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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import random
   5 import time
   6 import re
   7
   8 from .common import InfoExtractor
   9 from ..utils import (
  10     sanitized_Request,
  11     strip_jsonp,
  12     unescapeHTML,
  13     clean_html,
  14     ExtractorError,
  15 )
  16
  17
  18 class QQMusicIE(InfoExtractor):
  19     IE_NAME = 'qqmusic'
  20     IE_DESC = 'QQ音乐'
  21     _VALID_URL = r'https?://y\.qq\.com/#type=song&mid=(?P<id>[0-9A-Za-z]+)'
  22     _TESTS = [{
  23         'url': 'http://y.qq.com/#type=song&mid=004295Et37taLD',
  24         'md5': '9ce1c1c8445f561506d2e3cfb0255705',
  25         'info_dict': {
  26             'id': '004295Et37taLD',
  27             'ext': 'mp3',
  28             'title': '可惜没如果',
  29             'release_date': '20141227',
  30             'creator': '林俊杰',
  31             'description': 'md5:d327722d0361576fde558f1ac68a7065',
  32             'thumbnail': 're:^https?://.*\.jpg$',
  33         }
  34     }, {
  35         'note': 'There is no mp3-320 version of this song.',
  36         'url': 'http://y.qq.com/#type=song&mid=004MsGEo3DdNxV',
  37         'md5': 'fa3926f0c585cda0af8fa4f796482e3e',
  38         'info_dict': {
  39             'id': '004MsGEo3DdNxV',
  40             'ext': 'mp3',
  41             'title': '如果',
  42             'release_date': '20050626',
  43             'creator': '李季美',
  44             'description': 'md5:46857d5ed62bc4ba84607a805dccf437',
  45             'thumbnail': 're:^https?://.*\.jpg$',
  46         }
  47     }, {
  48         'note': 'lyrics not in .lrc format',
  49         'url': 'http://y.qq.com/#type=song&mid=001JyApY11tIp6',
  50         'info_dict': {
  51             'id': '001JyApY11tIp6',
  52             'ext': 'mp3',
  53             'title': 'Shadows Over Transylvania',
  54             'release_date': '19970225',
  55             'creator': 'Dark Funeral',
  56             'description': 'md5:ed14d5bd7ecec19609108052c25b2c11',
  57             'thumbnail': 're:^https?://.*\.jpg$',
  58         },
  59         'params': {
  60             'skip_download': True,
  61         },
  62     }]
  63
  64     _FORMATS = {
  65         'mp3-320': {'prefix': 'M800', 'ext': 'mp3', 'preference': 40, 'abr': 320},
  66         'mp3-128': {'prefix': 'M500', 'ext': 'mp3', 'preference': 30, 'abr': 128},
  67         'm4a': {'prefix': 'C200', 'ext': 'm4a', 'preference': 10}
  68     }
  69
  70     # Reference: m_r_GetRUin() in top_player.js
  71     # http://imgcache.gtimg.cn/music/portal_v3/y/top_player.js
  72     @staticmethod
  73     def m_r_get_ruin():
  74         curMs = int(time.time() * 1000) % 1000
  75         return int(round(random.random() * 2147483647) * curMs % 1E10)
  76
  77     def _real_extract(self, url):
  78         mid = self._match_id(url)
  79
  80         detail_info_page = self._download_webpage(
  81             'http://s.plcloud.music.qq.com/fcgi-bin/fcg_yqq_song_detail_info.fcg?songmid=%s&play=0' % mid,
  82             mid, note='Download song detail info',
  83             errnote='Unable to get song detail info', encoding='gbk')
  84
  85         song_name = self._html_search_regex(
  86             r"songname:\s*'([^']+)'", detail_info_page, 'song name')
  87
  88         publish_time = self._html_search_regex(
  89             r'发行时间：(\d{4}-\d{2}-\d{2})', detail_info_page,
  90             'publish time', default=None)
  91         if publish_time:
  92             publish_time = publish_time.replace('-', '')
  93
  94         singer = self._html_search_regex(
  95             r"singer:\s*'([^']+)", detail_info_page, 'singer', default=None)
  96
  97         lrc_content = self._html_search_regex(
  98             r'<div class="content" id="lrc_content"[^<>]*>([^<>]+)</div>',
  99             detail_info_page, 'LRC lyrics', default=None)
 100         if lrc_content:
 101             lrc_content = lrc_content.replace('\\n', '\n')
 102
 103         thumbnail_url = None
 104         albummid = self._search_regex(
 105             [r'albummid:\'([0-9a-zA-Z]+)\'', r'"albummid":"([0-9a-zA-Z]+)"'],
 106             detail_info_page, 'album mid', default=None)
 107         if albummid:
 108             thumbnail_url = "http://i.gtimg.cn/music/photo/mid_album_500/%s/%s/%s.jpg" \
 109                             % (albummid[-2:-1], albummid[-1], albummid)
 110
 111         guid = self.m_r_get_ruin()
 112
 113         vkey = self._download_json(
 114             'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
 115             mid, note='Retrieve vkey', errnote='Unable to get vkey',
 116             transform_source=strip_jsonp)['key']
 117
 118         formats = []
 119         for format_id, details in self._FORMATS.items():
