Added a util method to auto-fix invalid URIs; use that in Normalizer.

json-converter/src/test/java/org/biopax/paxtools/io/jsonld/JsonldBiopaxConverterTest.java

@@ -31,10 +31,10 @@ final void testSomePc14DemoPathway() throws IOException {
 
 		// convert owl test file in resource directory to jsonld format
 		InputStream in = getClass().getResourceAsStream("/demo-pathway.owl");
-		//- there is no rdf:datatype=... anymore; should be fine as the datatypes are defined in the biopax-level3.owl spec!
-		//todo: for some reason, Jena libs v4 or v5 fail at e.g. rdf:about="TEST_CHEBI:cs_26d67131a0608673ae6a683d1dad18f7",
+		//there is no rdf:datatype=... anymore; should be fine as the datatypes are defined in the biopax-level3.owl spec;
+		//Jena libs v4, v5 fail at e.g. rdf:about="TEST_CHEBI:cs_26d67131a0608673ae6a683d1dad18f7" (not a valid URI due to '_' in the prefix),
 		//but jena v3 just prints warnings, e.g.: org.apache.jena.riot - [line: 155, col: 82] {W107} Bad URI: <TEST_CHEBI:cs_26d67131a0608673ae6a683d1dad18f7> Code: 0/ILLEGAL_CHARACTER in SCHEME: The character violates the grammar rules for URIs/IRIs.
-		//howver, removing the underscore from TEST_CHEBI - makes those warning/errors go away...
+		//however, removing the underscore (or replacing with '.') from TEST_CHEBI - makes those warning/errors go away - so I did in demo-pathway.owl
 
 		ByteArrayOutputStream baos = new ByteArrayOutputStream();
 		converter.convertToJsonld(in, baos);

json-converter/src/test/resources/demo-pathway.owl

@@ -147,7 +147,7 @@
  <bp:db>hgnc.symbol</bp:db>
 </bp:RelationshipXref>
 
-<bp:ChemicalStructure rdf:about="TEST_CHEBI:cs_26d67131a0608673ae6a683d1dad18f7">
+<bp:ChemicalStructure rdf:about="TESTCHEBI:cs_26d67131a0608673ae6a683d1dad18f7">
  <bp:structureFormat>InChI</bp:structureFormat>
  <bp:structureData>InChI=1S/C10H18O/c1-5-10(4,11)8-6-7-9(2)3/h5,7,11H,1,6,8H2,2-4H3/t10-/m0/s1</bp:structureData>
 </bp:ChemicalStructure>
@@ -158,7 +158,7 @@
  <bp:db>ncbigene</bp:db>
 </bp:RelationshipXref>
 
-<bp:ChemicalStructure rdf:about="TESTCHEBI:cs_d763940a9c24185a3d1806945a0beb4a">
+<bp:ChemicalStructure rdf:about="TEST.CHEBI:cs_d763940a9c24185a3d1806945a0beb4a">
  <bp:structureFormat>InChI</bp:structureFormat>
  <bp:structureData>InChI=1S/C10H16/c1-7-8-4-5-9(6-8)10(7,2)3/h8-9H,1,4-6H2,2-3H3/t8-,9+/m0/s1</bp:structureData>
 </bp:ChemicalStructure>
@@ -452,7 +452,7 @@
  <bp:comment>A linalool that has formula C10H18O.</bp:comment>
  <bp:comment>is_conjugate_acid_of 422</bp:comment>
  <bp:comment>is_enantiomer_of 98</bp:comment>
- <bp:structure rdf:resource="TEST_CHEBI:cs_26d67131a0608673ae6a683d1dad18f7" />
+ <bp:structure rdf:resource="TESTCHEBI:cs_26d67131a0608673ae6a683d1dad18f7" />
 </bp:SmallMoleculeReference>
 
