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|player)\.youku\.com/(?:v_show/id_|player\.php/sid/)|youku:)(?P<ID>[A-Za-z0-9]+)(?:\.html|/v\.swf|)'
  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             error_code = config['data'][0].get('error_code')
  70             if error_code:
  71                 # -8 means blocked outside China.
  72                 error = config['data'][0].get('error')  # Chinese and English, separated by newline.
  73                 raise ExtractorError(error or u'Server reported error %i' % error_code,
  74                     expected=True)
  75
  76             video_title =  config['data'][0]['title']
  77             seed = config['data'][0]['seed']
  78
  79             format = self._downloader.params.get('format', None)
  80             supported_format = list(config['data'][0]['streamfileids'].keys())
  81
  82             if format is None or format == 'best':
  83                 if 'hd2' in supported_format:
  84                     format = 'hd2'
  85                 else:
  86                     format = 'flv'
  87                 ext = u'flv'
  88             elif format == 'worst':
  89                 format = 'mp4'
  90                 ext = u'mp4'
  91             else:
  92                 format = 'flv'
  93                 ext = u'flv'
  94
  95
  96             fileid = config['data'][0]['streamfileids'][format]
  97             keys = [s['k'] for s in config['data'][0]['segs'][format]]
  98             # segs is usually a dictionary, but an empty *list* if an error occured.
  99         except (UnicodeDecodeError, ValueError, KeyError):
 100             raise ExtractorError(u'Unable to extract info section')
 101
 102         files_info=[]
 103         sid = self._gen_sid()
 104         fileid = self._get_file_id(fileid, seed)
 105
 106         #column 8,9 of fileid represent the segment number
 107         #fileid[7:9] should be changed
 108         for index, key in enumerate(keys):
 109
 110             temp_fileid = '%s%02X%s' % (fileid[0:8], index, fileid[10:])
 111             download_url = 'http://f.youku.com/player/getFlvPath/sid/%s_%02X/st/flv/fileid/%s?k=%s' % (sid, index, temp_fileid, key)
 112
 113             info = {
 114                 'id': '%s_part%02d' % (video_id, index),
 115                 'url': download_url,
 116                 'uploader': None,
 117                 'upload_date': None,
 118                 'title': video_title,
 119                 'ext': ext,
 120             }
 121             files_info.append(info)
 122
 123         return files_info