1 | // © 2016 and later: Unicode, Inc. and others. |
2 | // License & terms of use: http://www.unicode.org/copyright.html |
3 | /* |
4 | ******************************************************************************* |
5 | * |
6 | * Copyright (C) 2002-2012, International Business Machines |
7 | * Corporation and others. All Rights Reserved. |
8 | * |
9 | ******************************************************************************* |
10 | * file name: utf_old.h |
11 | * encoding: UTF-8 |
12 | * tab size: 8 (not used) |
13 | * indentation:4 |
14 | * |
15 | * created on: 2002sep21 |
16 | * created by: Markus W. Scherer |
17 | */ |
18 | |
19 | /** |
20 | * \file |
21 | * \brief C API: Deprecated macros for Unicode string handling |
22 | */ |
23 | |
24 | /** |
25 | * |
26 | * The macros in utf_old.h are all deprecated and their use discouraged. |
27 | * Some of the design principles behind the set of UTF macros |
28 | * have changed or proved impractical. |
29 | * Almost all of the old "UTF macros" are at least renamed. |
30 | * If you are looking for a new equivalent to an old macro, please see the |
31 | * comment at the old one. |
32 | * |
33 | * Brief summary of reasons for deprecation: |
34 | * - Switch on UTF_SIZE (selection of UTF-8/16/32 default string processing) |
35 | * was impractical. |
36 | * - Switch on UTF_SAFE etc. (selection of unsafe/safe/strict default string processing) |
37 | * was of little use and impractical. |
38 | * - Whole classes of macros became obsolete outside of the UTF_SIZE/UTF_SAFE |
39 | * selection framework: UTF32_ macros (all trivial) |
40 | * and UTF_ default and intermediate macros (all aliases). |
41 | * - The selection framework also caused many macro aliases. |
42 | * - Change in Unicode standard: "irregular" sequences (3.0) became illegal (3.2). |
43 | * - Change of language in Unicode standard: |
44 | * Growing distinction between internal x-bit Unicode strings and external UTF-x |
45 | * forms, with the former more lenient. |
46 | * Suggests renaming of UTF16_ macros to U16_. |
47 | * - The prefix "UTF_" without a width number confused some users. |
48 | * - "Safe" append macros needed the addition of an error indicator output. |
49 | * - "Safe" UTF-8 macros used legitimate (if rarely used) code point values |
50 | * to indicate error conditions. |
51 | * - The use of the "_CHAR" infix for code point operations confused some users. |
52 | * |
53 | * More details: |
54 | * |
55 | * Until ICU 2.2, utf.h theoretically allowed to choose among UTF-8/16/32 |
56 | * for string processing, and among unsafe/safe/strict default macros for that. |
57 | * |
58 | * It proved nearly impossible to write non-trivial, high-performance code |
59 | * that is UTF-generic. |
60 | * Unsafe default macros would be dangerous for default string processing, |
61 | * and the main reason for the "strict" versions disappeared: |
62 | * Between Unicode 3.0 and 3.2 all "irregular" UTF-8 sequences became illegal. |
63 | * The only other conditions that "strict" checked for were non-characters, |
64 | * which are valid during processing. Only during text input/output should they |
65 | * be checked, and at that time other well-formedness checks may be |
66 | * necessary or useful as well. |
67 | * This can still be done by using U16_NEXT and U_IS_UNICODE_NONCHAR |
68 | * or U_IS_UNICODE_CHAR. |
69 | * |
70 | * The old UTF8_..._SAFE macros also used some normal Unicode code points |
71 | * to indicate malformed sequences. |
72 | * The new UTF8_ macros without suffix use negative values instead. |
73 | * |
74 | * The entire contents of utf32.h was moved here without replacement |
75 | * because all those macros were trivial and |
76 | * were meaningful only in the framework of choosing the UTF size. |
77 | * |
78 | * See Jitterbug 2150 and its discussion on the ICU mailing list |
79 | * in September 2002. |
80 | * |
81 | * <hr> |
82 | * |
