dashboard-nanobot/backend/core/settings.py

import os
import re
from pathlib import Path
from typing import Final
from urllib.parse import urlsplit, urlunsplit

from dotenv import dotenv_values, load_dotenv

BACKEND_ROOT: Final[Path] = Path(__file__).resolve().parents[1]
PROJECT_ROOT: Final[Path] = BACKEND_ROOT.parent

# Load env files used by this project.
# Priority (high -> low):
# 1) process environment
# 2) project/.env.prod
# 3) backend/.env
#
# We keep process-provided env untouched, while allowing .env.prod to override backend/.env.
_process_env_keys = set(os.environ.keys())
_backend_env_values = dotenv_values(BACKEND_ROOT / ".env")
_prod_env_values = dotenv_values(PROJECT_ROOT / ".env.prod")

load_dotenv(BACKEND_ROOT / ".env", override=False)
for _k, _v in _prod_env_values.items():
    if _v is None:
        continue
    if _k in _process_env_keys:
        continue
    os.environ[_k] = str(_v)


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


def _env_int(name: str, default: int, min_value: int, max_value: int) -> int:
    raw = os.getenv(name)
    if raw is None:
        return default
    try:
        value = int(str(raw).strip())
    except Exception:
        value = default
    return max(min_value, min(max_value, value))


def _normalize_extension(raw: str) -> str:
    text = str(raw or "").strip().lower()
    if not text:
        return ""
    if text.startswith("*."):
        text = text[1:]
    if not text.startswith("."):
        text = f".{text}"
    if not re.fullmatch(r"\.[a-z0-9][a-z0-9._+-]{0,31}", text):
        return ""
    return text


def _env_extensions(name: str, default: tuple[str, ...]) -> tuple[str, ...]:
    raw = os.getenv(name)
    if raw is None:
        source = list(default)
    else:
        source = re.split(r"[,;\s]+", str(raw))
    rows: list[str] = []
    for item in source:
        ext = _normalize_extension(item)
        if ext and ext not in rows:
            rows.append(ext)
    if raw is None:
        return tuple(rows or list(default))
    return tuple(rows)


def _normalize_dir_path(path_value: str) -> str:
    raw = str(path_value or "").strip()
    if not raw:
        return raw
    raw = os.path.expandvars(os.path.expanduser(raw))
    p = Path(raw)
    if p.is_absolute():
        return str(p)
    return str((BACKEND_ROOT / p).resolve())


DATA_ROOT: Final[str] = _normalize_dir_path(os.getenv("DATA_ROOT", str(PROJECT_ROOT / "data")))
BOTS_WORKSPACE_ROOT: Final[str] = _normalize_dir_path(
    os.getenv("BOTS_WORKSPACE_ROOT", str(PROJECT_ROOT / "workspace" / "bots"))
)


def _normalize_database_url(url: str) -> str:
    raw = str(url or "").strip()
    prefix = "sqlite:///"
    if not raw.startswith(prefix):
        return raw
    path_part = raw[len(prefix) :]
    if not path_part or path_part.startswith("/"):
        return raw
    abs_path = (BACKEND_ROOT / path_part).resolve()
    return f"{prefix}{abs_path.as_posix()}"


def _database_engine(url: str) -> str:
    raw = str(url or "").strip().lower()
    if raw.startswith("sqlite"):
        return "sqlite"
    if raw.startswith("postgresql"):
        return "postgresql"
    if raw.startswith("mysql"):
        return "mysql"
    if "+" in raw:
        return raw.split("+", 1)[0]
    if "://" in raw:
        return raw.split("://", 1)[0]
    return "unknown"


def _mask_database_url(url: str) -> str:
    raw = str(url or "").strip()
    if not raw or raw.startswith("sqlite"):
        return raw
    try:
        parsed = urlsplit(raw)
        if parsed.password is None:
            return raw
        host = parsed.hostname or ""
        if parsed.port:
            host = f"{host}:{parsed.port}"
        auth = parsed.username or ""
        if auth:
            auth = f"{auth}:***@{host}"
        else:
            auth = host
        netloc = auth
        return urlunsplit((parsed.scheme, netloc, parsed.path, parsed.query, parsed.fragment))
    except Exception:
        return raw


_db_env = str(os.getenv("DATABASE_URL") or "").strip()
if not _db_env:
    raise RuntimeError("DATABASE_URL is not set in environment. PostgreSQL is required.")

