diff --git a/recipes/mac_world_uk.recipe b/recipes/mac_world_uk.recipe index cf470578cc..5082e622dc 100644 --- a/recipes/mac_world_uk.recipe +++ b/recipes/mac_world_uk.recipe @@ -11,7 +11,6 @@ http://www.macworld.co.uk/ ''' from calibre.web.feeds.news import BasicNewsRecipe -from calibre.ptempfile import PersistentTemporaryFile class pcMag(BasicNewsRecipe): __author__ = 'Lorenzo Vigentini' @@ -25,55 +24,23 @@ class pcMag(BasicNewsRecipe): language = 'en_GB' timefmt = '[%a, %d %b, %Y]' - oldest_article = 15 max_articles_per_feed = 25 use_embedded_content = False recursion = 10 remove_javascript = True no_stylesheets = True - - temp_files = [] - articles_are_obfuscated = True - - def get_obfuscated_article(self, url): - br = self.get_browser() - br.open(url) - response = br.follow_link(url_regex='&print$', nr = 0) - html = response.read() - - self.temp_files.append(PersistentTemporaryFile('_fa.html')) - self.temp_files[-1].write(html) - self.temp_files[-1].close() - return self.temp_files[-1].name - - keep_only_tags = [ - dict(name='div', attrs={'id':'wrapper'}) - ] - - remove_tags = [ - dict(name='div', attrs={'class':'bannerContainer'}), - dict(name='p', attrs={'class':'breadcrumbs'}), - dict(name='ul', attrs={'id':'articleIconsList'}) - - ] - - remove_tags_after = [ - dict(name='p', attrs={'id':'articlePageList'}), - ] + auto_cleanup = True feeds = [ - (u'MacWorld Headlines', u'http://www.macworld.co.uk/rss/macworld.xml'), - (u'Reviews', u'http://www.macworld.co.uk/rss/reviews.xml'), - (u'Masterclass', u'http://www.macworld.co.uk/rss/masterclasses.xml'), - (u'MacWorld Team', u'http://www.macworld.co.uk/rss/blog8.xml'), - (u'Andy Ihnatko', u'http://www.macworld.co.uk/rss/blog7.xml'), - (u'Andy Penfold', u'http://www.macworld.co.uk/rss/blog11.xml'), - (u'Jonny Evans', u'http://www.macworld.co.uk/rss/blog1.xml'), - (u'Karen Haslam', u'http://www.macworld.co.uk/rss/blog4.xml'), - (u'Mark Hattersley', u'http://www.macworld.co.uk/rss/blog2.xml'), - (u'Nick Spence', u'http://www.macworld.co.uk/rss/blog12.xml'), - (u'Simon Iary', u'http://www.macworld.co.uk/rss/blog3.xml') + (u'MacWorld Headlines', u'http://www.macworld.co.uk/latest/rss'), + (u'Reviews', u'http://www.macworld.co.uk/news/rss'), + (u'Masterclass', u'http://www.macworld.co.uk/how-to/rss'), + (u'MacWorld Team', u'http://www.macworld.co.uk/review/rss'), + (u'Andy Ihnatko', u'http://www.macworld.co.uk/latest/iphone/rss'), + (u'Andy Penfold', u'http://www.macworld.co.uk/latest/ipad/rss'), + (u'Jonny Evans', u'http://www.macworld.co.uk/latest/mac/rss'), + (u'Karen Haslam', u'http://www.macworld.co.uk/latest/apple/rss') ] extra_css = '''