_ 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_str,
  13     compat_urlparse,
  14     compat_urllib_parse_urlencode,
  15 )
  16 from ..utils import (
  17     ExtractorError,
  18     int_or_none,
  19     try_get,
  20     unified_timestamp,
  21     update_url_query,
  22     url_or_none,
  23 )
  24
  25
  26 class SoundcloudIE(InfoExtractor):
  27     """Information extractor for soundcloud.com
  28        To access the media, the uid of the song and a stream token
  29        must be extracted from the page source and the script must make
  30        a request to media.soundcloud.com/crossdomain.xml. Then
  31        the media can be grabbed by requesting from an url composed
  32        of the stream token and uid
  33      """
  34
  35     _VALID_URL = r'''(?x)^(?:https?://)?
  36                     (?:(?:(?:www\.|m\.)?soundcloud\.com/
  37                             (?!stations/track)
  38                             (?P<uploader>[\w\d-]+)/
  39                             (?!(?:tracks|albums|sets(?:/.+?)?|reposts|likes|spotlight)/?(?:$|[?#]))
  40                             (?P<title>[\w\d-]+)/?
  41                             (?P<token>[^?]+?)?(?:[?].*)?$)
  42                        |(?:api\.soundcloud\.com/tracks/(?P<track_id>\d+)
  43                           (?:/?\?secret_token=(?P<secret_token>[^&]+))?)
  44                        |(?P<player>(?:w|player|p.)\.soundcloud\.com/player/?.*?url=.*)
  45                     )
  46                     '''
  47     IE_NAME = 'soundcloud'
  48     _TESTS = [
  49         {
  50             'url': 'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
  51             'md5': 'ebef0a451b909710ed1d7787dddbf0d7',
  52             'info_dict': {
  53                 'id': '62986583',
  54                 'ext': 'mp3',
  55                 'title': 'Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1',
  56                 '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',
  57                 'uploader': 'E.T. ExTerrestrial Music',
  58                 'timestamp': 1349920598,
  59                 'upload_date': '20121011',
  60                 'duration': 143,
  61                 'license': 'all-rights-reserved',
  62                 'view_count': int,
  63                 'like_count': int,
  64                 'comment_count': int,
  65                 'repost_count': int,
  66             }
  67         },
  68         # not streamable song
  69         {
  70             'url': 'https://soundcloud.com/the-concept-band/goldrushed-mastered?in=the-concept-band/sets/the-royal-concept-ep',
  71             'info_dict': {
  72                 'id': '47127627',
  73                 'ext': 'mp3',
  74                 'title': 'Goldrushed',
  75                 'description': 'From Stockholm Sweden\r\nPovel / Magnus / Filip / David\r\nwww.theroyalconcept.com',
  76                 'uploader': 'The Royal Concept',
  77                 'timestamp': 1337635207,
  78                 'upload_date': '20120521',
  79                 'duration': 30,
  80                 'license': 'all-rights-reserved',
  81                 'view_count': int,
  82                 'like_count': int,
  83                 'comment_count': int,
  84                 'repost_count': int,
  85             },
  86             'params': {
  87                 # rtmp
  88                 'skip_download': True,
  89             },
  90         },
  91         # private link
  92         {
  93             'url': 'https://soundcloud.com/jaimemf/youtube-dl-test-video-a-y-baw/s-8Pjrp',
  94             'md5': 'aa0dd32bfea9b0c5ef4f02aacd080604',
  95             'info_dict': {
  96                 'id': '123998367',
  97                 'ext': 'mp3',
  98                 'title': 'Youtube - Dl Test Video \'\' Ä↭',
  99                 'description': 'test chars:  \"\'/\\ä↭',
 100                 'uploader': 'jaimeMF',
 101                 'timestamp': 1386604920,
 102                 'upload_date': '20131209',
 103                 'duration': 9,
 104                 'license': 'all-rights-reserved',
 105                 'view_count': int,
 106                 'like_count': int,
 107                 'comment_count': int,
