[youtube] Fix extraction.

[youtube-dl] / youtube_dl / extractor / worldstarhiphop.py

diff --git a/youtube_dl/extractor/worldstarhiphop.py b/youtube_dl/extractor/worldstarhiphop.py
index bda3870db9f16e12c721c361696656df8be8a1b3..82587b4ce94c0f1cbaeba0beb88b7dcb02e67e40 100644 (file)
--- a/youtube_dl/extractor/worldstarhiphop.py
+++ b/youtube_dl/extractor/worldstarhiphop.py
@@ -1,54 +1,40 @@
 from __future__ import unicode_literals
 
-import re
-
 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",
-        "md5": "9d04de741161603bf7071bbf4e883186",
-        "info_dict": {
-            "id": "wshh6a7q1ny0G34ZwuIO",
-            "ext": "mp4",
-            "title": "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):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        m_vevo_id = re.search(r'videoId=(.*?)&amp?', webpage)
-        if m_vevo_id is not None:
-            return self.url_result('vevo:%s' % m_vevo_id.group(1), ie='Vevo')
-
-        video_url = self._search_regex(
-            r'so\.addVariable\("file","(.*?)"\)', webpage, 'video URL')
+        entries = self._parse_html5_media_entries(url, webpage, video_id)
 
-        if 'youtube' in video_url:
-            return self.url_result(video_url, ie='Youtube')
+        if not entries:
+            return self.url_result(url, 'Generic')
 
-        video_title = self._html_search_regex(
-            r'(?s)<div class="content-heading">\s*<h1>(.*?)</h1>',
+        title = self._html_search_regex(
+            [r'(?s)<div class="content-heading">\s*<h1>(.*?)</h1>',
+             r'<span[^>]+class="tc-sp-pinned-title">(.*)</span>'],
             webpage, 'title')
 
-        # Getting thumbnail and if not thumbnail sets correct title for WSHH candy video.
-        thumbnail = self._html_search_regex(
-            r'rel="image_src" href="(.*)" />', webpage, 'thumbnail',
-            fatal=False)
-        if not thumbnail:
-            _title = r'candytitles.*>(.*)</span>'
-            mobj = re.search(_title, webpage)
-            if mobj is not None:
-                video_title = mobj.group(1)
-
-        return {
+        info = entries[0]
+        info.update({
             'id': video_id,
-            'url': video_url,
-            'title': video_title,
-            'thumbnail': thumbnail,
-        }
-
+            'title': title,
+        })
+        return info