)
args[name] = text
}
inResponse
@@ -112,6 +110,59 @@ class SoapClient(
else -> inResponse
}
/**
* Read the content of the currently-started element. Try nextText() first
* (works when the content is text -- escaped XML included). If that throws
* (because the content is nested elements), manually walk to the matching
* END_TAG, accumulating text and re-serializing child elements.
*
* Some Sonos firmware sends the GetZoneGroupState payload as nested XML
* elements without escaping; this fallback recovers that path.
*/
private fun readElementContent(parser: XmlPullParser, tagName: String): String {
return runCatching { parser.nextText() }.getOrElse {
readUntilEndTag(parser, tagName)
}
}
private fun readUntilEndTag(parser: XmlPullParser, tagName: String): String {
val sb = StringBuilder()
var depth = 1
var done = false
while (!done && depth > 0) {
when (parser.next()) {
XmlPullParser.START_TAG -> {
appendStartTag(sb, parser)
depth += 1
}
XmlPullParser.END_TAG -> {
depth -= 1
if (depth == 0 && parser.name == tagName) {
done = true
} else {
sb.append("</").append(parser.name).append('>')
}
}
XmlPullParser.TEXT -> sb.append(parser.text.orEmpty())
XmlPullParser.END_DOCUMENT -> done = true
else -> Unit
}
}
return sb.toString()
}
private fun appendStartTag(sb: StringBuilder, parser: XmlPullParser) {
sb.append('<').append(parser.name)
for (i in 0 until parser.attributeCount) {
sb.append(' ')
.append(parser.getAttributeName(i))
.append("=\"")
.append(parser.getAttributeValue(i))
.append('"')
}
sb.append('>')
}
private fun faultCodeOf(body: String): String =
extractBetween(body, "<errorCode>", "</errorCode>") ?: "unknown"
@@ -138,3 +189,42 @@ class SoapClient(
*/
class SoapFaultException(val code: String, val description: String) :
Exception("SOAP fault $code: $description")
private const val MAX_DIAGNOSTIC_RESPONSES = 6 // 3 polls x 2 action types
private const val MAX_BODY_LOG_CHARS = 2048
/**
* Returns a [SoapClient] that logs the raw response body for the first
* [MAX_DIAGNOSTIC_RESPONSES] calls for actions in [DIAGNOSTIC_ACTIONS]
* (GetPositionInfo, GetTransportInfo, GetZoneGroupState). After that the
* callback is a no-op so there is no persistent log spam. Logged at WARN
* so release builds capture it without a separate log-level override.
*/
internal fun loggingSoapClient(okHttp: OkHttpClient, label: String): SoapClient {
val counter = AtomicInteger(0)
return SoapClient(okHttp) { action, body ->
if (action !in DIAGNOSTIC_ACTIONS) return@SoapClient
val n = counter.incrementAndGet()
if (n <= MAX_DIAGNOSTIC_RESPONSES) {
Timber.w(
"UPnP %s response #%d (%s): %s",
label, n, action,
body.take(MAX_BODY_LOG_CHARS),
)
}
}
}
private val DIAGNOSTIC_ACTIONS = setOf(
"GetPositionInfo",
"GetTransportInfo",
"GetZoneGroupState",
)
/** XML-escapes a string value for embedding as text content inside a SOAP envelope. */
internal fun xmlEscape(v: String): String = v
.replace("&", "&amp;")
.replace("<", "&lt;")
.replace(">", "&gt;")
.replace("\"", "&quot;")
.replace("'", "&apos;")
@@ -1,10 +1,14 @@
@file:Suppress("TooManyFunctions") // discovery + Sonos topology + transport-lookup density
package com.fabledsword.minstrel.player.output.upnp
import android.content.Context
import com.fabledsword.minstrel.di.ApplicationScope
import com.fabledsword.minstrel.player.output.upnp.sonos.SonosZoneGroup
import com.fabledsword.minstrel.player.output.upnp.sonos.ZoneGroupTopologyClient
import dagger.hilt.android.qualifiers.ApplicationContext
import kotlinx.coroutines.CoroutineScope
import kotlinx.coroutines.Dispatchers
import kotlinx.coroutines.delay
import kotlinx.coroutines.flow.MutableStateFlow
import kotlinx.coroutines.flow.StateFlow
import kotlinx.coroutines.flow.asStateFlow
@@ -13,6 +17,9 @@ import okhttp3.HttpUrl
import okhttp3.HttpUrl.Companion.toHttpUrlOrNull
import okhttp3.OkHttpClient
import okhttp3.Request
import timber.log.Timber
import java.util.concurrent.ConcurrentHashMap
import java.util.concurrent.TimeUnit
import javax.inject.Inject
import javax.inject.Singleton
@@ -41,9 +48,27 @@ class UpnpDiscoveryController @Inject constructor(
) {
private val ssdp = SsdpDiscovery(context)
// Transport commands (play/pause/seek/next) and the 1 Hz poll go through
// this client. A dead renderer should fail fast so OutputPickerController's
// drop-recovery reverts to the phone in ~2 s instead of the shared client's
// 10 s connect timeout. Derived via newBuilder() so the connection pool +
// dispatcher stay shared with the app client.
private val controlHttp: OkHttpClient = okHttp.newBuilder()
.connectTimeout(CONTROL_CONNECT_TIMEOUT_SECONDS, TimeUnit.SECONDS)
.build()
private val routesInternal = MutableStateFlow<List<UpnpRoute>>(emptyList())
val routes: StateFlow<List<UpnpRoute>> = routesInternal.asStateFlow()
// Wall-clock of the last NOTIFY / M-SEARCH response per route id. Drives
// post-active-scan pruning so a powered-off renderer leaves the picker
// (and can't pin a stale selection) instead of lingering for the process
// lifetime -- upsert only ever added before.
private val lastSeenMs = ConcurrentHashMap<String, Long>()
private val sonosTopologyInternal = MutableStateFlow<List<SonosZoneGroup>>(emptyList())
val sonosTopology: StateFlow<List<SonosZoneGroup>> = sonosTopologyInternal.asStateFlow()
init {
ssdp.start(appScope)
// appScope is process-lifetime (SupervisorJob + Dispatchers.Default),
@@ -61,7 +86,31 @@ class UpnpDiscoveryController @Inject constructor(
* that just opened).
*/
fun upgradeDiscovery() {
appScope.launch { ssdp.requestActiveScan() }
appScope.launch {
val scanStartMs = System.currentTimeMillis()
ssdp.requestActiveScan()
// Give present devices time to answer the M-SEARCH (+ our follow-up
// description fetch), then drop any route that didn't re-announce.
// Pruning is tied to picker-open scans only -- no background timer,
// so route rows never flicker while the user isn't looking.
delay(ACTIVE_SCAN_PRUNE_DELAY_MS)
pruneRoutesNotSeenSince(scanStartMs)
}
}
/**
* Drop routes whose last NOTIFY / M-SEARCH response predates [cutoffMs]
* -- i.e. they didn't answer the active scan that just ran, so they're
* gone from the LAN. Called only after [upgradeDiscovery]'s scan window.
