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 New York Magazine

Browse Source
This commit is contained in:
Kovid Goyal 2023-02-24 21:21:16 +05:30
parent 815cd507b9
commit f851d78e6a
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 5 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
recipes/nymag.recipe
View File

@ -24,12 +24,10 @@ class NewYorkMagazine(BasicNewsRecipe):
remove_javascript = True remove_javascript = True
encoding = 'utf-8' encoding = 'utf-8'
keep_only_tags = [ keep_only_tags = [
classes('lede-text headline-primary article-timestamp by-authors'), dict(name='article', attrs={'class':lambda x: x and 'article' in x.split()})
dict(id='main'),
dict(itemprop='articleBody'),
] ]
remove_tags = [ remove_tags = [
classes('related-stories start-discussion'), classes('related-stories start-discussion newsletter-flex-text comments-link tags related secondary-area'),
dict(id=['minibrowserbox', 'article-related', 'article-tools']) dict(id=['minibrowserbox', 'article-related', 'article-tools'])
] ]
remove_attributes = ['srcset'] remove_attributes = ['srcset']
@ -70,6 +68,9 @@ class NewYorkMagazine(BasicNewsRecipe):
return feeds return feeds
def preprocess_html(self, soup): def preprocess_html(self, soup):
if lede := soup.findAll('div', attrs={'class':lambda x: x and 'lede-image-wrapper' in x.split()}):
if len(lede) > 1:
lede[1].extract()
for img in soup.findAll('img', attrs={'data-src': True}): for img in soup.findAll('img', attrs={'data-src': True}):
img['src'] = img['data-src'] img['src'] = img['data-src']
return soup return soup