+// © 2016 and later: Unicode, Inc. and others.
+// License & terms of use: http://www.unicode.org/copyright.html
/*
*******************************************************************************
*
-* Copyright (C) 1999-2010, International Business Machines
+* Copyright (C) 1999-2012, International Business Machines
* Corporation and others. All Rights Reserved.
*
*******************************************************************************
* file name: utf16.h
-* encoding: US-ASCII
+* encoding: UTF-8
* tab size: 8 (not used)
* indentation:4
*
* \brief C API: 16-bit Unicode handling macros
*
* This file defines macros to deal with 16-bit Unicode (UTF-16) code units and strings.
- * utf16.h is included by utf.h after unicode/umachine.h
- * and some common definitions.
*
* For more information see utf.h and the ICU User Guide Strings chapter
- * (http://icu-project.org/userguide/strings.html).
+ * (http://userguide.icu-project.org/strings).
*
* <em>Usage:</em>
* ICU coding guidelines for if() statements should be followed when using these macros.
#ifndef __UTF16_H__
#define __UTF16_H__
-/* utf.h must be included first. */
+#include "unicode/umachine.h"
#ifndef __UTF_H__
# include "unicode/utf.h"
#endif
* @see U16_GET
* @stable ICU 2.4
*/
-#define U16_GET_UNSAFE(s, i, c) { \
+#define U16_GET_UNSAFE(s, i, c) UPRV_BLOCK_MACRO_BEGIN { \
(c)=(s)[i]; \
if(U16_IS_SURROGATE(c)) { \
if(U16_IS_SURROGATE_LEAD(c)) { \
(c)=U16_GET_SUPPLEMENTARY((s)[(i)-1], (c)); \
} \
} \
-}
+} UPRV_BLOCK_MACRO_END
/**
* Get a code point from a string at a random-access offset,
* The offset may point to either the lead or trail surrogate unit
* for a supplementary code point, in which case the macro will read
* the adjacent matching surrogate as well.
- * If the offset points to a single, unpaired surrogate, then that itself
- * will be returned as the code point.
+ *
+ * The length can be negative for a NUL-terminated string.
+ *
+ * If the offset points to a single, unpaired surrogate, then
+ * c is set to that unpaired surrogate.
* Iteration through a string is more efficient with U16_NEXT_UNSAFE or U16_NEXT.
*
* @param s const UChar * string
* @see U16_GET_UNSAFE
* @stable ICU 2.4
*/
-#define U16_GET(s, start, i, length, c) { \
+#define U16_GET(s, start, i, length, c) UPRV_BLOCK_MACRO_BEGIN { \
+ (c)=(s)[i]; \
+ if(U16_IS_SURROGATE(c)) { \
+ uint16_t __c2 = 0; \
+ if(U16_IS_SURROGATE_LEAD(c)) { \
+ if((i)+1!=(length) && U16_IS_TRAIL(__c2=(s)[(i)+1])) { \
+ (c)=U16_GET_SUPPLEMENTARY((c), __c2); \
+ } \
+ } else { \
+ if((i)>(start) && U16_IS_LEAD(__c2=(s)[(i)-1])) { \
+ (c)=U16_GET_SUPPLEMENTARY(__c2, (c)); \
+ } \
+ } \
+ } \
+} UPRV_BLOCK_MACRO_END
+
+/**
+ * Get a code point from a string at a random-access offset,
+ * without changing the offset.
+ * "Safe" macro, handles unpaired surrogates and checks for string boundaries.
+ *
+ * The offset may point to either the lead or trail surrogate unit
+ * for a supplementary code point, in which case the macro will read
+ * the adjacent matching surrogate as well.
+ *
+ * The length can be negative for a NUL-terminated string.
+ *
+ * If the offset points to a single, unpaired surrogate, then
+ * c is set to U+FFFD.
+ * Iteration through a string is more efficient with U16_NEXT_UNSAFE or U16_NEXT_OR_FFFD.
