jars/icu4j-52_1/tools/misc/src/com/ibm/icu/dev/tool/layout/ArabicCharacterData.java

   1 /*
   2  *******************************************************************************
   3  * Copyright (C) 1998-2004, International Business Machines Corporation and    *
   4  * others. All Rights Reserved.                                                *
   5  *******************************************************************************
   6  *
   7  * Created on Dec 3, 2003
   8  *
   9  *******************************************************************************
  10  */
  11 package com.ibm.icu.dev.tool.layout;
  12
  13 import com.ibm.icu.lang.UCharacter;
  14 import com.ibm.icu.lang.UProperty;
  15 import com.ibm.icu.text.Normalizer;
  16 import com.ibm.icu.text.UnicodeSet;
  17
  18 public class ArabicCharacterData
  19 {
  20     public class Record
  21     {
  22         public int getCodePoint()
  23         {
  24             return codePoint;
  25         }
  26
  27         public int getGeneralCategory()
  28         {
  29             return generalCategory;
  30         }
  31
  32         public int getDecompositionType()
  33         {
  34             return decompositionType;
  35         }
  36
  37         public String getDecomposition()
  38         {
  39             return decomposition;
  40         }
  41
  42         private Record(int character)
  43         {
  44             codePoint         = character;
  45             generalCategory   = UCharacter.getType(character);
  46             decompositionType = UCharacter.getIntPropertyValue(character, UProperty.DECOMPOSITION_TYPE);
  47
  48             switch (decompositionType) {
  49             case UCharacter.DecompositionType.FINAL:
  50             case UCharacter.DecompositionType.INITIAL:
  51             case UCharacter.DecompositionType.ISOLATED:
  52             case UCharacter.DecompositionType.MEDIAL:
  53                 decomposition = Normalizer.compose(UCharacter.toString(character), true);
  54                 break;
  55
  56             case UCharacter.DecompositionType.CANONICAL:
  57                 decomposition = Normalizer.decompose(UCharacter.toString(character), true);
  58                 break;
  59
  60             default:
  61                 decomposition = null;
  62             }
  63         }
  64
  65         private int codePoint;
  66         private int generalCategory;
  67         private int decompositionType;
  68         private String decomposition;
  69     }
  70
  71     private ArabicCharacterData(int charCount)
  72     {
  73         records = new Record[charCount];
  74     }
  75
  76     private void add(int character)
  77     {
  78         records[recordIndex++] = new Record(character);
  79     }
  80
  81     public Record getRecord(int index)
  82     {
  83         if (index < 0 || index >= records.length) {
  84             return null;
  85         }
  86
  87         return records[index];
  88     }
  89
  90     public int countRecords()
  91     {
  92         return records.length;
  93     }
  94
  95     // TODO: do we need to change this to use UnicodeSetIterator?
  96     // That will mean not knowing the number of characters until
  97     // after the iteration is done, so we'd have to use a vector
  98     // to hold the Records at first and copy it to an array
  99     // when we're done...
 100     public static ArabicCharacterData factory(UnicodeSet characterSet)
 101     {
 102         int charCount = characterSet.size();
 103         ArabicCharacterData data = new ArabicCharacterData(charCount);
 104
 105         for (int i = 0; i < charCount; i += 1) {
 106             data.add(characterSet.charAt(i));
 107         }
 108
 109         return data;
 110     }
 111
 112     private Record[] records;
 113     private int recordIndex = 0;
 114 }