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

   1 import binascii
   2 import base64
   3 import hashlib
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     compat_ord,
   9     compat_urllib_parse,
  10
  11     ExtractorError,
  12 )
  13
  14
  15
  16 class MyVideoIE(InfoExtractor):
  17     """Information Extractor for myvideo.de."""
  18
  19     _VALID_URL = r'(?:http://)?(?:www\.)?myvideo\.de/watch/([0-9]+)/([^?/]+).*'
  20     IE_NAME = u'myvideo'
  21
  22     # Original Code from: https://github.com/dersphere/plugin.video.myvideo_de.git
  23     # Released into the Public Domain by Tristan Fischer on 2013-05-19
  24     # https://github.com/rg3/youtube-dl/pull/842
  25     def __rc4crypt(self,data, key):
  26         x = 0
  27         box = list(range(256))
  28         for i in list(range(256)):
  29             x = (x + box[i] + compat_ord(key[i % len(key)])) % 256
  30             box[i], box[x] = box[x], box[i]
  31         x = 0
  32         y = 0
  33         out = ''
  34         for char in data:
  35             x = (x + 1) % 256
  36             y = (y + box[x]) % 256
  37             box[x], box[y] = box[y], box[x]
  38             out += chr(compat_ord(char) ^ box[(box[x] + box[y]) % 256])
  39         return out
  40
  41     def __md5(self,s):
  42         return hashlib.md5(s).hexdigest().encode()
  43
  44     def _real_extract(self,url):
  45         mobj = re.match(self._VALID_URL, url)
  46         if mobj is None:
  47             raise ExtractorError(u'invalid URL: %s' % url)
  48
  49         video_id = mobj.group(1)
  50
  51         GK = (
  52           b'WXpnME1EZGhNRGhpTTJNM01XVmhOREU0WldNNVpHTTJOakpt'
  53           b'TW1FMU5tVTBNR05pWkRaa05XRXhNVFJoWVRVd1ptSXhaVEV3'
  54           b'TnpsbA0KTVRkbU1tSTRNdz09'
  55         )
  56
  57         # Get video webpage
  58         webpage_url = 'http://www.myvideo.de/watch/%s' % video_id
  59         webpage = self._download_webpage(webpage_url, video_id)
  60
  61         mobj = re.search('source src=\'(.+?)[.]([^.]+)\'', webpage)
  62         if mobj is not None:
  63             self.report_extraction(video_id)
  64             video_url = mobj.group(1) + '.flv'
  65
  66             video_title = self._html_search_regex('<title>([^<]+)</title>',
  67                 webpage, u'title')
  68
  69             video_ext = self._search_regex('[.](.+?)$', video_url, u'extension')
  70
  71             return [{
  72                 'id':       video_id,
  73                 'url':      video_url,
  74                 'uploader': None,
  75                 'upload_date':  None,
  76                 'title':    video_title,
  77                 'ext':      video_ext,
  78             }]
  79
  80         # try encxml
  81         mobj = re.search('var flashvars={(.+?)}', webpage)
  82         if mobj is None:
  83             raise ExtractorError(u'Unable to extract video')
  84
  85         params = {}
  86         encxml = ''
  87         sec = mobj.group(1)
  88         for (a, b) in re.findall('(.+?):\'(.+?)\',?', sec):
  89             if not a == '_encxml':
  90                 params[a] = b
  91             else:
  92                 encxml = compat_urllib_parse.unquote(b)
  93         if not params.get('domain'):
  94             params['domain'] = 'www.myvideo.de'
  95         xmldata_url = '%s?%s' % (encxml, compat_urllib_parse.urlencode(params))
  96         if 'flash_playertype=MTV' in xmldata_url:
  97             self._downloader.report_warning(u'avoiding MTV player')
  98             xmldata_url = (
  99                 'http://www.myvideo.de/dynamic/get_player_video_xml.php'
 100                 '?flash_playertype=D&ID=%s&_countlimit=4&autorun=yes'
 101             ) % video_id
 102
 103         # get enc data
 104         enc_data = self._download_webpage(xmldata_url, video_id).split('=')[1]
 105         enc_data_b = binascii.unhexlify(enc_data)
 106         sk = self.__md5(
 107             base64.b64decode(base64.b64decode(GK)) +
 108             self.__md5(
 109                 str(video_id).encode('utf-8')
 110             )
 111         )
 112         dec_data = self.__rc4crypt(enc_data_b, sk)
 113
 114         # extracting infos
 115         self.report_extraction(video_id)
 116
 117         video_url = None
 118         mobj = re.search('connectionurl=\'(.*?)\'', dec_data)
 119         if mobj:
 120             video_url = compat_urllib_parse.unquote(mobj.group(1))
 121             if 'myvideo2flash' in video_url:
 122                 self._downloader.report_warning(u'forcing RTMPT ...')
 123                 video_url = video_url.replace('rtmpe://', 'rtmpt://')
 124
 125         if not video_url:
 126             # extract non rtmp videos
 127             mobj = re.search('path=\'(http.*?)\' source=\'(.*?)\'', dec_data)
 128             if mobj is None:
 129                 raise ExtractorError(u'unable to extract url')
 130             video_url = compat_urllib_parse.unquote(mobj.group(1)) + compat_urllib_parse.unquote(mobj.group(2))
 131
 132         video_file = self._search_regex('source=\'(.*?)\'', dec_data, u'video file')
 133         video_file = compat_urllib_parse.unquote(video_file)
 134
 135         if not video_file.endswith('f4m'):
 136             ppath, prefix = video_file.split('.')
 137             video_playpath = '%s:%s' % (prefix, ppath)
 138             video_hls_playlist = ''
 139         else:
 140             video_playpath = ''
 141             video_hls_playlist = (
 142                 video_file
 143             ).replace('.f4m', '.m3u8')
 144
 145         video_swfobj = self._search_regex('swfobject.embedSWF\(\'(.+?)\'', webpage, u'swfobj')
 146         video_swfobj = compat_urllib_parse.unquote(video_swfobj)
 147
 148         video_title = self._html_search_regex("<h1(?: class='globalHd')?>(.*?)</h1>",
 149             webpage, u'title')
 150
 151         return [{
 152             'id':                 video_id,
 153             'url':                video_url,
 154             'tc_url':             video_url,
 155             'uploader':           None,
 156             'upload_date':        None,
 157             'title':              video_title,
 158             'ext':                u'flv',
 159             'play_path':          video_playpath,
 160             'video_file':         video_file,
 161             'video_hls_playlist': video_hls_playlist,
 162             'player_url':         video_swfobj,
 163         }]
 164