154 lines
5.6 KiB
Java
154 lines
5.6 KiB
Java
|
|
package eu.dnetlib.dhp.oa.graph.dump.subset;
|
|
|
|
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
|
|
|
|
import java.io.Serializable;
|
|
import java.util.Optional;
|
|
|
|
import org.apache.commons.io.IOUtils;
|
|
import org.apache.spark.SparkConf;
|
|
import org.apache.spark.api.java.function.ForeachFunction;
|
|
import org.apache.spark.api.java.function.MapFunction;
|
|
import org.apache.spark.sql.Dataset;
|
|
import org.apache.spark.sql.Encoders;
|
|
import org.apache.spark.sql.SaveMode;
|
|
import org.apache.spark.sql.SparkSession;
|
|
import org.slf4j.Logger;
|
|
import org.slf4j.LoggerFactory;
|
|
|
|
import com.fasterxml.jackson.databind.ObjectMapper;
|
|
|
|
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
|
|
import eu.dnetlib.dhp.oa.graph.dump.Utils;
|
|
import eu.dnetlib.dhp.oa.model.graph.GraphResult;
|
|
import eu.dnetlib.dhp.oa.model.graph.Relation;
|
|
import eu.dnetlib.dhp.oa.model.graph.ResearchCommunity;
|
|
import scala.Tuple2;
|
|
|
|
/**
|
|
* @author miriam.baglioni
|
|
* @Date 15/11/22
|
|
*/
|
|
public class SparkSelectValidRelationContext implements Serializable {
|
|
private static final Logger log = LoggerFactory.getLogger(SparkSelectValidRelationContext.class);
|
|
|
|
public static void main(String[] args) throws Exception {
|
|
String jsonConfiguration = IOUtils
|
|
.toString(
|
|
SparkSelectValidRelationContext.class
|
|
.getResourceAsStream(
|
|
"/eu/dnetlib/dhp/oa/graph/dump/input_select_valid_relation_context_parameters.json"));
|
|
|
|
final ArgumentApplicationParser parser = new ArgumentApplicationParser(jsonConfiguration);
|
|
parser.parseArgument(args);
|
|
|
|
Boolean isSparkSessionManaged = Optional
|
|
.ofNullable(parser.get("isSparkSessionManaged"))
|
|
.map(Boolean::valueOf)
|
|
.orElse(Boolean.TRUE);
|
|
log.info("isSparkSessionManaged: {}", isSparkSessionManaged);
|
|
|
|
// results dumped
|
|
final String inputPath = parser.get("sourcePath");
|
|
log.info("inputPath: {}", inputPath);
|
|
|
|
final String contextRelationPath = parser.get("contextRelationPath");
|
|
log.info("contextRelationPath: {}", contextRelationPath);
|
|
|
|
SparkConf conf = new SparkConf();
|
|
|
|
runWithSparkSession(
|
|
conf,
|
|
isSparkSessionManaged,
|
|
spark -> {
|
|
selectValidRelation(spark, inputPath, contextRelationPath);
|
|
|
|
});
|
|
|
|
}
|
|
|
|
private static void selectValidRelation(SparkSession spark, String inputPath,
|
|
String contextRelationPath) {
|
|
// read the results
|
|
Dataset<String> dumpedIds = Utils
|
|
.readPath(spark, inputPath + "/publication", GraphResult.class)
|
|
.map((MapFunction<GraphResult, String>) r -> r.getId(), Encoders.STRING())
|
|
.union(
|
|
Utils
|
|
.readPath(spark, inputPath + "/dataset", GraphResult.class)
|
|
.map((MapFunction<GraphResult, String>) r -> r.getId(), Encoders.STRING()))
|
|
.union(
|
|
Utils
|
|
.readPath(spark, inputPath + "/software", GraphResult.class)
|
|
