_ Git - youtube-dl/blob - youtube_dl/extractor/ustream.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_urlparse,
   8 )
   9 from ..utils import (
  10     ExtractorError,
  11     int_or_none,
  12     float_or_none,
  13 )
  14
  15
  16 class UstreamIE(InfoExtractor):
  17     _VALID_URL = r'https?://www\.ustream\.tv/(?P<type>recorded|embed|embed/recorded)/(?P<id>\d+)'
  18     IE_NAME = 'ustream'
  19     _TESTS = [{
  20         'url': 'http://www.ustream.tv/recorded/20274954',
  21         'md5': '088f151799e8f572f84eb62f17d73e5c',
  22         'info_dict': {
  23             'id': '20274954',
  24             'ext': 'flv',
  25             'uploader': 'Young Americans for Liberty',
  26             'title': 'Young Americans for Liberty February 7, 2012 2:28 AM',
  27         },
  28     }, {
  29         # From http://sportscanada.tv/canadagames/index.php/week2/figure-skating/444
  30         # Title and uploader available only from params JSON
  31         'url': 'http://www.ustream.tv/embed/recorded/59307601?ub=ff0000&lc=ff0000&oc=ffffff&uc=ffffff&v=3&wmode=direct',
  32         'md5': '5a2abf40babeac9812ed20ae12d34e10',
  33         'info_dict': {
  34             'id': '59307601',
  35             'ext': 'flv',
  36             'title': '-CG11- Canada Games Figure Skating',
  37             'uploader': 'sportscanadatv',
  38         }
  39     }]
  40
  41     def _real_extract(self, url):
  42         m = re.match(self._VALID_URL, url)
  43         video_id = m.group('id')
  44
  45         # some sites use this embed format (see: http://github.com/rg3/youtube-dl/issues/2990)
  46         if m.group('type') == 'embed/recorded':
  47             video_id = m.group('id')
  48             desktop_url = 'http://www.ustream.tv/recorded/' + video_id
  49             return self.url_result(desktop_url, 'Ustream')
  50         if m.group('type') == 'embed':
  51             video_id = m.group('id')
  52             webpage = self._download_webpage(url, video_id)
  53             desktop_video_id = self._html_search_regex(
  54                 r'ContentVideoIds=\["([^"]*?)"\]', webpage, 'desktop_video_id')
  55             desktop_url = 'http://www.ustream.tv/recorded/' + desktop_video_id
  56             return self.url_result(desktop_url, 'Ustream')
  57
  58         params = self._download_json(
  59             'https://api.ustream.tv/videos/%s.json' % video_id, video_id)
  60
  61         error = params.get('error')
  62         if error:
  63             raise ExtractorError(
  64                 '%s returned error: %s' % (self.IE_NAME, error), expected=True)
  65
  66         video = params['video']
  67
  68         formats = [{
  69             'id': format_id,
  70             'url': video_url,
  71             'ext': format_id,
  72         } for format_id, video_url in video['media_urls'].items()]
  73         self._sort_formats(formats)
  74
  75         title = video['title']
  76         description = video.get('description')
  77         timestamp = int_or_none(video.get('created_at'))
  78         duration = float_or_none(video.get('length'))
  79         filesize = float_or_none(video.get('file_size'))
  80         view_count = int_or_none(video.get('views'))
  81
  82         uploader = video.get('owner', {}).get('username')
  83         uploader_id = video.get('owner', {}).get('id')
  84
  85         thumbnails = [{
  86             'id': thumbnail_id,
  87             'url': thumbnail_url,
  88         } for thumbnail_id, thumbnail_url in video.get('thumbnail', {}).items()]
  89
  90         return {
  91             'id': video_id,
  92             'title': title,
  93             'description': description,
  94             'thumbnails': thumbnails,
  95             'timestamp': timestamp,
  96             'duration': duration,
  97             'filesize': filesize,
  98             'view_count': view_count,
  99             'uploader': uploader,
 100             'uploader_id': uploader_id,
 101             'formats': formats,
 102         }
 103
 104
 105 class UstreamChannelIE(InfoExtractor):
 106     _VALID_URL = r'https?://www\.ustream\.tv/channel/(?P<slug>.+)'
 107     IE_NAME = 'ustream:channel'
 108     _TEST = {
 109         'url': 'http://www.ustream.tv/channel/channeljapan',
 110         'info_dict': {
 111             'id': '10874166',
 112         },
 113         'playlist_mincount': 17,
 114     }
 115
 116     def _real_extract(self, url):
 117         m = re.match(self._VALID_URL, url)
 118         display_id = m.group('slug')
 119         webpage = self._download_webpage(url, display_id)
 120         channel_id = self._html_search_meta('ustream:channel_id', webpage)
 121
 122         BASE = 'http://www.ustream.tv'
 123         next_url = '/ajax/socialstream/videos/%s/1.json' % channel_id
 124         video_ids = []
 125         while next_url:
 126             reply = self._download_json(
 127                 compat_urlparse.urljoin(BASE, next_url), display_id,
 128                 note='Downloading video information (next: %d)' % (len(video_ids) + 1))
 129             video_ids.extend(re.findall(r'data-content-id="(\d.*)"', reply['data']))
 130             next_url = reply['nextUrl']
 131
 132         entries = [
 133             self.url_result('http://www.ustream.tv/recorded/' + vid, 'Ustream')
 134             for vid in video_ids]
 135         return {
 136             '_type': 'playlist',
 137             'id': channel_id,
 138             'display_id': display_id,
 139             'entries': entries,
 140         }