Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/neteasemusic.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from hashlib import md5
   5 from base64 import b64encode
   6 from datetime import datetime
   7 import re
   8
   9 from .common import InfoExtractor
  10 from ..compat import (
  11     compat_urllib_parse,
  12     compat_str,
  13     compat_itertools_count,
  14 )
  15 from ..utils import (
  16     sanitized_Request,
  17     float_or_none,
  18 )
  19
  20
  21 class NetEaseMusicBaseIE(InfoExtractor):
  22     _FORMATS = ['bMusic', 'mMusic', 'hMusic']
  23     _NETEASE_SALT = '3go8&$8*3*3h0k(2)2'
  24     _API_BASE = 'http://music.163.com/api/'
  25
  26     @classmethod
  27     def _encrypt(cls, dfsid):
  28         salt_bytes = bytearray(cls._NETEASE_SALT.encode('utf-8'))
  29         string_bytes = bytearray(compat_str(dfsid).encode('ascii'))
  30         salt_len = len(salt_bytes)
  31         for i in range(len(string_bytes)):
  32             string_bytes[i] = string_bytes[i] ^ salt_bytes[i % salt_len]
  33         m = md5()
  34         m.update(bytes(string_bytes))
  35         result = b64encode(m.digest()).decode('ascii')
  36         return result.replace('/', '_').replace('+', '-')
  37
  38     def extract_formats(self, info):
  39         formats = []
  40         for song_format in self._FORMATS:
  41             details = info.get(song_format)
  42             if not details:
  43                 continue
  44             song_file_path = '/%s/%s.%s' % (
  45                 self._encrypt(details['dfsId']), details['dfsId'], details['extension'])
  46
  47             # 203.130.59.9, 124.40.233.182, 115.231.74.139, etc is a reverse proxy-like feature
  48             # from NetEase's CDN provider that can be used if m5.music.126.net does not
  49             # work, especially for users outside of Mainland China
  50             # via: https://github.com/JixunMoe/unblock-163/issues/3#issuecomment-163115880
  51             for host in ('http://m5.music.126.net', 'http://115.231.74.139/m1.music.126.net',
  52                          'http://124.40.233.182/m1.music.126.net', 'http://203.130.59.9/m1.music.126.net'):
  53                 song_url = host + song_file_path
  54                 if self._is_valid_url(song_url, info['id'], 'song'):
  55                     formats.append({
  56                         'url': song_url,
  57                         'ext': details.get('extension'),
  58                         'abr': float_or_none(details.get('bitrate'), scale=1000),
  59                         'format_id': song_format,
  60                         'filesize': details.get('size'),
  61                         'asr': details.get('sr')
  62                     })
  63                     break
  64         return formats
  65
  66     @classmethod
  67     def convert_milliseconds(cls, ms):
  68         return int(round(ms / 1000.0))
  69
  70     def query_api(self, endpoint, video_id, note):
  71         req = sanitized_Request('%s%s' % (self._API_BASE, endpoint))
  72         req.add_header('Referer', self._API_BASE)
  73         return self._download_json(req, video_id, note)
  74
  75
  76 class NetEaseMusicIE(NetEaseMusicBaseIE):
  77     IE_NAME = 'netease:song'
  78     IE_DESC = '网易云音乐'
  79     _VALID_URL = r'https?://music\.163\.com/(#/)?song\?id=(?P<id>[0-9]+)'
  80     _TESTS = [{
  81         'url': 'http://music.163.com/#/song?id=32102397',
  82         'md5': 'f2e97280e6345c74ba9d5677dd5dcb45',
  83         'info_dict': {
  84             'id': '32102397',
  85             'ext': 'mp3',
  86             'title': 'Bad Blood (feat. Kendrick Lamar)',
  87             'creator': 'Taylor Swift / Kendrick Lamar',
  88             'upload_date': '20150517',
  89             'timestamp': 1431878400,
  90             'description': 'md5:a10a54589c2860300d02e1de821eb2ef',
  91         },
  92     }, {
  93         'note': 'No lyrics translation.',
  94         'url': 'http://music.163.com/#/song?id=29822014',
  95         'info_dict': {
  96             'id': '29822014',
  97             'ext': 'mp3',
  98             'title': '听见下雨的声音',
  99             'creator': '周杰伦',
 100             'upload_date': '20141225',
 101             'timestamp': 1419523200,
 102             'description': 'md5:a4d8d89f44656af206b7b2555c0bce6c',
 103         },
 104     }, {
 105         'note': 'No lyrics.',
 106         'url': 'http://music.163.com/song?id=17241424',
 107         'info_dict': {
 108             'id': '17241424',
 109             'ext': 'mp3',
 110             'title': 'Opus 28',
 111             'creator': 'Dustin O\'Halloran',
