From 0d97ef43bec006157870fd4a5cedfac1eaebf3a9 Mon Sep 17 00:00:00 2001 From: Naglis Jonaitis Date: Thu, 26 Feb 2015 23:45:54 +0200 Subject: [PATCH] [kaltura] Add new extractor --- youtube_dl/extractor/__init__.py | 1 + youtube_dl/extractor/kaltura.py | 134 +++++++++++++++++++++++++++++++ 2 files changed, 135 insertions(+) create mode 100644 youtube_dl/extractor/kaltura.py diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index ddb9d6670..e3b2cb54f 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -227,6 +227,7 @@ from .jove import JoveIE from .jukebox import JukeboxIE from .jpopsukitv import JpopsukiIE +from .kaltura import KalturaIE from .kankan import KankanIE from .karaoketv import KaraoketvIE from .keezmovies import KeezMoviesIE diff --git a/youtube_dl/extractor/kaltura.py b/youtube_dl/extractor/kaltura.py new file mode 100644 index 000000000..2aff410c5 --- /dev/null +++ b/youtube_dl/extractor/kaltura.py @@ -0,0 +1,134 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..compat import compat_urllib_parse +from ..utils import ( + ExtractorError, + int_or_none, +) + + +class KalturaIE(InfoExtractor): + _VALID_URL = r'''(?x) + (?:kaltura:| + https?://(:?www\.)?kaltura\.com/index\.php/kwidget/(?:[^/]+/)*?wid/_ + )(?P\d+) + (?::| + /(?:[^/]+/)*?entry_id/ + )(?P[0-9a-z_]+)''' + _API_BASE = 'http://cdnapi.kaltura.com/api_v3/index.php?' + _TESTS = [ + { + 'url': 'kaltura:269692:1_1jc2y3e4', + 'md5': '3adcbdb3dcc02d647539e53f284ba171', + 'info_dict': { + 'id': '1_1jc2y3e4', + 'ext': 'mp4', + 'title': 'Track 4', + 'upload_date': '20131219', + 'uploader_id': 'mlundberg@wolfgangsvault.com', + 'description': 'The Allman Brothers Band, 12/16/1981', + 'thumbnail': 're:^https?://.*/thumbnail/.*', + 'timestamp': int, + }, + }, + { + 'url': 'http://www.kaltura.com/index.php/kwidget/cache_st/1300318621/wid/_269692/uiconf_id/3873291/entry_id/1_1jc2y3e4', + 'only_matching': True, + }, + ] + + def _kaltura_api_call(self, video_id, actions, *args, **kwargs): + params = actions[0] + if len(actions) > 1: + for i, a in enumerate(actions[1:], start=1): + for k, v in a.items(): + params['%d:%s' % (i, k)] = v + + query = compat_urllib_parse.urlencode(params) + url = self._API_BASE + query + data = self._download_json(url, video_id, *args, **kwargs) + + status = data if len(actions) == 1 else data[0] + if status.get('objectType') == 'KalturaAPIException': + raise ExtractorError( + '%s said: %s' % (self.IE_NAME, status['message'])) + + return data + + def _get_kaltura_signature(self, video_id, partner_id): + actions = [{ + 'apiVersion': '3.1', + 'expiry': 86400, + 'format': 1, + 'service': 'session', + 'action': 'startWidgetSession', + 'widgetId': '_%s' % partner_id, + }] + return self._kaltura_api_call( + video_id, actions, note='Downloading Kaltura signature')['ks'] + + def _get_video_info(self, video_id, partner_id): + signature = self._get_kaltura_signature(video_id, partner_id) + actions = [ + { + 'action': 'null', + 'apiVersion': '3.1.5', + 'clientTag': 'kdp:v3.8.5', + 'format': 1, # JSON, 2 = XML, 3 = PHP + 'service': 'multirequest', + 'ks': signature, + }, + { + 'action': 'get', + 'entryId': video_id, + 'service': 'baseentry', + 'version': '-1', + }, + { + 'action': 'getContextData', + 'contextDataParams:objectType': 'KalturaEntryContextDataParams', + 'contextDataParams:referrer': 'http://www.kaltura.com/', + 'contextDataParams:streamerType': 'http', + 'entryId': video_id, + 'service': 'baseentry', + }, + ] + return self._kaltura_api_call( + video_id, actions, note='Downloading video info JSON') + + def _real_extract(self, url): + video_id = self._match_id(url) + mobj = re.match(self._VALID_URL, url) + partner_id, entry_id = mobj.group('partner_id'), mobj.group('id') + + info, source_data = self._get_video_info(entry_id, partner_id) + + formats = [{ + 'format_id': '%(fileExt)s-%(bitrate)s' % f, + 'ext': f['fileExt'], + 'tbr': f['bitrate'], + 'fps': f.get('frameRate'), + 'filesize_approx': int_or_none(f.get('size'), invscale=1024), + 'container': f.get('containerFormat'), + 'vcodec': f.get('videoCodecId'), + 'height': f.get('height'), + 'width': f.get('width'), + 'url': '%s/flavorId/%s' % (info['dataUrl'], f['id']), + } for f in source_data['flavorAssets']] + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': info['name'], + 'formats': formats, + 'description': info.get('description'), + 'thumbnail': info.get('thumbnailUrl'), + 'duration': info.get('duration'), + 'timestamp': info.get('createdAt'), + 'uploader_id': info.get('userId'), + 'view_count': info.get('plays'), + }