_ Git - youtube-dl/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     strip_jsonp,
  11     unescapeHTML,
  12     js_to_json,
  13 )
  14 from ..compat import compat_urllib_request
  15
  16
  17 class QQMusicIE(InfoExtractor):
  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': 'bed90b6db2a7a7a7e11bc585f471f63a',
  22         'info_dict': {
  23             'id': '004295Et37taLD',
  24             'ext': 'm4a',
  25             'title': '可惜没如果',
  26             'upload_date': '20141227',
  27             'creator': '林俊杰',
  28             'description': 'md5:4348ff1dd24036906baa7b6f973f8d30',
  29         }
  30     }]
  31
  32     # Reference: m_r_GetRUin() in top_player.js
  33     # http://imgcache.gtimg.cn/music/portal_v3/y/top_player.js
  34     @staticmethod
  35     def m_r_get_ruin():
  36         curMs = int(time.time() * 1000) % 1000
  37         return int(round(random.random() * 2147483647) * curMs % 1E10)
  38
  39     def _real_extract(self, url):
  40         mid = self._match_id(url)
  41
  42         detail_info_page = self._download_webpage(
  43             'http://s.plcloud.music.qq.com/fcgi-bin/fcg_yqq_song_detail_info.fcg?songmid=%s&play=0' % mid,
  44             mid, note='Download song detail info',
  45             errnote='Unable to get song detail info', encoding='gbk')
  46
  47         song_name = self._html_search_regex(
  48             r"songname:\s*'([^']+)'", detail_info_page, 'song name')
  49
  50         publish_time = self._html_search_regex(
  51             r'发行时间：(\d{4}-\d{2}-\d{2})', detail_info_page,
  52             'publish time', default=None)
  53         if publish_time:
  54             publish_time = publish_time.replace('-', '')
  55
  56         singer = self._html_search_regex(
  57             r"singer:\s*'([^']+)", detail_info_page, 'singer', default=None)
  58
  59         lrc_content = self._html_search_regex(
  60             r'<div class="content" id="lrc_content"[^<>]*>([^<>]+)</div>',
  61             detail_info_page, 'LRC lyrics', default=None)
  62
  63         guid = self.m_r_get_ruin()
  64
  65         vkey = self._download_json(
  66             'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
  67             mid, note='Retrieve vkey', errnote='Unable to get vkey',
  68             transform_source=strip_jsonp)['key']
  69         song_url = 'http://cc.stream.qqmusic.qq.com/C200%s.m4a?vkey=%s&guid=%s&fromtag=0' % (mid, vkey, guid)
  70
  71         return {
  72             'id': mid,
  73             'url': song_url,
  74             'title': song_name,
  75             'upload_date': publish_time,
  76             'creator': singer,
  77             'description': lrc_content,
  78         }
  79
  80
  81 class QQPlaylistBaseIE(InfoExtractor):
  82     @staticmethod
  83     def qq_static_url(category, mid):
  84         return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
  85
  86     @classmethod
  87     def get_entries_from_page(cls, page):
  88         entries = []
  89
  90         for item in re.findall(r'class="data"[^<>]*>([^<>]+)</', page):
  91             song_mid = unescapeHTML(item).split('|')[-5]
  92             entries.append(cls.url_result(
  93                 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
  94                 song_mid))
  95
  96         return entries
  97
  98
  99 class QQMusicSingerIE(QQPlaylistBaseIE):
 100     _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
 101     _TEST = {
 102         'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
 103         'info_dict': {
 104             'id': '001BLpXF2DyJe2',
 105             'title': '林俊杰',
 106             'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
 107         },
 108         'playlist_count': 12,
 109     }
 110
 111     def _real_extract(self, url):
 112         mid = self._match_id(url)
 113
 114         singer_page = self._download_webpage(
 115             self.qq_static_url('singer', mid), mid, 'Download singer page')
 116
 117         entries = self.get_entries_from_page(singer_page)
 118
 119         singer_name = self._html_search_regex(
 120             r"singername\s*:\s*'([^']+)'", singer_page, 'singer name',
 121             default=None)
 122
 123         singer_id = self._html_search_regex(
