yt_dlp/extractor/neteasemusic.py

   1 import itertools
   2 import json
   3 import re
   4 import time
   5 from base64 import b64encode
   6 from binascii import hexlify
   7 from datetime import datetime
   8 from hashlib import md5
   9 from random import randint
  10
  11 from .common import InfoExtractor
  12 from ..aes import aes_ecb_encrypt, pkcs7_padding
  13 from ..compat import compat_urllib_parse_urlencode
  14 from ..utils import (
  15     ExtractorError,
  16     bytes_to_intlist,
  17     error_to_compat_str,
  18     float_or_none,
  19     int_or_none,
  20     intlist_to_bytes,
  21     sanitized_Request,
  22     try_get,
  23 )
  24
  25
  26 class NetEaseMusicBaseIE(InfoExtractor):
  27     _FORMATS = ['bMusic', 'mMusic', 'hMusic']
  28     _NETEASE_SALT = '3go8&$8*3*3h0k(2)2'
  29     _API_BASE = 'http://music.163.com/api/'
  30
  31     @classmethod
  32     def _encrypt(cls, dfsid):
  33         salt_bytes = bytearray(cls._NETEASE_SALT.encode('utf-8'))
  34         string_bytes = bytearray(str(dfsid).encode('ascii'))
  35         salt_len = len(salt_bytes)
  36         for i in range(len(string_bytes)):
  37             string_bytes[i] = string_bytes[i] ^ salt_bytes[i % salt_len]
  38         m = md5()
  39         m.update(bytes(string_bytes))
  40         result = b64encode(m.digest()).decode('ascii')
  41         return result.replace('/', '_').replace('+', '-')
  42
  43     def make_player_api_request_data_and_headers(self, song_id, bitrate):
  44         KEY = b'e82ckenh8dichen8'
  45         URL = '/api/song/enhance/player/url'
  46         now = int(time.time() * 1000)
  47         rand = randint(0, 1000)
  48         cookie = {
  49             'osver': None,
  50             'deviceId': None,
  51             'appver': '8.0.0',
  52             'versioncode': '140',
  53             'mobilename': None,
  54             'buildver': '1623435496',
  55             'resolution': '1920x1080',
  56             '__csrf': '',
  57             'os': 'pc',
  58             'channel': None,
  59             'requestId': '{0}_{1:04}'.format(now, rand),
  60         }
  61         request_text = json.dumps(
  62             {'ids': '[{0}]'.format(song_id), 'br': bitrate, 'header': cookie},
  63             separators=(',', ':'))
  64         message = 'nobody{0}use{1}md5forencrypt'.format(
  65             URL, request_text).encode('latin1')
  66         msg_digest = md5(message).hexdigest()
  67
  68         data = '{0}-36cd479b6b5-{1}-36cd479b6b5-{2}'.format(
  69             URL, request_text, msg_digest)
  70         data = pkcs7_padding(bytes_to_intlist(data))
  71         encrypted = intlist_to_bytes(aes_ecb_encrypt(data, bytes_to_intlist(KEY)))
  72         encrypted_params = hexlify(encrypted).decode('ascii').upper()
  73
  74         cookie = '; '.join(
  75             ['{0}={1}'.format(k, v if v is not None else 'undefined')
  76              for [k, v] in cookie.items()])
  77
  78         headers = {
  79             'User-Agent': self.extractor.get_param('http_headers')['User-Agent'],
  80             'Content-Type': 'application/x-www-form-urlencoded',
  81             'Referer': 'https://music.163.com',
  82             'Cookie': cookie,
  83         }
  84         return ('params={0}'.format(encrypted_params), headers)
  85
  86     def _call_player_api(self, song_id, bitrate):
  87         url = 'https://interface3.music.163.com/eapi/song/enhance/player/url'
  88         data, headers = self.make_player_api_request_data_and_headers(song_id, bitrate)
  89         try:
  90             msg = 'empty result'
  91             result = self._download_json(
  92                 url, song_id, data=data.encode('ascii'), headers=headers)
  93             if result:
  94                 return result
  95         except ExtractorError as e:
  96             if type(e.cause) in (ValueError, TypeError):
  97                 # JSON load failure
  98                 raise
  99         except Exception as e:
 100             msg = error_to_compat_str(e)
 101             self.report_warning('%s API call (%s) failed: %s' % (
 102                 song_id, bitrate, msg))
 103         return {}
 104
 105     def extract_formats(self, info):
 106         err = 0
 107         formats = []
 108         song_id = info['id']
 109         for song_format in self._FORMATS:
