FileDocCategorySizeDatePackage
ToHTMLStream.javaAPI DocJava SE 6 API82214Tue Jun 10 00:23:08 BST 2008com.sun.org.apache.xml.internal.serializer

ToHTMLStream

public final class ToHTMLStream extends ToStream
This serializer takes a series of SAX or SAX-like events and writes its output to the given stream. This class is not a public API, it is public because it is used from another package.
xsl.usage
internal

Fields Summary
protected boolean
m_inDTD
This flag is set while receiving events from the DTD
private boolean
m_inBlockElem
True if the current element is a block element. (seems like this needs to be a stack. -sb).
private static final CharInfo
m_htmlcharInfo
Map that tells which XML characters should have special treatment, and it provides character to entity name lookup.
static final Trie
m_elementFlags
A digital search trie for fast, case insensitive lookup of ElemDesc objects.
private static final ElemDesc
m_dummy
Dummy element for elements not found.
private boolean
m_specialEscapeURLs
True if URLs should be specially escaped with the %xx form.
private boolean
m_omitMetaTag
True if the META tag should be omitted.
private Trie
m_htmlInfo
A Trie that is just a copy of the "static" one. We need this one to be able to use the faster, but not thread-safe method Trie.get2(name)
Constructors Summary
public ToHTMLStream()
Default constructor.


        super();
        m_charInfo = m_htmlcharInfo;
        // initialize namespaces
        m_prefixMap = new NamespaceMappings();
Methods Summary
public voidaddUniqueAttribute(java.lang.String name, java.lang.String value, int flags)
This method is used to add an attribute to the currently open element. The caller has guaranted that this attribute is unique, which means that it not been seen before and will not be seen again.

param
name the qualified name of the attribute
param
value the value of the attribute which can contain only ASCII printable characters characters in the range 32 to 127 inclusive.
param
flags the bit values of this integer give optimization information.

        try
        {
            final java.io.Writer writer = m_writer;
            if ((flags & NO_BAD_CHARS) > 0 && m_htmlcharInfo.onlyQuotAmpLtGt)
            {
                // "flags" has indicated that the characters
                // '>'  '<'   '&'  and '"' are not in the value and
                // m_htmlcharInfo has recorded that there are no other
                // entities in the range 0 to 127 so we write out the
                // value directly
                writer.write(' ");
                writer.write(name);
                writer.write("=\"");
                writer.write(value);
                writer.write('"");
            }
            else if (
                (flags & HTML_ATTREMPTY) > 0
                    && (value.length() == 0 || value.equalsIgnoreCase(name)))
            {
                writer.write(' ");
                writer.write(name);
            }
            else
            {
                writer.write(' ");
                writer.write(name);
                writer.write("=\"");
                if ((flags & HTML_ATTRURL) > 0)
                {
                    writeAttrURI(writer, value, m_specialEscapeURLs);
                }
                else
                {
                    writeAttrString(writer, value, this.getEncoding());
                }
                writer.write('"");
            }
        } catch (IOException e) {
            throw new SAXException(e);
        }
public voidattributeDecl(java.lang.String eName, java.lang.String aName, java.lang.String type, java.lang.String valueDefault, java.lang.String value)
This method does nothing.

        // The internal DTD subset is not serialized by the ToHTMLStream serializer
public final voidcdata(char[] ch, int start, int length)
Receive notification of cdata.

The Parser will call this method to report each chunk of character data. SAX parsers may return all contiguous character data in a single chunk, or they may split it into several chunks; however, all of the characters in any single event must come from the same external entity, so that the Locator provides useful information.

The application must not attempt to read from the array outside of the specified range.

Note that some parsers will report whitespace using the ignorableWhitespace() method rather than this one (validating parsers must do so).

param
ch The characters from the XML document.
param
start The start position in the array.
param
length The number of characters to read from the array.
throws
org.xml.sax.SAXException Any SAX exception, possibly wrapping another exception.
see
#ignorableWhitespace
see
org.xml.sax.Locator
throws
org.xml.sax.SAXException


        if ((null != m_elemContext.m_elementName)
            && (m_elemContext.m_elementName.equalsIgnoreCase("SCRIPT")
                || m_elemContext.m_elementName.equalsIgnoreCase("STYLE")))
        {
            try
            {
                if (m_elemContext.m_startTagOpen)
                {
                    closeStartTag();
                    m_elemContext.m_startTagOpen = false;
                }

                m_ispreserve = true;

                if (shouldIndent())
                    indent();

                // writer.write(ch, start, length);
                writeNormalizedChars(ch, start, length, true, m_lineSepUse);
            }
            catch (IOException ioe)
            {
                throw new org.xml.sax.SAXException(
                    Utils.messages.createMessage(
                        MsgKey.ER_OIERROR,
                        null),
                    ioe);
                //"IO error", ioe);
            }
        }
        else
        {
            super.cdata(ch, start, length);
        }
public final voidcharacters(char[] chars, int start, int length)
Receive notification of character data.

The Parser will call this method to report each chunk of character data. SAX parsers may return all contiguous character data in a single chunk, or they may split it into several chunks; however, all of the characters in any single event must come from the same external entity, so that the Locator provides useful information.

The application must not attempt to read from the array outside of the specified range.

Note that some parsers will report whitespace using the ignorableWhitespace() method rather than this one (validating parsers must do so).

param
chars The characters from the XML document.
param
start The start position in the array.
param
length The number of characters to read from the array.
throws
org.xml.sax.SAXException Any SAX exception, possibly wrapping another exception.
see
#ignorableWhitespace
see
org.xml.sax.Locator
throws
org.xml.sax.SAXException


        if (m_elemContext.m_isRaw)
        {
            try
            {
                if (m_elemContext.m_startTagOpen)
                {
                    closeStartTag();
                    m_elemContext.m_startTagOpen = false;
                }
                m_ispreserve = true;
                
//              With m_ispreserve just set true it looks like shouldIndent()
//              will always return false, so drop any possible indentation.
//              if (shouldIndent())
//                  indent();

                // writer.write("<![CDATA[");
                // writer.write(chars, start, length);
                writeNormalizedChars(chars, start, length, false, m_lineSepUse);

                // writer.write("]]>");
                
                // time to generate characters event
                if (m_tracer != null)
                    super.fireCharEvent(chars, start, length);
                
                return;
            }
            catch (IOException ioe)
            {
                throw new org.xml.sax.SAXException(
                    Utils.messages.createMessage(
                        MsgKey.ER_OIERROR,
                        null),
                    ioe);
                //"IO error", ioe);
            }
        }
        else
        {
            super.characters(chars, start, length);
        }
protected voidcloseStartTag()
For the enclosing elements starting tag write out out any attributes followed by ">"

throws
org.xml.sax.SAXException

            try
            {

            // finish processing attributes, time to fire off the start element event
            if (m_tracer != null)
                super.fireStartElem(m_elemContext.m_elementName);  
            
            int nAttrs = m_attributes.getLength();   
            if (nAttrs>0)
            {
                processAttributes(m_writer, nAttrs);
                // clear attributes object for re-use with next element
                m_attributes.clear();
            }

            m_writer.write('>");

            /* whether Xalan or XSLTC, we have the prefix mappings now, so
             * lets determine if the current element is specified in the cdata-
             * section-elements list.
             */
            if (m_cdataSectionElements != null) 
                m_elemContext.m_isCdataSection = isCdataSection();
            if (m_doIndent)
            {
                m_isprevtext = false;
                m_preserves.push(m_ispreserve);
            }
            
            }
            catch(IOException e)
            {
                throw new SAXException(e);
            }
public voidcomment(char[] ch, int start, int length)

        // The internal DTD subset is not serialized by the ToHTMLStream serializer
        if (m_inDTD)
            return;
        super.comment(ch, start, length);
public voidelementDecl(java.lang.String name, java.lang.String model)
This method does nothing.

