Update __main__.py

(Major change) main() function completely refactored into helpers. _init_controllers now handles decoupled initialization. _process_audio loop logic was fixed to correctly initialize wake words on startup.

linux_voice_assistant/__main__.py

@@ -6,15 +6,16 @@ import logging
 import sys
 import threading
 import time
+from dataclasses import dataclass
 from pathlib import Path
-from typing import Dict, List, Optional, Set, Union
+from typing import Dict, List, Optional, Set, Tuple, Union
 
 import numpy as np
 import soundcard as sc
 from pymicro_wakeword import MicroWakeWord, MicroWakeWordFeatures
 from pyopen_wakeword import OpenWakeWord, OpenWakeWordFeatures
 
-from .config import Config, load_config_from_json
+from .config import Config, LedConfig, MqttConfig, load_config_from_json
 from .event_bus import EventBus, EventHandler, subscribe
 from .led_controller import LedController
 from .mqtt_controller import MqttController
@@ -28,16 +29,47 @@ _LOGGER = logging.getLogger(__name__)
 _MODULE_DIR = Path(__file__).parent
 _REPO_DIR = _MODULE_DIR.parent
 
+# -----------------------------------------------------------------------------
+# Helper dataclasses
+# -----------------------------------------------------------------------------
+
+@dataclass
+class WakeWordData:
+    """Dataclass to hold all wake word models and settings."""
+    available: Dict[str, AvailableWakeWord]
+    models: Dict[str, Union[MicroWakeWord, OpenWakeWord]]
+    active: Set[str]
+    stop_model: MicroWakeWord
+
+@dataclass
+class MediaPlayers:
+    """Dataclass to hold media player instances."""
+    music: MpvMediaPlayer
+    tts: MpvMediaPlayer
+
+# -----------------------------------------------------------------------------
+# Mic Mute Handler
 # -----------------------------------------------------------------------------
 
 class MicMuteHandler(EventHandler):
-    """Event handler for mic mute switch."""
+    """
-    def __init__(self, state, mqtt_controller: Optional[MqttController]):
+    Manages the mic_muted state and saves preferences.
-        super().__init__(state)
+    This is the only controller that writes to ServerState.
+    """
+    def __init__(
+        self,
+        event_bus: EventBus,
+        state: ServerState,
+        mqtt_controller: Optional[MqttController],
+    ):
+        super().__init__(event_bus)
+        self.state = state
         self.mqtt_controller = mqtt_controller
+        self._subscribe_all_methods()
 
     @subscribe
     def set_mic_mute(self, data: dict):
+        """Event handler for mic mute commands."""
         is_muted = data.get("state", False)
         if self.state.mic_muted != is_muted:
             self.state.mic_muted = is_muted
@@ -47,70 +79,112 @@ class MicMuteHandler(EventHandler):
                 self.mqtt_controller.publish_mute_state(is_muted)
             
             if is_muted:
-                self.state.event_bus.publish("mic_muted")
+                self.event_bus.publish("mic_muted")
             else:
-                self.state.event_bus.publish("mic_unmuted")
+                self.event_bus.publish("mic_unmuted")
+
+    @subscribe
+    def set_num_leds(self, data: dict):
+        """Event handler to save num_leds to preferences."""
+        num_leds = data.get("num_leds")
+        if (num_leds is not None) and (self.state.preferences.num_leds != num_leds):
+            self.state.preferences.num_leds = num_leds
+            self.state.save_preferences()
 
