Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/audiomack.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import itertools
   5 import time
   6
   7 from .common import InfoExtractor
   8 from .soundcloud import SoundcloudIE
   9 from ..utils import (
  10     ExtractorError,
  11     url_basename,
  12 )
  13
  14
  15 class AudiomackIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?audiomack\.com/song/(?P<id>[\w/-]+)'
  17     IE_NAME = 'audiomack'
  18     _TESTS = [
  19         # hosted on audiomack
  20         {
  21             'url': 'http://www.audiomack.com/song/roosh-williams/extraordinary',
  22             'info_dict':
  23             {
  24                 'id': '310086',
  25                 'ext': 'mp3',
  26                 'uploader': 'Roosh Williams',
  27                 'title': 'Extraordinary'
  28             }
  29         },
  30         # audiomack wrapper around soundcloud song
  31         {
  32             'add_ie': ['Soundcloud'],
  33             'url': 'http://www.audiomack.com/song/xclusiveszone/take-kare',
  34             'info_dict': {
  35                 'id': '172419696',
  36                 'ext': 'mp3',
  37                 'description': 'md5:1fc3272ed7a635cce5be1568c2822997',
  38                 'title': 'Young Thug ft Lil Wayne - Take Kare',
  39                 'uploader': 'Young Thug World',
  40                 'upload_date': '20141016',
  41             }
  42         },
  43     ]
  44
  45     def _real_extract(self, url):
  46         # URLs end with [uploader name]/[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)
  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 {'_type': 'url', 'url': api_response['url'], 'ie_key': 'Soundcloud'}
  65
  66         return {
  67             'id': 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/(?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': 15,
  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) - 9. Heaven or Hell (CHIMACA) ft Zuse (prod by DJ FU)',
  98                     'id': '837577',
  99                     'ext': 'mp3',
 100                     'uploader': 'Lil Herb a.k.a. G Herbo',
 101                 }
 102             }],
 103             'params': {
 104                 'playliststart': 9,
 105                 'playlistend': 9,
 106             }
 107         }
 108     ]
 109
 110     def _real_extract(self, url):
 111         # URLs end with [uploader name]/[uploader title]
 112         # this title is whatever the user types in, and is rarely
 113         # the proper song title.  Real metadata is in the api response
 114         album_url_tag = self._match_id(url)
 115         result = {'_type': 'playlist', 'entries': []}
 116         # There is no one endpoint for album metadata - instead it is included/repeated in each song's metadata
 117         # Therefore we don't know how many songs the album has and must infi-loop until failure
 118         for track_no in itertools.count():
 119             # Get song's metadata
 120             api_response = self._download_json(
 121                 'http://www.audiomack.com/api/music/url/album/%s/%d?extended=1&_=%d'
 122                 % (album_url_tag, track_no, time.time()), album_url_tag,
 123                 note='Querying song information (%d)' % (track_no + 1))
 124
 125             # Total failure, only occurs when url is totally wrong
 126             # Won't happen in middle of valid playlist (next case)
 127             if 'url' not in api_response or 'error' in api_response:
 128                 raise ExtractorError('Invalid url for track %d of album url %s' % (track_no, url))
 129             # URL is good but song id doesn't exist - usually means end of playlist
 130             elif not api_response['url']:
 131                 break
 132             else:
 133                 # Pull out the album metadata and add to result (if it exists)
 134                 for resultkey, apikey in [('id', 'album_id'), ('title', 'album_title')]:
 135                     if apikey in api_response and resultkey not in result:
 136                         result[resultkey] = api_response[apikey]
 137                 song_id = url_basename(api_response['url']).rpartition('.')[0]
 138                 result['entries'].append({
 139                     'id': api_response.get('id', song_id),
 140                     'uploader': api_response.get('artist'),
 141                     'title': api_response.get('title', song_id),
 142                     'url': api_response['url'],
 143                 })
 144         return result