yt_dlp/extractor/qqmusic.py

   1 import random
   2 import re
   3 import time
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     ExtractorError,
   8     clean_html,
   9     strip_jsonp,
  10     unescapeHTML,
  11 )
  12
  13
  14 class QQMusicIE(InfoExtractor):
  15     IE_NAME = 'qqmusic'
  16     IE_DESC = 'QQ音乐'
  17     _VALID_URL = r'https?://y\.qq\.com/n/yqq/song/(?P<id>[0-9A-Za-z]+)\.html'
  18     _TESTS = [{
  19         'url': 'https://y.qq.com/n/yqq/song/004295Et37taLD.html',
  20         'md5': '5f1e6cea39e182857da7ffc5ef5e6bb8',
  21         'info_dict': {
  22             'id': '004295Et37taLD',
  23             'ext': 'mp3',
  24             'title': '可惜没如果',
  25             'release_date': '20141227',
  26             'creator': '林俊杰',
  27             'description': 'md5:d85afb3051952ecc50a1ee8a286d1eac',
  28             'thumbnail': r're:^https?://.*\.jpg$',
  29         }
  30     }, {
  31         'note': 'There is no mp3-320 version of this song.',
  32         'url': 'https://y.qq.com/n/yqq/song/004MsGEo3DdNxV.html',
  33         'md5': 'fa3926f0c585cda0af8fa4f796482e3e',
  34         'info_dict': {
  35             'id': '004MsGEo3DdNxV',
  36             'ext': 'mp3',
  37             'title': '如果',
  38             'release_date': '20050626',
  39             'creator': '李季美',
  40             'description': 'md5:46857d5ed62bc4ba84607a805dccf437',
  41             'thumbnail': r're:^https?://.*\.jpg$',
  42         }
  43     }, {
  44         'note': 'lyrics not in .lrc format',
  45         'url': 'https://y.qq.com/n/yqq/song/001JyApY11tIp6.html',
  46         'info_dict': {
  47             'id': '001JyApY11tIp6',
  48             'ext': 'mp3',
  49             'title': 'Shadows Over Transylvania',
  50             'release_date': '19970225',
  51             'creator': 'Dark Funeral',
  52             'description': 'md5:c9b20210587cbcd6836a1c597bab4525',
  53             'thumbnail': r're:^https?://.*\.jpg$',
  54         },
  55         'params': {
  56             'skip_download': True,
  57         },
  58     }]
  59
  60     _FORMATS = {
  61         'mp3-320': {'prefix': 'M800', 'ext': 'mp3', 'preference': 40, 'abr': 320},
  62         'mp3-128': {'prefix': 'M500', 'ext': 'mp3', 'preference': 30, 'abr': 128},
  63         'm4a': {'prefix': 'C200', 'ext': 'm4a', 'preference': 10}
  64     }
  65
  66     # Reference: m_r_GetRUin() in top_player.js
  67     # http://imgcache.gtimg.cn/music/portal_v3/y/top_player.js
  68     @staticmethod
  69     def m_r_get_ruin():
  70         curMs = int(time.time() * 1000) % 1000
  71         return int(round(random.random() * 2147483647) * curMs % 1E10)
  72
  73     def _real_extract(self, url):
  74         mid = self._match_id(url)
  75
  76         detail_info_page = self._download_webpage(
  77             'http://s.plcloud.music.qq.com/fcgi-bin/fcg_yqq_song_detail_info.fcg?songmid=%s&play=0' % mid,
  78             mid, note='Download song detail info',
  79             errnote='Unable to get song detail info', encoding='gbk')
  80
  81         song_name = self._html_search_regex(
  82             r"songname:\s*'([^']+)'", detail_info_page, 'song name')
  83
  84         publish_time = self._html_search_regex(
  85             r'发行时间：(\d{4}-\d{2}-\d{2})', detail_info_page,
  86             'publish time', default=None)
  87         if publish_time:
  88             publish_time = publish_time.replace('-', '')
  89
  90         singer = self._html_search_regex(
  91             r"singer:\s*'([^']+)", detail_info_page, 'singer', default=None)
  92
  93         lrc_content = self._html_search_regex(
  94             r'<div class="content" id="lrc_content"[^<>]*>([^<>]+)</div>',
