// © 2016 and later: Unicode, Inc. and others. // License & terms of use: http://www.unicode.org/copyright.html /* ******************************************************************************* * * Copyright (C) 2001-2015, International Business Machines * Corporation and others. All Rights Reserved. * ******************************************************************************* * file name: ustrcase.cpp * encoding: UTF-8 * tab size: 8 (not used) * indentation:4 * * created on: 2002feb20 * created by: Markus W. Scherer * * Implementation file for string casing C API functions. * Uses functions from uchar.c for basic functionality that requires access * to the Unicode Character Database (uprops.dat). */ #include "unicode/utypes.h" #include "unicode/brkiter.h" #include "unicode/casemap.h" #include "unicode/edits.h" #include "unicode/stringoptions.h" #include "unicode/ustring.h" #include "unicode/ucasemap.h" #include "unicode/ubrk.h" #include "unicode/utf.h" #include "unicode/utf16.h" #include "cmemory.h" #include "ucase.h" #include "ucasemap_imp.h" #include "ustr_imp.h" #include "uassert.h" /** * Code point for COMBINING ACUTE ACCENT * @internal */ #define ACUTE … U_NAMESPACE_BEGIN namespace { int32_t checkOverflowAndEditsError(int32_t destIndex, int32_t destCapacity, Edits *edits, UErrorCode &errorCode) { … } /* Appends a full case mapping result, see UCASE_MAX_STRING_LENGTH. */ inline int32_t appendResult(char16_t *dest, int32_t destIndex, int32_t destCapacity, int32_t result, const char16_t *s, int32_t cpLength, uint32_t options, icu::Edits *edits) { … } inline int32_t appendUChar(char16_t *dest, int32_t destIndex, int32_t destCapacity, char16_t c) { … } int32_t appendNonEmptyUnchanged(char16_t *dest, int32_t destIndex, int32_t destCapacity, const char16_t *s, int32_t length, uint32_t options, icu::Edits *edits) { … } inline int32_t appendUnchanged(char16_t *dest, int32_t destIndex, int32_t destCapacity, const char16_t *s, int32_t length, uint32_t options, icu::Edits *edits) { … } UChar32 U_CALLCONV utf16_caseContextIterator(void *context, int8_t dir) { … } /** * caseLocale >= 0: Lowercases [srcStart..srcLimit[ but takes context [0..srcLength[ into account. * caseLocale < 0: Case-folds [srcStart..srcLimit[. */ int32_t toLower(int32_t caseLocale, uint32_t options, char16_t *dest, int32_t destCapacity, const char16_t *src, UCaseContext *csc, int32_t srcStart, int32_t srcLimit, icu::Edits *edits, UErrorCode &errorCode) { … } int32_t toUpper(int32_t caseLocale, uint32_t options, char16_t *dest, int32_t destCapacity, const char16_t *src, UCaseContext *csc, int32_t srcLength, icu::Edits *edits, UErrorCode &errorCode) { … } } // namespace U_NAMESPACE_END U_NAMESPACE_USE #if !UCONFIG_NO_BREAK_ITERATION namespace { /** * Input: c is a letter I with or without acute accent. * start is the index in src after c, and is less than segmentLimit. * If a plain i/I is followed by a plain j/J, * or an i/I with acute (precomposed or decomposed) is followed by a j/J with acute, * then we output accordingly. * * @return the src index after the titlecased sequence, or the start index if no Dutch IJ */ int32_t maybeTitleDutchIJ(const char16_t *src, UChar32 c, int32_t start, int32_t segmentLimit, char16_t *dest, int32_t &destIndex, int32_t destCapacity, uint32_t options, icu::Edits *edits) { … } } // namespace U_CFUNC int32_t U_CALLCONV ustrcase_internalToTitle(int32_t caseLocale, uint32_t options, BreakIterator *iter, char16_t *dest, int32_t destCapacity, const char16_t *src, int32_t srcLength, icu::Edits *edits, UErrorCode &errorCode) { … } #endif // !UCONFIG_NO_BREAK_ITERATION U_NAMESPACE_BEGIN namespace GreekUpper { // Data generated by prototype code, see // https://icu.unicode.org/design/case/greek-upper // TODO: Move this data into ucase.icu. static const uint16_t data0370[] = …; static const uint16_t data1F00[] = …; // U+2126 Ohm sign static const uint16_t data2126 = …; uint32_t getLetterData(UChar32 c) { … } uint32_t getDiacriticData(UChar32 c) { … } UBool isFollowedByCasedLetter(const char16_t *s, int32_t i, int32_t length) { … } /** * Greek string uppercasing with a state machine. * Probably simpler than a stateless function that has to figure out complex context-before * for