yt_dlp/extractor/soundcloud.py

   1 import itertools
   2 import re
   3 import json
   4 # import random
   5
   6 from .common import (
   7     InfoExtractor,
   8     SearchInfoExtractor
   9 )
  10 from ..compat import (
  11     compat_HTTPError,
  12     compat_str,
  13 )
  14 from ..utils import (
  15     error_to_compat_str,
  16     ExtractorError,
  17     float_or_none,
  18     HEADRequest,
  19     int_or_none,
  20     KNOWN_EXTENSIONS,
  21     mimetype2ext,
  22     remove_end,
  23     parse_qs,
  24     str_or_none,
  25     try_get,
  26     unified_timestamp,
  27     update_url_query,
  28     url_or_none,
  29     urlhandle_detect_ext,
  30     sanitized_Request,
  31 )
  32
  33
  34 class SoundcloudEmbedIE(InfoExtractor):
  35     _VALID_URL = r'https?://(?:w|player|p)\.soundcloud\.com/player/?.*?\burl=(?P<id>.+)'
  36     _TEST = {
  37         # from https://www.soundi.fi/uutiset/ennakkokuuntelussa-timo-kaukolammen-station-to-station-to-station-julkaisua-juhlitaan-tanaan-g-livelabissa/
  38         'url': 'https://w.soundcloud.com/player/?visual=true&url=https%3A%2F%2Fapi.soundcloud.com%2Fplaylists%2F922213810&show_artwork=true&maxwidth=640&maxheight=960&dnt=1&secret_token=s-ziYey',
  39         'only_matching': True,
  40     }
  41
  42     @staticmethod
  43     def _extract_urls(webpage):
  44         return [m.group('url') for m in re.finditer(
  45             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?://)?(?:w\.)?soundcloud\.com/player.+?)\1',
  46             webpage)]
  47
  48     def _real_extract(self, url):
  49         query = parse_qs(url)
  50         api_url = query['url'][0]
  51         secret_token = query.get('secret_token')
  52         if secret_token:
  53             api_url = update_url_query(api_url, {'secret_token': secret_token[0]})
  54         return self.url_result(api_url)
  55
  56
  57 class SoundcloudBaseIE(InfoExtractor):
  58     _NETRC_MACHINE = 'soundcloud'
  59
  60     _API_V2_BASE = 'https://api-v2.soundcloud.com/'
  61     _BASE_URL = 'https://soundcloud.com/'
  62     _USER_AGENT = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.105 Safari/537.36'
  63     _API_AUTH_QUERY_TEMPLATE = '?client_id=%s'
  64     _API_AUTH_URL_PW = 'https://api-auth.soundcloud.com/web-auth/sign-in/password%s'
  65     _API_VERIFY_AUTH_TOKEN = 'https://api-auth.soundcloud.com/connect/session%s'
  66     _access_token = None
  67     _HEADERS = {}
  68
  69     def _store_client_id(self, client_id):
  70         self.cache.store('soundcloud', 'client_id', client_id)
  71
  72     def _update_client_id(self):
  73         webpage = self._download_webpage('https://soundcloud.com/', None)
  74         for src in reversed(re.findall(r'<script[^>]+src="([^"]+)"', webpage)):
  75             script = self._download_webpage(src, None, fatal=False)
  76             if script:
  77                 client_id = self._search_regex(
  78                     r'client_id\s*:\s*"([0-9a-zA-Z]{32})"',
  79                     script, 'client id', default=None)
  80                 if client_id:
  81                     self._CLIENT_ID = client_id
  82                     self._store_client_id(client_id)
  83                     return
  84         raise ExtractorError('Unable to extract client id')
  85
  86     def _download_json(self, *args, **kwargs):
  87         non_fatal = kwargs.get('fatal') is False
  88         if non_fatal:
  89             del kwargs['fatal']
  90         query = kwargs.get('query', {}).copy()
  91         for _ in range(2):
  92             query['client_id'] = self._CLIENT_ID
  93             kwargs['query'] = query
  94             try:
  95                 return super()._download_json(*args, **kwargs)
  96             except ExtractorError as e:
  97                 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (401, 403):
  98                     self._store_client_id(None)
  99                     self._update_client_id()
 100                     continue
 101                 elif non_fatal:
 102                     self.report_warning(error_to_compat_str(e))
 103                     return False
 104                 raise
 105
 106     def _initialize_pre_login(self):
 107         self._CLIENT_ID = self.cache.load('soundcloud', 'client_id') or 'a3e059563d7fd3372b49b37f00a00bcf'
 108
 109     def _perform_login(self, username, password):
 110         if username != 'oauth':
 111             self.report_warning(
