diff --git a/yt_dlp/extractor/neteasemusic.py b/yt_dlp/extractor/neteasemusic.py index dd50efe51..0399efb1c 100644 --- a/yt_dlp/extractor/neteasemusic.py +++ b/yt_dlp/extractor/neteasemusic.py @@ -22,12 +22,12 @@ class NetEaseMusicBaseIE(InfoExtractor): - _FORMATS = ['bMusic', 'mMusic', 'hMusic'] + _LEVELS = ('standard', 'exhigh', 'lossless', 'hires', 'jyeffect', 'sky', 'jymaster') _API_BASE = 'http://music.163.com/api/' _GEO_BYPASS = False @staticmethod - def kilo_or_none(value): + def _kilo_or_none(value): return int_or_none(value, scale=1000) def _create_eapi_cipher(self, api_path, query_body, cookies): @@ -66,45 +66,36 @@ def _download_eapi_json(self, path, video_id, query_body, headers={}, **kwargs): **headers, }, **kwargs) - def _call_player_api(self, song_id, bitrate): + def _call_player_api(self, song_id, level): return self._download_eapi_json( - '/song/enhance/player/url', song_id, {'ids': f'[{song_id}]', 'br': bitrate}, - note=f'Downloading song URL info: bitrate {bitrate}') + '/song/enhance/player/url/v1', song_id, {'ids': f'[{song_id}]', 'level': level, 'encodeType': 'flac'}, + note=f'Downloading song URL info: level {level}') - def extract_formats(self, info): - err = 0 + def _extract_formats(self, info): formats = [] song_id = info['id'] - for song_format in self._FORMATS: - details = info.get(song_format) - if not details: + for level in self._LEVELS: + song = traverse_obj(self._call_player_api(song_id, level), ('data', 0, {dict})) or {} + if song.get('level') != level: + break # We have already extracted the highest level the user has access to + if not url_or_none(song.get('url')): continue - bitrate = int_or_none(details.get('bitrate')) or 999000 - for song in traverse_obj(self._call_player_api(song_id, bitrate), ('data', lambda _, v: url_or_none(v['url']))): - song_url = song['url'] - if self._is_valid_url(song_url, info['id'], 'song'): - formats.append({ - 'url': song_url, - 'format_id': song_format, - 'asr': traverse_obj(details, ('sr', {int_or_none})), - **traverse_obj(song, { - 'ext': ('type', {str}), - 'abr': ('br', {self.kilo_or_none}), - 'filesize': ('size', {int_or_none}), - }), - }) - elif err == 0: - err = traverse_obj(song, ('code', {int})) or 0 - + formats.append({ + 'url': song['url'], + 'format_id': level, + 'vcodec': 'none', + **traverse_obj(song, { + 'ext': ('type', {str}), + 'abr': ('br', {self._kilo_or_none}), + 'filesize': ('size', {int_or_none}), + }), + }) if not formats: - if err != 0 and (err < 200 or err >= 400): - raise ExtractorError(f'No media links found (site code {err})', expected=True) - else: - self.raise_geo_restricted( - 'No media links found: probably due to geo restriction.', countries=['CN']) + self.raise_geo_restricted( + 'No media links found; possibly due to geo restriction', countries=['CN']) return formats - def query_api(self, endpoint, video_id, note): + def _query_api(self, endpoint, video_id, note): result = self._download_json( f'{self._API_BASE}{endpoint}', video_id, note, headers={'Referer': self._API_BASE}) code = traverse_obj(result, ('code', {int})) @@ -146,8 +137,8 @@ class NetEaseMusicIE(NetEaseMusicBaseIE): 'average_rating': int, 'album_artist': '偶像练习生', }, + 'skip': 'No longer available', }, { - 'note': 'No lyrics.', 'url': 'http://music.163.com/song?id=17241424', 'info_dict': { 'id': '17241424', @@ -161,10 +152,14 @@ class NetEaseMusicIE(NetEaseMusicBaseIE): 'album': 'Piano Solos Vol. 2', 'album_artist': 'Dustin O\'Halloran', 'average_rating': int, + 'description': '[00:05.00]纯音乐，请欣赏\n', + 'album_artists': ['Dustin O\'Halloran'], + 'creators': ['Dustin O\'Halloran'], + 'subtitles': {'lyrics': [{'ext': 'lrc'}]}, }, }, { 'url': 'https://y.music.163.com/m/song?app_version=8.8.45&id=95670&uct2=sKnvS4+0YStsWkqsPhFijw%3D%3D&dlt=0846', - 'md5': '95826c73ea50b1c288b22180ec9e754d', + 'md5': 'b896be78d8d34bd7bb665b26710913ff', 'info_dict': { 'id': '95670', 'ext': 'mp3', @@ -180,6 +175,8 @@ class NetEaseMusicIE(NetEaseMusicBaseIE): 'average_rating': int, 'album': '红色摇滚', 