+import functools
import itertools
-import re
import json
-# import random
+import re
-from .common import (
- InfoExtractor,
- SearchInfoExtractor
-)
-from ..compat import compat_str
-from ..networking import HEADRequest, Request
+from .common import InfoExtractor, SearchInfoExtractor
+from ..networking import HEADRequest
from ..networking.exceptions import HTTPError
from ..utils import (
- error_to_compat_str,
+ KNOWN_EXTENSIONS,
ExtractorError,
float_or_none,
int_or_none,
- KNOWN_EXTENSIONS,
+ join_nonempty,
mimetype2ext,
parse_qs,
str_or_none,
+ try_call,
unified_timestamp,
update_url_query,
url_or_none,
_API_AUTH_QUERY_TEMPLATE = '?client_id=%s'
_API_AUTH_URL_PW = 'https://api-auth.soundcloud.com/web-auth/sign-in/password%s'
_API_VERIFY_AUTH_TOKEN = 'https://api-auth.soundcloud.com/connect/session%s'
- _access_token = None
_HEADERS = {}
_IMAGE_REPL_RE = r'-([0-9a-z]+)\.jpg'
'original': 0,
}
+ _DEFAULT_FORMATS = ['http_aac', 'hls_aac', 'http_opus', 'hls_opus', 'http_mp3', 'hls_mp3']
+
+ @functools.cached_property
+ def _is_requested(self):
+ return re.compile(r'|'.join(set(
+ re.escape(pattern).replace(r'\*', r'.*') if pattern != 'default'
+ else '|'.join(map(re.escape, self._DEFAULT_FORMATS))
+ for pattern in self._configuration_arg('formats', ['default'], ie_key=SoundcloudIE)
+ ))).fullmatch
+
def _store_client_id(self, client_id):
self.cache.store('soundcloud', 'client_id', client_id)
return
raise ExtractorError('Unable to extract client id')
- def _download_json(self, *args, **kwargs):
+ def _call_api(self, *args, **kwargs):
non_fatal = kwargs.get('fatal') is False
if non_fatal:
del kwargs['fatal']
query['client_id'] = self._CLIENT_ID
kwargs['query'] = query
try:
- return super()._download_json(*args, **kwargs)
+ return self._download_json(*args, **kwargs)
except ExtractorError as e:
if isinstance(e.cause, HTTPError) and e.cause.status in (401, 403):
self._store_client_id(None)
self._update_client_id()
continue
elif non_fatal:
- self.report_warning(error_to_compat_str(e))
+ self.report_warning(str(e))
return False
raise
def _initialize_pre_login(self):
self._CLIENT_ID = self.cache.load('soundcloud', 'client_id') or 'a3e059563d7fd3372b49b37f00a00bcf'
+ def _verify_oauth_token(self, token):
+ if self._request_webpage(
+ self._API_VERIFY_AUTH_TOKEN % (self._API_AUTH_QUERY_TEMPLATE % self._CLIENT_ID),
+ None, note='Verifying login token...', fatal=False,
+ data=json.dumps({'session': {'access_token': token}}).encode()):
+ self._HEADERS['Authorization'] = f'OAuth {token}'
+ self.report_login()
+ else:
+ self.report_warning('Provided authorization token is invalid. Continuing as guest')
+
+ def _real_initialize(self):
+ if self._HEADERS:
+ return
+ if token := try_call(lambda: self._get_cookies(self._BASE_URL)['oauth_token'].value):
+ self._verify_oauth_token(token)
+
def _perform_login(self, username, password):
if username != 'oauth':
- self.report_warning(
+ raise ExtractorError(
'Login using username and password is not currently supported. '
- 'Use "--username oauth --password <oauth_token>" to login using an oauth token')
- self._access_token = password
- query = self._API_AUTH_QUERY_TEMPLATE % self._CLIENT_ID
- payload = {'session': {'access_token': self._access_token}}
- token_verification = Request(self._API_VERIFY_AUTH_TOKEN % query, json.dumps(payload).encode('utf-8'))
- response = self._download_json(token_verification, None, note='Verifying login token...', fatal=False)
- if response is not False:
- self._HEADERS = {'Authorization': 'OAuth ' + self._access_token}
- self.report_login()
- else:
- self.report_warning('Provided authorization token seems to be invalid. Continue as guest')
+ 'Use "--username oauth --password <oauth_token>" to login using an oauth token, '
+ f'or else {self._login_hint(method="cookies")}', expected=True)
+ if self._HEADERS:
+ return
+ self._verify_oauth_token(password)
r'''
def genDevId():
'user_agent': self._USER_AGENT
}
- query = self._API_AUTH_QUERY_TEMPLATE % self._CLIENT_ID
- login = sanitized_Request(self._API_AUTH_URL_PW % query, json.dumps(payload).encode('utf-8'))
- response = self._download_json(login, None)
- self._access_token = response.get('session').get('access_token')
- if not self._access_token:
- self.report_warning('Unable to get access token, login may has failed')
- else:
- self._HEADERS = {'Authorization': 'OAuth ' + self._access_token}
+ response = self._call_api(
+ self._API_AUTH_URL_PW % (self._API_AUTH_QUERY_TEMPLATE % self._CLIENT_ID),
+ None, note='Verifying login token...', fatal=False,
+ data=json.dumps(payload).encode())
+
+ if token := traverse_obj(response, ('session', 'access_token', {str})):
+ self._HEADERS['Authorization'] = f'OAuth {token}'
+ self.report_login()
+ return
+
+ raise ExtractorError('Unable to get access token, login may have failed', expected=True)
'''
# signature generation
t = clid # _CLIENT_ID
d = '-'.join([str(mInt) for mInt in [a, i, s, w, u, l, b, k]])
- p = n + y + d + r + e + t + d + n
- h = p
+ h = n + y + d + r + e + t + d + n
m = 8011470
- f = 0
- for f in range(f, len(h)):
+ for f in range(len(h)):
m = (m >> 1) + ((1 & m) << 23)
m += ord(h[f])
m &= 16777215
# c is not even needed
- out = str(y) + ':' + str(d) + ':' + format(m, 'x') + ':' + str(c)
-
- return out
+ return f'{y}:{d}:{m:x}:{c}'
def _extract_info_dict(self, info, full_title=None, secret_token=None, extract_flat=False):
- track_id = compat_str(info['id'])
+ track_id = str(info['id'])
title = info['title']
format_urls = set()
query['secret_token'] = secret_token
if not extract_flat and info.get('downloadable') and info.get('has_downloads_left'):
- download_url = update_url_query(
- self._API_V2_BASE + 'tracks/' + track_id + '/download', query)
- redirect_url = (self._download_json(download_url, track_id, fatal=False) or {}).get('redirectUri')
- if redirect_url:
+ try:
+ # Do not use _call_api(); HTTP Error codes have different meanings for this request
+ download_data = self._download_json(
+ f'{self._API_V2_BASE}tracks/{track_id}/download', track_id,
+ 'Downloading original download format info JSON', query=query, headers=self._HEADERS)
+ except ExtractorError as e:
+ if isinstance(e.cause, HTTPError) and e.cause.status == 401:
+ self.report_warning(
+ 'Original download format is only available '
+ f'for registered users. {self._login_hint()}')
+ elif isinstance(e.cause, HTTPError) and e.cause.status == 403:
+ self.write_debug('Original download format is not available for this client')
+ else:
+ self.report_warning(e.msg)
+ download_data = None
+
+ if redirect_url := traverse_obj(download_data, ('redirectUri', {url_or_none})):
urlh = self._request_webpage(
- HEADRequest(redirect_url), track_id, fatal=False)
+ HEADRequest(redirect_url), track_id, 'Checking original download format availability',
+ 'Original download format is not available', fatal=False)
if urlh:
format_url = urlh.url
format_urls.add(format_url)
abr = f.get('abr')
if abr:
f['abr'] = int(abr)
- if protocol == 'hls':
+ if protocol in ('hls', 'hls-aes'):
protocol = 'm3u8' if ext == 'aac' else 'm3u8_native'
else:
protocol = 'http'
if extract_flat:
break
format_url = t['url']
- stream = None
+ protocol = traverse_obj(t, ('format', 'protocol', {str}))
