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     _VALID_URL = r'http://y.qq.com/#type=song&mid=(?P<id>[0-9A-Za-z]+)'
  18     _TESTS = [{
  19         'url': 'http://y.qq.com/#type=song&mid=004295Et37taLD',
  20         'md5': 'bed90b6db2a7a7a7e11bc585f471f63a',
  21         'info_dict': {
  22             'id': '004295Et37taLD',
  23             'ext': 'm4a',
  24             'title': '可惜没如果',
  25             'upload_date': '20141227',
  26             'creator': '林俊杰',
  27             'description': 'md5:4348ff1dd24036906baa7b6f973f8d30',
  28         }
  29     }]
  30
  31     # Reference: m_r_GetRUin() in top_player.js
  32     # http://imgcache.gtimg.cn/music/portal_v3/y/top_player.js
  33     @staticmethod
  34     def m_r_get_ruin():
  35         curMs = int(time.time() * 1000) % 1000
  36         return int(round(random.random() * 2147483647) * curMs % 1E10)
  37
  38     def _real_extract(self, url):
  39         mid = self._match_id(url)
  40
  41         detail_info_page = self._download_webpage(
  42             'http://s.plcloud.music.qq.com/fcgi-bin/fcg_yqq_song_detail_info.fcg?songmid=%s&play=0' % mid,
  43             mid, note='Download song detail info',
  44             errnote='Unable to get song detail info', encoding='gbk')
  45
  46         song_name = self._html_search_regex(
  47             r"songname:\s*'([^']+)'", detail_info_page, 'song name')
  48
  49         publish_time = self._html_search_regex(
  50             r'发行时间：(\d{4}-\d{2}-\d{2})', detail_info_page,
  51             'publish time', default=None)
  52         if publish_time:
  53             publish_time = publish_time.replace('-', '')
  54
  55         singer = self._html_search_regex(
  56             r"singer:\s*'([^']+)", detail_info_page, 'singer', default=None)
  57
  58         lrc_content = self._html_search_regex(
  59             r'<div class="content" id="lrc_content"[^<>]*>([^<>]+)</div>',
  60             detail_info_page, 'LRC lyrics', default=None)
  61
  62         guid = self.m_r_get_ruin()
  63
  64         vkey = self._download_json(
  65             'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
  66             mid, note='Retrieve vkey', errnote='Unable to get vkey',
  67             transform_source=strip_jsonp)['key']
  68         song_url = 'http://cc.stream.qqmusic.qq.com/C200%s.m4a?vkey=%s&guid=%s&fromtag=0' % (mid, vkey, guid)
  69
  70         return {
  71             'id': mid,
  72             'url': song_url,
  73             'title': song_name,
  74             'upload_date': publish_time,
  75             'creator': singer,
  76             'description': lrc_content,
  77         }
  78
  79
  80 class QQPlaylistBaseIE(InfoExtractor):
  81     @staticmethod
  82     def qq_static_url(category, mid):
  83         return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
  84
  85     @classmethod
  86     def get_entries_from_page(cls, page):
  87         entries = []
  88
  89         for item in re.findall(r'class="data"[^<>]*>([^<>]+)</', page):
  90             song_mid = unescapeHTML(item).split('|')[-5]
  91             entries.append(cls.url_result(
  92                 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
  93                 song_mid))
  94
  95         return entries
  96
  97
  98 class QQMusicSingerIE(QQPlaylistBaseIE):
  99     _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
 100     _TEST = {
 101         'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
 102         'info_dict': {
 103             'id': '001BLpXF2DyJe2',
 104             'title': '林俊杰',
 105             'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
 106         },
 107         'playlist_count': 12,
 108     }
 109
 110     def _real_extract(self, url):
 111         mid = self._match_id(url)
 112
 113         singer_page = self._download_webpage(
 114             self.qq_static_url('singer', mid), mid, 'Download singer page')
 115
 116         entries = self.get_entries_from_page(singer_page)
 117
 118         singer_name = self._html_search_regex(
 119             r"singername\s*:\s*'([^']+)'", singer_page, 'singer name',
 120             default=None)
 121
 122         singer_id = self._html_search_regex(
 123             r"singerid\s*:\s*'([0-9]+)'", singer_page, 'singer id',
 124             default=None)
 125
 126         singer_desc = None
 127
 128         if singer_id:
 129             req = compat_urllib_request.Request(
 130                 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
 131             req.add_header(
 132                 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
 133             singer_desc_page = self._download_xml(
 134                 req, mid, 'Donwload singer description XML')
 135
 136             singer_desc = singer_desc_page.find('./data/info/desc').text
 137
 138         return self.playlist_result(entries, mid, singer_name, singer_desc)
 139
 140
 141 class QQMusicAlbumIE(QQPlaylistBaseIE):
 142     _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
 143
 144     _TEST = {
 145         'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1&play=0',
 146         'info_dict': {
 147             'id': '000gXCTb2AhRR1',
 148             'title': '我们都是这样长大的',
 149             'description': 'md5:d216c55a2d4b3537fe4415b8767d74d6',
 150         },
 151         'playlist_count': 4,
 152     }
 153
 154     def _real_extract(self, url):
 155         mid = self._match_id(url)
 156
 157         album_page = self._download_webpage(
 158             self.qq_static_url('album', mid), mid, 'Download album page')
 159
 160         entries = self.get_entries_from_page(album_page)
 161
 162         album_name = self._html_search_regex(
 163             r"albumname\s*:\s*'([^']+)',", album_page, 'album name',
 164             default=None)
 165
 166         album_detail = self._html_search_regex(
 167             r'<div class="album_detail close_detail">\s*<p>((?:[^<>]+(?:<br />)?)+)</p>',
 168             album_page, 'album details', default=None)
 169
 170         return self.playlist_result(entries, mid, album_name, album_detail)