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 compat_str
  11 from ..networking import HEADRequest, Request
  12 from ..networking.exceptions import HTTPError
  13 from ..utils import (
  14     error_to_compat_str,
  15     ExtractorError,
  16     float_or_none,
  17     int_or_none,
  18     KNOWN_EXTENSIONS,
  19     mimetype2ext,
  20     parse_qs,
  21     str_or_none,
  22     try_get,
  23     unified_timestamp,
  24     update_url_query,
  25     url_or_none,
  26     urlhandle_detect_ext,
  27 )
  28
  29
  30 class SoundcloudEmbedIE(InfoExtractor):
  31     _VALID_URL = r'https?://(?:w|player|p)\.soundcloud\.com/player/?.*?\burl=(?P<id>.+)'
  32     _EMBED_REGEX = [r'<iframe[^>]+src=(["\'])(?P<url>(?:https?://)?(?:w\.)?soundcloud\.com/player.+?)\1']
  33     _TEST = {
  34         # from https://www.soundi.fi/uutiset/ennakkokuuntelussa-timo-kaukolammen-station-to-station-to-station-julkaisua-juhlitaan-tanaan-g-livelabissa/
  35         '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',
  36         'only_matching': True,
  37     }
  38
  39     def _real_extract(self, url):
  40         query = parse_qs(url)
  41         api_url = query['url'][0]
  42         secret_token = query.get('secret_token')
  43         if secret_token:
  44             api_url = update_url_query(api_url, {'secret_token': secret_token[0]})
  45         return self.url_result(api_url)
  46
  47
  48 class SoundcloudBaseIE(InfoExtractor):
  49     _NETRC_MACHINE = 'soundcloud'
  50
  51     _API_V2_BASE = 'https://api-v2.soundcloud.com/'
  52     _BASE_URL = 'https://soundcloud.com/'
  53     _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'
  54     _API_AUTH_QUERY_TEMPLATE = '?client_id=%s'
  55     _API_AUTH_URL_PW = 'https://api-auth.soundcloud.com/web-auth/sign-in/password%s'
  56     _API_VERIFY_AUTH_TOKEN = 'https://api-auth.soundcloud.com/connect/session%s'
  57     _access_token = None
  58     _HEADERS = {}
  59
  60     _IMAGE_REPL_RE = r'-([0-9a-z]+)\.jpg'
  61
  62     _ARTWORK_MAP = {
  63         'mini': 16,
  64         'tiny': 20,
  65         'small': 32,
  66         'badge': 47,
  67         't67x67': 67,
  68         'large': 100,
  69         't300x300': 300,
  70         'crop': 400,
  71         't500x500': 500,
  72         'original': 0,
  73     }
  74
  75     def _store_client_id(self, client_id):
  76         self.cache.store('soundcloud', 'client_id', client_id)
  77
  78     def _update_client_id(self):
  79         webpage = self._download_webpage('https://soundcloud.com/', None)
  80         for src in reversed(re.findall(r'<script[^>]+src="([^"]+)"', webpage)):
  81             script = self._download_webpage(src, None, fatal=False)
  82             if script:
  83                 client_id = self._search_regex(
  84                     r'client_id\s*:\s*"([0-9a-zA-Z]{32})"',
  85                     script, 'client id', default=None)
  86                 if client_id:
  87                     self._CLIENT_ID = client_id
  88                     self._store_client_id(client_id)
  89                     return
  90         raise ExtractorError('Unable to extract client id')
  91
  92     def _download_json(self, *args, **kwargs):
  93         non_fatal = kwargs.get('fatal') is False
  94         if non_fatal:
  95             del kwargs['fatal']
  96         query = kwargs.get('query', {}).copy()
  97         for _ in range(2):
  98             query['client_id'] = self._CLIENT_ID
  99             kwargs['query'] = query
 100             try:
 101                 return super()._download_json(*args, **kwargs)
 102             except ExtractorError as e:
 103                 if isinstance(e.cause, HTTPError) and e.cause.status in (401, 403):
 104                     self._store_client_id(None)
 105                     self._update_client_id()
 106                     continue
 107                 elif non_fatal:
 108                     self.report_warning(error_to_compat_str(e))
 109                     return False
 110                 raise
 111
 112     def _initialize_pre_login(self):
 113         self._CLIENT_ID = self.cache.load('soundcloud', 'client_id') or 'a3e059563d7fd3372b49b37f00a00bcf'
 114
 115     def _perform_login(self, username, password):
 116         if username != 'oauth':
 117             self.report_warning(
 118                 'Login using username and password is not currently supported. '
 119                 'Use "--username oauth --password <oauth_token>" to login using an oauth token')
 120         self._access_token = password
 121         query = self._API_AUTH_QUERY_TEMPLATE % self._CLIENT_ID
 122         payload = {'session': {'access_token': self._access_token}}
 123         token_verification = Request(self._API_VERIFY_AUTH_TOKEN % query, json.dumps(payload).encode('utf-8'))
 124         response = self._download_json(token_verification, None, note='Verifying login token...', fatal=False)
 125         if response is not False:
 126             self._HEADERS = {'Authorization': 'OAuth ' + self._access_token}
 127             self.report_login()
 128         else:
