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

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