Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. ItĀ collects links to all the places you might be looking at while hunting down a tough bug.
And, if youāre still stuck at the end, weāre happy to hop on a call to see how we can help out.
move_to_element_with_offset sometimes moves to wrong location
See original GitHub issueš Bug Report
Itās possible for a call like chain.move_to_element_with_offset(driver.find_element(By.TAG_NAME, "canvas"), 0, 0) to miss the canvas entirely and move to a different element.
To Reproduce
HTML:
<!DOCTYPE html>
<html>
<head>
<meta charset="UTF-8" />
<meta name="viewport" content="width=device-width" />
<title>Repro</title>
</head>
<style type="text/css">
html * {
box-sizing: border-box;
}
</style>
<body style="padding: 0; margin: 0;">
<div style="display: grid; grid-template-rows: 50px; position: absolute; left: 0; top: 0; right: 0; bottom: 0;">
<div>
Header
</div>
<div style="position: relative;">
<canvas style="padding: 0px; margin: 0px; border: 0px; position: absolute; top: 0px; left: 0px; width: 500px; height: 500px;">
</canvas>
</div>
</div>
</body>
</html>
Python script (using selenium==4.0.0a7):
from time import sleep
from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver import ActionChains
driver = webdriver.Chrome() # Or Firefox
driver.get("http://localhost:8002/repro.html") # The HTML file above
driver.set_window_size(780, 600) # Broken
#driver.set_window_size(780, 670) # Works
# Log the locations of all the clicks so we can see if things are behaving
driver.execute_script("""
window.clicks = [];
document.addEventListener("click", (event) => {
window.clicks.push([event.clientX, event.clientY]);
});
""")
# Almost certainly not necessary, just to demonstrate it's not a race condition
sleep(2)
WebDriverWait(driver, 10).until(EC.presence_of_element_located((By.TAG_NAME, "canvas")))
# The canvas rect is always correct
rect = driver.find_element(By.TAG_NAME, "canvas").rect
chain = ActionChains(driver)
chain.move_to_element_with_offset(driver.find_element(By.TAG_NAME, "canvas"), 0, 0)
chain.click()
chain.move_by_offset(200, 200)
chain.click()
chain.perform()
clicks = driver.execute_script("return window.clicks")
print("Rect", rect)
print("Clicks", clicks)
# Depending on the window size, this is correct or not
assert clicks[0] == [rect['x'], rect['y']]
assert clicks[1] == [clicks[0][0] + 200, clicks[0][1] + 200]
Expected behavior
The script should click the top left of the canvas and then the point (200, 200) from the top left of the canvas.
Instead, the actual points clicked vary.
With a large window size, the behaviour is correct. With smaller window sizes (it seems that the existence of a vertical scrollbar on the page can trigger the issue), different coordinates are clicked. The x-coordinate is always correct, the y-coordinate varies but is always smaller than the real y-coordinate of the canvas, so that the header is clicked and not the canvas.
This can be verified using document.addEventListener("click") as above, and also by observing that the text āHeaderā on the page gets selected if a drag operation is used.
In all cases the actual position of the canvas is identical, and Selenium reports the rect of the canvas correctly.
Reproduced with Chrome and Firefox, though they report different y-coordinates.
Environment
OS: Ubuntu 20.04 Browser: Reproduced on Firefox and Chrome Browser version: Firefox 82.0.3, Chrome 86.0.4240.198 Browser Driver version: ChromeDriver 86.0.4240.22, geckodriver 0.28.0 Language Bindings version: Python: selenium==4.0.0a7
Issue Analytics
- State:
- Created 3 years ago
- Comments:7 (4 by maintainers)
Top Results From Across the Web
Top Related Medium Post
Top Related StackOverflow Question
Troubleshoot Live Code
Top Related Reddit Thread
Top Related Hackernoon Post
Top Related Tweet
Top Related Dev.to Post
Top Related Hashnode Post
Hi @amagee ,
I am able to reproduce the same behaviour with ruby bindings as well. This could very well be a chromedriver issue and not a selenium one, will verify certain combinations (w3c on/off) and try it out
` require āselenium-webdriverā require ātest/unitā extend Test::Unit::Assertions capabilities = { āchromeOptionsā => {āw3cā => false} }
opts = Selenium::WebDriver::Chrome::Options.new(args: options) driver = Selenium::WebDriver.for(:chrome)
driver.get āhttp://localhost:8000/repro.htmlā sleep 1 driver.manage.window.resize_to(780, 600)
driver.execute_script(ā window.clicks = []; document.addEventListener(āclickā, (event) => { window.clicks.push([event.clientX, event.clientY]); }); ') sleep 1
rect = driver.find_element(:tag_name, ācanvasā).rect puts rect sleep 1
driver.action.move_to(driver.find_element(:tag_name, ācanvasā), 0, 0).click.move_by(200, 200).click.perform
clicks = driver.execute_script(āreturn window.clicksā) puts rect puts clicks[0] assert_equal(clicks[0], [rect[āxā], rect[āyā]]) assert_equal(clicks[1], [clicks[0][0] + 200, clicks[0][1] + 200])
sleep 2 driver.quit `
This issue has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.