Arbortext Publishing Engine Server Composer
Support for Arbortext Publishing Engine publishing on the Arbortext PE server is called the PE Server Composer.
The PE Server Composer is simply an Arbortext PE Application implemented in Java. It runs in an Arbortext PE sub-process like any other Arbortext PE Application.
The Server Composer application is invoked by an Arbortext PE sub-process at the request of the Arbortext PE Request Manager. It receives the request from the Arbortext Editor client, which includes the input directory zipped as a JAR file. It opens the JAR file, which contains information about what operation to perform along with the data required by the operation. It performs the requested publishing operation, and places the results in a temporary directory known as the output directory. Then it zips the output directory into a JAR file and returns the JAR file to the Arbortext PE Request Manager, which returns it to the Arbortext Editor client.
The PE Server Composer application consists of three components:
• The Composition Application, which is a standard Arbortext PE Application. It receives control from the Arbortext PE Request Manager and invokes the next layer, which is the PE Server Composer.
• The PE Server Composer performs operation-independent setup processing, then loads and calls an operation-specific PE Server Composition Extension.
• The PE Server Composition Extension does the real work by invoking code (usually the outer layer of the publishing framework) that knows how to run the content pipelines and content composers.