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

   1 # coding: utf-8
   2
   3 import json
   4 import math
   5 import random
   6 import re
   7 import time
   8
   9 from .common import InfoExtractor
  10 from ..utils import (
  11     ExtractorError,
  12 )
  13
  14
  15 class YoukuIE(InfoExtractor):
  16     _VALID_URL =  r'(?:http://)?v\.youku\.com/v_show/id_(?P<ID>[A-Za-z0-9]+)\.html'
  17     _TEST =   {
  18         u"url": u"http://v.youku.com/v_show/id_XNDgyMDQ2NTQw.html",
  19         u"file": u"XNDgyMDQ2NTQw_part00.flv",
  20         u"md5": u"ffe3f2e435663dc2d1eea34faeff5b5b",
  21         u"params": { u"test": False },
  22         u"info_dict": {
  23             u"title": u"youtube-dl test video \"'/\\ä↭𝕐"
  24         }
  25     }
  26
  27
  28     def _gen_sid(self):
  29         nowTime = int(time.time() * 1000)
  30         random1 = random.randint(1000,1998)
  31         random2 = random.randint(1000,9999)
  32
  33         return "%d%d%d" %(nowTime,random1,random2)
  34
  35     def _get_file_ID_mix_string(self, seed):
  36         mixed = []
  37         source = list("abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890")
  38         seed = float(seed)
  39         for i in range(len(source)):
  40             seed  =  (seed * 211 + 30031 ) % 65536
  41             index  =  math.floor(seed / 65536 * len(source) )
  42             mixed.append(source[int(index)])
  43             source.remove(source[int(index)])
  44         #return ''.join(mixed)
  45         return mixed
  46
  47     def _get_file_id(self, fileId, seed):
  48         mixed = self._get_file_ID_mix_string(seed)
  49         ids = fileId.split('*')
  50         realId = []
  51         for ch in ids:
  52             if ch:
  53                 realId.append(mixed[int(ch)])
  54         return ''.join(realId)
  55
  56     def _real_extract(self, url):
  57         mobj = re.match(self._VALID_URL, url)
  58         if mobj is None:
  59             raise ExtractorError(u'Invalid URL: %s' % url)
  60         video_id = mobj.group('ID')
  61
  62         info_url = 'http://v.youku.com/player/getPlayList/VideoIDS/' + video_id
  63
  64         jsondata = self._download_webpage(info_url, video_id)
  65
  66         self.report_extraction(video_id)
  67         try:
  68             config = json.loads(jsondata)
  69
  70             video_title =  config['data'][0]['title']
  71             seed = config['data'][0]['seed']
  72
  73             format = self._downloader.params.get('format', None)
  74             supported_format = list(config['data'][0]['streamfileids'].keys())
  75
  76             if format is None or format == 'best':
  77                 if 'hd2' in supported_format:
  78                     format = 'hd2'
  79                 else:
  80                     format = 'flv'
  81                 ext = u'flv'
  82             elif format == 'worst':
  83                 format = 'mp4'
  84                 ext = u'mp4'
  85             else:
  86                 format = 'flv'
  87                 ext = u'flv'
  88
  89
  90             fileid = config['data'][0]['streamfileids'][format]
  91             keys = [s['k'] for s in config['data'][0]['segs'][format]]
  92         except (UnicodeDecodeError, ValueError, KeyError):
  93             raise ExtractorError(u'Unable to extract info section')
  94
  95         files_info=[]
  96         sid = self._gen_sid()
  97         fileid = self._get_file_id(fileid, seed)
  98
  99         #column 8,9 of fileid represent the segment number
 100         #fileid[7:9] should be changed
 101         for index, key in enumerate(keys):
 102
 103             temp_fileid = '%s%02X%s' % (fileid[0:8], index, fileid[10:])
 104             download_url = 'http://f.youku.com/player/getFlvPath/sid/%s_%02X/st/flv/fileid/%s?k=%s' % (sid, index, temp_fileid, key)
 105
 106             info = {
 107                 'id': '%s_part%02d' % (video_id, index),
 108                 'url': download_url,
 109                 'uploader': None,
 110                 'upload_date': None,
 111                 'title': video_title,
 112                 'ext': ext,
 113             }
 114             files_info.append(info)
 115
 116         return files_info