.map((MapFunction<GraphResult, String>) r -> r.getId(), Encoders.STRING()))
|
|
.union(
|
|
Utils
|
|
.readPath(spark, inputPath + "/otherresearchproduct", GraphResult.class)
|
|
.map((MapFunction<GraphResult, String>) r -> r.getId(), Encoders.STRING()))
|
|
.union(
|
|
Utils
|
|
.readPath(spark, inputPath + "/organization", eu.dnetlib.dhp.oa.model.graph.Organization.class)
|
|
.map(
|
|
(MapFunction<eu.dnetlib.dhp.oa.model.graph.Organization, String>) o -> o.getId(),
|
|
Encoders.STRING()))
|
|
.union(
|
|
Utils
|
|
.readPath(spark, inputPath + "/project", eu.dnetlib.dhp.oa.model.graph.Project.class)
|
|
.map(
|
|
(MapFunction<eu.dnetlib.dhp.oa.model.graph.Project, String>) o -> o.getId(), Encoders.STRING()))
|
|
.union(
|
|
Utils
|
|
.readPath(spark, inputPath + "/datasource", eu.dnetlib.dhp.oa.model.graph.Datasource.class)
|
|
.map(
|
|
(MapFunction<eu.dnetlib.dhp.oa.model.graph.Datasource, String>) o -> o.getId(),
|
|
Encoders.STRING()));
|
|
|
|
Dataset<Tuple2<String, Relation>> relationSource = Utils
|
|
.readPath(spark, contextRelationPath + "/context", Relation.class)
|
|
.union(Utils.readPath(spark, contextRelationPath + "/contextOrg", Relation.class))
|
|
.map(
|
|
(MapFunction<Relation, Tuple2<String, Relation>>) r -> new Tuple2<>(r.getSource(), r),
|
|
Encoders.tuple(Encoders.STRING(), Encoders.bean(Relation.class)));
|
|
|
|
Dataset<ResearchCommunity> allowedContext = Utils
|
|
.readPath(spark, inputPath + "/communities_infrastructures", ResearchCommunity.class);
|
|
|
|
Dataset<Tuple2<String, Relation>> relJoinSource = relationSource
|
|
.joinWith(dumpedIds, relationSource.col("_1").equalTo(dumpedIds.col("value")))
|
|
.map(
|
|
(MapFunction<Tuple2<Tuple2<String, Relation>, String>, Tuple2<String, Relation>>) t2 -> new Tuple2<>(
|
|
t2._1()._2().getTarget(), t2._1()._2()),
|
|
Encoders.tuple(Encoders.STRING(), Encoders.bean(Relation.class)));
|
|
|
|
relJoinSource
|
|
.joinWith(allowedContext, relJoinSource.col("_1").equalTo(allowedContext.col("id")))
|
|
.map(
|
|
(MapFunction<Tuple2<Tuple2<String, Relation>, ResearchCommunity>, Relation>) t2 -> t2._1()._2(),
|
|
Encoders.bean(eu.dnetlib.dhp.oa.model.graph.Relation.class))
|
|
.write()
|
|
.mode(SaveMode.Append)
|
|
.option("compression", "gzip")
|
|
.json(inputPath + "/relation");
|
|
|
|
relJoinSource = relationSource
|
|
.joinWith(allowedContext, relationSource.col("_1").equalTo(allowedContext.col("id")))
|
|
.map(
|
|
(MapFunction<Tuple2<Tuple2<String, Relation>, ResearchCommunity>, Tuple2<String, Relation>>) t2 -> new Tuple2<>(
|
|
t2._1()._2().getTarget(), t2._1()._2()),
|
|
Encoders.tuple(Encoders.STRING(), Encoders.bean(Relation.class)));
|
|
|
|
relJoinSource
|
|
.joinWith(dumpedIds, relJoinSource.col("_1").equalTo(dumpedIds.col("value")))
|
|
.map(
|
|
(MapFunction<Tuple2<Tuple2<String, Relation>, String>, Relation>) t2 -> t2._1()._2(),
|
|
Encoders.bean(Relation.class))
|
|
.write()
|
|
.mode(SaveMode.Append)
|
|
.option("compression", "gzip")
|
|
.json(inputPath + "/relation");
|
|
|
|
}
|
|
|
|
}
|