|
sl@0
|
1 |
/*
|
|
sl@0
|
2 |
*******************************************************************************
|
|
sl@0
|
3 |
*
|
|
sl@0
|
4 |
* Copyright (C) 1999-2005, International Business Machines
|
|
sl@0
|
5 |
* Corporation and others. All Rights Reserved.
|
|
sl@0
|
6 |
*
|
|
sl@0
|
7 |
*******************************************************************************
|
|
sl@0
|
8 |
* file name: utf16.h
|
|
sl@0
|
9 |
* encoding: US-ASCII
|
|
sl@0
|
10 |
* tab size: 8 (not used)
|
|
sl@0
|
11 |
* indentation:4
|
|
sl@0
|
12 |
*
|
|
sl@0
|
13 |
* created on: 1999sep09
|
|
sl@0
|
14 |
* created by: Markus W. Scherer
|
|
sl@0
|
15 |
*/
|
|
sl@0
|
16 |
|
|
sl@0
|
17 |
/**
|
|
sl@0
|
18 |
* \file
|
|
sl@0
|
19 |
* \brief C API: 16-bit Unicode handling macros
|
|
sl@0
|
20 |
*
|
|
sl@0
|
21 |
* This file defines macros to deal with 16-bit Unicode (UTF-16) code units and strings.
|
|
sl@0
|
22 |
* utf16.h is included by utf.h after unicode/umachine.h
|
|
sl@0
|
23 |
* and some common definitions.
|
|
sl@0
|
24 |
*
|
|
sl@0
|
25 |
* For more information see utf.h and the ICU User Guide Strings chapter
|
|
sl@0
|
26 |
* (http://icu.sourceforge.net/userguide/strings.html).
|
|
sl@0
|
27 |
*
|
|
sl@0
|
28 |
* <em>Usage:</em>
|
|
sl@0
|
29 |
* ICU coding guidelines for if() statements should be followed when using these macros.
|
|
sl@0
|
30 |
* Compound statements (curly braces {}) must be used for if-else-while...
|
|
sl@0
|
31 |
* bodies and all macro statements should be terminated with semicolon.
|
|
sl@0
|
32 |
*/
|
|
sl@0
|
33 |
|
|
sl@0
|
34 |
#ifndef __UTF16_H__
|
|
sl@0
|
35 |
#define __UTF16_H__
|
|
sl@0
|
36 |
|
|
sl@0
|
37 |
/* utf.h must be included first. */
|
|
sl@0
|
38 |
#ifndef __UTF_H__
|
|
sl@0
|
39 |
# include "unicode/utf.h"
|
|
sl@0
|
40 |
#endif
|
|
sl@0
|
41 |
|
|
sl@0
|
42 |
/* single-code point definitions -------------------------------------------- */
|
|
sl@0
|
43 |
|
|
sl@0
|
44 |
/**
|
|
sl@0
|
45 |
* Does this code unit alone encode a code point (BMP, not a surrogate)?
|
|
sl@0
|
46 |
* @param c 16-bit code unit
|
|
sl@0
|
47 |
* @return TRUE or FALSE
|
|
sl@0
|
48 |
* @stable ICU 2.4
|
|
sl@0
|
49 |
*/
|
|
sl@0
|
50 |
#define U16_IS_SINGLE(c) !U_IS_SURROGATE(c)
|
|
sl@0
|
51 |
|
|
sl@0
|
52 |
/**
|
|
sl@0
|
53 |
* Is this code unit a lead surrogate (U+d800..U+dbff)?
|
|
sl@0
|
54 |
* @param c 16-bit code unit
|
|
sl@0
|
55 |
* @return TRUE or FALSE
|
|
sl@0
|
56 |
* @stable ICU 2.4
|
|
sl@0
|
57 |
*/
|
|
sl@0
|
58 |
#define U16_IS_LEAD(c) (((c)&0xfffffc00)==0xd800)
|
|
sl@0
|
59 |
|
|
sl@0
|
60 |
/**
|
|
sl@0
|
61 |
* Is this code unit a trail surrogate (U+dc00..U+dfff)?
|
|
sl@0
|
62 |
* @param c 16-bit code unit
|
|
sl@0
|
63 |
* @return TRUE or FALSE
|
|
sl@0
|
64 |
* @stable ICU 2.4
|
|
sl@0
|
65 |
*/
|
|
sl@0
|
66 |
#define U16_IS_TRAIL(c) (((c)&0xfffffc00)==0xdc00)
|
|
sl@0
|
67 |
|
|
sl@0
|
68 |
/**
|
|
sl@0
|
69 |
* Is this code unit a surrogate (U+d800..U+dfff)?
|
|
sl@0
|
70 |
* @param c 16-bit code unit
|
|
sl@0
|
71 |
* @return TRUE or FALSE
|
|
sl@0
|
72 |
* @stable ICU 2.4
|
|
sl@0
|
73 |
*/
|
|
sl@0
|
74 |
#define U16_IS_SURROGATE(c) U_IS_SURROGATE(c)
|
|
sl@0
|
75 |
|
|
sl@0
|
76 |
/**
|
|
sl@0
|
77 |
* Assuming c is a surrogate code point (U16_IS_SURROGATE(c)),
|
|
sl@0
|
78 |
* is it a lead surrogate?
