Raphaël G. Git Repositories - youtubedl/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     urlhandle_detect_ext,
  31 )
  32
  33
  34 class SoundcloudEmbedIE(InfoExtractor):
  35     _VALID_URL = r'https?://(?:w|player|p)\.soundcloud\.com/player/?.*?\burl=(?P<id>.+)'
  36     _TEST = {
  37         # from https://www.soundi.fi/uutiset/ennakkokuuntelussa-timo-kaukolammen-station-to-station-to-station-julkaisua-juhlitaan-tanaan-g-livelabissa/
  38         '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',
  39         'only_matching': True,
  40     }
  41
  42     @staticmethod
  43     def _extract_urls(webpage):
  44         return [m.group('url') for m in re.finditer(
  45             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?://)?(?:w\.)?soundcloud\.com/player.+?)\1',
  46             webpage)]
  47
  48     def _real_extract(self, url):
  49         query = compat_urlparse.parse_qs(
  50             compat_urlparse.urlparse(url).query)
  51         api_url = query['url'][0]
  52         secret_token = query.get('secret_token')
  53         if secret_token:
  54             api_url = update_url_query(api_url, {'secret_token': secret_token[0]})
  55         return self.url_result(api_url)
  56
  57
  58 class SoundcloudIE(InfoExtractor):
  59     """Information extractor for soundcloud.com
  60        To access the media, the uid of the song and a stream token
  61        must be extracted from the page source and the script must make
  62        a request to media.soundcloud.com/crossdomain.xml. Then
  63        the media can be grabbed by requesting from an url composed
  64        of the stream token and uid
  65      """
  66
  67     _VALID_URL = r'''(?x)^(?:https?://)?
  68                     (?:(?:(?:www\.|m\.)?soundcloud\.com/
  69                             (?!stations/track)
  70                             (?P<uploader>[\w\d-]+)/
  71                             (?!(?:tracks|albums|sets(?:/.+?)?|reposts|likes|spotlight)/?(?:$|[?#]))
  72                             (?P<title>[\w\d-]+)/?
  73                             (?P<token>[^?]+?)?(?:[?].*)?$)
  74                        |(?:api(?:-v2)?\.soundcloud\.com/tracks/(?P<track_id>\d+)
  75                           (?:/?\?secret_token=(?P<secret_token>[^&]+))?)
  76                     )
  77                     '''
  78     IE_NAME = 'soundcloud'
  79     _TESTS = [
  80         {
  81             'url': 'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
  82             'md5': 'ebef0a451b909710ed1d7787dddbf0d7',
  83             'info_dict': {
  84                 'id': '62986583',
  85                 'ext': 'mp3',
  86                 'title': 'Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1',
  87                 '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',
  88                 'uploader': 'E.T. ExTerrestrial Music',
  89                 'uploader_id': '1571244',
  90                 'timestamp': 1349920598,
  91                 'upload_date': '20121011',
  92                 'duration': 143.216,
  93                 'license': 'all-rights-reserved',
  94                 'view_count': int,
  95                 'like_count': int,
  96                 'comment_count': int,
  97                 'repost_count': int,
  98             }
  99         },
 100         # geo-restricted
 101         {
 102             'url': 'https://soundcloud.com/the-concept-band/goldrushed-mastered?in=the-concept-band/sets/the-royal-concept-ep',
 103             'info_dict': {
 104                 'id': '47127627',
 105                 'ext': 'mp3',
 106                 'title': 'Goldrushed',
 107                 'description': 'From Stockholm Sweden\r\nPovel / Magnus / Filip / David\r\nwww.theroyalconcept.com',
 108                 'uploader': 'The Royal Concept',
 109                 'uploader_id': '9615865',
 110                 'timestamp': 1337635207,
 111                 'upload_date': '20120521',
 112                 'duration': 227.155,
 113                 'license': 'all-rights-reserved',
 114                 'view_count': int,
 115                 'like_count': int,
 116                 'comment_count': int,
 117                 'repost_count': int,
 118             },
 119         },
 120         # private link
 121         {
 122             'url': 'https://soundcloud.com/jaimemf/youtube-dl-test-video-a-y-baw/s-8Pjrp',