 108                 'repost_count': int,
 109             },
 110         },
 111         # private link (alt format)
 112         {
 113             'url': 'https://api.soundcloud.com/tracks/123998367?secret_token=s-8Pjrp',
 114             'md5': 'aa0dd32bfea9b0c5ef4f02aacd080604',
 115             'info_dict': {
 116                 'id': '123998367',
 117                 'ext': 'mp3',
 118                 'title': 'Youtube - Dl Test Video \'\' Ä↭',
 119                 'description': 'test chars:  \"\'/\\ä↭',
 120                 'uploader': 'jaimeMF',
 121                 'timestamp': 1386604920,
 122                 'upload_date': '20131209',
 123                 'duration': 9,
 124                 'license': 'all-rights-reserved',
 125                 'view_count': int,
 126                 'like_count': int,
 127                 'comment_count': int,
 128                 'repost_count': int,
 129             },
 130         },
 131         # downloadable song
 132         {
 133             'url': 'https://soundcloud.com/oddsamples/bus-brakes',
 134             'md5': '7624f2351f8a3b2e7cd51522496e7631',
 135             'info_dict': {
 136                 'id': '128590877',
 137                 'ext': 'mp3',
 138                 'title': 'Bus Brakes',
 139                 'description': 'md5:0053ca6396e8d2fd7b7e1595ef12ab66',
 140                 'uploader': 'oddsamples',
 141                 'timestamp': 1389232924,
 142                 'upload_date': '20140109',
 143                 'duration': 17,
 144                 'license': 'cc-by-sa',
 145                 'view_count': int,
 146                 'like_count': int,
 147                 'comment_count': int,
 148                 'repost_count': int,
 149             },
 150         },
 151         # private link, downloadable format
 152         {
 153             'url': 'https://soundcloud.com/oriuplift/uponly-238-no-talking-wav/s-AyZUd',
 154             'md5': '64a60b16e617d41d0bef032b7f55441e',
 155             'info_dict': {
 156                 'id': '340344461',
 157                 'ext': 'wav',
 158                 'title': 'Uplifting Only 238 [No Talking] (incl. Alex Feed Guestmix) (Aug 31, 2017) [wav]',
 159                 'description': 'md5:fa20ee0fca76a3d6df8c7e57f3715366',
 160                 'uploader': 'Ori Uplift Music',
 161                 'timestamp': 1504206263,
 162                 'upload_date': '20170831',
 163                 'duration': 7449,
 164                 'license': 'all-rights-reserved',
 165                 'view_count': int,
 166                 'like_count': int,
 167                 'comment_count': int,
 168                 'repost_count': int,
 169             },
 170         },
 171         # no album art, use avatar pic for thumbnail
 172         {
 173             'url': 'https://soundcloud.com/garyvee/sideways-prod-mad-real',
 174             'md5': '59c7872bc44e5d99b7211891664760c2',
 175             'info_dict': {
 176                 'id': '309699954',
 177                 'ext': 'mp3',
 178                 'title': 'Sideways (Prod. Mad Real)',
 179                 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
 180                 'uploader': 'garyvee',
 181                 'timestamp': 1488152409,
 182                 'upload_date': '20170226',
 183                 'duration': 207,
 184                 'thumbnail': r're:https?://.*\.jpg',
 185                 'license': 'all-rights-reserved',
 186                 'view_count': int,
 187                 'like_count': int,
 188                 'comment_count': int,
 189                 'repost_count': int,
 190             },
 191             'params': {
 192                 'skip_download': True,
 193             },
 194         },
 195     ]
 196
 197     _CLIENT_ID = 'NmW1FlPaiL94ueEu7oziOWjYEzZzQDcK'
 198
 199     @staticmethod
 200     def _extract_urls(webpage):
 201         return [m.group('url') for m in re.finditer(
 202             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?://)?(?:w\.)?soundcloud\.com/player.+?)\1',
 203             webpage)]
 204
 205     def report_resolve(self, video_id):
 206         """Report information extraction."""
