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