app/utils/config.py

from pydantic_settings import BaseSettings
from pydantic import Field, field_validator, ConfigDict
from typing import List, Any, Optional, Dict
import os
import logging
import sys
from pathlib import Path
from dotenv import load_dotenv

# Load environment variables from .env file
load_dotenv()

def setup_logging(log_level: str = "INFO", log_file: str = None) -> None:
    """Configure logging for the application."""
    log_format = "%(asctime)s - %(name)s - %(levelname)s - %(message)s"
    
    # Configure root logger
    logging.basicConfig(
        level=log_level,
        format=log_format,
        handlers=[logging.StreamHandler(sys.stdout)]
    )
    
    # Add file handler if log file is specified
    if log_file:
        log_file = Path(log_file)
        log_file.parent.mkdir(parents=True, exist_ok=True)
        file_handler = logging.FileHandler(log_file)
        file_handler.setFormatter(logging.Formatter(log_format))
        logging.getLogger().addHandler(file_handler)
    
    # Set log level for specific loggers
    logging.getLogger("httpx").setLevel(logging.WARNING)
    logging.getLogger("openai").setLevel(logging.WARNING)
    logging.getLogger("urllib3").setLevel(logging.WARNING)


class Settings(BaseSettings):
    model_config = ConfigDict(
        env_file=".env",
        env_file_encoding="utf-8",
        extra="allow",
        env_nested_delimiter="__"
    )
    
    # App configuration
    APP_NAME: str = Field(default="Carehelp")
    ENVIRONMENT: str = Field(default="development")
    
    # Logging configuration
    LOG_LEVEL: str = Field(default="INFO")
    LOG_FILE: Optional[str] = Field(default=None)
    
    def configure_logging(self) -> None:
        """Configure logging based on settings."""
        setup_logging(
            log_level=self.LOG_LEVEL,
            log_file=self.LOG_FILE
        )
    PORT: int = Field(default=8000)

    # CORS configuration
    CORS_ALLOW_ORIGINS: List[str] = Field(default_factory=lambda: ["*"])

    # API Keys
    GOOGLE_MAPS_API_KEY: str | None = None
    SERPAPI_API_KEY: str | None = None
    
    # Supabase configuration
    SUPABASE_URL: str | None = None
    SUPABASE_ANON_KEY: str | None = None
    SUPABASE_SERVICE_ROLE_KEY: str | None = None
    SUPABASE_DB_PASSWORD: str | None = None
    
    # OpenAI configuration
    OPENAI_API_KEY: str | None = None
    OPENAI_MODEL: str = Field(default="gpt-4")
    OPENAI_WHISPER_MODEL: str = Field(default="whisper-1")
    OPENAI_WHISPER_FALLBACK_MODEL: str | None = Field(default="gpt-4-transcribe")
    OPENAI_WHISPER_MAX_CHUNK_SECS: int = Field(default=120)

    # JWT configuration
    JWT_SECRET: str = Field(default="change_this_secret")
    JWT_ALGORITHM: str = Field(default="HS256")
    ACCESS_TOKEN_EXPIRE_MINUTES: int = Field(default=60 * 24 * 30)

    # Hugging Face configuration
    HF_API_TOKEN: str | None = None
    HF_TRANSLATION_MODEL: str = Field(default="facebook/nllb-200-distilled-600M")

    # Provider selection and models
    # Options: 'hf' (Hugging Face Inference), 'ollama' (local), 'lmstudio' (local OpenAI-compatible)
    AI_PROVIDER: str = Field(default="hf")

    # Text generation (chat) models
    HF_TEXT_MODEL: str = Field(default="meta-llama/Meta-Llama-3-8B-Instruct")
    OLLAMA_MODEL: str = Field(default="llama3.1:8b")
    OLLAMA_BASE_URL: str = Field(default="http://localhost:11434")
    LMSTUDIO_MODEL: str | None = Field(default=None)
    LMSTUDIO_BASE_URL: str = Field(default="http://localhost:1234/v1")

    # Vision caption model (used to assist image analysis with HF)
    HF_VISION_CAPTION_MODEL: str = Field(default="Salesforce/blip-image-captioning-large")
    # Optional local vision model for Ollama (e.g., 'llava:latest')
    OLLAMA_VISION_MODEL: str = Field(default="llava:latest")

    # Automatic Speech Recognition (ASR)
    HF_ASR_MODEL: str = Field(default="distil-whisper/distil-large-v3")

    # Data paths
    PATIENT_DATA_PATH: str = Field(default="../patient_records.json")

    # Cameroon data configuration
    CAMEROON_DATA_CSV: str = Field(default="../../clinical_summaries.csv")
    CAMEROON_EMBEDDINGS_MODEL: str = Field(
        default="sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2"
    )
    CAMEROON_CACHE_DIR: str = Field(default=".cache_cameroon")
    CAMEROON_MAX_ROWS: int | None = None  # limit for dev/testing

    @field_validator("CORS_ALLOW_ORIGINS", mode="before")
    @classmethod
    def parse_cors_origins(cls, v: Any) -> Any:
        # Accept JSON array, comma-separated string, or single "*"
        if isinstance(v, list):
            return v
        if isinstance(v, str):
            s = v.strip()
            if s == "":
                return ["*"]
            if s == "*":
                return ["*"]
            # comma separated
            return [x.strip() for x in s.split(",") if x.strip()]
        return v


settings = Settings()