 112                 'Login using username and password is not currently supported. '
 113                 'Use "--username oauth --password <oauth_token>" to login using an oauth token')
 114         self._access_token = password
 115         query = self._API_AUTH_QUERY_TEMPLATE % self._CLIENT_ID
 116         payload = {'session': {'access_token': self._access_token}}
 117         token_verification = sanitized_Request(self._API_VERIFY_AUTH_TOKEN % query, json.dumps(payload).encode('utf-8'))
 118         response = self._download_json(token_verification, None, note='Verifying login token...', fatal=False)
 119         if response is not False:
 120             self._HEADERS = {'Authorization': 'OAuth ' + self._access_token}
 121             self.report_login()
 122         else:
 123             self.report_warning('Provided authorization token seems to be invalid. Continue as guest')
 124
 125         r'''
 126         def genDevId():
 127             def genNumBlock():
 128                 return ''.join([str(random.randrange(10)) for i in range(6)])
 129             return '-'.join([genNumBlock() for i in range(4)])
 130
 131         payload = {
 132             'client_id': self._CLIENT_ID,
 133             'recaptcha_pubkey': 'null',
 134             'recaptcha_response': 'null',
 135             'credentials': {
 136                 'identifier': username,
 137                 'password': password
 138             },
 139             'signature': self.sign(username, password, self._CLIENT_ID),
 140             'device_id': genDevId(),
 141             'user_agent': self._USER_AGENT
 142         }
 143
 144         query = self._API_AUTH_QUERY_TEMPLATE % self._CLIENT_ID
 145         login = sanitized_Request(self._API_AUTH_URL_PW % query, json.dumps(payload).encode('utf-8'))
 146         response = self._download_json(login, None)
 147         self._access_token = response.get('session').get('access_token')
 148         if not self._access_token:
 149             self.report_warning('Unable to get access token, login may has failed')
 150         else:
 151             self._HEADERS = {'Authorization': 'OAuth ' + self._access_token}
 152         '''
 153
 154     # signature generation
 155     def sign(self, user, pw, clid):
 156         a = 33
 157         i = 1
 158         s = 440123
 159         w = 117
 160         u = 1800000
 161         l = 1042
 162         b = 37
 163         k = 37
 164         c = 5
 165         n = '0763ed7314c69015fd4a0dc16bbf4b90'  # _KEY
 166         y = '8'  # _REV
 167         r = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.105 Safari/537.36'  # _USER_AGENT
 168         e = user  # _USERNAME
 169         t = clid  # _CLIENT_ID
 170
 171         d = '-'.join([str(mInt) for mInt in [a, i, s, w, u, l, b, k]])
 172         p = n + y + d + r + e + t + d + n
 173         h = p
 174
 175         m = 8011470
 176         f = 0
 177
 178         for f in range(f, len(h)):
 179             m = (m >> 1) + ((1 & m) << 23)
 180             m += ord(h[f])
 181             m &= 16777215
 182
 183         # c is not even needed
 184         out = str(y) + ':' + str(d) + ':' + format(m, 'x') + ':' + str(c)
 185
 186         return out
 187
 188     @classmethod
 189     def _resolv_url(cls, url):
 190         return cls._API_V2_BASE + 'resolve?url=' + url
 191
 192
 193 class SoundcloudIE(SoundcloudBaseIE):
 194     """Information extractor for soundcloud.com
 195        To access the media, the uid of the song and a stream token
 196        must be extracted from the page source and the script must make
 197        a request to media.soundcloud.com/crossdomain.xml. Then
 198        the media can be grabbed by requesting from an url composed
 199        of the stream token and uid
 200      """
 201
 202     _VALID_URL = r'''(?x)^(?:https?://)?
 203                     (?:(?:(?:www\.|m\.)?soundcloud\.com/
 204                             (?!stations/track)
 205                             (?P<uploader>[\w\d-]+)/
 206                             (?!(?:tracks|albums|sets(?:/.+?)?|reposts|likes|spotlight)/?(?:$|[?#]))
 207                             (?P<title>[\w\d-]+)
 208                             (?:/(?P<token>(?!(?:albums|sets|recommended))[^?]+?))?
 209                             (?:[?].*)?$)
 210                        |(?:api(?:-v2)?\.soundcloud\.com/tracks/(?P<track_id>\d+)
 211                           (?:/?\?secret_token=(?P<secret_token>[^&]+))?)
