Merge branch 'develop' into dependabot/maven/org.apache.maven.plugins…

…-maven-compiler-plugin-3.13.0

.github/workflows/java-build.yml

@@ -26,8 +26,10 @@ jobs:
       - name: Build with Maven
         run: mvn ${{ inputs.maven-options }} -B package --file pom.xml
       - name: Prepare artifact for upload
+        if: github.actor != 'dependabot[bot]'
         run: mkdir staging && cp target/*.zip staging
       - name: Upload artifact
+        if: github.actor != 'dependabot[bot]'
         uses: actions/upload-artifact@v4
         with:
           name: maven-build

README.md

@@ -87,9 +87,9 @@ SYNOPSIS
        neo2rdf dump [--deriveClassHierarchyByLabelSubsetCheck]
        [--derivePropertyHierarchyByRelationshipSubsetCheck] [--includeDeletedNeo4jLabels]
        [--includeDeletedPropertyKeys] [--includeDeletedRelationshipTypes]
-       [--basePrefix=<basePrefix>] [-cfg=<conversionConfigFile>] [-d=<neo4jDBDumpPath>]
-       -db=<neo4jDBDirectory> -o=<outputPath>
-       [--reificationVocabulary=<reificationVocabulary>]
+       [--reifyOnlyStatementsWithAnnotations] [--basePrefix=<basePrefix>]
+       [-cfg=<conversionConfigFile>] [-d=<neo4jDBDumpPath>] -db=<neo4jDBDirectory>
+       -o=<outputPath> [--reificationVocabulary=<reificationVocabulary>]
        [--schemaOutputPath=<schemaOutputPath>]
        [--sequenceConversionType=<sequenceConversionType>]
 
@@ -103,6 +103,8 @@ OPTIONS
 	   Prefix that is used for all converted nodes, property keys, relationship types,
 	   and relationships.
 
+	       Default: https://www.example.org/
+
        -cfg, --config=<conversionConfigFile>
 	   Instead of specifying non-mandatory options in terms of command line parameters,
 	   it is also possible to use an external YAML configuration. The YAML keys must
@@ -112,54 +114,52 @@ OPTIONS
        -d, --neo4jDBDumpPath=<neo4jDBDumpPath>
 	   If a path to a Neo4j dump has been specified using this parameter, the DB dump is
 	   extracted to the appropriate target
-
 	   Neo4j DB directory first, and subsequently, the conversion procedure gets
 	   executed as usual.
 
        -db, --neo4jDBDirectory=<neo4jDBDirectory>, --deriveClassHierarchyByLabelSubsetCheck
 	   Indicates whether the RDF class hierarchy should be derived.
-
 	   For this purpose, it is examined which sets of Neo4j nodes with an assigned label
 	   are a subset of one another.
 
        --derivePropertyHierarchyByRelationshipSubsetCheck
 	   Indicates whether the RDF property hierarchy should be derived.
-
 	   For this, the node-node combinations for each relationship type are initially
 	   collected in a set.
-
 	   Subsequently, for every pair of sets, it is examined whether they are a subset of
 	   each other.
 
        --includeDeletedNeo4jLabels, --includeDeletedPropertyKeys,
        --includeDeletedRelationshipTypes, -o, --outputPath=<outputPath>,
        --reificationVocabulary=<reificationVocabulary>
 	   The reification vocabulary defines how a quadruple (sbj, pred, obj, statementID)
-	   should be reified in RDF.
-
+	   should be reified.
 	   Options:
+	   •   RDF_REIFICATION: uses the RDF reification vocabulary, i.e.,  rdf:Statement,
+	       rdf:subject, rdf:predicate, and rdf:object
+		   (cf. https://www.w3.org/TR/rdf11-mt/#reification)
+	   •   OWL_REIFICATION: uses the OWL vocabulary, i.e.,	owl:Axiom,
+	       rdf:annotatedSource, owl:annotatedProperty, and owl:annotatedTarget
+		   (cf. https://www.w3.org/TR/owl2-quick-reference/#Annotations)
 
-	   •   RDF_COLLECTION: Neo4j sequences are converted into open lists in RDF.
+		   Default: OWL_REIFICATION
 
-	   •   SEPARATE_LITERALS: Neo4j sequences are converted into separate assertions,
-	       e.g., (x { has: [1, 2] }) is converted to (:x, :has, 1) and (:x, :has, 2).
+       --reifyOnlyStatementsWithAnnotations
+	   If enabled, whether only statements with annotations should be reified.
 
        --schemaOutputPath=<schemaOutputPath>
 	   If the RDF schema is derived from the Neo4j dataset, e.g., the class or property
-	   hierarchy,
-
-	   an additional path can be specified to store it separately on disk.
-
+	   hierarchy, an additional path can be specified to store it separately on disk.
 	   If the value is null or left out, the derived schema is stored along with the
 	   data.
 
