Attempt to fix macOS tray menu items doing nothing: a bare `cargo run`
binary is not a bundled .app, so macOS does not dispatch NSStatusItem
menu-item target-actions (the menu shows, items enabled, but clicks fire
nothing). Force the app to a Regular, active application at tray creation
via objc2-app-kit setActivationPolicy(.regular) + activate(), and switch
menu handling to MenuEvent::set_event_handler with a diagnostic log.
NOTE: NOT verified at runtime on macOS yet (pending the user's GUI test). If
the menu still does nothing, look for the '[tray] menu item activated' log
line to tell whether the handler fires.
cc
Add the NSStatusItem (macOS) / Shell_NotifyIcon (Windows) tray via the
tray-icon crate: the same 5-item menu, left-click window toggle, a themed
icon (auto uses a macOS template image that adapts to the menu bar), a
static QBZ tooltip, and live icon-variant switching. tray_icon::TrayIcon is
!Send and (on macOS) must be built on the main thread with NSApplication
running, so it lives in a thread_local created via invoke_from_event_loop;
its events are drained from a slint::Timer.
macOS close-to-tray reuses the cross-platform Slint hide/show path. The
opt-in Hide Dock icon toggle switches the NSApplication activation policy to
.accessory while closed to the menu bar (.regular by default, Spotify-style)
via objc2-app-kit.
cc
Port the Tauri tray to Slint on Linux via ksni: themed icon (auto/mono/
color), left-click window toggle, middle-click play/pause, scroll volume,
the 5-item menu, and a live tooltip reflecting the now-playing track. Tray
actions call the playback controller directly (QConnect-aware) instead of
emitting webview events. Settings reuse the shared qbz-app tray store
(per-user) wired through AppearanceState.
Window hide/show uses Slint window.hide()/show() (winit set_visible is a
no-op on Wayland) and the loop runs via run_event_loop_until_quit so a
tray-hide keeps the app alive. close-to-tray is honored on both the custom
titlebar close and the WM close (on_close_requested). minimize-to-tray is
hidden until custom window chrome exists.
cc
Add a mac_hide_dock column (with migration) to the shared tray settings
store for the upcoming macOS close-to-tray Dock-icon toggle, and flip the
close_to_tray default to true (opt-out: closing sends the window to the
tray and keeps the app running).
cc
Add drag-and-drop reordering to the Queue sidebar: a per-row drag handle, a
floating drag ghost, and a drop indicator. When a QConnect transport is
connected the reorder is routed to the cloud (CtrlSrvrQueueReorderTracks) so
order stays WS-authoritative; otherwise it moves the local queue track via
core().move_track. The Flickable is non-interactive so a stale pressed_time
cannot strand the gesture and leave phantom scroll after release.
cc
The take-renderer-when-idle path never fired: capture_session_state_takeover_input
parsed active_renderer_id raw. The cloud encodes "nobody active" as -1, which was
read as Some(-1) and classified as a peer being active (OtherPaused), so
compute_connection_state returned should_set_active_renderer: false and the
auto-take was suppressed. Normalize identically to the apply path
(normalize_active_renderer_id filters >= 0 -> None) so -1 maps to
ServerActiveState::None and the idle auto-take claims the render. Adds the
auto_take_when_idle decision to compute_connection_state plus regression tests
for the -1 / active-peer classification.
cc
Local toggle_play always called resume(), which fails with "cannot resume -
no audio data available" when the player holds no loaded stream — e.g. a
freshly materialized QConnect renderer queue whose cursor sits on a track that
was never loaded. Detect the no-loaded-audio case via Player::has_loaded_audio()
and load+play the current queue track instead of resuming. A normal pause keeps
the stream loaded, so the pause/resume path is unchanged.
cc
Shuffle and Repeat now go to the CLOUD whenever connected (renderer OR controller),
gating on transport_connected like Tauri's v2_toggle_shuffle / v2_set_repeat_mode,
instead of the old peer-active gate that fell through to a LOCAL reshuffle when QBZ
was the active renderer — that did nothing visible AND risked a divergent local
order. QBZ sends only the seed + pivot; the cloud owns the order; the toggle reads
the cloud-authoritative queue.shuffle_mode. The local shuffle path is now reachable
only when NOT connected.
- shuffle/repeat button reflects the cloud queue flag (queue.shuffle_mode for the
controller branch, core queue state via refresh_transport_modes for the renderer
branch), not the per-renderer field the cloud never populates for a peer — fixes
the iOS-reflects / WebPlayer-doesn't asymmetry.
