Just hacked together this recipe to get the non-multimedia parts of the new journalistic project called
Animal Político by the same guys of the
@pajaropolitico twitter user. (In case you don't know them and the name isn't obvious enough, this is mostly political news and opinion, with a very interactive approach)
Posting here in case anyone finds useful.
You can find the latest version in my github:
https://github.com/leamsi/Animal-Pol...olitico.recipe
Code:
#!/usr/bin/python
# encoding: utf-8
from calibre.web.feeds.news import BasicNewsRecipe
class AdvancedUserRecipe1290663986(BasicNewsRecipe):
title = u'Animal Pol\u00EDtico'
publisher = u'Animal Pol\u00EDtico'
category = u'News, Mexico'
description = u'Noticias Pol\u00EDticas'
masthead_url = 'http://www.animalpolitico.com/wp-content/themes/animal_mu/images/logo.png'
oldest_article = 1
max_articles_per_feed = 100
language = 'es'
#feeds = [(u'Animal Politico', u'http://www.animalpolitico.com/feed/')]
remove_tags_before = dict(name='div', id='main')
remove_tags = [dict(name='div', attrs={'class':'fb-like-button'})]
keep_only_tags = [dict(name='h1', attrs={'class':'entry-title'}),
dict(name='div', attrs={'class':'entry-content'})]
remove_javascript = True
INDEX = 'http://www.animalpolitico.com/'
def generic_parse(self, soup):
articles = []
for entry in soup.findAll(lambda tag: tag.name == 'li' and tag.has_key('class') and tag['class'].find('hentry') != -1): #soup.findAll('li', 'hentry'):
article_url = entry.a['href'] + '?print=yes'
article_title= entry.find('h3', 'entry-title')
article_title= self.tag_to_string(article_title)
article_date = entry.find('span', 'the-time')
article_date = self.tag_to_string(article_date)
article_desc = self.tag_to_string(entry.find('p'))
#print 'Article:',article_title, article_date,article_url
#print entry['class']
articles.append({'title' : article_title,
'date' : article_date,
'description' : article_desc,
'url' : article_url})
# Avoid including the multimedia stuff.
if entry['class'].find('last') != -1:
break
return articles
def plumaje_parse(self, soup):
articles = []
blogs_soup = soup.find(lambda tag: tag.name == 'ul' and tag.has_key('class') and tag['class'].find('bloglist-fecha') != -1)
for entry in blogs_soup.findAll('li'):
article_title = entry.p
article_url = article_title.a['href'] + '?print=yes'
article_date = article_title.nextSibling
article_title = self.tag_to_string(article_title)
article_date = self.tag_to_string(article_date).replace(u'Last Updated: ', '')
article_desc = self.tag_to_string(entry.find('h4'))
#print 'Article:',article_title, article_date,article_url
articles.append({'title' : article_title,
'date' : article_date,
'description' : article_desc,
'url' : article_url})
return articles
def boca_parse(self, soup):
articles = []
for entry in soup.findAll(lambda tag: tag.name == 'div' and tag.has_key('class') and tag['class'].find('hentry') != -1): #soup.findAll('li', 'hentry'):
article_title= entry.find('h2', 'entry-title')
article_url = article_title.a['href'] + '?print=yes'
article_title= self.tag_to_string(article_title)
article_date = entry.find('span', 'entry-date')
article_date = self.tag_to_string(article_date)
article_desc = self.tag_to_string(entry.find('div', 'entry-content'))
#print 'Article:',article_title, article_date,article_url
#print entry['class']
articles.append({'title' : article_title,
'date' : article_date,
'description' : article_desc,
'url' : article_url})
# Avoid including the multimedia stuff.
if entry['class'].find('last') != -1:
break
return articles
def parse_index(self):
gobierno_soup = self.index_to_soup(self.INDEX+'gobierno/')
congreso_soup = self.index_to_soup(self.INDEX+'congreso/')
seguridad_soup = self.index_to_soup(self.INDEX+'seguridad/')
comunidad_soup = self.index_to_soup(self.INDEX+'comunidad/')
plumaje_soup = self.index_to_soup(self.INDEX+'plumaje/')
la_boca_del_lobo_soup = self.index_to_soup(self.INDEX+'category/la-boca-del-lobo/')
gobierno_articles = self.generic_parse(gobierno_soup)
congreso_articles = self.generic_parse(congreso_soup)
seguridad_articles = self.generic_parse(seguridad_soup)
comunidad_articles = self.generic_parse(comunidad_soup)
plumaje_articles = self.plumaje_parse(plumaje_soup)
la_boca_del_lobo_articles = self.boca_parse(la_boca_del_lobo_soup)
return [ (u'Gobierno', gobierno_articles), (u'Congreso', congreso_articles), (u'Seguridad', seguridad_articles),
(u'Comunidad', comunidad_articles), (u'Plumaje', plumaje_articles), (u'La Boca del Lobo', la_boca_del_lobo_articles), ]
Btw, licensed under the
WTFPL (so I honestly don't care what you do with it).