Selenium - MoveTargetOutOfBoundsException with Firefox

Question

I'v got problem with function move_to_element on Firefox Webdriver (Chrome, IE works well)

driver = webdriver.Firefox()
driver.get("https://stackoverflow.com")
time.sleep(5)
source_element = driver.find_element_by_xpath('//*[@id="footer"]/div/ul/li[1]/a')
ActionChains(driver).move_to_element(source_element).perform()

I am working with these versions: geckodriver - 0.17.0 // Firefox - 54.0 // selenium - 3.4.3

After running this script, on output shows:

selenium.common.exceptions.MoveTargetOutOfBoundsException: Message: (134.96666717529297, 8682.183013916016) is out of bounds of viewport width (1268) and height (854)

Cynic · Answer 1 · 2018-08-29T22:23:30.423

34

I think the correct answer here got lucky that the element they were looking for happened to be at the bottom of the page and didn't really explain why this occurs in Firefox commonly.

Browsers other than Firefox treat Webdrivers move_to_element action as scroll to part of page with element then hover over it. Firefox seems to have taken a hardline stance that move_to_element is just hover and are waiting for a scroll action to fix this.

For now you have to workaround this bug using javascript as mentioned in previous answer, but I suggest using something like this instead of arbitrarily (well I guess the example was a footer) scrolling to bottom of page and hoping object is still in view.

    def scroll_shim(passed_in_driver, object):
        x = object.location['x']
        y = object.location['y']
        scroll_by_coord = 'window.scrollTo(%s,%s);' % (
            x,
            y
        )
        scroll_nav_out_of_way = 'window.scrollBy(0, -120);'
        passed_in_driver.execute_script(scroll_by_coord)
        passed_in_driver.execute_script(scroll_nav_out_of_way)

Then later

source_element = driver.find_element_by_xpath('//*[@id="footer"]/div/ul/li[1]/a')
if 'firefox' in driver.capabilities['browserName']:
    scroll_shim(driver, source_element)
# scroll_shim is just scrolling it into view, you still need to hover over it to click using an action chain.
actions = ActionChains(driver)
actions.move_to_element(source_element)
actions.click()
actions.perform()

edited Aug 29 '18 at 22:23

answered Aug 27 '18 at 18:40

Cynic

6,779
2
30
49

Oh man, they might finally fix this: https://www.w3.org/2018/10/26-webdriver-minutes.html#resolution01 – Cynic Nov 07 '18 at 16:22
I tried using your function, but it isn't scrolling at all. I then get the expected MoveTargetOutOfBoundsException when I try to access the element. Do I need to change anything for Python 3? – Eliezer Miron Dec 19 '18 at 20:01
You need to pass in the element and driver but it should just work. What browser are you using? Also try doing something like `import time` and `time.sleep(3)` after you call it so you can see it scroll – Cynic Dec 19 '18 at 20:13
I'm using Firefox. I'll try the sleeps. EDIT: Sleeps didn't help. I'm passing it the driver and element, and I know the function is running, it just doesn't scroll at all. I'm in a unique position where only part of the frame is scrollable, so I'm clicking in that area before trying this function. I guess that part may be failing. – Eliezer Miron Dec 19 '18 at 20:46
3

I ended up using and it actually worked fine. Not sure why that worked but your function didn't. – Eliezer Miron Dec 19 '18 at 20:53
Yeah, the exception is happening when you click. So if you don't use the function to scroll to it before you click you would get the exception. The function works fine in Firefox. Asked about browser as I haven't tested it IE but it's basic javascript so I would expect it should work. – Cynic Dec 19 '18 at 23:30
I thought your scroll_shim function was supposed to the scrolling? – Eliezer Miron Dec 20 '18 at 18:37
It does but you have to scroll before you click. So it wasn't working because you were hitting the error before you were calling the function. I am using this with Firefox and it works. – Cynic Dec 20 '18 at 20:37
I had a time.sleep between scroll_shim and the click call. It's probably just related to the weirdness of the HTML I was working with, so I wouldn't worry about it. – Eliezer Miron Dec 20 '18 at 22:51
@Cynic I see that w3 has updated some documents on this here: https://drafts.csswg.org/cssom-view/#dom-element-scrollintoview Do you know if there is any implementation of this yet? I've got a bunch of code for testing in Chrome that I was trying to use for FireFox, and I don't want to rewrite the existing code if I can help it. – Erin B Jul 29 '19 at 20:02
1

Eliezer Miron in a previous comment mentioned how you could use `driver.execute_script("arguments[0].scrollIntoView();` instead. But as far as them fixing it in selenium, I doubt anything would happen before Selenium 4, even with that, it doesn't sound like Firefox would make `move_to_element` consistent. We'd probably just have to add something like `scroll_to_element` to all our action chains to support all browsers. So the fix would likely be similar to creating your own scroll shim and pasting it in before your clicks. – Cynic Jul 29 '19 at 22:59
works for me, thank you! – Kiki Du Jul 12 '23 at 22:20

undetected Selenium · Accepted Answer · 2019-08-12T11:25:22.260

This error...

selenium.common.exceptions.MoveTargetOutOfBoundsException: Message: (134.96666717529297, 8682.183013916016) is out of bounds of Viewport width (1268) and height (854)