        --sequenceConversionType=<sequenceConversionType>
 	   Options:
-
 	   •   RDF_COLLECTION: Neo4j sequences are converted into open lists in RDF.
-
 	   •   SEPARATE_LITERALS: Neo4j sequences are converted into separate assertions,
 	       e.g., (x { has: [1, 2] }) is converted to (:x, :has, 1) and (:x, :has, 2).
+
+		   Default: RDF_COLLECTION
 ```
 
 
@@ -176,8 +176,9 @@ SYNOPSIS
        neo2rdf server [--deriveClassHierarchyByLabelSubsetCheck]
        [--derivePropertyHierarchyByRelationshipSubsetCheck] [--includeDeletedNeo4jLabels]
        [--includeDeletedPropertyKeys] [--includeDeletedRelationshipTypes]
-       [--basePrefix=<basePrefix>] [-cfg=<conversionConfigFile>] [-d=<neo4jDBDumpPath>]
-       -db=<neo4jDBDirectory> -p=<port> [--reificationVocabulary=<reificationVocabulary>]
+       [--reifyOnlyStatementsWithAnnotations] [--basePrefix=<basePrefix>]
+       [-cfg=<conversionConfigFile>] [-d=<neo4jDBDumpPath>] -db=<neo4jDBDirectory> -p=<port>
+       [--reificationVocabulary=<reificationVocabulary>]
        [--schemaOutputPath=<schemaOutputPath>]
        [--sequenceConversionType=<sequenceConversionType>] [-t=<numberOfServerThreads>]
 
@@ -192,6 +193,8 @@ OPTIONS
 	   Prefix that is used for all converted nodes, property keys, relationship types,
 	   and relationships.
 
+	       Default: https://www.example.org/
+
        -cfg, --config=<conversionConfigFile>
 	   Instead of specifying non-mandatory options in terms of command line parameters,
 	   it is also possible to use an external YAML configuration. The YAML keys must
@@ -200,55 +203,58 @@ OPTIONS
 
        -d, --neo4jDBDumpPath=<neo4jDBDumpPath>
 	   If a path to a Neo4j dump has been specified using this parameter, the DB dump is
-	   extracted to the appropriate target
-
-	   Neo4j DB directory first, and subsequently, the conversion procedure gets
-	   executed as usual.
+	   extracted to the appropriate target Neo4j DB directory first, and subsequently,
+	   the conversion procedure gets executed as usual.
 
        -db, --neo4jDBDirectory=<neo4jDBDirectory>, --deriveClassHierarchyByLabelSubsetCheck
 	   Indicates whether the RDF class hierarchy should be derived.
-
 	   For this purpose, it is examined which sets of Neo4j nodes with an assigned label
 	   are a subset of one another.
 
        --derivePropertyHierarchyByRelationshipSubsetCheck
 	   Indicates whether the RDF property hierarchy should be derived.
-
 	   For this, the node-node combinations for each relationship type are initially
 	   collected in a set.
-
 	   Subsequently, for every pair of sets, it is examined whether they are a subset of
 	   each other.
 
        --includeDeletedNeo4jLabels, --includeDeletedPropertyKeys,
-       --includeDeletedRelationshipTypes, -p, --port=<port>,
+       --includeDeletedRelationshipTypes 
+       
+       -p, --port=<port>
+	      Default: 8080
+
        --reificationVocabulary=<reificationVocabulary>
 	   The reification vocabulary defines how a quadruple (sbj, pred, obj, statementID)
-	   should be reified in RDF.
+	   should be reified.
 
 	   Options:
+	   •   RDF_REIFICATION: uses the RDF reification vocabulary, i.e.,  rdf:Statement,
+	       rdf:subject, rdf:predicate, and rdf:object
+		   (cf. https://www.w3.org/TR/rdf11-mt/#reification)
+	   •   OWL_REIFICATION: uses the OWL vocabulary, i.e.,	owl:Axiom,
+	       rdf:annotatedSource, owl:annotatedProperty, and owl:annotatedTarget
+		   (cf. https://www.w3.org/TR/owl2-quick-reference/#Annotations)
 
-	   •   RDF_COLLECTION: Neo4j sequences are converted into open lists in RDF.
+		   Default: OWL_REIFICATION
 
-	   •   SEPARATE_LITERALS: Neo4j sequences are converted into separate assertions,
-	       e.g., (x { has: [1, 2] }) is converted to (:x, :has, 1) and (:x, :has, 2).
+       --reifyOnlyStatementsWithAnnotations
+	   If this option is set, only Neo4j relationships with properties will be reified in RDF.
 
        --schemaOutputPath=<schemaOutputPath>
 	   If the RDF schema is derived from the Neo4j dataset, e.g., the class or property
-	   hierarchy,
-
-	   an additional path can be specified to store it separately on disk.
-
+	   hierarchy, an additional path can be specified to store it separately on disk.
 	   If the value is null or left out, the derived schema is stored along with the
 	   data.
 
        --sequenceConversionType=<sequenceConversionType>
 	   Options:
-
 	   •   RDF_COLLECTION: Neo4j sequences are converted into open lists in RDF.
-
 	   •   SEPARATE_LITERALS: Neo4j sequences are converted into separate assertions,
 	       e.g., (x { has: [1, 2] }) is converted to (:x, :has, 1) and (:x, :has, 2).
 
+		   Default: RDF_COLLECTION
+
        -t, --numberOfServerThreads=<numberOfServerThreads>
+	      Default: 2
 ```