- pause/play to a peer renderer: send a BARE SetPlayerState (playing_state only).
iOS reports current_queue_item_id=null while playing, so resolving the qid landed
on queue_item_id 0 — the protocol's "unknown" placeholder — and iOS could not
anchor the pause, auto-resuming. A bare transport toggle is spec-sanctioned: the
renderer pauses/resumes its own current item in place.
- skip_serializing_if on the SetPlayerState request structs so a None field is
OMITTED, not serialized as null; and the mapper now treats a null current_queue_item
as absent. Without this the bare command failed to encode and was never sent.
- log the outbound SetPlayerState payload for controller diagnostics.
cc
Version bump across Cargo/tauri/npm/snap/flake, metainfo release entry,
and TAG_DETAILS for 1.2.15. flake.nix hashes cleared (filled post-tag
via nix build once v1.2.15 exists).
cc
Discover and other image-heavy views created a fresh
reqwest::blocking::Client per image (and per notification artwork),
exhausting the process file-descriptor limit and crashing the app with
"Too many open files". Reuse a single shared client across downloads.
cc
Startup blocked on Qobuz's ~7 MB web bundle download with no cache or
timeout, so a slow CDN left the app on a blank screen for 25-30s before
the login screen appeared. Cache the extracted tokens on disk (keyed by
bundle version), apply them instantly on warm starts and refresh in the
background, and bound the cold fetch with a per-request timeout + retry.
The legacy api client shares the same cache. Emits BundleFetchStarted so
the UI can show a connecting state on the first run.
cc
- badge: gate is-remote on transport_connected so it clears on every session-end
path (transport drop / watchdog / reconnect-exhausted), not only toggle-off
- multi-track: route a whole album/playlist/favorites enqueue to the peer in one
QueueAddTracks/QueueInsertTracks (all-or-nothing admission), not single-track only
- shuffle inbound: SetShuffleMode renderer arm is flag-only (set_shuffle_flag) so the
renderer never invents a local order; the cloud's echoed order is authoritative
- shuffle outbound: send shuffle_seed + shuffle_pivot_queue_item_id (the cloud needs
the seed to build the order); hoist resolve_qconnect_shuffle_pivot to qconnect-app
- start index: resolve_remote_start_index validates the track_id before trusting a
qid, so a freshly-pushed album no longer inherits a stale cursor (wrong NowPlayingBar track)
- history: play_index records history only when actually moving, so the current track
is no longer duplicated in the History tab
- pause-to-peer: resolve current_queue_item from the core-aligned cursor instead of the
stale renderer snapshot, so pausing an iOS renderer (reports null qid) is accepted
cc
Restore ListScrollbar, TrackRow and AlbumPageView to their pre-scrollbar
state (always-visible scrollbars). The per-row hover-reveal only revealed
the bar over track rows, not over the whole scrollable viewport, so on
mixed-content views (Artist, Label) it felt incomplete. Card sizing
(Album/Playlist/Radio at 200px) is kept.
cc
- RadioCard enlarged to 200x246 to match Album/Playlist cards; ForYouView +
Spotlight row-width math updated to 200.
- Scrollbar hover-reveal: ListScrollbar gains a reveal prop (default true, no
regression) and fades on opacity; TrackRow exposes hover-changed; AlbumPageView
wired as the verified template (row-hover count -> reveal).
cc
AlbumCard and PlaylistCard are now 200x246 (was 220x266), with the larger
texts -2px and the smaller ones -1px. All album/playlist carousels and grids
follow (card-width/height 200/246) so nothing reflows: Carousel,
PlaylistCarousel, AlbumCollectionView/AlbumGrid, ReleaseGrid, Spotlight,
HomeSkeleton, DiscoverBrowseView, SearchResultsView, FavoritesView. Artist /
radio cards are untouched.
cc
NowPlayingBar / SongCard:
- Audio stamp homologated New<->Classic + vertically centred (alignment:center
on the stamp column; New had a binding-loop card height, now explicit).
- Dot-LED output indicators (circle-dot/circle) in both modes; render-target
badge moved into the LED row when a peer/cast device owns playback.
- Right-cluster buttons made square (32x32); context glyph -> disc-album,
flush-left; classic stamp right margin; quality tooltip = the quality text;
heart tooltip = Add to library; Qobuz Connect tooltip always shows On/Off.
Qobuz Connect:
- Clear the session topology on disconnect so a late event cannot resurrect
the golden render badge (badge-persists-after-disconnect fix).
Queue / Lyrics right panel:
- Queue and Lyrics can both be open at once, stacked vertically in one column
(independent queue-open/lyrics-open flags replace the exclusive right-panel
int); each closable from its bar button and its own X.