  95             detail_info_page, 'LRC lyrics', default=None)
  96         if lrc_content:
  97             lrc_content = lrc_content.replace('\\n', '\n')
  98
  99         thumbnail_url = None
 100         albummid = self._search_regex(
 101             [r'albummid:\'([0-9a-zA-Z]+)\'', r'"albummid":"([0-9a-zA-Z]+)"'],
 102             detail_info_page, 'album mid', default=None)
 103         if albummid:
 104             thumbnail_url = 'http://i.gtimg.cn/music/photo/mid_album_500/%s/%s/%s.jpg' \
 105                             % (albummid[-2:-1], albummid[-1], albummid)
 106
 107         guid = self.m_r_get_ruin()
 108
 109         vkey = self._download_json(
 110             'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
 111             mid, note='Retrieve vkey', errnote='Unable to get vkey',
 112             transform_source=strip_jsonp)['key']
 113
 114         formats = []
 115         for format_id, details in self._FORMATS.items():
 116             formats.append({
 117                 'url': 'http://cc.stream.qqmusic.qq.com/%s%s.%s?vkey=%s&guid=%s&fromtag=0'
 118                        % (details['prefix'], mid, details['ext'], vkey, guid),
 119                 'format': format_id,
 120                 'format_id': format_id,
 121                 'quality': details['preference'],
 122                 'abr': details.get('abr'),
 123             })
 124         self._check_formats(formats, mid)
 125
 126         actual_lrc_lyrics = ''.join(
 127             line + '\n' for line in re.findall(
 128                 r'(?m)^(\[[0-9]{2}:[0-9]{2}(?:\.[0-9]{2,})?\][^\n]*|\[[^\]]*\])', lrc_content))
 129
 130         info_dict = {
 131             'id': mid,
 132             'formats': formats,
 133             'title': song_name,
 134             'release_date': publish_time,
 135             'creator': singer,
 136             'description': lrc_content,
 137             'thumbnail': thumbnail_url
 138         }
 139         if actual_lrc_lyrics:
 140             info_dict['subtitles'] = {
 141                 'origin': [{
 142                     'ext': 'lrc',
 143                     'data': actual_lrc_lyrics,
 144                 }]
 145             }
 146         return info_dict
 147
 148
 149 class QQPlaylistBaseIE(InfoExtractor):
 150     @staticmethod
 151     def qq_static_url(category, mid):
 152         return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
 153
 154     def get_singer_all_songs(self, singmid, num):
 155         return self._download_webpage(
 156             r'https://c.y.qq.com/v8/fcg-bin/fcg_v8_singer_track_cp.fcg', singmid,
 157             query={
 158                 'format': 'json',
 159                 'inCharset': 'utf8',
 160                 'outCharset': 'utf-8',
 161                 'platform': 'yqq',
 162                 'needNewCode': 0,
 163                 'singermid': singmid,
 164                 'order': 'listen',
 165                 'begin': 0,
 166                 'num': num,
 167                 'songstatus': 1,
 168             })
 169
 170     def get_entries_from_page(self, singmid):
 171         entries = []
 172
 173         default_num = 1
 174         json_text = self.get_singer_all_songs(singmid, default_num)
 175         json_obj_all_songs = self._parse_json(json_text, singmid)
 176
 177         if json_obj_all_songs['code'] == 0:
 178             total = json_obj_all_songs['data']['total']
 179             json_text = self.get_singer_all_songs(singmid, total)
 180             json_obj_all_songs = self._parse_json(json_text, singmid)
 181
 182         for item in json_obj_all_songs['data']['list']:
 183             if item['musicData'].get('songmid') is not None:
 184                 songmid = item['musicData']['songmid']
 185                 entries.append(self.url_result(
 186                     r'https://y.qq.com/n/yqq/song/%s.html' % songmid, 'QQMusic', songmid))
 187
 188         return entries
 189
 190
 191 class QQMusicSingerIE(QQPlaylistBaseIE):
 192     IE_NAME = 'qqmusic:singer'
 193     IE_DESC = 'QQ音乐 - 歌手'
 194     _VALID_URL = r'https?://y\.qq\.com/n/yqq/singer/(?P<id>[0-9A-Za-z]+)\.html'
 195     _TEST = {
 196         'url': 'https://y.qq.com/n/yqq/singer/001BLpXF2DyJe2.html',
 197         'info_dict': {
 198             'id': '001BLpXF2DyJe2',
 199             'title': '林俊杰',
 200             'description': 'md5:870ec08f7d8547c29c93010899103751',
 201         },
 202         'playlist_mincount': 12,
 203     }
 204
 205     def _real_extract(self, url):
 206         mid = self._match_id(url)
 207
 208         entries = self.get_entries_from_page(mid)
 209         singer_page = self._download_webpage(url, mid, 'Download singer page')
 210         singer_name = self._html_search_regex(
 211             r"singername\s*:\s*'(.*?)'", singer_page, 'singer name', default=None)
 212         singer_desc = None
 213
 214         if mid:
 215             singer_desc_page = self._download_xml(
 216                 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg', mid,
 217                 'Donwload singer description XML',
 218                 query={'utf8': 1, 'outCharset': 'utf-8', 'format': 'xml', 'singermid': mid},
 219                 headers={'Referer': 'https://y.qq.com/n/yqq/singer/'})
 220
 221             singer_desc = singer_desc_page.find('./data/info/desc').text
 222
 223         return self.playlist_result(entries, mid, singer_name, singer_desc)
 224
 225
 226 class QQMusicAlbumIE(QQPlaylistBaseIE):
 227     IE_NAME = 'qqmusic:album'
 228     IE_DESC = 'QQ音乐 - 专辑'
 229     _VALID_URL = r'https?://y\.qq\.com/n/yqq/album/(?P<id>[0-9A-Za-z]+)\.html'
 230
 231     _TESTS = [{
 232         'url': 'https://y.qq.com/n/yqq/album/000gXCTb2AhRR1.html',
 233         'info_dict': {
 234             'id': '000gXCTb2AhRR1',
 235             'title': '我们都是这样长大的',
 236             'description': 'md5:179c5dce203a5931970d306aa9607ea6',
 237         },
 238         'playlist_count': 4,
 239     }, {
 240         'url': 'https://y.qq.com/n/yqq/album/002Y5a3b3AlCu3.html',
 241         'info_dict': {
 242             'id': '002Y5a3b3AlCu3',
 243             'title': '그리고...',
 244             'description': 'md5:a48823755615508a95080e81b51ba729',
 245         },
 246         'playlist_count': 8,
 247     }]
 248
 249     def _real_extract(self, url):
 250         mid = self._match_id(url)
 251
 252         album = self._download_json(
 253             'http://i.y.qq.com/v8/fcg-bin/fcg_v8_album_info_cp.fcg?albummid=%s&format=json' % mid,
 254             mid, 'Download album page')['data']
 255
 256         entries = [
 257             self.url_result(
 258                 'https://y.qq.com/n/yqq/song/' + song['songmid'] + '.html', 'QQMusic', song['songmid']
 259             ) for song in album['list']
 260         ]
 261         album_name = album.get('name')
 262         album_detail = album.get('desc')
 263         if album_detail is not None:
 264             album_detail = album_detail.strip()
 265
 266         return self.playlist_result(entries, mid, album_name, album_detail)
 267
 268
 269 class QQMusicToplistIE(QQPlaylistBaseIE):
 270     IE_NAME = 'qqmusic:toplist'
 271     IE_DESC = 'QQ音乐 - 排行榜'
 272     _VALID_URL = r'https?://y\.qq\.com/n/yqq/toplist/(?P<id>[0-9]+)\.html'
 273
 274     _TESTS = [{
 275         'url': 'https://y.qq.com/n/yqq/toplist/123.html',
 276         'info_dict': {
 277             'id': '123',
 278             'title': '美国iTunes榜',
