Update Prospect Magazine

recipes/prospectmaguk.recipe

@@ -1,4 +1,4 @@
-#!/usr/bin/env  python2
+#!/usr/bin/env python2
 __copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
 
 __license__ = 'GPL v3'
@@ -7,8 +7,8 @@ __license__ = 'GPL v3'
 calibre recipe for prospectmagazine.co.uk (subscription)
 '''
 
-import re
 from calibre.web.feeds.recipes import BasicNewsRecipe
+from css_selectors import Select
 
 
 class ProspectMagUK(BasicNewsRecipe):
@@ -18,16 +18,13 @@ class ProspectMagUK(BasicNewsRecipe):
     timefmt = ' [%d %B %Y]'
     no_stylesheets = True
     publication_type = 'magazine'
-    masthead_url = 'http://www.prospectmagazine.co.uk/wp-content/themes/prospect/images/titleMain.jpg'
     category = 'news, UK'
     language = 'en_GB'
     max_articles_per_feed = 100
-    auto_cleanup = True
     needs_subscription = True
 
-    auto_cleanup_keep = '//div[@class="lead_image"]'
-
     INDEX = 'http://www.prospectmagazine.co.uk/issue/'
+    keep_only_tags = [dict(id='post_content')]
 
     def get_browser(self):
         br = BasicNewsRecipe.get_browser(self)
@@ -40,41 +37,19 @@ class ProspectMagUK(BasicNewsRecipe):
         return br
 
     def parse_index(self):
-        soup = self.index_to_soup(self.INDEX)
-        div = soup.find('div', id='cover_image')
-        if div is not None:
-            img = div.find('img', src=True)
-            if img is not None:
-                src = img['src']
-                if src.startswith('/'):
-                    src = 'http://www.prospectmagazine.co.uk' + src
-                self.cover_url = src
+        root = self.index_to_soup(self.INDEX, as_tree=True)
+        sel = Select(root)
+        for img in sel('.block_this_month .img_wrap img'):
+            self.cover_url = img.get('src').partition('?')[0]
         feeds = []
-        # loop through sections
-        for sect in soup.findAll('div', attrs={'class': 'sectionheading'}):
-            fname = self.tag_to_string(sect).replace('>', '').strip()
-            self.log('Found section', fname)
+        for h2 in sel('h2.block-title'):
+            current_section = self.tag_to_string(h2)
             articles = []
-
-            # note: can't just find siblings with class='post' because that will also
-            # grab all the articles belonging to the sections that follow.
-            for item in sect.findNextSiblings('div', attrs={'class': True}):
-                if 'post' not in item['class']:
-                    break
-                a = item.find('a', href=True)
-                if a is None:
-                    continue
-                url = a['href']
-                title = self.tag_to_string(a)
-                p = item.find('p')
-                desc = self.tag_to_string(p) if p is not None else ''
-                art = {'title': title, 'description': desc,
-                       'date': ' ', 'url': url}
-                p = item.find(attrs={'class': re.compile('author')})
-                self.log('\tFound article:', title, '::', url)
-                if p is not None:
-                    art['author'] = self.tag_to_string(p).strip()
-                articles.append(art)
-
-            feeds.append((fname, articles))
+            self.log('Found section:', current_section)
+            for div in sel('div.block_home_post', h2.getparent()):
+                for a in sel('div.title a[href]', div):
+                    articles.append({'title':self.tag_to_string(a), 'url':a.get('href')})
+                    self.log('\tFound article:', articles[-1]['title'])
+            if articles:
+                feeds.append((current_section, articles))
         return feeds