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

WIP SparkCreateMergeRels distinct relations

This commit is contained in:
Claudio Atzori 2020-07-13 15:54:51 +02:00
parent 7dd91edf43
commit c8284bab06
3 changed files with 7 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
dhp-workflows/dhp-dedup-openaire/src/main/java/eu/dnetlib/dhp/oa/dedup/RelationAggregator.java
View File

@ -36,11 +36,11 @@ public class RelationAggregator extends Aggregator<Relation, Relation, Relation>
@Override
public Encoder<Relation> bufferEncoder() {
return Encoders.bean(Relation.class);
return Encoders.kryo(Relation.class);
}
@Override
public Encoder<Relation> outputEncoder() {
return Encoders.bean(Relation.class);
return Encoders.kryo(Relation.class);
}
}

4
dhp-workflows/dhp-dedup-openaire/src/main/java/eu/dnetlib/dhp/oa/dedup/SparkPropagateRelation.java
View File

@ -98,10 +98,10 @@ public class SparkPropagateRelation extends AbstractSparkAction {
getDeletedFn());
save(
newRels
distinctRelations(newRels
.union(updated)
.union(mergeRels)
.map((MapFunction<Relation, Relation>) r -> r, Encoders.kryo(Relation.class)),
.map((MapFunction<Relation, Relation>) r -> r, Encoders.kryo(Relation.class))),
outputRelationPath, SaveMode.Overwrite);
}

4
dhp-workflows/dhp-dedup-openaire/src/test/java/eu/dnetlib/dhp/oa/dedup/SparkDedupTest.java
View File

@ -77,11 +77,13 @@ public class SparkDedupTest implements Serializable {
FileUtils.deleteDirectory(new File(testOutputBasePath));
FileUtils.deleteDirectory(new File(testDedupGraphBasePath));
final SparkConf conf = new SparkConf();
conf.set("spark.sql.shuffle.partitions", "200");
spark = SparkSession
.builder()
.appName(SparkDedupTest.class.getSimpleName())
.master("local[*]")
.config(new SparkConf())
.config(conf)
.getOrCreate();
jsc = JavaSparkContext.fromSparkContext(spark.sparkContext());