_ 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             'title': 'Young Americans for Liberty February 7, 2012 2:28 AM',
  26             'description': 'Young Americans for Liberty February 7, 2012 2:28 AM',
  27             'timestamp': 1328577035,
  28             'upload_date': '20120207',
  29             'uploader': 'yaliberty',
  30             'uploader_id': '6780869',
  31         },
  32     }, {
  33         # From http://sportscanada.tv/canadagames/index.php/week2/figure-skating/444
  34         # Title and uploader available only from params JSON
  35         'url': 'http://www.ustream.tv/embed/recorded/59307601?ub=ff0000&lc=ff0000&oc=ffffff&uc=ffffff&v=3&wmode=direct',
  36         'md5': '5a2abf40babeac9812ed20ae12d34e10',
  37         'info_dict': {
  38             'id': '59307601',
  39             'ext': 'flv',
  40             'title': '-CG11- Canada Games Figure Skating',
  41             'uploader': 'sportscanadatv',
  42         },
  43         'skip': 'This Pro Broadcaster has chosen to remove this video from the ustream.tv site.',
  44     }, {
  45         'url': 'http://www.ustream.tv/embed/10299409',
  46         'info_dict': {
  47             'id': '10299409',
  48         },
  49         'playlist_count': 3,
  50     }]
  51
  52     def _real_extract(self, url):
  53         m = re.match(self._VALID_URL, url)
  54         video_id = m.group('id')
  55
  56         # some sites use this embed format (see: https://github.com/rg3/youtube-dl/issues/2990)
  57         if m.group('type') == 'embed/recorded':
  58             video_id = m.group('id')
  59             desktop_url = 'http://www.ustream.tv/recorded/' + video_id
  60             return self.url_result(desktop_url, 'Ustream')
  61         if m.group('type') == 'embed':
  62             video_id = m.group('id')
  63             webpage = self._download_webpage(url, video_id)
  64             content_video_ids = self._parse_json(self._search_regex(
  65                 r'ustream\.vars\.offAirContentVideoIds=([^;]+);', webpage,
  66                 'content video IDs'), video_id)
  67             return self.playlist_result(
  68                 map(lambda u: self.url_result('http://www.ustream.tv/recorded/' + u, 'Ustream'), content_video_ids),
  69                 video_id)
  70
  71         params = self._download_json(
  72             'https://api.ustream.tv/videos/%s.json' % video_id, video_id)
  73
  74         error = params.get('error')
  75         if error:
  76             raise ExtractorError(
  77                 '%s returned error: %s' % (self.IE_NAME, error), expected=True)
  78
  79         video = params['video']
  80
  81         title = video['title']
  82         filesize = float_or_none(video.get('file_size'))
  83
  84         formats = [{
  85             'id': video_id,
  86             'url': video_url,
  87             'ext': format_id,
  88             'filesize': filesize,
  89         } for format_id, video_url in video['media_urls'].items()]
  90         self._sort_formats(formats)
  91
  92         description = video.get('description')
  93         timestamp = int_or_none(video.get('created_at'))
  94         duration = float_or_none(video.get('length'))
  95         view_count = int_or_none(video.get('views'))
  96
  97         uploader = video.get('owner', {}).get('username')
  98         uploader_id = video.get('owner', {}).get('id')
  99
 100         thumbnails = [{
 101             'id': thumbnail_id,
 102             'url': thumbnail_url,
 103         } for thumbnail_id, thumbnail_url in video.get('thumbnail', {}).items()]
 104
 105         return {
 106             'id': video_id,
 107             'title': title,
 108             'description': description,
 109             'thumbnails': thumbnails,
 110             'timestamp': timestamp,
 111             'duration': duration,
 112             'view_count': view_count,
 113             'uploader': uploader,
 114             'uploader_id': uploader_id,
 115             'formats': formats,
 116         }
 117
 118
 119 class UstreamChannelIE(InfoExtractor):
 120     _VALID_URL = r'https?://www\.ustream\.tv/channel/(?P<slug>.+)'
 121     IE_NAME = 'ustream:channel'
 122     _TEST = {
 123         'url': 'http://www.ustream.tv/channel/channeljapan',
 124         'info_dict': {
 125             'id': '10874166',
 126         },
 127         'playlist_mincount': 17,
 128     }
 129
 130     def _real_extract(self, url):
 131         m = re.match(self._VALID_URL, url)
 132         display_id = m.group('slug')
 133         webpage = self._download_webpage(url, display_id)
 134         channel_id = self._html_search_meta('ustream:channel_id', webpage)
 135
 136         BASE = 'http://www.ustream.tv'
 137         next_url = '/ajax/socialstream/videos/%s/1.json' % channel_id
 138         video_ids = []
 139         while next_url:
 140             reply = self._download_json(
 141                 compat_urlparse.urljoin(BASE, next_url), display_id,
 142                 note='Downloading video information (next: %d)' % (len(video_ids) + 1))
 143             video_ids.extend(re.findall(r'data-content-id="(\d.*)"', reply['data']))
 144             next_url = reply['nextUrl']
 145
 146         entries = [
 147             self.url_result('http://www.ustream.tv/recorded/' + vid, 'Ustream')
 148             for vid in video_ids]
 149         return {
 150             '_type': 'playlist',
 151             'id': channel_id,
 152             'display_id': display_id,
 153             'entries': entries,
 154         }