calibre/recipes/newsweek.recipe

77 lines
2.7 KiB
Plaintext

import string
from calibre.web.feeds.news import BasicNewsRecipe
class Newsweek(BasicNewsRecipe):
title = 'Newsweek'
__author__ = 'Kovid Goyal'
description = 'Weekly news and current affairs in the US'
language = 'en'
encoding = 'utf-8'
no_stylesheets = True
BASE_URL = 'http://www.newsweek.com'
INDEX = BASE_URL+'/topics.html'
keep_only_tags = dict(name='article', attrs={'class':'article-text'})
remove_tags = [dict(attrs={'data-dartad':True})]
remove_attributes = ['property']
def postprocess_html(self, soup, first):
for tag in soup.findAll(name=['article', 'header']):
tag.name = 'div'
return soup
def newsweek_sections(self):
soup = self.index_to_soup(self.INDEX)
for a in soup.findAll('a', title='Primary tag', href=True):
yield (string.capitalize(self.tag_to_string(a)),
self.BASE_URL+a['href'])
def newsweek_parse_section_page(self, soup):
for article in soup.findAll('article', about=True,
attrs={'class':'stream-item'}):
title = article.find(attrs={'property': 'dc:title'})
if title is None: continue
title = self.tag_to_string(title)
url = self.BASE_URL + article['about']
desc = ''
author = article.find({'property':'dc:creator'})
if author:
desc = u'by %s. '%self.tag_to_string(author)
p = article.find(attrs={'property':'dc:abstract'})
if p is not None:
for a in p.find('a'): a.extract()
desc += self.tag_to_string(p)
t = article.find('time', attrs={'property':'dc:created'})
date = ''
if t is not None:
date = u' [%s]'%self.tag_to_string(t)
self.log('\tFound article:', title, 'at', url)
self.log('\t\t', desc)
yield {'title':title, 'url':url, 'description':desc, 'date':date}
def parse_index(self):
sections = []
for section, shref in self.newsweek_sections():
self.log('Processing section', section, shref)
articles = []
soups = [self.index_to_soup(shref)]
na = soups[0].find('a', rel='next')
if na:
soups.append(self.index_to_soup(self.BASE_URL+na['href']))
for soup in soups:
articles.extend(self.newsweek_parse_section_page(soup))
if self.test and len(articles) > 1:
break
if articles:
sections.append((section, articles))
if self.test and len(sections) > 1:
break
return sections