Java XML Numeric Character References

Question

I am having an issue when I parse an XML document that has numeric character references (ie &#xA0). The problem I am running into is that when the document is parsed, the & is replaced with &amp ; (without the space before the ;), so my parsed document will contain &amp ;#xA0;. How do I stop this from happening? I have tried using xmlDoc.setExpandEntityReferences(false), but that doesnt seem to change anything.

Here is my code for parsing the document:

public static Document getXmlDoc(File xmlFile) throws ParserConfigurationException, SAXExeption, IOException {
    DocumentBuilderFactory factory = DocumentBuilderFactory.newInstance();
    factory.setIgnoringElementContentWhitespace(true);
    factory.setExpandEntityReferences(false);
    DocumentBuilder builder = factory.newDocumentBuilder();
    return builder.parse(xmlFile);
}

Any help would be greatly appreciated.

EDIT:

The XML that is parsed form the above code is modified and then written back to a file. The code to do this is below:

public static File saveXmlDoc(Document xmlDocument, String outputToDir, String outputFilename) throws IOException {
    String outputDir = outputToDir;
    if (!outputDir.endWith(File.separator)) outputDir += File.separator;
    if (!new FIle(outputDir).exists()) new File(outputDir).mkdir();
    File xmlFile = new File(outputDir + outputFilename);
    Transformer transformer = TransformerFactory.newInstance().newTransformer();
    transformer.setOutputProperty(OutputKeys.INDENT, "no");
    StreamResult saveResult = new StreamResult(outputDir + outputFilename);
    DOMSource source = new DOMSource(xmlDocument);
    transformer.transform(source, saveResult);

    return xmlFile;
}

EDIT 2:

Fixed a typo for factory.setIgnoringElementContentWhitespace(true);.

EDIT 3 - My Solution:

Since my reputation is too low to answer my own question, here is the solution I used to fix all of this.

Here are the functions I changed in order to resolve this issue:

To get the XML Document:

    public static Document getXmlDoc(File xmlFile) throws ParserConfigurationException, SAXException, IOException {
        DocumentBuilderFactory factory = DocumentBuilderFactory.newInstance();
        factory.setIgnoringElementContentWhitespace(true);
        factory.setExpandEntityReferences(false);
        DocumentBuilder builder = factory.newDocumentBuilder();
        return builder.parse(xmlFile);
    }

To save the XML Document:

    public static File saveXmlDoc(Document xmlDocument, String outputToDir, String outputFilename) throws Exception {
        readNodesForHexConversion(xmlDocument.getChildNodes());
        String xml = getXmlAsString(xmlDocument);

        // write the xml out to a file
        Exception writeError = null;
        File xmlFile = null;
        FileOutputStream fos = null;
        try {
            if (!new File(outputToDir).exists()) new File(outputToDir).mkdir();
            xmlFile = new File(outputToDir + outputFilename);
            if (!xmlFile.exists()) xmlFile.createNewFile();
            fos = new FileOutputStream(xmlFile);

            byte[] xmlBytes = xml.getBytes("UTF-8");
            fos.write(xmlBytes);
            fos.flush();
        } catch (Exception ex) {
            ex.printStackTrace();
            writeError = ex;
        } finally {
            if (fos != null) fos.close();
            if (writeError != null) throw writeError;
        }

        return xmlFile;
    }

To convert the XML Document to String:

        public static String getXmlAsString(Document xmlDocument) throws TransformerFactoryConfigurationError, TransformerException {
    DOMSource domSource = new DOMSource(xmlDocument);
    StringWriter writer = new StringWriter();
    StreamResult result = new StreamResult(writer);
    Transformer transformer;
    transformer = TransformerFactory.newInstance().newTransformer();
    transformer.transform(domSource, result);
    return writer.toString();
}

Answer 1

I can't reproduce the problem at the moment. Here's a short but complete program which tries to:

import org.w3c.dom.*;
import java.io.*;
import javax.xml.*;
import javax.xml.parsers.*;

public class Test {
    public static void main (String[] args) throws Exception {
        byte[] xml = "<foo>&#xA0;</foo>".getBytes("UTF-8");
        DocumentBuilderFactory factory = DocumentBuilderFactory.newInstance();
        factory.setIgnoringElementContentWhitespace(true);
        factory.setExpandEntityReferences(false);
        DocumentBuilder builder = factory.newDocumentBuilder();
        Document document = builder.parse(new ByteArrayInputStream(xml));
        Element element = document.getDocumentElement();
        String text = element.getFirstChild().getNodeValue();
        System.out.println(text.length()); // Prints 1
        System.out.println((int) text.charAt(0)); // Prints 160
    }
}

Now it's not clear from the above XML would be written out again - and it would help if you'd show the code you're using to do that - but it's clear that the single-character value of the text node is not being read as an ampersand followed by "#xA0;" separately, as I believe your question describes it, so I'd be really surprised to see it written out as " ".

Can you write a similar short but complete program which does demonstrate the problem? Will continue to try to do so myself.