aboutsummaryrefslogtreecommitdiffstats
path: root/hypervideo_dl/extractor/kaltura.py
diff options
context:
space:
mode:
Diffstat (limited to 'hypervideo_dl/extractor/kaltura.py')
-rw-r--r--hypervideo_dl/extractor/kaltura.py265
1 files changed, 215 insertions, 50 deletions
diff --git a/hypervideo_dl/extractor/kaltura.py b/hypervideo_dl/extractor/kaltura.py
index f6dfc9c..95e2dee 100644
--- a/hypervideo_dl/extractor/kaltura.py
+++ b/hypervideo_dl/extractor/kaltura.py
@@ -1,8 +1,6 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-import re
import base64
+import json
+import re
from .common import InfoExtractor
from ..compat import (
@@ -16,13 +14,15 @@ from ..utils import (
int_or_none,
unsmuggle_url,
smuggle_url,
+ traverse_obj,
+ remove_start
)
class KalturaIE(InfoExtractor):
_VALID_URL = r'''(?x)
(?:
- kaltura:(?P<partner_id>\d+):(?P<id>[0-9a-z_]+)|
+ kaltura:(?P<partner_id>\w+):(?P<id>\w+)(?::(?P<player_type>\w+))?|
https?://
(:?(?:www|cdnapi(?:sec)?)\.)?kaltura\.com(?::\d+)?/
(?:
@@ -36,7 +36,7 @@ class KalturaIE(InfoExtractor):
)
'''
_SERVICE_URL = 'http://cdnapi.kaltura.com'
- _SERVICE_BASE = '/api_v3/index.php'
+ _SERVICE_BASE = '/api_v3/service/multirequest'
# See https://github.com/kaltura/server/blob/master/plugins/content/caption/base/lib/model/enums/CaptionType.php
_CAPTION_TYPES = {
1: 'srt',
@@ -57,6 +57,7 @@ class KalturaIE(InfoExtractor):
'thumbnail': 're:^https?://.*/thumbnail/.*',
'timestamp': int,
},
+ 'skip': 'The access to this service is forbidden since the specified partner is blocked'
},
{
'url': 'http://www.kaltura.com/index.php/kwidget/cache_st/1300318621/wid/_269692/uiconf_id/3873291/entry_id/1_1jc2y3e4',
@@ -109,16 +110,85 @@ class KalturaIE(InfoExtractor):
# unavailable source format
'url': 'kaltura:513551:1_66x4rg7o',
'only_matching': True,
+ },
+ {
+ # html5lib URL using kwidget player
+ 'url': 'https://cdnapisec.kaltura.com/html5/html5lib/v2.46/mwEmbedFrame.php/p/691292/uiconf_id/20499062/entry_id/0_c076mna6?wid=_691292&iframeembed=true&playerId=kaltura_player_1420508608&entry_id=0_c076mna6&flashvars%5BakamaiHD.loadingPolicy%5D=preInitialize&flashvars%5BakamaiHD.asyncInit%5D=true&flashvars%5BstreamerType%5D=hdnetwork',
+ 'info_dict': {
+ 'id': '0_c076mna6',
+ 'ext': 'mp4',
+ 'title': 'md5:4883e7acbcbf42583a2dddc97dee4855',
+ 'duration': 3608,
+ 'uploader_id': 'commons@swinburne.edu.au',
+ 'timestamp': 1408086874,
+ 'view_count': int,
+ 'upload_date': '20140815',
+ 'thumbnail': 'http://cfvod.kaltura.com/p/691292/sp/69129200/thumbnail/entry_id/0_c076mna6/version/100022',
+ }
+ },
+ {
+ # html5lib playlist URL using kwidget player
+ 'url': 'https://cdnapisec.kaltura.com/html5/html5lib/v2.89/mwEmbedFrame.php/p/2019031/uiconf_id/40436601?wid=1_4j3m32cv&iframeembed=true&playerId=kaltura_player_&flashvars[playlistAPI.kpl0Id]=1_jovey5nu&flashvars[ks]=&&flashvars[imageDefaultDuration]=30&flashvars[localizationCode]=en&flashvars[leadWithHTML5]=true&flashvars[forceMobileHTML5]=true&flashvars[nextPrevBtn.plugin]=true&flashvars[hotspots.plugin]=true&flashvars[sideBarContainer.plugin]=true&flashvars[sideBarContainer.position]=left&flashvars[sideBarContainer.clickToClose]=true&flashvars[chapters.plugin]=true&flashvars[chapters.layout]=vertical&flashvars[chapters.thumbnailRotator]=false&flashvars[streamSelector.plugin]=true&flashvars[EmbedPlayer.SpinnerTarget]=videoHolder&flashvars[dualScreen.plugin]=true&flashvars[playlistAPI.playlistUrl]=https://canvasgatechtest.kaf.kaltura.com/playlist/details/{playlistAPI.kpl0Id}/categoryid/126428551',