 110             details = info.get(song_format)
 111             if not details:
 112                 continue
 113
 114             bitrate = int_or_none(details.get('bitrate')) or 999000
 115             data = self._call_player_api(song_id, bitrate)
 116             for song in try_get(data, lambda x: x['data'], list) or []:
 117                 song_url = try_get(song, lambda x: x['url'])
 118                 if not song_url:
 119                     continue
 120                 if self._is_valid_url(song_url, info['id'], 'song'):
 121                     formats.append({
 122                         'url': song_url,
 123                         'ext': details.get('extension'),
 124                         'abr': float_or_none(song.get('br'), scale=1000),
 125                         'format_id': song_format,
 126                         'filesize': int_or_none(song.get('size')),
 127                         'asr': int_or_none(details.get('sr')),
 128                     })
 129                 elif err == 0:
 130                     err = try_get(song, lambda x: x['code'], int)
 131
 132         if not formats:
 133             msg = 'No media links found'
 134             if err != 0 and (err < 200 or err >= 400):
 135                 raise ExtractorError(
 136                     '%s (site code %d)' % (msg, err, ), expected=True)
 137             else:
 138                 self.raise_geo_restricted(
 139                     msg + ': probably this video is not available from your location due to geo restriction.',
 140                     countries=['CN'])
 141
 142         return formats
 143
 144     @classmethod
 145     def convert_milliseconds(cls, ms):
 146         return int(round(ms / 1000.0))
 147
 148     def query_api(self, endpoint, video_id, note):
 149         req = sanitized_Request('%s%s' % (self._API_BASE, endpoint))
 150         req.add_header('Referer', self._API_BASE)
 151         return self._download_json(req, video_id, note)
 152
 153
 154 class NetEaseMusicIE(NetEaseMusicBaseIE):
 155     IE_NAME = 'netease:song'
 156     IE_DESC = '网易云音乐'
 157     _VALID_URL = r'https?://(y\.)?music\.163\.com/(?:[#m]/)?song\?.*?\bid=(?P<id>[0-9]+)'
 158     _TESTS = [{
 159         'url': 'http://music.163.com/#/song?id=32102397',
 160         'md5': '3e909614ce09b1ccef4a3eb205441190',
 161         'info_dict': {
 162             'id': '32102397',
 163             'ext': 'mp3',
 164             'title': 'Bad Blood',
 165             'creator': 'Taylor Swift / Kendrick Lamar',
 166             'upload_date': '20150516',
 167             'timestamp': 1431792000,
 168             'description': 'md5:25fc5f27e47aad975aa6d36382c7833c',
 169         },
 170     }, {
 171         'note': 'No lyrics.',
 172         'url': 'http://music.163.com/song?id=17241424',
 173         'info_dict': {
 174             'id': '17241424',
 175             'ext': 'mp3',
 176             'title': 'Opus 28',
 177             'creator': 'Dustin O\'Halloran',
 178             'upload_date': '20080211',
 179             'description': 'md5:f12945b0f6e0365e3b73c5032e1b0ff4',
 180             'timestamp': 1202745600,
 181         },
 182     }, {
 183         'note': 'Has translated name.',
 184         'url': 'http://music.163.com/#/song?id=22735043',
 185         'info_dict': {
 186             'id': '22735043',
 187             'ext': 'mp3',
 188             'title': '소원을 말해봐 (Genie)',
 189             'creator': '少女时代',
 190             'description': 'md5:79d99cc560e4ca97e0c4d86800ee4184',
 191             'upload_date': '20100127',
 192             'timestamp': 1264608000,
 193             'alt_title': '说出愿望吧(Genie)',
 194         },
 195     }, {
 196         'url': 'https://y.music.163.com/m/song?app_version=8.8.45&id=95670&uct2=sKnvS4+0YStsWkqsPhFijw%3D%3D&dlt=0846',
 197         'md5': '95826c73ea50b1c288b22180ec9e754d',
 198         'info_dict': {
 199             'id': '95670',
 200             'ext': 'mp3',
 201             'title': '国际歌',
 202             'creator': '马备',
 203             'upload_date': '19911130',
 204             'timestamp': 691516800,
 205             'description': 'md5:1ba2f911a2b0aa398479f595224f2141',
 206         },
 207     }]
 208
 209     def _process_lyrics(self, lyrics_info):
 210         original = lyrics_info.get('lrc', {}).get('lyric')
 211         translated = lyrics_info.get('tlyric', {}).get('lyric')