 129             self.report_warning('Provided authorization token seems to be invalid. Continue as guest')
 130
 131         r'''
 132         def genDevId():
 133             def genNumBlock():
 134                 return ''.join([str(random.randrange(10)) for i in range(6)])
 135             return '-'.join([genNumBlock() for i in range(4)])
 136
 137         payload = {
 138             'client_id': self._CLIENT_ID,
 139             'recaptcha_pubkey': 'null',
 140             'recaptcha_response': 'null',
 141             'credentials': {
 142                 'identifier': username,
 143                 'password': password
 144             },
 145             'signature': self.sign(username, password, self._CLIENT_ID),
 146             'device_id': genDevId(),
 147             'user_agent': self._USER_AGENT
 148         }
 149
 150         query = self._API_AUTH_QUERY_TEMPLATE % self._CLIENT_ID
 151         login = sanitized_Request(self._API_AUTH_URL_PW % query, json.dumps(payload).encode('utf-8'))
 152         response = self._download_json(login, None)
 153         self._access_token = response.get('session').get('access_token')
 154         if not self._access_token:
 155             self.report_warning('Unable to get access token, login may has failed')
 156         else:
 157             self._HEADERS = {'Authorization': 'OAuth ' + self._access_token}
 158         '''
 159
 160     # signature generation
 161     def sign(self, user, pw, clid):
 162         a = 33
 163         i = 1
 164         s = 440123
 165         w = 117
 166         u = 1800000
 167         l = 1042
 168         b = 37
 169         k = 37
 170         c = 5
 171         n = '0763ed7314c69015fd4a0dc16bbf4b90'  # _KEY
 172         y = '8'  # _REV
 173         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
 174         e = user  # _USERNAME
 175         t = clid  # _CLIENT_ID
 176
 177         d = '-'.join([str(mInt) for mInt in [a, i, s, w, u, l, b, k]])
 178         p = n + y + d + r + e + t + d + n
 179         h = p
 180
 181         m = 8011470
 182         f = 0
 183
 184         for f in range(f, len(h)):
 185             m = (m >> 1) + ((1 & m) << 23)
 186             m += ord(h[f])
 187             m &= 16777215
 188
 189         # c is not even needed
 190         out = str(y) + ':' + str(d) + ':' + format(m, 'x') + ':' + str(c)
 191
 192         return out
 193
 194     def _extract_info_dict(self, info, full_title=None, secret_token=None, extract_flat=False):
 195         track_id = compat_str(info['id'])
 196         title = info['title']
 197
 198         format_urls = set()
 199         formats = []
 200         query = {'client_id': self._CLIENT_ID}
 201         if secret_token:
 202             query['secret_token'] = secret_token
 203
 204         if not extract_flat and info.get('downloadable') and info.get('has_downloads_left'):
 205             download_url = update_url_query(
 206                 self._API_V2_BASE + 'tracks/' + track_id + '/download', query)
 207             redirect_url = (self._download_json(download_url, track_id, fatal=False) or {}).get('redirectUri')
 208             if redirect_url:
 209                 urlh = self._request_webpage(
 210                     HEADRequest(redirect_url), track_id, fatal=False)
 211                 if urlh:
 212                     format_url = urlh.url
 213                     format_urls.add(format_url)
 214                     formats.append({
 215                         'format_id': 'download',
 216                         'ext': urlhandle_detect_ext(urlh) or 'mp3',
 217                         'filesize': int_or_none(urlh.headers.get('Content-Length')),
 218                         'url': format_url,
 219                         'quality': 10,
 220                     })
 221
 222         def invalid_url(url):
 223             return not url or url in format_urls
 224
 225         def add_format(f, protocol, is_preview=False):
 226             mobj = re.search(r'\.(?P<abr>\d+)\.(?P<ext>[0-9a-z]{3,4})(?=[/?])', stream_url)
 227             if mobj:
 228                 for k, v in mobj.groupdict().items():
 229                     if not f.get(k):
 230                         f[k] = v
 231             format_id_list = []
 232             if protocol:
 233                 format_id_list.append(protocol)
 234             ext = f.get('ext')
 235             if ext == 'aac':
 236                 f['abr'] = '256'
 237             for k in ('ext', 'abr'):
 238                 v = f.get(k)
 239                 if v:
 240                     format_id_list.append(v)
 241             preview = is_preview or re.search(r'/(?:preview|playlist)/0/30/', f['url'])
 242             if preview:
 243                 format_id_list.append('preview')
 244             abr = f.get('abr')
 245             if abr:
 246                 f['abr'] = int(abr)
 247             if protocol == 'hls':
 248                 protocol = 'm3u8' if ext == 'aac' else 'm3u8_native'
 249             else:
 250                 protocol = 'http'
 251             f.update({
 252                 'format_id': '_'.join(format_id_list),
 253                 'protocol': protocol,
