calibre/recipes/grantland.recipe
2019-09-08 14:07:59 +05:30

80 lines
2.5 KiB
Plaintext

import re
from calibre.web.feeds.news import BasicNewsRecipe
class GrantLand(BasicNewsRecipe):
title = u"Grantland"
description = 'Writings on Sports & Pop Culture'
language = 'en'
__author__ = 'barty on mobileread.com forum'
max_articles_per_feed = 100
no_stylesheets = True
# auto_cleanup is too aggressive sometimes and we end up with blank
# articles
auto_cleanup = False
timefmt = ' [%a, %d %b %Y]'
oldest_article = 90
cover_url = 'http://cdn0.sbnation.com/imported_assets/740965/blog_grantland_grid_3.jpg'
masthead_url = 'http://a1.espncdn.com/prod/assets/grantland/grantland-logo.jpg'
INDEX = 'http://www.grantland.com'
CATEGORIES = [
# comment out second line if you don't want older articles
# (user friendly name, url suffix, max number of articles to load)
('Today in Grantland', '', 20),
('In Case You Missed It', 'incaseyoumissedit', 35),
]
remove_tags = [
{'name': ['style', 'aside', 'nav', 'footer', 'script']},
{'name': 'h1', 'text': 'Grantland'},
{'id': ['header', 'col-right']},
{'class': ['connect_widget']},
{'name': 'section', 'class': re.compile(r'\b(ad|module)\b')},
]
preprocess_regexps = [
# remove blog banners
(re.compile(r'<a href="/blog/(?:(?!</a>).)+</a>',
re.DOTALL | re.IGNORECASE), lambda m: ''),
]
def parse_index(self):
feeds = []
seen_urls = set()
for category in self.CATEGORIES:
(cat_name, tag, max_articles) = category
self.log('Reading category:', cat_name)
articles = []
page = "%s/%s" % (self.INDEX, tag)
soup = self.index_to_soup(page)
main = soup.find('div', id='col-main')
if main is None:
main = soup
for tag in main.findAll('a', href=re.compile(r'(story|post)/_/id/\d+')):
url = tag['href']
if url in seen_urls:
continue
title = tag.string
# blank title probably means <a href=".."><img /></a>. skip
if not title:
continue
self.log('\tFound article:', title)
self.log('\t', url)
articles.append({'title': title, 'url': url})
seen_urls.add(url)
if len(articles) >= max_articles:
break
if articles:
feeds.append((cat_name, articles))
return feeds