Raphaël G. Git Repositories - youtubedl/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     compat_urllib_request,
  13 )
  14
  15
  16 class YoukuIE(InfoExtractor):
  17     IE_NAME = 'youku'
  18     IE_DESC = '优酷'
  19     _VALID_URL = r'''(?x)
  20         (?:
  21             http://(?:v|player)\.youku\.com/(?:v_show/id_|player\.php/sid/)|
  22             youku:)
  23         (?P<id>[A-Za-z0-9]+)(?:\.html|/v\.swf|)
  24     '''
  25
  26     _TESTS = [{
  27         'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html',
  28         'md5': '5f3af4192eabacc4501508d54a8cabd7',
  29         'info_dict': {
  30             'id': 'XMTc1ODE5Njcy_part1',
  31             'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.',
  32             'ext': 'flv'
  33         }
  34     }, {
  35         'url': 'http://player.youku.com/player.php/sid/XNDgyMDQ2NTQw/v.swf',
  36         'only_matching': True,
  37     }, {
  38         'url': 'http://v.youku.com/v_show/id_XODgxNjg1Mzk2_ev_1.html',
  39         'info_dict': {
  40             'id': 'XODgxNjg1Mzk2',
  41             'title': '武媚娘传奇 85',
  42         },
  43         'playlist_count': 11,
  44     }, {
  45         'url': 'http://v.youku.com/v_show/id_XMTI1OTczNDM5Mg==.html',
  46         'info_dict': {
  47             'id': 'XMTI1OTczNDM5Mg',
  48             'title': '花千骨 04',
  49         },
  50         'playlist_count': 13,
  51         'skip': 'Available in China only',
  52     }]
  53
  54     def construct_video_urls(self, data1, data2):
  55         # get sid, token
  56         def yk_t(s1, s2):
  57             ls = list(range(256))
  58             t = 0
  59             for i in range(256):
  60                 t = (t + ls[i] + compat_ord(s1[i % len(s1)])) % 256
  61                 ls[i], ls[t] = ls[t], ls[i]
  62             s = bytearray()
  63             x, y = 0, 0
  64             for i in range(len(s2)):
  65                 y = (y + 1) % 256
  66                 x = (x + ls[y]) % 256
  67                 ls[x], ls[y] = ls[y], ls[x]
  68                 s.append(compat_ord(s2[i]) ^ ls[(ls[x] + ls[y]) % 256])
  69             return bytes(s)
  70
  71         sid, token = yk_t(
  72             b'becaf9be', base64.b64decode(data2['ep'].encode('ascii'))
  73         ).decode('ascii').split('_')
  74
  75         # get oip
  76         oip = data2['ip']
  77
  78         # get fileid
  79         string_ls = list(
  80             'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890')
  81         shuffled_string_ls = []
  82         seed = data1['seed']
  83         N = len(string_ls)
  84         for ii in range(N):
  85             seed = (seed * 0xd3 + 0x754f) % 0x10000
  86             idx = seed * len(string_ls) // 0x10000
  87             shuffled_string_ls.append(string_ls[idx])
  88             del string_ls[idx]
  89
  90         fileid_dict = {}
  91         for format in data1['streamtypes']:
  92             streamfileid = [
  93                 int(i) for i in data1['streamfileids'][format].strip('*').split('*')]
  94             fileid = ''.join(
  95                 [shuffled_string_ls[i] for i in streamfileid])
  96             fileid_dict[format] = fileid[:8] + '%s' + fileid[10:]
  97
  98         def get_fileid(format, n):
  99             fileid = fileid_dict[format] % hex(int(n))[2:].upper().zfill(2)
 100             return fileid
 101
 102         # get ep
 103         def generate_ep(format, n):
 104             fileid = get_fileid(format, n)
 105             ep_t = yk_t(
 106                 b'bf7e5f01',
 107                 ('%s_%s_%s' % (sid, fileid, token)).encode('ascii')
 108             )
 109             ep = base64.b64encode(ep_t).decode('ascii')
 110             return ep
 111
 112         # generate video_urls
 113         video_urls_dict = {}
 114         for format in data1['streamtypes']:
 115             video_urls = []
 116             for dt in data1['segs'][format]:
 117                 n = str(int(dt['no']))
 118                 param = {
 119                     'K': dt['k'],
 120                     'hd': self.get_hd(format),
 121                     'myp': 0,
 122                     'ts': dt['seconds'],
 123                     'ypp': 0,
 124                     'ctype': 12,
