Add a "CleanRelation" action after the PropagateRelation to filter out all relations that have been deleted by inference or that are pointing to dangling entities #328

Merged
miriam.baglioni merged 2 commits from cleanup_relations_after_dedup into beta 2023-08-08 09:49:13 +02:00
4 changed files with 218 additions and 39 deletions

View File

@ -0,0 +1,78 @@
package eu.dnetlib.dhp.oa.dedup
import eu.dnetlib.dhp.application.ArgumentApplicationParser
import eu.dnetlib.dhp.common.HdfsSupport
import eu.dnetlib.dhp.schema.oaf.Relation
import eu.dnetlib.dhp.utils.ISLookupClientFactory
import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpService
import org.apache.commons.io.IOUtils
import org.apache.spark.SparkConf
import org.apache.spark.sql._
import org.apache.spark.sql.functions.col
import org.apache.spark.sql.types.{DataTypes, StructField, StructType}
import org.slf4j.LoggerFactory
object SparkCleanRelation {
private val log = LoggerFactory.getLogger(classOf[SparkCleanRelation])
@throws[Exception]
def main(args: Array[String]): Unit = {
val parser = new ArgumentApplicationParser(
IOUtils.toString(
classOf[SparkCleanRelation].getResourceAsStream("/eu/dnetlib/dhp/oa/dedup/cleanRelation_parameters.json")
)
)
parser.parseArgument(args)
val conf = new SparkConf
new SparkCleanRelation(parser, AbstractSparkAction.getSparkSession(conf))
.run(ISLookupClientFactory.getLookUpService(parser.get("isLookUpUrl")))
}
}
class SparkCleanRelation(parser: ArgumentApplicationParser, spark: SparkSession)
extends AbstractSparkAction(parser, spark) {
override def run(isLookUpService: ISLookUpService): Unit = {
val graphBasePath = parser.get("graphBasePath")
val inputPath = parser.get("inputPath")
val outputPath = parser.get("outputPath")
SparkCleanRelation.log.info("graphBasePath: '{}'", graphBasePath)
SparkCleanRelation.log.info("inputPath: '{}'", inputPath)
SparkCleanRelation.log.info("outputPath: '{}'", outputPath)
AbstractSparkAction.removeOutputDir(spark, outputPath)
val entities =
Seq("datasource", "project", "organization", "publication", "dataset", "software", "otherresearchproduct")
val idsSchema = StructType.fromDDL("`id` STRING, `dataInfo` STRUCT<`deletedbyinference`:BOOLEAN,`invisible`:BOOLEAN>")
val emptyIds = spark.createDataFrame(spark.sparkContext.emptyRDD[Row].setName("empty"),
idsSchema)
val ids = entities
.foldLeft(emptyIds)((ds, entity) => {
val entityPath = graphBasePath + '/' + entity
if (HdfsSupport.exists(entityPath, spark.sparkContext.hadoopConfiguration)) {
ds.union(spark.read.schema(idsSchema).json(entityPath))
} else {
ds
}
})
.filter("dataInfo.deletedbyinference != true AND dataInfo.invisible != true")
.select("id")
.distinct()
val relations = spark.read.schema(Encoders.bean(classOf[Relation]).schema).json(inputPath)
.filter("dataInfo.deletedbyinference != true AND dataInfo.invisible != true")
AbstractSparkAction.save(
relations
.join(ids, col("source") === ids("id"), "leftsemi")
.join(ids, col("target") === ids("id"), "leftsemi"),
outputPath,
SaveMode.Overwrite
)
}
}

View File

@ -0,0 +1,20 @@
[
{
"paramName": "i",
"paramLongName": "graphBasePath",
"paramDescription": "the base path of raw graph",
"paramRequired": true
},
{
"paramName": "w",
"paramLongName": "inputPath",
"paramDescription": "the path to the input relation to cleanup",
"paramRequired": true
},
{
"paramName": "o",
"paramLongName": "outputPath",
"paramDescription": "the path of the output relation cleaned",
"paramRequired": true
}
]

