Refactor project into two standalone components

Split monorepo into separate units for future independent repositories:
- media-server/: Standalone FastAPI server with own README, requirements,
  config example, and CLAUDE.md
- haos-integration/: HACS-ready Home Assistant integration with hacs.json,
  own README, and CLAUDE.md

Both components now have their own .gitignore files and can be easily
extracted into separate repositories.

Also adds custom icon support for scripts configuration.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This commit is contained in:
2026-02-04 14:36:23 +03:00
parent 5519e449cd
commit e26df64e4b
44 changed files with 367 additions and 105 deletions

View File

@@ -0,0 +1,75 @@
"""Media controller services."""
import os
import platform
from pathlib import Path
from typing import TYPE_CHECKING
if TYPE_CHECKING:
from .media_controller import MediaController
_controller_instance: "MediaController | None" = None
def _is_android() -> bool:
"""Check if running on Android (e.g., via Termux)."""
# Check for Android-specific paths and environment
android_indicators = [
Path("/system/build.prop").exists(),
Path("/data/data/com.termux").exists(),
"ANDROID_ROOT" in os.environ,
"TERMUX_VERSION" in os.environ,
]
return any(android_indicators)
def get_media_controller() -> "MediaController":
"""Get the platform-specific media controller instance.
Returns:
The media controller for the current platform
Raises:
RuntimeError: If the platform is not supported
"""
global _controller_instance
if _controller_instance is not None:
return _controller_instance
system = platform.system()
if system == "Windows":
from .windows_media import WindowsMediaController
_controller_instance = WindowsMediaController()
elif system == "Linux":
# Check if running on Android
if _is_android():
from .android_media import AndroidMediaController
_controller_instance = AndroidMediaController()
else:
from .linux_media import LinuxMediaController
_controller_instance = LinuxMediaController()
elif system == "Darwin": # macOS
from .macos_media import MacOSMediaController
_controller_instance = MacOSMediaController()
else:
raise RuntimeError(f"Unsupported platform: {system}")
return _controller_instance
def get_current_album_art() -> bytes | None:
"""Get the current album art bytes (Windows only for now)."""
system = platform.system()
if system == "Windows":
from .windows_media import get_current_album_art as _get_art
return _get_art()
return None
__all__ = ["get_media_controller", "get_current_album_art"]

View File

@@ -0,0 +1,232 @@
"""Android media controller using Termux:API.
This controller is designed to run on Android devices using Termux.
It requires the Termux:API app and termux-api package to be installed.
Installation:
1. Install Termux from F-Droid (not Play Store)
2. Install Termux:API from F-Droid
3. In Termux: pkg install termux-api
4. Grant necessary permissions to Termux:API
"""
import asyncio
import json
import logging
import subprocess
from typing import Optional, Any
from ..models import MediaState, MediaStatus
from .media_controller import MediaController
logger = logging.getLogger(__name__)
def _check_termux_api() -> bool:
"""Check if termux-api is available."""
try:
result = subprocess.run(
["which", "termux-media-player"],
capture_output=True,
timeout=5,
)
return result.returncode == 0
except Exception:
return False
TERMUX_API_AVAILABLE = _check_termux_api()
class AndroidMediaController(MediaController):
"""Media controller for Android using Termux:API.
Requires:
- Termux app
- Termux:API app
- termux-api package (pkg install termux-api)
"""
def __init__(self):
if not TERMUX_API_AVAILABLE:
logger.warning(
"Termux:API not available. Install with: pkg install termux-api"
)
def _run_termux_command(
self, command: list[str], timeout: int = 10
) -> Optional[str]:
"""Run a termux-api command and return the output."""
try:
result = subprocess.run(
command,
capture_output=True,
text=True,
timeout=timeout,
)
if result.returncode == 0:
return result.stdout.strip()
logger.error(f"Termux command failed: {result.stderr}")
return None
except subprocess.TimeoutExpired:
logger.error(f"Termux command timed out: {command}")
return None
except Exception as e:
logger.error(f"Termux command error: {e}")
