from calibre.web.feeds.news import BasicNewsRecipe
import re


class Gildia(BasicNewsRecipe):
    title = u'Gildia.pl'
    __author__ = 'fenuks'
    description = u'Fantastyczny Portal Kulturalny - newsy, recenzje, galerie, wywiady. Literatura, film, gry komputerowe i planszowe, komiks, RPG, sklep. Nie lekceważ potęgi wyobraźni!'  # noqa
    cover_url = 'http://www.film.gildia.pl/_n_/portal/redakcja/logo/logo-gildia.pl-500.jpg'
    category = 'culture'
    cover_url = 'http://portal.gildia.pl/images/logo-main.png'
    language = 'pl'
    oldest_article = 8
    max_articles_per_feed = 100
    remove_empty_feeds = True
    no_stylesheets = True
    use_embedded_content = False
    ignore_duplicate_articles = {'title', 'url'}
    preprocess_regexps = [(re.compile(u'</?sup>'), lambda match: '')]
    ignore_duplicate_articles = {'title', 'url'}
    remove_tags = [dict(name='div', attrs={'class': [
                        'backlink', 'im_img', 'addthis_toolbox addthis_default_style', 'banner-bottom']})]
    keep_only_tags = [dict(name='div', attrs={'class': 'widetext'}), dict(name='article', attrs={'id': re.compile(r'post-\d+')})]
    feeds = [(u'Gry', u'http://www.gry.gildia.pl/rss'),
             (u'Literatura', u'http://www.literatura.gildia.pl/rss'),
             (u'Film', u'http://www.film.gildia.pl/rss'),
             (u'Horror', u'http://www.horror.gildia.pl/rss'),
             (u'Konwenty', u'http://www.konwenty.gildia.pl/rss'),
             (u'Plansz\xf3wki', u'http://www.planszowki.gildia.pl/rss'),
             (u'Manga i anime', u'http://www.manga.gildia.pl/rss'),
             (u'Star Wars', u'http://www.starwars.gildia.pl/rss'),
             (u'Techno', u'http://www.techno.gildia.pl/rss'),
             (u'Historia', u'http://www.historia.gildia.pl/rss'),
             (u'Magia', u'http://www.magia.gildia.pl/rss'),
             (u'Bitewniaki', u'http://www.bitewniaki.gildia.pl/rss'),
             (u'RPG', u'http://www.rpg.gildia.pl/rss'),
             (u'LARP', u'http://www.larp.gildia.pl/rss'),
             (u'Muzyka', u'http://www.muzyka.gildia.pl/rss'),
             (u'Nauka', u'http://www.nauka.gildia.pl/rss'),
             ]

    def skip_ad_pages(self, soup):
        content = soup.find('div', attrs={'class': 'news'})
        if content is None:
            return

        words = ('recenzj', 'zapowied', 'fragmen',
                 'relacj', 'wywiad', 'nominacj')
        document_title = soup.title.renderContents().decode('utf-8').lower()
        for word in words:
            if word in document_title:
                for link in content.findAll(name='a'):
                    if word in link['href'] or (link.string and word in link.string):
                        return self.index_to_soup(link['href'], raw=True)
        for tag in content.findAll(name='a', href=re.compile('/publicystyka/')):
            if 'Wi&#281;cej...' == tag.string:
                return self.index_to_soup(tag['href'], raw=True)

    def preprocess_html(self, soup):
        title = soup.title.renderContents().decode('utf-8').lower()
        for a in soup('a', href=True):
            if not a['href'].startswith('http'):
                if '/gry/' in a['href']:
                    a['href'] = 'http://www.gry.gildia.pl' + a['href']
                elif u'książk' in title or u'komiks' in title:
                    a['href'] = 'http://www.literatura.gildia.pl' + a['href']
                elif u'komiks' in title:
                    a['href'] = 'http://www.literatura.gildia.pl' + a['href']
                else:
                    a['href'] = 'http://www.gildia.pl' + a['href']
        return soup