|
|
sl@0
|
79 |
* @param c 16-bit code unit
|
|
sl@0
|
80 |
* @return TRUE or FALSE
|
|
sl@0
|
81 |
* @stable ICU 2.4
|
|
sl@0
|
82 |
*/
|
|
sl@0
|
83 |
#define U16_IS_SURROGATE_LEAD(c) (((c)&0x400)==0)
|
|
sl@0
|
84 |
|
|
sl@0
|
85 |
/**
|
|
sl@0
|
86 |
* Helper constant for U16_GET_SUPPLEMENTARY.
|
|
sl@0
|
87 |
* @internal
|
|
sl@0
|
88 |
*/
|
|
sl@0
|
89 |
#define U16_SURROGATE_OFFSET ((0xd800<<10UL)+0xdc00-0x10000)
|
|
sl@0
|
90 |
|
|
sl@0
|
91 |
/**
|
|
sl@0
|
92 |
* Get a supplementary code point value (U+10000..U+10ffff)
|
|
sl@0
|
93 |
* from its lead and trail surrogates.
|
|
sl@0
|
94 |
* The result is undefined if the input values are not
|
|
sl@0
|
95 |
* lead and trail surrogates.
|
|
sl@0
|
96 |
*
|
|
sl@0
|
97 |
* @param lead lead surrogate (U+d800..U+dbff)
|
|
sl@0
|
98 |
* @param trail trail surrogate (U+dc00..U+dfff)
|
|
sl@0
|
99 |
* @return supplementary code point (U+10000..U+10ffff)
|
|
sl@0
|
100 |
* @stable ICU 2.4
|
|
sl@0
|
101 |
*/
|
|
sl@0
|
102 |
#define U16_GET_SUPPLEMENTARY(lead, trail) \
|
|
sl@0
|
103 |
(((UChar32)(lead)<<10UL)+(UChar32)(trail)-U16_SURROGATE_OFFSET)
|
|
sl@0
|
104 |
|
|
sl@0
|
105 |
|
|
sl@0
|
106 |
/**
|
|
sl@0
|
107 |
* Get the lead surrogate (0xd800..0xdbff) for a
|
|
sl@0
|
108 |
* supplementary code point (0x10000..0x10ffff).
|
|
sl@0
|
109 |
* @param supplementary 32-bit code point (U+10000..U+10ffff)
|
|
sl@0
|
110 |
* @return lead surrogate (U+d800..U+dbff) for supplementary
|
|
sl@0
|
111 |
* @stable ICU 2.4
|
|
sl@0
|
112 |
*/
|
|
sl@0
|
113 |
#define U16_LEAD(supplementary) (UChar)(((supplementary)>>10)+0xd7c0)
|
|
sl@0
|
114 |
|
|
sl@0
|
115 |
/**
|
|
sl@0
|
116 |
* Get the trail surrogate (0xdc00..0xdfff) for a
|
|
sl@0
|
117 |
* supplementary code point (0x10000..0x10ffff).
|
|
sl@0
|
118 |
* @param supplementary 32-bit code point (U+10000..U+10ffff)
|
|
sl@0
|
119 |
* @return trail surrogate (U+dc00..U+dfff) for supplementary
|
|
sl@0
|
120 |
* @stable ICU 2.4
|
|
sl@0
|
121 |
*/
|
|
sl@0
|
122 |
#define U16_TRAIL(supplementary) (UChar)(((supplementary)&0x3ff)|0xdc00)
|
|
sl@0
|
123 |
|
|
sl@0
|
124 |
/**
|
|
sl@0
|
125 |
* How many 16-bit code units are used to encode this Unicode code point? (1 or 2)
|
|
sl@0
|
126 |
* The result is not defined if c is not a Unicode code point (U+0000..U+10ffff).
|
|
sl@0
|
127 |
* @param c 32-bit code point
|
|
sl@0
|
128 |
* @return 1 or 2
|
|
sl@0
|
129 |
* @stable ICU 2.4
|
|
sl@0
|
130 |
*/
|
|
sl@0
|
131 |
#define U16_LENGTH(c) ((uint32_t)(c)<=0xffff ? 1 : 2)
|
|
sl@0
|
132 |
|
|
sl@0
|
133 |
/**
|
|
sl@0
|
134 |
* The maximum number of 16-bit code units per Unicode code point (U+0000..U+10ffff).
|
|
sl@0
|
135 |
* @return 2
|
|
sl@0
|
136 |
* @stable ICU 2.4
|
|
sl@0
|
137 |
*/
|
|
sl@0
|
138 |
#define U16_MAX_LENGTH 2
|
|
sl@0
|
139 |
|
|
sl@0
|
140 |
/**
|
|
sl@0
|
141 |
* Get a code point from a string at a random-access offset,
|
|
sl@0
|
142 |
* without changing the offset.
|
|
sl@0
|
143 |
* "Unsafe" macro, assumes well-formed UTF-16.
|
|
sl@0
|
144 |
*
|
|
sl@0
|
145 |
* The offset may point to either the lead or trail surrogate unit
|
|
sl@0
|
146 |
* for a supplementary code point, in which case the macro will read
|
|
sl@0
|
147 |
* the adjacent matching surrogate as well.
|
|
sl@0
|
148 |
* The result is undefined if the offset points to a single, unpaired surrogate.
|
|
sl@0
|
149 |
* Iteration through a string is more efficient with U16_NEXT_UNSAFE or U16_NEXT.
