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

Update San Francisco Chronicle

Browse Source
This commit is contained in:
Kovid Goyal 2013-04-09 15:00:31 +05:30
parent d449d6b0bb
commit 43331acbe5
1 changed files with 8 additions and 38 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

46
recipes/san_fran_chronicle.recipe
View File

@ -7,7 +7,6 @@ sfgate.com
'''
from calibre.web.feeds.news import BasicNewsRecipe
import re
class SanFranciscoChronicle(BasicNewsRecipe):
title = u'San Francisco Chronicle'
@ -19,16 +18,7 @@ class SanFranciscoChronicle(BasicNewsRecipe):
max_articles_per_feed = 100
no_stylesheets = True
use_embedded_content = False
remove_tags_before = {'id':'printheader'}
remove_tags = [
dict(name='div',attrs={'id':'printheader'})
,dict(name='a', attrs={'href':re.compile('http://ads\.pheedo\.com.*')})
,dict(name='div',attrs={'id':'footer'})
]
auto_cleanup = True
extra_css = '''
h1{font-family :Arial,Helvetica,sans-serif; font-size:large;}
@ -43,33 +33,13 @@ class SanFranciscoChronicle(BasicNewsRecipe):
'''
feeds = [
(u'Top News Stories', u'http://www.sfgate.com/rss/feeds/news.xml')
(u'Bay Area News', u'http://www.sfgate.com/bayarea/feed/Bay-Area-News-429.php'),
(u'City Insider', u'http://www.sfgate.com/default/feed/City-Insider-Blog-573.php'),
(u'Crime Scene', u'http://www.sfgate.com/rss/feed/Crime-Scene-Blog-599.php'),
(u'Education News', u'http://www.sfgate.com/education/feed/Education-News-from-SFGate-430.php'),
(u'National News', u'http://www.sfgate.com/rss/feed/National-News-RSS-Feed-435.php'),
(u'Weird News', u'http://www.sfgate.com/weird/feed/Weird-News-RSS-Feed-433.php'),
(u'World News', u'http://www.sfgate.com/rss/feed/World-News-From-SFGate-432.php'),
]
def print_version(self,url):
url= url +"&type=printable"
return url
def get_article_url(self, article):
print str(article['title_detail']['value'])
url = article.get('guid',None)
url = "http://www.sfgate.com/cgi-bin/article.cgi?f="+url
if "Presented By:" in str(article['title_detail']['value']):
url = ''
return url