148 lines
5.4 KiB
Python
148 lines
5.4 KiB
Python
import json
|
|
import sqlite3
|
|
from contextlib import contextmanager
|
|
from datetime import datetime
|
|
from typing import Any, Iterator
|
|
|
|
from config import Config
|
|
|
|
|
|
def now_iso() -> str:
|
|
return datetime.now().astimezone().isoformat(timespec="seconds")
|
|
|
|
|
|
@contextmanager
|
|
def connection() -> Iterator[sqlite3.Connection]:
|
|
conn = sqlite3.connect(Config.DATABASE_PATH, timeout=30)
|
|
conn.row_factory = sqlite3.Row
|
|
conn.execute("PRAGMA foreign_keys = ON")
|
|
try:
|
|
yield conn
|
|
conn.commit()
|
|
finally:
|
|
conn.close()
|
|
|
|
|
|
def init_database() -> None:
|
|
"""仅在服务启动时创建表,接口调用过程中不会重复执行建表。"""
|
|
|
|
Config.DATABASE_PATH.parent.mkdir(parents=True, exist_ok=True)
|
|
Config.FILE_STORAGE_PATH.mkdir(parents=True, exist_ok=True)
|
|
with connection() as conn:
|
|
conn.executescript(
|
|
"""
|
|
CREATE TABLE IF NOT EXISTS research_tasks (
|
|
id INTEGER PRIMARY KEY AUTOINCREMENT,
|
|
public_task_id TEXT NOT NULL UNIQUE,
|
|
name TEXT NOT NULL,
|
|
platform TEXT NOT NULL,
|
|
mode TEXT NOT NULL,
|
|
status TEXT NOT NULL,
|
|
execution_stage TEXT NOT NULL DEFAULT 'queued',
|
|
progress_percent INTEGER NOT NULL DEFAULT 0 CHECK(progress_percent BETWEEN 0 AND 100),
|
|
created_by_id TEXT NOT NULL,
|
|
created_by_name TEXT NOT NULL,
|
|
retry_of_task_id INTEGER,
|
|
crawler_task_id TEXT,
|
|
error_code TEXT,
|
|
error_message TEXT,
|
|
result_json TEXT,
|
|
cancel_requested_at TEXT,
|
|
started_at TEXT,
|
|
completed_at TEXT,
|
|
created_at TEXT NOT NULL,
|
|
updated_at TEXT NOT NULL,
|
|
version INTEGER NOT NULL DEFAULT 0,
|
|
FOREIGN KEY(retry_of_task_id) REFERENCES research_tasks(id)
|
|
);
|
|
CREATE INDEX IF NOT EXISTS idx_tasks_status_created
|
|
ON research_tasks(status, created_at DESC);
|
|
CREATE INDEX IF NOT EXISTS idx_tasks_creator_created
|
|
ON research_tasks(created_by_id, created_at DESC);
|
|
|
|
CREATE TABLE IF NOT EXISTS task_keywords (
|
|
id INTEGER PRIMARY KEY AUTOINCREMENT,
|
|
task_id INTEGER NOT NULL,
|
|
keyword_order INTEGER NOT NULL,
|
|
keyword TEXT NOT NULL,
|
|
normalized_keyword TEXT NOT NULL,
|
|
created_at TEXT NOT NULL,
|
|
UNIQUE(task_id, keyword_order),
|
|
UNIQUE(task_id, normalized_keyword),
|
|
FOREIGN KEY(task_id) REFERENCES research_tasks(id) ON DELETE CASCADE
|
|
);
|
|
|
|
CREATE TABLE IF NOT EXISTS task_events (
|
|
id INTEGER PRIMARY KEY AUTOINCREMENT,
|
|
task_id INTEGER NOT NULL,
|
|
event_type TEXT NOT NULL,
|
|
stage TEXT,
|
|
progress_percent INTEGER,
|
|
actor_type TEXT NOT NULL,
|
|
actor_id TEXT,
|
|
payload_json TEXT,
|
|
occurred_at TEXT NOT NULL,
|
|
FOREIGN KEY(task_id) REFERENCES research_tasks(id) ON DELETE CASCADE
|
|
);
|
|
|
|
CREATE TABLE IF NOT EXISTS task_files (
|
|
id INTEGER PRIMARY KEY AUTOINCREMENT,
|
|
public_file_id TEXT NOT NULL UNIQUE,
|
|
task_id INTEGER NOT NULL,
|
|
name TEXT NOT NULL,
|
|
stage TEXT NOT NULL,
|
|
category TEXT NOT NULL,
|
|
format TEXT NOT NULL,
|
|
record_count INTEGER NOT NULL DEFAULT 0,
|
|
size_bytes INTEGER NOT NULL DEFAULT 0,
|
|
storage_key TEXT NOT NULL,
|
|
sha256 TEXT NOT NULL,
|
|
schema_version TEXT NOT NULL DEFAULT '1.0',
|
|
generated_at TEXT NOT NULL,
|
|
available INTEGER NOT NULL DEFAULT 1,
|
|
UNIQUE(task_id, name),
|
|
FOREIGN KEY(task_id) REFERENCES research_tasks(id) ON DELETE CASCADE
|
|
);
|
|
|
|
CREATE TABLE IF NOT EXISTS auth_users (
|
|
id INTEGER PRIMARY KEY AUTOINCREMENT,
|
|
ding_user_id TEXT NOT NULL UNIQUE,
|
|
union_id TEXT,
|
|
open_id TEXT,
|
|
name TEXT NOT NULL,
|
|
avatar TEXT,
|
|
mobile TEXT,
|
|
email TEXT,
|
|
raw_json TEXT,
|
|
last_login_at TEXT NOT NULL,
|
|
created_at TEXT NOT NULL,
|
|
updated_at TEXT NOT NULL
|
|
);
|
|
CREATE INDEX IF NOT EXISTS idx_auth_users_name
|
|
ON auth_users(name);
|
|
|
|
CREATE TABLE IF NOT EXISTS auth_sessions (
|
|
id INTEGER PRIMARY KEY AUTOINCREMENT,
|
|
session_hash TEXT NOT NULL UNIQUE,
|
|
user_id INTEGER NOT NULL,
|
|
expires_at TEXT NOT NULL,
|
|
created_at TEXT NOT NULL,
|
|
last_seen_at TEXT NOT NULL,
|
|
revoked_at TEXT,
|
|
FOREIGN KEY(user_id) REFERENCES auth_users(id) ON DELETE CASCADE
|
|
);
|
|
CREATE INDEX IF NOT EXISTS idx_auth_sessions_user
|
|
ON auth_sessions(user_id, expires_at);
|
|
"""
|
|
)
|
|
|
|
|
|
def json_dumps(value: Any) -> str:
|
|
return json.dumps(value, ensure_ascii=False, separators=(",", ":"))
|
|
|
|
|
|
def json_loads(value: str | None) -> Any:
|
|
if not value:
|
|
return None
|
|
return json.loads(value)
|