added SparkResolveEntities node to the oozie wf

dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/resolution/SparkResolveEntities.scala

@@ -2,9 +2,11 @@ package eu.dnetlib.dhp.oa.graph.resolution
 
 import com.fasterxml.jackson.databind.ObjectMapper
 import eu.dnetlib.dhp.application.ArgumentApplicationParser
+import eu.dnetlib.dhp.common.HdfsSupport
 import eu.dnetlib.dhp.schema.common.EntityType
 import eu.dnetlib.dhp.schema.oaf.{OtherResearchProduct, Publication, Result, Software, Dataset => OafDataset}
 import org.apache.commons.io.IOUtils
+import org.apache.hadoop.fs.{FileSystem, Path}
 import org.apache.spark.SparkConf
 import org.apache.spark.sql._
 import org.slf4j.{Logger, LoggerFactory}
@@ -34,13 +36,22 @@ object SparkResolveEntities {
     val unresolvedPath = parser.get("unresolvedPath")
     log.info(s"unresolvedPath  -> $unresolvedPath")
 
+    val fs = FileSystem.get(spark.sparkContext.hadoopConfiguration)
+    fs.mkdirs(new Path(workingPath))
 
     resolveEntities(spark, workingPath, unresolvedPath)
-
-
     generateResolvedEntities(spark, workingPath, graphBasePath)
 
-  }
+    // TO BE conservative we keep the original entities in the working dir
+    // and save the resolved entities on the graphBasePath
+    //In future these lines of code should be removed
+    entities.foreach {
+      e =>
+        fs.rename(new Path(s"$graphBasePath/$e"), new Path(s"$workingPath/${e}_old"))
+        fs.rename(new Path(s"$workingPath/resolvedGraph/$e"), new Path(s"$graphBasePath/$e"))
+    }
+
+}
 
 
 def resolveEntities(spark: SparkSession, workingPath: String, unresolvedPath: String) = {

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

@@ -4,6 +4,10 @@
             <name>graphBasePath</name>
             <description>the path of the graph</description>
         </property>
+        <property>
+            <name>unresolvedPath</name>
+            <description>the path of the unresolved Entities</description>
+        </property>
     </parameters>
 
     <start to="ResolveRelations"/>
@@ -36,5 +40,33 @@
         <ok to="End"/>
         <error to="Kill"/>
     </action>
+
+    <action name="ResolveEntities">
+        <spark xmlns="uri:oozie:spark-action:0.2">
+            <master>yarn</master>
+            <mode>cluster</mode>
+            <name>Resolve Relations in raw graph</name>
+            <class>eu.dnetlib.dhp.oa.graph.resolution.SparkResolveEntities</class>
+            <jar>dhp-graph-mapper-${projectVersion}.jar</jar>
+            <spark-opts>
+                --executor-memory=${sparkExecutorMemory}
+                --executor-cores=${sparkExecutorCores}
+                --driver-memory=${sparkDriverMemory}
+                --conf spark.extraListeners=${spark2ExtraListeners}
+                --conf spark.sql.shuffle.partitions=10000
+                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
+                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
+                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
+            </spark-opts>
+            <arg>--master</arg><arg>yarn</arg>
+            <arg>--graphBasePath</arg><arg>${graphBasePath}</arg>
+            <arg>--unresolvedPath</arg><arg>${unresolvedPath}</arg>
+            <arg>--workingPath</arg><arg>${workingDir}</arg>
+        </spark>
+        <ok to="End"/>
+        <error to="Kill"/>
+    </action>
+
+
     <end name="End"/>
 </workflow-app>