+ _VALID_URL = r'''(?x)^(?:https?://)?
+ (?:(?:(?:www\.|m\.)?soundcloud\.com/
+ (?!stations/track)
+ (?P<uploader>[\w\d-]+)/
+ (?!(?:tracks|albums|sets(?:/.+?)?|reposts|likes|spotlight)/?(?:$|[?#]))
+ (?P<title>[\w\d-]+)/?
+ (?P<token>[^?]+?)?(?:[?].*)?$)
+ |(?:api(?:-v2)?\.soundcloud\.com/tracks/(?P<track_id>\d+)
+ (?:/?\?secret_token=(?P<secret_token>[^&]+))?)
+ )
+ '''
+ IE_NAME = 'soundcloud'
+ _TESTS = [
+ {
+ 'url': 'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
+ 'md5': 'ebef0a451b909710ed1d7787dddbf0d7',
+ 'info_dict': {
+ 'id': '62986583',
+ 'ext': 'mp3',
+ 'title': 'Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1',
+ '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',
+ 'uploader': 'E.T. ExTerrestrial Music',
+ 'uploader_id': '1571244',
+ 'timestamp': 1349920598,
+ 'upload_date': '20121011',
+ 'duration': 143.216,
+ 'license': 'all-rights-reserved',
+ 'view_count': int,
+ 'like_count': int,
+ 'comment_count': int,
+ 'repost_count': int,
+ }
+ },
+ # not streamable song, preview
+ {
+ 'url': 'https://soundcloud.com/the-concept-band/goldrushed-mastered?in=the-concept-band/sets/the-royal-concept-ep',
+ 'info_dict': {
+ 'id': '47127627',
+ 'ext': 'mp3',
+ 'title': 'Goldrushed',
+ 'description': 'From Stockholm Sweden\r\nPovel / Magnus / Filip / David\r\nwww.theroyalconcept.com',
+ 'uploader': 'The Royal Concept',
+ 'uploader_id': '9615865',
+ 'timestamp': 1337635207,
+ 'upload_date': '20120521',
+ 'duration': 30,
+ 'license': 'all-rights-reserved',
+ 'view_count': int,
+ 'like_count': int,
+ 'comment_count': int,
+ 'repost_count': int,
+ },
+ 'params': {
+ # rtmp
+ 'skip_download': True,
+ },
+ },
+ # private link
+ {
+ 'url': 'https://soundcloud.com/jaimemf/youtube-dl-test-video-a-y-baw/s-8Pjrp',
+ 'md5': 'aa0dd32bfea9b0c5ef4f02aacd080604',
+ 'info_dict': {
+ 'id': '123998367',
+ 'ext': 'mp3',
+ 'title': 'Youtube - Dl Test Video \'\' Ä↭',
+ 'description': 'test chars: \"\'/\\ä↭',
+ 'uploader': 'jaimeMF',
+ 'uploader_id': '69767071',
+ 'timestamp': 1386604920,
+ 'upload_date': '20131209',
+ 'duration': 9.927,
+ 'license': 'all-rights-reserved',
+ 'view_count': int,
+ 'like_count': int,
+ 'comment_count': int,
+ 'repost_count': int,
+ },
+ },
+ # private link (alt format)
+ {
+ 'url': 'https://api.soundcloud.com/tracks/123998367?secret_token=s-8Pjrp',
+ 'md5': 'aa0dd32bfea9b0c5ef4f02aacd080604',
+ 'info_dict': {
+ 'id': '123998367',
+ 'ext': 'mp3',
+ 'title': 'Youtube - Dl Test Video \'\' Ä↭',
+ 'description': 'test chars: \"\'/\\ä↭',
+ 'uploader': 'jaimeMF',
+ 'uploader_id': '69767071',
+ 'timestamp': 1386604920,
+ 'upload_date': '20131209',
+ 'duration': 9.927,
+ 'license': 'all-rights-reserved',
+ 'view_count': int,
+ 'like_count': int,
+ 'comment_count': int,
+ 'repost_count': int,
+ },
+ },
+ # downloadable song
+ {
+ 'url': 'https://soundcloud.com/oddsamples/bus-brakes',
+ 'md5': '7624f2351f8a3b2e7cd51522496e7631',
+ 'info_dict': {
+ 'id': '128590877',
+ 'ext': 'mp3',
+ 'title': 'Bus Brakes',
+ 'description': 'md5:0053ca6396e8d2fd7b7e1595ef12ab66',
+ 'uploader': 'oddsamples',
+ 'uploader_id': '73680509',
+ 'timestamp': 1389232924,
+ 'upload_date': '20140109',
+ 'duration': 17.346,
+ 'license': 'cc-by-sa',
+ 'view_count': int,
+ 'like_count': int,
+ 'comment_count': int,
+ 'repost_count': int,
+ },
+ },
+ # private link, downloadable format
+ {
+ 'url': 'https://soundcloud.com/oriuplift/uponly-238-no-talking-wav/s-AyZUd',
+ 'md5': '64a60b16e617d41d0bef032b7f55441e',
+ 'info_dict': {
+ 'id': '340344461',
+ 'ext': 'wav',
+ 'title': 'Uplifting Only 238 [No Talking] (incl. Alex Feed Guestmix) (Aug 31, 2017) [wav]',
+ 'description': 'md5:fa20ee0fca76a3d6df8c7e57f3715366',
