jars/icu4j-52_1/main/classes/translit/src/com/ibm/icu/text/CaseFoldTransliterator.java

   1 /*
   2  *******************************************************************************
   3  * Copyright (C) 2009-2011, Google, International Business Machines Corporation
   4  * and others. All Rights Reserved.
   5  *******************************************************************************
   6  */
   7 package com.ibm.icu.text;
   8
   9 import com.ibm.icu.impl.UCaseProps;
  10 import com.ibm.icu.lang.UCharacter;
  11
  12 /**
  13  * A transliterator that performs locale-sensitive toLower()
  14  * case mapping.
  15  */
  16 class CaseFoldTransliterator extends Transliterator{
  17
  18     /**
  19      * Package accessible ID.
  20      */
  21     static final String _ID = "Any-CaseFold";
  22
  23     // TODO: Add variants for tr, az, lt, default = default locale
  24
  25     /**
  26      * System registration hook.
  27      */
  28     static void register() {
  29         Transliterator.registerFactory(_ID, new Transliterator.Factory() {
  30             public Transliterator getInstance(String ID) {
  31                 return new CaseFoldTransliterator();
  32             }
  33         });
  34
  35         Transliterator.registerSpecialInverse("CaseFold", "Upper", false);
  36     }
  37
  38     private UCaseProps csp;
  39     private ReplaceableContextIterator iter;
  40     private StringBuilder result;
  41
  42     /**
  43      * Constructs a transliterator.
  44      */
  45
  46     public CaseFoldTransliterator() {
  47         super(_ID, null);
  48         csp=UCaseProps.INSTANCE;
  49         iter=new ReplaceableContextIterator();
  50         result = new StringBuilder();
  51     }
  52
  53     /**
  54      * Implements {@link Transliterator#handleTransliterate}.
  55      */
  56     protected synchronized void handleTransliterate(Replaceable text,
  57                                        Position offsets, boolean isIncremental) {
  58         if(csp==null) {
  59             return;
  60         }
  61
  62         if(offsets.start >= offsets.limit) {
  63             return;
  64         }
  65
  66         iter.setText(text);
  67         result.setLength(0);
  68         int c, delta;
  69
  70         // Walk through original string
  71         // If there is a case change, modify corresponding position in replaceable
  72
  73         iter.setIndex(offsets.start);
  74         iter.setLimit(offsets.limit);
  75         iter.setContextLimits(offsets.contextStart, offsets.contextLimit);
  76         while((c=iter.nextCaseMapCP())>=0) {
  77             c=csp.toFullFolding(c, result, 0); // toFullFolding(int c, StringBuffer out, int options)
  78
  79             if(iter.didReachLimit() && isIncremental) {
  80                 // the case mapping function tried to look beyond the context limit
  81                 // wait for more input
  82                 offsets.start=iter.getCaseMapCPStart();
  83                 return;
  84             }
  85
  86             /* decode the result */
  87             if(c<0) {
  88                 /* c mapped to itself, no change */
  89                 continue;
  90             } else if(c<=UCaseProps.MAX_STRING_LENGTH) {
  91                 /* replace by the mapping string */
  92                 delta=iter.replace(result.toString());
  93                 result.setLength(0);
  94             } else {
  95                 /* replace by single-code point mapping */
  96                 delta=iter.replace(UTF16.valueOf(c));
  97             }
  98
  99             if(delta!=0) {
 100                 offsets.limit += delta;
 101                 offsets.contextLimit += delta;
 102             }
 103         }
 104         offsets.start = offsets.limit;
 105     }
 106
 107     static SourceTargetUtility sourceTargetUtility = null;
 108
 109     /* (non-Javadoc)
 110      * @see com.ibm.icu.text.Transliterator#addSourceTargetSet(com.ibm.icu.text.UnicodeSet, com.ibm.icu.text.UnicodeSet, com.ibm.icu.text.UnicodeSet)
 111      */
 112     @Override
 113     public void addSourceTargetSet(UnicodeSet inputFilter, UnicodeSet sourceSet, UnicodeSet targetSet) {
 114         synchronized (UppercaseTransliterator.class) {
 115             if (sourceTargetUtility == null) {
 116                 sourceTargetUtility = new SourceTargetUtility(new Transform<String,String>() {
 117                     public String transform(String source) {
 118                         return UCharacter.foldCase(source, true);
 119                     }
 120                 });
 121             }
 122         }
 123         sourceTargetUtility.addSourceTargetSet(this, inputFilter, sourceSet, targetSet);
 124     }
 125 }