 123             'md5': 'aa0dd32bfea9b0c5ef4f02aacd080604',
 124             'info_dict': {
 125                 'id': '123998367',
 126                 'ext': 'mp3',
 127                 'title': 'Youtube - Dl Test Video \'\' Ä↭',
 128                 'description': 'test chars:  \"\'/\\ä↭',
 129                 'uploader': 'jaimeMF',
 130                 'uploader_id': '69767071',
 131                 'timestamp': 1386604920,
 132                 'upload_date': '20131209',
 133                 'duration': 9.927,
 134                 'license': 'all-rights-reserved',
 135                 'view_count': int,
 136                 'like_count': int,
 137                 'comment_count': int,
 138                 'repost_count': int,
 139             },
 140         },
 141         # private link (alt format)
 142         {
 143             'url': 'https://api.soundcloud.com/tracks/123998367?secret_token=s-8Pjrp',
 144             'md5': 'aa0dd32bfea9b0c5ef4f02aacd080604',
 145             'info_dict': {
 146                 'id': '123998367',
 147                 'ext': 'mp3',
 148                 'title': 'Youtube - Dl Test Video \'\' Ä↭',
 149                 'description': 'test chars:  \"\'/\\ä↭',
 150                 'uploader': 'jaimeMF',
 151                 'uploader_id': '69767071',
 152                 'timestamp': 1386604920,
 153                 'upload_date': '20131209',
 154                 'duration': 9.927,
 155                 'license': 'all-rights-reserved',
 156                 'view_count': int,
 157                 'like_count': int,
 158                 'comment_count': int,
 159                 'repost_count': int,
 160             },
 161         },
 162         # downloadable song
 163         {
 164             'url': 'https://soundcloud.com/oddsamples/bus-brakes',
 165             'md5': '7624f2351f8a3b2e7cd51522496e7631',
 166             'info_dict': {
 167                 'id': '128590877',
 168                 'ext': 'mp3',
 169                 'title': 'Bus Brakes',
 170                 'description': 'md5:0053ca6396e8d2fd7b7e1595ef12ab66',
 171                 'uploader': 'oddsamples',
 172                 'uploader_id': '73680509',
 173                 'timestamp': 1389232924,
 174                 'upload_date': '20140109',
 175                 'duration': 17.346,
 176                 'license': 'cc-by-sa',
 177                 'view_count': int,
 178                 'like_count': int,
 179                 'comment_count': int,
 180                 'repost_count': int,
 181             },
 182         },
 183         # private link, downloadable format
 184         {
 185             'url': 'https://soundcloud.com/oriuplift/uponly-238-no-talking-wav/s-AyZUd',
 186             'md5': '64a60b16e617d41d0bef032b7f55441e',
 187             'info_dict': {
 188                 'id': '340344461',
 189                 'ext': 'wav',
 190                 'title': 'Uplifting Only 238 [No Talking] (incl. Alex Feed Guestmix) (Aug 31, 2017) [wav]',
 191                 'description': 'md5:fa20ee0fca76a3d6df8c7e57f3715366',
 192                 'uploader': 'Ori Uplift Music',
 193                 'uploader_id': '12563093',
 194                 'timestamp': 1504206263,
 195                 'upload_date': '20170831',
 196                 'duration': 7449.096,
 197                 'license': 'all-rights-reserved',
 198                 'view_count': int,
 199                 'like_count': int,
 200                 'comment_count': int,
 201                 'repost_count': int,
 202             },
 203         },
 204         # no album art, use avatar pic for thumbnail
 205         {
 206             'url': 'https://soundcloud.com/garyvee/sideways-prod-mad-real',
 207             'md5': '59c7872bc44e5d99b7211891664760c2',
 208             'info_dict': {
 209                 'id': '309699954',
 210                 'ext': 'mp3',
 211                 'title': 'Sideways (Prod. Mad Real)',
 212                 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
 213                 'uploader': 'garyvee',
 214                 'uploader_id': '2366352',
 215                 'timestamp': 1488152409,
 216                 'upload_date': '20170226',
 217                 'duration': 207.012,
 218                 'thumbnail': r're:https?://.*\.jpg',
 219                 'license': 'all-rights-reserved',
 220                 'view_count': int,
 221                 'like_count': int,
