/*

 * reserved comment block

 * DO NOT REMOVE OR ALTER!

 */

/*

 * Licensed to the Apache Software Foundation (ASF) under one or more

 * contributor license agreements.  See the NOTICE file distributed with

 * this work for additional information regarding copyright ownership.

 * The ASF licenses this file to You 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 comments to preserve whitespaces and add a line break

//  when indenting. Reported by Gervase Markham <gerv@gerv.net>

// Sep 14, 2000:

//  Fixed serializer to report IO exception directly, instead at

//  the end of document processing.

//  Reported by Patrick Higgins <phiggins@transzap.com>

// Sep 13, 2000:

//   CR in character data will print as &#0D;

// Aug 25, 2000:

//   Fixed processing instruction printing inside element content

//   to not escape content. Reported by Mikael Staldal

//   <d96-mst@d.kth.se>

// Aug 25, 2000:

//   Added ability to omit comments.

//   Contributed by Anupam Bagchi <abagchi@jtcsv.com>

// Aug 26, 2000:

//   Fixed bug in newline handling when preserving spaces.

//   Contributed by Mike Dusseault <mdusseault@home.com>

// Aug 29, 2000:

//   Fixed state.unescaped not being set to false when

//   entering element state.

//   Reported by Lowell Vaughn <lvaughn@agillion.com>

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

import java.io.IOException;

import java.io.OutputStream;

import java.io.Writer;

import java.util.Hashtable;

import java.util.Vector;

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

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

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

import com.sun.org.apache.xerces.internal.util.XMLChar;

import org.w3c.dom.DOMError;

import org.w3c.dom.DOMErrorHandler;

import org.w3c.dom.Document;

import org.w3c.dom.DocumentFragment;

import org.w3c.dom.DocumentType;

import org.w3c.dom.Element;

import org.w3c.dom.Node;

import org.w3c.dom.ls.LSException;

import org.w3c.dom.ls.LSSerializerFilter;

import org.w3c.dom.traversal.NodeFilter;

import org.xml.sax.ContentHandler;

import org.xml.sax.DTDHandler;

import org.xml.sax.DocumentHandler;

import org.xml.sax.Locator;

import org.xml.sax.SAXException;

import org.xml.sax.ext.DeclHandler;

import org.xml.sax.ext.LexicalHandler;

/**

 * Base class for a serializer supporting both DOM and SAX pretty

 * serializing of XML/HTML/XHTML documents. Derives classes perform

 * the method-specific serializing, this class provides the common

 * serializing mechanisms.

 * <p>

 * The serializer must be initialized with the proper writer and

 * output format before it can be used by calling {@link #setOutputCharStream}

 * or {@link #setOutputByteStream} for the writer and {@link #setOutputFormat}

 * for the output format.

 * <p>

 * The serializer can be reused any number of times, but cannot

 * be used concurrently by two threads.

 * <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(Document)} and SAX serializing is done by firing

 * SAX events and using the serializer as a document handler.

 * This also applies to derived class.

 * <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>

 * When indenting, the serializer is capable of detecting seemingly

 * element content, and serializing these elements indented on separate

 * lines. An element is serialized indented when it is the first or

 * last child of an element, or immediate following or preceding

 * another element.

 *

 *

 * @author <a href="mailto:arkin@intalio.com">Assaf Arkin</a>

 * @author <a href="mailto:rahul.srivastava@sun.com">Rahul Srivastava</a>

 * @author Elena Litani, IBM

 * @author Sunitha Reddy, Sun Microsystems

 * @see Serializer

 * @see org.w3c.dom.ls.LSSerializer

 */

public abstract class BaseMarkupSerializer

    implements ContentHandler, DocumentHandler, LexicalHandler,

               DTDHandler, DeclHandler, DOMSerializer, Serializer

{

    // DOM L3 implementation

    protected short features = 0xFFFFFFFF;

    protected DOMErrorHandler fDOMErrorHandler;

    protected final DOMErrorImpl fDOMError = new DOMErrorImpl();

    protected LSSerializerFilter fDOMFilter;

    protected EncodingInfo _encodingInfo;

    /**

     * Holds array of all element states that have been entered.

     * The array is automatically resized. When leaving an element,

     * it's state is not removed but reused when later returning

     * to the same nesting level.

