这是本节的多页打印视图。点击此处打印.

返回本页常规视图.

入门指南

如果你是Selenium的新手, 我们有一些资源帮助你快速入门.

1: 安装Selenium类库
2: 编写第一个Selenium脚本
3: Organizing and Executing Selenium Code

Selenium 通过使用 WebDriver 支持市场上所有主流浏览器的自动化。 Webdriver 是一个 API 和协议，它定义了一个语言中立的接口，用于控制 web 浏览器的行为。每个浏览器都有一个特定的 WebDriver 实现，称为驱动程序。驱动程序是负责委派给浏览器的组件，并处理与 Selenium 和浏览器之间的通信。

这种分离是有意识地努力让浏览器供应商为其浏览器的实现负责的一部分。 Selenium 在可能的情况下使用这些第三方驱动程序，但是在这些驱动程序不存在的情况下，它也提供了由项目自己维护的驱动程序。

Selenium 框架通过一个面向用户的界面将所有这些部分连接在一起，该界面允许透明地使用不同的浏览器后端，从而实现跨浏览器和跨平台自动化。

Selenium setup is quite different from the setup of other commercial tools. Before you can start writing Selenium code, you have to install the language bindings libraries for your language of choice, the browser you want to use, and the driver for that browser.

Follow the links below to get up and going with Selenium WebDriver.

If you wish to start with a low-code/record and playback tool, please check Selenium IDE

Once you get things working, if you want to scale up your tests, check out the Selenium Grid.

1 - 安装Selenium类库

配置自动化的浏览器.

首先，您需要为自动化项目安装 Selenium 绑定库。库的安装过程取决于您选择使用的语言。

请求对应的程序语言

查看该库所支持java的最低版本 here.

应熟练掌握build tool以安装支持java的Selenium库

Maven

具体的依赖位于项目中的 pom.xml 文件:

        <dependency>
            <groupId>org.seleniumhq.selenium</groupId>
            <artifactId>selenium-java</artifactId>
            <version>4.15.0</version>
        </dependency>

View full example on GitHub

Gradle

具体的依赖位于项目中的 build.gradle 文件中的 testImplementation:

    testImplementation 'org.seleniumhq.selenium:selenium-java:4.15.0'
    testImplementation 'org.junit.jupiter:junit-jupiter-engine:5.10.0'

View full example on GitHub

该库所支持的Python版本最低版本可以在 支持的Python版本 章节中找到 PyPi

这里提供了几种不同的方式来安装 Selenium .

Pip

pip install selenium

下载

此外你可以从这里下载 PyPI source archive (selenium-x.x.x.tar.gz) 并通过: setup.py 文件安装:

python setup.py install

在项目中使用

为了在项目中使用它,需要将它添加到 requirements.txt 文件中:

selenium==4.15.2

View full example on GitHub

Selenium 所支持的所有平台的列表一览见诸于 Nuget

该处阐述了一些安装Selenium的选项.

包管理器

Install-Package Selenium.WebDriver

.NET CLI

dotnet add package Selenium.WebDriver

CSProj

在 csproj 文件里, 具体的依赖 PackageReference(包参数) 位于 ItemGroup (项目组)中:

      <PackageReference Include="Selenium.WebDriver" Version="4.15.0" />

View full example on GitHub

其他附加思虑事项

更多的注意事项,适用于使用 Visual Studio Code (vscode) 和 C#

安装兼容的 .NET SDK 作为章节的先决条件同时安装 vscode 的扩展 (Ctrl-Shift-X)以适配 C# 和 NuGet 可以遵照此处进行操作指南创建 C# 控制台项目并运行 “Hello World”. 你也可以用命令行 dotnet new NUnit 创建NUnit初阶项目. 确保文件 %appdata%\NuGet\nuget.config 已经配置完成,就像某位开发者报告的问题一样,它可能因为某种因素被自动清空. 如果 nuget.config 是空的,或者未配置的,那么 .NET 创建的Selenium项目可能失败. 加入如下章节到文件 nuget.config 如果出现清空的情况:

<configuration>
  <packageSources>
    <add key="nuget.org" value="https://api.nuget.org/v3/index.json" protocolVersion="3" />
    <add key="nuget.org" value="https://www.nuget.org/api/v2/" />   
  </packageSources>
...

更多关于 nuget.config 的信息点击. 你可能需要按照自己的需求配置 nuget.config .

现在,返回 vscode ,按下 Ctrl-Shift-P, 然后键入 “NuGet Add Package”, 并选择自己需要的 Selenium 包,例如 Selenium.WebDriver. 按下回车并选择版本. 现在你可以使用说明文档中关于 C# vscode下的案例了.

