_ 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 ..compat import (
   8     compat_urllib_parse,
   9     compat_ord,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13     sanitized_Request,
  14 )
  15
  16
  17 class YoukuIE(InfoExtractor):
  18     IE_NAME = 'youku'
  19     IE_DESC = '优酷'
  20     _VALID_URL = r'''(?x)
  21         (?:
  22             http://(?:v|player)\.youku\.com/(?:v_show/id_|player\.php/sid/)|
  23             youku:)
  24         (?P<id>[A-Za-z0-9]+)(?:\.html|/v\.swf|)
  25     '''
  26
  27     _TESTS = [{
  28         'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html',
  29         'md5': '5f3af4192eabacc4501508d54a8cabd7',
  30         'info_dict': {
  31             'id': 'XMTc1ODE5Njcy_part1',
  32             'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.',
  33             'ext': 'flv'
  34         }
  35     }, {
  36         'url': 'http://player.youku.com/player.php/sid/XNDgyMDQ2NTQw/v.swf',
  37         'only_matching': True,
  38     }, {
  39         'url': 'http://v.youku.com/v_show/id_XODgxNjg1Mzk2_ev_1.html',
  40         'info_dict': {
  41             'id': 'XODgxNjg1Mzk2',
  42             'title': '武媚娘传奇 85',
  43         },
  44         'playlist_count': 11,
  45     }, {
  46         'url': 'http://v.youku.com/v_show/id_XMTI1OTczNDM5Mg==.html',
  47         'info_dict': {
  48             'id': 'XMTI1OTczNDM5Mg',
  49             'title': '花千骨 04',
  50         },
  51         'playlist_count': 13,
  52         'skip': 'Available in China only',
  53     }, {
  54         'url': 'http://v.youku.com/v_show/id_XNjA1NzA2Njgw.html',
  55         'note': 'Video protected with password',
  56         'info_dict': {
  57             'id': 'XNjA1NzA2Njgw',
  58             'title': '邢義田复旦讲座之想象中的胡人—从“左衽孔子”说起',
  59         },
  60         'playlist_count': 19,
  61         'params': {
  62             'videopassword': '100600',
  63         },
  64     }]
  65
  66     def construct_video_urls(self, data1, data2):
  67         # get sid, token
  68         def yk_t(s1, s2):
  69             ls = list(range(256))
  70             t = 0
  71             for i in range(256):
  72                 t = (t + ls[i] + compat_ord(s1[i % len(s1)])) % 256
  73                 ls[i], ls[t] = ls[t], ls[i]
  74             s = bytearray()
  75             x, y = 0, 0
  76             for i in range(len(s2)):
  77                 y = (y + 1) % 256
  78                 x = (x + ls[y]) % 256
  79                 ls[x], ls[y] = ls[y], ls[x]
  80                 s.append(compat_ord(s2[i]) ^ ls[(ls[x] + ls[y]) % 256])
  81             return bytes(s)
  82
  83         sid, token = yk_t(
  84             b'becaf9be', base64.b64decode(data2['security']['encrypt_string'].encode('ascii'))
  85         ).decode('ascii').split('_')
  86
  87         # get oip
  88         oip = data1['security']['ip']
  89
  90         # get fileid
  91         string_ls = list(
  92             'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890')
  93         shuffled_string_ls = []
  94         seed = data1.get('seed')
  95         seed = 0
  96         N = len(string_ls)
  97         for ii in range(N):
  98             seed = (seed * 0xd3 + 0x754f) % 0x10000
  99             idx = seed * len(string_ls) // 0x10000
 100             shuffled_string_ls.append(string_ls[idx])
 101             del string_ls[idx]
 102
 103         fileid_dict = {}
 104         for stream in data1['stream']:
 105             format = stream.get('stream_type')
 106             #streamfileid = [
 107             #    int(i) for i in data1['stream']['streamfileids']#[format].strip('*').split('*')]
 108             #fileid = ''.join(
 109             #    [shuffled_string_ls[i] for i in streamfileid])
 110             fileid = stream['stream_fileid']
 111             fileid_dict[format] = fileid
 112             #fileid_dict[format] = fileid[:8] + '%s' + fileid[10:]
 113
 114         def get_fileid(format, n):
 115             number = hex(int(str(n), 10))[2:].upper()
 116             if len(number) == 1:
 117                 number = '0' + number
 118             streamfileids = fileid_dict[format]
 119             fileid = streamfileids[0:8] + number + streamfileids[10:]
 120             #fileid = fileid_dict[format] % hex(int(n))[2:].upper().zfill(2)
 121             #fileid = fileid_dict[format]
 122             return fileid
 123
 124         # get ep
 125         def generate_ep(format, n):
 126             fileid = get_fileid(format, n)
 127             ep_t = yk_t(
 128                 b'bf7e5f01',
 129                 ('%s_%s_%s' % (sid, fileid, token)).encode('ascii')
 130             )
 131             ep = base64.b64encode(ep_t).decode('ascii')
 132             return ep
 133
 134         # generate video_urls
 135         video_urls_dict = {}
 136         for stream in data1['stream']:
 137             format = stream.get('stream_type')
 138             video_urls = []
 139             for dt in stream['segs']:
 140                 #n = str(int(dt['size']))
 141                 n = str(stream['segs'].index(dt))
 142                 param = {
 143                     'K': dt['key'],
 144                     'hd': self.get_hd(format),
