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 )
  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, preview
 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         },
 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 _store_client_id(self, client_id):
 276         self._downloader.cache.store('soundcloud', 'client_id', client_id)
 277
 278     def _update_client_id(self):
 279         webpage = self._download_webpage('https://soundcloud.com/', None)
 280         for src in reversed(re.findall(r'<script[^>]+src="([^"]+)"', webpage)):
 281             script = self._download_webpage(src, None, fatal=False)
 282             if script:
 283                 client_id = self._search_regex(
 284                     r'client_id\s*:\s*"([0-9a-zA-Z]{32})"',
 285                     script, 'client id', default=None)
 286                 if client_id:
 287                     self._CLIENT_ID = client_id
 288                     self._store_client_id(client_id)
 289                     return
 290         raise ExtractorError('Unable to extract client id')
 291
 292     def _download_json(self, *args, **kwargs):
 293         non_fatal = kwargs.get('fatal') is False
 294         if non_fatal:
 295             del kwargs['fatal']
 296         query = kwargs.get('query', {}).copy()
 297         for _ in range(2):
 298             query['client_id'] = self._CLIENT_ID
 299             kwargs['query'] = query
 300             try:
 301                 return super(SoundcloudIE, self)._download_json(*args, **compat_kwargs(kwargs))
 302             except ExtractorError as e:
 303                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
 304                     self._store_client_id(None)
 305                     self._update_client_id()
 306                     continue
 307                 elif non_fatal:
 308                     self._downloader.report_warning(error_to_compat_str(e))
 309                     return False
 310                 raise
 311
 312     def _real_initialize(self):
 313         self._CLIENT_ID = self._downloader.cache.load('soundcloud', 'client_id') or 'YUKXoArFcqrlQn9tfNHvvyfnDISj04zk'
 314
 315     @classmethod
 316     def _resolv_url(cls, url):
 317         return SoundcloudIE._API_V2_BASE + 'resolve?url=' + url
 318
 319     def _extract_info_dict(self, info, full_title=None, secret_token=None, version=2):
 320         track_id = compat_str(info['id'])
 321         title = info['title']
 322         track_base_url = self._API_BASE + 'tracks/%s' % track_id
 323
 324         format_urls = set()
 325         formats = []
 326         query = {'client_id': self._CLIENT_ID}
 327         if secret_token:
 328             query['secret_token'] = secret_token
 329
 330         if info.get('downloadable') and info.get('has_downloads_left'):
 331             format_url = update_url_query(
 332                 info.get('download_url') or track_base_url + '/download', query)
 333             format_urls.add(format_url)
 334             if version == 2:
 335                 v1_info = self._download_json(
 336                     track_base_url, track_id, query=query, fatal=False) or {}
 337             else:
 338                 v1_info = info
 339             formats.append({
 340                 'format_id': 'download',
 341                 'ext': v1_info.get('original_format') or 'mp3',
 342                 'filesize': int_or_none(v1_info.get('original_content_size')),
 343                 'url': format_url,
 344                 'preference': 10,
 345             })
 346
 347         def invalid_url(url):
 348             return not url or url in format_urls
 349
 350         def add_format(f, protocol, is_preview=False):
 351             mobj = re.search(r'\.(?P<abr>\d+)\.(?P<ext>[0-9a-z]{3,4})(?=[/?])', stream_url)
 352             if mobj:
 353                 for k, v in mobj.groupdict().items():
 354                     if not f.get(k):
 355                         f[k] = v
 356             format_id_list = []
 357             if protocol:
 358                 format_id_list.append(protocol)
 359             for k in ('ext', 'abr'):
 360                 v = f.get(k)
 361                 if v:
 362                     format_id_list.append(v)
 363             preview = is_preview or re.search(r'/(?:preview|playlist)/0/30/', f['url'])
 364             if preview:
 365                 format_id_list.append('preview')
 366             abr = f.get('abr')
 367             if abr:
 368                 f['abr'] = int(abr)
 369             f.update({
 370                 'format_id': '_'.join(format_id_list),
 371                 'protocol': 'm3u8_native' if protocol == 'hls' else 'http',
 372                 'preference': -10 if preview else None,
 373             })
 374             formats.append(f)
 375
 376         # New API
 377         transcodings = try_get(
 378             info, lambda x: x['media']['transcodings'], list) or []
 379         for t in transcodings:
 380             if not isinstance(t, dict):
 381                 continue
 382             format_url = url_or_none(t.get('url'))
 383             if not format_url:
 384                 continue
 385             stream = self._download_json(
 386                 format_url, track_id, query=query, fatal=False)
 387             if not isinstance(stream, dict):
 388                 continue
 389             stream_url = url_or_none(stream.get('url'))
 390             if invalid_url(stream_url):
 391                 continue
 392             format_urls.add(stream_url)
 393             stream_format = t.get('format') or {}
 394             protocol = stream_format.get('protocol')
 395             if protocol != 'hls' and '/hls' in format_url:
 396                 protocol = 'hls'
 397             ext = None
 398             preset = str_or_none(t.get('preset'))
 399             if preset:
 400                 ext = preset.split('_')[0]
 401             if ext not in KNOWN_EXTENSIONS:
 402                 ext = mimetype2ext(stream_format.get('mime_type'))
 403             add_format({
 404                 'url': stream_url,
 405                 'ext': ext,
 406             }, 'http' if protocol == 'progressive' else protocol,
 407                 t.get('snipped') or '/preview/' in format_url)
 408
 409         if not formats:
 410             # Old API, does not work for some tracks (e.g.
