yt_dlp/extractor/soundcloud.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import itertools
   5 import re
   6 import json
   7 # import random
   8
   9 from .common import (
  10     InfoExtractor,
  11     SearchInfoExtractor
  12 )
  13 from ..compat import (
  14     compat_HTTPError,
  15     compat_kwargs,
  16     compat_str,
  17 )
  18 from ..utils import (
  19     error_to_compat_str,
  20     ExtractorError,
  21     float_or_none,
  22     HEADRequest,
  23     int_or_none,
  24     KNOWN_EXTENSIONS,
  25     mimetype2ext,
  26     remove_end,
  27     parse_qs,
  28     str_or_none,
  29     try_get,
  30     unified_timestamp,
  31     update_url_query,
  32     url_or_none,
  33     urlhandle_detect_ext,
  34     sanitized_Request,
  35 )
  36
  37
  38 class SoundcloudEmbedIE(InfoExtractor):
  39     _VALID_URL = r'https?://(?:w|player|p)\.soundcloud\.com/player/?.*?\burl=(?P<id>.+)'
  40     _TEST = {
  41         # from https://www.soundi.fi/uutiset/ennakkokuuntelussa-timo-kaukolammen-station-to-station-to-station-julkaisua-juhlitaan-tanaan-g-livelabissa/
  42         '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',
  43         'only_matching': True,
  44     }
  45
  46     @staticmethod
  47     def _extract_urls(webpage):
  48         return [m.group('url') for m in re.finditer(
  49             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?://)?(?:w\.)?soundcloud\.com/player.+?)\1',
  50             webpage)]
  51
  52     def _real_extract(self, url):
  53         query = parse_qs(url)
  54         api_url = query['url'][0]
  55         secret_token = query.get('secret_token')
  56         if secret_token:
  57             api_url = update_url_query(api_url, {'secret_token': secret_token[0]})
  58         return self.url_result(api_url)
  59
  60
  61 class SoundcloudBaseIE(InfoExtractor):
  62     _NETRC_MACHINE = 'soundcloud'
  63
  64     _API_V2_BASE = 'https://api-v2.soundcloud.com/'
  65     _BASE_URL = 'https://soundcloud.com/'
  66     _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'
  67     _API_AUTH_QUERY_TEMPLATE = '?client_id=%s'
  68     _API_AUTH_URL_PW = 'https://api-auth.soundcloud.com/web-auth/sign-in/password%s'
  69     _API_VERIFY_AUTH_TOKEN = 'https://api-auth.soundcloud.com/connect/session%s'
  70     _access_token = None
  71     _HEADERS = {}
  72
  73     def _store_client_id(self, client_id):
  74         self._downloader.cache.store('soundcloud', 'client_id', client_id)
  75
  76     def _update_client_id(self):
  77         webpage = self._download_webpage('https://soundcloud.com/', None)
  78         for src in reversed(re.findall(r'<script[^>]+src="([^"]+)"', webpage)):
  79             script = self._download_webpage(src, None, fatal=False)
  80             if script:
  81                 client_id = self._search_regex(
  82                     r'client_id\s*:\s*"([0-9a-zA-Z]{32})"',
  83                     script, 'client id', default=None)
  84                 if client_id:
  85                     self._CLIENT_ID = client_id
  86                     self._store_client_id(client_id)
  87                     return
  88         raise ExtractorError('Unable to extract client id')
  89
  90     def _download_json(self, *args, **kwargs):
  91         non_fatal = kwargs.get('fatal') is False
  92         if non_fatal:
  93             del kwargs['fatal']
  94         query = kwargs.get('query', {}).copy()
  95         for _ in range(2):
  96             query['client_id'] = self._CLIENT_ID
  97             kwargs['query'] = query
  98             try:
  99                 return super()._download_json(*args, **compat_kwargs(kwargs))
 100             except ExtractorError as e:
 101                 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (401, 403):
 102                     self._store_client_id(None)
 103                     self._update_client_id()
 104                     continue
 105                 elif non_fatal:
 106                     self.report_warning(error_to_compat_str(e))
 107                     return False
 108                 raise
 109
 110     def _real_initialize(self):