 212                     )
 213                     '''
 214     IE_NAME = 'soundcloud'
 215     _TESTS = [
 216         {
 217             'url': 'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
 218             'md5': 'ebef0a451b909710ed1d7787dddbf0d7',
 219             'info_dict': {
 220                 'id': '62986583',
 221                 'ext': 'mp3',
 222                 'title': 'Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1',
 223                 'description': 'No Downloads untill we record the finished version this weekend, i was too pumped n i had to post it , earl is prolly gonna b hella p.o\'d',
 224                 'uploader': 'E.T. ExTerrestrial Music',
 225                 'uploader_id': '1571244',
 226                 'timestamp': 1349920598,
 227                 'upload_date': '20121011',
 228                 'duration': 143.216,
 229                 'license': 'all-rights-reserved',
 230                 'view_count': int,
 231                 'like_count': int,
 232                 'comment_count': int,
 233                 'repost_count': int,
 234             }
 235         },
 236         # geo-restricted
 237         {
 238             'url': 'https://soundcloud.com/the-concept-band/goldrushed-mastered?in=the-concept-band/sets/the-royal-concept-ep',
 239             'info_dict': {
 240                 'id': '47127627',
 241                 'ext': 'mp3',
 242                 'title': 'Goldrushed',
 243                 'description': 'From Stockholm Sweden\r\nPovel / Magnus / Filip / David\r\nwww.theroyalconcept.com',
 244                 'uploader': 'The Royal Concept',
 245                 'uploader_id': '9615865',
 246                 'timestamp': 1337635207,
 247                 'upload_date': '20120521',
 248                 'duration': 227.155,
 249                 'license': 'all-rights-reserved',
 250                 'view_count': int,
 251                 'like_count': int,
 252                 'comment_count': int,
 253                 'repost_count': int,
 254             },
 255         },
 256         # private link
 257         {
 258             'url': 'https://soundcloud.com/jaimemf/youtube-dl-test-video-a-y-baw/s-8Pjrp',
 259             'md5': 'aa0dd32bfea9b0c5ef4f02aacd080604',
 260             'info_dict': {
 261                 'id': '123998367',
 262                 'ext': 'mp3',
 263                 'title': 'Youtube - Dl Test Video \'\' Ä↭',
 264                 'description': 'test chars:  \"\'/\\ä↭',
 265                 'uploader': 'jaimeMF',
 266                 'uploader_id': '69767071',
 267                 'timestamp': 1386604920,
 268                 'upload_date': '20131209',
 269                 'duration': 9.927,
 270                 'license': 'all-rights-reserved',
 271                 'view_count': int,
 272                 'like_count': int,
 273                 'comment_count': int,
 274                 'repost_count': int,
 275             },
 276         },
 277         # private link (alt format)
 278         {
 279             'url': 'https://api.soundcloud.com/tracks/123998367?secret_token=s-8Pjrp',
 280             'md5': 'aa0dd32bfea9b0c5ef4f02aacd080604',
 281             'info_dict': {
 282                 'id': '123998367',
 283                 'ext': 'mp3',
 284                 'title': 'Youtube - Dl Test Video \'\' Ä↭',
 285                 'description': 'test chars:  \"\'/\\ä↭',
 286                 'uploader': 'jaimeMF',
 287                 'uploader_id': '69767071',
 288                 'timestamp': 1386604920,
 289                 'upload_date': '20131209',
 290                 'duration': 9.927,
 291                 'license': 'all-rights-reserved',
 292                 'view_count': int,
 293                 'like_count': int,
 294                 'comment_count': int,
 295                 'repost_count': int,
 296             },
 297         },
 298         # downloadable song
 299         {
 300             'url': 'https://soundcloud.com/the80m/the-following',
 301             'md5': '9ffcddb08c87d74fb5808a3c183a1d04',
 302             'info_dict': {
 303                 'id': '343609555',
 304                 'ext': 'wav',
 305             },
 306         },
 307         # private link, downloadable format
 308         {
 309             'url': 'https://soundcloud.com/oriuplift/uponly-238-no-talking-wav/s-AyZUd',
 310             'md5': '64a60b16e617d41d0bef032b7f55441e',
 311             'info_dict': {
 312                 'id': '340344461',
 313                 'ext': 'wav',
 314                 'title': 'Uplifting Only 238 [No Talking] (incl. Alex Feed Guestmix) (Aug 31, 2017) [wav]',
 315                 'description': 'md5:fa20ee0fca76a3d6df8c7e57f3715366',
 316                 'uploader': 'Ori Uplift Music',
 317                 'uploader_id': '12563093',
 318                 'timestamp': 1504206263,
 319                 'upload_date': '20170831',
 320                 'duration': 7449.096,
 321                 'license': 'all-rights-reserved',
 322                 'view_count': int,
 323                 'like_count': int,
 324                 'comment_count': int,
 325                 'repost_count': int,
 326             },
 327         },
 328         # no album art, use avatar pic for thumbnail
 329         {
 330             'url': 'https://soundcloud.com/garyvee/sideways-prod-mad-real',
 331             'md5': '59c7872bc44e5d99b7211891664760c2',
 332             'info_dict': {
 333                 'id': '309699954',
 334                 'ext': 'mp3',
 335                 'title': 'Sideways (Prod. Mad Real)',
 336                 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
 337                 'uploader': 'garyvee',
 338                 'uploader_id': '2366352',
 339                 'timestamp': 1488152409,
