charinfo.java

来自「JAVA的一些源码 JAVA2 STANDARD EDITION DEVELO」· Java 代码 · 共 610 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: CharInfo.java,v 1.11 2004/02/23 10:29:37 aruny Exp $
 */
package com.sun.org.apache.xml.internal.serializer;

import java.io.BufferedReader;
import java.io.InputStream;
import java.io.InputStreamReader;
import java.io.UnsupportedEncodingException;
import java.net.URL;
import java.util.Hashtable;
import java.util.PropertyResourceBundle;
import java.util.Enumeration;
import java.util.ResourceBundle;

import javax.xml.transform.TransformerException;

import com.sun.org.apache.xml.internal.res.XMLErrorResources;
import com.sun.org.apache.xml.internal.res.XMLMessages;
import com.sun.org.apache.xml.internal.utils.CharKey;
import com.sun.org.apache.xml.internal.utils.SystemIDResolver;
import com.sun.org.apache.xml.internal.utils.WrappedRuntimeException;

/**
 * This class provides services that tell if a character should have
 * special treatement, such as entity reference substitution or normalization
 * of a newline character.  It also provides character to entity reference
 * lookup.
 *
 * DEVELOPERS: See Known Issue in the constructor.
 * 
 * @xsl.usage internal
 */
class CharInfo
{
    /** Lookup table for characters to entity references. */
    private Hashtable m_charToEntityRef = new Hashtable();

    /**
     * The name of the HTML entities file.
     * If specified, the file will be resource loaded with the default class loader.
     */
    public static String HTML_ENTITIES_RESOURCE = "com.sun.org.apache.xml.internal.serializer.HTMLEntities";

    /**
     * The name of the XML entities file.
     * If specified, the file will be resource loaded with the default class loader.
     */
    public static String XML_ENTITIES_RESOURCE = "com.sun.org.apache.xml.internal.serializer.XMLEntities";

    /** The horizontal tab character, which the parser should always normalize. */
    public static final char S_HORIZONAL_TAB = 0x09;

    /** The linefeed character, which the parser should always normalize. */
    public static final char S_LINEFEED = 0x0A;

    /** The carriage return character, which the parser should always normalize. */
    public static char S_CARRIAGERETURN = 0x0D;
    
    /** This flag is an optimization for HTML entities. It false if entities 
     * other than quot (34), amp (38), lt (60) and gt (62) are defined
     * in the range 0 to 127.
     * @xsl.usage internal
     */    
    final boolean onlyQuotAmpLtGt;
    
    /** Copy the first 0,1 ... ASCII_MAX values into an array */
    private static final int ASCII_MAX = 128;
    
    /** Array of values is faster access than a set of bits 
     * to quickly check ASCII characters in attribute values. 
     */
    private boolean[] isSpecialAttrASCII = new boolean[ASCII_MAX];
    
    /** Array of values is faster access than a set of bits 
     * to quickly check ASCII characters in text nodes. 
     */
    private boolean[] isSpecialTextASCII = new boolean[ASCII_MAX];

    private boolean[] isCleanTextASCII = new boolean[ASCII_MAX];

    /** An array of bits to record if the character is in the set.
     * Although information in this array is complete, the
     * isSpecialAttrASCII array is used first because access to its values
     * is common and faster.
     */   
    private int array_of_bits[] = createEmptySetOfIntegers(65535);
     
    
    // 5 for 32 bit words,  6 for 64 bit words ...
    /*
     * This constant is used to shift an integer to quickly
     * calculate which element its bit is stored in.
     * 5 for 32 bit words (int) ,  6 for 64 bit words (long)
     */
    private static final int SHIFT_PER_WORD = 5;
    
    /*
     * A mask to get the low order bits which are used to
     * calculate the value of the bit within a given word,
     * that will represent the presence of the integer in the 
     * set.
     * 
     * 0x1F for 32 bit words (int),
     * or 0x3F for 64 bit words (long) 
     */
    private static final int LOW_ORDER_BITMASK = 0x1f;
    
    /*
     * This is used for optimizing the lookup of bits representing
     * the integers in the set. It is the index of the first element
     * in the array array_of_bits[] that is not used.
     */
    private int firstWordNotUsed;


    /**
     * Constructor that reads in a resource file that describes the mapping of
     * characters to entity references.
     * This constructor is private, just to force the use
     * of the getCharInfo(entitiesResource) factory
     *
     * Resource files must be encoded in UTF-8 and can either be properties
     * files with a .properties extension assumed.  Alternatively, they can
     * have the following form, with no particular extension assumed:
     *
     * <pre>
     * # First char # is a comment
     * Entity numericValue
     * quot 34
     * amp 38
     * </pre>
     *    
     * @param entitiesResource Name of properties or resource file that should
     * be loaded, which describes that mapping of characters to entity
     * references.
     */
    private CharInfo(String entitiesResource, String method)
    {
        this(entitiesResource, method, false);
    }

    private CharInfo(String entitiesResource, String method, boolean internal)
    {
        ResourceBundle entities = null;
        boolean noExtraEntities = true;

