Python, Selenium : 'Element is no longer attached to the DOM'

问题

I am scraping a website, www.lipperleaders.com. I want to extract the funds detail of Singapore. I have successfully implemented drop-down selection and extracted the content of the first page appearing after submission of the options. But when I try to go to next pages (by making the code to click next button) I am getting error 'Element is no longer attached to the DOM'.

My code is about 100 lines but I can give a general idea of the flow of execution of my code:

...                    # creating driver object and all the imports
def main():
    ...
    result = find_elements_by_tag_name('span')  
    ...
    driver.find_element_by_id("ctl00_ContentPlaceHolder1_ucDataPager_btnNext").click()
    main()
main()

This code works fine for the 1st page but when main() is called again after clicking of the next button. Before this recursive method, I also tried putting this inside a loop, then also same error.

And if I write the same code like:

# some code
result = find_elements_by_tag_name('span')  
driver.find_element_by_id("ctl00_ContentPlaceHolder1_ucDataPager_btnNext").click()
# some code
driver.find_element_by_id("ctl00_ContentPlaceHolder1_ucDataPager_btnNext").click()
.
.

This code works fine w/o any error the next page loads and executes the code written after that. But I cannot write the same driver.find_element_by_id().click() for 500 pages, even I will have to repeat the rest of the code associated with each page. That's why I am trying for loop or recursion, but its not working for me.

Please let me know what is the problem with my approach.

回答1:

The problem is that the element is being detached by some javascript. So you should make the driver wait for the element: This is done by setting implicitly_wait, see:

from selenium import webdriver

ff = webdriver.Firefox()
ff.implicitly_wait(10) # seconds
...
myDynamicElement = ff.find_element_by_id("myDynamicElement")

from http://docs.seleniumhq.org/docs/04_webdriver_advanced.jsp#implicit-waits

回答2:

It seems this is an stale element exception generally it occurs when you try to find some element. Which gets loaded every time but you found it earlier, so this is a stale.

I'll suggest to use some customized method to avoid this, One of the simplest solution:

void clickOnStaleElement(String id, WebDriver driver) {
    try {
        driver.find_element_by_id(id).click();
    } catch (StaleElementReferenceException e) {
        // Trying to find element stale element
        clickOnStaleElement(id, driver);
    } catch (NoSuchElementException ele) {
        clickOnStaleElement(id, driver);
    }
 }

来源：https://stackoverflow.com/questions/23869119/python-selenium-element-is-no-longer-attached-to-the-dom