_ Git - youtube-dl/blob - youtube_dl/extractor/soundcloud.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import itertools
   5 import re
   6
   7 from .common import (
   8     InfoExtractor,
   9     SearchInfoExtractor
  10 )
  11 from ..compat import (
  12     compat_HTTPError,
  13     compat_kwargs,
  14     compat_str,
  15     compat_urlparse,
  16 )
  17 from ..utils import (
  18     error_to_compat_str,
  19     ExtractorError,
  20     float_or_none,
  21     HEADRequest,
  22     int_or_none,
  23     KNOWN_EXTENSIONS,
  24     mimetype2ext,
  25     str_or_none,
  26     try_get,
  27     unified_timestamp,
  28     update_url_query,
  29     url_or_none,
  30 )
  31
  32
  33 class SoundcloudEmbedIE(InfoExtractor):
  34     _VALID_URL = r'https?://(?:w|player|p)\.soundcloud\.com/player/?.*?\burl=(?P<id>.+)'
  35     _TEST = {
  36         # from https://www.soundi.fi/uutiset/ennakkokuuntelussa-timo-kaukolammen-station-to-station-to-station-julkaisua-juhlitaan-tanaan-g-livelabissa/
  37         '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',
  38         'only_matching': True,
  39     }
  40
  41     @staticmethod
  42     def _extract_urls(webpage):
  43         return [m.group('url') for m in re.finditer(
  44             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?://)?(?:w\.)?soundcloud\.com/player.+?)\1',
  45             webpage)]
  46
  47     def _real_extract(self, url):
  48         query = compat_urlparse.parse_qs(
  49             compat_urlparse.urlparse(url).query)
  50         api_url = query['url'][0]
  51         secret_token = query.get('secret_token')
  52         if secret_token:
  53             api_url = update_url_query(api_url, {'secret_token': secret_token[0]})
  54         return self.url_result(api_url)
  55
  56
  57 class SoundcloudIE(InfoExtractor):
  58     """Information extractor for soundcloud.com
  59        To access the media, the uid of the song and a stream token
  60        must be extracted from the page source and the script must make
  61        a request to media.soundcloud.com/crossdomain.xml. Then
  62        the media can be grabbed by requesting from an url composed
  63        of the stream token and uid
  64      """
  65
  66     _VALID_URL = r'''(?x)^(?:https?://)?
  67                     (?:(?:(?:www\.|m\.)?soundcloud\.com/
  68                             (?!stations/track)
  69                             (?P<uploader>[\w\d-]+)/
  70                             (?!(?:tracks|albums|sets(?:/.+?)?|reposts|likes|spotlight)/?(?:$|[?#]))
  71                             (?P<title>[\w\d-]+)/?
  72                             (?P<token>[^?]+?)?(?:[?].*)?$)
  73                        |(?:api(?:-v2)?\.soundcloud\.com/tracks/(?P<track_id>\d+)
  74                           (?:/?\?secret_token=(?P<secret_token>[^&]+))?)