 222                 'comment_count': int,
 223                 'repost_count': int,
 224             },
 225             'params': {
 226                 'skip_download': True,
 227             },
 228         },
 229         {
 230             'url': 'https://soundcloud.com/giovannisarani/mezzo-valzer',
 231             'md5': 'e22aecd2bc88e0e4e432d7dcc0a1abf7',
 232             'info_dict': {
 233                 'id': '583011102',
 234                 'ext': 'mp3',
 235                 'title': 'Mezzo Valzer',
 236                 'description': 'md5:4138d582f81866a530317bae316e8b61',
 237                 'uploader': 'Micronie',
 238                 'uploader_id': '3352531',
 239                 'timestamp': 1551394171,
 240                 'upload_date': '20190228',
 241                 'duration': 180.157,
 242                 'thumbnail': r're:https?://.*\.jpg',
 243                 'license': 'all-rights-reserved',
 244                 'view_count': int,
 245                 'like_count': int,
 246                 'comment_count': int,
 247                 'repost_count': int,
 248             },
 249         },
 250         {
 251             # with AAC HQ format available via OAuth token
 252             'url': 'https://soundcloud.com/wandw/the-chainsmokers-ft-daya-dont-let-me-down-ww-remix-1',
 253             'only_matching': True,
 254         },
 255     ]
 256
 257     _API_V2_BASE = 'https://api-v2.soundcloud.com/'
 258     _BASE_URL = 'https://soundcloud.com/'
 259     _IMAGE_REPL_RE = r'-([0-9a-z]+)\.jpg'
 260
 261     _ARTWORK_MAP = {
 262         'mini': 16,
 263         'tiny': 20,
 264         'small': 32,
 265         'badge': 47,
 266         't67x67': 67,
 267         'large': 100,
 268         't300x300': 300,
 269         'crop': 400,
 270         't500x500': 500,
 271         'original': 0,
 272     }
 273
 274     def _store_client_id(self, client_id):
 275         self._downloader.cache.store('soundcloud', 'client_id', client_id)
 276
 277     def _update_client_id(self):
 278         webpage = self._download_webpage('https://soundcloud.com/', None)
 279         for src in reversed(re.findall(r'<script[^>]+src="([^"]+)"', webpage)):
 280             script = self._download_webpage(src, None, fatal=False)
 281             if script:
 282                 client_id = self._search_regex(
 283                     r'client_id\s*:\s*"([0-9a-zA-Z]{32})"',
 284                     script, 'client id', default=None)
 285                 if client_id:
 286                     self._CLIENT_ID = client_id
 287                     self._store_client_id(client_id)
 288                     return
 289         raise ExtractorError('Unable to extract client id')
 290
 291     def _download_json(self, *args, **kwargs):
 292         non_fatal = kwargs.get('fatal') is False
 293         if non_fatal:
 294             del kwargs['fatal']
 295         query = kwargs.get('query', {}).copy()
 296         for _ in range(2):
 297             query['client_id'] = self._CLIENT_ID
 298             kwargs['query'] = query
 299             try:
 300                 return super(SoundcloudIE, self)._download_json(*args, **compat_kwargs(kwargs))
 301             except ExtractorError as e:
 302                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
 303                     self._store_client_id(None)
 304                     self._update_client_id()
 305                     continue
 306                 elif non_fatal:
 307                     self._downloader.report_warning(error_to_compat_str(e))
 308                     return False
 309                 raise
 310
 311     def _real_initialize(self):
 312         self._CLIENT_ID = self._downloader.cache.load('soundcloud', 'client_id') or 'YUKXoArFcqrlQn9tfNHvvyfnDISj04zk'
 313
 314     @classmethod
 315     def _resolv_url(cls, url):
 316         return SoundcloudIE._API_V2_BASE + 'resolve?url=' + url
 317
 318     def _extract_info_dict(self, info, full_title=None, secret_token=None):
 319         track_id = compat_str(info['id'])
 320         title = info['title']
 321
 322         format_urls = set()
 323         formats = []
 324         query = {'client_id': self._CLIENT_ID}
 325         if secret_token:
 326             query['secret_token'] = secret_token
 327
 328         if info.get('downloadable') and info.get('has_downloads_left'):
 329             download_url = update_url_query(
