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 #5403 (Articles to PC Mag not downloaded)

Browse Source
This commit is contained in:
Kovid Goyal 2010-04-29 18:30:44 -06:00
parent 6d5d22d692
commit 54e524c7e1
1 changed files with 10 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
resources/recipes/pc_mag.recipe
View File

@ -9,8 +9,9 @@ __description__ = 'PCMag (www.pcmag.com) delivers authoritative, labs-based comp
'''
http://www.pcmag.com/
'''
import re
from calibre.web.feeds.news import BasicNewsRecipe
from calibre.ebooks.BeautifulSoup import Comment
class pcMag(BasicNewsRecipe):
__author__ = 'Lorenzo Vigentini'
@ -33,9 +34,6 @@ class pcMag(BasicNewsRecipe):
remove_javascript = True
no_stylesheets = True
keep_only_tags = [
dict(name='div', attrs={'id':'articleContent'})
]
feeds = [
(u'Tech Commentary from the Editors of PC Magazine', u'http://rssnewsapps.ziffdavis.com/PCMAG_commentary.xml'),
@ -49,8 +47,13 @@ class pcMag(BasicNewsRecipe):
(u'Technology News from Ziff Davis', u'http://rssnewsapps.ziffdavis.com/pcmagbreakingnews.xml')
]
keep_only_tags = [dict(attrs={'class':'content-page'})]
remove_tags = [
dict(name='div', attrs={'id':['microAd','intellitxt','articleDeckTalkback','inlineDigg','underArticleLinks','w_talkback']}),
dict(name='span', attrs={'id':['highlights_content','yahooBuzzBadge-48558872521263350499378']})
]
dict(attrs={'class':['control-side','comment','highlights_content','btn-holder','subscribe-panel',
'grey-box comments-box']}),
dict(id=['inlineDigg']),
dict(text=lambda text:isinstance(text, Comment)),
dict(name='img', width='1'),
]
preprocess_regexps = [(re.compile(r"<img '[^']+?'"), lambda m : '<img ')]