 254                 'preference': -10 if preview else None,
 255             })
 256             formats.append(f)
 257
 258         # New API
 259         transcodings = try_get(
 260             info, lambda x: x['media']['transcodings'], list) or []
 261         for t in transcodings:
 262             if not isinstance(t, dict):
 263                 continue
 264             format_url = url_or_none(t.get('url'))
 265             if not format_url:
 266                 continue
 267             stream = None if extract_flat else self._download_json(
 268                 format_url, track_id, query=query, fatal=False, headers=self._HEADERS)
 269             if not isinstance(stream, dict):
 270                 continue
 271             stream_url = url_or_none(stream.get('url'))
 272             if invalid_url(stream_url):
 273                 continue
 274             format_urls.add(stream_url)
 275             stream_format = t.get('format') or {}
 276             protocol = stream_format.get('protocol')
 277             if protocol != 'hls' and '/hls' in format_url:
 278                 protocol = 'hls'
 279             ext = None
 280             preset = str_or_none(t.get('preset'))
 281             if preset:
 282                 ext = preset.split('_')[0]
 283             if ext not in KNOWN_EXTENSIONS:
 284                 ext = mimetype2ext(stream_format.get('mime_type'))
 285             add_format({
 286                 'url': stream_url,
 287                 'ext': ext,
 288             }, 'http' if protocol == 'progressive' else protocol,
 289                 t.get('snipped') or '/preview/' in format_url)
 290
 291         for f in formats:
 292             f['vcodec'] = 'none'
 293
 294         if not formats and info.get('policy') == 'BLOCK':
 295             self.raise_geo_restricted(metadata_available=True)
 296
 297         user = info.get('user') or {}
 298
 299         thumbnails = []
 300         artwork_url = info.get('artwork_url')
 301         thumbnail = artwork_url or user.get('avatar_url')
 302         if isinstance(thumbnail, compat_str):
 303             if re.search(self._IMAGE_REPL_RE, thumbnail):
 304                 for image_id, size in self._ARTWORK_MAP.items():
 305                     i = {
 306                         'id': image_id,
 307                         'url': re.sub(self._IMAGE_REPL_RE, '-%s.jpg' % image_id, thumbnail),
 308                     }
 309                     if image_id == 'tiny' and not artwork_url:
 310                         size = 18
 311                     elif image_id == 'original':
 312                         i['preference'] = 10
 313                     if size:
 314                         i.update({
 315                             'width': size,
 316                             'height': size,
 317                         })
 318                     thumbnails.append(i)
 319             else:
 320                 thumbnails = [{'url': thumbnail}]
 321
 322         def extract_count(key):
 323             return int_or_none(info.get('%s_count' % key))
 324
 325         return {
 326             'id': track_id,
 327             'uploader': user.get('username'),
 328             'uploader_id': str_or_none(user.get('id')) or user.get('permalink'),
 329             'uploader_url': user.get('permalink_url'),
 330             'timestamp': unified_timestamp(info.get('created_at')),
 331             'title': title,
 332             'description': info.get('description'),
 333             'thumbnails': thumbnails,
 334             'duration': float_or_none(info.get('duration'), 1000),
 335             'webpage_url': info.get('permalink_url'),
 336             'license': info.get('license'),
 337             'view_count': extract_count('playback'),
 338             'like_count': extract_count('favoritings') or extract_count('likes'),
 339             'comment_count': extract_count('comment'),
 340             'repost_count': extract_count('reposts'),
 341             'genre': info.get('genre'),
 342             'formats': formats if not extract_flat else None
 343         }
 344
 345     @classmethod
 346     def _resolv_url(cls, url):
 347         return cls._API_V2_BASE + 'resolve?url=' + url
 348
 349
 350 class SoundcloudIE(SoundcloudBaseIE):
 351     """Information extractor for soundcloud.com
 352        To access the media, the uid of the song and a stream token
 353        must be extracted from the page source and the script must make
 354        a request to media.soundcloud.com/crossdomain.xml. Then
 355        the media can be grabbed by requesting from an url composed
 356        of the stream token and uid
 357      """
 358
 359     _VALID_URL = r'''(?x)^(?:https?://)?
 360                     (?:(?:(?:www\.|m\.)?soundcloud\.com/
 361                             (?!stations/track)
 362                             (?P<uploader>[\w\d-]+)/
 363                             (?!(?:tracks|albums|sets(?:/.+?)?|reposts|likes|spotlight)/?(?:$|[?#]))
 364                             (?P<title>[\w\d-]+)
 365                             (?:/(?P<token>(?!(?:albums|sets|recommended))[^?]+?))?
