Unify the two Atlantic recipes

recipes/atlantic.recipe

@@ -3,13 +3,14 @@ from __future__ import unicode_literals
 __license__ = 'GPL v3'
 __copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
 '''
-theatlantic.com
+Recipe for web and magazine versions of the The Atlantic
 '''
-import html5lib
-from lxml import html
 from calibre.web.feeds.news import BasicNewsRecipe
 
 
+web_version = False
+
+
 def classes(classes):
     q = frozenset(classes.split(' '))
     return dict(
@@ -19,10 +20,15 @@ def classes(classes):
 
 class TheAtlantic(BasicNewsRecipe):
 
+    if web_version:
+        title = 'TheAtlantic.com'
+        description = 'News and editorial about politics, culture, entertainment, tech, etc. Contains many articles not seen in The Atlantic magazine'
+    else:
         title = 'The Atlantic'
-    __author__ = 'Kovid Goyal'
         description = 'Current affairs and politics focussed on the US'
         INDEX = 'http://www.theatlantic.com/magazine/'
+
+    __author__ = 'Kovid Goyal'
     language = 'en'
     encoding = 'utf-8'
 
@@ -71,21 +77,39 @@ class TheAtlantic(BasicNewsRecipe):
         br.set_cookie('inEuropeanUnion', '0', '.theatlantic.com')
         return br
 
-    def preprocess_raw_html(self, raw, url):
-        return html.tostring(
-            html5lib.parse(raw, treebuilder='lxml', namespaceHTMLElements=False),
-            method='html',
-            encoding=unicode
-        )
-
-    def print_version(self, url):
-        return url + '?single_page=true'
-
     def preprocess_html(self, soup):
         for img in soup.findAll('img', attrs={'data-srcset': True}):
             img['src'] = img['data-srcset'].split()[0]
         return soup
 
+    def print_version(self, url):
+        return url.partition('?')[0] + '?single_page=true'
+
+    if web_version:
+
+        use_embedded_content = False
+
+        feeds = [
+            ('The Atlantic', 'http://www.theatlantic.com/feed/all/'),
+            ('Best of The Atlantic', 'http://www.theatlantic.com/feed/best-of/'),
+            ('Politics | The Atlantic', 'http://www.theatlantic.com/feed/channel/politics/'),
+            ('Business | The Atlantic', 'http://www.theatlantic.com/feed/channel/business/'),
+            ('Culture | The Atlantic', 'http://www.theatlantic.com/feed/channel/entertainment/'),
+            ('Global | The Atlantic', 'http://www.theatlantic.com/feed/channel/international/'),
+            ('Technology | The Atlantic', 'http://www.theatlantic.com/feed/channel/technology/'),
+            ('U.S. | The Atlantic', 'http://www.theatlantic.com/feed/channel/national/'),
+            ('Health | The Atlantic', 'http://www.theatlantic.com/feed/channel/health/'),
+            ('Video | The Atlantic', 'http://www.theatlantic.com/feed/channel/video/'),
+            ('Sexes | The Atlantic', 'http://www.theatlantic.com/feed/channel/sexes/'),
+            ('Education | The Atlantic', 'http://www.theatlantic.com/feed/channel/education/'),
+            ('Science | The Atlantic', 'http://www.theatlantic.com/feed/channel/science/'),
+            ('News | The Atlantic', 'http://www.theatlantic.com/feed/channel/news/'),
+            ('Press Releases | The Atlantic', 'http://www.theatlantic.com/feed/channel/press-releases/'),
+            ('Newsletters | The Atlantic', 'http://www.theatlantic.com/feed/channel/newsletters/'),
+            ('The Atlantic Photo', 'http://feeds.feedburner.com/theatlantic/infocus'),
+            ('Notes | The Atlantic', 'http://feeds.feedburner.com/TheAtlanticNotes'),
+        ]
+    else:
         def parse_index(self):
             soup = self.index_to_soup(self.INDEX)
             figure = soup.find('figure', id='cover-image')

recipes/atlantic_com.recipe

@@ -1,23 +1,94 @@
 #!/usr/bin/env  python2
-# vim:fileencoding=utf-8
+from __future__ import unicode_literals
-from __future__ import unicode_literals, division, absolute_import, print_function
+__license__ = 'GPL v3'
+__copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
+'''
+Recipe for web and magazine versions of the The Atlantic
+'''
 from calibre.web.feeds.news import BasicNewsRecipe
 
