Backup_Repos/calibre
1
0
Fork 0
mirror of git://github.com/kovidgoyal/calibre.git synced 2025-12-29 08:36:25 +01:00
Code Issues Projects Releases Wiki Activity

Fix Financial Times Int'l categories

Browse source
This commit is contained in:
bobbysteel 2017-08-27 14:38:42 +01:00 committed by GitHub
parent fe9f9ac723
commit 1e04233e2c
1 changed files with 14 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

20
recipes/financial_times_us.recipe
View file

@ -70,12 +70,20 @@ def get_browser(self):
def parse_index(self):
articles = []
soup = self.index_to_soup(self.INDEX)
for article in soup.findAll('a', href=True, attrs={'data-trackable':'main-link'}):
url = self.PREFIX + article['href']
title = self.tag_to_string(article)
articles.append({'title': title, 'url': url, 'description': '', 'date': ''})
return [("Articles", articles)]
totalfeeds = []
current_section = []
div = []
for div in soup.findAll('div', attrs={'data-trackable': 'list'}):
articles = []
current_section = self.tag_to_string(div.find('h2'))
self.log('in section: ', current_section)
for article in div.findAll('a', href=True, attrs={'data-trackable':'main-link'}):
url = self.PREFIX + article['href']
title = self.tag_to_string(article)
articles.append({'title': title, 'url': url, 'description': '', 'date': ''})
self.log('title: ', title, ' url: ', url)
totalfeeds.append((current_section,articles))
return totalfeeds
def preprocess_html(self, soup):
for img in soup.findAll('img', srcset=True):