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

[EOSC TAG] added logic in case subject is null

This commit is contained in:
Miriam Baglioni 2022-04-22 18:34:03 +02:00
parent 27c85e901a
commit dfbd2bcbea
1 changed files with 22 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
dhp-workflows/dhp-enrichment/src/main/java/eu/dnetlib/dhp/bulktag/SparkEoscTag.java
View File

@ -14,6 +14,9 @@ import org.apache.spark.sql.SparkSession;
import org.slf4j.Logger; import org.slf4j.Logger;
import org.slf4j.LoggerFactory; import org.slf4j.LoggerFactory;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.List;
import java.util.Optional; import java.util.Optional;
import static eu.dnetlib.dhp.PropagationConstant.readPath; import static eu.dnetlib.dhp.PropagationConstant.readPath;
@ -72,11 +75,17 @@ public class SparkEoscTag {
readPath(spark, inputPath + "/software", Software.class) readPath(spark, inputPath + "/software", Software.class)
.map((MapFunction<Software, Software>) s -> { .map((MapFunction<Software, Software>) s -> {
List<StructuredProperty> sbject;
if (!Optional.ofNullable(s.getSubject()).isPresent())
s.setSubject(new ArrayList<>());
sbject = s.getSubject();
if(containsCriteriaNotebook(s)){ if(containsCriteriaNotebook(s)){
s.getSubject().add(EOSC_NOTEBOOK); sbject.add(EOSC_NOTEBOOK);
} }
if(containsCriteriaGalaxy(s)){ if(containsCriteriaGalaxy(s)){
s.getSubject().add(EOSC_GALAXY); sbject.add(EOSC_GALAXY);
} }
return s; return s;
}, Encoders.bean(Software.class) ) }, Encoders.bean(Software.class) )
@ -94,11 +103,15 @@ public class SparkEoscTag {
readPath(spark, inputPath + "/otherresearchproduct", OtherResearchProduct.class) readPath(spark, inputPath + "/otherresearchproduct", OtherResearchProduct.class)
.map((MapFunction<OtherResearchProduct, OtherResearchProduct>) orp -> .map((MapFunction<OtherResearchProduct, OtherResearchProduct>) orp ->
{ {
List<StructuredProperty> sbject;
if (!Optional.ofNullable(orp.getSubject()).isPresent())
orp.setSubject(new ArrayList<>());
sbject = orp.getSubject();
if(containsCriteriaGalaxy(orp)){ if(containsCriteriaGalaxy(orp)){
orp.getSubject().add(EOSC_GALAXY); sbject.add(EOSC_GALAXY);
} }
if(containscriteriaTwitter(orp)){ if(containscriteriaTwitter(orp)){
orp.getSubject().add(EOSC_TWITTER); sbject.add(EOSC_TWITTER);
} }
return orp; return orp;
}, Encoders.bean(OtherResearchProduct.class)) }, Encoders.bean(OtherResearchProduct.class))
@ -115,8 +128,12 @@ public class SparkEoscTag {
readPath(spark, inputPath + "/dataset", Dataset.class) readPath(spark, inputPath + "/dataset", Dataset.class)
.map((MapFunction<Dataset, Dataset>) d -> { .map((MapFunction<Dataset, Dataset>) d -> {
List<StructuredProperty> sbject;
if (!Optional.ofNullable(d.getSubject()).isPresent())
d.setSubject(new ArrayList<>());
sbject = d.getSubject();
if(containscriteriaTwitter(d)){ if(containscriteriaTwitter(d)){
d.getSubject().add(EOSC_TWITTER); sbject.add(EOSC_TWITTER);
} }
return d; return d;
} , Encoders.bean(Dataset.class) ) } , Encoders.bean(Dataset.class) )