I have been trying to understand the usage of enableClasspathModules, clusters, and enableModules, but they do not seem to work as I expect.

package p;
import junit.framework.Test;
import org.netbeans.junit.NbModuleSuite;
import org.netbeans.junit.NbTestCase;
public class XTest extends NbTestCase {
    public XTest(String testName) {
        super(testName);
    }
    public static Test suite() {
        return NbModuleSuite.create(NbModuleSuite.createConfiguration(XTest.class)
                .enableModules(".*", ".*javascript.*")
                );
    }
    public void testStuff() {}
}

in a standalone module (set to an IDE as the target platform) loads all modules in the platform cluster, and no other clusters (except for the standalone module itself).

If I add .clusters(".*") then all IDE modules in all clusters are loaded.

Similarly, .clusters("ide").enableModules("ide", ".*javascript.*") loads all modules in the platform and ide clusters. In no case is just part of a cluster loaded.

I am not sure why filtering does not work, but it is clear why just .enableModules(".*") does not load any non-platform clusters: the findClusters method checks clusterRegExp but ignores moduleRegExp. So, if you pass just enableModules, runInRuntimeContainer will call turnModules on only the platform cluster.