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

Update LifeHacker

Browse Source
This commit is contained in:
Kovid Goyal 2020-10-21 11:15:53 +05:30
parent 7fe84df779
commit 5f4c2ce051
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 12 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
recipes/lifehacker.recipe
View File

@ -24,13 +24,14 @@ class LifeHacker(BasicNewsRecipe):
no_stylesheets = True no_stylesheets = True
use_embedded_content = False use_embedded_content = False
remove_empty_feeds = True remove_empty_feeds = True
remove_attributes = ['style']
keep_only_tags = [ keep_only_tags = [
dict(name='h1'), dict(name='h1'),
classes('headline author post-content js_post-content'), classes('headline author post-content js_post-content'),
] ]
remove_tags = [ remove_tags = [
classes('magnifier commerce-inset gmg-avatar js_ad-dynamic instream-native-video inset--story'), classes('magnifier commerce-inset gmg-avatar js_ad-dynamic instream-native-video inset--story js_commerce-inset-permalink'),
dict(id='lifehacker-avatar-svg'), dict(id='lifehacker-avatar-svg'),
dict(name='aside'), dict(name='aside'),
] ]
@ -38,8 +39,14 @@ class LifeHacker(BasicNewsRecipe):
feeds = [(u'Articles', u'https://lifehacker.com/rss')] feeds = [(u'Articles', u'https://lifehacker.com/rss')]
def preprocess_soup(self, soup): def preprocess_html(self, soup):
for source in soup.findAll('source', attrs={'data-srcset':True}):
for img in source.findAll('img'): def choose_from_srcset(raw):
img['src'] = source['data-srcset'] parts = raw.split()
if len(parts) > 2:
return parts[2]
return parts[0]
for img in soup.findAll(attrs={'data-srcset':True}):
img['src'] = choose_from_srcset(img['data-srcset'])
return soup return soup