2 from __future__ import unicode_literals
11 from ..compat import (
14 compat_urllib_parse_urlencode,
31 class SoundcloudIE(InfoExtractor):
32 """Information extractor for soundcloud.com
33 To access the media, the uid of the song and a stream token
34 must be extracted from the page source and the script must make
35 a request to media.soundcloud.com/crossdomain.xml. Then
36 the media can be grabbed by requesting from an url composed
37 of the stream token and uid
40 _VALID_URL = r'''(?x)^(?:https?://)?
41 (?:(?:(?:www\.|m\.)?soundcloud\.com/
43 (?P<uploader>[\w\d-]+)/
44 (?!(?:tracks|albums|sets(?:/.+?)?|reposts|likes|spotlight)/?(?:$|[?#]))
46 (?P<token>[^?]+?)?(?:[?].*)?$)
47 |(?:api\.soundcloud\.com/tracks/(?P<track_id>\d+)
48 (?:/?\?secret_token=(?P<secret_token>[^&]+))?)
49 |(?P<player>(?:w|player|p.)\.soundcloud\.com/player/?.*?url=.*)
52 IE_NAME = 'soundcloud'
55 'url': 'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
56 'md5': 'ebef0a451b909710ed1d7787dddbf0d7',
60 'title': 'Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1',
61 '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',
62 'uploader': 'E.T. ExTerrestrial Music',
63 'timestamp': 1349920598,
64 'upload_date': '20121011',
66 'license': 'all-rights-reserved',
75 'url': 'https://soundcloud.com/the-concept-band/goldrushed-mastered?in=the-concept-band/sets/the-royal-concept-ep',
79 'title': 'Goldrushed',
80 'description': 'From Stockholm Sweden\r\nPovel / Magnus / Filip / David\r\nwww.theroyalconcept.com',
81 'uploader': 'The Royal Concept',
82 'timestamp': 1337635207,
83 'upload_date': '20120521',
85 'license': 'all-rights-reserved',
93 'skip_download': True,
98 'url': 'https://soundcloud.com/jaimemf/youtube-dl-test-video-a-y-baw/s-8Pjrp',
99 'md5': 'aa0dd32bfea9b0c5ef4f02aacd080604',
103 'title': 'Youtube - Dl Test Video \'\' Ä↭',
104 'description': 'test chars: \"\'/\\ä↭',
105 'uploader': 'jaimeMF',
106 'timestamp': 1386604920,
107 'upload_date': '20131209',
109 'license': 'all-rights-reserved',
112 'comment_count': int,
116 # private link (alt format)
118 'url': 'https://api.soundcloud.com/tracks/123998367?secret_token=s-8Pjrp',
119 'md5': 'aa0dd32bfea9b0c5ef4f02aacd080604',
123 'title': 'Youtube - Dl Test Video \'\' Ä↭',
124 'description': 'test chars: \"\'/\\ä↭',
125 'uploader': 'jaimeMF',
126 'timestamp': 1386604920,
127 'upload_date': '20131209',
129 'license': 'all-rights-reserved',
132 'comment_count': int,
138 'url': 'https://soundcloud.com/oddsamples/bus-brakes',
139 'md5': '7624f2351f8a3b2e7cd51522496e7631',
143 'title': 'Bus Brakes',
144 'description': 'md5:0053ca6396e8d2fd7b7e1595ef12ab66',
145 'uploader': 'oddsamples',
146 'timestamp': 1389232924,
147 'upload_date': '20140109',
149 'license': 'cc-by-sa',
152 'comment_count': int,
156 # private link, downloadable format
158 'url': 'https://soundcloud.com/oriuplift/uponly-238-no-talking-wav/s-AyZUd',
159 'md5': '64a60b16e617d41d0bef032b7f55441e',
163 'title': 'Uplifting Only 238 [No Talking] (incl. Alex Feed Guestmix) (Aug 31, 2017) [wav]',
164 'description': 'md5:fa20ee0fca76a3d6df8c7e57f3715366',
165 'uploader': 'Ori Uplift Music',
