+// © 2016 and later: Unicode, Inc. and others.
+// License & terms of use: http://www.unicode.org/copyright.html
/*
**********************************************************************
-* Copyright (C) 1999-2011, International Business Machines
+* Copyright (C) 1999-2015, International Business Machines
* Corporation and others. All Rights Reserved.
**********************************************************************
* file name: ustr_imp.h
-* encoding: US-ASCII
+* encoding: UTF-8
* tab size: 8 (not used)
* indentation:4
*
#define __USTR_IMP_H__
#include "unicode/utypes.h"
-#include "unicode/uiter.h"
-#include "ucase.h"
-
-/** Simple declaration to avoid including unicode/ubrk.h. */
-#ifndef UBRK_TYPEDEF_UBREAK_ITERATOR
-# define UBRK_TYPEDEF_UBREAK_ITERATOR
- typedef struct UBreakIterator UBreakIterator;
-#endif
-
-#ifndef U_COMPARE_IGNORE_CASE
-/* see also unorm.h */
-/**
- * Option bit for unorm_compare:
- * Perform case-insensitive comparison.
- */
-#define U_COMPARE_IGNORE_CASE 0x10000
-#endif
+#include "unicode/utf8.h"
/**
* Internal option for unorm_cmpEquivFold() for strncmp style.
const UChar *s2, int32_t length2,
UBool strncmpStyle, UBool codePointOrder);
-/**
- * Internal API, used by u_strcasecmp() etc.
- * Compare strings case-insensitively,
- * in code point order or code unit order.
- */
-U_CFUNC int32_t
-u_strcmpFold(const UChar *s1, int32_t length1,
- const UChar *s2, int32_t length2,
- uint32_t options,
- UErrorCode *pErrorCode);
-
-/**
- * Are the Unicode properties loaded?
- * This must be used before internal functions are called that do
- * not perform this check.
- * Generate a debug assertion failure if data is not loaded.
- */
-U_CFUNC UBool
-uprv_haveProperties(UErrorCode *pErrorCode);
-
-/**
- * Load the Unicode property data.
- * Intended primarily for use from u_init().
- * Has no effect if property data is already loaded.
- * NOT thread safe.
- */
-/*U_CFUNC int8_t
-uprv_loadPropsData(UErrorCode *errorCode);*/
-
-/*
- * Internal string casing functions implementing
- * ustring.h/ustrcase.c and UnicodeString case mapping functions.
- */
-
-struct UCaseMap {
- const UCaseProps *csp;
-#if !UCONFIG_NO_BREAK_ITERATION
- UBreakIterator *iter; /* We adopt the iterator, so we own it. */
-#endif
- char locale[32];
- int32_t locCache;
- uint32_t options;
-};
-
-#ifndef __UCASEMAP_H__
-typedef struct UCaseMap UCaseMap;
-#endif
-
-#if UCONFIG_NO_BREAK_ITERATION
-# define UCASEMAP_INITIALIZER { NULL, { 0 }, 0, 0 }
-#else
-# define UCASEMAP_INITIALIZER { NULL, NULL, { 0 }, 0, 0 }
-#endif
-
-U_CFUNC void
-ustrcase_setTempCaseMapLocale(UCaseMap *csm, const char *locale);
-
-#ifndef U_STRING_CASE_MAPPER_DEFINED
-#define U_STRING_CASE_MAPPER_DEFINED
-
-/**
- * String case mapping function type, used by ustrcase_map().
- * All error checking must be done.
- * The UCaseMap must be fully initialized, with locale and/or iter set as needed.
- * src and dest must not overlap.
- */
-typedef int32_t U_CALLCONV
-UStringCaseMapper(const UCaseMap *csm,
- UChar *dest, int32_t destCapacity,
- const UChar *src, int32_t srcLength,
- UErrorCode *pErrorCode);
-
-#endif
-
-/** Implements UStringCaseMapper. */
-U_CFUNC int32_t U_CALLCONV
-ustrcase_internalToLower(const UCaseMap *csm,
- UChar *dest, int32_t destCapacity,
- const UChar *src, int32_t srcLength,
- UErrorCode *pErrorCode);
-
-/** Implements UStringCaseMapper. */
-U_CFUNC int32_t U_CALLCONV
-ustrcase_internalToUpper(const UCaseMap *csm,
- UChar *dest, int32_t destCapacity,
- const UChar *src, int32_t srcLength,
- UErrorCode *pErrorCode);
-
-#if !UCONFIG_NO_BREAK_ITERATION
-
-/** Implements UStringCaseMapper. */
-U_CFUNC int32_t U_CALLCONV
-ustrcase_internalToTitle(const UCaseMap *csm,
- UChar *dest, int32_t destCapacity,
- const UChar *src, int32_t srcLength,
- UErrorCode *pErrorCode);
-
-#endif
-
-/** Implements UStringCaseMapper. */
-U_CFUNC int32_t U_CALLCONV
-ustrcase_internalFold(const UCaseMap *csm,
- UChar *dest, int32_t destCapacity,
- const UChar *src, int32_t srcLength,
- UErrorCode *pErrorCode);
-
-/**
- * Implements argument checking and buffer handling
- * for string case mapping as a common function.
