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

...

Browse Source
This commit is contained in:
Kovid Goyal 2010-03-17 05:48:15 +05:30
parent 6fd9ed3be3
commit 2803d7f95f
1 changed files with 7 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
resources/recipes/elsevier.recipe
View File

@ -9,16 +9,16 @@ from calibre.web.feeds.news import BasicNewsRecipe
class Pagina12(BasicNewsRecipe):
title = 'Elsevier.nl'
__author__ = 'Darko Miletic'
description = 'News from Denmark'
description = 'News from Holland'
publisher = 'elsevier.nl'
category = 'news, politics, Denmark'
category = 'news, politics, Holland'
oldest_article = 2
max_articles_per_feed = 200
no_stylesheets = True
encoding = 'utf-8'
use_embedded_content = False
language = 'nl'
country = 'NL'
country = 'NL'
remove_empty_feeds = True
masthead_url = 'http://www.elsevier.nl/static/elsevier/stdimg/logo.gif'
extra_css = ' body{font-family: Arial,Helvetica,sans-serif } img{margin-bottom: 0.4em} '
@ -29,7 +29,7 @@ class Pagina12(BasicNewsRecipe):
, 'publisher' : publisher
, 'language' : language
}
keep_only_tags = dict(attrs={'id':'artikel_container'})
remove_tags_before = dict(attrs={'id':'breadcrumb_container'})
remove_tags_after = dict(attrs={'class':'author_link'})
@ -50,7 +50,7 @@ class Pagina12(BasicNewsRecipe):
,(u'Cultuur & Televisie' , u'http://www.elsevier.nl/web/RSS/Cultuur-Televisie-RSS.htm?output=xml')
,(u'Society' , u'http://www.elsevier.nl/web/RSS/Society-RSS.htm?output=xml' )
,(u'Internet&/Gadgets' , u'http://www.elsevier.nl/web/RSS/Internet-Gadgets-RSS.htm?output=xml' )
,(u'Comentaren' , u'http://www.elsevier.nl/web/RSS/Commentaren-RSS.htm?output=xml' )
,(u'Comentaren' , u'http://www.elsevier.nl/web/RSS/Commentaren-RSS.htm?output=xml' )
]
def print_version(self, url):
@ -58,8 +58,8 @@ class Pagina12(BasicNewsRecipe):
def get_article_url(self, article):
return article.get('guid', None).rpartition('?')[0]
def preprocess_html(self, soup):
for item in soup.findAll(style=True):
del item['style']
return soup
return soup