icuSources/i18n/regeximp.cpp

   1 //
   2 //   Copyright (C) 2012 International Business Machines Corporation
   3 //   and others. All rights reserved.
   4 //
   5 //   file:  regeximp.cpp
   6 //
   7 //           ICU Regular Expressions,
   8 //             miscellaneous implementation functions.
   9 //
  10
  11 #include "unicode/utypes.h"
  12
  13 #if !UCONFIG_NO_REGULAR_EXPRESSIONS
  14 #include "regeximp.h"
  15 #include "unicode/utf16.h"
  16
  17 U_NAMESPACE_BEGIN
  18
  19 CaseFoldingUTextIterator::CaseFoldingUTextIterator(UText &text) :
  20    fUText(text), fcsp(NULL), fFoldChars(NULL), fFoldLength(0) {
  21    fcsp = ucase_getSingleton();
  22 }
  23
  24 CaseFoldingUTextIterator::~CaseFoldingUTextIterator() {}
  25
  26 UChar32 CaseFoldingUTextIterator::next() {
  27     UChar32  foldedC;
  28     UChar32  originalC;
  29     if (fFoldChars == NULL) {
  30         // We are not in a string folding of an earlier character.
  31         // Start handling the next char from the input UText.
  32         originalC = UTEXT_NEXT32(&fUText);
  33         if (originalC == U_SENTINEL) {
  34             return originalC;
  35         }
  36         fFoldLength = ucase_toFullFolding(fcsp, originalC, &fFoldChars, U_FOLD_CASE_DEFAULT);
  37         if (fFoldLength >= UCASE_MAX_STRING_LENGTH || fFoldLength < 0) {
  38             // input code point folds to a single code point, possibly itself.
  39             // See comment in ucase.h for explanation of return values from ucase_toFullFoldings.
  40             if (fFoldLength < 0) {
  41                 fFoldLength = ~fFoldLength;
  42             }
  43             foldedC = (UChar32)fFoldLength;
  44             fFoldChars = NULL;
  45             return foldedC;
  46         }
  47         // String foldings fall through here.
  48         fFoldIndex = 0;
  49     }
  50
  51     U16_NEXT(fFoldChars, fFoldIndex, fFoldLength, foldedC);
  52     if (fFoldIndex >= fFoldLength) {
  53         fFoldChars = NULL;
  54     }
  55     return foldedC;
  56 }
  57
  58
  59 UBool CaseFoldingUTextIterator::inExpansion() {
  60     return fFoldChars != NULL;
  61 }
  62
  63
  64
  65 CaseFoldingUCharIterator::CaseFoldingUCharIterator(const UChar *chars, int64_t start, int64_t limit) :
  66    fChars(chars), fIndex(start), fLimit(limit), fcsp(NULL), fFoldChars(NULL), fFoldLength(0) {
  67    fcsp = ucase_getSingleton();
  68 }
  69
  70
  71 CaseFoldingUCharIterator::~CaseFoldingUCharIterator() {}
  72
  73
  74 UChar32 CaseFoldingUCharIterator::next() {
  75     UChar32  foldedC;
  76     UChar32  originalC;
  77     if (fFoldChars == NULL) {
  78         // We are not in a string folding of an earlier character.
  79         // Start handling the next char from the input UText.
  80         if (fIndex >= fLimit) {
  81             return U_SENTINEL;
  82         }
  83         U16_NEXT(fChars, fIndex, fLimit, originalC);
  84
  85         fFoldLength = ucase_toFullFolding(fcsp, originalC, &fFoldChars, U_FOLD_CASE_DEFAULT);
  86         if (fFoldLength >= UCASE_MAX_STRING_LENGTH || fFoldLength < 0) {
  87             // input code point folds to a single code point, possibly itself.
  88             // See comment in ucase.h for explanation of return values from ucase_toFullFoldings.
  89             if (fFoldLength < 0) {
  90                 fFoldLength = ~fFoldLength;
  91             }
  92             foldedC = (UChar32)fFoldLength;
  93             fFoldChars = NULL;
  94             return foldedC;
  95         }
  96         // String foldings fall through here.
  97         fFoldIndex = 0;
  98     }
  99
 100     U16_NEXT(fFoldChars, fFoldIndex, fFoldLength, foldedC);
 101     if (fFoldIndex >= fFoldLength) {
 102         fFoldChars = NULL;
 103     }
 104     return foldedC;
 105 }
 106
 107
 108 UBool CaseFoldingUCharIterator::inExpansion() {
 109     return fFoldChars != NULL;
 110 }
 111
 112 int64_t CaseFoldingUCharIterator::getIndex() {
 113     return fIndex;
 114 }
 115
 116
 117 U_NAMESPACE_END
 118
 119 #endif
 120