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         'url': 'http://v.youku.com/v_show/id_XNjA1NzA2Njgw.html',
  54         'note': 'Video protected with password',
  55         'info_dict': {
  56             'id': 'XNjA1NzA2Njgw',
  57             'title': '邢義田复旦讲座之想象中的胡人—从“左衽孔子”说起',
  58         },
  59         'playlist_count': 19,
  60         'params': {
  61             'videopassword': '100600',
  62         },
  63     }]
  64
  65     def construct_video_urls(self, data1, data2):
  66         # get sid, token
  67         def yk_t(s1, s2):
  68             ls = list(range(256))
  69             t = 0
  70             for i in range(256):
  71                 t = (t + ls[i] + compat_ord(s1[i % len(s1)])) % 256
  72                 ls[i], ls[t] = ls[t], ls[i]
  73             s = bytearray()
  74             x, y = 0, 0
  75             for i in range(len(s2)):
  76                 y = (y + 1) % 256
  77                 x = (x + ls[y]) % 256
  78                 ls[x], ls[y] = ls[y], ls[x]
  79                 s.append(compat_ord(s2[i]) ^ ls[(ls[x] + ls[y]) % 256])
  80             return bytes(s)
  81
  82         sid, token = yk_t(
  83             b'becaf9be', base64.b64decode(data2['ep'].encode('ascii'))
  84         ).decode('ascii').split('_')
  85
  86         # get oip
  87         oip = data2['ip']
  88
  89         # get fileid
  90         string_ls = list(
  91             'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890')
  92         shuffled_string_ls = []
  93         seed = data1['seed']
  94         N = len(string_ls)
  95         for ii in range(N):
  96             seed = (seed * 0xd3 + 0x754f) % 0x10000
  97             idx = seed * len(string_ls) // 0x10000
  98             shuffled_string_ls.append(string_ls[idx])
  99             del string_ls[idx]
 100
 101         fileid_dict = {}
 102         for format in data1['streamtypes']:
 103             streamfileid = [
 104                 int(i) for i in data1['streamfileids'][format].strip('*').split('*')]
 105             fileid = ''.join(
 106                 [shuffled_string_ls[i] for i in streamfileid])
 107             fileid_dict[format] = fileid[:8] + '%s' + fileid[10:]
 108
 109         def get_fileid(format, n):
 110             fileid = fileid_dict[format] % hex(int(n))[2:].upper().zfill(2)
 111             return fileid
 112
 113         # get ep
 114         def generate_ep(format, n):
 115             fileid = get_fileid(format, n)
 116             ep_t = yk_t(
 117                 b'bf7e5f01',
 118                 ('%s_%s_%s' % (sid, fileid, token)).encode('ascii')
 119             )
 120             ep = base64.b64encode(ep_t).decode('ascii')
 121             return ep
 122
 123         # generate video_urls
 124         video_urls_dict = {}
 125         for format in data1['streamtypes']:
 126             video_urls = []
 127             for dt in data1['segs'][format]:
 128                 n = str(int(dt['no']))
 129                 param = {
 130                     'K': dt['k'],
 131                     'hd': self.get_hd(format),
 132                     'myp': 0,
 133                     'ts': dt['seconds'],
 134                     'ypp': 0,
 135                     'ctype': 12,
 136                     'ev': 1,
 137                     'token': token,
 138                     'oip': oip,
 139                     'ep': generate_ep(format, n)
 140                 }
 141                 video_url = \
 142                     'http://k.youku.com/player/getFlvPath/' + \
 143                     'sid/' + sid + \
 144                     '_' + str(int(n) + 1).zfill(2) + \
 145                     '/st/' + self.parse_ext_l(format) + \
 146                     '/fileid/' + get_fileid(format, n) + '?' + \
 147                     compat_urllib_parse.urlencode(param)
 148                 video_urls.append(video_url)
 149             video_urls_dict[format] = video_urls
 150
 151         return video_urls_dict
 152
 153     def get_hd(self, fm):
 154         hd_id_dict = {
 155             'flv': '0',
 156             'mp4': '1',
 157             'hd2': '2',
 158             'hd3': '3',
 159             '3gp': '0',
 160             '3gphd': '1'
 161         }
 162         return hd_id_dict[fm]
 163
 164     def parse_ext_l(self, fm):
 165         ext_dict = {
 166             'flv': 'flv',
 167             'mp4': 'mp4',
 168             'hd2': 'flv',
 169             'hd3': 'flv',
 170             '3gp': 'flv',
 171             '3gphd': 'mp4'
 172         }
 173         return ext_dict[fm]
 174
 175     def get_format_name(self, fm):
 176         _dict = {
 177             '3gp': 'h6',
 178             '3gphd': 'h5',
 179             'flv': 'h4',
 180             'mp4': 'h3',
 181             'hd2': 'h2',
 182             'hd3': 'h1'
 183         }
 184         return _dict[fm]
 185
 186     def _real_extract(self, url):
 187         video_id = self._match_id(url)
 188
 189         def retrieve_data(req_url, note):
 190             req = compat_urllib_request.Request(req_url)
 191
 192             cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
 193             if cn_verification_proxy:
 194                 req.add_header('Ytdl-request-proxy', cn_verification_proxy)
 195
 196             raw_data = self._download_json(req, video_id, note=note)
 197             return raw_data['data'][0]
 198
 199         video_password = self._downloader.params.get('videopassword', None)
 200
 201         # request basic data
 202         basic_data_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id
 203         if video_password:
 204             basic_data_url += '?password=%s' % video_password
 205
 206         data1 = retrieve_data(
 207             basic_data_url,
 208             'Downloading JSON metadata 1')
 209         data2 = retrieve_data(
 210             'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id,
 211             'Downloading JSON metadata 2')
 212
 213         error_code = data1.get('error_code')
 214         if error_code:
 215             error = data1.get('error')
 216             if error is not None and '因版权原因无法观看此视频' in error:
 217                 raise ExtractorError(
 218                     'Youku said: Sorry, this video is available in China only', expected=True)
 219             else:
 220                 msg = 'Youku server reported error %i' % error_code
 221                 if error is not None:
 222                     msg += ': ' + error
 223                 raise ExtractorError(msg)
 224
 225         title = data1['title']
 226
 227         # generate video_urls_dict
 228         video_urls_dict = self.construct_video_urls(data1, data2)
 229
 230         # construct info
 231         entries = [{
 232             'id': '%s_part%d' % (video_id, i + 1),
 233             'title': title,
 234             'formats': [],
 235             # some formats are not available for all parts, we have to detect
 236             # which one has all
 237         } for i in range(max(len(v) for v in data1['segs'].values()))]
 238         for fm in data1['streamtypes']:
 239             video_urls = video_urls_dict[fm]
 240             for video_url, seg, entry in zip(video_urls, data1['segs'][fm], entries):
 241                 entry['formats'].append({
 242                     'url': video_url,
 243                     'format_id': self.get_format_name(fm),
 244                     'ext': self.parse_ext_l(fm),
 245                     'filesize': int(seg['size']),
 246                 })
 247
 248         return {
 249             '_type': 'multi_video',
 250             'id': video_id,
 251             'title': title,
 252             'entries': entries,
 253         }