 411             # https://soundcloud.com/giovannisarani/mezzo-valzer)
 412             # and might serve preview URLs (e.g.
 413             # http://www.soundcloud.com/snbrn/ele)
 414             format_dict = self._download_json(
 415                 track_base_url + '/streams', track_id,
 416                 'Downloading track url', query=query, fatal=False) or {}
 417
 418             for key, stream_url in format_dict.items():
 419                 if invalid_url(stream_url):
 420                     continue
 421                 format_urls.add(stream_url)
 422                 mobj = re.search(r'(http|hls)_([^_]+)_(\d+)_url', key)
 423                 if mobj:
 424                     protocol, ext, abr = mobj.groups()
 425                     add_format({
 426                         'abr': abr,
 427                         'ext': ext,
 428                         'url': stream_url,
 429                     }, protocol)
 430
 431         if not formats:
 432             # We fallback to the stream_url in the original info, this
 433             # cannot be always used, sometimes it can give an HTTP 404 error
 434             urlh = self._request_webpage(
 435                 HEADRequest(info.get('stream_url') or track_base_url + '/stream'),
 436                 track_id, query=query, fatal=False)
 437             if urlh:
 438                 stream_url = urlh.geturl()
 439                 if not invalid_url(stream_url):
 440                     add_format({'url': stream_url}, 'http')
 441
 442         for f in formats:
 443             f['vcodec'] = 'none'
 444
 445         self._sort_formats(formats)
 446
 447         user = info.get('user') or {}
 448
 449         thumbnails = []
 450         artwork_url = info.get('artwork_url')
 451         thumbnail = artwork_url or user.get('avatar_url')
 452         if isinstance(thumbnail, compat_str):
 453             if re.search(self._IMAGE_REPL_RE, thumbnail):
 454                 for image_id, size in self._ARTWORK_MAP.items():
 455                     i = {
 456                         'id': image_id,
 457                         'url': re.sub(self._IMAGE_REPL_RE, '-%s.jpg' % image_id, thumbnail),
 458                     }
 459                     if image_id == 'tiny' and not artwork_url:
 460                         size = 18
 461                     elif image_id == 'original':
 462                         i['preference'] = 10
 463                     if size:
 464                         i.update({
 465                             'width': size,
 466                             'height': size,
 467                         })
 468                     thumbnails.append(i)
 469             else:
 470                 thumbnails = [{'url': thumbnail}]
 471
 472         def extract_count(key):
 473             return int_or_none(info.get('%s_count' % key))
 474
 475         return {
 476             'id': track_id,
 477             'uploader': user.get('username'),
 478             'uploader_id': str_or_none(user.get('id')) or user.get('permalink'),
 479             'uploader_url': user.get('permalink_url'),
 480             'timestamp': unified_timestamp(info.get('created_at')),
 481             'title': title,
 482             'description': info.get('description'),
 483             'thumbnails': thumbnails,
 484             'duration': float_or_none(info.get('duration'), 1000),
 485             'webpage_url': info.get('permalink_url'),
 486             'license': info.get('license'),
 487             'view_count': extract_count('playback'),
 488             'like_count': extract_count('favoritings') or extract_count('likes'),
 489             'comment_count': extract_count('comment'),
 490             'repost_count': extract_count('reposts'),
 491             'genre': info.get('genre'),
 492             'formats': formats
 493         }
 494
 495     def _real_extract(self, url):
 496         mobj = re.match(self._VALID_URL, url)
 497
 498         track_id = mobj.group('track_id')
 499
 500         query = {}
 501         if track_id:
 502             info_json_url = self._API_V2_BASE + 'tracks/' + track_id
 503             full_title = track_id
 504             token = mobj.group('secret_token')
 505             if token:
 506                 query['secret_token'] = token
 507         else:
 508             full_title = resolve_title = '%s/%s' % mobj.group('uploader', 'title')
 509             token = mobj.group('token')
 510             if token:
 511                 resolve_title += '/%s' % token
 512             info_json_url = self._resolv_url(self._BASE_URL + resolve_title)
 513
 514         version = 2
 515         info = self._download_json(
 516             info_json_url, full_title, 'Downloading info JSON', query=query, fatal=False)
 517         if not info:
 518             info = self._download_json(
 519                 info_json_url.replace(self._API_V2_BASE, self._API_BASE),
 520                 full_title, 'Downloading info JSON', query=query)
 521             version = 1
 522
 523         return self._extract_info_dict(info, full_title, token, version)
 524
 525
 526 class SoundcloudPlaylistBaseIE(SoundcloudIE):
 527     def _extract_track_entries(self, tracks, token=None):
 528         entries = []
 529         for track in tracks:
 530             track_id = str_or_none(track.get('id'))
 531             url = track.get('permalink_url')
 532             if not url:
 533                 if not track_id:
 534                     continue
 535                 url = self._API_V2_BASE + 'tracks/' + track_id
 536                 if token:
 537                     url += '?secret_token=' + token
 538             entries.append(self.url_result(
 539                 url, SoundcloudIE.ie_key(), track_id))
 540         return entries
 541
 542
 543 class SoundcloudSetIE(SoundcloudPlaylistBaseIE):
 544     _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/(?P<uploader>[\w\d-]+)/sets/(?P<slug_title>[\w\d-]+)(?:/(?P<token>[^?/]+))?'
 545     IE_NAME = 'soundcloud:set'
 546     _TESTS = [{
 547         'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep',
 548         'info_dict': {
 549             'id': '2284613',
 550             'title': 'The Royal Concept EP',
 551         },
 552         'playlist_mincount': 5,
 553     }, {
 554         'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep/token',
 555         'only_matching': True,
 556     }]
 557
 558     def _real_extract(self, url):
 559         mobj = re.match(self._VALID_URL, url)
 560
 561         full_title = '%s/sets/%s' % mobj.group('uploader', 'slug_title')
 562         token = mobj.group('token')
 563         if token:
 564             full_title += '/' + token
 565
 566         info = self._download_json(self._resolv_url(
 567             self._BASE_URL + full_title), full_title)
 568
 569         if 'errors' in info:
 570             msgs = (compat_str(err['error_message']) for err in info['errors'])
 571             raise ExtractorError('unable to download video webpage: %s' % ','.join(msgs))
 572
 573         entries = self._extract_track_entries(info['tracks'], token)
 574
 575         return self.playlist_result(
 576             entries, str_or_none(info.get('id')), info.get('title'))
 577
 578
 579 class SoundcloudPagedPlaylistBaseIE(SoundcloudPlaylistBaseIE):
 580     def _extract_playlist(self, base_url, playlist_id, playlist_title):
 581         COMMON_QUERY = {
 582             'limit': 2000000000,
 583             'linked_partitioning': '1',
 584         }
 585
 586         query = COMMON_QUERY.copy()
 587         query['offset'] = 0
 588
 589         next_href = base_url
 590
 591         entries = []
 592         for i in itertools.count():
 593             response = self._download_json(
 594                 next_href, playlist_id,
 595                 'Downloading track page %s' % (i + 1), query=query)
 596
 597             collection = response['collection']
 598
 599             if not isinstance(collection, list):
 600                 collection = []
 601
 602             # Empty collection may be returned, in this case we proceed
 603             # straight to next_href
 604
 605             def resolve_entry(candidates):
 606                 for cand in candidates:
 607                     if not isinstance(cand, dict):
 608                         continue
 609                     permalink_url = url_or_none(cand.get('permalink_url'))
 610                     if not permalink_url:
 611                         continue
