-# coding: utf-8
-from __future__ import unicode_literals
-
import itertools
import re
import json
-import random
+# import random
from .common import (
InfoExtractor,
SearchInfoExtractor
)
-from ..compat import (
- compat_HTTPError,
- compat_kwargs,
- compat_str,
- compat_urlparse,
-)
+from ..compat import compat_str
+from ..networking import HEADRequest, Request
+from ..networking.exceptions import HTTPError
from ..utils import (
error_to_compat_str,
ExtractorError,
float_or_none,
- HEADRequest,
int_or_none,
KNOWN_EXTENSIONS,
mimetype2ext,
+ parse_qs,
str_or_none,
try_get,
unified_timestamp,
update_url_query,
url_or_none,
urlhandle_detect_ext,
- sanitized_Request,
)
class SoundcloudEmbedIE(InfoExtractor):
_VALID_URL = r'https?://(?:w|player|p)\.soundcloud\.com/player/?.*?\burl=(?P<id>.+)'
+ _EMBED_REGEX = [r'<iframe[^>]+src=(["\'])(?P<url>(?:https?://)?(?:w\.)?soundcloud\.com/player.+?)\1']
_TEST = {
# from https://www.soundi.fi/uutiset/ennakkokuuntelussa-timo-kaukolammen-station-to-station-to-station-julkaisua-juhlitaan-tanaan-g-livelabissa/
'url': 'https://w.soundcloud.com/player/?visual=true&url=https%3A%2F%2Fapi.soundcloud.com%2Fplaylists%2F922213810&show_artwork=true&maxwidth=640&maxheight=960&dnt=1&secret_token=s-ziYey',
'only_matching': True,
}
- @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)]
-
def _real_extract(self, url):
- query = compat_urlparse.parse_qs(
- compat_urlparse.urlparse(url).query)
+ query = parse_qs(url)
api_url = query['url'][0]
secret_token = query.get('secret_token')
if secret_token:
return self.url_result(api_url)
-class SoundcloudIE(InfoExtractor):
- """Information extractor for soundcloud.com
- To access the media, the uid of the song and a stream token
- must be extracted from the page source and the script must make
- a request to media.soundcloud.com/crossdomain.xml. Then
- the media can be grabbed by requesting from an url composed
- of the stream token and uid
- """
-
- _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,
- }
- },
- # 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',
- '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': 227.155,
- 'license': 'all-rights-reserved',
- 'view_count': int,
- 'like_count': int,
- 'comment_count': int,
- 'repost_count': int,
- },
- },
- # 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,
- },
- },
- {
- 'url': 'https://soundcloud.com/giovannisarani/mezzo-valzer',
- 'md5': 'e22aecd2bc88e0e4e432d7dcc0a1abf7',
- 'info_dict': {
- 'id': '583011102',
- 'ext': 'mp3',
- 'title': 'Mezzo Valzer',
- 'description': 'md5:4138d582f81866a530317bae316e8b61',
- 'uploader': 'Micronie',
- '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,
- },
- },
- {
- # AAC HQ format available (account with active subscription needed)
- 'url': 'https://soundcloud.com/wandw/the-chainsmokers-ft-daya-dont-let-me-down-ww-remix-1',
- 'only_matching': True,
- },
- {
- # Go+ (account with active subscription needed)
- 'url': 'https://soundcloud.com/taylorswiftofficial/look-what-you-made-me-do',
- 'only_matching': True,
- },
- ]
+class SoundcloudBaseIE(InfoExtractor):
+ _NETRC_MACHINE = 'soundcloud'
_API_V2_BASE = 'https://api-v2.soundcloud.com/'
_BASE_URL = 'https://soundcloud.com/'
+ _USER_AGENT = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.105 Safari/537.36'
+ _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'
_ARTWORK_MAP = {
}
def _store_client_id(self, client_id):
- self._downloader.cache.store('soundcloud', 'client_id', client_id)
+ self.cache.store('soundcloud', 'client_id', client_id)
def _update_client_id(self):
webpage = self._download_webpage('https://soundcloud.com/', None)
query['client_id'] = self._CLIENT_ID
kwargs['query'] = query
try:
- return super(SoundcloudIE, self)._download_json(*args, **compat_kwargs(kwargs))
+ return super()._download_json(*args, **kwargs)
except ExtractorError as e:
- if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
+ if isinstance(e.cause, HTTPError) and e.cause.status in (401, 403):
self._store_client_id(None)
self._update_client_id()
continue
return False
raise
- def _real_initialize(self):
- self._CLIENT_ID = self._downloader.cache.load('soundcloud', 'client_id') or "T5R4kgWS2PRf6lzLyIravUMnKlbIxQag" # 'EXLwg5lHTO2dslU5EePe3xkw0m1h86Cd' # 'YUKXoArFcqrlQn9tfNHvvyfnDISj04zk'
- self._login()
-
- _USER_AGENT = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.105 Safari/537.36"
- _API_AUTH_QUERY_TEMPLATE = '?client_id=%s'
- _API_AUTH_URL_PW = 'https://api-auth.soundcloud.com/web-auth/sign-in/password%s'
- _access_token = None
- _HEADERS = {}
- _NETRC_MACHINE = 'soundcloud'
+ def _initialize_pre_login(self):
+ self._CLIENT_ID = self.cache.load('soundcloud', 'client_id') or 'a3e059563d7fd3372b49b37f00a00bcf'
- def _login(self):
- username, password = self._get_login_info()
- if username is None:
- return
+ def _perform_login(self, username, password):
+ if username != 'oauth':
+ self.report_warning(
+ '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')
+ r'''
def genDevId():
def genNumBlock():
return ''.join([str(random.randrange(10)) for i in range(6)])
self.report_warning('Unable to get access token, login may has failed')
else:
self._HEADERS = {'Authorization': 'OAuth ' + self._access_token}
+ '''
# signature generation
def sign(self, user, pw, clid):
b = 37
k = 37
c = 5
- n = "0763ed7314c69015fd4a0dc16bbf4b90" # _KEY
- y = "8" # _REV
- r = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.105 Safari/537.36" # _USER_AGENT
+ n = '0763ed7314c69015fd4a0dc16bbf4b90' # _KEY
+ y = '8' # _REV
+ r = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.105 Safari/537.36' # _USER_AGENT
e = user # _USERNAME
t = clid # _CLIENT_ID
return out
- @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):
+ def _extract_info_dict(self, info, full_title=None, secret_token=None, extract_flat=False):
track_id = compat_str(info['id'])
title = info['title']
if secret_token:
query['secret_token'] = secret_token
- if info.get('downloadable') and info.get('has_downloads_left'):
+ 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')
urlh = self._request_webpage(
HEADRequest(redirect_url), track_id, fatal=False)
if urlh:
- format_url = urlh.geturl()
+ format_url = urlh.url
format_urls.add(format_url)
formats.append({
'format_id': 'download',
format_url = url_or_none(t.get('url'))
if not format_url:
continue
- stream = self._download_json(
+ stream = None if extract_flat else self._download_json(
format_url, track_id, query=query, fatal=False, headers=self._HEADERS)
if not isinstance(stream, dict):
continue
f['vcodec'] = 'none'
if not formats and info.get('policy') == 'BLOCK':
- self.raise_geo_restricted()
- self._sort_formats(formats)
+ self.raise_geo_restricted(metadata_available=True)
user = info.get('user') or {}
'comment_count': extract_count('comment'),
'repost_count': extract_count('reposts'),
'genre': info.get('genre'),
- 'formats': formats
+ 'formats': formats if not extract_flat else None
}
+ @classmethod
+ def _resolv_url(cls, url):
+ return cls._API_V2_BASE + 'resolve?url=' + url
+
+
+class SoundcloudIE(SoundcloudBaseIE):
+ """Information extractor for soundcloud.com
+ To access the media, the uid of the song and a stream token
+ must be extracted from the page source and the script must make
+ a request to media.soundcloud.com/crossdomain.xml. Then
+ the media can be grabbed by requesting from an url composed
+ of the stream token and uid
+ """
+
+ _VALID_URL = r'''(?x)^(?:https?://)?
