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     js_to_json,
  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         }
  31     }]
  32
  33     _FORMATS = {
  34         'mp3-320': {'prefix': 'M800', 'ext': 'mp3', 'preference': 40, 'abr': 320},
  35         'mp3-128': {'prefix': 'M500', 'ext': 'mp3', 'preference': 30, 'abr': 128},
  36         'm4a': {'prefix': 'C200', 'ext': 'm4a', 'preference': 10}
  37     }
  38
  39     # Reference: m_r_GetRUin() in top_player.js
  40     # http://imgcache.gtimg.cn/music/portal_v3/y/top_player.js
  41     @staticmethod
  42     def m_r_get_ruin():
  43         curMs = int(time.time() * 1000) % 1000
  44         return int(round(random.random() * 2147483647) * curMs % 1E10)
  45
  46     def _real_extract(self, url):
  47         mid = self._match_id(url)
  48
  49         detail_info_page = self._download_webpage(
  50             'http://s.plcloud.music.qq.com/fcgi-bin/fcg_yqq_song_detail_info.fcg?songmid=%s&play=0' % mid,
  51             mid, note='Download song detail info',
  52             errnote='Unable to get song detail info', encoding='gbk')
  53
  54         song_name = self._html_search_regex(
  55             r"songname:\s*'([^']+)'", detail_info_page, 'song name')
  56
  57         publish_time = self._html_search_regex(
  58             r'发行时间：(\d{4}-\d{2}-\d{2})', detail_info_page,
  59             'publish time', default=None)
  60         if publish_time:
  61             publish_time = publish_time.replace('-', '')
  62
  63         singer = self._html_search_regex(
  64             r"singer:\s*'([^']+)", detail_info_page, 'singer', default=None)
  65
  66         lrc_content = self._html_search_regex(
  67             r'<div class="content" id="lrc_content"[^<>]*>([^<>]+)</div>',
  68             detail_info_page, 'LRC lyrics', default=None)
  69         if lrc_content:
  70             lrc_content = lrc_content.replace('\\n', '\n')
  71
  72         guid = self.m_r_get_ruin()
  73
  74         vkey = self._download_json(
  75             'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
  76             mid, note='Retrieve vkey', errnote='Unable to get vkey',
  77             transform_source=strip_jsonp)['key']
  78
  79         formats = []
  80         for k, f in self._FORMATS.items():
  81             formats.append({
  82                 'url': 'http://cc.stream.qqmusic.qq.com/%s%s.%s?vkey=%s&guid=%s&fromtag=0'
  83                        % (f['prefix'], mid, f['ext'], vkey, guid),
  84                 'format': k, 'format_id': k, 'preference': f['preference'],
  85                 'abr': f.get('abr')
  86             })
  87         self._sort_formats(formats)
  88
  89         return {
  90             'id': mid,
  91             'formats': formats,
  92             'title': song_name,
  93             'upload_date': publish_time,
  94             'creator': singer,
  95             'description': lrc_content,
  96         }
  97
  98
  99 class QQPlaylistBaseIE(InfoExtractor):
 100     @staticmethod
 101     def qq_static_url(category, mid):
 102         return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
 103
 104     @classmethod
 105     def get_entries_from_page(cls, page):
 106         entries = []
 107
 108         for item in re.findall(r'class="data"[^<>]*>([^<>]+)</', page):
 109             song_mid = unescapeHTML(item).split('|')[-5]
 110             entries.append(cls.url_result(
 111                 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
 112                 song_mid))
 113
 114         return entries
 115
 116
 117 class QQMusicSingerIE(QQPlaylistBaseIE):
 118     IE_NAME = 'qqmusic:singer'
 119     _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
 120     _TEST = {
 121         'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
 122         'info_dict': {
 123             'id': '001BLpXF2DyJe2',
 124             'title': '林俊杰',
 125             'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
 126         },
 127         'playlist_count': 12,
 128     }
 129
 130     def _real_extract(self, url):
 131         mid = self._match_id(url)
 132
 133         singer_page = self._download_webpage(
 134             self.qq_static_url('singer', mid), mid, 'Download singer page')
 135
 136         entries = self.get_entries_from_page(singer_page)
 137
 138         singer_name = self._html_search_regex(
