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