 366                             (?:[?].*)?$)
 367                        |(?:api(?:-v2)?\.soundcloud\.com/tracks/(?P<track_id>\d+)
 368                           (?:/?\?secret_token=(?P<secret_token>[^&]+))?)
 369                     )
 370                     '''
 371     IE_NAME = 'soundcloud'
 372     _TESTS = [
 373         {
 374             'url': 'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
 375             'md5': 'ebef0a451b909710ed1d7787dddbf0d7',
 376             'info_dict': {
 377                 'id': '62986583',
 378                 'ext': 'mp3',
 379                 'title': 'Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1',
 380                 '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',
 381                 'uploader': 'E.T. ExTerrestrial Music',
 382                 'uploader_id': '1571244',
 383                 'timestamp': 1349920598,
 384                 'upload_date': '20121011',
 385                 'duration': 143.216,
 386                 'license': 'all-rights-reserved',
 387                 'view_count': int,
 388                 'like_count': int,
 389                 'comment_count': int,
 390                 'repost_count': int,
 391             }
 392         },
 393         # geo-restricted
 394         {
 395             'url': 'https://soundcloud.com/the-concept-band/goldrushed-mastered?in=the-concept-band/sets/the-royal-concept-ep',
 396             'info_dict': {
 397                 'id': '47127627',
 398                 'ext': 'mp3',
 399                 'title': 'Goldrushed',
 400                 'description': 'From Stockholm Sweden\r\nPovel / Magnus / Filip / David\r\nwww.theroyalconcept.com',
 401                 'uploader': 'The Royal Concept',
 402                 'uploader_id': '9615865',
 403                 'timestamp': 1337635207,
 404                 'upload_date': '20120521',
 405                 'duration': 227.155,
 406                 'license': 'all-rights-reserved',
 407                 'view_count': int,
 408                 'like_count': int,
 409                 'comment_count': int,
 410                 'repost_count': int,
 411             },
 412         },
 413         # private link
 414         {
 415             'url': 'https://soundcloud.com/jaimemf/youtube-dl-test-video-a-y-baw/s-8Pjrp',
 416             'md5': 'aa0dd32bfea9b0c5ef4f02aacd080604',
 417             'info_dict': {
 418                 'id': '123998367',
 419                 'ext': 'mp3',
 420                 'title': 'Youtube - Dl Test Video \'\' Ä↭',
 421                 'description': 'test chars:  \"\'/\\ä↭',
 422                 'uploader': 'jaimeMF',
 423                 'uploader_id': '69767071',
 424                 'timestamp': 1386604920,
 425                 'upload_date': '20131209',
 426                 'duration': 9.927,
 427                 'license': 'all-rights-reserved',
 428                 'view_count': int,
 429                 'like_count': int,
 430                 'comment_count': int,
 431                 'repost_count': int,
 432             },
 433         },
 434         # private link (alt format)
 435         {
 436             'url': 'https://api.soundcloud.com/tracks/123998367?secret_token=s-8Pjrp',
 437             'md5': 'aa0dd32bfea9b0c5ef4f02aacd080604',
 438             'info_dict': {
 439                 'id': '123998367',
 440                 'ext': 'mp3',
 441                 'title': 'Youtube - Dl Test Video \'\' Ä↭',
 442                 'description': 'test chars:  \"\'/\\ä↭',
 443                 'uploader': 'jaimeMF',
 444                 'uploader_id': '69767071',
 445                 'timestamp': 1386604920,
 446                 'upload_date': '20131209',
 447                 'duration': 9.927,
 448                 'license': 'all-rights-reserved',
 449                 'view_count': int,
 450                 'like_count': int,
 451                 'comment_count': int,
 452                 'repost_count': int,
 453             },
 454         },
 455         # downloadable song
 456         {
 457             'url': 'https://soundcloud.com/the80m/the-following',
 458             'md5': '9ffcddb08c87d74fb5808a3c183a1d04',
 459             'info_dict': {
 460                 'id': '343609555',
 461                 'ext': 'wav',
 462             },
 463         },
 464         # private link, downloadable format
 465         {
 466             'url': 'https://soundcloud.com/oriuplift/uponly-238-no-talking-wav/s-AyZUd',
 467             'md5': '64a60b16e617d41d0bef032b7f55441e',
 468             'info_dict': {
 469                 'id': '340344461',
 470                 'ext': 'wav',
 471                 'title': 'Uplifting Only 238 [No Talking] (incl. Alex Feed Guestmix) (Aug 31, 2017) [wav]',
 472                 'description': 'md5:fa20ee0fca76a3d6df8c7e57f3715366',
 473                 'uploader': 'Ori Uplift Music',
 474                 'uploader_id': '12563093',
 475                 'timestamp': 1504206263,
 476                 'upload_date': '20170831',
 477                 'duration': 7449.096,
 478                 'license': 'all-rights-reserved',
 479                 'view_count': int,
 480                 'like_count': int,