|
|
sl@0
|
150 |
*
|
|
sl@0
|
151 |
* @param s const UChar * string
|
|
sl@0
|
152 |
* @param i string offset
|
|
sl@0
|
153 |
* @param c output UChar32 variable
|
|
sl@0
|
154 |
* @see U16_GET
|
|
sl@0
|
155 |
* @stable ICU 2.4
|
|
sl@0
|
156 |
*/
|
|
sl@0
|
157 |
#define U16_GET_UNSAFE(s, i, c) { \
|
|
sl@0
|
158 |
(c)=(s)[i]; \
|
|
sl@0
|
159 |
if(U16_IS_SURROGATE(c)) { \
|
|
sl@0
|
160 |
if(U16_IS_SURROGATE_LEAD(c)) { \
|
|
sl@0
|
161 |
(c)=U16_GET_SUPPLEMENTARY((c), (s)[(i)+1]); \
|
|
sl@0
|
162 |
} else { \
|
|
sl@0
|
163 |
(c)=U16_GET_SUPPLEMENTARY((s)[(i)-1], (c)); \
|
|
sl@0
|
164 |
} \
|
|
sl@0
|
165 |
} \
|
|
sl@0
|
166 |
}
|
|
sl@0
|
167 |
|
|
sl@0
|
168 |
/**
|
|
sl@0
|
169 |
* Get a code point from a string at a random-access offset,
|
|
sl@0
|
170 |
* without changing the offset.
|
|
sl@0
|
171 |
* "Safe" macro, handles unpaired surrogates and checks for string boundaries.
|
|
sl@0
|
172 |
*
|
|
sl@0
|
173 |
* The offset may point to either the lead or trail surrogate unit
|
|
sl@0
|
174 |
* for a supplementary code point, in which case the macro will read
|
|
sl@0
|
175 |
* the adjacent matching surrogate as well.
|
|
sl@0
|
176 |
* If the offset points to a single, unpaired surrogate, then that itself
|
|
sl@0
|
177 |
* will be returned as the code point.
|
|
sl@0
|
178 |
* Iteration through a string is more efficient with U16_NEXT_UNSAFE or U16_NEXT.
|
|
sl@0
|
179 |
*
|
|
sl@0
|
180 |
* @param s const UChar * string
|
|
sl@0
|
181 |
* @param start starting string offset (usually 0)
|
|
sl@0
|
182 |
* @param i string offset, start<=i<length
|
|
sl@0
|
183 |
* @param length string length
|
|
sl@0
|
184 |
* @param c output UChar32 variable
|
|
sl@0
|
185 |
* @see U16_GET_UNSAFE
|
|
sl@0
|
186 |
* @stable ICU 2.4
|
|
sl@0
|
187 |
*/
|
|
sl@0
|
188 |
#define U16_GET(s, start, i, length, c) { \
|
|
sl@0
|
189 |
(c)=(s)[i]; \
|
|
sl@0
|
190 |
if(U16_IS_SURROGATE(c)) { \
|
|
sl@0
|
191 |
uint16_t __c2; \
|
|
sl@0
|
192 |
if(U16_IS_SURROGATE_LEAD(c)) { \
|
|
sl@0
|
193 |
if((i)+1<(length) && U16_IS_TRAIL(__c2=(s)[(i)+1])) { \
|
|
sl@0
|
194 |
(c)=U16_GET_SUPPLEMENTARY((c), __c2); \
|
|
sl@0
|
195 |
} \
|
|
sl@0
|
196 |
} else { \
|
|
sl@0
|
197 |
if((i)-1>=(start) && U16_IS_LEAD(__c2=(s)[(i)-1])) { \
|
|
sl@0
|
198 |
(c)=U16_GET_SUPPLEMENTARY(__c2, (c)); \
|
|
sl@0
|
199 |
} \
|
|
sl@0
|
200 |
} \
|
|
sl@0
|
201 |
} \
|
|
sl@0
|
202 |
}
|
|
sl@0
|
203 |
|
|
sl@0
|
204 |
/* definitions with forward iteration --------------------------------------- */
|
|
sl@0
|
205 |
|
|
sl@0
|
206 |
/**
|
|
sl@0
|
207 |
* Get a code point from a string at a code point boundary offset,
|
|
sl@0
|
208 |
* and advance the offset to the next code point boundary.
|
|
sl@0
|
209 |
* (Post-incrementing forward iteration.)
|
|
sl@0
|
210 |
* "Unsafe" macro, assumes well-formed UTF-16.
|
|
sl@0
|
211 |
*
|
|
sl@0
|
212 |
* The offset may point to the lead surrogate unit
|
|
sl@0
|
213 |
* for a supplementary code point, in which case the macro will read
|
|
sl@0
|
214 |
* the following trail surrogate as well.
|
|
sl@0
|
215 |
* If the offset points to a trail surrogate, then that itself
|
|
sl@0
|
216 |
* will be returned as the code point.
|
|
sl@0
|
217 |
* The result is undefined if the offset points to a single, unpaired lead surrogate.