 111         self._CLIENT_ID = self._downloader.cache.load('soundcloud', 'client_id') or 'a3e059563d7fd3372b49b37f00a00bcf'
 112         self._login()
 113
 114     def _login(self):
 115         username, password = self._get_login_info()
 116         if username is None:
 117             return
 118
 119         if username == 'oauth' and password is not None:
 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 = sanitized_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         elif username is not None:
 131             self.report_warning(
 132                 'Login using username and password is not currently supported. '
 133                 'Use "--username oauth --password <oauth_token>" to login using an oauth token')
 134
 135         r'''
 136         def genDevId():
 137             def genNumBlock():
 138                 return ''.join([str(random.randrange(10)) for i in range(6)])
 139             return '-'.join([genNumBlock() for i in range(4)])
 140
 141         payload = {
 142             'client_id': self._CLIENT_ID,
 143             'recaptcha_pubkey': 'null',
 144             'recaptcha_response': 'null',
 145             'credentials': {
 146                 'identifier': username,
 147                 'password': password
 148             },
 149             'signature': self.sign(username, password, self._CLIENT_ID),
 150             'device_id': genDevId(),
 151             'user_agent': self._USER_AGENT
 152         }
 153
 154         query = self._API_AUTH_QUERY_TEMPLATE % self._CLIENT_ID
 155         login = sanitized_Request(self._API_AUTH_URL_PW % query, json.dumps(payload).encode('utf-8'))
 156         response = self._download_json(login, None)
 157         self._access_token = response.get('session').get('access_token')
 158         if not self._access_token:
 159             self.report_warning('Unable to get access token, login may has failed')
 160         else:
 161             self._HEADERS = {'Authorization': 'OAuth ' + self._access_token}
 162         '''
 163
 164     # signature generation
 165     def sign(self, user, pw, clid):
 166         a = 33
 167         i = 1
 168         s = 440123
 169         w = 117
 170         u = 1800000
 171         l = 1042
 172         b = 37
 173         k = 37
 174         c = 5
 175         n = '0763ed7314c69015fd4a0dc16bbf4b90'  # _KEY
 176         y = '8'  # _REV
 177         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
 178         e = user  # _USERNAME
 179         t = clid  # _CLIENT_ID
 180
 181         d = '-'.join([str(mInt) for mInt in [a, i, s, w, u, l, b, k]])
 182         p = n + y + d + r + e + t + d + n
 183         h = p
 184
 185         m = 8011470
 186         f = 0
 187
 188         for f in range(f, len(h)):
 189             m = (m >> 1) + ((1 & m) << 23)
 190             m += ord(h[f])
 191             m &= 16777215
 192
 193         # c is not even needed
 194         out = str(y) + ':' + str(d) + ':' + format(m, 'x') + ':' + str(c)
 195
 196         return out
 197
 198     @classmethod
 199     def _resolv_url(cls, url):
 200         return cls._API_V2_BASE + 'resolve?url=' + url
 201
 202
 203 class SoundcloudIE(SoundcloudBaseIE):
 204     """Information extractor for soundcloud.com
 205        To access the media, the uid of the song and a stream token
 206        must be extracted from the page source and the script must make
 207        a request to media.soundcloud.com/crossdomain.xml. Then
 208        the media can be grabbed by requesting from an url composed
 209        of the stream token and uid
 210      """
 211
 212     _VALID_URL = r'''(?x)^(?:https?://)?
 213                     (?:(?:(?:www\.|m\.)?soundcloud\.com/
 214                             (?!stations/track)
 215                             (?P<uploader>[\w\d-]+)/
 216                             (?!(?:tracks|albums|sets(?:/.+?)?|reposts|likes|spotlight)/?(?:$|[?#]))
 217                             (?P<title>[\w\d-]+)
 218                             (?:/(?P<token>(?!(?:albums|sets|recommended))[^?]+?))?
 219                             (?:[?].*)?$)
 220                        |(?:api(?:-v2)?\.soundcloud\.com/tracks/(?P<track_id>\d+)
 221                           (?:/?\?secret_token=(?P<secret_token>[^&]+))?)