- Panel width 280; compact rows (zebra), now-playing card homologated with a
row, flexible footer search, Lyrics title sized like the Queue tabs.
Shell layout:
- Pin AppShell to the window and size the content area arithmetically
(window - header - player-bar - padding). A Slint VerticalLayout ignores an
explicit height and does not auto-fill its parent Rectangle, so the content
row collapsed once the right panels became conditionally mounted; a Rectangle
takes the explicit height and it cascades up. Bar height references the single
Layout.player-bar-height source (note left for the future Small/Large bars).
cc
When QBZ becomes the active renderer (takeback), force a stream of the
current track and resume at the peer's position instead of failing or
restarting from 0:
- add QconnectRendererEngine::has_loaded_audio so the orchestration can
tell a stale track_id left by stop() from audio that is actually
buffered. stop() clears the buffer but leaves current_track_id, so the
track-id guard alone skipped the reload and the next resume hit
no audio data available.
- force_remote_track_stream on SetActive(true): reload even when the
track id matches (unless already streaming it), at the handed-off
position.
- thread the SetState position into ensure_remote_track_loaded so the
first takeback of a session (whose load lands in the SetState path)
also resumes at the peer's position rather than streaming from 0 and
issuing a forward seek the audio thread ignores past the watermark.
cc
deferred_renderer_join joined every connect with is_active=true, so QBZ
grabbed playback from whatever peer was rendering the instant it came
online, and the self-state echo from the post-join AskForRendererState
then reset the cursor to the queue head. Join as an available renderer
(is_active=false) on a fresh connect; only a post-drop RECONNECTION
rejoins as active, so a network blip mid-render does not lose the render.
Taking over is now explicit (device-picker Play here, or the phone
selecting QBZ, both arriving as SET_ACTIVE) - matching the official
clients.
cc
slint-dev.sh / slint-dev-fast.sh assume a Linux toolchain (mold linker,
x86 AES-NI features). On macOS clang rejects -fuse-ld=mold and the build
fails at the first link. This adds slint-dev-mac.sh: same nightly parallel
frontend and release-by-default intent, with --fast for a quick debug
build, the AES-NI/SSSE3 features scoped to x86_64 only, and the default
Apple linker (ld-prime). Verified on Apple Silicon: builds, runs, plays audio.
From the controller-mode debug session. All peer-facing paths gate on
is_peer_renderer_active, so renderer / local / disconnected playback is
unaffected.
- Now-playing bar + queue follow the peer authoritatively: reflect the peer's
current track / position / volume, refresh on the peer's track changes, and
fetch a fresh renderer state on becoming a controller (so the bar isn't stale
until the next track change; this also fixed remote pause, which had been
sending a stale current_queue_item).
- Route play-next / add-to-queue (single track) to the peer via
QueueInsertTracks / QueueAddTracks (insert-after the peer's current item);
multi-track album/playlist enqueue stays local for now. Clear the superseded
SetVolume pending to stop the volume-drag command spam.
- Clean disconnect: disarm the liveness watchdog + clear the device list / cast
state, so no spurious "renderer unreachable" toast and no lingering badge.
- Volume lock only for local ALSA-Direct bit-perfect (backend Alsa + plugin Hw),
lifted while controlling; reflect the peer's volume with a 50% safety clamp
when unknown.
- Connect button golden whenever a session is on; +2px now-playing bar height.
cc
Add `is_set_volume_action` to PendingQueueAction (set for CtrlSrvrSetVolume) and
complete it on the SrvrCtrlVolumeChanged echo, so rapid volume changes can
supersede the in-flight one instead of colliding on the single pending slot (the
"pending queue action already active" spam). The frontend adapter clears the
superseded pending before sending the next SetVolume (lands with the qbz-slint
controller changes).
cc
Mirror Qobuz Connect's device icons in the connect flyout: map each renderer's
device_type to an icon (6 -> smartphone, 5 -> monitor or globe when the name is a
web player/browser, else -> speaker), matching the Tauri QconnectBadge
resolveDeviceType. Adds lucide monitor/smartphone/speaker/globe svgs, an icon key
on the QconnectDevice model computed in the event sink, and a per-row icon in the
PlayerBar device list.
cc
Port the rest of Tauri's volume logic to the Slint bar.
PlayerBar gains a derived vol-locked reproducing Tauri's handleVolumeChange
formula exactly: (alsaDirectHw && !isRemote) || volumeLocked. The local
ALSA-Direct (bit-perfect) lock is lifted while controlling a peer (the
!is-remote term) so the user can adjust the remote renderer; the sink-pushed
NowPlayingState.volume-locked keeps meaning only the peer-disallow half. All
five volume-interaction gates and the slider opacity now read vol-locked.