        // Make various attempts to interpret the parameter as a properties
        // file or resource file, as follows:
        //
        //   1) attempt to load .properties file using ResourceBundle
        //   2) try using the class loader to find the specified file a resource
        //      file
        //   3) try treating the resource a URI

        if (internal) { 
            try {
                // Load entity property files by using PropertyResourceBundle,
                // cause of security issure for applets
                entities = PropertyResourceBundle.getBundle(entitiesResource);
            } catch (Exception e) {}
        }

        if (entities != null) {
            Enumeration keys = entities.getKeys();
            while (keys.hasMoreElements()){
                String name = (String) keys.nextElement();
                String value = entities.getString(name);
                int code = Integer.parseInt(value);
                defineEntity(name, (char) code);
                if (extraEntity(code))
                    noExtraEntities = false;
            }
            set(S_LINEFEED);
            set(S_CARRIAGERETURN);
        } else {
            InputStream is = null;

            // Load user specified resource file by using URL loading, it
            // requires a valid URI as parameter
            try {
                if (internal) {
                    is = CharInfo.class.getResourceAsStream(entitiesResource);
                } else {
                    ClassLoader cl = ObjectFactory.findClassLoader();
                    if (cl == null) {
                        is = ClassLoader.getSystemResourceAsStream(entitiesResource);
                    } else {
                        is = cl.getResourceAsStream(entitiesResource);
                    }

                    if (is == null) {
                        try {
                            URL url = new URL(entitiesResource);
                            is = url.openStream();
                        } catch (Exception e) {}
                    }
                }

                if (is == null) {
                    throw new RuntimeException(
                        XMLMessages.createXMLMessage(
                            XMLErrorResources.ER_RESOURCE_COULD_NOT_FIND,
                            new Object[] {entitiesResource, entitiesResource}));
                }

                // Fix Bugzilla#4000: force reading in UTF-8
                //  This creates the de facto standard that Xalan's resource 
                //  files must be encoded in UTF-8. This should work in all
                // JVMs.
                //
                // %REVIEW% KNOWN ISSUE: IT FAILS IN MICROSOFT VJ++, which
                // didn't implement the UTF-8 encoding. Theoretically, we should
                // simply let it fail in that case, since the JVM is obviously
                // broken if it doesn't support such a basic standard.  But
                // since there are still some users attempting to use VJ++ for
                // development, we have dropped in a fallback which makes a
                // second attempt using the platform's default encoding. In VJ++
                // this is apparently ASCII, which is subset of UTF-8... and
                // since the strings we'll be reading here are also primarily
                // limited to the 7-bit ASCII range (at least, in English
                // versions of Xalan), this should work well enough to keep us
                // on the air until we're ready to officially decommit from
                // VJ++.

                BufferedReader reader;
                try {
                    reader = new BufferedReader(new InputStreamReader(is, "UTF-8"));
                } catch (UnsupportedEncodingException e) {
                    reader = new BufferedReader(new InputStreamReader(is));
                }

                String line = reader.readLine();

                while (line != null) {
                    if (line.length() == 0 || line.charAt(0) == '#') {
                        line = reader.readLine();

                        continue;
                    }

                    int index = line.indexOf(' ');

                    if (index > 1) {
                        String name = line.substring(0, index);

                        ++index;

                        if (index < line.length()) {
                            String value = line.substring(index);
                            index = value.indexOf(' ');

                            if (index > 0) {
                                value = value.substring(0, index);
                            }

                            int code = Integer.parseInt(value);

                            defineEntity(name, (char) code);
                            if (extraEntity(code))
                                noExtraEntities = false;
                        }
                    }

                    line = reader.readLine();
                }

                is.close();
                set(S_LINEFEED);
                set(S_CARRIAGERETURN);
            } catch (Exception e) {
                throw new RuntimeException(
                    XMLMessages.createXMLMessage(
                        XMLErrorResources.ER_RESOURCE_COULD_NOT_LOAD,
                        new Object[] { entitiesResource,
                                       e.toString(),
                                       entitiesResource,
                                       e.toString()}));
            } finally {
                if (is != null) {
                    try {
                        is.close();
                    } catch (Exception except) {}
                }
            }
        }
          
        /* initialize the array isCleanTextASCII[] with a cache of values
         * for use by ToStream.character(char[], int , int)
         * and the array isSpecialTextASCII[] with the opposite values