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 ..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['ep'].encode('ascii'))
  85         ).decode('ascii').split('_')
  86
  87         # get oip
  88         oip = data2['ip']
  89
  90         # get fileid
  91         string_ls = list(
  92             'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890')
  93         shuffled_string_ls = []
  94         seed = data1['seed']
  95         N = len(string_ls)
  96         for ii in range(N):
  97             seed = (seed * 0xd3 + 0x754f) % 0x10000
  98             idx = seed * len(string_ls) // 0x10000
  99             shuffled_string_ls.append(string_ls[idx])
 100             del string_ls[idx]
 101
 102         fileid_dict = {}
 103         for format in data1['streamtypes']:
 104             streamfileid = [
 105                 int(i) for i in data1['streamfileids'][format].strip('*').split('*')]
 106             fileid = ''.join(
 107                 [shuffled_string_ls[i] for i in streamfileid])
 108             fileid_dict[format] = fileid[:8] + '%s' + fileid[10:]
 109
 110         def get_fileid(format, n):
 111             fileid = fileid_dict[format] % hex(int(n))[2:].upper().zfill(2)
 112             return fileid
 113
 114         # get ep
 115         def generate_ep(format, n):
 116             fileid = get_fileid(format, n)
 117             ep_t = yk_t(
 118                 b'bf7e5f01',
 119                 ('%s_%s_%s' % (sid, fileid, token)).encode('ascii')
 120             )
 121             ep = base64.b64encode(ep_t).decode('ascii')
 122             return ep
 123
 124         # generate video_urls
 125         video_urls_dict = {}
 126         for format in data1['streamtypes']:
 127             video_urls = []
 128             for dt in data1['segs'][format]:
 129                 n = str(int(dt['no']))
 130                 param = {
 131                     'K': dt['k'],
 132                     'hd': self.get_hd(format),
 133                     'myp': 0,
 134                     'ts': dt['seconds'],
 135                     'ypp': 0,
 136                     'ctype': 12,
 137                     'ev': 1,
 138                     'token': token,
 139                     'oip': oip,
 140                     'ep': generate_ep(format, n)
 141                 }
 142                 video_url = \
 143                     'http://k.youku.com/player/getFlvPath/' + \
 144                     'sid/' + sid + \
 145                     '_' + str(int(n) + 1).zfill(2) + \
 146                     '/st/' + self.parse_ext_l(format) + \
 147                     '/fileid/' + get_fileid(format, n) + '?' + \
 148                     compat_urllib_parse.urlencode(param)
 149                 video_urls.append(video_url)
 150             video_urls_dict[format] = video_urls
 151
 152         return video_urls_dict
 153
 154     def get_hd(self, fm):
 155         hd_id_dict = {
 156             'flv': '0',
 157             'mp4': '1',
 158             'hd2': '2',
 159             'hd3': '3',
 160             '3gp': '0',
 161             '3gphd': '1'
 162         }
 163         return hd_id_dict[fm]
 164
 165     def parse_ext_l(self, fm):
 166         ext_dict = {
 167             'flv': 'flv',
 168             'mp4': 'mp4',
 169             'hd2': 'flv',
 170             'hd3': 'flv',
 171             '3gp': 'flv',
 172             '3gphd': 'mp4'
 173         }
 174         return ext_dict[fm]
 175
 176     def get_format_name(self, fm):
 177         _dict = {
 178             '3gp': 'h6',
 179             '3gphd': 'h5',
 180             'flv': 'h4',
 181             'mp4': 'h3',
 182             'hd2': 'h2',
 183             'hd3': 'h1'
 184         }
 185         return _dict[fm]
 186
 187     def _real_extract(self, url):
 188         video_id = self._match_id(url)
 189
 190         def retrieve_data(req_url, note):
 191             req = sanitized_Request(req_url)
 192
 193             cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
 194             if cn_verification_proxy:
 195                 req.add_header('Ytdl-request-proxy', cn_verification_proxy)
 196
 197             raw_data = self._download_json(req, video_id, note=note)
 198             return raw_data['data'][0]
 199
 200         video_password = self._downloader.params.get('videopassword', None)
 201
 202         # request basic data
 203         basic_data_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id
 204         if video_password:
 205             basic_data_url += '?password=%s' % video_password
 206
 207         data1 = retrieve_data(
 208             basic_data_url,
 209             'Downloading JSON metadata 1')
 210         data2 = retrieve_data(
 211             'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id,
 212             'Downloading JSON metadata 2')
 213
 214         error_code = data1.get('error_code')
 215         if error_code:
 216             error = data1.get('error')
 217             if error is not None and '因版权原因无法观看此视频' in error:
 218                 raise ExtractorError(
 219                     'Youku said: Sorry, this video is available in China only', expected=True)
 220             else:
 221                 msg = 'Youku server reported error %i' % error_code
 222                 if error is not None:
 223                     msg += ': ' + error
 224                 raise ExtractorError(msg)
 225
 226         title = data1['title']
 227
 228         # generate video_urls_dict
 229         video_urls_dict = self.construct_video_urls(data1, data2)
 230
 231         # construct info
 232         entries = [{
 233             'id': '%s_part%d' % (video_id, i + 1),
 234             'title': title,
 235             'formats': [],
 236             # some formats are not available for all parts, we have to detect
 237             # which one has all
 238         } for i in range(max(len(v) for v in data1['segs'].values()))]
 239         for fm in data1['streamtypes']:
 240             video_urls = video_urls_dict[fm]
 241             for video_url, seg, entry in zip(video_urls, data1['segs'][fm], entries):
 242                 entry['formats'].append({
 243                     'url': video_url,
 244                     'format_id': self.get_format_name(fm),
 245                     'ext': self.parse_ext_l(fm),
 246                     'filesize': int(seg['size']),
 247                 })
 248
 249         return {
 250             '_type': 'multi_video',
 251             'id': video_id,
 252             'title': title,
 253             'entries': entries,
 254         }