General Objects

Filter Objects

Array Objects


String Enums Constants

Kaltura API

Service Name uploadToken
Name Description
addAdds new upload token to upload a file
getGet upload token by id
uploadUpload a file using the upload token id, returns an error on failure (an exception will be thrown when using one of the Kaltura clients)
Chunks can be uploaded in parallel and they will be appended according to their resumeAt position.
A parallel upload session should have three stages:
1. A single upload with resume=false and finalChunk=false
2. Parallel upload requests each with resume=true,finalChunk=false and the expected resumetAt position.
If a chunk fails to upload it can be re-uploaded.
3. After all of the chunks have been uploaded a final chunk (can be of zero size) should be uploaded
with resume=true, finalChunk=true and the expected resumeAt position. In case an UPLOAD_TOKEN_CANNOT_MATCH_EXPECTED_SIZE exception
has been returned (indicating not all of the chunks were appended yet) the final request can be retried.
deleteDeletes the upload token by upload token id
listList upload token by filter with pager support.
When using a user session the service will be restricted to users objects only.