return None
def _send_media_key(self, key: str) -> bool:
"""Send a media key event.
Args:
key: One of: play, pause, play-pause, stop, next, previous
"""
# termux-media-player command
result = self._run_termux_command(["termux-media-player", key])
return result is not None
def _get_media_info(self) -> dict[str, Any]:
"""Get current media playback info using termux-media-player."""
result = self._run_termux_command(["termux-media-player", "info"])
if result:
try:
return json.loads(result)
except json.JSONDecodeError:
pass
return {}
def _get_volume(self) -> tuple[int, bool]:
"""Get current volume using termux-volume."""
result = self._run_termux_command(["termux-volume"])
if result:
try:
volumes = json.loads(result)
# Find music stream
for stream in volumes:
if stream.get("stream") == "music":
volume = stream.get("volume", 0)
max_volume = stream.get("max_volume", 15)
# Convert to 0-100 scale
percent = int((volume / max_volume) * 100) if max_volume > 0 else 0
return percent, False
except (json.JSONDecodeError, KeyError):
pass
return 100, False
def _set_volume_internal(self, volume: int) -> bool:
"""Set volume using termux-volume."""
# termux-volume expects stream name and volume level
# Convert 0-100 to device scale (usually 0-15)
result = self._run_termux_command(["termux-volume"])
if result:
try:
volumes = json.loads(result)
for stream in volumes:
if stream.get("stream") == "music":
max_volume = stream.get("max_volume", 15)
device_volume = int((volume / 100) * max_volume)
self._run_termux_command(
["termux-volume", "music", str(device_volume)]
)
return True
except (json.JSONDecodeError, KeyError):
pass
return False
async def get_status(self) -> MediaStatus:
"""Get current media playback status."""
status = MediaStatus()
# Get volume
volume, muted = self._get_volume()
status.volume = volume
status.muted = muted
# Get media info
info = self._get_media_info()
if not info:
status.state = MediaState.IDLE
return status
# Parse playback status
playback_status = info.get("status", "").lower()
if playback_status == "playing":
status.state = MediaState.PLAYING
elif playback_status == "paused":
status.state = MediaState.PAUSED
elif playback_status == "stopped":
status.state = MediaState.STOPPED
else:
status.state = MediaState.IDLE
# Parse track info
status.title = info.get("title") or info.get("Track") or None
status.artist = info.get("artist") or info.get("Artist") or None
status.album = info.get("album") or info.get("Album") or None
# Duration and position (in milliseconds from some sources)
duration = info.get("duration", 0)
if duration > 1000: # Likely milliseconds
duration = duration / 1000
status.duration = duration if duration > 0 else None
position = info.get("position", info.get("current_position", 0))
if position > 1000: # Likely milliseconds
position = position / 1000
status.position = position if position > 0 else None
status.source = "Android"
return status
async def play(self) -> bool:
"""Resume playback."""
return self._send_media_key("play")
async def pause(self) -> bool:
"""Pause playback."""
return self._send_media_key("pause")
async def stop(self) -> bool:
"""Stop playback."""
return self._send_media_key("stop")
async def next_track(self) -> bool:
"""Skip to next track."""
return self._send_media_key("next")
async def previous_track(self) -> bool:
"""Go to previous track."""
return self._send_media_key("previous")
async def set_volume(self, volume: int) -> bool:
"""Set system volume."""
return self._set_volume_internal(volume)
async def toggle_mute(self) -> bool:
"""Toggle mute state.
Note: Android doesn't have a simple mute toggle via termux-api,
so we set volume to 0 or restore previous volume.
"""
volume, _ = self._get_volume()
if volume > 0:
# Store current volume and mute
self._previous_volume = volume
self._set_volume_internal(0)
return True
else:
# Restore previous volume
prev = getattr(self, "_previous_volume", 50)
self._set_volume_internal(prev)
return False
async def seek(self, position: float) -> bool:
"""Seek to position in seconds.
Note: Seek functionality may be limited depending on the media player.
"""
# termux-media-player doesn't support seek directly
# This is a limitation of the API
logger.warning("Seek not fully supported on Android via Termux:API")
return False

