X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/c512650955de0b16d37e7fa7fb29ea0985e415bb..139d14b198add5b26a11ed5dca022feaa6581d98:/youtube_dl/extractor/ustream.py?ds=sidebyside diff --git a/youtube_dl/extractor/ustream.py b/youtube_dl/extractor/ustream.py index 488b10d..c39c278 100644 --- a/youtube_dl/extractor/ustream.py +++ b/youtube_dl/extractor/ustream.py @@ -4,16 +4,16 @@ import json import re from .common import InfoExtractor -from ..utils import ( +from ..compat import ( compat_urlparse, - get_meta_content, ) +from ..utils import ExtractorError class UstreamIE(InfoExtractor): _VALID_URL = r'https?://www\.ustream\.tv/(?Precorded|embed|embed/recorded)/(?P\d+)' IE_NAME = 'ustream' - _TEST = { + _TESTS = [{ 'url': 'http://www.ustream.tv/recorded/20274954', 'md5': '088f151799e8f572f84eb62f17d73e5c', 'info_dict': { @@ -22,7 +22,18 @@ class UstreamIE(InfoExtractor): 'uploader': 'Young Americans for Liberty', 'title': 'Young Americans for Liberty February 7, 2012 2:28 AM', }, - } + }, { + # From http://sportscanada.tv/canadagames/index.php/week2/figure-skating/444 + # Title and uploader available only from params JSON + 'url': 'http://www.ustream.tv/embed/recorded/59307601?ub=ff0000&lc=ff0000&oc=ffffff&uc=ffffff&v=3&wmode=direct', + 'md5': '5a2abf40babeac9812ed20ae12d34e10', + 'info_dict': { + 'id': '59307601', + 'ext': 'flv', + 'title': '-CG11- Canada Games Figure Skating', + 'uploader': 'sportscanadatv', + } + }] def _real_extract(self, url): m = re.match(self._VALID_URL, url) @@ -41,19 +52,45 @@ class UstreamIE(InfoExtractor): desktop_url = 'http://www.ustream.tv/recorded/' + desktop_video_id return self.url_result(desktop_url, 'Ustream') - video_url = 'http://tcdn.ustream.tv/video/%s' % video_id + params = self._download_json( + 'http://cdngw.ustream.tv/rgwjson/Viewer.getVideo/' + json.dumps({ + 'brandId': 1, + 'videoId': int(video_id), + 'autoplay': False, + }), video_id) + + if 'error' in params: + raise ExtractorError(params['error']['message'], expected=True) + + video_url = params['flv'] + webpage = self._download_webpage(url, video_id) self.report_extraction(video_id) video_title = self._html_search_regex(r'data-title="(?P.+)"', - webpage, 'title') + webpage, 'title', default=None) + + if not video_title: + try: + video_title = params['moduleConfig']['meta']['title'] + except KeyError: + pass + + if not video_title: + video_title = 'Ustream video ' + video_id uploader = self._html_search_regex(r'data-content-type="channel".*?>(?P<uploader>.*?)</a>', - webpage, 'uploader', fatal=False, flags=re.DOTALL) + webpage, 'uploader', fatal=False, flags=re.DOTALL, default=None) + + if not uploader: + try: + uploader = params['moduleConfig']['meta']['userName'] + except KeyError: + uploader = None thumbnail = self._html_search_regex(r'<link rel="image_src" href="(?P<thumb>.*?)"', - webpage, 'thumbnail', fatal=False) + webpage, 'thumbnail', fatal=False) return { 'id': video_id, @@ -68,21 +105,36 @@ class UstreamIE(InfoExtractor): class UstreamChannelIE(InfoExtractor): _VALID_URL = r'https?://www\.ustream\.tv/channel/(?P<slug>.+)' IE_NAME = 'ustream:channel' + _TEST = { + 'url': 'http://www.ustream.tv/channel/channeljapan', + 'info_dict': { + 'id': '10874166', + }, + 'playlist_mincount': 17, + } def _real_extract(self, url): m = re.match(self._VALID_URL, url) - slug = m.group('slug') - webpage = self._download_webpage(url, slug) - channel_id = get_meta_content('ustream:channel_id', webpage) + display_id = m.group('slug') + webpage = self._download_webpage(url, display_id) + channel_id = self._html_search_meta('ustream:channel_id', webpage) BASE = 'http://www.ustream.tv' next_url = '/ajax/socialstream/videos/%s/1.json' % channel_id video_ids = [] while next_url: - reply = json.loads(self._download_webpage(compat_urlparse.urljoin(BASE, next_url), channel_id)) + reply = self._download_json( + compat_urlparse.urljoin(BASE, next_url), display_id, + note='Downloading video information (next: %d)' % (len(video_ids) + 1)) video_ids.extend(re.findall(r'data-content-id="(\d.*)"', reply['data'])) next_url = reply['nextUrl'] - urls = ['http://www.ustream.tv/recorded/' + vid for vid in video_ids] - url_entries = [self.url_result(eurl, 'Ustream') for eurl in urls] - return self.playlist_result(url_entries, channel_id) + entries = [ + self.url_result('http://www.ustream.tv/recorded/' + vid, 'Ustream') + for vid in video_ids] + return { + '_type': 'playlist', + 'id': channel_id, + 'display_id': display_id, + 'entries': entries, + }