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_urlencode,
  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         'skip': 'Blocked outside Mainland China',
  93     }, {
  94         'note': 'No lyrics translation.',
  95         'url': 'http://music.163.com/#/song?id=29822014',
  96         'info_dict': {
  97             'id': '29822014',
  98             'ext': 'mp3',
  99             'title': '听见下雨的声音',
 100             'creator': '周杰伦',
 101             'upload_date': '20141225',
 102             'timestamp': 1419523200,
 103             'description': 'md5:a4d8d89f44656af206b7b2555c0bce6c',
 104         },
 105         'skip': 'Blocked outside Mainland China',
 106     }, {
 107         'note': 'No lyrics.',
 108         'url': 'http://music.163.com/song?id=17241424',
 109         'info_dict': {
 110             'id': '17241424',
 111             'ext': 'mp3',
 112             'title': 'Opus 28',
 113             'creator': 'Dustin O\'Halloran',
 114             'upload_date': '20080211',
 115             'timestamp': 1202745600,
 116         },
 117         'skip': 'Blocked outside Mainland China',
 118     }, {
 119         'note': 'Has translated name.',
 120         'url': 'http://music.163.com/#/song?id=22735043',
 121         'info_dict': {
 122             'id': '22735043',
 123             'ext': 'mp3',
 124             'title': '소원을 말해봐 (Genie)',
 125             'creator': '少女时代',
 126             'description': 'md5:79d99cc560e4ca97e0c4d86800ee4184',
 127             'upload_date': '20100127',
 128             'timestamp': 1264608000,
 129             'alt_title': '说出愿望吧(Genie)',
 130         },
 131         'skip': 'Blocked outside Mainland China',
 132     }]
 133
 134     def _process_lyrics(self, lyrics_info):
 135         original = lyrics_info.get('lrc', {}).get('lyric')
 136         translated = lyrics_info.get('tlyric', {}).get('lyric')
 137
 138         if not translated:
 139             return original
 140
 141         lyrics_expr = r'(\[[0-9]{2}:[0-9]{2}\.[0-9]{2,}\])([^\n]+)'
 142         original_ts_texts = re.findall(lyrics_expr, original)
 143         translation_ts_dict = dict(
 144             (time_stamp, text) for time_stamp, text in re.findall(lyrics_expr, translated)
 145         )
 146         lyrics = '\n'.join([
 147             '%s%s / %s' % (time_stamp, text, translation_ts_dict.get(time_stamp, ''))
 148             for time_stamp, text in original_ts_texts
 149         ])
 150         return lyrics
 151
 152     def _real_extract(self, url):
 153         song_id = self._match_id(url)
 154
 155         params = {
 156             'id': song_id,
 157             'ids': '[%s]' % song_id
 158         }
 159         info = self.query_api(
 160             'song/detail?' + compat_urllib_parse_urlencode(params),
 161             song_id, 'Downloading song info')['songs'][0]
 162
 163         formats = self.extract_formats(info)
 164         self._sort_formats(formats)
 165
 166         lyrics_info = self.query_api(
 167             'song/lyric?id=%s&lv=-1&tv=-1' % song_id,
 168             song_id, 'Downloading lyrics data')
 169         lyrics = self._process_lyrics(lyrics_info)
 170
 171         alt_title = None
 172         if info.get('transNames'):
 173             alt_title = '/'.join(info.get('transNames'))
 174
 175         return {
 176             'id': song_id,
 177             'title': info['name'],
 178             'alt_title': alt_title,
 179             'creator': ' / '.join([artist['name'] for artist in info.get('artists', [])]),
 180             'timestamp': self.convert_milliseconds(info.get('album', {}).get('publishTime')),
 181             'thumbnail': info.get('album', {}).get('picUrl'),
 182             'duration': self.convert_milliseconds(info.get('duration', 0)),
 183             'description': lyrics,
 184             'formats': formats,
 185         }