Force the local player to unity gain (set_volume(1.0)) in bit-perfect, mirroring
Tauri's playerSetVolume(100): when the backend is ALSA + the hw plugin AND a
peer is not the active renderer, set core volume to 1.0 and push
NowPlayingState.volume = 1.0. Applied (a) at the startup audio-settings load and
(b) on backend / alsa-plugin change. Gated on a new is_peer_active() so the
force-100 is skipped while controlling a peer. core().set_volume is the safe
seam; the protected device-init is untouched.
cc
When controlling a peer renderer, the volume slider must reflect the PEER's
actual volume, never QBZ's local 100, so a drag starts from a safe level and
never nukes the AVR. Extend RemoteNowPlaying with the peer's reported volume
(sourced from the effective remote renderer snapshot) and, in the poll loop's
peer-reflection branch, push it to NowPlayingState.volume as
v/100 (clamped 0..1), falling back to 0.5 (50%) when the peer's volume is
unknown. This branch only runs when a peer is active, so local/renderer
playback is unaffected.
cc
Playing a new album/track on QBZ while controlling a peer renderer played
it LOCALLY instead of casting it to the peer. Add
SlintQconnectService::play_on_peer_if_active(track_id): gated on
is_peer_renderer_active, it pushes the current core queue to the peer
unconditionally (same all-or-nothing Qobuz-castable admission as
sync_local_queue_if_changed; a mixed/local/Plex queue is refused with a
toast and treated as handled so it is not double-played locally), then
calls play_remote_renderer_track_if_active to SetPlayerState the peer to
the requested track. Send errors are logged but still return handled — a
peer owns playback, so never fall back to local audio (double-play).
play_audible (the single local-play chokepoint for play-all / track play)
now tries play_on_peer_if_active first and returns early when handled. When
no peer is active it returns false and the existing local play runs
byte-unchanged, so renderer / local playback do not regress.
cc
When QBZ is CONTROLLING a peer renderer the local player is stopped, so
the poll loop saw track_id == 0 and the seek bar froze. Add
RemoteNowPlaying + SlintQconnectService::remote_now_playing(), sourced
from effective_remote_renderer_snapshot (peer position / updated_at /
playing). The poll loop now branches at the top of each tick: when a peer
is active it drives position/duration/progress/playing from the peer
(duration from the core queue's current track, position extrapolated and
clamped), resets the local edge trackers, and skips the local body. In
every non-controller situation remote_now_playing returns None and the
existing local path runs byte-unchanged, so renderer mode does not regress.
cc
Add the CONTROLLER-mode device picker. New QconnectDevice struct + devices /
active-renderer-id / set-active on the QconnectDevState global (re-exported via
app.slint so include_modules! generates the type). The connect flyout replaces
the "No devices found" placeholder with a plain-row list (ADR-008: no pills)
when session renderers exist: each row taps to set-active and closes the menu;
the local device renders as "Play here" to pull playback back to QBZ. A
"Playing on: {name}" header shows while a peer is active.
The event sink rebuilds the device list and pushes the cast-aware now-playing
state (is-remote / cast-target / volume-locked) from session topology after
every event, replacing the prior TODO. main.rs wires on_set_active to the new
SlintQconnectService::set_active_renderer (over send_set_active_renderer).
Wired but unverified (blind build, no phone+speaker); device round-trips need a
smoke test. Slice 4 (peer-state reflection in the poll loop) deferred.
cc
Wire the eight NowPlayingState transport callbacks (toggle_play, next,
previous, seek, set_volume, toggle_mute, toggle_shuffle, cycle_repeat) with the
"try remote -> fall back to local" pattern: each spawns a task that calls the
matching *_if_remote method; Ok(true) short-circuits (forwarded to the active
peer), Ok(false)/no-service falls through to the existing playback::* local
call unchanged. Includes the seek fraction->ms (via local duration), volume
0..1->0..100, and mute-negation conversions.
Adds playback::is_muted() to read the authoritative local mute flag for the
mute handoff. No regression to renderer/local playback: the remote gate is
is_peer_renderer_active, so every non-controller branch returns Ok(false).
cc
Port the controller-mode transport seam from the Tauri adapter onto
SlintQconnectService: effective_active/remote_renderer_snapshot, the prime_*
optimistic-state helpers, send_command (with the superseded-pending clear),
and all *_if_remote methods (toggle_play, skip next/previous, play_track,
toggle_shuffle, cycle_repeat, set_volume, mute, set_autoplay_mode,
autoplay_load_tracks, stop, set_position) plus set_active_renderer. Each gates
on a peer renderer being active and returns Ok(false) to fall back to local.
