Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-06-23 15:30:45 -04:00
Code Issues Packages Projects Releases Wiki Activity

Fix Anchorage Daily News

Browse Source
This commit is contained in:
Kovid Goyal 2012-12-10 18:03:44 +05:30
parent ecb661b220
commit cfed6ac42f
1 changed files with 18 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

34
recipes/anchorage_daily.recipe
View File

@ -5,14 +5,16 @@ class AdvancedUserRecipe1278347258(BasicNewsRecipe):
__author__ = 'rty'
oldest_article = 7
max_articles_per_feed = 100
auto_cleanup = True
feeds = [(u'Alaska News', u'http://www.adn.com/news/alaska/index.xml'),
(u'Business', u'http://www.adn.com/money/index.xml'),
(u'Sports', u'http://www.adn.com/sports/index.xml'),
(u'Politics', u'http://www.adn.com/politics/index.xml'),
(u'Lifestyles', u'http://www.adn.com/life/index.xml'),
(u'Iditarod', u'http://www.adn.com/iditarod/index.xml')
]
(u'Business', u'http://www.adn.com/money/index.xml'),
(u'Sports', u'http://www.adn.com/sports/index.xml'),
(u'Politics', u'http://www.adn.com/politics/index.xml'),
(u'Lifestyles', u'http://www.adn.com/life/index.xml'),
(u'Iditarod', u'http://www.adn.com/iditarod/index.xml')
]
description = ''''Alaska's Newspaper'''
publisher = 'http://www.adn.com'
category = 'news, Alaska, Anchorage'
@ -28,13 +30,13 @@ class AdvancedUserRecipe1278347258(BasicNewsRecipe):
conversion_options = {'linearize_tables':True}
masthead_url = 'http://media.adn.com/includes/assets/images/adn_logo.2.gif'
keep_only_tags = [
dict(name='div', attrs={'class':'left_col story_mainbar'}),
]
remove_tags = [
dict(name='div', attrs={'class':'story_tools'}),
dict(name='p', attrs={'class':'ad_label'}),
]
remove_tags_after = [
dict(name='div', attrs={'class':'advertisement'}),
]
#keep_only_tags = [
#dict(name='div', attrs={'class':'left_col story_mainbar'}),
#]
#remove_tags = [
#dict(name='div', attrs={'class':'story_tools'}),
#dict(name='p', attrs={'class':'ad_label'}),
#]
#remove_tags_after = [
#dict(name='div', attrs={'class':'advertisement'}),
#]