 112             'upload_date': '20080211',
 113             'timestamp': 1202745600,
 114         },
 115     }, {
 116         'note': 'Has translated name.',
 117         'url': 'http://music.163.com/#/song?id=22735043',
 118         'info_dict': {
 119             'id': '22735043',
 120             'ext': 'mp3',
 121             'title': '소원을 말해봐 (Genie)',
 122             'creator': '少女时代',
 123             'description': 'md5:79d99cc560e4ca97e0c4d86800ee4184',
 124             'upload_date': '20100127',
 125             'timestamp': 1264608000,
 126             'alt_title': '说出愿望吧(Genie)',
 127         }
 128     }]
 129
 130     def _process_lyrics(self, lyrics_info):
 131         original = lyrics_info.get('lrc', {}).get('lyric')
 132         translated = lyrics_info.get('tlyric', {}).get('lyric')
 133
 134         if not translated:
 135             return original
 136
 137         lyrics_expr = r'(\[[0-9]{2}:[0-9]{2}\.[0-9]{2,}\])([^\n]+)'
 138         original_ts_texts = re.findall(lyrics_expr, original)
 139         translation_ts_dict = dict(
 140             (time_stamp, text) for time_stamp, text in re.findall(lyrics_expr, translated)
 141         )
 142         lyrics = '\n'.join([
 143             '%s%s / %s' % (time_stamp, text, translation_ts_dict.get(time_stamp, ''))
 144             for time_stamp, text in original_ts_texts
 145         ])
 146         return lyrics
 147
 148     def _real_extract(self, url):
 149         song_id = self._match_id(url)
 150
 151         params = {
 152             'id': song_id,
 153             'ids': '[%s]' % song_id
 154         }
 155         info = self.query_api(
 156             'song/detail?' + compat_urllib_parse.urlencode(params),
 157             song_id, 'Downloading song info')['songs'][0]
 158
 159         formats = self.extract_formats(info)
 160         self._sort_formats(formats)
 161
 162         lyrics_info = self.query_api(
 163             'song/lyric?id=%s&lv=-1&tv=-1' % song_id,
 164             song_id, 'Downloading lyrics data')
 165         lyrics = self._process_lyrics(lyrics_info)
 166
 167         alt_title = None
 168         if info.get('transNames'):
 169             alt_title = '/'.join(info.get('transNames'))
 170
 171         return {
 172             'id': song_id,
 173             'title': info['name'],
 174             'alt_title': alt_title,
 175             'creator': ' / '.join([artist['name'] for artist in info.get('artists', [])]),
 176             'timestamp': self.convert_milliseconds(info.get('album', {}).get('publishTime')),
 177             'thumbnail': info.get('album', {}).get('picUrl'),
 178             'duration': self.convert_milliseconds(info.get('duration', 0)),
 179             'description': lyrics,
 180             'formats': formats,
 181         }
 182
 183
 184 class NetEaseMusicAlbumIE(NetEaseMusicBaseIE):
 185     IE_NAME = 'netease:album'
 186     IE_DESC = '网易云音乐 - 专辑'
 187     _VALID_URL = r'https?://music\.163\.com/(#/)?album\?id=(?P<id>[0-9]+)'
 188     _TEST = {
 189         'url': 'http://music.163.com/#/album?id=220780',
 190         'info_dict': {
 191             'id': '220780',
 192             'title': 'B\'day',
 193         },
 194         'playlist_count': 23,
 195     }
 196
 197     def _real_extract(self, url):
 198         album_id = self._match_id(url)
 199
 200         info = self.query_api(
 201             'album/%s?id=%s' % (album_id, album_id),
 202             album_id, 'Downloading album data')['album']
 203
 204         name = info['name']
 205         desc = info.get('description')
 206         entries = [
 207             self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
 208                             'NetEaseMusic', song['id'])
 209             for song in info['songs']
 210         ]
 211         return self.playlist_result(entries, album_id, name, desc)
 212
 213
 214 class NetEaseMusicSingerIE(NetEaseMusicBaseIE):
 215     IE_NAME = 'netease:singer'
 216     IE_DESC = '网易云音乐 - 歌手'
 217     _VALID_URL = r'https?://music\.163\.com/(#/)?artist\?id=(?P<id>[0-9]+)'
 218     _TESTS = [{
 219         'note': 'Singer has aliases.',
 220         'url': 'http://music.163.com/#/artist?id=10559',
 221         'info_dict': {
 222             'id': '10559',
 223             'title': '张惠妹 - aMEI;阿密特',
 224         },
 225         'playlist_count': 50,
 226     }, {
 227         'note': 'Singer has translated name.',
 228         'url': 'http://music.163.com/#/artist?id=124098',
 229         'info_dict': {
 230             'id': '124098',
 231             'title': '李昇基 - 이승기',
 232         },