Adds the controller request DTOs (set-player-state/volume/mute + the pure
build_set_position_player_state_request) to qconnect_transport, mirroring the
Tauri types byte-for-byte. Diagnostics route through log + dev_push_event;
cursor alignment uses the core's silent sync_current_to_id. Not yet wired to
the transport callbacks (Slice 2) or the device picker (Slice 3); dead-code
warnings are expected until then.
cc
Move the pure build_effective_renderer_snapshot helper from the Tauri adapter
into qconnect-app/src/session.rs (next to build_session_renderer_snapshot),
export it from the crate root, and repoint the Tauri side to the re-export.
Byte-identical behavior; enables the Slint controller path to reuse it.
cc
Relocate the QConnect diagnostics launcher out of the now-playing bar flyout
into a new opt-in Settings > Developer section (the modal + DEV log/status code
is unchanged). Also make the status block selectable (read-only TextEdit) so
session_uuid etc. can be copied, and relabel the event log.
cc
Port the QConnect adapter to the Slint frontend so qbz acts as a full peer
(renderer + controller), inheriting the shared qconnect-app hardening.
- Device-info / join payloads + transport config (piece a).
- SlintQconnectEventSink: queue materialize / renderer-command apply /
session-management via the shared qconnect-app orchestration; renderer
ready-report on SetActive(true).
- SlintSessionLoopHost + service connect recipe (subscribe-before-spawn) +
controller bootstrap + deferred renderer-join; connect/disconnect toggle
wired in the now-playing bar flyout.
- Renderer sync: outbound RndrSrvrStateUpdated reports (position / current
queue item / file+device audio quality) while qbz is the active local
renderer, and inbound UI refresh (now-playing card + queue sidebar) after
remote queue / state changes.
- Controller-side queue push: when the local queue changes while connected,
push it to the session (QueueLoadTracks) with all-or-nothing admission
(refuse mixed local/Plex queues; offline qobuz_download is eligible) and
echo suppression against the last-applied / last-pushed queue.
- Error + lifecycle toasts; DEV diagnostics modal (live session status +
selectable rolling event log) opened from the connect flyout.
cc
Move build_session_renderer_snapshot, queue_item_snapshot_for_cursor and
cache_renderer_snapshot into qconnect-app (session.rs / sync_state.rs) and
re-export them; repoint the Tauri qconnect adapter to consume them. This lets
the Slint adapter share one definition (ADR-006) instead of duplicating the
projection logic. Behavior is byte-identical.
cc
Ports the Tauri qconnect/transport.rs config layer to Slint: resolve_transport_config
(env overrides -> qws/createToken auto-discovery -> WsTransportConfig with require_jwt
+ 60s idle-retry + default channels) and fetch_qconnect_transport_credentials (POST
/qws/createToken with X-App-Id / X-User-Auth-Token), plus the device-identity helpers
(uuid via shared qbz_app::qconnect_identity, hostname, friendly name, persisted
device-name in qconnect_settings.db) and the hex-channel parsers.
Frontend difference handled: the Qobuz client is reached via runtime.core().client()
(Arc<RwLock<Option<QobuzClient>>>, may be None before API init) and is gated/ok_or'd,
unlike Tauri's always-present AppState.client. Uses qbz_qobuz::endpoints::build_url
(same base) instead of the Tauri-local one.
Deps added: qbz-qobuz (build_url), qconnect-transport-ws (WsTransportConfig /
NativeWsTransport). allow(dead_code) until the service wires it. cargo check
-p qbz-slint green (only the pre-existing unrelated rename_folder warning).
cc
Phase S steps 9-10. Makes qbz-Slint a QConnect renderer: SlintRendererEngine
implements qconnect_app::QconnectRendererEngine over runtime.core(), so the Slint
frontend inherits the shared echo/cursor/materialize/shuffle orchestration in
qconnect_app::renderer instead of re-deriving it.
- Deps added to qbz-slint: qconnect-app, explicit qbz-player, futures-util, uuid.
- 16 trait forwards through runtime.core() (CoreError mapped to String) +
current_output_format reading player().state.
- start_track_stream + the HTTP feeder (probe FLAC -> play_streaming_dynamic,
fallback download -> play_data) ported from the Tauri CoreBridge impl. The
protected bit-perfect seams are only called, never modified; the probe-derived
sample_rate/channels/bit_depth flow straight through.