  75                     )
  76                     '''
  77     IE_NAME = 'soundcloud'
  78     _TESTS = [
  79         {
  80             'url': 'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
  81             'md5': 'ebef0a451b909710ed1d7787dddbf0d7',
  82             'info_dict': {
  83                 'id': '62986583',
  84                 'ext': 'mp3',
  85                 'title': 'Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1',
  86                 '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',
  87                 'uploader': 'E.T. ExTerrestrial Music',
  88                 'uploader_id': '1571244',
  89                 'timestamp': 1349920598,
  90                 'upload_date': '20121011',
  91                 'duration': 143.216,
  92                 'license': 'all-rights-reserved',
  93                 'view_count': int,
  94                 'like_count': int,
  95                 'comment_count': int,
  96                 'repost_count': int,
  97             }
  98         },
  99         # not streamable song
 100         {
 101             'url': 'https://soundcloud.com/the-concept-band/goldrushed-mastered?in=the-concept-band/sets/the-royal-concept-ep',
 102             'info_dict': {
 103                 'id': '47127627',
 104                 'ext': 'mp3',
 105                 'title': 'Goldrushed',
 106                 'description': 'From Stockholm Sweden\r\nPovel / Magnus / Filip / David\r\nwww.theroyalconcept.com',
 107                 'uploader': 'The Royal Concept',
 108                 'uploader_id': '9615865',
 109                 'timestamp': 1337635207,
 110                 'upload_date': '20120521',
 111                 'duration': 30,
 112                 'license': 'all-rights-reserved',
 113                 'view_count': int,
 114                 'like_count': int,
 115                 'comment_count': int,
 116                 'repost_count': int,
 117             },
 118             'params': {
 119                 # rtmp
 120                 'skip_download': True,
 121             },
 122             'skip': 'Preview',
 123         },
 124         # private link
 125         {
 126             'url': 'https://soundcloud.com/jaimemf/youtube-dl-test-video-a-y-baw/s-8Pjrp',
 127             'md5': 'aa0dd32bfea9b0c5ef4f02aacd080604',
 128             'info_dict': {
 129                 'id': '123998367',
 130                 'ext': 'mp3',
 131                 'title': 'Youtube - Dl Test Video \'\' Ä↭',
 132                 'description': 'test chars:  \"\'/\\ä↭',
 133                 'uploader': 'jaimeMF',
 134                 'uploader_id': '69767071',
 135                 'timestamp': 1386604920,
 136                 'upload_date': '20131209',
 137                 'duration': 9.927,
 138                 'license': 'all-rights-reserved',
 139                 'view_count': int,
 140                 'like_count': int,
 141                 'comment_count': int,
 142                 'repost_count': int,
 143             },
 144         },
 145         # private link (alt format)
 146         {
 147             'url': 'https://api.soundcloud.com/tracks/123998367?secret_token=s-8Pjrp',
 148             'md5': 'aa0dd32bfea9b0c5ef4f02aacd080604',
 149             'info_dict': {
 150                 'id': '123998367',
 151                 'ext': 'mp3',
 152                 'title': 'Youtube - Dl Test Video \'\' Ä↭',
 153                 'description': 'test chars:  \"\'/\\ä↭',
 154                 'uploader': 'jaimeMF',
 155                 'uploader_id': '69767071',
 156                 'timestamp': 1386604920,
 157                 'upload_date': '20131209',
 158                 'duration': 9.927,
 159                 'license': 'all-rights-reserved',
 160                 'view_count': int,
 161                 'like_count': int,
 162                 'comment_count': int,
 163                 'repost_count': int,
 164             },
 165         },
 166         # downloadable song
 167         {
 168             'url': 'https://soundcloud.com/oddsamples/bus-brakes',
 169             'md5': '7624f2351f8a3b2e7cd51522496e7631',
 170             'info_dict': {
 171                 'id': '128590877',
 172                 'ext': 'mp3',
 173                 'title': 'Bus Brakes',
 174                 'description': 'md5:0053ca6396e8d2fd7b7e1595ef12ab66',
 175                 'uploader': 'oddsamples',
 176                 'uploader_id': '73680509',
 177                 'timestamp': 1389232924,
 178                 'upload_date': '20140109',
 179                 'duration': 17.346,
 180                 'license': 'cc-by-sa',
 181                 'view_count': int,
 182                 'like_count': int,
 183                 'comment_count': int,
 184                 'repost_count': int,
 185             },