View File

@@ -0,0 +1,295 @@
"""Linux media controller using MPRIS D-Bus interface."""
import asyncio
import logging
import subprocess
from typing import Optional, Any
from ..models import MediaState, MediaStatus
from .media_controller import MediaController
logger = logging.getLogger(__name__)
# Linux-specific imports
try:
import dbus
from dbus.mainloop.glib import DBusGMainLoop
DBUS_AVAILABLE = True
except ImportError:
DBUS_AVAILABLE = False
logger.warning("D-Bus libraries not available")
class LinuxMediaController(MediaController):
"""Media controller for Linux using MPRIS D-Bus interface."""
MPRIS_PATH = "/org/mpris/MediaPlayer2"
MPRIS_INTERFACE = "org.mpris.MediaPlayer2.Player"
MPRIS_PREFIX = "org.mpris.MediaPlayer2."
PROPERTIES_INTERFACE = "org.freedesktop.DBus.Properties"
def __init__(self):
if not DBUS_AVAILABLE:
raise RuntimeError(
"Linux media control requires dbus-python package. "
"Install with: sudo apt-get install python3-dbus"
)
DBusGMainLoop(set_as_default=True)
self._bus = dbus.SessionBus()
def _get_active_player(self) -> Optional[str]:
"""Find an active MPRIS media player on the bus."""
try:
bus_names = self._bus.list_names()
mpris_players = [
name for name in bus_names if name.startswith(self.MPRIS_PREFIX)
]
if not mpris_players:
return None
# Prefer players that are currently playing
for player in mpris_players:
try:
proxy = self._bus.get_object(player, self.MPRIS_PATH)
props = dbus.Interface(proxy, self.PROPERTIES_INTERFACE)
status = props.Get(self.MPRIS_INTERFACE, "PlaybackStatus")
if status == "Playing":
return player
except Exception:
continue
# Return the first available player
return mpris_players[0]
except Exception as e:
logger.error(f"Failed to get active player: {e}")
return None
def _get_player_interface(self, player_name: str):
"""Get the MPRIS player interface."""
proxy = self._bus.get_object(player_name, self.MPRIS_PATH)
return dbus.Interface(proxy, self.MPRIS_INTERFACE)
def _get_properties_interface(self, player_name: str):
"""Get the properties interface for a player."""
proxy = self._bus.get_object(player_name, self.MPRIS_PATH)
return dbus.Interface(proxy, self.PROPERTIES_INTERFACE)
def _get_property(self, player_name: str, property_name: str) -> Any:
"""Get a property from the player."""
try:
props = self._get_properties_interface(player_name)
return props.Get(self.MPRIS_INTERFACE, property_name)
except Exception as e:
logger.debug(f"Failed to get property {property_name}: {e}")
return None
def _get_volume_pulseaudio(self) -> tuple[int, bool]:
"""Get volume using pactl (PulseAudio/PipeWire)."""
try:
# Get default sink volume
result = subprocess.run(
["pactl", "get-sink-volume", "@DEFAULT_SINK@"],
capture_output=True,
text=True,
timeout=5,
)
if result.returncode == 0:
# Parse volume from output like "Volume: front-left: 65536 / 100% / 0.00 dB"
for part in result.stdout.split("/"):
if "%" in part:
volume = int(part.strip().rstrip("%"))
break
else:
volume = 100
else:
volume = 100
# Get mute status
result = subprocess.run(
["pactl", "get-sink-mute", "@DEFAULT_SINK@"],
capture_output=True,
text=True,
timeout=5,
)
muted = "yes" in result.stdout.lower() if result.returncode == 0 else False
return volume, muted
except Exception as e:
logger.error(f"Failed to get volume via pactl: {e}")
return 100, False
def _set_volume_pulseaudio(self, volume: int) -> bool:
"""Set volume using pactl."""
try:
result = subprocess.run(
["pactl", "set-sink-volume", "@DEFAULT_SINK@", f"{volume}%"],
capture_output=True,
timeout=5,
)
return result.returncode == 0
except Exception as e:
logger.error(f"Failed to set volume: {e}")
return False
def _toggle_mute_pulseaudio(self) -> bool:
"""Toggle mute using pactl, returns new mute state."""
try:
result = subprocess.run(
["pactl", "set-sink-mute", "@DEFAULT_SINK@", "toggle"],
capture_output=True,
timeout=5,
)
if result.returncode == 0:
_, muted = self._get_volume_pulseaudio()
return muted
return False
except Exception as e:
logger.error(f"Failed to toggle mute: {e}")
return False
async def get_status(self) -> MediaStatus:
"""Get current media playback status."""
status = MediaStatus()
# Get system volume
volume, muted = self._get_volume_pulseaudio()
status.volume = volume
status.muted = muted
# Get active player
player_name = self._get_active_player()
if player_name is None:
status.state = MediaState.IDLE
return status
# Get playback status
playback_status = self._get_property(player_name, "PlaybackStatus")
if playback_status == "Playing":
status.state = MediaState.PLAYING
elif playback_status == "Paused":
status.state = MediaState.PAUSED
elif playback_status == "Stopped":
status.state = MediaState.STOPPED
else:
status.state = MediaState.IDLE
# Get metadata
metadata = self._get_property(player_name, "Metadata")
if metadata:
status.title = str(metadata.get("xesam:title", "")) or None
artists = metadata.get("xesam:artist", [])
if artists:
status.artist = str(artists[0]) if isinstance(artists, list) else str(artists)
status.album = str(metadata.get("xesam:album", "")) or None
status.album_art_url = str(metadata.get("mpris:artUrl", "")) or None
# Duration in microseconds
length = metadata.get("mpris:length", 0)
if length:
status.duration = int(length) / 1_000_000
# Get position (in microseconds)
position = self._get_property(player_name, "Position")
if position is not None:
status.position = int(position) / 1_000_000
# Get source name
status.source = player_name.replace(self.MPRIS_PREFIX, "")
return status
async def play(self) -> bool:
"""Resume playback."""
player_name = self._get_active_player()
if player_name is None:
return False
try:
player = self._get_player_interface(player_name)
player.Play()
return True
except Exception as e:
logger.error(f"Failed to play: {e}")
return False
async def pause(self) -> bool:
"""Pause playback."""
player_name = self._get_active_player()
if player_name is None:
return False
try:
player = self._get_player_interface(player_name)
player.Pause()
return True
except Exception as e:
logger.error(f"Failed to pause: {e}")
return False
async def stop(self) -> bool:
"""Stop playback."""
player_name = self._get_active_player()
if player_name is None:
return False
try:
player = self._get_player_interface(player_name)
player.Stop()
return True
except Exception as e:
logger.error(f"Failed to stop: {e}")
return False
async def next_track(self) -> bool:
"""Skip to next track."""
player_name = self._get_active_player()
if player_name is None:
return False
try:
player = self._get_player_interface(player_name)
player.Next()
return True
except Exception as e:
logger.error(f"Failed to skip next: {e}")
return False
async def previous_track(self) -> bool:
"""Go to previous track."""
player_name = self._get_active_player()
if player_name is None:
return False
try:
player = self._get_player_interface(player_name)
player.Previous()
return True
except Exception as e:
logger.error(f"Failed to skip previous: {e}")
return False
async def set_volume(self, volume: int) -> bool:
"""Set system volume."""
return self._set_volume_pulseaudio(volume)
async def toggle_mute(self) -> bool:
"""Toggle mute state."""
return self._toggle_mute_pulseaudio()
async def seek(self, position: float) -> bool:
"""Seek to position in seconds."""
player_name = self._get_active_player()
if player_name is None:
return False
try:
player = self._get_player_interface(player_name)
# MPRIS expects position in microseconds
player.SetPosition(
self._get_property(player_name, "Metadata").get("mpris:trackid", "/"),
int(position * 1_000_000),
)
return True
except Exception as e:
logger.error(f"Failed to seek: {e}")
return False