 233         'playlist_count': 50,
 234     }]
 235
 236     def _real_extract(self, url):
 237         singer_id = self._match_id(url)
 238
 239         info = self.query_api(
 240             'artist/%s?id=%s' % (singer_id, singer_id),
 241             singer_id, 'Downloading singer data')
 242
 243         name = info['artist']['name']
 244         if info['artist']['trans']:
 245             name = '%s - %s' % (name, info['artist']['trans'])
 246         if info['artist']['alias']:
 247             name = '%s - %s' % (name, ';'.join(info['artist']['alias']))
 248
 249         entries = [
 250             self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
 251                             'NetEaseMusic', song['id'])
 252             for song in info['hotSongs']
 253         ]
 254         return self.playlist_result(entries, singer_id, name)
 255
 256
 257 class NetEaseMusicListIE(NetEaseMusicBaseIE):
 258     IE_NAME = 'netease:playlist'
 259     IE_DESC = '网易云音乐 - 歌单'
 260     _VALID_URL = r'https?://music\.163\.com/(#/)?(playlist|discover/toplist)\?id=(?P<id>[0-9]+)'
 261     _TESTS = [{
 262         'url': 'http://music.163.com/#/playlist?id=79177352',
 263         'info_dict': {
 264             'id': '79177352',
 265             'title': 'Billboard 2007 Top 100',
 266             'description': 'md5:12fd0819cab2965b9583ace0f8b7b022'
 267         },
 268         'playlist_count': 99,
 269     }, {
 270         'note': 'Toplist/Charts sample',
 271         'url': 'http://music.163.com/#/discover/toplist?id=3733003',
 272         'info_dict': {
 273             'id': '3733003',
 274             'title': 're:韩国Melon排行榜周榜 [0-9]{4}-[0-9]{2}-[0-9]{2}',
 275             'description': 'md5:73ec782a612711cadc7872d9c1e134fc',
 276         },
 277         'playlist_count': 50,
 278     }]
 279
 280     def _real_extract(self, url):
 281         list_id = self._match_id(url)
 282
 283         info = self.query_api(
 284             'playlist/detail?id=%s&lv=-1&tv=-1' % list_id,
 285             list_id, 'Downloading playlist data')['result']
 286
 287         name = info['name']
 288         desc = info.get('description')
 289
 290         if info.get('specialType') == 10:  # is a chart/toplist
 291             datestamp = datetime.fromtimestamp(
 292                 self.convert_milliseconds(info['updateTime'])).strftime('%Y-%m-%d')
 293             name = '%s %s' % (name, datestamp)
 294
 295         entries = [
 296             self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
 297                             'NetEaseMusic', song['id'])
 298             for song in info['tracks']
 299         ]
 300         return self.playlist_result(entries, list_id, name, desc)
 301
 302
 303 class NetEaseMusicMvIE(NetEaseMusicBaseIE):
 304     IE_NAME = 'netease:mv'
 305     IE_DESC = '网易云音乐 - MV'
 306     _VALID_URL = r'https?://music\.163\.com/(#/)?mv\?id=(?P<id>[0-9]+)'
 307     _TEST = {
 308         'url': 'http://music.163.com/#/mv?id=415350',
 309         'info_dict': {
 310             'id': '415350',
 311             'ext': 'mp4',
 312             'title': '이럴거면 그러지말지',
 313             'description': '白雅言自作曲唱甜蜜爱情',
 314             'creator': '白雅言',
 315             'upload_date': '20150520',
 316         },
 317     }
 318
 319     def _real_extract(self, url):
 320         mv_id = self._match_id(url)
 321
 322         info = self.query_api(
 323             'mv/detail?id=%s&type=mp4' % mv_id,
 324             mv_id, 'Downloading mv info')['data']
 325
 326         formats = [
 327             {'url': mv_url, 'ext': 'mp4', 'format_id': '%sp' % brs, 'height': int(brs)}
 328             for brs, mv_url in info['brs'].items()
 329         ]
 330         self._sort_formats(formats)
 331
 332         return {
 333             'id': mv_id,
 334             'title': info['name'],
 335             'description': info.get('desc') or info.get('briefDesc'),
 336             'creator': info['artistName'],
 337             'upload_date': info['publishTime'].replace('-', ''),
 338             'formats': formats,
 339             'thumbnail': info.get('cover'),
 340             'duration': self.convert_milliseconds(info.get('duration', 0)),
 341         }
 342
 343
 344 class NetEaseMusicProgramIE(NetEaseMusicBaseIE):
 345     IE_NAME = 'netease:program'
 346     IE_DESC = '网易云音乐 - 电台节目'
 347     _VALID_URL = r'https?://music\.163\.com/(#/?)program\?id=(?P<id>[0-9]+)'
 348     _TESTS = [{
 349         'url': 'http://music.163.com/#/program?id=10109055',
 350         'info_dict': {
 351             'id': '10109055',
 352             'ext': 'mp3',
 353             'title': '不丹足球背后的故事',
