__license__ = 'GPL v3' __copyright__ = '2010, Larry Chan ' ''' oriental daily ''' from calibre.web.feeds.recipes import BasicNewsRecipe class OrientalDaily(BasicNewsRecipe): title = 'Oriental Dailly' __author__ = 'Larry Chan' description = 'News from HK' oldest_article = 2 max_articles_per_feed = 100 simultaneous_downloads = 5 no_stylesheets = True use_embedded_content = False encoding = 'utf8' publisher = 'Oriental Daily' category = 'news, HK, world' language = 'zh' publication_type = 'newsportal' extra_css = ' body{ font-family: Verdana,Helvetica,Arial,sans-serif } .introduction{font-weight: bold} .story-feature{display: block; padding: 0; border: 1px solid; width: 40%; font-size: small} .story-feature h2{text-align: center; text-transform: uppercase} ' # noqa conversion_options = { 'comments': description, 'tags': category, 'language': language, 'publisher': publisher, 'linearize_tables': True } masthead_url = 'https://upload.wikimedia.org/wikipedia/commons/4/4f/Orisun.JPG' remove_tags_after = dict(id='bottomNavCTN') keep_only_tags = [ dict(name='div', attrs={'id': ['leadin', 'contentCTN-right']}) ] remove_tags = [ dict(name='div', attrs={ 'class': ['largeAdsCTN', 'contentCTN-left', 'textAdsCTN', 'footerAds clear']}), dict(name='div', attrs={'id': ['articleNav']}) ] remove_attributes = ['width', 'height', 'href'] summary_length = 0 feeds = [(u'Oriental Daily', u'http://orientaldaily.on.cc/rss/news.xml')]