+ *
+ * @param s const UChar * string
+ * @param start starting string offset (usually 0)
+ * @param i string offset, must be start<=i<length
+ * @param length string length
+ * @param c output UChar32 variable
+ * @see U16_GET_UNSAFE
+ * @stable ICU 60
+ */
+#define U16_GET_OR_FFFD(s, start, i, length, c) UPRV_BLOCK_MACRO_BEGIN { \
(c)=(s)[i]; \
if(U16_IS_SURROGATE(c)) { \
- uint16_t __c2; \
+ uint16_t __c2 = 0; \
if(U16_IS_SURROGATE_LEAD(c)) { \
- if((i)+1<(length) && U16_IS_TRAIL(__c2=(s)[(i)+1])) { \
+ if((i)+1!=(length) && U16_IS_TRAIL(__c2=(s)[(i)+1])) { \
(c)=U16_GET_SUPPLEMENTARY((c), __c2); \
+ } else { \
+ (c)=0xfffd; \
} \
} else { \
if((i)>(start) && U16_IS_LEAD(__c2=(s)[(i)-1])) { \
(c)=U16_GET_SUPPLEMENTARY(__c2, (c)); \
+ } else { \
+ (c)=0xfffd; \
} \
} \
} \
-}
+} UPRV_BLOCK_MACRO_END
/* definitions with forward iteration --------------------------------------- */
* @see U16_NEXT
* @stable ICU 2.4
*/
-#define U16_NEXT_UNSAFE(s, i, c) { \
+#define U16_NEXT_UNSAFE(s, i, c) UPRV_BLOCK_MACRO_BEGIN { \
(c)=(s)[(i)++]; \
if(U16_IS_LEAD(c)) { \
(c)=U16_GET_SUPPLEMENTARY((c), (s)[(i)++]); \
} \
-}
+} UPRV_BLOCK_MACRO_END
/**
* Get a code point from a string at a code point boundary offset,
* (Post-incrementing forward iteration.)
* "Safe" macro, handles unpaired surrogates and checks for string boundaries.
*
+ * The length can be negative for a NUL-terminated string.
+ *
* The offset may point to the lead surrogate unit
* for a supplementary code point, in which case the macro will read
* the following trail surrogate as well.
* If the offset points to a trail surrogate or
- * to a single, unpaired lead surrogate, then that itself
- * will be returned as the code point.
+ * to a single, unpaired lead surrogate, then c is set to that unpaired surrogate.
*
* @param s const UChar * string
* @param i string offset, must be i<length
* @see U16_NEXT_UNSAFE
* @stable ICU 2.4
*/
-#define U16_NEXT(s, i, length, c) { \
+#define U16_NEXT(s, i, length, c) UPRV_BLOCK_MACRO_BEGIN { \
(c)=(s)[(i)++]; \
if(U16_IS_LEAD(c)) { \
- uint16_t __c2; \
- if((i)<(length) && U16_IS_TRAIL(__c2=(s)[(i)])) { \
+ uint16_t __c2 = 0; \
+ if((i)!=(length) && U16_IS_TRAIL(__c2=(s)[(i)])) { \
++(i); \
(c)=U16_GET_SUPPLEMENTARY((c), __c2); \
} \
} \
-}
+} UPRV_BLOCK_MACRO_END
+
+/**
+ * Get a code point from a string at a code point boundary offset,
+ * and advance the offset to the next code point boundary.
+ * (Post-incrementing forward iteration.)
+ * "Safe" macro, handles unpaired surrogates and checks for string boundaries.
+ *
+ * The length can be negative for a NUL-terminated string.
+ *
+ * The offset may point to the lead surrogate unit
+ * for a supplementary code point, in which case the macro will read
+ * the following trail surrogate as well.
+ * If the offset points to a trail surrogate or
+ * to a single, unpaired lead surrogate, then c is set to U+FFFD.
+ *
+ * @param s const UChar * string
+ * @param i string offset, must be i<length
+ * @param length string length
+ * @param c output UChar32 variable
+ * @see U16_NEXT_UNSAFE
+ * @stable ICU 60
+ */
+#define U16_NEXT_OR_FFFD(s, i, length, c) UPRV_BLOCK_MACRO_BEGIN { \
+ (c)=(s)[(i)++]; \
+ if(U16_IS_SURROGATE(c)) { \
+ uint16_t __c2 = 0; \
+ if(U16_IS_SURROGATE_LEAD(c) && (i)!=(length) && U16_IS_TRAIL(__c2=(s)[(i)])) { \
+ ++(i); \
+ (c)=U16_GET_SUPPLEMENTARY((c), __c2); \
+ } else { \
+ (c)=0xfffd; \
+ } \
+ } \
+} UPRV_BLOCK_MACRO_END
/**
* Append a code point to a string, overwriting 1 or 2 code units.
