_ 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 from ..utils import ExtractorError
  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         'md5': '088f151799e8f572f84eb62f17d73e5c',
  19         'info_dict': {
  20             'id': '20274954',
  21             'ext': 'flv',
  22             'uploader': 'Young Americans for Liberty',
  23             'title': 'Young Americans for Liberty February 7, 2012 2:28 AM',
  24         },
  25     }
  26
  27     def _real_extract(self, url):
  28         m = re.match(self._VALID_URL, url)
  29         video_id = m.group('videoID')
  30
  31         # some sites use this embed format (see: http://github.com/rg3/youtube-dl/issues/2990)
  32         if m.group('type') == 'embed/recorded':
  33             video_id = m.group('videoID')
  34             desktop_url = 'http://www.ustream.tv/recorded/' + video_id
  35             return self.url_result(desktop_url, 'Ustream')
  36         if m.group('type') == 'embed':
  37             video_id = m.group('videoID')
  38             webpage = self._download_webpage(url, video_id)
  39             desktop_video_id = self._html_search_regex(
  40                 r'ContentVideoIds=\["([^"]*?)"\]', webpage, 'desktop_video_id')
  41             desktop_url = 'http://www.ustream.tv/recorded/' + desktop_video_id
  42             return self.url_result(desktop_url, 'Ustream')
  43
  44         params = self._download_json(
  45             'http://cdngw.ustream.tv/rgwjson/Viewer.getVideo/' + json.dumps({
  46                 'brandId': 1,
  47                 'videoId': int(video_id),
  48                 'autoplay': False,
  49             }), video_id)
  50
  51         if 'error' in params:
  52             raise ExtractorError(params['error']['message'], expected=True)
  53
  54         video_url = params['flv']
  55
  56         webpage = self._download_webpage(url, video_id)
  57
  58         self.report_extraction(video_id)
  59
  60         video_title = self._html_search_regex(r'data-title="(?P<title>.+)"',
  61                                               webpage, 'title')
  62
  63         uploader = self._html_search_regex(r'data-content-type="channel".*?>(?P<uploader>.*?)</a>',
  64                                            webpage, 'uploader', fatal=False, flags=re.DOTALL)
  65
  66         thumbnail = self._html_search_regex(r'<link rel="image_src" href="(?P<thumb>.*?)"',
  67                                             webpage, 'thumbnail', fatal=False)
  68
  69         return {
  70             'id': video_id,
  71             'url': video_url,
  72             'ext': 'flv',
  73             'title': video_title,
  74             'uploader': uploader,
  75             'thumbnail': thumbnail,
  76         }
  77
  78
  79 class UstreamChannelIE(InfoExtractor):
  80     _VALID_URL = r'https?://www\.ustream\.tv/channel/(?P<slug>.+)'
  81     IE_NAME = 'ustream:channel'
  82     _TEST = {
  83         'url': 'http://www.ustream.tv/channel/channeljapan',
  84         'info_dict': {
  85             'id': '10874166',
  86         },
  87         'playlist_mincount': 17,
  88     }
  89
  90     def _real_extract(self, url):
  91         m = re.match(self._VALID_URL, url)
  92         display_id = m.group('slug')
  93         webpage = self._download_webpage(url, display_id)
  94         channel_id = self._html_search_meta('ustream:channel_id', webpage)
  95
  96         BASE = 'http://www.ustream.tv'
  97         next_url = '/ajax/socialstream/videos/%s/1.json' % channel_id
  98         video_ids = []
  99         while next_url:
 100             reply = self._download_json(
 101                 compat_urlparse.urljoin(BASE, next_url), display_id,
 102                 note='Downloading video information (next: %d)' % (len(video_ids) + 1))
 103             video_ids.extend(re.findall(r'data-content-id="(\d.*)"', reply['data']))
 104             next_url = reply['nextUrl']
 105
 106         entries = [
 107             self.url_result('http://www.ustream.tv/recorded/' + vid, 'Ustream')
 108             for vid in video_ids]
 109         return {
 110             '_type': 'playlist',
 111             'id': channel_id,
 112             'display_id': display_id,
 113             'entries': entries,
 114         }