source: trunk/gli/src/org/greenstone/gatherer/util/XMLTools.java@ 8766

package org.greenstone.gatherer.util;


import java.io.*;
import java.net.*;
import java.util.*;
import org.apache.xerces.parsers.*;
import org.apache.xml.serialize.*;
import org.greenstone.gatherer.DebugStream;
import org.w3c.dom.*;
import org.xml.sax.*;


/** This class is a static class containing useful XML functions */
public class XMLTools
{
    /** Remove all of the child nodes from a certain node. */
    static final public void clear(Node node)
    {
    while (node.hasChildNodes()) {
        node.removeChild(node.getFirstChild());
    }
    }


    static public ArrayList getChildElementsByTagName(Element parent_element, String element_name)
    {
    ArrayList child_elements = new ArrayList();

    NodeList children_nodelist = parent_element.getChildNodes();
    for (int i = 0; i < children_nodelist.getLength(); i++) {
        Node child_node = children_nodelist.item(i);
        if (child_node.getNodeType() == Node.ELEMENT_NODE && child_node.getNodeName().equals(element_name)) {
        child_elements.add(child_node);
        }
    }

    return child_elements;
    }


    static public String getElementTextValue(Element element)
    {
    // Find the first text node child
    NodeList children_nodelist = element.getChildNodes();
    for (int i = 0; i < children_nodelist.getLength(); i++) {
        Node child_node = children_nodelist.item(i);
        if (child_node.getNodeType() == Node.TEXT_NODE) {
        return child_node.getNodeValue();
        }
    }

    // None found
    return "";
    }


    /** Method to retrieve the value of a given node.
     * @param element The <strong>Element</strong> whose value we wish to find.
     * Soon to be deprecated!
     */
    static final public String getValue(Node element) {
    // If we've been given a subject node first retrieve its value node.
    if(element.getNodeName().equals("Subject")) {
        element = getNodeFromNamed(element, "Value");
    }
    // If we've got a value node, then reconstruct the text. Remember that DOM will split text over 256 characters into several text nodes
    if(element != null && element.hasChildNodes()) {
        StringBuffer text_buffer = new StringBuffer();
        NodeList text_nodes = element.getChildNodes();
        for(int i = 0; i < text_nodes.getLength(); i++) {
        Node possible_text = text_nodes.item(i);
        if(possible_text.getNodeName().equals(StaticStrings.TEXT_NODE)) {
            text_buffer.append(possible_text.getNodeValue());
        }
        }
        return text_buffer.toString();
    }
    return "";
    }


    /** Method to retrieve from the node given, a certain child node with the specified name.
     * @param parent The <strong>Node</strong> whose children should be searched.
     * @param name The required nodes name as a <strong>String</strong>.
     * @return The requested <strong>Node</strong> if it is found, <i>null</i> otherwise.
     * Soon to be deprecated!
     */ 
    static final public Node getNodeFromNamed(Node parent, String name) {
    Node child = null;
    for(Node i = parent.getFirstChild(); i != null && child == null; 
        i = i.getNextSibling()) {
        if(i.getNodeName().equals(name)) {
        child = i;
        }
    }
    return child;
    }


    /** Parse an XML document from a given file */
    static public Document parseXMLFile(File xml_file)
    {
    Document document = null;

    try {
        FileInputStream fis   = new FileInputStream(xml_file);
        InputStreamReader isr = new InputStreamReader(fis, "UTF-8");
        Reader r              = new BufferedReader(isr);
        InputSource isc       = new InputSource(r);
        DOMParser parser      = new DOMParser();
        parser.setFeature("http://xml.org/sax/features/validation", false);
        parser.setFeature("http://apache.org/xml/features/nonvalidating/load-external-dtd", false);
        // May or may not be ignored, the documentation for Xerces is contradictory. If it works then parsing -should- be faster.
        parser.setFeature("http://apache.org/xml/features/dom/defer-node-expansion", true);
        parser.setFeature("http://apache.org/xml/features/dom/include-ignorable-whitespace", false);
        parser.parse(isc);
        document = parser.getDocument();
        isr.close();
        fis.close();
    }
    catch (Exception exception) {
        DebugStream.printStackTrace(exception);
    }

    return document;
    }


    /** Removes characters that are invalid in XML (see http://www.w3.org/TR/2000/REC-xml-20001006#charsets) */
    static public String removeInvalidCharacters(String text)
    {
    char[] safe_characters = new char[text.length()];
    int j = 0;

    char[] raw_characters = new char[text.length()];
    text.getChars(0, text.length(), raw_characters, 0);
    for (int i = 0; i < raw_characters.length; i++) {
        char character = raw_characters[i];
        if ((character >= 0x20 && character <= 0xD7FF) || character == 0x09 || character == 0x0A || character == 0x0D || (character >= 0xE000 && character <= 0xFFFD) || (character >= 0x10000 && character <= 0x10FFFF)) {
        safe_characters[j] = character;
        j++;
        }
    }

    return new String(safe_characters, 0, j);
    }


    /** Set the #text node value of some element.
     * @param element the Element whose value we wish to set
     * @param value the new value for the element as a String
     * Soon to be deprecated!
     */
    static final public void setValue(Element element, String value) {
    // Remove any existing child node(s)
    clear(element);
    // Add new text node.
    if (value != null) {
        element.appendChild(element.getOwnerDocument().createTextNode(value));
    }
    }


    /** Write an XML document to a given file */
    static public void writeXMLFile(File xml_file, Document document)
    {
    try {
        OutputStream os = new FileOutputStream(xml_file);
        // Create an output format for our document.
        OutputFormat f = new OutputFormat(document);
        f.setEncoding("UTF-8");
        f.setIndenting(true);
        f.setLineWidth(0); // Why isn't this working!
        f.setPreserveSpace(false);
        // Create the necessary writer stream for serialization.
        OutputStreamWriter osw = new OutputStreamWriter(os, "UTF-8");
        Writer w               = new BufferedWriter(osw);
        // Generate a new serializer from the above.
        XMLSerializer s        = new XMLSerializer(w, f);
        s.asDOMSerializer();
        // Finally serialize the document to file.
        s.serialize(document);
        // And close.
        os.close();
    }
    catch (Exception exception) {
        DebugStream.printStackTrace(exception);
    }
    }


    /** ------------ OLD FUNCTIONS FROM UTILITY ---------------
    /** Using this method we can request that a certain document be written, as valid XML, to a certain output stream. This makes use of the Xerces Serialization suite, which should in no way be confused with the usual method of Serialization used by Java. */
    static public boolean export(Document document, String filename) {
    return export(document, new File(filename));
    }

    static public boolean export(Document document, File file) {
    try {
        OutputStream os = new FileOutputStream(file);
        // Create an output format for our document.
        OutputFormat f = new OutputFormat(document);
        f.setEncoding("UTF-8");
        f.setIndenting(true);
        f.setLineWidth(0); // Why isn't this working!
        f.setPreserveSpace(false);
        // Create the necessary writer stream for serialization.
        OutputStreamWriter osw = new OutputStreamWriter(os, "UTF-8");
        Writer w               = new BufferedWriter(osw);
        // Generate a new serializer from the above.
        XMLSerializer s        = new XMLSerializer(w, f);
        s.asDOMSerializer();
        // Finally serialize the document to file.
        s.serialize(document);
        // And close.
        os.close();
        return true;
    }
    // A file not found exception is most likely thrown because the directory the metadata.xml file is attempting to be written to no longer has any files in it. I'll add a test in MetadataXMLFile to test for this, but if it still happens ignore it (a non-existant directory can't really have metadata added to it any way.
    catch (Exception exception) {
        if(!file.getName().endsWith(StaticStrings.METADATA_XML)) {
        DebugStream.printStackTrace(exception);
        return false;
        }
        return true;
    }
    }

    /** Parse in a xml document from a given file. */
    static public Document parse(File file) {
    return parse(file, true);
    }

    /** Parse in a xml document from a given file. */
    static public Document parse(File file, boolean noisey) 
    {
    Document document = null;
    try {
        if (file.exists()) {
        DebugStream.println("Parsing XML file: " + file);
        FileInputStream fis = new FileInputStream(file);
        document = parse(fis, noisey);
        }
    }
    catch (Exception error) {
        if(noisey) {
        error.printStackTrace();
        DebugStream.println("Exception in Utility.parse() - Unexpected");
        }
        else {
        DebugStream.println("Exception in Utility.parse() - Expected");
        DebugStream.printStackTrace(error);
        }
    }

    return document;
    }

    /** Parse in a xml document from a given URL. */
    static public Document parse(URL url, boolean noisey) 
    {
    Document document = null;
    try {
        
        URLConnection connection = url.openConnection();
        InputStream is = connection.getInputStream();
        document = parse(is,noisey);
    }
    catch (Exception error) {
        if(noisey) {
        error.printStackTrace();
        DebugStream.println("Exception in Utility.parse() - Unexpected");
        }
        else {
        DebugStream.println("Exception in Utility.parse() - Expected");
        DebugStream.printStackTrace(error);
        }
    }

    return document;
    }

    /** Parse in a xml document from a given file. */
    static public Document parse(InputStream is, boolean noisey) {
    Document document = null;
    try {
        InputStreamReader isr = new InputStreamReader(is, "UTF-8");
        Reader r              = new BufferedReader(isr);
        InputSource isc       = new InputSource(r);
        DOMParser parser      = new DOMParser();
        parser.setFeature("http://xml.org/sax/features/validation", false);
        parser.setFeature("http://apache.org/xml/features/nonvalidating/load-external-dtd", false);
                // May or may not be ignored, the documentation for Xerces is contradictory. If it works then parsing -should- be faster.
        parser.setFeature("http://apache.org/xml/features/dom/defer-node-expansion", true);
        parser.setFeature("http://apache.org/xml/features/dom/include-ignorable-whitespace", false);
        parser.parse(isc);
        document = parser.getDocument();
        isr.close();
        is.close();
        parser = null;
        isc = null;
        r = null;
        isr = null;
        is = null;
    }
    catch (Exception error) {
        if(noisey) {
        error.printStackTrace();
        DebugStream.println("Exception in Utility.parse() - Unexpected");
        }
        else {
        DebugStream.println("Exception in Utility.parse() - Expected");
        }
        DebugStream.printStackTrace(error);
    }
    return document;
    }
}