Dataset based implementation for SparkCreateDedupRecord phase, fixed datasource entity dump supplementing dedup unit tests

dhp-workflows/dhp-dedup-openaire/src/main/java/eu/dnetlib/dhp/oa/dedup/SparkCreateDedupRecord.java

@@ -11,8 +11,6 @@ import eu.dnetlib.pace.config.DedupConfig;
 import java.io.IOException;
 import org.apache.commons.io.IOUtils;
 import org.apache.spark.SparkConf;
-import org.apache.spark.api.java.function.MapFunction;
-import org.apache.spark.sql.Encoders;
 import org.apache.spark.sql.SaveMode;
 import org.apache.spark.sql.SparkSession;
 import org.dom4j.DocumentException;
@@ -72,12 +70,9 @@ public class SparkCreateDedupRecord extends AbstractSparkAction {
             Class<OafEntity> clazz = ModelSupport.entityTypes.get(EntityType.valueOf(subEntity));
 
             DedupRecordFactory.createDedupRecord(spark, mergeRelPath, entityPath, clazz)
-                    .map(
-                            (MapFunction<OafEntity, String>)
-                                    value -> OBJECT_MAPPER.writeValueAsString(value),
-                            Encoders.STRING())
                     .write()
                     .mode(SaveMode.Overwrite)
+                    .option("compression", "gzip")
                     .json(outputPath);
         }
     }

dhp-workflows/dhp-dedup-openaire/src/main/java/eu/dnetlib/dhp/oa/dedup/SparkUpdateEntity.java

@@ -33,7 +33,7 @@ public class SparkUpdateEntity extends AbstractSparkAction {
 
     private static final Logger log = LoggerFactory.getLogger(SparkUpdateEntity.class);
 
-    final String IDJSONPATH = "$.id";
+    private static final String IDJSONPATH = "$.id";
 
     public SparkUpdateEntity(ArgumentApplicationParser parser, SparkSession spark) {
         super(parser, spark);
@@ -65,27 +65,25 @@ public class SparkUpdateEntity extends AbstractSparkAction {
         log.info("workingPath:    '{}'", workingPath);
         log.info("dedupGraphPath: '{}'", dedupGraphPath);
 
-        final JavaSparkContext sc = new JavaSparkContext(spark.sparkContext());
+        final JavaSparkContext sc = JavaSparkContext.fromSparkContext(spark.sparkContext());
 
         // for each entity
         ModelSupport.entityTypes.forEach(
-                (entity, clazz) -> {
-                    final String outputPath = dedupGraphPath + "/" + entity;
+                (type, clazz) -> {
+                    final String outputPath = dedupGraphPath + "/" + type;
                     removeOutputDir(spark, outputPath);
 
                     JavaRDD<String> sourceEntity =
                             sc.textFile(
-                                    DedupUtility.createEntityPath(
-                                            graphBasePath, entity.toString()));
+                                    DedupUtility.createEntityPath(graphBasePath, type.toString()));
 
-                    if (mergeRelExists(workingPath, entity.toString())) {
+                    if (mergeRelExists(workingPath, type.toString())) {
 
                         final String mergeRelPath =
-                                DedupUtility.createMergeRelPath(
-                                        workingPath, "*", entity.toString());
+                                DedupUtility.createMergeRelPath(workingPath, "*", type.toString());
                         final String dedupRecordPath =
                                 DedupUtility.createDedupRecordPath(
-                                        workingPath, "*", entity.toString());
+                                        workingPath, "*", type.toString());
 
                         final Dataset<Relation> rel =
                                 spark.read().load(mergeRelPath).as(Encoders.bean(Relation.class));
@@ -107,7 +105,6 @@ public class SparkUpdateEntity extends AbstractSparkAction {
                                                                 MapDocumentUtil.getJPathString(
                                                                         IDJSONPATH, s),
                                                                 s));
-
                         JavaRDD<String> map =
                                 entitiesWithId
                                         .leftOuterJoin(mergedIds)

dhp-workflows/dhp-dedup-openaire/src/test/java/eu/dnetlib/dhp/oa/dedup/SparkDedupTest.java

@@ -72,7 +72,7 @@ public class SparkDedupTest implements Serializable {
 
         spark =
                 SparkSession.builder()
-                        .appName(SparkCreateSimRels.class.getSimpleName())
+                        .appName(SparkDedupTest.class.getSimpleName())
                         .master("local[*]")
                         .config(new SparkConf())
                         .getOrCreate();
@@ -300,8 +300,8 @@ public class SparkDedupTest implements Serializable {
 
         long deletedSw =
                 jsc.textFile(testDedupGraphBasePath + "/software")
-                .filter(this::isDeletedByInference)
-                .count();
+                        .filter(this::isDeletedByInference)
+                        .count();
 
         assertEquals(mergedOrgs, deletedOrgs);
         assertEquals(mergedPubs, deletedPubs);