__license__ = 'GPL v3'
__copyright__ = '2012'
'''
lemonde.fr
'''
from calibre.web.feeds.recipes import BasicNewsRecipe


def classes(classes):
    q = frozenset(classes.split(' '))
    return dict(attrs={
        'class': lambda x: x and frozenset(x.split()).intersection(q)})


class LeMonde(BasicNewsRecipe):
    title = 'Le Monde'
    __author__ = 'veezh'
    description = u'Actualités'
    oldest_article = 1
    max_articles_per_feed = 100
    no_stylesheets = True
    use_embedded_content = False
    encoding = 'utf-8'
    publisher = 'lemonde.fr'
    category = 'news, France, world'
    language = 'fr'
    extra_css = '''
                img{max-width:100%}
                h1{font-size:1.2em !important; line-height:1.2em !important; }
                h2{font-size:1em !important; line-height:1em !important; }
                h3{font-size:1em !important; text-transform:uppercase !important; color:#666;}
                #photo{text-align:center !important; margin:10px 0 -8px;}
                #lgd{font-size:1em !important; line-height:1em !important;  font-style:italic; color:#333;} '''

    keep_only_tags = [
        dict(itemprop=['Headline', 'description']),
        classes('bloc_signature'),
        dict(itemprop=['articleBody']),
    ]
    remove_empty_feeds = True

    def preprocess_html(self, soup):
        for lgd in soup.findAll(id="lgd"):
            lgd.contents[-1].extract()
        for img in soup.findAll('img', attrs={'data-src': True}):
            img['src'] = img['data-src']
        return soup

    def get_article_url(self, article):
        url = article.get('guid', None)
        if '/chat/' in url or '.blog' in url or '/video/' in url or '/sport/' in url or '/portfolio/' in url or '/visuel/' in url:
            url = None
        return url

    feeds = [
        ('A la une', 'http://www.lemonde.fr/rss/une.xml'),
        ('International', 'http://www.lemonde.fr/rss/tag/international.xml'),
        ('Europe', 'http://www.lemonde.fr/rss/tag/europe.xml'),
        (u'Société', 'http://www.lemonde.fr/rss/tag/societe.xml'),
        ('Economie', 'http://www.lemonde.fr/rss/tag/economie.xml'),
        (u'Médias', 'http://www.lemonde.fr/rss/tag/actualite-medias.xml'),
        (u'Planète', 'http://www.lemonde.fr/rss/tag/planete.xml'),
        ('Culture', 'http://www.lemonde.fr/rss/tag/culture.xml'),
        ('Technologies', 'http://www.lemonde.fr/rss/tag/technologies.xml'),
        ('Livres', 'http://www.lemonde.fr/rss/tag/livres.xml'),

    ]

    def get_cover_url(self):
        cover_url = None
        soup = self.index_to_soup(
            'http://www.lemonde.fr/web/monde_pdf/0,33-0,1-0,0.html')
        link_item = soup.find('div', attrs={'class': 'pg-gch'})

        if link_item and link_item.img:
            cover_url = link_item.img['src']

        return cover_url