From 25267c16898f0452789f1613b54b3366469c38ec Mon Sep 17 00:00:00 2001
From: "miriam.baglioni" <miriam.baglioni@isti.cnr.it>
Date: Wed, 25 Oct 2023 11:48:42 +0200
Subject: [PATCH] extended the workflow to add the dump for the relations

---
 .../oa/graph/dump/eosc/oozie_app/workflow.xml | 26 ++++++++++++++++++-
 1 file changed, 25 insertions(+), 1 deletion(-)

diff --git a/dump/src/main/resources/eu/dnetlib/dhp/oa/graph/dump/eosc/oozie_app/workflow.xml b/dump/src/main/resources/eu/dnetlib/dhp/oa/graph/dump/eosc/oozie_app/workflow.xml
index def9f0c..077bb4f 100644
--- a/dump/src/main/resources/eu/dnetlib/dhp/oa/graph/dump/eosc/oozie_app/workflow.xml
+++ b/dump/src/main/resources/eu/dnetlib/dhp/oa/graph/dump/eosc/oozie_app/workflow.xml
@@ -603,7 +603,31 @@
     </action>
 
 
-    <join name="join_extend_relation" to="make_archive"/>
+    <join name="join_extend_relation" to="dump_organization_project_relations"/>
+
+    <action name="dump_organization_project_relations">
+        <spark xmlns="uri:oozie:spark-action:0.2">
+            <master>yarn</master>
+            <mode>cluster</mode>
+            <name>Dump for the relations between organization and projects in the subset of entities relevant for EOSC</name>
+            <class>eu.dnetlib.dhp.oa.graph.dump.eosc.SparkDumpOrganizationProject</class>
+            <jar>dump-${projectVersion}.jar</jar>
+            <spark-opts>
+                --executor-memory=${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>${sourcePath}</arg>
+            <arg>--outputPath</arg><arg>${outputPath}/dump/</arg>
+        </spark>
+        <ok to="make_archive"/>
+        <error to="Kill"/>
+    </action>
     <action name="make_archive">
         <java>
             <main-class>eu.dnetlib.dhp.oa.graph.dump.MakeTar</main-class>