 120             formats.append({
 121                 'url': 'http://cc.stream.qqmusic.qq.com/%s%s.%s?vkey=%s&guid=%s&fromtag=0'
 122                        % (details['prefix'], mid, details['ext'], vkey, guid),
 123                 'format': format_id,
 124                 'format_id': format_id,
 125                 'preference': details['preference'],
 126                 'abr': details.get('abr'),
 127             })
 128         self._check_formats(formats, mid)
 129         self._sort_formats(formats)
 130
 131         actual_lrc_lyrics = ''.join(
 132             line + '\n' for line in re.findall(
 133                 r'(?m)^(\[[0-9]{2}:[0-9]{2}(?:\.[0-9]{2,})?\][^\n]*|\[[^\]]*\])', lrc_content))
 134
 135         info_dict = {
 136             'id': mid,
 137             'formats': formats,
 138             'title': song_name,
 139             'release_date': publish_time,
 140             'creator': singer,
 141             'description': lrc_content,
 142             'thumbnail': thumbnail_url
 143         }
 144         if actual_lrc_lyrics:
 145             info_dict['subtitles'] = {
 146                 'origin': [{
 147                     'ext': 'lrc',
 148                     'data': actual_lrc_lyrics,
 149                 }]
 150             }
 151         return info_dict
 152
 153
 154 class QQPlaylistBaseIE(InfoExtractor):
 155     @staticmethod
 156     def qq_static_url(category, mid):
 157         return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
 158
 159     @classmethod
 160     def get_entries_from_page(cls, page):
 161         entries = []
 162
 163         for item in re.findall(r'class="data"[^<>]*>([^<>]+)</', page):
 164             song_mid = unescapeHTML(item).split('|')[-5]
 165             entries.append(cls.url_result(
 166                 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
 167                 song_mid))
 168
 169         return entries
 170
 171
 172 class QQMusicSingerIE(QQPlaylistBaseIE):
 173     IE_NAME = 'qqmusic:singer'
 174     IE_DESC = 'QQ音乐 - 歌手'
 175     _VALID_URL = r'https?://y\.qq\.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
 176     _TEST = {
 177         'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
 178         'info_dict': {
 179             'id': '001BLpXF2DyJe2',
 180             'title': '林俊杰',
 181             'description': 'md5:870ec08f7d8547c29c93010899103751',
 182         },
 183         'playlist_count': 12,
 184     }
 185
 186     def _real_extract(self, url):
 187         mid = self._match_id(url)
 188
 189         singer_page = self._download_webpage(
 190             self.qq_static_url('singer', mid), mid, 'Download singer page')
 191
 192         entries = self.get_entries_from_page(singer_page)
 193
 194         singer_name = self._html_search_regex(
 195             r"singername\s*:\s*'([^']+)'", singer_page, 'singer name',
 196             default=None)
 197
 198         singer_id = self._html_search_regex(
 199             r"singerid\s*:\s*'([0-9]+)'", singer_page, 'singer id',
 200             default=None)
 201
 202         singer_desc = None
 203
 204         if singer_id:
 205             req = sanitized_Request(
 206                 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
 207             req.add_header(
 208                 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
 209             singer_desc_page = self._download_xml(
 210                 req, mid, 'Donwload singer description XML')
 211
 212             singer_desc = singer_desc_page.find('./data/info/desc').text
 213
 214         return self.playlist_result(entries, mid, singer_name, singer_desc)
 215
 216
 217 class QQMusicAlbumIE(QQPlaylistBaseIE):
 218     IE_NAME = 'qqmusic:album'
 219     IE_DESC = 'QQ音乐 - 专辑'
 220     _VALID_URL = r'https?://y\.qq\.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
 221
 222     _TESTS = [{
 223         'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1',
 224         'info_dict': {
 225             'id': '000gXCTb2AhRR1',
 226             'title': '我们都是这样长大的',
 227             'description': 'md5:179c5dce203a5931970d306aa9607ea6',
 228         },
 229         'playlist_count': 4,
 230     }, {
 231         'url': 'http://y.qq.com/#type=album&mid=002Y5a3b3AlCu3',
 232         'info_dict': {
 233             'id': '002Y5a3b3AlCu3',
 234             'title': '그리고...',
 235             'description': 'md5:a48823755615508a95080e81b51ba729',
 236         },
 237         'playlist_count': 8,
 238     }]
 239
 240     def _real_extract(self, url):
 241         mid = self._match_id(url)
 242
 243         album = self._download_json(
 244             'http://i.y.qq.com/v8/fcg-bin/fcg_v8_album_info_cp.fcg?albummid=%s&format=json' % mid,
