perf: 优化请求鉴权链路并批量化统计/调度查询

- 为 Pipeline/Context 增加按需读取请求体能力,支持 async 懒加载 JSON body
- 为 chat/cli/video/claude/openai-cli 适配器关闭默认预读,减少无效 body 读取与超时风险
- 将本地登录、JWT 用户加载、API Key 鉴权迁移到线程池隔离会话执行,避免阻塞事件循环
- 为 API Key 鉴权返回结构化余额结果,并在主请求会话中重新绑定 user/api_key 后再校验状态、过期和锁定信息
- 为 management/user token 前缀认证引入独立会话与结果回绑,避免跨会话对象写入失效

- 为 Usage 余额检查补充结构化返回,统一透出 remaining 与欠费/不可用文案映射
- 为用户与管理端活跃请求查询增加 maintain_status 开关,避免轮询指定 id 时误触发状态修复
- 重写 user_me usage 汇总逻辑,支持 group_by=None 的粗粒度聚合
- 修正 provider 维度成功率与平均响应时间统计,基于 success_count 和成功响应耗时汇总计算
- 前端 Usage 轮询由 setInterval 改为串行 setTimeout,避免并发轮询叠加

- 为 StatsAggregator 增加按本地日期批量计算百分位能力,替代逐天 fan-out 查询
- 为混合统计查询合并连续实时日期区间,并批量读取 StatsDaily,减少逐日查询次数
- 为用户日统计增加批量聚合入口,替代逐用户循环聚合
- 为系统配置导出改用 selectinload 预加载 provider 关联数据,减少 N+1 查询
- 为管理员用户列表增加钱包批量查询,避免逐用户回表

- 为调度器增加 provider 轻量引用预过滤,先按 allowed_providers 缩小范围再加载完整 provider 图
- 为 CandidateBuilder 增加 provider refs/provider_ids 查询能力,保留分页顺序
- 为模型缓存增加 provider_model_mappings 索引缓存与 model_mappings 规则缓存,减少重复全量扫描
- 为请求候选中间态改为 flush/batch commit,降低 pending/streaming 状态切换的事务往返
- 为钱包访问结果补充 balance_snapshot,并抽取余额快照复用逻辑

- 补充 pipeline、auth、admin users、user_me usage、stats aggregator、model cache、
  scheduler、wallet、request candidate 等回归与契约测试
This commit is contained in:
AAEE86
2026-03-09 22:57:23 +08:00
parent bf818e3b61
commit 57c7cca556
37 changed files with 2490 additions and 498 deletions

View File

@@ -0,0 +1,93 @@
from __future__ import annotations
from datetime import datetime, timezone
from types import SimpleNamespace
from typing import Any
from unittest.mock import MagicMock
import pytest
from fastapi import FastAPI
from fastapi.testclient import TestClient
from src.api.admin.users.routes import router as admin_users_router
from src.database import get_db
def _build_admin_users_app(db: MagicMock, monkeypatch: pytest.MonkeyPatch) -> TestClient:
app = FastAPI()
app.include_router(admin_users_router)
app.dependency_overrides[get_db] = lambda: db
async def _fake_pipeline_run(
*, adapter: Any, http_request: object, db: MagicMock, mode: object
) -> Any:
_ = http_request, mode
context = SimpleNamespace(
db=db,
request=SimpleNamespace(state=SimpleNamespace()),
user=SimpleNamespace(id="admin-1"),
ensure_json_body=lambda: {},
add_audit_metadata=lambda **_: None,
)
return await adapter.handle(context)
monkeypatch.setattr("src.api.admin.users.routes.pipeline.run", _fake_pipeline_run)
return TestClient(app)
def test_list_users_uses_wallet_batch_lookup(monkeypatch: pytest.MonkeyPatch) -> None:
db = MagicMock()
client = _build_admin_users_app(db, monkeypatch)
now = datetime.now(timezone.utc)
users = [
SimpleNamespace(
id="user-1",
email="u1@example.com",
username="user1",
role=SimpleNamespace(value="user"),
allowed_providers=None,
allowed_api_formats=None,
allowed_models=None,
is_active=True,
created_at=now,
updated_at=now,
last_login_at=None,
),
SimpleNamespace(
id="user-2",
email="u2@example.com",
username="user2",
role=SimpleNamespace(value="admin"),
allowed_providers=None,
allowed_api_formats=None,
allowed_models=None,
is_active=True,
created_at=now,
updated_at=None,
last_login_at=None,
),
]
wallets_by_user_id = {
"user-1": SimpleNamespace(limit_mode="unlimited"),
}
batch_getter = MagicMock(return_value=wallets_by_user_id)
monkeypatch.setattr(
"src.api.admin.users.routes.UserService.list_users", lambda *_a, **_k: users
)
monkeypatch.setattr(
"src.api.admin.users.routes.WalletService.get_wallets_by_user_ids",
batch_getter,
)
monkeypatch.setattr(
"src.api.admin.users.routes.WalletService.get_wallet",
lambda *_a, **_k: (_ for _ in ()).throw(AssertionError("不应回退到逐个钱包查询")),
)
response = client.get("/api/admin/users")
assert response.status_code == 200
assert response.json()[0]["unlimited"] is True
assert response.json()[1]["unlimited"] is False
batch_getter.assert_called_once()
assert batch_getter.call_args.args[1] == ["user-1", "user-2"]