dnet-hadoop/dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/actionmanager/bipfinder/oozie_app/workflow.xml

<workflow-app name="BipFinderScore" xmlns="uri:oozie:workflow:0.5">
    <parameters>
        <property>
            <name>inputPath</name>
            <description>the input path of the resources to be extended</description>
        </property>

        <property>
            <name>bipScorePath</name>
            <description>the path where to find the bipFinder scores</description>
        </property>
        <property>
            <name>outputPath</name>
            <description>the path where to store the actionset</description>
        </property>
    </parameters>

    <start to="deleteoutputpath"/>
    <kill name="Kill">
        <message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
    </kill>
    <action name="deleteoutputpath">
        <fs>
            <delete path="${outputPath}"/>
            <mkdir path="${outputPath}"/>
            <delete path="${workingDir}"/>
            <mkdir path="${workingDir}"/>
        </fs>
        <ok to="atomicactions"/>
        <error to="Kill"/>
    </action>

    <fork name="atomicactions">
        <path start="atomicactions_publication"/>
        <path start="atomicactions_dataset"/>
        <path start="atomicactions_orp"/>
        <path start="atomicactions_software"/>
    </fork>

    <action name="atomicactions_publication">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Produces the atomic action with the bip finder scores for publications</name>
            <class>eu.dnetlib.dhp.actionmanager.bipfinder.SparkAtomicActionScoreJob</class>
            <jar>dhp-aggregation-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-memory=${sparkExecutorMemory}
                --executor-cores=${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
            </spark-opts>
            <arg>--inputPath</arg><arg>${inputPath}/publication</arg>
            <arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
            <arg>--outputPath</arg><arg>${workingDir}/publication</arg>
            <arg>--bipScorePath</arg><arg>${bipScorePath}</arg>
        </spark>
        <ok to="join_aa"/>
        <error to="Kill"/>
    </action>

    <action name="atomicactions_dataset">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Produces the atomic action with the bip finder scores for datasets</name>
            <class>eu.dnetlib.dhp.actionmanager.bipfinder.SparkAtomicActionScoreJob</class>
            <jar>dhp-aggregation-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-memory=${sparkExecutorMemory}
                --executor-cores=${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
            </spark-opts>
            <arg>--inputPath</arg><arg>${inputPath}/dataset</arg>
            <arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
            <arg>--outputPath</arg><arg>${workingDir}/dataset</arg>
            <arg>--bipScorePath</arg><arg>${bipScorePath}</arg>
        </spark>
        <ok to="join_aa"/>
        <error to="Kill"/>
    </action>

    <action name="atomicactions_orp">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Produces the atomic action with the bip finder scores for orp</name>
            <class>eu.dnetlib.dhp.actionmanager.bipfinder.SparkAtomicActionScoreJob</class>
            <jar>dhp-aggregation-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-memory=${sparkExecutorMemory}
                --executor-cores=${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
            </spark-opts>
            <arg>--inputPath</arg><arg>${inputPath}/otherresearchproduct</arg>
            <arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
            <arg>--outputPath</arg><arg>${workingDir}/otherresearchproduct</arg>
            <arg>--bipScorePath</arg><arg>${bipScorePath}</arg>
        </spark>
        <ok to="join_aa"/>
        <error to="Kill"/>
    </action>

    <action name="atomicactions_software">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Produces the atomic action with the bip finder scores for software</name>
            <class>eu.dnetlib.dhp.actionmanager.bipfinder.SparkAtomicActionScoreJob</class>
            <jar>dhp-aggregation-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-memory=${sparkExecutorMemory}
                --executor-cores=${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
            </spark-opts>
            <arg>--inputPath</arg><arg>${inputPath}/software</arg>
            <arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
            <arg>--outputPath</arg><arg>${workingDir}/software</arg>
            <arg>--bipScorePath</arg><arg>${bipScorePath}</arg>
        </spark>
        <ok to="join_aa"/>
        <error to="Kill"/>
    </action>

    <join name="join_aa" to="collectandsave"/>

    <action name="collectandsave">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>saves all the aa produced for the several types of results in the as output path</name>
            <class>eu.dnetlib.dhp.actionmanager.bipfinder.CollectAndSave</class>
            <jar>dhp-aggregation-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-memory=${sparkExecutorMemory}
                --executor-cores=${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
            </spark-opts>
            <arg>--inputPath</arg><arg>${workingDir}</arg>
            <arg>--outputPath</arg><arg>${outputPath}</arg>
        </spark>
        <ok to="End"/>
        <error to="Kill"/>
    </action>

    <end name="End"/>
</workflow-app>