youtube_dl/extractor/ustream.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     compat_urlparse,
   8     get_meta_content,
   9 )
  10
  11
  12 class UstreamIE(InfoExtractor):
  13     _VALID_URL = r'https?://www\.ustream\.tv/(?P<type>recorded|embed|embed/recorded)/(?P<videoID>\d+)'
  14     IE_NAME = 'ustream'
  15     _TEST = {
  16         'url': 'http://www.ustream.tv/recorded/20274954',
  17         'md5': '088f151799e8f572f84eb62f17d73e5c',
  18         'info_dict': {
  19             'id': '20274954',
  20             'ext': 'flv',
  21             'uploader': 'Young Americans for Liberty',
  22             'title': 'Young Americans for Liberty February 7, 2012 2:28 AM',
  23         },
  24     }
  25
  26     def _real_extract(self, url):
  27         m = re.match(self._VALID_URL, url)
  28         video_id = m.group('videoID')
  29
  30         # some sites use this embed format (see: http://github.com/rg3/youtube-dl/issues/2990)
  31         if m.group('type') == 'embed/recorded':
  32             video_id = m.group('videoID')
  33             desktop_url = 'http://www.ustream.tv/recorded/' + video_id
  34             return self.url_result(desktop_url, 'Ustream')
  35         if m.group('type') == 'embed':
  36             video_id = m.group('videoID')
  37             webpage = self._download_webpage(url, video_id)
  38             desktop_video_id = self._html_search_regex(
  39                 r'ContentVideoIds=\["([^"]*?)"\]', webpage, 'desktop_video_id')
  40             desktop_url = 'http://www.ustream.tv/recorded/' + desktop_video_id
  41             return self.url_result(desktop_url, 'Ustream')
  42
  43         video_url = 'http://tcdn.ustream.tv/video/%s' % video_id
  44         webpage = self._download_webpage(url, video_id)
  45
  46         self.report_extraction(video_id)
  47
  48         video_title = self._html_search_regex(r'data-title="(?P<title>.+)"',
  49             webpage, 'title')
  50
  51         uploader = self._html_search_regex(r'data-content-type="channel".*?>(?P<uploader>.*?)</a>',
  52             webpage, 'uploader', fatal=False, flags=re.DOTALL)
  53
  54         thumbnail = self._html_search_regex(r'<link rel="image_src" href="(?P<thumb>.*?)"',
  55             webpage, 'thumbnail', fatal=False)
  56
  57         return {
  58             'id': video_id,
  59             'url': video_url,
  60             'ext': 'flv',
  61             'title': video_title,
  62             'uploader': uploader,
  63             'thumbnail': thumbnail,
  64         }
  65
  66
  67 class UstreamChannelIE(InfoExtractor):
  68     _VALID_URL = r'https?://www\.ustream\.tv/channel/(?P<slug>.+)'
  69     IE_NAME = 'ustream:channel'
  70     _TEST = {
  71         'url': 'http://www.ustream.tv/channel/channeljapan',
  72         'info_dict': {
  73             'id': '10874166',
  74         },
  75         'playlist_mincount': 54,
  76     }
  77
  78     def _real_extract(self, url):
  79         m = re.match(self._VALID_URL, url)
  80         display_id = m.group('slug')
  81         webpage = self._download_webpage(url, display_id)
  82         channel_id = get_meta_content('ustream:channel_id', webpage)
  83
  84         BASE = 'http://www.ustream.tv'
  85         next_url = '/ajax/socialstream/videos/%s/1.json' % channel_id
  86         video_ids = []
  87         while next_url:
  88             reply = self._download_json(
  89                 compat_urlparse.urljoin(BASE, next_url), display_id,
  90                 note='Downloading video information (next: %d)' % (len(video_ids) + 1))
  91             video_ids.extend(re.findall(r'data-content-id="(\d.*)"', reply['data']))
  92             next_url = reply['nextUrl']
  93
  94         entries = [
  95             self.url_result('http://www.ustream.tv/recorded/' + vid, 'Ustream')
  96             for vid in video_ids]
  97         return {
  98             '_type': 'playlist',
  99             'id': channel_id,
 100             'display_id': display_id,
 101             'entries': entries,
 102         }