Raphaël G. Git Repositories - youtubedl/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     _TESTS = [{
  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         # From http://sportscanada.tv/canadagames/index.php/week2/figure-skating/444
  27         # Title and uploader available only from params JSON
  28         'url': 'http://www.ustream.tv/embed/recorded/59307601?ub=ff0000&lc=ff0000&oc=ffffff&uc=ffffff&v=3&wmode=direct',
  29         'md5': '5a2abf40babeac9812ed20ae12d34e10',
  30         'info_dict': {
  31             'id': '59307601',
  32             'ext': 'flv',
  33             'title': '-CG11- Canada Games Figure Skating',
  34             'uploader': 'sportscanadatv',
  35         }
  36     }]
  37
  38     def _real_extract(self, url):
  39         m = re.match(self._VALID_URL, url)
  40         video_id = m.group('videoID')
  41
  42         # some sites use this embed format (see: http://github.com/rg3/youtube-dl/issues/2990)
  43         if m.group('type') == 'embed/recorded':
  44             video_id = m.group('videoID')
  45             desktop_url = 'http://www.ustream.tv/recorded/' + video_id
  46             return self.url_result(desktop_url, 'Ustream')
  47         if m.group('type') == 'embed':
  48             video_id = m.group('videoID')
  49             webpage = self._download_webpage(url, video_id)
  50             desktop_video_id = self._html_search_regex(
  51                 r'ContentVideoIds=\["([^"]*?)"\]', webpage, 'desktop_video_id')
  52             desktop_url = 'http://www.ustream.tv/recorded/' + desktop_video_id
  53             return self.url_result(desktop_url, 'Ustream')
  54
  55         params = self._download_json(
  56             'http://cdngw.ustream.tv/rgwjson/Viewer.getVideo/' + json.dumps({
  57                 'brandId': 1,
  58                 'videoId': int(video_id),
  59                 'autoplay': False,
  60             }), video_id)
  61
  62         if 'error' in params:
  63             raise ExtractorError(params['error']['message'], expected=True)
  64
  65         video_url = params['flv']
  66
  67         webpage = self._download_webpage(url, video_id)
  68
  69         self.report_extraction(video_id)
  70
  71         video_title = self._html_search_regex(r'data-title="(?P<title>.+)"',
  72                                               webpage, 'title', default=None)
  73
  74         if not video_title:
  75             try:
  76                 video_title = params['moduleConfig']['meta']['title']
  77             except KeyError:
  78                 pass
  79
  80         if not video_title:
  81             video_title = 'Ustream video ' + video_id
  82
  83         uploader = self._html_search_regex(r'data-content-type="channel".*?>(?P<uploader>.*?)</a>',
  84                                            webpage, 'uploader', fatal=False, flags=re.DOTALL, default=None)
  85
  86         if not uploader:
  87             try:
  88                 uploader = params['moduleConfig']['meta']['userName']
  89             except KeyError:
  90                 uploader = None
  91
  92         thumbnail = self._html_search_regex(r'<link rel="image_src" href="(?P<thumb>.*?)"',
  93                                             webpage, 'thumbnail', fatal=False)
  94
  95         return {
  96             'id': video_id,
  97             'url': video_url,
  98             'ext': 'flv',
  99             'title': video_title,
 100             'uploader': uploader,
 101             'thumbnail': thumbnail,
 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         }