youtube_dl/extractor/qqmusic.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import random
   5 import re
   6 import time
   7
   8 from .common import InfoExtractor
   9 from ..utils import (
  10     clean_html,
  11     ExtractorError,
  12     strip_jsonp,
  13     unescapeHTML,
  14 )
  15
  16
  17 class QQMusicIE(InfoExtractor):
  18     IE_NAME = 'qqmusic'
  19     IE_DESC = 'QQ音乐'
  20     _VALID_URL = r'https?://y\.qq\.com/n/yqq/song/(?P<id>[0-9A-Za-z]+)\.html'
  21     _TESTS = [{
  22         'url': 'https://y.qq.com/n/yqq/song/004295Et37taLD.html',
  23         'md5': '5f1e6cea39e182857da7ffc5ef5e6bb8',
  24         'info_dict': {
  25             'id': '004295Et37taLD',
  26             'ext': 'mp3',
  27             'title': '可惜没如果',
  28             'release_date': '20141227',
  29             'creator': '林俊杰',
  30             'description': 'md5:d85afb3051952ecc50a1ee8a286d1eac',
  31             'thumbnail': r're:^https?://.*\.jpg$',
  32         }
  33     }, {
  34         'note': 'There is no mp3-320 version of this song.',
  35         'url': 'https://y.qq.com/n/yqq/song/004MsGEo3DdNxV.html',
  36         'md5': 'fa3926f0c585cda0af8fa4f796482e3e',
  37         'info_dict': {
  38             'id': '004MsGEo3DdNxV',
  39             'ext': 'mp3',
  40             'title': '如果',
  41             'release_date': '20050626',
  42             'creator': '李季美',
  43             'description': 'md5:46857d5ed62bc4ba84607a805dccf437',
  44             'thumbnail': r're:^https?://.*\.jpg$',
  45         }
  46     }, {
  47         'note': 'lyrics not in .lrc format',
  48         'url': 'https://y.qq.com/n/yqq/song/001JyApY11tIp6.html',
  49         'info_dict': {
  50             'id': '001JyApY11tIp6',
  51             'ext': 'mp3',
  52             'title': 'Shadows Over Transylvania',
  53             'release_date': '19970225',
  54             'creator': 'Dark Funeral',
  55             'description': 'md5:c9b20210587cbcd6836a1c597bab4525',
  56             'thumbnail': r're:^https?://.*\.jpg$',
  57         },
  58         'params': {
  59             'skip_download': True,
  60         },
  61     }]
  62
  63     _FORMATS = {
  64         'mp3-320': {'prefix': 'M800', 'ext': 'mp3', 'preference': 40, 'abr': 320},
  65         'mp3-128': {'prefix': 'M500', 'ext': 'mp3', 'preference': 30, 'abr': 128},
  66         'm4a': {'prefix': 'C200', 'ext': 'm4a', 'preference': 10}
  67     }
  68
  69     # Reference: m_r_GetRUin() in top_player.js
  70     # http://imgcache.gtimg.cn/music/portal_v3/y/top_player.js
  71     @staticmethod
  72     def m_r_get_ruin():
  73         curMs = int(time.time() * 1000) % 1000
  74         return int(round(random.random() * 2147483647) * curMs % 1E10)
  75
  76     def _real_extract(self, url):
  77         mid = self._match_id(url)
  78
  79         detail_info_page = self._download_webpage(
  80             'http://s.plcloud.music.qq.com/fcgi-bin/fcg_yqq_song_detail_info.fcg?songmid=%s&play=0' % mid,
  81             mid, note='Download song detail info',
  82             errnote='Unable to get song detail info', encoding='gbk')
  83
  84         song_name = self._html_search_regex(
  85             r"songname:\s*'([^']+)'", detail_info_page, 'song name')
  86
  87         publish_time = self._html_search_regex(
  88             r'发行时间：(\d{4}-\d{2}-\d{2})', detail_info_page,
  89             'publish time', default=None)
  90         if publish_time:
  91             publish_time = publish_time.replace('-', '')
  92
  93         singer = self._html_search_regex(
  94             r"singer:\s*'([^']+)", detail_info_page, 'singer', default=None)
  95
  96         lrc_content = self._html_search_regex(