 186         },
 187         # private link, downloadable format
 188         {
 189             'url': 'https://soundcloud.com/oriuplift/uponly-238-no-talking-wav/s-AyZUd',
 190             'md5': '64a60b16e617d41d0bef032b7f55441e',
 191             'info_dict': {
 192                 'id': '340344461',
 193                 'ext': 'wav',
 194                 'title': 'Uplifting Only 238 [No Talking] (incl. Alex Feed Guestmix) (Aug 31, 2017) [wav]',
 195                 'description': 'md5:fa20ee0fca76a3d6df8c7e57f3715366',
 196                 'uploader': 'Ori Uplift Music',
 197                 'uploader_id': '12563093',
 198                 'timestamp': 1504206263,
 199                 'upload_date': '20170831',
 200                 'duration': 7449.096,
 201                 'license': 'all-rights-reserved',
 202                 'view_count': int,
 203                 'like_count': int,
 204                 'comment_count': int,
 205                 'repost_count': int,
 206             },
 207         },
 208         # no album art, use avatar pic for thumbnail
 209         {
 210             'url': 'https://soundcloud.com/garyvee/sideways-prod-mad-real',
 211             'md5': '59c7872bc44e5d99b7211891664760c2',
 212             'info_dict': {
 213                 'id': '309699954',
 214                 'ext': 'mp3',
 215                 'title': 'Sideways (Prod. Mad Real)',
 216                 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
 217                 'uploader': 'garyvee',
 218                 'uploader_id': '2366352',
 219                 'timestamp': 1488152409,
 220                 'upload_date': '20170226',
 221                 'duration': 207.012,
 222                 'thumbnail': r're:https?://.*\.jpg',
 223                 'license': 'all-rights-reserved',
 224                 'view_count': int,
 225                 'like_count': int,
 226                 'comment_count': int,
 227                 'repost_count': int,
 228             },
 229             'params': {
 230                 'skip_download': True,
 231             },
 232         },
 233         # not available via api.soundcloud.com/i1/tracks/id/streams
 234         {
 235             'url': 'https://soundcloud.com/giovannisarani/mezzo-valzer',
 236             'md5': 'e22aecd2bc88e0e4e432d7dcc0a1abf7',
 237             'info_dict': {
 238                 'id': '583011102',
 239                 'ext': 'mp3',
 240                 'title': 'Mezzo Valzer',
 241                 'description': 'md5:4138d582f81866a530317bae316e8b61',
 242                 'uploader': 'Giovanni Sarani',
 243                 'uploader_id': '3352531',
 244                 'timestamp': 1551394171,
 245                 'upload_date': '20190228',
 246                 'duration': 180.157,
 247                 'thumbnail': r're:https?://.*\.jpg',
 248                 'license': 'all-rights-reserved',
 249                 'view_count': int,
 250                 'like_count': int,
 251                 'comment_count': int,
 252                 'repost_count': int,
 253             },
 254             'expected_warnings': ['Unable to download JSON metadata'],
 255         }
 256     ]
 257
 258     _API_BASE = 'https://api.soundcloud.com/'
 259     _API_V2_BASE = 'https://api-v2.soundcloud.com/'
 260     _BASE_URL = 'https://soundcloud.com/'
 261     _IMAGE_REPL_RE = r'-([0-9a-z]+)\.jpg'
 262
 263     _ARTWORK_MAP = {
 264         'mini': 16,
 265         'tiny': 20,
 266         'small': 32,
 267         'badge': 47,
 268         't67x67': 67,
 269         'large': 100,
 270         't300x300': 300,
 271         'crop': 400,
 272         't500x500': 500,
 273         'original': 0,
 274     }
 275
 276     def _store_client_id(self, client_id):
 277         self._downloader.cache.store('soundcloud', 'client_id', client_id)
 278
 279     def _update_client_id(self):
 280         webpage = self._download_webpage('https://soundcloud.com/', None)
 281         for src in reversed(re.findall(r'<script[^>]+src="([^"]+)"', webpage)):
 282             script = self._download_webpage(src, None, fatal=False)
 283             if script:
 284                 client_id = self._search_regex(
 285                     r'client_id\s*:\s*"([0-9a-zA-Z]{32})"',
 286                     script, 'client id', default=None)
 287                 if client_id:
 288                     self._CLIENT_ID = client_id
 289                     self._store_client_id(client_id)
 290                     return
 291         raise ExtractorError('Unable to extract client id')
 292
 293     def _download_json(self, *args, **kwargs):
 294         non_fatal = kwargs.get('fatal') is False
 295         if non_fatal:
 296             del kwargs['fatal']
 297         query = kwargs.get('query', {}).copy()
 298         for _ in range(2):
