htmlserializer.java

来自「JAVA 所有包」· Java 代码 · 共 883 行 · 第 1/3 页

/*
 * Copyright 1999-2004 The Apache Software Foundation.
 * 
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 * 
 *      http://www.apache.org/licenses/LICENSE-2.0
 * 
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */


// Sep 14, 2000:
//  Fixed serializer to report IO exception directly, instead at
//  the end of document processing.
//  Reported by Patrick Higgins <[e-mail]>
// Aug 21, 2000:
//  Fixed bug in startDocument not calling prepare.
//  Reported by Mikael Staldal <[e-mail]>
// Aug 21, 2000:
//  Added ability to omit DOCTYPE declaration.
// Sep 1, 2000:
//   If no output format is provided the serializer now defaults
//   to ISO-8859-1 encoding. Reported by Mikael Staldal
//   <[e-mail]>


package com.sun.org.apache.xml.internal.serialize;

import com.sun.org.apache.xerces.internal.dom.DOMMessageFormatter;

import java.io.IOException;
import java.io.OutputStream;
import java.io.Writer;
import java.util.Enumeration;
import java.util.Locale;

import org.w3c.dom.Attr;
import org.w3c.dom.Element;
import org.w3c.dom.NamedNodeMap;
import org.w3c.dom.Node;
import org.xml.sax.AttributeList;
import org.xml.sax.Attributes;
import org.xml.sax.SAXException;


/**
 * Implements an HTML/XHTML serializer supporting both DOM and SAX
 * pretty serializing. HTML/XHTML mode is determined in the
 * constructor.  For usage instructions see {@link Serializer}.
 * <p>
 * If an output stream is used, the encoding is taken from the
 * output format (defaults to <tt>UTF-8</tt>). If a writer is
 * used, make sure the writer uses the same encoding (if applies)
 * as specified in the output format.
 * <p>
 * The serializer supports both DOM and SAX. DOM serializing is done
 * by calling {@link #serialize} and SAX serializing is done by firing
 * SAX events and using the serializer as a document handler.
 * <p>
 * If an I/O exception occurs while serializing, the serializer
 * will not throw an exception directly, but only throw it
 * at the end of serializing (either DOM or SAX's {@link
 * org.xml.sax.DocumentHandler#endDocument}.
 * <p>
 * For elements that are not specified as whitespace preserving,
 * the serializer will potentially break long text lines at space
 * boundaries, indent lines, and serialize elements on separate
 * lines. Line terminators will be regarded as spaces, and
 * spaces at beginning of line will be stripped.
 * <p>
 * XHTML is slightly different than HTML:
 * <ul>
 * <li>Element/attribute names are lower case and case matters
 * <li>Attributes must specify value, even if empty string
 * <li>Empty elements must have '/' in empty tag
 * <li>Contents of SCRIPT and STYLE elements serialized as CDATA
 * </ul>
 *
 * @deprecated This class was deprecated in Xerces 2.6.2. It is
 * recommended that new applications use JAXP's Transformation API 
 * for XML (TrAX) for serializing HTML. See the Xerces documentation
 * for more information.
 * @version $Revision: 1.2.6.1 $ $Date: 2005/09/09 07:26:14 $
 * @author <a href="mailto:[e-mail]">Assaf Arkin</a>
 * @see Serializer
 */
public class HTMLSerializer
    extends BaseMarkupSerializer
{


    /**
     * True if serializing in XHTML format.
     */
    private boolean _xhtml;


    public static final String XHTMLNamespace = "http://www.w3.org/1999/xhtml";

    // for users to override XHTMLNamespace if need be.
    private String fUserXHTMLNamespace = null;


    /**
     * Constructs a new HTML/XHTML serializer depending on the value of
     * <tt>xhtml</tt>. The serializer cannot be used without calling
     * {@link #setOutputCharStream} or {@link #setOutputByteStream} first.
     *
     * @param xhtml True if XHTML serializing
     */
    protected HTMLSerializer( boolean xhtml, OutputFormat format )
    {
        super( format );
        _xhtml = xhtml;
    }


    /**
     * Constructs a new serializer. The serializer cannot be used without
     * calling {@link #setOutputCharStream} or {@link #setOutputByteStream}
     * first.
     */
    public HTMLSerializer()
    {
        this( false, new OutputFormat( Method.HTML, "ISO-8859-1", false ) );
    }


    /**
     * Constructs a new serializer. The serializer cannot be used without
     * calling {@link #setOutputCharStream} or {@link #setOutputByteStream}
     * first.
     */
    public HTMLSerializer( OutputFormat format )
    {
        this( false, format != null ? format : new OutputFormat( Method.HTML, "ISO-8859-1", false ) );
    }