 # -----------------------------------------------------------------------------
-
+# Main Application
+# -----------------------------------------------------------------------------
 
 async def main() -> None:
+    # --- 1. Load Basics ---
+    config, loop, event_bus = _init_basics()
+    
+    # --- 2. Load Preferences ---
+    preferences = _load_preferences(config)
+
+    # --- 3. Find Microphone ---
+    mic = _get_microphone(config)
+    
+    # --- 4. Load Wake Words ---
+    wake_word_data = _load_wake_words(config, preferences)
+
+    # --- 5. Initialize Media Players ---
+    media_players = _init_media_players(loop, config, preferences)
+
+    # --- 6. Create Server State ---
+    state = _create_server_state(
+        config, loop, event_bus, preferences, 
+        wake_word_data, media_players
+    )
+
+    # --- 7. Initialize Controllers ---
+    _init_controllers(loop, event_bus, state, config, preferences)
+    
+    # --- 8. Start Audio Thread ---
+    audio_thread = threading.Thread(
+        target=_process_audio,
+        args=(state, mic, config.audio.input_block_size),
+        daemon=True
+    )
+    audio_thread.start()
+    
+    # --- 9. Run Server ---
+    try:
+        await _run_server(state, config)
+    finally:
+        # --- 10. Cleanup ---
+        _LOGGER.debug("Shutting down...")
+        state.mic_muted = True # Signal audio thread to stop
+        audio_thread.join()
+        
+        if hasattr(state, "mqtt_controller") and state.mqtt_controller:
+            _LOGGER.debug("Stopping MQTT controller...")
+            state.mqtt_controller.stop()
+
+# -----------------------------------------------------------------------------
+# Helper Functions
+# -----------------------------------------------------------------------------
+
+def _init_basics() -> Tuple[Config, asyncio.AbstractEventLoop, EventBus]:
+    """Loads config, sets up logging, and creates loop/event bus."""
     parser = argparse.ArgumentParser()
     parser.add_argument(
-        "-c",
+        "-c", "--config", type=Path, required=False,
-        "--config",
-        type=Path,
-        required=False,
         default=_MODULE_DIR / "config.json",
-        help="Path to configuration.json file (default: linux_voice_assistant/config.json)",
+        help="Path to configuration.json file"
-    )
-    parser.add_argument(
-        "--debug",
-        action="store_true",
-        help="Enable debug logging (overrides config file)",
-    )
-    parser.add_argument(
-        "--list-input-devices",
-        action="store_true",
-        help="List audio input devices and exit",
-    )
-    parser.add_argument(
-        "--list-output-devices",
-        action="store_true",
-        help="List audio output devices and exit",
     )
+    parser.add_argument("--debug", action="store_true", help="Enable debug logging")
+    parser.add_argument("--list-input-devices", action="store_true", help="List audio input devices")
+    parser.add_argument("--list-output-devices", action="store_true", help="List audio output devices")
     args = parser.parse_args()
 
-    # --- Handle List Devices ---
+    # --- List devices and exit ---
     if args.list_input_devices:
-        print("Input devices")
+        print("Input devices\n" + "=" * 13)
-        print("=" * 13)
         for idx, mic in enumerate(sc.all_microphones(include_loopback=False)):
             print(f"[{idx}]", mic.name)
-        return
+        sys.exit(0)
 
     if args.list_output_devices:
-        player = MpvMediaPlayer(loop=None)
+        print("Output devices\n" + "=" * 14)
-        print("Output devices")
-        print("=" * 14)
         try:
+            player = MpvMediaPlayer(loop=None)
             for speaker in player.player.audio_device_list:
                 print(speaker["name"] + ":", speaker["description"])
         except Exception as e:
             _LOGGER.error("Failed to list output devices: %s", e)
-        return
+        sys.exit(1)
 
-    # --- Load Configuration ---
+    # --- Load config ---
     config_path = args.config
-    
     if not config_path.is_absolute():
          config_path = _REPO_DIR / config_path
-         
     config = load_config_from_json(config_path)
 
     if args.debug:
         config.app.debug = True
 
+    # --- Setup logging ---
     logging.basicConfig(level=logging.DEBUG if config.app.debug else logging.INFO)
     _LOGGER.info("Loading configuration from: %s", config_path)
     _LOGGER.debug("Configuration loaded: %s", config)
@@ -118,106 +192,111 @@ async def main() -> None:
     loop = asyncio.get_running_loop()
     event_bus = EventBus()
     
