Scholexplorer Datasource Aggregation:

- Added collectedfrom in the inverse relation generated
Relation resolution:
- increased number of partitions in workflow.xml
- using classid instead of classname to build the pid-dnetId mapping

dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/collection/CollectionUtils.scala

@@ -34,6 +34,12 @@ object CollectionUtils {
         inverse.setRelType(currentRel.getRelType)
         inverse.setSubRelType(currentRel.getSubReltype)
         inverse.setRelClass(currentRel.getInverseRelClass)
+        inverse.setCollectedfrom(r.getCollectedfrom)
+        inverse.setDataInfo(r.getDataInfo)
+        inverse.setProperties(r.getProperties)
+        inverse.setLastupdatetimestamp(r.getLastupdatetimestamp)
+        inverse.setValidated(r.getValidated)
+        inverse.setValidationDate(r.getValidationDate)
         return List(r, inverse)
       }
     }

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

@@ -104,14 +104,14 @@ object SparkResolveRelation {
       JObject(pids) <- json \\ "instance" \ "pid"
       JField("value", JString(pidValue)) <- pids
       JField("qualifier", JObject(qualifier)) <- pids
-      JField("classname", JString(pidType)) <- qualifier
+      JField("classid", JString(pidType)) <- qualifier
     } yield (pidValue, pidType)
 
     val alternateIds: List[(String, String)] = for {
       JObject(pids) <- json \\ "alternateIdentifier"
       JField("value", JString(pidValue)) <- pids
       JField("qualifier", JObject(qualifier)) <- pids
-      JField("classname", JString(pidType)) <- qualifier
+      JField("classid", JString(pidType)) <- qualifier
     } yield (pidValue, pidType)
 
     (id, result ::: alternateIds)

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

@@ -24,7 +24,7 @@
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
-                --conf spark.sql.shuffle.partitions=3000
+                --conf spark.sql.shuffle.partitions=8000
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                 --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}