feat(player): webview audio backend for platforms without a native one

Adds WebviewAudioBackend, used on non-Linux/non-Android targets (e.g. Windows) where there is no libmpv/ExoPlayer. Instead of decoding, it emits a WebviewAudioLoad event with the stream URL; a frontend <audio> element (WebviewAudioAdapter + webviewAudio service) plays it and reports state/position back through the existing player_report_* round-trip, so the Rust PlayerController stays the single source of truth. Play/pause/ seek reach the element via the existing ControlCommand event. All video already renders in the webview on every platform, so this completes audio-only playback for Windows (video via WebView2, audio via <audio>). Pure Rust + Tauri events, so it still cross-compiles from Linux. Regenerates bindings.ts (adds webview_audio_load; also carries the equalizer EQ bindings). TRACES: UR-003, UR-004, UR-005 | DR-004
2026-07-24 23:49:23 +02:00
parent c543f90ad3
commit d4e2cd120c
8 changed files with 643 additions and 5 deletions
@@ -156,6 +156,25 @@ pub enum PlayerStatusEvent {
        /// Target position in seconds (only meaningful for "seek").
        position: Option<f64>,
    },
+    /// Ask the frontend webview `<audio>` element to load and play a stream.
+    ///
+    /// Emitted by `WebviewAudioBackend` on platforms with no native audio
+    /// backend (e.g. Windows): audio-only playback is rendered by an `<audio>`
+    /// element in the webview, mirroring how all video already renders through
+    /// the webview `<video>`. The element then reports its state/position back
+    /// through the `player_report_*` commands, so the Rust controller stays the
+    /// single source of truth. Subsequent play/pause/seek/stop reach the element
+    /// via `ControlCommand`.
+    WebviewAudioLoad {
+        /// Stream URL for the `<audio>` element to play.
+        url: String,
+        /// Jellyfin item id, used as the media_id when reporting state back.
+        media_id: Option<String>,
+        /// Resume position in seconds (0 = start from the beginning).
+        position: f64,
+        /// Whether to begin playing immediately after loading.
+        autoplay: bool,
+    },
 }

 /// Trait for emitting player events to the frontend.
@@ -22,6 +22,11 @@ pub mod android;
 #[cfg(target_os = "linux")]
 pub mod mpv_backend;

+// Platforms with no native audio backend (e.g. Windows) render audio-only
+// playback through a webview <audio> element, mirroring how all video renders.
+#[cfg(not(any(target_os = "linux", target_os = "android")))]
+pub mod webview_audio_backend;
+
 // Re-export commonly used types
 pub use autoplay::{AutoplayDecision, AutoplaySettings};
 pub use backend::{NullBackend, PlayerBackend, PlayerError};
@@ -40,6 +45,9 @@ pub use android::ExoPlayerBackend;
 #[cfg(target_os = "linux")]
 pub use mpv_backend::MpvBackend;

+#[cfg(not(any(target_os = "linux", target_os = "android")))]
+pub use webview_audio_backend::WebviewAudioBackend;
+
 #[cfg(target_os = "android")]
 pub use android::{
    disable_remote_volume, enable_remote_volume, get_detected_codecs, set_media_command_handler,
@@ -0,0 +1,295 @@
+//! Webview audio backend — audio-only playback for platforms without a native
+//! audio backend (currently Windows).
+//!
+//! ## Why this exists
+//! All *video* already renders through the webview HTML5 `<video>` element on
+//! every platform (see `VideoPlayer.svelte`); libmpv/ExoPlayer only ever drive
+//! *audio-only* (music) playback. On Windows there is no native audio backend,
+//! so `create_player_backend()` used to fall back to `NullBackend` and music was
+//! silent.
+//!
+//! This backend fills that gap without any C dependency (so it still
+//! cross-compiles from Linux): instead of decoding audio itself, it hands the
+//! stream URL to a frontend `<audio>` element via a `WebviewAudioLoad` event and
+//! then drives play/pause/seek/stop through `ControlCommand` events — exactly the
+//! round-trip the HTML5 video path already uses. The `<audio>` element reports
+//! its real state/position back through the `player_report_*` commands, so the
+//! Rust `PlayerController` remains the single source of truth (the controller's
+//! `report_html5_*` methods fold those reports into the normal event pipeline).
+//!
+//! Because the reported state flows through the event pipeline (not through this
+//! backend's `position()`/`state()` pollers — the timer loop does not poll the
+//! backend for HTML5-rendered media), this backend only needs to keep a
+//! best-effort local mirror for direct `player_get_state` queries.
+//!