        // The internal DTD subset is not serialized by the ToHTMLStream serializer
public voidendDTD()
Report the end of DTD declarations.

throws
org.xml.sax.SAXException The application may raise an exception.
see
#startDTD

        m_inDTD = false;
        /* for ToHTMLStream the DOCTYPE is entirely output in the
         * startDocumentInternal() method, so don't do anything here
         */
public final voidendDocument()
Receive notification of the end of a document.

throws
org.xml.sax.SAXException Any SAX exception, possibly wrapping another exception.
throws
org.xml.sax.SAXException

        
        flushPending();
        if (m_doIndent && !m_isprevtext)
        {
            try
            {
            outputLineSep();
            }
            catch(IOException e)
            {
                throw new SAXException(e);
            }
        }

        flushWriter();
        if (m_tracer != null)
            super.fireEndDoc();
public final voidendElement(java.lang.String namespaceURI, java.lang.String localName, java.lang.String name)
Receive notification of the end of an element.

param
namespaceURI
param
localName
param
name The element type name
throws
org.xml.sax.SAXException Any SAX exception, possibly wrapping another exception.

        // deal with any pending issues
        if (m_cdataTagOpen)
            closeCDATA();

        // if the element has a namespace, treat it like XML, not HTML
        if (null != namespaceURI && namespaceURI.length() > 0)
        {
            super.endElement(namespaceURI, localName, name);

            return;
        }

        try
        {

            ElemContext elemContext = m_elemContext;
            final ElemDesc elemDesc = elemContext.m_elementDesc;
            final int elemFlags = elemDesc.getFlags();
            final boolean elemEmpty = (elemFlags & ElemDesc.EMPTY) != 0;

            // deal with any indentation issues
            if (m_doIndent)
            {
                final boolean isBlockElement = (elemFlags&ElemDesc.BLOCK) != 0;
                boolean shouldIndent = false;

                if (m_ispreserve)
                {
                    m_ispreserve = false;
                }
                else if (m_doIndent && (!m_inBlockElem || isBlockElement))
                {
                    m_startNewLine = true;
                    shouldIndent = true;
                }
                if (!elemContext.m_startTagOpen && shouldIndent)
                    indent(elemContext.m_currentElemDepth - 1);
                m_inBlockElem = !isBlockElement;
            }

            final java.io.Writer writer = m_writer;
            if (!elemContext.m_startTagOpen)
            {
                writer.write("</");
                writer.write(name);
                writer.write('>");
            }
            else
            {
                // the start-tag open when this method was called,
                // so we need to process it now.
                
                if (m_tracer != null)
                    super.fireStartElem(name);

                // the starting tag was still open when we received this endElement() call
                // so we need to process any gathered attributes NOW, before they go away.
                int nAttrs = m_attributes.getLength();
                if (nAttrs > 0)
                {
                    processAttributes(m_writer, nAttrs);
                    // clear attributes object for re-use with next element
                    m_attributes.clear();
                }
                if (!elemEmpty)
                {
                    // As per Dave/Paul recommendation 12/06/2000
                    // if (shouldIndent)
                    // writer.write('>');
                    //  indent(m_currentIndent);

                    writer.write("></");
                    writer.write(name);
                    writer.write('>");
                }
                else
                {
                    writer.write('>");
                }
            }
            
            // clean up because the element has ended
            if ((elemFlags & ElemDesc.WHITESPACESENSITIVE) != 0)
                m_ispreserve = true;
            m_isprevtext = false;

            // fire off the end element event
            if (m_tracer != null)
                super.fireEndElem(name);            
                           
            // OPTIMIZE-EMPTY                
            if (elemEmpty)
            {
                // a quick exit if the HTML element had no children.
                // This block of code can be removed if the corresponding block of code
                // in startElement() also labeled with "OPTIMIZE-EMPTY" is also removed
                m_elemContext = elemContext.m_prev;
                return;
            }

            // some more clean because the element has ended. 
            if (!elemContext.m_startTagOpen)
            {
                if (m_doIndent && !m_preserves.isEmpty())
                    m_preserves.pop();
            }
            m_elemContext = elemContext.m_prev;
//            m_isRawStack.pop();
        }
        catch (IOException e)
        {
            throw new SAXException(e);
        }
public final voidendElement(java.lang.String elemName)

see
ExtendedContentHandler#endElement(String)

        endElement(null, null, elemName);
public final voidentityReference(java.lang.String name)
Receive notivication of a entityReference.

param
name non-null reference to entity name string.
throws
org.xml.sax.SAXException

        try
        {

        final java.io.Writer writer = m_writer;
        writer.write('&");
        writer.write(name);
        writer.write(';");
        
        } catch(IOException e)
        {
            throw new SAXException(e);
        }
public voidexternalEntityDecl(java.lang.String name, java.lang.String publicId, java.lang.String systemId)
This method does nothing.

        // The internal DTD subset is not serialized by the ToHTMLStream serializer
public static final com.sun.org.apache.xml.internal.serializer.ElemDescgetElemDesc(java.lang.String name)
Get a description of the given element.

param
name non-null name of element, case insensitive.
return
non-null reference to ElemDesc, which may be m_dummy if no element description matches the given name.

        /* this method used to return m_dummy  when name was null
         * but now it doesn't check and and requires non-null name.
         */
        Object obj = m_elementFlags.get(name);
        if (null != obj)
            return (ElemDesc)obj;
        return m_dummy;
private com.sun.org.apache.xml.internal.serializer.ElemDescgetElemDesc2(java.lang.String name)
Calls to this method could be replaced with calls to getElemDesc(name), but this one should be faster.

                          
       
    
        Object obj = m_htmlInfo.get2(name);
        if (null != obj)
            return (ElemDesc)obj;
        return m_dummy;
private final booleangetOmitMetaTag()
Tells if the formatter should omit the META tag.

return
True if the META tag should be omitted.

        return m_omitMetaTag;
private final booleangetSpecialEscapeURLs()
Tells if the formatter should use special URL escaping.

return
True if URLs should be specially escaped with the %xx form.

        return m_specialEscapeURLs;
protected synchronized voidinit(java.io.OutputStream output, java.util.Properties format)
Initialize the serializer with the specified output stream and output format. Must be called before calling any of the serialize methods.

param
output The output stream to use
param
format The output format
throws
UnsupportedEncodingException The encoding specified in the output format is not supported

        if (null == format)
        {
            format = OutputPropertiesFactory.getDefaultMethodProperties(Method.HTML);
         }
        super.init(output,format, false);
static voidinitTagReference(com.sun.org.apache.xml.internal.serializer.ToHTMLStream$Trie m_elementFlags)


     
        initTagReference(m_elementFlags);
    