'album_artist': '侯牧人', + 'creators': ['马备'], + 'album_artists': ['侯牧人'], }, }, { 'url': 'http://music.163.com/#/song?id=32102397', @@ -251,12 +248,12 @@ def _process_lyrics(self, lyrics_info): def _real_extract(self, url): song_id = self._match_id(url) - info = self.query_api( + info = self._query_api( f'song/detail?id={song_id}&ids=%5B{song_id}%5D', song_id, 'Downloading song info')['songs'][0] - formats = self.extract_formats(info) + formats = self._extract_formats(info) - lyrics = self._process_lyrics(self.query_api( + lyrics = self._process_lyrics(self._query_api( f'song/lyric?id={song_id}&lv=-1&tv=-1', song_id, 'Downloading lyrics data')) lyric_data = { 'description': traverse_obj(lyrics, (('lyrics_merged', 'lyrics'), 0, 'data'), get_all=False), @@ -272,9 +269,9 @@ def _real_extract(self, url): **lyric_data, **traverse_obj(info, { 'title': ('name', {str}), - 'timestamp': ('album', 'publishTime', {self.kilo_or_none}), + 'timestamp': ('album', 'publishTime', {self._kilo_or_none}), 'thumbnail': ('album', 'picUrl', {url_or_none}), - 'duration': ('duration', {self.kilo_or_none}), + 'duration': ('duration', {self._kilo_or_none}), 'album': ('album', 'name', {str}), 'average_rating': ('score', {int_or_none}), }), @@ -358,7 +355,7 @@ class NetEaseMusicSingerIE(NetEaseMusicBaseIE): def _real_extract(self, url): singer_id = self._match_id(url) - info = self.query_api( + info = self._query_api( f'artist/{singer_id}?id={singer_id}', singer_id, note='Downloading singer data') name = join_nonempty( @@ -426,7 +423,7 @@ def _real_extract(self, url): 'tags': ('tags', ..., {str}), 'uploader': ('creator', 'nickname', {str}), 'uploader_id': ('creator', 'userId', {str_or_none}), - 'timestamp': ('updateTime', {self.kilo_or_none}), + 'timestamp': ('updateTime', {self._kilo_or_none}), })) if traverse_obj(info, ('playlist', 'specialType')) == 10: metainfo['title'] = f'{metainfo.get("title")} {strftime_or_none(metainfo.get("timestamp"), "%Y-%m-%d")}' @@ -473,7 +470,7 @@ class NetEaseMusicMvIE(NetEaseMusicBaseIE): def _real_extract(self, url): mv_id = self._match_id(url) - info = self.query_api( + info = self._query_api( f'mv/detail?id={mv_id}&type=mp4', mv_id, 'Downloading mv info')['data'] formats = [ @@ -490,7 +487,7 @@ def _real_extract(self, url): 'creator': ('artistName', {str}), 'upload_date': ('publishTime', {unified_strdate}), 'thumbnail': ('cover', {url_or_none}), - 'duration': ('duration', {self.kilo_or_none}), + 'duration': ('duration', {self._kilo_or_none}), 'view_count': ('playCount', {int_or_none}), 'like_count': ('likeCount', {int_or_none}), 'comment_count': ('commentCount', {int_or_none}), @@ -550,7 +547,7 @@ class NetEaseMusicProgramIE(NetEaseMusicBaseIE): def _real_extract(self, url): program_id = self._match_id(url) - info = self.query_api( + info = self._query_api( f'dj/program/detail?id={program_id}', program_id, note='Downloading program info')['program'] metainfo = traverse_obj(info, { @@ -558,17 +555,17 @@ def _real_extract(self, url): 'description': ('description', {str}), 'creator': ('dj', 'brand', {str}), 'thumbnail': ('coverUrl', {url_or_none}), - 'timestamp': ('createTime', {self.kilo_or_none}), + 'timestamp': ('createTime', {self._kilo_or_none}), }) if not self._yes_playlist( info['songs'] and program_id, info['mainSong']['id'], playlist_label='program', video_label='song'): - formats = self.extract_formats(info['mainSong']) + formats = self._extract_formats(info['mainSong']) return { 'id': str(info['mainSong']['id']), 'formats': formats, - 'duration': traverse_obj(info, ('mainSong', 'duration', {self.kilo_or_none})), + 'duration': traverse_obj(info, ('mainSong', 'duration', {self._kilo_or_none})), **metainfo, } @@ -597,7 +594,7 @@ def _real_extract(self, url): metainfo = {} entries = [] for offset in itertools.count(start=0, step=self._PAGE_SIZE): - info = self.query_api( + info = self._query_api( f'dj/program/byradio?asc=false&limit={self._PAGE_SIZE}&radioId={dj_id}&offset={offset}', dj_id, note=f'Downloading dj programs - {offset}')