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

Updated recipe for Telegraph UK

Browse Source
This commit is contained in:
Kovid Goyal 2009-09-03 10:32:25 -06:00
parent 1698282f2c
commit 264d032646
1 changed files with 58 additions and 38 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

96
src/calibre/web/feeds/recipes/recipe_telegraph_uk.py
View File

@ -1,38 +1,58 @@
#!/usr/bin/env python #!/usr/bin/env python
__license__ = 'GPL v3'
__license__ = 'GPL v3' __copyright__ = '2008, Darko Miletic <darko.miletic at gmail.com>'
__copyright__ = '2008, Darko Miletic <darko.miletic at gmail.com>' '''
''' telegraph.co.uk
telegraph.co.uk '''
'''
from calibre.web.feeds.news import BasicNewsRecipe
from calibre.web.feeds.news import BasicNewsRecipe
class TelegraphUK(BasicNewsRecipe):
class TelegraphUK(BasicNewsRecipe): title = u'Telegraph.co.uk'
title = u'Telegraph.co.uk' __author__ = 'Darko Miletic and Sujata Raman'
__author__ = 'Darko Miletic' description = 'News from United Kingdom'
description = 'News from United Kingdom' oldest_article = 7
oldest_article = 7 max_articles_per_feed = 100
max_articles_per_feed = 100 no_stylesheets = True
no_stylesheets = True language = _('English')
language = _('English') use_embedded_content = False
use_embedded_content = False
extra_css = '''
keep_only_tags = [ h1{font-family :Arial,Helvetica,sans-serif; font-size:large; color:#666666}
dict(name='div', attrs={'class':'storyHead'}) h2{font-family :Arial,Helvetica,sans-serif; font-size:small; color:#444444}
,dict(name='div', attrs={'class':'story' }) .story{font-family :Arial,Helvetica,sans-serif; font-size: x-small; color:#444444 }
] .byline{color:#666666; font-family :Arial,Helvetica,sans-serif; font-size: xx-small;}
remove_tags = [dict(name='div', attrs={'class':'slideshow'})] a{color:#234B7B; }
.imageExtras{color:#666666; font-family :Arial,Helvetica,sans-serif; font-size: xx-small;}
feeds = [ '''
(u'UK News' , u'http://www.telegraph.co.uk/news/uknews/rss' )
,(u'World News' , u'http://www.telegraph.co.uk/news/worldnews/rss' ) keep_only_tags = [
,(u'Politics' , u'http://www.telegraph.co.uk/news/newstopics/politics/rss' ) dict(name='div', attrs={'class':'storyHead'})
,(u'Technology News', u'http://www.telegraph.co.uk/scienceandtechnology/technology/technologynews/rss' ) ,dict(name='div', attrs={'class':'story' })
,(u'UK News' , u'http://www.telegraph.co.uk/scienceandtechnology/technology/technologyreviews/rss') #,dict(name='div', attrs={'class':['slideshowHD gutterUnder',"twoThirds gutter","caption" ] })
,(u'Science News' , u'http://www.telegraph.co.uk/scienceandtechnology/science/sciencenews/rss' ) ]
,(u'Sport' , u'http://www.telegraph.co.uk/sport/rss' ) remove_tags = [dict(name='div', attrs={'class':['related_links_inline',"imgindex","next","prev","gutterUnder"]})]
,(u'Earth News' , u'http://www.telegraph.co.uk/earth/earthnews/rss' )
,(u'Comment' , u'http://www.telegraph.co.uk/comment/rss' ) feeds = [
,(u'How about that?', u'http://www.telegraph.co.uk/news/newstopics/howaboutthat/rss' ) (u'UK News' , u'http://www.telegraph.co.uk/news/uknews/rss' )
] ,(u'World News' , u'http://www.telegraph.co.uk/news/worldnews/rss' )
,(u'Politics' , u'http://www.telegraph.co.uk/news/newstopics/politics/rss' )
,(u'Technology News', u'http://www.telegraph.co.uk/scienceandtechnology/technology/technologynews/rss' )
,(u'UK News' , u'http://www.telegraph.co.uk/scienceandtechnology/technology/technologyreviews/rss')
,(u'Science News' , u'http://www.telegraph.co.uk/scienceandtechnology/science/sciencenews/rss' )
,(u'Sport' , u'http://www.telegraph.co.uk/sport/rss' )
,(u'Earth News' , u'http://www.telegraph.co.uk/earth/earthnews/rss' )
,(u'Comment' , u'http://www.telegraph.co.uk/comment/rss' )
,(u'How about that?', u'http://www.telegraph.co.uk/news/newstopics/howaboutthat/rss' )
]
def get_article_url(self, article):
url = article.get('guid', None)
if 'picture-galleries' in url or 'pictures' in url or 'picturegalleries' in url :
url = None
return url