yt_dlp/extractor/jiosaavn.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     int_or_none,
   4     js_to_json,
   5     url_or_none,
   6     urlencode_postdata,
   7     urljoin,
   8 )
   9 from ..utils.traversal import traverse_obj
  10
  11
  12 class JioSaavnBaseIE(InfoExtractor):
  13     def _extract_initial_data(self, url, audio_id):
  14         webpage = self._download_webpage(url, audio_id)
  15         return self._search_json(
  16             r'window\.__INITIAL_DATA__\s*=', webpage,
  17             'init json', audio_id, transform_source=js_to_json)
  18
  19
  20 class JioSaavnSongIE(JioSaavnBaseIE):
  21     _VALID_URL = r'https?://(?:www\.)?(?:jiosaavn\.com/song/[^/?#]+/|saavn\.com/s/song/(?:[^/?#]+/){3})(?P<id>[^/?#]+)'
  22     _TESTS = [{
  23         'url': 'https://www.jiosaavn.com/song/leja-re/OQsEfQFVUXk',
  24         'md5': '3b84396d15ed9e083c3106f1fa589c04',
  25         'info_dict': {
  26             'id': 'OQsEfQFVUXk',
  27             'ext': 'mp4',
  28             'title': 'Leja Re',
  29             'album': 'Leja Re',
  30             'thumbnail': 'https://c.saavncdn.com/258/Leja-Re-Hindi-2018-20181124024539-500x500.jpg',
  31             'duration': 205,
  32             'view_count': int,
  33             'release_year': 2018,
  34         },
  35     }, {
  36         'url': 'https://www.saavn.com/s/song/hindi/Saathiya/O-Humdum-Suniyo-Re/KAMiazoCblU',
  37         'only_matching': True,
  38     }]
  39
  40     _VALID_BITRATES = ('16', '32', '64', '128', '320')
  41
  42     def _real_extract(self, url):
  43         audio_id = self._match_id(url)
  44         extract_bitrates = self._configuration_arg('bitrate', ['128', '320'], ie_key='JioSaavn')
  45         if invalid_bitrates := [br for br in extract_bitrates if br not in self._VALID_BITRATES]:
  46             raise ValueError(
  47                 f'Invalid bitrate(s): {", ".join(invalid_bitrates)}. '
  48                 + f'Valid bitrates are: {", ".join(self._VALID_BITRATES)}')
  49
  50         song_data = self._extract_initial_data(url, audio_id)['song']['song']
  51         formats = []
  52         for bitrate in extract_bitrates:
  53             media_data = self._download_json(
  54                 'https://www.jiosaavn.com/api.php', audio_id, f'Downloading format info for {bitrate}',
  55                 fatal=False, data=urlencode_postdata({
  56                     '__call': 'song.generateAuthToken',
  57                     '_format': 'json',
  58                     'bitrate': bitrate,
  59                     'url': song_data['encrypted_media_url'],
  60                 }))
  61             if not media_data.get('auth_url'):
  62                 self.report_warning(f'Unable to extract format info for {bitrate}')
  63                 continue
  64             formats.append({
  65                 'url': media_data['auth_url'],
  66                 'ext': media_data.get('type'),
  67                 'format_id': bitrate,
  68                 'abr': int(bitrate),
  69                 'vcodec': 'none',
  70             })
  71
  72         return {
  73             'id': audio_id,
  74             'formats': formats,
  75             **traverse_obj(song_data, {
  76                 'title': ('title', 'text'),
  77                 'album': ('album', 'text'),
  78                 'thumbnail': ('image', 0, {url_or_none}),
  79                 'duration': ('duration', {int_or_none}),
  80                 'view_count': ('play_count', {int_or_none}),
  81                 'release_year': ('year', {int_or_none}),
  82             }),
  83         }
  84
  85
  86 class JioSaavnAlbumIE(JioSaavnBaseIE):
  87     _VALID_URL = r'https?://(?:www\.)?(?:jio)?saavn\.com/album/[^/?#]+/(?P<id>[^/?#]+)'
  88     _TESTS = [{
  89         'url': 'https://www.jiosaavn.com/album/96/buIOjYZDrNA_',
  90         'info_dict': {
  91             'id': 'buIOjYZDrNA_',
  92             'title': '96',
  93         },
  94         'playlist_count': 10,
  95     }]
  96
  97     def _real_extract(self, url):
  98         album_id = self._match_id(url)
  99         album_view = self._extract_initial_data(url, album_id)['albumView']
 100
 101         return self.playlist_from_matches(
 102             traverse_obj(album_view, (
 103                 'modules', lambda _, x: x['key'] == 'list', 'data', ..., 'title', 'action', {str})),
 104             album_id, traverse_obj(album_view, ('album', 'title', 'text', {str})), ie=JioSaavnSongIE,
 105             getter=lambda x: urljoin('https://www.jiosaavn.com/', x))