dnet-hadoop/dhp-workflows/dhp-enrichment/src/main/resources/eu/dnetlib/dhp/wf/subworkflows/resulttocommunityfromsemrel/oozie_app/workflow.xml

<workflow-app name="result_to_community_from_semrel_propagation" xmlns="uri:oozie:workflow:0.5">
    <parameters>
        <property>
            <name>sourcePath</name>
            <description>the source path</description>
        </property>
        <property>
            <name>allowedsemrels</name>
            <description>the semantic relationships allowed for propagation</description>
        </property>
        <property>
            <name>baseURL</name>
            <description>the isLookup service endpoint</description>
        </property>
        <property>
            <name>outputPath</name>
            <description>the output path</description>
        </property>
    </parameters>

    <start to="reset_outputpath"/>

    <kill name="Kill">
        <message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
    </kill>

    <action name="reset_outputpath">
        <fs>
            <delete path="${workingDir}"/>
            <mkdir path="${workingDir}"/>
        </fs>
        <ok to="fork_prepare_assoc_step1"/>
        <error to="Kill"/>
    </action>

    <fork name="fork_prepare_assoc_step1">
        <path start="join_prepare_publication"/>
        <path start="join_prepare_dataset"/>
        <path start="join_prepare_otherresearchproduct"/>
        <path start="join_prepare_software"/>
    </fork>

    <action name="join_prepare_publication">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>ResultToCommunitySemRel-PreparePhase1-Publications</name>
            <class>eu.dnetlib.dhp.resulttocommunityfromsemrel.PrepareResultCommunitySetStep1</class>
            <jar>dhp-enrichment-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-cores=6
                --executor-memory=5G
                --conf spark.executor.memoryOverhead=3g
                --conf spark.sql.shuffle.partitions=3284
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.dynamicAllocation.enabled=true
                --conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
            </spark-opts>
            <arg>--sourcePath</arg><arg>${sourcePath}</arg>
            <arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
            <arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
            <arg>--outputPath</arg><arg>${workingDir}/communitysemrel/preparedInfo/targetCommunityAssoc</arg>
            <arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
            <arg>--baseURL</arg><arg>${baseURL}</arg>
        </spark>
        <ok to="wait"/>
        <error to="Kill"/>
    </action>

    <action name="join_prepare_dataset">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>ResultToCommunitySemRel-PreparePhase1-Dataset</name>
            <class>eu.dnetlib.dhp.resulttocommunityfromsemrel.PrepareResultCommunitySetStep1</class>
            <jar>dhp-enrichment-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-cores=6
                --executor-memory=5G
                --conf spark.executor.memoryOverhead=3g
                --conf spark.sql.shuffle.partitions=3284
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.dynamicAllocation.enabled=true
                --conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
            </spark-opts>
            <arg>--sourcePath</arg><arg>${sourcePath}</arg>
            <arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
            <arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
            <arg>--outputPath</arg><arg>${workingDir}/communitysemrel/preparedInfo/targetCommunityAssoc</arg>
            <arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
            <arg>--baseURL</arg><arg>${baseURL}</arg>
        </spark>
        <ok to="wait"/>
        <error to="Kill"/>
    </action>

    <action name="join_prepare_otherresearchproduct">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>ResultToCommunitySemRel-PreparePhase1-ORP</name>
            <class>eu.dnetlib.dhp.resulttocommunityfromsemrel.PrepareResultCommunitySetStep1</class>
            <jar>dhp-enrichment-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-cores=6
                --executor-memory=5G
                --conf spark.executor.memoryOverhead=3g
                --conf spark.sql.shuffle.partitions=3284
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.dynamicAllocation.enabled=true
                --conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
            </spark-opts>
            <arg>--sourcePath</arg><arg>${sourcePath}</arg>
            <arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
            <arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
            <arg>--outputPath</arg><arg>${workingDir}/communitysemrel/preparedInfo/targetCommunityAssoc</arg>
            <arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
            <arg>--baseURL</arg><arg>${baseURL}</arg>
        </spark>
        <ok to="wait"/>
        <error to="Kill"/>
    </action>

    <action name="join_prepare_software">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>ResultToCommunitySemRel-PreparePhase1-Software</name>
            <class>eu.dnetlib.dhp.resulttocommunityfromsemrel.PrepareResultCommunitySetStep1</class>
            <jar>dhp-enrichment-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-cores=6
                --executor-memory=5G
                --conf spark.executor.memoryOverhead=3g
                --conf spark.sql.shuffle.partitions=3284
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.dynamicAllocation.enabled=true
                --conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
            </spark-opts>
            <arg>--sourcePath</arg><arg>${sourcePath}</arg>
            <arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
            <arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
            <arg>--outputPath</arg><arg>${workingDir}/communitysemrel/preparedInfo/targetCommunityAssoc</arg>
            <arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
            <arg>--baseURL</arg><arg>${baseURL}</arg>
        </spark>
        <ok to="wait"/>
        <error to="Kill"/>
    </action>

