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