 207         self.to_screen('%s: Resolving id' % video_id)
 208
 209     @classmethod
 210     def _resolv_url(cls, url):
 211         return 'https://api.soundcloud.com/resolve.json?url=' + url + '&client_id=' + cls._CLIENT_ID
 212
 213     def _extract_info_dict(self, info, full_title=None, quiet=False, secret_token=None):
 214         track_id = compat_str(info['id'])
 215         title = info['title']
 216         name = full_title or track_id
 217         if quiet:
 218             self.report_extraction(name)
 219         thumbnail = info.get('artwork_url') or info.get('user', {}).get('avatar_url')
 220         if isinstance(thumbnail, compat_str):
 221             thumbnail = thumbnail.replace('-large', '-t500x500')
 222         username = try_get(info, lambda x: x['user']['username'], compat_str)
 223
 224         def extract_count(key):
 225             return int_or_none(info.get('%s_count' % key))
 226
 227         result = {
 228             'id': track_id,
 229             'uploader': username,
 230             'timestamp': unified_timestamp(info.get('created_at')),
 231             'title': title,
 232             'description': info.get('description'),
 233             'thumbnail': thumbnail,
 234             'duration': int_or_none(info.get('duration'), 1000),
 235             'webpage_url': info.get('permalink_url'),
 236             'license': info.get('license'),
 237             'view_count': extract_count('playback'),
 238             'like_count': extract_count('favoritings'),
 239             'comment_count': extract_count('comment'),
 240             'repost_count': extract_count('reposts'),
 241             'genre': info.get('genre'),
 242         }
 243         formats = []
 244         query = {'client_id': self._CLIENT_ID}
 245         if secret_token is not None:
 246             query['secret_token'] = secret_token
 247         if info.get('downloadable', False):
 248             # We can build a direct link to the song
 249             format_url = update_url_query(
 250                 'https://api.soundcloud.com/tracks/%s/download' % track_id, query)
 251             formats.append({
 252                 'format_id': 'download',
 253                 'ext': info.get('original_format', 'mp3'),
 254                 'url': format_url,
 255                 'vcodec': 'none',
 256                 'preference': 10,
 257             })
 258
 259         # We have to retrieve the url
 260         format_dict = self._download_json(
 261             'https://api.soundcloud.com/i1/tracks/%s/streams' % track_id,
 262             track_id, 'Downloading track url', query=query)
 263
 264         for key, stream_url in format_dict.items():
 265             ext, abr = 'mp3', None
 266             mobj = re.search(r'_([^_]+)_(\d+)_url', key)
 267             if mobj:
 268                 ext, abr = mobj.groups()
 269                 abr = int(abr)
 270             if key.startswith('http'):
 271                 stream_formats = [{
 272                     'format_id': key,
 273                     'ext': ext,
 274                     'url': stream_url,
 275                 }]
 276             elif key.startswith('rtmp'):
 277                 # The url doesn't have an rtmp app, we have to extract the playpath
 278                 url, path = stream_url.split('mp3:', 1)
 279                 stream_formats = [{
 280                     'format_id': key,
 281                     'url': url,
 282                     'play_path': 'mp3:' + path,
 283                     'ext': 'flv',
 284                 }]
 285             elif key.startswith('hls'):
 286                 stream_formats = self._extract_m3u8_formats(
 287                     stream_url, track_id, ext, entry_protocol='m3u8_native',
 288                     m3u8_id=key, fatal=False)
 289             else:
 290                 continue
 291
 292             if abr:
 293                 for f in stream_formats:
 294                     f['abr'] = abr