 299             query['client_id'] = self._CLIENT_ID
 300             kwargs['query'] = query
 301             try:
 302                 return super(SoundcloudIE, self)._download_json(*args, **compat_kwargs(kwargs))
 303             except ExtractorError as e:
 304                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
 305                     self._store_client_id(None)
 306                     self._update_client_id()
 307                     continue
 308                 elif non_fatal:
 309                     self._downloader.report_warning(error_to_compat_str(e))
 310                     return False
 311                 raise
 312
 313     def _real_initialize(self):
 314         self._CLIENT_ID = self._downloader.cache.load('soundcloud', 'client_id') or 'YUKXoArFcqrlQn9tfNHvvyfnDISj04zk'
 315
 316     @classmethod
 317     def _resolv_url(cls, url):
 318         return SoundcloudIE._API_V2_BASE + 'resolve?url=' + url
 319
 320     def _extract_info_dict(self, info, full_title=None, secret_token=None, version=2):
 321         track_id = compat_str(info['id'])
 322         title = info['title']
 323         track_base_url = self._API_BASE + 'tracks/%s' % track_id
 324
 325         format_urls = set()
 326         formats = []
 327         query = {'client_id': self._CLIENT_ID}
 328         if secret_token:
 329             query['secret_token'] = secret_token
 330
 331         if info.get('downloadable') and info.get('has_downloads_left'):
 332             format_url = update_url_query(
 333                 info.get('download_url') or track_base_url + '/download', query)
 334             format_urls.add(format_url)
 335             if version == 2:
 336                 v1_info = self._download_json(
 337                     track_base_url, track_id, query=query, fatal=False) or {}
 338             else:
 339                 v1_info = info
 340             formats.append({
 341                 'format_id': 'download',
 342                 'ext': v1_info.get('original_format') or 'mp3',
 343                 'filesize': int_or_none(v1_info.get('original_content_size')),
 344                 'url': format_url,
 345                 'preference': 10,
 346             })
 347
 348         def invalid_url(url):
 349             return not url or url in format_urls or re.search(r'/(?:preview|playlist)/0/30/', url)
 350
 351         def add_format(f, protocol):
 352             mobj = re.search(r'\.(?P<abr>\d+)\.(?P<ext>[0-9a-z]{3,4})(?=[/?])', stream_url)
 353             if mobj:
 354                 for k, v in mobj.groupdict().items():
 355                     if not f.get(k):
 356                         f[k] = v
 357             format_id_list = []
 358             if protocol:
 359                 format_id_list.append(protocol)
 360             for k in ('ext', 'abr'):
 361                 v = f.get(k)
 362                 if v:
 363                     format_id_list.append(v)
 364             abr = f.get('abr')
 365             if abr:
 366                 f['abr'] = int(abr)
 367             f.update({
 368                 'format_id': '_'.join(format_id_list),
 369                 'protocol': 'm3u8_native' if protocol == 'hls' else 'http',
 370             })
 371             formats.append(f)
 372
 373         # New API
 374         transcodings = try_get(
 375             info, lambda x: x['media']['transcodings'], list) or []
 376         for t in transcodings:
 377             if not isinstance(t, dict):
 378                 continue
 379             format_url = url_or_none(t.get('url'))
 380             if not format_url or t.get('snipped') or '/preview/' in format_url:
 381                 continue
 382             stream = self._download_json(
 383                 format_url, track_id, query=query, fatal=False)
 384             if not isinstance(stream, dict):
 385                 continue
 386             stream_url = url_or_none(stream.get('url'))
 387             if invalid_url(stream_url):
 388                 continue
 389             format_urls.add(stream_url)
 390             stream_format = t.get('format') or {}
 391             protocol = stream_format.get('protocol')
 392             if protocol != 'hls' and '/hls' in format_url:
 393                 protocol = 'hls'
 394             ext = None
 395             preset = str_or_none(t.get('preset'))
 396             if preset:
 397                 ext = preset.split('_')[0]
 398             if ext not in KNOWN_EXTENSIONS:
 399                 ext = mimetype2ext(stream_format.get('mime_type'))
 400             add_format({
 401                 'url': stream_url,
 402                 'ext': ext,
 403             }, 'http' if protocol == 'progressive' else protocol)
 404
 405         if not formats:
 406             # Old API, does not work for some tracks (e.g.
