apacheGH-1773: Convert xmlinput to use IRIx

build-files/rat-exclusions.txt

@@ -75,6 +75,7 @@ archive.md
 **/src/test/files/**/*
 **/src/test/resources/**/*
 **/testing/**/*
+jena-core/testing0/**
 
 ## Databases
 **/DB/**/*

jena-arq/src/main/java/org/apache/jena/riot/lang/ReaderRIOTRDFXML.java

@@ -23,7 +23,6 @@
 import java.io.Reader ;
 import java.util.Map;
 
-import org.apache.jena.JenaRuntime;
 import org.apache.jena.atlas.lib.Pair ;
 import org.apache.jena.atlas.logging.Log;
 import org.apache.jena.atlas.web.ContentType;
@@ -33,7 +32,6 @@
 import org.apache.jena.graph.NodeFactory ;
 import org.apache.jena.graph.Triple ;
 import org.apache.jena.irix.IRIs;
-import org.apache.jena.irix.SetupJenaIRI;
 import org.apache.jena.rdf.model.RDFErrorHandler ;
 import org.apache.jena.rdfxml.xmlinput.* ;
 import org.apache.jena.rdfxml.xmlinput.impl.ARPSaxErrorHandler ;
@@ -98,7 +96,7 @@ public void read(Reader reader, String baseURI, ContentType ct, StreamRDF output
     // RDF 1.0 (and RDF/XML) was based on "RDF URI References" which did allow spaces.
 
     // Use with TDB requires this to be "true" - it is set by InitTDB.
-    public static boolean RiotUniformCompatibility = false ;
+    public static boolean RiotUniformCompatibility = true ;
     // Warnings in ARP that should be errors to be compatible with
     // non-XML-based languages.  e.g. language tags should be
     // syntactically valid.
@@ -162,9 +160,6 @@ public void parse() {
                 arpOptions.setErrorMode(code, ARPErrorNumbers.EM_ERROR) ;
         }
 
-        if ( JenaRuntime.isRDF11 )
-            arp.getOptions().setIRIFactory(SetupJenaIRI.iriFactory_RDFXML());
-
         if ( context != null ) {
             Map<String, Object> properties = null;
             try {
@@ -201,8 +196,8 @@ public void parse() {
     private static String baseURI_RDFXML(String baseIRI) {
         if ( baseIRI == null )
             return IRIs.getBaseStr();
-        else
-            return IRIs.toBase(baseIRI) ;
+        // RDFParserBuidler resolved the baseIRI
+        return baseIRI;
     }
 
     private static class HandlerSink extends ARPSaxErrorHandler implements StatementHandler, NamespaceHandler {

jena-arq/src/main/java/org/apache/jena/riot/system/ParserProfileStd.java

@@ -113,6 +113,12 @@ private IRIx internalMakeIRI(String uriStr, long line, long col) {
     }
 
     private void doChecking(IRIx irix, String uriStr, long line, long col) {
+        // Should become ...
+//        irix.handleViolations((isError, message)->{
+//            if ( isError )
+//                errorHandler.error(message, line, col);
+//        });
+
         IRI iri;
         if ( irix instanceof IRIProviderJenaIRI.IRIxJena )
             iri = (IRI)irix.getImpl();

jena-arq/src/main/java/org/apache/jena/sparql/util/NodeUtils.java

@@ -26,7 +26,7 @@
 import org.apache.jena.datatypes.xsd.XSDDatatype ;
 import org.apache.jena.graph.Node ;
 import org.apache.jena.graph.NodeFactory ;
-import org.apache.jena.iri.IRI ;
+import org.apache.jena.irix.IRIx;
 import org.apache.jena.rdf.model.impl.Util;
 import org.apache.jena.sparql.expr.ExprEvalException ;
 import org.apache.jena.sparql.expr.NodeValue ;
@@ -40,8 +40,12 @@
  */
 public class NodeUtils
 {
-    /** IRI to Node */
-    public static Node asNode(IRI iri) {
+    /**
+     * IRI to Node
+     * @deprecated Do not use org.apache.jena.iri.IRI. Use {@link IRIx}.
+     */
+    @Deprecated
+    public static Node asNode(org.apache.jena.iri.IRI iri) {
         return NodeFactory.createURI(iri.toString()) ;
     }
 

