How to use ProxyNodeWebsockets class of org.openqa.selenium.grid.node package

Use Actions class to perform mouse, keyboard actions on WebElements using WebDriver.

Actions action = new Actions(driver);
WebElement element=driver.findElement(By.linkText("TEST"));

//Double click
action.doubleClick(element).perform();

//Mouse over
action.moveToElement(element).perform();

//Right Click
action.contextClick(element).perform();

This is a common source of XPath confusion. First the straightforward parts:

• //a selects all a elements in the document.
• //a//b selects all b elements in the document that are descendants of a elements.

Normal stuff so far. Next is the tricky part:

1. To select the last b elements among siblings (beneath a elements):

   //a//b[last()]
   

   Here, the filtering is a part of the b selection criteria because [] has a higher precedence than //.

2. To select the last b element in the document (beneath a elements):

   (//a//b)[last()]
   

   Here, the last() is an index on the list of all selected b elements because () is used to override the default precedence.

If you have the DOM parsed already into JSoup, then I would recommend using JSoup. It is much faster than selenium, since it does not need to bother with a "living" DOM. Selenium must always check if the element handles are still valid before doing any operations with them.

If you can, avoid selenium altogether, since its overhead is really noticeable when you do serious scraping. Selenium shines, however, if your content is dynamically generated by JavaScript in the client. JSoup can't handle this at all, since it does not execute JavaScript.

Addendum to answer a comment

Short answer : It depends!

Longer: If the website you are scraping is generated by JavaScript and it does not change after generation, it is perfectly fine to access it with selenium, especially, if the DOM is complex and would take long to read into JSoup, although JSoup is fairly fast. However, JSoup will generate the DOM in memory again, so if your DOM is huge you will not only have it in a memory consuming way in selenium, but also in JSoup. This may or may not be an issue in your case, but it is worth keeping in mind.

From my personal experience I would kill the selenium process as soon as possible after getting the final HTML and parse this in JSoup again, since it is as you say: Jsoup scraping is way easier than the corresponding selenium selector constructs, especially if you are sure that any changes in the DOM after the initial creation are irrelevant to your scraping.

I ran across this same problem and could not find any solutions. Came up with a solution and posting it here, hope this helps someone with the same problem. I created a class to handle stale elements depending on their type, cssselector, id, etc and simply call it like I would any other page object.

public void StaleElementHandleByID (String elementID)
{
    int count = 0;
    boolean clicked = false;
    while (count < 4 && !clicked)
    {
        try 
        {
            WebElement yourSlipperyElement= driver.findElement(By.id(elementID));
            yourSlipperyElement.click(); 
            clicked = true;
        } 
        catch (StaleElementReferenceException e)
        {
            e.toString();
            System.out.println("Trying to recover from a stale element :" + e.getMessage());
            count = count+1;
        }
    }
}

I'd recommend only using this on elements you know cause problems for WebDriver.

Chances are this problem is caused by an already-running instance of the Selenium server. The new instance needs to listen on the same port number, but can't, because the port is already in use.

Let's say your Selenium server is configured to start on port 4444. Determine if the port is in use using the 'netstat' command:

On Windows: netstat -an | find "4444"

Expect to see output like this:

  TCP    0.0.0.0:4444           0.0.0.0:0              LISTENING
  TCP    [::]:4444              [::]:0                 LISTENING

On Linux, use: netstat -anp | grep 4444

(No Linux box to hand, so can't show example output!)

If you see any output, you need to kill the process that's listening on the port that Selenium wants to use. On Windows, use netstat -anb to find the process name (it'll be listed after the line specifying the port number). Kill it using the Task Manager. On Linux, the process PID and name will be listed by the command above - kill it using kill <PID>.

Use full resource ID ...It worked for me....

below line worked for me.... driver.findElement(MobileBy.id("com.android.packageinstaller:id/permission_allow_button")).click();

java -jar seleniumjar.jar -Dwebdriver.chrome.driver=./chromedriver 

If you want to specifiy the port do

java -jar seleniumjar.version.jar -Dwebdriver.chrome.driver=./chromedriver -port 4545

As of 3.8.1, specifying the driver is no longer necessary:

java -jar seleniumjar.jar

The classpath is needed also when running the Jar file so that the JVM could find library classes. However, when using the jar attribute in the java task, "all classpath settings are ignored" (see https://ant.apache.org/manual/Tasks/java.html). The simplest way is to specify the main class and add the compiled Jar along with the library Jars located in the RunningPowerJars directory:

<target name="run" description="Run the jar file" depends="jar" >
    <java classname="RunningPower" fork="true">
       <classpath>
            <pathelement location="build/jar/RunningPower.jar"/>
            <fileset dir="${lib.dir}">
                <include name="**/*.jar" />
            </fileset>
        </classpath>
   </java>
</target>

An unrelated note is that you could use the properties defined in init throughout the buildfile. For example, in the jar target you can use ${classes.dir} instead of repeating build/class.

It still doesn't make sense why the build would fail if the classpath element is added inside the java task when using the jar attribute. Although it is ignored, it is weird that it is failing with the "failed to create task or type" error.

You could essentially turn the body of the loop to a lambda expression and collect the the results

List<String> testing = 
    measureKeys.stream()
               .map(e -> element.getText().split("\\(")[0])
               .collect(Collectors.toList());