 407             # https://soundcloud.com/giovannisarani/mezzo-valzer)
 408             # and might serve preview URLs (e.g.
 409             # http://www.soundcloud.com/snbrn/ele)
 410             format_dict = self._download_json(
 411                 track_base_url + '/streams', track_id,
 412                 'Downloading track url', query=query, fatal=False) or {}
 413
 414             for key, stream_url in format_dict.items():
 415                 if invalid_url(stream_url):
 416                     continue
 417                 format_urls.add(stream_url)
 418                 mobj = re.search(r'(http|hls)_([^_]+)_(\d+)_url', key)
 419                 if mobj:
 420                     protocol, ext, abr = mobj.groups()
 421                     add_format({
 422                         'abr': abr,
 423                         'ext': ext,
 424                         'url': stream_url,
 425                     }, protocol)
 426
 427         if not formats:
 428             # We fallback to the stream_url in the original info, this
 429             # cannot be always used, sometimes it can give an HTTP 404 error
 430             urlh = self._request_webpage(
 431                 HEADRequest(info.get('stream_url') or track_base_url + '/stream'),
 432                 track_id, query=query, fatal=False)
 433             if urlh:
 434                 stream_url = urlh.geturl()
 435                 if not invalid_url(stream_url):
 436                     add_format({'url': stream_url}, 'http')
 437
 438         for f in formats:
 439             f['vcodec'] = 'none'
 440
 441         self._sort_formats(formats)
 442
 443         user = info.get('user') or {}
 444
 445         thumbnails = []
 446         artwork_url = info.get('artwork_url')
 447         thumbnail = artwork_url or user.get('avatar_url')
 448         if isinstance(thumbnail, compat_str):
 449             if re.search(self._IMAGE_REPL_RE, thumbnail):
 450                 for image_id, size in self._ARTWORK_MAP.items():
 451                     i = {
 452                         'id': image_id,
 453                         'url': re.sub(self._IMAGE_REPL_RE, '-%s.jpg' % image_id, thumbnail),
 454                     }
 455                     if image_id == 'tiny' and not artwork_url:
 456                         size = 18
 457                     elif image_id == 'original':
 458                         i['preference'] = 10
 459                     if size:
 460                         i.update({
 461                             'width': size,
 462                             'height': size,
 463                         })
 464                     thumbnails.append(i)
 465             else:
 466                 thumbnails = [{'url': thumbnail}]
 467
 468         def extract_count(key):
 469             return int_or_none(info.get('%s_count' % key))
 470
 471         return {
 472             'id': track_id,
 473             'uploader': user.get('username'),
 474             'uploader_id': str_or_none(user.get('id')) or user.get('permalink'),
 475             'uploader_url': user.get('permalink_url'),
 476             'timestamp': unified_timestamp(info.get('created_at')),
 477             'title': title,
 478             'description': info.get('description'),
 479             'thumbnails': thumbnails,
 480             'duration': float_or_none(info.get('duration'), 1000),
 481             'webpage_url': info.get('permalink_url'),
 482             'license': info.get('license'),
 483             'view_count': extract_count('playback'),
 484             'like_count': extract_count('favoritings') or extract_count('likes'),
 485             'comment_count': extract_count('comment'),
 486             'repost_count': extract_count('reposts'),
 487             'genre': info.get('genre'),
 488             'formats': formats
 489         }
 490
 491     def _real_extract(self, url):
 492         mobj = re.match(self._VALID_URL, url)
 493
 494         track_id = mobj.group('track_id')
 495
 496         query = {}
 497         if track_id:
 498             info_json_url = self._API_V2_BASE + 'tracks/' + track_id
 499             full_title = track_id
 500             token = mobj.group('secret_token')
 501             if token:
 502                 query['secret_token'] = token
 503         else:
 504             full_title = resolve_title = '%s/%s' % mobj.group('uploader', 'title')
 505             token = mobj.group('token')
 506             if token:
 507                 resolve_title += '/%s' % token
 508             info_json_url = self._resolv_url(self._BASE_URL + resolve_title)
 509
 510         version = 2
 511         info = self._download_json(
 512             info_json_url, full_title, 'Downloading info JSON', query=query, fatal=False)
 513         if not info:
 514             info = self._download_json(
 515                 info_json_url.replace(self._API_V2_BASE, self._API_BASE),
 516                 full_title, 'Downloading info JSON', query=query)
