__license__ = 'GPL v3' __copyright__ = '2010, Darko Miletic ' ''' gamasutra.com ''' from calibre.web.feeds.news import BasicNewsRecipe class Gamasutra(BasicNewsRecipe): title = 'Gamasutra News' __author__ = 'Darko Miletic' description = 'The Art and Business of Making Games' publisher = 'Gamasutra' category = 'news, games, IT' oldest_article = 2 max_articles_per_feed = 200 no_stylesheets = True encoding = 'cp1252' use_embedded_content = False language = 'en' remove_empty_feeds = True masthead_url = 'http://www.gamasutra.com/images/gamasutra_logo.gif' extra_css = ' body{font-family: Verdana,Arial,Helvetica,sans-serif } img{margin-bottom: 0.4em} .newsTitle{font-size: xx-large; font-weight: bold} ' conversion_options = { 'comment' : description , 'tags' : category , 'publisher' : publisher , 'language' : language , 'linearize_tables' : True } remove_tags = [dict(attrs={'class':['relatedNews','adBox']})] keep_only_tags = [dict(attrs={'class':['newsTitle','newsAuth','newsDate','newsText']})] remove_attributes = ['width','height'] feeds = [(u'News', u'http://feeds.feedburner.com/GamasutraNews')] def get_article_url(self, article): return article.get('guid', None) def preprocess_html(self, soup): for item in soup.findAll(style=True): del item['style'] return self.adeify_images(soup)