你可以查看 Selenium 对 Ruby 版本支持和最低支持. 具体位于 rubygems.org

Selenium 可以使用两种不同方法安装.

手动安装

gem install selenium-webdriver

加入项目的 gemfile

gem 'selenium-devtools', '~> 0.119'

View full example on GitHub

You can find the minimum required version of Node for any given version of Selenium in the 你可以在此查看 Selenium 对 Node 的版本支持情况位于 Node Support Policy 中的相关章节 npmjs

Selenium is typically installed using npm.

本地安装

npm install selenium-webdriver

加入项目

在你的项目 package.json, 必须加入到 dependencies:

        "mocha": "^10.2.0"

View full example on GitHub

Use the Java bindings for Kotlin.

下一步

创建你的第一个Selenium脚本

2 - 编写第一个Selenium脚本

逐步构建一个Selenium脚本的说明

当你完成 Selenium安装后, 便可以开始书写Selenium脚本了.

八个基本组成部分

Selenium所做的一切, 就是发送给浏览器命令, 用以执行某些操作或为信息发送请求. 您将使用Selenium执行的大部分操作, 都是以下基本命令的组合

Click on the link to “View full example on GitHub” to see the code in context.

1. 使用驱动实例开启会话

关于如何启动会话，请浏览我们的文档驱动会话

        WebDriver driver = new ChromeDriver();

View full example on GitHub

driver = webdriver.Chrome()

View full example on GitHub

        IWebDriver driver = new ChromeDriver();

View full example on GitHub

driver = Selenium::WebDriver.for :chrome

View full example on GitHub

    driver = await new Builder().forBrowser('chrome').build();

View full example on GitHub

        driver = ChromeDriver()

View full example on GitHub

2. 在浏览器上执行操作

在本例中, 我们导航到一个网页.

        driver.get("https://www.selenium.dev/selenium/web/web-form.html");

View full example on GitHub

driver.get("https://www.selenium.dev/selenium/web/web-form.html")

View full example on GitHub

        driver.Navigate().GoToUrl("https://www.selenium.dev/selenium/web/web-form.html");

View full example on GitHub

driver.get('https://www.selenium.dev/selenium/web/web-form.html')

View full example on GitHub

    await driver.get('https://www.selenium.dev/selenium/web/web-form.html');

View full example on GitHub

        driver.get("https://www.selenium.dev/selenium/web/web-form.html")

View full example on GitHub

3. 请求浏览器信息

您可以请求一系列关于浏览器的信息 , 包括窗口句柄、浏览器尺寸/位置、cookie、警报等.

        driver.getTitle();

View full example on GitHub

title = driver.title

View full example on GitHub

        var title = driver.Title;

View full example on GitHub

driver.title

View full example on GitHub

    let title = await driver.getTitle();

View full example on GitHub

        val title = driver.title

View full example on GitHub

4. 建立等待策略

将代码与浏览器的当前状态同步是Selenium面临的最大挑战之一, 做好它是一个高级主题.

基本上, 您希望在尝试定位元素之前, 确保该元素位于页面上, 并且在尝试与该元素交互之前, 该元素处于可交互状态.

隐式等待很少是最好的解决方案, 但在这里最容易演示, 所以我们将使用它作为占位符.

阅读更多关于等待策略的信息.

        driver.manage().timeouts().implicitlyWait(Duration.ofMillis(500));

View full example on GitHub

driver.implicitly_wait(0.5)

View full example on GitHub

        driver.Manage().Timeouts().ImplicitWait = TimeSpan.FromMilliseconds(500);

View full example on GitHub

driver.manage.timeouts.implicit_wait = 500

View full example on GitHub

    await driver.manage().setTimeouts({implicit: 500});

View full example on GitHub

        driver.manage().timeouts().implicitlyWait(Duration.ofMillis(500))

View full example on GitHub

5. 发送命令查找元素

大多数Selenium会话中的主要命令都与元素相关, 如果不先找到元素, 就无法与之交互.

