#!/usr/bin/env python __license__ = 'GPL v3' __author__ = 'Lorenzo Vigentini' __copyright__ = '2009, Lorenzo Vigentini ' __version__ = 'v1.01' __date__ = '14, January 2010' __description__ = 'Macworld is a publication of IDG Communication in the UK specifically on the Apple Mac.' ''' http://www.macworld.co.uk/ ''' from calibre.web.feeds.news import BasicNewsRecipe from calibre.ptempfile import PersistentTemporaryFile class pcMag(BasicNewsRecipe): __author__ = 'Lorenzo Vigentini' description = 'Macworld is a publication of IDG Communication in the UK specifically on the Apple Mac.' cover_url = 'http://media.macworld.co.uk/images/masthead.jpg' title = 'Mac World UK ' publisher = 'IDG Communication' category = 'Apple, Mac, computing, product reviews, UK' 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'}), ] 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') ] extra_css = ''' h1 {color:#0066CC;font-family:Arial,Helvetica,sans-serif; font-size:20px; font-size-adjust:none; font-stretch:normal; font-style:normal; font-variant:normal; font-weight:bold; line-height:20px;} h2 {color:#4D4D4D;font-family:Arial,Helvetica,sans-serif; font-size:16px; font-size-adjust:none; font-stretch:normal; font-style:normal; font-variant:normal; font-weight:bold; line-height:16px; } h3 {color:#4D4D4D;font-family:Arial,Helvetica,sans-serif; font-size:15px; font-size-adjust:none; font-stretch:normal; font-style:normal; font-variant:normal; font-weight:bold; line-height:15px;} h4 {color:#333333; font-family:Arial,Helvetica,sans-serif;font-size:13px; font-size-adjust:none; font-stretch:normal; font-style:normal; font-variant:normal; font-weight:bold; line-height:13px; } h5 {color:#333333; font-family:Arial,Helvetica,sans-serif; font-size:11px; font-size-adjust:none; font-stretch:normal; font-style:normal; font-variant:normal; font-weight:bold; line-height:11px; text-transform:uppercase;} p.authorCredit {-x-system-font:none;font-family:Arial,sans-serif;font-size:10pt;font-size-adjust:none;font-stretch:normal;font-style:normal;font-variant:normal;font-weight:normal;line-height:1.1em;} p.date {font-size:10pt;margin-bottom:0;} img {align:left;} '''