Detailed explanation of design ideas of webUI test framework

前言

自动化测试中,The most important thing is to build the framework,A good automated testing framework can greatly reduce maintenance costs,提高效率.这篇文章就简单介绍一下UIThe idea of ​​building an automated testing framework.
基于POM+DDT+pytest+selenium+allure搭建的web自动化测试框架：https://github.com/tdx1997tdx/web_auto_testing

POM设计模式

POM,全程Page Object Model,is the testing industryUIMainstream design patterns for automation frameworks.The entire framework separates code and data,逻辑代码与测试代码分离,POM的结构如下：

1. 基类：Mainly produces a series of functions that can be called in the page object.The existence of itself as a tool library.
2. 页面对象类：逻辑代码部分.结合实际业务,Extract all pages that can be automated.The class contains the core elements of the page and the core process of the page.
3. 测试用例类：测试代码部分.Used to stitch various page objects,Implement the final test flow.
4. 测试数据类：It is used to extract all the data content that needs to be applied in the actual test process.

DDT数据驱动测试

数据驱动测试,Test cases are driven by external data sets,A different set of data to perform the same operation(a script),Test data and test operations are completely separated from the scripting design pattern,从数据文件读取输入数据,通过变量的参数化,将测试数据传入测试脚本,不同的数据文件对应不同的测试用例.

ddt优势如下：

• 代码复用率高,一次编写多条数据复用逻辑

• 异常排查效率高,测试执行隔离,数据间无影响

• 代码可维护性高,提高代码的易读性

项目实战讲解

项目结构介绍

• base_page:The base class mainly produces a series of functions that can be called in the page object.The existence of itself as a tool library.
• page_object:页面对象类,逻辑代码部分.结合实际业务,Extract all pages that can be automated.The class contains the core elements of the page and the core process of the page.
• test_case:Store the test case code implementation,Used to stitch various page objects,Implement the final test flow.
• data:存放测试数据,ddt实现的核心.
• config:Store general configuration
• report:存放allure测试报告
• log:存放日志
• utils:Store some general tool kits

base_page

POMmiddle class,Mainly encapsulate someselenium中常用的方法

base_page.py

from selenium.webdriver import Keys

from config.config import url as index_url


class BasePage:

    # 构造函数
    def __int__(self, driver):
        self.driver = driver

    # 访问url
    def visit(self, url):
        self.driver.get(url)

    def visit_index(self):
        self.driver.get(index_url)

    # 元素定位
    def locate(self, loc):
        return self.driver.find_element(*loc)

    # 输入文本
    def input(self, loc, text):
        self.locate(loc).send_keys(text)

    # 单击
    def click(self, loc):
        self.locate(loc).click()

    # refresh
    def refresh_page(self):
        self.driver.refresh()

    # 按下回车键
    def enter(self, loc):
        self.locate(loc).send_keys(Keys.ENTER)

page_object

Stores page objects and the core business flow of the page

业务主页面：

index_page.py

from selenium.webdriver.common.by import By

from base_page.base_page import BasePage


class IndexPage(BasePage):
    uri = "/"
    # 核心元素
    search_input = (By.XPATH, '//*[@id="layout"]/div/div[1]/div[2]/div/span/div[1]/span/span[1]/input')

    def __init__(self, driver):
        super().__int__(driver)

    # core business flow

    def search(self, txt):
        self.visit_index()
        self.input(self.search_input, txt)
        self.enter(self.search_input)

测试用例中,We only relate to the search function,So abstract core elements and core business flows

• 核心元素：The framed part of the picture,That is, the search input box
• core business flow：
  1. 进入页面
  2. 定位输入框,并输入内容
  3. Click Enter to search for content

test_case

测试用例类,使用pytest框架搭建

from selenium import webdriver
from config.chrome_options import options
from page_object.login_page import LoginPage
from page_object.index_page import IndexPage
import allure
import pytest
from utils.json_tool import load_json
from utils.log_tool import logger
import time


@allure.feature('Search for module tests')
class TestSearch:

    def setup_class(self):
        self.driver = webdriver.Chrome(options=options())
        lp = LoginPage(self.driver)
        lp.login_with_cookie()

    def teardown_class(self):
        self.driver.quit()

    @pytest.mark.parametrize("test_input,story,title", load_json("./data/search_data.json"))
    def test_search(self, test_input, story, title):
        """ 用例描述：搜索测试 """
        # Dynamically add modules and headers
        allure.dynamic.story(story)
        allure.dynamic.title(title)
        ip = IndexPage(self.driver)
        ip.search(test_input["search_txt"])
        logger.info(str(test_input) + "finish")
        time.sleep(2)

@pytest.mark.parametrize和load_json组成ddt的核心,下面看看data中json文件的定义：

[
  {
    
    "data": {
    
      "search_txt": "tdx"
    },
    "story": "用例1：输入tdx",
    "title": "输入测试tdx"
  },
  {
    
    "data": {
    
      "search_txt": "mytdx"
    },
    "story": "用例2：输入mytdx",
    "title": "输入测试mytdx"
  },
  {
    
    "data": {
    
      "search_txt": "三国"
    },
    "story": "用例3：Enter the Three Kingdoms",
    "title": "Enter Test Three Kingdoms"
  }
]

我们封装个load_jsonmethod to read data for data-driven testing

import json


# 读取json文件内容,Convert to a data-driven format
def load_json(path):
    with open(path) as f:
        data = json.load(f)
    res = []
    for i in data:
        res.append((i["data"], i["story"], i["title"]))
    return res

data

data中存放ddt的相关数据,Separate data and code
search_data.json

[
  {
    
    "data": {
    
      "search_txt": "tdx"
    },
    "story": "用例1：输入tdx",
    "title": "输入测试tdx"
  },
  {
    
    "data": {
    
      "search_txt": "mytdx"
    },
    "story": "用例2：输入mytdx",
    "title": "输入测试mytdx"
  },
  {
    
    "data": {
    
      "search_txt": "三国"
    },
    "story": "用例3：Enter the Three Kingdoms",
    "title": "Enter Test Three Kingdoms"
  }
]

成果展示

自动化测试流程

测试报告