{
'url': 'https://cdnapisec.kaltura.com/html5/html5lib/v2.30.2/mwEmbedFrame.php/p/1337/uiconf_id/20540612/entry_id/1_sf5ovm7u?wid=_243342',
'only_matching': True,
+ },
+ {
+ # video with subtitles
+ 'url': 'kaltura:111032:1_cw786r8q',
+ 'only_matching': True,
+ },
+ {
+ # video with ttml subtitles (no fileExt)
+ 'url': 'kaltura:1926081:0_l5ye1133',
+ 'info_dict': {
+ 'id': '0_l5ye1133',
+ 'ext': 'mp4',
+ 'title': 'What Can You Do With Python?',
+ 'upload_date': '20160221',
+ 'uploader_id': 'stork',
+ 'thumbnail': 're:^https?://.*/thumbnail/.*',
+ 'timestamp': int,
+ 'subtitles': {
+ 'en': [{
+ 'ext': 'ttml',
+ }],
+ },
+ },
+ 'params': {
+ 'skip_download': True,
+ },
}
]
if mobj:
embed_info = mobj.groupdict()
url = 'kaltura:%(partner_id)s:%(id)s' % embed_info
+ escaped_pid = re.escape(embed_info['partner_id'])
service_url = re.search(
- '<script[^>]+src=(?:["\'])((?:https?:)?//.+?)/p/%(partner_id)s/sp/%(partner_id)s00/embedIframeJs' % embed_info,
+ r'<script[^>]+src=["\']((?:https?:)?//.+?)/p/%s/sp/%s00/embedIframeJs' % (escaped_pid, escaped_pid),
webpage)
if service_url:
url = smuggle_url(url, {'service_url': service_url.group(1)})
return url
- def _kaltura_api_call(self, video_id, actions, *args, **kwargs):
+ def _kaltura_api_call(self, video_id, actions, service_url=None, *args, **kwargs):
params = actions[0]
if len(actions) > 1:
for i, a in enumerate(actions[1:], start=1):
params['%d:%s' % (i, k)] = v
data = self._download_json(
- self._SERVICE_URL + self._SERVICE_BASE,
+ (service_url or self._SERVICE_URL) + self._SERVICE_BASE,
video_id, query=params, *args, **kwargs)
status = data if len(actions) == 1 else data[0]
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)
+ def _get_video_info(self, video_id, partner_id, service_url=None):
actions = [
{
'action': 'null',
'clientTag': 'kdp:v3.8.5',
'format': 1, # JSON, 2 = XML, 3 = PHP
'service': 'multirequest',
- 'ks': signature,
+ },
+ {
+ 'expiry': 86400,
+ 'service': 'session',
+ 'action': 'startWidgetSession',
+ 'widgetId': '_%s' % partner_id,
},
{
'action': 'get',
'entryId': video_id,
'service': 'baseentry',
- 'version': '-1',
+ 'ks': '{1:result:ks}',
},
{
'action': 'getbyentryid',
'entryId': video_id,
'service': 'flavorAsset',
+ 'ks': '{1:result:ks}',
+ },
+ {
+ 'action': 'list',
+ 'filter:entryIdEqual': video_id,
+ 'service': 'caption_captionasset',
+ 'ks': '{1:result:ks}',
},
]
return self._kaltura_api_call(
- video_id, actions, note='Downloading video info JSON')
+ video_id, actions, service_url, note='Downloading video info JSON')
def _real_extract(self, url):
url, smuggled_data = unsmuggle_url(url, {})
- service_url = smuggled_data.get('service_url')
- if service_url:
- self._SERVICE_URL = service_url
mobj = re.match(self._VALID_URL, url)
partner_id, entry_id = mobj.group('partner_id', 'id')
ks = None
+ captions = None
if partner_id and entry_id:
- info, flavor_assets = self._get_video_info(entry_id, partner_id)
+ _, info, flavor_assets, captions = self._get_video_info(entry_id, partner_id, smuggled_data.get('service_url'))
else:
path, query = mobj.group('path', 'query')
if not path and not query:
raise ExtractorError('Invalid URL', expected=True)
if 'entry_id' in params:
entry_id = params['entry_id'][0]
- info, flavor_assets = self._get_video_info(entry_id, partner_id)
+ _, info, flavor_assets, captions = self._get_video_info(entry_id, partner_id)
elif 'uiconf_id' in params and 'flashvars[referenceId]' in params:
reference_id = params['flashvars[referenceId]'][0]
webpage = self._download_webpage(url, reference_id)
reference_id)['entryResult']
info, flavor_assets = entry_data['meta'], entry_data['contextData']['flavorAssets']
entry_id = info['id']
+ # Unfortunately, data returned in kalturaIframePackageData lacks
+ # captions so we will try requesting the complete data using
+ # regular approach since we now know the entry_id
+ try:
+ _, info, flavor_assets, captions = self._get_video_info(
+ entry_id, partner_id)
+ except ExtractorError:
+ # Regular scenario failed but we already have everything
+ # extracted apart from captions and can process at least
+ # with this
+ pass
else:
raise ExtractorError('Invalid URL', expected=True)
ks = params.get('flashvars[ks]', [None])[0]
formats = []
for f in flavor_assets:
# Continue if asset is not ready
- if f['status'] != 2:
+ if f.get('status') != 2:
continue
video_url = sign_url(
'%s/flavorId/%s' % (data_url, f['id']))
m3u8_url, entry_id, 'mp4', 'm3u8_native',
m3u8_id='hls', fatal=False))
- self._check_formats(formats, entry_id)
self._sort_formats(formats)
+ subtitles = {}
+ if captions:
+ for caption in captions.get('objects', []):
+ # Continue if caption is not ready
+ if f.get('status') != 2:
+ continue
+ subtitles.setdefault(caption.get('languageCode') or caption.get('language'), []).append({
+ 'url': '%s/api_v3/service/caption_captionasset/action/serve/captionAssetId/%s' % (self._SERVICE_URL, caption['id']),
+ 'ext': caption.get('fileExt', 'ttml'),
+ })
+
return {
'id': entry_id,
'title': info['name'],
'formats': formats,
+ 'subtitles': subtitles,
'description': clean_html(info.get('description')),
'thumbnail': info.get('thumbnailUrl'),
'duration': info.get('duration'),