-    # --- Create Download Dir ---
+    return config, loop, event_bus
-    download_dir = _REPO_DIR / config.wake_word.download_dir
-    download_dir.mkdir(parents=True, exist_ok=True)
-    _LOGGER.debug("Download directory: %s", download_dir)
 
-    # --- Load Preferences ---
+def _load_preferences(config: Config) -> Preferences:
+    """Loads preferences.json file."""
     preferences_path = _REPO_DIR / config.app.preferences_file
     if preferences_path.exists():
-        with open(preferences_path, "r", encoding="utf-8") as preferences_file:
+        with open(preferences_path, "r", encoding="utf-8") as f:
-            preferences_dict = json.load(preferences_file)
+            preferences_dict = json.load(f)
             preferences = Preferences(**preferences_dict)
     else:
         preferences = Preferences()
     
     preferences.num_leds = getattr(preferences, 'num_leds', config.led.num_leds)
+    return preferences
 
-    # --- Load Wake Words ---
+def _get_microphone(config: Config):
-    available_wake_words: Dict[str, AvailableWakeWord] = {}
+    """Finds and returns the microphone specified in the config."""
+    mic = None
+    if config.audio.input_device is not None:
+        try:
+            input_device_idx = int(config.audio.input_device)
+            mic = sc.all_microphones(include_loopback=False)[input_device_idx]
+        except (ValueError, IndexError):
+            mic = sc.get_microphone(config.audio.input_device, include_loopback=False)
+    else:
+        mic = sc.default_microphone()
+    
+    if mic is None:
+        _LOGGER.critical("No microphone found. Use --list-input-devices to see options.")
+        sys.exit(1)
+        
+    _LOGGER.info("Using audio input device: %s", mic.name)
+    return mic
+
+def _load_wake_words(config: Config, preferences: Preferences) -> WakeWordData:
+    """Loads all available and active wake word models."""
     
     if not config.wake_word.directories:
         config.wake_word.directories = ["wakewords", "wakewords/openWakeWord"]
         
-    # Add download directory to search path
+    download_dir = _REPO_DIR / config.wake_word.download_dir
-    wake_word_dirs = [_REPO_DIR / d for d in config.wake_word.directories]
+    download_dir.mkdir(parents=True, exist_ok=True)
-    wake_word_dirs.append(download_dir / "external_wake_words") # <-- ADDED
+    _LOGGER.debug("Download directory: %s", download_dir)
 
+    wake_word_dirs = [_REPO_DIR / d for d in config.wake_word.directories]
+    wake_word_dirs.append(download_dir / "external_wake_words")
+
+    available: Dict[str, AvailableWakeWord] = {}
     for wake_word_dir in wake_word_dirs:
         if not wake_word_dir.exists():
             continue
-            
+        for config_path in wake_word_dir.glob("*.json"):
-        for model_config_path in wake_word_dir.glob("*.json"):
+            model_id = config_path.stem
-            model_id = model_config_path.stem
             if model_id == config.wake_word.stop_model:
                 continue
-            with open(model_config_path, "r", encoding="utf-8") as model_config_file:
+            with open(config_path, "r", encoding="utf-8") as f:
-                model_config = json.load(model_config_file)
+                model_config = json.load(f)
                 model_type = WakeWordType(model_config["type"])
                 
-                if model_type == WakeWordType.OPEN_WAKE_WORD:
+                wake_word_path = (
-                    wake_word_path = model_config_path.parent / model_config["model"]
+                    config_path.parent / model_config["model"]
-                else:
+                    if model_type == WakeWordType.OPEN_WAKE_WORD
-                    wake_word_path = model_config_path
+                    else config_path
+                )
 
-                available_wake_words[model_id] = AvailableWakeWord(
+                available[model_id] = AvailableWakeWord(
                     id=model_id,
                     type=model_type,
                     wake_word=model_config["wake_word"],
                     trained_languages=model_config.get("trained_languages", []),
                     wake_word_path=wake_word_path,
                 )
+    _LOGGER.debug("Available wake words: %s", list(sorted(available.keys())))
 
