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