Add action to check conformance with GPAD ShEx on each change.

geneontology · Jul 18, 2024 · c7426ed · c7426ed
1 parent dc14000
commit c7426ed
Show file tree

Hide file tree

Showing 5 changed files with 96 additions and 22 deletions.
diff --git a/minerva-converter/src/main/java/org/geneontology/minerva/legacy/sparql/GPADSPARQLExport.java b/minerva-converter/src/main/java/org/geneontology/minerva/legacy/sparql/GPADSPARQLExport.java
@@ -12,8 +12,8 @@
 import org.apache.jena.rdf.model.Resource;
 import org.apache.jena.sparql.core.Var;
 import org.apache.jena.sparql.engine.binding.Binding;
+import org.apache.jena.sparql.engine.binding.BindingBuilder;
 import org.apache.jena.sparql.engine.binding.BindingFactory;
-import org.apache.jena.sparql.engine.binding.BindingMap;
 import org.apache.jena.vocabulary.RDF;
 import org.apache.log4j.Logger;
 import org.geneontology.minerva.curie.CurieHandler;
@@ -288,11 +288,11 @@ private Map<Triple, Set<GPADEvidence>> evidencesForFacts(Set<Triple> facts, Mode
 
     @SafeVarargs
     private final Binding createBinding(Pair<Var, Node>... bindings) {
-        BindingMap map = BindingFactory.create();
+        BindingBuilder builder = BindingFactory.builder();
         for (Pair<Var, Node> binding : bindings) {
-            map.add(binding.getLeft(), binding.getRight());
+            builder.add(binding.getLeft(), binding.getRight());
         }
-        return map;
+        return builder.build();
     }
 
     private Set<AnnotationExtension> possibleExtensions(Set<BasicGPADData> basicAnnotations, Model model) {

diff --git a/minerva-core/pom.xml b/minerva-core/pom.xml
@@ -92,6 +92,11 @@
             <artifactId>commons-math3</artifactId>
             <version>3.6.1</version>
         </dependency>
+        <dependency>
+            <groupId>commons-codec</groupId>
+            <artifactId>commons-codec</artifactId>
+            <version>1.15</version>
+        </dependency>
         <dependency>
             <groupId>org.obolibrary.robot</groupId>
             <artifactId>robot-core</artifactId>

