X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fgamestar.py;h=e607d6ab8215db56afd3810613f24bb2debf63f5;hb=ca127ab2c174cdee4428eb4e192393c6ca942ac8;hp=5c9decdea96b5730352f250a0fe1c9039892b230;hpb=8646eb790e2352ec08c50357b9957cbe5548ddaa;p=youtube-dl

diff --git a/youtube_dl/extractor/gamestar.py b/youtube_dl/extractor/gamestar.py
index 5c9decdea..e607d6ab8 100644
--- a/youtube_dl/extractor/gamestar.py
+++ b/youtube_dl/extractor/gamestar.py
@@ -1,12 +1,15 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    remove_end,
+)
+
 
 class GameStarIE(InfoExtractor):
-    _VALID_URL = r'http://www\.gamestar\.de/videos/.*,(?P<id>[0-9]+)\.html'
+    _VALID_URL = r'https?://(?:www\.)?gamestar\.de/videos/.*,(?P<id>[0-9]+)\.html'
     _TEST = {
         'url': 'http://www.gamestar.de/videos/trailer,3/hobbit-3-die-schlacht-der-fuenf-heere,76110.html',
         'md5': '96974ecbb7fd8d0d20fca5a00810cea7',
@@ -14,55 +17,39 @@ class GameStarIE(InfoExtractor):
             'id': '76110',
             'ext': 'mp4',
             'title': 'Hobbit 3: Die Schlacht der Fünf Heere - Teaser-Trailer zum dritten Teil',
-            'description': 'Der Teaser-Trailer zu Hobbit 3: Die Schlacht der Fünf Heere zeigt einige Szenen aus dem dritten Teil der Saga und kündigt den vollständigen Trailer an.',
-            'thumbnail': 'http://images.gamestar.de/images/idgwpgsgp/bdb/2494525/600x.jpg',
+            'description': 'Der Teaser-Trailer zu Hobbit 3: Die Schlacht der Fünf Heere zeigt einige Szenen aus dem dritten Teil der Saga und kündigt den...',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'timestamp': 1406542020,
             'upload_date': '20140728',
             'duration': 17
         }
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        og_title = self._og_search_title(webpage)
-        title = og_title.replace(' - Video bei GameStar.de', '').strip()
-
         url = 'http://gamestar.de/_misc/videos/portal/getVideoUrl.cfm?premium=0&videoId=' + video_id
 
-        description = self._og_search_description(webpage).strip()
-
-        og_thumbnail = self._og_search_thumbnail(webpage)
-        thumbnail = 'http:' + og_thumbnail
+        # TODO: there are multiple ld+json objects in the webpage,
+        # while _search_json_ld finds only the first one
+        json_ld = self._parse_json(self._search_regex(
+            r'(?s)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>[^<]+VideoObject[^<]+)</script>',
+            webpage, 'JSON-LD', group='json_ld'), video_id)
+        info_dict = self._json_ld(json_ld, video_id)
+        info_dict['title'] = remove_end(info_dict['title'], ' - GameStar')
 
-        upload_date_raw = self._html_search_regex(
-            r'<span style="float:left;font-size:11px;">Datum: ([0-9]+\.[0-9]+\.[0-9]+)&nbsp;&nbsp;',
-            webpage, 'upload_date').split('.')
-        upload_date = upload_date_raw[2] + upload_date_raw[1] + upload_date_raw[0]
+        view_count = json_ld.get('interactionCount')
+        comment_count = int_or_none(self._html_search_regex(
+            r'([0-9]+) Kommentare</span>', webpage, 'comment_count',
+            fatal=False))
 
-        duration_raw = self._html_search_regex(
-            r'&nbsp;&nbsp;Länge: ([0-9]+:[0-9]+)</span>', webpage, 'duration').split(':')
-        duration = int(duration_raw[0])*60 + int(duration_raw[1])
-
-        view_count_raw = self._html_search_regex(
-            r'&nbsp;&nbsp;Zuschauer: ([0-9\.]+)&nbsp;&nbsp;', webpage, 'view_count')
-        view_count = int(view_count_raw.replace('.', ''))
-
-        comment_count_raw = self._html_search_regex(
-            r'>Kommentieren \(([0-9]+)\)</a>', webpage, 'comment_count')
-        comment_count = int(comment_count_raw)
-
-        return {
+        info_dict.update({
             'id': video_id,
-            'title': title,
             'url': url,
             'ext': 'mp4',
-            'thumbnail': thumbnail,
-            'description': description,
-            'upload_date': upload_date,
-            'duration': duration,
             'view_count': view_count,
             'comment_count': comment_count
-        }
+        })
+
+        return info_dict