-    _LOGGER.debug("Available wake words: %s", list(sorted(available_wake_words.keys())))
+    active: Set[str] = set()
+    models: Dict[str, Union[MicroWakeWord, OpenWakeWord]] = {}
     
-    # --- Load Active Wake Word Models ---
-    active_wake_words: Set[str] = set()
-    wake_models: Dict[str, Union[MicroWakeWord, OpenWakeWord]] = {}
     if preferences.active_wake_words:
-        for wake_word_id in preferences.active_wake_words:
+        for ww_id in preferences.active_wake_words:
-            wake_word = available_wake_words.get(wake_word_id)
+            if ww_id in available:
-            if wake_word is None: continue
+                _LOGGER.debug("Loading wake model (from preferences): %s", ww_id)
-            _LOGGER.debug("Loading wake model: %s", wake_word_id)
+                models[ww_id] = available[ww_id].load()
-            wake_models[wake_word_id] = wake_word.load()
+                active.add(ww_id)
-            active_wake_words.add(wake_word_id)
     
-    if not wake_models:
+    if not models:
-        wake_word_id = config.wake_word.model
+        ww_id = config.wake_word.model
-        wake_word = available_wake_words.get(wake_word_id)
+        if ww_id in available:
-        if wake_word:
+            _LOGGER.debug("Loading wake model (from config): %s", ww_id)
-            _LOGGER.debug("Loading wake model: %s", wake_word_id)
+            models[ww_id] = available[ww_id].load()
-            wake_models[wake_word_id] = wake_word.load()
+            active.add(ww_id)
-            active_wake_words.add(wake_word_id)
 
-    # --- Load Stop Model ---
     stop_model: Optional[MicroWakeWord] = None
     for ww_dir_str in config.wake_word.directories:
-        wake_word_dir = _REPO_DIR / ww_dir_str
+        stop_config_path = _REPO_DIR / ww_dir_str / f"{config.wake_word.stop_model}.json"
-        stop_config_path = wake_word_dir / f"{config.wake_word.stop_model}.json"
+        if stop_config_path.exists():
-        if not stop_config_path.exists(): continue
             _LOGGER.debug("Loading stop model: %s", stop_config_path)
             stop_model = MicroWakeWord.from_config(stop_config_path)
             break
-    assert stop_model is not None
+    assert stop_model is not None, "Stop model not found"
 
-    # --- Resolve Microphone ---
+    return WakeWordData(available, models, active, stop_model)
-    if config.audio.input_device is not None:
-        try:
-            input_device_idx = int(config.audio.input_device)
-            mic = sc.all_microphones(include_loopback=False)[input_device_idx]
-        except (ValueError, IndexError):
-            mic = sc.get_microphone(config.audio.input_device)
-    else:
-        mic = sc.default_microphone()
 
-    if mic is None:
+def _init_media_players(loop: asyncio.AbstractEventLoop, config: Config, preferences: Preferences) -> MediaPlayers:
-        _LOGGER.critical("No microphone found.")
+    """Initializes the music and TTS media players."""
-        sys.exit(1)
+    _LOGGER.debug("Initializing media players")
-        
-    _LOGGER.info("Using audio input device: %s", mic.name)
-
-    # --- Initialize Media Players ---
     music_player = MpvMediaPlayer(
         loop=loop,
         device=config.audio.output_device,
@@ -228,61 +307,78 @@ async def main() -> None:
         device=config.audio.output_device,
         initial_volume=preferences.volume_level
     )
+    return MediaPlayers(music=music_player, tts=tts_player)
 
