This document provides basic information on data samples and how they are useful when building integration flows.
A data sample is an example of connector output data. It is essential in building integration flows, because it allows you to see what sort of data the next connector in the flow will receive from the previous one. With this information you can configure the next connector to act properly on the input it receives.
You can generate more than one sample from a connector. By default, the first one in the list will be used. Also, you can manually select the proper sample from the list. All unused samples will be deleted once you make a choice.
Samples in Integration Flows
Each connector in a flow has its own input/output data standards. For proper flow operation, the connector need data mapping. It is a process of data conversion from one connector’s standard to the next one’s. Basically, data mapper retrieves data from one connector and relays it to the next connector in a compatible form. In order to see what sort of output data the first connector produces we use data sample.
A data sample can be retrieved from a connector or written manually by the integrator. In any case, the sample is then given as input to the next connector. The receiving connector may not require all the data from the sample or may require it in completely different order. That’s when you configure proper mapping so that the receiving connector gets input in accordance with its standard.
Please note that samples exist for flow building and testing purposes. Retrieving a sample for an action creates new objects in the target system. If you want to avoid this, you can use Generate Stub Sample function. It retrieves the connector’s output template and fills it with fake data that you can discard afterwards.
Let’s observe the following process example:
- Connector Baboon is followed by Connector Toucan.
- When we select Connector Toucan, we can see what sort of data it accepts as input.
- By retrieving a data sample from Connector Baboon, we can see what sort of data it provides as output.
- Then we configure the data mapper to relay Baboon data to Toucan, converting it first into a form acceptable by Toucan. The mapping process may include ignoring some of the output fields, switching field places, etc.
- On the example scheme we can see, that Baboon sends out the following data:
Name and Surnametogether in one field. Data mapper makes the necessary changes into Baboon output, so that the data fits Toucan input Standard.