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         self._sort_formats(formats)
 240
 241         lyrics_info = self.query_api(
 242             'song/lyric?id=%s&lv=-1&tv=-1' % song_id,
 243             song_id, 'Downloading lyrics data')
 244         lyrics = self._process_lyrics(lyrics_info)
 245
 246         alt_title = None
 247         if info.get('transNames'):
 248             alt_title = '/'.join(info.get('transNames'))
 249
 250         return {
 251             'id': song_id,
 252             'title': info['name'],
 253             'alt_title': alt_title,
 254             'creator': ' / '.join([artist['name'] for artist in info.get('artists', [])]),
 255             'timestamp': self.convert_milliseconds(info.get('album', {}).get('publishTime')),
 256             'thumbnail': info.get('album', {}).get('picUrl'),
 257             'duration': self.convert_milliseconds(info.get('duration', 0)),
 258             'description': lyrics,
 259             'formats': formats,
 260         }
 261
 262
 263 class NetEaseMusicAlbumIE(NetEaseMusicBaseIE):
 264     IE_NAME = 'netease:album'
 265     IE_DESC = '网易云音乐 - 专辑'
 266     _VALID_URL = r'https?://music\.163\.com/(#/)?album\?id=(?P<id>[0-9]+)'
 267     _TEST = {
 268         'url': 'http://music.163.com/#/album?id=220780',
 269         'info_dict': {
 270             'id': '220780',
 271             'title': 'B\'day',
 272         },
 273         'playlist_count': 23,
 274         'skip': 'Blocked outside Mainland China',
 275     }
 276
 277     def _real_extract(self, url):
 278         album_id = self._match_id(url)
 279
 280         info = self.query_api(
 281             'album/%s?id=%s' % (album_id, album_id),
 282             album_id, 'Downloading album data')['album']
 283
 284         name = info['name']
 285         desc = info.get('description')
 286         entries = [
 287             self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
 288                             'NetEaseMusic', song['id'])
 289             for song in info['songs']
 290         ]
 291         return self.playlist_result(entries, album_id, name, desc)
 292
 293
 294 class NetEaseMusicSingerIE(NetEaseMusicBaseIE):
 295     IE_NAME = 'netease:singer'
 296     IE_DESC = '网易云音乐 - 歌手'
 297     _VALID_URL = r'https?://music\.163\.com/(#/)?artist\?id=(?P<id>[0-9]+)'
 298     _TESTS = [{
 299         'note': 'Singer has aliases.',
 300         'url': 'http://music.163.com/#/artist?id=10559',
 301         'info_dict': {
 302             'id': '10559',
 303             'title': '张惠妹 - aMEI;阿密特',
 304         },
 305         'playlist_count': 50,
 306         'skip': 'Blocked outside Mainland China',
 307     }, {
 308         'note': 'Singer has translated name.',
 309         'url': 'http://music.163.com/#/artist?id=124098',
 310         'info_dict': {
 311             'id': '124098',
 312             'title': '李昇基 - 이승기',
 313         },
 314         'playlist_count': 50,
 315         'skip': 'Blocked outside Mainland China',
 316     }]
 317
 318     def _real_extract(self, url):
 319         singer_id = self._match_id(url)
 320
 321         info = self.query_api(
 322             'artist/%s?id=%s' % (singer_id, singer_id),
 323             singer_id, 'Downloading singer data')
 324
 325         name = info['artist']['name']
 326         if info['artist']['trans']:
 327             name = '%s - %s' % (name, info['artist']['trans'])
 328         if info['artist']['alias']:
 329             name = '%s - %s' % (name, ';'.join(info['artist']['alias']))
 330
 331         entries = [
 332             self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
 333                             'NetEaseMusic', song['id'])
 334             for song in info['hotSongs']
 335         ]
 336         return self.playlist_result(entries, singer_id, name)
 337
 338
 339 class NetEaseMusicListIE(NetEaseMusicBaseIE):
 340     IE_NAME = 'netease:playlist'
 341     IE_DESC = '网易云音乐 - 歌单'
 342     _VALID_URL = r'https?://music\.163\.com/(#/)?(playlist|discover/toplist)\?id=(?P<id>[0-9]+)'
 343     _TESTS = [{
