How to use RetryRequest class of org.openqa.selenium.remote.http package

This error message...

An internal error occurred during: "Launching NewTest"
org.testng.eclipse.maven.MavenTestNGLaunchConfigurationProvider.getClasspath(Lorg/eclipse/debug/core/ILaunchConfiguration;)Ljava/util/List;

...implies that there was incompatibility between TestNG and Maven related classfiles.

Historically, this error would occur if while installing testng you have kept the checkbox for M2E Maven as checked.

The easiest approach to address this issue would be to reinstall the TestNG ensuring that the checkbox for M2E Maven is kept unchecked.

Create cookies using the Java API as follows:

Cookie ck = new Cookie("name", "value");
driver.manage().addCookie(ck);

Create cookies using the Python API as follows:

driver.add_cookie({'name': 'foo', 'value': 'bar'})

Use the xpath as shown below:

driver.findElements(By.xpath("//input[contains(@text,\"WE'd\")]"));

Hope this helps.

You don't have your XPath syntax right. You need quotes round the text attribute values you're matching against. Try:

d.findElement(By.xpath("//select[@id='category']/option[@id='cat2']")).click();

Java String(s) can't be done in single quotes. This,

By.cssSelector('[data-testing-id="data-id1"]')

Should be (escaping the double quotes),

By.cssSelector("[data-testing-id=\"data-id1\"]")

or with single quotes inside the double quotes, like

By.cssSelector("[data-testing-id='data-id1']")

You can either use the crx file of the extension or point to a directory it's installed in. Assuming you have it installed, you'll need to add ChromeOptions to your new ChromeDriver:

ChromeOptions options = new ChromeOptions();
options.addArguments("load-extension=/path/to/extension");
DesiredCapabilities capabilities = new DesiredCapabilities();
capabilities.setCapability(ChromeOptions.CAPABILITY, options);
ChromeDriver driver = new ChromeDriver(capabilities);

See here for more details and walkthrough.

EDIT: to see your extensions path, navigate to chrome://version and look at 'Profile Path'. In that folder look for 'Extensions' folder. The extension ID for AdBlock Plus is cfhdojbkjhnklbpkdaibdccddilifddb. So the path/to/extension should be something like (Windows):

C:\Users\<user>\AppData\Local\Google\Chrome\User Data\<profile>\Extensions\cfhdojbkjhnklbpkdaibdccddilifddb

I was finally able to get RSelenium to work by piecing together information from a number of different sources. I think it would be helpful to have all of this information in one location, so here is the process that I went through to get RSelenium to work on Windows 7 (64-bit) with Chrome as the browser:

1. Download the 64-bit version of Java. I could not get anything to work with the standard download.
2. Download ChromeDriver.
3. Download the Selenium Standalone Server or run checkForServer() from R.

4. Create a batch file to start the Selenium server. I initially tried to use startServer() from an R script, but it would frequently get stuck and not carry on to the next line in the script. Here is the batch file that I created:

   java -jar C:\path\to\selenium-server-standalone.jar -Dwebdriver.chrome.driver=C:\path\to\chromedriver.exe
   

   ChromeDriver can be put in the PATH environmental variable, but I decided to add in the path to ChromeDriver to the batch file (which accomplishes the same goal).

5. Run the R script. Here is my final script:

   library(RSelenium)
   shell.exec(paste0("C:\\path\\to\\yourbatchfile.bat"))
   Sys.sleep(5)
   
   remDr <- remoteDriver(browserName = "chrome")
   remDr$open(silent = TRUE)
   remDr$navigate("http://www.google.com")
   

   The Sys.sleep() call was necessary because I would get an error in the remoteDriver() call if it ran before the Selenium Server had finished starting.

There's probably a click event transmitted to the <input type="file"> element, which opens the file dialog.

One way to overcome this issue is to disable the internal click by overriding the method. This piece of JavaScript will disable the internal click for each <input type="file"> element:

((JavascriptExecutor)driver).executeScript(
    "HTMLInputElement.prototype.click = function() {                     " +
    "  if(this.type !== 'file') HTMLElement.prototype.click.call(this);  " +
    "};                                                                  " );

Then upload the file by calling .sendKeys("full file path") on the <input type="file"> element.

In some cases, the <input type="file"> only resides in memory. If that's the case, you'll have to insert it in the page:

((JavascriptExecutor)driver).executeScript(
    "HTMLInputElement.prototype.click = function () {" +
    "    if (this.type !== 'file') {" +
    "        HTMLElement.prototype.click.call(this);" +
    "    }" +
    "    else if (!this.parentNode) {" +
    "        this.style.display = 'none';" +
    "        this.ownerDocument.documentElement.appendChild(this);" +
    "        this.addEventListener('change', () => this.remove());" +
    "    }" +
    "}");

See https://gist.github.com/florentbr/d7bee068932bdf6e00af3b9f01af9af6

The answer to this was frustratingly simple. The XPATH I was using was not specific enough and returning multiple objects when I thought it was only returning one, it then was directing Seleniums attention to the first in the array which was not visible (as it correctly reporting).

Once I changed my method to return the array of matching objects and simply cycle through until I found one that was both enabled and displayed to use Robert was my mothers brother.

The dodgy XPATH issue comes from the fact that I checked the XPATH within firepath (in firefox) and this was correctly highlighting the object I wanted but selenium could somehow actually multiples of the same object (despite the page source not showing more than 1 object) and this wasn't highlighted to me until I used the firefox Selenium addin to record my actions.

So the lesson learned here is to not trust an XPATH generated via firepath 100% if it doesnt appear to be working without checking in the Selenium IDE first...!