  97             r'<div class="content" id="lrc_content"[^<>]*>([^<>]+)</div>',
  98             detail_info_page, 'LRC lyrics', default=None)
  99         if lrc_content:
 100             lrc_content = lrc_content.replace('\\n', '\n')
 101
 102         thumbnail_url = None
 103         albummid = self._search_regex(
 104             [r'albummid:\'([0-9a-zA-Z]+)\'', r'"albummid":"([0-9a-zA-Z]+)"'],
 105             detail_info_page, 'album mid', default=None)
 106         if albummid:
 107             thumbnail_url = 'http://i.gtimg.cn/music/photo/mid_album_500/%s/%s/%s.jpg' \
 108                             % (albummid[-2:-1], albummid[-1], albummid)
 109
 110         guid = self.m_r_get_ruin()
 111
 112         vkey = self._download_json(
 113             'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
 114             mid, note='Retrieve vkey', errnote='Unable to get vkey',
 115             transform_source=strip_jsonp)['key']
 116
 117         formats = []
 118         for format_id, details in self._FORMATS.items():
 119             formats.append({
 120                 'url': 'http://cc.stream.qqmusic.qq.com/%s%s.%s?vkey=%s&guid=%s&fromtag=0'
 121                        % (details['prefix'], mid, details['ext'], vkey, guid),
 122                 'format': format_id,
 123                 'format_id': format_id,
 124                 'preference': details['preference'],
 125                 'abr': details.get('abr'),
 126             })
 127         self._check_formats(formats, mid)
 128         self._sort_formats(formats)
 129
 130         actual_lrc_lyrics = ''.join(
 131             line + '\n' for line in re.findall(
 132                 r'(?m)^(\[[0-9]{2}:[0-9]{2}(?:\.[0-9]{2,})?\][^\n]*|\[[^\]]*\])', lrc_content))
 133
 134         info_dict = {
 135             'id': mid,
 136             'formats': formats,
 137             'title': song_name,
 138             'release_date': publish_time,
 139             'creator': singer,
 140             'description': lrc_content,
 141             'thumbnail': thumbnail_url
 142         }
 143         if actual_lrc_lyrics:
 144             info_dict['subtitles'] = {
 145                 'origin': [{
 146                     'ext': 'lrc',
 147                     'data': actual_lrc_lyrics,
 148                 }]
 149             }
 150         return info_dict
 151
 152
 153 class QQPlaylistBaseIE(InfoExtractor):
 154     @staticmethod
 155     def qq_static_url(category, mid):
 156         return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
 157
 158     def get_singer_all_songs(self, singmid, num):
 159         return self._download_webpage(
 160             r'https://c.y.qq.com/v8/fcg-bin/fcg_v8_singer_track_cp.fcg', singmid,
 161             query={
 162                 'format': 'json',
 163                 'inCharset': 'utf8',
 164                 'outCharset': 'utf-8',
 165                 'platform': 'yqq',
 166                 'needNewCode': 0,
 167                 'singermid': singmid,
 168                 'order': 'listen',
 169                 'begin': 0,
 170                 'num': num,
 171                 'songstatus': 1,
 172             })
 173
 174     def get_entries_from_page(self, singmid):
 175         entries = []
 176
 177         default_num = 1
 178         json_text = self.get_singer_all_songs(singmid, default_num)
 179         json_obj_all_songs = self._parse_json(json_text, singmid)
 180
 181         if json_obj_all_songs['code'] == 0:
 182             total = json_obj_all_songs['data']['total']
 183             json_text = self.get_singer_all_songs(singmid, total)
 184             json_obj_all_songs = self._parse_json(json_text, singmid)
 185
 186         for item in json_obj_all_songs['data']['list']:
 187             if item['musicData'].get('songmid') is not None:
 188                 songmid = item['musicData']['songmid']
 189                 entries.append(self.url_result(
 190                     r'https://y.qq.com/n/yqq/song/%s.html' % songmid, 'QQMusic', songmid))
 191
 192         return entries
 193
 194
 195 class QQMusicSingerIE(QQPlaylistBaseIE):
 196     IE_NAME = 'qqmusic:singer'
 197     IE_DESC = 'QQ音乐 - 歌手'
 198     _VALID_URL = r'https?://y\.qq\.com/n/yqq/singer/(?P<id>[0-9A-Za-z]+)\.html'
 199     _TEST = {
 200         'url': 'https://y.qq.com/n/yqq/singer/001BLpXF2DyJe2.html',
 201         'info_dict': {
 202             'id': '001BLpXF2DyJe2',
 203             'title': '林俊杰',
 204             'description': 'md5:870ec08f7d8547c29c93010899103751',
 205         },
 206         'playlist_mincount': 12,
 207     }
 208
 209     def _real_extract(self, url):
 210         mid = self._match_id(url)
 211
 212         entries = self.get_entries_from_page(mid)
 213         singer_page = self._download_webpage(url, mid, 'Download singer page')
 214         singer_name = self._html_search_regex(
 215             r"singername\s*:\s*'(.*?)'", singer_page, 'singer name', default=None)
 216         singer_desc = None
 217
 218         if mid:
 219             singer_desc_page = self._download_xml(
 220                 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg', mid,
 221                 'Donwload singer description XML',
 222                 query={'utf8': 1, 'outCharset': 'utf-8', 'format': 'xml', 'singermid': mid},
 223                 headers={'Referer': 'https://y.qq.com/n/yqq/singer/'})
 224
 225             singer_desc = singer_desc_page.find('./data/info/desc').text
 226
 227         return self.playlist_result(entries, mid, singer_name, singer_desc)
 228
 229
 230 class QQMusicAlbumIE(QQPlaylistBaseIE):
 231     IE_NAME = 'qqmusic:album'
 232     IE_DESC = 'QQ音乐 - 专辑'
 233     _VALID_URL = r'https?://y\.qq\.com/n/yqq/album/(?P<id>[0-9A-Za-z]+)\.html'
 234
 235     _TESTS = [{
 236         'url': 'https://y.qq.com/n/yqq/album/000gXCTb2AhRR1.html',
 237         'info_dict': {
 238             'id': '000gXCTb2AhRR1',
 239             'title': '我们都是这样长大的',
 240             'description': 'md5:179c5dce203a5931970d306aa9607ea6',
 241         },
 242         'playlist_count': 4,
 243     }, {
 244         'url': 'https://y.qq.com/n/yqq/album/002Y5a3b3AlCu3.html',
 245         'info_dict': {
 246             'id': '002Y5a3b3AlCu3',
 247             'title': '그리고...',
 248             'description': 'md5:a48823755615508a95080e81b51ba729',
 249         },
 250         'playlist_count': 8,
 251     }]
 252
 253     def _real_extract(self, url):
 254         mid = self._match_id(url)
 255
 256         album = self._download_json(
 257             'http://i.y.qq.com/v8/fcg-bin/fcg_v8_album_info_cp.fcg?albummid=%s&format=json' % mid,
 258             mid, 'Download album page')['data']
 259
 260         entries = [
 261             self.url_result(
 262                 'https://y.qq.com/n/yqq/song/' + song['songmid'] + '.html', 'QQMusic', song['songmid']
 263             ) for song in album['list']
 264         ]
 265         album_name = album.get('name')
 266         album_detail = album.get('desc')
 267         if album_detail is not None:
 268             album_detail = album_detail.strip()
 269
 270         return self.playlist_result(entries, mid, album_name, album_detail)
 271
 272
 273 class QQMusicToplistIE(QQPlaylistBaseIE):
 274     IE_NAME = 'qqmusic:toplist'
 275     IE_DESC = 'QQ音乐 - 排行榜'
 276     _VALID_URL = r'https?://y\.qq\.com/n/yqq/toplist/(?P<id>[0-9]+)\.html'
 277
 278     _TESTS = [{
 279         'url': 'https://y.qq.com/n/yqq/toplist/123.html',
 280         'info_dict': {
