How the Business Process Component works

This topic provides a generic description of what happens when a Branch Transformation Toolkit application needs to perform one or more activities in the business logic tier.

  1. The application presentation layer creates a WSIF request message and sends it to the application logic layer through a WSIF Access action, or calls a parameterized EJB method directly or through the EJB Access action.
  2. As a process engine starts the process, the ProcessStarter of the Abstract layer starts automatically.
  3. The Abstract Layer retrieves the session data (session ID and sub-session ID if used) from the request message or from the work area.
  4. The Abstract Layer creates the process CHA context and uses a request mapper to map the data in the request message or method call to the context.
  5. The Abstract Layer chains the process context to the session context in the CHA hierarchy. The exception is a login process, which chains the process context to a specified context in the hierarchy.
  6. The Abstract Layer caches the contexts to improve performance.
  7. The process engine proceeds through the activities in the process flow.
  8. At the end of a process flow, the ProcessTerminator of the Abstract Layer consults the MapList attribute to determine which data goes in response and uses response mappers to format the data. The ProcessTerminator then places the formatted data into the response message.
  9. When the process engine has completed the flow, the Abstract Layer cleans up after the process. The Abstract Layer clears caches and removes the process context from the context hierarchy.
  10. The application logic layer returns the response message to the application presentation layer.

For more details and the variations that can occur within the process, see the descriptions of the default processes.

Related concepts
Architecture
Related reference
Default processes