        WebElement textBox = driver.findElement(By.name("my-text"));
        WebElement submitButton = driver.findElement(By.cssSelector("button"));

View full example on GitHub

text_box = driver.find_element(by=By.NAME, value="my-text")
submit_button = driver.find_element(by=By.CSS_SELECTOR, value="button")

View full example on GitHub

        var textBox = driver.FindElement(By.Name("my-text"));
        var submitButton = driver.FindElement(By.TagName("button"));

View full example on GitHub

text_box = driver.find_element(name: 'my-text')
submit_button = driver.find_element(tag_name: 'button')

View full example on GitHub

    let textBox = await driver.findElement(By.name('my-text'));
    let submitButton = await driver.findElement(By.css('button'));

View full example on GitHub

        var textBox = driver.findElement(By.name("my-text"))
        val submitButton = driver.findElement(By.cssSelector("button"))

View full example on GitHub

6. 操作元素

对于一个元素, 只有少数几个操作可以执行, 但您将经常使用它们.

        textBox.sendKeys("Selenium");
        submitButton.click();

View full example on GitHub

text_box.send_keys("Selenium")
submit_button.click()

View full example on GitHub

        textBox.SendKeys("Selenium");
        submitButton.Click();

View full example on GitHub

text_box.send_keys('Selenium')
submit_button.click

View full example on GitHub

    await textBox.sendKeys('Selenium');
    await submitButton.click();

View full example on GitHub

        textBox.sendKeys("Selenium")
        submitButton.click()

View full example on GitHub

7. 获取元素信息

元素存储了很多被请求的信息.

        message.getText();

View full example on GitHub

text = message.text

View full example on GitHub

        var value = message.Text;

View full example on GitHub

message.text

View full example on GitHub

    let value = await message.getText();

View full example on GitHub

        val value = message.getText()

View full example on GitHub

8. 结束会话

这将结束驱动程序进程, 默认情况下, 该进程也会关闭浏览器. 无法向此驱动程序实例发送更多命令.

See Quitting Sessions.

        driver.quit();

View full example on GitHub

driver.quit()

View full example on GitHub

        driver.Quit();

View full example on GitHub

driver.quit

View full example on GitHub

    await driver.quit();

View full example on GitHub

        driver.quit()

View full example on GitHub

Running Selenium File

Coding Help

Note: This section could use some updated code examples

Add example for how to execute each of the above pages via command line

Check our contribution guidelines and code example formats if you’d like to help.

接下来的步骤

Most Selenium users execute many sessions and need to organize them to minimize duplication and keep the code more maintainable. Read on to learn about how to put this code into context for your use case with Using Selenium.

3 - Organizing and Executing Selenium Code

Scaling Selenium execution with an IDE and a Test Runner library

If you want to run more than a handful of one-off scripts, you need to be able to organize and work with your code. This page should give you ideas for how to actually do productive things with your Selenium code.

Common Uses

Most people use Selenium to execute automated tests for web applications, but Selenium support any use case of browser automation.

Repetitive Tasks

Perhaps you need to log into a website and download something, or submit a form. You can create a Selenium script to run with a service at preset times.

Web Scraping

Are you looking to collect data from a site that doesn’t have an API? Selenium will let you do this, but please make sure you are familiar with the website’s terms of service as some websites do not permit it and others will even block Selenium.

Testing

Running Selenium for testing requires making assertions on actions taken by Selenium. So a good assertion library is required. Additional features to provide structure for tests require use of Test Runner.

IDEs

Regardless of how you use Selenium code, you won’t be very effective writing or executing it without a good Integrated Developer Environment. Here are some common options…

Test Runner

Even if you aren’t using Selenium for testing, if you have advanced use cases, it might make sense to use a test runner to better organize your code. Being able to use before/after hooks and run things in groups or in parallel can be very useful.

Choosing

There are many different test runners available.

All the code examples in this documentation can be found in (or is being moved to) our example directories that use test runners and get executed every release to ensure all the code is correct and updated. Here is a list of test runners with links. The first item is the one that is used by this repository and the one that will be used for all examples on this page.

JUnit - A widely-used testing framework for Java-based Selenium tests.
TestNG - Offers extra features like parallel test execution and parameterized tests.

pytest - A preferred choice for many, thanks to its simplicity and powerful plugins.
unittest - Python’s standard library testing framework.

NUnit - A popular unit-testing framework for .NET.
MS Test - Microsoft’s own unit testing framework.

RSpec - The most widely used testing library for running Selenium tests in Ruby.
Minitest - A lightweight testing framework that comes with Ruby standard library.

Jest - Primarily known as a testing framework for React, it can also be used for Selenium tests.
Mocha - The most common JS library for running Selenium tests.

Installing

This is very similar to what was required in Install a Selenium Library. This code is only showing examples for what is being used in our Documentation Examples project.