 186
 187
 188 class NetEaseMusicAlbumIE(NetEaseMusicBaseIE):
 189     IE_NAME = 'netease:album'
 190     IE_DESC = '网易云音乐 - 专辑'
 191     _VALID_URL = r'https?://music\.163\.com/(#/)?album\?id=(?P<id>[0-9]+)'
 192     _TEST = {
 193         'url': 'http://music.163.com/#/album?id=220780',
 194         'info_dict': {
 195             'id': '220780',
 196             'title': 'B\'day',
 197         },
 198         'playlist_count': 23,
 199         'skip': 'Blocked outside Mainland China',
 200     }
 201
 202     def _real_extract(self, url):
 203         album_id = self._match_id(url)
 204
 205         info = self.query_api(
 206             'album/%s?id=%s' % (album_id, album_id),
 207             album_id, 'Downloading album data')['album']
 208
 209         name = info['name']
 210         desc = info.get('description')
 211         entries = [
 212             self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
 213                             'NetEaseMusic', song['id'])
 214             for song in info['songs']
 215         ]
 216         return self.playlist_result(entries, album_id, name, desc)
 217
 218
 219 class NetEaseMusicSingerIE(NetEaseMusicBaseIE):
 220     IE_NAME = 'netease:singer'
 221     IE_DESC = '网易云音乐 - 歌手'
 222     _VALID_URL = r'https?://music\.163\.com/(#/)?artist\?id=(?P<id>[0-9]+)'
 223     _TESTS = [{
 224         'note': 'Singer has aliases.',
 225         'url': 'http://music.163.com/#/artist?id=10559',
 226         'info_dict': {
 227             'id': '10559',
 228             'title': '张惠妹 - aMEI;阿密特',
 229         },
 230         'playlist_count': 50,
 231         'skip': 'Blocked outside Mainland China',
 232     }, {
 233         'note': 'Singer has translated name.',
 234         'url': 'http://music.163.com/#/artist?id=124098',
 235         'info_dict': {
 236             'id': '124098',
 237             'title': '李昇基 - 이승기',
 238         },
 239         'playlist_count': 50,
 240         'skip': 'Blocked outside Mainland China',
 241     }]
 242
 243     def _real_extract(self, url):
 244         singer_id = self._match_id(url)
 245
 246         info = self.query_api(
 247             'artist/%s?id=%s' % (singer_id, singer_id),
 248             singer_id, 'Downloading singer data')
 249
 250         name = info['artist']['name']
 251         if info['artist']['trans']:
 252             name = '%s - %s' % (name, info['artist']['trans'])
 253         if info['artist']['alias']:
 254             name = '%s - %s' % (name, ';'.join(info['artist']['alias']))
 255
 256         entries = [
 257             self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
 258                             'NetEaseMusic', song['id'])
 259             for song in info['hotSongs']
 260         ]
 261         return self.playlist_result(entries, singer_id, name)
 262
 263
 264 class NetEaseMusicListIE(NetEaseMusicBaseIE):
 265     IE_NAME = 'netease:playlist'
 266     IE_DESC = '网易云音乐 - 歌单'
 267     _VALID_URL = r'https?://music\.163\.com/(#/)?(playlist|discover/toplist)\?id=(?P<id>[0-9]+)'
 268     _TESTS = [{
 269         'url': 'http://music.163.com/#/playlist?id=79177352',
 270         'info_dict': {
 271             'id': '79177352',
 272             'title': 'Billboard 2007 Top 100',
 273             'description': 'md5:12fd0819cab2965b9583ace0f8b7b022'
 274         },
 275         'playlist_count': 99,
 276         'skip': 'Blocked outside Mainland China',
 277     }, {
 278         'note': 'Toplist/Charts sample',
 279         'url': 'http://music.163.com/#/discover/toplist?id=3733003',
 280         'info_dict': {
 281             'id': '3733003',
 282             'title': 're:韩国Melon排行榜周榜 [0-9]{4}-[0-9]{2}-[0-9]{2}',
 283             'description': 'md5:73ec782a612711cadc7872d9c1e134fc',
 284         },
 285         'playlist_count': 50,
 286         'skip': 'Blocked outside Mainland China',
