import csv
import pytest

def load_test_data(file_path):
    """从CSV加载测试数据"""
    test_data = []
    with open(file_path, 'r') as file:
        reader = csv.DictReader(file)
        for row in reader:
            test_data.append(row)
    return test_data

@pytest.mark.parametrize("username,password,expected", 
    load_test_data("login_data.csv"))
def test_login(username, password, expected):
    result = login_api(username, password)
    assert result["status"] == expected

# 错误示范：数据逻辑混在测试里
def test_login_with_data():
    for row in load_test_data("login_data.csv"):
        result = login_api(row["username"], row["password"])
        # 这里处理了数据转换，污染了测试逻辑

# 正确示范：数据逻辑完全剥离
def test_login_execution(test_case):
    # test_case 已经由数据提供层处理好
    result = login_api(test_case.username, test_case.password)
    assert result.status == test_case.expected_status

case_id,username,email,password,age,expected_code,expected_msg
001,test_user1,user1@test.com,Pass123!,25,200,success
002,,user1@test.com,Pass123!,25,400,username_required
003,test_user2,invalid-email,Pass123!,25,400,invalid_email
004,test_user3,user3@test.com,123,25,400,weak_password
005,test_user4,user4@test.com,Pass123!,-1,400,invalid_age
006,test_user5,user5@test.com,Pass123!,151,400,invalid_age
...

import pytest
import requests
from dataclasses import dataclass
import csv

@dataclass
class RegisterCase:
    case_id: str
    username: str
    email: str
    password: str
    age: int
    expected_code: int
    expected_msg: str

def load_register_cases():
    cases = []
    with open('register_cases.csv', 'r') as f:
        reader = csv.DictReader(f)
        for row in reader:
            cases.append(RegisterCase(
                case_id=row['case_id'],
                username=row['username'],
                email=row['email'],
                password=row['password'],
                age=int(row['age']),
                expected_code=int(row['expected_code']),
                expected_msg=row['expected_msg']
            ))
    return cases

@pytest.mark.parametrize("case", load_register_cases())
def test_user_registration(case):
    """数据驱动注册测试"""
    # 跳过空值字段（不发送）
    payload = {}
    if case.username:
        payload['username'] = case.username
    if case.email:
        payload['email'] = case.email
    if case.password:
        payload['password'] = case.password
    if case.age:
        payload['age'] = case.age
    
    response = requests.post(
        "https://api.example.com/register",
        json=payload
    )
    
    assert response.status_code == case.expected_code
    assert case.expected_msg in response.json().get('message', '')

<div style="margin:24px 0;border:1px solid #E3F2FD;border-radius:8px;background:#FAFAFA;overflow:hidden;"><div style="display:flex;align-items:center;gap:12px;padding:12px 20px;background:linear-gradient(135deg,#E3F2FD,#BBDEFB);"><span style="width:36px;height:36px;background:#2196F3;color:#fff;border-radius:50%;display:inline-flex;align-items:center;justify-content:center;font-weight:bold;font-size:16px;flex-shrink:0;">2. </span><span style="font-size:17px;font-weight:700;color:#1565C0;">组合测试：Pairwise方法</span></div><div style="padding:16px 20px;"><p>当你有5个参数，每个参数有3-4个取值时，全组合测试用例数量是 4^5 = 1024个。使用Pairwise（成对组合）可以缩减到20-30个。</p></div></div>
# 使用allpairspy库
from allpairspy import AllPairs

parameters = [
    ["valid", "invalid", "empty"],           # username
    ["valid", "invalid", "empty"],           # email
    ["strong", "weak", "empty"],             # password
    ["adult", "minor", "invalid"],           # age
    ["accepted", "rejected"],                # terms
]

for i, combo in enumerate(AllPairs(parameters)):
    print(f"Case {i+1}: {combo}")
    # 输出只有16个组合，覆盖所有成对交互

<blockquote style="margin:16px 0;padding:12px 20px;background:#F5F5F5;border-left:4px solid #9E9E9E;color:#616161;font-style:italic;border-radius:0 6px 6px 0;">数据是测试的活水，代码是固定的管道。管道不动，水流动。</blockquote>

## 05 陷阱：数据驱动测试的5个常见错误

### 错误1：数据与代码耦合

**解决方案**：数据必须从外部文件或数据库读取。

### 错误2：无意义的数据命名

**解决方案**：使用有意义的列名和case_id，每个case标注测试目的。

### 错误3：忽略数据版本控制
测试数据和代码一样需要版本管理。把CSV文件提交到Git仓库，记录每次修改的原因。

### 错误4：过度参数化

**解决方案**：使用数据类（dataclass）或字典封装参数，保持测试方法简洁。

### 错误5：不处理数据错误

**解决方案**：添加数据验证和错误处理。