1
0
mirror of https://codeberg.org/polarisfm/youtube-dl synced 2024-11-18 06:54:35 +01:00
youtube-dl/youtube_dl/extractor/kaltura.py

214 lines
7.9 KiB
Python
Raw Normal View History

2015-02-26 22:45:54 +01:00
# coding: utf-8
from __future__ import unicode_literals
import re
import base64
2015-02-26 22:45:54 +01:00
from .common import InfoExtractor
from ..compat import (
compat_urllib_parse_urlencode,
compat_urlparse,
compat_parse_qs,
)
2015-02-26 22:45:54 +01:00
from ..utils import (
2015-11-20 20:39:29 +01:00
clean_html,
2015-02-26 22:45:54 +01:00
ExtractorError,
int_or_none,
unsmuggle_url,
2015-02-26 22:45:54 +01:00
)
class KalturaIE(InfoExtractor):
_VALID_URL = r'''(?x)
2015-08-29 16:21:59 +02:00
(?:
kaltura:(?P<partner_id>\d+):(?P<id>[0-9a-z_]+)|
2015-08-29 16:21:59 +02:00
https?://
2015-11-09 15:50:43 +01:00
(:?(?:www|cdnapi(?:sec)?)\.)?kaltura\.com/
2015-08-29 16:21:59 +02:00
(?:
(?:
# flash player
index\.php/kwidget|
2015-08-29 16:23:06 +02:00
# html5 player
html5/html5lib/[^/]+/mwEmbedFrame\.php
2015-08-29 16:26:51 +02:00
)
)(?:/(?P<path>[^?]+))?(?:\?(?P<query>.*))?
2015-08-29 16:21:59 +02:00
)
'''
2015-02-26 22:45:54 +01:00
_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',
2015-12-11 23:40:12 +01:00
'title': 'Straight from the Heart',
2015-02-26 22:45:54 +01:00
'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,
},
{
'url': 'https://cdnapisec.kaltura.com/index.php/kwidget/wid/_557781/uiconf_id/22845202/entry_id/1_plr1syf3',
'only_matching': True,
},
2015-08-29 08:12:15 +02:00
{
'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,
}
2015-02-26 22:45:54 +01:00
]
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)
2015-02-26 22:45:54 +01:00
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',
},
{
2015-12-11 23:40:12 +01:00
'action': 'getbyentryid',
2015-02-26 22:45:54 +01:00
'entryId': video_id,
2015-12-11 23:40:12 +01:00
'service': 'flavorAsset',
2015-02-26 22:45:54 +01:00
},
]
return self._kaltura_api_call(
video_id, actions, note='Downloading video info JSON')
def _real_extract(self, url):
url, smuggled_data = unsmuggle_url(url, {})
2015-02-26 22:45:54 +01:00
mobj = re.match(self._VALID_URL, url)
partner_id, entry_id = mobj.group('partner_id', 'id')
2016-03-12 21:01:10 +01:00
ks = None
if partner_id and entry_id:
info, flavor_assets = self._get_video_info(entry_id, partner_id)
else:
path, query = mobj.group('path', 'query')
if not path and not query:
raise ExtractorError('Invalid URL', expected=True)
params = {}
if query:
params = compat_parse_qs(query)
if path:
splitted_path = path.split('/')
2016-03-12 18:43:45 +01:00
params.update(dict((zip(splitted_path[::2], [[v] for v in splitted_path[1::2]]))))
if 'wid' in params:
partner_id = params['wid'][0][1:]
elif 'p' in params:
partner_id = params['p'][0]
else:
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)
elif 'uiconf_id' in params and 'flashvars[referenceId]' in params:
reference_id = params['flashvars[referenceId]'][0]
webpage = self._download_webpage(url, reference_id)
entry_data = self._parse_json(self._search_regex(
r'window\.kalturaIframePackageData\s*=\s*({.*});',
webpage, 'kalturaIframePackageData'),
reference_id)['entryResult']
info, flavor_assets = entry_data['meta'], entry_data['contextData']['flavorAssets']
entry_id = info['id']
else:
raise ExtractorError('Invalid URL', expected=True)
2016-03-12 21:01:10 +01:00
ks = params.get('flashvars[ks]', [None])[0]
2015-02-26 22:45:54 +01:00
source_url = smuggled_data.get('source_url')
if source_url:
referrer = base64.b64encode(
'://'.join(compat_urlparse.urlparse(source_url)[:2])
.encode('utf-8')).decode('utf-8')
else:
referrer = None
2016-03-12 21:01:10 +01:00
def sign_url(unsigned_url):
if ks:
unsigned_url += '/ks/%s' % ks
if referrer:
unsigned_url += '?referrer=%s' % referrer
return unsigned_url
formats = []
2015-12-11 23:40:12 +01:00
for f in flavor_assets:
# Continue if asset is not ready
if f['status'] != 2:
continue
2016-03-12 21:01:10 +01:00
video_url = sign_url('%s/flavorId/%s' % (info['dataUrl'], f['id']))
formats.append({
'format_id': '%(fileExt)s-%(bitrate)s' % f,
2015-11-20 20:38:08 +01:00
'ext': f.get('fileExt'),
'tbr': int_or_none(f['bitrate']),
'fps': int_or_none(f.get('frameRate')),
'filesize_approx': int_or_none(f.get('size'), invscale=1024),
'container': f.get('containerFormat'),
'vcodec': f.get('videoCodecId'),
2015-11-20 20:38:08 +01:00
'height': int_or_none(f.get('height')),
'width': int_or_none(f.get('width')),
'url': video_url,
})
2016-03-12 21:01:10 +01:00
m3u8_url = sign_url(info['dataUrl'].replace('format/url', 'format/applehttp'))
formats.extend(self._extract_m3u8_formats(
m3u8_url, entry_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
2015-12-11 23:40:12 +01:00
self._check_formats(formats, entry_id)
2015-02-26 22:45:54 +01:00
self._sort_formats(formats)
return {
2015-08-29 16:21:59 +02:00
'id': entry_id,
2015-02-26 22:45:54 +01:00
'title': info['name'],
'formats': formats,
2015-11-20 20:39:29 +01:00
'description': clean_html(info.get('description')),
2015-02-26 22:45:54 +01:00
'thumbnail': info.get('thumbnailUrl'),
'duration': info.get('duration'),
'timestamp': info.get('createdAt'),
'uploader_id': info.get('userId'),
'view_count': info.get('plays'),
}