Using the Integrated Ant Task

The Abator plugin includes an integrated Ant task for running Abator. This Ant task is different from the Ant task included in the basic Abator distribution in that it supports Java file merging, and is integrated with the eclipse user interface, logging system, and progress notification system.

It is useful to use the integrated Ant task in the following situations:

When you want to automate an Abator run, and support merging Java files. In this case, you should setup an Ant build as demonstrated below, then use an eclipse "headless Ant build" to run Abator automatically. The standard eclipse help includes directions for running a headless Ant build. Search in the eclipse help for the word "headless".
When you want to provide a custom implementation of one of the core Abator interfaces. For example, if you want to provide a custom implementation of the JavaTypeResolver interface. When running Abator from the simple menu option, you have no control over the classpath of the Abator run. When you run from Ant, you have much more control over the runtime classpath./li>

Integrated Ant Task Detail

The Abator plug-in declares an Ant task for use within the Eclipse JRE. The task is called "abator.generate". Here is en example Ant build file:

<project default="runAbator">
  <target name="runAbator">
    <eclipse.convertPath resourcepath="AbatorTest/src/config/abatorConfig.xml"
                         property="thePath"/>
    <eclipse.convertPath resourcepath="AbatorTest/lib/hsqldb.jar"
                         property="jdbc.driver.jar"/>
    <abator.generate configfile="${thePath}" >
      <propertyset>
        <propertyref name="jdbc.driver.jar"/>
      </propertyset>
    </abator.generate>
  </target>
</project>

This file uses the Eclipse supplied "eclipse.convertPath" task to calculate the actual path of the configuration file and the JDBC driver. Then we use the "abator.generate" task to run Abator. When running this from Eclipse interactively, the Ant launch must be setup to run in the same JRE as the workspace. This setting is a standard configuration setting for running Ant in Eclipse.

The task supports an optional nested <propertyset> element which is the standard Ant property set type. This can be used to pass parameters into a configuration file. For example, the above property jdbc.driver.jar can be accessed in the Abator configuration file with the escape sequence ${jdbc.driver.jar}

The integrated Ant task supports the following attributes:

Attribute	Description
configfile	This is the Abator configuration file to use. Use the "eclipse.convertPath" task to convert a workspace path to an absolute path before setting this attribute (as shown above) This attribute is required.
contextIds	If specified, then this is a comma delimited list of contexts to use in the current run of Abator. Any id specified in the list must exactly match the value of the `id` attribute of an <abatorContext> configuration element. Only ids specified in this list will be active for this run of Abator. If this attribute is not specified, then all contexts will be active.
fullyQualifiedTableNames	If specified, then this is a comma delimited list of tables to use in the current run of Abator. Any table specified in the list must exactly match the fully qualified table name specified in a <table> configuration element. Only tables specified in this list will be active for this run of Abator. If this attribute is not specified, then all tables will be active. Specify table names as: `table` `schema.table` `catalog..table` etc.