Update Technology Review (DE)

recipes/technology_review_de.recipe

@@ -1,24 +1,66 @@
-from calibre.web.feeds.recipes import BasicNewsRecipe
+#!/usr/bin/env python2
-class AdvancedUserRecipe1303841067(BasicNewsRecipe):
+# vim:fileencoding=utf-8
+from __future__ import unicode_literals, division, absolute_import, print_function
 
-    title          = u'Technology Review'
+__license__   = 'GPL v3'
-    __author__  = 'schuster'
+__copyright__ = '2010, Anton Gillert <atx at binaryninja.de>'
-    remove_tags_before = dict(id='keywords')
+
-    remove_tags_after  = dict(id='kommentar')
+'''
-    remove_tags = [dict(attrs={'class':['navi_oben_pvg', 'navi_oben_tarifr', 'navi_oben_itm', 'navi_oben_eve', 'navi_oben_whi', 'navi_oben_abo', 'navi_oben_shop', 'navi_top_logo', 'navi_top_abschnitt', 'first']}),
+Technology Review (deutsch) - heise.de/tr
-               dict(id=['footer', 'toolsRight', 'articleInline', 'navigation', 'archive', 'side_search', 'blog_sidebar', 'side_tool', 'side_index']),
+'''
-               dict(name=['script', 'noscript', 'style'])]
+
-    oldest_article = 4
+import re
-    max_articles_per_feed = 100
+from calibre.web.feeds.news import BasicNewsRecipe
-    no_stylesheets         = True
+
-    use_embedded_content   = False
+class TechnologyReviewDe(BasicNewsRecipe):
+    title       = 'Technology Review'
+    __author__  = 'Anton Gillert, schuster'
+    description = 'Technology news from Germany'
     language    = 'de'
+
+    oldest_article        = 14
+    max_articles_per_feed = 50
+    use_embedded_content  = False
+    no_stylesheets        = True
     remove_javascript     = True
 
+    masthead_url = 'http://1.f.ix.de/imgs/02/3/0/8/5/2/8/tr_logo-544bd18881c81263.png'
+
+    feeds = [
+        ('News', 'http://www.heise.de/tr/rss/news-atom.xml'),
+        ('Blog', 'http://www.heise.de/tr/rss/blog-atom.xml')
+    ]
+
+    keep_only_tags = [
+        dict(name='article')
+    ]
+
+    remove_tags = [
+        dict(name='nav'),
+        dict(name='figure', attrs={'class':'logo'}),
+        dict(name='hr')
+    ]
+
+    extra_css = '.bild_zentriert {font-size: 0.6em} \
+                 .source {font-size: 0.6em}'
+
+    def get_cover_url(self):
+        self.cover_url = ''
+        soup = self.index_to_soup('http://www.heise.de/tr/magazin/')
+        img = soup.find('img', alt=re.compile('Titelbild Technology Review'), src=True)
+        if img:
+            self.cover_url = 'http://www.heise.de' + img['src']
+        return self.cover_url
+
     def print_version(self, url):
         return url + '?view=print'
 
-
+    def preprocess_html(self, soup):
-    feeds          = [
+        # remove style attributes
-    (u'Technik News', u'http://www.heise.de/tr/news-atom.xml') ]
+        for item in soup.findAll(attrs={'style':True}):
-
+            del item['style']
+        # remove reference to article source
+        for p in soup.findAll('p'):
+            if 'URL dieses Artikels:' in self.tag_to_string(p):
+                p.extract()
+        return soup

recipes/tr.recipe

@@ -1,37 +0,0 @@
-__license__   = 'GPL v3'
-__copyright__ = '2010, Anton Gillert <atx at binaryninja.de>'
-
-'''
-Fetch Technology Review.
-'''
-from time import strftime
-from calibre.web.feeds.news import BasicNewsRecipe
-
-
-class TechnologyReviewDe(BasicNewsRecipe):
-
-    title = 'Technology Review'
-    description = 'Technology news from Germany'
-    __author__ = 'Anton Gillert'
-    use_embedded_content   = False
-    language = 'de'
-    timefmt = ' [%d %b %Y]'
-    max_articles_per_feed = 40
-    no_stylesheets = True
-
-    feeds =  [ ('Technology Review', 'http://www.heise.de/tr/news-atom.xml') ]
-
-    def print_version(self, url):
-        return url + '?view=print'
-
-    remove_tags = [dict(id='navi_top'),
-		   dict(id='navi_bottom'),
-		   dict(name='div', attrs={'class':'navi_top_logo'}),
-		   dict(name='img', attrs={'src':'/tr/icons/tr_logo2006.gif'}),
-		   dict(name='p', attrs={'class':'size80'})]
-    remove_tags_after = [dict(name='p', attrs={'class':'size80'})]
-
-    def get_cover_url(self):
-        return 'http://www.heise-medien.de/presseinfo/bilder/tr/' + strftime("%y/tr%m%Y.jpg")
-
-