 340                 'upload_date': '20170226',
 341                 'duration': 207.012,
 342                 'thumbnail': r're:https?://.*\.jpg',
 343                 'license': 'all-rights-reserved',
 344                 'view_count': int,
 345                 'like_count': int,
 346                 'comment_count': int,
 347                 'repost_count': int,
 348             },
 349             'params': {
 350                 'skip_download': True,
 351             },
 352         },
 353         {
 354             'url': 'https://soundcloud.com/giovannisarani/mezzo-valzer',
 355             'md5': 'e22aecd2bc88e0e4e432d7dcc0a1abf7',
 356             'info_dict': {
 357                 'id': '583011102',
 358                 'ext': 'mp3',
 359                 'title': 'Mezzo Valzer',
 360                 'description': 'md5:4138d582f81866a530317bae316e8b61',
 361                 'uploader': 'Micronie',
 362                 'uploader_id': '3352531',
 363                 'timestamp': 1551394171,
 364                 'upload_date': '20190228',
 365                 'duration': 180.157,
 366                 'thumbnail': r're:https?://.*\.jpg',
 367                 'license': 'all-rights-reserved',
 368                 'view_count': int,
 369                 'like_count': int,
 370                 'comment_count': int,
 371                 'repost_count': int,
 372             },
 373         },
 374         {
 375             # AAC HQ format available (account with active subscription needed)
 376             'url': 'https://soundcloud.com/wandw/the-chainsmokers-ft-daya-dont-let-me-down-ww-remix-1',
 377             'only_matching': True,
 378         },
 379         {
 380             # Go+ (account with active subscription needed)
 381             'url': 'https://soundcloud.com/taylorswiftofficial/look-what-you-made-me-do',
 382             'only_matching': True,
 383         },
 384     ]
 385
 386     _IMAGE_REPL_RE = r'-([0-9a-z]+)\.jpg'
 387
 388     _ARTWORK_MAP = {
 389         'mini': 16,
 390         'tiny': 20,
 391         'small': 32,
 392         'badge': 47,
 393         't67x67': 67,
 394         'large': 100,
 395         't300x300': 300,
 396         'crop': 400,
 397         't500x500': 500,
 398         'original': 0,
 399     }
 400
 401     def _extract_info_dict(self, info, full_title=None, secret_token=None):
 402         track_id = compat_str(info['id'])
 403         title = info['title']
 404
 405         format_urls = set()
 406         formats = []
 407         query = {'client_id': self._CLIENT_ID}
 408         if secret_token:
 409             query['secret_token'] = secret_token
 410
 411         if info.get('downloadable') and info.get('has_downloads_left'):
 412             download_url = update_url_query(
 413                 self._API_V2_BASE + 'tracks/' + track_id + '/download', query)
 414             redirect_url = (self._download_json(download_url, track_id, fatal=False) or {}).get('redirectUri')
 415             if redirect_url:
 416                 urlh = self._request_webpage(
 417                     HEADRequest(redirect_url), track_id, fatal=False)
 418                 if urlh:
 419                     format_url = urlh.geturl()
 420                     format_urls.add(format_url)
 421                     formats.append({
 422                         'format_id': 'download',
 423                         'ext': urlhandle_detect_ext(urlh) or 'mp3',
 424                         'filesize': int_or_none(urlh.headers.get('Content-Length')),
 425                         'url': format_url,
 426                         'quality': 10,
 427                     })
 428
 429         def invalid_url(url):
 430             return not url or url in format_urls
 431
 432         def add_format(f, protocol, is_preview=False):
 433             mobj = re.search(r'\.(?P<abr>\d+)\.(?P<ext>[0-9a-z]{3,4})(?=[/?])', stream_url)
 434             if mobj:
 435                 for k, v in mobj.groupdict().items():
 436                     if not f.get(k):
 437                         f[k] = v
 438             format_id_list = []
 439             if protocol:
 440                 format_id_list.append(protocol)
 441             ext = f.get('ext')
 442             if ext == 'aac':
 443                 f['abr'] = '256'
 444             for k in ('ext', 'abr'):
 445                 v = f.get(k)
 446                 if v:
 447                     format_id_list.append(v)
 448             preview = is_preview or re.search(r'/(?:preview|playlist)/0/30/', f['url'])
 449             if preview:
 450                 format_id_list.append('preview')
 451             abr = f.get('abr')
 452             if abr:
 453                 f['abr'] = int(abr)
 454             if protocol == 'hls':
 455                 protocol = 'm3u8' if ext == 'aac' else 'm3u8_native'
 456             else:
 457                 protocol = 'http'
 458             f.update({
 459                 'format_id': '_'.join(format_id_list),
 460                 'protocol': protocol,
 461                 'preference': -10 if preview else None,
 462             })
 463             formats.append(f)
 464
 465         # New API
 466         transcodings = try_get(
 467             info, lambda x: x['media']['transcodings'], list) or []
 468         for t in transcodings:
 469             if not isinstance(t, dict):
 470                 continue
 471             format_url = url_or_none(t.get('url'))
 472             if not format_url:
 473                 continue
 474             stream = self._download_json(
 475                 format_url, track_id, query=query, fatal=False, headers=self._HEADERS)
 476             if not isinstance(stream, dict):
 477                 continue
 478             stream_url = url_or_none(stream.get('url'))
 479             if invalid_url(stream_url):
 480                 continue