        // HTML 4.0 loose DTD
        m_elementFlags.put("BASEFONT", new ElemDesc(0 | ElemDesc.EMPTY));
        m_elementFlags.put(
            "FRAME",
            new ElemDesc(0 | ElemDesc.EMPTY | ElemDesc.BLOCK));
        m_elementFlags.put("FRAMESET", new ElemDesc(0 | ElemDesc.BLOCK));
        m_elementFlags.put("NOFRAMES", new ElemDesc(0 | ElemDesc.BLOCK));
        m_elementFlags.put(
            "ISINDEX",
            new ElemDesc(0 | ElemDesc.EMPTY | ElemDesc.BLOCK));
        m_elementFlags.put(
            "APPLET",
            new ElemDesc(0 | ElemDesc.WHITESPACESENSITIVE));
        m_elementFlags.put("CENTER", new ElemDesc(0 | ElemDesc.BLOCK));
        m_elementFlags.put("DIR", new ElemDesc(0 | ElemDesc.BLOCK));
        m_elementFlags.put("MENU", new ElemDesc(0 | ElemDesc.BLOCK));

        // HTML 4.0 strict DTD
        m_elementFlags.put("TT", new ElemDesc(0 | ElemDesc.FONTSTYLE));
        m_elementFlags.put("I", new ElemDesc(0 | ElemDesc.FONTSTYLE));
        m_elementFlags.put("B", new ElemDesc(0 | ElemDesc.FONTSTYLE));
        m_elementFlags.put("BIG", new ElemDesc(0 | ElemDesc.FONTSTYLE));
        m_elementFlags.put("SMALL", new ElemDesc(0 | ElemDesc.FONTSTYLE));
        m_elementFlags.put("EM", new ElemDesc(0 | ElemDesc.PHRASE));
        m_elementFlags.put("STRONG", new ElemDesc(0 | ElemDesc.PHRASE));
        m_elementFlags.put("DFN", new ElemDesc(0 | ElemDesc.PHRASE));
        m_elementFlags.put("CODE", new ElemDesc(0 | ElemDesc.PHRASE));
        m_elementFlags.put("SAMP", new ElemDesc(0 | ElemDesc.PHRASE));
        m_elementFlags.put("KBD", new ElemDesc(0 | ElemDesc.PHRASE));
        m_elementFlags.put("VAR", new ElemDesc(0 | ElemDesc.PHRASE));
        m_elementFlags.put("CITE", new ElemDesc(0 | ElemDesc.PHRASE));
        m_elementFlags.put("ABBR", new ElemDesc(0 | ElemDesc.PHRASE));
        m_elementFlags.put("ACRONYM", new ElemDesc(0 | ElemDesc.PHRASE));
        m_elementFlags.put(
            "SUP",
            new ElemDesc(0 | ElemDesc.SPECIAL | ElemDesc.ASPECIAL));
        m_elementFlags.put(
            "SUB",
            new ElemDesc(0 | ElemDesc.SPECIAL | ElemDesc.ASPECIAL));
        m_elementFlags.put(
            "SPAN",
            new ElemDesc(0 | ElemDesc.SPECIAL | ElemDesc.ASPECIAL));
        m_elementFlags.put(
            "BDO",
            new ElemDesc(0 | ElemDesc.SPECIAL | ElemDesc.ASPECIAL));
        m_elementFlags.put(
            "BR",
            new ElemDesc(
                0
                    | ElemDesc.SPECIAL
                    | ElemDesc.ASPECIAL
                    | ElemDesc.EMPTY
                    | ElemDesc.BLOCK));
        m_elementFlags.put("BODY", new ElemDesc(0 | ElemDesc.BLOCK));
        m_elementFlags.put(
            "ADDRESS",
            new ElemDesc(
                0
                    | ElemDesc.BLOCK
                    | ElemDesc.BLOCKFORM
                    | ElemDesc.BLOCKFORMFIELDSET));
        m_elementFlags.put(
            "DIV",
            new ElemDesc(
                0
                    | ElemDesc.BLOCK
                    | ElemDesc.BLOCKFORM
                    | ElemDesc.BLOCKFORMFIELDSET));
        m_elementFlags.put("A", new ElemDesc(0 | ElemDesc.SPECIAL));
        m_elementFlags.put(
            "MAP",
            new ElemDesc(
                0 | ElemDesc.SPECIAL | ElemDesc.ASPECIAL | ElemDesc.BLOCK));
        m_elementFlags.put(
            "AREA",
            new ElemDesc(0 | ElemDesc.EMPTY | ElemDesc.BLOCK));
        m_elementFlags.put(
            "LINK",
            new ElemDesc(
                0 | ElemDesc.HEADMISC | ElemDesc.EMPTY | ElemDesc.BLOCK));
        m_elementFlags.put(
            "IMG",
            new ElemDesc(
                0
                    | ElemDesc.SPECIAL
                    | ElemDesc.ASPECIAL
                    | ElemDesc.EMPTY
                    | ElemDesc.WHITESPACESENSITIVE));
        m_elementFlags.put(
            "OBJECT",
            new ElemDesc(
                0
                    | ElemDesc.SPECIAL
                    | ElemDesc.ASPECIAL
                    | ElemDesc.HEADMISC
                    | ElemDesc.WHITESPACESENSITIVE));
        m_elementFlags.put("PARAM", new ElemDesc(0 | ElemDesc.EMPTY));
        m_elementFlags.put(
            "HR",
            new ElemDesc(
                0
                    | ElemDesc.BLOCK
                    | ElemDesc.BLOCKFORM
                    | ElemDesc.BLOCKFORMFIELDSET
                    | ElemDesc.EMPTY));
        m_elementFlags.put(
            "P",
            new ElemDesc(
                0
                    | ElemDesc.BLOCK
                    | ElemDesc.BLOCKFORM
                    | ElemDesc.BLOCKFORMFIELDSET));
        m_elementFlags.put(
            "H1",
            new ElemDesc(0 | ElemDesc.HEAD | ElemDesc.BLOCK));
        m_elementFlags.put(
            "H2",
            new ElemDesc(0 | ElemDesc.HEAD | ElemDesc.BLOCK));
        m_elementFlags.put(
            "H3",
            new ElemDesc(0 | ElemDesc.HEAD | ElemDesc.BLOCK));
        m_elementFlags.put(
            "H4",
            new ElemDesc(0 | ElemDesc.HEAD | ElemDesc.BLOCK));
        m_elementFlags.put(
            "H5",
            new ElemDesc(0 | ElemDesc.HEAD | ElemDesc.BLOCK));
        m_elementFlags.put(
            "H6",
            new ElemDesc(0 | ElemDesc.HEAD | ElemDesc.BLOCK));
        m_elementFlags.put(
            "PRE",
            new ElemDesc(0 | ElemDesc.PREFORMATTED | ElemDesc.BLOCK));
        m_elementFlags.put(
            "Q",
            new ElemDesc(0 | ElemDesc.SPECIAL | ElemDesc.ASPECIAL));
        m_elementFlags.put(
            "BLOCKQUOTE",
            new ElemDesc(
                0
                    | ElemDesc.BLOCK
                    | ElemDesc.BLOCKFORM
                    | ElemDesc.BLOCKFORMFIELDSET));
        m_elementFlags.put("INS", new ElemDesc(0));
        m_elementFlags.put("DEL", new ElemDesc(0));
        m_elementFlags.put(
            "DL",
            new ElemDesc(
                0
                    | ElemDesc.BLOCK
                    | ElemDesc.BLOCKFORM
                    | ElemDesc.BLOCKFORMFIELDSET));
        m_elementFlags.put("DT", new ElemDesc(0 | ElemDesc.BLOCK));
        m_elementFlags.put("DD", new ElemDesc(0 | ElemDesc.BLOCK));
        m_elementFlags.put(
            "OL",
            new ElemDesc(0 | ElemDesc.LIST | ElemDesc.BLOCK));
        m_elementFlags.put(
            "UL",
            new ElemDesc(0 | ElemDesc.LIST | ElemDesc.BLOCK));
        m_elementFlags.put("LI", new ElemDesc(0 | ElemDesc.BLOCK));
        m_elementFlags.put("FORM", new ElemDesc(0 | ElemDesc.BLOCK));
        m_elementFlags.put("LABEL", new ElemDesc(0 | ElemDesc.FORMCTRL));
        m_elementFlags.put(
            "INPUT",
            new ElemDesc(
                0 | ElemDesc.FORMCTRL | ElemDesc.INLINELABEL | ElemDesc.EMPTY));
        m_elementFlags.put(
            "SELECT",
            new ElemDesc(0 | ElemDesc.FORMCTRL | ElemDesc.INLINELABEL));
        m_elementFlags.put("OPTGROUP", new ElemDesc(0));
        m_elementFlags.put("OPTION", new ElemDesc(0));
        m_elementFlags.put(
            "TEXTAREA",
            new ElemDesc(0 | ElemDesc.FORMCTRL | ElemDesc.INLINELABEL));
        m_elementFlags.put(
            "FIELDSET",
            new ElemDesc(0 | ElemDesc.BLOCK | ElemDesc.BLOCKFORM));
        m_elementFlags.put("LEGEND", new ElemDesc(0));
        m_elementFlags.put(
            "BUTTON",
            new ElemDesc(0 | ElemDesc.FORMCTRL | ElemDesc.INLINELABEL));
        m_elementFlags.put(
            "TABLE",
            new ElemDesc(
                0
                    | ElemDesc.BLOCK
                    | ElemDesc.BLOCKFORM
                    | ElemDesc.BLOCKFORMFIELDSET));
        m_elementFlags.put("CAPTION", new ElemDesc(0 | ElemDesc.BLOCK));
        m_elementFlags.put("THEAD", new ElemDesc(0 | ElemDesc.BLOCK));
        m_elementFlags.put("TFOOT", new ElemDesc(0 | ElemDesc.BLOCK));
        m_elementFlags.put("TBODY", new ElemDesc(0 | ElemDesc.BLOCK));
        m_elementFlags.put("COLGROUP", new ElemDesc(0 | ElemDesc.BLOCK));
        m_elementFlags.put(
            "COL",
            new ElemDesc(0 | ElemDesc.EMPTY | ElemDesc.BLOCK));
        m_elementFlags.put("TR", new ElemDesc(0 | ElemDesc.BLOCK));
        m_elementFlags.put("TH", new ElemDesc(0));
        m_elementFlags.put("TD", new ElemDesc(0));
        m_elementFlags.put(
            "HEAD",
            new ElemDesc(0 | ElemDesc.BLOCK | ElemDesc.HEADELEM));
        m_elementFlags.put("TITLE", new ElemDesc(0 | ElemDesc.BLOCK));
        m_elementFlags.put(
            "BASE",
            new ElemDesc(0 | ElemDesc.EMPTY | ElemDesc.BLOCK));
        m_elementFlags.put(
            "META",
            new ElemDesc(
                0 | ElemDesc.HEADMISC | ElemDesc.EMPTY | ElemDesc.BLOCK));
        m_elementFlags.put(
            "STYLE",
            new ElemDesc(
                0 | ElemDesc.HEADMISC | ElemDesc.RAW | ElemDesc.BLOCK));
        m_elementFlags.put(
            "SCRIPT",
            new ElemDesc(
                0
                    | ElemDesc.SPECIAL
                    | ElemDesc.ASPECIAL
                    | ElemDesc.HEADMISC
                    | ElemDesc.RAW));
        m_elementFlags.put(
            "NOSCRIPT",
            new ElemDesc(
                0
                    | ElemDesc.BLOCK
                    | ElemDesc.BLOCKFORM
                    | ElemDesc.BLOCKFORMFIELDSET));
        m_elementFlags.put("HTML", new ElemDesc(0 | ElemDesc.BLOCK));

