new implementation for result to project propagation. Use the prepared info in propagation

This commit is contained in:
Miriam Baglioni 2020-04-14 16:46:07 +02:00
parent ceb1f299bf
commit cadab9b81d
1 changed files with 120 additions and 98 deletions

View File

@ -1,143 +1,160 @@
package eu.dnetlib.dhp.projecttoresult;
import com.fasterxml.jackson.databind.ObjectMapper;
import com.google.gson.Gson;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.countrypropagation.DatasourceCountry;
import eu.dnetlib.dhp.countrypropagation.PrepareDatasourceCountryAssociation;
import eu.dnetlib.dhp.countrypropagation.ResultCountrySet;
import eu.dnetlib.dhp.schema.oaf.Relation;
import org.apache.arrow.flatbuf.Bool;
import org.apache.commons.io.IOUtils;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.io.compress.GzipCodec;
import org.apache.spark.SparkConf;
import org.apache.spark.api.java.JavaRDD;
import org.apache.spark.api.java.JavaSparkContext;
import org.apache.spark.sql.Dataset;
import org.apache.spark.sql.Encoders;
import org.apache.spark.sql.Row;
import org.apache.spark.sql.SparkSession;
import org.apache.spark.api.java.function.FlatMapFunction;
import org.apache.spark.api.java.function.FlatMapFunction2;
import org.apache.spark.api.java.function.MapFunction;
import org.apache.spark.sql.*;
import org.omg.CORBA.OBJ_ADAPTER;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import scala.Tuple2;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.List;
import java.util.Optional;
import java.util.stream.Collectors;
import static eu.dnetlib.dhp.PropagationConstant.*;
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
public class SparkResultToProjectThroughSemRelJob3 {
private static final Logger log = LoggerFactory.getLogger(PrepareDatasourceCountryAssociation.class);
private static final ObjectMapper OBJECT_MAPPER = new ObjectMapper();
public static void main(String[] args) throws Exception {
String jsonConfiguration = IOUtils.toString(SparkResultToProjectThroughSemRelJob3.class
.getResourceAsStream("/eu/dnetlib/dhp/projecttoresult/input_projecttoresult_parameters.json"));
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
IOUtils.toString(SparkResultToProjectThroughSemRelJob3.class
.getResourceAsStream("/eu/dnetlib/dhp/projecttoresult/input_projecttoresult_parameters.json")));
jsonConfiguration);
parser.parseArgument(args);
Boolean isSparkSessionManaged = isSparkSessionManaged(parser);
log.info("isSparkSessionManaged: {}", isSparkSessionManaged);
String inputPath = parser.get("sourcePath");
log.info("inputPath: {}", inputPath);
final String outputPath = parser.get("outputPath");
log.info("outputPath {}: ", outputPath);
final String potentialUpdatePath = parser.get("potentialUpdatePath");
log.info("potentialUpdatePath {}: ", potentialUpdatePath);
final String alreadyLinkedPath = parser.get("alreadyLinkedPath");
log.info("alreadyLinkedPath {}: ", alreadyLinkedPath);
final Boolean writeUpdates = Boolean.valueOf(parser.get("writeUpdate"));
log.info("writeUpdate: {}", writeUpdates);
final Boolean saveGraph = Boolean.valueOf(parser.get("saveGraph"));
log.info("saveGraph: {}", saveGraph);
SparkConf conf = new SparkConf();
conf.set("hive.metastore.uris", parser.get("hive_metastore_uris"));
final SparkSession spark = SparkSession
.builder()
.appName(SparkResultToProjectThroughSemRelJob3.class.getSimpleName())
.master(parser.get("master"))
.config(conf)
.enableHiveSupport()
.getOrCreate();
final JavaSparkContext sc = new JavaSparkContext(spark.sparkContext());