 481             format_urls.add(stream_url)
 482             stream_format = t.get('format') or {}
 483             protocol = stream_format.get('protocol')
 484             if protocol != 'hls' and '/hls' in format_url:
 485                 protocol = 'hls'
 486             ext = None
 487             preset = str_or_none(t.get('preset'))
 488             if preset:
 489                 ext = preset.split('_')[0]
 490             if ext not in KNOWN_EXTENSIONS:
 491                 ext = mimetype2ext(stream_format.get('mime_type'))
 492             add_format({
 493                 'url': stream_url,
 494                 'ext': ext,
 495             }, 'http' if protocol == 'progressive' else protocol,
 496                 t.get('snipped') or '/preview/' in format_url)
 497
 498         for f in formats:
 499             f['vcodec'] = 'none'
 500
 501         if not formats and info.get('policy') == 'BLOCK':
 502             self.raise_geo_restricted(metadata_available=True)
 503         self._sort_formats(formats)
 504
 505         user = info.get('user') or {}
 506
 507         thumbnails = []
 508         artwork_url = info.get('artwork_url')
 509         thumbnail = artwork_url or user.get('avatar_url')
 510         if isinstance(thumbnail, compat_str):
 511             if re.search(self._IMAGE_REPL_RE, thumbnail):
 512                 for image_id, size in self._ARTWORK_MAP.items():
 513                     i = {
 514                         'id': image_id,
 515                         'url': re.sub(self._IMAGE_REPL_RE, '-%s.jpg' % image_id, thumbnail),
 516                     }
 517                     if image_id == 'tiny' and not artwork_url:
 518                         size = 18
 519                     elif image_id == 'original':
 520                         i['preference'] = 10
 521                     if size:
 522                         i.update({
 523                             'width': size,
 524                             'height': size,
 525                         })
 526                     thumbnails.append(i)
 527             else:
 528                 thumbnails = [{'url': thumbnail}]
 529
 530         def extract_count(key):
 531             return int_or_none(info.get('%s_count' % key))
 532
 533         return {
 534             'id': track_id,
 535             'uploader': user.get('username'),
 536             'uploader_id': str_or_none(user.get('id')) or user.get('permalink'),
 537             'uploader_url': user.get('permalink_url'),
 538             'timestamp': unified_timestamp(info.get('created_at')),
 539             'title': title,
 540             'description': info.get('description'),
 541             'thumbnails': thumbnails,
 542             'duration': float_or_none(info.get('duration'), 1000),
 543             'webpage_url': info.get('permalink_url'),
 544             'license': info.get('license'),
 545             'view_count': extract_count('playback'),
 546             'like_count': extract_count('favoritings') or extract_count('likes'),
 547             'comment_count': extract_count('comment'),
 548             'repost_count': extract_count('reposts'),
 549             'genre': info.get('genre'),
 550             'formats': formats
 551         }
 552
 553     def _real_extract(self, url):
 554         mobj = self._match_valid_url(url)
 555
 556         track_id = mobj.group('track_id')
 557
 558         query = {}
 559         if track_id:
 560             info_json_url = self._API_V2_BASE + 'tracks/' + track_id
 561             full_title = track_id
 562             token = mobj.group('secret_token')
 563             if token:
 564                 query['secret_token'] = token
 565         else:
 566             full_title = resolve_title = '%s/%s' % mobj.group('uploader', 'title')
 567             token = mobj.group('token')
 568             if token:
 569                 resolve_title += '/%s' % token
 570             info_json_url = self._resolv_url(self._BASE_URL + resolve_title)
 571
 572         info = self._download_json(
 573             info_json_url, full_title, 'Downloading info JSON', query=query, headers=self._HEADERS)
 574
 575         return self._extract_info_dict(info, full_title, token)
 576
 577
 578 class SoundcloudPlaylistBaseIE(SoundcloudBaseIE):
 579     def _extract_set(self, playlist, token=None):
 580         playlist_id = compat_str(playlist['id'])
 581         tracks = playlist.get('tracks') or []
 582         if not all([t.get('permalink_url') for t in tracks]) and token:
 583             tracks = self._download_json(
 584                 self._API_V2_BASE + 'tracks', playlist_id,
 585                 'Downloading tracks', query={
 586                     'ids': ','.join([compat_str(t['id']) for t in tracks]),
 587                     'playlistId': playlist_id,
 588                     'playlistSecretToken': token,
 589                 }, headers=self._HEADERS)
 590         entries = []
 591         for track in tracks:
 592             track_id = str_or_none(track.get('id'))
 593             url = track.get('permalink_url')
 594             if not url:
 595                 if not track_id:
 596                     continue
 597                 url = self._API_V2_BASE + 'tracks/' + track_id
 598                 if token:
 599                     url += '?secret_token=' + token
 600             entries.append(self.url_result(
 601                 url, SoundcloudIE.ie_key(), track_id))
 602         return self.playlist_result(
 603             entries, playlist_id,
 604             playlist.get('title'),
 605             playlist.get('description'))
 606
 607
 608 class SoundcloudSetIE(SoundcloudPlaylistBaseIE):
 609     _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/(?P<uploader>[\w\d-]+)/sets/(?P<slug_title>[:\w\d-]+)(?:/(?P<token>[^?/]+))?'
