_ 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 ..compat import (
   8     compat_urlparse,
   9 )
  10
  11
  12 class UstreamIE(InfoExtractor):
  13     _VALID_URL = r'https?://www\.ustream\.tv/(?P<type>recorded|embed|embed/recorded)/(?P<videoID>\d+)'
  14     IE_NAME = 'ustream'
  15     _TEST = {
  16         'url': 'http://www.ustream.tv/recorded/20274954',
  17         'md5': '088f151799e8f572f84eb62f17d73e5c',
  18         'info_dict': {
  19             'id': '20274954',
  20             'ext': 'flv',
  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         # some sites use this embed format (see: http://github.com/rg3/youtube-dl/issues/2990)
  31         if m.group('type') == 'embed/recorded':
  32             video_id = m.group('videoID')
  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(
  39                 r'ContentVideoIds=\["([^"]*?)"\]', webpage, 'desktop_video_id')
  40             desktop_url = 'http://www.ustream.tv/recorded/' + desktop_video_id
  41             return self.url_result(desktop_url, 'Ustream')
  42
  43         params = self._download_json(
  44             'http://cdngw.ustream.tv/rgwjson/Viewer.getVideo/' + json.dumps({
  45                 'brandId': 1,
  46                 'videoId': int(video_id),
  47                 'autoplay': False,
  48             }), video_id)
  49
  50         video_url = params['flv']
  51
  52         webpage = self._download_webpage(url, video_id)
  53
  54         self.report_extraction(video_id)
  55
  56         video_title = self._html_search_regex(r'data-title="(?P<title>.+)"',
  57                                               webpage, 'title')
  58
  59         uploader = self._html_search_regex(r'data-content-type="channel".*?>(?P<uploader>.*?)</a>',
  60                                            webpage, 'uploader', fatal=False, flags=re.DOTALL)
  61
  62         thumbnail = self._html_search_regex(r'<link rel="image_src" href="(?P<thumb>.*?)"',
  63                                             webpage, 'thumbnail', fatal=False)
  64
  65         return {
  66             'id': video_id,
  67             'url': video_url,
  68             'ext': 'flv',
  69             'title': video_title,
  70             'uploader': uploader,
  71             'thumbnail': thumbnail,
  72         }
  73
  74
  75 class UstreamChannelIE(InfoExtractor):
  76     _VALID_URL = r'https?://www\.ustream\.tv/channel/(?P<slug>.+)'
  77     IE_NAME = 'ustream:channel'
  78     _TEST = {
  79         'url': 'http://www.ustream.tv/channel/channeljapan',
  80         'info_dict': {
  81             'id': '10874166',
  82         },
  83         'playlist_mincount': 17,
  84     }
  85
  86     def _real_extract(self, url):
  87         m = re.match(self._VALID_URL, url)
  88         display_id = m.group('slug')
  89         webpage = self._download_webpage(url, display_id)
  90         channel_id = self._html_search_meta('ustream:channel_id', webpage)
  91
  92         BASE = 'http://www.ustream.tv'
  93         next_url = '/ajax/socialstream/videos/%s/1.json' % channel_id
  94         video_ids = []
  95         while next_url:
  96             reply = self._download_json(
  97                 compat_urlparse.urljoin(BASE, next_url), display_id,
  98                 note='Downloading video information (next: %d)' % (len(video_ids) + 1))
  99             video_ids.extend(re.findall(r'data-content-id="(\d.*)"', reply['data']))
 100             next_url = reply['nextUrl']
 101
 102         entries = [
 103             self.url_result('http://www.ustream.tv/recorded/' + vid, 'Ustream')
 104             for vid in video_ids]
 105         return {
 106             '_type': 'playlist',
 107             'id': channel_id,
 108             'display_id': display_id,
 109             'entries': entries,
 110         }