From f9ee343fc081ab8715d39305ac44ba79e2038ec7 Mon Sep 17 00:00:00 2001
From: "miriam.baglioni" <miriam.baglioni@isti.cnr.it>
Date: Mon, 27 Apr 2020 10:48:31 +0200
Subject: [PATCH] new parametrized workflow with preparation steps and new
 parameter input files

---
 .../input_communitytoresult_parameters.json   |   6 +
 ..._preparecommunitytoresult2_parameters.json |  32 +-
 ...t_preparecommunitytoresult_parameters.json |  29 +-
 .../oozie_app/config-default.xml              |  36 ++
 .../oozie_app/workflow.xml                    | 315 ++++++++++++++++--
 5 files changed, 338 insertions(+), 80 deletions(-)

diff --git a/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/input_communitytoresult_parameters.json b/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/input_communitytoresult_parameters.json
index 27bb097f9d..2baec0e687 100644
--- a/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/input_communitytoresult_parameters.json
+++ b/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/input_communitytoresult_parameters.json
@@ -35,6 +35,12 @@
     "paramLongName": "preparedInfoPath",
     "paramDescription": "the path where prepared info have been stored",
     "paramRequired": true
+  },
+  {
+    "paramName":"test",
+    "paramLongName":"isTest",
+    "paramDescription": "true if it is executing a test",
+    "paramRequired": false
   }
 
 ]
\ No newline at end of file
diff --git a/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/input_preparecommunitytoresult2_parameters.json b/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/input_preparecommunitytoresult2_parameters.json
index f61db6f346..3ba3c8e9c7 100644
--- a/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/input_preparecommunitytoresult2_parameters.json
+++ b/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/input_preparecommunitytoresult2_parameters.json
@@ -1,34 +1,10 @@
 [
-  {
-    "paramName":"is",
-    "paramLongName":"isLookupUrl",
-    "paramDescription": "URL of the isLookUp Service",
-    "paramRequired": true
-  },
   {
     "paramName":"s",
     "paramLongName":"sourcePath",
     "paramDescription": "the path of the sequencial file to read",
     "paramRequired": true
   },
-  {
-    "paramName":"as",
-    "paramLongName":"allowedsemrels",
-    "paramDescription": "the allowed semantic relations for propagation",
-    "paramRequired": true
-  },
-  {
-    "paramName":"h",
-    "paramLongName":"hive_metastore_uris",
-    "paramDescription": "the hive metastore uris",
-    "paramRequired": true
-  },
-  {
-    "paramName":"sg",
-    "paramLongName":"saveGraph",
-    "paramDescription": "true if the new version of the graph must be saved",
-    "paramRequired": false
-  },
   {
     "paramName": "ssm",
     "paramLongName": "isSparkSessionManaged",
@@ -40,11 +16,5 @@
     "paramLongName": "outputPath",
     "paramDescription": "the path used to store temporary output files",
     "paramRequired": true
-  },
-  {
-  "paramName":"tn",
-  "paramLongName":"resultTableName",
-  "paramDescription": "the name of the result table we are currently working on",
-  "paramRequired": true
- }
+  }
 ]
\ No newline at end of file
diff --git a/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/input_preparecommunitytoresult_parameters.json b/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/input_preparecommunitytoresult_parameters.json
index 90b5974ea4..a5dfefc0a0 100644
--- a/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/input_preparecommunitytoresult_parameters.json
+++ b/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/input_preparecommunitytoresult_parameters.json
@@ -5,12 +5,6 @@
     "paramDescription": "URL of the isLookUp Service",
     "paramRequired": true
   },
-  {
-    "paramName":"mt",
-    "paramLongName":"master",
-    "paramDescription": "should be local or yarn",
-    "paramRequired": true
-  },
   {
     "paramName":"s",
     "paramLongName":"sourcePath",
@@ -30,16 +24,21 @@
     "paramRequired": true
   },
   {
-    "paramName":"wu",
-    "paramLongName":"writeUpdate",
-    "paramDescription": "true if the update must be writte. No double check if information is already present",
+    "paramName": "ssm",
+    "paramLongName": "isSparkSessionManaged",
+    "paramDescription": "true if the spark session is managed, false otherwise",
+    "paramRequired": false
+  },
+  {
+    "paramName": "out",
+    "paramLongName": "outputPath",
+    "paramDescription": "the path used to store temporary output files",
     "paramRequired": true
   },
   {
-    "paramName":"sg",
-    "paramLongName":"saveGraph",
-    "paramDescription": "true if the new version of the graph must be saved",
-    "paramRequired": true
-  }
-
+  "paramName":"tn",
+  "paramLongName":"resultTableName",
+  "paramDescription": "the name of the result table we are currently working on",
+  "paramRequired": true
+ }
 ]
\ No newline at end of file
diff --git a/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/oozie_app/config-default.xml b/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/oozie_app/config-default.xml
index ea3a4d9223..2744ea92ba 100644
--- a/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/oozie_app/config-default.xml
+++ b/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/oozie_app/config-default.xml
@@ -19,4 +19,40 @@
         <name>hive_metastore_uris</name>
         <value>thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083</value>
     </property>
+    <property>
+        <name>spark2YarnHistoryServerAddress</name>
+        <value>http://iis-cdh5-test-gw.ocean.icm.edu.pl:18089</value>
+    </property>
+    <property>
+        <name>spark2EventLogDir</name>
+        <value>/user/spark/spark2ApplicationHistory</value>
+    </property>
+    <property>
+        <name>spark2ExtraListeners</name>
+        <value>com.cloudera.spark.lineage.NavigatorAppListener</value>
+    </property>
+    <property>
+        <name>spark2SqlQueryExecutionListeners</name>
+        <value>com.cloudera.spark.lineage.NavigatorQueryListener</value>
+    </property>
+    <property>
+        <name>sparkExecutorNumber</name>
+        <value>4</value>
+    </property>
+    <property>
+        <name>sparkDriverMemory</name>
+        <value>15G</value>
+    </property>
+    <property>
+        <name>sparkExecutorMemory</name>
+        <value>6G</value>
+    </property>
+    <property>
+        <name>sparkExecutorCores</name>
+        <value>1</value>
+    </property>
+    <property>
+        <name>spark2MaxExecutors</name>
+        <value>50</value>
+    </property>
 </configuration>
\ No newline at end of file
diff --git a/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/oozie_app/workflow.xml b/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/oozie_app/workflow.xml
index 4312ec0682..d320bc9eb4 100644
--- a/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/oozie_app/workflow.xml
+++ b/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/oozie_app/workflow.xml
@@ -8,54 +8,301 @@
         <name>allowedsemrels</name>
         <description>the semantic relationships allowed for propagation</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>isLookupUrl</name>
         <description>the isLookup service endpoint</description>
     </property>
 </parameters>
 
