_ 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     HEADRequest,
  13     ExtractorError,
  14 )
  15 from ..compat import (
  16     compat_urllib_request,
  17     compat_HTTPError,
  18 )
  19
  20
  21 class QQMusicIE(InfoExtractor):
  22     IE_NAME = 'qqmusic'
  23     _VALID_URL = r'http://y.qq.com/#type=song&mid=(?P<id>[0-9A-Za-z]+)'
  24     _TESTS = [{
  25         'url': 'http://y.qq.com/#type=song&mid=004295Et37taLD',
  26         'md5': '9ce1c1c8445f561506d2e3cfb0255705',
  27         'info_dict': {
  28             'id': '004295Et37taLD',
  29             'ext': 'mp3',
  30             'title': '可惜没如果',
  31             'upload_date': '20141227',
  32             'creator': '林俊杰',
  33             'description': 'md5:d327722d0361576fde558f1ac68a7065',
  34             'thumbnail': 'http://i.gtimg.cn/music/photo/mid_album_500/7/p/001IV22P1RDX7p.jpg',
  35         }
  36     }, {
  37         'note': 'There is no mp3-320 version of this song.',
  38         'url': 'http://y.qq.com/#type=song&mid=004MsGEo3DdNxV',
  39         'md5': 'fa3926f0c585cda0af8fa4f796482e3e',
  40         'info_dict': {
  41             'id': '004MsGEo3DdNxV',
  42             'ext': 'mp3',
  43             'title': '如果',
  44             'upload_date': '20050626',
  45             'creator': '李季美',
  46             'description': 'md5:46857d5ed62bc4ba84607a805dccf437',
  47             'thumbnail': 'http://i.gtimg.cn/music/photo/mid_album_500/r/Q/0042owYj46IxrQ.jpg',
  48         }
  49     }]
  50
  51     _FORMATS = {
  52         'mp3-320': {'prefix': 'M800', 'ext': 'mp3', 'preference': 40, 'abr': 320},
  53         'mp3-128': {'prefix': 'M500', 'ext': 'mp3', 'preference': 30, 'abr': 128},
  54         'm4a': {'prefix': 'C200', 'ext': 'm4a', 'preference': 10}
  55     }
  56
  57     # Reference: m_r_GetRUin() in top_player.js
  58     # http://imgcache.gtimg.cn/music/portal_v3/y/top_player.js
  59     @staticmethod
  60     def m_r_get_ruin():
  61         curMs = int(time.time() * 1000) % 1000
  62         return int(round(random.random() * 2147483647) * curMs % 1E10)
  63
  64     def _real_extract(self, url):
  65         mid = self._match_id(url)
  66
  67         detail_info_page = self._download_webpage(
  68             'http://s.plcloud.music.qq.com/fcgi-bin/fcg_yqq_song_detail_info.fcg?songmid=%s&play=0' % mid,
  69             mid, note='Download song detail info',
  70             errnote='Unable to get song detail info', encoding='gbk')
  71
  72         song_name = self._html_search_regex(
  73             r"songname:\s*'([^']+)'", detail_info_page, 'song name')
  74
  75         publish_time = self._html_search_regex(
  76             r'发行时间：(\d{4}-\d{2}-\d{2})', detail_info_page,
  77             'publish time', default=None)
  78         if publish_time:
  79             publish_time = publish_time.replace('-', '')
  80
  81         singer = self._html_search_regex(
  82             r"singer:\s*'([^']+)", detail_info_page, 'singer', default=None)
  83
  84         lrc_content = self._html_search_regex(
  85             r'<div class="content" id="lrc_content"[^<>]*>([^<>]+)</div>',
  86             detail_info_page, 'LRC lyrics', default=None)
  87         if lrc_content:
  88             lrc_content = lrc_content.replace('\\n', '\n')
  89
  90         thumbnail_url = None
  91         albummid = self._search_regex(
  92             [r'albummid:\'([0-9a-zA-Z]+)\'', r'"albummid":"([0-9a-zA-Z]+)"'],
  93             detail_info_page, 'album mid', default=None)
  94         if albummid:
  95             thumbnail_url = "http://i.gtimg.cn/music/photo/mid_album_500/%s/%s/%s.jpg" \
  96                             % (albummid[-2:-1], albummid[-1], albummid)
  97
  98         guid = self.m_r_get_ruin()
  99
 100         vkey = self._download_json(
 101             'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
 102             mid, note='Retrieve vkey', errnote='Unable to get vkey',
 103             transform_source=strip_jsonp)['key']
 104
 105         formats = []
 106         for format_id, details in self._FORMATS.items():
 107             video_url = 'http://cc.stream.qqmusic.qq.com/%s%s.%s?vkey=%s&guid=%s&fromtag=0' \
 108                         % (details['prefix'], mid, details['ext'], vkey, guid)
 109             req = HEADRequest(video_url)
 110             try:
 111                 res = self._request_webpage(
 112                     req, mid, note='Testing %s video URL' % format_id, fatal=False)
 113             except ExtractorError as e:
 114                 if isinstance(e.cause, compat_HTTPError) and e.cause.code in [400, 404]:
 115                     self.report_warning('Invalid %s video URL' % format_id, mid)
 116             else:
 117                 if res:
 118                     formats.append({
 119                         'url': video_url,
 120                         'format': format_id,
 121                         'format_id': format_id,
 122                         'preference': details['preference'],
 123                         'abr': details.get('abr'),
 124                     })
 125         self._sort_formats(formats)
 126
 127         return {
 128             'id': mid,
 129             'formats': formats,
 130             'title': song_name,
 131             'upload_date': publish_time,
 132             'creator': singer,
 133             'description': lrc_content,
 134             'thumbnail': thumbnail_url,
 135         }
 136
 137
 138 class QQPlaylistBaseIE(InfoExtractor):
 139     @staticmethod