View File

@ -92,9 +92,34 @@
--conf spark.sql.shuffle.partitions=15000
</spark-opts>
<arg>--graphBasePath</arg><arg>${graphBasePath}</arg>
<arg>--o</arg><arg>${graphOutputPath}</arg>
<arg>--outputPath</arg><arg>${workingPath}/propagaterelation/</arg>
<arg>--workingPath</arg><arg>${workingPath}</arg>
</spark>
<ok to="CleanRelation"/>
<error to="Kill"/>
</action>
<action name="CleanRelation">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Clean Relations</name>
<class>eu.dnetlib.dhp.oa.dedup.SparkCleanRelation</class>
<jar>dhp-dedup-openaire-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory=${sparkExecutorMemory}
--executor-cores=${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.shuffle.partitions=15000
</spark-opts>
<arg>--graphBasePath</arg><arg>${graphBasePath}</arg>
<arg>--inputPath</arg><arg>${workingPath}/propagaterelation/relation</arg>
<arg>--outputPath</arg><arg>${graphOutputPath}/relation</arg>
</spark>
<ok to="group_entities"/>
<error to="Kill"/>
</action>

View File

@ -1,11 +1,32 @@
package eu.dnetlib.dhp.oa.dedup;
import static java.nio.file.Files.createTempDirectory;
import static org.apache.spark.sql.functions.count;
import static org.junit.jupiter.api.Assertions.*;
import static org.mockito.Mockito.lenient;
import com.fasterxml.jackson.databind.DeserializationFeature;
import com.fasterxml.jackson.databind.ObjectMapper;
import com.google.common.collect.Sets;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.schema.common.ModelConstants;
import eu.dnetlib.dhp.schema.oaf.*;
import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpException;
import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpService;
import eu.dnetlib.pace.util.MapDocumentUtil;
import org.apache.commons.io.FileUtils;
import org.apache.commons.io.IOUtils;
import org.apache.spark.SparkConf;
import org.apache.spark.api.java.JavaPairRDD;
import org.apache.spark.api.java.JavaRDD;
import org.apache.spark.api.java.JavaSparkContext;
import org.apache.spark.api.java.function.FilterFunction;
import org.apache.spark.api.java.function.MapFunction;
import org.apache.spark.api.java.function.PairFunction;
import org.apache.spark.sql.*;
import org.apache.spark.sql.Dataset;
import org.junit.jupiter.api.*;
import org.junit.jupiter.api.extension.ExtendWith;
import org.mockito.Mock;
import org.mockito.Mockito;
import org.mockito.junit.jupiter.MockitoExtension;
import scala.Tuple2;
import java.io.File;
import java.io.FileReader;
@ -19,36 +40,11 @@ import java.util.Optional;
import java.util.Set;
import java.util.stream.Collectors;
import org.apache.commons.io.FileUtils;
import org.apache.commons.io.IOUtils;
import org.apache.spark.SparkConf;
import org.apache.spark.api.java.JavaPairRDD;
import org.apache.spark.api.java.JavaRDD;
import org.apache.spark.api.java.JavaSparkContext;
import org.apache.spark.api.java.function.FilterFunction;
import org.apache.spark.api.java.function.MapFunction;
import org.apache.spark.api.java.function.PairFunction;
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.junit.jupiter.api.*;
import org.junit.jupiter.api.extension.ExtendWith;
import org.mockito.Mock;
import org.mockito.Mockito;
import org.mockito.junit.jupiter.MockitoExtension;
import com.fasterxml.jackson.databind.DeserializationFeature;
import com.fasterxml.jackson.databind.ObjectMapper;
import com.google.common.collect.Sets;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.schema.common.ModelConstants;
import eu.dnetlib.dhp.schema.oaf.*;
import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpException;
import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpService;
import eu.dnetlib.pace.util.MapDocumentUtil;
import scala.Tuple2;
import static java.nio.file.Files.createTempDirectory;
import static org.apache.spark.sql.functions.col;