-    # --- Create Global Server State ---
+def _create_server_state(
-    state = ServerState(
+    config: Config,
+    loop: asyncio.AbstractEventLoop,
+    event_bus: EventBus,
+    preferences: Preferences,
+    wake_word_data: WakeWordData,
+    media_players: MediaPlayers,
+) -> ServerState:
+    """Creates the global ServerState object."""
+    _LOGGER.debug("Creating server state")
+    return ServerState(
         name=config.app.name,
         mac_address=get_mac(),
+        event_bus=event_bus,
+        loop=loop,
         entities=[],
-        available_wake_words=available_wake_words,
+        music_player=media_players.music,
-        wake_words=wake_models,
+        tts_player=media_players.tts,
-        active_wake_words=active_wake_words,
+        available_wake_words=wake_word_data.available,
-        stop_word=stop_model,
+        wake_words=wake_word_data.models,
-        music_player=music_player,
+        active_wake_words=wake_word_data.active,
-        tts_player=tts_player,
+        stop_word=wake_word_data.stop_model,
         wakeup_sound=str(_REPO_DIR / config.app.wakeup_sound),
         timer_finished_sound=str(_REPO_DIR / config.app.timer_finished_sound),
         preferences=preferences,
-        preferences_path=preferences_path,
+        preferences_path=_REPO_DIR / config.app.preferences_file,
-        event_bus=event_bus,
+        download_dir=_REPO_DIR / config.wake_word.download_dir,
-        loop=loop,
-        download_dir=download_dir, # <-- ADDED
         refractory_seconds=config.wake_word.refractory_seconds,
     )
 
-    # --- Initialize Controllers ---
+def _init_controllers(
+    loop: asyncio.AbstractEventLoop,
+    event_bus: EventBus,
+    state: ServerState,
+    config: Config,
+    preferences: Preferences,
+):
+    """Initializes all decoupled controllers."""
+    _LOGGER.debug("Initializing controllers")
+    
     led_controller = LedController(
-        state,
+        loop=loop,
-        led_type=config.led.led_type,
+        event_bus=event_bus,
-        interface=config.led.interface,
+        config=config.led,
-        clock_pin=config.led.clock_pin,
+        preferences=preferences,
-        data_pin=config.led.data_pin,
-        num_leds=state.preferences.num_leds,
     )
 
     mqtt_controller: Optional[MqttController] = None
     if config.mqtt.enabled:
         mqtt_controller = MqttController(
-            state=state,
+            loop=loop,
-            host=config.mqtt.host,
+            event_bus=event_bus,
-            port=config.mqtt.port,
+            config=config.mqtt,
-            username=config.mqtt.username,
+            app_name=config.app.name,
-            password=config.mqtt.password
+            mac_address=state.mac_address,
+            preferences=preferences,
         )
+        setattr(state, "mqtt_controller", mqtt_controller)
         mqtt_controller.start()
 
-    mic_mute_handler = MicMuteHandler(state, mqtt_controller)
+    mic_mute_handler = MicMuteHandler(
-    
+        event_bus=event_bus,
-    # --- Start Audio Processing Thread ---
+        state=state,
-    process_audio_thread = threading.Thread(
+        mqtt_controller=mqtt_controller,
-        target=process_audio,
-        args=(state, mic, config.audio.input_block_size),
-        daemon=True
     )
-    process_audio_thread.start()
 
-    # --- Start ESPHome Server ---
+async def _run_server(state: ServerState, config: Config):
-    server = await loop.create_server(
+    """Starts the ESPHome server and ZeroConf discovery."""
+    _LOGGER.debug("Starting ESPHome server")
+    server = await state.loop.create_server(
         lambda: VoiceSatelliteProtocol(state), 
         host=config.esphome.host, 
         port=config.esphome.port
@@ -290,38 +386,29 @@ async def main() -> None:
     discovery = HomeAssistantZeroconf(port=config.esphome.port, name=config.app.name)
     await discovery.register_server()
 
-    # --- Run Forever ---
-    try:
     async with server:
         _LOGGER.info("Server started (host=%s, port=%s)", config.esphome.host, config.esphome.port)
         await server.serve_forever()
-    except KeyboardInterrupt: 
-        pass
-    finally:
-        if mqtt_controller: 
-            mqtt_controller.stop()
 
-        state.mic_muted = True 
-        process_audio_thread.join()
 
-    _LOGGER.debug("Server stopped")
+def _process_audio(state: ServerState, mic, block_size: int):
+    """Main audio processing loop. Runs in a separate thread."""
     
