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

Function to get bing cached version of URL

Browse Source
This commit is contained in:
Kovid Goyal 2024-09-26 12:36:32 +05:30
parent df7344be33
commit f07335f107
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 10 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
src/calibre/ebooks/metadata/sources/search_engines.py
View File

@ -217,6 +217,12 @@ def bing_url_processor(url):
return url return url
def bing_cached_url(url):
results, search_url = bing_search(['url:' + url])
for result in results:
return result.cached_url
def resolve_bing_wrapper_page(url, br, log): def resolve_bing_wrapper_page(url, br, log):
raw = br.open_novisit(url).read().decode('utf-8', 'replace') raw = br.open_novisit(url).read().decode('utf-8', 'replace')
m = re.search(r'var u = "(.+)"', raw) m = re.search(r'var u = "(.+)"', raw)
@ -259,10 +265,10 @@ def bing_search(terms, site=None, br=None, log=prints, safe_search=False, dump_r
d, w = div.get('u').split('|')[-2:] d, w = div.get('u').split('|')[-2:]
cached_url = 'https://cc.bingj.com/cache.aspx?q={q}&d={d}&mkt=en-US&setlang=en-US&w={w}'.format( cached_url = 'https://cc.bingj.com/cache.aspx?q={q}&d={d}&mkt=en-US&setlang=en-US&w={w}'.format(
q=q, d=d, w=w) q=q, d=d, w=w)
url = a.get('href') ans_url = a.get('href')
if url.startswith('https://www.bing.com/'): if ans_url.startswith('https://www.bing.com/'):
url = resolve_bing_wrapper_page(url, br, log) ans_url = resolve_bing_wrapper_page(ans_url, br, log)
ans.append(Result(url, title, cached_url)) ans.append(Result(ans_url, 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)