*/
private fun pruneRoutesNotSeenSince(cutoffMs: Long) {
val before = routesInternal.value
val survivors = before.filter { (lastSeenMs[it.id] ?: 0L) >= cutoffMs }
if (survivors.size == before.size) return
val removed = before.filterNot { it in survivors }.map { it.id }
removed.forEach { lastSeenMs.remove(it) }
Timber.w("UPnP discovery: pruned %d stale route(s): %s", removed.size, removed)
routesInternal.value = survivors
}
/**
@@ -83,13 +132,78 @@ class UpnpDiscoveryController @Inject constructor(
*/
fun transportFor(routeId: String): AVTransportClient? {
val route = routesInternal.value.firstOrNull { it.id == routeId } ?: return null
return AVTransportClient(SoapClient(okHttp), route.avTransportControlUrl)
return AVTransportClient(
loggingSoapClient(controlHttp, route.name),
route.avTransportControlUrl,
)
}
private suspend fun handleDiscovery(locationUrl: String) {
val route = fetchRoute(locationUrl) ?: return
routesInternal.value =
routesInternal.value.filterNot { it.id == route.id } + route
upsertRoute(route)
if (route.manufacturer.contains("Sonos", ignoreCase = true)) {
refreshSonosTopology(route)
}
}
private fun upsertRoute(route: UpnpRoute) {
lastSeenMs[route.id] = System.currentTimeMillis()
val current = routesInternal.value
val idx = current.indexOfFirst { it.id == route.id }
routesInternal.value = if (idx < 0) {
current + route
} else {
current.toMutableList().also { it[idx] = route }
}
}
private suspend fun refreshSonosTopology(anySonos: UpnpRoute) {
val zgtUrl = anySonos.zoneGroupTopologyControlUrl ?: run {
Timber.w(
"Sonos %s has no ZoneGroupTopology URL -- topology grouping disabled",
anySonos.id,
)
return
}
val groups = runCatching {
ZoneGroupTopologyClient(
loggingSoapClient(okHttp, "ZGT-${anySonos.id}"),
zgtUrl,
).getZoneGroupState()
}
.onFailure { Timber.w(it, "refreshSonosTopology failed for %s", anySonos.id) }
.getOrNull() ?: return
Timber.w(
"Sonos topology refreshed for %s: %d group(s)",
anySonos.id,
groups.size,
)
sonosTopologyInternal.value = groups
}
/**
* Returns the coordinator UDN's full route for the group [udn] belongs
* to, or null if topology hasn't loaded / the udn is in no group. UDN
* normalization strips the "uuid:" prefix because DeviceDescription's
* <UDN> carries it but Sonos's ZoneGroupState Coordinator attr does not.
*/
fun coordinatorRouteFor(udn: String): UpnpRoute? {
val bare = udn.bareUdn()
val coord = sonosTopologyInternal.value
.firstOrNull { g -> g.members.any { it.udn.bareUdn() == bare } }
?.coordinatorUdn ?: return null
return routesInternal.value.firstOrNull { it.id.bareUdn() == coord.bareUdn() }
}
/**
* UDNs of every NON-coordinator Sonos group member. The picker
* controller suppresses these from the visible list.
*/
fun nonCoordinatorMemberUdns(): Set<String> {
return sonosTopologyInternal.value.flatMap { g ->
g.members.map { it.udn.bareUdn() }
.filter { it != g.coordinatorUdn.bareUdn() }
}.toSet()
}
/**
@@ -111,6 +225,7 @@ class UpnpDiscoveryController @Inject constructor(
modelName = desc.modelName,
avTransportControlUrl = desc.avTransportControlUrl,
renderingControlUrl = desc.renderingControlUrl,
zoneGroupTopologyControlUrl = desc.zoneGroupTopologyControlUrl,
)
}
}
@@ -153,5 +268,32 @@ class UpnpDiscoveryController @Inject constructor(
// devices append. Anchored to end-of-string so it never eats
// a legitimate parenthetical inside a name.
val IP_SUFFIX_REGEX = Regex("""\s*\(\d+\.\d+\.\d+\.\d+\)$""")
// Aggressive connect timeout for transport SOAP so a dead renderer
// trips the drop-recovery quickly. Conservative enough not to false-drop
// a slow-but-alive LAN; read timeout still inherits the shared client.
const val CONTROL_CONNECT_TIMEOUT_SECONDS = 2L
// Grace window after an M-SEARCH burst before pruning non-responders.
// Covers the SSDP round-trip plus our per-device description fetch;
// generous enough that a present-but-slow renderer isn't false-pruned.
const val ACTIVE_SCAN_PRUNE_DELAY_MS = 4_000L
}
}
/**
* Normalize a Sonos UDN to its bare RINCON form for cross-comparison.
*
* Three places use UDN strings with different conventions:
* - DeviceDescription's <UDN> tag carries the `uuid:` prefix.
* - Sonos exposes one UDN per embedded device. The MediaRenderer adds
* a `_MR` suffix and the MediaServer adds `_MS`.
* - The ZGT GetZoneGroupState response uses bare `RINCON_xxx` with
* neither the `uuid:` prefix nor any device suffix.
*
* To compare any pair of those, strip the prefix and the suffix so
* everything reduces to the underlying speaker identity.
*/
internal fun String.bareUdn(): String = removePrefix("uuid:")
.removeSuffix("_MR")
.removeSuffix("_MS")
@@ -7,10 +7,10 @@ import okhttp3.HttpUrl
* Bose SoundTouch, generic DLNA renderers). Lifted out of the SOAP /
* SSDP details so the picker UI consumes a narrow domain shape.
*
* Generic UPnP only for THIS slice — Sonos-specific grouping value-adds
* (group join/leave, zone topology) live in a separate Sonos extension
* scoped in
* docs/superpowers/specs/2026-06-03-android-output-picker-upnp-scope.md.
* Sonos devices additionally populate [zoneGroupTopologyControlUrl] from the
* ZoneGroupTopology service in their device description; non-Sonos devices
* leave it null. The discovery controller uses that URL to aggregate stereo
* pairs and multi-speaker groups into single picker rows.
*
* [id] is the device UDN (e.g. `uuid:RINCON_ABC...`). [name] is the
* raw `<friendlyName>` straight from the device description — callers
@@ -24,4 +24,5 @@ data class UpnpRoute(
val modelName: String,
val avTransportControlUrl: HttpUrl,
val renderingControlUrl: HttpUrl?,
val zoneGroupTopologyControlUrl: HttpUrl? = null,
)
@@ -0,0 +1,90 @@
package com.fabledsword.minstrel.player.output.upnp.sonos
import org.xmlpull.v1.XmlPullParser
import org.xmlpull.v1.XmlPullParserFactory
/**
* One Sonos zone group as exposed by ZoneGroupTopology.GetZoneGroupState.
* Stereo pairs and multi-speaker groups all appear as one group with
* multiple members; one member is the coordinator we send SOAP to.
*/
data class SonosZoneGroup(
val coordinatorUdn: String,
val name: String,
val members: List<SonosZoneMember>,
)
data class SonosZoneMember(
val udn: String,
val roomName: String,
val location: String?,
val channelMapSet: String?,
)
object SonosTopology {
fun parse(xml: String): List<SonosZoneGroup> {
val effective = if (xml.contains("&lt;ZoneGroup")) {
unescapeXmlEntities(xml)
} else {
xml
}
return runCatching { parseStrict(effective) }.getOrDefault(emptyList())
}
private fun unescapeXmlEntities(s: String): String = s
.replace("&lt;", "<")
.replace("&gt;", ">")
.replace("&quot;", "\"")
.replace("&apos;", "'")
.replace("&amp;", "&") // must be last to avoid double-decoding
private fun parseStrict(xml: String): List<SonosZoneGroup> {
val parser = XmlPullParserFactory.newInstance().newPullParser().apply {
setFeature(XmlPullParser.FEATURE_PROCESS_NAMESPACES, true)
setInput(xml.reader())
}
val groups = mutableListOf<SonosZoneGroup>()
var currentCoordinator: String? = null
var currentMembers: MutableList<SonosZoneMember>? = null
while (parser.eventType != XmlPullParser.END_DOCUMENT) {
when (parser.eventType) {
XmlPullParser.START_TAG -> when (parser.name) {
TAG_ZONE_GROUP -> {
currentCoordinator = parser.getAttributeValue(null, ATTR_COORDINATOR)
currentMembers = mutableListOf()
}
TAG_ZONE_GROUP_MEMBER -> currentMembers?.add(memberOf(parser))
}
XmlPullParser.END_TAG -> if (parser.name == TAG_ZONE_GROUP) {
val members = currentMembers ?: emptyList()
val coord = currentCoordinator
if (coord != null && members.isNotEmpty()) {
val name = members.firstOrNull { it.udn == coord }?.roomName
?: members.first().roomName
groups.add(SonosZoneGroup(coord, name, members))
}
currentCoordinator = null
currentMembers = null
}
}
parser.next()
}
return groups
}
private fun memberOf(parser: XmlPullParser): SonosZoneMember = SonosZoneMember(
udn = parser.getAttributeValue(null, ATTR_UUID).orEmpty(),
roomName = parser.getAttributeValue(null, ATTR_ZONE_NAME).orEmpty(),
location = parser.getAttributeValue(null, ATTR_LOCATION),
channelMapSet = parser.getAttributeValue(null, ATTR_CHANNEL_MAP_SET),
)
private const val TAG_ZONE_GROUP = "ZoneGroup"
private const val TAG_ZONE_GROUP_MEMBER = "ZoneGroupMember"
private const val ATTR_COORDINATOR = "Coordinator"
private const val ATTR_UUID = "UUID"
private const val ATTR_ZONE_NAME = "ZoneName"
private const val ATTR_LOCATION = "Location"
private const val ATTR_CHANNEL_MAP_SET = "ChannelMapSet"
}
@@ -0,0 +1,37 @@
package com.fabledsword.minstrel.player.output.upnp.sonos
import com.fabledsword.minstrel.player.output.upnp.SoapClient
import okhttp3.HttpUrl
import timber.log.Timber
/**
* Sonos's proprietary ZoneGroupTopology service. Same SOAP shape as a
* standard UPnP service, exposed on port 1400 at
* /ZoneGroupTopology/Control. GetZoneGroupState returns the full
* network topology as one XML doc wrapped inside a SOAP arg.