 295
 296             formats.extend(stream_formats)
 297
 298         if not formats:
 299             # We fallback to the stream_url in the original info, this
 300             # cannot be always used, sometimes it can give an HTTP 404 error
 301             formats.append({
 302                 'format_id': 'fallback',
 303                 'url': update_url_query(info['stream_url'], query),
 304                 'ext': 'mp3',
 305             })
 306
 307         for f in formats:
 308             f['vcodec'] = 'none'
 309
 310         self._check_formats(formats, track_id)
 311         self._sort_formats(formats)
 312         result['formats'] = formats
 313
 314         return result
 315
 316     def _real_extract(self, url):
 317         mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
 318         if mobj is None:
 319             raise ExtractorError('Invalid URL: %s' % url)
 320
 321         track_id = mobj.group('track_id')
 322
 323         if track_id is not None:
 324             info_json_url = 'https://api.soundcloud.com/tracks/' + track_id + '.json?client_id=' + self._CLIENT_ID
 325             full_title = track_id
 326             token = mobj.group('secret_token')
 327             if token:
 328                 info_json_url += '&secret_token=' + token
 329         elif mobj.group('player'):
 330             query = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
 331             real_url = query['url'][0]
 332             # If the token is in the query of the original url we have to
 333             # manually add it
 334             if 'secret_token' in query:
 335                 real_url += '?secret_token=' + query['secret_token'][0]
 336             return self.url_result(real_url)
 337         else:
 338             # extract uploader (which is in the url)
 339             uploader = mobj.group('uploader')
 340             # extract simple title (uploader + slug of song title)
 341             slug_title = mobj.group('title')
 342             token = mobj.group('token')
 343             full_title = resolve_title = '%s/%s' % (uploader, slug_title)
 344             if token:
 345                 resolve_title += '/%s' % token
 346
 347             self.report_resolve(full_title)
 348
 349             url = 'https://soundcloud.com/%s' % resolve_title
 350             info_json_url = self._resolv_url(url)
 351         info = self._download_json(info_json_url, full_title, 'Downloading info JSON')
 352
 353         return self._extract_info_dict(info, full_title, secret_token=token)
 354
 355
 356 class SoundcloudPlaylistBaseIE(SoundcloudIE):
 357     @staticmethod
 358     def _extract_id(e):
 359         return compat_str(e['id']) if e.get('id') else None
 360
 361     def _extract_track_entries(self, tracks):
 362         return [
 363             self.url_result(
 364                 track['permalink_url'], SoundcloudIE.ie_key(),
 365                 video_id=self._extract_id(track))
 366             for track in tracks if track.get('permalink_url')]
 367
 368
 369 class SoundcloudSetIE(SoundcloudPlaylistBaseIE):
 370     _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/(?P<uploader>[\w\d-]+)/sets/(?P<slug_title>[\w\d-]+)(?:/(?P<token>[^?/]+))?'
 371     IE_NAME = 'soundcloud:set'
 372     _TESTS = [{
 373         'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep',
 374         'info_dict': {
 375             'id': '2284613',
 376             'title': 'The Royal Concept EP',
 377         },
 378         'playlist_mincount': 5,
 379     }, {
 380         'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep/token',
 381         'only_matching': True,
 382     }]
 383
 384     def _real_extract(self, url):
 385         mobj = re.match(self._VALID_URL, url)
 386
 387         # extract uploader (which is in the url)
 388         uploader = mobj.group('uploader')
 389         # extract simple title (uploader + slug of song title)