|
|
sl@0
|
218 |
*
|
|
sl@0
|
219 |
* @param s const UChar * string
|
|
sl@0
|
220 |
* @param i string offset
|
|
sl@0
|
221 |
* @param c output UChar32 variable
|
|
sl@0
|
222 |
* @see U16_NEXT
|
|
sl@0
|
223 |
* @stable ICU 2.4
|
|
sl@0
|
224 |
*/
|
|
sl@0
|
225 |
#define U16_NEXT_UNSAFE(s, i, c) { \
|
|
sl@0
|
226 |
(c)=(s)[(i)++]; \
|
|
sl@0
|
227 |
if(U16_IS_LEAD(c)) { \
|
|
sl@0
|
228 |
(c)=U16_GET_SUPPLEMENTARY((c), (s)[(i)++]); \
|
|
sl@0
|
229 |
} \
|
|
sl@0
|
230 |
}
|
|
sl@0
|
231 |
|
|
sl@0
|
232 |
/**
|
|
sl@0
|
233 |
* Get a code point from a string at a code point boundary offset,
|
|
sl@0
|
234 |
* and advance the offset to the next code point boundary.
|
|
sl@0
|
235 |
* (Post-incrementing forward iteration.)
|
|
sl@0
|
236 |
* "Safe" macro, handles unpaired surrogates and checks for string boundaries.
|
|
sl@0
|
237 |
*
|
|
sl@0
|
238 |
* The offset may point to the lead surrogate unit
|
|
sl@0
|
239 |
* for a supplementary code point, in which case the macro will read
|
|
sl@0
|
240 |
* the following trail surrogate as well.
|
|
sl@0
|
241 |
* If the offset points to a trail surrogate or
|
|
sl@0
|
242 |
* to a single, unpaired lead surrogate, then that itself
|
|
sl@0
|
243 |
* will be returned as the code point.
|
|
sl@0
|
244 |
*
|
|
sl@0
|
245 |
* @param s const UChar * string
|
|
sl@0
|
246 |
* @param i string offset, i<length
|
|
sl@0
|
247 |
* @param length string length
|
|
sl@0
|
248 |
* @param c output UChar32 variable
|
|
sl@0
|
249 |
* @see U16_NEXT_UNSAFE
|
|
sl@0
|
250 |
* @stable ICU 2.4
|
|
sl@0
|
251 |
*/
|
|
sl@0
|
252 |
#define U16_NEXT(s, i, length, c) { \
|
|
sl@0
|
253 |
(c)=(s)[(i)++]; \
|
|
sl@0
|
254 |
if(U16_IS_LEAD(c)) { \
|
|
sl@0
|
255 |
uint16_t __c2; \
|
|
sl@0
|
256 |
if((i)<(length) && U16_IS_TRAIL(__c2=(s)[(i)])) { \
|
|
sl@0
|
257 |
++(i); \
|
|
sl@0
|
258 |
(c)=U16_GET_SUPPLEMENTARY((c), __c2); \
|
|
sl@0
|
259 |
} \
|
|
sl@0
|
260 |
} \
|
|
sl@0
|
261 |
}
|
|
sl@0
|
262 |
|
|
sl@0
|
263 |
/**
|
|
sl@0
|
264 |
* Append a code point to a string, overwriting 1 or 2 code units.
|
|
sl@0
|
265 |
* The offset points to the current end of the string contents
|
|
sl@0
|
266 |
* and is advanced (post-increment).
|
|
sl@0
|
267 |
* "Unsafe" macro, assumes a valid code point and sufficient space in the string.
|
|
sl@0
|
268 |
* Otherwise, the result is undefined.
|
|
sl@0
|
269 |
*
|
|
sl@0
|
270 |
* @param s const UChar * string buffer
|
|
sl@0
|
271 |
* @param i string offset
|
|
sl@0
|
272 |
* @param c code point to append
|
|
sl@0
|
273 |
* @see U16_APPEND
|
|
sl@0
|
274 |
* @stable ICU 2.4
|
|
sl@0
|
275 |
*/
|
|
sl@0
|
276 |
#define U16_APPEND_UNSAFE(s, i, c) { \
|
|
sl@0
|
277 |
if((uint32_t)(c)<=0xffff) { \
|
|
sl@0
|
278 |
(s)[(i)++]=(uint16_t)(c); \
|
|
sl@0
|
279 |
} else { \
|
|
sl@0
|
280 |
(s)[(i)++]=(uint16_t)(((c)>>10)+0xd7c0); \
|
|
sl@0
|
281 |
(s)[(i)++]=(uint16_t)(((c)&0x3ff)|0xdc00); \
|
|
sl@0
|
282 |
} \
|
|
sl@0
|
283 |
}
|
|
sl@0
|
284 |
|
|
sl@0
|
285 |
/**
|
|
sl@0
|
286 |
* Append a code point to a string, overwriting 1 or 2 code units.
|
|
sl@0
|
287 |
* The offset points to the current end of the string contents
|
|
sl@0
|
288 |
* and is advanced (post-increment).
|
|
sl@0
|
289 |
* "Safe" macro, checks for a valid code point.
|
|
sl@0
|
290 |
* If a surrogate pair is written, checks for sufficient space in the string.
|
|
sl@0
|
291 |
* If the code point is not valid or a trail surrogate does not fit,
|
|
sl@0
|
292 |
* then isError is set to TRUE.