 281             'id': '123',
 282             'title': '美国iTunes榜',
 283             'description': 'md5:89db2335fdbb10678dee2d43fe9aba08',
 284         },
 285         'playlist_count': 100,
 286     }, {
 287         'url': 'https://y.qq.com/n/yqq/toplist/3.html',
 288         'info_dict': {
 289             'id': '3',
 290             'title': '巅峰榜·欧美',
 291             'description': 'md5:5a600d42c01696b26b71f8c4d43407da',
 292         },
 293         'playlist_count': 100,
 294     }, {
 295         'url': 'https://y.qq.com/n/yqq/toplist/106.html',
 296         'info_dict': {
 297             'id': '106',
 298             'title': '韩国Mnet榜',
 299             'description': 'md5:cb84b325215e1d21708c615cac82a6e7',
 300         },
 301         'playlist_count': 50,
 302     }]
 303
 304     def _real_extract(self, url):
 305         list_id = self._match_id(url)
 306
 307         toplist_json = self._download_json(
 308             'http://i.y.qq.com/v8/fcg-bin/fcg_v8_toplist_cp.fcg', list_id,
 309             note='Download toplist page',
 310             query={'type': 'toplist', 'topid': list_id, 'format': 'json'})
 311
 312         entries = [self.url_result(
 313             'https://y.qq.com/n/yqq/song/' + song['data']['songmid'] + '.html', 'QQMusic',
 314             song['data']['songmid'])
 315             for song in toplist_json['songlist']]
 316
 317         topinfo = toplist_json.get('topinfo', {})
 318         list_name = topinfo.get('ListName')
 319         list_description = topinfo.get('info')
 320         return self.playlist_result(entries, list_id, list_name, list_description)
 321
 322
 323 class QQMusicPlaylistIE(QQPlaylistBaseIE):
 324     IE_NAME = 'qqmusic:playlist'
 325     IE_DESC = 'QQ音乐 - 歌单'
 326     _VALID_URL = r'https?://y\.qq\.com/n/yqq/playlist/(?P<id>[0-9]+)\.html'
 327
 328     _TESTS = [{
 329         'url': 'http://y.qq.com/n/yqq/playlist/3462654915.html',
 330         'info_dict': {
 331             'id': '3462654915',
 332             'title': '韩国5月新歌精选下旬',
 333             'description': 'md5:d2c9d758a96b9888cf4fe82f603121d4',
 334         },
 335         'playlist_count': 40,
 336         'skip': 'playlist gone',
 337     }, {
 338         'url': 'https://y.qq.com/n/yqq/playlist/1374105607.html',
 339         'info_dict': {
 340             'id': '1374105607',
 341             'title': '易入人心的华语民谣',
 342             '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。',
 343         },
 344         'playlist_count': 20,
 345     }]
 346
 347     def _real_extract(self, url):
 348         list_id = self._match_id(url)
 349
 350         list_json = self._download_json(
 351             'http://i.y.qq.com/qzone-music/fcg-bin/fcg_ucc_getcdinfo_byids_cp.fcg',
 352             list_id, 'Download list page',
 353             query={'type': 1, 'json': 1, 'utf8': 1, 'onlysong': 0, 'disstid': list_id},
 354             transform_source=strip_jsonp)
 355         if not len(list_json.get('cdlist', [])):
 356             if list_json.get('code'):
 357                 raise ExtractorError(
 358                     'QQ Music said: error %d in fetching playlist info' % list_json['code'],
 359                     expected=True)
 360             raise ExtractorError('Unable to get playlist info')
 361
 362         cdlist = list_json['cdlist'][0]
 363         entries = [self.url_result(
 364             'https://y.qq.com/n/yqq/song/' + song['songmid'] + '.html', 'QQMusic', song['songmid'])
 365             for song in cdlist['songlist']]
 366
 367         list_name = cdlist.get('dissname')
 368         list_description = clean_html(unescapeHTML(cdlist.get('desc')))
 369         return self.playlist_result(entries, list_id, list_name, list_description)