Selenium WebDriver in C#: How to use the existing window of Chrome Browser.

Selenium WebDriver with all its advantages has some pain points. One of them is to reuse the existing window of your opened browser so that you can continue your test cases from where they are failed. Unfortunately there is no direct method to use the browser window. You have to be a little creative to come up with a solution.

Luckily, after spending hours of internet search and trial and error, I am finally able to devise a solution for this problem.

I have also posted a solution about re-using the existing window of Firefox here.

This solution is specifically for chrome browser. I have also attached the source code.

The steps are as follows.

1. Create a class which is derived from RemoteWebDriver. We name this class as CustomRemoteDriver.cs.
2. Override the constructor and Execute() method in the new class like this .

 public class CustomRemoteWebDriver : RemoteWebDriver
    {
        public static bool newSession = false;
        public static string capPath = @"c:\automation\sessionCap";
        public static string sessiondIdPath = @"c:\automation\sessionid";

        public CustomRemoteWebDriver(Uri remoteAddress, DesiredCapabilities desiredCapabilities)
            : base(remoteAddress, desiredCapabilities)
        {


        }




        /// 

        /// Store for the name property.
        /// 
        /// A  value representing the command to execute.
        /// A  containing the names and values of the parameters of the command.
        /// 
        /// A  containing information about the success or failure of the command and any data returned by the command.
        /// 
        protected override Response Execute(string driverCommandToExecute, Dictionary < string, object > parameters)
        {
            if (driverCommandToExecute == DriverCommand.NewSession)
            {
                if (!newSession)
                {
                   
                    var sidText = File.ReadAllText(sessiondIdPath);

                  
                    return new Response
                    {
                        SessionId = sidText,
                        
                    };
                }
                else
                {
                    var response = base.Execute(driverCommandToExecute, parameters);
                  
                    File.WriteAllText(sessiondIdPath, response.SessionId);
                    return response;
                }
            }
            else
            {
                var response = base.Execute(driverCommandToExecute, parameters);
                return response;
            }
        }
    }

This above code is just basically saving the session id in a file when new session is created and in case this is not a new session, it is reading the session id  from the saved file and returning the response based on this session id.

1. In your [TestInitialize] method, initialize your CustomRemoteWebDriver. Start the chromedriver with Process.Start(). For every test case, check if process is already started. If yes, then Attach the custom remote webdriver to this process, otherwise start the new process. Like this.

  [TestInitialize]
        public void MyTestInitialize()
        {
            if (ConfigurationManager.AppSettings["UseChromeDebugging"] == "true")
            {

                var pr = Process.GetProcessesByName("chromedriver");
                if (pr.Length > 0)
                {
                    CustomRemoteWebDriver.newSession = false;
                    var driver = new CustomRemoteWebDriver(new Uri("http://localhost:9515"), DesiredCapabilities.Chrome());
                    WebDriver = driver;
                }

                else
                {
                    Process.Start(ConfigurationManager.AppSettings["ChromeDriverPath"]);
                    CustomRemoteWebDriver.newSession = true;
                    var driver = new CustomRemoteWebDriver(new Uri("http://localhost:9515"), DesiredCapabilities.Chrome());
                    WebDriver = driver;

                }
            }

            else
            {
                //normal execution
                WebDriver = new ChromeDriver(DRIVER_PATH);                
            }
            
            
        }

• You are good to go. Now create two test methods. In the first method navigate to google.com and in the second method, search something in google search box.

• Build the solution, in the test view window, you will be able to see two test methods

• Execute the first method, chrome driver command prompt window will open,  a chrome window will open and it will navigate to google.com. Test case will be passed.

• Now the real magic will happen, Execute the second method, It will use the existing window of chrome, No new window will open and google will search the given text.

Full source code will be attached soon. Please do give your feedback. I hope I can make lives of some test automation engineers easier with this approach :)