        // From "John Ky" <hand@syd.speednet.com.au
        // Transitional Document Type Definition ()
        // file:///C:/Documents%20and%20Settings/sboag.BOAG600E/My%20Documents/html/sgml/loosedtd.html#basefont
        m_elementFlags.put("FONT", new ElemDesc(0 | ElemDesc.FONTSTYLE));

        // file:///C:/Documents%20and%20Settings/sboag.BOAG600E/My%20Documents/html/present/graphics.html#edef-STRIKE
        m_elementFlags.put("S", new ElemDesc(0 | ElemDesc.FONTSTYLE));
        m_elementFlags.put("STRIKE", new ElemDesc(0 | ElemDesc.FONTSTYLE));

        // file:///C:/Documents%20and%20Settings/sboag.BOAG600E/My%20Documents/html/present/graphics.html#edef-U
        m_elementFlags.put("U", new ElemDesc(0 | ElemDesc.FONTSTYLE));

        // From "John Ky" <hand@syd.speednet.com.au
        m_elementFlags.put("NOBR", new ElemDesc(0 | ElemDesc.FONTSTYLE));

        // HTML 4.0, section 16.5
        m_elementFlags.put(
            "IFRAME",
            new ElemDesc(
                0
                    | ElemDesc.BLOCK
                    | ElemDesc.BLOCKFORM
                    | ElemDesc.BLOCKFORMFIELDSET));
                    
        // Netscape 4 extension
        m_elementFlags.put(
            "LAYER",
            new ElemDesc(
                0
                    | ElemDesc.BLOCK
                    | ElemDesc.BLOCKFORM
                    | ElemDesc.BLOCKFORMFIELDSET));
        // Netscape 4 extension                    
        m_elementFlags.put(
            "ILAYER",
            new ElemDesc(
                0
                    | ElemDesc.BLOCK
                    | ElemDesc.BLOCKFORM
                    | ElemDesc.BLOCKFORMFIELDSET));


        // NOW FOR ATTRIBUTE INFORMATION . . .
        ElemDesc elemDesc;


        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("A");
        elemDesc.setAttr("HREF", ElemDesc.ATTRURL);
        elemDesc.setAttr("NAME", ElemDesc.ATTRURL);
        
        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("AREA");
        elemDesc.setAttr("HREF", ElemDesc.ATTRURL);
        elemDesc.setAttr("NOHREF", ElemDesc.ATTREMPTY);

        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("BASE");
        elemDesc.setAttr("HREF", ElemDesc.ATTRURL);

        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("BUTTON");
        elemDesc.setAttr("DISABLED", ElemDesc.ATTREMPTY);

        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("BLOCKQUOTE");
        elemDesc.setAttr("CITE", ElemDesc.ATTRURL);

        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("DEL");
        elemDesc.setAttr("CITE", ElemDesc.ATTRURL);
     
        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("DIR");
        elemDesc.setAttr("COMPACT", ElemDesc.ATTREMPTY);
           