jena-core/src/main/java/org/apache/jena/datatypes/xsd/impl/XMLLiteralType.java

@@ -38,17 +38,20 @@
  * by use of <code>rdf:parsetype='literal'</code>.
  */
 public class XMLLiteralType extends BaseDatatype implements RDFDatatype {
+
+
+    public static String XMLLiteralTypeURI = "http://www.w3.org/1999/02/22-rdf-syntax-ns#XMLLiteral";
     /** Singleton instance */
     // Include the string for the RDF namespace, not use RDF.getURI(), to avoid an initializer circularity
-    public static final RDFDatatype theXMLLiteralType = new XMLLiteralType("http://www.w3.org/1999/02/22-rdf-syntax-ns#XMLLiteral");
-    
+    public static final RDFDatatype theXMLLiteralType = new XMLLiteralType(XMLLiteralTypeURI);
+
     /**
      * Private constructor.
      */
     private XMLLiteralType(String uri) {
         super(uri);
     }
-    
+
     /**
      * Convert a serialize a value of this datatype out
      * to lexical form.
@@ -57,7 +60,7 @@ private XMLLiteralType(String uri) {
     public String unparse(Object value) {
         return value.toString();
     }
-    
+
     /**
      * Parse a lexical form of this datatype to a value
      * @throws DatatypeFormatException if the lexical form is not legal
@@ -68,7 +71,7 @@ public Object parse(String lexicalForm) throws DatatypeFormatException {
           throw new DatatypeFormatException("Bad rdf:XMLLiteral");
         return lexicalForm;
     }
-    
+
     /**
      * Test whether the given string is a legal lexical form
      * of this datatype.
@@ -89,9 +92,9 @@ public boolean isValid(final String lexicalForm) {
         // status[0] true on error or other reason to know that this is not well-formed
         // status[1] true once first triple found
         // status[2] the result (good if status[1] and not status[0]).
-        
+
         ARP arp = new ARP();
-        
+
         arp.getHandlers().setErrorHandler(new ErrorHandler(){
         	@Override
             public void fatalError(SAXParseException e){
@@ -112,7 +115,7 @@ public void statement(AResource a, AResource b, ALiteral l){
         	/* this method is invoked exactly once
         	 * while parsing the dummy document.
         	 * The l argument is in exclusive canonical XML and
-        	 * corresponds to where the lexical form has been 
+        	 * corresponds to where the lexical form has been
         	 * in the dummy document. The lexical form is valid
         	 * iff it is unchanged.
         	 */
@@ -130,24 +133,24 @@ public void statement(AResource a, AResource b, AResource l){
 	    }
         });
         try {
-        
+
         arp.load(new StringReader(
         "<rdf:RDF  xmlns:rdf='http://www.w3.org/1999/02/22-rdf-syntax-ns#'>\n"
         +"<rdf:Description><rdf:value rdf:parseType='Literal'>"
         +lexicalForm+"</rdf:value>\n"
         +"</rdf:Description></rdf:RDF>"
         ));
-        
+
         }
         catch (IOException ioe){
-           throw new BrokenException(ioe);	
+           throw new BrokenException(ioe);
         }
         catch (SAXException s){
         	return false;
         }
-        
-        
+
+
         return (!status[0])&&status[1]&&status[2];
-    }    
+    }
 
 }

jena-core/src/main/java/org/apache/jena/irix/IRIProviderJenaIRI.java

@@ -195,8 +195,6 @@ public String toString() {
         return this.getClass().getSimpleName();
     }
 
-    private static IRI baseIRI = null;
-
     // -----------------------------
 
     private static IRIFactory iriFactory() {

jena-core/src/main/java/org/apache/jena/irix/IRIs.java

@@ -78,6 +78,8 @@ public static String getBaseStr() {
      * turn it into a IRI suitable as a base IRI.
      */
     public static String toBase(String baseURI) {
+        if ( baseURI == null )
+            return getBaseStr();
         String scheme = scheme(baseURI);
         if ( Sys.isWindows ) {
             // Assume a scheme of one letter is a Windows drive letter.

