_ 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)/(?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         if m.group('type') == 'embed':
  29             video_id = m.group('videoID')
  30             webpage = self._download_webpage(url, video_id)
  31             desktop_video_id = self._html_search_regex(r'ContentVideoIds=\["([^"]*?)"\]', webpage, 'desktop_video_id')
  32             desktop_url = 'http://www.ustream.tv/recorded/' + desktop_video_id
  33             return self.url_result(desktop_url, 'Ustream')
  34
  35         video_id = m.group('videoID')
  36
  37         video_url = 'http://tcdn.ustream.tv/video/%s' % video_id
  38         webpage = self._download_webpage(url, video_id)
  39
  40         self.report_extraction(video_id)
  41
  42         video_title = self._html_search_regex(r'data-title="(?P<title>.+)"',
  43             webpage, 'title')
  44
  45         uploader = self._html_search_regex(r'data-content-type="channel".*?>(?P<uploader>.*?)</a>',
  46             webpage, 'uploader', fatal=False, flags=re.DOTALL)
  47
  48         thumbnail = self._html_search_regex(r'<link rel="image_src" href="(?P<thumb>.*?)"',
  49             webpage, 'thumbnail', fatal=False)
  50
  51         return {
  52             'id': video_id,
  53             'url': video_url,
  54             'ext': 'flv',
  55             'title': video_title,
  56             'uploader': uploader,
  57             'thumbnail': thumbnail,
  58         }
  59
  60
  61 class UstreamChannelIE(InfoExtractor):
  62     _VALID_URL = r'https?://www\.ustream\.tv/channel/(?P<slug>.+)'
  63     IE_NAME = 'ustream:channel'
  64
  65     def _real_extract(self, url):
  66         m = re.match(self._VALID_URL, url)
  67         slug = m.group('slug')
  68         webpage = self._download_webpage(url, slug)
  69         channel_id = get_meta_content('ustream:channel_id', webpage)
  70
  71         BASE = 'http://www.ustream.tv'
  72         next_url = '/ajax/socialstream/videos/%s/1.json' % channel_id
  73         video_ids = []
  74         while next_url:
  75             reply = json.loads(self._download_webpage(compat_urlparse.urljoin(BASE, next_url), channel_id))
  76             video_ids.extend(re.findall(r'data-content-id="(\d.*)"', reply['data']))
  77             next_url = reply['nextUrl']
  78
  79         urls = ['http://www.ustream.tv/recorded/' + vid for vid in video_ids]
  80         url_entries = [self.url_result(eurl, 'Ustream') for eurl in urls]
  81         return self.playlist_result(url_entries, channel_id)