final String inputPath = parser.get("sourcePath");
final String outputPath = "/tmp/provision/propagation/projecttoresult";
boolean writeUpdates = "true".equals(parser.get("writeUpdate"));
boolean saveGraph = "true".equals(parser.get("saveGraph"));
runWithSparkSession(conf, isSparkSessionManaged,
spark -> {
//createOutputDirs(outputPath, FileSystem.get(spark.sparkContext().hadoopConfiguration()));
if(isTest(parser)) {
removeOutputDir(spark, outputPath);
}
execPropagation(spark, inputPath, outputPath, alreadyLinkedPath, potentialUpdatePath, writeUpdates, saveGraph);
});
}
final List<String> allowedsemrel = Arrays.asList(parser.get("allowedsemrels").split(";"));
createOutputDirs(outputPath, FileSystem.get(spark.sparkContext().hadoopConfiguration()));
private static void execPropagation(SparkSession spark, String inputPath, String outputPath, String alreadyLinkedPath, String potentialUpdatePath,
Boolean writeUpdate, Boolean saveGraph){
Dataset<Relation> relation = spark.createDataset(sc.textFile(inputPath + "/relation")
.map(item -> new ObjectMapper().readValue(item, Relation.class)).rdd(), Encoders.bean(Relation.class));
relation.createOrReplaceTempView("relation");
Dataset<ProjectResultSet> toaddrelations = readAssocProjectResults(spark, potentialUpdatePath);
Dataset<ProjectResultSet> alreadyLinked = readAssocProjectResults(spark, alreadyLinkedPath);
String query = "Select source, target " +
"from relation " +
"where datainfo.deletedbyinference = false and relClass = '" + RELATION_RESULT_PROJECT_REL_CLASS + "'";
if(writeUpdate){
toaddrelations
.toJSON()
.write()
.mode(SaveMode.Overwrite)
.option("compression","gzip")
.text(outputPath +"/potential_updates");
//writeUpdates(toaddrelations.toJavaRDD(), outputPath + "/potential_updates");
}
if (saveGraph){
getNewRelations(alreadyLinked, toaddrelations)
.toJSON()
.write()
.mode(SaveMode.Append)
.option("compression", "gzip")
.text(outputPath);
// JavaSparkContext sc = new JavaSparkContext(spark.sparkContext());
// sc.textFile(inputPath)
// .map(item -> OBJECT_MAPPER.readValue(item, Relation.class))
// .union(getNewRelations(alreadyLinked, toaddrelations)
// .toJavaRDD())
// .map(r -> OBJECT_MAPPER.writeValueAsString(r))
// .saveAsTextFile(outputPath , GzipCodec.class);
}
Dataset<Row> resproj_relation = spark.sql(query);
query = "Select source, target " +
"from relation " +
"where datainfo.deletedbyinference = false " + getConstraintList(" relClass = '", allowedsemrel );
}
Dataset<Row> resres_relation = spark.sql(query);
resres_relation.createOrReplaceTempView("resres_relation");
resproj_relation.createOrReplaceTempView("resproj_relation");
private static Dataset<Relation> getNewRelations(Dataset<ProjectResultSet> alreadyLinked,
Dataset<ProjectResultSet> toaddrelations){
query ="SELECT proj, collect_set(r1target) result_set " +
"FROM (" +
" SELECT r1.source as source, r1.target as r1target, r2.target as proj " +
" FROM resres_relation r1 " +
" JOIN resproj_relation r2 " +
" ON r1.source = r2.source " +
" ) tmp " +
"GROUP BY proj ";
Dataset<Row> toaddrelations = spark.sql(query);
query = "select target, collect_set(source) result_list from " +
"resproj_relation " +
"group by target";
Dataset<Row> project_resultlist = spark.sql(query);
//if (writeUpdaes){
toaddrelations.toJavaRDD().map(r->new ObjectMapper().writeValueAsString(r))
.saveAsTextFile(outputPath + "/toupdaterelations");
//}
if(saveGraph){
JavaRDD<Relation> new_relations = toaddrelations.toJavaRDD().mapToPair(r -> new Tuple2<>(r.getString(0), r.getList(1)))
