Update CNN

recipes/cnn.recipe

@@ -18,8 +18,9 @@ class CNN(BasicNewsRecipe):
     no_stylesheets = True
     use_embedded_content   = False
     oldest_article        = 15
-    #recursions = 1
-    #match_regexps = [r'http://sportsillustrated.cnn.com/.*/[1-9].html']
+    ignore_duplicate_articles = {'url'}
+    # recursions = 1
+    # match_regexps = [r'http://sportsillustrated.cnn.com/.*/[1-9].html']
     max_articles_per_feed = 25
 
     extra_css = '''
@@ -36,25 +37,18 @@ class CNN(BasicNewsRecipe):
         (re.compile(r'<style.*?</style>', re.DOTALL), lambda m: ''),
     ]
 
-    keep_only_tags = [dict(id=['cnnContentContainer', 'storycontent'])]
-    remove_tags = [
-            {'class':['cnn_strybtntools', 'cnn_strylftcntnt',
-                'cnn_strybtntools', 'cnn_strybtntoolsbttm', 'cnn_strybtmcntnt',
-                'cnn_strycntntrgt', 'hed_side', 'foot', 'cnn_strylftcntnt cnn_strylftcexpbx']},
-            {'class':['cnn_html_media_title_new', 'cnn_html_media_title_new cnn_html_media_title_none',
-                'cnnArticleGalleryCaptionControlText', 'articleGalleryNavContainer']},
-            {'id':['articleGalleryNav00JumpPrev', 'articleGalleryNav00Prev',
-                'articleGalleryNav00Next', 'articleGalleryNav00JumpNext']},
-            {'style':['display:none']},
-            dict(id=['ie_column']),
+    keep_only_tags = [
+        dict(id=['body-text', 'storycontent']),
+        dict(attrs={'class':['pg-headline', 'metadata']}),
+    ]
+    remove_tags = [
     ]
-
 
     feeds =  [
              ('Top News', 'http://rss.cnn.com/rss/cnn_topstories.rss'),
              ('World', 'http://rss.cnn.com/rss/cnn_world.rss'),
              ('U.S.', 'http://rss.cnn.com/rss/cnn_us.rss'),
-             #('Sports', 'http://rss.cnn.com/rss/si_topstories.rss'),
+             # ('Sports', 'http://rss.cnn.com/rss/si_topstories.rss'),
              ('Business', 'http://rss.cnn.com/rss/money_latest.rss'),
              ('Politics', 'http://rss.cnn.com/rss/cnn_allpolitics.rss'),
              ('Law', 'http://rss.cnn.com/rss/cnn_law.rss'),
@@ -67,9 +61,21 @@ class CNN(BasicNewsRecipe):
              ('Most Popular', 'http://rss.cnn.com/rss/cnn_mostpopular.rss')
              ]
 
+    def preprocess_html(self, soup):
+        body = soup.find('body')
+        for h2 in soup.findAll(attrs={'class':'pg-headline'}):
+            h2.extract()
+            body.insert(0, h2)
+        for img in soup.findAll('img', attrs={'data-src-medium':True}):
+            img['src'] = img['data-src-medium']
+        return soup
+
     def get_article_url(self, article):
         ans = BasicNewsRecipe.get_article_url(self, article)
-        return ans.partition('?')[0]
+        ans = ans.partition('?')[0]
+        if '.com/videos/' in ans:
+            ans = None
+        return ans
 
     def get_masthead_url(self):
         masthead = 'http://i.cdn.turner.com/cnn/.element/img/3.0/global/header/intl/hdr-globe-central.gif'