 612                     return self.url_result(
 613                         permalink_url,
 614                         SoundcloudIE.ie_key() if SoundcloudIE.suitable(permalink_url) else None,
 615                         str_or_none(cand.get('id')), cand.get('title'))
 616
 617             for e in collection:
 618                 entry = resolve_entry((e, e.get('track'), e.get('playlist')))
 619                 if entry:
 620                     entries.append(entry)
 621
 622             next_href = response.get('next_href')
 623             if not next_href:
 624                 break
 625
 626             next_href = response['next_href']
 627             parsed_next_href = compat_urlparse.urlparse(next_href)
 628             query = compat_urlparse.parse_qs(parsed_next_href.query)
 629             query.update(COMMON_QUERY)
 630
 631         return {
 632             '_type': 'playlist',
 633             'id': playlist_id,
 634             'title': playlist_title,
 635             'entries': entries,
 636         }
 637
 638
 639 class SoundcloudUserIE(SoundcloudPagedPlaylistBaseIE):
 640     _VALID_URL = r'''(?x)
 641                         https?://
 642                             (?:(?:www|m)\.)?soundcloud\.com/
 643                             (?P<user>[^/]+)
 644                             (?:/
 645                                 (?P<rsrc>tracks|albums|sets|reposts|likes|spotlight)
 646                             )?
 647                             /?(?:[?#].*)?$
 648                     '''
 649     IE_NAME = 'soundcloud:user'
 650     _TESTS = [{
 651         'url': 'https://soundcloud.com/soft-cell-official',
 652         'info_dict': {
 653             'id': '207965082',
 654             'title': 'Soft Cell (All)',
 655         },
 656         'playlist_mincount': 28,
 657     }, {
 658         'url': 'https://soundcloud.com/soft-cell-official/tracks',
 659         'info_dict': {
 660             'id': '207965082',
 661             'title': 'Soft Cell (Tracks)',
 662         },
 663         'playlist_mincount': 27,
 664     }, {
 665         'url': 'https://soundcloud.com/soft-cell-official/albums',
 666         'info_dict': {
 667             'id': '207965082',
 668             'title': 'Soft Cell (Albums)',
 669         },
 670         'playlist_mincount': 1,
 671     }, {
 672         'url': 'https://soundcloud.com/jcv246/sets',
 673         'info_dict': {
 674             'id': '12982173',
 675             'title': 'Jordi / cv (Sets)',
 676         },
 677         'playlist_mincount': 2,
 678     }, {
 679         'url': 'https://soundcloud.com/jcv246/reposts',
 680         'info_dict': {
 681             'id': '12982173',
 682             'title': 'Jordi / cv (Reposts)',
 683         },
 684         'playlist_mincount': 6,
 685     }, {
 686         'url': 'https://soundcloud.com/clalberg/likes',
 687         'info_dict': {
 688             'id': '11817582',
 689             'title': 'clalberg (Likes)',
 690         },
 691         'playlist_mincount': 5,
 692     }, {
 693         'url': 'https://soundcloud.com/grynpyret/spotlight',
 694         'info_dict': {
 695             'id': '7098329',
 696             'title': 'Grynpyret (Spotlight)',
 697         },
 698         'playlist_mincount': 1,
 699     }]
 700
 701     _BASE_URL_MAP = {
 702         'all': 'stream/users/%s',
 703         'tracks': 'users/%s/tracks',
 704         'albums': 'users/%s/albums',
 705         'sets': 'users/%s/playlists',
 706         'reposts': 'stream/users/%s/reposts',
 707         'likes': 'users/%s/likes',
 708         'spotlight': 'users/%s/spotlight',
 709     }
 710
 711     def _real_extract(self, url):
 712         mobj = re.match(self._VALID_URL, url)
 713         uploader = mobj.group('user')
 714
 715         user = self._download_json(
 716             self._resolv_url(self._BASE_URL + uploader),
 717             uploader, 'Downloading user info')
 718
 719         resource = mobj.group('rsrc') or 'all'
 720
 721         return self._extract_playlist(
 722             self._API_V2_BASE + self._BASE_URL_MAP[resource] % user['id'],
 723             str_or_none(user.get('id')),
 724             '%s (%s)' % (user['username'], resource.capitalize()))
 725
 726