+ 'uploader': 'Ori Uplift Music',
+ 'uploader_id': '12563093',
+ 'timestamp': 1504206263,
+ 'upload_date': '20170831',
+ 'duration': 7449.096,
+ 'license': 'all-rights-reserved',
+ 'view_count': int,
+ 'like_count': int,
+ 'comment_count': int,
+ 'repost_count': int,
+ },
+ },
+ # no album art, use avatar pic for thumbnail
+ {
+ 'url': 'https://soundcloud.com/garyvee/sideways-prod-mad-real',
+ 'md5': '59c7872bc44e5d99b7211891664760c2',
+ 'info_dict': {
+ 'id': '309699954',
+ 'ext': 'mp3',
+ 'title': 'Sideways (Prod. Mad Real)',
+ 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
+ 'uploader': 'garyvee',
+ 'uploader_id': '2366352',
+ 'timestamp': 1488152409,
+ 'upload_date': '20170226',
+ 'duration': 207.012,
+ 'thumbnail': r're:https?://.*\.jpg',
+ 'license': 'all-rights-reserved',
+ 'view_count': int,
+ 'like_count': int,
+ 'comment_count': int,
+ 'repost_count': int,
+ },
+ 'params': {
+ 'skip_download': True,
+ },
+ },
+ # not available via api.soundcloud.com/i1/tracks/id/streams
+ {
+ 'url': 'https://soundcloud.com/giovannisarani/mezzo-valzer',
+ 'md5': 'e22aecd2bc88e0e4e432d7dcc0a1abf7',
+ 'info_dict': {
+ 'id': '583011102',
+ 'ext': 'mp3',
+ 'title': 'Mezzo Valzer',
+ 'description': 'md5:4138d582f81866a530317bae316e8b61',
+ 'uploader': 'Giovanni Sarani',
+ 'uploader_id': '3352531',
+ 'timestamp': 1551394171,
+ 'upload_date': '20190228',
+ 'duration': 180.157,
+ 'thumbnail': r're:https?://.*\.jpg',
+ 'license': 'all-rights-reserved',
+ 'view_count': int,
+ 'like_count': int,
+ 'comment_count': int,
+ 'repost_count': int,
+ },
+ 'expected_warnings': ['Unable to download JSON metadata'],
+ }
+ ]
+
+ _API_BASE = 'https://api.soundcloud.com/'
+ _API_V2_BASE = 'https://api-v2.soundcloud.com/'
+ _BASE_URL = 'https://soundcloud.com/'
+ _IMAGE_REPL_RE = r'-([0-9a-z]+)\.jpg'
+
+ _ARTWORK_MAP = {
+ 'mini': 16,
+ 'tiny': 20,
+ 'small': 32,
+ 'badge': 47,
+ 't67x67': 67,
+ 'large': 100,
+ 't300x300': 300,
+ 'crop': 400,
+ 't500x500': 500,
+ 'original': 0,
+ }
+
+ def _store_client_id(self, client_id):
+ self._downloader.cache.store('soundcloud', 'client_id', client_id)
+
+ def _update_client_id(self):
+ webpage = self._download_webpage('https://soundcloud.com/', None)
+ for src in reversed(re.findall(r'<script[^>]+src="([^"]+)"', webpage)):
+ script = self._download_webpage(src, None, fatal=False)
+ if script:
+ client_id = self._search_regex(
+ r'client_id\s*:\s*"([0-9a-zA-Z]{32})"',
+ script, 'client id', default=None)
+ if client_id:
+ self._CLIENT_ID = client_id
+ self._store_client_id(client_id)
+ return
+ raise ExtractorError('Unable to extract client id')
+
+ def _download_json(self, *args, **kwargs):
+ non_fatal = kwargs.get('fatal') is False
+ if non_fatal:
+ del kwargs['fatal']
+ query = kwargs.get('query', {}).copy()
+ for _ in range(2):
+ query['client_id'] = self._CLIENT_ID
+ kwargs['query'] = query
+ try:
+ return super(SoundcloudIE, self)._download_json(*args, **compat_kwargs(kwargs))
+ except ExtractorError as e:
+ if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
+ self._store_client_id(None)
+ self._update_client_id()
+ continue
+ elif non_fatal:
+ self._downloader.report_warning(error_to_compat_str(e))
+ return False
+ raise
+
+ def _real_initialize(self):
+ self._CLIENT_ID = self._downloader.cache.load('soundcloud', 'client_id') or 'YUKXoArFcqrlQn9tfNHvvyfnDISj04zk'
+
+ @classmethod
+ def _resolv_url(cls, url):
+ return SoundcloudIE._API_V2_BASE + 'resolve?url=' + url
+
+ def _extract_info_dict(self, info, full_title=None, secret_token=None, version=2):
+ track_id = compat_str(info['id'])
+ title = info['title']
+ track_base_url = self._API_BASE + 'tracks/%s' % track_id
