Mapping input and output data for an activity or step

In Process Designer, to pass variables to an activity or a step you have to set the input and output data mapping.

Before you begin

You must have a business process definition (BPD) that contains a set of declared variables and an activity, or a service that contains a set of declared variables and a step. The activity or step must implement a service or linked process which also contains a set of declared variables that match the ones in the main BPD in the case of an activity, or the service in the case of a step.

Restriction: In the case of a client-side human service, mapping input and output data is applicable only if the client-side human service is used within a BPD or a case type activity. Data mapping is not applicable if the client-side human service is used for a dashboard or a case instance.

About this task

When developing processes in Process Designer, you must set the input and output mapping for each activity included in a BPD so that the variable values received and generated by the subprocesses and services that implement the activities map to the variables from the main BPD. Similarly, if you have a service, you must set the input and output mapping for each step included in the service.

Note: This task is not relevant if you use the Activity Wizard to create a service to implement an activity. Also, Case management functions are only available if you have IBM BPM Advanced with the Basic Case Management feature installed.

The following procedure describes how to map both input and output data for an activity or a step. Depending on the logic of your BPD or service, an activity or step may only require input or output data and not both.

Procedure

Open the diagram of your BPD or service.
Click the activity in the BPD diagram or the step in the service diagram, and then click the Data Mapping tab in the properties. The Data Mapping tab displays the variables that are available from the service implemented in the activity or step.
- For receive events that are triggered using an undercover agent (UCA), both identifier and non-identifier variables can be used for correlation.
- For receive events that are triggered using a Service Component Architecture (SCA) service call, only variables that are marked as process instance identifiers can be used for correlation.
  Tip: If a selected variable is subsequently changed to not be a process instance identifier, an error icon is displayed by the variable on this tab.
To complete the data mapping, click the Auto-map input properties with variables from service icon () in both the Input Mapping and Output Mapping sections.
Note: Auto-mapping works only when variable names and types match exactly. You should always use an identical name and data type for a set of input and output variables that are passed in, processed, and then passed back.

If your variable is a shared business object then the output mapping from the activity or step is not required.
Click Save to save the configuration.

What to do next

The data mapping is complete, and you can test it.