    <join name="wait" to="prepare_assoc_step2"/>

    <action name="prepare_assoc_step2">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>ResultToCommunityEmRelPropagation-PreparePhase2</name>
            <class>eu.dnetlib.dhp.resulttocommunityfromsemrel.PrepareResultCommunitySetStep2</class>
            <jar>dhp-enrichment-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-cores=6
                --executor-memory=5G
                --conf spark.executor.memoryOverhead=3g
                --conf spark.sql.shuffle.partitions=3284
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.dynamicAllocation.enabled=true
                --conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
            </spark-opts>
            <arg>--sourcePath</arg><arg>${workingDir}/communitysemrel/preparedInfo/targetCommunityAssoc</arg>
            <arg>--outputPath</arg><arg>${workingDir}/communitysemrel/preparedInfo/mergedCommunityAssoc</arg>
        </spark>
        <ok to="fork-join-exec-propagation"/>
        <error to="Kill"/>
    </action>

    <fork name="fork-join-exec-propagation">
        <path start="join_propagate_publication"/>
        <path start="join_propagate_dataset"/>
        <path start="join_propagate_otherresearchproduct"/>
        <path start="join_propagate_software"/>
    </fork>

    <action name="join_propagate_publication">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Result2CommunitySemRelPropagation-Publication</name>
            <class>eu.dnetlib.dhp.resulttocommunityfromsemrel.SparkResultToCommunityThroughSemRelJob</class>
            <jar>dhp-enrichment-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-cores=6
                --executor-memory=5G
                --conf spark.executor.memoryOverhead=3g
                --conf spark.sql.shuffle.partitions=3284
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.dynamicAllocation.enabled=true
                --conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
            </spark-opts>
            <arg>--preparedInfoPath</arg><arg>${workingDir}/communitysemrel/preparedInfo/mergedCommunityAssoc</arg>
            <arg>--sourcePath</arg><arg>${sourcePath}/publication</arg>
            <arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
            <arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
            <arg>--outputPath</arg><arg>${workingDir}/communitysemrel/publication</arg>

        </spark>
        <ok to="wait2"/>
        <error to="Kill"/>
    </action>

    <action name="join_propagate_dataset">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Result2CommunitySemRelPropagation-Dataset</name>
            <class>eu.dnetlib.dhp.resulttocommunityfromsemrel.SparkResultToCommunityThroughSemRelJob</class>
            <jar>dhp-enrichment-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-cores=6
                --executor-memory=5G
                --conf spark.executor.memoryOverhead=3g
                --conf spark.sql.shuffle.partitions=3284
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.dynamicAllocation.enabled=true
                --conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
            </spark-opts>
            <arg>--preparedInfoPath</arg><arg>${workingDir}/communitysemrel/preparedInfo/mergedCommunityAssoc</arg>
            <arg>--sourcePath</arg><arg>${sourcePath}/dataset</arg>
            <arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
            <arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
            <arg>--outputPath</arg><arg>${workingDir}/communitysemrel/dataset</arg>

        </spark>
        <ok to="wait2"/>
        <error to="Kill"/>
    </action>

    <action name="join_propagate_otherresearchproduct">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Result2CommunitySemRelPropagation-ORP</name>
            <class>eu.dnetlib.dhp.resulttocommunityfromsemrel.SparkResultToCommunityThroughSemRelJob</class>
            <jar>dhp-enrichment-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-cores=6
                --executor-memory=5G
                --conf spark.executor.memoryOverhead=3g
                --conf spark.sql.shuffle.partitions=3284
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.dynamicAllocation.enabled=true
                --conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
            </spark-opts>
            <arg>--preparedInfoPath</arg><arg>${workingDir}/communitysemrel/preparedInfo/mergedCommunityAssoc</arg>
            <arg>--sourcePath</arg><arg>${sourcePath}/otherresearchproduct</arg>
            <arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
            <arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
            <arg>--outputPath</arg><arg>${workingDir}/communitysemrel/otherresearchproduct</arg>

        </spark>
        <ok to="wait2"/>
        <error to="Kill"/>
    </action>

    <action name="join_propagate_software">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Result2CommunitySemRelPropagation-Software</name>
            <class>eu.dnetlib.dhp.resulttocommunityfromsemrel.SparkResultToCommunityThroughSemRelJob</class>
            <jar>dhp-enrichment-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-cores=6
                --executor-memory=5G
                --conf spark.executor.memoryOverhead=3g
                --conf spark.sql.shuffle.partitions=3284
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.dynamicAllocation.enabled=true
                --conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
            </spark-opts>
            <arg>--preparedInfoPath</arg><arg>${workingDir}/communitysemrel/preparedInfo/mergedCommunityAssoc</arg>
            <arg>--sourcePath</arg><arg>${sourcePath}/software</arg>
            <arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
            <arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
            <arg>--outputPath</arg><arg>${workingDir}/communitysemrel/software</arg>

        </spark>
        <ok to="wait2"/>
        <error to="Kill"/>
    </action>

    <join name="wait2" to="End"/>

    <end name="End"/>

</workflow-app>