yt_dlp/extractor/mixcloud.py

   1 from __future__ import unicode_literals
   2
   3 import itertools
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_b64decode,
   9     compat_chr,
  10     compat_ord,
  11     compat_str,
  12     compat_urllib_parse_unquote,
  13     compat_zip
  14 )
  15 from ..utils import (
  16     int_or_none,
  17     parse_iso8601,
  18     strip_or_none,
  19     try_get,
  20 )
  21
  22
  23 class MixcloudBaseIE(InfoExtractor):
  24     def _call_api(self, object_type, object_fields, display_id, username, slug=None):
  25         lookup_key = object_type + 'Lookup'
  26         return self._download_json(
  27             'https://www.mixcloud.com/graphql', display_id, query={
  28                 'query': '''{
  29   %s(lookup: {username: "%s"%s}) {
  30     %s
  31   }
  32 }''' % (lookup_key, username, ', slug: "%s"' % slug if slug else '', object_fields)
  33             })['data'][lookup_key]
  34
  35
  36 class MixcloudIE(MixcloudBaseIE):
  37     _VALID_URL = r'https?://(?:(?:www|beta|m)\.)?mixcloud\.com/([^/]+)/(?!stream|uploads|favorites|listens|playlists)([^/]+)'
  38     IE_NAME = 'mixcloud'
  39
  40     _TESTS = [{
  41         'url': 'http://www.mixcloud.com/dholbach/cryptkeeper/',
  42         'info_dict': {
  43             'id': 'dholbach_cryptkeeper',
  44             'ext': 'm4a',
  45             'title': 'Cryptkeeper',
  46             'description': 'After quite a long silence from myself, finally another Drum\'n\'Bass mix with my favourite current dance floor bangers.',
  47             'uploader': 'Daniel Holbach',
  48             'uploader_id': 'dholbach',
  49             'thumbnail': r're:https?://.*\.jpg',
  50             'view_count': int,
  51             'timestamp': 1321359578,
  52             'upload_date': '20111115',
  53         },
  54     }, {
  55         'url': 'http://www.mixcloud.com/gillespeterson/caribou-7-inch-vinyl-mix-chat/',
  56         'info_dict': {
  57             'id': 'gillespeterson_caribou-7-inch-vinyl-mix-chat',
  58             'ext': 'mp3',
  59             'title': 'Caribou 7 inch Vinyl Mix & Chat',
  60             'description': 'md5:2b8aec6adce69f9d41724647c65875e8',
  61             'uploader': 'Gilles Peterson Worldwide',
  62             'uploader_id': 'gillespeterson',
  63             'thumbnail': 're:https?://.*',
  64             'view_count': int,
  65             'timestamp': 1422987057,
  66             'upload_date': '20150203',
  67         },
  68     }, {
  69         'url': 'https://beta.mixcloud.com/RedLightRadio/nosedrip-15-red-light-radio-01-18-2016/',
  70         'only_matching': True,
  71     }]
  72     _DECRYPTION_KEY = 'IFYOUWANTTHEARTISTSTOGETPAIDDONOTDOWNLOADFROMMIXCLOUD'
  73
  74     @staticmethod
  75     def _decrypt_xor_cipher(key, ciphertext):
  76         """Encrypt/Decrypt XOR cipher. Both ways are possible because it's XOR."""