 <bp:RelationshipXref rdf:ID="RX_chebi_CHEBI_17580_multiple_parent_reference">
@@ -490,7 +490,7 @@
  <bp:name>CC1(C)[C@@H]2CC[C@@H](C2)C1=C</bp:name>
  <bp:comment>is_enantiomer_of 89</bp:comment>
  <bp:comment>A camphene that has formula C10H16.</bp:comment>
- <bp:structure rdf:resource="TESTCHEBI:cs_d763940a9c24185a3d1806945a0beb4a" />
+ <bp:structure rdf:resource="TEST.CHEBI:cs_d763940a9c24185a3d1806945a0beb4a" />
 </bp:SmallMoleculeReference>
 
 <bp:UnificationXref rdf:ID="UX_protein_modification_ontology_MOD_00046">

normalizer/src/main/java/org/biopax/paxtools/normalizer/Normalizer.java

@@ -484,13 +484,15 @@ public void normalize(Model model) {
 	 */
 	public void normalize(Model model, boolean usePrefixAsDbName) {
 
-		if(model.getLevel() != BioPAXLevel.L3)
+		if(model.getLevel() != BioPAXLevel.L3) {
 			throw new IllegalArgumentException("Not Level3 model. " +
-				"Consider converting it first (e.g., with the PaxTools).");
+					"Consider converting it first (e.g., with the PaxTools).");
+		}
 
 		//if set, update the xml:base
-		if(xmlBase != null && !xmlBase.isEmpty())
+		if(xmlBase != null && !xmlBase.isEmpty()) {
 			model.setXmlBase(xmlBase);
+		}
 
 		// Normalize/merge xrefs first and then - CVs
 		// (xrefs could have URIs that should be instead used for CV, PR, SMR or BS biopax types)
@@ -518,15 +520,15 @@ public void normalize(Model model, boolean usePrefixAsDbName) {
 
 		log.info("Normalizing entity references..." + description);
 		normalizeERs(model);
+
+		log.info("Fixing invalid URIs if any...");
+		ModelUtils.fixInvalidUris(model);
 
 		// find/add lost (in replace) children
 		log.info("Repairing..." + description);
 		model.repair(); // it does not remove dangling utility class objects (can be done separately, later, if needed)
-
-		log.info("Optional tasks (reasoning)..." + description);
 	}
 
-
 	private void normalizeCVs(Model model) {
 
 		NormalizerMap map = new NormalizerMap(model);

normalizer/src/test/java/org/biopax/paxtools/normalizer/NormalizerTest.java

@@ -1,6 +1,5 @@
 package org.biopax.paxtools.normalizer;
 
-
 import org.apache.commons.lang3.StringUtils;
 import org.biopax.paxtools.io.SimpleIOHandler;
 import org.biopax.paxtools.model.BioPAXElement;
@@ -366,23 +365,4 @@ void normalizeInoh() {
 		e = model.getByID(model.getXmlBase() + "IMR_0100366_G_alpha_s_Canonical");
 		assertTrue(e instanceof ProteinReference);
 	}
-
-	private void print(XReferrable xr, Model m) {
-		System.out.println();
-		System.out.println("model=" + m.contains(xr) + ":\t"
-				+ xr.getUri() +
-				" is " + xr.getModelInterface().getSimpleName()
-				+ " and has xrefs: ");
-		for(Xref x : xr.getXref()) {
-			System.out.println("model=" + m.contains(x) + ":\t"
-					+"  " + x + " is "
-					+ x.getModelInterface().getSimpleName()
-					+ " - " + x.getUri() + ", db=" + x.getDb()
-					+ ", id=" + x.getId() + ", idVer=" + x.getIdVersion());
-			for(XReferrable rx : x.getXrefOf()) {
-				System.out.println("model=" + m.contains(rx) + ":\t"
-						+ "    xrefOf: " + rx);
-			}
-		}
-	}
 }

paxtools-core/src/main/java/org/biopax/paxtools/controller/ModelUtils.java

@@ -1,5 +1,6 @@
 package org.biopax.paxtools.controller;
 
