[heise] Fix description, thumbnail and format ID

youtube_dl/extractor/common.py

@@ -404,7 +404,7 @@ class InfoExtractor(object):
             video_info['title'] = playlist_title
         return video_info
 
-    def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0):
+    def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0, group=None):
         """
         Perform a regex search on the given string, using a single or a list of
         patterns returning the first matching group.
@@ -425,8 +425,11 @@ class InfoExtractor(object):
             _name = name
 
         if mobj:
-            # return the first matching group
-            return next(g for g in mobj.groups() if g is not None)
+            if group is None:
+                # return the first matching group
+                return next(g for g in mobj.groups() if g is not None)
+            else:
+                return mobj.group(group)
         elif default is not _NO_DEFAULT:
             return default
         elif fatal:
@@ -436,11 +439,11 @@ class InfoExtractor(object):
                 'please report this issue on http://yt-dl.org/bug' % _name)
             return None
 
-    def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0):
+    def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0, group=None):
         """
         Like _search_regex, but strips HTML tags and unescapes entities.
         """
-        res = self._search_regex(pattern, string, name, default, fatal, flags)
+        res = self._search_regex(pattern, string, name, default, fatal, flags, group)
         if res:
             return clean_html(res).strip()
         else:
@@ -534,9 +537,9 @@ class InfoExtractor(object):
             display_name = name
         return self._html_search_regex(
             r'''(?ix)<meta
-                    (?=[^>]+(?:itemprop|name|property)=["\']?%s["\']?)
-                    [^>]+content=["\']([^"\']+)["\']''' % re.escape(name),
-            html, display_name, fatal=fatal, **kwargs)
+                    (?=[^>]+(?:itemprop|name|property)=(["\']?)%s\1)
+                    [^>]+content=(["\'])(?P<content>.*?)\1''' % re.escape(name),
+            html, display_name, fatal=fatal, group='content', **kwargs)
 
     def _dc_search_uploader(self, html):
         return self._html_search_meta('dc.creator', html, 'uploader')

youtube_dl/extractor/heise.py

@@ -3,7 +3,7 @@ from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from ..utils import (
-    get_meta_content,
+    determine_ext,
     int_or_none,
     parse_iso8601,
 )
@@ -25,11 +25,11 @@ class HeiseIE(InfoExtractor):
             'title': (
                 "Podcast: c't uplink 3.3 – Owncloud / Tastaturen / Peilsender Smartphone"
             ),
-            'format_id': 'mp4_720',
+            'format_id': 'mp4_720p',
             'timestamp': 1411812600,
             'upload_date': '20140927',
             'description': 'In uplink-Episode 3.3 geht es darum, wie man sich von Cloud-Anbietern emanzipieren kann, worauf man beim Kauf einer Tastatur achten sollte und was Smartphones über uns verraten.',
-            'thumbnail': 're:https?://.*\.jpg$',
+            'thumbnail': 're:^https?://.*\.jpe?g$',
         }
     }
 
@@ -49,11 +49,12 @@ class HeiseIE(InfoExtractor):
         info = {
             'id': video_id,
             'thumbnail': self._og_search_thumbnail(webpage),
-            'timestamp': parse_iso8601(get_meta_content('date', webpage)),
+            'timestamp': parse_iso8601(
+                self._html_search_meta('date', webpage)),
             'description': self._og_search_description(webpage),
         }
 
-        title = get_meta_content('fulltitle', webpage)
+        title = self._html_search_meta('fulltitle', webpage)
         if title:
             info['title'] = title
         else:
@@ -64,9 +65,12 @@ class HeiseIE(InfoExtractor):
             label = source_node.attrib['label']
             height = int_or_none(self._search_regex(
                 r'^(.*?_)?([0-9]+)p$', label, 'height', default=None))
+            video_url = source_node.attrib['file']
+            ext = determine_ext(video_url, '')
             formats.append({
-                'url': source_node.attrib['file'],
+                'url': video_url,
                 'format_note': label,
+                'format_id': '%s_%s' % (ext, label),
                 'height': height,
             })
         self._sort_formats(formats)