youtube_dl/extractor/ustream.py

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