    /**
     * Constructs a new serializer that writes to the specified writer
     * using the specified output format. If <tt>format</tt> is null,
     * will use a default output format.
     *
     * @param writer The writer to use
     * @param format The output format to use, null for the default
     */
    public HTMLSerializer( Writer writer, OutputFormat format )
    {
        this( false, format != null ? format : new OutputFormat( Method.HTML, "ISO-8859-1", false ) );
        setOutputCharStream( writer );
    }


    /**
     * Constructs a new serializer that writes to the specified output
     * stream using the specified output format. If <tt>format</tt>
     * is null, will use a default output format.
     *
     * @param output The output stream to use
     * @param format The output format to use, null for the default
     */
    public HTMLSerializer( OutputStream output, OutputFormat format )
    {
        this( false, format != null ? format : new OutputFormat( Method.HTML, "ISO-8859-1", false ) );
        setOutputByteStream( output );
    }


    public void setOutputFormat( OutputFormat format )
    {
        super.setOutputFormat( format != null ? format : new OutputFormat( Method.HTML, "ISO-8859-1", false ) );
    }

    // Set  value for alternate XHTML namespace.
    public void setXHTMLNamespace(String newNamespace) {
        fUserXHTMLNamespace = newNamespace;
    } // setXHTMLNamespace(String)

    //-----------------------------------------//
    // SAX content handler serializing methods //
    //-----------------------------------------//


    public void startElement( String namespaceURI, String localName,
                              String rawName, Attributes attrs )
        throws SAXException
    {
        int          i;
        boolean      preserveSpace;
        ElementState state;
        String       name;
        String       value;
        String       htmlName;
        boolean      addNSAttr = false;

        try {
            if ( _printer == null )
            	throw new IllegalStateException( 
				    DOMMessageFormatter.formatMessage(
				    DOMMessageFormatter.SERIALIZER_DOMAIN,
                    "NoWriterSupplied", null));

            state = getElementState();
            if ( isDocumentState() ) {
                // If this is the root element handle it differently.
                // If the first root element in the document, serialize
                // the document's DOCTYPE. Space preserving defaults
                // to that of the output format.
                if ( ! _started )
                    startDocument( (localName == null || localName.length() == 0) 
                        ? rawName : localName );
            } else {
                // For any other element, if first in parent, then
                // close parent's opening tag and use the parnet's
                // space preserving.
                if ( state.empty )
                    _printer.printText( '>' );
                // Indent this element on a new line if the first
                // content of the parent element or immediately
                // following an element.
                if ( _indenting && ! state.preserveSpace &&
                     ( state.empty || state.afterElement ) )
                    _printer.breakLine();
            }
            preserveSpace = state.preserveSpace;

            // Do not change the current element state yet.
            // This only happens in endElement().
            
            // As per SAX2, the namespace URI is an empty string if the element has no
            // namespace URI, or namespaces is turned off. The check against null protects
            // against broken SAX implementations, so I've left it there. - mrglavas
            boolean hasNamespaceURI = (namespaceURI != null && namespaceURI.length() != 0);

            // SAX2: rawName (QName) could be empty string if 
            // namespace-prefixes property is false.
            if ( rawName == null || rawName.length() == 0) {
                rawName = localName;
                if ( hasNamespaceURI ) {
                    String prefix;
                    prefix = getPrefix( namespaceURI );
                    if ( prefix != null && prefix.length() != 0 )
                        rawName = prefix + ":" + localName;
                }
                addNSAttr = true;
            }
            if ( !hasNamespaceURI )
                htmlName = rawName;
            else {
                if ( namespaceURI.equals( XHTMLNamespace ) ||
                        (fUserXHTMLNamespace != null && fUserXHTMLNamespace.equals(namespaceURI)) )
                    htmlName = localName;
                else
                    htmlName = null;
            }

            // XHTML: element names are lower case, DOM will be different
            _printer.printText( '<' );
            if ( _xhtml )
                _printer.printText( rawName.toLowerCase(Locale.ENGLISH) );
            else
                _printer.printText( rawName );
            _printer.indent();

            // For each attribute serialize it's name and value as one part,
            // separated with a space so the element can be broken on
            // multiple lines.
            if ( attrs != null ) {
                for ( i = 0 ; i < attrs.getLength() ; ++i ) {
                    _printer.printSpace();
                    name = attrs.getQName( i ).toLowerCase(Locale.ENGLISH);
                    value = attrs.getValue( i );
                    if ( _xhtml || hasNamespaceURI ) {
                        // XHTML: print empty string for null values.
                        if ( value == null ) {
                            _printer.printText( name );
                            _printer.printText( "=\"\"" );
                        } else {
                            _printer.printText( name );
                            _printer.printText( "=\"" );
                            printEscaped( value );
                            _printer.printText( '"' );
                        }
                    } else {
                        // HTML: Empty values print as attribute name, no value.
                        // HTML: URI attributes will print unescaped
                        if ( value == null ) {