*/
class ZoneGroupTopologyClient(
private val soap: SoapClient,
private val controlUrl: HttpUrl,
) {
suspend fun getZoneGroupState(): List<SonosZoneGroup> {
val args = soap.call(
controlUrl = controlUrl,
serviceType = SERVICE_TYPE,
action = "GetZoneGroupState",
args = emptyMap(),
)
val inner = args["ZoneGroupState"].orEmpty()
Timber.w(
"ZGT extracted ZoneGroupState (%d chars): %s",
inner.length,
inner.take(ZGT_LOG_TRUNCATE_CHARS),
)
return SonosTopology.parse(inner)
}
private companion object {
const val SERVICE_TYPE = "urn:schemas-upnp-org:service:ZoneGroupTopology:1"
const val ZGT_LOG_TRUNCATE_CHARS = 2048
}
}
@@ -17,6 +17,7 @@ import androidx.compose.foundation.layout.padding
import androidx.compose.foundation.layout.size
import androidx.compose.foundation.layout.width
import androidx.compose.foundation.shape.RoundedCornerShape
import androidx.compose.material3.CircularProgressIndicator
import androidx.compose.material3.Icon
import androidx.compose.material3.IconButton
import androidx.compose.material3.MaterialTheme
@@ -164,6 +165,7 @@ fun MiniPlayer(
MiniRow(
track = track,
isPlaying = state.isPlaying,
isUpnpLoading = state.isUpnpLoading,
isLiked = isLiked,
onExpandClick = onExpandClick,
onPlayPause = { if (state.isPlaying) viewModel.pause() else viewModel.play() },
@@ -204,6 +206,7 @@ private fun MiniProgressFill(positionMs: Long, durationMs: Long) {
private fun MiniRow(
track: TrackRef,
isPlaying: Boolean,
isUpnpLoading: Boolean,
isLiked: Boolean,
onExpandClick: () -> Unit,
onPlayPause: () -> Unit,
@@ -248,15 +251,39 @@ private fun MiniRow(
}
LikeButton(liked = isLiked, onToggle = onToggleLike)
TransportButton(icon = Lucide.SkipBack, description = "Previous", onClick = onPrev)
TransportButton(
icon = if (isPlaying) Lucide.Pause else Lucide.Play,
description = if (isPlaying) "Pause" else "Play",
MiniPlayPauseButton(
isPlaying = isPlaying,
isUpnpLoading = isUpnpLoading,
onClick = onPlayPause,
)
TransportButton(icon = Lucide.SkipForward, description = "Next", onClick = onNext)
}
}
@Composable
private fun MiniPlayPauseButton(
isPlaying: Boolean,
isUpnpLoading: Boolean,
onClick: () -> Unit,
) {
IconButton(onClick = onClick, enabled = !isUpnpLoading) {
if (isUpnpLoading) {
CircularProgressIndicator(
modifier = Modifier.size(MINI_PLAY_PAUSE_SPINNER_DP.dp),
strokeWidth = 2.dp,
)
} else {
Icon(
imageVector = if (isPlaying) Lucide.Pause else Lucide.Play,
contentDescription = if (isPlaying) "Pause" else "Play",
tint = MaterialTheme.colorScheme.onSurface,
)
}
}
}
private const val MINI_PLAY_PAUSE_SPINNER_DP = 24
@Composable
private fun TransportButton(
icon: androidx.compose.ui.graphics.vector.ImageVector,
@@ -8,6 +8,7 @@ import androidx.activity.compose.rememberLauncherForActivityResult
import androidx.activity.result.contract.ActivityResultContracts
import androidx.compose.animation.ExperimentalSharedTransitionApi
import androidx.compose.foundation.background
import androidx.compose.foundation.focusable
import androidx.compose.foundation.rememberScrollState
import androidx.compose.foundation.verticalScroll
import androidx.compose.foundation.layout.Arrangement
@@ -25,6 +26,7 @@ import androidx.compose.foundation.layout.size
import androidx.compose.foundation.layout.widthIn
import androidx.compose.foundation.shape.CircleShape
import androidx.compose.foundation.shape.RoundedCornerShape
import androidx.compose.material3.CircularProgressIndicator
import androidx.compose.material3.Icon
import androidx.compose.material3.IconButton
import androidx.compose.material3.MaterialTheme
@@ -50,10 +52,19 @@ import androidx.compose.runtime.LaunchedEffect
import androidx.compose.runtime.getValue
import androidx.compose.runtime.mutableStateOf
import androidx.compose.runtime.remember
import androidx.compose.runtime.rememberCoroutineScope
import androidx.compose.runtime.setValue
import androidx.compose.ui.Alignment
import androidx.compose.ui.focus.FocusRequester
import androidx.compose.ui.focus.focusRequester
import androidx.compose.ui.Modifier
import androidx.compose.ui.draw.clip
import androidx.compose.ui.input.key.Key
import androidx.compose.ui.input.key.KeyEvent
import androidx.compose.ui.input.key.KeyEventType
import androidx.compose.ui.input.key.key
import androidx.compose.ui.input.key.onKeyEvent
import androidx.compose.ui.input.key.type
import androidx.compose.ui.platform.LocalContext
import androidx.compose.ui.text.style.TextAlign
import androidx.compose.ui.text.style.TextOverflow
@@ -74,11 +85,13 @@ import com.composables.icons.lucide.Shuffle
import com.composables.icons.lucide.SkipBack
import com.composables.icons.lucide.SkipForward
import com.fabledsword.minstrel.player.RepeatMode
import com.fabledsword.minstrel.player.output.ActiveUpnp
import com.fabledsword.minstrel.player.output.DeviceChip
import com.fabledsword.minstrel.player.output.OutputPickerSheet
import com.fabledsword.minstrel.player.output.OutputPickerViewModel
import com.fabledsword.minstrel.player.output.OutputRoute
import com.fabledsword.minstrel.player.output.RouteSnapshot
import kotlinx.coroutines.launch
import com.fabledsword.minstrel.nav.AlbumDetail
import com.fabledsword.minstrel.nav.ArtistDetail
import com.fabledsword.minstrel.nav.HERO_KEY_NOW_PLAYING_COVER
@@ -129,27 +142,30 @@ fun NowPlayingScreen(
snackbarHostState.showSnackbar(msg)
}
}
LaunchedEffect(Unit) {
viewModel.dropEvents.collect { snackbarHostState.showSnackbar("Disconnected from $it") }
}
val track = state.currentTrack
if (track == null) {
// Session torn down (queue finished + auto-stop, or user cleared
// the queue from elsewhere). Pop back to whichever shell screen
// launched NowPlaying rather than stranding the user on an
// EmptyState with no escape. A short delay swallows the
// momentary null during MediaController IPC bind on cold-mount.