+ (?:(?:(?:www\.|m\.)?soundcloud\.com/
+ (?!stations/track)
+ (?P<uploader>[\w\d-]+)/
+ (?!(?:tracks|albums|sets(?:/.+?)?|reposts|likes|spotlight)/?(?:$|[?#]))
+ (?P<title>[\w\d-]+)
+ (?:/(?P<token>(?!(?:albums|sets|recommended))[^?]+?))?
+ (?:[?].*)?$)
+ |(?: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,
+ }
+ },
+ # 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',
+ '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': 227.155,
+ 'license': 'all-rights-reserved',
+ 'view_count': int,
+ 'like_count': int,
+ 'comment_count': int,
+ 'repost_count': int,
+ },
+ },
+ # 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/the80m/the-following',
+ 'md5': '9ffcddb08c87d74fb5808a3c183a1d04',
+ 'info_dict': {
+ 'id': '343609555',
+ 'ext': 'wav',
+ },
+ },
+ # 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,
+ },
+ },
+ {
+ 'url': 'https://soundcloud.com/giovannisarani/mezzo-valzer',
+ 'md5': 'e22aecd2bc88e0e4e432d7dcc0a1abf7',
+ 'info_dict': {
+ 'id': '583011102',
+ 'ext': 'mp3',
+ 'title': 'Mezzo Valzer',
+ 'description': 'md5:4138d582f81866a530317bae316e8b61',
+ 'uploader': 'Micronie',
+ '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,
+ },
+ },
+ {
+ # AAC HQ format available (account with active subscription needed)
+ 'url': 'https://soundcloud.com/wandw/the-chainsmokers-ft-daya-dont-let-me-down-ww-remix-1',
+ 'only_matching': True,
+ },
+ {
+ # Go+ (account with active subscription needed)
+ 'url': 'https://soundcloud.com/taylorswiftofficial/look-what-you-made-me-do',
+ 'only_matching': True,
+ },
+ ]
+
def _real_extract(self, url):
- mobj = re.match(self._VALID_URL, url)
+ mobj = self._match_valid_url(url)
track_id = mobj.group('track_id')
return self._extract_info_dict(info, full_title, token)
-class SoundcloudPlaylistBaseIE(SoundcloudIE):
+class SoundcloudPlaylistBaseIE(SoundcloudBaseIE):
def _extract_set(self, playlist, token=None):
playlist_id = compat_str(playlist['id'])
tracks = playlist.get('tracks') or []
}]
def _real_extract(self, url):
- mobj = re.match(self._VALID_URL, url)
+ mobj = self._match_valid_url(url)
full_title = '%s/sets/%s' % mobj.group('uploader', 'slug_title')
token = mobj.group('token')
return self._extract_set(info, token)
-class SoundcloudPagedPlaylistBaseIE(SoundcloudIE):
+class SoundcloudPagedPlaylistBaseIE(SoundcloudBaseIE):
def _extract_playlist(self, base_url, playlist_id, playlist_title):
+ return {
+ '_type': 'playlist',
+ 'id': playlist_id,
+ 'title': playlist_title,
+ 'entries': self._entries(base_url, playlist_id),
+ }
+
+ def _entries(self, url, playlist_id):