 517             version = 1
 518
 519         return self._extract_info_dict(info, full_title, token, version)
 520
 521
 522 class SoundcloudPlaylistBaseIE(SoundcloudIE):
 523     def _extract_track_entries(self, tracks, token=None):
 524         entries = []
 525         for track in tracks:
 526             track_id = str_or_none(track.get('id'))
 527             url = track.get('permalink_url')
 528             if not url:
 529                 if not track_id:
 530                     continue
 531                 url = self._API_V2_BASE + 'tracks/' + track_id
 532                 if token:
 533                     url += '?secret_token=' + token
 534             entries.append(self.url_result(
 535                 url, SoundcloudIE.ie_key(), track_id))
 536         return entries
 537
 538
 539 class SoundcloudSetIE(SoundcloudPlaylistBaseIE):
 540     _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/(?P<uploader>[\w\d-]+)/sets/(?P<slug_title>[\w\d-]+)(?:/(?P<token>[^?/]+))?'
 541     IE_NAME = 'soundcloud:set'
 542     _TESTS = [{
 543         'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep',
 544         'info_dict': {
 545             'id': '2284613',
 546             'title': 'The Royal Concept EP',
 547         },
 548         'playlist_mincount': 5,
 549     }, {
 550         'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep/token',
 551         'only_matching': True,
 552     }]
 553
 554     def _real_extract(self, url):
 555         mobj = re.match(self._VALID_URL, url)
 556
 557         full_title = '%s/sets/%s' % mobj.group('uploader', 'slug_title')
 558         token = mobj.group('token')
 559         if token:
 560             full_title += '/' + token
 561
 562         info = self._download_json(self._resolv_url(
 563             self._BASE_URL + full_title), full_title)
 564
 565         if 'errors' in info:
 566             msgs = (compat_str(err['error_message']) for err in info['errors'])
 567             raise ExtractorError('unable to download video webpage: %s' % ','.join(msgs))
 568
 569         entries = self._extract_track_entries(info['tracks'], token)
 570
 571         return self.playlist_result(
 572             entries, str_or_none(info.get('id')), info.get('title'))
 573
 574
 575 class SoundcloudPagedPlaylistBaseIE(SoundcloudPlaylistBaseIE):
 576     def _extract_playlist(self, base_url, playlist_id, playlist_title):
 577         COMMON_QUERY = {
 578             'limit': 2000000000,
 579             'linked_partitioning': '1',
 580         }
 581
 582         query = COMMON_QUERY.copy()
 583         query['offset'] = 0
 584
 585         next_href = base_url
 586
 587         entries = []
 588         for i in itertools.count():
 589             response = self._download_json(
 590                 next_href, playlist_id,
 591                 'Downloading track page %s' % (i + 1), query=query)
 592
 593             collection = response['collection']
 594
 595             if not isinstance(collection, list):
 596                 collection = []
 597
 598             # Empty collection may be returned, in this case we proceed
 599             # straight to next_href
 600
 601             def resolve_entry(candidates):
 602                 for cand in candidates:
 603                     if not isinstance(cand, dict):
 604                         continue
 605                     permalink_url = url_or_none(cand.get('permalink_url'))
 606                     if not permalink_url:
 607                         continue
 608                     return self.url_result(
 609                         permalink_url,
 610                         SoundcloudIE.ie_key() if SoundcloudIE.suitable(permalink_url) else None,
 611                         str_or_none(cand.get('id')), cand.get('title'))
 612
 613             for e in collection:
 614                 entry = resolve_entry((e, e.get('track'), e.get('playlist')))
 615                 if entry:
 616                     entries.append(entry)
 617
 618             next_href = response.get('next_href')
 619             if not next_href:
 620                 break
 621
 622             next_href = response['next_href']
 623             parsed_next_href = compat_urlparse.urlparse(next_href)
 624             query = compat_urlparse.parse_qs(parsed_next_href.query)
 625             query.update(COMMON_QUERY)
 626
 627         return {
 628             '_type': 'playlist',
 629             'id': playlist_id,
 630             'title': playlist_title,
 631             'entries': entries,
 632         }
 633
 634
 635 class SoundcloudUserIE(SoundcloudPagedPlaylistBaseIE):
 636     _VALID_URL = r'''(?x)
 637                         https?://
 638                             (?:(?:www|m)\.)?soundcloud\.com/
 639                             (?P<user>[^/]+)
 640                             (?:/
 641                                 (?P<rsrc>tracks|albums|sets|reposts|likes|spotlight)
 642                             )?
