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     clean_html,
  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 format_id, details 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                        % (details['prefix'], mid, details['ext'], vkey, guid),
  84                 'format': format_id,
  85                 'format_id': format_id,
  86                 'preference': details['preference'],
  87                 'abr': details.get('abr'),
  88             })
  89         self._sort_formats(formats)
  90
  91         return {
  92             'id': mid,
  93             'formats': formats,
  94             'title': song_name,
  95             'upload_date': publish_time,
  96             'creator': singer,
  97             'description': lrc_content,
  98         }
  99
 100
 101 class QQPlaylistBaseIE(InfoExtractor):
 102     @staticmethod
 103     def qq_static_url(category, mid):
 104         return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
 105
 106     @classmethod
 107     def get_entries_from_page(cls, page):
 108         entries = []
 109
 110         for item in re.findall(r'class="data"[^<>]*>([^<>]+)</', page):
 111             song_mid = unescapeHTML(item).split('|')[-5]
 112             entries.append(cls.url_result(
 113                 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
 114                 song_mid))
 115
 116         return entries
 117
 118
 119 class QQMusicSingerIE(QQPlaylistBaseIE):
 120     IE_NAME = 'qqmusic:singer'
 121     _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
 122     _TEST = {
 123         'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
 124         'info_dict': {
 125             'id': '001BLpXF2DyJe2',
 126             'title': '林俊杰',
 127             'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
 128         },
 129         'playlist_count': 12,
 130     }
 131
 132     def _real_extract(self, url):
 133         mid = self._match_id(url)
 134
 135         singer_page = self._download_webpage(
 136             self.qq_static_url('singer', mid), mid, 'Download singer page')
 137
 138         entries = self.get_entries_from_page(singer_page)
 139
 140         singer_name = self._html_search_regex(
 141             r"singername\s*:\s*'([^']+)'", singer_page, 'singer name',
 142             default=None)
 143
 144         singer_id = self._html_search_regex(
 145             r"singerid\s*:\s*'([0-9]+)'", singer_page, 'singer id',
 146             default=None)
 147
 148         singer_desc = None
 149
 150         if singer_id:
 151             req = compat_urllib_request.Request(
 152                 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
 153             req.add_header(
 154                 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
 155             singer_desc_page = self._download_xml(
 156                 req, mid, 'Donwload singer description XML')
 157
 158             singer_desc = singer_desc_page.find('./data/info/desc').text
 159
 160         return self.playlist_result(entries, mid, singer_name, singer_desc)
 161
 162
 163 class QQMusicAlbumIE(QQPlaylistBaseIE):
 164     IE_NAME = 'qqmusic:album'
 165     _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
 166
 167     _TEST = {
 168         'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1&play=0',
 169         'info_dict': {
 170             'id': '000gXCTb2AhRR1',
 171             'title': '我们都是这样长大的',
 172             'description': 'md5:d216c55a2d4b3537fe4415b8767d74d6',
 173         },
 174         'playlist_count': 4,
 175     }
 176
 177     def _real_extract(self, url):
 178         mid = self._match_id(url)
 179
 180         album_page = self._download_webpage(
 181             self.qq_static_url('album', mid), mid, 'Download album page')
 182
 183         entries = self.get_entries_from_page(album_page)
 184
 185         album_name = self._html_search_regex(
 186             r"albumname\s*:\s*'([^']+)',", album_page, 'album name',
 187             default=None)
 188
 189         album_detail = self._html_search_regex(
 190             r'<div class="album_detail close_detail">\s*<p>((?:[^<>]+(?:<br />)?)+)</p>',
 191             album_page, 'album details', default=None)
 192
 193         return self.playlist_result(entries, mid, album_name, album_detail)
 194
 195
 196 class QQMusicToplistIE(QQPlaylistBaseIE):
 197     IE_NAME = 'qqmusic:toplist'
 198     _VALID_URL = r'http://y\.qq\.com/#type=toplist&p=(?P<id>(top|global)_[0-9]+)'
 199
 200     _TESTS = [{
 201         'url': 'http://y.qq.com/#type=toplist&p=global_123',
 202         'info_dict': {
 203             'id': 'global_123',
 204             'title': '美国iTunes榜',
 205         },
 206         'playlist_count': 10,
 207     }, {
 208         'url': 'http://y.qq.com/#type=toplist&p=top_3',
 209         'info_dict': {
 210             'id': 'top_3',
 211             'title': 'QQ音乐巅峰榜·欧美',
 212             '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'
 213                            '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'
 214                            '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：'
 215                            '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'
 216         },
 217         'playlist_count': 100,
 218     }, {
 219         'url': 'http://y.qq.com/#type=toplist&p=global_106',
 220         'info_dict': {
 221             'id': 'global_106',
 222             'title': '韩国Mnet榜',
 223         },
 224         'playlist_count': 50,
 225     }]
 226
 227     def _real_extract(self, url):
 228         list_id = self._match_id(url)
 229
 230         list_type, num_id = list_id.split("_")
 231
 232         toplist_json = self._download_json(
 233             'http://i.y.qq.com/v8/fcg-bin/fcg_v8_toplist_cp.fcg?type=%s&topid=%s&format=json'
 234             % (list_type, num_id),
 235             list_id, 'Download toplist page')
 236
 237         entries = [
 238             self.url_result(
 239                 'http://y.qq.com/#type=song&mid=' + song['data']['songmid'], 'QQMusic', song['data']['songmid']
 240             ) for song in toplist_json['songlist']
 241         ]
 242
 243         topinfo = toplist_json.get('topinfo', {})
 244         list_name = topinfo.get('ListName')
 245         list_description = topinfo.get('info')
 246         return self.playlist_result(entries, list_id, list_name, list_description)
 247
 248
 249 class QQMusicPlaylistIE(QQPlaylistBaseIE):
 250     IE_NAME = 'qqmusic:playlist'
 251     _VALID_URL = r'http://y\.qq\.com/#type=taoge&id=(?P<id>[0-9]+)'
 252
 253     _TEST = {
 254         'url': 'http://y.qq.com/#type=taoge&id=3462654915',
 255         'info_dict': {
 256             'id': '3462654915',
 257             'title': '韩国5月新歌精选下旬',
 258             'description': 'md5:d2c9d758a96b9888cf4fe82f603121d4',
 259         },
 260         'playlist_count': 40,
 261     }
 262
 263     def _real_extract(self, url):
 264         list_id = self._match_id(url)
 265
 266         list_json = self._download_json(
 267             'http://i.y.qq.com/qzone-music/fcg-bin/fcg_ucc_getcdinfo_byids_cp.fcg?type=1&json=1&utf8=1&onlysong=0&disstid=%s'
 268             % list_id, list_id, 'Download list page',
 269             transform_source=strip_jsonp)['cdlist'][0]
 270
 271         entries = [
 272             self.url_result(
 273                 'http://y.qq.com/#type=song&mid=' + song['songmid'], 'QQMusic', song['songmid']
 274             ) for song in list_json['songlist']
 275         ]
 276
 277         list_name = list_json.get('dissname')
 278         list_description = clean_html(unescapeHTML(list_json.get('desc')))
 279         return self.playlist_result(entries, list_id, list_name, list_description)