From eb25901073c5c6162761f1e49ed9a86a418798ad Mon Sep 17 00:00:00 2001 From: unkernet Date: Fri, 16 Jun 2023 16:40:58 +0700 Subject: [PATCH] [soundcloud] Download Soundcloud GO+ high quality streams and private playlists --- youtube_dl/extractor/soundcloud.py | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) diff --git a/youtube_dl/extractor/soundcloud.py b/youtube_dl/extractor/soundcloud.py index abb85e1e5..ad61fa618 100644 --- a/youtube_dl/extractor/soundcloud.py +++ b/youtube_dl/extractor/soundcloud.py @@ -289,10 +289,13 @@ class SoundcloudIE(InfoExtractor): 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'] + non_fatal = not kwargs.get('fatal', True) query = kwargs.get('query', {}).copy() + oauth_token = self._get_cookies(self._BASE_URL).get('oauth_token') + if oauth_token: + if not kwargs.get('headers'): + kwargs['headers'] = {} + kwargs['headers']['authorization'] = 'OAuth ' + oauth_token.value for _ in range(2): query['client_id'] = self._CLIENT_ID kwargs['query'] = query @@ -319,6 +322,7 @@ class SoundcloudIE(InfoExtractor): track_id = compat_str(info['id']) title = info['title'] + duration = info.get('duration') format_urls = set() formats = [] query = {'client_id': self._CLIENT_ID} @@ -335,12 +339,14 @@ class SoundcloudIE(InfoExtractor): if urlh: format_url = urlh.geturl() format_urls.add(format_url) + filesize = int_or_none(urlh.headers.get('Content-Length')) formats.append({ 'format_id': 'download', 'ext': urlhandle_detect_ext(urlh) or 'mp3', - 'filesize': int_or_none(urlh.headers.get('Content-Length')), + 'filesize': filesize, 'url': format_url, 'preference': 10, + 'abr': int_or_none(filesize, invscale=8, scale=duration), }) def invalid_url(url): @@ -456,7 +462,7 @@ class SoundcloudIE(InfoExtractor): 'title': title, 'description': info.get('description'), 'thumbnails': thumbnails, - 'duration': float_or_none(info.get('duration'), 1000), + 'duration': float_or_none(duration, 1000), 'webpage_url': info.get('permalink_url'), 'license': info.get('license'), 'view_count': extract_count('playback'),