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 Sun UK

Browse Source
This commit is contained in:
Kovid Goyal 2012-04-07 17:26:19 +05:30
parent 0f9137a4ef
commit a689e82b56
1 changed files with 27 additions and 24 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

51
recipes/the_sun.recipe
View File

@ -1,57 +1,60 @@
import re import re
from calibre.web.feeds.recipes import BasicNewsRecipe from calibre.web.feeds.recipes import BasicNewsRecipe
class AdvancedUserRecipe1325006965(BasicNewsRecipe): class AdvancedUserRecipe1325006965(BasicNewsRecipe):
title = u'The Sun UK' title = u'The Sun UK'
cover_url = 'http://www.thesun.co.uk/img/global/new-masthead-logo.png' cover_url = 'http://www.thesun.co.uk/img/global/new-masthead-logo.png'
description = 'A Recipe for The Sun tabloid UK - uses feed43' description = 'A Recipe for The Sun tabloid UK'
__author__ = 'Dave Asbury' __author__ = 'Dave Asbury'
# last updated 20/2/12 # last updated 7/4/12
language = 'en_GB' language = 'en_GB'
oldest_article = 1 oldest_article = 1
max_articles_per_feed = 15 max_articles_per_feed = 15
remove_empty_feeds = True remove_empty_feeds = True
no_stylesheets = True no_stylesheets = True
#auto_cleanup = True
#articles_are_obfuscated = True
masthead_url = 'http://www.thesun.co.uk/sol/img/global/Sun-logo.gif' masthead_url = 'http://www.thesun.co.uk/sol/img/global/Sun-logo.gif'
encoding = 'cp1251' encoding = 'UTF-8'
encoding = 'cp1252'
remove_empty_feeds = True remove_empty_feeds = True
remove_javascript = True remove_javascript = True
no_stylesheets = True no_stylesheets = True
extra_css = ''' 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;} 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;}
''' '''
preprocess_regexps = [ preprocess_regexps = [
(re.compile(r'<div class="foot-copyright".*?</div>', re.IGNORECASE | re.DOTALL), lambda match: '')] (re.compile(r'<div class="foot-copyright".*?</div>', re.IGNORECASE | re.DOTALL), lambda match: '')]
keep_only_tags = [ keep_only_tags = [
dict(name='h1'),dict(name='h2',attrs={'class' : 'medium centered'}), dict(name='h1'),dict(name='h2',attrs={'class' : 'medium centered'}),
dict(name='div',attrs={'class' : 'text-center'}), dict(name='div',attrs={'class' : 'text-center'}),
dict(name='div',attrs={'id' : 'bodyText'}) dict(name='div',attrs={'id' : 'bodyText'})
# dict(name='p') # dict(name='p')
] ]
remove_tags=[ remove_tags=[
#dict(name='head'), #dict(name='head'),
dict(attrs={'class' : ['mystery-meat-link','ltbx-container','ltbx-var ltbx-hbxpn','ltbx-var ltbx-nav-loop','ltbx-var ltbx-url']}), dict(attrs={'class' : ['mystery-meat-link','ltbx-container','ltbx-var ltbx-hbxpn','ltbx-var ltbx-nav-loop','ltbx-var ltbx-url']}),
dict(name='div',attrs={'class' : 'cf'}), dict(name='div',attrs={'class' : 'cf'}),
dict(attrs={'title' : 'download flash'}), dict(attrs={'title' : 'download flash'}),
dict(attrs={'style' : 'padding: 5px'}) dict(attrs={'style' : 'padding: 5px'})
] ]
feeds = [ feeds = [
(u'News','http://feed43.com/2517447382644748.xml'), #(u'News', u'http://www.thesun.co.uk/sol/homepage/news/rss'),
(u'Sport', u'http://feed43.com/4283846255668687.xml'), (u'News','http://feed43.com/2517447382644748.xml'),
(u'Bizarre', u'http://feed43.com/0233840304242011.xml'), (u'Sport', u'http://feed43.com/4283846255668687.xml'),
(u'Film',u'http://feed43.com/1307545221226200.xml'), (u'Bizarre', u'http://feed43.com/0233840304242011.xml'),
(u'Music',u'http://feed43.com/1701513435064132.xml'), (u'Film',u'http://feed43.com/1307545221226200.xml'),
(u'Sun Woman',u'http://feed43.com/0022626854226453.xml'), (u'Music',u'http://feed43.com/1701513435064132.xml'),
] (u'Sun Woman',u'http://feed43.com/0022626854226453.xml'),
]