* @see U16_APPEND
* @stable ICU 2.4
*/
-#define U16_APPEND_UNSAFE(s, i, c) { \
+#define U16_APPEND_UNSAFE(s, i, c) UPRV_BLOCK_MACRO_BEGIN { \
if((uint32_t)(c)<=0xffff) { \
(s)[(i)++]=(uint16_t)(c); \
} else { \
(s)[(i)++]=(uint16_t)(((c)>>10)+0xd7c0); \
(s)[(i)++]=(uint16_t)(((c)&0x3ff)|0xdc00); \
} \
-}
+} UPRV_BLOCK_MACRO_END
/**
* Append a code point to a string, overwriting 1 or 2 code units.
* @see U16_APPEND_UNSAFE
* @stable ICU 2.4
*/
-#define U16_APPEND(s, i, capacity, c, isError) { \
+#define U16_APPEND(s, i, capacity, c, isError) UPRV_BLOCK_MACRO_BEGIN { \
if((uint32_t)(c)<=0xffff) { \
(s)[(i)++]=(uint16_t)(c); \
} else if((uint32_t)(c)<=0x10ffff && (i)+1<(capacity)) { \
} else /* c>0x10ffff or not enough space */ { \
(isError)=TRUE; \
} \
-}
+} UPRV_BLOCK_MACRO_END
/**
* Advance the string offset from one code point boundary to the next.
* @see U16_FWD_1
* @stable ICU 2.4
*/
-#define U16_FWD_1_UNSAFE(s, i) { \
+#define U16_FWD_1_UNSAFE(s, i) UPRV_BLOCK_MACRO_BEGIN { \
if(U16_IS_LEAD((s)[(i)++])) { \
++(i); \
} \
-}
+} UPRV_BLOCK_MACRO_END
/**
* Advance the string offset from one code point boundary to the next.
* (Post-incrementing iteration.)
* "Safe" macro, handles unpaired surrogates and checks for string boundaries.
*
+ * The length can be negative for a NUL-terminated string.
+ *
* @param s const UChar * string
* @param i string offset, must be i<length
* @param length string length
* @see U16_FWD_1_UNSAFE
* @stable ICU 2.4
*/
-#define U16_FWD_1(s, i, length) { \
- if(U16_IS_LEAD((s)[(i)++]) && (i)<(length) && U16_IS_TRAIL((s)[i])) { \
+#define U16_FWD_1(s, i, length) UPRV_BLOCK_MACRO_BEGIN { \
+ if(U16_IS_LEAD((s)[(i)++]) && (i)!=(length) && U16_IS_TRAIL((s)[i])) { \
++(i); \
} \
-}
+} UPRV_BLOCK_MACRO_END
/**
* Advance the string offset from one code point boundary to the n-th next one,
* @see U16_FWD_N
* @stable ICU 2.4
*/
-#define U16_FWD_N_UNSAFE(s, i, n) { \
+#define U16_FWD_N_UNSAFE(s, i, n) UPRV_BLOCK_MACRO_BEGIN { \
int32_t __N=(n); \
while(__N>0) { \
U16_FWD_1_UNSAFE(s, i); \
--__N; \
} \
-}
+} UPRV_BLOCK_MACRO_END
/**
* Advance the string offset from one code point boundary to the n-th next one,
* (Post-incrementing iteration.)
* "Safe" macro, handles unpaired surrogates and checks for string boundaries.
*
+ * The length can be negative for a NUL-terminated string.