 
-class AdvancedUserRecipe1421956712(BasicNewsRecipe):
+web_version = True
+
+
+def classes(classes):
+    q = frozenset(classes.split(' '))
+    return dict(
+        attrs={'class': lambda x: x and frozenset(x.split()).intersection(q)}
+    )
+
+
+class TheAtlantic(BasicNewsRecipe):
+
+    if web_version:
         title = 'TheAtlantic.com'
-    __author__ = 'ebrandon'
-    language = 'en'
         description = 'News and editorial about politics, culture, entertainment, tech, etc. Contains many articles not seen in The Atlantic magazine'
-    oldest_article = 7
+    else:
-    max_articles_per_feed = 100
+        title = 'The Atlantic'
-    auto_cleanup = True
+        description = 'Current affairs and politics focussed on the US'
-    ignore_duplicate_articles = {'title', 'url'}
+        INDEX = 'http://www.theatlantic.com/magazine/'
+
+    __author__ = 'Kovid Goyal'
+    language = 'en'
+    encoding = 'utf-8'
+
+    keep_only_tags = [
+        classes(
+            'article-header lead-img article-cover-extra article-body article-magazine article-cover-content'
+        ),
+        {
+            'name': ['img']
+        },
+    ]
+    remove_tags = [
+        classes(
+            'social-kit-top letter-writer-info callout secondary-byline embed-wrapper offset-wrapper boxtop-most-popular'
+        ),
+        {
+            'name': ['meta', 'link', 'noscript', 'aside', 'h3']
+        },
+        {
+            'attrs': {
+                'class': ['offset-wrapper', 'boxtop-most-popular']
+            }
+        },
+        {
+            'attrs': {
+                'class': lambda x: x and 'article-tools' in x
+            }
+        },
+        {
+            'src': lambda x: x and 'spotxchange.com' in x
+        },
+    ]
+    remove_tags_after = classes('article-body')
+
+    no_stylesheets = True
+    remove_attributes = ['style']
+    extra_css = '''
+        .credit { text-align: right; font-size: 75%; display: block }
+        .figcaption { font-size: 75% }
+        .caption { font-size: 75% }
+        .lead-img { display: block }
+    '''
+
+    def get_browser(self):
+        br = BasicNewsRecipe.get_browser(self)
+        br.set_cookie('inEuropeanUnion', '0', '.theatlantic.com')
+        return br
+
+    def preprocess_html(self, soup):
+        for img in soup.findAll('img', attrs={'data-srcset': True}):
+            img['src'] = img['data-srcset'].split()[0]
+        return soup
 
     def print_version(self, url):
-        return url.replace('/archive/', '/print/')
+        return url.partition('?')[0] + '?single_page=true'
+
+    if web_version:
+
+        use_embedded_content = False
 
-    # Feed are found here: http://www.theatlantic.com/follow-the-atlantic/#follow-rssfeeds
         feeds = [
             ('The Atlantic', 'http://www.theatlantic.com/feed/all/'),
             ('Best of The Atlantic', 'http://www.theatlantic.com/feed/best-of/'),
@@ -38,3 +109,53 @@ class AdvancedUserRecipe1421956712(BasicNewsRecipe):
             ('The Atlantic Photo', 'http://feeds.feedburner.com/theatlantic/infocus'),
             ('Notes | The Atlantic', 'http://feeds.feedburner.com/TheAtlanticNotes'),
         ]
+    else:
+        def parse_index(self):
+            soup = self.index_to_soup(self.INDEX)
+            figure = soup.find('figure', id='cover-image')
+            if figure is not None:
+                img = figure.find('img', src=True)
+                if img:
+                    self.cover_url = img['src']
+            current_section, current_articles = 'Cover Story', []
+            feeds = []
+            for div in soup.findAll('div', attrs={'class': lambda x: x and set(x.split()).intersection({'top-sections', 'bottom-sections'})}):
+                for h2 in div.findAll('h2', attrs={'class': True}):
+                    if 'section-name' in h2['class'].split():
+                        if current_articles:
+                            feeds.append((current_section, current_articles))
+                        current_articles = []
+                        current_section = self.tag_to_string(h2)
+                        self.log('\nFound section:', current_section)
+                    elif 'hed' in h2['class'].split():
+                        title = self.tag_to_string(h2)
+                        a = h2.findParent('a', href=True)
+                        url = a['href']
+                        if url.startswith('/'):
+                            url = 'http://www.theatlantic.com' + url
+                        li = a.findParent(
+                            'li',
+                            attrs={'class': lambda x: x and 'article' in x.split()}
+                        )
+                        desc = ''
+                        dek = li.find(
+                            attrs={'class': lambda x: x and 'dek' in x.split()}
+                        )
+                        if dek is not None:
+                            desc += self.tag_to_string(dek)
+                        byline = li.find(
+                            attrs={'class': lambda x: x and 'byline' in x.split()}
+                        )
+                        if byline is not None:
+                            desc += ' -- ' + self.tag_to_string(byline)
+                        self.log('\t', title, 'at', url)
+                        if desc:
+                            self.log('\t\t', desc)
+                        current_articles.append({
+                            'title': title,
+                            'url': url,
+                            'description': desc
+                        })
+            if current_articles:
+                feeds.append((current_section, current_articles))
+            return feeds