Use when establishing tests, m…

# 订单服务质量门禁方案

一、示例场景

订单服务核心接口 POST /api/order/create，依赖库存服务RPC和MySQL，逻辑：校验商品→锁库存→生成订单→扣款。

二、单元测试（pytest，覆盖率82%）

```python

# test_order.py

import pytest

from unittest.mock import patch, MagicMock

from order_service import create_order, OrderError

@pytest.fixture

def mock_deps():

with patch("order_service.stock_client") as s, \

patch("order_service.db") as db:

yield s, db

def test_create_order_success(mock_deps):

s, db = mock_deps

s.lock_stock.return_value = True

db.insert.return_value = "ORD001"

order = create_order(user_id=1, sku="A100", qty=2, price=99.0)

assert order["order_id"] == "ORD001"

assert order["amount"] == 198.0

def test_create_order_insufficient_stock(mock_deps):

s, _ = mock_deps

s.lock_stock.return_value = False

with pytest.raises(OrderError, match="库存不足"):

create_order(1, "A100", 100, 99.0)

def test_create_order_invalid_qty(mock_deps):

with pytest.raises(OrderError, match="数量非法"):

create_order(1, "A100", 0, 99.0)

def test_create_order_db_rollback(mock_deps):

s, db = mock_deps

s.lock_stock.return_value = True

db.insert.side_effect = Exception("DB down")

with pytest.raises(OrderError, match="下单失败"):

create_order(1, "A100", 1, 99.0)

assert s.release_stock.called # 验证库存回滚

```

运行：pytest --cov=order_service --cov-fail-under=80 test_order.py

三、Prometheus监控

应用埋点（prometheus_client）：

```python

from prometheus_client import Histogram

LATENCY = Histogram("order_create_latency_seconds", "下单延迟",

buckets=[0.05, 0.1, 0.25, 0.5,