|
|
sl@0
|
293 |
*
|
|
sl@0
|
294 |
* @param s const UChar * string buffer
|
|
sl@0
|
295 |
* @param i string offset, i<length
|
|
sl@0
|
296 |
* @param capacity size of the string buffer
|
|
sl@0
|
297 |
* @param c code point to append
|
|
sl@0
|
298 |
* @param isError output UBool set to TRUE if an error occurs, otherwise not modified
|
|
sl@0
|
299 |
* @see U16_APPEND_UNSAFE
|
|
sl@0
|
300 |
* @stable ICU 2.4
|
|
sl@0
|
301 |
*/
|
|
sl@0
|
302 |
#define U16_APPEND(s, i, capacity, c, isError) { \
|
|
sl@0
|
303 |
if((uint32_t)(c)<=0xffff) { \
|
|
sl@0
|
304 |
(s)[(i)++]=(uint16_t)(c); \
|
|
sl@0
|
305 |
} else if((uint32_t)(c)<=0x10ffff && (i)+1<(capacity)) { \
|
|
sl@0
|
306 |
(s)[(i)++]=(uint16_t)(((c)>>10)+0xd7c0); \
|
|
sl@0
|
307 |
(s)[(i)++]=(uint16_t)(((c)&0x3ff)|0xdc00); \
|
|
sl@0
|
308 |
} else /* c>0x10ffff or not enough space */ { \
|
|
sl@0
|
309 |
(isError)=TRUE; \
|
|
sl@0
|
310 |
} \
|
|
sl@0
|
311 |
}
|
|
sl@0
|
312 |
|
|
sl@0
|
313 |
/**
|
|
sl@0
|
314 |
* Advance the string offset from one code point boundary to the next.
|
|
sl@0
|
315 |
* (Post-incrementing iteration.)
|
|
sl@0
|
316 |
* "Unsafe" macro, assumes well-formed UTF-16.
|
|
sl@0
|
317 |
*
|
|
sl@0
|
318 |
* @param s const UChar * string
|
|
sl@0
|
319 |
* @param i string offset
|
|
sl@0
|
320 |
* @see U16_FWD_1
|
|
sl@0
|
321 |
* @stable ICU 2.4
|
|
sl@0
|
322 |
*/
|
|
sl@0
|
323 |
#define U16_FWD_1_UNSAFE(s, i) { \
|
|
sl@0
|
324 |
if(U16_IS_LEAD((s)[(i)++])) { \
|
|
sl@0
|
325 |
++(i); \
|
|
sl@0
|
326 |
} \
|
|
sl@0
|
327 |
}
|
|
sl@0
|
328 |
|
|
sl@0
|
329 |
/**
|
|
sl@0
|
330 |
* Advance the string offset from one code point boundary to the next.
|
|
sl@0
|
331 |
* (Post-incrementing iteration.)
|
|
sl@0
|
332 |
* "Safe" macro, handles unpaired surrogates and checks for string boundaries.
|
|
sl@0
|
333 |
*
|
|
sl@0
|
334 |
* @param s const UChar * string
|
|
sl@0
|
335 |
* @param i string offset, i<length
|
|
sl@0
|
336 |
* @param length string length
|
|
sl@0
|
337 |
* @see U16_FWD_1_UNSAFE
|
|
sl@0
|
338 |
* @stable ICU 2.4
|
|
sl@0
|
339 |
*/
|
|
sl@0
|
340 |
#define U16_FWD_1(s, i, length) { \
|
|
sl@0
|
341 |
if(U16_IS_LEAD((s)[(i)++]) && (i)<(length) && U16_IS_TRAIL((s)[i])) { \
|
|
sl@0
|
342 |
++(i); \
|
|
sl@0
|
343 |
} \
|
|
sl@0
|
344 |
}
|
|
sl@0
|
345 |
|
|
sl@0
|
346 |
/**
|
|
sl@0
|
347 |
* Advance the string offset from one code point boundary to the n-th next one,
|
|
sl@0
|
348 |
* i.e., move forward by n code points.
|
|
sl@0
|
349 |
* (Post-incrementing iteration.)
|
|
sl@0
|
350 |
* "Unsafe" macro, assumes well-formed UTF-16.
|
|
sl@0
|
351 |
*
|
|
sl@0
|
352 |
* @param s const UChar * string
|
|
sl@0
|
353 |
* @param i string offset
|
|
sl@0
|
354 |
* @param n number of code points to skip
|
|
sl@0
|
355 |
* @see U16_FWD_N
|
|
sl@0
|
356 |
* @stable ICU 2.4
|
|
sl@0
|
357 |
*/
|
|
sl@0
|
358 |
#define U16_FWD_N_UNSAFE(s, i, n) { \
|
|
sl@0
|
359 |
int32_t __N=(n); \
|
|
sl@0
|
360 |
while(__N>0) { \
|
|
sl@0
|
361 |
U16_FWD_1_UNSAFE(s, i); \
|
|
sl@0
|
362 |
--__N; \
|
|
sl@0
|
363 |
} \
|
|
sl@0
|
364 |
}
|
|
sl@0
|
365 |
|
|
sl@0
|
366 |
/**
|
|
sl@0
|
367 |
* Advance the string offset from one code point boundary to the n-th next one,
|
|
sl@0
|
368 |
* i.e., move forward by n code points.
|
|
sl@0
|
369 |
* (Post-incrementing iteration.)
|
|
sl@0
|
370 |
* "Safe" macro, handles unpaired surrogates and checks for string boundaries.
