One of Dynamo’s most important tasks is handling HTTP requests. In handling these requests, Dynamo uses session tracking, page compilation, JSPs, and other powerful extensions to the basic Web server model.
Request handling can usually be broken down into a series of independent steps. Each step might depend on additional information being available about the request, so order does matter. However, the individual steps are separable. For example, a typical request might go through these steps:
Compare the request URI against a list of restricted directories, to make sure that the user has permission to access the specified directory.
Translate the request URI into a real file name, taking index files into account when the file name refers to a directory.
Given the file name’s extension, determine the MIME type of the file.
From the MIME type, dispatch the request to the appropriate handler.
This is only one of many possible request-handling configurations. Other configurations might dispatch based on a beginning path such as /cgibin. Other configurations might move the session-tracking step to be performed only for files with the MIME type text/session-tracked.
In Nucleus-based applications, each of these steps is represented by a single Nucleus service that also implements the Servlet interface. Most requests handled by Nucleus are dispatched to the PageFilter filter or DynamoProxyServlet, either of which start the servlet pipeline. Each servlet in turn performs its specific function on the request. Each servlet is also configured with a reference to the next servlet in the pipeline. When a servlet is done acting on the request, it can pass the request to the next servlet. This linked list of servlets is called a request-handling pipeline.
Treating the request-handling process as a pipeline of independent elements allows request-handling to be treated in a component-oriented manner. This lets you use Nucleus to customize the request-handling process through configurable components, thereby giving you the flexibility that is often required by enterprise applications.
The request-handling pipeline in Dynamo lets you do the following:
Send requests to a Java servlet that you implemented.
Example: running Java classFooServletwhen the user enters the path/foo.Execute code for all requests that Dynamo processes.
Example: performing authentication on requests.Change Dynamo’s default request processing behavior.
Dynamo uses two request-handling pipelines:
DAS servlet pipeline for JHTML requests
DAF servlet pipeline for JSP requests
When a JHTML page is requested, the DAS servlet pipeline runs through the set of servlets appropriate for conditions of the request as would the DAF servlet pipeline for JSP requests. Because JHTML is a proprietary language, it relies on the page compiler provided in the DAS servlet pipeline to generate JHTML into a servlet that is rendered as HTML by the application server. Other minor differences exist between the request-handling pipelines, but many of the classes invoked for each are the same.
In this chapter
This chapter includes the following sections that describe how to use and customize the DAF servlet pipeline:
This chapter does not address any of the features provided by the ATG Portal module. See the Portal Requests and the Filter Chain chapter of the ATG Portal Development Guide for specific information.
