+ _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\.soundcloud\.com/tracks/(?P<track_id>\d+)
+ (?:/?\?secret_token=(?P<secret_token>[^&]+))?)
+ |(?P<player>(?:w|player|p.)\.soundcloud\.com/player/?.*?url=.*)
+ )
+ '''
+ 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',
+ '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
+ {
+ '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',
+ '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',
+ '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',
+ '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',
+ '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',
+ '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',
+ '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 avaialble 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',
+ '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'],
+ }
+ ]
+
+ _CLIENT_ID = 'BeGVhOrGmfboy1LtiHTQF6Ejpt9ULJCI'
+
+ @staticmethod
+ def _extract_urls(webpage):
+ return [m.group('url') for m in re.finditer(
+ r'<iframe[^>]+src=(["\'])(?P<url>(?:https?://)?(?:w\.)?soundcloud\.com/player.+?)\1',
+ webpage)]
+
+ @classmethod
+ def _resolv_url(cls, url):
+ return 'https://api.soundcloud.com/resolve.json?url=' + url + '&client_id=' + cls._CLIENT_ID
+
+ def _extract_info_dict(self, info, full_title=None, quiet=False, secret_token=None):
+ track_id = compat_str(info['id'])
+ title = info['title']
+ name = full_title or track_id
+ if quiet:
+ self.report_extraction(name)
+ thumbnail = info.get('artwork_url') or info.get('user', {}).get('avatar_url')
+ if isinstance(thumbnail, compat_str):
+ thumbnail = thumbnail.replace('-large', '-t500x500')
+ username = try_get(info, lambda x: x['user']['username'], compat_str)
+
+ def extract_count(key):
+ return int_or_none(info.get('%s_count' % key))
+
+ like_count = extract_count('favoritings')
+ if like_count is None:
+ like_count = extract_count('likes')
+
+ result = {
+ 'id': track_id,
+ 'uploader': username,
+ 'timestamp': unified_timestamp(info.get('created_at')),
+ 'title': title,
+ 'description': info.get('description'),
+ 'thumbnail': thumbnail,
+ 'duration': float_or_none(info.get('duration'), 1000),
+ 'webpage_url': info.get('permalink_url'),
+ 'license': info.get('license'),
+ 'view_count': extract_count('playback'),
+ 'like_count': like_count,
+ 'comment_count': extract_count('comment'),
+ 'repost_count': extract_count('reposts'),
+ 'genre': info.get('genre'),
+ }
+
+ format_urls = set()
+ formats = []
+ query = {'client_id': self._CLIENT_ID}
+ if secret_token is not None:
+ query['secret_token'] = secret_token
+ if info.get('downloadable', False):
+ # We can build a direct link to the song
+ format_url = update_url_query(
+ 'https://api.soundcloud.com/tracks/%s/download' % track_id, query)
+ format_urls.add(format_url)
+ formats.append({
+ 'format_id': 'download',
+ 'ext': info.get('original_format', 'mp3'),
+ 'url': format_url,
+ 'vcodec': 'none',
+ 'preference': 10,
+ })
+
+ # Old API, does not work for some tracks (e.g.
+ # https://soundcloud.com/giovannisarani/mezzo-valzer)
+ format_dict = self._download_json(
+ 'https://api.soundcloud.com/i1/tracks/%s/streams' % track_id,
+ track_id, 'Downloading track url', query=query, fatal=False)
+
+ if format_dict:
+ for key, stream_url in format_dict.items():
+ if stream_url in format_urls:
+ continue
+ format_urls.add(stream_url)
+ ext, abr = 'mp3', None
+ mobj = re.search(r'_([^_]+)_(\d+)_url', key)
+ if mobj:
+ ext, abr = mobj.groups()
+ abr = int(abr)
+ if key.startswith('http'):
+ stream_formats = [{
+ 'format_id': key,
+ 'ext': ext,
+ 'url': stream_url,
+ }]
+ elif key.startswith('rtmp'):
+ # The url doesn't have an rtmp app, we have to extract the playpath
+ url, path = stream_url.split('mp3:', 1)
+ stream_formats = [{
+ 'format_id': key,
+ 'url': url,
+ 'play_path': 'mp3:' + path,
+ 'ext': 'flv',
+ }]
+ elif key.startswith('hls'):
+ stream_formats = self._extract_m3u8_formats(
+ stream_url, track_id, ext, entry_protocol='m3u8_native',
+ m3u8_id=key, fatal=False)
+ else:
+ continue
+
+ if abr:
+ for f in stream_formats:
+ f['abr'] = abr
+
+ formats.extend(stream_formats)
+
+ # 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(
+ update_url_query(format_url, query), track_id, fatal=False)
+ if not isinstance(stream, dict):
+ continue
+ stream_url = url_or_none(stream.get('url'))
+ if not stream_url:
+ continue
+ if stream_url in format_urls:
+ continue
+ format_urls.add(stream_url)
+ protocol = try_get(t, lambda x: x['format']['protocol'], compat_str)
+ 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:
+ mimetype = try_get(
+ t, lambda x: x['format']['mime_type'], compat_str)
+ ext = mimetype2ext(mimetype) or 'mp3'
+ format_id_list = []
+ if protocol:
+ format_id_list.append(protocol)
+ format_id_list.append(ext)
+ format_id = '_'.join(format_id_list)
+ formats.append({
+ 'url': stream_url,
+ 'format_id': format_id,
+ 'ext': ext,
+ 'protocol': 'm3u8_native' if protocol == 'hls' else 'http',
+ })
+
+ 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
+ formats.append({
+ 'format_id': 'fallback',
+ 'url': update_url_query(info['stream_url'], query),
+ 'ext': 'mp3',
+ })
+ self._check_formats(formats, track_id)
+
+ for f in formats:
+ f['vcodec'] = 'none'
+
+ self._sort_formats(formats)
+ result['formats'] = formats
Rapsys Git / youtubedl / blobdiff