-<start to="ResultToCommunityFromSemRelPropagation"/>
+    <start to="reset-outputpath"/>
 
-<kill name="Kill">
-    <message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
-</kill>
+    <kill name="Kill">
+        <message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
+    </kill>
+    <action name="reset-outputpath">
+        <fs>
+            <delete path='${workingDir}/resulttocommunitysemrel_propagation'/>
+        </fs>
+        <ok to="copy_relation"/>
+        <error to="Kill"/>
+    </action>
+    <action name="copy_relation">
+        <distcp xmlns="uri:oozie:distcp-action:0.2">
+            <job-tracker>${jobTracker}</job-tracker>
+            <name-node>${nameNode}</name-node>
+            <arg>${nameNode}/${sourcePath}/relation</arg>
+            <arg>${nameNode}/${workingDir}/projecttoresult_propagation/relation</arg>
+        </distcp>
+        <ok to="fork_prepare_assoc_step1"/>
+        <error to="Kill"/>
+    </action>
 
-<action name="ResultToCommunityFromSemRelPropagation">
-    <spark xmlns="uri:oozie:spark-action:0.2">
-        <job-tracker>${jobTracker}</job-tracker>
-        <name-node>${nameNode}</name-node>
-        <master>yarn-cluster</master>
+    <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-propagation-${projectVersion}.jar</jar>
+            <spark-opts>
+                --executor-cores=${sparkExecutorCores}
+                --executor-memory=${sparkExecutorMemory}
+                --driver-memory=${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>${workingDir}/preparedInfo/targetCommunityAssoc</arg>
+            <arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
+            <arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg>
+        </spark>
+        <ok to="wait"/>
+        <error to="Kill"/>
+    </action>
+    <action name="join_prepare_dataset">
+        <spark xmlns="uri:oozie:spark-action:0.2">
+            <master>yarn</master>
+            <mode>cluster</mode>
+            <name>ResultToCommunitySemRel-PreparePhase1-Dataset</name>
+            <class>eu.dnetlib.dhp.resulttocommunityfromsemrel.PrepareResultCommunitySetStep1</class>
+            <jar>dhp-propagation-${projectVersion}.jar</jar>
+            <spark-opts>
+                --executor-cores=${sparkExecutorCores}
+                --executor-memory=${sparkExecutorMemory}
+                --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}/preparedInfo/targetCommunityAssoc</arg>
+            <arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
+            <arg>--isLookupUrl</arg><arg>${isLookupUrl}</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-propagation-${projectVersion}.jar</jar>
+            <spark-opts>
+                --executor-cores=${sparkExecutorCores}
+                --executor-memory=${sparkExecutorMemory}
+                --driver-memory=${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>${workingDir}/preparedInfo/targetCommunityAssoc</arg>
+            <arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
+            <arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg>
+        </spark>
+        <ok to="wait"/>
+        <error to="Kill"/>
+    </action>
+    <action name="join_prepare_software">
+        <spark xmlns="uri:oozie:spark-action:0.2">
+            <master>yarn</master>
+            <mode>cluster</mode>
+            <name>ResultToCommunitySemRel-PreparePhase1-Software</name>
+            <class>eu.dnetlib.dhp.resulttocommunityfromsemrel.PrepareResultCommunitySetStep1</class>
+            <jar>dhp-propagation-${projectVersion}.jar</jar>
+            <spark-opts>
+                --executor-cores=${sparkExecutorCores}
+                --executor-memory=${sparkExecutorMemory}
+                --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}/preparedInfo/targetCommunityAssoc</arg>
+            <arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
