D-Net
/
dnet-hadoop
14
5
Fork 11
Code Issues 14 Pull Requests 12 Projects 2 Releases 2 Wiki Activity

[graph dedup] consistency wf should not remove the relations while dispatching the entities

This commit is contained in:
Claudio Atzori 2023-09-12 14:34:28 +02:00
parent 265180bfd2
commit dc80ab14d3
1 changed files with 5 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
dhp-common/src/main/java/eu/dnetlib/dhp/oa/merge/DispatchEntitiesSparkJob.java
View File

@ -47,17 +47,14 @@ public class DispatchEntitiesSparkJob {
String outputPath = parser.get("outputPath");
log.info("outputPath: {}", outputPath);
boolean filterInvisible = Boolean.valueOf(parser.get("filterInvisible"));
boolean filterInvisible = Boolean.parseBoolean(parser.get("filterInvisible"));
log.info("filterInvisible: {}", filterInvisible);
SparkConf conf = new SparkConf();
runWithSparkSession(
conf,
isSparkSessionManaged,
spark -> {
HdfsSupport.remove(outputPath, spark.sparkContext().hadoopConfiguration());
dispatchEntities(spark, inputPath, outputPath, filterInvisible);
});
spark -> dispatchEntities(spark, inputPath, outputPath, filterInvisible));
}
private static void dispatchEntities(
@ -72,7 +69,9 @@ public class DispatchEntitiesSparkJob {
String entityType = entry.getKey();
Class<?> clazz = entry.getValue();
final String entityPath = outputPath + "/" + entityType;
if (!entityType.equalsIgnoreCase("relation")) {
HdfsSupport.remove(entityPath, spark.sparkContext().hadoopConfiguration());
Dataset<Row> entityDF = spark
.read()
.schema(Encoders.bean(clazz).schema())
@ -91,7 +90,7 @@ public class DispatchEntitiesSparkJob {
.write()
.mode(SaveMode.Overwrite)
.option("compression", "gzip")
.json(outputPath + "/" + entityType);
.json(entityPath);
}
});
}