144 lines
5.1 KiB
Java
144 lines
5.1 KiB
Java
|
|
package eu.dnetlib.dhp.oa.graph.dump;
|
|
|
|
import java.io.BufferedReader;
|
|
import java.io.IOException;
|
|
import java.io.InputStreamReader;
|
|
|
|
import org.apache.hadoop.fs.FileSystem;
|
|
import org.apache.hadoop.fs.Path;
|
|
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 com.fasterxml.jackson.databind.ObjectMapper;
|
|
import com.google.gson.Gson;
|
|
|
|
import eu.dnetlib.dhp.common.HdfsSupport;
|
|
import eu.dnetlib.dhp.oa.graph.dump.community.CommunityMap;
|
|
import eu.dnetlib.dhp.oa.graph.dump.complete.Constants;
|
|
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 eu.dnetlib.dhp.utils.DHPUtils;
|
|
import eu.dnetlib.dhp.utils.ISLookupClientFactory;
|
|
import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpService;
|
|
import scala.Tuple2;
|
|
|
|
public class Utils {
|
|
public static final ObjectMapper OBJECT_MAPPER = new ObjectMapper();
|
|
|
|
private Utils() {
|
|
}
|
|
|
|
public static void removeOutputDir(SparkSession spark, String path) {
|
|
HdfsSupport.remove(path, spark.sparkContext().hadoopConfiguration());
|
|
}
|
|
|
|
public static <R> Dataset<R> readPath(
|
|
SparkSession spark, String inputPath, Class<R> clazz) {
|
|
return spark
|
|
.read()
|
|
.textFile(inputPath)
|
|
.map((MapFunction<String, R>) value -> OBJECT_MAPPER.readValue(value, clazz), Encoders.bean(clazz));
|
|
}
|
|
|
|
public static ISLookUpService getIsLookUpService(String isLookUpUrl) {
|
|
return ISLookupClientFactory.getLookUpService(isLookUpUrl);
|
|
}
|
|
|
|
public static String getContextId(String id) {
|
|
|
|
return String
|
|
.format(
|
|
"%s|%s::%s", Constants.CONTEXT_ID, Constants.CONTEXT_NS_PREFIX,
|
|
DHPUtils.md5(id));
|
|
}
|
|
|
|
public static CommunityMap getCommunityMap(SparkSession spark, String communityMapPath) {
|
|
|
|
return new Gson().fromJson(spark.read().textFile(communityMapPath).collectAsList().get(0), CommunityMap.class);
|
|
|
|
}
|
|
|
|
public static CommunityMap readCommunityMap(FileSystem fileSystem, String communityMapPath) throws IOException {
|
|
BufferedReader br = new BufferedReader(new InputStreamReader(fileSystem.open(new Path(communityMapPath))));
|
|
StringBuilder sb = new StringBuilder();
|
|
try {
|
|
String line;
|
|
while ((line = br.readLine()) != null) {
|
|
sb.append(line);
|
|
}
|
|
} finally {
|
|
br.close();
|
|
|
|
}
|
|
|
|
return new Gson().fromJson(sb.toString(), CommunityMap.class);
|
|
}
|
|
|
|
public static Dataset<String> getEntitiesId(SparkSession spark, String inputPath) {
|
|
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()))
|
|
.union(
|
|
Utils
|
|
.readPath(spark, inputPath + "/communities_infrastructures", ResearchCommunity.class)
|
|
.map((MapFunction<ResearchCommunity, String>) c -> c.getId(), Encoders.STRING()));
|
|
return dumpedIds;
|
|
}
|
|
|
|
public static Dataset<Relation> getValidRelations(SparkSession spark, Dataset<Relation> relations,
|
|
Dataset<String> entitiesIds) {
|
|
Dataset<Tuple2<String, Relation>> relationSource = relations
|
|
.map(
|
|
(MapFunction<Relation, Tuple2<String, Relation>>) r -> new Tuple2<>(r.getSource().getId(), r),
|
|
Encoders.tuple(Encoders.STRING(), Encoders.bean(Relation.class)));
|
|
|
|
Dataset<Tuple2<String, Relation>> relJoinSource = relationSource
|
|
.joinWith(entitiesIds, relationSource.col("_1").equalTo(entitiesIds.col("value")))
|
|
.map(
|
|
(MapFunction<Tuple2<Tuple2<String, Relation>, String>, Tuple2<String, Relation>>) t2 -> new Tuple2<>(
|
|
t2._1()._2().getTarget().getId(), t2._1()._2()),
|
|
Encoders.tuple(Encoders.STRING(), Encoders.bean(Relation.class)));
|
|
|
|
return relJoinSource
|
|
.joinWith(entitiesIds, relJoinSource.col("_1").equalTo(entitiesIds.col("value")))
|
|
.map(
|
|
(MapFunction<Tuple2<Tuple2<String, Relation>, String>, Relation>) t2 -> t2._1()._2(),
|
|
Encoders.bean(Relation.class));
|
|
}
|
|
|
|
}
|