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 The Daily Mail

Browse Source
This commit is contained in:
Kovid Goyal 2009-10-09 10:58:57 -06:00
parent dfe6c1208a
commit a9d86fd6c9
1 changed files with 25 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

34
src/calibre/web/feeds/recipes/recipe_daily_mail.py
View File

@ -5,15 +5,28 @@ class TheDailyMail(BasicNewsRecipe):
oldest_article = 2
language = 'en_GB'
author = 'RufusA'
author = 'RufusA and Sujata Raman'
simultaneous_downloads= 1
max_articles_per_feed = 50
extra_css = 'h1 {text-align: left;}'
extra_css = '''#js-article-text{font-family:Arial,Helvetica,sans-serif;}
h1{font-size:large; font-weight:bold;}
a.author{color:#003580;}
.js-article-text{font-size:50%;}
.imageCaption{font-size:x-small; font-weight:bold}
'''
remove_tags = [ dict(name='div', attrs={'class':['article-icon-links-container','print-or-mail-links cleared',
'social-links cleared','explore-links','relatedItems','intellicrumbs box','travel','moduleHalf']}),
dict(name='div', attrs={'id':['link-unit-wrapper','pushdown-ad','reader-comments','googleAds',]}),
dict(name='h3', attrs={'class':['social-links-title']}),
dict(name='span', attrs={'class':['clickToEnlargeTop']}),
]
#remove_tags_before = dict(name='div', attrs={'id':'content'})
keep_only_tags = [dict(name='div', attrs={'id':'js-article-text'})]
remove_tags = [ dict(name='ul', attrs={'class':'article-icons-links'}) ]
remove_tags_after = dict(name='h3', attrs={'class':'social-links-title'})
remove_tags_before = dict(name='div', attrs={'id':'content'})
no_stylesheets = True
feeds = [
@ -27,8 +40,11 @@ class TheDailyMail(BasicNewsRecipe):
(u'Money', u'http://www.dailymail.co.uk/money/index.rss'),
(u'Property', u'http://www.dailymail.co.uk/property/index.rss'),
(u'Motoring', u'http://www.dailymail.co.uk/motoring/index.rss'),
(u'Travel', u'http://www.dailymail.co.uk/travel/index.rss')]
(u'Travel', u'http://www.dailymail.co.uk/travel/index.rss')
]
#def print_version(self, url):
# main = url.partition('?')[0]
# return main + '?printingPage=true'
def print_version(self, url):
main = url.partition('?')[0]
return main + '?printingPage=true'