This commit is contained in:
Miriam Baglioni 2020-04-30 20:22:07 +02:00
parent 43053a286d
commit dbf3ba051a
5 changed files with 84 additions and 74 deletions

View File

@ -11,6 +11,7 @@
<artifactId>dhp-blacklist</artifactId>
<dependencies>
<dependency>
<groupId>eu.dnetlib.dhp</groupId>
<artifactId>dhp-graph-mapper</artifactId>

View File

@ -1,9 +1,10 @@
package eu.dnetlib.dhp.blacklist;
import com.fasterxml.jackson.databind.ObjectMapper;
import com.google.gson.Gson;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.schema.oaf.Relation;
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkHiveSession;
import java.util.Optional;
import org.apache.commons.io.IOUtils;
import org.apache.spark.SparkConf;
import org.apache.spark.api.java.function.MapFunction;
@ -14,11 +15,10 @@ import org.apache.spark.sql.SparkSession;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import java.util.Arrays;
import java.util.List;
import java.util.Optional;
import com.fasterxml.jackson.databind.ObjectMapper;
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkHiveSession;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.schema.oaf.Relation;
public class PrepareMergedRelationJob {
@ -56,8 +56,6 @@ public class PrepareMergedRelationJob {
conf,
isSparkSessionManaged,
spark -> {
// removeOutputDir(spark, potentialUpdatePath);
// removeOutputDir(spark, alreadyLinkedPath);
selectMergesRelations(
spark,
inputPath,
@ -70,7 +68,10 @@ public class PrepareMergedRelationJob {
Dataset<Relation> relation = readRelations(spark, inputPath);
relation.createOrReplaceTempView("relation");
spark.sql("Select * from relation where relclass = 'merges' and datainfo.deletedbyinference = false")
spark
.sql("Select * from relation " +
"where relclass = 'merges' " +
"and datainfo.deletedbyinference = false")
.as(Encoders.bean(Relation.class))
.toJSON()
.write()

View File

@ -55,7 +55,6 @@ public class SparkRemoveBlacklistedRelationJob {
log.info("mergesPath {}: ", mergesPath);
SparkConf conf = new SparkConf();
conf.set("hive.metastore.uris", parser.get("hive_metastore_uris"));
runWithSparkSession(
conf,
@ -68,6 +67,8 @@ public class SparkRemoveBlacklistedRelationJob {
outputPath,
mergesPath);
});
}
private static void removeBlacklistedRelations(SparkSession spark, String blacklistPath, String inputPath,
@ -78,7 +79,6 @@ public class SparkRemoveBlacklistedRelationJob {
Dataset<Relation> dedupSource = blackListed
.joinWith(mergesRelation, blackListed.col("source").equalTo(mergesRelation.col("target")), "left_outer")
// .joinWith(inputRelation,blackListed.col("target").equalTo(inputRelation.col("target")),"left_outer")
.map(c -> {
Optional<Relation> merged = Optional.ofNullable(c._2());
Relation bl = c._1();

View File

@ -63,7 +63,7 @@
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
<arg>--outputPath</arg><arg>${workingDir}/relation</arg>
<arg>--outputPath</arg><arg>${workingDir}/mergesRelation</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
</spark>
<ok to="apply_blacklist"/>
@ -89,6 +89,7 @@
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
<arg>--outputPath</arg><arg>${workingDir}/relation</arg>
<arg>--hdfsPath</arg><arg>${workingDir}/blacklist</arg>
<arg>--mergesPath</arg><arg>${workingDir}/mergesRelation</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>

View File

@ -22,5 +22,12 @@
"paramLongName": "isSparkSessionManaged",
"paramDescription": "true if the spark session is managed",
"paramRequired": false
},
{
"paramName": "m",
"paramLongName": "mergesPath",
"paramDescription": "true if the spark session is managed",
"paramRequired": true
}
]