 643                             /?(?:[?#].*)?$
 644                     '''
 645     IE_NAME = 'soundcloud:user'
 646     _TESTS = [{
 647         'url': 'https://soundcloud.com/soft-cell-official',
 648         'info_dict': {
 649             'id': '207965082',
 650             'title': 'Soft Cell (All)',
 651         },
 652         'playlist_mincount': 28,
 653     }, {
 654         'url': 'https://soundcloud.com/soft-cell-official/tracks',
 655         'info_dict': {
 656             'id': '207965082',
 657             'title': 'Soft Cell (Tracks)',
 658         },
 659         'playlist_mincount': 27,
 660     }, {
 661         'url': 'https://soundcloud.com/soft-cell-official/albums',
 662         'info_dict': {
 663             'id': '207965082',
 664             'title': 'Soft Cell (Albums)',
 665         },
 666         'playlist_mincount': 1,
 667     }, {
 668         'url': 'https://soundcloud.com/jcv246/sets',
 669         'info_dict': {
 670             'id': '12982173',
 671             'title': 'Jordi / cv (Sets)',
 672         },
 673         'playlist_mincount': 2,
 674     }, {
 675         'url': 'https://soundcloud.com/jcv246/reposts',
 676         'info_dict': {
 677             'id': '12982173',
 678             'title': 'Jordi / cv (Reposts)',
 679         },
 680         'playlist_mincount': 6,
 681     }, {
 682         'url': 'https://soundcloud.com/clalberg/likes',
 683         'info_dict': {
 684             'id': '11817582',
 685             'title': 'clalberg (Likes)',
 686         },
 687         'playlist_mincount': 5,
 688     }, {
 689         'url': 'https://soundcloud.com/grynpyret/spotlight',
 690         'info_dict': {
 691             'id': '7098329',
 692             'title': 'Grynpyret (Spotlight)',
 693         },
 694         'playlist_mincount': 1,
 695     }]
 696
 697     _BASE_URL_MAP = {
 698         'all': 'stream/users/%s',
 699         'tracks': 'users/%s/tracks',
 700         'albums': 'users/%s/albums',
 701         'sets': 'users/%s/playlists',
 702         'reposts': 'stream/users/%s/reposts',
 703         'likes': 'users/%s/likes',
 704         'spotlight': 'users/%s/spotlight',
 705     }
 706
 707     def _real_extract(self, url):
 708         mobj = re.match(self._VALID_URL, url)
 709         uploader = mobj.group('user')
 710
 711         user = self._download_json(
 712             self._resolv_url(self._BASE_URL + uploader),
 713             uploader, 'Downloading user info')
 714
 715         resource = mobj.group('rsrc') or 'all'
 716
 717         return self._extract_playlist(
 718             self._API_V2_BASE + self._BASE_URL_MAP[resource] % user['id'],
 719             str_or_none(user.get('id')),
 720             '%s (%s)' % (user['username'], resource.capitalize()))
 721
 722
 723 class SoundcloudTrackStationIE(SoundcloudPagedPlaylistBaseIE):
 724     _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/stations/track/[^/]+/(?P<id>[^/?#&]+)'
 725     IE_NAME = 'soundcloud:trackstation'
 726     _TESTS = [{
 727         'url': 'https://soundcloud.com/stations/track/officialsundial/your-text',
 728         'info_dict': {
 729             'id': '286017854',
 730             'title': 'Track station: your text',
 731         },
 732         'playlist_mincount': 47,
 733     }]
 734
 735     def _real_extract(self, url):
 736         track_name = self._match_id(url)
 737
 738         track = self._download_json(self._resolv_url(url), track_name)
 739         track_id = self._search_regex(