 140     def qq_static_url(category, mid):
 141         return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
 142
 143     @classmethod
 144     def get_entries_from_page(cls, page):
 145         entries = []
 146
 147         for item in re.findall(r'class="data"[^<>]*>([^<>]+)</', page):
 148             song_mid = unescapeHTML(item).split('|')[-5]
 149             entries.append(cls.url_result(
 150                 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
 151                 song_mid))
 152
 153         return entries
 154
 155
 156 class QQMusicSingerIE(QQPlaylistBaseIE):
 157     IE_NAME = 'qqmusic:singer'
 158     _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
 159     _TEST = {
 160         'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
 161         'info_dict': {
 162             'id': '001BLpXF2DyJe2',
 163             'title': '林俊杰',
 164             'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
 165         },
 166         'playlist_count': 12,
 167     }
 168
 169     def _real_extract(self, url):
 170         mid = self._match_id(url)
 171
 172         singer_page = self._download_webpage(
 173             self.qq_static_url('singer', mid), mid, 'Download singer page')
 174
 175         entries = self.get_entries_from_page(singer_page)
 176
 177         singer_name = self._html_search_regex(
 178             r"singername\s*:\s*'([^']+)'", singer_page, 'singer name',
 179             default=None)
 180
 181         singer_id = self._html_search_regex(
 182             r"singerid\s*:\s*'([0-9]+)'", singer_page, 'singer id',
 183             default=None)
 184
 185         singer_desc = None
 186
 187         if singer_id:
 188             req = compat_urllib_request.Request(
 189                 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
 190             req.add_header(
 191                 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
 192             singer_desc_page = self._download_xml(
 193                 req, mid, 'Donwload singer description XML')
 194
 195             singer_desc = singer_desc_page.find('./data/info/desc').text
 196
 197         return self.playlist_result(entries, mid, singer_name, singer_desc)
 198
 199
 200 class QQMusicAlbumIE(QQPlaylistBaseIE):
 201     IE_NAME = 'qqmusic:album'
 202     _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
 203
 204     _TEST = {
 205         'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1&play=0',
 206         'info_dict': {
 207             'id': '000gXCTb2AhRR1',
 208             'title': '我们都是这样长大的',
 209             'description': 'md5:d216c55a2d4b3537fe4415b8767d74d6',
 210         },
 211         'playlist_count': 4,
 212     }
 213
 214     def _real_extract(self, url):
 215         mid = self._match_id(url)
 216
 217         album_page = self._download_webpage(
 218             self.qq_static_url('album', mid), mid, 'Download album page')
 219
 220         entries = self.get_entries_from_page(album_page)
 221
 222         album_name = self._html_search_regex(
 223             r"albumname\s*:\s*'([^']+)',", album_page, 'album name',
 224             default=None)
 225
 226         album_detail = self._html_search_regex(
 227             r'<div class="album_detail close_detail">\s*<p>((?:[^<>]+(?:<br />)?)+)</p>',
 228             album_page, 'album details', default=None)
 229
 230         return self.playlist_result(entries, mid, album_name, album_detail)
 231
 232
 233 class QQMusicToplistIE(QQPlaylistBaseIE):
 234     IE_NAME = 'qqmusic:toplist'
 235     _VALID_URL = r'http://y\.qq\.com/#type=toplist&p=(?P<id>(top|global)_[0-9]+)'
 236
 237     _TESTS = [{
 238         'url': 'http://y.qq.com/#type=toplist&p=global_123',
 239         'info_dict': {
 240             'id': 'global_123',
 241             'title': '美国iTunes榜',
 242         },
 243         'playlist_count': 10,
 244     }, {
 245         'url': 'http://y.qq.com/#type=toplist&p=top_3',
 246         'info_dict': {
 247             'id': 'top_3',
 248             'title': 'QQ音乐巅峰榜·欧美',
 249             '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'
 250                            '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'
 251                            '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：'
 252                            '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'
 253         },
 254         'playlist_count': 100,
 255     }, {
 256         'url': 'http://y.qq.com/#type=toplist&p=global_106',
 257         'info_dict': {
 258             'id': 'global_106',
 259             'title': '韩国Mnet榜',
 260         },
 261         'playlist_count': 50,
 262     }]
 263
 264     def _real_extract(self, url):
 265         list_id = self._match_id(url)
 266
 267         list_type, num_id = list_id.split("_")
 268
 269         toplist_json = self._download_json(
 270             'http://i.y.qq.com/v8/fcg-bin/fcg_v8_toplist_cp.fcg?type=%s&topid=%s&format=json'
 271             % (list_type, num_id),
 272             list_id, 'Download toplist page')
 273
 274         entries = [
 275             self.url_result(
 276                 'http://y.qq.com/#type=song&mid=' + song['data']['songmid'], 'QQMusic', song['data']['songmid']
 277             ) for song in toplist_json['songlist']
 278         ]
 279
 280         topinfo = toplist_json.get('topinfo', {})
 281         list_name = topinfo.get('ListName')
 282         list_description = topinfo.get('info')
 283         return self.playlist_result(entries, list_id, list_name, list_description)