 330                 self._API_V2_BASE + 'tracks/' + track_id + '/download', query)
 331             redirect_url = (self._download_json(download_url, track_id, fatal=False) or {}).get('redirectUri')
 332             if redirect_url:
 333                 urlh = self._request_webpage(
 334                     HEADRequest(redirect_url), track_id, fatal=False)
 335                 if urlh:
 336                     format_url = urlh.geturl()
 337                     format_urls.add(format_url)
 338                     formats.append({
 339                         'format_id': 'download',
 340                         'ext': urlhandle_detect_ext(urlh) or 'mp3',
 341                         'filesize': int_or_none(urlh.headers.get('Content-Length')),
 342                         'url': format_url,
 343                         'preference': 10,
 344                     })
 345
 346         def invalid_url(url):
 347             return not url or url in format_urls
 348
 349         def add_format(f, protocol, is_preview=False):
 350             mobj = re.search(r'\.(?P<abr>\d+)\.(?P<ext>[0-9a-z]{3,4})(?=[/?])', stream_url)
 351             if mobj:
 352                 for k, v in mobj.groupdict().items():
 353                     if not f.get(k):
 354                         f[k] = v
 355             format_id_list = []
 356             if protocol:
 357                 format_id_list.append(protocol)
 358             ext = f.get('ext')
 359             if ext == 'aac':
 360                 f['abr'] = '256'
 361             for k in ('ext', 'abr'):
 362                 v = f.get(k)
 363                 if v:
 364                     format_id_list.append(v)
 365             preview = is_preview or re.search(r'/(?:preview|playlist)/0/30/', f['url'])
 366             if preview:
 367                 format_id_list.append('preview')
 368             abr = f.get('abr')
 369             if abr:
 370                 f['abr'] = int(abr)
 371             if protocol == 'hls':
 372                 protocol = 'm3u8' if ext == 'aac' else 'm3u8_native'
 373             else:
 374                 protocol = 'http'
 375             f.update({
 376                 'format_id': '_'.join(format_id_list),
 377                 'protocol': protocol,
 378                 'preference': -10 if preview else None,
 379             })
 380             formats.append(f)
 381
 382         # New API
 383         transcodings = try_get(
 384             info, lambda x: x['media']['transcodings'], list) or []
 385         for t in transcodings:
 386             if not isinstance(t, dict):
 387                 continue
 388             format_url = url_or_none(t.get('url'))
 389             if not format_url:
 390                 continue
 391             stream = self._download_json(
 392                 format_url, track_id, query=query, fatal=False)
 393             if not isinstance(stream, dict):
 394                 continue
 395             stream_url = url_or_none(stream.get('url'))
 396             if invalid_url(stream_url):
 397                 continue
 398             format_urls.add(stream_url)
 399             stream_format = t.get('format') or {}
 400             protocol = stream_format.get('protocol')
 401             if protocol != 'hls' and '/hls' in format_url:
 402                 protocol = 'hls'
 403             ext = None
 404             preset = str_or_none(t.get('preset'))
 405             if preset:
 406                 ext = preset.split('_')[0]
 407             if ext not in KNOWN_EXTENSIONS:
 408                 ext = mimetype2ext(stream_format.get('mime_type'))
 409             add_format({
 410                 'url': stream_url,
 411                 'ext': ext,
 412             }, 'http' if protocol == 'progressive' else protocol,
 413                 t.get('snipped') or '/preview/' in format_url)
 414
 415         for f in formats:
 416             f['vcodec'] = 'none'
 417
 418         if not formats and info.get('policy') == 'BLOCK':
 419             self.raise_geo_restricted()
 420         self._sort_formats(formats)
 421
 422         user = info.get('user') or {}
 423
 424         thumbnails = []
 425         artwork_url = info.get('artwork_url')
 426         thumbnail = artwork_url or user.get('avatar_url')
 427         if isinstance(thumbnail, compat_str):
 428             if re.search(self._IMAGE_REPL_RE, thumbnail):
