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