166 'timestamp': 1504206263,
167 'upload_date': '20170831',
168 'duration': 7449.096,
169 'license': 'all-rights-reserved',
172 'comment_count': int,
176 # no album art, use avatar pic for thumbnail
178 'url': 'https://soundcloud.com/garyvee/sideways-prod-mad-real',
179 'md5': '59c7872bc44e5d99b7211891664760c2',
183 'title': 'Sideways (Prod. Mad Real)',
184 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
185 'uploader': 'garyvee',
186 'timestamp': 1488152409,
187 'upload_date': '20170226',
189 'thumbnail': r're:https?://.*\.jpg',
190 'license': 'all-rights-reserved',
193 'comment_count': int,
197 'skip_download': True,
200 # not avaialble via api.soundcloud.com/i1/tracks/id/streams
202 'url': 'https://soundcloud.com/giovannisarani/mezzo-valzer',
203 'md5': 'e22aecd2bc88e0e4e432d7dcc0a1abf7',
207 'title': 'Mezzo Valzer',
208 'description': 'md5:4138d582f81866a530317bae316e8b61',
209 'uploader': 'Giovanni Sarani',
210 'timestamp': 1551394171,
211 'upload_date': '20190228',
213 'thumbnail': r're:https?://.*\.jpg',
214 'license': 'all-rights-reserved',
217 'comment_count': int,
220 'expected_warnings': ['Unable to download JSON metadata'],
224 _CLIENT_ID = 'FweeGBOOEOYJWLJN3oEyToGLKhmSz0I7'
227 def _extract_urls(webpage):
228 return [m.group('url') for m in re.finditer(
229 r'<iframe[^>]+src=(["\'])(?P<url>(?:https?://)?(?:w\.)?soundcloud\.com/player.+?)\1',
233 def _resolv_url(cls, url):
234 return 'https://api.soundcloud.com/resolve.json?url=' + url + '&client_id=' + cls._CLIENT_ID
236 def _extract_info_dict(self, info, full_title=None, quiet=False, secret_token=None):
237 track_id = compat_str(info['id'])
238 title = info['title']
239 name = full_title or track_id
241 self.report_extraction(name)
242 thumbnail = info.get('artwork_url') or info.get('user', {}).get('avatar_url')
243 if isinstance(thumbnail, compat_str):
244 thumbnail = thumbnail.replace('-large', '-t500x500')
245 username = try_get(info, lambda x: x['user']['username'], compat_str)
247 def extract_count(key):
248 return int_or_none(info.get('%s_count' % key))
250 like_count = extract_count('favoritings')
251 if like_count is None:
252 like_count = extract_count('likes')
256 'uploader': username,
257 'timestamp': unified_timestamp(info.get('created_at')),
259 'description': info.get('description'),
260 'thumbnail': thumbnail,
261 'duration': float_or_none(info.get('duration'), 1000),
262 'webpage_url': info.get('permalink_url'),
263 'license': info.get('license'),
264 'view_count': extract_count('playback'),
265 'like_count': like_count,
266 'comment_count': extract_count('comment'),
267 'repost_count': extract_count('reposts'),
268 'genre': info.get('genre'),
273 query = {'client_id': self._CLIENT_ID}
274 if secret_token is not None:
275 query['secret_token'] = secret_token
276 if info.get('downloadable', False):
277 # We can build a direct link to the song
278 format_url = update_url_query(
279 'https://api.soundcloud.com/tracks/%s/download' % track_id, query)
280 format_urls.add(format_url)
282 'format_id': 'download',
283 'ext': info.get('original_format', 'mp3'),
289 # Old API, does not work for some tracks (e.g.
