import re
class LinuxFR(BasicNewsRecipe):
    title                 = 'LinuxFR'
    __author__            = 'Christophe Chailloleau'
    description           = u'Articles et journaux du site d\'actualit\xe9 autour du logiciel libre LinuxFR.org'
    language              = 'fr'
    encoding              = 'utf8'
    cover_url             = 'http://linuxfr.org/images/logos/linuxfr2_gnu_approved.png'
    oldest_article        = 1
    timefmt = ' [%a %d %b %Y]'
    max_articles_per_feed = 100
    use_embedded_content  = False
    no_stylesheets        = True
    remove_javascript     = True
    feeds                 = [('Articles', 'http://linuxfr.org/backend/news/rss20.rss'), ('Journaux', 'http://linuxfr.org/backend/journaux/rss20.rss')]
    category              = u'Linux, GNU, logiciel, libre, actualit\xe9, news, free, software, DLFP'
    remove_tags           = [dict(name='div', attrs={'class':['menubartop','menubar','leftcol','rightbox','footer','commentsreply','signature']}),
                             dict(name='div', attrs={'style':'clear:both'}),
                             dict(name='p', attrs={'class':'hautpage'}),
                             dict(name='a', attrs={'href':re.compile('#reply')}),
                             dict(name='p', text=re.compile(u'^\s*[\[\]]\s*$')),
                             dict(name='a', text=['[^]','[+]','[-]']),
                             dict(name='a', attrs={'rel':'tag'}),
                             dict(name='i'),
                             dict(name='h1', text=[' : ','Journal : ']),
                             dict(name='span', attrs={'class':'content-score'}),
                            ]
    extra_css             = "h1 { font-size: inherit; font-weight: bold; }"

    def preprocess_html(self, soup):
      for tag in soup.findAll(name='a',attrs={'href':re.compile('^\.\.')}):
        if tag.has_key('title'):
          tag['href']=tag['title']
          p=tag.parent
          if p.name=='li' and len(p.contents)==4:
            p.contents[3].extract()
        else: 
          p=tag.parent
          index=p.contents.index(tag)
          if len(tag.contents)>0:
            text=tag.contents[0]
          tag.extract()
          if not text is None:
            if text=='Lire le journal' or text=='Lire la suite' or text=='Lire les commentaires':
              p.extract()
            else:
              p.insert(index,text)
      first=soup.find(name='div', attrs={'class':'comments'})
      if not first is None:
        p=first.parent
        index=p.contents.index(first)
        p.insert(index,"<h1 style='font-style: italic;'>Commentaires</h1>")

      return soup