This is the multi-page printable view of this section. Click here to print.

Return to the regular view of this page.

Chrome DevTools Protocol

Examples of working with Chrome DevTools Protocol in Selenium. CDP support is temporary until WebDriver BiDi has been implemented.

Many browsers provide “DevTools” – a set of tools that are integrated with the browser that developers can use to debug web apps and explore the performance of their pages. Google Chrome’s DevTools make use of a protocol called the Chrome DevTools Protocol (or “CDP” for short). As the name suggests, this is not designed for testing, nor to have a stable API, so functionality is highly dependent on the version of the browser.

Selenium is working to implement a standards-based, cross-browser, stable alternative to CDP called [WebDriver BiDi]. Until the support for this new protocol has finished, Selenium plans to provide access to CDP features where applicable.

Using Chrome DevTools Protocol with Selenium

Chrome and Edge have a method to send basic CDP commands. This does not work for features that require bidirectional communication, and you need to know what domains to enable when and the exact names and types of domains/methods/parameters.

    Map<String, Object> cookie = new HashMap<>();
    cookie.put("name", "cheese");
    cookie.put("value", "gouda");
    cookie.put("domain", "www.selenium.dev");
    cookie.put("secure", true);
    ((HasCdp) driver).executeCdpCommand("Network.setCookie", cookie);
    cookie = {'name': 'cheese',
              'value': 'gouda',
              'domain': 'www.selenium.dev',
              'secure': True}

    driver.execute_cdp_cmd('Network.setCookie', cookie)
            var cookie = new Dictionary<string, object>
            {
                { "name", "cheese" },
                { "value", "gouda" },
                { "domain", "www.selenium.dev" },
                { "secure", true }
            };
            ((ChromeDriver)driver).ExecuteCdpCommand("Network.setCookie", cookie);
    driver.execute_cdp('Network.setCookie',
                       name: 'cheese',
                       value: 'gouda',
                       domain: 'www.selenium.dev',
                       secure: true)

To make working with CDP easier, and to provide access to the more advanced features, Selenium bindings automatically generate classes and methods for the most common domains. CDP methods and implementations can change from version to version, though, so you want to keep the version of Chrome and the version of DevTools matching. Selenium supports the 3 most recent versions of Chrome at any given time, and tries to time releases to ensure that access to the latest versions are available.

This limitation provides additional challenges for several bindings, where dynamically generated CDP support requires users to regularly update their code to reference the proper version of CDP. In some cases an idealized implementation has been created that should work for any version of CDP without the user needing to change their code, but that is not always available.

Examples of how to use CDP in your Selenium tests can be found on the following pages, but we want to call out a couple commonly cited examples that are of limited practical value.

  • Geo Location — almost all sites use the IP address to determine physical location, so setting an emulated geolocation rarely has the desired effect.
  • Overriding Device Metrics — Chrome provides a great API for setting Mobile Emulation in the Options classes, which is generally superior to attempting to do this with CDP.

1 - Chrome DevTools Logging Features

Logging features using CDP.

While Selenium 4 provides direct access to the Chrome DevTools Protocol, these methods will eventually be removed when WebDriver BiDi implemented.

Console Logs

    ((HasLogEvents) driver).onLogEvent(consoleEvent(e -> messages.add(e.getMessages().get(0))));
    async with driver.bidi_connection() as session:
        async with Log(driver, session).add_listener(Console.ALL) as messages:
            using IJavaScriptEngine monitor = new JavaScriptEngine(driver);
            var messages = new List<string>();
            monitor.JavaScriptConsoleApiCalled += (_, e) =>
            {
                messages.Add(e.MessageContent);
            };
            await monitor.StartEventMonitoring();
    driver.on_log_event(:console) { |log| logs << log.args.first }

JavaScript Exceptions

    async with driver.bidi_connection() as session:
        async with Log(driver, session).add_js_error_listener() as messages:
            using IJavaScriptEngine monitor = new JavaScriptEngine(driver);
            var messages = new List<string>();
            monitor.JavaScriptExceptionThrown += (_, e) =>
            {
                messages.Add(e.Message);
            };
            await monitor.StartEventMonitoring();
    driver.on_log_event(:exception) { |exception| exceptions << exception }

2 - Chrome DevTools Network Features

Network features using CDP.

