feat(HEL-382): 搭建 datahub 底座和盘后正式数据链路

新增独立 xiaobai-datahub 服务(SQLite WAL、Tushare 盘后发布、/v1 契约和管理后台),不改现站页面与数据链路。

Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: multica-agent <github@multica.ai>
This commit is contained in:
总工
2026-09-02 12:05:26 +08:00
co-authored by Cursor multica-agent
parent c2ebc0ab91
commit 3498dd7a4b
52 changed files with 4259 additions and 0 deletions
+72
View File
@@ -0,0 +1,72 @@
from __future__ import annotations
import json
import os
from dataclasses import dataclass, field
from pathlib import Path
from typing import Any
ROOT = Path(__file__).resolve().parents[1]
DEFAULT_DB_PATH = Path(os.environ.get("DATAHUB_DB_PATH") or (ROOT / "data" / "datahub.db"))
DEFAULT_BACKUP_DIR = Path(os.environ.get("DATAHUB_BACKUP_DIR") or (ROOT / "data" / "backups"))
DEFAULT_CONFIG_PATH = ROOT / "config" / "hub-quality.config.json"
def _load_quality(path: Path) -> dict[str, Any]:
if not path.is_file():
return {}
return json.loads(path.read_text(encoding="utf-8"))
@dataclass
class Settings:
host: str = "127.0.0.1"
port: int = 8766
encryption_key: str = ""
api_token: str = ""
admin_password: str = ""
tushare_token: str = ""
db_path: Path = DEFAULT_DB_PATH
backup_dir: Path = DEFAULT_BACKUP_DIR
quality: dict[str, Any] = field(default_factory=dict)
log_level: str = "INFO"
scheduler_enabled: bool = True
@property
def tushare_rate_per_minute(self) -> int:
return int(self.quality.get("tushare_rate_per_minute") or 300)
@property
def max_publish_attempts(self) -> int:
return int(self.quality.get("max_publish_attempts") or 5)
@property
def list_limit_default(self) -> int:
return int(self.quality.get("list_limit_default") or 5000)
@property
def list_limit_max(self) -> int:
return int(self.quality.get("list_limit_max") or 5000)
def load_settings(
env: dict[str, str] | None = None,
config_path: Path | None = None,
) -> Settings:
environ = env if env is not None else dict(os.environ)
quality_path = config_path or DEFAULT_CONFIG_PATH
db_path = Path(environ.get("DATAHUB_DB_PATH") or DEFAULT_DB_PATH)
backup_dir = Path(environ.get("DATAHUB_BACKUP_DIR") or DEFAULT_BACKUP_DIR)
return Settings(
host=environ.get("DATAHUB_HOST") or "127.0.0.1",
port=int(environ.get("DATAHUB_PORT") or 8766),
encryption_key=str(environ.get("DATAHUB_ENCRYPTION_KEY") or "").strip(),
api_token=str(environ.get("DATAHUB_TOKEN") or "").strip(),
admin_password=str(environ.get("DATAHUB_ADMIN_PASSWORD") or "").strip(),
tushare_token=str(environ.get("TUSHARE_TOKEN") or "").strip(),
db_path=db_path,
backup_dir=backup_dir,
quality=_load_quality(quality_path),
log_level=environ.get("DATAHUB_LOG_LEVEL") or "INFO",
scheduler_enabled=str(environ.get("DATAHUB_SCHEDULER") or "1") not in {"0", "false", "False"},
)