 429                 for image_id, size in self._ARTWORK_MAP.items():
 430                     i = {
 431                         'id': image_id,
 432                         'url': re.sub(self._IMAGE_REPL_RE, '-%s.jpg' % image_id, thumbnail),
 433                     }
 434                     if image_id == 'tiny' and not artwork_url:
 435                         size = 18
 436                     elif image_id == 'original':
 437                         i['preference'] = 10
 438                     if size:
 439                         i.update({
 440                             'width': size,
 441                             'height': size,
 442                         })
 443                     thumbnails.append(i)
 444             else:
 445                 thumbnails = [{'url': thumbnail}]
 446
 447         def extract_count(key):
 448             return int_or_none(info.get('%s_count' % key))
 449
 450         return {
 451             'id': track_id,
 452             'uploader': user.get('username'),
 453             'uploader_id': str_or_none(user.get('id')) or user.get('permalink'),
 454             'uploader_url': user.get('permalink_url'),
 455             'timestamp': unified_timestamp(info.get('created_at')),
 456             'title': title,
 457             'description': info.get('description'),
 458             'thumbnails': thumbnails,
 459             'duration': float_or_none(info.get('duration'), 1000),
 460             'webpage_url': info.get('permalink_url'),
 461             'license': info.get('license'),
 462             'view_count': extract_count('playback'),
 463             'like_count': extract_count('favoritings') or extract_count('likes'),
 464             'comment_count': extract_count('comment'),
 465             'repost_count': extract_count('reposts'),
 466             'genre': info.get('genre'),
 467             'formats': formats
 468         }
 469
 470     def _real_extract(self, url):
 471         mobj = re.match(self._VALID_URL, url)
 472
 473         track_id = mobj.group('track_id')
 474
 475         query = {}
 476         if track_id:
 477             info_json_url = self._API_V2_BASE + 'tracks/' + track_id
 478             full_title = track_id
 479             token = mobj.group('secret_token')
 480             if token:
 481                 query['secret_token'] = token
 482         else:
 483             full_title = resolve_title = '%s/%s' % mobj.group('uploader', 'title')
 484             token = mobj.group('token')
 485             if token:
 486                 resolve_title += '/%s' % token
 487             info_json_url = self._resolv_url(self._BASE_URL + resolve_title)
 488
 489         info = self._download_json(
 490             info_json_url, full_title, 'Downloading info JSON', query=query)
 491
 492         return self._extract_info_dict(info, full_title, token)
 493
 494
 495 class SoundcloudPlaylistBaseIE(SoundcloudIE):
 496     def _extract_set(self, playlist, token=None):
 497         playlist_id = compat_str(playlist['id'])
 498         tracks = playlist.get('tracks') or []
 499         if not all([t.get('permalink_url') for t in tracks]) and token:
 500             tracks = self._download_json(
 501                 self._API_V2_BASE + 'tracks', playlist_id,
 502                 'Downloading tracks', query={
 503                     'ids': ','.join([compat_str(t['id']) for t in tracks]),
 504                     'playlistId': playlist_id,
 505                     'playlistSecretToken': token,
 506                 })
 507         entries = []
 508         for track in tracks:
 509             track_id = str_or_none(track.get('id'))
 510             url = track.get('permalink_url')
 511             if not url:
 512                 if not track_id:
 513                     continue
 514                 url = self._API_V2_BASE + 'tracks/' + track_id
 515                 if token:
 516                     url += '?secret_token=' + token
 517             entries.append(self.url_result(
 518                 url, SoundcloudIE.ie_key(), track_id))
 519         return self.playlist_result(
 520             entries, playlist_id,
 521             playlist.get('title'),
 522             playlist.get('description'))
 523
 524
 525 class SoundcloudSetIE(SoundcloudPlaylistBaseIE):
 526     _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/(?P<uploader>[\w\d-]+)/sets/(?P<slug_title>[\w\d-]+)(?:/(?P<token>[^?/]+))?'
