Btukfyl

I'm trying to get the text for the login error message in twitter.

HTML code for this part of Twitter:

<div class="message-inside">

    <span class="message-text">The email or password doesn't match.</span>

  </div>

I then used:

String message = driver.findElement(By.xpath("/html/body/div[2]/div/div/span")).getText();

Where the xpath is the xpath given by the inspect element of Mozilla. However, the String message is empty.

PS: At the moment getText is called, the message is not hidden, so I don't think that is the problem

EDIT: I have also tried .getAttribute("innerHTML") instead of .getText(), which also failed

Never resolve such issues with sleep: it's unreliable (what if element appeared 1ms later than sleep length?) and slows your tests (what if element appeared 10ms after sleep start? you wasted 990ms, and on many tests it adds up).

Instead, use explicit wait pattern to wait for text to appear:

WebDriverWait wait = new WebDriverWait(driver, 10); // wait for max 10 sec

String message =  wait.until(ExpectedConditions. visibilityOfElementLocated(By.xpath("/html/body/div[2]/div/div/span")).getText();

This way you tell selenium to wait for element for up to 10 seconds, but it will return faster if element appears before that.

Another thing: xpath like /html/body/div[2]/div/div/span is not something you should normally use (there are situations when it's inevitable, but rarely). Why:

• First of all from looking at the path, you can't tell what is it that you are choosing. So if HTML changes, you will have a hard time figuring out what this statement did before.


• Selection by index (e.g. div[2]) is very breakable: all it takes is for one more div to appear above the one you selected, and your xpath will break


• You rarely need to select absolute path. Concentrate on parts that matter.

For instance what you are trying to select, looks like this on Twitter:

 <div class="alert-messages" id="message-drawer" style="top: -40px;">

     <div class="message ">

         <div class="message-inside">

             <span class="message-text"></span>

         </div>

     </div>

 </div>

So a more meaningful xpath would be

//div[@id='message-drawer']//span[@class='message-text']

which tells you which part of the page you are looking at - message-drawer (and id is best to identify things - it's always unique). And which element inside that part you are interested in. This is it. If anything on page changes, you mostly won't care. And even if you do, xpath clearly tells you the purpose of your selection, so easier to understand what it meant.

Try changing the path to div[1] or removing one of the "/div"'s in the path.

Ok, so apparently the error message WAS hidden when I tried to get the text. I told Selenium to wait a little before getting the text (and after inputing a wrong login) with:

Thread.sleep(1000);

and it was able to get the text without troubles. If you're trying to get a text that was previously hidden, tell Selenium to wait a little, so you can make sure it is no longer hidden.