youtube_dl/extractor/vimple.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3 import re
   4 import zlib
   5 import base64
   6 import xml.etree.ElementTree
   7
   8 from .common import InfoExtractor
   9
  10
  11 class VimpleIE(InfoExtractor):
  12     IE_DESC = 'Vimple.ru'
  13     _VALID_URL = r'https?://(player.vimple.ru/iframe|vimple.ru)/(?P<id>[a-f0-9]{10,})'
  14     _TESTS = [
  15         # Quality: Large, from iframe
  16         {
  17             'url': 'http://player.vimple.ru/iframe/b132bdfd71b546d3972f9ab9a25f201c',
  18             'info_dict': {
  19                 'id': 'b132bdfd71b546d3972f9ab9a25f201c',
  20                 'title': 'great-escape-minecraft.flv',
  21                 'ext': 'mp4',
  22                 'duration': 352,
  23                 'webpage_url': 'http://vimple.ru/b132bdfd71b546d3972f9ab9a25f201c',
  24             },
  25         },
  26         # Quality: Medium, from mainpage
  27         {
  28             'url': 'http://vimple.ru/a15950562888453b8e6f9572dc8600cd',
  29             'info_dict': {
  30                 'id': 'a15950562888453b8e6f9572dc8600cd',
  31                 'title': 'DB 01',
  32                 'ext': 'flv',
  33                 'duration': 1484,
  34                 'webpage_url': 'http://vimple.ru/a15950562888453b8e6f9572dc8600cd',
  35             }
  36         },
  37     ]
  38
  39     # http://jsunpack-n.googlecode.com/svn-history/r63/trunk/swf.py
  40
  41     def _real_extract(self, url):
  42         mobj = re.match(self._VALID_URL, url)
  43         video_id = mobj.group('id')
  44
  45         iframe_url = 'http://player.vimple.ru/iframe/%s' % video_id
  46
  47         iframe = self._download_webpage(iframe_url, video_id, note='Downloading iframe', errnote='unable to fetch iframe')
  48         player_url = self._html_search_regex(r'"(http://player.vimple.ru/flash/.+?)"', iframe, 'player url')
  49
  50         player = self._request_webpage(player_url, video_id, note='Downloading swf player').read()
  51
  52         # http://stackoverflow.com/a/6804758
  53         # http://stackoverflow.com/a/12073686
  54         player = zlib.decompress(player[8:])
  55
  56         xml_pieces = re.findall(b'([a-zA-Z0-9 =+/]{500})', player)
  57         xml_pieces = [piece[1:-1] for piece in xml_pieces]
  58
  59         xml_data = b''.join(xml_pieces)
  60         xml_data = base64.b64decode(xml_data)
  61
  62         xml_data = xml.etree.ElementTree.fromstring(xml_data)
  63
  64         video = xml_data.find('Video')
  65         quality = video.get('quality')
  66         q_tag = video.find(quality.capitalize())
  67
  68         formats = [
  69             {
  70                 'url': q_tag.get('url'),
  71                 'tbr': int(q_tag.get('bitrate')),
  72                 'filesize': int(q_tag.get('filesize')),
  73                 'format_id': quality,
  74             },
  75         ]
  76
  77         return {
  78             'id': video_id,
  79             'title': video.find('Title').text,
  80             'formats': formats,
  81             'thumbnail': video.find('Poster').get('url'),
  82             'duration': int(video.get('duration')),
  83             'webpage_url': video.find('Share').get('videoPageUrl'),
  84         }