D-Net
/
dnet-hadoop
14
5
Fork 11
Code Issues 14 Pull Requests 13 Projects 2 Releases 2 Wiki Activity

code refactor

This commit is contained in:
Sandro La Bruzzo 2021-07-08 09:08:25 +02:00
parent a01dbe0ab0
commit a4a54a3786
1 changed files with 1 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/sx/graph/SparkCreateScholix.scala
View File

@ -81,9 +81,6 @@ object SparkCreateScholix {
val scholix_final:Dataset[Scholix] = spark.read.load(s"$targetPath/scholix").as[Scholix]
val stats:Dataset[(String,String,Long)]= scholix_final.map(s => (s.getSource.getDnetIdentifier, s.getTarget.getObjectType)).groupBy("_1", "_2").agg(count("_1")).as[(String,String,Long)]
@ -94,11 +91,8 @@ object SparkCreateScholix {
.map(_._2)
.write.mode(SaveMode.Overwrite).save(s"$targetPath/related_entities")
val relatedEntitiesDS:Dataset[RelatedEntities] = spark.read.load(s"$targetPath/related_entities").as[RelatedEntities].filter(r => r.relatedPublication>0 || r.relatedDataset > 0)
relatedEntitiesDS.joinWith(summaryDS, relatedEntitiesDS("id").equalTo(summaryDS("_1")), "inner").map{i =>
val re = i._1
val sum = i._2._2
@ -109,8 +103,4 @@ object SparkCreateScholix {
}.write.mode(SaveMode.Overwrite).save(s"${summaryPath}_filtered")
}
}