 527     IE_NAME = 'soundcloud:set'
 528     _TESTS = [{
 529         'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep',
 530         'info_dict': {
 531             'id': '2284613',
 532             'title': 'The Royal Concept EP',
 533             'description': 'md5:71d07087c7a449e8941a70a29e34671e',
 534         },
 535         'playlist_mincount': 5,
 536     }, {
 537         'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep/token',
 538         'only_matching': True,
 539     }]
 540
 541     def _real_extract(self, url):
 542         mobj = re.match(self._VALID_URL, url)
 543
 544         full_title = '%s/sets/%s' % mobj.group('uploader', 'slug_title')
 545         token = mobj.group('token')
 546         if token:
 547             full_title += '/' + token
 548
 549         info = self._download_json(self._resolv_url(
 550             self._BASE_URL + full_title), full_title)
 551
 552         if 'errors' in info:
 553             msgs = (compat_str(err['error_message']) for err in info['errors'])
 554             raise ExtractorError('unable to download video webpage: %s' % ','.join(msgs))
 555
 556         return self._extract_set(info, token)
 557
 558
 559 class SoundcloudPagedPlaylistBaseIE(SoundcloudIE):
 560     def _extract_playlist(self, base_url, playlist_id, playlist_title):
 561         # Per the SoundCloud documentation, the maximum limit for a linked partioning query is 200.
 562         # https://developers.soundcloud.com/blog/offset-pagination-deprecated
 563         COMMON_QUERY = {
 564             'limit': 200,
 565             'linked_partitioning': '1',
 566         }
 567
 568         query = COMMON_QUERY.copy()
 569         query['offset'] = 0
 570
 571         next_href = base_url
 572
 573         entries = []
 574         for i in itertools.count():
 575             response = self._download_json(
 576                 next_href, playlist_id,
 577                 'Downloading track page %s' % (i + 1), query=query)
 578
 579             collection = response['collection']
 580
 581             if not isinstance(collection, list):
 582                 collection = []
 583
 584             # Empty collection may be returned, in this case we proceed
 585             # straight to next_href
 586
 587             def resolve_entry(candidates):
 588                 for cand in candidates:
 589                     if not isinstance(cand, dict):
 590                         continue
 591                     permalink_url = url_or_none(cand.get('permalink_url'))
 592                     if not permalink_url:
 593                         continue
 594                     return self.url_result(
 595                         permalink_url,
 596                         SoundcloudIE.ie_key() if SoundcloudIE.suitable(permalink_url) else None,
 597                         str_or_none(cand.get('id')), cand.get('title'))
 598
 599             for e in collection:
 600                 entry = resolve_entry((e, e.get('track'), e.get('playlist')))
 601                 if entry:
 602                     entries.append(entry)
 603
 604             next_href = response.get('next_href')
 605             if not next_href:
 606                 break
 607
 608             next_href = response['next_href']
 609             parsed_next_href = compat_urlparse.urlparse(next_href)
 610             query = compat_urlparse.parse_qs(parsed_next_href.query)
 611             query.update(COMMON_QUERY)
 612
 613         return {
 614             '_type': 'playlist',
 615             'id': playlist_id,
 616             'title': playlist_title,
 617             'entries': entries,
 618         }
 619
 620
 621 class SoundcloudUserIE(SoundcloudPagedPlaylistBaseIE):
 622     _VALID_URL = r'''(?x)
 623                         https?://
 624                             (?:(?:www|m)\.)?soundcloud\.com/
 625                             (?P<user>[^/]+)
 626                             (?:/
 627                                 (?P<rsrc>tracks|albums|sets|reposts|likes|spotlight)
 628                             )?