 481                 'comment_count': int,
 482                 'repost_count': int,
 483             },
 484         },
 485         # no album art, use avatar pic for thumbnail
 486         {
 487             'url': 'https://soundcloud.com/garyvee/sideways-prod-mad-real',
 488             'md5': '59c7872bc44e5d99b7211891664760c2',
 489             'info_dict': {
 490                 'id': '309699954',
 491                 'ext': 'mp3',
 492                 'title': 'Sideways (Prod. Mad Real)',
 493                 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
 494                 'uploader': 'garyvee',
 495                 'uploader_id': '2366352',
 496                 'timestamp': 1488152409,
 497                 'upload_date': '20170226',
 498                 'duration': 207.012,
 499                 'thumbnail': r're:https?://.*\.jpg',
 500                 'license': 'all-rights-reserved',
 501                 'view_count': int,
 502                 'like_count': int,
 503                 'comment_count': int,
 504                 'repost_count': int,
 505             },
 506             'params': {
 507                 'skip_download': True,
 508             },
 509         },
 510         {
 511             'url': 'https://soundcloud.com/giovannisarani/mezzo-valzer',
 512             'md5': 'e22aecd2bc88e0e4e432d7dcc0a1abf7',
 513             'info_dict': {
 514                 'id': '583011102',
 515                 'ext': 'mp3',
 516                 'title': 'Mezzo Valzer',
 517                 'description': 'md5:4138d582f81866a530317bae316e8b61',
 518                 'uploader': 'Micronie',
 519                 'uploader_id': '3352531',
 520                 'timestamp': 1551394171,
 521                 'upload_date': '20190228',
 522                 'duration': 180.157,
 523                 'thumbnail': r're:https?://.*\.jpg',
 524                 'license': 'all-rights-reserved',
 525                 'view_count': int,
 526                 'like_count': int,
 527                 'comment_count': int,
 528                 'repost_count': int,
 529             },
 530         },
 531         {
 532             # AAC HQ format available (account with active subscription needed)
 533             'url': 'https://soundcloud.com/wandw/the-chainsmokers-ft-daya-dont-let-me-down-ww-remix-1',
 534             'only_matching': True,
 535         },
 536         {
 537             # Go+ (account with active subscription needed)
 538             'url': 'https://soundcloud.com/taylorswiftofficial/look-what-you-made-me-do',
 539             'only_matching': True,
 540         },
 541     ]
 542
 543     def _real_extract(self, url):
 544         mobj = self._match_valid_url(url)
 545
 546         track_id = mobj.group('track_id')
 547
 548         query = {}
 549         if track_id:
 550             info_json_url = self._API_V2_BASE + 'tracks/' + track_id
 551             full_title = track_id
 552             token = mobj.group('secret_token')
 553             if token:
 554                 query['secret_token'] = token
 555         else:
 556             full_title = resolve_title = '%s/%s' % mobj.group('uploader', 'title')
 557             token = mobj.group('token')
 558             if token:
 559                 resolve_title += '/%s' % token
 560             info_json_url = self._resolv_url(self._BASE_URL + resolve_title)
 561
 562         info = self._download_json(
 563             info_json_url, full_title, 'Downloading info JSON', query=query, headers=self._HEADERS)
 564
 565         return self._extract_info_dict(info, full_title, token)
 566
 567
 568 class SoundcloudPlaylistBaseIE(SoundcloudBaseIE):
 569     def _extract_set(self, playlist, token=None):
 570         playlist_id = compat_str(playlist['id'])
 571         tracks = playlist.get('tracks') or []
 572         if not all([t.get('permalink_url') for t in tracks]) and token:
 573             tracks = self._download_json(
 574                 self._API_V2_BASE + 'tracks', playlist_id,
 575                 'Downloading tracks', query={
 576                     'ids': ','.join([compat_str(t['id']) for t in tracks]),
 577                     'playlistId': playlist_id,
 578                     'playlistSecretToken': token,
 579                 }, headers=self._HEADERS)
 580         entries = []
 581         for track in tracks:
 582             track_id = str_or_none(track.get('id'))
 583             url = track.get('permalink_url')
 584             if not url:
 585                 if not track_id:
 586                     continue
 587                 url = self._API_V2_BASE + 'tracks/' + track_id
 588                 if token:
 589                     url += '?secret_token=' + token
 590             entries.append(self.url_result(
 591                 url, SoundcloudIE.ie_key(), track_id))
 592         return self.playlist_result(
 593             entries, playlist_id,
 594             playlist.get('title'),
 595             playlist.get('description'))
 596
 597
 598 class SoundcloudSetIE(SoundcloudPlaylistBaseIE):
 599     _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/(?P<uploader>[\w\d-]+)/sets/(?P<slug_title>[:\w\d-]+)(?:/(?P<token>[^?/]+))?'