LaunchedEffect(Unit) {
kotlinx.coroutines.delay(POP_GRACE_MS)
if (viewModel.uiState.value.currentTrack == null) {
navController.popBackStack()
}
}
NowPlayingNullTrackGuard(navController, viewModel)
return
}
val outputViewModel: OutputPickerViewModel = hiltViewModel()
val activeUpnp by outputViewModel.activeUpnp.collectAsStateWithLifecycle()
val onKeyEvent = rememberUpnpVolumeKeyHandler(activeUpnp)
val focusRequester = remember { FocusRequester() }
LaunchedEffect(activeUpnp) { if (activeUpnp != null) focusRequester.requestFocus() }
val dominant = rememberDominantColor(track.coverUrl)
val dismissConnection = rememberDragDismissConnection(
onDismiss = { navController.popBackStack() },
)
Scaffold(
modifier = Modifier.fillMaxSize().nestedScroll(dismissConnection),
modifier = Modifier
.fillMaxSize()
.nestedScroll(dismissConnection)
.focusRequester(focusRequester)
.focusable()
.onKeyEvent(onKeyEvent),
topBar = { NowPlayingTopBar(onClose = { navController.popBackStack() }) },
snackbarHost = { SnackbarHost(snackbarHostState) },
containerColor = Color.Transparent,
@@ -166,11 +182,32 @@ fun NowPlayingScreen(
navController = navController,
viewModel = viewModel,
trackActionsViewModel = trackActionsViewModel,
outputViewModel = outputViewModel,
)
}
}
}
/**
* Null-track guard extracted from [NowPlayingScreen] to keep that
* function under detekt's LongMethod ceiling. Session torn down
* (queue finished + auto-stop, or user cleared the queue from
* elsewhere). Pops back after a short grace delay so a momentary
* null during MediaController IPC bind on cold-mount doesn't flash.
*/
@Composable
private fun NowPlayingNullTrackGuard(
navController: NavHostController,
viewModel: PlayerViewModel,
) {
LaunchedEffect(Unit) {
kotlinx.coroutines.delay(POP_GRACE_MS)
if (viewModel.uiState.value.currentTrack == null) {
navController.popBackStack()
}
}
}
@Composable
private fun dominantGradient(top: Color): Brush {
val base = MaterialTheme.colorScheme.background
@@ -270,6 +307,7 @@ private fun NowPlayingTopBar(onClose: () -> Unit) {
)
}
@Suppress("LongParameterList") // Compose screen wiring — layout args, not logic
@Composable
private fun NowPlayingBody(
inner: androidx.compose.foundation.layout.PaddingValues,
@@ -278,10 +316,10 @@ private fun NowPlayingBody(
navController: NavHostController,
viewModel: PlayerViewModel,
trackActionsViewModel: TrackActionsViewModel,
outputViewModel: OutputPickerViewModel,
) {
val isLiked by trackActionsViewModel.isLikedFlow(track.id)
.collectAsStateWithLifecycle(initialValue = false)
val outputViewModel: OutputPickerViewModel = hiltViewModel()
val routes by outputViewModel.routes.collectAsStateWithLifecycle()
val sheetVisible by outputViewModel.sheetVisible.collectAsStateWithLifecycle()
val permissionDenied = rememberBluetoothPermissionState(sheetVisible)
@@ -387,6 +425,7 @@ private fun PlaybackControlsBlock(
Spacer(Modifier.height(4.dp))
TransportRow(
isPlaying = state.isPlaying,
isUpnpLoading = state.isUpnpLoading,
onPrev = viewModel::skipToPrevious,
onPlayPause = { if (state.isPlaying) viewModel.pause() else viewModel.play() },
onNext = viewModel::skipToNext,
@@ -667,6 +706,7 @@ private fun ScrubTrack(fraction: Float, accent: Color) {
@Composable
private fun TransportRow(
isPlaying: Boolean,
isUpnpLoading: Boolean,
onPrev: () -> Unit,
onPlayPause: () -> Unit,
onNext: () -> Unit,
@@ -685,7 +725,13 @@ private fun TransportRow(
modifier = Modifier.size(TRANSPORT_ICON_DP.dp),
)
}
IconButton(onClick = onPlayPause) {
IconButton(onClick = onPlayPause, enabled = !isUpnpLoading) {
if (isUpnpLoading) {
CircularProgressIndicator(
modifier = Modifier.size(PLAY_PAUSE_ICON_DP.dp),
strokeWidth = 3.dp,
)
} else {
Icon(
imageVector = if (isPlaying) Lucide.Pause else Lucide.Play,
contentDescription = if (isPlaying) "Pause" else "Play",
@@ -693,6 +739,7 @@ private fun TransportRow(
modifier = Modifier.size(PLAY_PAUSE_ICON_DP.dp),
)
}
}
IconButton(onClick = onNext) {
Icon(
imageVector = Lucide.SkipForward,
@@ -703,3 +750,40 @@ private fun TransportRow(
}
}
}
/**
* Returns a key-event handler that intercepts volume-up/down when a UPnP
* route is active and routes the step through [ActiveUpnp.rendering].
* Volume is cached locally so rapid key presses don't each wait on a
* getVolume() round-trip. Returns false (not consumed) for every event
* when no UPnP route is active so the system handles volume normally.
*/
@Composable
private fun rememberUpnpVolumeKeyHandler(activeUpnp: ActiveUpnp?): (KeyEvent) -> Boolean {
val scope = rememberCoroutineScope()
val cache = remember(activeUpnp?.routeId) { VolumeCache() }
return remember(activeUpnp) {
handler@{ event: KeyEvent ->
val rc = activeUpnp?.rendering ?: return@handler false
if (event.type != KeyEventType.KeyDown) return@handler false
val delta = when (event.key) {
Key.VolumeUp -> VOLUME_KEY_STEP
Key.VolumeDown -> -VOLUME_KEY_STEP
else -> return@handler false
}
scope.launch {
val current = cache.value ?: runCatching { rc.getVolume() }.getOrNull() ?: 0
val next = (current + delta).coerceIn(VOLUME_MIN_PERCENT, VOLUME_MAX_PERCENT)
cache.value = next
runCatching { rc.setVolume(next) }
}
true
}
}
}
private class VolumeCache(var value: Int? = null)
private const val VOLUME_KEY_STEP = 5
private const val VOLUME_MIN_PERCENT = 0
private const val VOLUME_MAX_PERCENT = 100
@@ -4,6 +4,7 @@ import androidx.lifecycle.ViewModel
import com.fabledsword.minstrel.player.PlayerController
import com.fabledsword.minstrel.player.PlayerUiState
import dagger.hilt.android.lifecycle.HiltViewModel
import kotlinx.coroutines.flow.SharedFlow
import kotlinx.coroutines.flow.StateFlow
import javax.inject.Inject
@@ -24,6 +25,7 @@ class PlayerViewModel @Inject constructor(
) : ViewModel() {
val uiState: StateFlow<PlayerUiState> = controller.uiState
val dropEvents: SharedFlow<String> = controller.dropEvents
fun play() = controller.play()
fun pause() = controller.pause()
@@ -0,0 +1,71 @@
package com.fabledsword.minstrel.playlists.data
import com.fabledsword.minstrel.models.PlaylistRef
import com.fabledsword.minstrel.player.PlayerController
import com.fabledsword.minstrel.api.ErrorCopy
import kotlinx.coroutines.TimeoutCancellationException
import kotlinx.coroutines.withTimeout
private const val PLAYLIST_FETCH_TIMEOUT_MS = 8_000L
/**
* Fetch a playlist and hand it to the player as a shuffled queue.