        // ----------------------------------------------
        
        elemDesc = (ElemDesc) m_elementFlags.get("DIV");
        elemDesc.setAttr("SRC", ElemDesc.ATTRURL); // Netscape 4 extension
        elemDesc.setAttr("NOWRAP", ElemDesc.ATTREMPTY); // Internet-Explorer extension
   
        // ----------------------------------------------        
        elemDesc = (ElemDesc) m_elementFlags.get("DL");
        elemDesc.setAttr("COMPACT", ElemDesc.ATTREMPTY);
           
        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("FORM");
        elemDesc.setAttr("ACTION", ElemDesc.ATTRURL);

        // ----------------------------------------------
        // Attribution to: "Voytenko, Dimitry" <DVoytenko@SECTORBASE.COM>
        elemDesc = (ElemDesc) m_elementFlags.get("FRAME");
        elemDesc.setAttr("SRC", ElemDesc.ATTRURL);
        elemDesc.setAttr("LONGDESC", ElemDesc.ATTRURL);
        elemDesc.setAttr("NORESIZE",ElemDesc.ATTREMPTY);

        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("HEAD");
        elemDesc.setAttr("PROFILE", ElemDesc.ATTRURL);

        // ----------------------------------------------        
        elemDesc = (ElemDesc) m_elementFlags.get("HR");
        elemDesc.setAttr("NOSHADE", ElemDesc.ATTREMPTY);
        
        // ----------------------------------------------
        // HTML 4.0, section 16.5
        elemDesc = (ElemDesc) m_elementFlags.get("IFRAME");
        elemDesc.setAttr("SRC", ElemDesc.ATTRURL);
        elemDesc.setAttr("LONGDESC", ElemDesc.ATTRURL);

        // ----------------------------------------------
        // Netscape 4 extension
        elemDesc = (ElemDesc) m_elementFlags.get("ILAYER");
        elemDesc.setAttr("SRC", ElemDesc.ATTRURL);

        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("IMG");
        elemDesc.setAttr("SRC", ElemDesc.ATTRURL);
        elemDesc.setAttr("LONGDESC", ElemDesc.ATTRURL);
        elemDesc.setAttr("USEMAP", ElemDesc.ATTRURL);
        elemDesc.setAttr("ISMAP", ElemDesc.ATTREMPTY);

        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("INPUT");
        elemDesc.setAttr("SRC", ElemDesc.ATTRURL);
        elemDesc.setAttr("USEMAP", ElemDesc.ATTRURL);
        elemDesc.setAttr("CHECKED", ElemDesc.ATTREMPTY);
        elemDesc.setAttr("DISABLED", ElemDesc.ATTREMPTY);
        elemDesc.setAttr("ISMAP", ElemDesc.ATTREMPTY);
        elemDesc.setAttr("READONLY", ElemDesc.ATTREMPTY);
        
        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("INS");
        elemDesc.setAttr("CITE", ElemDesc.ATTRURL);

        // ----------------------------------------------
        // Netscape 4 extension
        elemDesc = (ElemDesc) m_elementFlags.get("LAYER");
        elemDesc.setAttr("SRC", ElemDesc.ATTRURL);

        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("LINK");
        elemDesc.setAttr("HREF", ElemDesc.ATTRURL);
       
        // ----------------------------------------------       
        elemDesc = (ElemDesc) m_elementFlags.get("MENU");
        elemDesc.setAttr("COMPACT", ElemDesc.ATTREMPTY);
        
        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("OBJECT");
        elemDesc.setAttr("CLASSID", ElemDesc.ATTRURL);
        elemDesc.setAttr("CODEBASE", ElemDesc.ATTRURL);
        elemDesc.setAttr("DATA", ElemDesc.ATTRURL);
        elemDesc.setAttr("ARCHIVE", ElemDesc.ATTRURL);
        elemDesc.setAttr("USEMAP", ElemDesc.ATTRURL);
        elemDesc.setAttr("DECLARE", ElemDesc.ATTREMPTY);
        
        // ----------------------------------------------        
        elemDesc = (ElemDesc) m_elementFlags.get("OL");
        elemDesc.setAttr("COMPACT", ElemDesc.ATTREMPTY);
        
        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("OPTGROUP");
        elemDesc.setAttr("DISABLED", ElemDesc.ATTREMPTY);

        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("OPTION");
        elemDesc.setAttr("SELECTED", ElemDesc.ATTREMPTY);
        elemDesc.setAttr("DISABLED", ElemDesc.ATTREMPTY);
        
        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("Q");
        elemDesc.setAttr("CITE", ElemDesc.ATTRURL);

        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("SCRIPT");
        elemDesc.setAttr("SRC", ElemDesc.ATTRURL);
        elemDesc.setAttr("FOR", ElemDesc.ATTRURL);
        elemDesc.setAttr("DEFER", ElemDesc.ATTREMPTY);

        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("SELECT");
        elemDesc.setAttr("DISABLED", ElemDesc.ATTREMPTY);
        elemDesc.setAttr("MULTIPLE", ElemDesc.ATTREMPTY);

        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("TABLE");
        elemDesc.setAttr("NOWRAP", ElemDesc.ATTREMPTY); // Internet-Explorer extension
        
        // ----------------------------------------------        
        elemDesc = (ElemDesc) m_elementFlags.get("TD");
        elemDesc.setAttr("NOWRAP", ElemDesc.ATTREMPTY);

        // ----------------------------------------------
        elemDesc = (ElemDesc) m_elementFlags.get("TEXTAREA");
        elemDesc.setAttr("DISABLED", ElemDesc.ATTREMPTY);
        elemDesc.setAttr("READONLY", ElemDesc.ATTREMPTY);
       
        // ----------------------------------------------                
        elemDesc = (ElemDesc) m_elementFlags.get("TH");
        elemDesc.setAttr("NOWRAP", ElemDesc.ATTREMPTY);
        
        // ----------------------------------------------
        // The nowrap attribute of a tr element is both
        // a Netscape and Internet-Explorer extension                
        elemDesc = (ElemDesc) m_elementFlags.get("TR");
        elemDesc.setAttr("NOWRAP", ElemDesc.ATTREMPTY);
        
        // ----------------------------------------------        
        elemDesc = (ElemDesc) m_elementFlags.get("UL");
        elemDesc.setAttr("COMPACT", ElemDesc.ATTREMPTY);
private voidinitToHTMLStream()

//        m_elementDesc = null;
        m_inBlockElem = false;
        m_inDTD = false;
//        m_isRawStack.clear();
        m_omitMetaTag = false;
        m_specialEscapeURLs = true;
public voidinternalEntityDecl(java.lang.String name, java.lang.String value)
This method does nothing.

        // The internal DTD subset is not serialized by the ToHTMLStream serializer
private booleanisASCIIDigit(char c)
Tell if a character is an ASCII digit.

        return (c >= '0" && c <= '9");
private booleanisHHSign(java.lang.String str)
Dmitri Ilyin: Makes sure if the String is HH encoded sign.

param
str must be 2 characters long
return
true or false

        boolean sign = true;
        try
        {
            char r = (char) Integer.parseInt(str, 16);
        }
        catch (NumberFormatException e)
        {
            sign = false;
        }
        return sign;
private static java.lang.StringmakeHHString(int i)
Make an integer into an HH hex value. Does no checking on the size of the input, since this is only meant to be used locally by writeAttrURI.

param
i must be a value less than 255.
return
should be a two character string.