 222                     )
 223                     '''
 224     IE_NAME = 'soundcloud'
 225     _TESTS = [
 226         {
 227             'url': 'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
 228             'md5': 'ebef0a451b909710ed1d7787dddbf0d7',
 229             'info_dict': {
 230                 'id': '62986583',
 231                 'ext': 'mp3',
 232                 'title': 'Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1',
 233                 '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',
 234                 'uploader': 'E.T. ExTerrestrial Music',
 235                 'uploader_id': '1571244',
 236                 'timestamp': 1349920598,
 237                 'upload_date': '20121011',
 238                 'duration': 143.216,
 239                 'license': 'all-rights-reserved',
 240                 'view_count': int,
 241                 'like_count': int,
 242                 'comment_count': int,
 243                 'repost_count': int,
 244             }
 245         },
 246         # geo-restricted
 247         {
 248             'url': 'https://soundcloud.com/the-concept-band/goldrushed-mastered?in=the-concept-band/sets/the-royal-concept-ep',
 249             'info_dict': {
 250                 'id': '47127627',
 251                 'ext': 'mp3',
 252                 'title': 'Goldrushed',
 253                 'description': 'From Stockholm Sweden\r\nPovel / Magnus / Filip / David\r\nwww.theroyalconcept.com',
 254                 'uploader': 'The Royal Concept',
 255                 'uploader_id': '9615865',
 256                 'timestamp': 1337635207,
 257                 'upload_date': '20120521',
 258                 'duration': 227.155,
 259                 'license': 'all-rights-reserved',
 260                 'view_count': int,
 261                 'like_count': int,
 262                 'comment_count': int,
 263                 'repost_count': int,
 264             },
 265         },
 266         # private link
 267         {
 268             'url': 'https://soundcloud.com/jaimemf/youtube-dl-test-video-a-y-baw/s-8Pjrp',
 269             'md5': 'aa0dd32bfea9b0c5ef4f02aacd080604',
 270             'info_dict': {
 271                 'id': '123998367',
 272                 'ext': 'mp3',
 273                 'title': 'Youtube - Dl Test Video \'\' Ä↭',
 274                 'description': 'test chars:  \"\'/\\ä↭',
 275                 'uploader': 'jaimeMF',
 276                 'uploader_id': '69767071',
 277                 'timestamp': 1386604920,
 278                 'upload_date': '20131209',
 279                 'duration': 9.927,
 280                 'license': 'all-rights-reserved',
 281                 'view_count': int,
 282                 'like_count': int,
 283                 'comment_count': int,
 284                 'repost_count': int,
 285             },
 286         },
 287         # private link (alt format)
 288         {
 289             'url': 'https://api.soundcloud.com/tracks/123998367?secret_token=s-8Pjrp',
 290             'md5': 'aa0dd32bfea9b0c5ef4f02aacd080604',
 291             'info_dict': {
 292                 'id': '123998367',
 293                 'ext': 'mp3',
 294                 'title': 'Youtube - Dl Test Video \'\' Ä↭',
 295                 'description': 'test chars:  \"\'/\\ä↭',
 296                 'uploader': 'jaimeMF',
 297                 'uploader_id': '69767071',
 298                 'timestamp': 1386604920,
 299                 'upload_date': '20131209',
 300                 'duration': 9.927,
 301                 'license': 'all-rights-reserved',
 302                 'view_count': int,
 303                 'like_count': int,
 304                 'comment_count': int,
 305                 'repost_count': int,
 306             },
 307         },
 308         # downloadable song
 309         {
 310             'url': 'https://soundcloud.com/the80m/the-following',
 311             'md5': '9ffcddb08c87d74fb5808a3c183a1d04',
 312             'info_dict': {
 313                 'id': '343609555',
 314                 'ext': 'wav',
 315             },
 316         },
 317         # private link, downloadable format
 318         {
 319             'url': 'https://soundcloud.com/oriuplift/uponly-238-no-talking-wav/s-AyZUd',
 320             'md5': '64a60b16e617d41d0bef032b7f55441e',
 321             'info_dict': {
 322                 'id': '340344461',
 323                 'ext': 'wav',
 324                 'title': 'Uplifting Only 238 [No Talking] (incl. Alex Feed Guestmix) (Aug 31, 2017) [wav]',
 325                 'description': 'md5:fa20ee0fca76a3d6df8c7e57f3715366',
 326                 'uploader': 'Ori Uplift Music',
 327                 'uploader_id': '12563093',
 328                 'timestamp': 1504206263,
 329                 'upload_date': '20170831',
 330                 'duration': 7449.096,
 331                 'license': 'all-rights-reserved',
 332                 'view_count': int,
 333                 'like_count': int,
 334                 'comment_count': int,
 335                 'repost_count': int,
 336             },
 337         },
 338         # no album art, use avatar pic for thumbnail
 339         {
 340             'url': 'https://soundcloud.com/garyvee/sideways-prod-mad-real',
 341             'md5': '59c7872bc44e5d99b7211891664760c2',
 342             'info_dict': {
 343                 'id': '309699954',
 344                 'ext': 'mp3',
 345                 'title': 'Sideways (Prod. Mad Real)',
 346                 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
 347                 'uploader': 'garyvee',
 348                 'uploader_id': '2366352',
 349                 'timestamp': 1488152409,
 350                 'upload_date': '20170226',
 351                 'duration': 207.012,