|
|
sl@0
|
371 |
*
|
|
sl@0
|
372 |
* @param s const UChar * string
|
|
sl@0
|
373 |
* @param i string offset, i<length
|
|
sl@0
|
374 |
* @param length string length
|
|
sl@0
|
375 |
* @param n number of code points to skip
|
|
sl@0
|
376 |
* @see U16_FWD_N_UNSAFE
|
|
sl@0
|
377 |
* @stable ICU 2.4
|
|
sl@0
|
378 |
*/
|
|
sl@0
|
379 |
#define U16_FWD_N(s, i, length, n) { \
|
|
sl@0
|
380 |
int32_t __N=(n); \
|
|
sl@0
|
381 |
while(__N>0 && (i)<(length)) { \
|
|
sl@0
|
382 |
U16_FWD_1(s, i, length); \
|
|
sl@0
|
383 |
--__N; \
|
|
sl@0
|
384 |
} \
|
|
sl@0
|
385 |
}
|
|
sl@0
|
386 |
|
|
sl@0
|
387 |
/**
|
|
sl@0
|
388 |
* Adjust a random-access offset to a code point boundary
|
|
sl@0
|
389 |
* at the start of a code point.
|
|
sl@0
|
390 |
* If the offset points to the trail surrogate of a surrogate pair,
|
|
sl@0
|
391 |
* then the offset is decremented.
|
|
sl@0
|
392 |
* Otherwise, it is not modified.
|
|
sl@0
|
393 |
* "Unsafe" macro, assumes well-formed UTF-16.
|
|
sl@0
|
394 |
*
|
|
sl@0
|
395 |
* @param s const UChar * string
|
|
sl@0
|
396 |
* @param i string offset
|
|
sl@0
|
397 |
* @see U16_SET_CP_START
|
|
sl@0
|
398 |
* @stable ICU 2.4
|
|
sl@0
|
399 |
*/
|
|
sl@0
|
400 |
#define U16_SET_CP_START_UNSAFE(s, i) { \
|
|
sl@0
|
401 |
if(U16_IS_TRAIL((s)[i])) { \
|
|
sl@0
|
402 |
--(i); \
|
|
sl@0
|
403 |
} \
|
|
sl@0
|
404 |
}
|
|
sl@0
|
405 |
|
|
sl@0
|
406 |
/**
|
|
sl@0
|
407 |
* Adjust a random-access offset to a code point boundary
|
|
sl@0
|
408 |
* at the start of a code point.
|
|
sl@0
|
409 |
* If the offset points to the trail surrogate of a surrogate pair,
|
|
sl@0
|
410 |
* then the offset is decremented.
|
|
sl@0
|
411 |
* Otherwise, it is not modified.
|
|
sl@0
|
412 |
* "Safe" macro, handles unpaired surrogates and checks for string boundaries.
|
|
sl@0
|
413 |
*
|
|
sl@0
|
414 |
* @param s const UChar * string
|
|
sl@0
|
415 |
* @param start starting string offset (usually 0)
|
|
sl@0
|
416 |
* @param i string offset, start<=i
|
|
sl@0
|
417 |
* @see U16_SET_CP_START_UNSAFE
|
|
sl@0
|
418 |
* @stable ICU 2.4
|
|
sl@0
|
419 |
*/
|
|
sl@0
|
420 |
#define U16_SET_CP_START(s, start, i) { \
|
|
sl@0
|
421 |
if(U16_IS_TRAIL((s)[i]) && (i)>(start) && U16_IS_LEAD((s)[(i)-1])) { \
|
|
sl@0
|
422 |
--(i); \
|
|
sl@0
|
423 |
} \
|
|
sl@0
|
424 |
}
|
|
sl@0
|
425 |
|
|
sl@0
|
426 |
/* definitions with backward iteration -------------------------------------- */
|
|
sl@0
|
427 |
|
|
sl@0
|
428 |
/**
|
|
sl@0
|
429 |
* Move the string offset from one code point boundary to the previous one
|
|
sl@0
|
430 |
* and get the code point between them.
|
|
sl@0
|
431 |
* (Pre-decrementing backward iteration.)
|
|
sl@0
|
432 |
* "Unsafe" macro, assumes well-formed UTF-16.
|
|
sl@0
|
433 |
*
|
|
sl@0
|
434 |
* The input offset may be the same as the string length.
|
|
sl@0
|
435 |
* If the offset is behind a trail surrogate unit
|
|
sl@0
|
436 |
* for a supplementary code point, then the macro will read
|
|
sl@0
|
437 |
* the preceding lead surrogate as well.
|
|
sl@0
|
438 |
* If the offset is behind a lead surrogate, then that itself
|
|
sl@0
|
439 |
* will be returned as the code point.
|
|
sl@0
|
440 |
* The result is undefined if the offset is behind a single, unpaired trail surrogate.
|
|
sl@0
|
441 |
*
|
|
sl@0
|
442 |
* @param s const UChar * string
|
|
sl@0
|
443 |
* @param i string offset
|
|
sl@0
|
444 |
* @param c output UChar32 variable
|
|
sl@0
|
445 |
* @see U16_PREV
|
|
sl@0
|
446 |
* @stable ICU 2.4
|
|
sl@0
|
447 |
*/
|
|
sl@0
|
448 |
#define U16_PREV_UNSAFE(s, i, c) { \
|
|
sl@0
|
449 |
(c)=(s)[--(i)]; \
|
|
sl@0
|
450 |
if(U16_IS_TRAIL(c)) { \
|
|
sl@0
|
451 |
(c)=U16_GET_SUPPLEMENTARY((s)[--(i)], (c)); \
|
|
sl@0
|
452 |
} \
|
|
sl@0
|
453 |
}
|
|
sl@0
|
454 |
|
|
sl@0
|
455 |
/**
|
|
sl@0
|
456 |
* Move the string offset from one code point boundary to the previous one
|
|
sl@0
|
457 |
* and get the code point between them.