View File

@@ -0,0 +1,296 @@
"""macOS media controller using AppleScript and system commands."""
import asyncio
import logging
import subprocess
import json
from typing import Optional
from ..models import MediaState, MediaStatus
from .media_controller import MediaController
logger = logging.getLogger(__name__)
class MacOSMediaController(MediaController):
"""Media controller for macOS using osascript and system commands."""
def _run_osascript(self, script: str) -> Optional[str]:
"""Run an AppleScript and return the output."""
try:
result = subprocess.run(
["osascript", "-e", script],
capture_output=True,
text=True,
timeout=5,
)
if result.returncode == 0:
return result.stdout.strip()
return None
except Exception as e:
logger.error(f"osascript error: {e}")
return None
def _get_active_app(self) -> Optional[str]:
"""Get the currently active media application."""
# Check common media apps in order of preference
apps = ["Spotify", "Music", "TV", "VLC", "QuickTime Player"]
for app in apps:
script = f'''
tell application "System Events"
if exists (processes where name is "{app}") then
return "{app}"
end if
end tell
return ""
'''
result = self._run_osascript(script)
if result:
return result
return None
def _get_spotify_info(self) -> dict:
"""Get playback info from Spotify."""
script = '''
tell application "Spotify"
if player state is playing then
set currentState to "playing"
else if player state is paused then
set currentState to "paused"
else
set currentState to "stopped"
end if
try
set trackName to name of current track
set artistName to artist of current track
set albumName to album of current track
set trackDuration to duration of current track
set trackPosition to player position
set artUrl to artwork url of current track
on error
set trackName to ""
set artistName to ""
set albumName to ""
set trackDuration to 0
set trackPosition to 0
set artUrl to ""
end try
return currentState & "|" & trackName & "|" & artistName & "|" & albumName & "|" & trackDuration & "|" & trackPosition & "|" & artUrl
end tell
'''
result = self._run_osascript(script)
if result:
parts = result.split("|")
if len(parts) >= 7:
return {
"state": parts[0],
"title": parts[1] or None,
"artist": parts[2] or None,
"album": parts[3] or None,
"duration": float(parts[4]) / 1000 if parts[4] else None, # ms to seconds
"position": float(parts[5]) if parts[5] else None,
"art_url": parts[6] or None,
}
return {}
def _get_music_info(self) -> dict:
"""Get playback info from Apple Music."""
script = '''
tell application "Music"
if player state is playing then
set currentState to "playing"
else if player state is paused then
set currentState to "paused"
else
set currentState to "stopped"
end if
try
set trackName to name of current track
set artistName to artist of current track
set albumName to album of current track
set trackDuration to duration of current track
set trackPosition to player position
on error
set trackName to ""
set artistName to ""
set albumName to ""
set trackDuration to 0
set trackPosition to 0
end try
return currentState & "|" & trackName & "|" & artistName & "|" & albumName & "|" & trackDuration & "|" & trackPosition
end tell
'''
result = self._run_osascript(script)
if result:
parts = result.split("|")
if len(parts) >= 6:
return {
"state": parts[0],
"title": parts[1] or None,
"artist": parts[2] or None,
"album": parts[3] or None,
"duration": float(parts[4]) if parts[4] else None,
"position": float(parts[5]) if parts[5] else None,
}
return {}
def _get_volume(self) -> tuple[int, bool]:
"""Get system volume and mute state."""
try:
# Get volume level
result = self._run_osascript("output volume of (get volume settings)")
volume = int(result) if result else 100
# Get mute state
result = self._run_osascript("output muted of (get volume settings)")
muted = result == "true"
return volume, muted
except Exception as e:
logger.error(f"Failed to get volume: {e}")
return 100, False
async def get_status(self) -> MediaStatus:
"""Get current media playback status."""
status = MediaStatus()
# Get system volume
volume, muted = self._get_volume()
status.volume = volume
status.muted = muted
# Try to get info from active media app
active_app = self._get_active_app()
if active_app is None:
status.state = MediaState.IDLE
return status
status.source = active_app
if active_app == "Spotify":
info = self._get_spotify_info()
elif active_app == "Music":
info = self._get_music_info()
else:
info = {}
if info:
state = info.get("state", "stopped")
if state == "playing":
status.state = MediaState.PLAYING
elif state == "paused":
status.state = MediaState.PAUSED
else:
status.state = MediaState.STOPPED
status.title = info.get("title")
status.artist = info.get("artist")
status.album = info.get("album")
status.duration = info.get("duration")
status.position = info.get("position")
status.album_art_url = info.get("art_url")
else:
status.state = MediaState.IDLE
return status
async def play(self) -> bool:
"""Resume playback using media key simulation."""
# Use system media key
script = '''
tell application "System Events"
key code 16 using {command down, option down}
end tell
'''
# Fallback: try specific app
active_app = self._get_active_app()
if active_app == "Spotify":
self._run_osascript('tell application "Spotify" to play')
return True
elif active_app == "Music":
self._run_osascript('tell application "Music" to play')
return True
# Use media key simulation
result = subprocess.run(
["osascript", "-e", 'tell application "System Events" to key code 49'],
capture_output=True,
)
return result.returncode == 0
async def pause(self) -> bool:
"""Pause playback."""
active_app = self._get_active_app()
if active_app == "Spotify":
self._run_osascript('tell application "Spotify" to pause')
return True
elif active_app == "Music":
self._run_osascript('tell application "Music" to pause')
return True
return False
async def stop(self) -> bool:
"""Stop playback."""
active_app = self._get_active_app()
if active_app == "Spotify":
self._run_osascript('tell application "Spotify" to pause')
return True
elif active_app == "Music":
self._run_osascript('tell application "Music" to stop')
return True
return False
async def next_track(self) -> bool:
"""Skip to next track."""
active_app = self._get_active_app()
if active_app == "Spotify":
self._run_osascript('tell application "Spotify" to next track')
return True
elif active_app == "Music":
self._run_osascript('tell application "Music" to next track')
return True
return False
async def previous_track(self) -> bool:
"""Go to previous track."""
active_app = self._get_active_app()
if active_app == "Spotify":
self._run_osascript('tell application "Spotify" to previous track')
return True
elif active_app == "Music":
self._run_osascript('tell application "Music" to previous track')
return True
return False
async def set_volume(self, volume: int) -> bool:
"""Set system volume."""
result = self._run_osascript(f"set volume output volume {volume}")
return result is not None or True # osascript returns empty on success
async def toggle_mute(self) -> bool:
"""Toggle mute state."""
_, current_mute = self._get_volume()
new_mute = not current_mute
self._run_osascript(f"set volume output muted {str(new_mute).lower()}")
return new_mute
async def seek(self, position: float) -> bool:
"""Seek to position in seconds."""
active_app = self._get_active_app()
if active_app == "Spotify":
self._run_osascript(
f'tell application "Spotify" to set player position to {position}'
)
return True
elif active_app == "Music":
self._run_osascript(
f'tell application "Music" to set player position to {position}'
)
return True
return False

