chromium/third_party/icu/source/common/ucase.h

// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
*******************************************************************************
*
*   Copyright (C) 2004-2012, International Business Machines
*   Corporation and others.  All Rights Reserved.
*
*******************************************************************************
*   file name:  ucase.h
*   encoding:   UTF-8
*   tab size:   8 (not used)
*   indentation:4
*
*   created on: 2004aug30
*   created by: Markus W. Scherer
*
*   Low-level Unicode character/string case mapping code.
*/

#ifndef __UCASE_H__
#define __UCASE_H__

#include "unicode/utypes.h"
#include "unicode/uset.h"
#include "putilimp.h"
#include "uset_imp.h"
#include "udataswp.h"
#include "utrie2.h"

#ifdef __cplusplus
U_NAMESPACE_BEGIN

class UnicodeString;

U_NAMESPACE_END
#endif

/* library API -------------------------------------------------------------- */

U_CFUNC void U_EXPORT2
ucase_addPropertyStarts(const USetAdder *sa, UErrorCode *pErrorCode);

/**
 * Requires non-NULL locale ID but otherwise does the equivalent of
 * checking for language codes as if uloc_getLanguage() were called:
 * Accepts both 2- and 3-letter codes and accepts case variants.
 */
U_CFUNC int32_t
ucase_getCaseLocale(const char *locale);

/* Casing locale types for ucase_getCaseLocale */
enum {};

/**
 * Bit mask for getting just the options from a string compare options word
 * that are relevant for case-insensitive string comparison.
 * See stringoptions.h. Also include _STRNCMP_STYLE and U_COMPARE_CODE_POINT_ORDER.
 * @internal
 */
#define _STRCASECMP_OPTIONS_MASK

/**
 * Bit mask for getting just the options from a string compare options word
 * that are relevant for case folding (of a single string or code point).
 *
 * Currently only bit 0 for U_FOLD_CASE_EXCLUDE_SPECIAL_I.
 * It is conceivable that at some point we might use one more bit for using uppercase sharp s.
 * It is conceivable that at some point we might want the option to use only simple case foldings
 * when operating on strings.
 *
 * See stringoptions.h.
 * @internal
 */
#define _FOLD_CASE_OPTIONS_MASK

/* single-code point functions */

U_CAPI UChar32 U_EXPORT2
ucase_tolower(UChar32 c);

U_CAPI UChar32 U_EXPORT2
ucase_toupper(UChar32 c);

U_CAPI UChar32 U_EXPORT2
ucase_totitle(UChar32 c);

U_CAPI UChar32 U_EXPORT2
ucase_fold(UChar32 c, uint32_t options);

/**
 * Adds all simple case mappings and the full case folding for c to sa,
 * and also adds special case closure mappings.
 * c itself is not added.
 * For example, the mappings
 * - for s include long s
 * - for sharp s include ss
 * - for k include the Kelvin sign
 */
U_CFUNC void U_EXPORT2
ucase_addCaseClosure(UChar32 c, const USetAdder *sa);

/** Case closure with only scf=Simple_Case_Folding. */
U_CFUNC void U_EXPORT2
ucase_addSimpleCaseClosure(UChar32 c, const USetAdder *sa);

/**
 * Maps the string to single code points and adds the associated case closure
 * mappings.
 * The string is mapped to code points if it is their full case folding string.
 * In other words, this performs a reverse full case folding and then
 * adds the case closure items of the resulting code points.
 * If the string is found and its closure applied, then
 * the string itself is added as well as part of its code points' closure.
 * It must be length>=0.
 *
 * @return true if the string was found
 */
U_CFUNC UBool U_EXPORT2
ucase_addStringCaseClosure(const UChar *s, int32_t length, const USetAdder *sa);

#ifdef __cplusplus
U_NAMESPACE_BEGIN

/**
 * Iterator over characters with more than one code point in the full default Case_Folding.
 */
class U_COMMON_API FullCaseFoldingIterator {};

/**
 * Fast case mapping data for ASCII/Latin.
 * Linear arrays of delta bytes: 0=no mapping; EXC=exception.
 * Deltas must not cross the ASCII boundary, or else they cannot be easily used
 * in simple UTF-8 code.
 */
namespace LatinCase {

/** Case mapping/folding data for code points up to U+017F. */
constexpr char16_t LIMIT =;
/** U+017F case-folds and uppercases crossing the ASCII boundary. */
constexpr char16_t LONG_S =;
/** Exception: Complex mapping, or too-large delta. */
constexpr int8_t EXC =;

/** Deltas for lowercasing for most locales, and default case folding. */
extern const int8_t TO_LOWER_NORMAL[LIMIT];
/** Deltas for lowercasing for tr/az/lt, and Turkic case folding. */
extern const int8_t TO_LOWER_TR_LT[LIMIT];

/** Deltas for uppercasing for most locales. */
extern const int8_t TO_UPPER_NORMAL[LIMIT];
/** Deltas for uppercasing for tr/az. */
extern const int8_t TO_UPPER_TR[LIMIT];

}  // namespace LatinCase

U_NAMESPACE_END
#endif

/** @return UCASE_NONE, UCASE_LOWER, UCASE_UPPER, UCASE_TITLE */
U_CAPI int32_t U_EXPORT2
ucase_getType(UChar32 c);

/** @return like ucase_getType() but also sets UCASE_IGNORABLE if c is case-ignorable */
U_CAPI int32_t U_EXPORT2
ucase_getTypeOrIgnorable(UChar32 c);

U_CAPI UBool U_EXPORT2
ucase_isSoftDotted(UChar32 c);

U_CAPI UBool U_EXPORT2
ucase_isCaseSensitive(UChar32 c);

