Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/youku.py

   1 import json
   2 import math
   3 import random
   4 import re
   5 import time
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     ExtractorError,
  10 )
  11
  12
  13 class YoukuIE(InfoExtractor):
  14     _VALID_URL =  r'(?:http://)?v\.youku\.com/v_show/id_(?P<ID>[A-Za-z0-9]+)\.html'
  15
  16     def _gen_sid(self):
  17         nowTime = int(time.time() * 1000)
  18         random1 = random.randint(1000,1998)
  19         random2 = random.randint(1000,9999)
  20
  21         return "%d%d%d" %(nowTime,random1,random2)
  22
  23     def _get_file_ID_mix_string(self, seed):
  24         mixed = []
  25         source = list("abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890")
  26         seed = float(seed)
  27         for i in range(len(source)):
  28             seed  =  (seed * 211 + 30031 ) % 65536
  29             index  =  math.floor(seed / 65536 * len(source) )
  30             mixed.append(source[int(index)])
  31             source.remove(source[int(index)])
  32         #return ''.join(mixed)
  33         return mixed
  34
  35     def _get_file_id(self, fileId, seed):
  36         mixed = self._get_file_ID_mix_string(seed)
  37         ids = fileId.split('*')
  38         realId = []
  39         for ch in ids:
  40             if ch:
  41                 realId.append(mixed[int(ch)])
  42         return ''.join(realId)
  43
  44     def _real_extract(self, url):
  45         mobj = re.match(self._VALID_URL, url)
  46         if mobj is None:
  47             raise ExtractorError(u'Invalid URL: %s' % url)
  48         video_id = mobj.group('ID')
  49
  50         info_url = 'http://v.youku.com/player/getPlayList/VideoIDS/' + video_id
  51
  52         jsondata = self._download_webpage(info_url, video_id)
  53
  54         self.report_extraction(video_id)
  55         try:
  56             config = json.loads(jsondata)
  57
  58             video_title =  config['data'][0]['title']
  59             seed = config['data'][0]['seed']
  60
  61             format = self._downloader.params.get('format', None)
  62             supported_format = list(config['data'][0]['streamfileids'].keys())
  63
  64             if format is None or format == 'best':
  65                 if 'hd2' in supported_format:
  66                     format = 'hd2'
  67                 else:
  68                     format = 'flv'
  69                 ext = u'flv'
  70             elif format == 'worst':
  71                 format = 'mp4'
  72                 ext = u'mp4'
  73             else:
  74                 format = 'flv'
  75                 ext = u'flv'
  76
  77
  78             fileid = config['data'][0]['streamfileids'][format]
  79             keys = [s['k'] for s in config['data'][0]['segs'][format]]
  80         except (UnicodeDecodeError, ValueError, KeyError):
  81             raise ExtractorError(u'Unable to extract info section')
  82
  83         files_info=[]
  84         sid = self._gen_sid()
  85         fileid = self._get_file_id(fileid, seed)
  86
  87         #column 8,9 of fileid represent the segment number
  88         #fileid[7:9] should be changed
  89         for index, key in enumerate(keys):
  90
  91             temp_fileid = '%s%02X%s' % (fileid[0:8], index, fileid[10:])
  92             download_url = 'http://f.youku.com/player/getFlvPath/sid/%s_%02X/st/flv/fileid/%s?k=%s' % (sid, index, temp_fileid, key)
  93
  94             info = {
  95                 'id': '%s_part%02d' % (video_id, index),
  96                 'url': download_url,
  97                 'uploader': None,
  98                 'upload_date': None,
  99                 'title': video_title,
 100                 'ext': ext,
 101             }
 102             files_info.append(info)
 103
 104         return files_info