+ if protocol == 'progressive':
+ protocol = 'http'
+ if protocol != 'hls' and '/hls' in format_url:
+ protocol = 'hls'
+ if protocol == 'encrypted-hls' or '/encrypted-hls' in format_url:
+ protocol = 'hls-aes'
+
+ ext = None
+ if preset := traverse_obj(t, ('preset', {str_or_none})):
+ ext = preset.split('_')[0]
+ if ext not in KNOWN_EXTENSIONS:
+ ext = mimetype2ext(traverse_obj(t, ('format', 'mime_type', {str})))
+
+ identifier = join_nonempty(protocol, ext, delim='_')
+ if not self._is_requested(identifier):
+ self.write_debug(f'"{identifier}" is not a requested format, skipping')
+ continue
+
+ stream = None
for retry in self.RetryManager(fatal=False):
try:
- stream = self._download_json(format_url, track_id, query=query, headers=self._HEADERS)
+ stream = self._call_api(
+ format_url, track_id, f'Downloading {identifier} format info JSON',
+ query=query, headers=self._HEADERS)
except ExtractorError as e:
if isinstance(e.cause, HTTPError) and e.cause.status == 429:
self.report_warning(
else:
self.report_warning(e.msg)
- if not isinstance(stream, dict):
- continue
- stream_url = url_or_none(stream.get('url'))
+ stream_url = traverse_obj(stream, ('url', {url_or_none}))
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)
+ }, protocol, t.get('snipped') or '/preview/' in format_url)
for f in formats:
f['vcodec'] = 'none'
thumbnails = []
artwork_url = info.get('artwork_url')
thumbnail = artwork_url or user.get('avatar_url')
- if isinstance(thumbnail, compat_str):
+ if isinstance(thumbnail, str):
if re.search(self._IMAGE_REPL_RE, thumbnail):
for image_id, size in self._ARTWORK_MAP.items():
i = {
'id': image_id,
- 'url': re.sub(self._IMAGE_REPL_RE, '-%s.jpg' % image_id, thumbnail),
+ 'url': re.sub(self._IMAGE_REPL_RE, f'-{image_id}.jpg', thumbnail),
}
if image_id == 'tiny' and not artwork_url:
size = 18
thumbnails = [{'url': thumbnail}]
def extract_count(key):
- return int_or_none(info.get('%s_count' % key))
+ return int_or_none(info.get(f'{key}_count'))
return {
'id': track_id,
'like_count': extract_count('favoritings') or extract_count('likes'),
'comment_count': extract_count('comment'),
'repost_count': extract_count('reposts'),
- 'genre': info.get('genre'),
- 'formats': formats if not extract_flat else None
+ 'genres': traverse_obj(info, ('genre', {str}, {lambda x: x or None}, all)),
+ 'formats': formats if not extract_flat else None,
}
@classmethod
_TESTS = [
{
'url': 'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
- 'md5': 'ebef0a451b909710ed1d7787dddbf0d7',
+ 'md5': 'de9bac153e7427a7333b4b0c1b6a18d2',
'info_dict': {
'id': '62986583',
- 'ext': 'mp3',
+ 'ext': 'opus',
'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',
'like_count': int,
'comment_count': int,
'repost_count': int,
- }
+ 'thumbnail': 'https://i1.sndcdn.com/artworks-000031955188-rwb18x-original.jpg',
+ 'uploader_url': 'https://soundcloud.com/ethmusic',
+ 'genres': [],
+ },
},
# geo-restricted
{
'url': 'https://soundcloud.com/the-concept-band/goldrushed-mastered?in=the-concept-band/sets/the-royal-concept-ep',
'info_dict': {
'id': '47127627',
- 'ext': 'mp3',
+ 'ext': 'opus',
'title': 'Goldrushed',
'description': 'From Stockholm Sweden\r\nPovel / Magnus / Filip / David\r\nwww.theroyalconcept.com',
'uploader': 'The Royal Concept',
'like_count': int,
'comment_count': int,
'repost_count': int,
+ 'uploader_url': 'https://soundcloud.com/the-concept-band',
+ 'thumbnail': 'https://i1.sndcdn.com/artworks-v8bFHhXm7Au6-0-original.jpg',
+ 'genres': ['Alternative'],
},
},
# private link
'id': '123998367',
'ext': 'mp3',
'title': 'Youtube - Dl Test Video \'\' Ä↭',
