linebreakdata.java

来自「《移动Agent技术》一书的所有章节源代码。」· Java 代码 · 共 372 行 · 第 1/2 页

/*
 * @(#)LineBreakData.java	1.8 98/01/12
 *
 * (C) Copyright Taligent, Inc. 1996 - All Rights Reserved
 * (C) Copyright IBM Corp. 1996 - All Rights Reserved
 *
 * Portions copyright (c) 1996 Sun Microsystems, Inc. All Rights Reserved.
 *
 *   The original version of this source code and documentation is copyrighted
 * and owned by Taligent, Inc., a wholly-owned subsidiary of IBM. These
 * materials are provided under terms of a License Agreement between Taligent
 * and Sun. This technology is protected by multiple US and International
 * patents. This notice and attribution to Taligent may not be removed.
 *   Taligent is a registered trademark of Taligent, Inc.
 *
 * Permission to use, copy, modify, and distribute this software
 * and its documentation for NON-COMMERCIAL purposes and without
 * fee is hereby granted provided that this copyright notice
 * appears in all copies. Please refer to the file "copyright.html"
 * for further important copyright and licensing information.
 *
 * SUN MAKES NO REPRESENTATIONS OR WARRANTIES ABOUT THE SUITABILITY OF
 * THE SOFTWARE, EITHER EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED
 * TO THE IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A
 * PARTICULAR PURPOSE, OR NON-INFRINGEMENT. SUN SHALL NOT BE LIABLE FOR
 * ANY DAMAGES SUFFERED BY LICENSEE AS A RESULT OF USING, MODIFYING OR
 * DISTRIBUTING THIS SOFTWARE OR ITS DERIVATIVES.
 *
 */

package java.text;

/**
 * The LineBreakData contains data used by SimpleTextBoundary
 * to determine line breaks.
 * @see #BreakIterator
 */
final class LineBreakData extends TextBoundaryData
{
    private static final byte BREAK                 = 0;
    //always breaks (must be present as first item)
    private static final byte blank                 = 1;
    //spaces, tabs, nulls.
    private static final byte cr                    = 2;
    //carriage return
    private static final byte nonBlank              = 3;
    //everything not included elsewhere
    private static final byte op                    = 4;
    //hyphens....
    private static final byte jwrd                  = 5;
    //hiragana, katakana, and kanji
    private static final byte preJwrd               = 6;
    //characters that bind to the beginning of a Japanese word
    private static final byte postJwrd              = 7;
    //characters that bind to the end of a Japanese word
    private static final byte digit                 = 8;
    //digits
    private static final byte numPunct              = 9;
    //punctuation that can appear within a number
    private static final byte currency              = 10;
    //currency symbols that can precede a number
    private static final byte nsm                   = 11;
    // non-spacing marks
    private static final byte nbsp                  = 12;
    // non-breaking characters
    private static final byte EOS                   = 13;
    private static final int COL_COUNT = 14;

    private static final byte SI = (byte)0x80;
    private static final byte STOP = (byte) 0;
    private static final byte SI_STOP = (byte)SI + STOP;

