bellingcat/auto-archiver
1
0
Fork 0
mirror of https://github.com/bellingcat/auto-archiver.git synced 2026-06-12 21:28:29 +03:00
Code Issues Packages Projects Releases Wiki Activity

Add hash and screenshot methods; switch to more recent ytdl fork

Browse Source
This commit is contained in:
Logan Williams
2022-02-25 13:54:40 +01:00
parent d76e3bc7ec
commit 1eb17e4de5
7 changed files with 444 additions and 37 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
archivers/wayback_archiver.py
View File

@@ -8,8 +8,8 @@ from .base_archiver import Archiver, ArchiveResult
class WaybackArchiver(Archiver):
name = "wayback"
def __init__(self, storage: Storage):
super(WaybackArchiver, self).__init__(storage)
def __init__(self, storage: Storage, driver):
super(WaybackArchiver, self).__init__(storage, driver)
self.seen_urls = {}
def download(self, url, check_if_exists=False):
@@ -71,6 +71,7 @@ class WaybackArchiver(Archiver):
except:
title = "Could not get title"
result = ArchiveResult(status='Internet Archive fallback', cdn_url=archive_url, title=title)
screenshot = self.get_screenshot(url)
result = ArchiveResult(status='Internet Archive fallback', cdn_url=archive_url, title=title, screenshot=screenshot)
self.seen_urls[url] = result
return result