Workaround for broken XHTML in some Washington Post articles

This commit is contained in:
Kovid Goyal 2010-01-29 18:25:41 -07:00
parent bb11c80871
commit a5f13fa593

View File

@ -46,3 +46,10 @@ class WashingtonPost(BasicNewsRecipe):
div['style'] = '' div['style'] = ''
return soup return soup
def preprocess_html(self, soup):
for tag in soup.findAll('font'):
if tag.has_key('size'):
if tag['size'] == '+2':
if tag.b:
return soup
return None