Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

Dont use google webcache as google has discontinued it

Browse Source
This commit is contained in:
Kovid Goyal 2024-09-25 08:21:45 +05:30
parent 4c3a1606f2
commit b13307a960
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 4 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
src/calibre/ebooks/metadata/sources/search_engines.py
View File

@ -31,7 +31,7 @@ from calibre.ebooks.chardet import xml_to_unicode
from calibre.utils.lock import ExclusiveFile from calibre.utils.lock import ExclusiveFile
from calibre.utils.random_ua import accept_header_for_ua from calibre.utils.random_ua import accept_header_for_ua
current_version = (1, 2, 5) current_version = (1, 2, 6)
minimum_calibre_version = (2, 80, 0) minimum_calibre_version = (2, 80, 0)
webcache = {} webcache = {}
webcache_lock = Lock() webcache_lock = Lock()
@ -289,6 +289,7 @@ def google_cache_url_for_url(url):
def google_get_cached_url(url, br=None, log=prints, timeout=60): def google_get_cached_url(url, br=None, log=prints, timeout=60):
# Google's webcache was discontinued in september 2024
cached_url = google_cache_url_for_url(url) cached_url = google_cache_url_for_url(url)
br = google_specialize_browser(br or browser()) br = google_specialize_browser(br or browser())
try: try:
@ -327,8 +328,7 @@ def google_parse_results(root, raw, log=prints, ignore_uncached=True):
if curl in seen: if curl in seen:
continue continue
seen.add(curl) seen.add(curl)
cached_url = google_cache_url_for_url(curl) ans.append(Result(a.get('href'), title, curl))
ans.append(Result(a.get('href'), title, cached_url))
if not ans: if not ans:
title = ' '.join(root.xpath('//title/text()')) title = ' '.join(root.xpath('//title/text()'))
log('Failed to find any results on results page, with title:', title) log('Failed to find any results on results page, with title:', title)
@ -413,7 +413,7 @@ def google_develop(search_terms='1423146786', raw_from=''):
def get_cached_url(url, br=None, log=prints, timeout=60): def get_cached_url(url, br=None, log=prints, timeout=60):
return google_get_cached_url(url, br, log, timeout) or wayback_machine_cached_url(url, br, log, timeout) return wayback_machine_cached_url(url, br, log, timeout)
def get_data_for_cached_url(url): def get_data_for_cached_url(url):