This topic describes how Extract actions
and Insert actions are implemented in a recorded flow.
Recall that an Invoke screen operation
node specifies the information required to manage the transition from
one application screen to the next. The input actions (Input actions,
AID key actions, and Insert actions apply to the currently displayed
application screen. The output actions (Extract actions) apply to
the next application screen displayed. (See Screen operation.)
When an Invoke screen operation node is performed
the runtime environment processes the information in the associated
screen operation in the following order:
- It performs the input actions on the currently displayed screen:
- It performs the Insert actions (if any).
- It performs the Input actions (if any).
- It performs the AID key.
- It waits for the host to send the next application screen.
- It does screen recognition by comparing each screen in the next
screen list to the next application screen until it finds a match.
- It selects the output terminal associated with the recognized
screen description.
- It performs the output actions on the newly displayed application
screen:
- It performs the Extract actions (if any).
The flow of control then passes to the next node in the flow.
The input actions (Input actions, AID key action,
and Insert action) are implemented as mappings in the mapping routine
associated with the input terminal of the Invoke screen operation
node (see How actions are implemented in mappings: AID key, Input, Extract, and Insert).
The output actions (Extract actions) are implemented
as mappings in the mapping routine associated with selected output
terminal (see How actions are implemented in mappings: AID key, Input, Extract, and Insert).