_ Git - youtube-dl/blob - 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         'file': '20274954.flv',
  19         'md5': '088f151799e8f572f84eb62f17d73e5c',
  20         'info_dict': {
  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         if m.group('type') == 'embed/recorded': # some sites use this embed format (see: http://github.com/rg3/youtube-dl/issues/2990)
  31             video_id = m.group('videoID')
  32             webpage = self._download_webpage(url, video_id, note="Downloading embedded Ustream page")
  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(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
  70     def _real_extract(self, url):
  71         m = re.match(self._VALID_URL, url)
  72         slug = m.group('slug')
  73         webpage = self._download_webpage(url, slug)
  74         channel_id = get_meta_content('ustream:channel_id', webpage)
  75
  76         BASE = 'http://www.ustream.tv'
  77         next_url = '/ajax/socialstream/videos/%s/1.json' % channel_id
  78         video_ids = []
  79         while next_url:
  80             reply = json.loads(self._download_webpage(compat_urlparse.urljoin(BASE, next_url), channel_id))
  81             video_ids.extend(re.findall(r'data-content-id="(\d.*)"', reply['data']))
  82             next_url = reply['nextUrl']
  83
  84         urls = ['http://www.ustream.tv/recorded/' + vid for vid in video_ids]
  85         url_entries = [self.url_result(eurl, 'Ustream') for eurl in urls]
  86         return self.playlist_result(url_entries, channel_id)