290 # https://soundcloud.com/giovannisarani/mezzo-valzer)
291 format_dict = self._download_json(
292 'https://api.soundcloud.com/i1/tracks/%s/streams' % track_id,
293 track_id, 'Downloading track url', query=query, fatal=False)
296 for key, stream_url in format_dict.items():
297 if stream_url in format_urls:
299 format_urls.add(stream_url)
300 ext, abr = 'mp3', None
301 mobj = re.search(r'_([^_]+)_(\d+)_url', key)
303 ext, abr = mobj.groups()
305 if key.startswith('http'):
311 elif key.startswith('rtmp'):
312 # The url doesn't have an rtmp app, we have to extract the playpath
313 url, path = stream_url.split('mp3:', 1)
317 'play_path': 'mp3:' + path,
320 elif key.startswith('hls'):
321 stream_formats = self._extract_m3u8_formats(
322 stream_url, track_id, ext, entry_protocol='m3u8_native',
323 m3u8_id=key, fatal=False)
328 for f in stream_formats:
331 formats.extend(stream_formats)
334 transcodings = try_get(
335 info, lambda x: x['media']['transcodings'], list) or []
336 for t in transcodings:
337 if not isinstance(t, dict):
339 format_url = url_or_none(t.get('url'))
342 stream = self._download_json(
343 update_url_query(format_url, query), track_id, fatal=False)
344 if not isinstance(stream, dict):
346 stream_url = url_or_none(stream.get('url'))
349 if stream_url in format_urls:
351 format_urls.add(stream_url)
352 protocol = try_get(t, lambda x: x['format']['protocol'], compat_str)
353 if protocol != 'hls' and '/hls' in format_url:
356 preset = str_or_none(t.get('preset'))
358 ext = preset.split('_')[0]
359 if ext not in KNOWN_EXTENSIONS:
361 t, lambda x: x['format']['mime_type'], compat_str)
362 ext = mimetype2ext(mimetype) or 'mp3'
365 format_id_list.append(protocol)
366 format_id_list.append(ext)
367 format_id = '_'.join(format_id_list)
370 'format_id': format_id,
372 'protocol': 'm3u8_native' if protocol == 'hls' else 'http',
376 # We fallback to the stream_url in the original info, this
377 # cannot be always used, sometimes it can give an HTTP 404 error
379 'format_id': 'fallback',
380 'url': update_url_query(info['stream_url'], query),
383 self._check_formats(formats, track_id)
388 self._sort_formats(formats)
389 result['formats'] = formats
393 def _real_extract(self, url):
394 mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
396 raise ExtractorError('Invalid URL: %s' % url)
398 track_id = mobj.group('track_id')
401 if track_id is not None:
402 info_json_url = 'https://api.soundcloud.com/tracks/' + track_id + '.json?client_id=' + self._CLIENT_ID
403 full_title = track_id
404 token = mobj.group('secret_token')
406 info_json_url += '&secret_token=' + token
407 elif mobj.group('player'):
408 query = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
409 real_url = query['url'][0]
410 # If the token is in the query of the original url we have to
412 if 'secret_token' in query:
413 real_url += '?secret_token=' + query['secret_token'][0]
414 return self.url_result(real_url)
416 # extract uploader (which is in the url)
417 uploader = mobj.group('uploader')
418 # extract simple title (uploader + slug of song title)
419 slug_title = mobj.group('title')
420 token = mobj.group('token')
421 full_title = resolve_title = '%s/%s' % (uploader, slug_title)
423 resolve_title += '/%s' % token
425 webpage = self._download_webpage(url, full_title, fatal=False)
427 entries = self._parse_json(
429 r'var\s+c\s*=\s*(\[.+?\])\s*,\s*o\s*=Date\b', webpage,
430 'data', default='[]'), full_title, fatal=False)
433 if not isinstance(e, dict):
435 if e.get('id') != 67:
437 data = try_get(e, lambda x: x['data'][0], dict)
441 info_json_url = self._resolv_url(
442 'https://soundcloud.com/%s' % resolve_title)
444 # Contains some additional info missing from new_info
445 info = self._download_json(
446 info_json_url, full_title, 'Downloading info JSON')
448 return self._extract_info_dict(
449 merge_dicts(info, new_info), full_title, secret_token=token)
452 class SoundcloudPlaylistBaseIE(SoundcloudIE):
455 return compat_str(e['id']) if e.get('id') else None
457 def _extract_track_entries(self, tracks):
460 track['permalink_url'], SoundcloudIE.ie_key(),
461 video_id=self._extract_id(track))
462 for track in tracks if track.get('permalink_url')]
465 class SoundcloudSetIE(SoundcloudPlaylistBaseIE):
466 _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/(?P<uploader>[\w\d-]+)/sets/(?P<slug_title>[\w\d-]+)(?:/(?P<token>[^?/]+))?'
