selenium chrome 驱动程序的 getText() 方法有时会返回一个空字符串

Question

I have a curious case where the selenium chrome driver getText() method (java) returns an empty string for some elements, even though it returns a non-empty string for other elements with the same xpath . Here is a bit of the page.

<div __gwt_cell="cell-gwt-uid-223" style="outline-style:none;">
<div>Text_1</div>
<div>Text_2</div>
<div>Text_3</div>
<div>Text_4</div>
<div>Text_5</div>
<div>Text_6</div>
</div>

for each of the inner tags, I can get valid return values for getTagName() , getLocation() , isEnabled() , and isDisplayed() . However, getText() returns an empty string for some of the divs.

Further, I notice that if I use the mac chrome driver, it is consistently the 'Text_5' for which getText() returns an empty string. If I use the windows chrome driver, it is , it is consistently the 'Text_2' for which getText() returns an empty string. If I use the firefox driver, getText() returns the expected text from all the divs.

Has anyone else had this difficulty?

In my code, I use something like this…

ArrayList<WebElement> list = (ArrayList<WebElement>) driver.findElements(By.xpath(“my xPath here”));
for (WebElement e: list) System.out.println(e.getText());

As suggested below, here is the actual xPath I am using. The page snippet above deals with the last two divs.

//*[@class='gwt-DialogBox']//tr[contains(@class,'data-grid-table-row')]//td[contains(@class,'lms-assignment-selection-wizard-cell')]/div/div

Answer 1

Update: The textContent attribute is a better option and supported across the majority of browsers. The differences are explained in detail at this blog post:innerText vs. textContent

As an alternative, the innerText attribute will return the text content of an element which exists in the DOM.

element.getAttribute("innerText")

The isDisplayed() method can sometimes trip over when the element is not really hidden but outside the viewport; getText() returns an empty string for such an element.

You can also bring the element into the viewport by scrolling to it using javascript, as follows:

((JavaScriptExecutor)driver).executeScript("arguments[0].scrollIntoView(true);", element);

and then getText() should return the correct value.

Details on the isDisplayed() method can be found in this SO question:

How does Selenium WebDriver's isDisplayed() method work

Answer 2

WebElement.getAttribute("value") 应该可以帮助你！！

Answer 3

This is not a solution, so I don't know if it belongs in an answer, but it's too long for a comment and includes links, so I'm putting it an answer.

I have had this issue as well. After doing some digging, it seems that the problem arises when trying to get the text of an element that is not visible on the screen.(As @Faiz comments above.)This can happen if the element is not scrolled to, or if you scroll down and the element is near the top of the document and no longer visible after the scroll. I see you have a FindElements() call that gets a list of elements. At least some are probably not visible; you can check this by trying boolean b = webElement.isDisplayed(); on each element in the list and checking the result. (See here for a very long discussion of this issue that's a year old and still no resolution.)

Apparently, this is a deliberate design decision (see here ); gettext on invisible elements is supposed to return empty. Why they are so firm about this, I don't know. Various workarounds have been suggested, including clicking on the element before getting its text or scrolling to it. (See above link for example code for the latter.) I can't vouch for these because I haven't tried them, but they're just trying to bring the element into visiblity so the text will be available. Not sure how practical that is for your application; it wasn't for mine. For some reason, FirefoxDriver does not have this issue, so that's what I use.

I'm sorry I can't give you a better answer - perhaps if you submit a bug report on the issues page they'll see that many people find it to be a bug rather than a feature and they'll change the functionality.

Good luck! bsg

EDIT

See this question for a possible workaround. You won't be able to use it exactly as given if isDisplayed returns true, but if you know which element is causing the issue, or if the text is not normally blank and you can set an 'if string is empty' condition to catch it when it happens, you can still try it. It doesn't work for everyone, unfortunately.

NEW UPDATE I just tried the answer given below and it worked for me. So thanks, Faiz!

