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 )
  13 from ..compat import compat_urllib_request
  14
  15
  16 class QQMusicIE(InfoExtractor):
  17     IE_NAME = 'qqmusic'
  18     _VALID_URL = r'http://y.qq.com/#type=song&mid=(?P<id>[0-9A-Za-z]+)'
  19     _TESTS = [{
  20         'url': 'http://y.qq.com/#type=song&mid=004295Et37taLD',
  21         'md5': '9ce1c1c8445f561506d2e3cfb0255705',
  22         'info_dict': {
  23             'id': '004295Et37taLD',
  24             'ext': 'mp3',
  25             'title': '可惜没如果',
  26             'upload_date': '20141227',
  27             'creator': '林俊杰',
  28             'description': 'md5:d327722d0361576fde558f1ac68a7065',
  29         }
  30     }]
  31
  32     _FORMATS = {
  33         'mp3-320': {'prefix': 'M800', 'ext': 'mp3', 'preference': 40, 'abr': 320},
  34         'mp3-128': {'prefix': 'M500', 'ext': 'mp3', 'preference': 30, 'abr': 128},
  35         'm4a': {'prefix': 'C200', 'ext': 'm4a', 'preference': 10}
  36     }
  37
  38     # Reference: m_r_GetRUin() in top_player.js
  39     # http://imgcache.gtimg.cn/music/portal_v3/y/top_player.js
  40     @staticmethod
  41     def m_r_get_ruin():
  42         curMs = int(time.time() * 1000) % 1000
  43         return int(round(random.random() * 2147483647) * curMs % 1E10)
  44
  45     def _real_extract(self, url):
  46         mid = self._match_id(url)
  47
  48         detail_info_page = self._download_webpage(
  49             'http://s.plcloud.music.qq.com/fcgi-bin/fcg_yqq_song_detail_info.fcg?songmid=%s&play=0' % mid,
  50             mid, note='Download song detail info',
  51             errnote='Unable to get song detail info', encoding='gbk')
  52
  53         song_name = self._html_search_regex(
  54             r"songname:\s*'([^']+)'", detail_info_page, 'song name')
  55
  56         publish_time = self._html_search_regex(
  57             r'发行时间：(\d{4}-\d{2}-\d{2})', detail_info_page,
  58             'publish time', default=None)
  59         if publish_time:
  60             publish_time = publish_time.replace('-', '')
  61
  62         singer = self._html_search_regex(
  63             r"singer:\s*'([^']+)", detail_info_page, 'singer', default=None)
  64
  65         lrc_content = self._html_search_regex(
  66             r'<div class="content" id="lrc_content"[^<>]*>([^<>]+)</div>',
  67             detail_info_page, 'LRC lyrics', default=None)
  68         if lrc_content:
  69             lrc_content = lrc_content.replace('\\n', '\n')
  70
  71         guid = self.m_r_get_ruin()
  72
  73         vkey = self._download_json(
  74             'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
  75             mid, note='Retrieve vkey', errnote='Unable to get vkey',
  76             transform_source=strip_jsonp)['key']
  77
  78         formats = []
  79         for format_id, details in self._FORMATS.items():
  80             formats.append({
  81                 'url': 'http://cc.stream.qqmusic.qq.com/%s%s.%s?vkey=%s&guid=%s&fromtag=0'
  82                        % (details['prefix'], mid, details['ext'], vkey, guid),
  83                 'format': format_id,
  84                 'format_id': format_id,
  85                 'preference': details['preference'],
  86                 'abr': details.get('abr'),
  87             })
  88         self._sort_formats(formats)
  89
  90         return {
  91             'id': mid,
  92             'formats': formats,
  93             'title': song_name,
  94             'upload_date': publish_time,
  95             'creator': singer,
  96             'description': lrc_content,
  97         }
  98
  99
 100 class QQPlaylistBaseIE(InfoExtractor):
 101     @staticmethod
 102     def qq_static_url(category, mid):
 103         return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
 104
 105     @classmethod
 106     def get_entries_from_page(cls, page):
 107         entries = []
 108
 109         for item in re.findall(r'class="data"[^<>]*>([^<>]+)</', page):
 110             song_mid = unescapeHTML(item).split('|')[-5]
 111             entries.append(cls.url_result(
 112                 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
 113                 song_mid))
 114
 115         return entries
 116
 117
 118 class QQMusicSingerIE(QQPlaylistBaseIE):
 119     IE_NAME = 'qqmusic:singer'
 120     _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
 121     _TEST = {
 122         'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
 123         'info_dict': {
 124             'id': '001BLpXF2DyJe2',
 125             'title': '林俊杰',
 126             'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
 127         },
 128         'playlist_count': 12,
 129     }
 130
 131     def _real_extract(self, url):
 132         mid = self._match_id(url)
 133
 134         singer_page = self._download_webpage(
 135             self.qq_static_url('singer', mid), mid, 'Download singer page')
 136
 137         entries = self.get_entries_from_page(singer_page)
 138
 139         singer_name = self._html_search_regex(
 140             r"singername\s*:\s*'([^']+)'", singer_page, 'singer name',
 141             default=None)
 142
 143         singer_id = self._html_search_regex(
 144             r"singerid\s*:\s*'([0-9]+)'", singer_page, 'singer id',
 145             default=None)
 146
 147         singer_desc = None
 148
 149         if singer_id:
 150             req = compat_urllib_request.Request(
 151                 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
 152             req.add_header(
 153                 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
 154             singer_desc_page = self._download_xml(
 155                 req, mid, 'Donwload singer description XML')
 156
 157             singer_desc = singer_desc_page.find('./data/info/desc').text
 158
 159         return self.playlist_result(entries, mid, singer_name, singer_desc)
 160
 161
 162 class QQMusicAlbumIE(QQPlaylistBaseIE):
 163     IE_NAME = 'qqmusic:album'
 164     _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
 165
 166     _TEST = {
 167         'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1&play=0',
 168         'info_dict': {
 169             'id': '000gXCTb2AhRR1',
 170             'title': '我们都是这样长大的',
 171             'description': 'md5:d216c55a2d4b3537fe4415b8767d74d6',
 172         },
 173         'playlist_count': 4,
 174     }
 175
 176     def _real_extract(self, url):
 177         mid = self._match_id(url)
 178
 179         album_page = self._download_webpage(
 180             self.qq_static_url('album', mid), mid, 'Download album page')
 181
 182         entries = self.get_entries_from_page(album_page)
 183
 184         album_name = self._html_search_regex(
 185             r"albumname\s*:\s*'([^']+)',", album_page, 'album name',
 186             default=None)
 187
 188         album_detail = self._html_search_regex(
 189             r'<div class="album_detail close_detail">\s*<p>((?:[^<>]+(?:<br />)?)+)</p>',
 190             album_page, 'album details', default=None)
 191
 192         return self.playlist_result(entries, mid, album_name, album_detail)
 193
 194
 195 class QQMusicToplistIE(QQPlaylistBaseIE):
 196     IE_NAME = 'qqmusic:toplist'
 197     _VALID_URL = r'http://y\.qq\.com/#type=toplist&p=(?P<id>(top|global)_[0-9]+)'
 198
 199     _TESTS = [{
 200         'url': 'http://y.qq.com/#type=toplist&p=global_123',
 201         'info_dict': {
 202             'id': 'global_123',
 203             'title': '美国iTunes榜',
 204         },
 205         'playlist_count': 10,
 206     }, {
 207         'url': 'http://y.qq.com/#type=toplist&p=top_3',
 208         'info_dict': {
 209             'id': 'top_3',
 210             'title': 'QQ音乐巅峰榜·欧美',
 211             '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'
 212                            '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'
 213                            '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：'
 214                            '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'
 215         },
 216         'playlist_count': 100,
 217     }, {
 218         'url': 'http://y.qq.com/#type=toplist&p=global_106',
 219         'info_dict': {
 220             'id': 'global_106',
 221             'title': '韩国Mnet榜',
 222         },
 223         'playlist_count': 50,
 224     }]
 225
 226     def _real_extract(self, url):
 227         list_id = self._match_id(url)
 228
 229         list_type, num_id = list_id.split("_")
 230
 231         toplist_json = self._download_json(
 232             'http://i.y.qq.com/v8/fcg-bin/fcg_v8_toplist_cp.fcg?type=%s&topid=%s&format=json'
 233             % (list_type, num_id),
 234             list_id, 'Download toplist page')
 235
 236         entries = [
 237             self.url_result(
 238                 'http://y.qq.com/#type=song&mid=' + song['data']['songmid'], 'QQMusic', song['data']['songmid']
 239             ) for song in toplist_json['songlist']
 240         ]
 241
 242         topinfo = toplist_json.get('topinfo', {})
 243         list_name = topinfo.get('ListName')
 244         list_description = topinfo.get('info')
 245         return self.playlist_result(entries, list_id, list_name, list_description)