+            <arg>--isLookupUrl</arg><arg>${isLookupUrl}</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>ResultToCommunitySemRelPropagation</name>
-        <class>eu.dnetlib.dhp.resulttoorganizationfrominstrepo.SparkResultToOrganizationFromIstRepoJob</class>
+        <name>ResultToCommunityEmRelPropagation-PreparePhase2</name>
+        <class>eu.dnetlib.dhp.resulttocommunityfromsemrel.PrepareResultCommunitySetStep2</class>
         <jar>dhp-propagation-${projectVersion}.jar</jar>
-        <spark-opts>--executor-memory ${sparkExecutorMemory}
-            --executor-cores ${sparkExecutorCores}
+        <spark-opts>
+            --executor-cores=${sparkExecutorCores}
+            --executor-memory=${sparkExecutorMemory}
             --driver-memory=${sparkDriverMemory}
-            --conf spark.extraListeners="com.cloudera.spark.lineage.NavigatorAppListener"
-            --conf spark.sql.queryExecutionListeners="com.cloudera.spark.lineage.NavigatorQueryListener"
+            --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>-mt</arg> <arg>yarn-cluster</arg>
-        <arg>--sourcePath</arg><arg>${sourcePath}</arg>
-       <arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
-        <arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
-        <arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg>
+        <arg>--sourcePath</arg><arg>${workingDir}/preparedInfo/targetCommunityAssoc</arg>
+        <arg>--outputPath</arg><arg>${workingDir}/preparedInfo/mergedCommunityAssoc</arg>
     </spark>
-    <ok to="End"/>
+    <ok to="fork-join-exec-propagation"/>
+<!--            <ok to="End"/>-->
     <error to="Kill"/>
-</action>
+
+    </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.SparkResultToCommunityThroughSemRelJob4</class>
+            <jar>dhp-propagation-${projectVersion}.jar</jar>
+            <spark-opts>
+                --executor-cores=${sparkExecutorCores}
+                --executor-memory=${sparkExecutorMemory}
+                --driver-memory=${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>${workingDir}/preparedInfo/mergedCommunityAssoc</arg>
+            <arg>--sourcePath</arg><arg>${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}/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.SparkResultToCommunityThroughSemRelJob4</class>
+            <jar>dhp-propagation-${projectVersion}.jar</jar>
+            <spark-opts>
+                --executor-cores=${sparkExecutorCores}
+                --executor-memory=${sparkExecutorMemory}
+                --driver-memory=${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>${workingDir}/preparedInfo/mergedCommunityAssoc</arg>
+            <arg>--sourcePath</arg><arg>${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}/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.SparkResultToCommunityThroughSemRelJob4</class>
+            <jar>dhp-propagation-${projectVersion}.jar</jar>
+            <spark-opts>
+                --executor-cores=${sparkExecutorCores}
+                --executor-memory=${sparkExecutorMemory}
+                --driver-memory=${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>${workingDir}/preparedInfo/mergedCommunityAssoc</arg>
+            <arg>--sourcePath</arg><arg>${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}/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.SparkResultToCommunityThroughSemRelJob4</class>
+            <jar>dhp-propagation-${projectVersion}.jar</jar>
+            <spark-opts>
+                --executor-cores=${sparkExecutorCores}
+                --executor-memory=${sparkExecutorMemory}
+                --driver-memory=${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>${workingDir}/preparedInfo/mergedCommunityAssoc</arg>
+            <arg>--sourcePath</arg><arg>${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}/software</arg>
+        </spark>
+        <ok to="wait2"/>
+        <error to="Kill"/>
+    </action>
+
+    <join name="wait2" to="End"/>
 
 <end name="End"/>
 </workflow-app>
\ No newline at end of file