 600     IE_NAME = 'soundcloud:set'
 601     _TESTS = [{
 602         'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep',
 603         'info_dict': {
 604             'id': '2284613',
 605             'title': 'The Royal Concept EP',
 606             'description': 'md5:71d07087c7a449e8941a70a29e34671e',
 607         },
 608         'playlist_mincount': 5,
 609     }, {
 610         'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep/token',
 611         'only_matching': True,
 612     }, {
 613         'url': 'https://soundcloud.com/discover/sets/weekly::flacmatic',
 614         'only_matching': True,
 615     }, {
 616         'url': 'https://soundcloud.com/discover/sets/charts-top:all-music:de',
 617         'only_matching': True,
 618     }, {
 619         'url': 'https://soundcloud.com/discover/sets/charts-top:hiphoprap:kr',
 620         'only_matching': True,
 621     }]
 622
 623     def _real_extract(self, url):
 624         mobj = self._match_valid_url(url)
 625
 626         full_title = '%s/sets/%s' % mobj.group('uploader', 'slug_title')
 627         token = mobj.group('token')
 628         if token:
 629             full_title += '/' + token
 630
 631         info = self._download_json(self._resolv_url(
 632             self._BASE_URL + full_title), full_title, headers=self._HEADERS)
 633
 634         if 'errors' in info:
 635             msgs = (compat_str(err['error_message']) for err in info['errors'])
 636             raise ExtractorError('unable to download video webpage: %s' % ','.join(msgs))
 637
 638         return self._extract_set(info, token)
 639
 640
 641 class SoundcloudPagedPlaylistBaseIE(SoundcloudBaseIE):
 642     def _extract_playlist(self, base_url, playlist_id, playlist_title):
 643         return {
 644             '_type': 'playlist',
 645             'id': playlist_id,
 646             'title': playlist_title,
 647             'entries': self._entries(base_url, playlist_id),
 648         }
 649
 650     def _entries(self, url, playlist_id):
 651         # Per the SoundCloud documentation, the maximum limit for a linked partitioning query is 200.
 652         # https://developers.soundcloud.com/blog/offset-pagination-deprecated
 653         query = {
 654             'limit': 200,
 655             'linked_partitioning': '1',
 656             'offset': 0,
 657         }
 658
 659         for i in itertools.count():
 660             for retry in self.RetryManager():
 661                 try:
 662                     response = self._download_json(
 663                         url, playlist_id, query=query, headers=self._HEADERS,
 664                         note=f'Downloading track page {i + 1}')
 665                     break
 666                 except ExtractorError as e:
 667                     # Downloading page may result in intermittent 502 HTTP error
 668                     # See https://github.com/yt-dlp/yt-dlp/issues/872
 669                     if not isinstance(e.cause, HTTPError) or e.cause.status != 502:
 670                         raise
 671                     retry.error = e
 672                     continue
 673
 674             def resolve_entry(*candidates):
 675                 for cand in candidates:
 676                     if not isinstance(cand, dict):
 677                         continue
 678                     permalink_url = url_or_none(cand.get('permalink_url'))
 679                     if permalink_url:
 680                         return self.url_result(
 681                             permalink_url,
 682                             SoundcloudIE.ie_key() if SoundcloudIE.suitable(permalink_url) else None,
 683                             str_or_none(cand.get('id')), cand.get('title'))
 684
 685             for e in response['collection'] or []:
 686                 yield resolve_entry(e, e.get('track'), e.get('playlist'))
 687
 688             url = response.get('next_href')
 689             if not url:
 690                 break
 691             query.pop('offset', None)
 692
 693
 694 class SoundcloudUserIE(SoundcloudPagedPlaylistBaseIE):
 695     _VALID_URL = r'''(?x)
 696                         https?://
 697                             (?:(?:www|m)\.)?soundcloud\.com/
 698                             (?P<user>[^/]+)
 699                             (?:/
 700                                 (?P<rsrc>tracks|albums|sets|reposts|likes|spotlight)
 701                             )?
