Data Flow using Data Fabric API

Hi all

So with a client having SunSystems in the cloud, and the ability to access the SunSystems databases for pulling data via ION, I'm running into some trouble.

There is an API for the Data Fabric, but it seems convoluted when trying to string together the jobs, status, and results endpoints to obtain the query results.

Attempt1:

Using Workflow I seem to be able to get 90% of the way there; inputting the desired query into the jobs API, having it check for a running status and loop until it changes and kick off the results api step; however, there is no means to get the document/body out at the end of this workflow.

Attempt 2:

Publish the workflow, call it from a API flow, and then attempt to path the query ID from the first workflow to the second; there seems to be no means of doing so.

Surely it can't be this difficult for three APIs that are ostensibly supposed to work with one another?!

Find more posts tagged with

Infor Data Fabric

Accepted answers

Legacy Contributor

Hello Leith,
While using ION to orchestrate the Data Fabric APIs of jobs, status and results is a theoretical possibility, in practice this is not a recommended best practice. Because ION Workflows are meant for orchestration of business processes and workflows that involve assigning tasks to users or sending them alerts and notifications. From a load and throughput point of view, its not really designed for integration/ data retrieval patterns.

All comments

Legacy Contributor

Hi, @lgilroy

I happen to be using these data fabric api's with Infor CSD as follows.

Submit the query job and store the returned queryID in a variable. I also store it in a table, but this just helps me with logging. This api also returns the result rowcount to expect, which you can store in a variable for use in the results api.

Pass the queryID from variable into the status api, loop until FINISHED status is obtained. Branch to deal with a failure result to perhaps send an email or restart from the beginning.

Pass the queryID from variable to the results api, which returns the result body. The paging is maybe where you are having issues. I am not an advanced programmer and have a much easier time with this part by using the ZappySys SSIS PowerPack tool (which only works with SSIS,) which makes it relatively easy to do the result paging and pass on the results to the database. It has a component that easily does the upsert to the destination table as well without needing to manually construct a complex merge query. Whatever you do, you either have to use a tool or program the results paging by creating a variable for the offset and incrementing it on each loop through the results. I hope that helps. This api set is more complex but much more resilient than FetchWhere or the previous Compass api's. A failed result retrieval could also be restarted from where you left off if you are storing the offset as you go.

Legacy Contributor

Copyright © 2025 Infor. All rights reserved.