+ 'info_dict': {
+ 'id': '1_jovey5nu',
+ 'title': '00-00 Introduction'
+ },
+ 'playlist': [
+ {
+ 'info_dict': {
+ 'id': '1_b1y5hlvx',
+ 'ext': 'mp4',
+ 'title': 'CS7646_00-00 Introductio_Introduction',
+ 'duration': 91,
+ 'thumbnail': 'http://cfvod.kaltura.com/p/2019031/sp/201903100/thumbnail/entry_id/1_b1y5hlvx/version/100001',
+ 'view_count': int,
+ 'timestamp': 1533154447,
+ 'upload_date': '20180801',
+ 'uploader_id': 'djoyner3',
+ }
+ }, {
+ 'info_dict': {
+ 'id': '1_jfb7mdpn',
+ 'ext': 'mp4',
+ 'title': 'CS7646_00-00 Introductio_Three parts to the course',
+ 'duration': 63,
+ 'thumbnail': 'http://cfvod.kaltura.com/p/2019031/sp/201903100/thumbnail/entry_id/1_jfb7mdpn/version/100001',
+ 'view_count': int,
+ 'timestamp': 1533154489,
+ 'upload_date': '20180801',
+ 'uploader_id': 'djoyner3',
+ }
+ }, {
+ 'info_dict': {
+ 'id': '1_8xflxdp7',
+ 'ext': 'mp4',
+ 'title': 'CS7646_00-00 Introductio_Textbooks',
+ 'duration': 37,
+ 'thumbnail': 'http://cfvod.kaltura.com/p/2019031/sp/201903100/thumbnail/entry_id/1_8xflxdp7/version/100001',
+ 'view_count': int,
+ 'timestamp': 1533154512,
+ 'upload_date': '20180801',
+ 'uploader_id': 'djoyner3',
+ }
+ }, {
+ 'info_dict': {
+ 'id': '1_3hqew8kn',
+ 'ext': 'mp4',
+ 'title': 'CS7646_00-00 Introductio_Prerequisites',
+ 'duration': 49,
+ 'thumbnail': 'http://cfvod.kaltura.com/p/2019031/sp/201903100/thumbnail/entry_id/1_3hqew8kn/version/100001',
+ 'view_count': int,
+ 'timestamp': 1533154536,
+ 'upload_date': '20180801',
+ 'uploader_id': 'djoyner3',
+ }
+ }
+ ]
}
]
- @staticmethod
- def _extract_url(webpage):
- urls = KalturaIE._extract_urls(webpage)
- return urls[0] if urls else None
-
- @staticmethod
- def _extract_urls(webpage):
+ @classmethod
+ def _extract_embed_urls(cls, url, webpage):
# Embed codes: https://knowledge.kaltura.com/embedding-kaltura-media-players-your-site
finditer = (
list(re.finditer(
@@ -160,63 +230,80 @@ class KalturaIE(InfoExtractor):
for k, v in embed_info.items():
if v:
embed_info[k] = v.strip()
- url = 'kaltura:%(partner_id)s:%(id)s' % embed_info
+ embed_url = 'kaltura:%(partner_id)s:%(id)s' % embed_info
escaped_pid = re.escape(embed_info['partner_id'])
service_mobj = re.search(
r'<script[^>]+src=(["\'])(?P<id>(?:https?:)?//(?:(?!\1).)+)/p/%s/sp/%s00/embedIframeJs' % (escaped_pid, escaped_pid),
webpage)
if service_mobj:
- url = smuggle_url(url, {'service_url': service_mobj.group('id')})
- urls.append(url)
+ embed_url = smuggle_url(embed_url, {'service_url': service_mobj.group('id')})
+ urls.append(embed_url)
return urls
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):
- for k, v in a.items():
- params['%d:%s' % (i, k)] = v
+ params.update({i: a for i, a in enumerate(actions[1:], start=1)})
data = self._download_json(
(service_url or self._SERVICE_URL) + self._SERVICE_BASE,
- video_id, query=params, *args, **kwargs)
+ video_id, data=json.dumps(params).encode('utf-8'),
+ headers={
+ 'Content-Type': 'application/json',
+ 'Accept-Encoding': 'gzip, deflate, br',
+ }, *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']))
+ for idx, status in enumerate(data):
+ if not isinstance(status, dict):
+ continue
+ if status.get('objectType') == 'KalturaAPIException':
+ raise ExtractorError(
+ '%s said: %s (%d)' % (self.IE_NAME, status['message'], idx))
+
+ data[1] = traverse_obj(data, (1, 'objects', 0))
return data
- def _get_video_info(self, video_id, partner_id, service_url=None):