*
* Behavior matches the tile play-button contract used on Home and the
* Playlists list: refreshable system playlists go through systemShuffle
* (server-side rotation-aware order; tagging with the variant advances
* rotation), everything else uses refreshDetail. System playlists are
* then client-side shuffled so the tile feels random rather than
* "start at the rotation head". User playlists keep their authored
* order.
*
* Errors and empty mixes surface through [onMessage] for the caller to
* present as a snackbar / toast / etc. Returns when the player has
* accepted the queue (or an error path bailed).
*/
suspend fun playPlaylistShuffled(
playlist: PlaylistRef,
repository: PlaylistsRepository,
player: PlayerController,
onMessage: (String) -> Unit,
) {
val detail = fetchPlaylistDetail(playlist, repository, onMessage) ?: return
val tracks = detail.tracks.toPlayableTrackRefs()
if (tracks.isEmpty()) {
onMessage("Mix isn't ready yet - try again in a moment")
return
}
// Drift #564: bare systemVariant string (not "playlist:<variant>") --
// server's rotation matcher keys on the bare variant.
val source = if (playlist.refreshable) playlist.systemVariant else null
// System playlist tile play button == "pick a random song + shuffle
// the rest" UX. Server's rotation-aware order still drives rotation
// bookkeeping via `source`; the client shuffle just removes the
// deterministic "start at rotation head" feel.
val ordered = if (playlist.isSystem) tracks.shuffled() else tracks
player.setQueue(ordered, initialIndex = 0, source = source)
}
private suspend fun fetchPlaylistDetail(
playlist: PlaylistRef,
repository: PlaylistsRepository,
onMessage: (String) -> Unit,
): PlaylistDetailRef? = try {
withTimeout(PLAYLIST_FETCH_TIMEOUT_MS) {
if (playlist.refreshable && playlist.systemVariant != null) {
repository.systemShuffle(playlist.systemVariant)
} else {
repository.refreshDetail(playlist.id)
}
}
} catch (
@Suppress("SwallowedException") _: TimeoutCancellationException,
) {
onMessage("Couldn't load playlist - check your connection")
null
} catch (
@Suppress("TooGenericExceptionCaught") e: Throwable,
) {
onMessage("Couldn't load playlist: ${ErrorCopy.fromThrowable(e)}")
null
}
@@ -7,6 +7,7 @@ import retrofit2.HttpException
import java.net.HttpURLConnection
import com.fabledsword.minstrel.cache.db.dao.CachedPlaylistDao
import com.fabledsword.minstrel.cache.db.dao.CachedPlaylistTrackDao
import com.fabledsword.minstrel.cache.db.dao.PlaylistCachedCount
import com.fabledsword.minstrel.cache.db.entities.CachedPlaylistEntity
import com.fabledsword.minstrel.cache.db.entities.CachedPlaylistTrackEntity
import com.fabledsword.minstrel.cache.mutations.MutationQueue
@@ -18,6 +19,7 @@ import com.fabledsword.minstrel.models.wire.PlaylistTrackWire
import com.fabledsword.minstrel.models.wire.PlaylistWire
import com.fabledsword.minstrel.shared.resolveServerUrl
import kotlinx.coroutines.flow.Flow
import kotlinx.coroutines.flow.combine
import kotlinx.coroutines.flow.map
import retrofit2.Retrofit
import retrofit2.create
@@ -51,9 +53,16 @@ class PlaylistsRepository @Inject constructor(
// ── Reads (Flow, cache-first; the cache is the source of truth) ──
/** Owned + public; UI splits by isSystem / userId comparison. */
/**
* Owned + public; UI splits by isSystem / userId comparison. Joined with
* per-playlist cache counts so [PlaylistRef.fullyCached] drives the offline
* greying without each consumer re-querying the cache index.
*/
fun observeAll(): Flow<List<PlaylistRef>> =
playlistDao.observeAll().map { rows -> rows.map { it.toDomain() } }
combine(
playlistDao.observeAll(),
playlistDao.observeCachedCounts(),
) { rows, counts -> mergePlaylistsWithCache(rows, counts) }
/** User-owned playlists only (systemVariant IS NULL). */
fun observeUserPlaylists(): Flow<List<PlaylistRef>> =
@@ -233,6 +242,23 @@ fun List<PlaylistTrackRef>.toPlayableTrackRefs(): List<TrackRef> =
// ── Mappers (internal — keep Room + wire types out of the UI layer) ──
/**
* Maps cached playlist rows to domain refs, stamping [PlaylistRef.fullyCached]
* from the cache-index counts. A playlist is fully cached when it has tracks
* and every member track is resident (`cachedCount >= trackCount`). Pulled out
* of the Flow so the predicate is unit-testable without a Room database.
*/
internal fun mergePlaylistsWithCache(
rows: List<CachedPlaylistEntity>,
counts: List<PlaylistCachedCount>,
): List<PlaylistRef> {
val cachedById = counts.associate { it.playlistId to it.cachedCount }
return rows.map { row ->
val cached = cachedById[row.id] ?: 0
row.toDomain().copy(fullyCached = row.trackCount > 0 && cached >= row.trackCount)
}
}
private fun CachedPlaylistEntity.toDomain(): PlaylistRef =
PlaylistRef(
id = id,
@@ -28,15 +28,17 @@ import androidx.compose.material3.IconButton
import androidx.compose.material3.MaterialTheme
import androidx.compose.material3.OutlinedButton
import androidx.compose.material3.Scaffold
import androidx.compose.material3.SnackbarDuration
import androidx.compose.material3.SnackbarHost
import androidx.compose.material3.SnackbarHostState
import androidx.compose.material3.SnackbarResult
import androidx.compose.material3.Text
import androidx.compose.material3.TopAppBar
import androidx.compose.runtime.Composable
import androidx.compose.runtime.LaunchedEffect
import androidx.compose.runtime.collectAsState
import androidx.compose.runtime.getValue
import androidx.compose.runtime.mutableStateOf
import androidx.compose.runtime.remember
import androidx.compose.runtime.setValue
import androidx.compose.ui.Alignment
import androidx.compose.ui.Modifier
import androidx.compose.ui.draw.clip
@@ -56,6 +58,8 @@ import com.composables.icons.lucide.Play
import com.composables.icons.lucide.RefreshCw
import com.composables.icons.lucide.Shuffle
import com.fabledsword.minstrel.api.ErrorCopy
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.recoveries
import com.fabledsword.minstrel.events.EventsStream
import com.fabledsword.minstrel.events.LiveEvent
import com.fabledsword.minstrel.models.PlaylistRef
@@ -72,6 +76,7 @@ import com.fabledsword.minstrel.playlists.data.PlaylistsRepository
import com.fabledsword.minstrel.playlists.data.toPlayableTrackRefs
import com.fabledsword.minstrel.shared.formatDuration
import com.fabledsword.minstrel.shared.widgets.TrackRow
import com.fabledsword.minstrel.shared.widgets.ErrorRetry
import com.fabledsword.minstrel.shared.widgets.LikeButton
import com.fabledsword.minstrel.shared.widgets.PullToRefreshScaffold
import com.fabledsword.minstrel.shared.widgets.SkeletonTrackRow
@@ -113,6 +118,7 @@ class PlaylistDetailViewModel @Inject constructor(
private val player: PlayerController,
private val eventsStream: EventsStream,
private val seedCache: DetailSeedCache,
networkStatus: NetworkStatusController,
savedStateHandle: SavedStateHandle,
) : ViewModel() {
@@ -139,6 +145,15 @@ class PlaylistDetailViewModel @Inject constructor(
private val regeneratedChannel = Channel<String>(Channel.BUFFERED)
val regenerated: Flow<String> = regeneratedChannel.receiveAsFlow()
/**
* #980: a system-playlist rebuild landed (SSE) while this screen is open —
* its list is now stale and the uuid has rotated. Drives a "this mix was
* refreshed → Refresh" snackbar rather than yanking the user; Refresh
* re-resolves the variant via [reloadRebuilt].
