yt_dlp/extractor/audiomack.py

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