 610     IE_NAME = 'soundcloud:set'
 611     _TESTS = [{
 612         'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep',
 613         'info_dict': {
 614             'id': '2284613',
 615             'title': 'The Royal Concept EP',
 616             'description': 'md5:71d07087c7a449e8941a70a29e34671e',
 617         },
 618         'playlist_mincount': 5,
 619     }, {
 620         'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep/token',
 621         'only_matching': True,
 622     }, {
 623         'url': 'https://soundcloud.com/discover/sets/weekly::flacmatic',
 624         'only_matching': True,
 625     }, {
 626         'url': 'https://soundcloud.com/discover/sets/charts-top:all-music:de',
 627         'only_matching': True,
 628     }, {
 629         'url': 'https://soundcloud.com/discover/sets/charts-top:hiphoprap:kr',
 630         'only_matching': True,
 631     }]
 632
 633     def _real_extract(self, url):
 634         mobj = self._match_valid_url(url)
 635
 636         full_title = '%s/sets/%s' % mobj.group('uploader', 'slug_title')
 637         token = mobj.group('token')
 638         if token:
 639             full_title += '/' + token
 640
 641         info = self._download_json(self._resolv_url(
 642             self._BASE_URL + full_title), full_title, headers=self._HEADERS)
 643
 644         if 'errors' in info:
 645             msgs = (compat_str(err['error_message']) for err in info['errors'])
 646             raise ExtractorError('unable to download video webpage: %s' % ','.join(msgs))
 647
 648         return self._extract_set(info, token)
 649
 650
 651 class SoundcloudPagedPlaylistBaseIE(SoundcloudBaseIE):
 652     def _extract_playlist(self, base_url, playlist_id, playlist_title):
 653         return {
 654             '_type': 'playlist',
 655             'id': playlist_id,
 656             'title': playlist_title,
 657             'entries': self._entries(base_url, playlist_id),
 658         }
 659
 660     def _entries(self, url, playlist_id):
 661         # Per the SoundCloud documentation, the maximum limit for a linked partitioning query is 200.
 662         # https://developers.soundcloud.com/blog/offset-pagination-deprecated
 663         query = {
 664             'limit': 200,
 665             'linked_partitioning': '1',
 666             'offset': 0,
 667         }
 668
 669         retries = self.get_param('extractor_retries', 3)
 670
 671         for i in itertools.count():
 672             attempt, last_error = -1, None
 673             while attempt < retries:
 674                 attempt += 1
 675                 if last_error:
 676                     self.report_warning('%s. Retrying ...' % remove_end(last_error, '.'), playlist_id)
 677                 try:
 678                     response = self._download_json(
 679                         url, playlist_id, query=query, headers=self._HEADERS,
 680                         note='Downloading track page %s%s' % (i + 1, f' (retry #{attempt})' if attempt else ''))
 681                     break
 682                 except ExtractorError as e:
 683                     # Downloading page may result in intermittent 502 HTTP error
 684                     # See https://github.com/yt-dlp/yt-dlp/issues/872
 685                     if attempt >= retries or not isinstance(e.cause, compat_HTTPError) or e.cause.code != 502:
 686                         raise
 687                     last_error = str(e.cause or e.msg)
 688
 689             def resolve_entry(*candidates):
 690                 for cand in candidates:
 691                     if not isinstance(cand, dict):
 692                         continue
 693                     permalink_url = url_or_none(cand.get('permalink_url'))
 694                     if permalink_url:
 695                         return self.url_result(
 696                             permalink_url,
 697                             SoundcloudIE.ie_key() if SoundcloudIE.suitable(permalink_url) else None,
 698                             str_or_none(cand.get('id')), cand.get('title'))
 699
 700             for e in response['collection'] or []:
 701                 yield resolve_entry(e, e.get('track'), e.get('playlist'))
 702
 703             url = response.get('next_href')
 704             if not url:
 705                 break
 706             query.pop('offset', None)
 707
 708
 709 class SoundcloudUserIE(SoundcloudPagedPlaylistBaseIE):
 710     _VALID_URL = r'''(?x)
 711                         https?://
 712                             (?:(?:www|m)\.)?soundcloud\.com/
 713                             (?P<user>[^/]+)
 714                             (?:/
 715                                 (?P<rsrc>tracks|albums|sets|reposts|likes|spotlight)
 716                             )?