*/
private val staleInternal = MutableStateFlow(false)
val stale: StateFlow<Boolean> = staleInternal.asStateFlow()
val likedTrackIds: StateFlow<Set<String>> =
likes.observeLikedTrackIds()
.stateIn(
@@ -154,6 +169,14 @@ class PlaylistDetailViewModel @Inject constructor(
.filter { it.kind.startsWith("playlist.") }
.collect { handlePlaylistEvent(it) }
}
// Screen-level auto-recovery (issue #1245): a detail that failed to
// load while the server was unreachable re-fetches when health
// returns, instead of waiting for the Retry tap.
viewModelScope.launch {
networkStatus.recoveries().collect {
if (internal.value is PlaylistDetailUiState.Error) refresh()
}
}
}
/**
@@ -163,6 +186,12 @@ class PlaylistDetailViewModel @Inject constructor(
* ignored — the playlists-list screen handles those.
*/
private fun handlePlaylistEvent(event: LiveEvent) {
// #980: a system rebuild carries no playlist_id (it rebuilds all of the
// user's system mixes at once) — handle it before the per-id filter.
if (event.kind == "playlist.system_rebuilt") {
markStaleIfSystem()
return
}
val eventPlaylistId = event.data["playlist_id"]?.jsonPrimitive?.contentOrNull
if (eventPlaylistId != playlistId) return
when (event.kind) {
@@ -171,6 +200,14 @@ class PlaylistDetailViewModel @Inject constructor(
}
}
private fun markStaleIfSystem() {
val playlist = (internal.value as? PlaylistDetailUiState.Success)
?.detail?.playlist ?: return
if (playlist.systemVariant?.takeIf { playlist.refreshable } != null) {
staleInternal.value = true
}
}
fun toggleLikeTrack(trackId: String) {
val desired = trackId !in likedTrackIds.value
viewModelScope.launch {
@@ -220,6 +257,30 @@ class PlaylistDetailViewModel @Inject constructor(
}
}
/**
* #980: user tapped Refresh on the stale-view snackbar. The rebuild
* already ran server-side (uuid rotated) — re-resolve the variant to the
* fresh playlist id and hand it to [regenerated] for navigate-replace.
* Unlike [regenerate] this does NOT trigger another rebuild.
*/
fun reloadRebuilt() {
val playlist = (internal.value as? PlaylistDetailUiState.Success)
?.detail?.playlist ?: return
val variant = playlist.systemVariant?.takeIf { playlist.refreshable } ?: return
staleInternal.value = false
viewModelScope.launch {
runCatching { repository.systemShuffle(variant).playlist.id }
.onSuccess { newId ->
if (newId.isNotEmpty()) regeneratedChannel.trySend(newId)
}
}
}
/** #980: user dismissed the stale-view snackbar without refreshing. */
fun dismissStale() {
staleInternal.value = false
}
/** Play the available tracks starting at [startTrackId] (or first available). */
fun play(tracks: List<PlaylistTrackRef>, startTrackId: String?) {
val refs = tracks.toPlayableTrackRefs()
@@ -261,8 +322,25 @@ fun PlaylistDetailScreen(
}
}
}
// #980: a rebuild landed while this system-playlist screen is open. Offer a
// non-intrusive "refreshed → Refresh" snackbar; Refresh re-resolves the
// rotated uuid (via the regenerated navigate-replace flow).
val snackbarHostState = remember { SnackbarHostState() }
val stale by viewModel.stale.collectAsState()
LaunchedEffect(stale) {
if (stale) {
val result = snackbarHostState.showSnackbar(
message = "This mix was refreshed",
actionLabel = "Refresh",
duration = SnackbarDuration.Indefinite,
)
if (result == SnackbarResult.ActionPerformed) viewModel.reloadRebuilt()
else viewModel.dismissStale()
}
}
Scaffold(
modifier = Modifier.fillMaxSize(),
snackbarHost = { SnackbarHost(snackbarHostState) },
topBar = {
TopAppBar(
title = { Text(currentTitle(state)) },
@@ -301,7 +379,11 @@ private fun PlaylistDetailContent(
Crossfade(targetState = state::class, label = "playlist-detail") { _ -> when (val s = state) {
is PlaylistDetailUiState.Loading ->
s.seed?.let { SeededPlaylistLoading(it) } ?: SkeletonPlaylistTrackList()
is PlaylistDetailUiState.Error -> ErrorBlock(s.message, viewModel::refresh)
is PlaylistDetailUiState.Error -> ErrorRetry(
title = "Couldn't load playlist",
message = s.message,
onRetry = { viewModel.refresh() },
)
is PlaylistDetailUiState.Success -> {
val likedTrackIds by viewModel.likedTrackIds.collectAsState()
val playerState by playerViewModel.uiState.collectAsState()
@@ -585,33 +667,6 @@ private fun SeededPlaylistLoading(seed: PlaylistRef) {
}
}
@Composable
private fun ErrorBlock(message: String, onRetry: () -> Unit) {
var retried by remember { mutableStateOf(false) }
Column(
modifier = Modifier
.fillMaxSize()
.padding(24.dp),
verticalArrangement = Arrangement.Center,
horizontalAlignment = Alignment.CenterHorizontally,
) {
Text(
text = message,
color = MaterialTheme.colorScheme.error,
style = MaterialTheme.typography.bodyLarge,
)
Spacer(Modifier.height(12.dp))
OutlinedButton(
onClick = {
if (!retried) {
retried = true
onRetry()
}
},
) { Text("Retry") }
}
}
// ─── Helpers ─────────────────────────────────────────────────────────
private const val UNAVAILABLE_ALPHA = 0.4f
@@ -13,9 +13,13 @@ import androidx.compose.foundation.lazy.grid.LazyVerticalGrid
import androidx.compose.foundation.lazy.grid.items
import androidx.compose.material3.MaterialTheme
import androidx.compose.material3.Scaffold
import androidx.compose.material3.SnackbarHost
import androidx.compose.material3.SnackbarHostState
import androidx.compose.material3.Text
import androidx.compose.runtime.Composable
import androidx.compose.runtime.LaunchedEffect
import androidx.compose.runtime.getValue
import androidx.compose.runtime.remember
import androidx.compose.ui.Modifier
import androidx.compose.ui.unit.dp
import androidx.hilt.navigation.compose.hiltViewModel
@@ -23,26 +27,42 @@ import androidx.lifecycle.ViewModel
import androidx.lifecycle.compose.collectAsStateWithLifecycle
import androidx.lifecycle.viewModelScope
import androidx.navigation.NavHostController
import com.fabledsword.minstrel.api.ErrorCopy
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.ServerHealth
import com.fabledsword.minstrel.connectivity.recoveries
import com.fabledsword.minstrel.events.EventsStream
import com.fabledsword.minstrel.models.PlaylistRef
import com.fabledsword.minstrel.nav.PlaylistDetail
import com.fabledsword.minstrel.nav.Playlists
import com.fabledsword.minstrel.player.PlayerController
import com.fabledsword.minstrel.playlists.data.PlaylistsRepository
import com.fabledsword.minstrel.playlists.data.playPlaylistShuffled
import com.fabledsword.minstrel.shared.UiState
import com.fabledsword.minstrel.shared.asCacheFirstStateFlow
import com.fabledsword.minstrel.playlists.widgets.PlaylistCard
import com.fabledsword.minstrel.shared.widgets.EmptyState
import com.fabledsword.minstrel.shared.widgets.ErrorRetry
import com.fabledsword.minstrel.shared.widgets.LoadingCentered
import com.fabledsword.minstrel.shared.widgets.MinstrelTopAppBar
import com.fabledsword.minstrel.shared.widgets.PullToRefreshScaffold
import dagger.hilt.android.lifecycle.HiltViewModel
import kotlinx.coroutines.Job
import kotlinx.coroutines.channels.Channel
import kotlinx.coroutines.flow.Flow
import kotlinx.coroutines.flow.MutableStateFlow
import kotlinx.coroutines.flow.SharingStarted
import kotlinx.coroutines.flow.StateFlow
import kotlinx.coroutines.flow.combine
import kotlinx.coroutines.flow.filter
import kotlinx.coroutines.flow.map
import kotlinx.coroutines.flow.receiveAsFlow
import kotlinx.coroutines.flow.stateIn
import kotlinx.coroutines.launch
import javax.inject.Inject
private const val SHARE_STOP_TIMEOUT_MS = 5_000L
// ─── State ───────────────────────────────────────────────────────────
// ─── ViewModel ───────────────────────────────────────────────────────
@@ -50,9 +70,28 @@ import javax.inject.Inject
@HiltViewModel
class PlaylistsListViewModel @Inject constructor(
private val repository: PlaylistsRepository,
private val player: PlayerController,
private val eventsStream: EventsStream,
networkStatus: NetworkStatusController,
) : ViewModel() {
private val poolMessages = Channel<String>(Channel.BUFFERED)
/** Transient snackbar messages from playlist tile play taps. */
val transientMessages: Flow<String> = poolMessages.receiveAsFlow()
/** Cache-only: no link OR server unreachable (Unstable stays calm). Greys tiles. */
val offline: StateFlow<Boolean> = networkStatus.state
.map { it == ServerHealth.Offline || it == ServerHealth.ServerDown }
.stateIn(
scope = viewModelScope,
started = SharingStarted.WhileSubscribed(SHARE_STOP_TIMEOUT_MS),
initialValue = false,
)
/** Copy for the most recent list-refresh failure; null once one succeeds. */
private val refreshError = MutableStateFlow<String?>(null)
init {
refresh()
// Live updates: a playlist created/updated/deleted from another
@@ -62,20 +101,37 @@ class PlaylistsListViewModel @Inject constructor(
.filter { it.kind.startsWith("playlist.") }
.collect { refresh() }
}
// Screen-level auto-recovery (issue #1245): re-pull when server
// health returns, so a load that failed offline heals unprompted.