 629                             /?(?:[?#].*)?$
 630                     '''
 631     IE_NAME = 'soundcloud:user'
 632     _TESTS = [{
 633         'url': 'https://soundcloud.com/soft-cell-official',
 634         'info_dict': {
 635             'id': '207965082',
 636             'title': 'Soft Cell (All)',
 637         },
 638         'playlist_mincount': 28,
 639     }, {
 640         'url': 'https://soundcloud.com/soft-cell-official/tracks',
 641         'info_dict': {
 642             'id': '207965082',
 643             'title': 'Soft Cell (Tracks)',
 644         },
 645         'playlist_mincount': 27,
 646     }, {
 647         'url': 'https://soundcloud.com/soft-cell-official/albums',
 648         'info_dict': {
 649             'id': '207965082',
 650             'title': 'Soft Cell (Albums)',
 651         },
 652         'playlist_mincount': 1,
 653     }, {
 654         'url': 'https://soundcloud.com/jcv246/sets',
 655         'info_dict': {
 656             'id': '12982173',
 657             'title': 'Jordi / cv (Sets)',
 658         },
 659         'playlist_mincount': 2,
 660     }, {
 661         'url': 'https://soundcloud.com/jcv246/reposts',
 662         'info_dict': {
 663             'id': '12982173',
 664             'title': 'Jordi / cv (Reposts)',
 665         },
 666         'playlist_mincount': 6,
 667     }, {
 668         'url': 'https://soundcloud.com/clalberg/likes',
 669         'info_dict': {
 670             'id': '11817582',
 671             'title': 'clalberg (Likes)',
 672         },
 673         'playlist_mincount': 5,
 674     }, {
 675         'url': 'https://soundcloud.com/grynpyret/spotlight',
 676         'info_dict': {
 677             'id': '7098329',
 678             'title': 'Grynpyret (Spotlight)',
 679         },
 680         'playlist_mincount': 1,
 681     }]
 682
 683     _BASE_URL_MAP = {
 684         'all': 'stream/users/%s',
 685         'tracks': 'users/%s/tracks',
 686         'albums': 'users/%s/albums',
 687         'sets': 'users/%s/playlists',
 688         'reposts': 'stream/users/%s/reposts',
 689         'likes': 'users/%s/likes',
 690         'spotlight': 'users/%s/spotlight',
 691     }
 692
 693     def _real_extract(self, url):
 694         mobj = re.match(self._VALID_URL, url)
 695         uploader = mobj.group('user')
 696
 697         user = self._download_json(
 698             self._resolv_url(self._BASE_URL + uploader),
 699             uploader, 'Downloading user info')
 700
 701         resource = mobj.group('rsrc') or 'all'
 702
 703         return self._extract_playlist(
 704             self._API_V2_BASE + self._BASE_URL_MAP[resource] % user['id'],
 705             str_or_none(user.get('id')),
 706             '%s (%s)' % (user['username'], resource.capitalize()))
 707
 708
 709 class SoundcloudTrackStationIE(SoundcloudPagedPlaylistBaseIE):
 710     _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/stations/track/[^/]+/(?P<id>[^/?#&]+)'
 711     IE_NAME = 'soundcloud:trackstation'
 712     _TESTS = [{
 713         'url': 'https://soundcloud.com/stations/track/officialsundial/your-text',
 714         'info_dict': {
 715             'id': '286017854',
 716             'title': 'Track station: your text',
 717         },
 718         'playlist_mincount': 47,
 719     }]
 720
 721     def _real_extract(self, url):
 722         track_name = self._match_id(url)
 723
 724         track = self._download_json(self._resolv_url(url), track_name)