...implies that the element you are looking for is not within the Viewport. We need to scroll down to bring the element within the Viewport. Here is the working code:

from selenium import webdriver
from selenium.webdriver.firefox.firefox_binary import FirefoxBinary
from selenium.webdriver.common.desired_capabilities import DesiredCapabilities
from selenium.webdriver.common.action_chains import ActionChains

binary = FirefoxBinary('C:\\Program Files\\Mozilla Firefox\\firefox.exe')
caps = DesiredCapabilities().FIREFOX
caps["marionette"] = True
driver = webdriver.Firefox(capabilities=caps, firefox_binary=binary, executable_path="C:\\Utility\\BrowserDrivers\\geckodriver.exe")
driver.get("https://stackoverflow.com")
last_height = driver.execute_script("return document.body.scrollHeight")
driver.execute_script("window.scrollTo(0, document.body.scrollHeight);")
source_element = driver.find_element_by_xpath('//*[@id="footer"]/div/ul/li[1]/a')
ActionChains(driver).move_to_element(source_element).perform()

Let me know if this Answers your Question.

Adding a simple element.click() before the actions, worked for me :) — Nitin, Sep 15 '20 at 06:03

score 5 · Answer 3 · answered Aug 05 '19 at 08:24

You can try below while automating the script in Firefox when it usually throws MoveTargetOutOfBoundsException error:

One can transform/Zoom-in or out by

driver.execute_script("document.body.style.transform='scale(0.9)';")

Sometimes if you are running automation script in Jenkins(CI tools), you might also face the issue from above transform code where content of browser is scaled out not the actual browser, in those condition you can try out to resize the browser window:

driver.set_window_size(x, y)

or

driver.set_window_size(2000, 694)

score 1 · Answer 4 · answered Jun 12 '20 at 06:45

1

time.sleep() command can be used sometimes when an elements is taking time to switch.Scrolling always doesn't help you.

answered Jun 12 '20 at 06:45

Lokesh Murge

11
2

score 1 · Answer 5 · edited Jul 07 '22 at 14:50

I am using this to avoid getting MoveTargetOutOfBoundsException with running Firefox. (in C#):

public void ScrollToElement(IWebElement elementLocator)
{

    try
    {
        action.MoveToElement(elementLocator).Perform();
    }

    catch(MoveTargetOutOfBoundsException e)
    {
        Console.WriteLine(e);
        js.ExecuteScript("arguments[0].scrollIntoView(true);", elementLocator);
    }

}

In Python it would probably be like this:

def scroll_to_element(driver, element_locator):
    actions = ActionChains(driver)
    try:
        actions.move_to_element(element_locator).perform()
    except MoveTargetOutOfBoundsException as e:
        print(e)
        driver.execute_script("arguments[0].scrollIntoView(true);", element_locator)

score 1 · Answer 6 · answered Aug 15 '22 at 21:51

Note that if you're encountering this problem while using move_to_element to trigger an event like a mouseover or mouseenter, you can skip using move_to_element entirely.

You can instead use execute_script to call the event directly:

driver = webdriver.Firefox()
driver.get("https://stackoverflow.com")
source_element = driver.find_element_by_xpath('//*[@id="footer"]/div/ul/li[1]/a')
driver.execute_script("arguments[0].dispatchEvent(new Event('mouseover'));", source_element)

It works by using javascript to dispatch the event, and it will work even if the element isn't in the viewport.
See also the dispatchEvent docs for more detail.

score 0 · Answer 7 · answered Jul 23 '20 at 11:24

This is not a generic solution, but might help ring a bell. Thanks to the inputs in this thread, I could resolve the issue by tweaking the target page.

The exception, in my case, occurred :

while checking for existence of an element that happened to be at the top of the page,
while the page, when rendered on the browser, was getting auto-scrolled to the bottom due to "autofocus" that was set on a button element at the bottom of the page.

Because the test script was against my own development, I had the flexibility to review the need for the "autofocus" and avoid the auto-scroll to the bottom of the page when the page got rendered.

Alternatively, checking for existence of a different element, that gets rendered within the viewport, if feasible, could also resolve the issue.

oguzk · Answer 8 · 2022-07-25T08:49:37.533

You can also use the code below. I used it, the program works fine

lenOfPage = driver.execute_script("window.scrollTo(0, document.body.scrollHeight);var lenOfPage=document.body.scrollHeight;return lenOfPage;")
match=False
while(match==False):
    lastCount = lenOfPage
    time.sleep(3)
    lenOfPage = driver.execute_script("window.scrollTo(0, document.body.scrollHeight);var lenOfPage=document.body.scrollHeight;return lenOfPage;")
    if lastCount == lenOfPage:
        match=True
time.sleep(5)

Then later

actions = ActionChains(driver)
actions.move_to_element(source_element)
actions.click()
actions.perform()

Besides you can add your code:

from selenium.common.exceptions import StaleElementReferenceException, MoveTargetOutOfBoundsException

and use try except script.

For example:

try:
    'Main Code'
            
except (StaleElementReferenceException, MoveTargetOutOfBoundsException):
    pass

score -3 · Answer 9 · edited Apr 05 '20 at 13:49

-3

I was searching for the solution for last 2 hours and nothing was working. Then I just gave a sleep time time.sleep(5) after the window opened on which action is to be performed and it started working.

edited Apr 05 '20 at 13:49

colidyre

4,170
12
37
53

answered Apr 05 '20 at 12:37

Babita

9

1

That's just bad solution. – Mate Ajdukovic May 18 '20 at 13:50
1

Should be wrapped with WebDriverWait. – m3nda Feb 02 '21 at 12:22

Selenium - MoveTargetOutOfBoundsException with Firefox

9 Answers9

Linked

Related