        String s = Integer.toHexString(i).toUpperCase();
        if (s.length() == 1)
        {
            s = "0" + s;
        }
        return s;
public voidnamespaceAfterStartElement(java.lang.String prefix, java.lang.String uri)
This method is used when a prefix/uri namespace mapping is indicated after the element was started with a startElement() and before and endElement(). startPrefixMapping(prefix,uri) would be used before the startElement() call.

param
uri the URI of the namespace
param
prefix the prefix associated with the given URI.
see
ExtendedContentHandler#namespaceAfterStartElement(String, String)

            // hack for XSLTC with finding URI for default namespace
            if (m_elemContext.m_elementURI == null)
            {
                String prefix1 = getPrefixPart(m_elemContext.m_elementName);
                if (prefix1 == null && EMPTYSTRING.equals(prefix))
                {
                    // the elements URI is not known yet, and it
                    // doesn't have a prefix, and we are currently
                    // setting the uri for prefix "", so we have
                    // the uri for the element... lets remember it
                    m_elemContext.m_elementURI = uri;
                }
            }            
            startPrefixMapping(prefix,uri,false);
protected voidprocessAttribute(java.io.Writer writer, java.lang.String name, java.lang.String value, com.sun.org.apache.xml.internal.serializer.ElemDesc elemDesc)
Process an attribute.

param
writer The writer to write the processed output to.
param
name The name of the attribute.
param
value The value of the attribute.
param
elemDesc The description of the HTML element that has this attribute.
throws
org.xml.sax.SAXException

        writer.write(' ");

        if (   ((value.length() == 0) || value.equalsIgnoreCase(name))
            && elemDesc != null 
            && elemDesc.isAttrFlagSet(name, ElemDesc.ATTREMPTY))
        {
            writer.write(name);
        }
        else
        {
            // %REVIEW% %OPT%
            // Two calls to single-char write may NOT
            // be more efficient than one to string-write...
            writer.write(name);
            writer.write("=\"");
            if (   elemDesc != null
                && elemDesc.isAttrFlagSet(name, ElemDesc.ATTRURL))
                writeAttrURI(writer, value, m_specialEscapeURLs);
            else
                writeAttrString(writer, value, this.getEncoding());
            writer.write('"");

        }
public voidprocessAttributes(java.io.Writer writer, int nAttrs)
Process the attributes, which means to write out the currently collected attributes to the writer. The attributes are not cleared by this method

param
writer the writer to write processed attributes to.
param
nAttrs the number of attributes in m_attributes to be processed
throws
org.xml.sax.SAXException

            /* 
             * process the collected attributes
             */
            for (int i = 0; i < nAttrs; i++)
            {
                processAttribute(
                    writer,
                    m_attributes.getQName(i),
                    m_attributes.getValue(i),
                    m_elemContext.m_elementDesc);
            }
public voidprocessingInstruction(java.lang.String target, java.lang.String data)
Receive notification of a processing instruction.

param
target The processing instruction target.
param
data The processing instruction data, or null if none was supplied.
throws
org.xml.sax.SAXException Any SAX exception, possibly wrapping another exception.
throws
org.xml.sax.SAXException


        // Process any pending starDocument and startElement first.
        flushPending(); 
        
        // Use a fairly nasty hack to tell if the next node is supposed to be 
        // unescaped text.
        if (target.equals(Result.PI_DISABLE_OUTPUT_ESCAPING))
        {
            startNonEscaping();
        }
        else if (target.equals(Result.PI_ENABLE_OUTPUT_ESCAPING))
        {
            endNonEscaping();
        }
        else
        {
            try
            {
            if (m_elemContext.m_startTagOpen)
            {
                closeStartTag();
                m_elemContext.m_startTagOpen = false;
            }
            else if (m_needToCallStartDocument)
                startDocumentInternal();

            if (shouldIndent())
                indent();

            final java.io.Writer writer = m_writer;
            //writer.write("<?" + target);
            writer.write("<?");
            writer.write(target);

            if (data.length() > 0 && !Character.isSpaceChar(data.charAt(0)))
                writer.write(' "); 

            //writer.write(data + ">"); // different from XML
            writer.write(data); // different from XML
            writer.write('>"); // different from XML

            // Always output a newline char if not inside of an 
            // element. The whitespace is not significant in that
            // case.
            if (m_elemContext.m_currentElemDepth <= 0)
                outputLineSep();

            m_startNewLine = true;
            }
            catch(IOException e)
            {
                throw new SAXException(e);
            }
        }
               
        // now generate the PI event
        if (m_tracer != null)
            super.fireEscapingEvent(target, data);
public booleanreset()

        boolean ret = super.reset();
        if (!ret)
            return false;
        initToHTMLStream();
        return true;
public voidsetOmitMetaTag(boolean bool)
Tells if the formatter should omit the META tag.

param
bool True if the META tag should be omitted.

        m_omitMetaTag = bool;
public voidsetOutputFormat(java.util.Properties format)
Specifies an output format for this serializer. It the serializer has already been associated with an output format, it will switch to the new format. This method should not be called while the serializer is in the process of serializing a document. This method can be called multiple times before starting the serialization of a particular result-tree. In principle all serialization parameters can be changed, with the exception of method="html" (it must be method="html" otherwise we shouldn't even have a ToHTMLStream object here!)

param
format The output format or serialzation parameters to use.

 
        m_specialEscapeURLs =
            OutputPropertyUtils.getBooleanProperty(
                OutputPropertiesFactory.S_USE_URL_ESCAPING,
                format);

        m_omitMetaTag =
            OutputPropertyUtils.getBooleanProperty(
                OutputPropertiesFactory.S_OMIT_META_TAG,
                format);

        super.setOutputFormat(format);
public voidsetOutputStream(java.io.OutputStream output)
Specifies an output stream to which the document should be serialized. This method should not be called while the serializer is in the process of serializing a document.

The encoding specified in the output properties is used, or if no encoding was specified, the default for the selected output method.

param
output The output stream


            try
            {
                Properties format;
                if (null == m_format)
                    format = OutputPropertiesFactory.getDefaultMethodProperties(Method.HTML);
                else
                    format = m_format;
                init(output, format, true);
            }
            catch (UnsupportedEncodingException uee)
            {

                // Should have been warned in init, I guess...
            }
public voidsetSpecialEscapeURLs(boolean bool)
Tells if the formatter should use special URL escaping.

param
bool True if URLs should be specially escaped with the %xx form.