While Selenium 4 provides direct access to the Chrome DevTools Protocol, these methods will eventually be removed when WebDriver BiDi implemented.

Basic authentication

Some applications make use of browser authentication to secure pages. It used to be common to handle them in the URL, but browsers stopped supporting this. With this code you can insert the credentials into the header when necessary

    Predicate<URI> uriPredicate = uri -> uri.toString().contains("herokuapp.com");
    Supplier<Credentials> authentication = UsernameAndPassword.of("admin", "admin");
    ((HasAuthentication) driver).register(uriPredicate, authentication);
        credentials = base64.b64encode("admin:admin".encode()).decode()
        auth = {'authorization': 'Basic ' + credentials}
        await connection.session.execute(connection.devtools.network.set_extra_http_headers(Headers(auth)))
            var handler = new NetworkAuthenticationHandler()
            {
                UriMatcher = uri => uri.AbsoluteUri.Contains("herokuapp"),
                Credentials = new PasswordCredentials("admin", "admin")
            };
            var networkInterceptor = driver.Manage().Network;
            networkInterceptor.AddAuthenticationHandler(handler);
            await networkInterceptor.StartMonitoring();
    driver.register(username: 'admin',
                    password: 'admin',
                    uri: /herokuapp/)

Network Interception

Both requests and responses can be recorded or transformed.

Response information

    try (NetworkInterceptor ignored =
        new NetworkInterceptor(
            driver,
            (Filter)
                next ->
                    req -> {
                      HttpResponse res = next.execute(req);
                      contentType.add(res.getHeader("Content-Type"));
                      return res;
                    })) {
            INetwork networkInterceptor = driver.Manage().Network;
            networkInterceptor.NetworkResponseReceived += (_, e)  =>
            {
                contentType.Add(e.ResponseHeaders["content-type"]);
            };
            await networkInterceptor.StartMonitoring();
    driver.intercept do |request, &continue|
      continue.call(request) do |response|
        content_type << response.headers['content-type']
      end
    end

Response transformation

    try (NetworkInterceptor ignored =
        new NetworkInterceptor(
            driver,
            Route.matching(req -> true)
                .to(
                    () ->
                        req ->
                            new HttpResponse()
                                .setStatus(200)
                                .addHeader("Content-Type", MediaType.HTML_UTF_8.toString())
                                .setContent(Contents.utf8String("Creamy, delicious cheese!"))))) {
            var handler = new NetworkResponseHandler()
            {
                ResponseMatcher = _ => true,
                ResponseTransformer = _ => new HttpResponseData
                {
                    StatusCode = 200,
                    Body = "Creamy, delicious cheese!"
                }
            };
            INetwork networkInterceptor = driver.Manage().Network;
            networkInterceptor.AddResponseHandler(handler);
            await networkInterceptor.StartMonitoring();
    driver.intercept do |request, &continue|
      continue.call(request) do |response|
        response.body = 'Creamy, delicious cheese!' if request.url.include?('blank')
      end
    end

Request interception

    try (NetworkInterceptor ignored =
        new NetworkInterceptor(
            driver,
            (Filter)
                next ->
                    req -> {
                      if (req.getUri().contains("one.js")) {
                        req =
                            new HttpRequest(
                                HttpMethod.GET, req.getUri().replace("one.js", "two.js"));
                      }
                      completed.set(true);
                      return next.execute(req);
                    })) {
            var handler = new NetworkRequestHandler
            {
                RequestMatcher = request => request.Url.Contains("one.js"),
                RequestTransformer = request =>
                {
                    request.Url = request.Url.Replace("one", "two");

                    return request;
                }
            };
            INetwork networkInterceptor = driver.Manage().Network;
            networkInterceptor.AddRequestHandler(handler);
            await networkInterceptor.StartMonitoring();
    driver.intercept do |request, &continue|
      uri = URI(request.url)
      request.url = uri.to_s.gsub('one', 'two') if uri.path&.end_with?('one.js')
      continue.call(request)
    end

