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 from ..utils import ExtractorError
   8
   9
  10 class NewstubeIE(InfoExtractor):
  11     _VALID_URL = r'https?://(?:www\.)?newstube\.ru/media/(?P<id>.+)'
  12     _TEST = {
  13         'url': 'http://www.newstube.ru/media/telekanal-cnn-peremestil-gorod-slavyansk-v-krym',
  14         'info_dict': {
  15             'id': '728e0ef2-e187-4012-bac0-5a081fdcb1f6',
  16             'ext': 'flv',
  17             'title': 'Телеканал CNN переместил город Славянск в Крым',
  18             'description': 'md5:419a8c9f03442bc0b0a794d689360335',
  19             'duration': 31.05,
  20         },
  21         'params': {
  22             # rtmp download
  23             'skip_download': True,
  24         },
  25     }
  26
  27     def _real_extract(self, url):
  28         mobj = re.match(self._VALID_URL, url)
  29         video_id = mobj.group('id')
  30
  31         page = self._download_webpage(url, video_id, 'Downloading page')
  32
  33         video_guid = self._html_search_regex(
  34             r'<meta property="og:video:url" 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})',
  35             page, 'video GUID')
  36
  37         player = self._download_xml(
  38             'http://p.newstube.ru/v2/player.asmx/GetAutoPlayInfo6?state=&url=%s&sessionId=&id=%s&placement=profile&location=n2' % (url, video_guid),
  39             video_guid, 'Downloading player XML')
  40
  41         def ns(s):
  42             return s.replace('/', '/%(ns)s') % {'ns': '{http://app1.newstube.ru/N2SiteWS/player.asmx}'}
  43
  44         error_message = player.find(ns('./ErrorMessage'))
  45         if error_message is not None:
  46             raise ExtractorError('%s returned error: %s' % (self.IE_NAME, error_message.text), expected=True)
  47
  48         session_id = player.find(ns('./SessionId')).text
  49         media_info = player.find(ns('./Medias/MediaInfo'))
  50         title = media_info.find(ns('./Name')).text
  51         description = self._og_search_description(page)
  52         thumbnail = media_info.find(ns('./KeyFrame')).text
  53         duration = int(media_info.find(ns('./Duration')).text) / 1000.0
  54
  55         formats = []
  56
  57         for stream_info in media_info.findall(ns('./Streams/StreamInfo')):
  58             media_location = stream_info.find(ns('./MediaLocation'))
  59             if media_location is None:
  60                 continue
  61
  62             server = media_location.find(ns('./Server')).text
  63             app = media_location.find(ns('./App')).text
  64             media_id = stream_info.find(ns('./Id')).text
  65             quality_id = stream_info.find(ns('./QualityId')).text
  66             name = stream_info.find(ns('./Name')).text
  67             width = int(stream_info.find(ns('./Width')).text)
  68             height = int(stream_info.find(ns('./Height')).text)
  69
  70             formats.append({
  71                 'url': 'rtmp://%s/%s' % (server, app),
  72                 'app': app,
  73                 'play_path': '01/%s' % video_guid.upper(),
  74                 'rtmp_conn': ['S:%s' % session_id, 'S:%s' % media_id, 'S:n2'],
  75                 'page_url': url,
  76                 'ext': 'flv',
  77                 'format_id': quality_id,
  78                 'format_note': name,
  79                 'width': width,
  80                 'height': height,
  81             })
  82
  83         self._sort_formats(formats)
  84
  85         return {
  86             'id': video_guid,
  87             'title': title,
  88             'description': description,
  89             'thumbnail': thumbnail,
  90             'duration': duration,
  91             'formats': formats,
  92         }