+
+ format_urls = set()
+ formats = []
+ query = {'client_id': self._CLIENT_ID}
+ if secret_token:
+ query['secret_token'] = secret_token
+
+ if info.get('downloadable') and info.get('has_downloads_left'):
+ format_url = update_url_query(
+ info.get('download_url') or track_base_url + '/download', query)
+ format_urls.add(format_url)
+ if version == 2:
+ v1_info = self._download_json(
+ track_base_url, track_id, query=query, fatal=False) or {}
+ else:
+ v1_info = info
+ formats.append({
+ 'format_id': 'download',
+ 'ext': v1_info.get('original_format') or 'mp3',
+ 'filesize': int_or_none(v1_info.get('original_content_size')),
+ 'url': format_url,
+ 'preference': 10,
+ })
+
+ def invalid_url(url):
+ return not url or url in format_urls
+
+ def add_format(f, protocol, is_preview=False):
+ mobj = re.search(r'\.(?P<abr>\d+)\.(?P<ext>[0-9a-z]{3,4})(?=[/?])', stream_url)
+ if mobj:
+ for k, v in mobj.groupdict().items():
+ if not f.get(k):
+ f[k] = v
+ format_id_list = []
+ if protocol:
+ format_id_list.append(protocol)
+ for k in ('ext', 'abr'):
+ v = f.get(k)
+ if v:
+ format_id_list.append(v)
+ preview = is_preview or re.search(r'/(?:preview|playlist)/0/30/', f['url'])
+ if preview:
+ format_id_list.append('preview')
+ abr = f.get('abr')
+ if abr:
+ f['abr'] = int(abr)
+ f.update({
+ 'format_id': '_'.join(format_id_list),
+ 'protocol': 'm3u8_native' if protocol == 'hls' else 'http',
+ 'preference': -10 if preview else None,
+ })
+ formats.append(f)
+
+ # New API
+ transcodings = try_get(
+ info, lambda x: x['media']['transcodings'], list) or []
+ for t in transcodings:
+ if not isinstance(t, dict):
+ continue
+ format_url = url_or_none(t.get('url'))
+ if not format_url:
+ continue
+ stream = self._download_json(
+ format_url, track_id, query=query, fatal=False)
+ if not isinstance(stream, dict):
+ continue
+ stream_url = url_or_none(stream.get('url'))
+ if invalid_url(stream_url):
+ continue
+ format_urls.add(stream_url)
+ stream_format = t.get('format') or {}
+ protocol = stream_format.get('protocol')
+ if protocol != 'hls' and '/hls' in format_url:
+ protocol = 'hls'
+ ext = None
+ preset = str_or_none(t.get('preset'))
+ if preset:
+ ext = preset.split('_')[0]
+ if ext not in KNOWN_EXTENSIONS:
+ ext = mimetype2ext(stream_format.get('mime_type'))
+ add_format({
+ 'url': stream_url,
+ 'ext': ext,
+ }, 'http' if protocol == 'progressive' else protocol,
+ t.get('snipped') or '/preview/' in format_url)
+
+ if not formats:
+ # Old API, does not work for some tracks (e.g.
+ # https://soundcloud.com/giovannisarani/mezzo-valzer)
+ # and might serve preview URLs (e.g.
+ # http://www.soundcloud.com/snbrn/ele)
+ format_dict = self._download_json(
+ track_base_url + '/streams', track_id,
+ 'Downloading track url', query=query, fatal=False) or {}
+
+ for key, stream_url in format_dict.items():
+ if invalid_url(stream_url):
+ continue
+ format_urls.add(stream_url)
+ mobj = re.search(r'(http|hls)_([^_]+)_(\d+)_url', key)
+ if mobj:
+ protocol, ext, abr = mobj.groups()
+ add_format({
+ 'abr': abr,
+ 'ext': ext,
+ 'url': stream_url,
+ }, protocol)
+
+ if not formats:
+ # We fallback to the stream_url in the original info, this
+ # cannot be always used, sometimes it can give an HTTP 404 error
+ urlh = self._request_webpage(
+ HEADRequest(info.get('stream_url') or track_base_url + '/stream'),
+ track_id, query=query, fatal=False)
+ if urlh:
+ stream_url = urlh.geturl()
+ if not invalid_url(stream_url):
+ add_format({'url': stream_url}, 'http')
+
+ for f in formats:
+ f['vcodec'] = 'none'
+
+ self._sort_formats(formats)
+
+ user = info.get('user') or {}
Rapsys Git / youtubedl / blobdiff