- 'description': 'test chars: \"\'/\\ä↭',
+ 'description': 'test chars: "\'/\\ä↭',
'uploader': 'jaimeMF',
'uploader_id': '69767071',
'timestamp': 1386604920,
'like_count': int,
'comment_count': int,
'repost_count': int,
+ 'uploader_url': 'https://soundcloud.com/jaimemf',
+ 'thumbnail': 'https://a1.sndcdn.com/images/default_avatar_large.png',
+ 'genres': ['youtubedl'],
},
},
# private link (alt format)
'id': '123998367',
'ext': 'mp3',
'title': 'Youtube - Dl Test Video \'\' Ä↭',
- 'description': 'test chars: \"\'/\\ä↭',
+ 'description': 'test chars: "\'/\\ä↭',
'uploader': 'jaimeMF',
'uploader_id': '69767071',
'timestamp': 1386604920,
'like_count': int,
'comment_count': int,
'repost_count': int,
+ 'uploader_url': 'https://soundcloud.com/jaimemf',
+ 'thumbnail': 'https://a1.sndcdn.com/images/default_avatar_large.png',
+ 'genres': ['youtubedl'],
},
},
# downloadable song
'info_dict': {
'id': '343609555',
'ext': 'wav',
+ 'title': 'The Following',
+ 'description': '',
+ 'uploader': '80M',
+ 'uploader_id': '312384765',
+ 'uploader_url': 'https://soundcloud.com/the80m',
+ 'upload_date': '20170922',
+ 'timestamp': 1506120436,
+ 'duration': 397.228,
+ 'thumbnail': 'https://i1.sndcdn.com/artworks-000243916348-ktoo7d-original.jpg',
+ 'license': 'all-rights-reserved',
+ 'like_count': int,
+ 'comment_count': int,
+ 'repost_count': int,
+ 'view_count': int,
+ 'genres': ['Dance & EDM'],
},
},
# private link, downloadable format
'like_count': int,
'comment_count': int,
'repost_count': int,
+ 'thumbnail': 'https://i1.sndcdn.com/artworks-000240712245-kedn4p-original.jpg',
+ 'uploader_url': 'https://soundcloud.com/oriuplift',
+ 'genres': ['Trance'],
},
},
# no album art, use avatar pic for thumbnail
'like_count': int,
'comment_count': int,
'repost_count': int,
+ 'uploader_url': 'https://soundcloud.com/garyvee',
+ 'genres': [],
},
'params': {
'skip_download': True,
},
{
'url': 'https://soundcloud.com/giovannisarani/mezzo-valzer',
- 'md5': 'e22aecd2bc88e0e4e432d7dcc0a1abf7',
+ 'md5': '8227c3473a4264df6b02ad7e5b7527ac',
'info_dict': {
'id': '583011102',
- 'ext': 'mp3',
+ 'ext': 'opus',
'title': 'Mezzo Valzer',
- 'description': 'md5:4138d582f81866a530317bae316e8b61',
- 'uploader': 'Micronie',
+ 'description': 'md5:f4d5f39d52e0ccc2b4f665326428901a',
+ 'uploader': 'Giovanni Sarani',
'uploader_id': '3352531',
'timestamp': 1551394171,
'upload_date': '20190228',
'like_count': int,
'comment_count': int,
'repost_count': int,
+ 'genres': ['Piano'],
+ 'uploader_url': 'https://soundcloud.com/giovannisarani',
},
},
{
if token:
query['secret_token'] = token
else:
- full_title = resolve_title = '%s/%s' % mobj.group('uploader', 'title')
+ full_title = resolve_title = '{}/{}'.format(*mobj.group('uploader', 'title'))
token = mobj.group('token')
if token:
- resolve_title += '/%s' % token
+ resolve_title += f'/{token}'
info_json_url = self._resolv_url(self._BASE_URL + resolve_title)
- info = self._download_json(
+ info = self._call_api(
info_json_url, full_title, 'Downloading info JSON', query=query, headers=self._HEADERS)
return self._extract_info_dict(info, full_title, token)
class SoundcloudPlaylistBaseIE(SoundcloudBaseIE):
def _extract_set(self, playlist, token=None):
- playlist_id = compat_str(playlist['id'])
+ playlist_id = str(playlist['id'])
tracks = playlist.get('tracks') or []
- if not all([t.get('permalink_url') for t in tracks]) and token:
- tracks = self._download_json(
+ if not all(t.get('permalink_url') for t in tracks) and token:
+ tracks = self._call_api(
self._API_V2_BASE + 'tracks', playlist_id,
'Downloading tracks', query={
- 'ids': ','.join([compat_str(t['id']) for t in tracks]),
