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

[Actionmanager] zero function considers empty entity.id as well as rel.source/rel.target

This commit is contained in:
Claudio Atzori 2021-03-23 17:34:32 +01:00
parent 1e423fdc07
commit 751125fdf9
3 changed files with 7 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
dhp-workflows/dhp-actionmanager/src/main/java/eu/dnetlib/dhp/actionmanager/promote/PromoteActionPayloadForGraphTableJob.java
View File

@ -5,12 +5,12 @@ import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
import static eu.dnetlib.dhp.schema.common.ModelSupport.isSubClass; import static eu.dnetlib.dhp.schema.common.ModelSupport.isSubClass;
import java.io.IOException; import java.io.IOException;
import java.util.Objects;
import java.util.Optional; import java.util.Optional;
import java.util.function.BiFunction; import java.util.function.BiFunction;
import java.util.function.Function; import java.util.function.Function;
import org.apache.commons.io.IOUtils; import org.apache.commons.io.IOUtils;
import org.apache.commons.lang3.StringUtils;
import org.apache.spark.SparkConf; import org.apache.spark.SparkConf;
import org.apache.spark.api.java.function.MapFunction; import org.apache.spark.api.java.function.MapFunction;
import org.apache.spark.sql.Dataset; import org.apache.spark.sql.Dataset;
@ -194,7 +194,7 @@ public class PromoteActionPayloadForGraphTableJob {
SerializableSupplier<BiFunction<G, A, G>> mergeRowWithActionPayloadAndGetFn = MergeAndGet.functionFor(strategy); SerializableSupplier<BiFunction<G, A, G>> mergeRowWithActionPayloadAndGetFn = MergeAndGet.functionFor(strategy);
SerializableSupplier<BiFunction<G, G, G>> mergeRowsAndGetFn = MergeAndGet.functionFor(strategy); SerializableSupplier<BiFunction<G, G, G>> mergeRowsAndGetFn = MergeAndGet.functionFor(strategy);
SerializableSupplier<G> zeroFn = zeroFn(rowClazz); SerializableSupplier<G> zeroFn = zeroFn(rowClazz);
SerializableSupplier<Function<G, Boolean>> isNotZeroFn = PromoteActionPayloadForGraphTableJob::isNotZeroFnUsingIdOrSource; SerializableSupplier<Function<G, Boolean>> isNotZeroFn = PromoteActionPayloadForGraphTableJob::isNotZeroFnUsingIdOrSourceAndTarget;
Dataset<G> joinedAndMerged = PromoteActionPayloadFunctions Dataset<G> joinedAndMerged = PromoteActionPayloadFunctions
.joinGraphTableWithActionPayloadAndMerge( .joinGraphTableWithActionPayloadAndMerge(
@ -238,12 +238,13 @@ public class PromoteActionPayloadForGraphTableJob {
} }
} }
private static <T extends Oaf> Function<T, Boolean> isNotZeroFnUsingIdOrSource() { private static <T extends Oaf> Function<T, Boolean> isNotZeroFnUsingIdOrSourceAndTarget() {
return t -> { return t -> {
if (isSubClass(t, Relation.class)) { if (isSubClass(t, Relation.class)) {
return Objects.nonNull(((Relation) t).getSource()); final Relation rel = (Relation) t;
return StringUtils.isNotBlank(rel.getSource()) && StringUtils.isNotBlank(rel.getTarget());
} }
return Objects.nonNull(((OafEntity) t).getId()); return StringUtils.isNotBlank(((OafEntity) t).getId());
}; };
} }

1
dhp-workflows/dhp-actionmanager/src/main/java/eu/dnetlib/dhp/actionmanager/promote/PromoteActionPayloadFunctions.java
View File

@ -111,7 +111,6 @@ public class PromoteActionPayloadFunctions {
SerializableSupplier<Function<G, Boolean>> isNotZeroFn, SerializableSupplier<Function<G, Boolean>> isNotZeroFn,
Class<G> rowClazz) { Class<G> rowClazz) {
TypedColumn<G, G> aggregator = new TableAggregator<>(zeroFn, mergeAndGetFn, isNotZeroFn, rowClazz).toColumn(); TypedColumn<G, G> aggregator = new TableAggregator<>(zeroFn, mergeAndGetFn, isNotZeroFn, rowClazz).toColumn();
return rowDS return rowDS
.filter((FilterFunction<G>) o -> isNotZeroFn.get().apply(o)) .filter((FilterFunction<G>) o -> isNotZeroFn.get().apply(o))
.groupByKey((MapFunction<G, String>) x -> rowIdFn.get().apply(x), Encoders.STRING()) .groupByKey((MapFunction<G, String>) x -> rowIdFn.get().apply(x), Encoders.STRING())

2
dhp-workflows/dhp-graph-mapper/src/test/java/eu/dnetlib/dhp/oa/graph/raw/MappersTest.java
View File

@ -11,7 +11,6 @@ import java.io.IOException;
import java.util.List; import java.util.List;
import java.util.Optional; import java.util.Optional;
import eu.dnetlib.dhp.schema.oaf.utils.PidType;
import org.apache.commons.io.IOUtils; import org.apache.commons.io.IOUtils;
import org.apache.commons.lang3.StringUtils; import org.apache.commons.lang3.StringUtils;
import org.junit.jupiter.api.BeforeEach; import org.junit.jupiter.api.BeforeEach;
@ -33,6 +32,7 @@ import eu.dnetlib.dhp.schema.oaf.Publication;
import eu.dnetlib.dhp.schema.oaf.Relation; import eu.dnetlib.dhp.schema.oaf.Relation;
import eu.dnetlib.dhp.schema.oaf.Software; import eu.dnetlib.dhp.schema.oaf.Software;
import eu.dnetlib.dhp.schema.oaf.StructuredProperty; import eu.dnetlib.dhp.schema.oaf.StructuredProperty;
import eu.dnetlib.dhp.schema.oaf.utils.PidType;
import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpService; import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpService;
@ExtendWith(MockitoExtension.class) @ExtendWith(MockitoExtension.class)