each character. * TODO: Try to re-consolidate one way or another with the non-Greek function. */ int32_t toUpper(uint32_t options, char16_t *dest, int32_t destCapacity, const char16_t *src, int32_t srcLength, Edits *edits, UErrorCode &errorCode) { … } } // namespace GreekUpper U_NAMESPACE_END /* functions available in the common library (for unistr_case.cpp) */ U_CFUNC int32_t U_CALLCONV ustrcase_internalToLower(int32_t caseLocale, uint32_t options, UCASEMAP_BREAK_ITERATOR_UNUSED char16_t *dest, int32_t destCapacity, const char16_t *src, int32_t srcLength, icu::Edits *edits, UErrorCode &errorCode) { … } U_CFUNC int32_t U_CALLCONV ustrcase_internalToUpper(int32_t caseLocale, uint32_t options, UCASEMAP_BREAK_ITERATOR_UNUSED char16_t *dest, int32_t destCapacity, const char16_t *src, int32_t srcLength, icu::Edits *edits, UErrorCode &errorCode) { … } U_CFUNC int32_t U_CALLCONV ustrcase_internalFold(int32_t /* caseLocale */, uint32_t options, UCASEMAP_BREAK_ITERATOR_UNUSED char16_t *dest, int32_t destCapacity, const char16_t *src, int32_t srcLength, icu::Edits *edits, UErrorCode &errorCode) { … } U_CFUNC int32_t ustrcase_map(int32_t caseLocale, uint32_t options, UCASEMAP_BREAK_ITERATOR_PARAM char16_t *dest, int32_t destCapacity, const char16_t *src, int32_t srcLength, UStringCaseMapper *stringCaseMapper, icu::Edits *edits, UErrorCode &errorCode) { … } U_CFUNC int32_t ustrcase_mapWithOverlap(int32_t caseLocale, uint32_t options, UCASEMAP_BREAK_ITERATOR_PARAM char16_t *dest, int32_t destCapacity, const char16_t *src, int32_t srcLength, UStringCaseMapper *stringCaseMapper, UErrorCode &errorCode) { … } /* public API functions */ U_CAPI int32_t U_EXPORT2 u_strFoldCase(char16_t *dest, int32_t destCapacity, const char16_t *src, int32_t srcLength, uint32_t options, UErrorCode *pErrorCode) { … } U_NAMESPACE_BEGIN int32_t CaseMap::fold( uint32_t options, const char16_t *src, int32_t srcLength, char16_t *dest, int32_t destCapacity, Edits *edits, UErrorCode &errorCode) { … } U_NAMESPACE_END /* case-insensitive string comparisons -------------------------------------- */ /* * This function is a copy of unorm_cmpEquivFold() minus the parts for * canonical equivalence. * Keep the functions in sync, and see there for how this works. * The duplication is for modularization: * It makes caseless (but not canonical caseless) matches independent of * the normalization code. */ /* stack element for previous-level source/decomposition pointers */ struct CmpEquivLevel { … }; CmpEquivLevel; /** * Internal implementation code comparing string with case fold. * This function is called from u_strcmpFold() and u_caseInsensitivePrefixMatch(). * * @param s1 input string 1 * @param length1 length of string 1, or -1 (NUL terminated) * @param s2 input string 2 * @param length2 length of string 2, or -1 (NUL terminated) * @param options compare options * @param matchLen1 (output) length of partial prefix match in s1 * @param matchLen2 (output) length of partial prefix match in s2 * @param pErrorCode receives error status * @return The result of comparison */ static int32_t _cmpFold( const char16_t *s1, int32_t length1, const char16_t *s2, int32_t length2, uint32_t options, int32_t *matchLen1, int32_t *matchLen2, UErrorCode *pErrorCode) { … } /* internal function */ U_CFUNC int32_t u_strcmpFold(const char16_t *s1, int32_t length1, const char16_t *s2, int32_t length2, uint32_t options, UErrorCode *pErrorCode) { … } /* public API functions */ U_CAPI int32_t U_EXPORT2 u_strCaseCompare(const char16_t *s1, int32_t length1, const char16_t *s2, int32_t length2, uint32_t options, UErrorCode *pErrorCode) { … } U_CAPI int32_t U_EXPORT2 u_strcasecmp(const char16_t *s1, const char16_t *s2, uint32_t options) { … } U_CAPI int32_t U_EXPORT2 u_memcasecmp(const char16_t *s1, const char16_t *s2, int32_t length, uint32_t options) { … } U_CAPI int32_t U_EXPORT2 u_strncasecmp(const char16_t *s1, const char16_t *s2, int32_t n, uint32_t options) { … } /* internal API - detect length of shared prefix */ U_CAPI void u_caseInsensitivePrefixMatch(const char16_t *s1, int32_t length1, const char16_t *s2, int32_t length2, uint32_t options, int32_t *matchLen1, int32_t *matchLen2, UErrorCode *pErrorCode) { … }
Codebase Browser
chromium