DATABASE_URL: Final[str] = _normalize_database_url(_db_env)
DATABASE_ENGINE: Final[str] = _database_engine(DATABASE_URL)
DATABASE_URL_DISPLAY: Final[str] = _mask_database_url(DATABASE_URL)
DATABASE_ECHO: Final[bool] = _env_bool("DATABASE_ECHO", True)
DATABASE_POOL_SIZE: Final[int] = _env_int("DATABASE_POOL_SIZE", 20, 1, 200)
DATABASE_MAX_OVERFLOW: Final[int] = _env_int("DATABASE_MAX_OVERFLOW", 40, 0, 200)
DATABASE_POOL_TIMEOUT: Final[int] = _env_int("DATABASE_POOL_TIMEOUT", 30, 1, 300)
DATABASE_POOL_RECYCLE: Final[int] = _env_int("DATABASE_POOL_RECYCLE", 1800, 30, 86400)
DEFAULT_UPLOAD_MAX_MB: Final[int] = 100
DEFAULT_PAGE_SIZE: Final[int] = 10
DEFAULT_CHAT_PULL_PAGE_SIZE: Final[int] = 60
DEFAULT_COMMAND_AUTO_UNLOCK_SECONDS: Final[int] = _env_int("COMMAND_AUTO_UNLOCK_SECONDS", 10, 1, 600)
DEFAULT_BOT_SYSTEM_TIMEZONE: Final[str] = str(
    os.getenv("DEFAULT_BOT_SYSTEM_TIMEZONE") or os.getenv("TZ") or "Asia/Shanghai"
).strip() or "Asia/Shanghai"
DEFAULT_WORKSPACE_DOWNLOAD_EXTENSIONS: Final[tuple[str, ...]] = (
    ".pdf",
    ".doc",
    ".docx",
    ".xls",
    ".xlsx",
    ".xlsm",
    ".ppt",
    ".pptx",
    ".odt",
    ".ods",
    ".odp",
    ".wps",
)
STT_ENABLED_DEFAULT: Final[bool] = True
STT_MODEL: Final[str] = str(os.getenv("STT_MODEL") or "ggml-small-q8_0.bin").strip()
_DEFAULT_STT_MODEL_DIR: Final[Path] = (Path(DATA_ROOT) / "model").resolve()
_configured_stt_model_dir = _normalize_dir_path(os.getenv("STT_MODEL_DIR", str(_DEFAULT_STT_MODEL_DIR)))
if _configured_stt_model_dir and not Path(_configured_stt_model_dir).exists() and _DEFAULT_STT_MODEL_DIR.exists():
    STT_MODEL_DIR: Final[str] = str(_DEFAULT_STT_MODEL_DIR)
else:
    STT_MODEL_DIR: Final[str] = _configured_stt_model_dir
STT_DEVICE: Final[str] = str(os.getenv("STT_DEVICE") or "cpu").strip().lower() or "cpu"
DEFAULT_STT_MAX_AUDIO_SECONDS: Final[int] = 20
DEFAULT_STT_DEFAULT_LANGUAGE: Final[str] = "zh"
DEFAULT_STT_FORCE_SIMPLIFIED: Final[bool] = True
DEFAULT_STT_AUDIO_PREPROCESS: Final[bool] = True
DEFAULT_STT_AUDIO_FILTER: Final[str] = "highpass=f=120,lowpass=f=7600,afftdn=nf=-20"
DEFAULT_STT_INITIAL_PROMPT: Final[str] = (
    "以下内容可能包含简体中文和英文术语。请优先输出简体中文，英文单词、缩写、品牌名和数字保持原文，不要翻译。"
)

REDIS_ENABLED: Final[bool] = _env_bool("REDIS_ENABLED", False)
REDIS_URL: Final[str] = str(os.getenv("REDIS_URL") or "").strip()
REDIS_PREFIX: Final[str] = str(os.getenv("REDIS_PREFIX") or "dashboard_nanobot").strip() or "dashboard_nanobot"
REDIS_DEFAULT_TTL: Final[int] = _env_int("REDIS_DEFAULT_TTL", 60, 1, 86400)
PANEL_ACCESS_PASSWORD: Final[str] = str(os.getenv("PANEL_ACCESS_PASSWORD") or "").strip()

APP_HOST: Final[str] = str(os.getenv("APP_HOST") or "0.0.0.0").strip()
APP_PORT: Final[int] = _env_int("APP_PORT", 8000, 1, 65535)
APP_RELOAD: Final[bool] = _env_bool("APP_RELOAD", False)

TEMPLATE_ROOT: Final[Path] = (BACKEND_ROOT / "templates").resolve()
AGENT_MD_TEMPLATES_FILE: Final[Path] = TEMPLATE_ROOT / "agent_md_templates.json"
TOPIC_PRESETS_TEMPLATES_FILE: Final[Path] = TEMPLATE_ROOT / "topic_presets.json"