 390         slug_title = mobj.group('slug_title')
 391         full_title = '%s/sets/%s' % (uploader, slug_title)
 392         url = 'https://soundcloud.com/%s/sets/%s' % (uploader, slug_title)
 393
 394         token = mobj.group('token')
 395         if token:
 396             full_title += '/' + token
 397             url += '/' + token
 398
 399         self.report_resolve(full_title)
 400
 401         resolv_url = self._resolv_url(url)
 402         info = self._download_json(resolv_url, full_title)
 403
 404         if 'errors' in info:
 405             msgs = (compat_str(err['error_message']) for err in info['errors'])
 406             raise ExtractorError('unable to download video webpage: %s' % ','.join(msgs))
 407
 408         entries = self._extract_track_entries(info['tracks'])
 409
 410         return {
 411             '_type': 'playlist',
 412             'entries': entries,
 413             'id': '%s' % info['id'],
 414             'title': info['title'],
 415         }
 416
 417
 418 class SoundcloudPagedPlaylistBaseIE(SoundcloudPlaylistBaseIE):
 419     _API_V2_BASE = 'https://api-v2.soundcloud.com'
 420
 421     def _extract_playlist(self, base_url, playlist_id, playlist_title):
 422         COMMON_QUERY = {
 423             'limit': 50,
 424             'client_id': self._CLIENT_ID,
 425             'linked_partitioning': '1',
 426         }
 427
 428         query = COMMON_QUERY.copy()
 429         query['offset'] = 0
 430
 431         next_href = base_url + '?' + compat_urllib_parse_urlencode(query)
 432
 433         entries = []
 434         for i in itertools.count():
 435             response = self._download_json(
 436                 next_href, playlist_id, 'Downloading track page %s' % (i + 1))
 437
 438             collection = response['collection']
 439
 440             if not isinstance(collection, list):
 441                 collection = []
 442
 443             # Empty collection may be returned, in this case we proceed
 444             # straight to next_href
 445
 446             def resolve_entry(candidates):
 447                 for cand in candidates:
 448                     if not isinstance(cand, dict):
 449                         continue
 450                     permalink_url = url_or_none(cand.get('permalink_url'))
 451                     if not permalink_url:
 452                         continue
 453                     return self.url_result(
 454                         permalink_url,
 455                         ie=SoundcloudIE.ie_key() if SoundcloudIE.suitable(permalink_url) else None,
 456                         video_id=self._extract_id(cand),
 457                         video_title=cand.get('title'))
 458
 459             for e in collection:
 460                 entry = resolve_entry((e, e.get('track'), e.get('playlist')))
 461                 if entry:
 462                     entries.append(entry)
 463
 464             next_href = response.get('next_href')
 465             if not next_href:
 466                 break
 467
 468             parsed_next_href = compat_urlparse.urlparse(response['next_href'])
 469             qs = compat_urlparse.parse_qs(parsed_next_href.query)
 470             qs.update(COMMON_QUERY)
 471             next_href = compat_urlparse.urlunparse(
 472                 parsed_next_href._replace(query=compat_urllib_parse_urlencode(qs, True)))
 473
 474         return {
 475             '_type': 'playlist',
 476             'id': playlist_id,
 477             'title': playlist_title,
 478             'entries': entries,
 479         }
 480
 481
 482 class SoundcloudUserIE(SoundcloudPagedPlaylistBaseIE):
 483     _VALID_URL = r'''(?x)
 484                         https?://
 485                             (?:(?:www|m)\.)?soundcloud\.com/
 486                             (?P<user>[^/]+)
 487                             (?:/
 488                                 (?P<rsrc>tracks|albums|sets|reposts|likes|spotlight)
 489                             )?
