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

<workflow-app name="dump_funder_results" xmlns="uri:oozie:workflow:0.5">

    <parameters>
        <property>
            <name>sourcePath</name>
            <description>the source path</description>
        </property>
        <property>
            <name>isLookUpUrl</name>
            <description>the isLookup service endpoint</description>
        </property>
        <property>
            <name>outputPath</name>
            <description>the output path</description>
        </property>
        <property>
            <name>accessToken</name>
            <description>the access token used for the deposition in Zenodo</description>
        </property>
        <property>
            <name>connectionUrl</name>
            <description>the connection url for Zenodo</description>
        </property>
        <property>
            <name>metadata</name>
            <description> the metadata associated to the deposition</description>
        </property>
        <property>
            <name>depositionType</name>
            <description>the type of deposition we want to perform. "new" for brand new deposition, "version" for a new version of a published deposition (in this case the concept record id must be provided), "upload" to upload content to an open deposition for which we already have the deposition id (in this case the deposition id should be provided)</description>
        </property>
        <property>
            <name>conceptRecordId</name>
            <description>for new version, the id of the record for the old deposition</description>
        </property>
        <property>
            <name>depositionId</name>
            <description>the depositionId of a deposition open that has to be added content</description>
        </property>
        <property>
            <name>hiveDbName</name>
            <description>the target hive database name</description>
        </property>
        <property>
            <name>hiveJdbcUrl</name>
            <description>hive server jdbc url</description>
        </property>
        <property>
            <name>hiveMetastoreUris</name>
            <description>hive server metastore URIs</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="reset_outputpath"/>

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

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

    <action name="save_community_map">
        <java>
            <main-class>eu.dnetlib.dhp.oa.graph.dump.SaveCommunityMap</main-class>
            <arg>--outputPath</arg><arg>${workingDir}/communityMap</arg>
            <arg>--nameNode</arg><arg>${nameNode}</arg>
            <arg>--isLookUpUrl</arg><arg>${isLookUpUrl}</arg>
        </java>
        <ok to="dump_funder_results"/>
        <error to="Kill"/>
    </action>


    <action name="dump_funder_results">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Dump funder results </name>
            <class>eu.dnetlib.dhp.oa.graph.dump.funderresults.SparkPrepareResultProject</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}${spark2EventLogDir}
                --conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
            </spark-opts>
            <arg>--sourcePath</arg><arg>${sourcePath}</arg>
            <arg>--outputPath</arg><arg>${workingDir}/result</arg>
            <arg>--communityMapPath</arg><arg>${workingDir}/communityMap</arg>
        </spark>
        <ok to="End"/>
        <error to="Kill"/>
    </action>
    
    <action name="send_zenodo">
        <java>
            <main-class>eu.dnetlib.dhp.oa.graph.dump.SendToZenodoHDFS</main-class>
            <arg>--hdfsPath</arg><arg>${outputPath}</arg>
            <arg>--nameNode</arg><arg>${nameNode}</arg>
            <arg>--accessToken</arg><arg>${accessToken}</arg>
            <arg>--connectionUrl</arg><arg>${connectionUrl}</arg>
            <arg>--metadata</arg><arg>${metadata}</arg>
            <arg>--communityMapPath</arg><arg>${workingDir}/communityMap</arg>
            <arg>--conceptRecordId</arg><arg>${conceptRecordId}</arg>
            <arg>--depositionType</arg><arg>${depositionType}</arg>
            <arg>--depositionId</arg><arg>${depositionId}</arg>
        </java>
        <ok to="End"/>
        <error to="Kill"/>
    </action>

    <end name="End"/>

</workflow-app>