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