httprunner学习5-参数化与数据驱动

前言

参数化是自动化测试离不开的话题，httprunner里面只要把上一篇声明变量学会了，参数化也就自然会了。

不同的地方在于声明变量时对应值只有一个，参数化是多个值，存放在list里面。

httprunner==1.5.8

登录参数化

先准备测试数据，准备四组登录用的账号和密码，账号为test1,test2,test3,test4,密码统一设置为123456。

参数user账号数据，设置对应的值 ["test1", "test2", "test3", "test4"],定义单个参数用variables，定义参数化用parameters

    parameters:
        - user: ["test1", "test2", "test3", "test4"]  # 参数化
        - psw: ["123456"]

如果参数化里面的数据只有一个，比如psw对应的值只有一个，使用variables和parameters对应都可以

    parameters:
        - user: ["test1", "test2", "test3", "test4"]  # 参数化
    variables:
        psw: 123456

完整的test_parameters.yml脚本如下

# 上海悠悠，QQ交流群：750815713
- config:
    name: logincase
    parameters:
        - user: ["test1", "test2", "test3", "test4"]    # 参数化
        - psw: ["123456"]
- test:
    name: login case1
    request:
        url: http://127.0.0.1:8000/api/v1/login/
        method: POST
        headers:
            Content-Type: application/json
            User-Agent: python-requests/2.18.4
        json:
            username: $user               # 引用变量
            password: $psw
    extract:
        - token: content.token         # 提取token
    validate:
        - eq: [status_code, 200]
        - eq: [headers.Content-Type, application/json]
        - eq: [content.msg, login success!]
        - eq: [content.code, 0]

运行用例

运行用例，会自动生成四个测试用例

D:\soft\untitled>hrun test_parameters.yml
login case1
INFO     POST http://127.0.0.1:8000/api/v1/login/
INFO     status_code: 200, response_time(ms): 523.77 ms, response_length: 110 bytes
INFO     start to extract from response object.
INFO     start to validate.
.
login case1
INFO     POST http://127.0.0.1:8000/api/v1/login/
INFO     status_code: 200, response_time(ms): 525.67 ms, response_length: 110 bytes
INFO     start to extract from response object.
INFO     start to validate.
.
login case1
INFO     POST http://127.0.0.1:8000/api/v1/login/
INFO     status_code: 200, response_time(ms): 436.3 ms, response_length: 110 bytes
INFO     start to extract from response object.
INFO     start to validate.
.
login case1
INFO     POST http://127.0.0.1:8000/api/v1/login/
INFO     status_code: 200, response_time(ms): 487.6 ms, response_length: 110 bytes
INFO     start to extract from response object.
INFO     start to validate.
.

----------------------------------------------------------------------
Ran 4 tests in 2.037s

OK
INFO     Start to render Html report ...
INFO     Generated Html report: D:\soft\untitled\reports\1569118975.html

查看报告

关联参数

上面的案例多个账号，密码都是一样的，如果密码不一样呢？上面的就行不通了，所以针对于一个账号对应一个密码，这种一一对应的关系，可以用关联性的参数化

    parameters:
        - user-psw:
            - ["test1", "123456"]
            - ["test2", "123456"]
            - ["test3", "123456"]
            - ["test3", "123456"]

完整的测试用例test_param_psw.yml

# 上海悠悠，QQ交流群：750815713
- config:
    name: logincase
    parameters:
        - user-psw:
            - ["test1", "123456"]
            - ["test2", "123456"]
            - ["test3", "123456"]
            - ["test3", "123456"]
- test:
    name: login case1
    request:
        url: http://127.0.0.1:8000/api/v1/login/
        method: POST
        headers:
            Content-Type: application/json
            User-Agent: python-requests/2.18.4
        json:
            username: $user
            password: $psw
    extract:
        - token: content.token         # 提取token
    validate:
        - eq: [status_code, 200]
        - eq: [headers.Content-Type, application/json]
        - eq: [content.msg, login success!]
        - eq: [content.code, 0]

笛卡尔积组合

比如测试账号有四种["test1", "test2", "test3", "test4"],密码也有四种 ["123456", "123456", "123456", "123456"]

用笛卡尔积组合的话，就是4*4=16种组合

# 上海悠悠，QQ交流群：750815713
- config:
    name: logincase
    parameters:
        - user: ["test1", "test2", "test3", "test4"]
        - psw: ["123456", "123456", "123456", "123456"]

- test:
    name: login case1
    request:
        url: http://127.0.0.1:8000/api/v1/login/
        method: POST
        headers:
            Content-Type: application/json
            User-Agent: python-requests/2.18.4
        json:
            username: $user
            password: $psw
    extract:
        - token: content.token         # 提取token
    validate:
        - eq: [status_code, 200]
        - eq: [headers.Content-Type, application/json]
        - eq: [content.msg, login success!]
        - eq: [content.code, 0]

这样运行会生成16组用例，很显然，这种不适用与登录-密码一对一的情况。可以用在其它的测试场景