Arbortext Publishing Engine ships with a predefined Request Handler implemented by the Java class com.arbortext.e3.RequestHandler. It processes HTTP requests by invoking one of a set of predefined Java objects called Request Functions. For each HTTP request, the Request Handler looks for a query parameter named f and invokes a Request Function corresponding to the f parameter value. For example, to handle a request with parameter f=status it invokes the request function com.arbortext.e3.FunctionStatus.

At initialization, the Arbortext Publishing Engine Request Handler reads parameters from the e3config.xml configuration file. It looks for two parameters, query-function-name and function-prefix. The query-function-name parameter tells the Request Handler the query parameter to search for when it is offered an HTTP request to process. The default value is f. The function-prefix specifies an initial substring for other initialization parameters. Its default value is f-. Initialization parameter names beginning with the function-prefix string specify the function name as the rest of the initialization parameter name and the corresponding Request Function as the parameter value.

For example, an initialization parameter named f-status has a value of com.arbortext.e3.FunctionStatus. The parameter name begins with f-, so the Request Handler binds the remaining part of the parameter name, status, to the com.arbortext.e3c.FunctionStatus class during request processing.

You can extend the Arbortext PE Request Manager by developing your own Request Handlers, as already discussed. However, to avoid implementing a complete Request Handler, you can implement your own Request Functions and configure the predefined Request Handler to invoke them.