[hungama] Fix code and extract more metadata (closes #18771)

ytdl-org · Jan 8, 2019 · 06b4b90 · 06b4b90
1 parent 8cb5c21
commit 06b4b90
Showing 1 changed file with 32 additions and 9 deletions.
diff --git a/youtube_dl/extractor/hungama.py b/youtube_dl/extractor/hungama.py
@@ -2,31 +2,54 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..utils import int_or_none
 
 
 class HungamaIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)hungama\.com/song/[\w\d-]+/(?P<id>[0-9]+)'
+    _VALID_URL = r'https?://(?:www\.)?hungama\.com/song/[^/]+/(?P<id>[0-9]+)'
     _TEST = {
         'url': 'https://www.hungama.com/song/kitni-haseen-zindagi/2931166/',
-        'md5': '396fa7e8e7e67aa25da0edc4cac9b785',
+        'md5': 'a845a6d1ebd08d80c1035126d49bd6a0',
         'info_dict': {
             'id': '2931166',
             'ext': 'mp4',
-            'title': 'Kitni Haseen Zindagi',
+            'title': 'Lucky Ali - Kitni Haseen Zindagi',
+            'track': 'Kitni Haseen Zindagi',
+            'artist': 'Lucky Ali',
+            'album': 'Aks',
+            'release_year': 2000,
         }
     }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
 
-        player_data = self._download_json('https://www.hungama.com/audio-player-data/track/%s?_country=IN' % video_id, video_id)[0]
-        title = player_data.get('song_name') or self._og_search_title(webpage)
-        track_data = self._download_json(player_data['file'], video_id)
-        media_url = track_data['response']['media_url']
+        data = self._download_json(
+            'https://www.hungama.com/audio-player-data/track/%s' % video_id,
+            video_id, query={'_country': 'IN'})[0]
+
+        track = data['song_name']
+        artist = data.get('singer_name')
+
+        m3u8_url = self._download_json(
+            data.get('file') or data['preview_link'],
+            video_id)['response']['media_url']
+
+        formats = self._extract_m3u8_formats(
+            m3u8_url, video_id, ext='mp4', entry_protocol='m3u8_native',
+            m3u8_id='hls')
+        self._sort_formats(formats)
+
+        title = '%s - %s' % (artist, track) if artist else track
+        thumbnail = data.get('img_src') or data.get('album_image')
 
         return {
             'id': video_id,
             'title': title,
-            'formats': self._extract_m3u8_formats(media_url, video_id, ext='mp4'),
+            'thumbnail': thumbnail,
+            'track': track,
+            'artist': artist,
+            'album': data.get('album_name'),
+            'release_year': int_or_none(data.get('date')),
+            'formats': formats,
         }