enrico.ottonello
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

code formatting

This commit is contained in:
Claudio Atzori 2021-04-16 17:28:25 +02:00
parent 7ad573d023
commit 45057440c1
1 changed files with 12 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
dhp-workflows/dhp-dedup-openaire/src/main/java/eu/dnetlib/dhp/oa/dedup/SparkPropagateRelation.java
View File

@ -117,12 +117,13 @@ public class SparkPropagateRelation extends AbstractSparkAction {
private static Dataset<Relation> createNewRels(
Dataset<Relation> rels,
Dataset<Tuple2<String,String>> mergedIds,
Dataset<Tuple2<String, String>> mergedIds,
MapFunction<Tuple2<Tuple2<Tuple3<String, Relation, String>, Tuple2<String, String>>, Tuple2<String, String>>, Relation> mapRel) {
Dataset<Tuple3<String, Relation, String>> mapped = rels
.map(
(MapFunction<Relation, Tuple3<String, Relation, String>>) r -> new Tuple3<>(getId(r, FieldType.SOURCE), r, getId(r, FieldType.TARGET)),
(MapFunction<Relation, Tuple3<String, Relation, String>>) r -> new Tuple3<>(getId(r, FieldType.SOURCE),
r, getId(r, FieldType.TARGET)),
Encoders.tuple(Encoders.STRING(), Encoders.kryo(Relation.class), Encoders.STRING()));
Dataset<Tuple2<Tuple3<String, Relation, String>, Tuple2<String, String>>> relSource = mapped
@ -130,7 +131,9 @@ public class SparkPropagateRelation extends AbstractSparkAction {
return relSource
.joinWith(mergedIds, relSource.col("_1._3").equalTo(mergedIds.col("_2")), "left_outer")
.filter((FilterFunction<Tuple2<Tuple2<Tuple3<String, Relation, String>, Tuple2<String, String>>, Tuple2<String, String>>>) r -> r._2() != null || r._1() != null)
.filter(
(FilterFunction<Tuple2<Tuple2<Tuple3<String, Relation, String>, Tuple2<String, String>>, Tuple2<String, String>>>) r -> r
._2() != null || r._1() != null)
.map(mapRel, Encoders.bean(Relation.class))
.distinct();
}