_ 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     _TESTS = [{
  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         'url': 'http://www.ustream.tv/embed/17357891',
  27         'file': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman-45734260.flv',
  28         'md5': '27b99cdb639c9b12a79bca876a073417',
  29         'info_dict': {
  30             "uploader": "AU SPA: The NSA and Privacy",
  31             "title": "NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman"
  32     },
  33     }
  34     ]
  35
  36     def _real_extract(self, url):
  37         m = re.match(self._VALID_URL, url)
  38         if m.group('type') == 'embed':
  39             video_id = m.group('videoID')
  40             webpage = self._download_webpage(url, video_id)
  41             desktop_video_id = self._html_search_regex(r'ContentVideoIds=\["([^"]*?)"\]', webpage, 'desktop_video_id')
  42             desktop_url = 'http://www.ustream.tv/recorded/' + desktop_video_id
  43             return self.url_result(desktop_url, 'Ustream')
  44
  45         video_id = m.group('videoID')
  46
  47         video_url = 'http://tcdn.ustream.tv/video/%s' % video_id
  48         webpage = self._download_webpage(url, video_id)
  49
  50         self.report_extraction(video_id)
  51
  52         video_title = self._html_search_regex(r'data-title="(?P<title>.+)"',
  53             webpage, 'title')
  54
  55         uploader = self._html_search_regex(r'data-content-type="channel".*?>(?P<uploader>.*?)</a>',
  56             webpage, 'uploader', fatal=False, flags=re.DOTALL)
  57
  58         thumbnail = self._html_search_regex(r'<link rel="image_src" href="(?P<thumb>.*?)"',
  59             webpage, 'thumbnail', fatal=False)
  60
  61         return {
  62             'id': video_id,
  63             'url': video_url,
  64             'ext': 'flv',
  65             'title': video_title,
  66             'uploader': uploader,
  67             'thumbnail': thumbnail,
  68         }
  69
  70
  71 class UstreamChannelIE(InfoExtractor):
  72     _VALID_URL = r'https?://www\.ustream\.tv/channel/(?P<slug>.+)'
  73     IE_NAME = 'ustream:channel'
  74
  75     def _real_extract(self, url):
  76         m = re.match(self._VALID_URL, url)
  77         slug = m.group('slug')
  78         webpage = self._download_webpage(url, slug)
  79         channel_id = get_meta_content('ustream:channel_id', webpage)
  80
  81         BASE = 'http://www.ustream.tv'
  82         next_url = '/ajax/socialstream/videos/%s/1.json' % channel_id
  83         video_ids = []
  84         while next_url:
  85             reply = json.loads(self._download_webpage(compat_urlparse.urljoin(BASE, next_url), channel_id))
  86             video_ids.extend(re.findall(r'data-content-id="(\d.*)"', reply['data']))
  87             next_url = reply['nextUrl']
  88
  89         urls = ['http://www.ustream.tv/recorded/' + vid for vid in video_ids]
  90         url_entries = [self.url_result(eurl, 'Ustream') for eurl in urls]
  91         return self.playlist_result(url_entries, channel_id)