[myspace] Add support for song urls (fixes #2040)

youtube_dl/extractor/myspace.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 import json
 
@@ -8,41 +10,75 @@ from ..utils import (
 
 
 class MySpaceIE(InfoExtractor):
-    _VALID_URL = r'https?://myspace\.com/([^/]+)/video/[^/]+/(?P<id>\d+)'
+    _VALID_URL = r'https?://myspace\.com/([^/]+)/(?:video/[^/]+/|music/song/.*?)(?P<id>\d+)'
-
+
-    _TEST = {
+    _TESTS = [
-        u'url': u'https://myspace.com/coldplay/video/viva-la-vida/100008689',
+        {
-        u'info_dict': {
+            'url': 'https://myspace.com/coldplay/video/viva-la-vida/100008689',
-            u'id': u'100008689',
+            'info_dict': {
-            u'ext': u'flv',
+                'id': '100008689',
-            u'title': u'Viva La Vida',
+                'ext': 'flv',
-            u'description': u'The official Viva La Vida video, directed by Hype Williams',
+                'title': 'Viva La Vida',
-            u'uploader': u'Coldplay',
+                'description': 'The official Viva La Vida video, directed by Hype Williams',
-            u'uploader_id': u'coldplay',
+                'uploader': 'Coldplay',
+                'uploader_id': 'coldplay',
             },
-        u'params': {
+            'params': {
                 # rtmp download
-            u'skip_download': True,
+                'skip_download': True,
             },
-    }
+        },
+        # song
+        {
+            'url': 'https://myspace.com/spiderbags/music/song/darkness-in-my-heart-39008454-27041242',
+            'info_dict': {
+                'id': '39008454',
+                'ext': 'flv',
+                'title': 'Darkness In My Heart',
+                'uploader_id': 'spiderbags',
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+        },
+    ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
+
+        if 'music/song' in url:
+            # songs don't store any useful info in the 'context' variable
+            def search_data(name):
+                return self._search_regex(r'data-%s="(.*?)"' % name, webpage,
+                    name)
+            streamUrl = search_data('stream-url')
+            info = {
+                'id': video_id,
+                'title': self._og_search_title(webpage),
+                'uploader_id': search_data('artist-username'),
+                'thumbnail': self._og_search_thumbnail(webpage),
+            }
+        else:
             context = json.loads(self._search_regex(r'context = ({.*?});', webpage,
                 u'context'))
             video = context['video']
-        rtmp_url, play_path = video['streamUrl'].split(';', 1)
+            streamUrl = video['streamUrl']
-
+            info = {
-        return {
                 'id': compat_str(video['mediaId']),
                 'title': video['title'],
-            'url': rtmp_url,
-            'play_path': play_path,
-            'ext': 'flv',
                 'description': video['description'],
                 'thumbnail': video['imageUrl'],
                 'uploader': video['artistName'],
                 'uploader_id': video['artistUsername'],
             }
+
+        rtmp_url, play_path = streamUrl.split(';', 1)
+        info.update({
+            'url': rtmp_url,
+            'play_path': play_path,
+            'ext': 'flv',
+        })
+        return info