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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9     int_or_none,
  10 )
  11
  12
  13 class NewstubeIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?newstube\.ru/media/(?P<id>.+)'
  15     _TEST = {
  16         'url': 'http://www.newstube.ru/media/telekanal-cnn-peremestil-gorod-slavyansk-v-krym',
  17         'md5': '801eef0c2a9f4089fa04e4fe3533abdc',
  18         'info_dict': {
  19             'id': '728e0ef2-e187-4012-bac0-5a081fdcb1f6',
  20             'ext': 'mp4',
  21             'title': 'Телеканал CNN переместил город Славянск в Крым',
  22             'description': 'md5:419a8c9f03442bc0b0a794d689360335',
  23             'duration': 31.05,
  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             name = stream_info.find(ns('./Name')).text
  66             width = int(stream_info.find(ns('./Width')).text)
  67             height = int(stream_info.find(ns('./Height')).text)
  68
  69             formats.append({
  70                 'url': 'rtmp://%s/%s' % (server, app),
  71                 'app': app,
  72                 'play_path': '01/%s' % video_guid.upper(),
  73                 'rtmp_conn': ['S:%s' % session_id, 'S:%s' % media_id, 'S:n2'],
  74                 'page_url': url,
  75                 'ext': 'flv',
  76                 'format_id': 'rtmp' + ('-%s' % name if name else ''),
  77                 'width': width,
  78                 'height': height,
  79             })
  80
  81         sources_data = self._download_json(
  82             'http://www.newstube.ru/player2/getsources?guid=%s' % video_guid,
  83             video_guid, fatal=False)
  84         if sources_data:
  85             for source in sources_data.get('Sources', []):
  86                 source_url = source.get('Src')
  87                 if not source_url:
  88                     continue
  89                 height = int_or_none(source.get('Height'))
  90                 f = {
  91                     'format_id': 'http' + ('-%dp' % height if height else ''),
  92                     'url': source_url,
  93                     'width': int_or_none(source.get('Width')),
  94                     'height': height,
  95                 }
  96                 source_type = source.get('Type')
  97                 if source_type:
  98                     mobj = re.search(r'codecs="([^,]+),\s*([^"]+)"', source_type)
  99                     if mobj:
 100                         vcodec, acodec = mobj.groups()
 101                         f.update({
 102                             'vcodec': vcodec,
 103                             'acodec': acodec,
 104                         })
 105                 formats.append(f)
 106
 107         self._check_formats(formats, video_guid)
 108         self._sort_formats(formats)
 109
 110         return {
 111             'id': video_guid,
 112             'title': title,
 113             'description': description,
 114             'thumbnail': thumbnail,
 115             'duration': duration,
 116             'formats': formats,
 117         }