|
|
sl@0
|
458 |
* (Pre-decrementing backward iteration.)
|
|
sl@0
|
459 |
* "Safe" macro, handles unpaired surrogates and checks for string boundaries.
|
|
sl@0
|
460 |
*
|
|
sl@0
|
461 |
* The input offset may be the same as the string length.
|
|
sl@0
|
462 |
* If the offset is behind a trail surrogate unit
|
|
sl@0
|
463 |
* for a supplementary code point, then the macro will read
|
|
sl@0
|
464 |
* the preceding lead surrogate as well.
|
|
sl@0
|
465 |
* If the offset is behind a lead surrogate or behind a single, unpaired
|
|
sl@0
|
466 |
* trail surrogate, then that itself
|
|
sl@0
|
467 |
* will be returned as the code point.
|
|
sl@0
|
468 |
*
|
|
sl@0
|
469 |
* @param s const UChar * string
|
|
sl@0
|
470 |
* @param start starting string offset (usually 0)
|
|
sl@0
|
471 |
* @param i string offset, start<=i
|
|
sl@0
|
472 |
* @param c output UChar32 variable
|
|
sl@0
|
473 |
* @see U16_PREV_UNSAFE
|
|
sl@0
|
474 |
* @stable ICU 2.4
|
|
sl@0
|
475 |
*/
|
|
sl@0
|
476 |
#define U16_PREV(s, start, i, c) { \
|
|
sl@0
|
477 |
(c)=(s)[--(i)]; \
|
|
sl@0
|
478 |
if(U16_IS_TRAIL(c)) { \
|
|
sl@0
|
479 |
uint16_t __c2; \
|
|
sl@0
|
480 |
if((i)>(start) && U16_IS_LEAD(__c2=(s)[(i)-1])) { \
|
|
sl@0
|
481 |
--(i); \
|
|
sl@0
|
482 |
(c)=U16_GET_SUPPLEMENTARY(__c2, (c)); \
|
|
sl@0
|
483 |
} \
|
|
sl@0
|
484 |
} \
|
|
sl@0
|
485 |
}
|
|
sl@0
|
486 |
|
|
sl@0
|
487 |
/**
|
|
sl@0
|
488 |
* Move the string offset from one code point boundary to the previous one.
|
|
sl@0
|
489 |
* (Pre-decrementing backward iteration.)
|
|
sl@0
|
490 |
* The input offset may be the same as the string length.
|
|
sl@0
|
491 |
* "Unsafe" macro, assumes well-formed UTF-16.
|
|
sl@0
|
492 |
*
|
|
sl@0
|
493 |
* @param s const UChar * string
|
|
sl@0
|
494 |
* @param i string offset
|
|
sl@0
|
495 |
* @see U16_BACK_1
|
|
sl@0
|
496 |
* @stable ICU 2.4
|
|
sl@0
|
497 |
*/
|
|
sl@0
|
498 |
#define U16_BACK_1_UNSAFE(s, i) { \
|
|
sl@0
|
499 |
if(U16_IS_TRAIL((s)[--(i)])) { \
|
|
sl@0
|
500 |
--(i); \
|
|
sl@0
|
501 |
} \
|
|
sl@0
|
502 |
}
|
|
sl@0
|
503 |
|
|
sl@0
|
504 |
/**
|
|
sl@0
|
505 |
* Move the string offset from one code point boundary to the previous one.
|
|
sl@0
|
506 |
* (Pre-decrementing backward iteration.)
|
|
sl@0
|
507 |
* The input offset may be the same as the string length.
|
|
sl@0
|
508 |
* "Safe" macro, handles unpaired surrogates and checks for string boundaries.
|
|
sl@0
|
509 |
*
|
|
sl@0
|
510 |
* @param s const UChar * string
|
|
sl@0
|
511 |
* @param start starting string offset (usually 0)
|
|
sl@0
|
512 |
* @param i string offset, start<=i
|
|
sl@0
|
513 |
* @see U16_BACK_1_UNSAFE
|
|
sl@0
|
514 |
* @stable ICU 2.4
|
|
sl@0
|
515 |
*/
|
|
sl@0
|
516 |
#define U16_BACK_1(s, start, i) { \
|
|
sl@0
|
517 |
if(U16_IS_TRAIL((s)[--(i)]) && (i)>(start) && U16_IS_LEAD((s)[(i)-1])) { \
|
|
sl@0
|
518 |
--(i); \
|
|
sl@0
|
519 |
} \
|
|
sl@0
|
520 |
}
|
|
sl@0
|
521 |
|
|
sl@0
|
522 |
/**
|
|
sl@0
|
523 |
* Move the string offset from one code point boundary to the n-th one before it,
|
|
sl@0
|
524 |
* i.e., move backward by n code points.
|
|
sl@0
|
525 |
* (Pre-decrementing backward iteration.)
|
|
sl@0
|
526 |
* The input offset may be the same as the string length.
|
|
sl@0
|
527 |
* "Unsafe" macro, assumes well-formed UTF-16.
