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