  77         return ''.join([
  78             compat_chr(compat_ord(ch) ^ compat_ord(k))
  79             for ch, k in compat_zip(ciphertext, itertools.cycle(key))])
  80
  81     def _real_extract(self, url):
  82         username, slug = re.match(self._VALID_URL, url).groups()
  83         username, slug = compat_urllib_parse_unquote(username), compat_urllib_parse_unquote(slug)
  84         track_id = '%s_%s' % (username, slug)
  85
  86         cloudcast = self._call_api('cloudcast', '''audioLength
  87     comments(first: 100) {
  88       edges {
  89         node {
  90           comment
  91           created
  92           user {
  93             displayName
  94             username
  95           }
  96         }
  97       }
  98       totalCount
  99     }
 100     description
 101     favorites {
 102       totalCount
 103     }
 104     featuringArtistList
 105     isExclusive
 106     name
 107     owner {
 108       displayName
 109       url
 110       username
 111     }
 112     picture(width: 1024, height: 1024) {
 113         url
 114     }
 115     plays
 116     publishDate
 117     reposts {
 118       totalCount
 119     }
 120     streamInfo {
 121       dashUrl
 122       hlsUrl
 123       url
 124     }
 125     tags {
 126       tag {
 127         name
 128       }
 129     }''', track_id, username, slug)
 130
 131         title = cloudcast['name']
 132
 133         stream_info = cloudcast['streamInfo']
 134         formats = []
 135
 136         for url_key in ('url', 'hlsUrl', 'dashUrl'):
 137             format_url = stream_info.get(url_key)
 138             if not format_url:
 139                 continue
 140             decrypted = self._decrypt_xor_cipher(
 141                 self._DECRYPTION_KEY, compat_b64decode(format_url))
 142             if url_key == 'hlsUrl':
 143                 formats.extend(self._extract_m3u8_formats(
 144                     decrypted, track_id, 'mp4', entry_protocol='m3u8_native',
 145                     m3u8_id='hls', fatal=False))
 146             elif url_key == 'dashUrl':
 147                 formats.extend(self._extract_mpd_formats(
 148                     decrypted, track_id, mpd_id='dash', fatal=False))
 149             else:
 150                 formats.append({
 151                     'format_id': 'http',
 152                     'url': decrypted,
 153                     'downloader_options': {
 154                         # Mixcloud starts throttling at >~5M
 155                         'http_chunk_size': 5242880,
 156                     },
 157                 })
 158
 159         if not formats and cloudcast.get('isExclusive'):
 160             self.raise_login_required()
 161
 162         self._sort_formats(formats)
 163
 164         comments = []
 165         for edge in (try_get(cloudcast, lambda x: x['comments']['edges']) or []):
 166             node = edge.get('node') or {}
 167             text = strip_or_none(node.get('comment'))
 168             if not text:
 169                 continue
 170             user = node.get('user') or {}
 171             comments.append({
 172                 'author': user.get('displayName'),
 173                 'author_id': user.get('username'),
 174                 'text': text,
 175                 'timestamp': parse_iso8601(node.get('created')),
 176             })
 177
 178         tags = []
 179         for t in cloudcast.get('tags'):
 180             tag = try_get(t, lambda x: x['tag']['name'], compat_str)
 181             if not tag:
 182                 tags.append(tag)
 183
 184         get_count = lambda x: int_or_none(try_get(cloudcast, lambda y: y[x]['totalCount']))
 185
 186         owner = cloudcast.get('owner') or {}
 187
 188         return {
 189             'id': track_id,
 190             'title': title,
 191             'formats': formats,
 192             'description': cloudcast.get('description'),
 193             'thumbnail': try_get(cloudcast, lambda x: x['picture']['url'], compat_str),
 194             'uploader': owner.get('displayName'),
 195             'timestamp': parse_iso8601(cloudcast.get('publishDate')),
 196             'uploader_id': owner.get('username'),
 197             'uploader_url': owner.get('url'),
 198             'duration': int_or_none(cloudcast.get('audioLength')),
 199             'view_count': int_or_none(cloudcast.get('plays')),
 200             'like_count': get_count('favorites'),
 201             'repost_count': get_count('reposts'),
 202             'comment_count': get_count('comments'),
 203             'comments': comments,
 204             'tags': tags,
 205             'artist': ', '.join(cloudcast.get('featuringArtistList') or []) or None,
 206         }
 207
 208
 209 class MixcloudPlaylistBaseIE(MixcloudBaseIE):
 210     def _get_cloudcast(self, node):
 211         return node
 212
 213     def _get_playlist_title(self, title, slug):
 214         return title
 215
 216     def _real_extract(self, url):
 217         username, slug = re.match(self._VALID_URL, url).groups()
 218         username = compat_urllib_parse_unquote(username)
 219         if not slug:
 220             slug = 'uploads'
 221         else:
 222             slug = compat_urllib_parse_unquote(slug)
