__license__ = 'GPL v3' __copyright__ = '2010, Darko Miletic ' ''' emg.rs/en/news ''' from calibre.web.feeds.news import BasicNewsRecipe class emportal_en(BasicNewsRecipe): title = 'Ekonom:east News' __author__ = 'Darko Miletic' description = 'Daily business news from Serbia.' publisher = 'Ekonom:east Media Group' category = 'Business, SEE, Serbia, Belgrade, news, Ekonomist, EMportal' oldest_article = 2 max_articles_per_feed = 200 no_stylesheets = True encoding = 'utf8' use_embedded_content = False language = 'en' remove_empty_feeds = True masthead_url = 'http://www.emg.rs/img/emportal-rss.png' extra_css = ' body{font-family: Arial,Helvetica,sans-serif } ' conversion_options = { 'comment' : description , 'tags' : category , 'publisher' : publisher , 'language' : language } remove_tags = [dict(attrs={'class':['text-share']})] keep_only_tags = [dict(attrs={'class':'text'})] remove_tags_after = dict(attrs={'class':'text-share'}) remove_attributes = ['width','height'] feeds = [(u'Serbia' , u'http://www.emg.rs/en/news/serbia/rss.xml')] def print_version(self, url): return url.replace('.html','.print.html') def preprocess_html(self, soup): for item in soup.findAll(style=True): del item['style'] return soup