viewModelScope.launch {
networkStatus.recoveries().collect { refresh() }
}
}
/** Re-pull the playlists list. Returns the Job for pull-to-refresh awaits. */
fun refresh(): Job = viewModelScope.launch {
refreshError.value = null
runCatching { repository.refreshList() }
.onFailure { refreshError.value = ErrorCopy.fromThrowable(it) }
}
/** Tile play button: shuffle the playlist's tracks and start at index 0. */
suspend fun playPlaylist(playlist: PlaylistRef) {
viewModelScope.launch {
playPlaylistShuffled(playlist, repository, player) {
poolMessages.trySend(it)
}
}.join()
}
val uiState: StateFlow<UiState<List<PlaylistRef>>> =
repository.observeAll()
.map { list ->
if (list.isEmpty()) {
UiState.Empty
} else {
UiState.Success(list)
combine(repository.observeAll(), refreshError) { list, err ->
when {
list.isNotEmpty() -> UiState.Success(list)
// Empty cache + failed refresh is a load failure, not "no
// playlists yet" — surface Error so the user gets a Retry.
err != null -> UiState.Error(err)
else -> UiState.Empty
}
}
.asCacheFirstStateFlow(viewModelScope)
@@ -88,6 +144,10 @@ fun PlaylistsListScreen(
navController: NavHostController,
viewModel: PlaylistsListViewModel = hiltViewModel(),
) {
val snackbar = remember { SnackbarHostState() }
LaunchedEffect(Unit) {
viewModel.transientMessages.collect { snackbar.showSnackbar(it) }
}
Scaffold(
modifier = Modifier.fillMaxSize(),
topBar = {
@@ -97,8 +157,10 @@ fun PlaylistsListScreen(
currentRouteName = Playlists::class.qualifiedName,
)
},
snackbarHost = { SnackbarHost(snackbar) },
) { inner ->
val state by viewModel.uiState.collectAsStateWithLifecycle()
val offline by viewModel.offline.collectAsStateWithLifecycle()
PullToRefreshScaffold(
onRefresh = { viewModel.refresh().join() },
modifier = Modifier.fillMaxSize().padding(inner),
@@ -111,13 +173,16 @@ fun PlaylistsListScreen(
"appear once your library has enough plays. Create your " +
"own playlist from any album or track in the meantime.",
)
is UiState.Error -> EmptyState(
is UiState.Error -> ErrorRetry(
title = "Couldn't load playlists",
body = s.message,
message = s.message,
onRetry = { viewModel.refresh() },
)
is UiState.Success -> PlaylistsGrid(
playlists = s.data,
offline = offline,
onPlaylistClick = { id -> navController.navigate(PlaylistDetail(id)) },
onPlay = viewModel::playPlaylist,
)
}
}
@@ -127,7 +192,9 @@ fun PlaylistsListScreen(
@Composable
private fun PlaylistsGrid(
playlists: List<PlaylistRef>,
offline: Boolean,
onPlaylistClick: (String) -> Unit,
onPlay: suspend (PlaylistRef) -> Unit,
) {
val systemPlaylists = playlists.filter { it.isSystem }
val userPlaylists = playlists.filter { !it.isSystem }
@@ -142,7 +209,16 @@ private fun PlaylistsGrid(
SectionHeader("System playlists")
}
items(items = systemPlaylists, key = { it.id }) { playlist ->
PlaylistCard(playlist = playlist, onClick = { onPlaylistClick(playlist.id) })
// Greyed offline when it needs the live server or isn't fully
// cached — dimmed but still tappable into the detail.
val greyed = offline && playlist.unavailableOffline
PlaylistCard(
playlist = playlist,
onClick = { onPlaylistClick(playlist.id) },
onPlay = { onPlay(playlist) },
playEnabled = playlist.trackCount > 0 && !greyed,
greyed = greyed,
)
}
}
if (userPlaylists.isNotEmpty()) {
@@ -150,7 +226,14 @@ private fun PlaylistsGrid(
SectionHeader("Your playlists")
}
items(items = userPlaylists, key = { it.id }) { playlist ->
PlaylistCard(playlist = playlist, onClick = { onPlaylistClick(playlist.id) })
val greyed = offline && playlist.unavailableOffline
PlaylistCard(
playlist = playlist,
onClick = { onPlaylistClick(playlist.id) },
onPlay = { onPlay(playlist) },
playEnabled = playlist.trackCount > 0 && !greyed,
greyed = greyed,
)
}
}
}
@@ -15,6 +15,7 @@ import androidx.compose.material3.Text
import androidx.compose.runtime.Composable
import androidx.compose.ui.Alignment
import androidx.compose.ui.Modifier
import androidx.compose.ui.draw.alpha
import androidx.compose.ui.graphics.Color
import androidx.compose.ui.text.style.TextOverflow
import androidx.compose.ui.unit.dp
@@ -40,6 +41,10 @@ import com.fabledsword.minstrel.theme.FabledSwordFlatTokens
* [playEnabled] disables the overlay (50% alpha + ignore taps) — Home
* uses it for system playlists in offline mode (their server-side
* shuffle endpoint is unreachable) and for empty playlists.
*
* [greyed] dims the whole tile (it can't be reliably played offline) while
* keeping it tappable — the detail screen is the escape hatch for shuffling
* whatever subset is cached. Greyed always disables the play overlay too.
