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

Fix Harpers

Browse Source
This commit is contained in:
Kovid Goyal 2012-12-07 15:32:19 +05:30
parent db692f9f21
commit 7580d9c1c0
1 changed files with 14 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
recipes/harpers.recipe
View File

@ -16,6 +16,7 @@ class Harpers(BasicNewsRecipe):
max_articles_per_feed = 100 max_articles_per_feed = 100
no_stylesheets = True no_stylesheets = True
use_embedded_content = False use_embedded_content = False
auto_cleanup = True
conversion_options = { conversion_options = {
'comment' : description 'comment' : description
@ -31,14 +32,14 @@ class Harpers(BasicNewsRecipe):
.caption{font-family:Verdana,sans-serif;font-size:x-small;color:#666666;} .caption{font-family:Verdana,sans-serif;font-size:x-small;color:#666666;}
''' '''
keep_only_tags = [ dict(name='div', attrs={'id':'cached'}) ] #keep_only_tags = [ dict(name='div', attrs={'id':'cached'}) ]
remove_tags = [ #remove_tags = [
dict(name='table', attrs={'class':['rcnt','rcnt topline']}) #dict(name='table', attrs={'class':['rcnt','rcnt topline']})
,dict(name=['link','object','embed','meta','base']) #,dict(name=['link','object','embed','meta','base'])
] #]
remove_attributes = ['width','height'] #remove_attributes = ['width','height']
feeds = [(u"Harper's Magazine", u'http://www.harpers.org/rss/frontpage-rss20.xml')] feeds = [(u"Harper's Magazine", u'http://harpers.org/feed/')]
def get_cover_url(self): def get_cover_url(self):
cover_url = None cover_url = None
@ -49,9 +50,9 @@ class Harpers(BasicNewsRecipe):
cover_url = 'http://harpers.org' + link_item['src'] cover_url = 'http://harpers.org' + link_item['src']
return cover_url return cover_url
def preprocess_html(self, soup): #def preprocess_html(self, soup):
for item in soup.findAll(style=True): #for item in soup.findAll(style=True):
del item['style'] #del item['style']
for item in soup.findAll(xmlns=True): #for item in soup.findAll(xmlns=True):
del item['xmlns'] #del item['xmlns']
return soup #return soup