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_unquote,
  13     compat_urllib_parse_urlencode,
  14 )
  15 from ..utils import (
  16     ExtractorError,
  17     sanitized_Request,
  18 )
  19
  20
  21 class MyVideoIE(InfoExtractor):
  22     _WORKING = False
  23     _VALID_URL = r'https?://(?:www\.)?myvideo\.de/(?:[^/]+/)?watch/(?P<id>[0-9]+)/[^?/]+.*'
  24     IE_NAME = 'myvideo'
  25     _TEST = {
  26         'url': 'http://www.myvideo.de/watch/8229274/bowling_fail_or_win',
  27         'md5': '2d2753e8130479ba2cb7e0a37002053e',
  28         'info_dict': {
  29             'id': '8229274',
  30             'ext': 'flv',
  31             'title': 'bowling-fail-or-win',
  32         }
  33     }
  34
  35     # Original Code from: https://github.com/dersphere/plugin.video.myvideo_de.git
  36     # Released into the Public Domain by Tristan Fischer on 2013-05-19
  37     # https://github.com/rg3/youtube-dl/pull/842
  38     def __rc4crypt(self, data, key):
  39         x = 0
  40         box = list(range(256))
  41         for i in list(range(256)):
  42             x = (x + box[i] + compat_ord(key[i % len(key)])) % 256
  43             box[i], box[x] = box[x], box[i]
  44         x = 0
  45         y = 0
  46         out = ''
  47         for char in data:
  48             x = (x + 1) % 256
  49             y = (y + box[x]) % 256
  50             box[x], box[y] = box[y], box[x]
  51             out += chr(compat_ord(char) ^ box[(box[x] + box[y]) % 256])
  52         return out
  53
  54     def __md5(self, s):
  55         return hashlib.md5(s).hexdigest().encode()
  56
  57     def _real_extract(self, url):
  58         mobj = re.match(self._VALID_URL, url)
  59         video_id = mobj.group('id')
  60
  61         GK = (
  62             b'WXpnME1EZGhNRGhpTTJNM01XVmhOREU0WldNNVpHTTJOakpt'
  63             b'TW1FMU5tVTBNR05pWkRaa05XRXhNVFJoWVRVd1ptSXhaVEV3'
  64             b'TnpsbA0KTVRkbU1tSTRNdz09'
  65         )
  66
  67         # Get video webpage
  68         webpage_url = 'http://www.myvideo.de/watch/%s' % video_id
  69         webpage = self._download_webpage(webpage_url, video_id)
  70
  71         mobj = re.search('source src=\'(.+?)[.]([^.]+)\'', webpage)
  72         if mobj is not None:
  73             self.report_extraction(video_id)
  74             video_url = mobj.group(1) + '.flv'
  75
  76             video_title = self._html_search_regex('<title>([^<]+)</title>',
  77                                                   webpage, 'title')
  78
  79             return {
  80                 'id': video_id,
  81                 'url': video_url,
  82                 'title': video_title,
  83             }
  84
  85         mobj = re.search(r'data-video-service="/service/data/video/%s/config' % video_id, webpage)
  86         if mobj is not None:
  87             request = sanitized_Request('http://www.myvideo.de/service/data/video/%s/config' % video_id, '')
  88             response = self._download_webpage(request, video_id,
  89                                               'Downloading video info')
  90             info = json.loads(base64.b64decode(response).decode('utf-8'))
  91             return {
  92                 'id': video_id,
  93                 'title': info['title'],
  94                 'url': info['streaming_url'].replace('rtmpe', 'rtmpt'),
  95                 'play_path': info['filename'],
  96                 'ext': 'flv',
  97                 'thumbnail': info['thumbnail'][0]['url'],
  98             }
  99
 100         # try encxml
 101         mobj = re.search('var flashvars={(.+?)}', webpage)
 102         if mobj is None:
 103             raise ExtractorError('Unable to extract video')
 104
 105         params = {}
 106         encxml = ''
 107         sec = mobj.group(1)
 108         for (a, b) in re.findall('(.+?):\'(.+?)\',?', sec):
 109             if not a == '_encxml':
 110                 params[a] = b
 111             else:
 112                 encxml = compat_urllib_parse_unquote(b)
 113         if not params.get('domain'):
 114             params['domain'] = 'www.myvideo.de'
 115         xmldata_url = '%s?%s' % (encxml, compat_urllib_parse_urlencode(params))
 116         if 'flash_playertype=MTV' in xmldata_url:
 117             self._downloader.report_warning('avoiding MTV player')
 118             xmldata_url = (
 119                 'http://www.myvideo.de/dynamic/get_player_video_xml.php'
 120                 '?flash_playertype=D&ID=%s&_countlimit=4&autorun=yes'
 121             ) % video_id
 122
 123         # get enc data
 124         enc_data = self._download_webpage(xmldata_url, video_id).split('=')[1]
 125         enc_data_b = binascii.unhexlify(enc_data)
 126         sk = self.__md5(
 127             base64.b64decode(base64.b64decode(GK)) +
 128             self.__md5(
 129                 str(video_id).encode('utf-8')
 130             )
 131         )
 132         dec_data = self.__rc4crypt(enc_data_b, sk)
 133
 134         # extracting infos
 135         self.report_extraction(video_id)
 136
 137         video_url = None
 138         mobj = re.search('connectionurl=\'(.*?)\'', dec_data)
 139         if mobj:
 140             video_url = compat_urllib_parse_unquote(mobj.group(1))
 141             if 'myvideo2flash' in video_url:
 142                 self.report_warning(
 143                     'Rewriting URL to use unencrypted rtmp:// ...',
 144                     video_id)
 145                 video_url = video_url.replace('rtmpe://', 'rtmp://')
 146
 147         if not video_url:
 148             # extract non rtmp videos
 149             mobj = re.search('path=\'(http.*?)\' source=\'(.*?)\'', dec_data)
 150             if mobj is None:
 151                 raise ExtractorError('unable to extract url')
 152             video_url = compat_urllib_parse_unquote(mobj.group(1)) + compat_urllib_parse_unquote(mobj.group(2))
 153
 154         video_file = self._search_regex('source=\'(.*?)\'', dec_data, 'video file')
 155         video_file = compat_urllib_parse_unquote(video_file)
 156
 157         if not video_file.endswith('f4m'):
 158             ppath, prefix = video_file.split('.')
 159             video_playpath = '%s:%s' % (prefix, ppath)
 160         else:
 161             video_playpath = ''
 162
 163         video_swfobj = self._search_regex('swfobject.embedSWF\(\'(.+?)\'', webpage, 'swfobj')
 164         video_swfobj = compat_urllib_parse_unquote(video_swfobj)
 165
 166         video_title = self._html_search_regex("<h1(?: class='globalHd')?>(.*?)</h1>",
 167                                               webpage, 'title')
 168
 169         return {
 170             'id': video_id,
 171             'url': video_url,
 172             'tc_url': video_url,
 173             'title': video_title,
 174             'ext': 'flv',
 175             'play_path': video_playpath,
 176             'player_url': video_swfobj,
 177         }