.leftOuterJoin(project_resultlist.toJavaRDD().mapToPair(pr -> new Tuple2<>(pr.getString(0), pr.getList(1))))
.flatMap(c -> {
List<Object> toAddRel = new ArrayList<>();
toAddRel.addAll(c._2()._1());
if (c._2()._2().isPresent()) {
List<Object> originalRels = c._2()._2().get();
for (Object o : originalRels) {
if (toAddRel.contains(o)) {
toAddRel.remove(o);
return toaddrelations
.joinWith(alreadyLinked, toaddrelations.col("projectId").equalTo(alreadyLinked.col("projectId")), "left")
.flatMap(value -> {
List<Relation> new_relations = new ArrayList<>();
ProjectResultSet potential_update = value._1();
ProjectResultSet already_linked = value._2();
String projId = already_linked.getProjectId();
potential_update
.getResultSet()
.stream()
.forEach(rId -> {
if (!already_linked.getResultSet().contains(rId)){
new_relations.add(getRelation(rId, projId, RELATION_RESULT_PROJECT_REL_CLASS, RELATION_RESULTPROJECT_REL_TYPE,
RELATION_RESULTPROJECT_SUBREL_TYPE, PROPAGATION_DATA_INFO_TYPE,
PROPAGATION_RELATION_RESULT_PROJECT_SEM_REL_CLASS_ID,
PROPAGATION_RELATION_RESULT_PROJECT_SEM_REL_CLASS_NAME));
new_relations.add(getRelation(projId, rId, RELATION_PROJECT_RESULT_REL_CLASS, RELATION_RESULTPROJECT_REL_TYPE,
RELATION_RESULTPROJECT_SUBREL_TYPE, PROPAGATION_DATA_INFO_TYPE,
PROPAGATION_RELATION_RESULT_PROJECT_SEM_REL_CLASS_ID,
PROPAGATION_RELATION_RESULT_PROJECT_SEM_REL_CLASS_NAME));
}
}
}
List<Relation> relationList = new ArrayList<>();
String projId = c._1();
for (Object r : toAddRel) {
String rId = (String) r;
relationList.add(getRelation(rId, projId, RELATION_RESULT_PROJECT_REL_CLASS, RELATION_RESULTPROJECT_REL_TYPE,
RELATION_RESULTPROJECT_SUBREL_TYPE, PROPAGATION_DATA_INFO_TYPE,
PROPAGATION_RELATION_RESULT_PROJECT_SEM_REL_CLASS_ID,
PROPAGATION_RELATION_RESULT_PROJECT_SEM_REL_CLASS_NAME));
relationList.add(getRelation(projId, rId, RELATION_PROJECT_RESULT_REL_CLASS, RELATION_RESULTPROJECT_REL_TYPE,
RELATION_RESULTPROJECT_SUBREL_TYPE, PROPAGATION_DATA_INFO_TYPE,
PROPAGATION_RELATION_RESULT_PROJECT_SEM_REL_CLASS_ID,
PROPAGATION_RELATION_RESULT_PROJECT_SEM_REL_CLASS_NAME));
});
return new_relations.iterator();
}
if(relationList.size()==0){
return null;
}
return relationList.iterator();
}).filter(r -> !(r==null));
,Encoders.bean(Relation.class));
new_relations.map(r-> new ObjectMapper().writeValueAsString(r))
.saveAsTextFile(outputPath + "/new_relations" );
sc.textFile(inputPath + "/relation")
.map(item -> new ObjectMapper().readValue(item, Relation.class))
.union(new_relations)
.map(r -> new ObjectMapper().writeValueAsString(r))
.saveAsTextFile(outputPath + "/relation");
}
private static void writeUpdates(JavaRDD<ProjectResultSet> potentialUpdates, String outputPath){
potentialUpdates.map(u -> OBJECT_MAPPER.writeValueAsString(u))
.saveAsTextFile(outputPath, GzipCodec.class);
}
//JavaPairRDD<String, TypedRow> result_result = getResultResultSemRel(allowedsemrel, relations);
// JavaPairRDD<String, TypedRow> result_project = relations
@ -211,9 +228,14 @@ public class SparkResultToProjectThroughSemRelJob3 {
// newRels.union(relations).map(p -> new ObjectMapper().writeValueAsString(p))
// .saveAsTextFile(outputPath + "/relation");
//}
private static Dataset<ProjectResultSet> readAssocProjectResults(SparkSession spark, String potentialUpdatePath) {
return spark
.read()
.textFile(potentialUpdatePath)
.map(value -> OBJECT_MAPPER.readValue(value, ProjectResultSet.class), Encoders.bean(ProjectResultSet.class));
}
}