During this phase, the project team implements the core
runtime application (both the model transaction engine and the presentation
engine). To do this, the project team must do the following tasks:
- Code the startup process for the client workstation, both in the client workstation and in the server to which the workstation
is connected. Refer to Startup beans for information on how to code the startup
process, and see Flow processors for how to code the startup process for
Java clients.
- For each process of the application, code the main process
flows for both the client and the server to define the logic and parameters
of the process flows. This will involve doing the following:
- Defining which process the toolkit requires to execute to process the
process flow. You can use the Process Editor provided by WebSphere Studio
Application Developer and the toolkit BP Extension Tool to define business
processes.
- Defining which data elements the process flow uses. Reuse existing ones
or add new ones. Externalize the data for data elements using the process
definition file. You can use the CHA Editor to create new data elements and
CHA contexts.
- Defining which formatters the process flow uses. Reuse existing ones or
add new ones. You can use the Format Editor to create new formatters.
- Defining the process attributes. Reuse existing ones or add new ones.
- Code any new required services, such as a new communication adapter
or printing engine.
- Code the views.
- For Java clients, use the provided visual beans whenever possible. These
beans enable a developer to define the data management and process execution
processes required by the application. If new beans are required, see
Creating
a user-defined toolkit bean in the Visual beans documentation. For
these new beans, define the logic using visual composition tools such as the
Visual Composition Editor in WebSphere Studio Application Developer.
- For HTML clients, create JSP files based on the toolkit Struts
extensions. See JSPs for more information.