Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-07 10:14:46 -04:00
Code Issues Packages Projects Releases Wiki Activity

Fix #6302 (a change in the index-page broke the "Der Tagesspiegel"-recipe)

Browse Source
This commit is contained in:
Kovid Goyal 2010-07-26 09:06:39 -06:00
parent e2c048e23a
commit 2a39e43655
1 changed files with 10 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
resources/recipes/tagesspiegel.recipe
View File

@ -10,7 +10,7 @@ from calibre.web.feeds.news import BasicNewsRecipe
class TagesspiegelRSS(BasicNewsRecipe):
title = u'Der Tagesspiegel'
__author__ = 'ipaschke'
__author__ = 'Ingo Paschke'
language = 'de'
oldest_article = 7
max_articles_per_feed = 100
@ -40,24 +40,29 @@ class TagesspiegelRSS(BasicNewsRecipe):
dict(name='div', attrs={'class':["hcf-jump-to-comments","hcf-clear","hcf-magnify hcf-media-control"] }),
dict(name='span', attrs={'class':["hcf-mainsearch",] }),
dict(name='ul', attrs={'class':["hcf-tools"]}),
dict(name='ul', attrs={'class': re.compile('hcf-services')})
]
def parse_index(self):
soup = self.index_to_soup('http://www.tagesspiegel.de/zeitung/')
def feed_title(div):
return ''.join(div.findAll(text=True, recursive=False)).strip()
return ''.join(div.findAll(text=True, recursive=False)).strip() if div is not None else None
articles = {}
key = None
ans = []
maincol = soup.find('div', attrs={'class':re.compile('hcf-main-col')})
for div in soup.findAll(True, attrs={'class':['hcf-teaser', 'hcf-header', 'story headline']}):
for div in maincol.findAll(True, attrs={'class':['hcf-teaser', 'hcf-header', 'story headline']}):
if div['class'] == 'hcf-header':
try:
key = string.capwords(feed_title(div.em.a))
articles[key] = []
ans.append(key)
except:
continue
elif div['class'] == 'hcf-teaser' and getattr(div.contents[0],'name','') == 'h2':
a = div.find('a', href=True)
@ -84,3 +89,4 @@ class TagesspiegelRSS(BasicNewsRecipe):
return ans