 139             r"singername\s*:\s*'([^']+)'", singer_page, 'singer name',
 140             default=None)
 141
 142         singer_id = self._html_search_regex(
 143             r"singerid\s*:\s*'([0-9]+)'", singer_page, 'singer id',
 144             default=None)
 145
 146         singer_desc = None
 147
 148         if singer_id:
 149             req = compat_urllib_request.Request(
 150                 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
 151             req.add_header(
 152                 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
 153             singer_desc_page = self._download_xml(
 154                 req, mid, 'Donwload singer description XML')
 155
 156             singer_desc = singer_desc_page.find('./data/info/desc').text
 157
 158         return self.playlist_result(entries, mid, singer_name, singer_desc)
 159
 160
 161 class QQMusicAlbumIE(QQPlaylistBaseIE):
 162     IE_NAME = 'qqmusic:album'
 163     _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
 164
 165     _TEST = {
 166         'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1&play=0',
 167         'info_dict': {
 168             'id': '000gXCTb2AhRR1',
 169             'title': '我们都是这样长大的',
 170             'description': 'md5:d216c55a2d4b3537fe4415b8767d74d6',
 171         },
 172         'playlist_count': 4,
 173     }
 174
 175     def _real_extract(self, url):
 176         mid = self._match_id(url)
 177
 178         album_page = self._download_webpage(
 179             self.qq_static_url('album', mid), mid, 'Download album page')
 180
 181         entries = self.get_entries_from_page(album_page)
 182
 183         album_name = self._html_search_regex(
 184             r"albumname\s*:\s*'([^']+)',", album_page, 'album name',
 185             default=None)
 186
 187         album_detail = self._html_search_regex(
 188             r'<div class="album_detail close_detail">\s*<p>((?:[^<>]+(?:<br />)?)+)</p>',
 189             album_page, 'album details', default=None)
 190
 191         return self.playlist_result(entries, mid, album_name, album_detail)
 192
 193
 194 class QQMusicToplistIE(QQPlaylistBaseIE):
 195     IE_NAME = 'qqmusic:toplist'
 196     _VALID_URL = r'http://y\.qq\.com/#type=toplist&p=(?P<id>(top|global)_[0-9]+)'
 197
 198     _TESTS = [{
 199         'url': 'http://y.qq.com/#type=toplist&p=global_12',
 200         'info_dict': {
 201             'id': 'global_12',
 202             'title': 'itunes榜',
 203         },
 204         'playlist_count': 10,
 205     }, {
 206         'url': 'http://y.qq.com/#type=toplist&p=top_6',
 207         'info_dict': {
 208             'id': 'top_6',
 209             'title': 'QQ音乐巅峰榜·欧美',
 210         },
 211         'playlist_count': 100,
 212     }, {
 213         'url': 'http://y.qq.com/#type=toplist&p=global_5',
 214         'info_dict': {
 215             'id': 'global_5',
 216             'title': '韩国mnet排行榜',
 217         },
 218         'playlist_count': 50,
 219     }]
 220
 221     @staticmethod
 222     def strip_qq_jsonp(code):
 223         return js_to_json(re.sub(r'^MusicJsonCallback\((.*?)\)/\*.+?\*/$', r'\1', code))
 224
 225     def _real_extract(self, url):
 226         list_id = self._match_id(url)
 227
 228         list_type, num_id = list_id.split("_")
 229
 230         list_page = self._download_webpage(
 231             "http://y.qq.com/y/static/toplist/index/%s.html" % list_id,
 232             list_id, 'Download toplist page')
 233
 234         entries = []
 235         if list_type == 'top':
 236             jsonp_url = "http://y.qq.com/y/static/toplist/json/top/%s/1.js" % num_id
 237         else:
 238             jsonp_url = "http://y.qq.com/y/static/toplist/json/global/%s/1_1.js" % num_id
 239
 240         toplist_json = self._download_json(
 241             jsonp_url, list_id, note='Retrieve toplist json',
 242             errnote='Unable to get toplist json', transform_source=self.strip_qq_jsonp)
 243
 244         for song in toplist_json['l']:
 245             s = song['s']
 246             song_mid = s.split("|")[20]
 247             entries.append(self.url_result(
 248                 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
 249                 song_mid))
 250
 251         list_name = self._html_search_regex(
 252             r'<h2 id="top_name">([^\']+)</h2>', list_page, 'top list name',
 253             default=None)
 254
 255         return self.playlist_result(entries, list_id, list_name)