cleaning of subjects: perform the cleaning when the given value is equivalent to one of the terms in the vocabulary

dhp-common/src/main/java/eu/dnetlib/dhp/schema/oaf/utils/OafMapperUtils.java

@@ -11,6 +11,7 @@ import java.util.function.Function;
 import java.util.function.Predicate;
 import java.util.stream.Collectors;
 
+import eu.dnetlib.dhp.schema.common.ModelConstants;
 import org.apache.commons.lang3.StringUtils;
 
 import eu.dnetlib.dhp.schema.common.AccessRightComparator;
@@ -141,7 +142,7 @@ public class OafMapperUtils {
 	}
 
 	public static Qualifier unknown(final String schemeid, final String schemename) {
-		return qualifier("UNKNOWN", "Unknown", schemeid, schemename);
+		return qualifier(UNKNOWN, "Unknown", schemeid, schemename);
 	}
 
 	public static AccessRight accessRight(

dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/clean/CleaningRuleMap.java

@@ -46,7 +46,7 @@ public class CleaningRuleMap extends HashMap<Class<?>, SerializableConsumer<Obje
 				return;
 			}
 			Qualifier newValue = vocabulary.lookup(subject.getValue());
-			if (!subject.getValue().equals(newValue.getClassid())) {
+			if (!ModelConstants.UNKNOWN.equals(newValue.getClassid())) {
 				subject.setValue(newValue.getClassid());
 				subject.getQualifier().setClassid(vocabularyId);
 				subject.getQualifier().setClassname(vocabulary.getName());

dhp-workflows/dhp-graph-mapper/src/test/java/eu/dnetlib/dhp/oa/graph/clean/GraphCleaningFunctionsTest.java

@@ -260,7 +260,7 @@ public class GraphCleaningFunctionsTest {
 			.collect(Collectors.toList());
 
 		assertNotNull(fos_subjects);
-		assertEquals(2, fos_subjects.size());
+		assertEquals(3, fos_subjects.size());
 
 		assertTrue(
 			fos_subjects

dhp-workflows/dhp-graph-mapper/src/test/resources/eu/dnetlib/dhp/oa/graph/clean/result.json

@@ -787,12 +787,12 @@
         "trust": "0.9"
       },
       "qualifier": {
-        "classid": "",
-        "classname": "",
-        "schemeid": "",
-        "schemename": ""
+        "classid": "keyword",
+        "classname": "keyword",
+        "schemeid": "dnet:subject_classification_typologies",
+        "schemename": "dnet:subject_classification_typologies"
       },
-      "value": "silicon"
+      "value": "0101 mathematics"
     },
     {
       "dataInfo": {