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 ..compat import compat_str
  10 from ..utils import (
  11     ExtractorError,
  12     url_basename,
  13 )
  14
  15
  16 class AudiomackIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?audiomack\.com/song/(?P<id>[\w/-]+)'
  18     IE_NAME = 'audiomack'
  19     _TESTS = [
  20         # hosted on audiomack
  21         {
  22             'url': 'http://www.audiomack.com/song/roosh-williams/extraordinary',
  23             'info_dict':
  24             {
  25                 'id': '310086',
  26                 'ext': 'mp3',
  27                 'uploader': 'Roosh Williams',
  28                 'title': 'Extraordinary'
  29             }
  30         },
  31         # audiomack wrapper around soundcloud song
  32         {
  33             'add_ie': ['Soundcloud'],
  34             'url': 'http://www.audiomack.com/song/hip-hop-daily/black-mamba-freestyle',
  35             'info_dict': {
  36                 'id': '258901379',
  37                 'ext': 'mp3',
  38                 'description': 'mamba day freestyle for the legend Kobe Bryant ',
  39                 'title': 'Black Mamba Freestyle [Prod. By Danny Wolf]',
  40                 'uploader': 'ILOVEMAKONNEN',
  41                 'upload_date': '20160414',
  42             }
  43         },
  44     ]
  45
  46     def _real_extract(self, url):
  47         # URLs end with [uploader name]/[uploader title]
  48         # this title is whatever the user types in, and is rarely
  49         # the proper song title.  Real metadata is in the api response
  50         album_url_tag = self._match_id(url)
  51
  52         # Request the extended version of the api for extra fields like artist and title
  53         api_response = self._download_json(
  54             'http://www.audiomack.com/api/music/url/song/%s?extended=1&_=%d' % (
  55                 album_url_tag, time.time()),
  56             album_url_tag)
  57
  58         # API is inconsistent with errors
  59         if 'url' not in api_response or not api_response['url'] or 'error' in api_response:
  60             raise ExtractorError('Invalid url %s' % url)
  61
  62         # Audiomack wraps a lot of soundcloud tracks in their branded wrapper
  63         # if so, pass the work off to the soundcloud extractor
  64         if SoundcloudIE.suitable(api_response['url']):
  65             return {'_type': 'url', 'url': api_response['url'], 'ie_key': 'Soundcloud'}
  66
  67         return {
  68             'id': compat_str(api_response.get('id', album_url_tag)),
  69             'uploader': api_response.get('artist'),
  70             'title': api_response.get('title'),
  71             'url': api_response['url'],
  72         }
  73
  74
  75 class AudiomackAlbumIE(InfoExtractor):
  76     _VALID_URL = r'https?://(?:www\.)?audiomack\.com/album/(?P<id>[\w/-]+)'
  77     IE_NAME = 'audiomack:album'
  78     _TESTS = [
  79         # Standard album playlist
  80         {
  81             'url': 'http://www.audiomack.com/album/flytunezcom/tha-tour-part-2-mixtape',
  82             'playlist_count': 15,
  83             'info_dict':
  84             {
  85                 'id': '812251',
  86                 'title': 'Tha Tour: Part 2 (Official Mixtape)'
  87             }
  88         },
  89         # Album playlist ripped from fakeshoredrive with no metadata
  90         {
  91             'url': 'http://www.audiomack.com/album/fakeshoredrive/ppp-pistol-p-project',
  92             'info_dict': {
  93                 'title': 'PPP (Pistol P Project)',
  94                 'id': '837572',
  95             },
  96             'playlist': [{
  97                 'info_dict': {
  98                     'title': 'PPP (Pistol P Project) - 9. Heaven or Hell (CHIMACA) ft Zuse (prod by DJ FU)',
  99                     'id': '837577',
 100                     'ext': 'mp3',
 101                     'uploader': 'Lil Herb a.k.a. G Herbo',
 102                 }
 103             }],
 104             'params': {
 105                 'playliststart': 9,
 106                 'playlistend': 9,
 107             }
 108         }
 109     ]
 110
 111     def _real_extract(self, url):
 112         # URLs end with [uploader name]/[uploader title]
 113         # this title is whatever the user types in, and is rarely
 114         # the proper song title.  Real metadata is in the api response
 115         album_url_tag = self._match_id(url)
 116         result = {'_type': 'playlist', 'entries': []}
 117         # There is no one endpoint for album metadata - instead it is included/repeated in each song's metadata
 118         # Therefore we don't know how many songs the album has and must infi-loop until failure
 119         for track_no in itertools.count():
 120             # Get song's metadata
 121             api_response = self._download_json(
 122                 'http://www.audiomack.com/api/music/url/album/%s/%d?extended=1&_=%d'
 123                 % (album_url_tag, track_no, time.time()), album_url_tag,
 124                 note='Querying song information (%d)' % (track_no + 1))
 125
 126             # Total failure, only occurs when url is totally wrong
 127             # Won't happen in middle of valid playlist (next case)
 128             if 'url' not in api_response or 'error' in api_response:
 129                 raise ExtractorError('Invalid url for track %d of album url %s' % (track_no, url))
 130             # URL is good but song id doesn't exist - usually means end of playlist
 131             elif not api_response['url']:
 132                 break
 133             else:
 134                 # Pull out the album metadata and add to result (if it exists)
 135                 for resultkey, apikey in [('id', 'album_id'), ('title', 'album_title')]:
 136                     if apikey in api_response and resultkey not in result:
 137                         result[resultkey] = api_response[apikey]
 138                 song_id = url_basename(api_response['url']).rpartition('.')[0]
 139                 result['entries'].append({
 140                     'id': compat_str(api_response.get('id', song_id)),
 141                     'uploader': api_response.get('artist'),
 142                     'title': api_response.get('title', song_id),
 143                     'url': api_response['url'],
 144                 })
 145         return result