+//! TRACES: UR-003, UR-004, UR-005 | DR-004
+
+use std::sync::Arc;
+
+use log::{debug, info};
+
+use super::backend::{PlayerBackend, PlayerError};
+use super::events::{PlayerEventEmitter, PlayerStatusEvent};
+use super::media::{MediaItem, MediaSource};
+use super::state::PlayerState;
+use crate::settings::AudioSettings;
+use crate::utils::lock::MutexSafe;
+
+/// Extract a webview-playable URL from a media item's source.
+///
+/// Remote/DirectUrl are HTTP(S) URLs the `<audio>` element can play directly.
+/// Local files would need the Tauri asset protocol (`convertFileSrc`) on the
+/// frontend; for now we pass the path through and let the frontend resolve it.
+fn stream_url(media: &MediaItem) -> String {
+    match &media.source {
+        MediaSource::Remote { stream_url, .. } => stream_url.clone(),
+        MediaSource::DirectUrl { url } => url.clone(),
+        MediaSource::Local { file_path, .. } => file_path.to_string_lossy().to_string(),
+    }
+}
+
+struct InternalState {
+    current_media: Option<MediaItem>,
+    volume: f32,
+    position: f64,
+    duration: Option<f64>,
+    state: PlayerState,
+    audio_settings: AudioSettings,
+}
+
+pub struct WebviewAudioBackend {
+    emitter: Arc<dyn PlayerEventEmitter>,
+    state: Arc<std::sync::Mutex<InternalState>>,
+}
+
+impl WebviewAudioBackend {
+    pub fn new(emitter: Arc<dyn PlayerEventEmitter>) -> Result<Self, PlayerError> {
+        info!("[WebviewAudioBackend] Initializing (audio renders in webview <audio>)");
+        Ok(Self {
+            emitter,
+            state: Arc::new(std::sync::Mutex::new(InternalState {
+                current_media: None,
+                volume: 1.0,
+                position: 0.0,
+                duration: None,
+                state: PlayerState::Idle,
+                audio_settings: AudioSettings::default(),
+            })),
+        })
+    }
+
+    /// Emit a backend-originated control intent to the active frontend adapter
+    /// (the webview `<audio>` element, via `playerEvents.ts` -> active adapter).
+    fn emit_control(&self, action: &str, position: Option<f64>) {
+        self.emitter.emit(PlayerStatusEvent::ControlCommand {
+            action: action.to_string(),
+            position,
+        });
+    }
+}
+
+impl PlayerBackend for WebviewAudioBackend {
+    fn load(&mut self, media: &MediaItem) -> Result<(), PlayerError> {
+        let url = stream_url(media);
+        info!("[WebviewAudioBackend] load: {} - {}", media.title, url);
+
+        {
+            let mut st = self.state.lock_safe();
+            st.current_media = Some(media.clone());
+            st.position = 0.0;
+            st.duration = media.duration;
+            st.state = PlayerState::Loading {
+                media: media.clone(),
+            };
+        }
+
+        // Hand the URL to the frontend <audio> element. autoplay=true so a plain
+        // load-then-play (the common queue-advance path) starts immediately; an
+        // explicit pause afterwards is still honored via ControlCommand.
+        self.emitter.emit(PlayerStatusEvent::WebviewAudioLoad {
+            url,
+            media_id: media.jellyfin_id().map(|s| s.to_string()),
+            position: 0.0,
+            autoplay: true,
+        });
+        Ok(())
+    }
+
+    fn play(&mut self) -> Result<(), PlayerError> {
+        debug!("[WebviewAudioBackend] play");
+        self.emit_control("play", None);
+        Ok(())
+    }
+
+    fn pause(&mut self) -> Result<(), PlayerError> {
+        debug!("[WebviewAudioBackend] pause");
+        self.emit_control("pause", None);
+        Ok(())
+    }
+
+    fn stop(&mut self) -> Result<(), PlayerError> {
+        debug!("[WebviewAudioBackend] stop");
+        {
+            let mut st = self.state.lock_safe();
+            st.current_media = None;
+            st.position = 0.0;
+            st.duration = None;
+            st.state = PlayerState::Idle;
+        }
+        self.emit_control("stop", None);
+        Ok(())
+    }
+
+    fn seek(&mut self, position: f64) -> Result<(), PlayerError> {
+        debug!("[WebviewAudioBackend] seek: {}", position);
+        self.state.lock_safe().position = position;
+        self.emit_control("seek", Some(position));
+        Ok(())
+    }
+
+    fn set_volume(&mut self, volume: f32) -> Result<(), PlayerError> {
+        let clamped = volume.clamp(0.0, 1.0);
+        self.state.lock_safe().volume = clamped;
+        // Volume is applied on the element by the frontend, which observes the
+        // volume via the player store; no dedicated ControlCommand action yet.
