dnet-hadoop/dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/oa/graph/raw_hdfs_stores/oozie_app/workflow.xml

<workflow-app name="Test Import of Hdfs Stores" xmlns="uri:oozie:workflow:0.5">

    <parameters>
        <property>
            <name>graphOutputPath</name>
            <description>the target path to store raw graph</description>
        </property>
        <property>
            <name>contentPath</name>
            <description>path location to store (or reuse) content from the aggregator</description>
        </property>
        <property>
            <name>mdstoreManagerUrl</name>
            <description>the address of the Mdstore Manager</description>
        </property>
		<property>
            <name>isLookupUrl</name>
            <description>the address of the lookUp service</description>
        </property>
        <property>
            <name>sparkDriverMemory</name>
            <description>memory for driver process</description>
        </property>
        <property>
            <name>sparkExecutorMemory</name>
            <description>memory for individual executor</description>
        </property>
        <property>
            <name>sparkExecutorCores</name>
            <description>number of cores used by single executor</description>
        </property>
        <property>
            <name>oozieActionShareLibForSpark2</name>
            <description>oozie action sharelib for spark 2.*</description>
        </property>
        <property>
            <name>spark2ExtraListeners</name>
            <value>com.cloudera.spark.lineage.NavigatorAppListener</value>
            <description>spark 2.* extra listeners classname</description>
        </property>
        <property>
            <name>spark2SqlQueryExecutionListeners</name>
            <value>com.cloudera.spark.lineage.NavigatorQueryListener</value>
            <description>spark 2.* sql query execution listeners classname</description>
        </property>
        <property>
            <name>spark2YarnHistoryServerAddress</name>
            <description>spark 2.* yarn history server address</description>
        </property>
        <property>
            <name>spark2EventLogDir</name>
            <description>spark 2.* event log dir location</description>
        </property>
    </parameters>

    <global>
        <job-tracker>${jobTracker}</job-tracker>
        <name-node>${nameNode}</name-node>
        <configuration>
            <property>
                <name>mapreduce.job.queuename</name>
                <value>${queueName}</value>
            </property>
            <property>
                <name>oozie.launcher.mapred.job.queue.name</name>
                <value>${oozieLauncherQueueName}</value>
            </property>
            <property>
                <name>oozie.action.sharelib.for.spark</name>
                <value>${oozieActionShareLibForSpark2}</value>
            </property>
        </configuration>
    </global>

    <start to="ImportODF_hdfs"/>

    <kill name="Kill">
        <message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
    </kill>
    
    <action name="ImportODF_hdfs">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>ImportODF_hdfs</name>
            <class>eu.dnetlib.dhp.oa.graph.raw.MigrateHdfsMdstoresApplication</class>
            <jar>dhp-graph-mapper-${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}${spark2EventLogDir}
            </spark-opts>
            <arg>--hdfsPath</arg><arg>${contentPath}/odf_records_hdfs</arg>
            <arg>--mdstoreManagerUrl</arg><arg>${mdstoreManagerUrl}</arg>
            <arg>--mdFormat</arg><arg>ODF</arg>
            <arg>--mdLayout</arg><arg>store</arg>
            <arg>--mdInterpretation</arg><arg>cleaned</arg>
        </spark>
        <ok to="GenerateEntities"/>
        <error to="Kill"/>
    </action>

	<action name="GenerateEntities">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>GenerateEntities</name>
            <class>eu.dnetlib.dhp.oa.graph.raw.GenerateEntitiesApplication</class>
            <jar>dhp-graph-mapper-${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}${spark2EventLogDir}
            </spark-opts>
            <arg>--sourcePaths</arg><arg>${contentPath}/odf_records_hdfs</arg>
            <arg>--targetPath</arg><arg>${workingDir}/entities</arg>
            <arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg>
            <arg>--shouldHashId</arg><arg>${shouldHashId}</arg>
        </spark>
        <ok to="GenerateGraph"/>
        <error to="Kill"/>
    </action>
    
    <action name="GenerateGraph">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>GenerateGraph</name>
            <class>eu.dnetlib.dhp.oa.graph.raw.DispatchEntitiesApplication</class>
            <jar>dhp-graph-mapper-${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}${spark2EventLogDir}
                --conf spark.sql.shuffle.partitions=7680
            </spark-opts>
            <arg>--sourcePath</arg><arg>${workingDir}/entities</arg>
            <arg>--graphRawPath</arg><arg>${workingDir}/graph_raw</arg>
        </spark>
        <ok to="End"/>
        <error to="Kill"/>
    </action>
    
    <end name="End"/>
</workflow-app>