From beaf832c7a9d57833f365ce18f6115b88071b296 Mon Sep 17 00:00:00 2001 From: bashonly <88596187+bashonly@users.noreply.github.com> Date: Wed, 22 May 2024 17:20:29 -0500 Subject: [PATCH] [ie/soundcloud] Add `formats` extractor-arg (#10004) Authored by: bashonly --- README.md | 3 ++ yt_dlp/extractor/soundcloud.py | 58 +++++++++++++++++++++++----------- 2 files changed, 42 insertions(+), 19 deletions(-) diff --git a/README.md b/README.md index 2c909976a..887cfde23 100644 --- a/README.md +++ b/README.md @@ -1841,6 +1841,9 @@ #### jiosaavn #### afreecatvlive * `cdn`: One or more CDN IDs to use with the API call for stream URLs, e.g. `gcp_cdn`, `gs_cdn_pc_app`, `gs_cdn_mobile_web`, `gs_cdn_pc_web` +#### soundcloud +* `formats`: Formats to request from the API. Requested values should be in the format of `{protocol}_{extension}` (omitting the bitrate), e.g. `hls_opus,http_aac`. The `*` character functions as a wildcard, e.g. `*_mp3`, and can passed by itself to request all formats. Known protocols include `http`, `hls` and `hls-aes`; known extensions include `aac`, `opus` and `mp3`. Original `download` formats are always extracted. Default is `http_aac,hls_aac,http_opus,hls_opus,http_mp3,hls_mp3` + **Note**: These options may be changed/removed in the future without concern for backward compatibility diff --git a/yt_dlp/extractor/soundcloud.py b/yt_dlp/extractor/soundcloud.py index c9ca41a5c..358146171 100644 --- a/yt_dlp/extractor/soundcloud.py +++ b/yt_dlp/extractor/soundcloud.py @@ -1,3 +1,4 @@ +import functools import itertools import json import re @@ -12,6 +13,7 @@ error_to_compat_str, float_or_none, int_or_none, + join_nonempty, mimetype2ext, parse_qs, str_or_none, @@ -68,6 +70,16 @@ class SoundcloudBaseIE(InfoExtractor): '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) @@ -216,7 +228,7 @@ def _extract_info_dict(self, info, full_title=None, secret_token=None, extract_f redirect_url = (self._download_json(download_url, track_id, fatal=False) or {}).get('redirectUri') if redirect_url: urlh = self._request_webpage( - HEADRequest(redirect_url), track_id, fatal=False) + HEADRequest(redirect_url), track_id, 'Checking for original download format', fatal=False) if urlh: format_url = urlh.url format_urls.add(format_url) @@ -258,7 +270,7 @@ def add_format(f, protocol, is_preview=False): 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' @@ -274,11 +286,32 @@ def add_format(f, protocol, is_preview=False): 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._download_json( + 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( @@ -289,27 +322,14 @@ def add_format(f, protocol, is_preview=False): 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'