calibre/recipes/1843.recipe

54 lines
1.7 KiB
Python

#!/usr/bin/env python
# vim:fileencoding=utf-8
# License: GPLv3 Copyright: 2016, Kovid Goyal <kovid at kovidgoyal.net>
from __future__ import absolute_import, division, print_function, unicode_literals
from calibre.web.feeds.recipes import BasicNewsRecipe
def classes(classes):
q = frozenset(classes.split(' '))
return dict(attrs={
'class': lambda x: x and frozenset(x.split()).intersection(q)})
class E1843(BasicNewsRecipe):
title = '1843'
__author__ = 'Kovid Goyal'
description = 'The ideas, culture and lifestyle magazine from The Economist'
language = 'en_GB'
no_stylesheets = True
remove_javascript = True
encoding = 'utf-8'
keep_only_tags = [
dict(id='content')
]
remove_tags = [
classes('advert ad ds-share-list article__wordmark related-articles newsletter-signup')
]
def parse_index(self):
soup = self.index_to_soup('https://economist.com/1843')
ans = []
for a in soup.findAll(**classes('headline-link')):
url = a['href']
if url.startswith('/'):
url = 'https://economist.com' + url
title = self.tag_to_string(a)
self.log(title, ' at ', url)
desc = ''
d = a.parent.findNextSibling(itemprop='description')
if d is not None:
desc = self.tag_to_string(d)
ans.append({'title': title, 'url': url, 'description': desc})
return [('Articles', ans)]
def postprocess_html(self, soup, *a):
main = soup.find(id='content')
header = soup.find(**classes('article__header'))
header.extract()
main.insert(0, header)
return soup