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

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7
   8
   9 class NewstubeIE(InfoExtractor):
  10     _VALID_URL = r'https?://(?:www\.)?newstube\.ru/media/(?P<id>.+)'
  11     _TEST = {
  12         'url': 'http://newstube.ru/media/na-korable-progress-prodolzhaetsya-testirovanie-sistemy-kurs',
  13         'info_dict': {
  14             'id': 'd156a237-a6e9-4111-a682-039995f721f1',
  15             'ext': 'flv',
  16             'title': 'На корабле «Прогресс» продолжается тестирование системы «Курс»',
  17             'description': 'md5:d0cbe7b4a6f600552617e48548d5dc77',
  18             'duration': 20.04,
  19         },
  20         'params': {
  21             # rtmp download
  22             'skip_download': True,
  23         },
  24     }
  25
  26     def _real_extract(self, url):
  27         mobj = re.match(self._VALID_URL, url)
  28         video_id = mobj.group('id')
  29
  30         page = self._download_webpage(url, video_id, 'Downloading page')
  31
  32         video_guid = self._html_search_regex(
  33             r'<meta property="og:video" content="https?://(?:www\.)?newstube\.ru/freshplayer\.swf\?guid=(?P<guid>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})',
  34             page, 'video GUID')
  35
  36         player = self._download_xml(
  37             'http://p.newstube.ru/v2/player.asmx/GetAutoPlayInfo6?state=&url=%s&sessionId=&id=%s&placement=profile&location=n2' % (url, video_guid),
  38             video_guid, 'Downloading player XML')
  39
  40         def ns(s):
  41             return s.replace('/', '/%(ns)s') % {'ns': '{http://app1.newstube.ru/N2SiteWS/player.asmx}'}
  42
  43         session_id = player.find(ns('./SessionId')).text
  44         media_info = player.find(ns('./Medias/MediaInfo'))
  45         title = media_info.find(ns('./Name')).text
  46         description = self._og_search_description(page)
  47         thumbnail = media_info.find(ns('./KeyFrame')).text
  48         duration = int(media_info.find(ns('./Duration')).text) / 1000.0
  49
  50         formats = []
  51
  52         for stream_info in media_info.findall(ns('./Streams/StreamInfo')):
  53             media_location = stream_info.find(ns('./MediaLocation'))
  54             if media_location is None:
  55                 continue
  56
  57             server = media_location.find(ns('./Server')).text
  58             app = media_location.find(ns('./App')).text
  59             media_id = stream_info.find(ns('./Id')).text
  60             quality_id = stream_info.find(ns('./QualityId')).text
  61             name = stream_info.find(ns('./Name')).text
  62             width = int(stream_info.find(ns('./Width')).text)
  63             height = int(stream_info.find(ns('./Height')).text)
  64
  65             formats.append({
  66                 'url': 'rtmp://%s/%s' % (server, app),
  67                 'app': app,
  68                 'play_path': '01/%s' % video_guid.upper(),
  69                 'rtmp_conn': ['S:%s' % session_id, 'S:%s' % media_id, 'S:n2'],
  70                 'page_url': url,
  71                 'ext': 'flv',
  72                 'format_id': quality_id,
  73                 'format_note': name,
  74                 'width': width,
  75                 'height': height,
  76             })
  77
  78         self._sort_formats(formats)
  79
  80         return {
  81             'id': video_guid,
  82             'title': title,
  83             'description': description,
  84             'thumbnail': thumbnail,
  85             'duration': duration,
  86             'formats': formats,
  87         }