From ba6b0c8261e9f0a6373885736ff90a89dd1fb614 Mon Sep 17 00:00:00 2001 From: DmitryScaletta Date: Fri, 19 Jan 2024 06:16:21 +0300 Subject: [PATCH] [ie/chzzk] Add extractors (#8887) Closes #8804 Authored by: DmitryScaletta --- yt_dlp/extractor/_extractors.py | 4 + yt_dlp/extractor/chzzk.py | 139 ++++++++++++++++++++++++++++++++ 2 files changed, 143 insertions(+) create mode 100644 yt_dlp/extractor/chzzk.py diff --git a/yt_dlp/extractor/_extractors.py b/yt_dlp/extractor/_extractors.py index 7250ad5e0..3d360a52f 100644 --- a/yt_dlp/extractor/_extractors.py +++ b/yt_dlp/extractor/_extractors.py @@ -345,6 +345,10 @@ ChingariIE, ChingariUserIE, ) +from .chzzk import ( + CHZZKLiveIE, + CHZZKVideoIE, +) from .cinemax import CinemaxIE from .cinetecamilano import CinetecaMilanoIE from .cineverse import ( diff --git a/yt_dlp/extractor/chzzk.py b/yt_dlp/extractor/chzzk.py new file mode 100644 index 000000000..6894baea5 --- /dev/null +++ b/yt_dlp/extractor/chzzk.py @@ -0,0 +1,139 @@ +import functools + +from .common import InfoExtractor +from ..utils import ( + ExtractorError, + float_or_none, + int_or_none, + parse_iso8601, + url_or_none, +) +from ..utils.traversal import traverse_obj + + +class CHZZKLiveIE(InfoExtractor): + IE_NAME = 'chzzk:live' + _VALID_URL = r'https?://chzzk\.naver\.com/live/(?P[\da-f]+)' + _TESTS = [{ + 'url': 'https://chzzk.naver.com/live/c68b8ef525fb3d2fa146344d84991753', + 'info_dict': { + 'id': 'c68b8ef525fb3d2fa146344d84991753', + 'ext': 'mp4', + 'title': str, + 'channel': '진짜도현', + 'channel_id': 'c68b8ef525fb3d2fa146344d84991753', + 'channel_is_verified': False, + 'thumbnail': r're:^https?://.*\.jpg$', + 'timestamp': 1705510344, + 'upload_date': '20240117', + 'live_status': 'is_live', + 'view_count': int, + 'concurrent_view_count': int, + }, + 'skip': 'The channel is not currently live', + }] + + def _real_extract(self, url): + channel_id = self._match_id(url) + live_detail = self._download_json( + f'https://api.chzzk.naver.com/service/v2/channels/{channel_id}/live-detail', channel_id, + note='Downloading channel info', errnote='Unable to download channel info')['content'] + + if live_detail.get('status') == 'CLOSE': + raise ExtractorError('The channel is not currently live', expected=True) + + live_playback = self._parse_json(live_detail['livePlaybackJson'], channel_id) + + thumbnails = [] + thumbnail_template = traverse_obj( + live_playback, ('thumbnail', 'snapshotThumbnailTemplate', {url_or_none})) + if thumbnail_template and '{type}' in thumbnail_template: + for width in traverse_obj(live_playback, ('thumbnail', 'types', ..., {str})): + thumbnails.append({ + 'id': width, + 'url': thumbnail_template.replace('{type}', width), + 'width': int_or_none(width), + }) + + formats, subtitles = [], {} + for media in traverse_obj(live_playback, ('media', lambda _, v: url_or_none(v['path']))): + is_low_latency = media.get('mediaId') == 'LLHLS' + fmts, subs = self._extract_m3u8_formats_and_subtitles( + media['path'], channel_id, 'mp4', fatal=False, live=True, + m3u8_id='hls-ll' if is_low_latency else 'hls') + for f in fmts: + if is_low_latency: + f['source_preference'] = -2 + if '-afragalow.stream-audio.stream' in f['format_id']: + f['quality'] = -2 + formats.extend(fmts) + self._merge_subtitles(subs, target=subtitles) + + return { + 'id': channel_id, + 'is_live': True, + 'formats': formats, + 'subtitles': subtitles, + 'thumbnails': thumbnails, + **traverse_obj(live_detail, { + 'title': ('liveTitle', {str}), + 'timestamp': ('openDate', {functools.partial(parse_iso8601, delimiter=' ')}), + 'concurrent_view_count': ('concurrentUserCount', {int_or_none}), + 'view_count': ('accumulateCount', {int_or_none}), + 'channel': ('channel', 'channelName', {str}), + 'channel_id': ('channel', 'channelId', {str}), + 'channel_is_verified': ('channel', 'verifiedMark', {bool}), + }), + } + + +class CHZZKVideoIE(InfoExtractor): + IE_NAME = 'chzzk:video' + _VALID_URL = r'https?://chzzk\.naver\.com/video/(?P\d+)' + _TESTS = [{ + 'url': 'https://chzzk.naver.com/video/1754', + 'md5': 'b0c0c1bb888d913b93d702b1512c7f06', + 'info_dict': { + 'id': '1754', + 'ext': 'mp4', + 'title': '치지직 테스트 방송', + 'channel': '침착맨', + 'channel_id': 'bb382c2c0cc9fa7c86ab3b037fb5799c', + 'channel_is_verified': False, + 'thumbnail': r're:^https?://.*\.jpg$', + 'duration': 15577, + 'timestamp': 1702970505.417, + 'upload_date': '20231219', + 'view_count': int, + }, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + video_meta = self._download_json( + f'https://api.chzzk.naver.com/service/v2/videos/{video_id}', video_id, + note='Downloading video info', errnote='Unable to download video info')['content'] + formats, subtitles = self._extract_mpd_formats_and_subtitles( + f'https://apis.naver.com/neonplayer/vodplay/v1/playback/{video_meta["videoId"]}', video_id, + query={ + 'key': video_meta['inKey'], + 'env': 'real', + 'lc': 'en_US', + 'cpl': 'en_US', + }, note='Downloading video playback', errnote='Unable to download video playback') + + return { + 'id': video_id, + 'formats': formats, + 'subtitles': subtitles, + **traverse_obj(video_meta, { + 'title': ('videoTitle', {str}), + 'thumbnail': ('thumbnailImageUrl', {url_or_none}), + 'timestamp': ('publishDateAt', {functools.partial(float_or_none, scale=1000)}), + 'view_count': ('readCount', {int_or_none}), + 'duration': ('duration', {int_or_none}), + 'channel': ('channel', 'channelName', {str}), + 'channel_id': ('channel', 'channelId', {str}), + 'channel_is_verified': ('channel', 'verifiedMark', {bool}), + }), + }