build.xml

<project name="TitleHarvester" default="test" basedir=".">
    <description>
        Commoncrawl Title harvester
    </description>
    <taskdef resource="testngtasks" classpath="lib/testng-6.8.7.jar"/>

    <!-- set global properties for this build -->
    <property name="main.dir" location="src/main"/>
    <property name="test.dir"  location="src/test"/>
    <property name="build.dir"  location="build"/>
    <property name="build.main.dir" location="build/main"/>
    <property name="build.test.dir" location="build/test"/>
    <property name="dist.dir"  location="dist"/>
    <property name="test.output.dir" location="test_output"/>

    <target name="init">
        <!-- Create the time stamp -->
        <tstamp/>
        <!-- Create the build directory structure used by compile -->
        <mkdir dir="${build.main.dir}"/>
    </target>

    <target name="compile" depends="init"
            description="compile the source " >
        <javac srcdir="${main.dir}" destdir="${build.main.dir}" debug="true">
            <classpath>
                <pathelement path="${classpath}"/>
                <fileset dir="lib">
                    <include name="**/*.jar"/>
                </fileset>
            </classpath>
        </javac>
    </target>

    <target name="dist" depends="compile"
            description="generate the distribution" >
        <!-- Create the distribution directory -->
        <mkdir dir="${dist.dir}"/>
        <mkdir dir="${build.main.dir}/lib"/>

        <mkdir dir="${build.main.dir}/data"/>
        <mkdir dir="${build.main.dir}/data/domain_cat_index"/>
        <mkdir dir="${build.main.dir}/data/profiles"/>
        <mkdir dir="${build.main.dir}/data/stop-words"/>
        <!-- Put everything in ${build}/lib into the lib/ dir of build -->
        <copy todir="${build.main.dir}/lib">
            <fileset dir="lib" includes="*.jar" excludes="jcommander-1.32.jar,jmockit.jar,testng-6.8.7.jar,weka.jar,jsoup-1.6.1.jar,*-sources.jar"/>
        </copy>
        <copy todir="${build.main.dir}/data">
            <fileset dir="data" includes="*.json" />
        </copy>
        <copy todir="${build.main.dir}/data/domain_cat_index">
            <fileset dir="data/domain_cat_index" includes="*.json" />
        </copy>
        <copy todir="${build.main.dir}/data/profiles">
            <fileset dir="data/profiles" includes="*" />
        </copy>
        <copy todir="${build.main.dir}/data/stop-words">
            <fileset dir="data/stop-words" includes="*.txt" />
        </copy>

        <jar jarfile="${dist.dir}/TitleHarvester.jar" basedir="${build.main.dir}">
            <zipfileset includes="**/*.class **/*.json data/profiles/* data/stop-words/*.txt" src="lib/jsoup-1.6.1.jar"/>
        </jar>
    </target>


    <target name="test-compile" depends="compile"
            description="compile tests">
        <mkdir dir="${build.test.dir}"/>
        <javac srcdir="${test.dir}" destdir="${build.test.dir}" debug="true">
            <classpath>
                <pathelement path="${classpath}"/>
                <fileset dir="lib">
                    <include name="**/*.jar"/>
                </fileset>
            </classpath>
            <classpath>
                <pathelement path="${build.main.dir}"/>
            </classpath>
        </javac>
    </target>

    <target name="test" depends="test-compile"
            description="run test suite" >
        <mkdir dir="${test.output.dir}"/>
        <path id="test.classpath">
            <fileset dir="lib">
                <include name="**/*.jar"/>
            </fileset>
            <pathelement path="${build.main.dir}"/>
            <pathelement location="${build.test.dir}"/>
        </path>
        <testng classpathref="test.classpath"
                outputDir="${test.output.dir}"
                haltOnFailure="true" verbose="2">
            <classfileset dir="${build.test.dir}" includes="**/*.class" excludes="**/*$1.class"/>
        </testng>
    </target>

    <target name="clean"
            description="clean up" >
        <!-- Delete the ${build} and ${dist} directory trees -->
        <delete dir="${build.dir}"/>
        <delete dir="${dist.dir}"/>
        <delete dir="${test.output.dir}"/>
    </target>
</project>