/* string case mapping functions */

U_CDECL_BEGIN

/**
 * Iterator function for string case mappings, which need to look at the
 * context (surrounding text) of a given character for conditional mappings.
 *
 * The iterator only needs to go backward or forward away from the
 * character in question. It does not use any indexes on this interface.
 * It does not support random access or an arbitrary change of
 * iteration direction.
 *
 * The code point being case-mapped itself is never returned by
 * this iterator.
 *
 * @param context A pointer to the iterator's working data.
 * @param dir If <0 then start iterating backward from the character;
 *            if >0 then start iterating forward from the character;
 *            if 0 then continue iterating in the current direction.
 * @return Next code point, or <0 when the iteration is done.
 */
UCaseContextIterator;

/**
 * Sample struct which may be used by some implementations of
 * UCaseContextIterator.
 */
struct UCaseContext {};
UCaseContext;

U_CDECL_END

#define UCASECONTEXT_INITIALIZER

enum {};

/**
 * Get the full lowercase mapping for c.
 *
 * @param csp Case mapping properties.
 * @param c Character to be mapped.
 * @param iter Character iterator, used for context-sensitive mappings.
 *             See UCaseContextIterator for details.
 *             If iter==NULL then a context-independent result is returned.
 * @param context Pointer to be passed into iter.
 * @param pString If the mapping result is a string, then the pointer is
 *                written to *pString.
 * @param caseLocale Case locale value from ucase_getCaseLocale().
 * @return Output code point or string length, see UCASE_MAX_STRING_LENGTH.
 *
 * @see UCaseContextIterator
 * @see UCASE_MAX_STRING_LENGTH
 * @internal
 */
U_CAPI int32_t U_EXPORT2
ucase_toFullLower(UChar32 c,
                  UCaseContextIterator *iter, void *context,
                  const UChar **pString,
                  int32_t caseLocale);

U_CAPI int32_t U_EXPORT2
ucase_toFullUpper(UChar32 c,
                  UCaseContextIterator *iter, void *context,
                  const UChar **pString,
                  int32_t caseLocale);

U_CAPI int32_t U_EXPORT2
ucase_toFullTitle(UChar32 c,
                  UCaseContextIterator *iter, void *context,
                  const UChar **pString,
                  int32_t caseLocale);

U_CAPI int32_t U_EXPORT2
ucase_toFullFolding(UChar32 c,
                    const UChar **pString,
                    uint32_t options);

U_CFUNC int32_t U_EXPORT2
ucase_hasBinaryProperty(UChar32 c, UProperty which);


U_CDECL_BEGIN

/**
 * @internal
 */
UCaseMapFull;

U_CDECL_END

/* for icuexportdata -------------------------------------------------------- */

struct UCaseProps {};

U_CAPI const struct UCaseProps * U_EXPORT2
ucase_getSingleton(int32_t *pExceptionsLength, int32_t *pUnfoldLength);

/* file definitions --------------------------------------------------------- */

#define UCASE_DATA_NAME
#define UCASE_DATA_TYPE

/* format "cAsE" */
#define UCASE_FMT_0
#define UCASE_FMT_1
#define UCASE_FMT_2
#define UCASE_FMT_3

/* indexes into indexes[] */
enum {};

/* definitions for 16-bit case properties word ------------------------------ */

U_CFUNC const UTrie2 * U_EXPORT2
ucase_getTrie();

/* 2-bit constants for types of cased characters */
#define UCASE_TYPE_MASK
enum {};

#define UCASE_GET_TYPE(props)
#define UCASE_GET_TYPE_AND_IGNORABLE(props)

#define UCASE_IS_UPPER_OR_TITLE(props)

#define UCASE_IGNORABLE
#define UCASE_EXCEPTION
#define UCASE_SENSITIVE

#define UCASE_HAS_EXCEPTION(props)

#define UCASE_DOT_MASK
enum {};

/* no exception: bits 15..7 are a 9-bit signed case mapping delta */
#define UCASE_DELTA_SHIFT
#define UCASE_DELTA_MASK
#define UCASE_MAX_DELTA
#define UCASE_MIN_DELTA

#if U_SIGNED_RIGHT_SHIFT_IS_ARITHMETIC
#define UCASE_GET_DELTA(props)
#else
#define UCASE_GET_DELTA
#endif

/* exception: bits 15..4 are an unsigned 12-bit index into the exceptions array */
#define UCASE_EXC_SHIFT
#define UCASE_EXC_MASK
#define UCASE_MAX_EXCEPTIONS

/* definitions for 16-bit main exceptions word ------------------------------ */

/* first 8 bits indicate values in optional slots */
enum {};

/* each slot is 2 uint16_t instead of 1 */
#define UCASE_EXC_DOUBLE_SLOTS

enum {};

/* UCASE_EXC_DOT_MASK=UCASE_DOT_MASK<<UCASE_EXC_DOT_SHIFT */
#define UCASE_EXC_DOT_SHIFT

/* normally stored in the main word, but pushed out for larger exception indexes */
#define UCASE_EXC_DOT_MASK
enum {};

/* complex/conditional mappings */
#define UCASE_EXC_CONDITIONAL_SPECIAL
#define UCASE_EXC_CONDITIONAL_FOLD

/* definitions for lengths word for full case mappings */
#define UCASE_FULL_LOWER
#define UCASE_FULL_FOLDING
#define UCASE_FULL_UPPER
#define UCASE_FULL_TITLE

/* maximum lengths */
#define UCASE_FULL_MAPPINGS_MAX_LENGTH
#define UCASE_CLOSURE_MAX_LENGTH

/* constants for reverse case folding ("unfold") data */
enum {};

#endif