+import org.apache.commons.lang3.StringUtils;
 import org.biopax.paxtools.impl.BioPAXElementImpl;
 import org.biopax.paxtools.io.BioPAXIOHandler;
 import org.biopax.paxtools.io.SimpleIOHandler;
@@ -15,6 +16,7 @@
 
 import java.io.*;
 import java.lang.reflect.Method;
+import java.net.URI;
 import java.security.MessageDigest;
 import java.security.NoSuchAlgorithmException;
 import java.util.*;
@@ -1290,7 +1292,6 @@ public static  void updateUri(Model model, BioPAXElement el, String newUri) {
 			m.setAccessible(true);
 			m.invoke(el, newUri);
 		} catch (Exception e) {
-			e.printStackTrace();
 			throw new RuntimeException(e);
 		}
 
@@ -1342,4 +1343,17 @@ public static boolean isGeneric(BioPAXElement e) {
 		);
 		//false when e==null
 	}
+
+	public static void fixInvalidUris(Model model) {
+		String prefix = StringUtils.isBlank(model.getXmlBase()) ? "" : model.getXmlBase();
+		for(BioPAXElement bpe : new HashSet<>(model.getObjects())) {
+			try {
+				URI.create(bpe.getUri());
+			} catch (IllegalArgumentException e) {
+				String uri = prefix + md5hex(bpe.getUri());
+				LOG.info("Replaced invalid URI: '{}' with generated: '{}'", bpe.getUri(), uri);
+				updateUri(model, bpe, uri);
+			}
+		}
+	}
 }

paxtools-core/src/main/java/org/biopax/paxtools/io/SimpleIOHandler.java

@@ -20,6 +20,7 @@
 import javax.xml.stream.XMLStreamException;
 import javax.xml.stream.XMLStreamReader;
 import java.io.*;
+import java.net.URI;
 import java.nio.charset.StandardCharsets;
 import java.util.*;
 
@@ -363,6 +364,11 @@ private String processIndividual(Model model) throws XMLStreamException
 			throw new BioPaxIOException(
 					String.format("Error processing %s%s (rdf:ID/rdf:about not found)", r.getNamespaceURI(), getXmlStreamInfo()));
 		}
+		try {
+			URI.create(id);
+		} catch (IllegalArgumentException e) {
+			log.error("Invalid URI '{}' at {}{}", id, r.getNamespaceURI(), getXmlStreamInfo());
+		}
 
 		Class<? extends BioPAXElement> type;
 		try {
@@ -387,7 +393,7 @@ private String processIndividual(Model model) throws XMLStreamException
 		} else
 		{
 			//abstract BioPAX types, e.g. Entity, UtilityClass, cannot be used directly in RDF+XML model/file!
-			log.error(String.format("Ignoring abstract %s, id: %s", (r.hasText()?r.getText():getXmlStreamInfo()), id));
+			log.error("Ignoring abstract {}, id: {}", (r.hasText()?r.getText():getXmlStreamInfo()), id);
 			//id = null; //todo: uncomment/test (currently, ignored object's uri can become parent's property value, e.g. CV term)
 			//skip(); //was a bug - throws a misleading exception at the next element in some cases
 			//todo: shall we instead throw an exception when e.g. <term><Entity rdf:ID="Gene"></Entity></term>?

paxtools-core/src/test/java/org/biopax/paxtools/controller/ModelUtilsTest.java

@@ -99,10 +99,10 @@ public final void mergeAndReplace() {
 		assertEquals(8, m.getObjects().size()); // + pr3
 		assertTrue(m.contains(pr3)); // added!
 
-		assertTrue(m.contains(pr2)); // not deleted (may be dangling now)!
-		assertTrue(m.contains(x2)); // not deleted (may be dangling now)!
-		assertTrue(m.contains(pr1)); // not deleted (may be dangling now)!
-		assertTrue(m.contains(x1)); // not deleted (may be dangling now)!
+		assertTrue(m.contains(pr2)); // not deleted (maybe dangling now)!
+		assertTrue(m.contains(x2)); // not deleted (maybe dangling now)!
+		assertTrue(m.contains(pr1)); // not deleted (maybe dangling now)!
+		assertTrue(m.contains(x1)); // not deleted (maybe dangling now)!
 
 		// delete dangling
 		ModelUtils.removeObjectsIfDangling(m, ProteinReference.class);