- TLS deviation: the crates-workspace reqwest ships rustls-tls (not native-tls),
so .use_native_tls() is dropped. TLS is transport-only; audio bytes identical.
allow(dead_code) until the Slint QConnect service + event sink wire it (next
Phase S step). Verified: cargo check -p qbz-slint green (only a pre-existing
unrelated rename_folder dead-code warning).
cc
Step 8 of the renderer-engine extraction — the payoff of sharing. A recording
MockRendererEngine drives the renderer orchestration end-to-end, proving the
hard-won behavior is engine-independent (any future Slint regression is a wiring
bug in its trait impl, not a re-derivation bug in the shared logic). This logic
previously could only be exercised through the Tauri adapter.
Six scenarios:
- load dedup window: two ensure_remote_track_loaded for the same track within
5s -> start_track_stream once; no reload when track_id unchanged.
- #387 echo-seek: same-track SetState at <=1s while local is ahead -> seek
rejected; a genuine far seek IS honored.
- stale-projected-track guard: state-only SetState (command.current_track=None)
-> no align / no load from renderer_state's stale track; pause still applied.
- shuffle deferral: first event (shuffle_mode=true, order=None) -> shuffle_enabled
false (no invented identity order); second event with order -> enabled.
qconnect-app 38 -> 44 tests; all green, no warnings.
cc
Step 7 of the renderer-engine extraction. Moves report_file/device_audio_
quality_if_changed (dedup-against-last-reported + build RendererReport + send)
out of the Tauri service into QconnectApp methods, so the Slint report-back loop
reuses them. The Tauri service methods are now thin delegates that extract
runtime.app and forward.
Byte-identical: runtime.sync_state and the Arc passed to QconnectApp::new (the
one sync_handle() returns) are clones of the SAME Mutex (service.rs:232/242/277),
so the dedup state and the report build/send are unchanged; only the owner of
the code moved. send_renderer_report_command + the last_reported_* fields were
already in qconnect-app.
Verified: cargo check -p qbz-nix clean (no warnings); qbz-nix qconnect 42/42 +
qconnect-app 38/38 green.
cc
Step 6 of the renderer-engine extraction. Moves the renderer orchestration
out of the Tauri adapter UP into frontend-agnostic qconnect-app, rewritten
against &impl QconnectRendererEngine instead of &CoreBridge, so the Slint
adapter reuses the hard-won echo/cursor/materialize/shuffle logic instead of
re-deriving it.
Moved into qconnect_app::renderer (byte-for-byte, bridge -> engine):
apply_renderer_command (incl. the #387 echo-seek gate, the stale-projected-
track guard, force-restart-removed reasoning), materialize_remote_queue (skip-
gate, dedupe+batch hydrate, 3-tier start-index, shuffle deferral
effective_shuffle_enabled = shuffle_mode && order.is_some(), cursor-align +
stale-local-stop), sync_remote_shuffle_projection, align_queue_cursor,
apply_remote_loop_mode, ensure_remote_track_loaded (+ should_reload /
is_recent / queue_state_needs_materialization, LOAD_ATTEMPT_DEDUP_WINDOW,
PLAYING_STATE_* consts). The sync_state lock-drop-across-await discipline is
preserved; last_load_attempt is still recorded before the load.
Stays adapter-side: corebridge.rs is now thin wrappers that unwrap the
Arc<RwLock<Option<CoreBridge>>> guard and delegate (event_sink/service call
sites unchanged). In operation the bridge is populated at startup before any
QConnect session opens, so the guard-before-skip reorder is unreachable;
behavior is byte-identical. ensure_remote_track_loaded now resolves duration
via get_track then calls engine.start_track_stream (which owns get_stream_url +
the protected feeder); the two independent catalog reads swap order, benign.
mod.rs re-exports PLAYING_STATE_* from qconnect-app.
Verified: cargo check -p qbz-nix clean (no warnings); qbz-nix qconnect 42/42 +
qconnect-app 38/38 green; Opus adversarial review CLEAN across byte-identity,
guard-reorder, lock-discipline, ensure-reorder, Send/generics, re-exports.
cc
Step 5 of the renderer-engine extraction. Adds the Tauri-side trait impl
so CoreBridge becomes a QConnect renderer engine; the hard-won orchestration
(echo gate, materialize, dedup) will be hoisted above it in step 6.
Every transport/queue/catalog method is the same one-line forward CoreBridge
already does to QbzCore, spelled CoreBridge::method(self, ..) to call the
inherent method unambiguously and never recurse into the trait impl. The audio
thread sees identical bytes.
start_track_stream is the verbatim load_remote_track_into_player +
stream_remote_track_into_player body with quality/duration resolution lifted to
the caller; the protected seams (play_streaming_dynamic / play_data) and the
HTTP feeder stay impl-side and are only called, never modified. Threaded
start_position_secs through stream_remote_track_into_player (existing caller
passes 0, byte-identical) so the trait param is honored for non-QConnect callers.
current_output_format reads player().state.{get_sample_rate,get_bit_depth} for
the report-back path.
