How to Capture Full-Page Screenshots in SeleniumBase?

[dsmabulage]

I'm currently using SeleniumBase for my automated tests, and I need to capture full-page screenshots of the pages I'm testing. However, whenever I use driver.save_screenshot(), the screenshots are cropped and only show the visible part of the page.

driver.save_screenshot(f"table_element{i}.png")

Any advice or solution would be greatly appreciated!

[mdmintz]

Set the window size before taking the screenshot if the current size isn't good enough:

driver.set_window_size(x, y)

If you're running tests in headless mode, the default starting size is larger than normal.

More info: #2469 (comment).

Useful examples:

• https://github.com/seleniumbase/SeleniumBase/blob/master/examples/test_image_saving.py
• https://github.com/seleniumbase/SeleniumBase/blob/master/examples/test_save_screenshots.py

The SeleniumBase formats have their own screenshot methods separate from the driver:

• self.save_screenshot(name, folder=None, selector=None, by="css selector")
• self.save_screenshot_to_logs(name=None, selector=None, by="css selector")
• self.save_element_as_image_file(selector, file_name, folder=None, overlay_text="")

Methods called directly from the driver use the raw selenium API: https://github.com/SeleniumHQ/selenium, so if you're looking for specific functionality, check those docs. (You might want to see SeleniumHQ/selenium#14116, as that is related to your question.)

[dsmabulage]

import time
from seleniumbase import Driver


driver = Driver(
    uc=True,
    headless=False,
)

driver.get(
    "https://stackoverflow.com/questions/43008666/entityframeworkcore-does-not-exist-in-the-namespace-microsoft"
)

time.sleep(5)

x = driver.execute_script("return document.documentElement.scrollWidth")
y = driver.execute_script("return document.documentElement.scrollHeight")

print(x, y)

driver.set_window_size(x, y)

driver.save_screenshot(f"ss.png")

driver.quit()

1249 8781

I tried this approach but still getting cropped screenshots (only the visible part of the page). Am I missing something here?

[mdmintz]

Use this:

from seleniumbase import SB

with SB(uc=True, test=True, ad_block=True, headless2=True) as sb:
    sb.set_window_size(1400, 8000)
    url = "https://stackoverflow.com/questions/43008666/entityframeworkcore-does-not-exist-in-the-namespace-microsoft"
    sb.activate_cdp_mode(url)
    sb.save_screenshot("stack.png")

[dsmabulage]

Use this:

from seleniumbase import SB

with SB(uc=True, test=True, ad_block=True, headless2=True) as sb:
    sb.set_window_size(1400, 8000)
    url = "https://stackoverflow.com/questions/43008666/entityframeworkcore-does-not-exist-in-the-namespace-microsoft"
    sb.activate_cdp_mode(url)
    sb.save_screenshot("stack.png")

Can't we execute this method without headless=True because there are some manual user interactions? I tried it, and it didn't work.

[mdmintz]

Set the selector. Eg: sb.save_screenshot("stack.png", selector="body")

from seleniumbase import SB

with SB(uc=True, test=True, ad_block=True) as sb:
    url = "https://stackoverflow.com/questions/43008666/entityframeworkcore-does-not-exist-in-the-namespace-microsoft"
    sb.activate_cdp_mode(url)
    sb.save_screenshot("stack.png", selector="body")

[skald]

What is the best way to get just base64 of a screenshot with save_screenshot without saving a file? Not a binary png/jpg data. just base64