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     strip_jsonp,
  11     unescapeHTML,
  12     clean_html,
  13 )
  14 from ..compat import compat_urllib_request
  15
  16
  17 class QQMusicIE(InfoExtractor):
  18     IE_NAME = 'qqmusic'
  19     _VALID_URL = r'http://y.qq.com/#type=song&mid=(?P<id>[0-9A-Za-z]+)'
  20     _TESTS = [{
  21         'url': 'http://y.qq.com/#type=song&mid=004295Et37taLD',
  22         'md5': '9ce1c1c8445f561506d2e3cfb0255705',
  23         'info_dict': {
  24             'id': '004295Et37taLD',
  25             'ext': 'mp3',
  26             'title': '可惜没如果',
  27             'upload_date': '20141227',
  28             'creator': '林俊杰',
  29             'description': 'md5:d327722d0361576fde558f1ac68a7065',
  30             'thumbnail': 're:^https?://.*\.jpg$',
  31         }
  32     }, {
  33         'note': 'There is no mp3-320 version of this song.',
  34         'url': 'http://y.qq.com/#type=song&mid=004MsGEo3DdNxV',
  35         'md5': 'fa3926f0c585cda0af8fa4f796482e3e',
  36         'info_dict': {
  37             'id': '004MsGEo3DdNxV',
  38             'ext': 'mp3',
  39             'title': '如果',
  40             'upload_date': '20050626',
  41             'creator': '李季美',
  42             'description': 'md5:46857d5ed62bc4ba84607a805dccf437',
  43             'thumbnail': 're:^https?://.*\.jpg$',
  44         }
  45     }]
  46
  47     _FORMATS = {
  48         'mp3-320': {'prefix': 'M800', 'ext': 'mp3', 'preference': 40, 'abr': 320},
  49         'mp3-128': {'prefix': 'M500', 'ext': 'mp3', 'preference': 30, 'abr': 128},
  50         'm4a': {'prefix': 'C200', 'ext': 'm4a', 'preference': 10}
  51     }
  52
  53     # Reference: m_r_GetRUin() in top_player.js
  54     # http://imgcache.gtimg.cn/music/portal_v3/y/top_player.js
  55     @staticmethod
  56     def m_r_get_ruin():
  57         curMs = int(time.time() * 1000) % 1000
  58         return int(round(random.random() * 2147483647) * curMs % 1E10)
  59
  60     def _real_extract(self, url):
  61         mid = self._match_id(url)
  62
  63         detail_info_page = self._download_webpage(
  64             'http://s.plcloud.music.qq.com/fcgi-bin/fcg_yqq_song_detail_info.fcg?songmid=%s&play=0' % mid,
  65             mid, note='Download song detail info',
  66             errnote='Unable to get song detail info', encoding='gbk')
  67
  68         song_name = self._html_search_regex(
  69             r"songname:\s*'([^']+)'", detail_info_page, 'song name')
  70
  71         publish_time = self._html_search_regex(
  72             r'发行时间：(\d{4}-\d{2}-\d{2})', detail_info_page,
  73             'publish time', default=None)
  74         if publish_time:
  75             publish_time = publish_time.replace('-', '')
  76
  77         singer = self._html_search_regex(
  78             r"singer:\s*'([^']+)", detail_info_page, 'singer', default=None)
  79
  80         lrc_content = self._html_search_regex(
  81             r'<div class="content" id="lrc_content"[^<>]*>([^<>]+)</div>',
  82             detail_info_page, 'LRC lyrics', default=None)
  83         if lrc_content:
  84             lrc_content = lrc_content.replace('\\n', '\n')
  85
  86         thumbnail_url = None
  87         albummid = self._search_regex(
  88             [r'albummid:\'([0-9a-zA-Z]+)\'', r'"albummid":"([0-9a-zA-Z]+)"'],
  89             detail_info_page, 'album mid', default=None)
  90         if albummid:
  91             thumbnail_url = "http://i.gtimg.cn/music/photo/mid_album_500/%s/%s/%s.jpg" \
  92                             % (albummid[-2:-1], albummid[-1], albummid)
  93
  94         guid = self.m_r_get_ruin()
  95
  96         vkey = self._download_json(
  97             'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
  98             mid, note='Retrieve vkey', errnote='Unable to get vkey',
  99             transform_source=strip_jsonp)['key']
 100
 101         formats = []
 102         for format_id, details in self._FORMATS.items():
 103             formats.append({
 104                 'url': 'http://cc.stream.qqmusic.qq.com/%s%s.%s?vkey=%s&guid=%s&fromtag=0'
 105                        % (details['prefix'], mid, details['ext'], vkey, guid),
 106                 'format': format_id,
 107                 'format_id': format_id,
 108                 'preference': details['preference'],
 109                 'abr': details.get('abr'),
 110             })
 111         self._check_formats(formats, mid)
 112         self._sort_formats(formats)
 113
 114         return {
 115             'id': mid,
 116             'formats': formats,
 117             'title': song_name,
 118             'upload_date': publish_time,
 119             'creator': singer,
 120             'description': lrc_content,
 121             'thumbnail': thumbnail_url,
 122         }
 123
 124
 125 class QQPlaylistBaseIE(InfoExtractor):
 126     @staticmethod
 127     def qq_static_url(category, mid):
 128         return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
 129
 130     @classmethod
 131     def get_entries_from_page(cls, page):
 132         entries = []
 133
 134         for item in re.findall(r'class="data"[^<>]*>([^<>]+)</', page):
 135             song_mid = unescapeHTML(item).split('|')[-5]
 136             entries.append(cls.url_result(
 137                 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
 138                 song_mid))
 139
 140         return entries
 141
 142
 143 class QQMusicSingerIE(QQPlaylistBaseIE):
 144     IE_NAME = 'qqmusic:singer'
 145     _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
 146     _TEST = {
 147         'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
 148         'info_dict': {
 149             'id': '001BLpXF2DyJe2',
 150             'title': '林俊杰',
 151             'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
 152         },
 153         'playlist_count': 12,
 154     }
 155
 156     def _real_extract(self, url):
 157         mid = self._match_id(url)
 158
 159         singer_page = self._download_webpage(
 160             self.qq_static_url('singer', mid), mid, 'Download singer page')
 161
 162         entries = self.get_entries_from_page(singer_page)
 163
 164         singer_name = self._html_search_regex(
 165             r"singername\s*:\s*'([^']+)'", singer_page, 'singer name',
 166             default=None)
 167
 168         singer_id = self._html_search_regex(
 169             r"singerid\s*:\s*'([0-9]+)'", singer_page, 'singer id',
 170             default=None)
 171
 172         singer_desc = None
 173
 174         if singer_id:
 175             req = compat_urllib_request.Request(
 176                 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
 177             req.add_header(
 178                 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
 179             singer_desc_page = self._download_xml(
 180                 req, mid, 'Donwload singer description XML')
 181
 182             singer_desc = singer_desc_page.find('./data/info/desc').text
 183
 184         return self.playlist_result(entries, mid, singer_name, singer_desc)
 185
 186
 187 class QQMusicAlbumIE(QQPlaylistBaseIE):
 188     IE_NAME = 'qqmusic:album'
 189     _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
 190
 191     _TESTS = [{
 192         'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1',
 193         'info_dict': {
 194             'id': '000gXCTb2AhRR1',
 195             'title': '我们都是这样长大的',
 196             'description': 'md5:179c5dce203a5931970d306aa9607ea6',
 197         },
 198         'playlist_count': 4,
 199     }, {
 200         'url': 'http://y.qq.com/#type=album&mid=002Y5a3b3AlCu3',
 201         'info_dict': {
 202             'id': '002Y5a3b3AlCu3',
 203             'title': '그리고...',
 204             'description': 'md5:a48823755615508a95080e81b51ba729',
 205         },
 206         'playlist_count': 8,
 207     }]
 208
 209     def _real_extract(self, url):
 210         mid = self._match_id(url)
 211
 212         album = self._download_json(
 213             'http://i.y.qq.com/v8/fcg-bin/fcg_v8_album_info_cp.fcg?albummid=%s&format=json' % mid,
 214             mid, 'Download album page')['data']
 215
 216         entries = [
 217             self.url_result(
 218                 'http://y.qq.com/#type=song&mid=' + song['songmid'], 'QQMusic', song['songmid']
 219             ) for song in album['list']
 220         ]
 221         album_name = album.get('name')
 222         album_detail = album.get('desc')
 223         if album_detail is not None:
 224             album_detail = album_detail.strip()
 225
 226         return self.playlist_result(entries, mid, album_name, album_detail)
 227
 228
 229 class QQMusicToplistIE(QQPlaylistBaseIE):
 230     IE_NAME = 'qqmusic:toplist'
 231     _VALID_URL = r'http://y\.qq\.com/#type=toplist&p=(?P<id>(top|global)_[0-9]+)'
 232
 233     _TESTS = [{
 234         'url': 'http://y.qq.com/#type=toplist&p=global_123',
 235         'info_dict': {
 236             'id': 'global_123',
 237             'title': '美国iTunes榜',
 238         },
 239         'playlist_count': 10,
 240     }, {
 241         'url': 'http://y.qq.com/#type=toplist&p=top_3',
 242         'info_dict': {
 243             'id': 'top_3',
 244             'title': 'QQ音乐巅峰榜·欧美',
 245             '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'
 246                            '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'
 247                            '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：'
 248                            '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'
 249         },
 250         'playlist_count': 100,
 251     }, {
 252         'url': 'http://y.qq.com/#type=toplist&p=global_106',
 253         'info_dict': {
 254             'id': 'global_106',
 255             'title': '韩国Mnet榜',
 256         },
 257         'playlist_count': 50,
 258     }]
 259
 260     def _real_extract(self, url):
 261         list_id = self._match_id(url)
 262
 263         list_type, num_id = list_id.split("_")
 264
 265         toplist_json = self._download_json(
 266             'http://i.y.qq.com/v8/fcg-bin/fcg_v8_toplist_cp.fcg?type=%s&topid=%s&format=json'
 267             % (list_type, num_id),
 268             list_id, 'Download toplist page')
 269
 270         entries = [
 271             self.url_result(
 272                 'http://y.qq.com/#type=song&mid=' + song['data']['songmid'], 'QQMusic', song['data']['songmid']
 273             ) for song in toplist_json['songlist']
 274         ]
 275
 276         topinfo = toplist_json.get('topinfo', {})
 277         list_name = topinfo.get('ListName')
 278         list_description = topinfo.get('info')
 279         return self.playlist_result(entries, list_id, list_name, list_description)
 280
 281
 282 class QQMusicPlaylistIE(QQPlaylistBaseIE):
 283     IE_NAME = 'qqmusic:playlist'
 284     _VALID_URL = r'http://y\.qq\.com/#type=taoge&id=(?P<id>[0-9]+)'
 285
 286     _TEST = {
 287         'url': 'http://y.qq.com/#type=taoge&id=3462654915',
 288         'info_dict': {
 289             'id': '3462654915',
 290             'title': '韩国5月新歌精选下旬',
 291             'description': 'md5:d2c9d758a96b9888cf4fe82f603121d4',
 292         },
 293         'playlist_count': 40,
 294     }
 295
 296     def _real_extract(self, url):
 297         list_id = self._match_id(url)
 298
 299         list_json = self._download_json(
 300             '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'
 301             % list_id, list_id, 'Download list page',
 302             transform_source=strip_jsonp)['cdlist'][0]
 303
 304         entries = [
 305             self.url_result(
 306                 'http://y.qq.com/#type=song&mid=' + song['songmid'], 'QQMusic', song['songmid']
 307             ) for song in list_json['songlist']
 308         ]
 309
 310         list_name = list_json.get('dissname')
 311         list_description = clean_html(unescapeHTML(list_json.get('desc')))
 312         return self.playlist_result(entries, list_id, list_name, list_description)