Types of variables

Global service variables

Service variables are available with workunits as they are created. They are persistent.

For example, service variable information associated with an Inbasket action is used to create the workunit (before and during execution) and still available for viewing even after the user has performed the requested action.

Process variables

Process variables are variables that you as a process designer create on the Start node. These variables are also persistent.

Activity variables

Variables can be created for some activity nodes (in addition to the Start activity). These variables are non-persistent in the sense that they last long enough to pass information to the next node. They are not available throughout execution of the entire process.

Return (or output) variables

Return variables are a special type of variable associated with synchronous triggers and the Return activity node.

You as a process designer can use the Return activity node to request information to be sent back to the calling (or triggering) application after your process has executed.

More information about triggers, including the difference between synchronous and asynchronous triggers, and about the Return activity node are discussed in other sections.

Specifying variables for Assign, Branch, and Start nodes

<!appProdline> is an example of a global service variable delivered by IPA. When you specify this variable in most activity nodes you spell it as shown above (the variable name is preceded by an exclamation point and the entire variable is surrounded in angle brackets).

However, when you use this variable with the Assign, Branch, or Start nodes, which accept JavaScript expressions (a special type of notation), you do not use "< >" or "!" with variables. In this case you simply specify the variable name. For example:

appProdline==PROD

More information about creating JavaScript expressions with these nodes is in other sections.