467 IE_NAME = 'soundcloud:set'
469 'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep',
472 'title': 'The Royal Concept EP',
474 'playlist_mincount': 5,
476 'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep/token',
477 'only_matching': True,
480 def _real_extract(self, url):
481 mobj = re.match(self._VALID_URL, url)
483 # extract uploader (which is in the url)
484 uploader = mobj.group('uploader')
485 # extract simple title (uploader + slug of song title)
486 slug_title = mobj.group('slug_title')
487 full_title = '%s/sets/%s' % (uploader, slug_title)
488 url = 'https://soundcloud.com/%s/sets/%s' % (uploader, slug_title)
490 token = mobj.group('token')
492 full_title += '/' + token
495 resolv_url = self._resolv_url(url)
496 info = self._download_json(resolv_url, full_title)
499 msgs = (compat_str(err['error_message']) for err in info['errors'])
500 raise ExtractorError('unable to download video webpage: %s' % ','.join(msgs))
502 entries = self._extract_track_entries(info['tracks'])
507 'id': '%s' % info['id'],
508 'title': info['title'],
512 class SoundcloudPagedPlaylistBaseIE(SoundcloudPlaylistBaseIE):
513 _API_V2_BASE = 'https://api-v2.soundcloud.com'
515 def _extract_playlist(self, base_url, playlist_id, playlist_title):
518 'client_id': self._CLIENT_ID,
519 'linked_partitioning': '1',
522 query = COMMON_QUERY.copy()
525 next_href = base_url + '?' + compat_urllib_parse_urlencode(query)
528 for i in itertools.count():
529 response = self._download_json(
530 next_href, playlist_id, 'Downloading track page %s' % (i + 1))
532 collection = response['collection']
534 if not isinstance(collection, list):
537 # Empty collection may be returned, in this case we proceed
538 # straight to next_href
540 def resolve_entry(candidates):
541 for cand in candidates:
542 if not isinstance(cand, dict):
544 permalink_url = url_or_none(cand.get('permalink_url'))
545 if not permalink_url:
547 return self.url_result(
549 ie=SoundcloudIE.ie_key() if SoundcloudIE.suitable(permalink_url) else None,
550 video_id=self._extract_id(cand),
551 video_title=cand.get('title'))
554 entry = resolve_entry((e, e.get('track'), e.get('playlist')))
556 entries.append(entry)
558 next_href = response.get('next_href')
562 parsed_next_href = compat_urlparse.urlparse(response['next_href'])
563 qs = compat_urlparse.parse_qs(parsed_next_href.query)
564 qs.update(COMMON_QUERY)
565 next_href = compat_urlparse.urlunparse(
566 parsed_next_href._replace(query=compat_urllib_parse_urlencode(qs, True)))
571 'title': playlist_title,
576 class SoundcloudUserIE(SoundcloudPagedPlaylistBaseIE):
577 _VALID_URL = r'''(?x)
579 (?:(?:www|m)\.)?soundcloud\.com/
582 (?P<rsrc>tracks|albums|sets|reposts|likes|spotlight)
586 IE_NAME = 'soundcloud:user'
588 'url': 'https://soundcloud.com/soft-cell-official',
591 'title': 'Soft Cell (All)',
593 'playlist_mincount': 28,
595 'url': 'https://soundcloud.com/soft-cell-official/tracks',
598 'title': 'Soft Cell (Tracks)',
600 'playlist_mincount': 27,
602 'url': 'https://soundcloud.com/soft-cell-official/albums',
605 'title': 'Soft Cell (Albums)',
607 'playlist_mincount': 1,
609 'url': 'https://soundcloud.com/jcv246/sets',
612 'title': 'Jordi / cv (Playlists)',
614 'playlist_mincount': 2,
616 'url': 'https://soundcloud.com/jcv246/reposts',
619 'title': 'Jordi / cv (Reposts)',
621 'playlist_mincount': 6,
623 'url': 'https://soundcloud.com/clalberg/likes',
626 'title': 'clalberg (Likes)',
628 'playlist_mincount': 5,
630 'url': 'https://soundcloud.com/grynpyret/spotlight',
633 'title': 'Grynpyret (Spotlight)',
635 'playlist_mincount': 1,
639 'all': '%s/stream/users/%%s' % SoundcloudPagedPlaylistBaseIE._API_V2_BASE,
640 'tracks': '%s/users/%%s/tracks' % SoundcloudPagedPlaylistBaseIE._API_V2_BASE,
641 'albums': '%s/users/%%s/albums' % SoundcloudPagedPlaylistBaseIE._API_V2_BASE,
642 'sets': '%s/users/%%s/playlists' % SoundcloudPagedPlaylistBaseIE._API_V2_BASE,
643 'reposts': '%s/stream/users/%%s/reposts' % SoundcloudPagedPlaylistBaseIE._API_V2_BASE,
