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 Consumerist

Browse Source
This commit is contained in:
Kovid Goyal 2013-08-22 22:10:42 +05:30
parent 596f58bbcd
commit eb0eb79801
1 changed files with 7 additions and 22 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
recipes/consumerist.recipe
View File

@ -17,6 +17,8 @@ class Consumerist(BasicNewsRecipe):
no_stylesheets = True no_stylesheets = True
encoding = 'utf-8' encoding = 'utf-8'
use_embedded_content = False use_embedded_content = False
auto_cleanup = True
auto_cleanup_keep = '//div[@class="wp-caption alignleft"]'
language = 'en' language = 'en'
masthead_url = 'http://consumerist.com/css/images/footer_man.gif' masthead_url = 'http://consumerist.com/css/images/footer_man.gif'
extra_css = ''' extra_css = '''
@ -32,22 +34,5 @@ class Consumerist(BasicNewsRecipe):
, 'language' : language , 'language' : language
} }
remove_attributes = ['width','height']
#keep_only_tags = [dict(attrs={'class':['', 'category-breadcrumb']}),]
remove_tags_before = dict(name='h2')
remove_tags = [
#dict(name='iframe'),
dict(name='div', attrs={'class':['e-comments', 'more-about', 'entry-tags']}),
#dict(name='div', attrs={'id':['IEContainer', 'clickIncludeBox']}),
#dict(name='ul', attrs={'class':'article-tools'}),
#dict(name='ul', attrs={'class':'articleTools'}),
]
remove_tags_after = dict(attrs={'class':'e-body'})
feeds = [(u'Articles', u'http://consumerist.com/index.xml')] feeds = [(u'Articles', u'http://consumerist.com/index.xml')]
def preprocess_html(self, soup):
return self.adeify_images(soup)