 490                             /?(?:[?#].*)?$
 491                     '''
 492     IE_NAME = 'soundcloud:user'
 493     _TESTS = [{
 494         'url': 'https://soundcloud.com/soft-cell-official',
 495         'info_dict': {
 496             'id': '207965082',
 497             'title': 'Soft Cell (All)',
 498         },
 499         'playlist_mincount': 28,
 500     }, {
 501         'url': 'https://soundcloud.com/soft-cell-official/tracks',
 502         'info_dict': {
 503             'id': '207965082',
 504             'title': 'Soft Cell (Tracks)',
 505         },
 506         'playlist_mincount': 27,
 507     }, {
 508         'url': 'https://soundcloud.com/soft-cell-official/albums',
 509         'info_dict': {
 510             'id': '207965082',
 511             'title': 'Soft Cell (Albums)',
 512         },
 513         'playlist_mincount': 1,
 514     }, {
 515         'url': 'https://soundcloud.com/jcv246/sets',
 516         'info_dict': {
 517             'id': '12982173',
 518             'title': 'Jordi / cv (Playlists)',
 519         },
 520         'playlist_mincount': 2,
 521     }, {
 522         'url': 'https://soundcloud.com/jcv246/reposts',
 523         'info_dict': {
 524             'id': '12982173',
 525             'title': 'Jordi / cv (Reposts)',
 526         },
 527         'playlist_mincount': 6,
 528     }, {
 529         'url': 'https://soundcloud.com/clalberg/likes',
 530         'info_dict': {
 531             'id': '11817582',
 532             'title': 'clalberg (Likes)',
 533         },
 534         'playlist_mincount': 5,
 535     }, {
 536         'url': 'https://soundcloud.com/grynpyret/spotlight',
 537         'info_dict': {
 538             'id': '7098329',
 539             'title': 'Grynpyret (Spotlight)',
 540         },
 541         'playlist_mincount': 1,
 542     }]
 543
 544     _BASE_URL_MAP = {
 545         'all': '%s/stream/users/%%s' % SoundcloudPagedPlaylistBaseIE._API_V2_BASE,
 546         'tracks': '%s/users/%%s/tracks' % SoundcloudPagedPlaylistBaseIE._API_V2_BASE,
 547         'albums': '%s/users/%%s/albums' % SoundcloudPagedPlaylistBaseIE._API_V2_BASE,
 548         'sets': '%s/users/%%s/playlists' % SoundcloudPagedPlaylistBaseIE._API_V2_BASE,
 549         'reposts': '%s/stream/users/%%s/reposts' % SoundcloudPagedPlaylistBaseIE._API_V2_BASE,
 550         'likes': '%s/users/%%s/likes' % SoundcloudPagedPlaylistBaseIE._API_V2_BASE,
 551         'spotlight': '%s/users/%%s/spotlight' % SoundcloudPagedPlaylistBaseIE._API_V2_BASE,
 552     }
 553
 554     _TITLE_MAP = {
 555         'all': 'All',
 556         'tracks': 'Tracks',
 557         'albums': 'Albums',
 558         'sets': 'Playlists',
 559         'reposts': 'Reposts',
 560         'likes': 'Likes',
 561         'spotlight': 'Spotlight',
 562     }
 563
 564     def _real_extract(self, url):
 565         mobj = re.match(self._VALID_URL, url)
 566         uploader = mobj.group('user')
 567
 568         url = 'https://soundcloud.com/%s/' % uploader
 569         resolv_url = self._resolv_url(url)
 570         user = self._download_json(
 571             resolv_url, uploader, 'Downloading user info')
 572
 573         resource = mobj.group('rsrc') or 'all'
 574
 575         return self._extract_playlist(
 576             self._BASE_URL_MAP[resource] % user['id'], compat_str(user['id']),
 577             '%s (%s)' % (user['username'], self._TITLE_MAP[resource]))
 578
 579
 580 class SoundcloudTrackStationIE(SoundcloudPagedPlaylistBaseIE):
 581     _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/stations/track/[^/]+/(?P<id>[^/?#&]+)'
 582     IE_NAME = 'soundcloud:trackstation'
 583     _TESTS = [{
 584         'url': 'https://soundcloud.com/stations/track/officialsundial/your-text',
 585         'info_dict': {
 586             'id': '286017854',
 587             'title': 'Track station: your-text',
 588         },
 589         'playlist_mincount': 47,
 590     }]
 591
 592     def _real_extract(self, url):
 593         track_name = self._match_id(url)
 594
 595         webpage = self._download_webpage(url, track_name)
