_ 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     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},
  35         'mp3-128': {'prefix': 'M500', 'ext': 'mp3', 'preference': 30},
  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, sf 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                        % (sf['prefix'], mid, sf['ext'], vkey, guid),
  84                 'format': k, 'format_id': k, 'preference': sf['preference']
  85             })
  86         self._sort_formats(formats)
  87
  88         return {
  89             'id': mid,
  90             'formats': formats,
  91             'title': song_name,
  92             'upload_date': publish_time,
  93             'creator': singer,
  94             'description': lrc_content,
  95         }
  96
  97
  98 class QQPlaylistBaseIE(InfoExtractor):
  99     @staticmethod
 100     def qq_static_url(category, mid):
 101         return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
 102
 103     @classmethod
 104     def get_entries_from_page(cls, page):
 105         entries = []
 106
 107         for item in re.findall(r'class="data"[^<>]*>([^<>]+)</', page):
 108             song_mid = unescapeHTML(item).split('|')[-5]
 109             entries.append(cls.url_result(
 110                 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
 111                 song_mid))
 112
 113         return entries
 114
 115
 116 class QQMusicSingerIE(QQPlaylistBaseIE):
 117     IE_NAME = 'qqmusic:singer'
 118     _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
 119     _TEST = {
 120         'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
 121         'info_dict': {
 122             'id': '001BLpXF2DyJe2',
 123             'title': '林俊杰',
 124             'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
 125         },
 126         'playlist_count': 12,
 127     }
 128
 129     def _real_extract(self, url):
 130         mid = self._match_id(url)
 131
 132         singer_page = self._download_webpage(
 133             self.qq_static_url('singer', mid), mid, 'Download singer page')
 134
 135         entries = self.get_entries_from_page(singer_page)
 136
 137         singer_name = self._html_search_regex(
 138             r"singername\s*:\s*'([^']+)'", singer_page, 'singer name',
 139             default=None)
 140
 141         singer_id = self._html_search_regex(
 142             r"singerid\s*:\s*'([0-9]+)'", singer_page, 'singer id',
 143             default=None)
 144
 145         singer_desc = None
 146
 147         if singer_id:
 148             req = compat_urllib_request.Request(
 149                 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
 150             req.add_header(
 151                 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
 152             singer_desc_page = self._download_xml(
 153                 req, mid, 'Donwload singer description XML')
 154
 155             singer_desc = singer_desc_page.find('./data/info/desc').text
 156
 157         return self.playlist_result(entries, mid, singer_name, singer_desc)
 158
 159
 160 class QQMusicAlbumIE(QQPlaylistBaseIE):
 161     IE_NAME = 'qqmusic:album'
 162     _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
 163
 164     _TEST = {
 165         'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1&play=0',
 166         'info_dict': {
 167             'id': '000gXCTb2AhRR1',
 168             'title': '我们都是这样长大的',
 169             'description': 'md5:d216c55a2d4b3537fe4415b8767d74d6',
 170         },
 171         'playlist_count': 4,
 172     }
 173
 174     def _real_extract(self, url):
 175         mid = self._match_id(url)
 176
 177         album_page = self._download_webpage(
 178             self.qq_static_url('album', mid), mid, 'Download album page')
 179
 180         entries = self.get_entries_from_page(album_page)
 181
 182         album_name = self._html_search_regex(
 183             r"albumname\s*:\s*'([^']+)',", album_page, 'album name',
 184             default=None)
 185
 186         album_detail = self._html_search_regex(
 187             r'<div class="album_detail close_detail">\s*<p>((?:[^<>]+(?:<br />)?)+)</p>',
 188             album_page, 'album details', default=None)
 189
 190         return self.playlist_result(entries, mid, album_name, album_detail)
 191
 192
 193 class QQMusicToplistIE(QQPlaylistBaseIE):
 194     IE_NAME = 'qqmusic:toplist'
 195     _VALID_URL = r'http://y\.qq\.com/#type=toplist&p=(?P<id>(top|global)_[0-9]+)'
 196
 197     _TESTS = [{
 198         'url': 'http://y.qq.com/#type=toplist&p=global_12',
 199         'info_dict': {
 200             'id': 'global_12',
 201             'title': 'itunes榜',
 202         },
 203         'playlist_count': 10,
 204     }, {
 205         'url': 'http://y.qq.com/#type=toplist&p=top_6',
 206         'info_dict': {
 207             'id': 'top_6',
 208             'title': 'QQ音乐巅峰榜·欧美',
 209         },
 210         'playlist_count': 100,
 211     }, {
 212         'url': 'http://y.qq.com/#type=toplist&p=global_5',
 213         'info_dict': {
 214             'id': 'global_5',
 215             'title': '韩国mnet排行榜',
 216         },
 217         'playlist_count': 50,
 218     }]
 219
 220     @staticmethod
 221     def strip_qq_jsonp(code):
 222         return js_to_json(re.sub(r'^MusicJsonCallback\((.*?)\)/\*.+?\*/$', r'\1', code))
 223
 224     def _real_extract(self, url):
 225         list_id = self._match_id(url)
 226
 227         list_type, num_id = list_id.split("_")
 228
 229         list_page = self._download_webpage(
 230             "http://y.qq.com/y/static/toplist/index/%s.html" % list_id,
 231             list_id, 'Download toplist page')
 232
 233         entries = []
 234         if list_type == 'top':
 235             jsonp_url = "http://y.qq.com/y/static/toplist/json/top/%s/1.js" % num_id
 236         else:
 237             jsonp_url = "http://y.qq.com/y/static/toplist/json/global/%s/1_1.js" % num_id
 238
 239         toplist_json = self._download_json(
 240             jsonp_url, list_id, note='Retrieve toplist json',
 241             errnote='Unable to get toplist json', transform_source=self.strip_qq_jsonp)
 242
 243         for song in toplist_json['l']:
 244             s = song['s']
 245             song_mid = s.split("|")[20]
 246             entries.append(self.url_result(
 247                 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
 248                 song_mid))
 249
 250         list_name = self._html_search_regex(
 251             r'<h2 id="top_name">([^\']+)</h2>', list_page, 'top list name',
 252             default=None)
 253
 254         return self.playlist_result(entries, list_id, list_name)