+ 'ids': ','.join([str(t['id']) for t in tracks]),
'playlistId': playlist_id,
'playlistSecretToken': token,
}, headers=self._HEADERS)
def _real_extract(self, url):
mobj = self._match_valid_url(url)
- full_title = '%s/sets/%s' % mobj.group('uploader', 'slug_title')
+ full_title = '{}/sets/{}'.format(*mobj.group('uploader', 'slug_title'))
token = mobj.group('token')
if token:
full_title += '/' + token
- info = self._download_json(self._resolv_url(
+ info = self._call_api(self._resolv_url(
self._BASE_URL + full_title), full_title, headers=self._HEADERS)
if 'errors' in info:
- msgs = (compat_str(err['error_message']) for err in info['errors'])
- raise ExtractorError('unable to download video webpage: %s' % ','.join(msgs))
+ msgs = (str(err['error_message']) for err in info['errors'])
+ raise ExtractorError('unable to download video webpage: {}'.format(','.join(msgs)))
return self._extract_set(info, token)
for i in itertools.count():
for retry in self.RetryManager():
try:
- response = self._download_json(
+ response = self._call_api(
url, playlist_id, query=query, headers=self._HEADERS,
note=f'Downloading track page {i + 1}')
break
mobj = self._match_valid_url(url)
uploader = mobj.group('user')
- user = self._download_json(
+ user = self._call_api(
self._resolv_url(self._BASE_URL + uploader),
uploader, 'Downloading user info', headers=self._HEADERS)
return self._extract_playlist(
self._API_V2_BASE + self._BASE_URL_MAP[resource] % user['id'],
str_or_none(user.get('id')),
- '%s (%s)' % (user['username'], resource.capitalize()))
+ '{} ({})'.format(user['username'], resource.capitalize()))
class SoundcloudUserPermalinkIE(SoundcloudPagedPlaylistBaseIE):
def _real_extract(self, url):
user_id = self._match_id(url)
- user = self._download_json(
+ user = self._call_api(
self._resolv_url(url), user_id, 'Downloading user info', headers=self._HEADERS)
return self._extract_playlist(
def _real_extract(self, url):
track_name = self._match_id(url)
- track = self._download_json(self._resolv_url(url), track_name, headers=self._HEADERS)
+ track = self._call_api(self._resolv_url(url), track_name, headers=self._HEADERS)
track_id = self._search_regex(
r'soundcloud:track-stations:(\d+)', track['id'], 'track id')
return self._extract_playlist(
- self._API_V2_BASE + 'stations/%s/tracks' % track['id'],
- track_id, 'Track station: %s' % track['title'])
+ self._API_V2_BASE + 'stations/{}/tracks'.format(track['id']),
+ track_id, 'Track station: {}'.format(track['title']))
class SoundcloudRelatedIE(SoundcloudPagedPlaylistBaseIE):
def _real_extract(self, url):
slug, relation = self._match_valid_url(url).group('slug', 'relation')
- track = self._download_json(
+ track = self._call_api(
self._resolv_url(self._BASE_URL + slug),
slug, 'Downloading track info', headers=self._HEADERS)
return self._extract_playlist(
self._API_V2_BASE + self._BASE_URL_MAP[relation] % track['id'], str(track['id']),
- '%s (%s)' % (track.get('title') or slug, relation.capitalize()))
+ '{} ({})'.format(track.get('title') or slug, relation.capitalize()))
class SoundcloudPlaylistIE(SoundcloudPlaylistBaseIE):
if token:
query['secret_token'] = token
- data = self._download_json(
+ data = self._call_api(
self._API_V2_BASE + 'playlists/' + playlist_id,
playlist_id, 'Downloading playlist', query=query, headers=self._HEADERS)
next_url = update_url_query(self._API_V2_BASE + endpoint, query)
for i in itertools.count(1):
- response = self._download_json(
+ response = self._call_api(
next_url, collection_id, f'Downloading page {i}',
'Unable to download API page', headers=self._HEADERS)