ChromeDriver - selenium - Information about the Chrome Driver - Browser automation framework - Google Project Hosting

ChromeDriver - selenium - Information about the Chrome Driver - Browser automation framework - Google Project Hosting

ChromeDriver

Information about the Chrome Driver

 

WebDriver

Updated
Today (8 hours ago)

 

by adamgouc...@gmail.com

Chrome Driver

Developed in collaboration with the Chromium team, the ChromeDriver is a standalone server which implements WebDriver's wire protocol.

The ChromeDriver consists of three separate pieces. There is the browser itself ("chrome"), the language bindings provided by the Selenium project ("the driver") and an executable downloaded from the Chromium project which acts as a bridge between "chrome" and the "driver". This executable is called "chromedriver", but we'll try and refer to it as the "server" in this page to reduce confusion.

Requirements

The ChromeDriver controls the browser using Chrome's automation proxy framework.

The server expects you to have Chrome installed in the default location for each system:

OS	Expected Location of Chrome
Linux	/usr/bin/google-chrome1
Mac	/Applications/Google\ Chrome.app/Contents/MacOS/Google\ Chrome
Windows XP	%HOMEPATH%\Local Settings\Application Data\Google\Chrome\Application\chrome.exe
Windows Vista	C:\Users\%USERNAME%\AppData\Local\Google\Chrome\Application\chrome.exe

¹ For Linux systems, the ChromeDriver expects /usr/bin/google-chrome to be a symlink to the actual Chrome binary. See also the section on overriding the Chrome binary location .

Getting Started

To get set up, first download the appropriate prebuilt server. Make sure the server can be located on your PATH or specify its location via the webdriver.chrome.driver system property. Finally, all you need to do is create a new ChromeDriver instance:

WebDriver driver = new ChromeDriver();
driver.get("http://www.google.com");

Running the server in a child process

You may notice that the ChromeDriver class is merely a convenience class that starts the server upon creation and shuts it down when you call quit. While the server is light weight, starting and stopping it multiple times will add a noticeable delay to a larger test suite. To compensate for this, you can directly control the life and death of the server using the ChromeDriverService:

import static org.junit.Assert.assertEquals;

import org.junit.After;
import org.junit.AfterClass;
import org.junit.Before;
import org.junit.BeforeClass;
import org.junit.runner.RunWith;
import org.junit.runners.BlockJUnit4ClassRunner
import org.openqa.selenium.chrome.ChromeDriverService;
import org.openqa.selenium.remote.DesiredCapabilities;
import org.openqa.selenium.remote.RemoteWebDriver;

@RunWith(BlockJUnit4ClassRunner.class)}
public class ChromeTest extends TestCase {

  private static ChromeDriverService service;
  private WebDriver driver;

  @BeforeClass
  public static void createAndStartService() {
    service = new ChromeDriverService.Builder()
        .usingChromeDriverExecutable(new File("path/to/my/chromedriver"))
        .usingAnyFreePort()
        .build();
    service.start();
  }

  @AfterClass
  public static void createAndStopService() {
    service.stop();
  }

  @Before
  public void createDriver() {
    driver = new RemoteWebDriver(service.getUrl(),
        DesiredCapabilities.chrome());
  }

  @After
  public void quitDriver() {
    driver.quit();
  }

  @Test
  public void testGoogleSearch() {
    driver.get("http://www.google.com");
    WebElement searchBox = driver.findElement(By.name("q"));
    searchBox.sendKeys("webdriver");
    searchBox.quit();
    assertEquals("webdriver - Google Search", driver.getTitle());
  }
}

Running the server as a standalone process

Since the ChromeDriver implements the wire protocol, it is fully compatible with any RemoteWebDriver client. Simply start up your server, create a client, and away you go:

WebDriver driver = new RemoteWebDriver("http://localhost:9515", DesiredCapabilities.chrome());
driver.get("http://www.google.com");

Advanced Usage

Starting Chromium with Specific Flags

The ChromeDriver can be made to start the browser with specific command line flags using the chrome.switches capability key; this key should define a list of command line flags that should be passed to the browser on start-up. For example, to start Chrome as a maximized window:

  DesiredCapabilities capabilities = DesiredCapabilities.chrome();
  capabilities.setCapability("chrome.switches", Arrays.asList("--start-maximized"));
  WebDriver driver = new ChromeDriver(capabilities);

Similarly, to load an extension when Chrome starts:

  DesiredCapabilities capabilities = DesiredCapabilities.chrome();
  capabilities.setCapability("chrome.switches", Arrays.asList("--load-extension=/path/to/extension/directory"));
  WebDriver driver = new ChromeDriver(capabilities);

Or to load with a specific profile (note that the default profile directories can be found here):

  DesiredCapabilities capabilities = DesiredCapabilities.chrome();
  capabilities.setCapability("chrome.switches", Arrays.asList("--user-data-dir=/path/to/profile/directory"));
  WebDriver driver = new ChromeDriver(capabilities);

The full list of flags can be found here.

Setting Chrome's proxy configuration

Since ChromeDriver 18.0.995.0, WebDriver's Proxy Desired Capability can be used. If you're using a previous version. The following can be used:

  DesiredCapabilities capabilities = DesiredCapabilities.chrome();
  capabilities.setCapability("chrome.switches", Arrays.asList("--proxy-server=http://your-proxy-domain:4443"));
  WebDriver driver = new ChromeDriver(capabilities);

Overriding the Chrome binary location

You can specify the location of the Chrome binary by passing the "chrome.binary" capability, e.g. with a typical Chromium install on Debian:

  DesiredCapabilities capabilities = DesiredCapabilities.chrome();
  capabilities.setCapability("chrome.binary", "/usr/lib/chromium-browser/chromium-browser");
  WebDriver driver = new ChromeDriver(capabilities);

Or in Ruby:

  Selenium::WebDriver::Chrome.path = "/usr/lib/chromium-browser/chromium-browser"
  driver = Selenium::WebDriver.for :chrome

Chrome Extensions

The ‘chrome.extensions’ capability can be used to specify a list of extensions, each in the form of a base64 encoded crx file, to install on startup.

Troubleshooting

If you are using the Remote WebDriver and you get the The path to the chromedriver executable must be set by the webdriver.chrome.driver system property error message you likely need to check that one of these conditions is met:

• The chromedriver binary is in the system path, or
• The Selenium Server was started with -Dwebdriver.chrome.driver=c:\path\to\your\chromedriver.exe

Known Issues

There are a handful of known issues with ChromeDriver, listed below:

1. Can only retrieve the name and value of set cookies (no domain, path, etc.)
2. Typing does not work with rich-text enabled documents
3. Cannot specify a custom profile
4. HTML 5 API not implemented
5. Advanced user interactions API not implemented