bellingcat/auto-archiver
1
0
Fork 0
mirror of https://github.com/bellingcat/auto-archiver.git synced 2026-06-12 13:18:28 +03:00
Code Issues Packages Projects Releases Wiki Activity

Add 'print_pdf' option to the screenshot enricher. Fixes #132

Browse Source
This commit is contained in:
Patrick Robertson
2024-12-18 13:37:44 +03:00
parent e49550163f
commit 663c8ad93a
3 changed files with 36 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
src/auto_archiver/utils/webdriver.py
View File

@@ -2,18 +2,24 @@ from __future__ import annotations
from selenium import webdriver
from selenium.common.exceptions import TimeoutException
from selenium.webdriver.common.proxy import Proxy, ProxyType
from selenium.webdriver.common.print_page_options import PrintOptions
from loguru import logger
from selenium.webdriver.common.by import By
import time
class Webdriver:
def __init__(self, width: int, height: int, timeout_seconds: int, facebook_accept_cookies: bool = False, http_proxy: str = "") -> webdriver:
def __init__(self, width: int, height: int, timeout_seconds: int, facebook_accept_cookies: bool = False, http_proxy: str = "", print_options: dict = {}) -> webdriver:
self.width = width
self.height = height
self.timeout_seconds = timeout_seconds
self.facebook_accept_cookies = facebook_accept_cookies
self.http_proxy = http_proxy
# create and set print options
self.print_options = PrintOptions()
for k, v in print_options.items():
setattr(self.print_options, k, v)
def __enter__(self) -> webdriver:
options = webdriver.FirefoxOptions()
@@ -24,6 +30,7 @@ class Webdriver:
self.driver = webdriver.Firefox(options=options)
self.driver.set_window_size(self.width, self.height)
self.driver.set_page_load_timeout(self.timeout_seconds)
self.driver.print_options = self.print_options
except TimeoutException as e:
logger.error(f"failed to get new webdriver, possibly due to insufficient system resources or timeout settings: {e}")