Raphaël G. Git Repositories - youtubedl/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 )
  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': '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:d327722d0361576fde558f1ac68a7065',
  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         if lrc_content:
  63             lrc_content = lrc_content.replace('\\n', '\n')
  64
  65         guid = self.m_r_get_ruin()
  66
  67         vkey = self._download_json(
  68             'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
  69             mid, note='Retrieve vkey', errnote='Unable to get vkey',
  70             transform_source=strip_jsonp)['key']
  71         song_url = 'http://cc.stream.qqmusic.qq.com/C200%s.m4a?vkey=%s&guid=%s&fromtag=0' % (mid, vkey, guid)
  72
  73         return {
  74             'id': mid,
  75             'url': song_url,
  76             'title': song_name,
  77             'upload_date': publish_time,
  78             'creator': singer,
  79             'description': lrc_content,
  80         }
  81
  82
  83 class QQPlaylistBaseIE(InfoExtractor):
  84     @staticmethod
  85     def qq_static_url(category, mid):
  86         return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
  87
  88     @classmethod
  89     def get_entries_from_page(cls, page):
  90         entries = []
  91
  92         for item in re.findall(r'class="data"[^<>]*>([^<>]+)</', page):
  93             song_mid = unescapeHTML(item).split('|')[-5]
  94             entries.append(cls.url_result(
  95                 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
  96                 song_mid))
  97
  98         return entries
  99
 100
 101 class QQMusicSingerIE(QQPlaylistBaseIE):
 102     IE_NAME = 'qqmusic:singer'
 103     _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
 104     _TEST = {
 105         'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
 106         'info_dict': {
 107             'id': '001BLpXF2DyJe2',
 108             'title': '林俊杰',
 109             'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
 110         },
 111         'playlist_count': 12,
 112     }
 113
 114     def _real_extract(self, url):
 115         mid = self._match_id(url)
 116
 117         singer_page = self._download_webpage(
 118             self.qq_static_url('singer', mid), mid, 'Download singer page')
 119
 120         entries = self.get_entries_from_page(singer_page)
 121
 122         singer_name = self._html_search_regex(
 123             r"singername\s*:\s*'([^']+)'", singer_page, 'singer name',
 124             default=None)
 125
 126         singer_id = self._html_search_regex(
 127             r"singerid\s*:\s*'([0-9]+)'", singer_page, 'singer id',
 128             default=None)
 129
 130         singer_desc = None
 131
 132         if singer_id:
 133             req = compat_urllib_request.Request(
 134                 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
 135             req.add_header(
 136                 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
 137             singer_desc_page = self._download_xml(
 138                 req, mid, 'Donwload singer description XML')
 139
 140             singer_desc = singer_desc_page.find('./data/info/desc').text
 141
 142         return self.playlist_result(entries, mid, singer_name, singer_desc)
 143
 144
 145 class QQMusicAlbumIE(QQPlaylistBaseIE):
 146     IE_NAME = 'qqmusic:album'
 147     _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
 148
 149     _TEST = {
 150         'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1&play=0',
 151         'info_dict': {
 152             'id': '000gXCTb2AhRR1',
 153             'title': '我们都是这样长大的',
 154             'description': 'md5:d216c55a2d4b3537fe4415b8767d74d6',
 155         },
 156         'playlist_count': 4,
 157     }
 158
 159     def _real_extract(self, url):
 160         mid = self._match_id(url)
 161
 162         album_page = self._download_webpage(
 163             self.qq_static_url('album', mid), mid, 'Download album page')
 164
 165         entries = self.get_entries_from_page(album_page)
 166
 167         album_name = self._html_search_regex(
 168             r"albumname\s*:\s*'([^']+)',", album_page, 'album name',
 169             default=None)
 170
 171         album_detail = self._html_search_regex(
 172             r'<div class="album_detail close_detail">\s*<p>((?:[^<>]+(?:<br />)?)+)</p>',
 173             album_page, 'album details', default=None)
 174
 175         return self.playlist_result(entries, mid, album_name, album_detail)
 176
 177
 178 class QQMusicToplistIE(QQPlaylistBaseIE):
 179     IE_NAME = 'qqmusic:toplist'
 180     _VALID_URL = r'http://y\.qq\.com/#type=toplist&p=(?P<id>(top|global)_[0-9]+)'
 181
 182     _TESTS = [{
 183         'url': 'http://y.qq.com/#type=toplist&p=global_123',
 184         'info_dict': {
 185             'id': 'global_123',
 186             'title': '美国iTunes榜',
 187         },
 188         'playlist_count': 10,
 189     }, {
 190         'url': 'http://y.qq.com/#type=toplist&p=top_3',
 191         'info_dict': {
 192             'id': 'top_3',
 193             'title': 'QQ音乐巅峰榜·欧美',
 194             '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'
 195                            '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'
 196                            '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：'
 197                            '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'
 198         },
 199         'playlist_count': 100,
 200     }, {
 201         'url': 'http://y.qq.com/#type=toplist&p=global_106',
 202         'info_dict': {
 203             'id': 'global_106',
 204             'title': '韩国Mnet榜',
 205         },
 206         'playlist_count': 50,
 207     }]
 208
 209     def _real_extract(self, url):
 210         list_id = self._match_id(url)
 211
 212         list_type, num_id = list_id.split("_")
 213
 214         toplist_json = self._download_json(
 215             'http://i.y.qq.com/v8/fcg-bin/fcg_v8_toplist_cp.fcg?type=%s&topid=%s&format=json'
 216             % (list_type, num_id),
 217             list_id, 'Download toplist page')
 218
 219         entries = [
 220             self.url_result(
 221                 'http://y.qq.com/#type=song&mid=' + song['data']['songmid'], 'QQMusic', song['data']['songmid']
 222             ) for song in toplist_json['songlist']
 223         ]
 224
 225         topinfo = toplist_json.get('topinfo', {})
 226         list_name = topinfo.get('ListName')
 227         list_description = topinfo.get('info')
 228         return self.playlist_result(entries, list_id, list_name, list_description)