Question

We have our large build broken into many build files with a master build that calls each of them with an ant target. In the case of unit tests, we want to be able to run all of the tests (so reports are generated) and then fail the entire build on an error or failure at the end. I understand errorProperty and failureProperty (doc) can be used to set a property to true to be referenced at the end of the build, but don’t know how to bubble that back up to the top level build. How can we bubble up a JUnit failure or error from a sub build, but only fail once all tests are done?

Answer 1

One option is to use AntUnit:

The <antunit> task drives the tests much like <junit> does for JUnit tests.

When called on a build file, the task will start a new Ant project for that build file and scan for targets with names that start with “test”. For each such target it then will

• Execute the target named setUp, if there is one.
• Execute the target itself – if this target depends on other targets the normal Ant rules apply and the dependent targets are executed first.
• Execute the target names tearDown, if there is one.

AntUnit provides a summary for each subproject tested and fails the entire build if one of the subprojects fails.

In master build.xml

<target name="test" depends="compile" description="Run tests.">            
  <antunit>
    <plainlistener loglevel="info" />
    <fileset dir="${subprojects.dir}" includes="**/*.xml" />
  </antunit>
</target>

Subproject build.xml

<target name="testJUnit">
  <junit printsummary="on" fork="true" forkmode="once" showoutput="true">
    ...
  </junit>
</target>