 279             'description': 'md5:89db2335fdbb10678dee2d43fe9aba08',
 280         },
 281         'playlist_count': 100,
 282     }, {
 283         'url': 'https://y.qq.com/n/yqq/toplist/3.html',
 284         'info_dict': {
 285             'id': '3',
 286             'title': '巅峰榜·欧美',
 287             'description': 'md5:5a600d42c01696b26b71f8c4d43407da',
 288         },
 289         'playlist_count': 100,
 290     }, {
 291         'url': 'https://y.qq.com/n/yqq/toplist/106.html',
 292         'info_dict': {
 293             'id': '106',
 294             'title': '韩国Mnet榜',
 295             'description': 'md5:cb84b325215e1d21708c615cac82a6e7',
 296         },
 297         'playlist_count': 50,
 298     }]
 299
 300     def _real_extract(self, url):
 301         list_id = self._match_id(url)
 302
 303         toplist_json = self._download_json(
 304             'http://i.y.qq.com/v8/fcg-bin/fcg_v8_toplist_cp.fcg', list_id,
 305             note='Download toplist page',
 306             query={'type': 'toplist', 'topid': list_id, 'format': 'json'})
 307
 308         entries = [self.url_result(
 309             'https://y.qq.com/n/yqq/song/' + song['data']['songmid'] + '.html', 'QQMusic',
 310             song['data']['songmid'])
 311             for song in toplist_json['songlist']]
 312
 313         topinfo = toplist_json.get('topinfo', {})
 314         list_name = topinfo.get('ListName')
 315         list_description = topinfo.get('info')
 316         return self.playlist_result(entries, list_id, list_name, list_description)
 317
 318
 319 class QQMusicPlaylistIE(QQPlaylistBaseIE):
 320     IE_NAME = 'qqmusic:playlist'
 321     IE_DESC = 'QQ音乐 - 歌单'
 322     _VALID_URL = r'https?://y\.qq\.com/n/yqq/playlist/(?P<id>[0-9]+)\.html'
 323
 324     _TESTS = [{
 325         'url': 'http://y.qq.com/n/yqq/playlist/3462654915.html',
 326         'info_dict': {
 327             'id': '3462654915',
 328             'title': '韩国5月新歌精选下旬',
 329             'description': 'md5:d2c9d758a96b9888cf4fe82f603121d4',
 330         },
 331         'playlist_count': 40,
 332         'skip': 'playlist gone',
 333     }, {
 334         'url': 'https://y.qq.com/n/yqq/playlist/1374105607.html',
 335         'info_dict': {
 336             'id': '1374105607',
 337             'title': '易入人心的华语民谣',
 338             'description': '民谣的歌曲易于传唱、、歌词朗朗伤口、旋律简单温馨。属于那种才入耳孔。却上心头的感觉。没有太多的复杂情绪。简单而直接地表达乐者的情绪，就是这样的简单才易入人心。',
 339         },
 340         'playlist_count': 20,
 341     }]
 342
 343     def _real_extract(self, url):
 344         list_id = self._match_id(url)
 345
 346         list_json = self._download_json(
 347             'http://i.y.qq.com/qzone-music/fcg-bin/fcg_ucc_getcdinfo_byids_cp.fcg',
 348             list_id, 'Download list page',
 349             query={'type': 1, 'json': 1, 'utf8': 1, 'onlysong': 0, 'disstid': list_id},
 350             transform_source=strip_jsonp)
 351         if not len(list_json.get('cdlist', [])):
 352             if list_json.get('code'):
 353                 raise ExtractorError(
 354                     'QQ Music said: error %d in fetching playlist info' % list_json['code'],
 355                     expected=True)
 356             raise ExtractorError('Unable to get playlist info')
 357
 358         cdlist = list_json['cdlist'][0]
 359         entries = [self.url_result(
 360             'https://y.qq.com/n/yqq/song/' + song['songmid'] + '.html', 'QQMusic', song['songmid'])
 361             for song in cdlist['songlist']]
 362
 363         list_name = cdlist.get('dissname')
 364         list_description = clean_html(unescapeHTML(cdlist.get('desc')))
 365         return self.playlist_result(entries, list_id, list_name, list_description)