 245             mid, 'Download album page')['data']
 246
 247         entries = [
 248             self.url_result(
 249                 'http://y.qq.com/#type=song&mid=' + song['songmid'], 'QQMusic', song['songmid']
 250             ) for song in album['list']
 251         ]
 252         album_name = album.get('name')
 253         album_detail = album.get('desc')
 254         if album_detail is not None:
 255             album_detail = album_detail.strip()
 256
 257         return self.playlist_result(entries, mid, album_name, album_detail)
 258
 259
 260 class QQMusicToplistIE(QQPlaylistBaseIE):
 261     IE_NAME = 'qqmusic:toplist'
 262     IE_DESC = 'QQ音乐 - 排行榜'
 263     _VALID_URL = r'https?://y\.qq\.com/#type=toplist&p=(?P<id>(top|global)_[0-9]+)'
 264
 265     _TESTS = [{
 266         'url': 'http://y.qq.com/#type=toplist&p=global_123',
 267         'info_dict': {
 268             'id': 'global_123',
 269             'title': '美国iTunes榜',
 270         },
 271         'playlist_count': 10,
 272     }, {
 273         'url': 'http://y.qq.com/#type=toplist&p=top_3',
 274         'info_dict': {
 275             'id': 'top_3',
 276             'title': '巅峰榜·欧美',
 277             'description': 'QQ音乐巅峰榜·欧美根据用户收听行为自动生成，集结当下最流行的欧美新歌！:更新时间：每周四22点|统'
 278                            '计周期：一周（上周四至本周三）|统计对象：三个月内发行的欧美歌曲|统计数量：100首|统计算法：根据'
 279                            '歌曲在一周内的有效播放次数，由高到低取前100名（同一歌手最多允许5首歌曲同时上榜）|有效播放次数：'
 280                            '登录用户完整播放一首歌曲，记为一次有效播放；同一用户收听同一首歌曲，每天记录为1次有效播放'
 281         },
 282         'playlist_count': 100,
 283     }, {
 284         'url': 'http://y.qq.com/#type=toplist&p=global_106',
 285         'info_dict': {
 286             'id': 'global_106',
 287             'title': '韩国Mnet榜',
 288         },
 289         'playlist_count': 50,
 290     }]
 291
 292     def _real_extract(self, url):
 293         list_id = self._match_id(url)
 294
 295         list_type, num_id = list_id.split("_")
 296
 297         toplist_json = self._download_json(
 298             'http://i.y.qq.com/v8/fcg-bin/fcg_v8_toplist_cp.fcg?type=%s&topid=%s&format=json'
 299             % (list_type, num_id),
 300             list_id, 'Download toplist page')
 301
 302         entries = [
 303             self.url_result(
 304                 'http://y.qq.com/#type=song&mid=' + song['data']['songmid'], 'QQMusic', song['data']['songmid']
 305             ) for song in toplist_json['songlist']
 306         ]
 307
 308         topinfo = toplist_json.get('topinfo', {})
 309         list_name = topinfo.get('ListName')
 310         list_description = topinfo.get('info')
 311         return self.playlist_result(entries, list_id, list_name, list_description)
 312
 313
 314 class QQMusicPlaylistIE(QQPlaylistBaseIE):
 315     IE_NAME = 'qqmusic:playlist'
 316     IE_DESC = 'QQ音乐 - 歌单'
 317     _VALID_URL = r'https?://y\.qq\.com/#type=taoge&id=(?P<id>[0-9]+)'
 318
 319     _TESTS = [{
 320         'url': 'http://y.qq.com/#type=taoge&id=3462654915',
 321         'info_dict': {
 322             'id': '3462654915',
 323             'title': '韩国5月新歌精选下旬',
 324             'description': 'md5:d2c9d758a96b9888cf4fe82f603121d4',
 325         },
 326         'playlist_count': 40,
 327         'skip': 'playlist gone',
 328     }, {
 329         'url': 'http://y.qq.com/#type=taoge&id=1374105607',
 330         'info_dict': {
 331             'id': '1374105607',
 332             'title': '易入人心的华语民谣',
 333             '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。',
 334         },
 335         'playlist_count': 20,
 336     }]
 337
 338     def _real_extract(self, url):
 339         list_id = self._match_id(url)
 340
 341         list_json = self._download_json(
 342             'http://i.y.qq.com/qzone-music/fcg-bin/fcg_ucc_getcdinfo_byids_cp.fcg?type=1&json=1&utf8=1&onlysong=0&disstid=%s'
 343             % list_id, list_id, 'Download list page',
 344             transform_source=strip_jsonp)
 345         if not len(list_json.get('cdlist', [])):
 346             if list_json.get('code'):
 347                 raise ExtractorError(
 348                     'QQ Music said: error %d in fetching playlist info' % list_json['code'],
 349                     expected=True)
 350             raise ExtractorError('Unable to get playlist info')
 351
 352         cdlist = list_json['cdlist'][0]
 353         entries = [
 354             self.url_result(
 355                 'http://y.qq.com/#type=song&mid=' + song['songmid'], 'QQMusic', song['songmid']
 356             ) for song in cdlist['songlist']
 357         ]
 358
 359         list_name = cdlist.get('dissname')
 360         list_description = clean_html(unescapeHTML(cdlist.get('desc')))
 361         return self.playlist_result(entries, list_id, list_name, list_description)