bellingcat/auto-archiver
1
0
Fork 0
mirror of https://github.com/bellingcat/auto-archiver.git synced 2026-06-13 13:48:28 +03:00
Code Issues Packages Projects Releases Wiki Activity

Add some code to attempt to click the cookies banners on various websites

Browse Source
This commit is contained in:
Patrick Robertson
2025-03-03 15:55:04 +00:00
parent dea0a49600
commit 0dfab2d1bc
1 changed files with 12 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
src/auto_archiver/utils/webdriver.py
View File

@@ -72,6 +72,8 @@ class CookieSettingDriver(webdriver.Firefox):
time.sleep(2) time.sleep(2)
except Exception as e: except Exception as e:
logger.warning(f'Failed on fb accept cookies.', e) logger.warning(f'Failed on fb accept cookies.', e)
# now get the actual URL # now get the actual URL
super(CookieSettingDriver, self).get(url) super(CookieSettingDriver, self).get(url)
if self.facebook_accept_cookies: if self.facebook_accept_cookies:
@@ -79,7 +81,17 @@ class CookieSettingDriver(webdriver.Firefox):
close_button = self.find_element(By.XPATH, "//div[@role='dialog']//div[@aria-label='Close']") close_button = self.find_element(By.XPATH, "//div[@role='dialog']//div[@aria-label='Close']")
if close_button: if close_button:
close_button.click() close_button.click()
else:
# for all other sites, try and use some common button text to reject/accept cookies
for text in ["Refuse non-essential cookies", "Decline optional cookies", "Reject additional cookies", "Accept all cookies"]:
try:
accept_button = self.find_element(By.XPATH, f"//*[contains(translate(text(), 'ABCDEFGHIJKLMNOPQRSTUVWXYZ', 'abcdefghijklmnopqrstuvwxyz'), '{text.lower()}')]")
if accept_button:
accept_button.click()
break
except Exception as e:
pass
class Webdriver: class Webdriver: