在Java中剥离无效的XML字符

import org.xml.sax.*;
import java.io.*;

public class StripInvalidXmlChars {
   public static void main(String[] args) throws Exception {
      String inputText = "Hello, world!";
      String strippedText = stripInvalidXmlChars(inputText);
      System.out.println("Original Text: \n" + inputText);
      System.out.println("Stripped Text: \n" + strippedText);
   }

   public static String stripInvalidXmlChars(String inputText) throws SAXException {
      XMLReader xmlReader = XMLReaderFactory.createXMLReader();
      InputSource inputSource = new InputSource(new StringReader(inputText));
      xmlReader.setContentHandler(new StripContentHandler(xmlReader));
      xmlReader.parse();
      return xmlReader.getXMLReader().getLexicalHandler().getDomNode().getStringValue();
   }
}

class StripContentHandler implements ContentHandler {
   private XMLReader xmlReader;

   public StripContentHandler(XMLReader xmlReader) {
      this.xmlReader = xmlReader;
   }

   public void startDocument() throws SAXException {
      xmlReader.getLexicalHandler().startDocument();
   }

   public void startElement(String uri, String localName, String qName, Attributes atts) throws SAXException {
      xmlReader.getLexicalHandler().startElement(uri, localName, qName, atts);
   }

   public void characters(char[] ch, int start, int length) throws SAXException {
      xmlReader.getLocator().setCharacterStream(new StringReader(new String(ch, start, length)));
      xmlReader.getLexicalHandler().characters();
   }

   public void ignorableWhitespace(char[] ch, int start, int length) throws SAXException {
      xmlReader.getLocator().setCharacterStream(new StringReader(new String(ch, start, length)));
      xmlReader.getLexicalHandler().ignorableWhitespace();
   }

   public void endElement(String uri, String localName, String qName) throws SAXException {
      boolean hasInvalidChars = false;
      String chars = xmlReader.getLexicalHandler().endElement(uri, localName, qName);
      for (int i = 0; i< chars.length(); i++) {
         char c = chars.charAt(i);
         if (!Character.isLetterOrDigit(c)
              || c == ':' || c == '.' || c == ',' || c == '='
              || (c >= '0' && c <= '9') && c != ']' && c != ']'
              || c >= '\u0080' && !Character.isISOControl(c)) {
            hasInvalidChars = true;
            break;
         }
      }
      if (hasInvalidChars) {
         xmlReader.getLocator().setCharacterStream(new StringReader(new String(chars).replaceAll("[^A-Za-z0-9:.,=]", ""))).findCharacterEncoding();
         xmlReader.setContentHandler(new StripContentHandler(xmlReader));
         xmlReader.parse();
         return;
      }
      xmlReader.getLexicalHandler().endElement(uri, localName, qName);
   }

   public void endDocument() throws SAXException {
      xmlReader.getLexicalHandler().endDocument();
   }
}