 344         'url': 'http://music.163.com/#/playlist?id=79177352',
 345         'info_dict': {
 346             'id': '79177352',
 347             'title': 'Billboard 2007 Top 100',
 348             'description': 'md5:12fd0819cab2965b9583ace0f8b7b022'
 349         },
 350         'playlist_count': 99,
 351         'skip': 'Blocked outside Mainland China',
 352     }, {
 353         'note': 'Toplist/Charts sample',
 354         'url': 'http://music.163.com/#/discover/toplist?id=3733003',
 355         'info_dict': {
 356             'id': '3733003',
 357             'title': 're:韩国Melon排行榜周榜 [0-9]{4}-[0-9]{2}-[0-9]{2}',
 358             'description': 'md5:73ec782a612711cadc7872d9c1e134fc',
 359         },
 360         'playlist_count': 50,
 361         'skip': 'Blocked outside Mainland China',
 362     }]
 363
 364     def _real_extract(self, url):
 365         list_id = self._match_id(url)
 366
 367         info = self.query_api(
 368             'playlist/detail?id=%s&lv=-1&tv=-1' % list_id,
 369             list_id, 'Downloading playlist data')['result']
 370
 371         name = info['name']
 372         desc = info.get('description')
 373
 374         if info.get('specialType') == 10:  # is a chart/toplist
 375             datestamp = datetime.fromtimestamp(
 376                 self.convert_milliseconds(info['updateTime'])).strftime('%Y-%m-%d')
 377             name = '%s %s' % (name, datestamp)
 378
 379         entries = [
 380             self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
 381                             'NetEaseMusic', song['id'])
 382             for song in info['tracks']
 383         ]
 384         return self.playlist_result(entries, list_id, name, desc)
 385
 386
 387 class NetEaseMusicMvIE(NetEaseMusicBaseIE):
 388     IE_NAME = 'netease:mv'
 389     IE_DESC = '网易云音乐 - MV'
 390     _VALID_URL = r'https?://music\.163\.com/(#/)?mv\?id=(?P<id>[0-9]+)'
 391     _TEST = {
 392         'url': 'http://music.163.com/#/mv?id=415350',
 393         'info_dict': {
 394             'id': '415350',
 395             'ext': 'mp4',
 396             'title': '이럴거면 그러지말지',
 397             'description': '白雅言自作曲唱甜蜜爱情',
 398             'creator': '白雅言',
 399             'upload_date': '20150520',
 400         },
 401         'skip': 'Blocked outside Mainland China',
 402     }
 403
 404     def _real_extract(self, url):
 405         mv_id = self._match_id(url)
 406
 407         info = self.query_api(
 408             'mv/detail?id=%s&type=mp4' % mv_id,
 409             mv_id, 'Downloading mv info')['data']
 410
 411         formats = [
 412             {'url': mv_url, 'ext': 'mp4', 'format_id': '%sp' % brs, 'height': int(brs)}
 413             for brs, mv_url in info['brs'].items()
 414         ]
 415         self._sort_formats(formats)
 416
 417         return {
 418             'id': mv_id,
 419             'title': info['name'],
 420             'description': info.get('desc') or info.get('briefDesc'),
 421             'creator': info['artistName'],
 422             'upload_date': info['publishTime'].replace('-', ''),
 423             'formats': formats,
 424             'thumbnail': info.get('cover'),
 425             'duration': self.convert_milliseconds(info.get('duration', 0)),
 426         }
 427
 428
 429 class NetEaseMusicProgramIE(NetEaseMusicBaseIE):
 430     IE_NAME = 'netease:program'
 431     IE_DESC = '网易云音乐 - 电台节目'
 432     _VALID_URL = r'https?://music\.163\.com/(#/?)program\?id=(?P<id>[0-9]+)'
 433     _TESTS = [{
 434         'url': 'http://music.163.com/#/program?id=10109055',
 435         'info_dict': {
 436             'id': '10109055',
 437             'ext': 'mp3',
 438             'title': '不丹足球背后的故事',
 439             'description': '喜马拉雅人的足球梦 ...',
 440             'creator': '大话西藏',
 441             'timestamp': 1434179342,
 442             'upload_date': '20150613',
 443             'duration': 900,
 444         },
 445         'skip': 'Blocked outside Mainland China',
 446     }, {
 447         'note': 'This program has accompanying songs.',
 448         'url': 'http://music.163.com/#/program?id=10141022',