 352                 'thumbnail': r're:https?://.*\.jpg',
 353                 'license': 'all-rights-reserved',
 354                 'view_count': int,
 355                 'like_count': int,
 356                 'comment_count': int,
 357                 'repost_count': int,
 358             },
 359             'params': {
 360                 'skip_download': True,
 361             },
 362         },
 363         {
 364             'url': 'https://soundcloud.com/giovannisarani/mezzo-valzer',
 365             'md5': 'e22aecd2bc88e0e4e432d7dcc0a1abf7',
 366             'info_dict': {
 367                 'id': '583011102',
 368                 'ext': 'mp3',
 369                 'title': 'Mezzo Valzer',
 370                 'description': 'md5:4138d582f81866a530317bae316e8b61',
 371                 'uploader': 'Micronie',
 372                 'uploader_id': '3352531',
 373                 'timestamp': 1551394171,
 374                 'upload_date': '20190228',
 375                 'duration': 180.157,
 376                 'thumbnail': r're:https?://.*\.jpg',
 377                 'license': 'all-rights-reserved',
 378                 'view_count': int,
 379                 'like_count': int,
 380                 'comment_count': int,
 381                 'repost_count': int,
 382             },
 383         },
 384         {
 385             # AAC HQ format available (account with active subscription needed)
 386             'url': 'https://soundcloud.com/wandw/the-chainsmokers-ft-daya-dont-let-me-down-ww-remix-1',
 387             'only_matching': True,
 388         },
 389         {
 390             # Go+ (account with active subscription needed)
 391             'url': 'https://soundcloud.com/taylorswiftofficial/look-what-you-made-me-do',
 392             'only_matching': True,
 393         },
 394     ]
 395
 396     _IMAGE_REPL_RE = r'-([0-9a-z]+)\.jpg'
 397
 398     _ARTWORK_MAP = {
 399         'mini': 16,
 400         'tiny': 20,
 401         'small': 32,
 402         'badge': 47,
 403         't67x67': 67,
 404         'large': 100,
 405         't300x300': 300,
 406         'crop': 400,
 407         't500x500': 500,
 408         'original': 0,
 409     }
 410
 411     def _extract_info_dict(self, info, full_title=None, secret_token=None):
 412         track_id = compat_str(info['id'])
 413         title = info['title']
 414
 415         format_urls = set()
 416         formats = []
 417         query = {'client_id': self._CLIENT_ID}
 418         if secret_token:
 419             query['secret_token'] = secret_token
 420
 421         if info.get('downloadable') and info.get('has_downloads_left'):
 422             download_url = update_url_query(
 423                 self._API_V2_BASE + 'tracks/' + track_id + '/download', query)
 424             redirect_url = (self._download_json(download_url, track_id, fatal=False) or {}).get('redirectUri')
 425             if redirect_url:
 426                 urlh = self._request_webpage(
 427                     HEADRequest(redirect_url), track_id, fatal=False)
 428                 if urlh:
 429                     format_url = urlh.geturl()
 430                     format_urls.add(format_url)
 431                     formats.append({
 432                         'format_id': 'download',
 433                         'ext': urlhandle_detect_ext(urlh) or 'mp3',
 434                         'filesize': int_or_none(urlh.headers.get('Content-Length')),
 435                         'url': format_url,
 436                         'quality': 10,
 437                     })
 438
 439         def invalid_url(url):
 440             return not url or url in format_urls
 441
 442         def add_format(f, protocol, is_preview=False):
 443             mobj = re.search(r'\.(?P<abr>\d+)\.(?P<ext>[0-9a-z]{3,4})(?=[/?])', stream_url)
 444             if mobj:
 445                 for k, v in mobj.groupdict().items():
 446                     if not f.get(k):
 447                         f[k] = v
 448             format_id_list = []
 449             if protocol:
 450                 format_id_list.append(protocol)
 451             ext = f.get('ext')
 452             if ext == 'aac':
 453                 f['abr'] = '256'
 454             for k in ('ext', 'abr'):
 455                 v = f.get(k)
 456                 if v:
 457                     format_id_list.append(v)
 458             preview = is_preview or re.search(r'/(?:preview|playlist)/0/30/', f['url'])
 459             if preview:
 460                 format_id_list.append('preview')
 461             abr = f.get('abr')
 462             if abr:
 463                 f['abr'] = int(abr)
 464             if protocol == 'hls':
 465                 protocol = 'm3u8' if ext == 'aac' else 'm3u8_native'
 466             else:
 467                 protocol = 'http'
 468             f.update({
 469                 'format_id': '_'.join(format_id_list),
 470                 'protocol': protocol,
 471                 'preference': -10 if preview else None,
 472             })
 473             formats.append(f)
 474
 475         # New API
 476         transcodings = try_get(
 477             info, lambda x: x['media']['transcodings'], list) or []
 478         for t in transcodings:
 479             if not isinstance(t, dict):
 480                 continue
 481             format_url = url_or_none(t.get('url'))
 482             if not format_url:
 483                 continue
 484             stream = self._download_json(
 485                 format_url, track_id, query=query, fatal=False, headers=self._HEADERS)
 486             if not isinstance(stream, dict):
 487                 continue
 488             stream_url = url_or_none(stream.get('url'))
 489             if invalid_url(stream_url):
 490                 continue
