uniset_closure.cpp | Explore in Territory

// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
*******************************************************************************
*
*   Copyright (C) 2011, International Business Machines
*   Corporation and others.  All Rights Reserved.
*
*******************************************************************************
*   file name:  uniset_closure.cpp
*   encoding:   UTF-8
*   tab size:   8 (not used)
*   indentation:4
*
*   created on: 2011may30
*   created by: Markus W. Scherer
*
*   UnicodeSet::closeOver() and related methods moved here from uniset_props.cpp
*   to simplify dependencies.
*   In particular, this depends on the BreakIterator, but the BreakIterator
*   code also builds UnicodeSets from patterns and needs uniset_props.
*/

#include "unicode/brkiter.h"
#include "unicode/locid.h"
#include "unicode/parsepos.h"
#include "unicode/uniset.h"
#include "unicode/utf16.h"
#include "cmemory.h"
#include "ruleiter.h"
#include "ucase.h"
#include "uprops.h"
#include "util.h"
#include "uvector.h"

U_NAMESPACE_BEGIN

// TODO memory debugging provided inside uniset.cpp
// could be made available here but probably obsolete with use of modern
// memory leak checker tools
#define _dbgct(me) …

//----------------------------------------------------------------
// Constructors &c
//----------------------------------------------------------------

UnicodeSet::UnicodeSet(const UnicodeString& pattern,
                       uint32_t options,
                       const SymbolTable* symbols,
                       UErrorCode& status) { … }

UnicodeSet::UnicodeSet(const UnicodeString& pattern, ParsePosition& pos,
                       uint32_t options,
                       const SymbolTable* symbols,
                       UErrorCode& status) { … }

//----------------------------------------------------------------
// Public API
//----------------------------------------------------------------

UnicodeSet& UnicodeSet::applyPattern(const UnicodeString& pattern,
                                     uint32_t options,
                                     const SymbolTable* symbols,
                                     UErrorCode& status) { … }

UnicodeSet& UnicodeSet::applyPattern(const UnicodeString& pattern,
                              ParsePosition& pos,
                              uint32_t options,
                              const SymbolTable* symbols,
                              UErrorCode& status) { … }

// USetAdder implementation
// Does not use uset.h to reduce code dependencies
static void U_CALLCONV
_set_add(USet *set, UChar32 c) { … }

static void U_CALLCONV
_set_addRange(USet *set, UChar32 start, UChar32 end) { … }

static void U_CALLCONV
_set_addString(USet *set, const char16_t *str, int32_t length) { … }

//----------------------------------------------------------------
// Case folding API
//----------------------------------------------------------------

// add the result of a full case mapping to the set
// use str as a temporary string to avoid constructing one
static inline void
addCaseMapping(UnicodeSet &set, int32_t result, const char16_t *full, UnicodeString &str) { … }

namespace {

/** For case closure on a large set, look only at code points with relevant properties. */
const UnicodeSet &maybeOnlyCaseSensitive(const UnicodeSet &src, UnicodeSet &subset) { … }

// Per-character scf = Simple_Case_Folding of a string.
// (Normally when we case-fold a string we use full case foldings.)
bool scfString(const UnicodeString &s, UnicodeString &scf) { … }

}  // namespace

UnicodeSet& UnicodeSet::closeOver(int32_t attribute) { … }

void UnicodeSet::closeOverCaseInsensitive(bool simple) { … }

void UnicodeSet::closeOverAddCaseMappings() { … }

U_NAMESPACE_END
chromium/third_party/icu/source/common/uniset_closure.cpp
