[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / worldstarhiphop.py
index 8a52c7a99147c4c871ac1cdf97272941effbc104..82587b4ce94c0f1cbaeba0beb88b7dcb02e67e40 100644 (file)
@@ -1,70 +1,40 @@
-import re
+from __future__ import unicode_literals
 
 from .common import InfoExtractor
 
 
 class WorldStarHipHopIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www|m)\.worldstar(?:candy|hiphop)\.com/videos/video\.php\?v=(?P<id>.*)'
-    _TEST = {
-        "url": "http://www.worldstarhiphop.com/videos/video.php?v=wshh6a7q1ny0G34ZwuIO",
-        "file": "wshh6a7q1ny0G34ZwuIO.mp4",
-        "md5": "9d04de741161603bf7071bbf4e883186",
-        "info_dict": {
-            "title": "Video: KO Of The Week: MMA Fighter Gets Knocked Out By Swift Head Kick!"
+    _VALID_URL = r'https?://(?:www|m)\.worldstar(?:candy|hiphop)\.com/(?:videos|android)/video\.php\?.*?\bv=(?P<id>[^&]+)'
+    _TESTS = [{
+        'url': 'http://www.worldstarhiphop.com/videos/video.php?v=wshh6a7q1ny0G34ZwuIO',
+        'md5': '9d04de741161603bf7071bbf4e883186',
+        'info_dict': {
+            'id': 'wshh6a7q1ny0G34ZwuIO',
+            'ext': 'mp4',
+            'title': 'KO Of The Week: MMA Fighter Gets Knocked Out By Swift Head Kick!'
         }
-    }
-
+    }, {
+        'url': 'http://m.worldstarhiphop.com/android/video.php?v=wshh6a7q1ny0G34ZwuIO',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
-        m = re.match(self._VALID_URL, url)
-        video_id = m.group('id')
-
-        webpage_src = self._download_webpage(url, video_id)
-
-        video_url = re.search(r'videoId=(.*?)&amp?',
-            webpage_src)
-        
-        if video_url:
-            self.to_screen(u'Vevo video detected:')
-            return self.url_result('vevo:%s' % video_url.group(1), ie='Vevo')
-
-        video_url = self._search_regex(r'so\.addVariable\("file","(.*?)"\)',
-            webpage_src, u'video URL')
-
-        if video_url is None:
-            video_url = self._search_regex(r'videoId=(.*?)&amp?',
-                webpage_src, u'video URL')
-            self.to_screen(u'Vevo video detected:')
-            vevo_id = 'vevo:%s' % video_url
-            return self.url_result(vevo_id, ie='Vevo')
-
-        if 'youtube' in video_url:
-            self.to_screen(u'Youtube video detected:')
-            return self.url_result(video_url, ie='Youtube')
-
-        if 'mp4' in video_url:
-            ext = 'mp4'
-        else:
-            ext = 'flv'
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
 
-        video_title = self._html_search_regex(r"<title>(.*)</title>",
-            webpage_src, u'title')
+        entries = self._parse_html5_media_entries(url, webpage, video_id)
 
-        # Getting thumbnail and if not thumbnail sets correct title for WSHH candy video.
-        thumbnail = self._html_search_regex(r'rel="image_src" href="(.*)" />',
-            webpage_src, u'thumbnail', fatal=False)
+        if not entries:
+            return self.url_result(url, 'Generic')
 
-        if not thumbnail:
-            _title = r"""candytitles.*>(.*)</span>"""
-            mobj = re.search(_title, webpage_src)
-            if mobj is not None:
-                video_title = mobj.group(1)
+        title = self._html_search_regex(
+            [r'(?s)<div class="content-heading">\s*<h1>(.*?)</h1>',
+             r'<span[^>]+class="tc-sp-pinned-title">(.*)</span>'],
+            webpage, 'title')
 
-        results = [{
-                    'id': video_id,
-                    'url' : video_url,
-                    'title' : video_title,
-                    'thumbnail' : thumbnail,
-                    'ext' : ext,
-                    }]
-        return results
+        info = entries[0]
+        info.update({
+            'id': video_id,
+            'title': title,
+        })
+        return info