calibre/recipes/cnetnews.recipe
2019-08-13 20:12:36 +05:30

103 lines
3.7 KiB
Plaintext

__license__ = 'GPL v3'
__copyright__ = '2009, Darko Miletic <darko.miletic at gmail.com>'
'''
Changelog:
2011-09-24
Changed cover (drMerry)
2011-10-13
Updated Cover (drMerry)
2014-03-28
Update by Armin Geller
news.cnet.com
further updated by Bonni Salles
My updates use the current rss feeds for CNet and eliminates extraneous areas and
sets the oldest_article to 1 as the feed generally has the present day and one day before.
You may want to set it to download daily for this reason.
'''
from calibre.web.feeds.news import BasicNewsRecipe
def classes(classes):
q = frozenset(classes.split(' '))
return dict(attrs={
'class': lambda x: x and frozenset(x.split()).intersection(q)})
class CnetNews(BasicNewsRecipe):
title = 'CNET News'
__author__ = 'Kovid Goyal'
description = 'Tech news and business reports by CNET News. Focused on information technology, core topics include computers, hardware, software, networking, and Internet media.' # noqa
publisher = 'CNET'
category = 'news, IT, USA'
encoding = 'utf-8'
language = 'en'
oldest_article = 7
max_articles_per_feed = 100
compress_news_images = True
ignore_duplicate_articles = {'title', 'url'}
remove_empty_feeds = True
use_embedded_content = False
no_stylesheets = True
recursions = 1
remove_tags = [
dict(name='div', attrs={'id': 'livefyreContainer'}),
dict(section=['tag', 'shortcodeRelatedLinks']),
dict(attrs={'class': ['ad-inline-product-carousel-top',
'author-social', 'row controls', 'launchGallery', 'topWrap']}),
dict(name='a', attrs={'class': 'subHead',
'data-component': 'imageGalleryModal'}),
dict(attrs={'data-component': 'sharebar'}),
dict(name=['link', 'meta']),
classes('playerControls video share-button'),
]
keep_only_tags = [
dict(name='h1'),
dict(section='author'),
dict(id=["article-body", 'cnetReview']),
dict(attrs={'class': 'deal-content'}),
]
feeds = [
# (u'All of CNET', u'http://www.cnet.com/rss/all/'),
(u'CNET News', u'http://www.cnet.com/rss/news/'),
(u'CNET Reviews', u'http://www.cnet.com/rss/reviews/'),
(u'CNET How To', u'http://www.cnet.com/rss/how-to/'),
(u'CNET iPhone Update', u'http://www.cnet.com/rss/iphone-update/'),
(u'CNET Crave', u'http://www.cnet.com/rss/crave/'),
(u'CNET Car Tech', u'http://www.cnet.com/rss/car-tech/'),
(u'CNET Android Update', u'http://www.cnet.com/rss/android-update/'),
(u'CNET Gaming', u'http://www.cnet.com/rss/gaming/'),
(u'CNET Cheapskate', u'http://www.cnet.com/rss/cheapskate/'),
]
def get_article_url(self, article):
ans = BasicNewsRecipe.get_article_url(self, article)
if ans and ('cnet.com/videos/' in ans or 'cnet.com/pictures/' in ans):
ans = None
return ans
def is_link_wanted(self, url, tag):
return tag.findParent(attrs={'class': 'pageNav'}) is not None
def preprocess_html(self, soup):
for attr in 'data-original data-src'.split():
for img in soup.findAll('img', attrs={attr: True}):
img['src'] = img[attr].strip()
return soup
def postprocess_html(self, soup, first_fetch):
for div in soup.findAll(attrs={'class': 'pagination'}):
div.extract()
for div in soup.findAll(attrs={'data-update-area': 'pagination'}):
div.extract()
for h1 in soup.findAll(itemprop='headline'):
h1.extract()
if first_fetch:
soup.find('body').insert(1, h1)
for img in soup.findAll('img'):
img['height'] = img['width'] = ''
return soup