     */

    private ElementState[]  _elementStates;

    /**

     * The index of the next state to place in the array,

     * or one plus the index of the current state. When zero,

     * we are in no state.

     */

    private int             _elementStateCount;

    /**

     * Vector holding comments and PIs that come before the root

     * element (even after it), see {@link #serializePreRoot}.

     */

    private Vector          _preRoot;

    /**

     * If the document has been started (header serialized), this

     * flag is set to true so it's not started twice.

     */

    protected boolean       _started;

    /**

     * True if the serializer has been prepared. This flag is set

     * to false when the serializer is reset prior to using it,

     * and to true after it has been prepared for usage.

     */

    private boolean         _prepared;

    /**

     * Association between namespace URIs (keys) and prefixes (values).

     * Accumulated here prior to starting an element and placing this

     * list in the element state.

     */

    protected Hashtable     _prefixes;

    /**

     * The system identifier of the document type, if known.

     */

    protected String        _docTypePublicId;

    /**

     * The system identifier of the document type, if known.

     */

    protected String        _docTypeSystemId;

    /**

     * The output format associated with this serializer. This will never

     * be a null reference. If no format was passed to the constructor,

     * the default one for this document type will be used. The format

     * object is never changed by the serializer.

     */

    protected OutputFormat   _format;

    /**

     * The printer used for printing text parts.

     */

    protected Printer       _printer;

    /**

     * True if indenting printer.

     */

    protected boolean       _indenting;

    /** Temporary buffer to store character data */

    protected final StringBuffer fStrBuffer = new StringBuffer(40);

    /**

     * The underlying writer.

     */

    private Writer          _writer;

    /**

     * The output stream.

     */

    private OutputStream    _output;

    /** Current node that is being processed  */

    protected Node fCurrentNode = null;

    //--------------------------------//

    // Constructor and initialization //

    //--------------------------------//

    /**

     * Protected constructor can only be used by derived class.

     * Must initialize the serializer before serializing any document,

     * by calling {@link #setOutputCharStream} or {@link #setOutputByteStream}

                 * first

     */

    protected BaseMarkupSerializer( OutputFormat format )

    {

        int i;

        _elementStates = new ElementState[ 10 ];

        for ( i = 0 ; i < _elementStates.length ; ++i )

            _elementStates[ i ] = new ElementState();

        _format = format;

    }

    public DocumentHandler asDocumentHandler()

        throws IOException

    {

        prepare();

        return this;

    }

    public ContentHandler asContentHandler()

        throws IOException

    {

        prepare();

        return this;

    }

    public DOMSerializer asDOMSerializer()

        throws IOException

    {

        prepare();

        return this;

    }

    public void setOutputByteStream( OutputStream output )

    {

        if ( output == null ) {

            String msg = DOMMessageFormatter.formatMessage(DOMMessageFormatter.SERIALIZER_DOMAIN,

                                                           "ArgumentIsNull", new Object[]{"output"});

            throw new NullPointerException(msg);

        }

        _output = output;

        _writer = null;

        reset();

    }

    public void setOutputCharStream( Writer writer )

    {

        if ( writer == null ) {

            String msg = DOMMessageFormatter.formatMessage(DOMMessageFormatter.SERIALIZER_DOMAIN,

                                                           "ArgumentIsNull", new Object[]{"writer"});

            throw new NullPointerException(msg);

        }

        _writer = writer;

        _output = null;

        reset();

    }

    public void setOutputFormat( OutputFormat format )

    {

        if ( format == null ) {

            String msg = DOMMessageFormatter.formatMessage(DOMMessageFormatter.SERIALIZER_DOMAIN,

                                                           "ArgumentIsNull", new Object[]{"format"});

            throw new NullPointerException(msg);

        }

        _format = format;

        reset();

    }

    public boolean reset()

    {

        if ( _elementStateCount > 1 ) {

            String msg = DOMMessageFormatter.formatMessage(DOMMessageFormatter.SERIALIZER_DOMAIN,

                                                           "ResetInMiddle", null);

            throw new IllegalStateException(msg);

        }

        _prepared = false;

        fCurrentNode = null;

        fStrBuffer.setLength(0);

        return true;

    }

    protected void cleanup() {

        fCurrentNode = null;

    }

    protected void prepare()

        throws IOException

    {

        if ( _prepared )

            return;

        if ( _writer == null && _output == null ) {

            String msg = DOMMessageFormatter.formatMessage(DOMMessageFormatter.SERIALIZER_DOMAIN,

                                                           "NoWriterSupplied", null);

            throw new IOException(msg);

        }

        // If the output stream has been set, use it to construct

        // the writer. It is possible that the serializer has been

        // reused with the same output stream and different encoding.