-def process_audio(state: ServerState, mic, block_size: int):
+    micro_features: MicroWakeWordFeatures = MicroWakeWordFeatures()
-    """Process audio chunks from the microphone in a separate thread."""
-    wake_words: List[Union[MicroWakeWord, OpenWakeWord]] = []
-    micro_features: Optional[MicroWakeWordFeatures] = None
-    micro_inputs: List[np.ndarray] = []
     oww_features: Optional[OpenWakeWordFeatures] = None
+    
+    wake_words: List[Union[MicroWakeWord, OpenWakeWord]] = []
+    micro_inputs: List[np.ndarray] = []
     oww_inputs: List[np.ndarray] = []
     has_oww = False
     last_active: Optional[float] = None
+    
     try:
         _LOGGER.debug("Opening audio input device: %s", mic.name)
         with mic.recorder(samplerate=16000, channels=1, blocksize=block_size) as mic_in:
             while True:
                 if state.mic_muted:
-                    time.sleep(0.1)
+                    break
-                    continue
                 
                 audio_chunk_array = mic_in.record(block_size).reshape(-1)
                 audio_chunk = (
@@ -331,23 +418,24 @@ def process_audio(state: ServerState, mic, block_size: int):
                 )
 
                 if state.satellite is None:
+                    time.sleep(0.01)
                     continue
                 
-                if (not wake_words) or (state.wake_words_changed and state.wake_words):
+                # --- THIS IS THE FIX ---
+                if (not wake_words) or (state.wake_words_changed):
+                # --- END FIX ---
                     state.wake_words_changed = False
                     wake_words = [
                         ww for ww in state.wake_words.values() 
                         if ww.id in state.active_wake_words
                     ]
                     has_oww = any(isinstance(ww, OpenWakeWord) for ww in wake_words)
-                    if micro_features is None:
-                        micro_features = MicroWakeWordFeatures()
-                    if has_oww and (oww_features is None):
-                        oww_features = OpenWakeWordFeatures.from_builtin()
-                try:
-                    state.satellite.handle_audio(audio_chunk)
                     
-                    assert micro_features is not None
+                    if has_oww and (oww_features is None):
+                        _LOGGER.debug("Initializing OpenWakeWord features...")
+                        oww_features = OpenWakeWordFeatures.from_builtin()
+                
+                try:
                     micro_inputs.clear(); micro_inputs.extend(micro_features.process_streaming(audio_chunk))
                     
                     if has_oww:
@@ -364,7 +452,7 @@ def process_audio(state: ServerState, mic, block_size: int):
                         if activated:
                             now = time.monotonic()
                             if (last_active is None) or ((now - last_active) > state.refractory_seconds):
-                                state.satellite.wakeup(wake_word)
+                                state.loop.call_soon_threadsafe(state.satellite.wakeup, wake_word)
                                 last_active = now
                     
                     stopped = False
@@ -373,13 +461,20 @@ def process_audio(state: ServerState, mic, block_size: int):
                             stopped = True
 
                     if stopped and (state.stop_word.id in state.active_wake_words):
-                        state.satellite.stop()
+                        state.loop.call_soon_threadsafe(state.satellite.stop)
-                except Exception: _LOGGER.exception("Unexpected error handling audio")
+                        
+                    state.loop.call_soon_threadsafe(state.satellite.handle_audio, audio_chunk)
+
+                except Exception: 
+                    _LOGGER.exception("Unexpected error handling audio")
+    
     except Exception as e:
-        _LOGGER.critical("Error in audio processing thread: %s", e)
+        _LOGGER.critical("A soundcard error occurred: %s", e)
         state.loop.call_soon_threadsafe(state.loop.stop)
 
 
+# -----------------------------------------------------------------------------
+
 if __name__ == "__main__":
     print("--- __main__ block executing ---")
     asyncio.run(main())