 717                             /?(?:[?#].*)?$
 718                     '''
 719     IE_NAME = 'soundcloud:user'
 720     _TESTS = [{
 721         'url': 'https://soundcloud.com/soft-cell-official',
 722         'info_dict': {
 723             'id': '207965082',
 724             'title': 'Soft Cell (All)',
 725         },
 726         'playlist_mincount': 28,
 727     }, {
 728         'url': 'https://soundcloud.com/soft-cell-official/tracks',
 729         'info_dict': {
 730             'id': '207965082',
 731             'title': 'Soft Cell (Tracks)',
 732         },
 733         'playlist_mincount': 27,
 734     }, {
 735         'url': 'https://soundcloud.com/soft-cell-official/albums',
 736         'info_dict': {
 737             'id': '207965082',
 738             'title': 'Soft Cell (Albums)',
 739         },
 740         'playlist_mincount': 1,
 741     }, {
 742         'url': 'https://soundcloud.com/jcv246/sets',
 743         'info_dict': {
 744             'id': '12982173',
 745             'title': 'Jordi / cv (Sets)',
 746         },
 747         'playlist_mincount': 2,
 748     }, {
 749         'url': 'https://soundcloud.com/jcv246/reposts',
 750         'info_dict': {
 751             'id': '12982173',
 752             'title': 'Jordi / cv (Reposts)',
 753         },
 754         'playlist_mincount': 6,
 755     }, {
 756         'url': 'https://soundcloud.com/clalberg/likes',
 757         'info_dict': {
 758             'id': '11817582',
 759             'title': 'clalberg (Likes)',
 760         },
 761         'playlist_mincount': 5,
 762     }, {
 763         'url': 'https://soundcloud.com/grynpyret/spotlight',
 764         'info_dict': {
 765             'id': '7098329',
 766             'title': 'Grynpyret (Spotlight)',
 767         },
 768         'playlist_mincount': 1,
 769     }]
 770
 771     _BASE_URL_MAP = {
 772         'all': 'stream/users/%s',
 773         'tracks': 'users/%s/tracks',
 774         'albums': 'users/%s/albums',
 775         'sets': 'users/%s/playlists',
 776         'reposts': 'stream/users/%s/reposts',
 777         'likes': 'users/%s/likes',
 778         'spotlight': 'users/%s/spotlight',
 779     }
 780
 781     def _real_extract(self, url):
 782         mobj = self._match_valid_url(url)
 783         uploader = mobj.group('user')
 784
 785         user = self._download_json(
 786             self._resolv_url(self._BASE_URL + uploader),
 787             uploader, 'Downloading user info', headers=self._HEADERS)
 788
 789         resource = mobj.group('rsrc') or 'all'
 790
 791         return self._extract_playlist(
 792             self._API_V2_BASE + self._BASE_URL_MAP[resource] % user['id'],
 793             str_or_none(user.get('id')),
 794             '%s (%s)' % (user['username'], resource.capitalize()))
 795
 796
 797 class SoundcloudTrackStationIE(SoundcloudPagedPlaylistBaseIE):
 798     _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/stations/track/[^/]+/(?P<id>[^/?#&]+)'
 799     IE_NAME = 'soundcloud:trackstation'
 800     _TESTS = [{
 801         'url': 'https://soundcloud.com/stations/track/officialsundial/your-text',
 802         'info_dict': {
 803             'id': '286017854',
 804             'title': 'Track station: your text',
 805         },
 806         'playlist_mincount': 47,
 807     }]
 808
 809     def _real_extract(self, url):
 810         track_name = self._match_id(url)
 811
 812         track = self._download_json(self._resolv_url(url), track_name, headers=self._HEADERS)
 813         track_id = self._search_regex(
 814             r'soundcloud:track-stations:(\d+)', track['id'], 'track id')
 815
 816         return self._extract_playlist(
 817             self._API_V2_BASE + 'stations/%s/tracks' % track['id'],
 818             track_id, 'Track station: %s' % track['title'])
 819
 820
 821 class SoundcloudRelatedIE(SoundcloudPagedPlaylistBaseIE):
 822     _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/(?P<slug>[\w\d-]+/[\w\d-]+)/(?P<relation>albums|sets|recommended)'
 823     IE_NAME = 'soundcloud:related'
 824     _TESTS = [{
 825         'url': 'https://soundcloud.com/wajang/sexapil-pingers-5/recommended',
 826         'info_dict': {
 827             'id': '1084577272',
 828             'title': 'Sexapil - Pingers 5 (Recommended)',
 829         },
 830         'playlist_mincount': 50,