 725         track_id = self._search_regex(
 726             r'soundcloud:track-stations:(\d+)', track['id'], 'track id')
 727
 728         return self._extract_playlist(
 729             self._API_V2_BASE + 'stations/%s/tracks' % track['id'],
 730             track_id, 'Track station: %s' % track['title'])
 731
 732
 733 class SoundcloudPlaylistIE(SoundcloudPlaylistBaseIE):
 734     _VALID_URL = r'https?://api(?:-v2)?\.soundcloud\.com/playlists/(?P<id>[0-9]+)(?:/?\?secret_token=(?P<token>[^&]+?))?$'
 735     IE_NAME = 'soundcloud:playlist'
 736     _TESTS = [{
 737         'url': 'https://api.soundcloud.com/playlists/4110309',
 738         'info_dict': {
 739             'id': '4110309',
 740             'title': 'TILT Brass - Bowery Poetry Club, August \'03 [Non-Site SCR 02]',
 741             'description': 're:.*?TILT Brass - Bowery Poetry Club',
 742         },
 743         'playlist_count': 6,
 744     }]
 745
 746     def _real_extract(self, url):
 747         mobj = re.match(self._VALID_URL, url)
 748         playlist_id = mobj.group('id')
 749
 750         query = {}
 751         token = mobj.group('token')
 752         if token:
 753             query['secret_token'] = token
 754
 755         data = self._download_json(
 756             self._API_V2_BASE + 'playlists/' + playlist_id,
 757             playlist_id, 'Downloading playlist', query=query)
 758
 759         return self._extract_set(data, token)
 760
 761
 762 class SoundcloudSearchIE(SearchInfoExtractor, SoundcloudIE):
 763     IE_NAME = 'soundcloud:search'
 764     IE_DESC = 'Soundcloud search'
 765     _MAX_RESULTS = float('inf')
 766     _TESTS = [{
 767         'url': 'scsearch15:post-avant jazzcore',
 768         'info_dict': {
 769             'title': 'post-avant jazzcore',
 770         },
 771         'playlist_count': 15,
 772     }]
 773
 774     _SEARCH_KEY = 'scsearch'
 775     _MAX_RESULTS_PER_PAGE = 200
 776     _DEFAULT_RESULTS_PER_PAGE = 50
 777
 778     def _get_collection(self, endpoint, collection_id, **query):
 779         limit = min(
 780             query.get('limit', self._DEFAULT_RESULTS_PER_PAGE),
 781             self._MAX_RESULTS_PER_PAGE)
 782         query.update({
 783             'limit': limit,
 784             'linked_partitioning': 1,
 785             'offset': 0,
 786         })
 787         next_url = update_url_query(self._API_V2_BASE + endpoint, query)
 788
 789         collected_results = 0
 790
 791         for i in itertools.count(1):
 792             response = self._download_json(
 793                 next_url, collection_id, 'Downloading page {0}'.format(i),
 794                 'Unable to download API page')
 795
 796             collection = response.get('collection', [])
 797             if not collection:
 798                 break
 799
 800             collection = list(filter(bool, collection))
 801             collected_results += len(collection)
 802
 803             for item in collection:
 804                 yield self.url_result(item['uri'], SoundcloudIE.ie_key())
 805
 806             if not collection or collected_results >= limit:
 807                 break
 808
 809             next_url = response.get('next_href')
 810             if not next_url:
 811                 break
 812
 813     def _get_n_results(self, query, n):
 814         tracks = self._get_collection('search/tracks', query, limit=n, q=query)
 815         return self.playlist_result(tracks, playlist_title=query)