View File

@@ -0,0 +1,96 @@
"""Abstract base class for media controllers."""
from abc import ABC, abstractmethod
from ..models import MediaStatus
class MediaController(ABC):
"""Abstract base class for platform-specific media controllers."""
@abstractmethod
async def get_status(self) -> MediaStatus:
"""Get the current media playback status.
Returns:
MediaStatus with current playback info
"""
pass
@abstractmethod
async def play(self) -> bool:
"""Resume or start playback.
Returns:
True if successful, False otherwise
"""
pass
@abstractmethod
async def pause(self) -> bool:
"""Pause playback.
Returns:
True if successful, False otherwise
"""
pass
@abstractmethod
async def stop(self) -> bool:
"""Stop playback.
Returns:
True if successful, False otherwise
"""
pass
@abstractmethod
async def next_track(self) -> bool:
"""Skip to the next track.
Returns:
True if successful, False otherwise
"""
pass
@abstractmethod
async def previous_track(self) -> bool:
"""Go to the previous track.
Returns:
True if successful, False otherwise
"""
pass
@abstractmethod
async def set_volume(self, volume: int) -> bool:
"""Set the system volume.
Args:
volume: Volume level (0-100)
Returns:
True if successful, False otherwise
"""
pass
@abstractmethod
async def toggle_mute(self) -> bool:
"""Toggle the mute state.
Returns:
The new mute state (True = muted)
"""
pass
@abstractmethod
async def seek(self, position: float) -> bool:
"""Seek to a position in the current track.
Args:
position: Position in seconds
Returns:
True if successful, False otherwise
"""
pass

View File

@@ -0,0 +1,189 @@
"""WebSocket connection manager and status broadcaster."""
import asyncio
import logging
import time
from typing import Any, Callable, Coroutine
from fastapi import WebSocket
logger = logging.getLogger(__name__)
class ConnectionManager:
"""Manages WebSocket connections and broadcasts status updates."""
def __init__(self) -> None:
"""Initialize the connection manager."""
self._active_connections: set[WebSocket] = set()
self._lock = asyncio.Lock()
self._last_status: dict[str, Any] | None = None
self._broadcast_task: asyncio.Task | None = None
self._poll_interval: float = 0.5 # Internal poll interval for change detection
self._position_broadcast_interval: float = 5.0 # Send position updates every 5s during playback
self._last_broadcast_time: float = 0.0
self._running: bool = False
async def connect(self, websocket: WebSocket) -> None:
"""Accept a new WebSocket connection."""
await websocket.accept()
async with self._lock:
self._active_connections.add(websocket)
logger.info(
"WebSocket client connected. Total: %d", len(self._active_connections)
)
# Send current status immediately upon connection
if self._last_status:
try:
await websocket.send_json({"type": "status", "data": self._last_status})
except Exception as e:
logger.debug("Failed to send initial status: %s", e)
async def disconnect(self, websocket: WebSocket) -> None:
"""Remove a WebSocket connection."""
async with self._lock:
self._active_connections.discard(websocket)
logger.info(
"WebSocket client disconnected. Total: %d", len(self._active_connections)
)
async def broadcast(self, message: dict[str, Any]) -> None:
"""Broadcast a message to all connected clients."""
async with self._lock:
connections = list(self._active_connections)
if not connections:
return
disconnected = []
for websocket in connections:
try:
await websocket.send_json(message)
except Exception as e:
logger.debug("Failed to send to client: %s", e)
disconnected.append(websocket)
# Clean up disconnected clients
for ws in disconnected:
await self.disconnect(ws)
def status_changed(
self, old: dict[str, Any] | None, new: dict[str, Any]
) -> bool:
"""Detect if media status has meaningfully changed.
Position is NOT included for normal playback (let HA interpolate).
But seeks (large unexpected jumps) are detected.
"""
if old is None:
return True
# Fields to compare for changes (NO position - let HA interpolate)
significant_fields = [
"state",
"title",
"artist",
"album",
"volume",
"muted",
"duration",
"source",
"album_art_url",
]
for field in significant_fields:
if old.get(field) != new.get(field):
return True
# Detect seeks - large position jumps that aren't normal playback
old_pos = old.get("position") or 0
new_pos = new.get("position") or 0
pos_diff = new_pos - old_pos
# During playback, position should increase by ~0.5s (our poll interval)
# A seek is when position jumps backwards OR forward by more than expected
if new.get("state") == "playing":
# Backward seek or forward jump > 3s indicates seek
if pos_diff < -1.0 or pos_diff > 3.0:
return True
else:
# When paused, any significant position change is a seek
if abs(pos_diff) > 1.0:
return True
return False
async def start_status_monitor(
self,
get_status_func: Callable[[], Coroutine[Any, Any, Any]],
) -> None:
"""Start the background status monitoring loop."""
if self._running:
return
self._running = True
self._broadcast_task = asyncio.create_task(
self._status_monitor_loop(get_status_func)
)
logger.info("WebSocket status monitor started")
async def stop_status_monitor(self) -> None:
"""Stop the background status monitoring loop."""
self._running = False
if self._broadcast_task:
self._broadcast_task.cancel()
try:
await self._broadcast_task
except asyncio.CancelledError:
pass
logger.info("WebSocket status monitor stopped")
async def _status_monitor_loop(
self,
get_status_func: Callable[[], Coroutine[Any, Any, Any]],
) -> None:
"""Background loop that polls for status changes and broadcasts."""
while self._running:
try:
# Only poll if we have connected clients
async with self._lock:
has_clients = len(self._active_connections) > 0
if has_clients:
status = await get_status_func()
status_dict = status.model_dump()
# Only broadcast on actual state changes
# Let HA handle position interpolation during playback
if self.status_changed(self._last_status, status_dict):
self._last_status = status_dict
self._last_broadcast_time = time.time()
await self.broadcast(
{"type": "status_update", "data": status_dict}
)
logger.debug("Broadcast sent: status change")
else:
# Update cached status even without broadcast
self._last_status = status_dict
else:
# Still update cache for when clients connect
status = await get_status_func()
self._last_status = status.model_dump()
await asyncio.sleep(self._poll_interval)
except asyncio.CancelledError:
break
except Exception as e:
logger.error("Error in status monitor: %s", e)
await asyncio.sleep(self._poll_interval)
@property
def client_count(self) -> int:
"""Return the number of connected clients."""
return len(self._active_connections)
# Global instance
ws_manager = ConnectionManager()

