Fix RDF issues testing 4.1.0 RC 1 #412

This commit adds a test for ill formed XML literals, which are expected to fail to serialize. The issue is that ill formed literals would create ontologies that cannot be parsed back.
owlcs · Jun 5, 2015 · aeb9738 · aeb9738
1 parent 7204fb6
commit aeb9738
Show file tree

Hide file tree

Showing 2 changed files with 34 additions and 4 deletions.
diff --git a/.../src/test/java/org/semanticweb/owlapi/api/test/literals/OWLLiteralCorruptionTestCase.java b/.../src/test/java/org/semanticweb/owlapi/api/test/literals/OWLLiteralCorruptionTestCase.java
@@ -52,6 +52,20 @@ public void shouldRoundTripXMLLiteral() throws OWLOntologyStorageException {
         assertTrue(string.contains(literal));
     }
 
+    @Test
+    public void shouldFailOnMalformedXMLLiteral() throws OWLOntologyCreationException, OWLOntologyStorageException {
+        String literal = "<ncicp:ComplexDefinition><ncicp:def-definition>A form of cancer that begins in melanocytes (cells that make the pigment melanin). It may begin in a mole (skin melanoma), but can also begin in other pigmented tissues, such as in the eye or in the intestines.</ncicp:def-definition><ncicp:def-source>NCI-GLOSS</ncicp:def-source></ncicp:ComplexDefinition>";
+        OWLOntology o = m.createOntology();
+        OWLDataProperty p = df.getOWLDataProperty(IRI.create("urn:test#p"));
+        OWLLiteral l = df.getOWLLiteral(literal, OWL2Datatype.RDF_XML_LITERAL);
+        OWLNamedIndividual i = df.getOWLNamedIndividual(IRI.create("urn:test#i"));
+        o.add( df.getOWLDataPropertyAssertionAxiom(p, i, l));
+        expectedException.expect(OWLOntologyStorageException.class);
+        expectedException.expectMessage(literal);
+        expectedException.expectMessage("XML literal is not self contained");
+        saveOntology(o).toString();
+    }
+
     @Test
     public void shouldRoundtripPaddedLiterals() throws OWLOntologyCreationException, OWLOntologyStorageException {
         String in = "Prefix(:=<urn:test#>)\n" + "Prefix(a:=<urn:test#>)\n"

diff --git a/parsers/src/main/java/org/semanticweb/owlapi/rdf/rdfxml/renderer/XMLWriterImpl.java b/parsers/src/main/java/org/semanticweb/owlapi/rdf/rdfxml/renderer/XMLWriterImpl.java
@@ -14,7 +14,9 @@
 
 import static org.semanticweb.owlapi.util.OWLAPIPreconditions.*;
 
+import java.io.IOException;
 import java.io.PrintWriter;
+import java.io.StringReader;
 import java.util.*;
 
 import javax.annotation.Nonnull;
@@ -23,7 +25,11 @@
 import org.semanticweb.owlapi.io.XMLUtils;
 import org.semanticweb.owlapi.model.IRI;
 import org.semanticweb.owlapi.model.OWLRuntimeException;
+import org.semanticweb.owlapi.util.SAXParsers;
 import org.semanticweb.owlapi.util.StringLengthComparator;
+import org.xml.sax.InputSource;
+import org.xml.sax.SAXException;
+import org.xml.sax.helpers.DefaultHandler;
 
 import com.google.common.base.Splitter;
 import com.google.common.collect.Lists;
@@ -59,10 +65,10 @@ public class XMLWriterImpl implements XMLWriter {
      *        xml writer preferences instance
      */
     public XMLWriterImpl(PrintWriter writer, XMLWriterNamespaceManager xmlWriterNamespaceManager, String xmlBase,
-            XMLWriterPreferences preferences) {
+        XMLWriterPreferences preferences) {
         this.writer = checkNotNull(writer, "writer cannot be null");
         this.xmlWriterNamespaceManager = checkNotNull(xmlWriterNamespaceManager,
-                "xmlWriterNamespaceManager cannot be null");
+            "xmlWriterNamespaceManager cannot be null");
         this.xmlBase = checkNotNull(xmlBase, "xmlBase cannot be null");
         xmlPreferences = checkNotNull(preferences, "preferences cannot be null");
         elementStack = new Stack<>();
@@ -237,8 +243,8 @@ public void startDocument(IRI rootElement) {
         }
         for (String curPrefix : xmlWriterNamespaceManager.getPrefixes()) {
             if (!curPrefix.isEmpty()) {
-                writeAttribute("xmlns:" + curPrefix,
-                        verifyNotNull(xmlWriterNamespaceManager.getNamespaceForPrefix(curPrefix)));
+                writeAttribute("xmlns:" + curPrefix, verifyNotNull(xmlWriterNamespaceManager.getNamespaceForPrefix(
+                    curPrefix)));
             }
         }
     }
@@ -404,13 +410,23 @@ private void writeTextContent() {
             if (textContent != null) {
                 // only escape the data if this is not an XML literal
                 if ("http://www.w3.org/1999/02/22-rdf-syntax-ns#XMLLiteral".equals(attributes.get("rdf:datatype"))) {
+                    checkProperXMLLiteral(textContent);
                     writer.write(textContent);
                 } else {
                     writer.write(XMLUtils.escapeXML(verifyNotNull(textContent)));
                 }
             }
         }
 
+        private void checkProperXMLLiteral(String text) {
+            try {
+                SAXParsers.initParserWithOWLAPIStandards(null).parse(new InputSource(new StringReader(text)),
+                    new DefaultHandler());
+            } catch (SAXException | IOException e) {
+                throw new OWLRuntimeException("XML literal is not self contained: \"" + text + "\"", e);
+            }
+        }
+
         private void insertIndentation() {
             if (xmlPreferences.isIndenting()) {
                 for (int i = 0; i < indentation * xmlPreferences.getIndentSize(); i++) {