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 Daily Mirror

Browse Source
This commit is contained in:
Kovid Goyal 2011-10-30 22:20:33 +05:30
parent 96e4bf02c4
commit 228fdb8d94
1 changed files with 21 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

37
recipes/daily_mirror.recipe
View File

@ -1,10 +1,11 @@
from calibre.web.feeds.news import BasicNewsRecipe from calibre.web.feeds.news import BasicNewsRecipe
import re
class AdvancedUserRecipe1306061239(BasicNewsRecipe): class AdvancedUserRecipe1306061239(BasicNewsRecipe):
title = u'The Daily Mirror' title = u'The Daily Mirror'
description = 'News as provide by The Daily Mirror -UK' description = 'News as provide by The Daily Mirror -UK'
__author__ = 'Dave Asbury' __author__ = 'Dave Asbury'
# last updated 30/10/11
language = 'en_GB' language = 'en_GB'
cover_url = 'http://yookeo.com/screens/m/i/mirror.co.uk.jpg' cover_url = 'http://yookeo.com/screens/m/i/mirror.co.uk.jpg'
@ -12,26 +13,30 @@ class AdvancedUserRecipe1306061239(BasicNewsRecipe):
masthead_url = 'http://www.nmauk.co.uk/nma/images/daily_mirror.gif' masthead_url = 'http://www.nmauk.co.uk/nma/images/daily_mirror.gif'
oldest_article = 1 oldest_article = 2
max_articles_per_feed = 100 max_articles_per_feed = 30
remove_empty_feeds = True remove_empty_feeds = True
remove_javascript = True remove_javascript = True
no_stylesheets = True no_stylesheets = True
extra_css = '''
body{ text-align: justify; font-family:Arial,Helvetica,sans-serif; font-size:11px; font-size-adjust:none; font-stretch:normal; font-style:normal; font-variant:normal; font-weight:normal;}
'''
keep_only_tags = [ keep_only_tags = [
dict(name='h1'), dict(name='div',attrs={'id' : 'body-content'})
dict(attrs={'class':['article-attr']}), ]
dict(name='div', attrs={'class' : [ 'article-body', 'crosshead']})
remove_tags_after = [dict (name='div',attrs={'class' : 'related'})]
]
remove_tags = [ remove_tags = [
dict(name='div', attrs={'class' : ['caption', 'article-resize']}), dict(name='div',attrs={'id' : ['sidebar','menu','search-box','roffers-top']}),
dict( attrs={'class':'append-html'}) dict(name='div',attrs={'class' :['inline-ad span-16 last','article-resize','related','list teasers']}),
] dict(attrs={'class' : ['channellink','article-tags','replace','append-html']}),
dict(name='div',attrs={'class' : 'span-12 last sl-others addthis_toolbox addthis_default_style'})
]
preprocess_regexps = [
(re.compile(r'<dl class="q-search">.*?</dl>', re.IGNORECASE | re.DOTALL), lambda match: '')]
feeds = [ feeds = [
@ -43,10 +48,10 @@ class AdvancedUserRecipe1306061239(BasicNewsRecipe):
,(u'Music News','http://www.mirror.co.uk/celebs/music/rss.xml') ,(u'Music News','http://www.mirror.co.uk/celebs/music/rss.xml')
,(u'Celebs and Tv Gossip','http://www.mirror.co.uk/celebs/tv/rss.xml') ,(u'Celebs and Tv Gossip','http://www.mirror.co.uk/celebs/tv/rss.xml')
,(u'Sport','http://www.mirror.co.uk/sport/rss.xml') ,(u'Sport','http://www.mirror.co.uk/sport/rss.xml')
,(u'Life Style','http://www.mirror.co.uk/life-style/rss.xml') ,(u'Life Style','http://www.mirror.co.uk/life-style/rss.xml')
,(u'Advice','http://www.mirror.co.uk/advice/rss.xml') ,(u'Advice','http://www.mirror.co.uk/advice/rss.xml')
,(u'Travel','http://www.mirror.co.uk/advice/travel/rss.xml') ,(u'Travel','http://www.mirror.co.uk/advice/travel/rss.xml')
# example of commented out feed not needed ,(u'Travel','http://www.mirror.co.uk/advice/travel/rss.xml') # example of commented out feed not needed ,(u'Travel','http://www.mirror.co.uk/advice/travel/rss.xml')
]
]