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 (
  11     ExtractorError,
  12     int_or_none,
  13     float_or_none,
  14 )
  15
  16
  17 class UstreamIE(InfoExtractor):
  18     _VALID_URL = r'https?://www\.ustream\.tv/(?P<type>recorded|embed|embed/recorded)/(?P<id>\d+)'
  19     IE_NAME = 'ustream'
  20     _TESTS = [{
  21         'url': 'http://www.ustream.tv/recorded/20274954',
  22         'md5': '088f151799e8f572f84eb62f17d73e5c',
  23         'info_dict': {
  24             'id': '20274954',
  25             'ext': 'flv',
  26             'uploader': 'Young Americans for Liberty',
  27             'title': 'Young Americans for Liberty February 7, 2012 2:28 AM',
  28         },
  29     }, {
  30         # From http://sportscanada.tv/canadagames/index.php/week2/figure-skating/444
  31         # Title and uploader available only from params JSON
  32         'url': 'http://www.ustream.tv/embed/recorded/59307601?ub=ff0000&lc=ff0000&oc=ffffff&uc=ffffff&v=3&wmode=direct',
  33         'md5': '5a2abf40babeac9812ed20ae12d34e10',
  34         'info_dict': {
  35             'id': '59307601',
  36             'ext': 'flv',
  37             'title': '-CG11- Canada Games Figure Skating',
  38             'uploader': 'sportscanadatv',
  39         }
  40     }]
  41
  42     def _real_extract(self, url):
  43         m = re.match(self._VALID_URL, url)
  44         video_id = m.group('id')
  45
  46         # some sites use this embed format (see: http://github.com/rg3/youtube-dl/issues/2990)
  47         if m.group('type') == 'embed/recorded':
  48             video_id = m.group('id')
  49             desktop_url = 'http://www.ustream.tv/recorded/' + video_id
  50             return self.url_result(desktop_url, 'Ustream')
  51         if m.group('type') == 'embed':
  52             video_id = m.group('id')
  53             webpage = self._download_webpage(url, video_id)
  54             desktop_video_id = self._html_search_regex(
  55                 r'ContentVideoIds=\["([^"]*?)"\]', webpage, 'desktop_video_id')
  56             desktop_url = 'http://www.ustream.tv/recorded/' + desktop_video_id
  57             return self.url_result(desktop_url, 'Ustream')
  58
  59         params = self._download_json(
  60             'https://api.ustream.tv/videos/%s.json' % video_id, video_id)
  61
  62         error = params.get('error')
  63         if error:
  64             raise ExtractorError(
  65                 '%s returned error: %s' % (self.IE_NAME, error), expected=True)
  66
  67         video = params['video']
  68
  69         formats = [{
  70             'id': format_id,
  71             'url': video_url,
  72             'ext': format_id,
  73         } for format_id, video_url in video['media_urls'].items()]
  74         self._sort_formats(formats)
  75
  76         title = video['title']
  77         description = video.get('description')
  78         timestamp = int_or_none(video.get('created_at'))
  79         duration = float_or_none(video.get('length'))
  80         filesize = float_or_none(video.get('file_size'))
  81         view_count = int_or_none(video.get('views'))
  82
  83         uploader = video.get('owner', {}).get('username')
  84         uploader_id = video.get('owner', {}).get('id')
  85
  86         thumbnails = [{
  87             'id': thumbnail_id,
  88             'url': thumbnail_url,
  89         } for thumbnail_id, thumbnail_url in video.get('thumbnail', {}).items()]
  90
  91         return {
  92             'id': video_id,
  93             'title': title,
  94             'description': description,
  95             'thumbnails': thumbnails,
  96             'timestamp': timestamp,
  97             'duration': duration,
  98             'filesize': filesize,
  99             'view_count': view_count,
 100             'uploader': uploader,
 101             'uploader_id': uploader_id,
 102             'formats': formats,
 103         }
 104
 105
 106 class UstreamChannelIE(InfoExtractor):
 107     _VALID_URL = r'https?://www\.ustream\.tv/channel/(?P<slug>.+)'
 108     IE_NAME = 'ustream:channel'
 109     _TEST = {
 110         'url': 'http://www.ustream.tv/channel/channeljapan',
 111         'info_dict': {
 112             'id': '10874166',
 113         },
 114         'playlist_mincount': 17,
 115     }
 116
 117     def _real_extract(self, url):
 118         m = re.match(self._VALID_URL, url)
 119         display_id = m.group('slug')
 120         webpage = self._download_webpage(url, display_id)
 121         channel_id = self._html_search_meta('ustream:channel_id', webpage)
 122
 123         BASE = 'http://www.ustream.tv'
 124         next_url = '/ajax/socialstream/videos/%s/1.json' % channel_id
 125         video_ids = []
 126         while next_url:
 127             reply = self._download_json(
 128                 compat_urlparse.urljoin(BASE, next_url), display_id,
 129                 note='Downloading video information (next: %d)' % (len(video_ids) + 1))
 130             video_ids.extend(re.findall(r'data-content-id="(\d.*)"', reply['data']))
 131             next_url = reply['nextUrl']
 132
 133         entries = [
 134             self.url_result('http://www.ustream.tv/recorded/' + vid, 'Ustream')
 135             for vid in video_ids]
 136         return {
 137             '_type': 'playlist',
 138             'id': channel_id,
 139             'display_id': display_id,
 140             'entries': entries,
 141         }