Maven

Gradle

To use it in a project, add it to the requirements.txt file:

in the project’s csproj file, specify the dependency as a PackageReference in ItemGroup:

Add to project’s gemfile

In your project’s package.json, add requirement to dependencies:

Asserting

Setting Up and Tearing Down

Executing

Maven

mvn clean test

Gradle

gradle clean test

mocha runningTests.spec.js

Examples

In First script, we saw each of the components of a Selenium script. Here’s an example of that code using a test runner:

package dev.selenium.getting_started;

import org.junit.jupiter.api.Test;
import org.openqa.selenium.By;
import org.openqa.selenium.WebDriver;
import org.openqa.selenium.WebElement;
import org.openqa.selenium.chrome.ChromeDriver;

import java.time.Duration;

import static org.junit.jupiter.api.Assertions.assertEquals;

public class UsingSeleniumTest {

    @Test
    public void eightComponents() {
        WebDriver driver = new ChromeDriver();
        driver.get("https://www.selenium.dev/selenium/web/web-form.html");

        String title = driver.getTitle();
        assertEquals("Web form", title);

        driver.manage().timeouts().implicitlyWait(Duration.ofMillis(500));

        WebElement textBox = driver.findElement(By.name("my-text"));
        WebElement submitButton = driver.findElement(By.cssSelector("button"));

        textBox.sendKeys("Selenium");
        submitButton.click();

        WebElement message = driver.findElement(By.id("message"));
        String value = message.getText();
        assertEquals("Received!", value);

        driver.quit();
    }

}

View full example on GitHub

from selenium import webdriver
from selenium.webdriver.common.by import By


def test_eight_components():
    driver = webdriver.Chrome()

    driver.get("https://www.selenium.dev/selenium/web/web-form.html")

    title = driver.title
    assert title == "Web form"

    driver.implicitly_wait(0.5)

    text_box = driver.find_element(by=By.NAME, value="my-text")
    submit_button = driver.find_element(by=By.CSS_SELECTOR, value="button")

    text_box.send_keys("Selenium")
    submit_button.click()

    message = driver.find_element(by=By.ID, value="message")
    value = message.text
    assert value == "Received!"

    driver.quit()

View full example on GitHub

using System;
using Microsoft.VisualStudio.TestTools.UnitTesting;
using OpenQA.Selenium;
using OpenQA.Selenium.Chrome;

namespace SeleniumDocs.GettingStarted
{
    [TestClass]
    public class UsingSeleniumTest
    {

        [TestMethod]
        public void EightComponents()
        {
            IWebDriver driver = new ChromeDriver();

            driver.Navigate().GoToUrl("https://www.selenium.dev/selenium/web/web-form.html");

            var title = driver.Title;
            Assert.AreEqual("Web form", title);

            driver.Manage().Timeouts().ImplicitWait = TimeSpan.FromMilliseconds(500);

            var textBox = driver.FindElement(By.Name("my-text"));
            var submitButton = driver.FindElement(By.TagName("button"));
            
            textBox.SendKeys("Selenium");
            submitButton.Click();
            
            var message = driver.FindElement(By.Id("message"));
            var value = message.Text;
            Assert.AreEqual("Received!", value);
            
            driver.Quit();
        }
    }
}

View full example on GitHub

# frozen_string_literal: true

require 'spec_helper'

RSpec.describe 'Using Selenium' do
  it 'uses eight components' do
    driver = Selenium::WebDriver.for :chrome

    driver.get('https://www.selenium.dev/selenium/web/web-form.html')

    title = driver.title
    expect(title).to eq('Web form')

    driver.manage.timeouts.implicit_wait = 500

    text_box = driver.find_element(name: 'my-text')
    submit_button = driver.find_element(tag_name: 'button')

    text_box.send_keys('Selenium')
    submit_button.click

    message = driver.find_element(id: 'message')
    value = message.text
    expect(value).to eq('Received!')

    driver.quit
  end
end

View full example on GitHub

const {By, Builder} = require('selenium-webdriver');
const assert = require("assert");

  describe('First script', function () {
    let driver;
    
    before(async function () {
      driver = await new Builder().forBrowser('chrome').build();
    });
    
    it('First Selenium script with mocha', async function () {
      await driver.get('https://www.selenium.dev/selenium/web/web-form.html');
      
      let title = await driver.getTitle();
      assert.equal("Web form", title);
      
      await driver.manage().setTimeouts({implicit: 500});
      
      let textBox = await driver.findElement(By.name('my-text'));
      let submitButton = await driver.findElement(By.css('button'));
      
      await textBox.sendKeys('Selenium');
      await submitButton.click();
      
      let message = await driver.findElement(By.id('message'));
      let value = await message.getText();
      assert.equal("Received!", value);
    });
  
    after(async () => await driver.quit());
  });

View full example on GitHub

Next Steps

Take what you’ve learned and build out your Selenium code!

As you find more functionality that you need, read up on the rest of our WebDriver documentation.

Selenium Level Sponsors

Support the Selenium Project

Want to support the Selenium project? Learn more or view the full list of sponsors.