|
|
sl@0
|
528 |
*
|
|
sl@0
|
529 |
* @param s const UChar * string
|
|
sl@0
|
530 |
* @param i string offset
|
|
sl@0
|
531 |
* @param n number of code points to skip
|
|
sl@0
|
532 |
* @see U16_BACK_N
|
|
sl@0
|
533 |
* @stable ICU 2.4
|
|
sl@0
|
534 |
*/
|
|
sl@0
|
535 |
#define U16_BACK_N_UNSAFE(s, i, n) { \
|
|
sl@0
|
536 |
int32_t __N=(n); \
|
|
sl@0
|
537 |
while(__N>0) { \
|
|
sl@0
|
538 |
U16_BACK_1_UNSAFE(s, i); \
|
|
sl@0
|
539 |
--__N; \
|
|
sl@0
|
540 |
} \
|
|
sl@0
|
541 |
}
|
|
sl@0
|
542 |
|
|
sl@0
|
543 |
/**
|
|
sl@0
|
544 |
* Move the string offset from one code point boundary to the n-th one before it,
|
|
sl@0
|
545 |
* i.e., move backward by n code points.
|
|
sl@0
|
546 |
* (Pre-decrementing backward iteration.)
|
|
sl@0
|
547 |
* The input offset may be the same as the string length.
|
|
sl@0
|
548 |
* "Safe" macro, handles unpaired surrogates and checks for string boundaries.
|
|
sl@0
|
549 |
*
|
|
sl@0
|
550 |
* @param s const UChar * string
|
|
sl@0
|
551 |
* @param start start of string
|
|
sl@0
|
552 |
* @param i string offset, i<length
|
|
sl@0
|
553 |
* @param n number of code points to skip
|
|
sl@0
|
554 |
* @see U16_BACK_N_UNSAFE
|
|
sl@0
|
555 |
* @stable ICU 2.4
|
|
sl@0
|
556 |
*/
|
|
sl@0
|
557 |
#define U16_BACK_N(s, start, i, n) { \
|
|
sl@0
|
558 |
int32_t __N=(n); \
|
|
sl@0
|
559 |
while(__N>0 && (i)>(start)) { \
|
|
sl@0
|
560 |
U16_BACK_1(s, start, i); \
|
|
sl@0
|
561 |
--__N; \
|
|
sl@0
|
562 |
} \
|
|
sl@0
|
563 |
}
|
|
sl@0
|
564 |
|
|
sl@0
|
565 |
/**
|
|
sl@0
|
566 |
* Adjust a random-access offset to a code point boundary after a code point.
|
|
sl@0
|
567 |
* If the offset is behind the lead surrogate of a surrogate pair,
|
|
sl@0
|
568 |
* then the offset is incremented.
|
|
sl@0
|
569 |
* Otherwise, it is not modified.
|
|
sl@0
|
570 |
* The input offset may be the same as the string length.
|
|
sl@0
|
571 |
* "Unsafe" macro, assumes well-formed UTF-16.
|
|
sl@0
|
572 |
*
|
|
sl@0
|
573 |
* @param s const UChar * string
|
|
sl@0
|
574 |
* @param i string offset
|
|
sl@0
|
575 |
* @see U16_SET_CP_LIMIT
|
|
sl@0
|
576 |
* @stable ICU 2.4
|
|
sl@0
|
577 |
*/
|
|
sl@0
|
578 |
#define U16_SET_CP_LIMIT_UNSAFE(s, i) { \
|
|
sl@0
|
579 |
if(U16_IS_LEAD((s)[(i)-1])) { \
|
|
sl@0
|
580 |
++(i); \
|
|
sl@0
|
581 |
} \
|
|
sl@0
|
582 |
}
|
|
sl@0
|
583 |
|
|
sl@0
|
584 |
/**
|
|
sl@0
|
585 |
* Adjust a random-access offset to a code point boundary after a code point.
|
|
sl@0
|
586 |
* If the offset is behind the lead surrogate of a surrogate pair,
|
|
sl@0
|
587 |
* then the offset is incremented.
|
|
sl@0
|
588 |
* Otherwise, it is not modified.
|
|
sl@0
|
589 |
* The input offset may be the same as the string length.
|
|
sl@0
|
590 |
* "Safe" macro, handles unpaired surrogates and checks for string boundaries.
|
|
sl@0
|
591 |
*
|
|
sl@0
|
592 |
* @param s const UChar * string
|
|
sl@0
|
593 |
* @param start starting string offset (usually 0)
|
|
sl@0
|
594 |
* @param i string offset, start<=i<=length
|
|
sl@0
|
595 |
* @param length string length
|
|
sl@0
|
596 |
* @see U16_SET_CP_LIMIT_UNSAFE
|
|
sl@0
|
597 |
* @stable ICU 2.4
|
|
sl@0
|
598 |
*/
|
|
sl@0
|
599 |
#define U16_SET_CP_LIMIT(s, start, i, length) { \
|
|
sl@0
|
600 |
if((start)<(i) && (i)<(length) && U16_IS_LEAD((s)[(i)-1]) && U16_IS_TRAIL((s)[i])) { \
|
|
sl@0
|
601 |
++(i); \
|
|
sl@0
|
602 |
} \
|
|
sl@0
|
603 |
}
|
|
sl@0
|
604 |
|
|
sl@0
|
605 |
#endif
|