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