 596
 597         track_id = self._search_regex(
 598             r'soundcloud:track-stations:(\d+)', webpage, 'track id')
 599
 600         return self._extract_playlist(
 601             '%s/stations/soundcloud:track-stations:%s/tracks'
 602             % (self._API_V2_BASE, track_id),
 603             track_id, 'Track station: %s' % track_name)
 604
 605
 606 class SoundcloudPlaylistIE(SoundcloudPlaylistBaseIE):
 607     _VALID_URL = r'https?://api\.soundcloud\.com/playlists/(?P<id>[0-9]+)(?:/?\?secret_token=(?P<token>[^&]+?))?$'
 608     IE_NAME = 'soundcloud:playlist'
 609     _TESTS = [{
 610         'url': 'https://api.soundcloud.com/playlists/4110309',
 611         'info_dict': {
 612             'id': '4110309',
 613             'title': 'TILT Brass - Bowery Poetry Club, August \'03 [Non-Site SCR 02]',
 614             'description': 're:.*?TILT Brass - Bowery Poetry Club',
 615         },
 616         'playlist_count': 6,
 617     }]
 618
 619     def _real_extract(self, url):
 620         mobj = re.match(self._VALID_URL, url)
 621         playlist_id = mobj.group('id')
 622         base_url = '%s//api.soundcloud.com/playlists/%s.json?' % (self.http_scheme(), playlist_id)
 623
 624         data_dict = {
 625             'client_id': self._CLIENT_ID,
 626         }
 627         token = mobj.group('token')
 628
 629         if token:
 630             data_dict['secret_token'] = token
 631
 632         data = compat_urllib_parse_urlencode(data_dict)
 633         data = self._download_json(
 634             base_url + data, playlist_id, 'Downloading playlist')
 635
 636         entries = self._extract_track_entries(data['tracks'])
 637
 638         return {
 639             '_type': 'playlist',
 640             'id': playlist_id,
 641             'title': data.get('title'),
 642             'description': data.get('description'),
 643             'entries': entries,
 644         }
 645
 646
 647 class SoundcloudSearchIE(SearchInfoExtractor, SoundcloudIE):
 648     IE_NAME = 'soundcloud:search'
 649     IE_DESC = 'Soundcloud search'
 650     _MAX_RESULTS = float('inf')
 651     _TESTS = [{
 652         'url': 'scsearch15:post-avant jazzcore',
 653         'info_dict': {
 654             'title': 'post-avant jazzcore',
 655         },
 656         'playlist_count': 15,
 657     }]
 658
 659     _SEARCH_KEY = 'scsearch'
 660     _MAX_RESULTS_PER_PAGE = 200
 661     _DEFAULT_RESULTS_PER_PAGE = 50
 662     _API_V2_BASE = 'https://api-v2.soundcloud.com'
 663
 664     def _get_collection(self, endpoint, collection_id, **query):
 665         limit = min(
 666             query.get('limit', self._DEFAULT_RESULTS_PER_PAGE),
 667             self._MAX_RESULTS_PER_PAGE)
 668         query['limit'] = limit
 669         query['client_id'] = self._CLIENT_ID
 670         query['linked_partitioning'] = '1'
 671         query['offset'] = 0
 672         data = compat_urllib_parse_urlencode(query)
 673         next_url = '{0}{1}?{2}'.format(self._API_V2_BASE, endpoint, data)
 674
 675         collected_results = 0
 676
 677         for i in itertools.count(1):
 678             response = self._download_json(
 679                 next_url, collection_id, 'Downloading page {0}'.format(i),
 680                 'Unable to download API page')
 681
 682             collection = response.get('collection', [])
 683             if not collection:
 684                 break
 685
 686             collection = list(filter(bool, collection))
 687             collected_results += len(collection)
 688
 689             for item in collection:
 690                 yield self.url_result(item['uri'], SoundcloudIE.ie_key())
 691
 692             if not collection or collected_results >= limit:
 693                 break
 694
 695             next_url = response.get('next_href')
 696             if not next_url:
 697                 break
 698
 699     def _get_n_results(self, query, n):
 700         tracks = self._get_collection('/search/tracks', query, limit=n, q=query)
 701         return self.playlist_result(tracks, playlist_title=query)