import static org.apache.spark.sql.functions.count;
import static org.junit.jupiter.api.Assertions.*;
import static org.mockito.Mockito.lenient;
@ExtendWith(MockitoExtension.class)
@TestMethodOrder(MethodOrderer.OrderAnnotation.class)
@ -681,15 +677,16 @@ public class SparkDedupTest implements Serializable {
ArgumentApplicationParser parser = new ArgumentApplicationParser(
classPathResourceAsString("/eu/dnetlib/dhp/oa/dedup/propagateRelation_parameters.json"));
String outputRelPath = testDedupGraphBasePath + "/propagaterelation";
parser
.parseArgument(
new String[] {
"-i", testGraphBasePath, "-w", testOutputBasePath, "-o", testDedupGraphBasePath
"-i", testGraphBasePath, "-w", testOutputBasePath, "-o", outputRelPath
});
new SparkPropagateRelation(parser, spark).run(isLookUpService);
long relations = jsc.textFile(testDedupGraphBasePath + "/relation").count();
long relations = jsc.textFile(outputRelPath + "/relation").count();
// assertEquals(4860, relations);
System.out.println("relations = " + relations);
@ -708,7 +705,7 @@ public class SparkDedupTest implements Serializable {
(PairFunction<Row, String, String>) r -> new Tuple2<String, String>(r.getString(0), "d"));
JavaRDD<String> toCheck = jsc
.textFile(testDedupGraphBasePath + "/relation")
.textFile(outputRelPath + "/relation")
.mapToPair(json -> new Tuple2<>(MapDocumentUtil.getJPathString("$.source", json), json))
.join(mergedIds)
.map(t -> t._2()._1())
@ -724,6 +721,65 @@ public class SparkDedupTest implements Serializable {
@Test
@Order(8)
void testCleanBaseRelations() throws Exception {
ArgumentApplicationParser parser = new ArgumentApplicationParser(
classPathResourceAsString("/eu/dnetlib/dhp/oa/dedup/cleanRelation_parameters.json"));
// append dangling relations to be cleaned up
Dataset<Row> df_before = spark.read().schema(Encoders.bean(Relation.class).schema()).json(testGraphBasePath + "/relation");
Dataset<Row> df_input =df_before
.unionByName(df_before.drop("source").withColumn("source", functions.lit("n/a")))
.unionByName(df_before.drop("target").withColumn("target", functions.lit("n/a")));
df_input.write().mode(SaveMode.Overwrite).json(testOutputBasePath + "_tmp");
parser
.parseArgument(
new String[]{
"--graphBasePath", testGraphBasePath,
"--inputPath", testGraphBasePath + "/relation",
"--outputPath", testDedupGraphBasePath + "/relation"
});
new SparkCleanRelation(parser, spark).run(isLookUpService);
Dataset<Row> df_after = spark.read().schema(Encoders.bean(Relation.class).schema()).json(testDedupGraphBasePath + "/relation");
assertNotEquals(df_before.count(), df_input.count());
assertNotEquals(df_input.count(), df_after.count());
assertEquals(5, df_after.count());
}
@Test
@Order(9)
void testCleanDedupedRelations() throws Exception {
ArgumentApplicationParser parser = new ArgumentApplicationParser(
classPathResourceAsString("/eu/dnetlib/dhp/oa/dedup/cleanRelation_parameters.json"));
String inputRelPath = testDedupGraphBasePath + "/propagaterelation/relation";
// append dangling relations to be cleaned up
Dataset<Row> df_before = spark.read().schema(Encoders.bean(Relation.class).schema()).json(inputRelPath);
df_before.filter(col("dataInfo.deletedbyinference").notEqual(true)).show(50, false);
parser
.parseArgument(
new String[]{
"--graphBasePath", testGraphBasePath,
"--inputPath", inputRelPath,
"--outputPath", testDedupGraphBasePath + "/relation"
});
new SparkCleanRelation(parser, spark).run(isLookUpService);
Dataset<Row> df_after = spark.read().schema(Encoders.bean(Relation.class).schema()).json(testDedupGraphBasePath + "/relation");
assertNotEquals(df_before.count(), df_after.count());
assertEquals(0, df_after.count());
}
@Test
@Order(10)
void testRelations() throws Exception {
testUniqueness("/eu/dnetlib/dhp/dedup/test/relation_1.json", 12, 10);
testUniqueness("/eu/dnetlib/dhp/dedup/test/relation_2.json", 10, 2);