 212
 213         if not translated:
 214             return original
 215
 216         lyrics_expr = r'(\[[0-9]{2}:[0-9]{2}\.[0-9]{2,}\])([^\n]+)'
 217         original_ts_texts = re.findall(lyrics_expr, original)
 218         translation_ts_dict = dict(
 219             (time_stamp, text) for time_stamp, text in re.findall(lyrics_expr, translated)
 220         )
 221         lyrics = '\n'.join([
 222             '%s%s / %s' % (time_stamp, text, translation_ts_dict.get(time_stamp, ''))
 223             for time_stamp, text in original_ts_texts
 224         ])
 225         return lyrics
 226
 227     def _real_extract(self, url):
 228         song_id = self._match_id(url)
 229
 230         params = {
 231             'id': song_id,
 232             'ids': '[%s]' % song_id
 233         }
 234         info = self.query_api(
 235             'song/detail?' + compat_urllib_parse_urlencode(params),
 236             song_id, 'Downloading song info')['songs'][0]
 237
 238         formats = self.extract_formats(info)
 239
 240         lyrics_info = self.query_api(
 241             'song/lyric?id=%s&lv=-1&tv=-1' % song_id,
 242             song_id, 'Downloading lyrics data')
 243         lyrics = self._process_lyrics(lyrics_info)
 244
 245         alt_title = None
 246         if info.get('transNames'):
 247             alt_title = '/'.join(info.get('transNames'))
 248
 249         return {
 250             'id': song_id,
 251             'title': info['name'],
 252             'alt_title': alt_title,
 253             'creator': ' / '.join([artist['name'] for artist in info.get('artists', [])]),
 254             'timestamp': self.convert_milliseconds(info.get('album', {}).get('publishTime')),
 255             'thumbnail': info.get('album', {}).get('picUrl'),
 256             'duration': self.convert_milliseconds(info.get('duration', 0)),
 257             'description': lyrics,
 258             'formats': formats,
 259         }
 260
 261
 262 class NetEaseMusicAlbumIE(NetEaseMusicBaseIE):
 263     IE_NAME = 'netease:album'
 264     IE_DESC = '网易云音乐 - 专辑'
 265     _VALID_URL = r'https?://music\.163\.com/(#/)?album\?id=(?P<id>[0-9]+)'
 266     _TEST = {
 267         'url': 'http://music.163.com/#/album?id=220780',
 268         'info_dict': {
 269             'id': '220780',
 270             'title': 'B\'day',
 271         },
 272         'playlist_count': 23,
 273         'skip': 'Blocked outside Mainland China',
 274     }
 275
 276     def _real_extract(self, url):
 277         album_id = self._match_id(url)
 278
 279         info = self.query_api(
 280             'album/%s?id=%s' % (album_id, album_id),
 281             album_id, 'Downloading album data')['album']
 282
 283         name = info['name']
 284         desc = info.get('description')
 285         entries = [
 286             self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
 287                             'NetEaseMusic', song['id'])
 288             for song in info['songs']
 289         ]
 290         return self.playlist_result(entries, album_id, name, desc)
 291
 292
 293 class NetEaseMusicSingerIE(NetEaseMusicBaseIE):
 294     IE_NAME = 'netease:singer'
 295     IE_DESC = '网易云音乐 - 歌手'
 296     _VALID_URL = r'https?://music\.163\.com/(#/)?artist\?id=(?P<id>[0-9]+)'
 297     _TESTS = [{
 298         'note': 'Singer has aliases.',
 299         'url': 'http://music.163.com/#/artist?id=10559',
 300         'info_dict': {
 301             'id': '10559',
 302             'title': '张惠妹 - aMEI;阿密特',
 303         },
 304         'playlist_count': 50,
 305         'skip': 'Blocked outside Mainland China',
 306     }, {
 307         'note': 'Singer has translated name.',
 308         'url': 'http://music.163.com/#/artist?id=124098',
 309         'info_dict': {
 310             'id': '124098',
 311             'title': '李昇基 - 이승기',
 312         },
 313         'playlist_count': 50,
 314         'skip': 'Blocked outside Mainland China',
 315     }]
 316
 317     def _real_extract(self, url):
 318         singer_id = self._match_id(url)
 319
 320         info = self.query_api(
 321             'artist/%s?id=%s' % (singer_id, singer_id),
 322             singer_id, 'Downloading singer data')
 323
 324         name = info['artist']['name']
 325         if info['artist']['trans']:
 326             name = '%s - %s' % (name, info['artist']['trans'])