 223         playlist_id = '%s_%s' % (username, slug)
 224
 225         is_playlist_type = self._ROOT_TYPE == 'playlist'
 226         playlist_type = 'items' if is_playlist_type else slug
 227         list_filter = ''
 228
 229         has_next_page = True
 230         entries = []
 231         while has_next_page:
 232             playlist = self._call_api(
 233                 self._ROOT_TYPE, '''%s
 234     %s
 235     %s(first: 100%s) {
 236       edges {
 237         node {
 238           %s
 239         }
 240       }
 241       pageInfo {
 242         endCursor
 243         hasNextPage
 244       }
 245     }''' % (self._TITLE_KEY, self._DESCRIPTION_KEY, playlist_type, list_filter, self._NODE_TEMPLATE),
 246                 playlist_id, username, slug if is_playlist_type else None)
 247
 248             items = playlist.get(playlist_type) or {}
 249             for edge in items.get('edges', []):
 250                 cloudcast = self._get_cloudcast(edge.get('node') or {})
 251                 cloudcast_url = cloudcast.get('url')
 252                 if not cloudcast_url:
 253                     continue
 254                 slug = try_get(cloudcast, lambda x: x['slug'], compat_str)
 255                 owner_username = try_get(cloudcast, lambda x: x['owner']['username'], compat_str)
 256                 video_id = '%s_%s' % (owner_username, slug) if slug and owner_username else None
 257                 entries.append(self.url_result(
 258                     cloudcast_url, MixcloudIE.ie_key(), video_id))
 259
 260             page_info = items['pageInfo']
 261             has_next_page = page_info['hasNextPage']
 262             list_filter = ', after: "%s"' % page_info['endCursor']
 263
 264         return self.playlist_result(
 265             entries, playlist_id,
 266             self._get_playlist_title(playlist[self._TITLE_KEY], slug),
 267             playlist.get(self._DESCRIPTION_KEY))
 268
 269
 270 class MixcloudUserIE(MixcloudPlaylistBaseIE):
 271     _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<id>[^/]+)/(?P<type>uploads|favorites|listens|stream)?/?$'
 272     IE_NAME = 'mixcloud:user'
 273
 274     _TESTS = [{
 275         'url': 'http://www.mixcloud.com/dholbach/',
 276         'info_dict': {
 277             'id': 'dholbach_uploads',
 278             'title': 'Daniel Holbach (uploads)',
 279             'description': 'md5:b60d776f0bab534c5dabe0a34e47a789',
 280         },
 281         'playlist_mincount': 36,
 282     }, {
 283         'url': 'http://www.mixcloud.com/dholbach/uploads/',
 284         'info_dict': {
 285             'id': 'dholbach_uploads',
 286             'title': 'Daniel Holbach (uploads)',
 287             'description': 'md5:b60d776f0bab534c5dabe0a34e47a789',
 288         },
 289         'playlist_mincount': 36,
 290     }, {
 291         'url': 'http://www.mixcloud.com/dholbach/favorites/',
 292         'info_dict': {
 293             'id': 'dholbach_favorites',
 294             'title': 'Daniel Holbach (favorites)',
 295             'description': 'md5:b60d776f0bab534c5dabe0a34e47a789',
 296         },
 297         # 'params': {
 298         #     'playlist_items': '1-100',
 299         # },
 300         'playlist_mincount': 396,
 301     }, {
 302         'url': 'http://www.mixcloud.com/dholbach/listens/',
 303         'info_dict': {
 304             'id': 'dholbach_listens',
 305             'title': 'Daniel Holbach (listens)',
 306             'description': 'md5:b60d776f0bab534c5dabe0a34e47a789',
 307         },
 308         # 'params': {
 309         #     'playlist_items': '1-100',
 310         # },
 311         'playlist_mincount': 1623,
 312         'skip': 'Large list',
 313     }, {
 314         'url': 'https://www.mixcloud.com/FirstEar/stream/',
 315         'info_dict': {
 316             'id': 'FirstEar_stream',
 317             'title': 'First Ear (stream)',
 318             'description': 'Curators of good music\r\n\r\nfirstearmusic.com',
 319         },
 320         'playlist_mincount': 271,
 321     }]
 322
 323     _TITLE_KEY = 'displayName'
 324     _DESCRIPTION_KEY = 'biog'
 325     _ROOT_TYPE = 'user'
 326     _NODE_TEMPLATE = '''slug
 327           url
 328           owner { username }'''
 329
 330     def _get_playlist_title(self, title, slug):
 331         return '%s (%s)' % (title, slug)
 332
 333
 334 class MixcloudPlaylistIE(MixcloudPlaylistBaseIE):
 335     _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/playlists/(?P<playlist>[^/]+)/?$'
 336     IE_NAME = 'mixcloud:playlist'
 337
 338     _TESTS = [{
 339         'url': 'https://www.mixcloud.com/maxvibes/playlists/jazzcat-on-ness-radio/',
 340         'info_dict': {
 341             'id': 'maxvibes_jazzcat-on-ness-radio',
 342             'title': 'Ness Radio sessions',
 343         },
 344         'playlist_mincount': 59,
 345     }]
 346     _TITLE_KEY = 'name'
 347     _DESCRIPTION_KEY = 'description'
 348     _ROOT_TYPE = 'playlist'
 349     _NODE_TEMPLATE = '''cloudcast {
 350             slug
 351             url
 352             owner { username }
 353           }'''
 354
 355     def _get_cloudcast(self, node):
 356         return node.get('cloudcast') or {}