Performance Metrics

    devTools.send(Performance.enable(Optional.empty()));
    List<Metric> metricList = devTools.send(Performance.getMetrics());
    async with driver.bidi_connection() as connection:
        await connection.session.execute(connection.devtools.performance.enable())
        metric_list = await connection.session.execute(connection.devtools.performance.get_metrics())
            await domains.Performance.Enable(new OpenQA.Selenium.DevTools.V130.Performance.EnableCommandSettings());
            var metricsResponse =
                await session.SendCommand<GetMetricsCommandSettings, GetMetricsCommandResponse>(
                    new GetMetricsCommandSettings()
                );
    driver.devtools.performance.enable
    metric_list = driver.devtools.performance.get_metrics.dig('result', 'metrics')

Setting Cookies

    devTools.send(
            Network.setCookie(
                    "cheese",
                    "gouda",
                    Optional.empty(),
                    Optional.of("www.selenium.dev"),
                    Optional.empty(),
                    Optional.of(true),
                    Optional.empty(),
                    Optional.empty(),
                    Optional.empty(),
                    Optional.empty(),
                    Optional.empty(),
                    Optional.empty(),
                    Optional.empty(),
                    Optional.empty()));
    async with driver.bidi_connection() as connection:
        execution = connection.devtools.network.set_cookie(
            name="cheese",
            value="gouda",
            domain="www.selenium.dev",
            secure=True
        )
        await connection.session.execute(execution)
            var cookieCommandSettings = new SetCookieCommandSettings
            {
                Name = "cheese",
                Value = "gouda",
                Domain = "www.selenium.dev",
                Secure = true
            };
            await domains.Network.SetCookie(cookieCommandSettings);
    driver.devtools.network.set_cookie(name: 'cheese',
                                       value: 'gouda',
                                       domain: 'www.selenium.dev',
                                       secure: true)

Waiting for Downloads

    devTools.send(
            Browser.setDownloadBehavior(
                    Browser.SetDownloadBehaviorBehavior.ALLOWANDNAME,
                    Optional.empty(),
                    Optional.of(""),
                    Optional.of(true)));
    driver.devtools.browser.set_download_behavior(behavior: 'allow',
                                                  download_path: '',
                                                  events_enabled: true)

    driver.devtools.browser.on(:download_progress) do |progress|
      @completed = progress['state'] == 'completed'
    end

3 - Chrome DevTools Script Features

Script features using CDP.

While Selenium 4 provides direct access to the Chrome DevTools Protocol, these methods will eventually be removed when WebDriver BiDi implemented.

Script Pinning

    ScriptKey key = ((JavascriptExecutor) driver).pin("return arguments;");
    List<Object> arguments =
        (List<Object>) ((JavascriptExecutor) driver).executeScript(key, 1, true, element);
            var key = await new JavaScriptEngine(driver).PinScript("return arguments;");
            var arguments = ((WebDriver)driver).ExecuteScript(key, 1, true, element);
    key = driver.pin_script('return arguments;')
    arguments = driver.execute_script(key, 1, true, element)

DOM Mutation Handlers

    async with driver.bidi_connection() as session:
        async with Log(driver, session).mutation_events() as event:
            using IJavaScriptEngine monitor = new JavaScriptEngine(driver);
            monitor.DomMutated += (_, e) =>
            {
                var locator = By.CssSelector($"*[data-__webdriver_id='{e.AttributeData.TargetId}']");
                mutations.Add(driver.FindElement(locator));
            };
            await monitor.StartEventMonitoring();
            await monitor.EnableDomMutationMonitoring();
    driver.on_log_event(:mutation) { |mutation| mutations << mutation.element }