Using annotation processors

This document describes how to configure and use annotation processors in TG applications. For processor-specific details, refer respective wiki pages.

Development of TG applications involves the following processors:

• ApplicationDomain generator
• Model verifier
• Meta-model generator

Each processor page provides its own configuration instructions. Combined together, a typical TG application should be configured by adding the following to pom.xml of a *-pojo-bl Maven module:

<plugins>
    <plugin>
        <artifactId>maven-compiler-plugin</artifactId>
        <configuration>
            <encoding>UTF-8</encoding>
            <!-- to enable output of annotation processors -->
            <showWarnings>true</showWarnings>
            <generatedSourcesDirectory>${project.build.directory}/generated-sources/</generatedSourcesDirectory>
            <annotationProcessorPaths>
                <path>
                    <groupId>fielden</groupId>
                    <artifactId>platform-annotation-processors</artifactId>
                    <version>${platform.version}</version>
                </path>
            </annotationProcessorPaths>
            <annotationProcessors>
                <processor>ua.com.fielden.platform.processors.appdomain.ApplicationDomainProcessor</processor>
                <processor>ua.com.fielden.platform.processors.verify.VerifyingProcessor</processor>
                <processor>ua.com.fielden.platform.processors.metamodel.MetaModelProcessor</processor>
            </annotationProcessors>
        </configuration>
    </plugin>
</plugins>

The order of processors, specified by the <processor> tag, matters significantly.

1. ApplicationDomainProcessor generates the ApplicationDomain class.

2. VerifyingProcessor verifies entity definitions. It must be aware of the generated ApplicationDomain, hence is placed after it.

3. MetaModelProcessor generates meta-models. However, if verification fails, the verifying processor can prevent any subsequent processing from taking place, hence the meta-model processor is placed after it.

   NOTE: Currently, the meta-model processor can't be prevented from running by the verifying processor. Therefore, even if verification fails, meta-model generation will nevertheless be attempted.

To use annotations related to annotation processing, an additional compile-time dependency should be included in the same pom.xml file:

<dependency>
  <groupId>fielden</groupId>
  <artifactId>platform-annotation-processors</artifactId>
  <version>${platform.version}</version>
  <scope>compile</scope>
</dependency>