build-resources/generate-manual/generated-manpages-txt.sh

@@ -1,5 +1,13 @@
 # Codegen for README CLI manual
 
+# Manual update workflow:
+# - build the project locally with maven - $BUILD_DIRECTORY/target/generated-docs now contains autogenerated man pages
+# - execute this script which will generate a single manual.txt file from the manual template and autogenerated man pages
+# - remove manually from "manual.txt":
+#   - undesired newline characters that get inserted by the conversion procedure
+#   - generation date
+# - copy "manual.txt" into the README.md (a "git diff" might be helpful)
+
 # adjust width to default maximum GitHub markdown code width
 MANWIDTH=95
 export MANWIDTH

pom.xml

@@ -27,7 +27,7 @@
         <revision>${version}-${sha1}${changelist}-SNAPSHOT</revision>
         <version.mvn.plugin.flatten>1.1.0</version.mvn.plugin.flatten>
         <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
-        <neo4j.version>5.15.0</neo4j.version>
+        <neo4j.version>5.18.1</neo4j.version>
         <rdf4j.version>5.0.0-M2</rdf4j.version>
     </properties>
 

src/main/java/de/derivo/neo2rdf/conversion/Neo4jToRDFConverter.java

@@ -101,7 +101,7 @@ public void startProcessing() {
 
         log.info("Processing relationships...");
         relationshipProcessor.startProcessing();
-        this.deployedRelationshipTypes = relationshipProcessor.getDeployedRelationshiptypes();
+        this.deployedRelationshipTypes = relationshipProcessor.getDeployedRelationshipTypes();
         this.annotationPropertyKeys = new UnifiedSet<>(indexedSchema.getPropertyKeyIDToStr().size());
         this.annotationPropertyKeys.addAll(relationshipProcessor.getObjectPropertyKeys());
         this.annotationPropertyKeys.addAll(relationshipProcessor.getDatatypePropertyKeys());

src/main/java/de/derivo/neo2rdf/conversion/RelationshipToRDFConverter.java

@@ -24,7 +24,7 @@ public class RelationshipToRDFConverter extends RelationshipProcessor {
     private final Neo4jToRDFMapper neo4jToRDFMapper;
     private final ValueFactory valueFactory = new Neo4jToRDFValueFactory();
 
-    private Set<Long> deployedRelationshiptypes;
+    private Set<Long> deployedRelationshipTypes;
     private Set<Long> datatypePropertyKeys;
     private Set<Long> objectPropertyKeys;
     private final ConversionConfig config;
@@ -40,7 +40,7 @@ public RelationshipToRDFConverter(NeoStores neoStores, Neo4jToRDFConverter neo4j
     }
 
     private void init() {
-        this.deployedRelationshiptypes = new UnifiedSet<>(neo4jToRDFConverter.getIndexedSchema().getRelationshipTypeIDToStr().size());
+        this.deployedRelationshipTypes = new UnifiedSet<>(neo4jToRDFConverter.getIndexedSchema().getRelationshipTypeIDToStr().size());
         this.datatypePropertyKeys = new UnifiedSet<>(neo4jToRDFConverter.getIndexedSchema().getPropertyKeyIDToStr().size());
         this.objectPropertyKeys = new UnifiedSet<>(neo4jToRDFConverter.getIndexedSchema().getPropertyKeyIDToStr().size());
 
@@ -55,14 +55,14 @@ protected void process(long relationshipID,
                            long targetID,
                            long typeID,
                            boolean statementHasAnnotations) {
-        deployedRelationshiptypes.add(relationshipID);
+        deployedRelationshipTypes.add(relationshipID);
         Statement statement = valueFactory.createStatement(
                 neo4jToRDFMapper.nodeIDToResource(sourceID),
                 neo4jToRDFMapper.relationshipTypeIDToIRI(typeID),
                 neo4jToRDFMapper.nodeIDToResource(targetID),
                 neo4jToRDFMapper.relationshipIDToResource(relationshipID)
         );
-        if (statementHasAnnotations) {
+        if (!config.isReifyOnlyStatementsWithAnnotations() || statementHasAnnotations) {
             neo4jToRDFConverter.processStatement(statement);
             neo4jToRDFMapper.statementToReificationTriples(statement, neo4jToRDFConverter::processStatement);
         } else {
@@ -113,8 +113,8 @@ public Map<Long, Roaring64Bitmap> getRelationshipIDToInstanceSet() {
         return relationshipIDToInstanceSet;
     }
 
-    public Set<Long> getDeployedRelationshiptypes() {
-        return deployedRelationshiptypes;
+    public Set<Long> getDeployedRelationshipTypes() {
+        return deployedRelationshipTypes;
     }
 
     public Set<Long> getDatatypePropertyKeys() {