Frequently Asked Questions
Here, you will find answers to some of the most common questions we receive from our users
Questions
If there have been modifications made to assignments that have not been communicated to the client, it is unclear which assignments will be returned by
GET /planning/requests/{id}/assignments
Let's not make it more complicated than it is. Essentially, REST operations are stateless, which means that a GET (A->B) should not depend on whether information about that resource (in this case, an assignment) has been shared via a PUT. This means that the relevant GET should always provide current information, regardless of whether a PUT has been sent or not.
Can you confirm whether there are any POST, PUT, GET, or DELETE operations available for specific planning lines in the PlanningAssignment message?
Indeed, the Open API specs currently do not include IDs at the planninglines (Periodic & Single line) for the Assignment message. However, this does not mean that it is not possible to make adjustments or deletions for the Assignment message. A decision was made at the time to assign responsibility for assigning a worker to a job to either the staffing supplier's planning system (Scenario 1) or the planning system of the staffing customer (Scenario 2). In both cases, the Assignment message is intended as a notification to a back-office system. Therefore, it is correct that the OpenAPI spec only allows POST, PUT, and DELETE operations at the main level and not at the planningline level. If specific planning lines need to be modified or deleted, the sender can use a PUT to notify the recipient.
What is the protocol for dealing with errors in message content during synchronous or asynchronous message processing?
This should not be an issue for synchronous processing of the SETU Planning and Scheduling messages. If the receiver is unable to process the /planning/..., this can be indicated through a response with a 400 series error code.
Regarding asynchronous processing of planning requests, it is currently not possible to indicate errors in processing. This has not come up before. Please reach out to us if you consider working out this issue necessary.
Is it also possible to perform modifications to existing resources using either the obtained UUID or the document ID??
Currently only the newly received UUID can be used by the creator to modify or delete their created resources using methods such as PUT, DELETE, or even a POST of a planningline. The SETU is exploring options to enable modifications using both identifiers. This would mean that the integrating party of the API endpoint would also need to store the document ID. One possible approach is to include both the document ID and the newly generated UUID associated with the resource in the API response. By including both identifiers, both organisations will have visibility of both identifiers and can use them accordingly.