 491             format_urls.add(stream_url)
 492             stream_format = t.get('format') or {}
 493             protocol = stream_format.get('protocol')
 494             if protocol != 'hls' and '/hls' in format_url:
 495                 protocol = 'hls'
 496             ext = None
 497             preset = str_or_none(t.get('preset'))
 498             if preset:
 499                 ext = preset.split('_')[0]
 500             if ext not in KNOWN_EXTENSIONS:
 501                 ext = mimetype2ext(stream_format.get('mime_type'))
 502             add_format({
 503                 'url': stream_url,
 504                 'ext': ext,
 505             }, 'http' if protocol == 'progressive' else protocol,
 506                 t.get('snipped') or '/preview/' in format_url)
 507
 508         for f in formats:
 509             f['vcodec'] = 'none'
 510
 511         if not formats and info.get('policy') == 'BLOCK':
 512             self.raise_geo_restricted(metadata_available=True)
 513         self._sort_formats(formats)
 514
 515         user = info.get('user') or {}
 516
 517         thumbnails = []
 518         artwork_url = info.get('artwork_url')
 519         thumbnail = artwork_url or user.get('avatar_url')
 520         if isinstance(thumbnail, compat_str):
 521             if re.search(self._IMAGE_REPL_RE, thumbnail):
 522                 for image_id, size in self._ARTWORK_MAP.items():
 523                     i = {
 524                         'id': image_id,
 525                         'url': re.sub(self._IMAGE_REPL_RE, '-%s.jpg' % image_id, thumbnail),
 526                     }
 527                     if image_id == 'tiny' and not artwork_url:
 528                         size = 18
 529                     elif image_id == 'original':
 530                         i['preference'] = 10
 531                     if size:
 532                         i.update({
 533                             'width': size,
 534                             'height': size,
 535                         })
 536                     thumbnails.append(i)
 537             else:
 538                 thumbnails = [{'url': thumbnail}]
 539
 540         def extract_count(key):
 541             return int_or_none(info.get('%s_count' % key))
 542
 543         return {
 544             'id': track_id,
 545             'uploader': user.get('username'),
 546             'uploader_id': str_or_none(user.get('id')) or user.get('permalink'),
 547             'uploader_url': user.get('permalink_url'),
 548             'timestamp': unified_timestamp(info.get('created_at')),
 549             'title': title,
 550             'description': info.get('description'),
 551             'thumbnails': thumbnails,
 552             'duration': float_or_none(info.get('duration'), 1000),
 553             'webpage_url': info.get('permalink_url'),
 554             'license': info.get('license'),
 555             'view_count': extract_count('playback'),
 556             'like_count': extract_count('favoritings') or extract_count('likes'),
 557             'comment_count': extract_count('comment'),
 558             'repost_count': extract_count('reposts'),
 559             'genre': info.get('genre'),
 560             'formats': formats
 561         }
 562
 563     def _real_extract(self, url):
 564         mobj = self._match_valid_url(url)
 565
 566         track_id = mobj.group('track_id')
 567
 568         query = {}
 569         if track_id:
 570             info_json_url = self._API_V2_BASE + 'tracks/' + track_id
 571             full_title = track_id
 572             token = mobj.group('secret_token')
 573             if token:
 574                 query['secret_token'] = token
 575         else:
 576             full_title = resolve_title = '%s/%s' % mobj.group('uploader', 'title')
 577             token = mobj.group('token')
 578             if token:
 579                 resolve_title += '/%s' % token
 580             info_json_url = self._resolv_url(self._BASE_URL + resolve_title)
 581
 582         info = self._download_json(
 583             info_json_url, full_title, 'Downloading info JSON', query=query, headers=self._HEADERS)
 584
 585         return self._extract_info_dict(info, full_title, token)
 586
 587
 588 class SoundcloudPlaylistBaseIE(SoundcloudBaseIE):
 589     def _extract_set(self, playlist, token=None):
 590         playlist_id = compat_str(playlist['id'])
 591         tracks = playlist.get('tracks') or []
 592         if not all([t.get('permalink_url') for t in tracks]) and token:
 593             tracks = self._download_json(
 594                 self._API_V2_BASE + 'tracks', playlist_id,
 595                 'Downloading tracks', query={
 596                     'ids': ','.join([compat_str(t['id']) for t in tracks]),
 597                     'playlistId': playlist_id,
 598                     'playlistSecretToken': token,
 599                 }, headers=self._HEADERS)
 600         entries = []
 601         for track in tracks:
 602             track_id = str_or_none(track.get('id'))
 603             url = track.get('permalink_url')
 604             if not url:
 605                 if not track_id:
 606                     continue
 607                 url = self._API_V2_BASE + 'tracks/' + track_id
 608                 if token:
 609                     url += '?secret_token=' + token
 610             entries.append(self.url_result(
 611                 url, SoundcloudIE.ie_key(), track_id))
 612         return self.playlist_result(
 613             entries, playlist_id,
 614             playlist.get('title'),
 615             playlist.get('description'))
 616
 617
 618 class SoundcloudSetIE(SoundcloudPlaylistBaseIE):
 619     _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/(?P<uploader>[\w\d-]+)/sets/(?P<slug_title>[:\w\d-]+)(?:/(?P<token>[^?/]+))?'
