backend/utils/env.py

import os
from dataclasses import dataclass
from typing import List

from dotenv import load_dotenv


load_dotenv(override=False)


def _clean_env_value(value: str) -> str:
    # Strip whitespace and optional surrounding quotes.
    v = value.strip()
    if len(v) >= 2 and ((v[0] == v[-1] == '"') or (v[0] == v[-1] == "'")):
        v = v[1:-1].strip()
    return v


def _get_env(name: str, default: str | None = None) -> str:
    value = os.getenv(name, default)
    if value is None or value == "":
        raise RuntimeError(f"Missing required environment variable: {name}")
    return _clean_env_value(value)


def _get_env_optional(name: str, default: str | None = None) -> str | None:
    value = os.getenv(name, default)
    if value is None or value == "":
        return None
    return _clean_env_value(value)


def _get_bool_env(name: str, default: bool = False) -> bool:
    raw = os.getenv(name)
    if raw is None:
        return default
    return raw.strip().lower() in {"1", "true", "yes", "y", "on"}


def _get_bool_env_optional(name: str) -> bool | None:
    raw = os.getenv(name)
    if raw is None or raw.strip() == "":
        return None
    return raw.strip().lower() in {"1", "true", "yes", "y", "on"}


def _parse_cors_origins(raw: str | None) -> List[str]:
    if not raw:
        return ["*"]
    parts = [p.strip() for p in raw.split(",")]
    return [p for p in parts if p]


@dataclass(frozen=True)
class Settings:
    groq_api_key: str
    groq_model: str
    voice_transcription_model: str
    google_api_key: str | None
    openrouter_api_key: str | None
    default_provider: str
    default_model: str
    mongodb_uri: str
    mongodb_tls: bool | None
    mongodb_db: str
    mongo_sessions_collection: str
    mongo_messages_collection: str
    mongo_users_collection: str
    max_history_messages: int
    max_attachment_chars_per_file: int
    max_total_attachment_chars: int
    temperature: float
    debug: bool
    cors_origins: List[str]
    jwt_secret: str
    jwt_algorithm: str
    jwt_exp_minutes: int
    tavily_api_key: str | None
    tavily_max_results: int
    google_client_id: str | None
    google_client_secret: str | None
    google_redirect_uri: str | None
    google_scope: str
    enable_vector_memory: bool
    enable_local_vision: bool


def get_settings() -> Settings:
    debug = _get_bool_env("DEBUG", False)
    jwt_secret = _get_env_optional("JWT_SECRET", None)
    if not jwt_secret and not debug:
        raise RuntimeError(
            "Missing JWT_SECRET. Set it in your .env (or enable DEBUG=true for a dev-only insecure secret)."
        )
    return Settings(
        groq_api_key=_get_env_optional("GROQ_API_KEY", None) or "",
        groq_model=os.getenv("GROQ_MODEL", "llama-3.3-70b-versatile"),
        voice_transcription_model=os.getenv("VOICE_TRANSCRIPTION_MODEL", "whisper-large-v3-turbo").strip(),
        google_api_key=_get_env_optional("GOOGLE_API_KEY", None),
        openrouter_api_key=_get_env_optional("OPENROUTER_API_KEY", None),
        default_provider=os.getenv("DEFAULT_LLM_PROVIDER", "groq").strip().lower(),
        default_model=os.getenv("DEFAULT_LLM_MODEL", "llama-3.3-70b-versatile").strip(),
        mongodb_uri=_get_env("MONGODB_URI", "mongodb://localhost:27017"),
        mongodb_tls=_get_bool_env_optional("MONGODB_TLS"),
        mongodb_db=os.getenv("MONGODB_DB", "own_gpt"),
        mongo_sessions_collection=os.getenv("MONGO_SESSIONS_COLLECTION", "sessions"),
        mongo_messages_collection=os.getenv("MONGO_MESSAGES_COLLECTION", "messages"),
        mongo_users_collection=os.getenv("MONGO_USERS_COLLECTION", "users"),
        max_history_messages=int(os.getenv("MAX_HISTORY_MESSAGES", "20")),
        max_attachment_chars_per_file=int(os.getenv("MAX_ATTACHMENT_CHARS_PER_FILE", "8000")),
        max_total_attachment_chars=int(os.getenv("MAX_TOTAL_ATTACHMENT_CHARS", "16000")),
        temperature=float(os.getenv("GROQ_TEMPERATURE", "0.2")),
        debug=debug,
        cors_origins=_parse_cors_origins(_get_env_optional("CORS_ORIGINS", "*")),
        jwt_secret=jwt_secret or "dev-only-insecure-jwt-secret",
        jwt_algorithm=os.getenv("JWT_ALGORITHM", "HS256"),
        jwt_exp_minutes=int(os.getenv("JWT_EXP_MINUTES", "60")),
        tavily_api_key=_get_env_optional("TAVILY_API_KEY", None),
        tavily_max_results=int(os.getenv("TAVILY_MAX_RESULTS", "5")),
        google_client_id=_get_env_optional("GOOGLE_CLIENT_ID", None),
        google_client_secret=_get_env_optional("GOOGLE_CLIENT_SECRET", None),
        google_redirect_uri=_get_env_optional("GOOGLE_REDIRECT_URI", None),
        google_scope=os.getenv("GOOGLE_SCOPE", "openid email profile"),
        enable_vector_memory=_get_bool_env("ENABLE_VECTOR_MEMORY", True),
        enable_local_vision=_get_bool_env("ENABLE_LOCAL_VISION", False),
    )