The test data and test code are in the actual test work,Often stored separately,Conducive to separate maintenance of test data and test scripts.比如,Add several new sets of test data for the test case,Only the test data file needs to be modified,The test code does not need to be moved;If it is a test case, a new checkpoint is added,Just need to modify the test script,No modification of test data files is required.

—、准备测试数据

现在在data/Create a directory to store test dataYaml文件test_in_theaters.yaml,内容如下:

---
tests:
- case: 验证响应中start和countConsistent with the parameters in the request
  input:
    method: GET
    path: /v2/movie/in_theaters
    headers:
      User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.149 Safari/537.36
    params:
      apikey: 0df993c66c0c636e29ecbb5344252a4a
      start: 0
      count: 10
  expected:
    response:
      title: 正在上映的电影-上海
      count: 10
      start: 0
- case: 验证响应中title"正在上映的电影-北京"
  input:
    method: GET
    path: /v2/movie/in_theaters
    headers:
      User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.149 Safari/537.36
    params:
      apikey: 0df993c66c0c636e29ecbb5344252a4a
      start: 1
      count: 5
  expected:
    response:
      title: 正在上映的电影-北京
      count: 5
      start: 1

熟悉Yaml格式的同学,It should be easy to understand the contents of the test data file above.in this test data file,有一个数组tests,It contains two complete test data.A complete test data consists of three parts：

1. case,表示测试用例名称.

2. input,表示输入参数.

3. expected,表示预期结果.

上面的input输入参数,是一个http请求对象,Contains all parameters of the interface under test,包括请求方法、请求路径、请求头、请求参数.expected表示预期结果,in the test data above,Only the expected values ​​for the request response are listed,实际测试中,You can also list expected values ​​for the database.

二、编写测试脚本

新建一个测试脚本test_paramtrize_by_data_driven.py,内容如下：

import pytest

class TestDataDriven:
    @pytest.mark.datafile('data/test_in_theaters.yaml')  # A relative path relative to the project root directory
    def test_data_driven(self, parameters):
        print(parameters['input'])
        print(parameters['expected'])

First look at the test function,Above the test function a method called datafile的markerto provide data for the test function.The path to the test data is relative to the project root path,This means that the test data is in the root path of the projectdata目录中的test_in_theaters.yaml.The test function passesparameters这个fixtureGet the test data content.

三、实现pytest_generate_tests Hook

The above test script uses the test data method,是在tests/conftest.py文件中的pytest_generate_tests函数中实现的.

import yaml
import json

def pytest_generate_tests(metafunc):
    ids = []
    markers = metafunc.definition.own_markers
    for marker in markers:
        if marker.name == 'datafile':  # Read foreign data
            test_data_path = os.path.join(metafunc.config.rootdir, marker.args[0])  # Splicing test data paths
            with open(test_data_path) as f:
                ext = os.path.splitext(test_data_path)[-1]
                if ext in ['.yaml', '.yml']:
                    test_data = yaml.safe_load(f)
                elif ext == '.json':
                    test_data = json.load(f)
                else:
                    raise TypeError('datafile must be yaml or json,root must be tests')
    if "parameters" in metafunc.fixturenames:   # Parameterize with external data
        for data in test_data['tests']:  # 用test_data中的caseas the test case name
            ids.append(data['case'])
        # 用test_datathis listparameters进行参数化.
        metafunc.parametrize("parameters", test_data['tests'], ids=ids, scope=

metafunc.definition.own_markersCan read all the test functionsmarker,如果存在datafile这个marker,Indicates that test data needs to be read externally,Test data can beYAML格式也可以是Json格式.If the test function'smetafunc.fixturenames中含有parameters这个fixture函数,Just parameterize it with external test data.

四、Parametric routines

有了pytest_generate_tests这个Hook函数后,Parameterizing the test function requires three changes：

• 在data目录下新加一个YAML或者Json格式文件.Refer to the example above for the file content format.

• Specifies the path to the test data file@pytest.mark.datafile('data/test_in_theaters.yaml')

• Provide a name for the test data parameters的fixture

现在,The directory structure of the entire project should be as follows：

$ tree
.
├── Pipfile
├── Pipfile.lock
├── data
│   └── test_in_theaters.yaml
├── tests
│   └── conftest.py
│   └── test_data_driven.py

五、总结

This paper implements the core of the separation of test scripts and test data,还是实现了Hook函数,在Hookfunction to read the test data file,and for the test functionfixture进行参数化.

Just done with a small examplepostmanAn introduction to parameterization,Of course, you need to do it yourself to deepen your memory

最后在我的QQ技术交流群里整理了我这10几年软件测试生涯整理的一些技术资料,包括：电子书,简历模块,各种工作模板,面试宝典,自学项目等.如果在学习或工作中遇到问题,群里也会有大神帮忙解答,群号 798478386 ( 备注 今日头条555 )

全套软件测试自动化测试教学视频

 300G教程资料下载【视频教程+PPT+项目源码】

  全套软件测试自动化测试大厂面经