View File

@@ -0,0 +1,596 @@
"""Windows media controller using WinRT APIs."""
import asyncio
import logging
from concurrent.futures import ThreadPoolExecutor
from typing import Optional, Any
from ..models import MediaState, MediaStatus
from .media_controller import MediaController
logger = logging.getLogger(__name__)
# Thread pool for WinRT operations (they don't play well with asyncio)
_executor = ThreadPoolExecutor(max_workers=2, thread_name_prefix="winrt")
# Global storage for current album art (as bytes)
_current_album_art_bytes: bytes | None = None
# Global storage for position tracking
import time as _time
_position_cache = {
"track_id": "",
"base_position": 0.0,
"base_time": 0.0,
"is_playing": False,
"duration": 0.0,
}
# Flag to force position to 0 after track skip (until title changes)
_track_skip_pending = {
"active": False,
"old_title": "",
"skip_time": 0.0,
"grace_until": 0.0, # After title changes, ignore stale SMTC positions
"stale_pos": -999, # The stale SMTC position we're ignoring
}
def get_current_album_art() -> bytes | None:
"""Get the current album art bytes."""
return _current_album_art_bytes
# Windows-specific imports
try:
from winsdk.windows.media.control import (
GlobalSystemMediaTransportControlsSessionManager as MediaManager,
GlobalSystemMediaTransportControlsSessionPlaybackStatus as PlaybackStatus,
)
WINSDK_AVAILABLE = True
except ImportError:
WINSDK_AVAILABLE = False
logger.warning("winsdk not available")
# Volume control imports
PYCAW_AVAILABLE = False
_volume_control = None
try:
from ctypes import cast, POINTER
from comtypes import CLSCTX_ALL, CoInitialize, CoUninitialize
from pycaw.pycaw import AudioUtilities, IAudioEndpointVolume
def _init_volume_control():
"""Initialize volume control interface."""
global _volume_control
if _volume_control is not None:
return _volume_control
try:
devices = AudioUtilities.GetSpeakers()
interface = devices.Activate(IAudioEndpointVolume._iid_, CLSCTX_ALL, None)
_volume_control = cast(interface, POINTER(IAudioEndpointVolume))
return _volume_control
except AttributeError:
# Try accessing the underlying device
try:
devices = AudioUtilities.GetSpeakers()
if hasattr(devices, '_dev'):
interface = devices._dev.Activate(IAudioEndpointVolume._iid_, CLSCTX_ALL, None)
_volume_control = cast(interface, POINTER(IAudioEndpointVolume))
return _volume_control
except Exception as e:
logger.debug(f"Volume control init failed: {e}")
except Exception as e:
logger.debug(f"Volume control init error: {e}")
return None
PYCAW_AVAILABLE = True
except ImportError as e:
logger.warning(f"pycaw not available: {e}")
def _init_volume_control():
return None
WINDOWS_AVAILABLE = WINSDK_AVAILABLE
def _sync_get_media_status() -> dict[str, Any]:
"""Synchronously get media status (runs in thread pool)."""
import asyncio
result = {
"state": "idle",
"title": None,
"artist": None,
"album": None,
"duration": None,
"position": None,
"source": None,
}
try:
# Create a new event loop for this thread
loop = asyncio.new_event_loop()
asyncio.set_event_loop(loop)
try:
# Get media session manager
manager = loop.run_until_complete(MediaManager.request_async())
if manager is None:
return result
session = _find_best_session(manager, loop)
if session is None:
return result
# Get playback status
playback_info = session.get_playback_info()
if playback_info:
status = playback_info.playback_status
if status == PlaybackStatus.PLAYING:
result["state"] = "playing"
elif status == PlaybackStatus.PAUSED:
result["state"] = "paused"
elif status == PlaybackStatus.STOPPED:
result["state"] = "stopped"
# Get media properties FIRST (needed for track ID)
media_props = loop.run_until_complete(
session.try_get_media_properties_async()
)
if media_props:
result["title"] = media_props.title or None
result["artist"] = media_props.artist or None
result["album"] = media_props.album_title or None
# Get timeline
timeline = session.get_timeline_properties()
if timeline:
try:
# end_time and position are datetime.timedelta objects
end_time = timeline.end_time
position = timeline.position
# Get duration
if hasattr(end_time, 'total_seconds'):
duration = end_time.total_seconds()
# Sanity check: duration should be positive and reasonable (< 24 hours)
if 0 < duration < 86400:
result["duration"] = duration
# Get position from SMTC and interpolate for smooth updates
if hasattr(position, 'total_seconds'):
smtc_pos = position.total_seconds()
current_time = _time.time()
is_playing = result["state"] == "playing"
current_title = result.get('title', '')
# Check if track skip is pending and title changed
skip_just_completed = False
if _track_skip_pending["active"]:
if current_title and current_title != _track_skip_pending["old_title"]:
# Title changed - clear the skip flag and start grace period
_track_skip_pending["active"] = False
_track_skip_pending["old_title"] = ""
_track_skip_pending["grace_until"] = current_time + 300.0 # Long grace period
_track_skip_pending["stale_pos"] = -999 # Reset stale position tracking
skip_just_completed = True
# Reset position cache for new track
new_track_id = f"{current_title}:{result.get('artist', '')}:{result.get('duration', 0)}"
_position_cache["track_id"] = new_track_id
_position_cache["base_position"] = 0.0
_position_cache["base_time"] = current_time
_position_cache["last_smtc_pos"] = -999 # Force fresh start
_position_cache["is_playing"] = is_playing
logger.debug(f"Track skip complete, new title: {current_title}, grace until: {_track_skip_pending['grace_until']}")
elif current_time - _track_skip_pending["skip_time"] > 5.0:
# Timeout after 5 seconds
_track_skip_pending["active"] = False
logger.debug("Track skip timeout")
# Check if we're in grace period (after skip, ignore high SMTC positions)
in_grace_period = current_time < _track_skip_pending.get("grace_until", 0)
# If track skip is pending or just completed, use cached/reset position
if _track_skip_pending["active"]:
pos = 0.0
_position_cache["base_position"] = 0.0
