encodinginfo.java

来自「JAVA 所有包」· Java 代码 · 共 509 行 · 第 1/2 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.
 */
/*
 * $Id: EncodingInfo.java,v 1.2.4.2 2005/09/15 12:01:24 suresh_emailid Exp $
 */
package com.sun.org.apache.xml.internal.serializer;

import java.io.UnsupportedEncodingException;

/**
 * Holds information about a given encoding, which is the Java name for the
 * encoding, the equivalent ISO name.
 * <p>
 * An object of this type has two useful methods
 * <pre>
 * isInEncoding(char ch);
 * </pre>
 * which can be called if the character is not the high one in
 * a surrogate pair and:
 * <pre>
 * isInEncoding(char high, char low);
 * </pre>
 * which can be called if the two characters from a high/low surrogate pair.
 * <p>
 * An EncodingInfo object is a node in a binary search tree. Such a node
 * will answer if a character is in the encoding, and do so for a given
 * range of unicode values (<code>m_first</code> to
 * <code>m_last</code>). It will handle a certain range of values
 * explicitly (<code>m_explFirst</code> to <code>m_explLast</code>).
 * If the unicode point is before that explicit range, that is it
 * is in the range <code>m_first <= value < m_explFirst</code>, then it will delegate to another EncodingInfo object for The root
 * of such a tree, m_before.  Likewise for values in the range 
 * <code>m_explLast < value <= m_last</code>, but delgating to <code>m_after</code>
 * <p>
 * Actually figuring out if a code point is in the encoding is expensive. So the
 * purpose of this tree is to cache such determinations, and not to build the
 * entire tree of information at the start, but only build up as much of the 
 * tree as is used during the transformation.
 * <p>
 * This Class is not a public API, and should only be used internally within
 * the serializer.
 * 
 * @xsl.usage internal
 */
public final class EncodingInfo extends Object
{

    /**
     * The ISO encoding name.
     */
    final String name;

    /**
     * The name used by the Java convertor.
     */
    final String javaName;

    /**
     * A helper object that we can ask if a
     * single char, or a surrogate UTF-16 pair
     * of chars that form a single character,
     * is in this encoding.
     */
    private InEncoding m_encoding;
    
    /**
     * This is not a public API. It returns true if the
     * char in question is in the encoding.
     * @param ch the char in question.
     * @xsl.usage internal
     */
    public boolean isInEncoding(char ch) {
        if (m_encoding == null) {
            m_encoding = new EncodingImpl();
            
            // One could put alternate logic in here to
            // instantiate another object that implements the
            // InEncoding interface. For example if the JRE is 1.4 or up
            // we could have an object that uses JRE 1.4 methods
        }
        return m_encoding.isInEncoding(ch); 
    }
    
    /**
     * This is not a public API. It returns true if the
     * character formed by the high/low pair is in the encoding.
     * @param high a char that the a high char of a high/low surrogate pair.
     * @param low a char that is the low char of a high/low surrogate pair.
     * @xsl.usage internal
     */
    public boolean isInEncoding(char high, char low) {
        if (m_encoding == null) {
            m_encoding = new EncodingImpl();
            
            // One could put alternate logic in here to
            // instantiate another object that implements the
            // InEncoding interface. For example if the JRE is 1.4 or up
            // we could have an object that uses JRE 1.4 methods
        }
        return m_encoding.isInEncoding(high, low); 
    }

    /**
     * Create an EncodingInfo object based on the ISO name and Java name.
     * If both parameters are null any character will be considered to
     * be in the encoding. This is useful for when the serializer is in
     * temporary output state, and has no assciated encoding.
     *
     * @param name reference to the ISO name.
     * @param javaName reference to the Java encoding name.
     */
    public EncodingInfo(String name, String javaName)
    {

        this.name = name;
        this.javaName = javaName;
    }
    
    
    
    /**
     * A simple interface to isolate the implementation.
     * We could also use some new JRE 1.4 methods in another implementation
     * provided we use reflection with them.
     * <p>
     * This interface is not a public API,
     * and should only be used internally within the serializer. 
     * @xsl.usage internal
     */
    private interface InEncoding {
        /**
         * Returns true if the char is in the encoding
         */
        public boolean isInEncoding(char ch);
        /**
         * Returns true if the high/low surrogate pair forms
         * a character that is in the encoding.
         */
        public boolean isInEncoding(char high, char low);
    }

    /**
     * This class implements the 
     */
    private class EncodingImpl implements InEncoding {
        


        public boolean isInEncoding(char ch1) {
            final boolean ret;
            int codePoint = Encodings.toCodePoint(ch1);
            if (codePoint < m_explFirst) {
                // The unicode value is before the range
                // that we explictly manage, so we delegate the answer.
                
                // If we don't have an m_before object to delegate to, make one.
                if (m_before == null)
                    m_before =
                        new EncodingImpl(
                            m_encoding,
                            m_first,
                            m_explFirst - 1,
                            codePoint);
                ret = m_before.isInEncoding(ch1);
            } else if (m_explLast < codePoint) {
                // The unicode value is after the range
                // that we explictly manage, so we delegate the answer.
                
                // If we don't have an m_after object to delegate to, make one.
                if (m_after == null)
                    m_after =
                        new EncodingImpl(
                            m_encoding,
                            m_explLast + 1,
                            m_last,
                            codePoint);
                ret = m_after.isInEncoding(ch1);
            } else {
                // The unicode value is in the range we explitly handle
                final int idx = codePoint - m_explFirst;
                
                // If we already know the answer, just return it.
                if (m_alreadyKnown[idx])
                    ret = m_isInEncoding[idx];
                else {
                    // We don't know the answer, so find out,
                    // which may be expensive, then cache the answer 
                    ret = inEncoding(ch1, m_encoding);
                    m_alreadyKnown[idx] = true;
                    m_isInEncoding[idx] = ret;
                }
            }
            return ret;
        }

        public boolean isInEncoding(char high, char low) {
            final boolean ret;
            int codePoint = Encodings.toCodePoint(high,low);
            if (codePoint < m_explFirst) {
                // The unicode value is before the range
                // that we explictly manage, so we delegate the answer.
                
                // If we don't have an m_before object to delegate to, make one.
                if (m_before == null)
                    m_before =
                        new EncodingImpl(
                            m_encoding,
                            m_first,
                            m_explFirst - 1,
                            codePoint);
                ret = m_before.isInEncoding(high,low);
            } else if (m_explLast < codePoint) {
                // The unicode value is after the range
                // that we explictly manage, so we delegate the answer.
                
                // If we don't have an m_after object to delegate to, make one.
                if (m_after == null)
                    m_after =
                        new EncodingImpl(
                            m_encoding,
                            m_explLast + 1,
                            m_last,
                            codePoint);
                ret = m_after.isInEncoding(high,low);
            } else {
                // The unicode value is in the range we explitly handle
                final int idx = codePoint - m_explFirst;
                
                // If we already know the answer, just return it.
                if (m_alreadyKnown[idx])
                    ret = m_isInEncoding[idx];
                else {
                    // We don't know the answer, so find out,
                    // which may be expensive, then cache the answer 
                    ret = inEncoding(high, low, m_encoding);
                    m_alreadyKnown[idx] = true;
                    m_isInEncoding[idx] = ret;
                }
            }
            return ret;
        }