_ 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 strip_jsonp
  10 from ..compat import compat_urllib_request
  11
  12
  13 class QQMusicIE(InfoExtractor):
  14     _VALID_URL = r'http://y.qq.com/#type=song&mid=(?P<id>[0-9A-Za-z]+)'
  15     _TESTS = [{
  16         'url': 'http://y.qq.com/#type=song&mid=004295Et37taLD',
  17         'md5': 'bed90b6db2a7a7a7e11bc585f471f63a',
  18         'info_dict': {
  19             'id': '004295Et37taLD',
  20             'ext': 'm4a',
  21             'title': '可惜没如果',
  22             'upload_date': '20141227',
  23             'creator': '林俊杰',
  24         }
  25     }]
  26
  27     # Reference: m_r_GetRUin() in top_player.js
  28     # http://imgcache.gtimg.cn/music/portal_v3/y/top_player.js
  29     @staticmethod
  30     def m_r_get_ruin():
  31         curMs = int(time.time() * 1000) % 1000
  32         return int(round(random.random() * 2147483647) * curMs % 1E10)
  33
  34     def _real_extract(self, url):
  35         mid = self._match_id(url)
  36
  37         detail_info_page = self._download_webpage(
  38             'http://s.plcloud.music.qq.com/fcgi-bin/fcg_yqq_song_detail_info.fcg?songmid=%s&play=0' % mid,
  39             mid, note='Download song detail info',
  40             errnote='Unable to get song detail info')
  41
  42         song_name = self._html_search_regex(
  43             r"songname:\s*'([^']+)'", detail_info_page, 'song name')
  44
  45         publish_time = self._html_search_regex(
  46             r'发行时间：(\d{4}-\d{2}-\d{2})', detail_info_page,
  47             'publish time').replace('-', '')
  48
  49         singer = self._html_search_regex(
  50             r"singer:\s*'([^']+)", detail_info_page, 'singer')
  51
  52         guid = self.m_r_get_ruin()
  53
  54         vkey = self._download_json(
  55             'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
  56             mid, note='Retrieve vkey', errnote='Unable to get vkey',
  57             transform_source=strip_jsonp)['key']
  58         song_url = 'http://cc.stream.qqmusic.qq.com/C200%s.m4a?vkey=%s&guid=%s&fromtag=0' % (mid, vkey, guid)
  59
  60         return {
  61             'id': mid,
  62             'url': song_url,
  63             'title': song_name,
  64             'upload_date': publish_time,
  65             'creator': singer,
  66         }
  67
  68
  69 class QQMusicSingerIE(InfoExtractor):
  70     _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
  71     _TEST = {
  72         'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
  73         'info_dict': {
  74             'id': '001BLpXF2DyJe2',
  75             'title': '林俊杰',
  76             'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
  77         },
  78         'playlist_count': 12,
  79     }
  80
  81     def _real_extract(self, url):
  82         mid = self._match_id(url)
  83
  84         singer_page = self._download_webpage(
  85             'http://y.qq.com/y/static/singer/%s/%s/%s.html' % (mid[-2], mid[-1], mid),
  86             'Download singer page')
  87
  88         entries = []
  89
  90         for item in re.findall(r'<span class="data">([^<>]+)</span>', singer_page):
  91             song_mid = item.split('|')[-5]
  92             entries.append(self.url_result(
  93                 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic', song_mid))
  94
  95         singer_name = self._html_search_regex(
  96             r"singername\s*:\s*'([^']+)'", singer_page, 'singer name',
  97             default=None)
  98
  99         singer_id = self._html_search_regex(
 100             r"singerid\s*:\s*'([0-9]+)'", singer_page, 'singer id',
 101             default=None)
 102
 103         singer_desc = None
 104
 105         if singer_id:
 106             req = compat_urllib_request.Request(
 107                 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
 108             req.add_header(
 109                 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
 110             singer_desc_page = self._download_xml(
 111                 req, 'Donwload singer description XML')
 112
 113             singer_desc = singer_desc_page.find('./data/info/desc').text
 114
 115         return self.playlist_result(entries, mid, singer_name, singer_desc)