                               
       
    
        m_specialEscapeURLs = bool;
public voidstartDTD(java.lang.String name, java.lang.String publicId, java.lang.String systemId)

        m_inDTD = true;
        super.startDTD(name, publicId, systemId);
protected voidstartDocumentInternal()
Receive notification of the beginning of a document.

throws
org.xml.sax.SAXException Any SAX exception, possibly wrapping another exception.
throws
org.xml.sax.SAXException

        super.startDocumentInternal();

        m_needToCallStartDocument = false; 
        m_needToOutputDocTypeDecl = true;
        m_startNewLine = false;
        setOmitXMLDeclaration(true);

        if (true == m_needToOutputDocTypeDecl)
        {
            String doctypeSystem = getDoctypeSystem();
            String doctypePublic = getDoctypePublic();
            if ((null != doctypeSystem) || (null != doctypePublic))
            {
                final java.io.Writer writer = m_writer;
                try
                {
                writer.write("<!DOCTYPE html");

                if (null != doctypePublic)
                {
                    writer.write(" PUBLIC \"");
                    writer.write(doctypePublic);
                    writer.write('"");
                }

                if (null != doctypeSystem)
                {
                    if (null == doctypePublic)
                        writer.write(" SYSTEM \"");
                    else
                        writer.write(" \"");

                    writer.write(doctypeSystem);
                    writer.write('"");
                }

                writer.write('>");
                outputLineSep();
                }
                catch(IOException e)
                {
                    throw new SAXException(e);
                }
            }
        }

        m_needToOutputDocTypeDecl = false;
public voidstartElement(java.lang.String namespaceURI, java.lang.String localName, java.lang.String name, org.xml.sax.Attributes atts)
Receive notification of the beginning of an element.

param
namespaceURI
param
localName
param
name The element type name.
param
atts The attributes attached to the element, if any.
throws
org.xml.sax.SAXException Any SAX exception, possibly wrapping another exception.
see
#endElement
see
org.xml.sax.AttributeList


        ElemContext elemContext = m_elemContext;

        // clean up any pending things first
        if (elemContext.m_startTagOpen)
        {
            closeStartTag();
            elemContext.m_startTagOpen = false;
        }
        else if (m_cdataTagOpen)
        {
            closeCDATA();
            m_cdataTagOpen = false;
        }
        else if (m_needToCallStartDocument)
        {
            startDocumentInternal();
            m_needToCallStartDocument = false;
        }


        // if this element has a namespace then treat it like XML
        if (null != namespaceURI && namespaceURI.length() > 0)
        {
            super.startElement(namespaceURI, localName, name, atts);

            return;
        }
        
        try
        {
            // getElemDesc2(name) is faster than getElemDesc(name)
            ElemDesc elemDesc = getElemDesc2(name);
            int elemFlags = elemDesc.getFlags();

            // deal with indentation issues first
            if (m_doIndent)
            {

                boolean isBlockElement = (elemFlags & ElemDesc.BLOCK) != 0;
                if (m_ispreserve)
                    m_ispreserve = false;
                else if (
                    (null != elemContext.m_elementName)
                    && (!m_inBlockElem
                        || isBlockElement) /* && !isWhiteSpaceSensitive */
                    )
                {
                    m_startNewLine = true;

                    indent();

                }
                m_inBlockElem = !isBlockElement;
            }

            // save any attributes for later processing
            if (atts != null)
                addAttributes(atts);            

            m_isprevtext = false;
            final java.io.Writer writer = m_writer;
            writer.write('<");
            writer.write(name);



            if (m_tracer != null)
                firePseudoAttributes();
            
            if ((elemFlags & ElemDesc.EMPTY) != 0)  
            {
                // an optimization for elements which are expected
                // to be empty.
                m_elemContext = elemContext.push();
                /* XSLTC sometimes calls namespaceAfterStartElement()
                 * so we need to remember the name
                 */
                m_elemContext.m_elementName = name;
                m_elemContext.m_elementDesc = elemDesc;
                return;                
            } 
            else
            {
                elemContext = elemContext.push(namespaceURI,localName,name);
                m_elemContext = elemContext;
                elemContext.m_elementDesc = elemDesc;
                elemContext.m_isRaw = (elemFlags & ElemDesc.RAW) != 0;
            }
            

            if ((elemFlags & ElemDesc.HEADELEM) != 0)
            {
                // This is the <HEAD> element, do some special processing
                closeStartTag();
                elemContext.m_startTagOpen = false;
                if (!m_omitMetaTag)
                {
                    if (m_doIndent)
                        indent();
                    writer.write(
                        "<META http-equiv=\"Content-Type\" content=\"text/html; charset=");
                    String encoding = getEncoding();
                    String encode = Encodings.getMimeEncoding(encoding);
                    writer.write(encode);
                    writer.write("\">");
                }
            }
        }
        catch (IOException e)
        {
            throw new SAXException(e);
        }
public voidwriteAttrString(java.io.Writer writer, java.lang.String string, java.lang.String encoding)
Writes the specified string after substituting specials, and UTF-16 surrogates for character references &#xnn.

param
string String to convert to XML format.
param
encoding CURRENTLY NOT IMPLEMENTED.
throws
org.xml.sax.SAXException

        final int end = string.length();
        if (end > m_attrBuff.length)
        {
            m_attrBuff = new char[end * 2 + 1];
        }
        string.getChars(0, end, m_attrBuff, 0);
        final char[] chars = m_attrBuff;

        

        int cleanStart = 0;
        int cleanLength = 0;

        char ch = 0;
        for (int i = 0; i < end; i++)
        {
            ch = chars[i];

            // System.out.println("SPECIALSSIZE: "+SPECIALSSIZE);
            // System.out.println("ch: "+(int)ch);
            // System.out.println("m_maxCharacter: "+(int)m_maxCharacter);
            // System.out.println("m_attrCharsMap[ch]: "+(int)m_attrCharsMap[ch]);
            if (escapingNotNeeded(ch) && (!m_charInfo.isSpecialAttrChar(ch)))
            {
                cleanLength++;
            }
            else if ('<" == ch || '>" == ch)
            {
                cleanLength++; // no escaping in this case, as specified in 15.2
            }
            else if (
                ('&" == ch) && ((i + 1) < end) && ('{" == chars[i + 1]))
            {
                cleanLength++; // no escaping in this case, as specified in 15.2
            }
            else
            {
                if (cleanLength > 0)
                {
                    writer.write(chars,cleanStart,cleanLength);
                    cleanLength = 0;
                }
                int pos = accumDefaultEntity(writer, ch, i, chars, end, false, true);

                if (i != pos)
                {
                    i = pos - 1;
                }
                else
                {
                    if (Encodings.isHighUTF16Surrogate(ch))
                    {
 
                            writeUTF16Surrogate(ch, chars, i, end);
                            i++; // two input characters processed
                                 // this increments by one and the for()
                                 // loop itself increments by another one.
                    }

                    // The next is kind of a hack to keep from escaping in the case 
                    // of Shift_JIS and the like.

                    /*
                    else if ((ch < m_maxCharacter) && (m_maxCharacter == 0xFFFF)
                    && (ch != 160))
                    {
                    writer.write(ch);  // no escaping in this case
                    }
                    else
                    */
                    String outputStringForChar = m_charInfo.getOutputStringForChar(ch);
                    if (null != outputStringForChar)
                    {
                        writer.write(outputStringForChar);
                    }
                    else if (escapingNotNeeded(ch))
                    {
                        writer.write(ch); // no escaping in this case
                    }
                    else
                    {
                        writer.write("&#");
                        writer.write(Integer.toString(ch));
                        writer.write(';");
                    }
                }
                cleanStart = i + 1;
            }
        } // end of for()
        
        // are there any clean characters at the end of the array
        // that we haven't processed yet?
        if (cleanLength > 1)
        {
            // if the whole string can be written out as-is do so
            // otherwise write out the clean chars at the end of the
            // array
            if (cleanStart == 0)
                writer.write(string);
            else
                writer.write(chars, cleanStart, cleanLength);
        }
        else if (cleanLength == 1)
        {
            // a little optimization for 1 clean character
            // (we could have let the previous if(...) handle them all)
            writer.write(ch);
        }
public voidwriteAttrURI(java.io.Writer writer, java.lang.String string, boolean doURLEscaping)
Write the specified string after substituting non ASCII characters, with %HH, where HH is the hex of the byte value.

param
string String to convert to XML format.
param
doURLEscaping True if we should try to encode as per http://www.ietf.org/rfc/rfc2396.txt.
throws
org.xml.sax.SAXException if a bad surrogate pair is detected.

