config/settings.py

"""Application-wide configuration settings."""

import os
from dataclasses import dataclass, field
from typing import Optional
from pathlib import Path
from dotenv import load_dotenv

load_dotenv()


@dataclass
class Settings:
    """Application-wide configuration settings."""

    # ============================================
    # LLM Provider Settings
    # ============================================
    llm_provider: str = os.getenv("LLM_PROVIDER", "auto")
    
    # Hugging Face settings
    hf_token: str = os.getenv("HF_TOKEN", "")
    hf_chat_model: str = os.getenv("HF_CHAT_MODEL", "Qwen/Qwen2.5-7B-Instruct")
    hf_temperature: float = float(os.getenv("HF_TEMPERATURE", "0.001"))
    hf_max_new_tokens: int = int(os.getenv("HF_MAX_NEW_TOKENS", "512"))

    # Model settings
    model_name: str = os.getenv("MODEL_NAME", "Qwen/Qwen2.5-7B-Instruct")

    # ============================================
    # Audio Provider Settings
    # ============================================
    audio_provider: str = os.getenv("AUDIO_PROVIDER", "auto")
    tts_model: str = os.getenv("TTS_MODEL", "hexgrad/Kokoro-82M")
    stt_model: str = os.getenv("STT_MODEL", "openai/whisper-large-v3")

    # ============================================
    # VAD (Voice Activity Detection) Settings
    # ============================================
    vad_enabled: bool = os.getenv("VAD_ENABLED", "true").lower() == "true"
    vad_sample_rate: int = int(os.getenv("VAD_SAMPLE_RATE", "16000"))
    vad_frame_duration_ms: int = int(os.getenv("VAD_FRAME_DURATION_MS", "30"))
    vad_aggressiveness: int = int(os.getenv("VAD_AGGRESSIVENESS", "2"))
    vad_speech_threshold: float = float(os.getenv("VAD_SPEECH_THRESHOLD", "0.5"))
    vad_silence_threshold: float = float(os.getenv("VAD_SILENCE_THRESHOLD", "0.3"))
    vad_min_speech_ms: int = int(os.getenv("VAD_MIN_SPEECH_MS", "300"))
    vad_max_speech_s: float = float(os.getenv("VAD_MAX_SPEECH_S", "30.0"))
    vad_post_speech_silence_ms: int = int(os.getenv("VAD_POST_SPEECH_SILENCE_MS", "800"))

    # ============================================
    # Screen/Vision Settings
    # ============================================
    screen_capture_interval: float = float(os.getenv("SCREEN_CAPTURE_INTERVAL", "1.0"))
    screen_compression_quality: int = int(os.getenv("SCREEN_COMPRESSION_QUALITY", "50"))
    max_width: int = int(os.getenv("SCREEN_MAX_WIDTH", "3440"))
    max_height: int = int(os.getenv("SCREEN_MAX_HEIGHT", "1440"))
    
    # Vision model (Nebius)
    NEBIUS_MODEL: str = os.getenv("NEBIUS_MODEL", "google/gemma-3-27b-it-fast")
    NEBIUS_API_KEY: str = os.getenv("NEBIUS_API_KEY", "")
    NEBIUS_BASE_URL: str = os.getenv("NEBIUS_BASE_URL", "https://api.studio.nebius.com/v1/")
    
    # Auto-enable vision when screen context is needed
    vision_auto_enabled: bool = os.getenv("VISION_AUTO_ENABLED", "true").lower() == "true"
    vision_fps: float = float(os.getenv("VISION_FPS", "0.05"))  # Frames per second

    # ============================================
    # MCP Server Settings
    # ============================================
    mcp_server_url: str = os.getenv("MCP_SERVER_URL", "http://localhost:8000")
    mcp_auto_start: bool = os.getenv("MCP_AUTO_START", "true").lower() == "true"

    # ============================================
    # CRM Data Settings
    # ============================================
    crm_data_dir: str = os.getenv("CRM_DATA_DIR", "./data")

    # ============================================
    # Hyper-V Settings (Legacy)
    # ============================================
    hyperv_enabled: bool = os.getenv("HYPERV_ENABLED", "false").lower() == "true"
    hyperv_host: str = os.getenv("HYPERV_HOST", "localhost")
    hyperv_username: Optional[str] = os.getenv("HYPERV_USERNAME")
    hyperv_password: Optional[str] = os.getenv("HYPERV_PASSWORD")

    # ============================================
    # Application Settings
    # ============================================
    max_conversation_history: int = int(os.getenv("MAX_CONVERSATION_HISTORY", "50"))
    temp_dir: str = os.getenv("TEMP_DIR", "./temp")
    log_level: str = os.getenv("LOG_LEVEL", "INFO")
    
    # Feature flags
    enable_screen_sharing_button: bool = os.getenv("ENABLE_SCREEN_SHARING_BUTTON", "true").lower() == "true"
    enable_voice_input: bool = os.getenv("ENABLE_VOICE_INPUT", "true").lower() == "true"

    def __post_init__(self):
        """Initialize directories and validate settings."""
        # Ensure necessary directories exist
        Path(self.temp_dir).mkdir(exist_ok=True, parents=True)
        Path("./config").mkdir(exist_ok=True, parents=True)
        Path("./logs").mkdir(exist_ok=True, parents=True)
        Path(self.crm_data_dir).mkdir(exist_ok=True, parents=True)

        # 🔁 Refresh dynamic, env-backed values so they pick up changes done at runtime
        self.hf_token = os.getenv("HF_TOKEN", self.hf_token)
        self.NEBIUS_API_KEY = os.getenv("NEBIUS_API_KEY", self.NEBIUS_API_KEY)


    def is_hf_token_valid(self) -> bool:
        """Check if HuggingFace token is set and looks like a real HF token."""
        token = os.getenv("HF_TOKEN", "")  # always read the latest env
        return bool(token and token.startswith("hf_") and len(token) > 20)

    @property
    def effective_llm_provider(self) -> str:
        if self.llm_provider == "auto":
            return "huggingface" if self.is_hf_token_valid() else "openai"
        return self.llm_provider

    @property
    def effective_audio_provider(self) -> str:
        if self.audio_provider == "auto":
            return "huggingface" if self.is_hf_token_valid() else "openai"
        return self.audio_provider

    @property
    def llm_endpoint(self) -> str:
        if self.effective_llm_provider == "huggingface":
            return f"https://api-inference.huggingface.co/models/{self.hf_chat_model}"
        return getattr(self, 'openai_endpoint', '')

    @property
    def llm_api_key(self) -> str:
        if self.effective_llm_provider == "huggingface":
            return os.getenv("HF_TOKEN", "")  # latest HF token
        return getattr(self, "openai_api_key", "")

    @property
    def effective_model_name(self) -> str:
        return self.hf_chat_model if self.effective_llm_provider == "huggingface" else self.model_name

    def get_vad_config(self) -> dict:
        """Get VAD configuration as a dictionary."""
        return {
            "sample_rate": self.vad_sample_rate,
            "frame_duration_ms": self.vad_frame_duration_ms,
            "aggressiveness": self.vad_aggressiveness,
            "speech_threshold": self.vad_speech_threshold,
            "silence_threshold": self.vad_silence_threshold,
            "min_speech_duration_ms": self.vad_min_speech_ms,
            "max_speech_duration_s": self.vad_max_speech_s,
            "post_speech_silence_ms": self.vad_post_speech_silence_ms,
        }