Answer 4

for (int count=0;count<=sizeofdd;count++)
{       
   String GetInnerHTML=getddvalue.get(count).getAttribute("innerHTML");
}

where, 1. getddvalue is the WebElement 2. sizeofdd is the size of getddvalue

Answer 5

element.getAttribute("innerText")为我工作，当getText()返回空时。

Answer 6

I encountered a similar issue recently.

I had to check that the menu tab "LIFE EVENTS" was present in the scroll box. The problem is that there are many menu tabs and you are required to scroll down to see the rest of the menu tabs. So my initial solution worked fine with the visible menu tabs but not the ones that were out of sight.

I used the xpath below to point selenium to the parent element of the entire scroll box.

@FindBy(xpath = "//div[contains(@class, 'menu-tree')]")
protected WebElement menuTree;

I then created a list of WebElements that I could increment through. The solution worked if the menu tab was visible, and returned a true. But if the menu tab was out of sight, it returned false

public boolean menuTabPresent(String theMenuTab) {
    List<WebElement> menuTabs = new ArrayList<WebElement>();
    menuTabs = menuTree.findElements(By.xpath("//i/following-sibling::span"));

    for(WebElement e: menuTabs) {
        System.out.println(e.getText());
        if(e.getText().contains(theMenuTab)) {
            return true;
        }
    }
    return false;
}

I found 2 solutions to the problem which both work equally well.

    for(WebElement e: menuTabs) {
        scrollElementIntoView(e); //Solution 1
        System.out.println(e.getAttribute("textContent")); //Solution 2
        if(e.getAttribute("textContent").contains(theMenuTab)) {
            return true;
        }
    }
    return false;

Solution 1 calls the method below. It results in the scroll box to physically move down while selenium is running.

protected void scrollElementIntoView(WebElement element) {
    ((JavascriptExecutor) driver).executeScript("arguments[0].scrollIntoView(true)", element);
}

Solution 2 gets the text content (even for the menu tabs not currently visible) of the attribute that you are pointing to. Thus doing the job properly that .getText() was not able to do in this situation.

Answer 7

if you don't care about isDisplayed or scrolling position, you can also write

String text = ((JavaScriptExecutor)driver).executeScript("return $(arguments[0]).text();", element);

or without jquery

String text = ((JavaScriptExecutor)driver).executeScript("return arguments[0].innerText;", element);

Answer 8

Related to getText() I have also an issue and I resolved so:

WebElement errMsg;
errMsg = driver.findElement(By.xpath("//div[@id='mbr-login-error']"));
WebElement parent = driver.findElement(By.xpath("//form[@id='mbr-login-form']"));
List<WebElement> children = parent.findElements(By.tagName("div")); 
System.out.println("Size is: "+children.size());
//((JavascriptExecutor)driver).executeScript("arguments[0].scrollIntoView(true);", children);
for(int i = 0;i<children.size();i++)
{
    System.out.println(i + " " + children.get(i).getText());
}
int indexErr = children.indexOf(errMsg);
System.out.println("index " + indexErr);
Assert.assertEquals(expected, children.get(indexErr).getText());

None of the above solutions worked for me.

Answer 9

Worked for me:

add as a predicate of xpath the length of string greater than 0:

String text = wait.until(ExpectedConditions.presenceOfElementLocated(By.xpath("//span[string-length(text()) > 0]"))).getText();

Answer 10

Mine is python , but the core logic is similar:

• webElement.text
• webElement.get_attribute("innerText")
• webElement.get_attribute("textContent")

Full code:

def getText(curElement):
    """
    Get Selenium element text

    Args:
        curElement (WebElement): selenium web element
    Returns:
        str
    Raises:
    """
    # # for debug
    # elementHtml = curElement.get_attribute("innerHTML")
    # print("elementHtml=%s" % elementHtml)

    elementText = curElement.text # sometime NOT work

    if not elementText:
        elementText = curElement.get_attribute("innerText")

    if not elementText:
        elementText = curElement.get_attribute("textContent")

    # print("elementText=%s" % elementText)
    return elementText

Calll it:

curTitle = getText(h2AElement)

hope is useful for you.