Verified: cargo check -p qbz-nix clean; qbz-nix qconnect 42/42 +
qconnect-app 38/38 green; Opus adversarial review CLEAN (recursion-safety,
byte-identity, protected-seam-untouched, type unification).
cc
Move the pure protocol mappers model_track_to_core_queue_track,
normalize_volume_to_fraction, qconnect_repeat_mode_from_loop_mode (+ the
QCONNECT_REMOTE_QUEUE_SOURCE tag) into qconnect-app/src/renderer.rs so the
renderer orchestration that calls them (hoisted in a later step) can live there
too. The src-tauri mod.rs re-exports them, so corebridge + tests compile
unchanged. The load-dedup predicates and audio-quality report helpers move
alongside their orchestration / report consumers in the next steps (cohesion +
avoids churning the AUDIO_QUALITY_* consts now).
src-tauri qconnect 42 green.
cc
queue_resolution.rs is pure protocol math (queue_item_id/track_id/order cursor
reconciliation) over QConnectQueueState/QConnectRendererState only -- no engine,
no Tauri. Relocate it to qconnect-app so the renderer orchestration that calls it
(materialize / cursor-align, hoisted in a later step) can live there too. The
src-tauri module becomes a `pub(super) use qconnect_app::queue_resolution::*;`
re-export shim, so every `super::queue_resolution::…` call site + the existing
tests compile unchanged. Visibility widened pub(super)->pub for the re-export.
src-tauri qconnect 42 green (tests now exercise the moved code via the shim),
qconnect-app check green.
cc
The renderer-engine seam: a 16-method trait both frontends implement with
zero-cost forwards to their QbzCore/Player, so the hard-won renderer
orchestration (echo-seek rejection, cursor align, queue materialize, shuffle
deferral, load dedup window) can move UP into qconnect-app and be written once
against the trait instead of re-derived per frontend (slice 6, Phase T 1-2).
- Adds qbz-player as a qconnect-app dep, for PlaybackState (read by the echo
gate + should_reload; it is not in qbz-models and not re-exported). qbz-player
does NOT depend on qconnect-app -> no dependency cycle.
- The two protected bit-perfect seams (play_data / play_streaming_dynamic) are
reached ONLY through the high-level start_track_stream method; BufferWriter +
reqwest + the detached feeder spawn stay impl-side and never cross the crate
boundary. play_data / play_streaming_dynamic are never modified.
No impls, no logic moved yet -- pure foundation. qconnect-app 38 green, src-tauri
cargo check green.
cc
Slice 5 moved the transport subscription INTO the spawned run_session_loop
(self.subscribe_transport_events()), which races the WS handshake:
NativeWsTransport::connect spawns the WS loop and returns WITHOUT awaiting it,
and tokio broadcast has no replay -- so the initial Connected / Subscribed /
SessionEstablished / SESSION_STATE events could be sent to zero receivers and
silently dropped before the spawned task is first polled (a lost
SessionEstablished leaves the UI stuck Connecting; a lost SESSION_STATE skips
the deferred renderer join). The 42 wire tests miss it: they subscribe before
connect against a non-self-driving InMemoryWsTransport.
Restore the prior guarantee -- subscribe synchronously in connect() right after
app.connect() returns and before the spawn / any further await, and pass the
broadcast::Receiver into run_session_loop. Found by adversarial review.
qconnect-app 38 green, src-tauri qconnect 42 green.
cc
Move the spawned transport event-loop body out of the Tauri adapter into the
shared QconnectApp::run_session_loop, so both frontends inherit takeover
arbitration / deferred renderer-join / reconnect resync / Lagged re-ask. The
adapter-coupled seams go through a new SessionLoopHost trait:
- update_lifecycle -- Tauri gates on a live runtime + dedups, emits via sink
- bootstrap_after_reconnect
- deferred_renderer_join -- R2: the CoreBridge track-duration read stays adapter-side
- on_reconnect_exhausted -- R3: Exhausted + last_error + runtime=None teardown +
the raw qconnect:status_changed emit; returns whether the loop should break
- on_loop_error -- the raw qconnect:error channel
TauriSessionLoopHost implements it over the existing service handles; connect()
is reduced to: build host -> spawn(app.run_session_loop(host, idle_retry)). The
diagnostic/lifecycle/resync emits still route through self.sink, so the Svelte
wire surface is byte-identical. hex_preview + queue_payload_track_preview + the
re-ask budget const relocate with the loop; the dead Tauri copies are removed.
