yt_dlp/extractor/audiomack.py

   1 import itertools
   2 import time
   3
   4 from .common import InfoExtractor
   5 from .soundcloud import SoundcloudIE
   6 from ..utils import (
   7     ExtractorError,
   8     url_basename,
   9 )
  10
  11
  12 class AudiomackIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?audiomack\.com/(?:song/|(?=.+/song/))(?P<id>[\w/-]+)'
  14     IE_NAME = 'audiomack'
  15     _TESTS = [
  16         # hosted on audiomack
  17         {
  18             'url': 'http://www.audiomack.com/song/roosh-williams/extraordinary',
  19             'info_dict':
  20             {
  21                 'id': '310086',
  22                 'ext': 'mp3',
  23                 'uploader': 'Roosh Williams',
  24                 'title': 'Extraordinary',
  25             },
  26         },
  27         # audiomack wrapper around soundcloud song
  28         # Needs new test URL.
  29         {
  30             'add_ie': ['Soundcloud'],
  31             'url': 'http://www.audiomack.com/song/hip-hop-daily/black-mamba-freestyle',
  32             'info_dict': {
  33                 'id': '258901379',
  34                 'ext': 'mp3',
  35                 'description': 'mamba day freestyle for the legend Kobe Bryant ',
  36                 'title': 'Black Mamba Freestyle [Prod. By Danny Wolf]',
  37                 'uploader': 'ILOVEMAKONNEN',
  38                 'upload_date': '20160414',
  39             },
  40             'skip': 'Song has been removed from the site',
  41         },
  42     ]
  43
  44     def _real_extract(self, url):
  45         # URLs end with [uploader name]/song/[uploader title]
  46         # this title is whatever the user types in, and is rarely
  47         # the proper song title.  Real metadata is in the api response
  48         album_url_tag = self._match_id(url).replace('/song/', '/')
  49
  50         # Request the extended version of the api for extra fields like artist and title
  51         api_response = self._download_json(
  52             'http://www.audiomack.com/api/music/url/song/%s?extended=1&_=%d' % (
  53                 album_url_tag, time.time()),
  54             album_url_tag)
  55
  56         # API is inconsistent with errors
  57         if 'url' not in api_response or not api_response['url'] or 'error' in api_response:
  58             raise ExtractorError(f'Invalid url {url}')
  59
  60         # Audiomack wraps a lot of soundcloud tracks in their branded wrapper
  61         # if so, pass the work off to the soundcloud extractor
  62         if SoundcloudIE.suitable(api_response['url']):
  63             return self.url_result(api_response['url'], SoundcloudIE.ie_key())
  64
  65         return {
  66             'id': str(api_response.get('id', album_url_tag)),
  67             'uploader': api_response.get('artist'),
  68             'title': api_response.get('title'),
  69             'url': api_response['url'],
  70         }
  71
  72
  73 class AudiomackAlbumIE(InfoExtractor):
  74     _VALID_URL = r'https?://(?:www\.)?audiomack\.com/(?:album/|(?=.+/album/))(?P<id>[\w/-]+)'
  75     IE_NAME = 'audiomack:album'
  76     _TESTS = [
  77         # Standard album playlist
  78         {
  79             'url': 'http://www.audiomack.com/album/flytunezcom/tha-tour-part-2-mixtape',
  80             'playlist_count': 11,
  81             'info_dict':
  82             {
  83                 'id': '812251',
  84                 'title': 'Tha Tour: Part 2 (Official Mixtape)',
  85             },
  86         },
  87         # Album playlist ripped from fakeshoredrive with no metadata
  88         {
  89             'url': 'http://www.audiomack.com/album/fakeshoredrive/ppp-pistol-p-project',
  90             'info_dict': {
  91                 'title': 'PPP (Pistol P Project)',
  92                 'id': '837572',
  93             },
  94             'playlist': [{
  95                 'info_dict': {
  96                     'title': 'PPP (Pistol P Project) - 8. Real (prod by SYK SENSE  )',
  97                     'id': '837576',
  98                     'ext': 'mp3',
  99                     'uploader': 'Lil Herb a.k.a. G Herbo',
 100                 },
 101             }, {
 102                 'info_dict': {
 103                     'title': 'PPP (Pistol P Project) - 10. 4 Minutes Of Hell Part 4 (prod by DY OF 808 MAFIA)',
 104                     'id': '837580',
 105                     'ext': 'mp3',
 106                     'uploader': 'Lil Herb a.k.a. G Herbo',
 107                 },
 108             }],
 109         },
 110     ]
 111
 112     def _real_extract(self, url):
 113         # URLs end with [uploader name]/album/[uploader title]
 114         # this title is whatever the user types in, and is rarely
 115         # the proper song title.  Real metadata is in the api response
 116         album_url_tag = self._match_id(url).replace('/album/', '/')
 117         result = {'_type': 'playlist', 'entries': []}
 118         # There is no one endpoint for album metadata - instead it is included/repeated in each song's metadata
 119         # Therefore we don't know how many songs the album has and must infi-loop until failure
 120         for track_no in itertools.count():
 121             # Get song's metadata
 122             api_response = self._download_json(
 123                 'http://www.audiomack.com/api/music/url/album/%s/%d?extended=1&_=%d'
 124                 % (album_url_tag, track_no, time.time()), album_url_tag,
 125                 note=f'Querying song information ({track_no + 1})')
 126
 127             # Total failure, only occurs when url is totally wrong
 128             # Won't happen in middle of valid playlist (next case)
 129             if 'url' not in api_response or 'error' in api_response:
 130                 raise ExtractorError(f'Invalid url for track {track_no} of album url {url}')
 131             # URL is good but song id doesn't exist - usually means end of playlist
 132             elif not api_response['url']:
 133                 break
 134             else:
 135                 # Pull out the album metadata and add to result (if it exists)
 136                 for resultkey, apikey in [('id', 'album_id'), ('title', 'album_title')]:
 137                     if apikey in api_response and resultkey not in result:
 138                         result[resultkey] = str(api_response[apikey])
 139                 song_id = url_basename(api_response['url']).rpartition('.')[0]
 140                 result['entries'].append({
 141                     'id': str(api_response.get('id', song_id)),
 142                     'uploader': api_response.get('artist'),
 143                     'title': api_response.get('title', song_id),
 144                     'url': api_response['url'],
 145                 })
 146         return result