 831     }, {
 832         'url': 'https://soundcloud.com/wajang/sexapil-pingers-5/albums',
 833         'info_dict': {
 834             'id': '1084577272',
 835             'title': 'Sexapil - Pingers 5 (Albums)',
 836         },
 837         'playlist_mincount': 1,
 838     }, {
 839         'url': 'https://soundcloud.com/wajang/sexapil-pingers-5/sets',
 840         'info_dict': {
 841             'id': '1084577272',
 842             'title': 'Sexapil - Pingers 5 (Sets)',
 843         },
 844         'playlist_mincount': 4,
 845     }]
 846
 847     _BASE_URL_MAP = {
 848         'albums': 'tracks/%s/albums',
 849         'sets': 'tracks/%s/playlists_without_albums',
 850         'recommended': 'tracks/%s/related',
 851     }
 852
 853     def _real_extract(self, url):
 854         slug, relation = self._match_valid_url(url).group('slug', 'relation')
 855
 856         track = self._download_json(
 857             self._resolv_url(self._BASE_URL + slug),
 858             slug, 'Downloading track info', headers=self._HEADERS)
 859
 860         if track.get('errors'):
 861             raise ExtractorError(f'{self.IE_NAME} said: %s' % ','.join(
 862                 str(err['error_message']) for err in track['errors']), expected=True)
 863
 864         return self._extract_playlist(
 865             self._API_V2_BASE + self._BASE_URL_MAP[relation] % track['id'], str(track['id']),
 866             '%s (%s)' % (track.get('title') or slug, relation.capitalize()))
 867
 868
 869 class SoundcloudPlaylistIE(SoundcloudPlaylistBaseIE):
 870     _VALID_URL = r'https?://api(?:-v2)?\.soundcloud\.com/playlists/(?P<id>[0-9]+)(?:/?\?secret_token=(?P<token>[^&]+?))?$'
 871     IE_NAME = 'soundcloud:playlist'
 872     _TESTS = [{
 873         'url': 'https://api.soundcloud.com/playlists/4110309',
 874         'info_dict': {
 875             'id': '4110309',
 876             'title': 'TILT Brass - Bowery Poetry Club, August \'03 [Non-Site SCR 02]',
 877             'description': 're:.*?TILT Brass - Bowery Poetry Club',
 878         },
 879         'playlist_count': 6,
 880     }]
 881
 882     def _real_extract(self, url):
 883         mobj = self._match_valid_url(url)
 884         playlist_id = mobj.group('id')
 885
 886         query = {}
 887         token = mobj.group('token')
 888         if token:
 889             query['secret_token'] = token
 890
 891         data = self._download_json(
 892             self._API_V2_BASE + 'playlists/' + playlist_id,
 893             playlist_id, 'Downloading playlist', query=query, headers=self._HEADERS)
 894
 895         return self._extract_set(data, token)
 896
 897
 898 class SoundcloudSearchIE(SoundcloudBaseIE, SearchInfoExtractor):
 899     IE_NAME = 'soundcloud:search'
 900     IE_DESC = 'Soundcloud search'
 901     _SEARCH_KEY = 'scsearch'
 902     _TESTS = [{
 903         'url': 'scsearch15:post-avant jazzcore',
 904         'info_dict': {
 905             'title': 'post-avant jazzcore',
 906         },
 907         'playlist_count': 15,
 908     }]
 909
 910     _MAX_RESULTS_PER_PAGE = 200
 911     _DEFAULT_RESULTS_PER_PAGE = 50
 912
 913     def _get_collection(self, endpoint, collection_id, **query):
 914         limit = min(
 915             query.get('limit', self._DEFAULT_RESULTS_PER_PAGE),
 916             self._MAX_RESULTS_PER_PAGE)
 917         query.update({
 918             'limit': limit,
 919             'linked_partitioning': 1,
 920             'offset': 0,
 921         })
 922         next_url = update_url_query(self._API_V2_BASE + endpoint, query)
 923
 924         for i in itertools.count(1):
 925             response = self._download_json(
 926                 next_url, collection_id, f'Downloading page {i}',
 927                 'Unable to download API page', headers=self._HEADERS)
 928
 929             for item in response.get('collection') or []:
 930                 if item:
 931                     yield self.url_result(item['uri'], SoundcloudIE.ie_key())
 932
 933             next_url = response.get('next_href')
 934             if not next_url:
 935                 break
 936
 937     def _get_n_results(self, query, n):
 938         return self.playlist_result(itertools.islice(
 939             self._get_collection('search/tracks', query, limit=n, q=query),
 940             0, None if n == float('inf') else n), query, query)