# Per the SoundCloud documentation, the maximum limit for a linked partitioning query is 200.
# https://developers.soundcloud.com/blog/offset-pagination-deprecated
- COMMON_QUERY = {
+ query = {
'limit': 200,
'linked_partitioning': '1',
+ 'offset': 0,
}
- query = COMMON_QUERY.copy()
- query['offset'] = 0
-
- next_href = base_url
-
- entries = []
for i in itertools.count():
- response = self._download_json(
- next_href, playlist_id,
- 'Downloading track page %s' % (i + 1), query=query, headers=self._HEADERS)
-
- collection = response['collection']
-
- if not isinstance(collection, list):
- collection = []
-
- # Empty collection may be returned, in this case we proceed
- # straight to next_href
+ for retry in self.RetryManager():
+ try:
+ response = self._download_json(
+ url, playlist_id, query=query, headers=self._HEADERS,
+ note=f'Downloading track page {i + 1}')
+ break
+ except ExtractorError as e:
+ # Downloading page may result in intermittent 502 HTTP error
+ # See https://github.com/yt-dlp/yt-dlp/issues/872
+ if not isinstance(e.cause, HTTPError) or e.cause.status != 502:
+ raise
+ retry.error = e
+ continue
- def resolve_entry(candidates):
+ def resolve_entry(*candidates):
for cand in candidates:
if not isinstance(cand, dict):
continue
permalink_url = url_or_none(cand.get('permalink_url'))
- if not permalink_url:
- continue
- return self.url_result(
- permalink_url,
- SoundcloudIE.ie_key() if SoundcloudIE.suitable(permalink_url) else None,
- str_or_none(cand.get('id')), cand.get('title'))
-
- for e in collection:
- entry = resolve_entry((e, e.get('track'), e.get('playlist')))
- if entry:
- entries.append(entry)
-
- next_href = response.get('next_href')
- if not next_href:
- break
+ if permalink_url:
+ return self.url_result(
+ permalink_url,
+ SoundcloudIE.ie_key() if SoundcloudIE.suitable(permalink_url) else None,
+ str_or_none(cand.get('id')), cand.get('title'))
- next_href = response['next_href']
- parsed_next_href = compat_urlparse.urlparse(next_href)
- query = compat_urlparse.parse_qs(parsed_next_href.query)
- query.update(COMMON_QUERY)
+ for e in response['collection'] or []:
+ yield resolve_entry(e, e.get('track'), e.get('playlist'))
- return {
- '_type': 'playlist',
- 'id': playlist_id,
- 'title': playlist_title,
- 'entries': entries,
- }
+ url = response.get('next_href')
+ if not url:
+ break
+ query.pop('offset', None)
class SoundcloudUserIE(SoundcloudPagedPlaylistBaseIE):
}
def _real_extract(self, url):
- mobj = re.match(self._VALID_URL, url)
+ mobj = self._match_valid_url(url)
uploader = mobj.group('user')
user = self._download_json(
'%s (%s)' % (user['username'], resource.capitalize()))
+class SoundcloudUserPermalinkIE(SoundcloudPagedPlaylistBaseIE):
+ _VALID_URL = r'https?://api\.soundcloud\.com/users/(?P<id>\d+)'
+ IE_NAME = 'soundcloud:user:permalink'
+ _TESTS = [{
+ 'url': 'https://api.soundcloud.com/users/30909869',
+ 'info_dict': {
+ 'id': '30909869',
+ 'title': 'neilcic',
+ },
+ 'playlist_mincount': 23,
+ }]
+
+ def _real_extract(self, url):
+ user_id = self._match_id(url)
+ user = self._download_json(
+ self._resolv_url(url), user_id, 'Downloading user info', headers=self._HEADERS)
+
+ return self._extract_playlist(
+ f'{self._API_V2_BASE}stream/users/{user["id"]}', str(user['id']), user.get('username'))
+
+
class SoundcloudTrackStationIE(SoundcloudPagedPlaylistBaseIE):
_VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/stations/track/[^/]+/(?P<id>[^/?#&]+)'
IE_NAME = 'soundcloud:trackstation'
track_id, 'Track station: %s' % track['title'])