644 'likes': '%s/users/%%s/likes' % SoundcloudPagedPlaylistBaseIE._API_V2_BASE,
645 'spotlight': '%s/users/%%s/spotlight' % SoundcloudPagedPlaylistBaseIE._API_V2_BASE,
653 'reposts': 'Reposts',
655 'spotlight': 'Spotlight',
658 def _real_extract(self, url):
659 mobj = re.match(self._VALID_URL, url)
660 uploader = mobj.group('user')
662 url = 'https://soundcloud.com/%s/' % uploader
663 resolv_url = self._resolv_url(url)
664 user = self._download_json(
665 resolv_url, uploader, 'Downloading user info')
667 resource = mobj.group('rsrc') or 'all'
669 return self._extract_playlist(
670 self._BASE_URL_MAP[resource] % user['id'], compat_str(user['id']),
671 '%s (%s)' % (user['username'], self._TITLE_MAP[resource]))
674 class SoundcloudTrackStationIE(SoundcloudPagedPlaylistBaseIE):
675 _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/stations/track/[^/]+/(?P<id>[^/?#&]+)'
676 IE_NAME = 'soundcloud:trackstation'
678 'url': 'https://soundcloud.com/stations/track/officialsundial/your-text',
681 'title': 'Track station: your-text',
683 'playlist_mincount': 47,
686 def _real_extract(self, url):
687 track_name = self._match_id(url)
689 webpage = self._download_webpage(url, track_name)
691 track_id = self._search_regex(
692 r'soundcloud:track-stations:(\d+)', webpage, 'track id')
694 return self._extract_playlist(
695 '%s/stations/soundcloud:track-stations:%s/tracks'
696 % (self._API_V2_BASE, track_id),
697 track_id, 'Track station: %s' % track_name)
700 class SoundcloudPlaylistIE(SoundcloudPlaylistBaseIE):
701 _VALID_URL = r'https?://api\.soundcloud\.com/playlists/(?P<id>[0-9]+)(?:/?\?secret_token=(?P<token>[^&]+?))?$'
702 IE_NAME = 'soundcloud:playlist'
704 'url': 'https://api.soundcloud.com/playlists/4110309',
707 'title': 'TILT Brass - Bowery Poetry Club, August \'03 [Non-Site SCR 02]',
708 'description': 're:.*?TILT Brass - Bowery Poetry Club',
713 def _real_extract(self, url):
714 mobj = re.match(self._VALID_URL, url)
715 playlist_id = mobj.group('id')
716 base_url = '%s//api.soundcloud.com/playlists/%s.json?' % (self.http_scheme(), playlist_id)
719 'client_id': self._CLIENT_ID,
721 token = mobj.group('token')
724 data_dict['secret_token'] = token
726 data = compat_urllib_parse_urlencode(data_dict)
727 data = self._download_json(
728 base_url + data, playlist_id, 'Downloading playlist')
730 entries = self._extract_track_entries(data['tracks'])
735 'title': data.get('title'),
736 'description': data.get('description'),
741 class SoundcloudSearchIE(SearchInfoExtractor, SoundcloudIE):
742 IE_NAME = 'soundcloud:search'
743 IE_DESC = 'Soundcloud search'
744 _MAX_RESULTS = float('inf')
746 'url': 'scsearch15:post-avant jazzcore',
748 'title': 'post-avant jazzcore',
750 'playlist_count': 15,
753 _SEARCH_KEY = 'scsearch'
754 _MAX_RESULTS_PER_PAGE = 200
755 _DEFAULT_RESULTS_PER_PAGE = 50
756 _API_V2_BASE = 'https://api-v2.soundcloud.com'
758 def _get_collection(self, endpoint, collection_id, **query):
760 query.get('limit', self._DEFAULT_RESULTS_PER_PAGE),
761 self._MAX_RESULTS_PER_PAGE)
762 query['limit'] = limit
763 query['client_id'] = self._CLIENT_ID
764 query['linked_partitioning'] = '1'
766 data = compat_urllib_parse_urlencode(query)
767 next_url = '{0}{1}?{2}'.format(self._API_V2_BASE, endpoint, data)
769 collected_results = 0
771 for i in itertools.count(1):
772 response = self._download_json(
773 next_url, collection_id, 'Downloading page {0}'.format(i),
774 'Unable to download API page')
776 collection = response.get('collection', [])
780 collection = list(filter(bool, collection))
781 collected_results += len(collection)
783 for item in collection:
784 yield self.url_result(item['uri'], SoundcloudIE.ie_key())
786 if not collection or collected_results >= limit:
789 next_url = response.get('next_href')
793 def _get_n_results(self, query, n):
794 tracks = self._get_collection('/search/tracks', query, limit=n, q=query)
795 return self.playlist_result(tracks, playlist_title=query)