 145                     'myp': 0,
 146                     #'ts': dt['total_milliseconds_video'],
 147                     'ypp': 0,
 148                     'ctype': 12,
 149                     'ev': 1,
 150                     'token': token,
 151                     'oip': oip,
 152                     'ep': generate_ep(format, n)
 153                 }
 154                 video_url = \
 155                     'http://k.youku.com/player/getFlvPath/' + \
 156                     'sid/' + sid + \
 157                     '_00'+ \
 158                     '/st/' + self.parse_ext_l(format) + \
 159                     '/fileid/' + get_fileid(format, n) + '?' + \
 160                     compat_urllib_parse.urlencode(param)
 161                 video_urls.append(video_url)
 162             video_urls_dict[format] = video_urls
 163
 164         return video_urls_dict
 165
 166     def get_hd(self, fm):
 167         hd_id_dict = {
 168             'flv': '0',
 169             'mp4': '1',
 170             'hd2': '2',
 171             'hd3': '3',
 172             '3gp': '0',
 173             '3gphd': '1',
 174             'flvhd': '0'
 175         }
 176         return hd_id_dict[fm]
 177
 178     def parse_ext_l(self, fm):
 179         ext_dict = {
 180             'flv': 'flv',
 181             'mp4': 'mp4',
 182             'mp4hd': 'mp4',
 183             'mp4hd2': 'mp4',
 184             'hd2': 'flv',
 185             'hd3': 'flv',
 186             '3gp': 'flv',
 187             '3gphd': 'mp4',
 188             'flvhd': 'flv'
 189         }
 190         return ext_dict[fm]
 191
 192     def get_format_name(self, fm):
 193         _dict = {
 194             '3gp': 'h6',
 195             '3gphd': 'h5',
 196             'flvhd': 'h4',
 197             'flv': 'h4',
 198             'mp4': 'h3',
 199             'hd2': 'h2',
 200             'hd3': 'h1'
 201         }
 202         return _dict[fm]
 203
 204     def _real_extract(self, url):
 205         video_id = self._match_id(url)
 206
 207         def retrieve_data(req_url, note):
 208
 209
 210             headers = {
 211                     'Referer': req_url,
 212                 }
 213             self._set_cookie('youku.com','xreferrer','http://www.youku.com')
 214             req = sanitized_Request(req_url,headers=headers)
 215
 216             cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
 217             if cn_verification_proxy:
 218                 req.add_header('Ytdl-request-proxy', cn_verification_proxy)
 219
 220             raw_data = self._download_json(req, video_id, note=note)
 221
 222             return raw_data['data']
 223
 224
 225         video_password = self._downloader.params.get('videopassword', None)
 226
 227         # request basic data
 228         #basic_data_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id
 229         basic_data_url = "http://play.youku.com/play/get.json?vid=%s&ct=12" % video_id
 230         if video_password:
 231             basic_data_url += '?password=%s' % video_password
 232
 233         data1 = retrieve_data(
 234             basic_data_url,
 235             'Downloading JSON metadata 1')
 236         data2 = retrieve_data(
 237             #'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id,
 238             "http://play.youku.com/play/get.json?vid=%s&ct=12" % video_id,
 239             'Downloading JSON metadata 2')
 240
 241         error_code = data1.get('error_code')
 242         if error_code:
 243             error = data1.get('error')
 244             if error is not None and '因版权原因无法观看此视频' in error:
 245                 raise ExtractorError(
 246                     'Youku said: Sorry, this video is available in China only', expected=True)
 247             else:
 248                 msg = 'Youku server reported error %i' % error_code
 249                 if error is not None:
 250                     msg += ': ' + error
 251                 raise ExtractorError(msg)
 252
 253         #title = data1['title']
 254         title = data1['video']['title']
 255
 256         # generate video_urls_dict
 257         video_urls_dict = self.construct_video_urls(data1, data2)
 258
 259         # construct info
 260         entries = [{
 261             'id': '%s_part%d' % (video_id, i + 1),
 262             'title': title,
 263             'formats': [],
 264             # some formats are not available for all parts, we have to detect
 265             # which one has all
 266         } for i in range(max(len(v.get('segs')) for v in data1['stream']))]
 267         for stream in data1['stream']:
 268             fm = stream.get('stream_type')
 269             video_urls = video_urls_dict[fm]
 270             for video_url, seg, entry in zip(video_urls, stream['segs'], entries):
 271                 entry['formats'].append({
 272                     'url': video_url,
 273                     'format_id': self.get_format_name(fm),
 274                     'ext': self.parse_ext_l(fm),
 275                     'filesize': int(seg['size']),
 276                 })
 277
 278         return {
 279             '_type': 'multi_video',
 280             'id': video_id,
 281             'title': title,
 282             'entries': entries,
 283         }