+class SoundcloudRelatedIE(SoundcloudPagedPlaylistBaseIE):
+ _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/(?P<slug>[\w\d-]+/[\w\d-]+)/(?P<relation>albums|sets|recommended)'
+ IE_NAME = 'soundcloud:related'
+ _TESTS = [{
+ 'url': 'https://soundcloud.com/wajang/sexapil-pingers-5/recommended',
+ 'info_dict': {
+ 'id': '1084577272',
+ 'title': 'Sexapil - Pingers 5 (Recommended)',
+ },
+ 'playlist_mincount': 50,
+ }, {
+ 'url': 'https://soundcloud.com/wajang/sexapil-pingers-5/albums',
+ 'info_dict': {
+ 'id': '1084577272',
+ 'title': 'Sexapil - Pingers 5 (Albums)',
+ },
+ 'playlist_mincount': 1,
+ }, {
+ 'url': 'https://soundcloud.com/wajang/sexapil-pingers-5/sets',
+ 'info_dict': {
+ 'id': '1084577272',
+ 'title': 'Sexapil - Pingers 5 (Sets)',
+ },
+ 'playlist_mincount': 4,
+ }]
+
+ _BASE_URL_MAP = {
+ 'albums': 'tracks/%s/albums',
+ 'sets': 'tracks/%s/playlists_without_albums',
+ 'recommended': 'tracks/%s/related',
+ }
+
+ def _real_extract(self, url):
+ slug, relation = self._match_valid_url(url).group('slug', 'relation')
+
+ track = self._download_json(
+ self._resolv_url(self._BASE_URL + slug),
+ slug, 'Downloading track info', headers=self._HEADERS)
+
+ if track.get('errors'):
+ raise ExtractorError(f'{self.IE_NAME} said: %s' % ','.join(
+ str(err['error_message']) for err in track['errors']), expected=True)
+
+ 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()))
+
+
class SoundcloudPlaylistIE(SoundcloudPlaylistBaseIE):
_VALID_URL = r'https?://api(?:-v2)?\.soundcloud\.com/playlists/(?P<id>[0-9]+)(?:/?\?secret_token=(?P<token>[^&]+?))?$'
IE_NAME = 'soundcloud:playlist'
}]
def _real_extract(self, url):
- mobj = re.match(self._VALID_URL, url)
+ mobj = self._match_valid_url(url)
playlist_id = mobj.group('id')
query = {}
return self._extract_set(data, token)
-class SoundcloudSearchIE(SearchInfoExtractor, SoundcloudIE):
+class SoundcloudSearchIE(SoundcloudBaseIE, SearchInfoExtractor):
IE_NAME = 'soundcloud:search'
IE_DESC = 'Soundcloud search'
- _MAX_RESULTS = float('inf')
+ _SEARCH_KEY = 'scsearch'
_TESTS = [{
'url': 'scsearch15:post-avant jazzcore',
'info_dict': {
+ 'id': 'post-avant jazzcore',
'title': 'post-avant jazzcore',
},
'playlist_count': 15,
}]
- _SEARCH_KEY = 'scsearch'
_MAX_RESULTS_PER_PAGE = 200
_DEFAULT_RESULTS_PER_PAGE = 50
})
next_url = update_url_query(self._API_V2_BASE + endpoint, query)
- collected_results = 0
-
for i in itertools.count(1):
response = self._download_json(
- next_url, collection_id, 'Downloading page {0}'.format(i),
+ next_url, collection_id, f'Downloading page {i}',
'Unable to download API page', headers=self._HEADERS)
- collection = response.get('collection', [])
- if not collection:
- break
-
- collection = list(filter(bool, collection))
- collected_results += len(collection)
-
- for item in collection:
- yield self.url_result(item['uri'], SoundcloudIE.ie_key())
-
- if not collection or collected_results >= limit:
- break
+ for item in response.get('collection') or []:
+ if item:
+ yield self.url_result(
+ item['uri'], SoundcloudIE.ie_key(), **self._extract_info_dict(item, extract_flat=True))
next_url = response.get('next_href')
if not next_url:
break
def _get_n_results(self, query, n):
- tracks = self._get_collection('search/tracks', query, limit=n, q=query)
- return self.playlist_result(tracks, playlist_title=query)
+ return self.playlist_result(itertools.islice(
+ self._get_collection('search/tracks', query, limit=n, q=query),
+ 0, None if n == float('inf') else n), query, query)