__license__ = 'GPL v3'
from calibre.web.feeds.news import BasicNewsRecipe

class FDBPl(BasicNewsRecipe):
    title          = u'Fdb.pl'
    __author__        = 'fenuks'
    description   = u'Wiadomości ze świata filmu, baza danych filmowych, recenzje, zwiastuny, boxoffice.'
    category       = 'film'
    #publication_type = ''
    language       = 'pl'
    #encoding = ''
    extra_css = '.options-left > li {display: inline;} em {display: block;}'
    cover_url = 'http://fdb.pl/assets/fdb2/logo.png'
    #masthead_url = ''
    use_embedded_content = False
    oldest_article = 7
    max_articles_per_feed = 100
    no_stylesheets = True
    remove_empty_feeds = True
    remove_javascript = True
    remove_attributes = ['style', 'font']
    ignore_duplicate_articles = {'title', 'url'}

    keep_only_tags = [dict(attrs={'class':'news-item news-first'})]
    remove_tags = [dict(attrs={'class':['dig dig-first', 'ads clearfix', 'comments']})]
    #remove_tags_after = dict()
    #remove_tags_before = dict()
    feeds = []

    def parse_index(self):
         feeds = []
         feeds.append((u'Wiadomości', self.get_articles('http://fdb.pl/wiadomosci?page={0}', 2)))
         return feeds

    def get_articles(self, url, pages=1):
        articles = []
        for nr in range(1, pages+1):
            soup = self.index_to_soup(url.format(nr))
            for tag in soup.findAll(attrs={'class':'news-item clearfix'}):
                node = tag.find('h2')
                title = node.a.string
                url = 'http://fdb.pl' + node.a['href']
                date = ''
                articles.append({'title' : title,
                       'url'   : url,
                       'date'  : date,
                       'description' : ''
                        })
        return articles