D-Net
/
dnet-hadoop
13
4
Fork 10
Code Issues 14 Pull Requests 7 Projects 2 Releases 1 Wiki Activity

subjects cleaning #239

Merged
claudio.atzori merged 15 commits from clean_subjects into beta 2022-09-09 15:17:09 +02:00
Conversation 6 Commits 15 Files Changed 29 +33 -13
2 changed files with 33 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files
Showing only changes of commit 4eaa063b1f - Show all commits

32
dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/clean/CleaningRuleMap.java
View File

@ -3,13 +3,12 @@ package eu.dnetlib.dhp.oa.graph.clean;
import java.io.Serializable;
import java.util.HashMap;
import java.util.concurrent.atomic.AtomicReference;
import javax.jws.WebParam;
import org.apache.commons.lang3.SerializationUtils;
import org.apache.commons.lang3.StringUtils;
import eu.dnetlib.dhp.common.FunctionalInterfaceSupport.SerializableConsumer;
import eu.dnetlib.dhp.common.vocabulary.Vocabulary;
import eu.dnetlib.dhp.common.vocabulary.VocabularyGroup;
import eu.dnetlib.dhp.schema.common.ModelConstants;
import eu.dnetlib.dhp.schema.oaf.*;
@ -31,23 +30,30 @@ public class CleaningRuleMap extends HashMap<Class<?>, SerializableConsumer<Obje
return mapping;
}
private static void cleanSubject(VocabularyGroup vocabularies, Subject s) {
// TODO cleaning based on different subject vocabs can be added here
cleanSubjectForVocabulary(ModelConstants.DNET_SUBJECT_FOS_CLASSID, vocabularies, s);
private static void cleanSubject(VocabularyGroup vocabularies, Subject subject) {
if (cleanSubjectForVocabulary(ModelConstants.DNET_SUBJECT_FOS_CLASSID, vocabularies, subject)) {
return;
} else {
// TODO cleaning based on different subject vocabs can be added here
}
}
private static void cleanSubjectForVocabulary(String vocabularyId, VocabularyGroup vocabularies, Subject s) {
private static boolean cleanSubjectForVocabulary(String vocabularyId, VocabularyGroup vocabularies,
Subject subject) {
AtomicReference<Boolean> modified = new AtomicReference<>(false);
vocabularies.find(vocabularyId).ifPresent(vocabulary -> {
if (!ModelConstants.DNET_SUBJECT_KEYWORD.equalsIgnoreCase(s.getQualifier().getClassid())) {
if (!ModelConstants.DNET_SUBJECT_KEYWORD.equalsIgnoreCase(subject.getQualifier().getClassid())) {
return;
}
Qualifier newValue = vocabulary.lookup(s.getValue());
if (!s.getValue().equals(newValue.getClassid())) {
s.setValue(newValue.getClassid());
s.getQualifier().setClassid(vocabularyId);
s.getQualifier().setClassname(vocabulary.getName());
Qualifier newValue = vocabulary.lookup(subject.getValue());
if (!subject.getValue().equals(newValue.getClassid())) {
miriam.baglioni commented 2022-08-08 10:52:28 +02:00
Outdated
Review
Copy Link

I do not understand why you compare subject.value and newValue.classid

I do not understand why you compare subject.value and newValue.classid
miriam.baglioni commented 2022-08-08 11:00:57 +02:00
Outdated
Review
Copy Link

I saw by myself. It is ok.

One thing: you will not change the classId of the subject if the value provided is equal to the term in the vocabulary.

I saw by myself. It is ok. One thing: you will not change the classId of the subject if the value provided is equal to the term in the vocabulary.
claudio.atzori commented 2022-08-08 12:41:34 +02:00
Outdated
Review
Copy Link

You are right. However, the vocabulary.lookup method already tryes to find match a synonym and in case it can't then it looks for a matching equivalent term:

public Qualifier lookup(String id) {
	return Optional
		.ofNullable(getSynonymAsQualifier(id))
		.orElse(getTermAsQualifier(id));
}

Then, in case a matching term cannot be found, the method returns a qualifier set to UNKNOWN

public Qualifier getTermAsQualifier(final String termId) {
	if (StringUtils.isBlank(termId)) {
		return OafMapperUtils.unknown(getId(), getName());
	} else if (termExists(termId)) {
		final VocabularyTerm t = getTerm(termId);
		return OafMapperUtils.qualifier(t.getId(), t.getName(), getId(), getName());
	} else {
		return OafMapperUtils.qualifier(termId, termId, getId(), getName());
	}
}

hence I could just exploit this to decide what to do when the value provided is equal to the term in the vocabulary.

You are right. However, the `vocabulary.lookup` method already tryes to find match a synonym and in case it can't then it looks for a matching equivalent term: ``` public Qualifier lookup(String id) { return Optional .ofNullable(getSynonymAsQualifier(id)) .orElse(getTermAsQualifier(id)); } ``` Then, in case a matching term cannot be found, the method returns a qualifier set to UNKNOWN ``` public Qualifier getTermAsQualifier(final String termId) { if (StringUtils.isBlank(termId)) { return OafMapperUtils.unknown(getId(), getName()); } else if (termExists(termId)) { final VocabularyTerm t = getTerm(termId); return OafMapperUtils.qualifier(t.getId(), t.getName(), getId(), getName()); } else { return OafMapperUtils.qualifier(termId, termId, getId(), getName()); } } ``` hence I could just exploit this to decide what to do when the value provided is equal to the term in the vocabulary.
miriam.baglioni commented 2022-08-08 14:12:32 +02:00
Outdated
Review
Copy Link

Yes, I do agree

Yes, I do agree
subject.setValue(newValue.getClassid());
subject.getQualifier().setClassid(vocabularyId);
subject.getQualifier().setClassname(vocabulary.getName());
modified.set(true);
}
});
return modified.get();
}
private static void cleanRelation(VocabularyGroup vocabularies, Relation r) {

14
dhp-workflows/dhp-graph-mapper/src/test/java/eu/dnetlib/dhp/oa/graph/clean/GraphCleaningFunctionsTest.java
View File

@ -262,6 +262,20 @@ public class GraphCleaningFunctionsTest {
assertNotNull(fos_subjects);
assertEquals(2, fos_subjects.size());
assertTrue(
fos_subjects
.stream()
.anyMatch(
s -> "0101 mathematics".equals(s.getValue()) &
ModelConstants.DNET_SUBJECT_FOS_CLASSID.equals(s.getQualifier().getClassid())));
assertTrue(
fos_subjects
.stream()
.anyMatch(
s -> "0102 computer and information sciences".equals(s.getValue()) &
ModelConstants.DNET_SUBJECT_FOS_CLASSID.equals(s.getQualifier().getClassid())));
// TODO add more assertions to verity the cleaned values
System.out.println(MAPPER.writeValueAsString(p_cleaned));
}