diff --git a/minerva-core/src/main/java/org/geneontology/minerva/BlazegraphMolecularModelManager.java b/minerva-core/src/main/java/org/geneontology/minerva/BlazegraphMolecularModelManager.java
@@ -7,11 +7,20 @@
 import info.aduna.iteration.Iterations;
 import org.apache.commons.io.FileUtils;
 import org.apache.commons.lang3.StringUtils;
+import org.apache.commons.rdf.jena.JenaGraph;
+import org.apache.commons.rdf.jena.JenaRDF;
+import org.apache.jena.graph.Graph;
+import org.apache.jena.shex.ShapeMap;
+import org.apache.jena.shex.ShexReport;
+import org.apache.jena.shex.ShexSchema;
+import org.apache.jena.shex.ShexValidator;
+import org.apache.jena.sys.JenaSystem;
 import org.apache.log4j.Logger;
 import org.geneontology.minerva.MolecularModelManager.UnknownIdentifierException;
 import org.geneontology.minerva.curie.CurieHandler;
 import org.geneontology.minerva.util.AnnotationShorthand;
 import org.geneontology.minerva.util.BlazegraphMutationCounter;
+import org.geneontology.minerva.util.JenaOwlTool;
 import org.geneontology.minerva.util.ReverseChangeGenerator;
 import org.openrdf.model.*;
 import org.openrdf.model.impl.URIImpl;
@@ -22,6 +31,7 @@
 import org.openrdf.repository.RepositoryException;
 import org.openrdf.repository.RepositoryResult;
 import org.openrdf.rio.*;
+import org.openrdf.rio.helpers.BasicWriterSettings;
 import org.openrdf.rio.helpers.RDFHandlerBase;
 import org.openrdf.rio.helpers.StatementCollector;
 import org.semanticweb.owlapi.apibinding.OWLManager;
@@ -67,6 +77,7 @@ public class BlazegraphMolecularModelManager<METADATA> extends CoreMolecularMode
     public BlazegraphMolecularModelManager(OWLOntology tbox, CurieHandler curieHandler, String modelIdPrefix, @Nonnull String pathToJournal, String pathToExportFolder, String pathToOntologyJournal, boolean downloadOntologyJournal)
             throws OWLOntologyCreationException, IOException {
         super(tbox, pathToOntologyJournal, downloadOntologyJournal);
+        JenaSystem.init();
         if (curieHandler == null) {
             LOG.error("curie handler required for blazegraph model manager startup ");
             System.exit(-1);
@@ -195,23 +206,14 @@ public void saveModel(ModelContainer m)
         final OWLOntologyManager manager = ont.getOWLOntologyManager();
         List<OWLOntologyChange> changes = preSaveFileHandler(ont);
         synchronized (ont) {
-            try {
-                this.writeModelToDatabase(ont, modelId);
-                // reset modified flag for abox after successful save
-                m.setAboxModified(false);
-                // dump stored model to export file
-                if (this.pathToExportFolder != null) {
-                    File folder = new File(this.pathToExportFolder);
-                    dumpStoredModel(modelId, folder);
-                }
-            } finally {
-                if (changes != null) {
-                    List<OWLOntologyChange> invertedChanges = ReverseChangeGenerator
-                            .invertChanges(changes);
-                    if (invertedChanges != null && !invertedChanges.isEmpty()) {
-                        manager.applyChanges(invertedChanges);
-                    }
-                }
+            manager.applyChanges(changes);
+            this.writeModelToDatabase(ont, modelId);
+            // reset modified flag for abox after successful save
+            m.setAboxModified(false);
+            // dump stored model to export file
+            if (this.pathToExportFolder != null) {
+                File folder = new File(this.pathToExportFolder);
+                dumpStoredModel(modelId, folder);
             }
         }
     }
@@ -994,4 +996,29 @@ public int addTaxonToDatabaseWithSparql(IRI model_iri, IRI taxon_iri) throws Rep
         return changes;
     }
 
+    public static class ShexValidatingPreSaveFileHandler implements PreFileSaveHandler {
+
+        private final ShexSchema schema;
+        private final ShapeMap shapeMap;
+        private final OWLAnnotationProperty conformanceProperty;
+        public ShexValidatingPreSaveFileHandler(ShexSchema schema, ShapeMap map, String propertyIRI) {
+            this.schema = schema;
+            this.shapeMap = map;
+            this.conformanceProperty = OWLManager.getOWLDataFactory().getOWLAnnotationProperty(IRI.create(propertyIRI));
+        }
+
+        @Override
+        public List<OWLOntologyChange> handle(OWLOntology modelOnt) throws UnknownIdentifierException {
+            OWLDataFactory factory = OWLManager.getOWLDataFactory();
+            Graph graph = JenaOwlTool.getJenaModel(modelOnt).getGraph();
+            ShexReport report = ShexValidator.get().validate(graph, schema, shapeMap);
+            boolean conforms = report.conforms();
+            System.out.println("CONFORMS? " + conforms);
+            System.out.println(factory.getOWLAnnotation(conformanceProperty, factory.getOWLLiteral(conforms)));
+            OWLOntologyChange remove = new RemoveOntologyAnnotation(modelOnt, factory.getOWLAnnotation(conformanceProperty, factory.getOWLLiteral(!conforms)));
+            OWLOntologyChange add = new AddOntologyAnnotation(modelOnt, factory.getOWLAnnotation(conformanceProperty, factory.getOWLLiteral(conforms)));
+            return Arrays.asList(remove, add);
+        }
+    }
+
 }
diff --git a/minerva-server/src/main/java/org/geneontology/minerva/server/StartUpTool.java b/minerva-server/src/main/java/org/geneontology/minerva/server/StartUpTool.java
@@ -1,12 +1,16 @@
 package org.geneontology.minerva.server;
 