 287     }]
 288
 289     def _real_extract(self, url):
 290         list_id = self._match_id(url)
 291
 292         info = self.query_api(
 293             'playlist/detail?id=%s&lv=-1&tv=-1' % list_id,
 294             list_id, 'Downloading playlist data')['result']
 295
 296         name = info['name']
 297         desc = info.get('description')
 298
 299         if info.get('specialType') == 10:  # is a chart/toplist
 300             datestamp = datetime.fromtimestamp(
 301                 self.convert_milliseconds(info['updateTime'])).strftime('%Y-%m-%d')
 302             name = '%s %s' % (name, datestamp)
 303
 304         entries = [
 305             self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
 306                             'NetEaseMusic', song['id'])
 307             for song in info['tracks']
 308         ]
 309         return self.playlist_result(entries, list_id, name, desc)
 310
 311
 312 class NetEaseMusicMvIE(NetEaseMusicBaseIE):
 313     IE_NAME = 'netease:mv'
 314     IE_DESC = '网易云音乐 - MV'
 315     _VALID_URL = r'https?://music\.163\.com/(#/)?mv\?id=(?P<id>[0-9]+)'
 316     _TEST = {
 317         'url': 'http://music.163.com/#/mv?id=415350',
 318         'info_dict': {
 319             'id': '415350',
 320             'ext': 'mp4',
 321             'title': '이럴거면 그러지말지',
 322             'description': '白雅言自作曲唱甜蜜爱情',
 323             'creator': '白雅言',
 324             'upload_date': '20150520',
 325         },
 326         'skip': 'Blocked outside Mainland China',
 327     }
 328
 329     def _real_extract(self, url):
 330         mv_id = self._match_id(url)
 331
 332         info = self.query_api(
 333             'mv/detail?id=%s&type=mp4' % mv_id,
 334             mv_id, 'Downloading mv info')['data']
 335
 336         formats = [
 337             {'url': mv_url, 'ext': 'mp4', 'format_id': '%sp' % brs, 'height': int(brs)}
 338             for brs, mv_url in info['brs'].items()
 339         ]
 340         self._sort_formats(formats)
 341
 342         return {
 343             'id': mv_id,
 344             'title': info['name'],
 345             'description': info.get('desc') or info.get('briefDesc'),
 346             'creator': info['artistName'],
 347             'upload_date': info['publishTime'].replace('-', ''),
 348             'formats': formats,
 349             'thumbnail': info.get('cover'),
 350             'duration': self.convert_milliseconds(info.get('duration', 0)),
 351         }
 352
 353
 354 class NetEaseMusicProgramIE(NetEaseMusicBaseIE):
 355     IE_NAME = 'netease:program'
 356     IE_DESC = '网易云音乐 - 电台节目'
 357     _VALID_URL = r'https?://music\.163\.com/(#/?)program\?id=(?P<id>[0-9]+)'
 358     _TESTS = [{
 359         'url': 'http://music.163.com/#/program?id=10109055',
 360         'info_dict': {
 361             'id': '10109055',
 362             'ext': 'mp3',
 363             'title': '不丹足球背后的故事',
 364             'description': '喜马拉雅人的足球梦 ...',
 365             'creator': '大话西藏',
 366             'timestamp': 1434179342,
 367             'upload_date': '20150613',
 368             'duration': 900,
 369         },
 370         'skip': 'Blocked outside Mainland China',
 371     }, {
 372         'note': 'This program has accompanying songs.',
 373         'url': 'http://music.163.com/#/program?id=10141022',
 374         'info_dict': {
 375             'id': '10141022',
 376             'title': '25岁，你是自在如风的少年<27°C>',
 377             'description': 'md5:8d594db46cc3e6509107ede70a4aaa3b',
 378         },
 379         'playlist_count': 4,
 380         'skip': 'Blocked outside Mainland China',
 381     }, {
 382         'note': 'This program has accompanying songs.',
 383         'url': 'http://music.163.com/#/program?id=10141022',
 384         'info_dict': {
 385             'id': '10141022',
 386             'ext': 'mp3',
 387             'title': '25岁，你是自在如风的少年<27°C>',