+ def _get_video_info(self, video_id, partner_id, service_url=None, player_type='html5'):
+ assert player_type in ('html5', 'kwidget')
+ if player_type == 'kwidget':
+ return self._get_video_info_kwidget(video_id, partner_id, service_url)
+
+ return self._get_video_info_html5(video_id, partner_id, service_url)
+
+ def _get_video_info_html5(self, video_id, partner_id, service_url=None):
actions = [
{
- 'action': 'null',
- 'apiVersion': '3.1.5',
- 'clientTag': 'kdp:v3.8.5',
+ 'apiVersion': '3.3.0',
+ 'clientTag': 'html5:v3.1.0',
'format': 1, # JSON, 2 = XML, 3 = PHP
- 'service': 'multirequest',
+ 'ks': '',
+ 'partnerId': partner_id,
},
{
'expiry': 86400,
'service': 'session',
'action': 'startWidgetSession',
- 'widgetId': '_%s' % partner_id,
+ 'widgetId': self._build_widget_id(partner_id),
},
+ # info
{
- 'action': 'get',
- 'entryId': video_id,
+ 'action': 'list',
+ 'filter': {'redirectFromEntryId': video_id},
'service': 'baseentry',
'ks': '{1:result:ks}',
- 'responseProfile:fields': 'createdAt,dataUrl,duration,name,plays,thumbnailUrl,userId',
- 'responseProfile:type': 1,
+ 'responseProfile': {
+ 'type': 1,
+ 'fields': 'createdAt,dataUrl,duration,name,plays,thumbnailUrl,userId',
+ },
},
+ # flavor_assets
{
'action': 'getbyentryid',
'entryId': video_id,
'service': 'flavorAsset',
'ks': '{1:result:ks}',
},
+ # captions
{
'action': 'list',
'filter:entryIdEqual': video_id,
@@ -225,17 +312,85 @@ class KalturaIE(InfoExtractor):
},
]
return self._kaltura_api_call(
- video_id, actions, service_url, note='Downloading video info JSON')
+ video_id, actions, service_url, note='Downloading video info JSON (Kaltura html5 player)')
+
+ def _get_video_info_kwidget(self, video_id, partner_id, service_url=None):
+ actions = [
+ {
+ 'service': 'multirequest',
+ 'apiVersion': '3.1',
+ 'expiry': 86400,
+ 'clientTag': 'kwidget:v2.89',
+ 'format': 1, # JSON, 2 = XML, 3 = PHP
+ 'ignoreNull': 1,
+ 'action': 'null',
+ },
+ # header
+ {
+ 'expiry': 86400,
+ 'service': 'session',
+ 'action': 'startWidgetSession',
+ 'widgetId': self._build_widget_id(partner_id),
+ },
+ # (empty)
+ {
+ 'expiry': 86400,
+ 'service': 'session',
+ 'action': 'startwidgetsession',
+ 'widgetId': self._build_widget_id(partner_id),
+ 'format': 9,
+ 'apiVersion': '3.1',
+ 'clientTag': 'kwidget:v2.89',
+ 'ignoreNull': 1,
+ 'ks': '{1:result:ks}'
+ },
+ # info
+ {
+ 'action': 'list',
+ 'filter': {'redirectFromEntryId': video_id},
+ 'service': 'baseentry',
+ 'ks': '{1:result:ks}',
+ 'responseProfile': {
+ 'type': 1,
+ 'fields': 'createdAt,dataUrl,duration,name,plays,thumbnailUrl,userId',
+ },
+ },
+ # flavor_assets
+ {
+ 'action': 'getbyentryid',
+ 'entryId': video_id,
+ 'service': 'flavorAsset',
+ 'ks': '{1:result:ks}',
+ },
+ # captions
+ {
+ 'action': 'list',
+ 'filter:entryIdEqual': video_id,
+ 'service': 'caption_captionasset',
+ 'ks': '{1:result:ks}',
+ },
+ ]
+ # second object (representing the second start widget session) is None
+ header, _, _info, flavor_assets, captions = self._kaltura_api_call(
+ video_id, actions, service_url, note='Downloading video info JSON (Kaltura kwidget player)')
+ info = _info['objects'][0]
+ return header, info, flavor_assets, captions
+
+ def _build_widget_id(self, partner_id):
+ return partner_id if '_' in partner_id else f'_{partner_id}'
+
+ IFRAME_PACKAGE_DATA_REGEX = r'window\.kalturaIframePackageData\s*='
def _real_extract(self, url):
url, smuggled_data = unsmuggle_url(url, {})
mobj = self._match_valid_url(url)
- partner_id, entry_id = mobj.group('partner_id', 'id')
- ks = None