 702                             /?(?:[?#].*)?$
 703                     '''
 704     IE_NAME = 'soundcloud:user'
 705     _TESTS = [{
 706         'url': 'https://soundcloud.com/soft-cell-official',
 707         'info_dict': {
 708             'id': '207965082',
 709             'title': 'Soft Cell (All)',
 710         },
 711         'playlist_mincount': 28,
 712     }, {
 713         'url': 'https://soundcloud.com/soft-cell-official/tracks',
 714         'info_dict': {
 715             'id': '207965082',
 716             'title': 'Soft Cell (Tracks)',
 717         },
 718         'playlist_mincount': 27,
 719     }, {
 720         'url': 'https://soundcloud.com/soft-cell-official/albums',
 721         'info_dict': {
 722             'id': '207965082',
 723             'title': 'Soft Cell (Albums)',
 724         },
 725         'playlist_mincount': 1,
 726     }, {
 727         'url': 'https://soundcloud.com/jcv246/sets',
 728         'info_dict': {
 729             'id': '12982173',
 730             'title': 'Jordi / cv (Sets)',
 731         },
 732         'playlist_mincount': 2,
 733     }, {
 734         'url': 'https://soundcloud.com/jcv246/reposts',
 735         'info_dict': {
 736             'id': '12982173',
 737             'title': 'Jordi / cv (Reposts)',
 738         },
 739         'playlist_mincount': 6,
 740     }, {
 741         'url': 'https://soundcloud.com/clalberg/likes',
 742         'info_dict': {
 743             'id': '11817582',
 744             'title': 'clalberg (Likes)',
 745         },
 746         'playlist_mincount': 5,
 747     }, {
 748         'url': 'https://soundcloud.com/grynpyret/spotlight',
 749         'info_dict': {
 750             'id': '7098329',
 751             'title': 'Grynpyret (Spotlight)',
 752         },
 753         'playlist_mincount': 1,
 754     }]
 755
 756     _BASE_URL_MAP = {
 757         'all': 'stream/users/%s',
 758         'tracks': 'users/%s/tracks',
 759         'albums': 'users/%s/albums',
 760         'sets': 'users/%s/playlists',
 761         'reposts': 'stream/users/%s/reposts',
 762         'likes': 'users/%s/likes',
 763         'spotlight': 'users/%s/spotlight',
 764     }
 765
 766     def _real_extract(self, url):
 767         mobj = self._match_valid_url(url)
 768         uploader = mobj.group('user')
 769
 770         user = self._download_json(
 771             self._resolv_url(self._BASE_URL + uploader),
 772             uploader, 'Downloading user info', headers=self._HEADERS)
 773
 774         resource = mobj.group('rsrc') or 'all'
 775
 776         return self._extract_playlist(
 777             self._API_V2_BASE + self._BASE_URL_MAP[resource] % user['id'],
 778             str_or_none(user.get('id')),
 779             '%s (%s)' % (user['username'], resource.capitalize()))
 780
 781
 782 class SoundcloudUserPermalinkIE(SoundcloudPagedPlaylistBaseIE):
 783     _VALID_URL = r'https?://api\.soundcloud\.com/users/(?P<id>\d+)'
 784     IE_NAME = 'soundcloud:user:permalink'
 785     _TESTS = [{
 786         'url': 'https://api.soundcloud.com/users/30909869',
 787         'info_dict': {
 788             'id': '30909869',
 789             'title': 'neilcic',
 790         },
 791         'playlist_mincount': 23,
 792     }]
 793
 794     def _real_extract(self, url):
 795         user_id = self._match_id(url)
 796         user = self._download_json(
 797             self._resolv_url(url), user_id, 'Downloading user info', headers=self._HEADERS)
 798
 799         return self._extract_playlist(
 800             f'{self._API_V2_BASE}stream/users/{user["id"]}', str(user['id']), user.get('username'))
 801
 802
 803 class SoundcloudTrackStationIE(SoundcloudPagedPlaylistBaseIE):
 804     _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/stations/track/[^/]+/(?P<id>[^/?#&]+)'
 805     IE_NAME = 'soundcloud:trackstation'
 806     _TESTS = [{
 807         'url': 'https://soundcloud.com/stations/track/officialsundial/your-text',
 808         'info_dict': {
 809             'id': '286017854',
 810             'title': 'Track station: your text',
 811         },
 812         'playlist_mincount': 47,
 813     }]
 814
 815     def _real_extract(self, url):
 816         track_name = self._match_id(url)
 817
 818         track = self._download_json(self._resolv_url(url), track_name, headers=self._HEADERS)
 819         track_id = self._search_regex(
 820             r'soundcloud:track-stations:(\d+)', track['id'], 'track id')
 821
 822         return self._extract_playlist(
 823             self._API_V2_BASE + 'stations/%s/tracks' % track['id'],
 824             track_id, 'Track station: %s' % track['title'])
 825
 826
 827 class SoundcloudRelatedIE(SoundcloudPagedPlaylistBaseIE):
 828     _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/(?P<slug>[\w\d-]+/[\w\d-]+)/(?P<relation>albums|sets|recommended)'