 620     IE_NAME = 'soundcloud:set'
 621     _TESTS = [{
 622         'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep',
 623         'info_dict': {
 624             'id': '2284613',
 625             'title': 'The Royal Concept EP',
 626             'description': 'md5:71d07087c7a449e8941a70a29e34671e',
 627         },
 628         'playlist_mincount': 5,
 629     }, {
 630         'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep/token',
 631         'only_matching': True,
 632     }, {
 633         'url': 'https://soundcloud.com/discover/sets/weekly::flacmatic',
 634         'only_matching': True,
 635     }, {
 636         'url': 'https://soundcloud.com/discover/sets/charts-top:all-music:de',
 637         'only_matching': True,
 638     }, {
 639         'url': 'https://soundcloud.com/discover/sets/charts-top:hiphoprap:kr',
 640         'only_matching': True,
 641     }]
 642
 643     def _real_extract(self, url):
 644         mobj = self._match_valid_url(url)
 645
 646         full_title = '%s/sets/%s' % mobj.group('uploader', 'slug_title')
 647         token = mobj.group('token')
 648         if token:
 649             full_title += '/' + token
 650
 651         info = self._download_json(self._resolv_url(
 652             self._BASE_URL + full_title), full_title, headers=self._HEADERS)
 653
 654         if 'errors' in info:
 655             msgs = (compat_str(err['error_message']) for err in info['errors'])
 656             raise ExtractorError('unable to download video webpage: %s' % ','.join(msgs))
 657
 658         return self._extract_set(info, token)
 659
 660
 661 class SoundcloudPagedPlaylistBaseIE(SoundcloudBaseIE):
 662     def _extract_playlist(self, base_url, playlist_id, playlist_title):
 663         return {
 664             '_type': 'playlist',
 665             'id': playlist_id,
 666             'title': playlist_title,
 667             'entries': self._entries(base_url, playlist_id),
 668         }
 669
 670     def _entries(self, url, playlist_id):
 671         # Per the SoundCloud documentation, the maximum limit for a linked partitioning query is 200.
 672         # https://developers.soundcloud.com/blog/offset-pagination-deprecated
 673         query = {
 674             'limit': 200,
 675             'linked_partitioning': '1',
 676             'offset': 0,
 677         }
 678
 679         retries = self.get_param('extractor_retries', 3)
 680
 681         for i in itertools.count():
 682             attempt, last_error = -1, None
 683             while attempt < retries:
 684                 attempt += 1
 685                 if last_error:
 686                     self.report_warning('%s. Retrying ...' % remove_end(last_error, '.'), playlist_id)
 687                 try:
 688                     response = self._download_json(
 689                         url, playlist_id, query=query, headers=self._HEADERS,
 690                         note='Downloading track page %s%s' % (i + 1, f' (retry #{attempt})' if attempt else ''))
 691                     break
 692                 except ExtractorError as e:
 693                     # Downloading page may result in intermittent 502 HTTP error
 694                     # See https://github.com/yt-dlp/yt-dlp/issues/872
 695                     if attempt >= retries or not isinstance(e.cause, compat_HTTPError) or e.cause.code != 502:
 696                         raise
 697                     last_error = str(e.cause or e.msg)
 698
 699             def resolve_entry(*candidates):
 700                 for cand in candidates:
 701                     if not isinstance(cand, dict):
 702                         continue
 703                     permalink_url = url_or_none(cand.get('permalink_url'))
 704                     if permalink_url:
 705                         return self.url_result(
 706                             permalink_url,
 707                             SoundcloudIE.ie_key() if SoundcloudIE.suitable(permalink_url) else None,
 708                             str_or_none(cand.get('id')), cand.get('title'))
 709
 710             for e in response['collection'] or []:
 711                 yield resolve_entry(e, e.get('track'), e.get('playlist'))
 712
 713             url = response.get('next_href')
 714             if not url:
 715                 break
 716             query.pop('offset', None)
 717
 718
 719 class SoundcloudUserIE(SoundcloudPagedPlaylistBaseIE):
 720     _VALID_URL = r'''(?x)
 721                         https?://
 722                             (?:(?:www|m)\.)?soundcloud\.com/
 723                             (?P<user>[^/]+)
 724                             (?:/
 725                                 (?P<rsrc>tracks|albums|sets|reposts|likes|spotlight)
 726                             )?