 449         'info_dict': {
 450             'id': '10141022',
 451             'title': '25岁，你是自在如风的少年<27°C>',
 452             'description': 'md5:8d594db46cc3e6509107ede70a4aaa3b',
 453         },
 454         'playlist_count': 4,
 455         'skip': 'Blocked outside Mainland China',
 456     }, {
 457         'note': 'This program has accompanying songs.',
 458         'url': 'http://music.163.com/#/program?id=10141022',
 459         'info_dict': {
 460             'id': '10141022',
 461             'ext': 'mp3',
 462             'title': '25岁，你是自在如风的少年<27°C>',
 463             'description': 'md5:8d594db46cc3e6509107ede70a4aaa3b',
 464             'timestamp': 1434450841,
 465             'upload_date': '20150616',
 466         },
 467         'params': {
 468             'noplaylist': True
 469         },
 470         'skip': 'Blocked outside Mainland China',
 471     }]
 472
 473     def _real_extract(self, url):
 474         program_id = self._match_id(url)
 475
 476         info = self.query_api(
 477             'dj/program/detail?id=%s' % program_id,
 478             program_id, 'Downloading program info')['program']
 479
 480         name = info['name']
 481         description = info['description']
 482
 483         if not self._yes_playlist(info['songs'] and program_id, info['mainSong']['id']):
 484             formats = self.extract_formats(info['mainSong'])
 485             self._sort_formats(formats)
 486
 487             return {
 488                 'id': info['mainSong']['id'],
 489                 'title': name,
 490                 'description': description,
 491                 'creator': info['dj']['brand'],
 492                 'timestamp': self.convert_milliseconds(info['createTime']),
 493                 'thumbnail': info['coverUrl'],
 494                 'duration': self.convert_milliseconds(info.get('duration', 0)),
 495                 'formats': formats,
 496             }
 497
 498         song_ids = [info['mainSong']['id']]
 499         song_ids.extend([song['id'] for song in info['songs']])
 500         entries = [
 501             self.url_result('http://music.163.com/#/song?id=%s' % song_id,
 502                             'NetEaseMusic', song_id)
 503             for song_id in song_ids
 504         ]
 505         return self.playlist_result(entries, program_id, name, description)
 506
 507
 508 class NetEaseMusicDjRadioIE(NetEaseMusicBaseIE):
 509     IE_NAME = 'netease:djradio'
 510     IE_DESC = '网易云音乐 - 电台'
 511     _VALID_URL = r'https?://music\.163\.com/(#/)?djradio\?id=(?P<id>[0-9]+)'
 512     _TEST = {
 513         'url': 'http://music.163.com/#/djradio?id=42',
 514         'info_dict': {
 515             'id': '42',
 516             'title': '声音蔓延',
 517             'description': 'md5:766220985cbd16fdd552f64c578a6b15'
 518         },
 519         'playlist_mincount': 40,
 520         'skip': 'Blocked outside Mainland China',
 521     }
 522     _PAGE_SIZE = 1000
 523
 524     def _real_extract(self, url):
 525         dj_id = self._match_id(url)
 526
 527         name = None
 528         desc = None
 529         entries = []
 530         for offset in itertools.count(start=0, step=self._PAGE_SIZE):
 531             info = self.query_api(
 532                 'dj/program/byradio?asc=false&limit=%d&radioId=%s&offset=%d'
 533                 % (self._PAGE_SIZE, dj_id, offset),
 534                 dj_id, 'Downloading dj programs - %d' % offset)
 535
 536             entries.extend([
 537                 self.url_result(
 538                     'http://music.163.com/#/program?id=%s' % program['id'],
 539                     'NetEaseMusicProgram', program['id'])
 540                 for program in info['programs']
 541             ])
 542
 543             if name is None:
 544                 radio = info['programs'][0]['radio']
 545                 name = radio['name']
 546                 desc = radio['desc']
 547
 548             if not info['more']:
 549                 break
 550
 551         return self.playlist_result(entries, dj_id, name, desc)