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

HACK: patch NULL values with defaults found in result.datainfo.deletedbyinference and result.context

This commit is contained in:
Claudio Atzori 2020-05-26 10:28:35 +02:00
parent 7b288a94cb
commit 55595d7235
1 changed files with 15 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
dhp-workflows/dhp-enrichment/src/main/java/eu/dnetlib/dhp/bulktag/SparkBulkTagJob.java
View File

@ -4,6 +4,7 @@ package eu.dnetlib.dhp.bulktag;
import static eu.dnetlib.dhp.PropagationConstant.removeOutputDir;
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
import java.util.ArrayList;
import java.util.Optional;
import org.apache.commons.io.IOUtils;
@ -100,6 +101,7 @@ public class SparkBulkTagJob {
ResultTagger resultTagger = new ResultTagger();
readPath(spark, inputPath, resultClazz)
.map(patchResult(), Encoders.bean(resultClazz))
.map(
(MapFunction<R, R>) value -> resultTagger
.enrichContextCriteria(
@ -119,4 +121,17 @@ public class SparkBulkTagJob {
.map((MapFunction<String, R>) value -> OBJECT_MAPPER.readValue(value, clazz), Encoders.bean(clazz));
}
// TODO remove this hack as soon as the values fixed by this method will be provided as NON null
private static <R extends Result> MapFunction<R, R> patchResult() {
return (MapFunction<R, R>) r -> {
if (r.getDataInfo().getDeletedbyinference() == null) {
r.getDataInfo().setDeletedbyinference(false);
}
if (r.getContext() == null) {
r.setContext(new ArrayList<>());
}
return r;
};
}
}