_position_cache["base_time"] = current_time
_position_cache["is_playing"] = is_playing
elif skip_just_completed:
# Just completed skip - interpolate from 0
if is_playing:
elapsed = current_time - _position_cache["base_time"]
pos = elapsed
else:
pos = 0.0
elif in_grace_period:
# Grace period after track skip
# SMTC position is stale (from old track) and won't update until seek/pause
# We interpolate from 0 and only trust SMTC when it changes or reports low value
# Calculate interpolated position from start of new track
if is_playing:
elapsed = current_time - _position_cache.get("base_time", current_time)
interpolated_pos = _position_cache.get("base_position", 0.0) + elapsed
else:
interpolated_pos = _position_cache.get("base_position", 0.0)
# Get the stale position we've been tracking
stale_pos = _track_skip_pending.get("stale_pos", -999)
# Detect if SMTC position changed significantly from the stale value (user seeked)
smtc_changed = stale_pos >= 0 and abs(smtc_pos - stale_pos) > 3.0
# Trust SMTC if:
# 1. It reports a low position (indicating new track started)
# 2. It changed from the stale value (user seeked)
if smtc_pos < 10.0 or smtc_changed:
# SMTC is now trustworthy
_position_cache["base_position"] = smtc_pos
_position_cache["base_time"] = current_time
_position_cache["last_smtc_pos"] = smtc_pos
_position_cache["is_playing"] = is_playing
pos = smtc_pos
_track_skip_pending["grace_until"] = 0
_track_skip_pending["stale_pos"] = -999
logger.debug(f"Grace period: accepting SMTC pos {smtc_pos} (low={smtc_pos < 10}, changed={smtc_changed})")
else:
# SMTC is stale - keep interpolating
pos = interpolated_pos
# Record the stale position for change detection
if stale_pos < 0:
_track_skip_pending["stale_pos"] = smtc_pos
# Keep grace period active indefinitely while SMTC is stale
_track_skip_pending["grace_until"] = current_time + 300.0
logger.debug(f"Grace period: SMTC stale ({smtc_pos}), using interpolated {interpolated_pos}")
else:
# Normal position tracking
# Create track ID from title + artist + duration
track_id = f"{current_title}:{result.get('artist', '')}:{result.get('duration', 0)}"
# Detect if SMTC position changed (new track, seek, or state change)
smtc_pos_changed = abs(smtc_pos - _position_cache.get("last_smtc_pos", -999)) > 0.5
track_changed = track_id != _position_cache.get("track_id", "")
if smtc_pos_changed or track_changed:
# SMTC updated - store new baseline
_position_cache["track_id"] = track_id
_position_cache["last_smtc_pos"] = smtc_pos
_position_cache["base_position"] = smtc_pos
_position_cache["base_time"] = current_time
_position_cache["is_playing"] = is_playing
pos = smtc_pos
elif is_playing:
# Interpolate position based on elapsed time
elapsed = current_time - _position_cache.get("base_time", current_time)
pos = _position_cache.get("base_position", smtc_pos) + elapsed
else:
# Paused - use base position
pos = _position_cache.get("base_position", smtc_pos)
# Update playing state
if _position_cache.get("is_playing") != is_playing:
_position_cache["base_position"] = pos if is_playing else _position_cache.get("base_position", smtc_pos)
_position_cache["base_time"] = current_time
_position_cache["is_playing"] = is_playing
# Sanity check: position should be non-negative and <= duration
if pos >= 0:
if result["duration"] and pos <= result["duration"]:
result["position"] = pos
elif result["duration"] and pos > result["duration"]:
result["position"] = result["duration"]
elif not result["duration"]:
result["position"] = pos
logger.debug(f"Timeline: duration={result['duration']}, position={result['position']}")
except Exception as e:
logger.debug(f"Timeline parse error: {e}")
# Try to get album art (requires media_props)
if media_props:
try:
thumbnail = media_props.thumbnail
if thumbnail:
stream = loop.run_until_complete(thumbnail.open_read_async())
if stream:
size = stream.size
if size > 0 and size < 10 * 1024 * 1024: # Max 10MB
from winsdk.windows.storage.streams import DataReader
reader = DataReader(stream)
loop.run_until_complete(reader.load_async(size))
buffer = bytearray(size)
reader.read_bytes(buffer)
reader.close()
stream.close()
global _current_album_art_bytes
_current_album_art_bytes = bytes(buffer)
result["album_art_url"] = "/api/media/artwork"
except Exception as e:
logger.debug(f"Failed to get album art: {e}")
result["source"] = session.source_app_user_model_id
finally:
loop.close()
except Exception as e:
logger.error(f"Error getting media status: {e}")
return result
def _find_best_session(manager, loop):
"""Find the best media session to control."""
# First try the current session
session = manager.get_current_session()
# Log all available sessions for debugging
sessions = manager.get_sessions()
if sessions:
logger.debug(f"Total sessions available: {sessions.size}")
for i in range(sessions.size):
s = sessions.get_at(i)
if s:
playback_info = s.get_playback_info()
status_name = "unknown"
if playback_info:
status_name = str(playback_info.playback_status)
logger.debug(f" Session {i}: {s.source_app_user_model_id} - status: {status_name}")
# If no current session, try to find any active session
if session is None:
if sessions and sessions.size > 0:
# Find a playing session, or use the first one
for i in range(sessions.size):
s = sessions.get_at(i)
if s:
playback_info = s.get_playback_info()
if playback_info and playback_info.playback_status == PlaybackStatus.PLAYING:
session = s
break
# If no playing session found, use the first available one
if session is None and sessions.size > 0:
session = sessions.get_at(0)
return session
def _sync_media_command(command: str) -> bool:
"""Synchronously execute a media command (runs in thread pool)."""
import asyncio
try:
loop = asyncio.new_event_loop()
asyncio.set_event_loop(loop)
try:
manager = loop.run_until_complete(MediaManager.request_async())
if manager is None:
return False
session = _find_best_session(manager, loop)
if session is None:
return False
if command == "play":
return loop.run_until_complete(session.try_play_async())
elif command == "pause":
return loop.run_until_complete(session.try_pause_async())
elif command == "stop":
return loop.run_until_complete(session.try_stop_async())
elif command == "next":
return loop.run_until_complete(session.try_skip_next_async())
elif command == "previous":
return loop.run_until_complete(session.try_skip_previous_async())
return False
finally:
loop.close()
except Exception as e:
logger.error(f"Error executing media command {command}: {e}")
return False
def _sync_seek(position: float) -> bool:
"""Synchronously seek to position."""
import asyncio
try:
loop = asyncio.new_event_loop()
asyncio.set_event_loop(loop)
try:
manager = loop.run_until_complete(MediaManager.request_async())
if manager is None:
return False
session = _find_best_session(manager, loop)
if session is None:
return False
position_ticks = int(position * 10_000_000)
return loop.run_until_complete(
session.try_change_playback_position_async(position_ticks)
)
finally:
loop.close()
except Exception as e:
logger.error(f"Error seeking: {e}")
return False
class WindowsMediaController(MediaController):
"""Media controller for Windows using WinRT and pycaw."""
def __init__(self):
if not WINDOWS_AVAILABLE:
raise RuntimeError(
"Windows media control requires winsdk, pycaw, and comtypes packages"
)
self._volume_interface = None
self._volume_init_attempted = False
def _get_volume_interface(self):
"""Get the audio endpoint volume interface."""
if not self._volume_init_attempted:
self._volume_init_attempted = True
self._volume_interface = _init_volume_control()
if self._volume_interface:
logger.info("Volume control initialized successfully")
else:
logger.warning("Volume control not available")
return self._volume_interface
async def get_status(self) -> MediaStatus:
"""Get current media playback status."""
status = MediaStatus()
# Get volume info (synchronous, fast)
volume_if = self._get_volume_interface()
if volume_if:
try:
volume_scalar = volume_if.GetMasterVolumeLevelScalar()
status.volume = int(volume_scalar * 100)
status.muted = bool(volume_if.GetMute())
except Exception as e:
logger.debug(f"Failed to get volume: {e}")
# Get media info in thread pool (avoids asyncio/WinRT issues)
try:
loop = asyncio.get_event_loop()
media_info = await asyncio.wait_for(
loop.run_in_executor(_executor, _sync_get_media_status),
timeout=5.0
)
state_map = {
"playing": MediaState.PLAYING,
"paused": MediaState.PAUSED,
"stopped": MediaState.STOPPED,
"idle": MediaState.IDLE,
}
status.state = state_map.get(media_info.get("state", "idle"), MediaState.IDLE)
status.title = media_info.get("title")
status.artist = media_info.get("artist")
status.album = media_info.get("album")
status.album_art_url = media_info.get("album_art_url")
status.duration = media_info.get("duration")
status.position = media_info.get("position")
status.source = media_info.get("source")
except asyncio.TimeoutError:
logger.warning("Media status request timed out")
status.state = MediaState.IDLE
except Exception as e:
logger.error(f"Error getting media status: {e}")
status.state = MediaState.IDLE
return status
async def _run_command(self, command: str) -> bool:
"""Run a media command in the thread pool."""
try:
loop = asyncio.get_event_loop()
return await asyncio.wait_for(
loop.run_in_executor(_executor, _sync_media_command, command),
timeout=5.0
)
except asyncio.TimeoutError:
logger.warning(f"Media command {command} timed out")
return False
except Exception as e:
logger.error(f"Error running media command {command}: {e}")
return False
async def play(self) -> bool:
"""Resume playback."""
return await self._run_command("play")
async def pause(self) -> bool:
"""Pause playback."""
return await self._run_command("pause")
async def stop(self) -> bool:
"""Stop playback."""
return await self._run_command("stop")
async def next_track(self) -> bool:
"""Skip to next track."""
# Get current title before skipping
try:
status = await self.get_status()
old_title = status.title or ""
except Exception:
old_title = ""
result = await self._run_command("next")
if result:
# Set flag to force position to 0 until title changes
_track_skip_pending["active"] = True
_track_skip_pending["old_title"] = old_title
_track_skip_pending["skip_time"] = _time.time()
logger.debug(f"Track skip initiated, old title: {old_title}")
return result
async def previous_track(self) -> bool:
"""Go to previous track."""
# Get current title before skipping
try:
status = await self.get_status()
old_title = status.title or ""
except Exception:
old_title = ""
result = await self._run_command("previous")
if result:
# Set flag to force position to 0 until title changes
_track_skip_pending["active"] = True
_track_skip_pending["old_title"] = old_title
_track_skip_pending["skip_time"] = _time.time()
logger.debug(f"Track skip initiated, old title: {old_title}")
return result
async def set_volume(self, volume: int) -> bool:
"""Set system volume."""
volume_if = self._get_volume_interface()
if volume_if is None:
return False
try:
volume_if.SetMasterVolumeLevelScalar(volume / 100.0, None)
return True
except Exception as e:
logger.error(f"Failed to set volume: {e}")
return False
async def toggle_mute(self) -> bool:
"""Toggle mute state."""
volume_if = self._get_volume_interface()
if volume_if is None:
return False
try:
current_mute = bool(volume_if.GetMute())
volume_if.SetMute(not current_mute, None)
return not current_mute
except Exception as e:
logger.error(f"Failed to toggle mute: {e}")
return False
async def seek(self, position: float) -> bool:
"""Seek to position in seconds."""
try:
loop = asyncio.get_event_loop()
return await asyncio.wait_for(
loop.run_in_executor(_executor, _sync_seek, position),
timeout=5.0
)
except asyncio.TimeoutError:
logger.warning("Seek command timed out")
return False
except Exception as e:
logger.error(f"Failed to seek: {e}")
return False