        _encodingInfo = _format.getEncodingInfo();

        if ( _output != null ) {

            _writer = _encodingInfo.getWriter(_output);

        }

        if ( _format.getIndenting() ) {

            _indenting = true;

            _printer = new IndentPrinter( _writer, _format );

        } else {

            _indenting = false;

            _printer = new Printer( _writer, _format );

        }

        ElementState state;

        _elementStateCount = 0;

        state = _elementStates[ 0 ];

        state.namespaceURI = null;

        state.localName = null;

        state.rawName = null;

        state.preserveSpace = _format.getPreserveSpace();

        state.empty = true;

        state.afterElement = false;

        state.afterComment = false;

        state.doCData = state.inCData = false;

        state.prefixes = null;

        _docTypePublicId = _format.getDoctypePublic();

        _docTypeSystemId = _format.getDoctypeSystem();

        _started = false;

        _prepared = true;

    }

    //----------------------------------//

    // DOM document serializing methods //

    //----------------------------------//

    /**

     * Serializes the DOM element using the previously specified

     * writer and output format. Throws an exception only if

     * an I/O exception occured while serializing.

     *

     * @param elem The element to serialize

     * @throws IOException An I/O exception occured while

     *   serializing

     */

    public void serialize( Element elem )

        throws IOException

    {

        reset();

        prepare();

        serializeNode( elem );

        cleanup();

        _printer.flush();

        if ( _printer.getException() != null )

            throw _printer.getException();

    }

    /**

     * Serializes a node using the previously specified

     * writer and output format. Throws an exception only if

     * an I/O exception occured while serializing.

     *

     * @param node Node to serialize

     * @throws IOException An I/O exception occured while serializing

     */

    public void serialize( Node node ) throws IOException {

        reset();

        prepare();

        serializeNode( node );

        //Print any PIs and Comments which appeared in 'node'

        serializePreRoot();

        _printer.flush();

        if ( _printer.getException() != null )

            throw _printer.getException();

    }

    /**

     * Serializes the DOM document fragmnt using the previously specified

     * writer and output format. Throws an exception only if

     * an I/O exception occured while serializing.

     *

     * @param frag The document fragment to serialize

     * @throws IOException An I/O exception occured while

     *   serializing

     */

    public void serialize( DocumentFragment frag )

        throws IOException

    {

        reset();

        prepare();

        serializeNode( frag );

        cleanup();

        _printer.flush();

        if ( _printer.getException() != null )

            throw _printer.getException();

    }

    /**

     * Serializes the DOM document using the previously specified

     * writer and output format. Throws an exception only if

     * an I/O exception occured while serializing.

     *

     * @param doc The document to serialize

     * @throws IOException An I/O exception occured while

     *   serializing

     */

    public void serialize( Document doc )

        throws IOException

    {

        reset();

        prepare();

        serializeNode( doc );

        serializePreRoot();

        cleanup();

        _printer.flush();

        if ( _printer.getException() != null )

            throw _printer.getException();

    }

    //------------------------------------------//

    // SAX document handler serializing methods //

    //------------------------------------------//

    public void startDocument()

        throws SAXException

    {

        try {

            prepare();

        } catch ( IOException except ) {

            throw new SAXException( except.toString() );

        }

        // Nothing to do here. All the magic happens in startDocument(String)

    }

    public void characters( char[] chars, int start, int length )

        throws SAXException

    {

        ElementState state;

        try {

        state = content();

        // Check if text should be print as CDATA section or unescaped

        // based on elements listed in the output format (the element

        // state) or whether we are inside a CDATA section or entity.

        if ( state.inCData || state.doCData ) {

            int          saveIndent;

            // Print a CDATA section. The text is not escaped, but ']]>'

            // appearing in the code must be identified and dealt with.

            // The contents of a text node is considered space preserving.

            if ( ! state.inCData ) {

                _printer.printText( "<![CDATA[" );

                state.inCData = true;

            }

            saveIndent = _printer.getNextIndent();

            _printer.setNextIndent( 0 );