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     _TEST = {
  22         u'url': u'http://www.myvideo.de/watch/8229274/bowling_fail_or_win',
  23         u'file': u'8229274.flv',
  24         u'md5': u'2d2753e8130479ba2cb7e0a37002053e',
  25         u'info_dict': {
  26             u"title": u"bowling-fail-or-win"
  27         }
  28     }
  29
  30     # Original Code from: https://github.com/dersphere/plugin.video.myvideo_de.git
  31     # Released into the Public Domain by Tristan Fischer on 2013-05-19
  32     # https://github.com/rg3/youtube-dl/pull/842
  33     def __rc4crypt(self,data, key):
  34         x = 0
  35         box = list(range(256))
  36         for i in list(range(256)):
  37             x = (x + box[i] + compat_ord(key[i % len(key)])) % 256
  38             box[i], box[x] = box[x], box[i]
  39         x = 0
  40         y = 0
  41         out = ''
  42         for char in data:
  43             x = (x + 1) % 256
  44             y = (y + box[x]) % 256
  45             box[x], box[y] = box[y], box[x]
  46             out += chr(compat_ord(char) ^ box[(box[x] + box[y]) % 256])
  47         return out
  48
  49     def __md5(self,s):
  50         return hashlib.md5(s).hexdigest().encode()
  51
  52     def _real_extract(self,url):
  53         mobj = re.match(self._VALID_URL, url)
  54         if mobj is None:
  55             raise ExtractorError(u'invalid URL: %s' % url)
  56
  57         video_id = mobj.group(1)
  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, u'title')
  76
  77             video_ext = self._search_regex('[.](.+?)$', video_url, u'extension')
  78
  79             return [{
  80                 'id':       video_id,
  81                 'url':      video_url,
  82                 'uploader': None,
  83                 'upload_date':  None,
  84                 'title':    video_title,
  85                 'ext':      video_ext,
  86             }]
  87
  88         # try encxml
  89         mobj = re.search('var flashvars={(.+?)}', webpage)
  90         if mobj is None:
  91             raise ExtractorError(u'Unable to extract video')
  92
  93         params = {}
  94         encxml = ''
  95         sec = mobj.group(1)
  96         for (a, b) in re.findall('(.+?):\'(.+?)\',?', sec):
  97             if not a == '_encxml':
  98                 params[a] = b
  99             else:
 100                 encxml = compat_urllib_parse.unquote(b)
 101         if not params.get('domain'):
 102             params['domain'] = 'www.myvideo.de'
 103         xmldata_url = '%s?%s' % (encxml, compat_urllib_parse.urlencode(params))
 104         if 'flash_playertype=MTV' in xmldata_url:
 105             self._downloader.report_warning(u'avoiding MTV player')
 106             xmldata_url = (
 107                 'http://www.myvideo.de/dynamic/get_player_video_xml.php'
 108                 '?flash_playertype=D&ID=%s&_countlimit=4&autorun=yes'
 109             ) % video_id
 110
 111         # get enc data
 112         enc_data = self._download_webpage(xmldata_url, video_id).split('=')[1]
 113         enc_data_b = binascii.unhexlify(enc_data)
 114         sk = self.__md5(
 115             base64.b64decode(base64.b64decode(GK)) +
 116             self.__md5(
 117                 str(video_id).encode('utf-8')
 118             )
 119         )
 120         dec_data = self.__rc4crypt(enc_data_b, sk)
 121
 122         # extracting infos
 123         self.report_extraction(video_id)
 124
 125         video_url = None
 126         mobj = re.search('connectionurl=\'(.*?)\'', dec_data)
 127         if mobj:
 128             video_url = compat_urllib_parse.unquote(mobj.group(1))
 129             if 'myvideo2flash' in video_url:
 130                 self._downloader.report_warning(u'forcing RTMPT ...')
 131                 video_url = video_url.replace('rtmpe://', 'rtmpt://')
 132
 133         if not video_url:
 134             # extract non rtmp videos
 135             mobj = re.search('path=\'(http.*?)\' source=\'(.*?)\'', dec_data)
 136             if mobj is None:
 137                 raise ExtractorError(u'unable to extract url')
 138             video_url = compat_urllib_parse.unquote(mobj.group(1)) + compat_urllib_parse.unquote(mobj.group(2))
 139
 140         video_file = self._search_regex('source=\'(.*?)\'', dec_data, u'video file')
 141         video_file = compat_urllib_parse.unquote(video_file)
 142
 143         if not video_file.endswith('f4m'):
 144             ppath, prefix = video_file.split('.')
 145             video_playpath = '%s:%s' % (prefix, ppath)
 146             video_hls_playlist = ''
 147         else:
 148             video_playpath = ''
 149             video_hls_playlist = (
 150                 video_file
 151             ).replace('.f4m', '.m3u8')
 152
 153         video_swfobj = self._search_regex('swfobject.embedSWF\(\'(.+?)\'', webpage, u'swfobj')
 154         video_swfobj = compat_urllib_parse.unquote(video_swfobj)
 155
 156         video_title = self._html_search_regex("<h1(?: class='globalHd')?>(.*?)</h1>",
 157             webpage, u'title')
 158
 159         return [{
 160             'id':                 video_id,
 161             'url':                video_url,
 162             'tc_url':             video_url,
 163             'uploader':           None,
 164             'upload_date':        None,
 165             'title':              video_title,
 166             'ext':                u'flv',
 167             'play_path':          video_playpath,
 168             'video_file':         video_file,
 169             'video_hls_playlist': video_hls_playlist,
 170             'player_url':         video_swfobj,
 171         }]
 172