 354             'description': '喜马拉雅人的足球梦 ...',
 355             'creator': '大话西藏',
 356             'timestamp': 1434179342,
 357             'upload_date': '20150613',
 358             'duration': 900,
 359         },
 360     }, {
 361         'note': 'This program has accompanying songs.',
 362         'url': 'http://music.163.com/#/program?id=10141022',
 363         'info_dict': {
 364             'id': '10141022',
 365             'title': '25岁，你是自在如风的少年<27°C>',
 366             'description': 'md5:8d594db46cc3e6509107ede70a4aaa3b',
 367         },
 368         'playlist_count': 4,
 369     }, {
 370         'note': 'This program has accompanying songs.',
 371         'url': 'http://music.163.com/#/program?id=10141022',
 372         'info_dict': {
 373             'id': '10141022',
 374             'ext': 'mp3',
 375             'title': '25岁，你是自在如风的少年<27°C>',
 376             'description': 'md5:8d594db46cc3e6509107ede70a4aaa3b',
 377             'timestamp': 1434450841,
 378             'upload_date': '20150616',
 379         },
 380         'params': {
 381             'noplaylist': True
 382         }
 383     }]
 384
 385     def _real_extract(self, url):
 386         program_id = self._match_id(url)
 387
 388         info = self.query_api(
 389             'dj/program/detail?id=%s' % program_id,
 390             program_id, 'Downloading program info')['program']
 391
 392         name = info['name']
 393         description = info['description']
 394
 395         if not info['songs'] or self._downloader.params.get('noplaylist'):
 396             if info['songs']:
 397                 self.to_screen(
 398                     'Downloading just the main audio %s because of --no-playlist'
 399                     % info['mainSong']['id'])
 400
 401             formats = self.extract_formats(info['mainSong'])
 402             self._sort_formats(formats)
 403
 404             return {
 405                 'id': program_id,
 406                 'title': name,
 407                 'description': description,
 408                 'creator': info['dj']['brand'],
 409                 'timestamp': self.convert_milliseconds(info['createTime']),
 410                 'thumbnail': info['coverUrl'],
 411                 'duration': self.convert_milliseconds(info.get('duration', 0)),
 412                 'formats': formats,
 413             }
 414
 415         self.to_screen(
 416             'Downloading playlist %s - add --no-playlist to just download the main audio %s'
 417             % (program_id, info['mainSong']['id']))
 418
 419         song_ids = [info['mainSong']['id']]
 420         song_ids.extend([song['id'] for song in info['songs']])
 421         entries = [
 422             self.url_result('http://music.163.com/#/song?id=%s' % song_id,
 423                             'NetEaseMusic', song_id)
 424             for song_id in song_ids
 425         ]
 426         return self.playlist_result(entries, program_id, name, description)
 427
 428
 429 class NetEaseMusicDjRadioIE(NetEaseMusicBaseIE):
 430     IE_NAME = 'netease:djradio'
 431     IE_DESC = '网易云音乐 - 电台'
 432     _VALID_URL = r'https?://music\.163\.com/(#/)?djradio\?id=(?P<id>[0-9]+)'
 433     _TEST = {
 434         'url': 'http://music.163.com/#/djradio?id=42',
 435         'info_dict': {
 436             'id': '42',
 437             'title': '声音蔓延',
 438             'description': 'md5:766220985cbd16fdd552f64c578a6b15'
 439         },
 440         'playlist_mincount': 40,
 441     }
 442     _PAGE_SIZE = 1000
 443
 444     def _real_extract(self, url):
 445         dj_id = self._match_id(url)
 446
 447         name = None
 448         desc = None
 449         entries = []
 450         for offset in compat_itertools_count(start=0, step=self._PAGE_SIZE):
 451             info = self.query_api(
 452                 'dj/program/byradio?asc=false&limit=%d&radioId=%s&offset=%d'
 453                 % (self._PAGE_SIZE, dj_id, offset),
 454                 dj_id, 'Downloading dj programs - %d' % offset)
 455
 456             entries.extend([
 457                 self.url_result(
 458                     'http://music.163.com/#/program?id=%s' % program['id'],
 459                     'NetEaseMusicProgram', program['id'])
 460                 for program in info['programs']
 461             ])
 462
 463             if name is None:
 464                 radio = info['programs'][0]['radio']
 465                 name = radio['name']
 466                 desc = radio['desc']
 467
 468             if not info['more']:
 469                 break
 470
 471         return self.playlist_result(entries, dj_id, name, desc)