_ Git - youtube-dl/blob - youtube_dl/extractor/youku.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import base64
   5
   6 from .common import InfoExtractor
   7 from ..utils import ExtractorError
   8
   9 from ..compat import (
  10     compat_urllib_parse,
  11     compat_ord,
  12 )
  13
  14
  15 class YoukuIE(InfoExtractor):
  16     IE_NAME = 'youku'
  17     _VALID_URL = r'''(?x)
  18         (?:
  19             http://(?:v|player)\.youku\.com/(?:v_show/id_|player\.php/sid/)|
  20             youku:)
  21         (?P<id>[A-Za-z0-9]+)(?:\.html|/v\.swf|)
  22     '''
  23
  24     _TEST = {
  25         'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html',
  26         'md5': '5f3af4192eabacc4501508d54a8cabd7',
  27         'info_dict': {
  28             'id': 'XMTc1ODE5Njcy',
  29             'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.',
  30             'ext': 'flv'
  31         }
  32     }
  33
  34     def construct_video_urls(self, data1, data2):
  35         # get sid, token
  36         def yk_t(s1, s2):
  37             ls = list(range(256))
  38             t = 0
  39             for i in range(256):
  40                 t = (t + ls[i] + compat_ord(s1[i % len(s1)])) % 256
  41                 ls[i], ls[t] = ls[t], ls[i]
  42             s = bytearray()
  43             x, y = 0, 0
  44             for i in range(len(s2)):
  45                 y = (y + 1) % 256
  46                 x = (x + ls[y]) % 256
  47                 ls[x], ls[y] = ls[y], ls[x]
  48                 s.append(compat_ord(s2[i]) ^ ls[(ls[x] + ls[y]) % 256])
  49             return bytes(s)
  50
  51         sid, token = yk_t(
  52             b'becaf9be', base64.b64decode(data2['ep'].encode('ascii'))
  53         ).decode('ascii').split('_')
  54
  55         # get oip
  56         oip = data2['ip']
  57
  58         # get fileid
  59         string_ls = list(
  60             'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890')
  61         shuffled_string_ls = []
  62         seed = data1['seed']
  63         N = len(string_ls)
  64         for ii in range(N):
  65             seed = (seed * 0xd3 + 0x754f) % 0x10000
  66             idx = seed * len(string_ls) // 0x10000
  67             shuffled_string_ls.append(string_ls[idx])
  68             del string_ls[idx]
  69
  70         fileid_dict = {}
  71         for format in data1['streamtypes']:
  72             streamfileid = [
  73                 int(i) for i in data1['streamfileids'][format].strip('*').split('*')]
  74             fileid = ''.join(
  75                 [shuffled_string_ls[i] for i in streamfileid])
  76             fileid_dict[format] = fileid[:8] + '%s' + fileid[10:]
  77
  78         def get_fileid(format, n):
  79             fileid = fileid_dict[format] % hex(int(n))[2:].upper().zfill(2)
  80             return fileid
  81
  82         # get ep
  83         def generate_ep(format, n):
  84             fileid = get_fileid(format, n)
  85             ep_t = yk_t(
  86                 b'bf7e5f01',
  87                 ('%s_%s_%s' % (sid, fileid, token)).encode('ascii')
  88             )
  89             ep = base64.b64encode(ep_t).decode('ascii')
  90             return ep
  91
  92         # generate video_urls
  93         video_urls_dict = {}
  94         for format in data1['streamtypes']:
  95             video_urls = []
  96             for dt in data1['segs'][format]:
  97                 n = str(int(dt['no']))
  98                 param = {
  99                     'K': dt['k'],
 100                     'hd': self.get_hd(format),
 101                     'myp': 0,
 102                     'ts': dt['seconds'],
 103                     'ypp': 0,
 104                     'ctype': 12,
 105                     'ev': 1,
 106                     'token': token,
 107                     'oip': oip,
 108                     'ep': generate_ep(format, n)
 109                 }
 110                 video_url = \
 111                     'http://k.youku.com/player/getFlvPath/' + \
 112                     'sid/' + sid + \
 113                     '_' + str(int(n) + 1).zfill(2) + \
 114                     '/st/' + self.parse_ext_l(format) + \
 115                     '/fileid/' + get_fileid(format, n) + '?' + \
 116                     compat_urllib_parse.urlencode(param)
 117                 video_urls.append(video_url)
 118             video_urls_dict[format] = video_urls
 119
 120         return video_urls_dict
 121
 122     def get_hd(self, fm):
 123         hd_id_dict = {
 124             'flv': '0',
 125             'mp4': '1',
 126             'hd2': '2',
 127             'hd3': '3',
 128             '3gp': '0',
 129             '3gphd': '1'
 130         }
 131         return hd_id_dict[fm]
 132
 133     def parse_ext_l(self, fm):
 134         ext_dict = {
 135             'flv': 'flv',
 136             'mp4': 'mp4',
 137             'hd2': 'flv',
 138             'hd3': 'flv',
 139             '3gp': 'flv',
 140             '3gphd': 'mp4'
 141         }
 142         return ext_dict[fm]
 143
 144     def get_format_name(self, fm):
 145         _dict = {
 146             '3gp': 'h6',
 147             '3gphd': 'h5',
 148             'flv': 'h4',
 149             'mp4': 'h3',
 150             'hd2': 'h2',
 151             'hd3': 'h1'
 152         }
 153         return _dict[fm]
 154
 155     def _real_extract(self, url):
 156         video_id = self._match_id(url)
 157
 158         # request basic data
 159         data1_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id
 160         data2_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id
 161
 162         raw_data1 = self._download_json(data1_url, video_id)
 163         raw_data2 = self._download_json(data2_url, video_id)
 164         data1 = raw_data1['data'][0]
 165         data2 = raw_data2['data'][0]
 166
 167         error_code = data1.get('error_code')
 168         if error_code:
 169             # -8 means blocked outside China.
 170             # Chinese and English, separated by newline.
 171             error = data1.get('error')
 172             raise ExtractorError(
 173                 error or 'Server reported error %i' %
 174                 error_code,
 175                 expected=True)
 176
 177         title = data1['title']
 178
 179         # generate video_urls_dict
 180         video_urls_dict = self.construct_video_urls(data1, data2)
 181
 182         # construct info
 183         entries = []
 184         for fm in data1['streamtypes']:
 185             video_urls = video_urls_dict[fm]
 186             for i in range(len(video_urls)):
 187                 if len(entries) < i + 1:
 188                     entries.append({'formats': []})
 189                 entries[i]['formats'].append(
 190                     {
 191                         'url': video_urls[i],
 192                         'format_id': self.get_format_name(fm),
 193                         'ext': self.parse_ext_l(fm),
 194                         'filesize': int(data1['segs'][fm][i]['size'])
 195                     }
 196                 )
 197
 198         for i in range(len(entries)):
 199             entries[i].update(
 200                 {
 201                     'id': '_part%d' % (i + 1),
 202                     'title': title,
 203                 }
 204             )
 205
 206         if len(entries) > 1:
 207             info = {
 208                 '_type': 'multi_video',
 209                 'id': video_id,
 210                 'title': title,
 211                 'entries': entries,
 212             }
 213         else:
 214             info = entries[0]
 215             info['id'] = video_id
 216
 217         return info