 125                     'ev': 1,
 126                     'token': token,
 127                     'oip': oip,
 128                     'ep': generate_ep(format, n)
 129                 }
 130                 video_url = \
 131                     'http://k.youku.com/player/getFlvPath/' + \
 132                     'sid/' + sid + \
 133                     '_' + str(int(n) + 1).zfill(2) + \
 134                     '/st/' + self.parse_ext_l(format) + \
 135                     '/fileid/' + get_fileid(format, n) + '?' + \
 136                     compat_urllib_parse.urlencode(param)
 137                 video_urls.append(video_url)
 138             video_urls_dict[format] = video_urls
 139
 140         return video_urls_dict
 141
 142     def get_hd(self, fm):
 143         hd_id_dict = {
 144             'flv': '0',
 145             'mp4': '1',
 146             'hd2': '2',
 147             'hd3': '3',
 148             '3gp': '0',
 149             '3gphd': '1'
 150         }
 151         return hd_id_dict[fm]
 152
 153     def parse_ext_l(self, fm):
 154         ext_dict = {
 155             'flv': 'flv',
 156             'mp4': 'mp4',
 157             'hd2': 'flv',
 158             'hd3': 'flv',
 159             '3gp': 'flv',
 160             '3gphd': 'mp4'
 161         }
 162         return ext_dict[fm]
 163
 164     def get_format_name(self, fm):
 165         _dict = {
 166             '3gp': 'h6',
 167             '3gphd': 'h5',
 168             'flv': 'h4',
 169             'mp4': 'h3',
 170             'hd2': 'h2',
 171             'hd3': 'h1'
 172         }
 173         return _dict[fm]
 174
 175     def _real_extract(self, url):
 176         video_id = self._match_id(url)
 177
 178         def retrieve_data(req_url, note):
 179             req = compat_urllib_request.Request(req_url)
 180
 181             cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
 182             if cn_verification_proxy:
 183                 req.add_header('Ytdl-request-proxy', cn_verification_proxy)
 184
 185             raw_data = self._download_json(req, video_id, note=note)
 186             return raw_data['data'][0]
 187
 188         # request basic data
 189         data1 = retrieve_data(
 190             'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id,
 191             'Downloading JSON metadata 1')
 192         data2 = retrieve_data(
 193             'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id,
 194             'Downloading JSON metadata 2')
 195
 196         error_code = data1.get('error_code')
 197         if error_code:
 198             error = data1.get('error')
 199             if error is not None and '因版权原因无法观看此视频' in error:
 200                 raise ExtractorError(
 201                     'Youku said: Sorry, this video is available in China only', expected=True)
 202             else:
 203                 msg = 'Youku server reported error %i' % error_code
 204                 if error is not None:
 205                     msg += ': ' + error
 206                 raise ExtractorError(msg)
 207
 208         title = data1['title']
 209
 210         # generate video_urls_dict
 211         video_urls_dict = self.construct_video_urls(data1, data2)
 212
 213         # construct info
 214         entries = [{
 215             'id': '%s_part%d' % (video_id, i + 1),
 216             'title': title,
 217             'formats': [],
 218             # some formats are not available for all parts, we have to detect
 219             # which one has all
 220         } for i in range(max(len(v) for v in data1['segs'].values()))]
 221         for fm in data1['streamtypes']:
 222             video_urls = video_urls_dict[fm]
 223             for video_url, seg, entry in zip(video_urls, data1['segs'][fm], entries):
 224                 entry['formats'].append({
 225                     'url': video_url,
 226                     'format_id': self.get_format_name(fm),
 227                     'ext': self.parse_ext_l(fm),
 228                     'filesize': int(seg['size']),
 229                 })
 230
 231         return {
 232             '_type': 'multi_video',
 233             'id': video_id,
 234             'title': title,
 235             'entries': entries,
 236         }