sabeel
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
0
Fork 0
Code Pull Requests Activity

Hosted By Map - refactoring

This commit is contained in:
Miriam Baglioni 2021-08-04 10:16:30 +02:00
parent 1e952cccf6
commit eccf3851b0
1 changed files with 1 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/hostedbymap/SparkProduceHostedByMap.scala
View File

@ -23,10 +23,6 @@ object SparkProduceHostedByMap {
implicit val tupleForJoinEncoder: Encoder[(String, HostedByItemType)] = Encoders.tuple(Encoders.STRING, Encoders.product[HostedByItemType])
def toHostedByItemType(input: ((HostedByInfo, HostedByInfo), HostedByInfo)) : HostedByItemType = {
val openaire: HostedByInfo = input._1._1
val doaj: HostedByInfo = input._1._2
@ -217,7 +213,7 @@ object SparkProduceHostedByMap {
.union(doajHostedByDataset(spark, workingDirPath + "/doaj"))
.flatMap(hbi => toList(hbi))).filter(hbi => hbi._2.id.startsWith("10|"))
.map(hbi => toHostedByMap(hbi))(Encoders.STRING)
.rdd.saveAsTextFile(outputPath + "/hostedByMap", classOf[GzipCodec])
.rdd.saveAsTextFile(outputPath , classOf[GzipCodec])
}