from calibre.web.feeds.news import BasicNewsRecipe
import re
class PB_PL(BasicNewsRecipe):
    title          = u'Puls Biznesu'
    __author__        = 'fenuks'
    language       = 'pl'
    description = u'Puls Biznesu - biznes, ekonomia, giełda, inwestycje'
    category = u'newspaper'
    publication_type = u'newspaper'
    encoding = 'utf-8'
    #masthead_url = 'http://www.pb.pl/img/pb.png'
    oldest_article = 7
    max_articles_per_feed = 100
    no_stylesheets = True
    remove_empty_feeds = True
    ignore_duplicate_articles = {'title', 'url'}
    remove_tags_after = dict(name='div', attrs={'class':'news_content'})
    feeds          = [(u'Wszystkie', u'http://www.pb.pl/atom'), (u'Puls inwestora', u'http://pulsinwestora.pb.pl/atom'), (u'Puls Firmy', u'http://firma.pb.pl/atom'), (u'PB Weekend', u'http://weekend.pb.pl/atom'), (u'Forum MPS', u'http://forummsp.pb.pl/atom'), (u'Moto', u'http://moto.pb.pl/atom'), (u'Kariera i praca', u'http://kariera.pb.pl/atom'),(u'Nieruchomości', u'http://nieruchomosci.pb.pl/atom'), (u'Samorządy', u'http://samorzady.pb.pl/atom'), (u'Tech', u'http://tech.pb.pl/atom'), (u'Energetyka', u'http://energetyka.pb.pl/atom'), (u'Retailing', u'http://retailing.pb.pl/atom'), (u'Puls medycyny', u'http://pulsmedycyny.pl/atom'), (u'Logistyka', u'http://logistyka.pb.pl/atom')]

    def print_version(self, url):
        article_id = re.search(r'(?P<id>\d+,\d+)', url)
        if article_id:
            return 'http://www.pb.pl/actionprint/' + article_id.group('id')
        else:
            return url

    def get_cover_url(self):
        soup = self.index_to_soup('http://archiwum.pb.pl/')
        cover = soup.find(name='img', attrs={'class':'cover_picture'})
        self.cover_url= cover['src']
        return getattr(self, 'cover_url', self.cover_url)