+import org.apache.jena.shex.ShapeMap;
+import org.apache.jena.shex.ShexSchema;
+import org.apache.jena.shex.parser.ShExC;
 import org.apache.log4j.Logger;
 import org.eclipse.jetty.server.HttpConfiguration;
 import org.eclipse.jetty.server.HttpConnectionFactory;
 import org.eclipse.jetty.server.Server;
 import org.eclipse.jetty.server.ServerConnector;
 import org.eclipse.jetty.servlet.ServletContextHandler;
 import org.eclipse.jetty.servlet.ServletHolder;
+import org.geneontology.minerva.BlazegraphMolecularModelManager;
 import org.geneontology.minerva.MinervaOWLGraphWrapper;
 import org.geneontology.minerva.ModelReaderHelper;
 import org.geneontology.minerva.UndoAwareMolecularModelManager;
@@ -93,6 +97,10 @@ public static class MinervaStartUpConfig {
         public MinervaShexValidator shex;
         public String pathToOntologyJournal;
 
+        public String pathToGPADSheX;
+
+        public String pathToGPADShapeMap;
+
     }
 
     public static void main(String[] args) throws Exception {
@@ -172,6 +180,10 @@ else if (opts.nextEq("--golr-labels")) {
                 conf.sparqlEndpointTimeout = Integer.parseInt(opts.nextOpt());
             } else if (opts.nextEq("--ontojournal")) {
                 conf.pathToOntologyJournal = opts.nextOpt();
+            } else if (opts.nextEq("--gpad-shex")) {
+                conf.pathToGPADSheX = opts.nextOpt();
+            } else if (opts.nextEq("--gpad-shape-map")) {
+                conf.pathToGPADShapeMap = opts.nextOpt();
             } else {
                 break;
             }
@@ -340,6 +352,11 @@ public static Server startUp(final MinervaStartUpConfig conf)
                 conf.curieHandler, conf.modelIdPrefix, conf.journalFile, conf.exportFolder, conf.pathToOntologyJournal, true);
         // set pre and post file handlers
         models.addPostLoadOntologyFilter(ModelReaderHelper.INSTANCE);
+        if (conf.pathToGPADSheX != null && conf.pathToGPADShapeMap != null) {
+            ShexSchema schema = ShExC.parse(conf.pathToGPADSheX);
+            ShapeMap shapeMap = ShExC.parseShapeMap(conf.pathToGPADShapeMap);
+            models.addPreFileSaveHandler(new BlazegraphMolecularModelManager.ShexValidatingPreSaveFileHandler(schema, shapeMap, "http://geneontology.org/lego/conforms_to_gpad"));
+        }
         //	conf.shex.tbox_reasoner = models.getTbox_reasoner();
         conf.shex.setGo_lego_repo(models.getGolego_repo());
         conf.shex.curieHandler = conf.curieHandler;

diff --git a/pom.xml b/pom.xml
@@ -325,7 +325,7 @@
             <dependency>
                 <groupId>org.apache.jena</groupId>
                 <artifactId>apache-jena-libs</artifactId>
-                <version>3.12.0</version>
+                <version>4.9.0</version>
                 <type>pom</type>
             </dependency>
             <dependency>
@@ -344,6 +344,7 @@
                 <artifactId>log4j-core</artifactId>
                 <version>${log4j.version}</version>
             </dependency>
+
             <dependency>
                 <groupId>org.apache.logging.log4j</groupId>
                 <artifactId>log4j-1.2-api</artifactId>
@@ -358,6 +359,30 @@
                         <groupId>log4j</groupId>
                         <artifactId>log4j</artifactId>
                     </exclusion>
+                    <exclusion>
+                        <groupId>org.apache.jena</groupId>
+                        <artifactId>jena-arq</artifactId>
+                    </exclusion>
+                    <exclusion>
+                        <groupId>org.apache.jena</groupId>
+                        <artifactId>jena-base</artifactId>
+                    </exclusion>
+                    <exclusion>
+                        <groupId>org.apache.jena</groupId>
+                        <artifactId>jena-core</artifactId>
+                    </exclusion>
+                    <exclusion>
+                        <groupId>org.apache.jena</groupId>
+                        <artifactId>jena-iri</artifactId>
+                    </exclusion>
+                    <exclusion>
+                        <groupId>org.apache.jena</groupId>
+                        <artifactId>jena-shaded-guava</artifactId>
+                    </exclusion>
+                    <exclusion>
+                        <groupId>org.apache.jena</groupId>
+                        <artifactId>jena-tdb</artifactId>
+                    </exclusion>
                 </exclusions>
             </dependency>
         </dependencies>