_ Git - youtube-dl/blob - youtube_dl/extractor/myvideo.py

   1 from __future__ import unicode_literals
   2
   3 import binascii
   4 import base64
   5 import hashlib
   6 import re
   7 import json
   8
   9 from .common import InfoExtractor
  10 from ..compat import (
  11     compat_ord,
  12     compat_urllib_parse,
  13     compat_urllib_request,
  14 )
  15 from ..utils import (
  16     ExtractorError,
  17 )
  18
  19
  20 class MyVideoIE(InfoExtractor):
  21     _VALID_URL = r'http://(?:www\.)?myvideo\.de/(?:[^/]+/)?watch/(?P<id>[0-9]+)/[^?/]+.*'
  22     IE_NAME = 'myvideo'
  23     _TEST = {
  24         'url': 'http://www.myvideo.de/watch/8229274/bowling_fail_or_win',
  25         'md5': '2d2753e8130479ba2cb7e0a37002053e',
  26         'info_dict': {
  27             'id': '8229274',
  28             'ext': 'flv',
  29             'title': 'bowling-fail-or-win',
  30         }
  31     }
  32
  33     # Original Code from: https://github.com/dersphere/plugin.video.myvideo_de.git
  34     # Released into the Public Domain by Tristan Fischer on 2013-05-19
  35     # https://github.com/rg3/youtube-dl/pull/842
  36     def __rc4crypt(self, data, key):
  37         x = 0
  38         box = list(range(256))
  39         for i in list(range(256)):
  40             x = (x + box[i] + compat_ord(key[i % len(key)])) % 256
  41             box[i], box[x] = box[x], box[i]
  42         x = 0
  43         y = 0
  44         out = ''
  45         for char in data:
  46             x = (x + 1) % 256
  47             y = (y + box[x]) % 256
  48             box[x], box[y] = box[y], box[x]
  49             out += chr(compat_ord(char) ^ box[(box[x] + box[y]) % 256])
  50         return out
  51
  52     def __md5(self, s):
  53         return hashlib.md5(s).hexdigest().encode()
  54
  55     def _real_extract(self, url):
  56         mobj = re.match(self._VALID_URL, url)
  57         video_id = mobj.group('id')
  58
  59         GK = (
  60           b'WXpnME1EZGhNRGhpTTJNM01XVmhOREU0WldNNVpHTTJOakpt'
  61           b'TW1FMU5tVTBNR05pWkRaa05XRXhNVFJoWVRVd1ptSXhaVEV3'
  62           b'TnpsbA0KTVRkbU1tSTRNdz09'
  63         )
  64
  65         # Get video webpage
  66         webpage_url = 'http://www.myvideo.de/watch/%s' % video_id
  67         webpage = self._download_webpage(webpage_url, video_id)
  68
  69         mobj = re.search('source src=\'(.+?)[.]([^.]+)\'', webpage)
  70         if mobj is not None:
  71             self.report_extraction(video_id)
  72             video_url = mobj.group(1) + '.flv'
  73
  74             video_title = self._html_search_regex('<title>([^<]+)</title>',
  75                                                   webpage, 'title')
  76
  77             return {
  78                 'id': video_id,
  79                 'url': video_url,
  80                 'title': video_title,
  81             }
  82
  83         mobj = re.search(r'data-video-service="/service/data/video/%s/config' % video_id, webpage)
  84         if mobj is not None:
  85             request = compat_urllib_request.Request('http://www.myvideo.de/service/data/video/%s/config' % video_id, '')
  86             response = self._download_webpage(request, video_id,
  87                                               'Downloading video info')
  88             info = json.loads(base64.b64decode(response).decode('utf-8'))
  89             return {
  90                 'id': video_id,
  91                 'title': info['title'],
  92                 'url': info['streaming_url'].replace('rtmpe', 'rtmpt'),
  93                 'play_path': info['filename'],
  94                 'ext': 'flv',
  95                 'thumbnail': info['thumbnail'][0]['url'],
  96             }
  97
  98         # try encxml
  99         mobj = re.search('var flashvars={(.+?)}', webpage)
 100         if mobj is None:
 101             raise ExtractorError('Unable to extract video')
 102
 103         params = {}
 104         encxml = ''
 105         sec = mobj.group(1)
 106         for (a, b) in re.findall('(.+?):\'(.+?)\',?', sec):
 107             if not a == '_encxml':
 108                 params[a] = b
 109             else:
 110                 encxml = compat_urllib_parse.unquote(b)
 111         if not params.get('domain'):
 112             params['domain'] = 'www.myvideo.de'
 113         xmldata_url = '%s?%s' % (encxml, compat_urllib_parse.urlencode(params))
 114         if 'flash_playertype=MTV' in xmldata_url:
 115             self._downloader.report_warning('avoiding MTV player')
 116             xmldata_url = (
 117                 'http://www.myvideo.de/dynamic/get_player_video_xml.php'
 118                 '?flash_playertype=D&ID=%s&_countlimit=4&autorun=yes'
 119             ) % video_id
 120
 121         # get enc data
 122         enc_data = self._download_webpage(xmldata_url, video_id).split('=')[1]
 123         enc_data_b = binascii.unhexlify(enc_data)
 124         sk = self.__md5(
 125             base64.b64decode(base64.b64decode(GK)) +
 126             self.__md5(
 127                 str(video_id).encode('utf-8')
 128             )
 129         )
 130         dec_data = self.__rc4crypt(enc_data_b, sk)
 131
 132         # extracting infos
 133         self.report_extraction(video_id)
 134
 135         video_url = None
 136         mobj = re.search('connectionurl=\'(.*?)\'', dec_data)
 137         if mobj:
 138             video_url = compat_urllib_parse.unquote(mobj.group(1))
 139             if 'myvideo2flash' in video_url:
 140                 self.report_warning(
 141                     'Rewriting URL to use unencrypted rtmp:// ...',
 142                     video_id)
 143                 video_url = video_url.replace('rtmpe://', 'rtmp://')
 144
 145         if not video_url:
 146             # extract non rtmp videos
 147             mobj = re.search('path=\'(http.*?)\' source=\'(.*?)\'', dec_data)
 148             if mobj is None:
 149                 raise ExtractorError('unable to extract url')
 150             video_url = compat_urllib_parse.unquote(mobj.group(1)) + compat_urllib_parse.unquote(mobj.group(2))
 151
 152         video_file = self._search_regex('source=\'(.*?)\'', dec_data, 'video file')
 153         video_file = compat_urllib_parse.unquote(video_file)
 154
 155         if not video_file.endswith('f4m'):
 156             ppath, prefix = video_file.split('.')
 157             video_playpath = '%s:%s' % (prefix, ppath)
 158         else:
 159             video_playpath = ''
 160
 161         video_swfobj = self._search_regex('swfobject.embedSWF\(\'(.+?)\'', webpage, 'swfobj')
 162         video_swfobj = compat_urllib_parse.unquote(video_swfobj)
 163
 164         video_title = self._html_search_regex("<h1(?: class='globalHd')?>(.*?)</h1>",
 165                                               webpage, 'title')
 166
 167         return {
 168             'id': video_id,
 169             'url': video_url,
 170             'tc_url': video_url,
 171             'title': video_title,
 172             'ext': 'flv',
 173             'play_path': video_playpath,
 174             'player_url': video_swfobj,
 175         }