_ 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     _TESTS = [{
  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         'url': 'http://player.youku.com/player.php/sid/XNDgyMDQ2NTQw/v.swf',
  34         'only_matching': True,
  35     }]
  36
  37     def construct_video_urls(self, data1, data2):
  38         # get sid, token
  39         def yk_t(s1, s2):
  40             ls = list(range(256))
  41             t = 0
  42             for i in range(256):
  43                 t = (t + ls[i] + compat_ord(s1[i % len(s1)])) % 256
  44                 ls[i], ls[t] = ls[t], ls[i]
  45             s = bytearray()
  46             x, y = 0, 0
  47             for i in range(len(s2)):
  48                 y = (y + 1) % 256
  49                 x = (x + ls[y]) % 256
  50                 ls[x], ls[y] = ls[y], ls[x]
  51                 s.append(compat_ord(s2[i]) ^ ls[(ls[x] + ls[y]) % 256])
  52             return bytes(s)
  53
  54         sid, token = yk_t(
  55             b'becaf9be', base64.b64decode(data2['ep'].encode('ascii'))
  56         ).decode('ascii').split('_')
  57
  58         # get oip
  59         oip = data2['ip']
  60
  61         # get fileid
  62         string_ls = list(
  63             'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890')
  64         shuffled_string_ls = []
  65         seed = data1['seed']
  66         N = len(string_ls)
  67         for ii in range(N):
  68             seed = (seed * 0xd3 + 0x754f) % 0x10000
  69             idx = seed * len(string_ls) // 0x10000
  70             shuffled_string_ls.append(string_ls[idx])
  71             del string_ls[idx]
  72
  73         fileid_dict = {}
  74         for format in data1['streamtypes']:
  75             streamfileid = [
  76                 int(i) for i in data1['streamfileids'][format].strip('*').split('*')]
  77             fileid = ''.join(
  78                 [shuffled_string_ls[i] for i in streamfileid])
  79             fileid_dict[format] = fileid[:8] + '%s' + fileid[10:]
  80
  81         def get_fileid(format, n):
  82             fileid = fileid_dict[format] % hex(int(n))[2:].upper().zfill(2)
  83             return fileid
  84
  85         # get ep
  86         def generate_ep(format, n):
  87             fileid = get_fileid(format, n)
  88             ep_t = yk_t(
  89                 b'bf7e5f01',
  90                 ('%s_%s_%s' % (sid, fileid, token)).encode('ascii')
  91             )
  92             ep = base64.b64encode(ep_t).decode('ascii')
  93             return ep
  94
  95         # generate video_urls
  96         video_urls_dict = {}
  97         for format in data1['streamtypes']:
  98             video_urls = []
  99             for dt in data1['segs'][format]:
 100                 n = str(int(dt['no']))
 101                 param = {
 102                     'K': dt['k'],
 103                     'hd': self.get_hd(format),
 104                     'myp': 0,
 105                     'ts': dt['seconds'],
 106                     'ypp': 0,
 107                     'ctype': 12,
 108                     'ev': 1,
 109                     'token': token,
 110                     'oip': oip,
 111                     'ep': generate_ep(format, n)
 112                 }
 113                 video_url = \
 114                     'http://k.youku.com/player/getFlvPath/' + \
 115                     'sid/' + sid + \
 116                     '_' + str(int(n) + 1).zfill(2) + \
 117                     '/st/' + self.parse_ext_l(format) + \
 118                     '/fileid/' + get_fileid(format, n) + '?' + \
 119                     compat_urllib_parse.urlencode(param)
 120                 video_urls.append(video_url)
 121             video_urls_dict[format] = video_urls
 122
 123         return video_urls_dict
 124
 125     def get_hd(self, fm):
 126         hd_id_dict = {
 127             'flv': '0',
 128             'mp4': '1',
 129             'hd2': '2',
 130             'hd3': '3',
 131             '3gp': '0',
 132             '3gphd': '1'
 133         }
 134         return hd_id_dict[fm]
 135
 136     def parse_ext_l(self, fm):
 137         ext_dict = {
 138             'flv': 'flv',
 139             'mp4': 'mp4',
 140             'hd2': 'flv',
 141             'hd3': 'flv',
 142             '3gp': 'flv',
 143             '3gphd': 'mp4'
 144         }
 145         return ext_dict[fm]
 146
 147     def get_format_name(self, fm):
 148         _dict = {
 149             '3gp': 'h6',
 150             '3gphd': 'h5',
 151             'flv': 'h4',
 152             'mp4': 'h3',
 153             'hd2': 'h2',
 154             'hd3': 'h1'
 155         }
 156         return _dict[fm]
 157
 158     def _real_extract(self, url):
 159         video_id = self._match_id(url)
 160
 161         # request basic data
 162         data1_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id
 163         data2_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id
 164
 165         raw_data1 = self._download_json(data1_url, video_id)
 166         raw_data2 = self._download_json(data2_url, video_id)
 167         data1 = raw_data1['data'][0]
 168         data2 = raw_data2['data'][0]
 169
 170         error_code = data1.get('error_code')
 171         if error_code:
 172             # -8 means blocked outside China.
 173             # Chinese and English, separated by newline.
 174             error = data1.get('error')
 175             raise ExtractorError(
 176                 error or 'Server reported error %i' %
 177                 error_code,
 178                 expected=True)
 179
 180         title = data1['title']
 181
 182         # generate video_urls_dict
 183         video_urls_dict = self.construct_video_urls(data1, data2)
 184
 185         # construct info
 186         entries = []
 187         for fm in data1['streamtypes']:
 188             video_urls = video_urls_dict[fm]
 189             for i in range(len(video_urls)):
 190                 if len(entries) < i + 1:
 191                     entries.append({'formats': []})
 192                 entries[i]['formats'].append(
 193                     {
 194                         'url': video_urls[i],
 195                         'format_id': self.get_format_name(fm),
 196                         'ext': self.parse_ext_l(fm),
 197                         'filesize': int(data1['segs'][fm][i]['size'])
 198                     }
 199                 )
 200
 201         for i in range(len(entries)):
 202             entries[i].update(
 203                 {
 204                     'id': '_part%d' % (i + 1),
 205                     'title': title,
 206                 }
 207             )
 208
 209         if len(entries) > 1:
 210             info = {
 211                 '_type': 'multi_video',
 212                 'id': video_id,
 213                 'title': title,
 214                 'entries': entries,
 215             }
 216         else:
 217             info = entries[0]
 218             info['id'] = video_id
 219
 220         return info