        // http://www.ietf.org/rfc/rfc2396.txt says:
        // A URI is always in an "escaped" form, since escaping or unescaping a
        // completed URI might change its semantics.  Normally, the only time
        // escape encodings can safely be made is when the URI is being created
        // from its component parts; each component may have its own set of
        // characters that are reserved, so only the mechanism responsible for
        // generating or interpreting that component can determine whether or
        // not escaping a character will change its semantics. Likewise, a URI
        // must be separated into its components before the escaped characters
        // within those components can be safely decoded.
        //
        // ...So we do our best to do limited escaping of the URL, without 
        // causing damage.  If the URL is already properly escaped, in theory, this 
        // function should not change the string value.

        final int end = string.length();
        if (end > m_attrBuff.length)
        {
           m_attrBuff = new char[end*2 + 1];               
        }
        string.getChars(0,end, m_attrBuff, 0); 
        final char[] chars = m_attrBuff;

        int cleanStart = 0;
        int cleanLength = 0;
        
        
        char ch = 0;
        for (int i = 0; i < end; i++)
        {
            ch = chars[i];

            if ((ch < 32) || (ch > 126))
            {
                if (cleanLength > 0)
                {
                    writer.write(chars, cleanStart, cleanLength);
                    cleanLength = 0;
                }
                if (doURLEscaping)
                {
                    // Encode UTF16 to UTF8.
                    // Reference is Unicode, A Primer, by Tony Graham.
                    // Page 92.

                    // Note that Kay doesn't escape 0x20...
                    //  if(ch == 0x20) // Not sure about this... -sb
                    //  {
                    //    writer.write(ch);
                    //  }
                    //  else 
                    if (ch <= 0x7F)
                    {
                        writer.write('%");
                        writer.write(makeHHString(ch));
                    }
                    else if (ch <= 0x7FF)
                    {
                        // Clear low 6 bits before rotate, put high 4 bits in low byte, 
                        // and set two high bits.
                        int high = (ch >> 6) | 0xC0;
                        int low = (ch & 0x3F) | 0x80;
                        // First 6 bits, + high bit
                        writer.write('%");
                        writer.write(makeHHString(high));
                        writer.write('%");
                        writer.write(makeHHString(low));
                    }
                    else if (Encodings.isHighUTF16Surrogate(ch)) // high surrogate
                    {
                        // I'm sure this can be done in 3 instructions, but I choose 
                        // to try and do it exactly like it is done in the book, at least 
                        // until we are sure this is totally clean.  I don't think performance 
                        // is a big issue with this particular function, though I could be 
                        // wrong.  Also, the stuff below clearly does more masking than 
                        // it needs to do.

                        // Clear high 6 bits.
                        int highSurrogate = ((int) ch) & 0x03FF;

                        // Middle 4 bits (wwww) + 1
                        // "Note that the value of wwww from the high surrogate bit pattern
                        // is incremented to make the uuuuu bit pattern in the scalar value 
                        // so the surrogate pair don't address the BMP."
                        int wwww = ((highSurrogate & 0x03C0) >> 6);
                        int uuuuu = wwww + 1;

                        // next 4 bits
                        int zzzz = (highSurrogate & 0x003C) >> 2;

                        // low 2 bits
                        int yyyyyy = ((highSurrogate & 0x0003) << 4) & 0x30;

                        // Get low surrogate character.
                        ch = chars[++i];

                        // Clear high 6 bits.
                        int lowSurrogate = ((int) ch) & 0x03FF;

                        // put the middle 4 bits into the bottom of yyyyyy (byte 3)
                        yyyyyy = yyyyyy | ((lowSurrogate & 0x03C0) >> 6);

                        // bottom 6 bits.
                        int xxxxxx = (lowSurrogate & 0x003F);

                        int byte1 = 0xF0 | (uuuuu >> 2); // top 3 bits of uuuuu
                        int byte2 =
                            0x80 | (((uuuuu & 0x03) << 4) & 0x30) | zzzz;
                        int byte3 = 0x80 | yyyyyy;
                        int byte4 = 0x80 | xxxxxx;

                        writer.write('%");
                        writer.write(makeHHString(byte1));
                        writer.write('%");
                        writer.write(makeHHString(byte2));
                        writer.write('%");
                        writer.write(makeHHString(byte3));
                        writer.write('%");
                        writer.write(makeHHString(byte4));
                    }
                    else
                    {
                        int high = (ch >> 12) | 0xE0; // top 4 bits
                        int middle = ((ch & 0x0FC0) >> 6) | 0x80;
                        // middle 6 bits
                        int low = (ch & 0x3F) | 0x80;
                        // First 6 bits, + high bit
                        writer.write('%");
                        writer.write(makeHHString(high));
                        writer.write('%");
                        writer.write(makeHHString(middle));
                        writer.write('%");
                        writer.write(makeHHString(low));
                    }

                }
                else if (escapingNotNeeded(ch))
                {
                    writer.write(ch);
                }
                else
                {
                    writer.write("&#");
                    writer.write(Integer.toString(ch));
                    writer.write(';");
                }
                // In this character range we have first written out any previously accumulated 
                // "clean" characters, then processed the current more complicated character,
                // which may have incremented "i".
                // We now we reset the next possible clean character.
                cleanStart = i + 1;
            }
            // Since http://www.ietf.org/rfc/rfc2396.txt refers to the URI grammar as
            // not allowing quotes in the URI proper syntax, nor in the fragment 
            // identifier, we believe that it's OK to double escape quotes.
            else if (ch == '"")
            {
                // If the character is a '%' number number, try to avoid double-escaping.
                // There is a question if this is legal behavior.

                // Dmitri Ilyin: to check if '%' number number is invalid. It must be checked if %xx is a sign, that would be encoded
                // The encoded signes are in Hex form. So %xx my be in form %3C that is "<" sign. I will try to change here a little.

                //        if( ((i+2) < len) && isASCIIDigit(stringArray[i+1]) && isASCIIDigit(stringArray[i+2]) )

                // We are no longer escaping '%'

                if (cleanLength > 0)
                {
                    writer.write(chars, cleanStart, cleanLength);
                    cleanLength = 0;
                }   
                
                
                // Mike Kay encodes this as ", so he may know something I don't?
                if (doURLEscaping)
                    writer.write("%22");
                else
                    writer.write("""); // we have to escape this, I guess.

                // We have written out any clean characters, then the escaped '%' and now we
                // We now we reset the next possible clean character.
                cleanStart = i + 1;    
            }
            else
            {
                // no processing for this character, just count how
                // many characters in a row that we have that need no processing
                cleanLength++;
            }
        }
        
        // are there any clean characters at the end of the array
        // that we haven't processed yet?
        if (cleanLength > 1)
        {
            // if the whole string can be written out as-is do so
            // otherwise write out the clean chars at the end of the
            // array
            if (cleanStart == 0)
                writer.write(string);
            else
                writer.write(chars, cleanStart, cleanLength);
        }
        else if (cleanLength == 1)
        {
            // a little optimization for 1 clean character
            // (we could have let the previous if(...) handle them all)
            writer.write(ch);
        }