jena-core/src/main/java/org/apache/jena/irix/IRIx.java

@@ -194,8 +194,8 @@ protected IRIx(String string) {
     public abstract boolean hasViolations();
 
     /**
-     * Handle violations by sending a boolean, indicating whether this is an error or
-     * a warning, and string message to a handler.
+     * Handle violations by sending a boolean, indicating whether this is an error (true) or
+     * a warning (false), and string message to a handler.
      */
     public abstract void handleViolations(BiConsumer<Boolean, String> violation);
 

jena-core/src/main/java/org/apache/jena/irix/SetupJenaIRI.java

@@ -27,10 +27,6 @@
 /** Setup of jena-iri package IRI Factory for parsing and for checking. */
 public class SetupJenaIRI {
 
-    // Currently, the same.
-    // The difference is the treatment in IRIProviderJenaIRI and ParserProfileStd.internalMakeIRI
-    // both of which can be scheme and component sensitive.
-
     private static final IRIFactory iriFactoryInst = setupIRIFactory();
     private static final IRIFactory iriCheckerInst = setupCheckerIRIFactory();
 
@@ -43,11 +39,6 @@ public static IRIFactory iriFactory() {
         return iriFactoryInst;
     }
 
-    public static IRIFactory iriFactory_RDFXML() {
-        // Used in ReaderRiotRDFXML
-        return iriFactory();
-    }
-
     /**
      * An IRIFactory with more detailed warnings.
      */

jena-core/src/main/java/org/apache/jena/irix/SystemIRIx.java

@@ -28,30 +28,39 @@
 public class SystemIRIx {
 
     // -- Providers
-    private static IRIProvider providerJenaIRI = new IRIProviderJenaIRI();
-    static {
-        providerJenaIRI.strictMode("urn",  false);
-        providerJenaIRI.strictMode("http", false);
-        providerJenaIRI.strictMode("file", false);
+    private static IRIProvider makeProviderJenaIRI() {
+        IRIProvider newProviderJenaIRI = new IRIProviderJenaIRI();
+        newProviderJenaIRI.strictMode("urn",  false);
+        newProviderJenaIRI.strictMode("http", false);
+        newProviderJenaIRI.strictMode("file", false);
+        return newProviderJenaIRI;
     }
+    private static IRIProvider providerJenaIRI = makeProviderJenaIRI();
 
-    private static IRIProvider providerJDK     = new IRIProviderJDK();
     // ** Do not use IRIProviderJDK in production. **
-    static {}
+    private static IRIProvider makeProviderJDK() { return new IRIProviderJDK(); }
 
-//    private static IRIProvider providerIRI3986 = new IRIProvider3986();
-//    static {
-//        providerIRI3986.strictMode("urn",  true);
-//        providerIRI3986.strictMode("http", true);
-//        providerIRI3986.strictMode("file", true);
+//    private static IRIProvider makeProviderIRI3986() {
+//        IRIProvider newProviderIRI3986 = new IRIProvider3986();
+//        newProviderIRI3986.strictMode("urn", true);
+//        newProviderIRI3986.strictMode("http", true);
+//        newProviderIRI3986.strictMode("file", true);
 //    }
+//    private static IRIProvider providerIRI3986 = makeProviderIRI3986();
 
     // -- System-wide provider.
 
-    private static IRIProvider provider = providerJenaIRI;
-    //private static IRIProvider provider = providerIRI3986;
+    public static IRIProvider makeFreshSystemProvider() {
+        // Choice point.
+        return makeProviderJenaIRI();
+    }
+
+    private static IRIProvider provider = makeFreshSystemProvider();
 
     public static void init() {}
+    public static void reset() {
+        provider = makeFreshSystemProvider();
+    }
 
     public static void setProvider(IRIProvider aProvider) {
         provider = aProvider;