 727                             /?(?:[?#].*)?$
 728                     '''
 729     IE_NAME = 'soundcloud:user'
 730     _TESTS = [{
 731         'url': 'https://soundcloud.com/soft-cell-official',
 732         'info_dict': {
 733             'id': '207965082',
 734             'title': 'Soft Cell (All)',
 735         },
 736         'playlist_mincount': 28,
 737     }, {
 738         'url': 'https://soundcloud.com/soft-cell-official/tracks',
 739         'info_dict': {
 740             'id': '207965082',
 741             'title': 'Soft Cell (Tracks)',
 742         },
 743         'playlist_mincount': 27,
 744     }, {
 745         'url': 'https://soundcloud.com/soft-cell-official/albums',
 746         'info_dict': {
 747             'id': '207965082',
 748             'title': 'Soft Cell (Albums)',
 749         },
 750         'playlist_mincount': 1,
 751     }, {
 752         'url': 'https://soundcloud.com/jcv246/sets',
 753         'info_dict': {
 754             'id': '12982173',
 755             'title': 'Jordi / cv (Sets)',
 756         },
 757         'playlist_mincount': 2,
 758     }, {
 759         'url': 'https://soundcloud.com/jcv246/reposts',
 760         'info_dict': {
 761             'id': '12982173',
 762             'title': 'Jordi / cv (Reposts)',
 763         },
 764         'playlist_mincount': 6,
 765     }, {
 766         'url': 'https://soundcloud.com/clalberg/likes',
 767         'info_dict': {
 768             'id': '11817582',
 769             'title': 'clalberg (Likes)',
 770         },
 771         'playlist_mincount': 5,
 772     }, {
 773         'url': 'https://soundcloud.com/grynpyret/spotlight',
 774         'info_dict': {
 775             'id': '7098329',
 776             'title': 'Grynpyret (Spotlight)',
 777         },
 778         'playlist_mincount': 1,
 779     }]
 780
 781     _BASE_URL_MAP = {
 782         'all': 'stream/users/%s',
 783         'tracks': 'users/%s/tracks',
 784         'albums': 'users/%s/albums',
 785         'sets': 'users/%s/playlists',
 786         'reposts': 'stream/users/%s/reposts',
 787         'likes': 'users/%s/likes',
 788         'spotlight': 'users/%s/spotlight',
 789     }
 790
 791     def _real_extract(self, url):
 792         mobj = self._match_valid_url(url)
 793         uploader = mobj.group('user')
 794
 795         user = self._download_json(
 796             self._resolv_url(self._BASE_URL + uploader),
 797             uploader, 'Downloading user info', headers=self._HEADERS)
 798
 799         resource = mobj.group('rsrc') or 'all'
 800
 801         return self._extract_playlist(
 802             self._API_V2_BASE + self._BASE_URL_MAP[resource] % user['id'],
 803             str_or_none(user.get('id')),
 804             '%s (%s)' % (user['username'], resource.capitalize()))
 805
 806
 807 class SoundcloudTrackStationIE(SoundcloudPagedPlaylistBaseIE):
 808     _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/stations/track/[^/]+/(?P<id>[^/?#&]+)'
 809     IE_NAME = 'soundcloud:trackstation'
 810     _TESTS = [{
 811         'url': 'https://soundcloud.com/stations/track/officialsundial/your-text',
 812         'info_dict': {
 813             'id': '286017854',
 814             'title': 'Track station: your text',
 815         },
 816         'playlist_mincount': 47,
 817     }]
 818
 819     def _real_extract(self, url):
 820         track_name = self._match_id(url)
 821
 822         track = self._download_json(self._resolv_url(url), track_name, headers=self._HEADERS)
 823         track_id = self._search_regex(
 824             r'soundcloud:track-stations:(\d+)', track['id'], 'track id')
 825
 826         return self._extract_playlist(
 827             self._API_V2_BASE + 'stations/%s/tracks' % track['id'],
 828             track_id, 'Track station: %s' % track['title'])
 829
 830
 831 class SoundcloudRelatedIE(SoundcloudPagedPlaylistBaseIE):
 832     _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/(?P<slug>[\w\d-]+/[\w\d-]+)/(?P<relation>albums|sets|recommended)'
 833     IE_NAME = 'soundcloud:related'
 834     _TESTS = [{
 835         'url': 'https://soundcloud.com/wajang/sexapil-pingers-5/recommended',
 836         'info_dict': {
 837             'id': '1084577272',
 838             'title': 'Sexapil - Pingers 5 (Recommended)',
 839         },
 840         'playlist_mincount': 50,