 327         if info['artist']['alias']:
 328             name = '%s - %s' % (name, ';'.join(info['artist']['alias']))
 329
 330         entries = [
 331             self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
 332                             'NetEaseMusic', song['id'])
 333             for song in info['hotSongs']
 334         ]
 335         return self.playlist_result(entries, singer_id, name)
 336
 337
 338 class NetEaseMusicListIE(NetEaseMusicBaseIE):
 339     IE_NAME = 'netease:playlist'
 340     IE_DESC = '网易云音乐 - 歌单'
 341     _VALID_URL = r'https?://music\.163\.com/(#/)?(playlist|discover/toplist)\?id=(?P<id>[0-9]+)'
 342     _TESTS = [{
 343         'url': 'http://music.163.com/#/playlist?id=79177352',
 344         'info_dict': {
 345             'id': '79177352',
 346             'title': 'Billboard 2007 Top 100',
 347             'description': 'md5:12fd0819cab2965b9583ace0f8b7b022'
 348         },
 349         'playlist_count': 99,
 350         'skip': 'Blocked outside Mainland China',
 351     }, {
 352         'note': 'Toplist/Charts sample',
 353         'url': 'http://music.163.com/#/discover/toplist?id=3733003',
 354         'info_dict': {
 355             'id': '3733003',
 356             'title': 're:韩国Melon排行榜周榜 [0-9]{4}-[0-9]{2}-[0-9]{2}',
 357             'description': 'md5:73ec782a612711cadc7872d9c1e134fc',
 358         },
 359         'playlist_count': 50,
 360         'skip': 'Blocked outside Mainland China',
 361     }]
 362
 363     def _real_extract(self, url):
 364         list_id = self._match_id(url)
 365
 366         info = self.query_api(
 367             'playlist/detail?id=%s&lv=-1&tv=-1' % list_id,
 368             list_id, 'Downloading playlist data')['result']
 369
 370         name = info['name']
 371         desc = info.get('description')
 372
 373         if info.get('specialType') == 10:  # is a chart/toplist
 374             datestamp = datetime.fromtimestamp(
 375                 self.convert_milliseconds(info['updateTime'])).strftime('%Y-%m-%d')
 376             name = '%s %s' % (name, datestamp)
 377
 378         entries = [
 379             self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
 380                             'NetEaseMusic', song['id'])
 381             for song in info['tracks']
 382         ]
 383         return self.playlist_result(entries, list_id, name, desc)
 384
 385
 386 class NetEaseMusicMvIE(NetEaseMusicBaseIE):
 387     IE_NAME = 'netease:mv'
 388     IE_DESC = '网易云音乐 - MV'
 389     _VALID_URL = r'https?://music\.163\.com/(#/)?mv\?id=(?P<id>[0-9]+)'
 390     _TEST = {
 391         'url': 'http://music.163.com/#/mv?id=415350',
 392         'info_dict': {
 393             'id': '415350',
 394             'ext': 'mp4',
 395             'title': '이럴거면 그러지말지',
 396             'description': '白雅言自作曲唱甜蜜爱情',
 397             'creator': '白雅言',
 398             'upload_date': '20150520',
 399         },
 400         'skip': 'Blocked outside Mainland China',
 401     }
 402
 403     def _real_extract(self, url):
 404         mv_id = self._match_id(url)
 405
 406         info = self.query_api(
 407             'mv/detail?id=%s&type=mp4' % mv_id,
 408             mv_id, 'Downloading mv info')['data']
 409
 410         formats = [
 411             {'url': mv_url, 'ext': 'mp4', 'format_id': '%sp' % brs, 'height': int(brs)}
 412             for brs, mv_url in info['brs'].items()
 413         ]
 414
 415         return {
 416             'id': mv_id,
 417             'title': info['name'],
 418             'description': info.get('desc') or info.get('briefDesc'),
 419             'creator': info['artistName'],
 420             'upload_date': info['publishTime'].replace('-', ''),
 421             'formats': formats,
 422             'thumbnail': info.get('cover'),
 423             'duration': self.convert_milliseconds(info.get('duration', 0)),
 424         }
 425
 426
 427 class NetEaseMusicProgramIE(NetEaseMusicBaseIE):
 428     IE_NAME = 'netease:program'
 429     IE_DESC = '网易云音乐 - 电台节目'
 430     _VALID_URL = r'https?://music\.163\.com/(#/?)program\?id=(?P<id>[0-9]+)'
 431     _TESTS = [{
 432         'url': 'http://music.163.com/#/program?id=10109055',
 433         'info_dict': {
 434             'id': '10109055',
 435             'ext': 'mp3',
 436             'title': '不丹足球背后的故事',
 437             'description': '喜马拉雅人的足球梦 ...',