+ *
* @param s const UChar * string
- * @param i string offset, must be i<length
- * @param length string length
+ * @param i int32_t string offset, must be i<length
+ * @param length int32_t string length
* @param n number of code points to skip
* @see U16_FWD_N_UNSAFE
* @stable ICU 2.4
*/
-#define U16_FWD_N(s, i, length, n) { \
+#define U16_FWD_N(s, i, length, n) UPRV_BLOCK_MACRO_BEGIN { \
int32_t __N=(n); \
- while(__N>0 && (i)<(length)) { \
+ while(__N>0 && ((i)<(length) || ((length)<0 && (s)[i]!=0))) { \
U16_FWD_1(s, i, length); \
--__N; \
} \
-}
+} UPRV_BLOCK_MACRO_END
/**
* Adjust a random-access offset to a code point boundary
* @see U16_SET_CP_START
* @stable ICU 2.4
*/
-#define U16_SET_CP_START_UNSAFE(s, i) { \
+#define U16_SET_CP_START_UNSAFE(s, i) UPRV_BLOCK_MACRO_BEGIN { \
if(U16_IS_TRAIL((s)[i])) { \
--(i); \
} \
-}
+} UPRV_BLOCK_MACRO_END
/**
* Adjust a random-access offset to a code point boundary
* @see U16_SET_CP_START_UNSAFE
* @stable ICU 2.4
*/
-#define U16_SET_CP_START(s, start, i) { \
+#define U16_SET_CP_START(s, start, i) UPRV_BLOCK_MACRO_BEGIN { \
if(U16_IS_TRAIL((s)[i]) && (i)>(start) && U16_IS_LEAD((s)[(i)-1])) { \
--(i); \
} \
-}
+} UPRV_BLOCK_MACRO_END
/* definitions with backward iteration -------------------------------------- */
* @see U16_PREV
* @stable ICU 2.4
*/
-#define U16_PREV_UNSAFE(s, i, c) { \
+#define U16_PREV_UNSAFE(s, i, c) UPRV_BLOCK_MACRO_BEGIN { \
(c)=(s)[--(i)]; \
if(U16_IS_TRAIL(c)) { \
(c)=U16_GET_SUPPLEMENTARY((s)[--(i)], (c)); \
} \
-}
+} UPRV_BLOCK_MACRO_END
/**
* Move the string offset from one code point boundary to the previous one
* for a supplementary code point, then the macro will read
* the preceding lead surrogate as well.
* If the offset is behind a lead surrogate or behind a single, unpaired
- * trail surrogate, then that itself
- * will be returned as the code point.
+ * trail surrogate, then c is set to that unpaired surrogate.
*
* @param s const UChar * string
* @param start starting string offset (usually 0)
* @see U16_PREV_UNSAFE
* @stable ICU 2.4
*/
-#define U16_PREV(s, start, i, c) { \
+#define U16_PREV(s, start, i, c) UPRV_BLOCK_MACRO_BEGIN { \
(c)=(s)[--(i)]; \
if(U16_IS_TRAIL(c)) { \
- uint16_t __c2; \
+ uint16_t __c2 = 0; \
if((i)>(start) && U16_IS_LEAD(__c2=(s)[(i)-1])) { \
--(i); \
(c)=U16_GET_SUPPLEMENTARY(__c2, (c)); \
} \
} \
-}
+} UPRV_BLOCK_MACRO_END
+
+/**
+ * Move the string offset from one code point boundary to the previous one
+ * and get the code point between them.
+ * (Pre-decrementing backward iteration.)
+ * "Safe" macro, handles unpaired surrogates and checks for string boundaries.
+ *
+ * The input offset may be the same as the string length.
+ * If the offset is behind a trail surrogate unit
+ * for a supplementary code point, then the macro will read
+ * the preceding lead surrogate as well.
+ * If the offset is behind a lead surrogate or behind a single, unpaired
+ * trail surrogate, then c is set to U+FFFD.
+ *
+ * @param s const UChar * string
+ * @param start starting string offset (usually 0)
+ * @param i string offset, must be start<i
+ * @param c output UChar32 variable
+ * @see U16_PREV_UNSAFE
+ * @stable ICU 60
+ */
+#define U16_PREV_OR_FFFD(s, start, i, c) UPRV_BLOCK_MACRO_BEGIN { \
+ (c)=(s)[--(i)]; \
+ if(U16_IS_SURROGATE(c)) { \
+ uint16_t __c2 = 0; \
+ if(U16_IS_SURROGATE_TRAIL(c) && (i)>(start) && U16_IS_LEAD(__c2=(s)[(i)-1])) { \
+ --(i); \
+ (c)=U16_GET_SUPPLEMENTARY(__c2, (c)); \
+ } else { \
+ (c)=0xfffd; \
+ } \
+ } \
+} UPRV_BLOCK_MACRO_END
/**
* Move the string offset from one code point boundary to the previous one.