    private static final byte kLineForwardData[] =
    {
        // brk         bl             cr             nBl
        // op          kan            prJ            poJ
        // dgt         np             curr           nsm
        // nbsp        EOS
        /*00*/
        STOP,          STOP,          STOP,          STOP,
        STOP,          STOP,          STOP,          STOP,
        STOP,          STOP,          STOP,          STOP,
        STOP,          STOP,
        /*01*/
        (byte)(SI+4),  (byte)(SI+2),  (byte)(SI+7),  (byte)(SI+3),
        (byte)(SI+6),  (byte)(SI+5),  (byte)(SI+1),  (byte)(SI+8),
        (byte)(SI+9),  (byte)(SI+8),  (byte)(SI+1),  (byte)(SI+1),
        (byte)(SI+1),  SI_STOP,
        /*02*/
        (byte)(SI+4),  (byte)(SI+2),  (byte)(SI+7),  SI_STOP,
        SI_STOP,       SI_STOP,       SI_STOP,       SI_STOP,
        SI_STOP,       SI_STOP,       SI_STOP,       (byte)(SI+2),
        (byte)(SI+1),  SI_STOP,
        /*03*/
        (byte)(SI+4),  (byte)(SI+2),  (byte)(SI+7),  (byte)(SI+3),
        (byte)(SI+6),  SI_STOP,       SI_STOP,       (byte)(SI+8),
        (byte)(SI+9),  (byte)(SI+8),  SI_STOP,       (byte)(SI+3),
        (byte)(SI+1),  SI_STOP,
        /*04*/
        SI_STOP,       SI_STOP,       SI_STOP,       SI_STOP,
        SI_STOP,       SI_STOP,       SI_STOP,       SI_STOP,
        SI_STOP,       SI_STOP,       SI_STOP,       SI_STOP,
        SI_STOP,       SI_STOP,
        /*05*/
        (byte)(SI+4),  (byte)(SI+2),  (byte)(SI+7),  SI_STOP,
        SI_STOP,       SI_STOP,       SI_STOP,       (byte)(SI+8),
        SI_STOP,       (byte)(SI+8),  SI_STOP,       (byte)(SI+5),
        (byte)(SI+1),  SI_STOP,
        /*06*/
        (byte)(SI+4),  SI_STOP,       (byte)(SI+7),  SI_STOP,
        SI_STOP,       SI_STOP,       SI_STOP,       SI_STOP,
        (byte)(SI+9),  SI_STOP,       (byte)(SI+11), (byte)(SI+6),
        (byte)(SI+1),  SI_STOP,
        /*07*/
        (byte)(SI+4),  SI_STOP,       SI_STOP,       SI_STOP,
        SI_STOP,       SI_STOP,       SI_STOP,       SI_STOP,
        SI_STOP,       SI_STOP,       SI_STOP,       SI_STOP,
        SI_STOP,       SI_STOP,
        /*08*/
        (byte)(SI+4),  (byte)(SI+2),  (byte)(SI+7),  SI_STOP,
        SI_STOP,       SI_STOP,       SI_STOP,       (byte)(SI+8),
        SI_STOP,       (byte)(SI+8),  SI_STOP,       (byte)(SI+8),
        (byte)(SI+1),  SI_STOP,
        /*09*/
        (byte)(SI+4),  (byte)(SI+2),  (byte)(SI+7),  (byte)(SI+3),
        (byte)(SI+6),  SI_STOP,       SI_STOP,       (byte)(SI+8),
        (byte)(SI+9),  (byte)(SI+10), (byte)(SI+10), (byte)(SI+9),
        (byte)(SI+1),  SI_STOP,
        /*10*/
        (byte)(SI+4),  (byte)(SI+2),  (byte)(SI+7),  SI_STOP,
        SI_STOP,       SI_STOP,       SI_STOP,       (byte)(SI+8),
        (byte)(SI+9),  (byte)(SI+8),  SI_STOP,       (byte)(SI+10),
        (byte)(SI+1),  SI_STOP,
        /*11*/
        STOP,          STOP,          STOP,          STOP,
        STOP,          STOP,          STOP,          STOP,
        (byte)(SI+9),  STOP,          STOP,          (byte)(11),
        (byte)(SI+1),  STOP
    };

    private static final WordBreakTable kLineForward
        = new WordBreakTable(COL_COUNT, kLineForwardData);

    private static final byte kLineBackwardData[] =
    {
        // brk         bl             cr             nBl
        // op          kan            prJ            poJ
        // dgt         np             curr           nsm
        // nbsp        EOS
        /*00*/
        STOP,          STOP,          STOP,          STOP,
        STOP,          STOP,          STOP,          STOP,
        STOP,          STOP,          STOP,          STOP,
        STOP,          STOP,
        /*01*/
        (byte)(SI+1),  (byte)(SI+1),  (byte)(SI+1),  (byte)(SI+2),
        (byte)(SI+2),  (byte)(SI+4),  (byte)(SI+2),  (byte)(SI+3),
        (byte)(SI+2),  (byte)(SI+3),  (byte)(SI+2),  (byte)(SI+1),
        (byte)(SI+2),  STOP,
        /*02*/
        STOP,          STOP,          STOP,          (byte)(SI+2),
        (byte)(SI+2),  STOP,          (byte)(SI+2),  (byte)(SI+3),
        (byte)(SI+2),  (byte)(SI+3),  (byte)(SI+2),  (byte)(SI+2),
        (byte)(SI+2),  STOP,
        /*03*/
        STOP,          STOP,          STOP,          (byte)(SI+2),
        STOP,          (byte)(SI+4),  (byte)(SI+2),  (byte)(SI+3),
        (byte)(SI+2),  (byte)(SI+3),  (byte)(SI+2),  (byte)(SI+3),
        (byte)(SI+2),  STOP,
        /*04*/
        STOP,          STOP,          STOP,          STOP,
        STOP,          STOP,          (byte)(SI+2),  STOP,
        STOP,          STOP,          (byte)(SI+2),  (byte)(SI+4),
        (byte)(SI+4),  STOP
    };

    private static final WordBreakTable kLineBackward
        = new WordBreakTable(COL_COUNT, kLineBackwardData);

    private static final int kRawMapping[] =
    {
        nonBlank, //UNASSIGNED             = 0,
        nonBlank, //UPPERCASE_LETTER       = 1,
        nonBlank, //LOWERCASE_LETTER       = 2,
        nonBlank, //TITLECASE_LETTER       = 3,
        nonBlank, //MODIFIER_LETTER        = 4,