 727 class SoundcloudTrackStationIE(SoundcloudPagedPlaylistBaseIE):
 728     _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/stations/track/[^/]+/(?P<id>[^/?#&]+)'
 729     IE_NAME = 'soundcloud:trackstation'
 730     _TESTS = [{
 731         'url': 'https://soundcloud.com/stations/track/officialsundial/your-text',
 732         'info_dict': {
 733             'id': '286017854',
 734             'title': 'Track station: your text',
 735         },
 736         'playlist_mincount': 47,
 737     }]
 738
 739     def _real_extract(self, url):
 740         track_name = self._match_id(url)
 741
 742         track = self._download_json(self._resolv_url(url), track_name)
 743         track_id = self._search_regex(
 744             r'soundcloud:track-stations:(\d+)', track['id'], 'track id')
 745
 746         return self._extract_playlist(
 747             self._API_V2_BASE + 'stations/%s/tracks' % track['id'],
 748             track_id, 'Track station: %s' % track['title'])
 749
 750
 751 class SoundcloudPlaylistIE(SoundcloudPlaylistBaseIE):
 752     _VALID_URL = r'https?://api(?:-v2)?\.soundcloud\.com/playlists/(?P<id>[0-9]+)(?:/?\?secret_token=(?P<token>[^&]+?))?$'
 753     IE_NAME = 'soundcloud:playlist'
 754     _TESTS = [{
 755         'url': 'https://api.soundcloud.com/playlists/4110309',
 756         'info_dict': {
 757             'id': '4110309',
 758             'title': 'TILT Brass - Bowery Poetry Club, August \'03 [Non-Site SCR 02]',
 759             'description': 're:.*?TILT Brass - Bowery Poetry Club',
 760         },
 761         'playlist_count': 6,
 762     }]
 763
 764     def _real_extract(self, url):
 765         mobj = re.match(self._VALID_URL, url)
 766         playlist_id = mobj.group('id')
 767
 768         query = {}
 769         token = mobj.group('token')
 770         if token:
 771             query['secret_token'] = token
 772
 773         data = self._download_json(
 774             self._API_V2_BASE + 'playlists/' + playlist_id,
 775             playlist_id, 'Downloading playlist', query=query)
 776
 777         entries = self._extract_track_entries(data['tracks'], token)
 778
 779         return self.playlist_result(
 780             entries, playlist_id, data.get('title'), data.get('description'))
 781
 782
 783 class SoundcloudSearchIE(SearchInfoExtractor, SoundcloudIE):
 784     IE_NAME = 'soundcloud:search'
 785     IE_DESC = 'Soundcloud search'
 786     _MAX_RESULTS = float('inf')
 787     _TESTS = [{
 788         'url': 'scsearch15:post-avant jazzcore',
 789         'info_dict': {
 790             'title': 'post-avant jazzcore',
 791         },
 792         'playlist_count': 15,
 793     }]
 794
 795     _SEARCH_KEY = 'scsearch'
 796     _MAX_RESULTS_PER_PAGE = 200
 797     _DEFAULT_RESULTS_PER_PAGE = 50
 798
 799     def _get_collection(self, endpoint, collection_id, **query):
 800         limit = min(
 801             query.get('limit', self._DEFAULT_RESULTS_PER_PAGE),
 802             self._MAX_RESULTS_PER_PAGE)
 803         query.update({
 804             'limit': limit,
 805             'linked_partitioning': 1,
 806             'offset': 0,
 807         })
 808         next_url = update_url_query(self._API_V2_BASE + endpoint, query)
 809
 810         collected_results = 0
 811
 812         for i in itertools.count(1):
 813             response = self._download_json(
 814                 next_url, collection_id, 'Downloading page {0}'.format(i),
 815                 'Unable to download API page')
 816
 817             collection = response.get('collection', [])
 818             if not collection:
 819                 break
 820
 821             collection = list(filter(bool, collection))
 822             collected_results += len(collection)
 823
 824             for item in collection:
 825                 yield self.url_result(item['uri'], SoundcloudIE.ie_key())
 826
 827             if not collection or collected_results >= limit:
 828                 break
 829
 830             next_url = response.get('next_href')
 831             if not next_url:
 832                 break
 833
 834     def _get_n_results(self, query, n):
 835         tracks = self._get_collection('search/tracks', query, limit=n, q=query)
 836         return self.playlist_result(tracks, playlist_title=query)