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 Times of India

Browse Source
This commit is contained in:
Kovid Goyal 2011-09-24 10:52:22 -06:00
parent 6bf872b5be
commit 0f138cd43c
1 changed files with 7 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
recipes/toi.recipe
View File

@ -9,11 +9,12 @@ class TimesOfIndia(BasicNewsRecipe):
max_articles_per_feed = 25
no_stylesheets = True
keep_only_tags = [{'class':['maintable12', 'prttabl']}]
remove_attributes = ['style']
keep_only_tags = [{'class':re.compile(r'maintable12|prttabl')}]
remove_tags = [
dict(style=lambda x: x and 'float' in x),
{'class':['prvnxtbg', 'footbdrin', 'bcclftr']},
]
{'class':re.compile('tabsintbgshow|prvnxtbg')},
{'id':['fbrecommend', 'relmaindiv']}
]
feeds = [
('Top Stories',
@ -41,6 +42,8 @@ class TimesOfIndia(BasicNewsRecipe):
]
def get_article_url(self, article):
# Times of India sometimes serves an ad page instead of the article,
# this code, detects and circumvents that
url = BasicNewsRecipe.get_article_url(self, article)
if '/0Ltimesofindia' in url:
url = url.partition('/0L')[-1]
@ -61,6 +64,3 @@ class TimesOfIndia(BasicNewsRecipe):
return url
def preprocess_html(self, soup):
return soup