Selenium Compound class names not permitted
Question:
I have the below code that clicks on an element to pop up a screen and copy the text in it
el1 = driver.find_element_by_id("keyDev-A")
el1.click()
el2 = driver.find_element_by_class_name("content")
print(el2.text)
However, when I tried to get selenium
to click on the button within that popup with
el3 = driver.find_element(By.CLASS_NAME, "action-btn cancel alert-display")
el3.click()
It produces an error message:
invalid selector: Compound class names not permitted
This is the HTML that I am trying to get selenium
to click on. The Close
button.
<div class="nav">
<span class="action-btn confirm prompt-display">Confirm</span>
<span class="action-btn cancel prompt-display">Cancel</span>
<span class="action-btn cancel alert-display">Close</span>
</div>
How should I be writing el3
in order to click on the Close button?
Answers:
Leon’s comment leads to the correct information that compound class names are no longer supported. What you could do instead is try using css selectors. In your case, the following line of code should help you get the element you want :
el3 = driver.find_element_by_css_selector(".action-btn.cancel.alert-display")
It finds the element with all three classes (action-btn, cancel and alert-display) in the class attribute. Do note that the order of the classes does not matter here and any of the classes may appear anywhere in the class attribute. As long as the element has all three classes, it will be selected.
If you want the order of the classes to be fixed, you can use the following xpath :
el3 = driver.find_element_by_xpath("//*[@class='action-btn cancel alert-display']")
I am late to this question. But I also found a work around by treating the compound classes as a String, using tag_name, and get_attribute(‘class’), when you are not familiar with Xpath. It needs some more lines of code but it’s straight forward and fit for beginners like me.
elements = driver.find_elements_by_tag_name('Tag Name Here')
for element in elments:
className = watchingTable.get_attribute('class')
print(className)
if className == 'Your Needed Classname':
#Do your things
The answer here are incorrect :
If you check the exception from the by_class_name:
you can see it is using css under the hood
by_class_name just adds ‘.’ infront of the locator provided so ‘a’ will be passed as ‘.a’ and a.b will be passed as ‘.a.b’
So You can use class_name for multiple classes , just need to replace space with ‘.’
so "a b c" should be passed as "a.b.c"
Eg:
Working example:
from selenium import webdriver
import time
from selenium.webdriver.support import expected_conditions as EC
driver = webdriver.Chrome()
driver.get("https://stackoverflow.com/questions/65579491/find-element-by-class-name-in-selenium-giving-error/65579606?noredirect=1#comment115946541_65579606")
time.sleep(5)
elem = driver.find_element_by_class_name('overflow-x-auto.ml-auto.-secondary.grid.ai-center.list-reset.h100')
print(elem.get_attribute("outerHTML"))
This error message…
invalid selector: Compound class names not permitted
…implies that locator strategies using Compound class names are not valid while using Selenium.
Traces of this change can be confirmed from the Selenium v2.40.0 changelist where the change log mentions about adding proper error code for compound class name usage:
- Implemented proper error code for the case of invalid css selector empty class name, and compound class name in atoms.
Solution
As an alternative you can use either of the following Locator Strategies:
-
Using CSS_SELECTOR
:
driver.find_element(By.CSS_SELECTOR, "span.action-btn.cancel.alert-display").click()
-
Using XPATH
:
driver.find_element(By.XPATH, "//span[@class='action-btn cancel alert-display']").click()
References
You can find a couple of relevant detailed discussion in:
Manually intersect the results for multiple classes.
def find_element_multi_class(driver, classes):
elements = []
for i,c in enumerate(classes):
if i == 0:
elements = driver.find_elements(By.CLASS_NAME, c)
else:
new_elems = driver.find_elements(By.CLASS_NAME, c)
elements = list(set(elements) & set(new_elems))
return elements
I have the below code that clicks on an element to pop up a screen and copy the text in it
el1 = driver.find_element_by_id("keyDev-A")
el1.click()
el2 = driver.find_element_by_class_name("content")
print(el2.text)
However, when I tried to get selenium
to click on the button within that popup with
el3 = driver.find_element(By.CLASS_NAME, "action-btn cancel alert-display")
el3.click()
It produces an error message:
invalid selector: Compound class names not permitted
This is the HTML that I am trying to get selenium
to click on. The Close
button.
<div class="nav">
<span class="action-btn confirm prompt-display">Confirm</span>
<span class="action-btn cancel prompt-display">Cancel</span>
<span class="action-btn cancel alert-display">Close</span>
</div>
How should I be writing el3
in order to click on the Close button?
Leon’s comment leads to the correct information that compound class names are no longer supported. What you could do instead is try using css selectors. In your case, the following line of code should help you get the element you want :
el3 = driver.find_element_by_css_selector(".action-btn.cancel.alert-display")
It finds the element with all three classes (action-btn, cancel and alert-display) in the class attribute. Do note that the order of the classes does not matter here and any of the classes may appear anywhere in the class attribute. As long as the element has all three classes, it will be selected.
If you want the order of the classes to be fixed, you can use the following xpath :
el3 = driver.find_element_by_xpath("//*[@class='action-btn cancel alert-display']")
I am late to this question. But I also found a work around by treating the compound classes as a String, using tag_name, and get_attribute(‘class’), when you are not familiar with Xpath. It needs some more lines of code but it’s straight forward and fit for beginners like me.
elements = driver.find_elements_by_tag_name('Tag Name Here')
for element in elments:
className = watchingTable.get_attribute('class')
print(className)
if className == 'Your Needed Classname':
#Do your things
The answer here are incorrect :
If you check the exception from the by_class_name:
you can see it is using css under the hood
by_class_name just adds ‘.’ infront of the locator provided so ‘a’ will be passed as ‘.a’ and a.b will be passed as ‘.a.b’
So You can use class_name for multiple classes , just need to replace space with ‘.’
so "a b c" should be passed as "a.b.c"
Eg:
Working example:
from selenium import webdriver
import time
from selenium.webdriver.support import expected_conditions as EC
driver = webdriver.Chrome()
driver.get("https://stackoverflow.com/questions/65579491/find-element-by-class-name-in-selenium-giving-error/65579606?noredirect=1#comment115946541_65579606")
time.sleep(5)
elem = driver.find_element_by_class_name('overflow-x-auto.ml-auto.-secondary.grid.ai-center.list-reset.h100')
print(elem.get_attribute("outerHTML"))
This error message…
invalid selector: Compound class names not permitted
…implies that locator strategies using Compound class names are not valid while using Selenium.
Traces of this change can be confirmed from the Selenium v2.40.0 changelist where the change log mentions about adding proper error code for compound class name usage:
- Implemented proper error code for the case of invalid css selector empty class name, and compound class name in atoms.
Solution
As an alternative you can use either of the following Locator Strategies:
-
Using
CSS_SELECTOR
:driver.find_element(By.CSS_SELECTOR, "span.action-btn.cancel.alert-display").click()
-
Using
XPATH
:driver.find_element(By.XPATH, "//span[@class='action-btn cancel alert-display']").click()
References
You can find a couple of relevant detailed discussion in:
Manually intersect the results for multiple classes.
def find_element_multi_class(driver, classes):
elements = []
for i,c in enumerate(classes):
if i == 0:
elements = driver.find_elements(By.CLASS_NAME, c)
else:
new_elems = driver.find_elements(By.CLASS_NAME, c)
elements = list(set(elements) & set(new_elems))
return elements