Quelle  regeximp.cpp   Sprache: C

// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
//
//   Copyright (C) 2012 International Business Machines Corporation
//   and others. All rights reserved.
//
//   file:  regeximp.cpp
//
//           ICU Regular Expressions,
//             miscellaneous implementation functions.
//

#include "unicode/utypes.h"

#if !UCONFIG_NO_REGULAR_EXPRESSIONS
#include "regeximp.h"
#include "unicode/utf16.h"

U_NAMESPACE_BEGIN

CaseFoldingUTextIterator::CaseFoldingUTextIterator(UText &text) :
   fUText(text), fFoldChars(nullptr), fFoldLength(0) {
}

CaseFoldingUTextIterator::~CaseFoldingUTextIterator() {}

UChar32 CaseFoldingUTextIterator::next() {
    UChar32  foldedC;
    UChar32  originalC;
    if (fFoldChars == nullptr) {
        // We are not in a string folding of an earlier character.
        // Start handling the next char from the input UText.
        originalC = UTEXT_NEXT32(&fUText);
        if (originalC == U_SENTINEL) {
            return originalC;
        }
        fFoldLength = ucase_toFullFolding(originalC, &fFoldChars, U_FOLD_CASE_DEFAULT);
        if (fFoldLength >= UCASE_MAX_STRING_LENGTH || fFoldLength < 0) {
            // input code point folds to a single code point, possibly itself.
            // See comment in ucase.h for explanation of return values from ucase_toFullFoldings.
            if (fFoldLength < 0) {
                fFoldLength = ~fFoldLength;
            }
            foldedC = static_cast<UChar32>(fFoldLength);
            fFoldChars = nullptr;
            return foldedC;
        }
        // String foldings fall through here.
        fFoldIndex = 0;
    }

    U16_NEXT(fFoldChars, fFoldIndex, fFoldLength, foldedC);
    if (fFoldIndex >= fFoldLength) {
        fFoldChars = nullptr;
    }
    return foldedC;
}
    

UBool CaseFoldingUTextIterator::inExpansion() {
    return fFoldChars != nullptr;
}



CaseFoldingUCharIterator::CaseFoldingUCharIterator(const char16_t *chars, int64_t start, int64_t limit) :
   fChars(chars), fIndex(start), fLimit(limit), fFoldChars(nullptr), fFoldLength(0) {
}


CaseFoldingUCharIterator::~CaseFoldingUCharIterator() {}


UChar32 CaseFoldingUCharIterator::next() {
    UChar32  foldedC;
    UChar32  originalC;
    if (fFoldChars == nullptr) {
        // We are not in a string folding of an earlier character.
        // Start handling the next char from the input UText.
        if (fIndex >= fLimit) {
            return U_SENTINEL;
        }
        U16_NEXT(fChars, fIndex, fLimit, originalC);

        fFoldLength = ucase_toFullFolding(originalC, &fFoldChars, U_FOLD_CASE_DEFAULT);
        if (fFoldLength >= UCASE_MAX_STRING_LENGTH || fFoldLength < 0) {
            // input code point folds to a single code point, possibly itself.
            // See comment in ucase.h for explanation of return values from ucase_toFullFoldings.
            if (fFoldLength < 0) {
                fFoldLength = ~fFoldLength;
            }
            foldedC = static_cast<UChar32>(fFoldLength);
            fFoldChars = nullptr;
            return foldedC;
        }
        // String foldings fall through here.
        fFoldIndex = 0;
    }

    U16_NEXT(fFoldChars, fFoldIndex, fFoldLength, foldedC);
    if (fFoldIndex >= fFoldLength) {
        fFoldChars = nullptr;
    }
    return foldedC;
}
    

UBool CaseFoldingUCharIterator::inExpansion() {
    return fFoldChars != nullptr;
}

int64_t CaseFoldingUCharIterator::getIndex() {
    return fIndex;
}


U_NAMESPACE_END

#endif