- captions = None
+ partner_id, entry_id, player_type = mobj.group('partner_id', 'id', 'player_type')
+ ks, captions = None, None
+ if not player_type:
+ player_type = 'kwidget' if 'html5lib/v2' in url else 'html5'
if partner_id and entry_id:
- _, info, flavor_assets, captions = self._get_video_info(entry_id, partner_id, smuggled_data.get('service_url'))
+ _, info, flavor_assets, captions = self._get_video_info(entry_id, partner_id, smuggled_data.get('service_url'), player_type=player_type)
else:
path, query = mobj.group('path', 'query')
if not path and not query:
@@ -247,7 +402,7 @@ class KalturaIE(InfoExtractor):
splitted_path = path.split('/')
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:]
+ partner_id = remove_start(params['wid'][0], '_')
elif 'p' in params:
partner_id = params['p'][0]
elif 'partner_id' in params:
@@ -256,14 +411,13 @@ class KalturaIE(InfoExtractor):
raise ExtractorError('Invalid URL', expected=True)
if 'entry_id' in params:
entry_id = params['entry_id'][0]
- _, info, flavor_assets, captions = self._get_video_info(entry_id, partner_id)
+ _, info, flavor_assets, captions = self._get_video_info(entry_id, partner_id, player_type=player_type)
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']
+ entry_data = self._search_json(
+ self.IFRAME_PACKAGE_DATA_REGEX, webpage,
+ 'kalturaIframePackageData', reference_id)['entryResult']
info, flavor_assets = entry_data['meta'], entry_data['contextData']['flavorAssets']
entry_id = info['id']
# Unfortunately, data returned in kalturaIframePackageData lacks
@@ -271,16 +425,29 @@ class KalturaIE(InfoExtractor):
# regular approach since we now know the entry_id
try:
_, info, flavor_assets, captions = self._get_video_info(
- entry_id, partner_id)
+ entry_id, partner_id, player_type=player_type)
except ExtractorError:
# Regular scenario failed but we already have everything
# extracted apart from captions and can process at least
# with this
pass
+ elif 'uiconf_id' in params and 'flashvars[playlistAPI.kpl0Id]' in params:
+ playlist_id = params['flashvars[playlistAPI.kpl0Id]'][0]
+ webpage = self._download_webpage(url, playlist_id)
+ playlist_data = self._search_json(
+ self.IFRAME_PACKAGE_DATA_REGEX, webpage,
+ 'kalturaIframePackageData', playlist_id)['playlistResult']
+ return self.playlist_from_matches(
+ traverse_obj(playlist_data, (playlist_id, 'items', ..., 'id')),
+ playlist_id, traverse_obj(playlist_data, (playlist_id, 'name')),
+ ie=KalturaIE, getter=lambda x: f'kaltura:{partner_id}:{x}:{player_type}')
else:
raise ExtractorError('Invalid URL', expected=True)
ks = params.get('flashvars[ks]', [None])[0]
+ return self._per_video_extract(smuggled_data, entry_id, info, ks, flavor_assets, captions)
+
+ def _per_video_extract(self, smuggled_data, entry_id, info, ks, flavor_assets, captions):
source_url = smuggled_data.get('source_url')
if source_url:
referrer = base64.b64encode(
@@ -351,8 +518,6 @@ class KalturaIE(InfoExtractor):
formats.extend(fmts)
self._merge_subtitles(subs, target=subtitles)
- self._sort_formats(formats)
-
if captions:
for caption in captions.get('objects', []):
# Continue if caption is not ready
@@ -376,5 +541,5 @@ class KalturaIE(InfoExtractor):
'duration': info.get('duration'),
'timestamp': info.get('createdAt'),
'uploader_id': format_field(info, 'userId', ignore=('None', None)),
- 'view_count': info.get('plays'),
+ 'view_count': int_or_none(info.get('plays')),
}
generated by cgit v1.2.3 (git 2.25.1) at 2024-05-20 03:20:08 +0000