 124             r"singerid\s*:\s*'([0-9]+)'", singer_page, 'singer id',
 125             default=None)
 126
 127         singer_desc = None
 128
 129         if singer_id:
 130             req = compat_urllib_request.Request(
 131                 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
 132             req.add_header(
 133                 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
 134             singer_desc_page = self._download_xml(
 135                 req, mid, 'Donwload singer description XML')
 136
 137             singer_desc = singer_desc_page.find('./data/info/desc').text
 138
 139         return self.playlist_result(entries, mid, singer_name, singer_desc)
 140
 141
 142 class QQMusicAlbumIE(QQPlaylistBaseIE):
 143     _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
 144
 145     _TEST = {
 146         'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1&play=0',
 147         'info_dict': {
 148             'id': '000gXCTb2AhRR1',
 149             'title': '我们都是这样长大的',
 150             'description': 'md5:d216c55a2d4b3537fe4415b8767d74d6',
 151         },
 152         'playlist_count': 4,
 153     }
 154
 155     def _real_extract(self, url):
 156         mid = self._match_id(url)
 157
 158         album_page = self._download_webpage(
 159             self.qq_static_url('album', mid), mid, 'Download album page')
 160
 161         entries = self.get_entries_from_page(album_page)
 162
 163         album_name = self._html_search_regex(
 164             r"albumname\s*:\s*'([^']+)',", album_page, 'album name',
 165             default=None)
 166
 167         album_detail = self._html_search_regex(
 168             r'<div class="album_detail close_detail">\s*<p>((?:[^<>]+(?:<br />)?)+)</p>',
 169             album_page, 'album details', default=None)
 170
 171         return self.playlist_result(entries, mid, album_name, album_detail)
 172
 173
 174 class QQMusicToplistIE(QQPlaylistBaseIE):
 175     _VALID_URL = r'http://y\.qq\.com/#type=toplist&p=(?P<id>(top|global)_[0-9]+)'
 176
 177     _TESTS = [{
 178         'url': 'http://y.qq.com/#type=toplist&p=global_12',
 179         'info_dict': {
 180             'id': 'global_12',
 181             'title': 'itunes榜',
 182         },
 183         'playlist_count': 10,
 184     }, {
 185         'url': 'http://y.qq.com/#type=toplist&p=top_6',
 186         'info_dict': {
 187             'id': 'top_6',
 188             'title': 'QQ音乐巅峰榜·欧美',
 189         },
 190         'playlist_count': 100,
 191     }, {
 192         'url': 'http://y.qq.com/#type=toplist&p=global_5',
 193         'info_dict': {
 194             'id': 'global_5',
 195             'title': '韩国mnet排行榜',
 196         },
 197         'playlist_count': 50,
 198     }]
 199
 200     @staticmethod
 201     def strip_qq_jsonp(code):
 202         return js_to_json(re.sub(r'^MusicJsonCallback\((.*?)\)/\*.+?\*/$', r'\1', code))
 203
 204     def _real_extract(self, url):
 205         list_id = self._match_id(url)
 206
 207         list_type = list_id.split("_")[0]
 208         num_id = list_id.split("_")[1]
 209
 210         list_page = self._download_webpage(
 211             "http://y.qq.com/y/static/toplist/index/%s.html" % list_id,
 212             list_id, 'Download toplist page')
 213
 214         entries = []
 215         jsonp_url = ""
 216         if list_type == 'top':
 217             jsonp_url = "http://y.qq.com/y/static/toplist/json/top/%s/1.js" % num_id
 218         elif list_type == 'global':
 219             jsonp_url = "http://y.qq.com/y/static/toplist/json/global/%s/1_1.js" % num_id
 220
 221         list = self._download_json(jsonp_url, list_id, note='Retrieve toplist json',
 222             errnote='Unable to get toplist json', transform_source=self.strip_qq_jsonp)
 223
 224         for song in list['l']:
 225             s = song['s']
 226             song_mid = s.split("|")[20]
 227             entries.append(self.url_result(
 228                 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
 229                 song_mid))
 230
 231         list_name = self._html_search_regex(
 232             r'<h2 id="top_name">([^\']+)</h2>', list_page, 'top list name',
 233             default=None)
 234         list_desc = None
 235
 236         return self.playlist_result(entries, list_id, list_name, list_desc)
 237