calibre/recipes/toi.recipe
2013-06-10 16:03:38 +05:30

59 lines
2.4 KiB
Plaintext

# vim:fileencoding=utf-8
from calibre.web.feeds.news import BasicNewsRecipe
from lxml import html
allowed_sections = {'Top Headlines', 'Opinion', 'Science', 'Education', 'US', 'Pakistan', 'India Business', 'Tech News', 'Cricket', 'Bollywood'}
class TimesOfIndia(BasicNewsRecipe):
title = u'Times of India Headlines'
language = 'en'
description = 'Headline news from the Indian daily Times of India'
__author__ = 'Kovid Goyal'
no_stylesheets = True
no_javascript = True
keep_only_tags = [dict(name='h1'), dict(id=['storydiv', 'contentarea'])]
remove_tags = [
dict(name='div', attrs={'class':['video_list', 'rightpart', 'clearfix mTop15', 'footer_slider', 'read_more', 'flR', 'hide_new']}),
dict(name='div', attrs={'id':[
'most_pop', 'relartstory', 'slidebox', 'tmpFbokk', 'twittersource',
'reportAbuseDiv', 'result', 'yahoobuzzsyn', 'fb-root']}),
dict(style='float:right;margin-left:5px;'),
]
def parse_index(self):
index = 'http://timesofindia.indiatimes.com/home/headlines'
raw = self.index_to_soup(index, raw=True)
root = html.fromstring(raw)
feeds = []
current_section = None
current_articles = []
toc = root.xpath('//div[@align="center"]/descendant::table[@class="cnt"]')[0]
for x in toc.xpath('descendant::*[name()="h3" or (name()="ul" and @class="content")]'):
if x.tag == 'h3':
if current_articles and current_section in allowed_sections:
feeds.append((current_section, current_articles))
current_section = html.tostring(x, method='text', encoding=unicode).strip()
current_articles = []
self.log(current_section)
else:
for a in x.xpath('descendant::li/descendant::a[@href]'):
title = html.tostring(a, method='text', encoding=unicode).strip()
url = a.get('href')
if url.startswith('/'):
url = 'http://timesofindia.indiatimes.com' + url
self.log(' ', title)
current_articles.append({'title':title, 'url':url})
self.log('')
if current_articles and current_section in allowed_sections:
feeds.append((current_section, current_articles))
return feeds