- */
-U_CFUNC int32_t
-ustrcase_map(const UCaseMap *csm,
- UChar *dest, int32_t destCapacity,
- const UChar *src, int32_t srcLength,
- UStringCaseMapper *stringCaseMapper,
- UErrorCode *pErrorCode);
-
-/**
- * UTF-8 string case mapping function type, used by ucasemap_mapUTF8().
- * UTF-8 version of UStringCaseMapper.
- * All error checking must be done.
- * The UCaseMap must be fully initialized, with locale and/or iter set as needed.
- * src and dest must not overlap.
- */
-typedef int32_t U_CALLCONV
-UTF8CaseMapper(const UCaseMap *csm,
- uint8_t *dest, int32_t destCapacity,
- const uint8_t *src, int32_t srcLength,
- UErrorCode *pErrorCode);
-
-/** Implements UTF8CaseMapper. */
-U_CFUNC int32_t U_CALLCONV
-ucasemap_internalUTF8ToTitle(const UCaseMap *csm,
- uint8_t *dest, int32_t destCapacity,
- const uint8_t *src, int32_t srcLength,
- UErrorCode *pErrorCode);
-
-/**
- * Implements argument checking and buffer handling
- * for UTF-8 string case mapping as a common function.
- */
-U_CFUNC int32_t
-ucasemap_mapUTF8(const UCaseMap *csm,
- uint8_t *dest, int32_t destCapacity,
- const uint8_t *src, int32_t srcLength,
- UTF8CaseMapper *stringCaseMapper,
- UErrorCode *pErrorCode);
-
U_CAPI int32_t U_EXPORT2
ustr_hashUCharsN(const UChar *str, int32_t length);
U_CAPI int32_t U_EXPORT2
u_terminateWChars(wchar_t *dest, int32_t destCapacity, int32_t length, UErrorCode *pErrorCode);
+/**
+ * Counts the bytes of any whole valid sequence for a UTF-8 lead byte.
+ * Returns 1 for ASCII 0..0x7f.
+ * Returns 0 for 0x80..0xc1 as well as for 0xf5..0xff.
+ * leadByte might be evaluated multiple times.
+ *
+ * @param leadByte The first byte of a UTF-8 sequence. Must be 0..0xff.
+ * @return 0..4
+ */
+#define U8_COUNT_BYTES(leadByte) \
+ (U8_IS_SINGLE(leadByte) ? 1 : U8_COUNT_BYTES_NON_ASCII(leadByte))
+
+/**
+ * Counts the bytes of any whole valid sequence for a UTF-8 lead byte.
+ * Returns 0 for 0x00..0xc1 as well as for 0xf5..0xff.
+ * leadByte might be evaluated multiple times.
+ *
+ * @param leadByte The first byte of a UTF-8 sequence. Must be 0..0xff.
+ * @return 0 or 2..4
+ */
+#define U8_COUNT_BYTES_NON_ASCII(leadByte) \
+ (U8_IS_LEAD(leadByte) ? ((uint8_t)(leadByte)>=0xe0)+((uint8_t)(leadByte)>=0xf0)+2 : 0)
+
+#ifdef __cplusplus
+
+U_NAMESPACE_BEGIN
+
+class UTF8 {
+public:
+ UTF8() = delete; // all static
+
+ /**
+ * Is t a valid UTF-8 trail byte?
+ *
+ * @param prev Must be the preceding lead byte if i==1 and length>=3;
+ * otherwise ignored.
+ * @param t The i-th byte following the lead byte.
+ * @param i The index (1..3) of byte t in the byte sequence. 0<i<length
+ * @param length The length (2..4) of the byte sequence according to the lead byte.
+ * @return TRUE if t is a valid trail byte in this context.
+ */
+ static inline UBool isValidTrail(int32_t prev, uint8_t t, int32_t i, int32_t length) {
+ // The first trail byte after a 3- or 4-byte lead byte
+ // needs to be validated together with its lead byte.
+ if (length <= 2 || i > 1) {
+ return U8_IS_TRAIL(t);
+ } else if (length == 3) {
+ return U8_IS_VALID_LEAD3_AND_T1(prev, t);
+ } else { // length == 4
+ return U8_IS_VALID_LEAD4_AND_T1(prev, t);
+ }
+ }
+};
+
+U_NAMESPACE_END
+
+#endif // __cplusplus
+
+/**
+ * Check whether a code point has the emoji property.
+ * (implemented in uchar.cpp as reduced version of u_hasBinaryProperty)
+ *
+ * @param c Code point to test.
+ * @return TRUE or FALSE according to whether c has the emoji property.
+ *
+ * @internal Apple only
+ */
+U_CAPI UBool U_EXPORT2
+u_isEmoji(UChar32 c);
+
+
#endif
projects / apple / icu.git / blobdiff