youtube_dl/extractor/youku.py

   1 # coding: utf-8
   2
   3 from __future__ import unicode_literals
   4
   5 import json
   6 import math
   7 import random
   8 import re
   9 import time
  10
  11 from .common import InfoExtractor
  12 from ..utils import (
  13     ExtractorError,
  14 )
  15
  16
  17 class YoukuIE(InfoExtractor):
  18     _VALID_URL = r'''(?x)
  19         (?:
  20             http://(?:v|player)\.youku\.com/(?:v_show/id_|player\.php/sid/)|
  21             youku:)
  22         (?P<id>[A-Za-z0-9]+)(?:\.html|/v\.swf|)
  23     '''
  24     _TEST = {
  25         'url': 'http://v.youku.com/v_show/id_XNDgyMDQ2NTQw.html',
  26         'md5': 'ffe3f2e435663dc2d1eea34faeff5b5b',
  27         'params': {
  28             'test': False
  29         },
  30         'info_dict': {
  31             'id': 'XNDgyMDQ2NTQw_part00',
  32             'ext': 'flv',
  33             'title': 'youtube-dl test video "\'/\\ä↭𝕐'
  34         }
  35     }
  36
  37     def _gen_sid(self):
  38         nowTime = int(time.time() * 1000)
  39         random1 = random.randint(1000,1998)
  40         random2 = random.randint(1000,9999)
  41
  42         return "%d%d%d" %(nowTime,random1,random2)
  43
  44     def _get_file_ID_mix_string(self, seed):
  45         mixed = []
  46         source = list("abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890")
  47         seed = float(seed)
  48         for i in range(len(source)):
  49             seed  =  (seed * 211 + 30031) % 65536
  50             index  =  math.floor(seed / 65536 * len(source))
  51             mixed.append(source[int(index)])
  52             source.remove(source[int(index)])
  53         #return ''.join(mixed)
  54         return mixed
  55
  56     def _get_file_id(self, fileId, seed):
  57         mixed = self._get_file_ID_mix_string(seed)
  58         ids = fileId.split('*')
  59         realId = []
  60         for ch in ids:
  61             if ch:
  62                 realId.append(mixed[int(ch)])
  63         return ''.join(realId)
  64
  65     def _real_extract(self, url):
  66         mobj = re.match(self._VALID_URL, url)
  67         video_id = mobj.group('id')
  68
  69         info_url = 'http://v.youku.com/player/getPlayList/VideoIDS/' + video_id
  70
  71         config = self._download_json(info_url, video_id)
  72
  73         error_code = config['data'][0].get('error_code')
  74         if error_code:
  75             # -8 means blocked outside China.
  76             error = config['data'][0].get('error')  # Chinese and English, separated by newline.
  77             raise ExtractorError(error or 'Server reported error %i' % error_code,
  78                 expected=True)
  79
  80         video_title = config['data'][0]['title']
  81         seed = config['data'][0]['seed']
  82
  83         format = self._downloader.params.get('format', None)
  84         supported_format = list(config['data'][0]['streamfileids'].keys())
  85
  86         # TODO proper format selection
  87         if format is None or format == 'best':
  88             if 'hd2' in supported_format:
  89                 format = 'hd2'
  90             else:
  91                 format = 'flv'
  92             ext = 'flv'
  93         elif format == 'worst':
  94             format = 'mp4'
  95             ext = 'mp4'
  96         else:
  97             format = 'flv'
  98             ext = 'flv'
  99
 100         fileid = config['data'][0]['streamfileids'][format]
 101         keys = [s['k'] for s in config['data'][0]['segs'][format]]
 102         # segs is usually a dictionary, but an empty *list* if an error occured.
 103
 104         files_info=[]
 105         sid = self._gen_sid()
 106         fileid = self._get_file_id(fileid, seed)
 107
 108         #column 8,9 of fileid represent the segment number
 109         #fileid[7:9] should be changed
 110         for index, key in enumerate(keys):
 111             temp_fileid = '%s%02X%s' % (fileid[0:8], index, fileid[10:])
 112             download_url = 'http://k.youku.com/player/getFlvPath/sid/%s_%02X/st/flv/fileid/%s?k=%s' % (sid, index, temp_fileid, key)
 113
 114             info = {
 115                 'id': '%s_part%02d' % (video_id, index),
 116                 'url': download_url,
 117                 'uploader': None,
 118                 'upload_date': None,
 119                 'title': video_title,
 120                 'ext': ext,
 121             }
 122             files_info.append(info)
 123
 124         return files_info