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