 438             'creator': '大话西藏',
 439             'timestamp': 1434179342,
 440             'upload_date': '20150613',
 441             'duration': 900,
 442         },
 443         'skip': 'Blocked outside Mainland China',
 444     }, {
 445         'note': 'This program has accompanying songs.',
 446         'url': 'http://music.163.com/#/program?id=10141022',
 447         'info_dict': {
 448             'id': '10141022',
 449             'title': '25岁，你是自在如风的少年<27°C>',
 450             'description': 'md5:8d594db46cc3e6509107ede70a4aaa3b',
 451         },
 452         'playlist_count': 4,
 453         'skip': 'Blocked outside Mainland China',
 454     }, {
 455         'note': 'This program has accompanying songs.',
 456         'url': 'http://music.163.com/#/program?id=10141022',
 457         'info_dict': {
 458             'id': '10141022',
 459             'ext': 'mp3',
 460             'title': '25岁，你是自在如风的少年<27°C>',
 461             'description': 'md5:8d594db46cc3e6509107ede70a4aaa3b',
 462             'timestamp': 1434450841,
 463             'upload_date': '20150616',
 464         },
 465         'params': {
 466             'noplaylist': True
 467         },
 468         'skip': 'Blocked outside Mainland China',
 469     }]
 470
 471     def _real_extract(self, url):
 472         program_id = self._match_id(url)
 473
 474         info = self.query_api(
 475             'dj/program/detail?id=%s' % program_id,
 476             program_id, 'Downloading program info')['program']
 477
 478         name = info['name']
 479         description = info['description']
 480
 481         if not self._yes_playlist(info['songs'] and program_id, info['mainSong']['id']):
 482             formats = self.extract_formats(info['mainSong'])
 483
 484             return {
 485                 'id': info['mainSong']['id'],
 486                 'title': name,
 487                 'description': description,
 488                 'creator': info['dj']['brand'],
 489                 'timestamp': self.convert_milliseconds(info['createTime']),
 490                 'thumbnail': info['coverUrl'],
 491                 'duration': self.convert_milliseconds(info.get('duration', 0)),
 492                 'formats': formats,
 493             }
 494
 495         song_ids = [info['mainSong']['id']]
 496         song_ids.extend([song['id'] for song in info['songs']])
 497         entries = [
 498             self.url_result('http://music.163.com/#/song?id=%s' % song_id,
 499                             'NetEaseMusic', song_id)
 500             for song_id in song_ids
 501         ]
 502         return self.playlist_result(entries, program_id, name, description)
 503
 504
 505 class NetEaseMusicDjRadioIE(NetEaseMusicBaseIE):
 506     IE_NAME = 'netease:djradio'
 507     IE_DESC = '网易云音乐 - 电台'
 508     _VALID_URL = r'https?://music\.163\.com/(#/)?djradio\?id=(?P<id>[0-9]+)'
 509     _TEST = {
 510         'url': 'http://music.163.com/#/djradio?id=42',
 511         'info_dict': {
 512             'id': '42',
 513             'title': '声音蔓延',
 514             'description': 'md5:766220985cbd16fdd552f64c578a6b15'
 515         },
 516         'playlist_mincount': 40,
 517         'skip': 'Blocked outside Mainland China',
 518     }
 519     _PAGE_SIZE = 1000
 520
 521     def _real_extract(self, url):
 522         dj_id = self._match_id(url)
 523
 524         name = None
 525         desc = None
 526         entries = []
 527         for offset in itertools.count(start=0, step=self._PAGE_SIZE):
 528             info = self.query_api(
 529                 'dj/program/byradio?asc=false&limit=%d&radioId=%s&offset=%d'
 530                 % (self._PAGE_SIZE, dj_id, offset),
 531                 dj_id, 'Downloading dj programs - %d' % offset)
 532
 533             entries.extend([
 534                 self.url_result(
 535                     'http://music.163.com/#/program?id=%s' % program['id'],
 536                     'NetEaseMusicProgram', program['id'])
 537                 for program in info['programs']
 538             ])
 539
 540             if name is None:
 541                 radio = info['programs'][0]['radio']
 542                 name = radio['name']
 543                 desc = radio['desc']
 544
 545             if not info['more']:
 546                 break
 547
 548         return self.playlist_result(entries, dj_id, name, desc)