 829     IE_NAME = 'soundcloud:related'
 830     _TESTS = [{
 831         'url': 'https://soundcloud.com/wajang/sexapil-pingers-5/recommended',
 832         'info_dict': {
 833             'id': '1084577272',
 834             'title': 'Sexapil - Pingers 5 (Recommended)',
 835         },
 836         'playlist_mincount': 50,
 837     }, {
 838         'url': 'https://soundcloud.com/wajang/sexapil-pingers-5/albums',
 839         'info_dict': {
 840             'id': '1084577272',
 841             'title': 'Sexapil - Pingers 5 (Albums)',
 842         },
 843         'playlist_mincount': 1,
 844     }, {
 845         'url': 'https://soundcloud.com/wajang/sexapil-pingers-5/sets',
 846         'info_dict': {
 847             'id': '1084577272',
 848             'title': 'Sexapil - Pingers 5 (Sets)',
 849         },
 850         'playlist_mincount': 4,
 851     }]
 852
 853     _BASE_URL_MAP = {
 854         'albums': 'tracks/%s/albums',
 855         'sets': 'tracks/%s/playlists_without_albums',
 856         'recommended': 'tracks/%s/related',
 857     }
 858
 859     def _real_extract(self, url):
 860         slug, relation = self._match_valid_url(url).group('slug', 'relation')
 861
 862         track = self._download_json(
 863             self._resolv_url(self._BASE_URL + slug),
 864             slug, 'Downloading track info', headers=self._HEADERS)
 865
 866         if track.get('errors'):
 867             raise ExtractorError(f'{self.IE_NAME} said: %s' % ','.join(
 868                 str(err['error_message']) for err in track['errors']), expected=True)
 869
 870         return self._extract_playlist(
 871             self._API_V2_BASE + self._BASE_URL_MAP[relation] % track['id'], str(track['id']),
 872             '%s (%s)' % (track.get('title') or slug, relation.capitalize()))
 873
 874
 875 class SoundcloudPlaylistIE(SoundcloudPlaylistBaseIE):
 876     _VALID_URL = r'https?://api(?:-v2)?\.soundcloud\.com/playlists/(?P<id>[0-9]+)(?:/?\?secret_token=(?P<token>[^&]+?))?$'
 877     IE_NAME = 'soundcloud:playlist'
 878     _TESTS = [{
 879         'url': 'https://api.soundcloud.com/playlists/4110309',
 880         'info_dict': {
 881             'id': '4110309',
 882             'title': 'TILT Brass - Bowery Poetry Club, August \'03 [Non-Site SCR 02]',
 883             'description': 're:.*?TILT Brass - Bowery Poetry Club',
 884         },
 885         'playlist_count': 6,
 886     }]
 887
 888     def _real_extract(self, url):
 889         mobj = self._match_valid_url(url)
 890         playlist_id = mobj.group('id')
 891
 892         query = {}
 893         token = mobj.group('token')
 894         if token:
 895             query['secret_token'] = token
 896
 897         data = self._download_json(
 898             self._API_V2_BASE + 'playlists/' + playlist_id,
 899             playlist_id, 'Downloading playlist', query=query, headers=self._HEADERS)
 900
 901         return self._extract_set(data, token)
 902
 903
 904 class SoundcloudSearchIE(SoundcloudBaseIE, SearchInfoExtractor):
 905     IE_NAME = 'soundcloud:search'
 906     IE_DESC = 'Soundcloud search'
 907     _SEARCH_KEY = 'scsearch'
 908     _TESTS = [{
 909         'url': 'scsearch15:post-avant jazzcore',
 910         'info_dict': {
 911             'id': 'post-avant jazzcore',
 912             'title': 'post-avant jazzcore',
 913         },
 914         'playlist_count': 15,
 915     }]
 916
 917     _MAX_RESULTS_PER_PAGE = 200
 918     _DEFAULT_RESULTS_PER_PAGE = 50
 919
 920     def _get_collection(self, endpoint, collection_id, **query):
 921         limit = min(
 922             query.get('limit', self._DEFAULT_RESULTS_PER_PAGE),
 923             self._MAX_RESULTS_PER_PAGE)
 924         query.update({
 925             'limit': limit,
 926             'linked_partitioning': 1,
 927             'offset': 0,
 928         })
 929         next_url = update_url_query(self._API_V2_BASE + endpoint, query)
 930
 931         for i in itertools.count(1):
 932             response = self._download_json(
 933                 next_url, collection_id, f'Downloading page {i}',
 934                 'Unable to download API page', headers=self._HEADERS)
 935
 936             for item in response.get('collection') or []:
 937                 if item:
 938                     yield self.url_result(
 939                         item['uri'], SoundcloudIE.ie_key(), **self._extract_info_dict(item, extract_flat=True))
 940
 941             next_url = response.get('next_href')
 942             if not next_url:
 943                 break
 944
 945     def _get_n_results(self, query, n):
 946         return self.playlist_result(itertools.islice(
 947             self._get_collection('search/tracks', query, limit=n, q=query),
 948             0, None if n == float('inf') else n), query, query)