 841     }, {
 842         'url': 'https://soundcloud.com/wajang/sexapil-pingers-5/albums',
 843         'info_dict': {
 844             'id': '1084577272',
 845             'title': 'Sexapil - Pingers 5 (Albums)',
 846         },
 847         'playlist_mincount': 1,
 848     }, {
 849         'url': 'https://soundcloud.com/wajang/sexapil-pingers-5/sets',
 850         'info_dict': {
 851             'id': '1084577272',
 852             'title': 'Sexapil - Pingers 5 (Sets)',
 853         },
 854         'playlist_mincount': 4,
 855     }]
 856
 857     _BASE_URL_MAP = {
 858         'albums': 'tracks/%s/albums',
 859         'sets': 'tracks/%s/playlists_without_albums',
 860         'recommended': 'tracks/%s/related',
 861     }
 862
 863     def _real_extract(self, url):
 864         slug, relation = self._match_valid_url(url).group('slug', 'relation')
 865
 866         track = self._download_json(
 867             self._resolv_url(self._BASE_URL + slug),
 868             slug, 'Downloading track info', headers=self._HEADERS)
 869
 870         if track.get('errors'):
 871             raise ExtractorError(f'{self.IE_NAME} said: %s' % ','.join(
 872                 str(err['error_message']) for err in track['errors']), expected=True)
 873
 874         return self._extract_playlist(
 875             self._API_V2_BASE + self._BASE_URL_MAP[relation] % track['id'], str(track['id']),
 876             '%s (%s)' % (track.get('title') or slug, relation.capitalize()))
 877
 878
 879 class SoundcloudPlaylistIE(SoundcloudPlaylistBaseIE):
 880     _VALID_URL = r'https?://api(?:-v2)?\.soundcloud\.com/playlists/(?P<id>[0-9]+)(?:/?\?secret_token=(?P<token>[^&]+?))?$'
 881     IE_NAME = 'soundcloud:playlist'
 882     _TESTS = [{
 883         'url': 'https://api.soundcloud.com/playlists/4110309',
 884         'info_dict': {
 885             'id': '4110309',
 886             'title': 'TILT Brass - Bowery Poetry Club, August \'03 [Non-Site SCR 02]',
 887             'description': 're:.*?TILT Brass - Bowery Poetry Club',
 888         },
 889         'playlist_count': 6,
 890     }]
 891
 892     def _real_extract(self, url):
 893         mobj = self._match_valid_url(url)
 894         playlist_id = mobj.group('id')
 895
 896         query = {}
 897         token = mobj.group('token')
 898         if token:
 899             query['secret_token'] = token
 900
 901         data = self._download_json(
 902             self._API_V2_BASE + 'playlists/' + playlist_id,
 903             playlist_id, 'Downloading playlist', query=query, headers=self._HEADERS)
 904
 905         return self._extract_set(data, token)
 906
 907
 908 class SoundcloudSearchIE(SoundcloudBaseIE, SearchInfoExtractor):
 909     IE_NAME = 'soundcloud:search'
 910     IE_DESC = 'Soundcloud search'
 911     _SEARCH_KEY = 'scsearch'
 912     _TESTS = [{
 913         'url': 'scsearch15:post-avant jazzcore',
 914         'info_dict': {
 915             'title': 'post-avant jazzcore',
 916         },
 917         'playlist_count': 15,
 918     }]
 919
 920     _MAX_RESULTS_PER_PAGE = 200
 921     _DEFAULT_RESULTS_PER_PAGE = 50
 922
 923     def _get_collection(self, endpoint, collection_id, **query):
 924         limit = min(
 925             query.get('limit', self._DEFAULT_RESULTS_PER_PAGE),
 926             self._MAX_RESULTS_PER_PAGE)
 927         query.update({
 928             'limit': limit,
 929             'linked_partitioning': 1,
 930             'offset': 0,
 931         })
 932         next_url = update_url_query(self._API_V2_BASE + endpoint, query)
 933
 934         for i in itertools.count(1):
 935             response = self._download_json(
 936                 next_url, collection_id, f'Downloading page {i}',
 937                 'Unable to download API page', headers=self._HEADERS)
 938
 939             for item in response.get('collection') or []:
 940                 if item:
 941                     yield self.url_result(item['uri'], SoundcloudIE.ie_key())
 942
 943             next_url = response.get('next_href')
 944             if not next_url:
 945                 break
 946
 947     def _get_n_results(self, query, n):
 948         return self.playlist_result(itertools.islice(
 949             self._get_collection('search/tracks', query, limit=n, q=query),
 950             0, None if n == float('inf') else n), query, query)