*/
@Composable
fun PlaylistCard(
@@ -48,11 +53,13 @@ fun PlaylistCard(
modifier: Modifier = Modifier,
onPlay: (suspend () -> Unit)? = null,
playEnabled: Boolean = true,
greyed: Boolean = false,
) {
val seedCache = LocalDetailSeedCache.current
Surface(
modifier = modifier
.width(176.dp)
.alpha(if (greyed) GREYED_ALPHA else 1f) // dimmed, still clickable
.clickable {
seedCache.stashPlaylist(playlist)
onClick()
@@ -63,7 +70,7 @@ fun PlaylistCard(
PlaylistCardCover(
playlist = playlist,
onPlay = onPlay,
playEnabled = playEnabled,
playEnabled = playEnabled && !greyed,
)
Spacer(Modifier.height(8.dp))
Text(
@@ -148,6 +155,7 @@ private fun VariantPill(label: String, modifier: Modifier = Modifier) {
}
private const val PILL_BG_ALPHA = 0.85f
private const val GREYED_ALPHA = 0.45f
private fun subtitleFor(playlist: PlaylistRef): String = when {
playlist.trackCount > 0 -> "${playlist.trackCount} tracks"
@@ -27,6 +27,7 @@ import com.composables.icons.lucide.Lucide
import com.fabledsword.minstrel.models.QuarantineRef
import com.fabledsword.minstrel.shared.UiState
import com.fabledsword.minstrel.shared.widgets.EmptyState
import com.fabledsword.minstrel.shared.widgets.ErrorRetry
import com.fabledsword.minstrel.shared.widgets.LoadingCentered
import com.fabledsword.minstrel.shared.widgets.PullToRefreshScaffold
import com.fabledsword.minstrel.shared.widgets.TrackCoverThumb
@@ -43,9 +44,10 @@ fun HiddenTab(viewModel: HiddenTabViewModel = hiltViewModel()) {
"and it'll show up here. The track stays out of system " +
"playlists until you unhide it.",
)
is UiState.Error -> EmptyState(
is UiState.Error -> ErrorRetry(
title = "Couldn't load hidden tracks",
body = s.message,
message = s.message,
onRetry = { viewModel.refresh() },
)
is UiState.Success -> HiddenList(
rows = s.data,
@@ -3,6 +3,8 @@ package com.fabledsword.minstrel.quarantine.ui
import androidx.lifecycle.ViewModel
import androidx.lifecycle.viewModelScope
import com.fabledsword.minstrel.api.ErrorCopy
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.recoveries
import com.fabledsword.minstrel.events.EventsStream
import com.fabledsword.minstrel.models.QuarantineRef
import com.fabledsword.minstrel.shared.UiState
@@ -24,6 +26,7 @@ private const val SHARE_STOP_TIMEOUT_MS = 5_000L
class HiddenTabViewModel @Inject constructor(
private val repository: QuarantineRepository,
private val eventsStream: EventsStream,
networkStatus: NetworkStatusController,
) : ViewModel() {
private val refreshError = MutableStateFlow<String?>(null)
@@ -61,6 +64,11 @@ class HiddenTabViewModel @Inject constructor(
.filter { it.kind.startsWith("quarantine.") }
.collect { refresh() }
}
// Screen-level auto-recovery (issue #1245): re-pull when server
// health returns, so a load that failed offline heals unprompted.
viewModelScope.launch {
networkStatus.recoveries().collect { refresh() }
}
}
fun refresh(): Job = viewModelScope.launch {
@@ -46,6 +46,7 @@ import com.fabledsword.minstrel.shared.UiState
import com.fabledsword.minstrel.nav.ArtistDetail
import com.fabledsword.minstrel.nav.Requests
import com.fabledsword.minstrel.shared.widgets.EmptyState
import com.fabledsword.minstrel.shared.widgets.ErrorRetry
import com.fabledsword.minstrel.shared.widgets.LoadingCentered
import com.fabledsword.minstrel.shared.widgets.MinstrelTopAppBar
import com.fabledsword.minstrel.shared.widgets.PullToRefreshScaffold
@@ -78,9 +79,10 @@ fun RequestsScreen(
body = "Use Discover to ask Lidarr for new music; your " +
"requests show up here.",
)
is UiState.Error -> EmptyState(
is UiState.Error -> ErrorRetry(
title = "Couldn't load requests",
body = s.message,
message = s.message,
onRetry = { viewModel.refresh() },
)
is UiState.Success -> RequestList(
rows = s.data,
@@ -3,13 +3,17 @@ package com.fabledsword.minstrel.requests.ui
import androidx.lifecycle.ViewModel
import androidx.lifecycle.viewModelScope
import com.fabledsword.minstrel.api.ErrorCopy
import com.fabledsword.minstrel.connectivity.NetworkStatusController
import com.fabledsword.minstrel.connectivity.recoveries
import com.fabledsword.minstrel.events.EventsStream
import com.fabledsword.minstrel.models.RequestRef
import com.fabledsword.minstrel.models.RequestStatus
import com.fabledsword.minstrel.requests.data.CancelOutcome
import com.fabledsword.minstrel.requests.data.RequestsRepository
import com.fabledsword.minstrel.shared.UiState
import dagger.hilt.android.lifecycle.HiltViewModel
import kotlinx.coroutines.Job
import kotlinx.coroutines.delay
import kotlinx.coroutines.flow.MutableStateFlow
import kotlinx.coroutines.flow.StateFlow
import kotlinx.coroutines.flow.asStateFlow
@@ -20,10 +24,15 @@ import javax.inject.Inject
private val RELEVANT_EVENT_KINDS = setOf("request.status_changed")
// Cadence for in-flight request polling — matches the web client (#369):
// fast enough to feel live, slow enough not to hammer the server.
private const val POLL_INTERVAL_MS = 12_000L
@HiltViewModel
class RequestsViewModel @Inject constructor(
private val repository: RequestsRepository,
private val eventsStream: EventsStream,
networkStatus: NetworkStatusController,
) : ViewModel() {
private val internal = MutableStateFlow<UiState<List<RequestRef>>>(UiState.Loading)
@@ -31,10 +40,21 @@ class RequestsViewModel @Inject constructor(
init {
refresh()
// SSE push: reconciler completions arrive here; reload silently so the
// row updates in place rather than flashing the loading spinner.
viewModelScope.launch {
eventsStream.events
.filter { it.kind in RELEVANT_EVENT_KINDS }
.collect { refresh() }
.collect { silentReload() }
}
viewModelScope.launch { pollWhileInFlight() }
// Screen-level auto-recovery (issue #1245): a load that failed while
// the server was unreachable heals the moment health returns, without
// waiting for the next poll tick.
viewModelScope.launch {
networkStatus.recoveries().collect {
if (internal.value is UiState.Error) silentReload()
}
}
}
@@ -54,6 +74,47 @@ class RequestsViewModel @Inject constructor(
}
}
/**
* Auto-poll while any request is mid-ingest (status APPROVED), matching the
* web client (#369). Reloads silently every [POLL_INTERVAL_MS] and pauses
* when nothing is in-flight — complements the SSE push so progress stays
* live even if the event stream drops.
*/
private suspend fun pollWhileInFlight() {
while (true) {
// Also retry from a failed load (issue #1245): from UiState.Error
// the in-flight predicate can never become true, so without this
// the poll never self-heals the screen — e.g. after a transient
// 500 that health monitoring (transport-level) never saw.
if (hasInFlightRequest() || internal.value is UiState.Error) {
silentReload()
}
delay(POLL_INTERVAL_MS)
}
}
private fun hasInFlightRequest(): Boolean {
val state = internal.value
return state is UiState.Success &&
state.data.any { it.status == RequestStatus.APPROVED }
}
/**
* Reload the list in place without flipping to Loading — used by the poll
* loop and the SSE collector so visible rows update without a spinner
* flash. A transient failure keeps the current view; the next reload wins.
*/
private suspend fun silentReload() {
try {
val rows = repository.listMine()
internal.value = if (rows.isEmpty()) UiState.Empty else UiState.Success(rows)
} catch (
@Suppress("TooGenericExceptionCaught", "SwallowedException") e: Throwable,
) {
// Keep the current view; the next poll / SSE / pull reconciles.
}
}
/**
* Optimistically removes [id] from the success list so the row
* disappears immediately; on server-side success the row is

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