Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-08 18:54:09 -04:00
Code Issues Packages Projects Releases Wiki Activity

Update Slate

Browse Source
This commit is contained in:
Kovid Goyal 2021-02-14 20:48:03 +05:30
parent a2fe4afafe
commit 4311cae31a
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 12 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
recipes/slate.recipe
View File

@ -32,7 +32,7 @@ class Slate(BasicNewsRecipe):
classes('article__header article__content'),
]
remove_tags = [
dict(name='ul', attrs={'class':"social-share"}),
classes('social-share slate-ad newsletter-signup in-article-recirc'),
]
def preprocess_html(self, soup):
@ -53,7 +53,7 @@ class Slate(BasicNewsRecipe):
('Double X', '/articles/double_x.html'),
):
url = self.INDEX + url
self.log('\nFound section:', sectitle)
self.log('\nFound section:', sectitle, url)
articles = self.slate_section_articles(self.index_to_soup(url))
if articles:
ans.append((sectitle, articles))
@ -63,24 +63,23 @@ class Slate(BasicNewsRecipe):
def slate_section_articles(self, soup):
ans = []
main = soup.find('article', attrs={'class': 'main'})
main = soup.find('section', **classes('main'))
if main is None:
return ans
for a in main.findAll('a', attrs={'class': 'primary'}):
for div in main.findAll(**classes('section-feed__item')):
a = div.find('a')
url = a['href']
if url.endswith('/'):
continue
p = a.parent
title = p.find(attrs={'class': 'hed'})
if title is None:
continue
title = self.tag_to_string(title)
span = p.find(attrs={'class': 'byline'})
h = a.find(['h2', 'h3', 'h4'])
title = self.tag_to_string(h)
desc = ''
if span is not None:
desc = self.tag_to_string(span)
for q in ('byline', 'dek'):
span = div.find(attrs={'class': lambda x: x and ('-' + q) in x})
if span is not None:
desc += self.tag_to_string(span).strip()
self.log('\t' + title)
self.log('\t\t' + url)
ans.append({'title': title, 'description': desc,
ans.append({'title': title, 'description': desc.strip(),
'date': '', 'url': url})
return ans