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

Fix Newsweek

Browse Source
This commit is contained in:
Kovid Goyal 2011-04-19 14:54:41 -06:00
parent e776b5b1f6
commit 3f71ad9420
1 changed files with 8 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
recipes/newsweek.recipe
View File

@ -1,4 +1,3 @@
import string
from calibre.web.feeds.news import BasicNewsRecipe
class Newsweek(BasicNewsRecipe):
@ -11,7 +10,6 @@ class Newsweek(BasicNewsRecipe):
no_stylesheets = True
BASE_URL = 'http://www.newsweek.com'
INDEX = BASE_URL+'/topics.html'
keep_only_tags = dict(name='article', attrs={'class':'article-text'})
remove_tags = [dict(attrs={'data-dartad':True})]
@ -23,11 +21,14 @@ class Newsweek(BasicNewsRecipe):
return soup
def newsweek_sections(self):
soup = self.index_to_soup(self.INDEX)
for a in soup.findAll('a', title='Primary tag', href=True):
yield (string.capitalize(self.tag_to_string(a)),
self.BASE_URL+a['href'])
return [
('Nation', 'http://www.newsweek.com/tag/nation.html'),
('Society', 'http://www.newsweek.com/tag/society.html'),
('Culture', 'http://www.newsweek.com/tag/culture.html'),
('World', 'http://www.newsweek.com/tag/world.html'),
('Politics', 'http://www.newsweek.com/tag/politics.html'),
('Business', 'http://www.newsweek.com/tag/business.html'),
]
def newsweek_parse_section_page(self, soup):
for article in soup.findAll('article', about=True,