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