R3 (QconnectRuntime/JoinHandle ownership, #358 latch, disconnect teardown) stays
adapter-side.
qconnect-app 38 green, src-tauri qconnect 42 green (the byte-identity fence).
cc
Relocate the transport-loop helpers out of the Tauri adapter into QconnectApp
methods + qconnect-app free items, so the shared session loop (slice 5.2) and a
future Slint adapter use one definition:
- QconnectApp::capture_session_state_takeover_input (locks self.sync)
- QconnectApp::clear_pending_if_matches (locks self.state)
- QconnectApp::send_set_active_renderer -- inline {"renderer_id": i32} payload,
byte-identical to the prior QconnectSetActiveRendererRequest serialization
- QconnectApp::ask_for_active_renderer_state / QconnectApp::ask_for_queue_state
- SessionStateTakeoverInput struct + queue_payload_track_preview (pure free fn)
Deleted the service.rs free fns (capture_session_state_takeover_input,
send_set_active_renderer_via_app, ask_for_active_renderer_state,
ask_for_queue_state, queue_payload_track_preview, clear_pending_if_matches) and
the SessionStateTakeoverInput struct; rewired every call site (the loop +
deferred_renderer_join + the reconnect/lagged paths) to the methods. The request
structs serialize identically to the inline json!, so the wire is unchanged.
qconnect-app 38 green, src-tauri qconnect 42 green.
cc
The Tauri event sink's apply_session_management_event is now a thin delegator:
it resolves this device's identity (resolve_local_identity) and calls
QconnectApp::apply_session_management_event, then runs the post-lock CoreBridge
work the returned SessionApplyOutcome asks for in the exact prior order
(loop mode -> local-playback handoff -> projection -> freeze -> watchdog).
Freeze + watchdog route through app.freeze_active_renderer_projection /
app.arm_renderer_watchdog (same shared sync Mutex, same epoch guard).
Deleted from the Tauri adapter (now solely in qconnect-app): the locked
session-management match, bump_watchdog_generation, the inline 12s watchdog
spawn, freeze_active_renderer_projection_to_unknown, and the duplicate
ensure_session_renderer_state / sync_session_renderer_active_flags. ensure_* is
re-exported from qconnect-app for service.rs; refresh_local_renderer_id wrapper
is replaced by resolve_local_identity (returns the injected LocalIdentity), and
the 3 renderer-id-resolution tests call the pure qconnect-app fn with it.
Byte-identity fence holds: src-tauri qconnect 42 green, qconnect-app 38 green.
cc
Relocate the session-management apply + renderer-liveness logic into
qconnect-app so both the Tauri and the future Slint adapter share it:
- apply_session_management_event(message_type, payload, identity)
-> SessionApplyOutcome: the locked critical section, byte-identical,
now locking self.sync. Local identity is injected as data (LocalIdentity)
so the crate stays frontend-agnostic (refresh_local_renderer_id takes it
instead of resolving device-info itself). The CoreBridge post-lock work is
RETURNED via the outcome rather than run inline -- materialization stays
adapter-side (renderer-engine seam, slice 6). Early returns yield a default
outcome, exactly as the prior `return;` skipped the post-lock block.
- arm_renderer_watchdog / run_renderer_watchdog: the generation-guarded 12s
liveness task, modeled on watch_pending_action_timeout (tokio::spawn + sleep
+ re-lock + 4-field epoch revalidation).
- freeze_active_renderer_projection: cached playing_state -> UNKNOWN, then
emit the freeze event through the sink (dedicated channel + blanket event
preserved).
- ensure_session_renderer_state + sync_session_renderer_active_flags moved to
qconnect-app (pure mutators over the relocated accumulator).
Adds 5 tests (SESSION_STATE outcome, watchdog arm/fire-after-12s/epoch-disarm,
ACTIVE_DISCONNECTED freeze) using start_paused + advance. The Tauri sink keeps
its own live copy this commit; slice 2+4.5 rewires it to delegate and deletes
the duplicate. qconnect-app 38 green, src-tauri qconnect 42 green.
cc
QconnectApp now holds the shared QconnectRemoteSyncState accumulator behind
its own Mutex, disjoint from `state` (the two are never co-locked). The Tauri
adapter still builds the Arc and passes the SAME clone to its event sink /
service loop and into new(), so there is exactly one lock backing the
session/liveness/renderer paths -- the app simply gains a handle to it via
sync_handle(). Pure wiring, byte-identical locking, no logic moved. Prepares
slices 2+4.4/5 to relocate the session-apply + watchdog logic into qconnect-app.
cc
vicrodh