* @see U16_BACK_1
* @stable ICU 2.4
*/
-#define U16_BACK_1_UNSAFE(s, i) { \
+#define U16_BACK_1_UNSAFE(s, i) UPRV_BLOCK_MACRO_BEGIN { \
if(U16_IS_TRAIL((s)[--(i)])) { \
--(i); \
} \
-}
+} UPRV_BLOCK_MACRO_END
/**
* Move the string offset from one code point boundary to the previous one.
* @see U16_BACK_1_UNSAFE
* @stable ICU 2.4
*/
-#define U16_BACK_1(s, start, i) { \
+#define U16_BACK_1(s, start, i) UPRV_BLOCK_MACRO_BEGIN { \
if(U16_IS_TRAIL((s)[--(i)]) && (i)>(start) && U16_IS_LEAD((s)[(i)-1])) { \
--(i); \
} \
-}
+} UPRV_BLOCK_MACRO_END
/**
* Move the string offset from one code point boundary to the n-th one before it,
* @see U16_BACK_N
* @stable ICU 2.4
*/
-#define U16_BACK_N_UNSAFE(s, i, n) { \
+#define U16_BACK_N_UNSAFE(s, i, n) UPRV_BLOCK_MACRO_BEGIN { \
int32_t __N=(n); \
while(__N>0) { \
U16_BACK_1_UNSAFE(s, i); \
--__N; \
} \
-}
+} UPRV_BLOCK_MACRO_END
/**
* Move the string offset from one code point boundary to the n-th one before it,
* @see U16_BACK_N_UNSAFE
* @stable ICU 2.4
*/
-#define U16_BACK_N(s, start, i, n) { \
+#define U16_BACK_N(s, start, i, n) UPRV_BLOCK_MACRO_BEGIN { \
int32_t __N=(n); \
while(__N>0 && (i)>(start)) { \
U16_BACK_1(s, start, i); \
--__N; \
} \
-}
+} UPRV_BLOCK_MACRO_END
/**
* Adjust a random-access offset to a code point boundary after a code point.
* @see U16_SET_CP_LIMIT
* @stable ICU 2.4
*/
-#define U16_SET_CP_LIMIT_UNSAFE(s, i) { \
+#define U16_SET_CP_LIMIT_UNSAFE(s, i) UPRV_BLOCK_MACRO_BEGIN { \
if(U16_IS_LEAD((s)[(i)-1])) { \
++(i); \
} \
-}
+} UPRV_BLOCK_MACRO_END
/**
* Adjust a random-access offset to a code point boundary after a code point.
* The input offset may be the same as the string length.
* "Safe" macro, handles unpaired surrogates and checks for string boundaries.
*
+ * The length can be negative for a NUL-terminated string.
+ *
* @param s const UChar * string
- * @param start starting string offset (usually 0)
- * @param i string offset, start<=i<=length
- * @param length string length
+ * @param start int32_t starting string offset (usually 0)
+ * @param i int32_t string offset, start<=i<=length
+ * @param length int32_t string length
* @see U16_SET_CP_LIMIT_UNSAFE
* @stable ICU 2.4
*/
-#define U16_SET_CP_LIMIT(s, start, i, length) { \
- if((start)<(i) && (i)<(length) && U16_IS_LEAD((s)[(i)-1]) && U16_IS_TRAIL((s)[i])) { \
+#define U16_SET_CP_LIMIT(s, start, i, length) UPRV_BLOCK_MACRO_BEGIN { \
+ if((start)<(i) && ((i)<(length) || (length)<0) && U16_IS_LEAD((s)[(i)-1]) && U16_IS_TRAIL((s)[i])) { \
++(i); \
} \
-}
+} UPRV_BLOCK_MACRO_END
#endif
projects / apple / icu.git / blobdiff