forked from D-Net/dnet-hadoop
[Resolve Entities] Change test dataset to mirror the modification in the creation of the map between the pids and the unresolved
This commit is contained in:
parent
49f897ef29
commit
6d4a1c57ee
|
@ -52,8 +52,6 @@ class ResolveEntitiesTest extends Serializable {
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
def generateUpdates(spark:SparkSession):Unit = {
|
def generateUpdates(spark:SparkSession):Unit = {
|
||||||
val template = Source.fromInputStream(this.getClass.getResourceAsStream("updates")).mkString
|
val template = Source.fromInputStream(this.getClass.getResourceAsStream("updates")).mkString
|
||||||
|
|
||||||
|
@ -152,7 +150,6 @@ class ResolveEntitiesTest extends Serializable {
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
val pubDS:Dataset[Result] = spark.read.text(s"$workingDir/work/resolvedGraph/publication").as[String].map(s => SparkResolveEntities.deserializeObject(s, EntityType.publication))
|
val pubDS:Dataset[Result] = spark.read.text(s"$workingDir/work/resolvedGraph/publication").as[String].map(s => SparkResolveEntities.deserializeObject(s, EntityType.publication))
|
||||||
val t = pubDS.filter(p => p.getTitle!=null && p.getSubject!=null).filter(p => p.getTitle.asScala.exists(t => t.getValue.equalsIgnoreCase("FAKETITLE"))).count()
|
val t = pubDS.filter(p => p.getTitle!=null && p.getSubject!=null).filter(p => p.getTitle.asScala.exists(t => t.getValue.equalsIgnoreCase("FAKETITLE"))).count()
|
||||||
|
|
||||||
|
|
File diff suppressed because one or more lines are too long
Loading…
Reference in New Issue