 388             'description': 'md5:8d594db46cc3e6509107ede70a4aaa3b',
 389             'timestamp': 1434450841,
 390             'upload_date': '20150616',
 391         },
 392         'params': {
 393             'noplaylist': True
 394         },
 395         'skip': 'Blocked outside Mainland China',
 396     }]
 397
 398     def _real_extract(self, url):
 399         program_id = self._match_id(url)
 400
 401         info = self.query_api(
 402             'dj/program/detail?id=%s' % program_id,
 403             program_id, 'Downloading program info')['program']
 404
 405         name = info['name']
 406         description = info['description']
 407
 408         if not info['songs'] or self._downloader.params.get('noplaylist'):
 409             if info['songs']:
 410                 self.to_screen(
 411                     'Downloading just the main audio %s because of --no-playlist'
 412                     % info['mainSong']['id'])
 413
 414             formats = self.extract_formats(info['mainSong'])
 415             self._sort_formats(formats)
 416
 417             return {
 418                 'id': program_id,
 419                 'title': name,
 420                 'description': description,
 421                 'creator': info['dj']['brand'],
 422                 'timestamp': self.convert_milliseconds(info['createTime']),
 423                 'thumbnail': info['coverUrl'],
 424                 'duration': self.convert_milliseconds(info.get('duration', 0)),
 425                 'formats': formats,
 426             }
 427
 428         self.to_screen(
 429             'Downloading playlist %s - add --no-playlist to just download the main audio %s'
 430             % (program_id, info['mainSong']['id']))
 431
 432         song_ids = [info['mainSong']['id']]
 433         song_ids.extend([song['id'] for song in info['songs']])
 434         entries = [
 435             self.url_result('http://music.163.com/#/song?id=%s' % song_id,
 436                             'NetEaseMusic', song_id)
 437             for song_id in song_ids
 438         ]
 439         return self.playlist_result(entries, program_id, name, description)
 440
 441
 442 class NetEaseMusicDjRadioIE(NetEaseMusicBaseIE):
 443     IE_NAME = 'netease:djradio'
 444     IE_DESC = '网易云音乐 - 电台'
 445     _VALID_URL = r'https?://music\.163\.com/(#/)?djradio\?id=(?P<id>[0-9]+)'
 446     _TEST = {
 447         'url': 'http://music.163.com/#/djradio?id=42',
 448         'info_dict': {
 449             'id': '42',
 450             'title': '声音蔓延',
 451             'description': 'md5:766220985cbd16fdd552f64c578a6b15'
 452         },
 453         'playlist_mincount': 40,
 454         'skip': 'Blocked outside Mainland China',
 455     }
 456     _PAGE_SIZE = 1000
 457
 458     def _real_extract(self, url):
 459         dj_id = self._match_id(url)
 460
 461         name = None
 462         desc = None
 463         entries = []
 464         for offset in compat_itertools_count(start=0, step=self._PAGE_SIZE):
 465             info = self.query_api(
 466                 'dj/program/byradio?asc=false&limit=%d&radioId=%s&offset=%d'
 467                 % (self._PAGE_SIZE, dj_id, offset),
 468                 dj_id, 'Downloading dj programs - %d' % offset)
 469
 470             entries.extend([
 471                 self.url_result(
 472                     'http://music.163.com/#/program?id=%s' % program['id'],
 473                     'NetEaseMusicProgram', program['id'])
 474                 for program in info['programs']
 475             ])
 476
 477             if name is None:
 478                 radio = info['programs'][0]['radio']
 479                 name = radio['name']
 480                 desc = radio['desc']
 481
 482             if not info['more']:
 483                 break
 484
 485         return self.playlist_result(entries, dj_id, name, desc)