+        Ok(())
+    }
+
+    fn position(&self) -> f64 {
+        self.state.lock_safe().position
+    }
+
+    fn duration(&self) -> Option<f64> {
+        self.state.lock_safe().duration
+    }
+
+    fn state(&self) -> PlayerState {
+        self.state.lock_safe().state.clone()
+    }
+
+    fn volume(&self) -> f32 {
+        self.state.lock_safe().volume
+    }
+
+    fn set_audio_settings(&mut self, settings: &AudioSettings) -> Result<(), PlayerError> {
+        self.state.lock_safe().audio_settings = settings.clone().with_crossfade_clamped();
+        Ok(())
+    }
+
+    fn audio_settings(&self) -> AudioSettings {
+        self.state.lock_safe().audio_settings.clone()
+    }
+}
+
+// TRACES: UR-003, UR-004, UR-005 | DR-004
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use crate::player::events::PlayerStatusEvent;
+    use crate::player::media::{MediaSource, MediaType};
+    use std::sync::Mutex as StdMutex;
+
+    /// Test emitter that records everything emitted.
+    struct RecordingEmitter {
+        events: Arc<StdMutex<Vec<PlayerStatusEvent>>>,
+    }
+
+    impl PlayerEventEmitter for RecordingEmitter {
+        fn emit(&self, event: PlayerStatusEvent) {
+            self.events.lock().unwrap().push(event);
+        }
+    }
+
+    fn test_media() -> MediaItem {
+        MediaItem {
+            id: "track1".to_string(),
+            title: "Song".to_string(),
+            name: Some("Song".to_string()),
+            artist: Some("Artist".to_string()),
+            album: Some("Album".to_string()),
+            album_name: Some("Album".to_string()),
+            album_id: None,
+            artist_items: None,
+            artists: Some(vec!["Artist".to_string()]),
+            primary_image_tag: None,
+            image_id: None,
+            item_type: Some("Audio".to_string()),
+            playlist_id: None,
+            duration: Some(200.0),
+            artwork_url: None,
+            media_type: MediaType::Audio,
+            source: MediaSource::DirectUrl {
+                url: "http://example.com/song.mp3".to_string(),
+            },
+            video_codec: None,
+            needs_transcoding: false,
+            video_width: None,
+            video_height: None,
+            subtitles: vec![],
+            series_id: None,
+            server_id: None,
+        }
+    }
+
+    fn backend() -> (WebviewAudioBackend, Arc<StdMutex<Vec<PlayerStatusEvent>>>) {
+        let events = Arc::new(StdMutex::new(Vec::new()));
+        let emitter = Arc::new(RecordingEmitter {
+            events: events.clone(),
+        });
+        (WebviewAudioBackend::new(emitter).unwrap(), events)
+    }
+
+    #[test]
+    fn load_emits_webview_audio_load_with_url() {
+        let (mut b, events) = backend();
+        b.load(&test_media()).unwrap();
+
+        let ev = events.lock().unwrap();
+        let load = ev
+            .iter()
+            .find(|e| matches!(e, PlayerStatusEvent::WebviewAudioLoad { .. }))
+            .expect("WebviewAudioLoad emitted");
+        if let PlayerStatusEvent::WebviewAudioLoad { url, autoplay, .. } = load {
+            assert_eq!(url, "http://example.com/song.mp3");
+            assert!(*autoplay);
+        }
+        assert!(matches!(b.state(), PlayerState::Loading { .. }));
+    }
+
+    #[test]
+    fn pause_and_seek_emit_control_commands() {
+        let (mut b, events) = backend();
+        b.load(&test_media()).unwrap();
+        b.pause().unwrap();
+        b.seek(42.0).unwrap();
+
+        let ev = events.lock().unwrap();
+        assert!(ev.iter().any(|e| matches!(
+            e,
+            PlayerStatusEvent::ControlCommand { action, .. } if action == "pause"
+        )));
+        assert!(ev.iter().any(|e| matches!(
+            e,
+            PlayerStatusEvent::ControlCommand { action, position: Some(p) }
+                if action == "seek" && (*p - 42.0).abs() < f64::EPSILON
+        )));
+        assert_eq!(b.position(), 42.0);
+    }
+
+    #[test]
+    fn volume_is_clamped_and_stored() {
+        let (mut b, _events) = backend();
+        b.set_volume(1.5).unwrap();
+        assert_eq!(b.volume(), 1.0);
+        b.set_volume(-0.2).unwrap();
+        assert_eq!(b.volume(), 0.0);
+    }
+
+    #[test]
+    fn stop_resets_to_idle() {
+        let (mut b, _events) = backend();
+        b.load(&test_media()).unwrap();
+        b.stop().unwrap();
+        assert!(matches!(b.state(), PlayerState::Idle));
+    }
+}