[malah]

Hello, since 2 days the printed version of mediapart have a small change and that break the recipe. I haven't find how to use it simply, and that because, the new recipe does not use the printed version, dont know if that the best way, but it work.

__license__   = 'GPL v3'
__copyright__ = '2009, Mathieu Godlewski <mathieu at godlewski.fr>; 2010-2012, Louis Gesbert <meta at antislash dot info>; 2013, Malah <malah at neuf dot fr>'
'''
Mediapart
'''

__author__ = '2009, Mathieu Godlewski <mathieu at godlewski.fr>; 2010-2012, Louis Gesbert <meta at antislash dot info>; 2013, Malah <malah at neuf dot fr>'

from calibre.ebooks.BeautifulSoup import BeautifulSoup, Tag
from calibre.web.feeds.news import BasicNewsRecipe

class Mediapart(BasicNewsRecipe):
    title          = 'Mediapart'
    __author__ = 'Mathieu Godlewski, Louis Gesbert, Malah'
    description = 'Global news in french from news site Mediapart'
    oldest_article = 7
    language = 'fr'
    needs_subscription = True
    max_articles_per_feed = 50

    use_embedded_content = False
    no_stylesheets = True

    masthead_url       = 'https://upload.wikimedia.org/wikipedia/fr/2/23/Mediapart.png'
    cover_url = 'http://static.mediapart.fr/files/pave_mediapart.jpg'

    feeds =  [
        ('Les articles', 'http://www.mediapart.fr/articles/feed'),
    ]

# -- full-page-version

    conversion_options = { 'smarten_punctuation' : True }

    keep_only_tags = [ 
        dict(name='div', attrs={'class':'col-left fractal-desktop fractal-10-desktop collapse-7-desktop fractal-tablet fractal-6-tablet collapse-4-tablet'}),
        dict(name='div', attrs={'id':'pageFirstContent'})
    ]
    remove_tags = [ 
        dict(name='div', attrs={'id':'lire-aussi'}),
        dict(name='div', attrs={'class':'col-right-content'})
    ]

    def print_version(self, url):
        raw = self.browser.open(url).read()
        soup = BeautifulSoup(raw.decode('utf8', 'replace'))
        link = soup.find('a', {'href':re.compile('^.*?onglet=full$')})
        if link is None:
            return None
        return link['href']

# -- Handle login

    def get_browser(self):
        br = BasicNewsRecipe.get_browser(self)
        if self.username is not None and self.password is not None:
            br.open('http://blogs.mediapart.fr/editions/guide-du-coordonnateur-d-edition')
            br.select_form(nr=1)
            br['name'] = self.username
            br['pass'] = self.password
            br.submit()
        return br

    def preprocess_html(self, soup):
        for title in soup.findAll('p', {'class':'titre_page'}):
            title.name = 'h3'
        for legend in soup.findAll('span', {'class':'legend'}):
            legend.insert(0, Tag(soup, 'br', []))
            legend.name = 'small'
        return soup

Or Download the recipe : https://www.dropbox.com/s/jwugf1oym5...0130725.recipe