 740             r'soundcloud:track-stations:(\d+)', track['id'], 'track id')
 741
 742         return self._extract_playlist(
 743             self._API_V2_BASE + 'stations/%s/tracks' % track['id'],
 744             track_id, 'Track station: %s' % track['title'])
 745
 746
 747 class SoundcloudPlaylistIE(SoundcloudPlaylistBaseIE):
 748     _VALID_URL = r'https?://api(?:-v2)?\.soundcloud\.com/playlists/(?P<id>[0-9]+)(?:/?\?secret_token=(?P<token>[^&]+?))?$'
 749     IE_NAME = 'soundcloud:playlist'
 750     _TESTS = [{
 751         'url': 'https://api.soundcloud.com/playlists/4110309',
 752         'info_dict': {
 753             'id': '4110309',
 754             'title': 'TILT Brass - Bowery Poetry Club, August \'03 [Non-Site SCR 02]',
 755             'description': 're:.*?TILT Brass - Bowery Poetry Club',
 756         },
 757         'playlist_count': 6,
 758     }]
 759
 760     def _real_extract(self, url):
 761         mobj = re.match(self._VALID_URL, url)
 762         playlist_id = mobj.group('id')
 763
 764         query = {}
 765         token = mobj.group('token')
 766         if token:
 767             query['secret_token'] = token
 768
 769         data = self._download_json(
 770             self._API_V2_BASE + 'playlists/' + playlist_id,
 771             playlist_id, 'Downloading playlist', query=query)
 772
 773         entries = self._extract_track_entries(data['tracks'], token)
 774
 775         return self.playlist_result(
 776             entries, playlist_id, data.get('title'), data.get('description'))
 777
 778
 779 class SoundcloudSearchIE(SearchInfoExtractor, SoundcloudIE):
 780     IE_NAME = 'soundcloud:search'
 781     IE_DESC = 'Soundcloud search'
 782     _MAX_RESULTS = float('inf')
 783     _TESTS = [{
 784         'url': 'scsearch15:post-avant jazzcore',
 785         'info_dict': {
 786             'title': 'post-avant jazzcore',
 787         },
 788         'playlist_count': 15,
 789     }]
 790
 791     _SEARCH_KEY = 'scsearch'
 792     _MAX_RESULTS_PER_PAGE = 200
 793     _DEFAULT_RESULTS_PER_PAGE = 50
 794
 795     def _get_collection(self, endpoint, collection_id, **query):
 796         limit = min(
 797             query.get('limit', self._DEFAULT_RESULTS_PER_PAGE),
 798             self._MAX_RESULTS_PER_PAGE)
 799         query.update({
 800             'limit': limit,
 801             'linked_partitioning': 1,
 802             'offset': 0,
 803         })
 804         next_url = update_url_query(self._API_V2_BASE + endpoint, query)
 805
 806         collected_results = 0
 807
 808         for i in itertools.count(1):
 809             response = self._download_json(
 810                 next_url, collection_id, 'Downloading page {0}'.format(i),
 811                 'Unable to download API page')
 812
 813             collection = response.get('collection', [])
 814             if not collection:
 815                 break
 816
 817             collection = list(filter(bool, collection))
 818             collected_results += len(collection)
 819
 820             for item in collection:
 821                 yield self.url_result(item['uri'], SoundcloudIE.ie_key())
 822
 823             if not collection or collected_results >= limit:
 824                 break
 825
 826             next_url = response.get('next_href')
 827             if not next_url:
 828                 break
 829
 830     def _get_n_results(self, query, n):
 831         tracks = self._get_collection('search/tracks', query, limit=n, q=query)
 832         return self.playlist_result(tracks, playlist_title=query)