83 | * <em>Obsolete part</em> of pre-ICU 2.4 utf.h file documentation: |
84 | * |
85 | * <p>The original concept for these files was for ICU to allow |
86 | * in principle to set which UTF (UTF-8/16/32) is used internally |
87 | * by defining UTF_SIZE to either 8, 16, or 32. utf.h would then define the UChar type |
88 | * accordingly. UTF-16 was the default.</p> |
89 | * |
90 | * <p>This concept has been abandoned. |
91 | * A lot of the ICU source code assumes UChar strings are in UTF-16. |
92 | * This is especially true for low-level code like |
93 | * conversion, normalization, and collation. |
94 | * The utf.h header enforces the default of UTF-16. |
95 | * The UTF-8 and UTF-32 macros remain for now for completeness and backward compatibility.</p> |
96 | * |
97 | * <p>Accordingly, utf.h defines UChar to be an unsigned 16-bit integer. If this matches wchar_t, then |
98 | * UChar is defined to be exactly wchar_t, otherwise uint16_t.</p> |
99 | * |
100 | * <p>UChar32 is defined to be a signed 32-bit integer (int32_t), large enough for a 21-bit |
101 | * Unicode code point (Unicode scalar value, 0..0x10ffff). |
102 | * Before ICU 2.4, the definition of UChar32 was similarly platform-dependent as |
103 | * the definition of UChar. For details see the documentation for UChar32 itself.</p> |
104 | * |
105 | * <p>utf.h also defines a number of C macros for handling single Unicode code points and |
106 | * for using UTF Unicode strings. It includes utf8.h, utf16.h, and utf32.h for the actual |
107 | * implementations of those macros and then aliases one set of them (for UTF-16) for general use. |
108 | * The UTF-specific macros have the UTF size in the macro name prefixes (UTF16_...), while |
109 | * the general alias macros always begin with UTF_...</p> |
110 | * |
111 | * <p>Many string operations can be done with or without error checking. |
112 | * Where such a distinction is useful, there are two versions of the macros, "unsafe" and "safe" |
113 | * ones with ..._UNSAFE and ..._SAFE suffixes. The unsafe macros are fast but may cause |
114 | * program failures if the strings are not well-formed. The safe macros have an additional, boolean |
115 | * parameter "strict". If strict is FALSE, then only illegal sequences are detected. |
116 | * Otherwise, irregular sequences and non-characters are detected as well (like single surrogates). |
117 | * Safe macros return special error code points for illegal/irregular sequences: |
118 | * Typically, U+ffff, or values that would result in a code unit sequence of the same length |
119 | * as the erroneous input sequence.<br> |
120 | * Note that _UNSAFE macros have fewer parameters: They do not have the strictness parameter, and |
121 | * they do not have start/length parameters for boundary checking.</p> |
122 | * |
123 | * <p>Here, the macros are aliased in two steps: |
124 | * In the first step, the UTF-specific macros with UTF16_ prefix and _UNSAFE and _SAFE suffixes are |
125 | * aliased according to the UTF_SIZE to macros with UTF_ prefix and the same suffixes and signatures. |
126 | * Then, in a second step, the default, general alias macros are set to use either the unsafe or |
127 | * the safe/not strict (default) or the safe/strict macro; |
128 | * these general macros do not have a strictness parameter.</p> |
129 | * |
130 | * <p>It is possible to change the default choice for the general alias macros to be unsafe, safe/not strict or safe/strict. |
131 | * The default is safe/not strict. It is not recommended to select the unsafe macros as the basis for |
132 | * Unicode string handling in ICU! To select this, define UTF_SAFE, UTF_STRICT, or UTF_UNSAFE.</p> |
133 | * |
134 | * <p>For general use, one should use the default, general macros with UTF_ prefix and no _SAFE/_UNSAFE suffix. |
135 | * Only in some cases it may be necessary to control the choice of macro directly and use a less generic alias. |
136 | * For example, if it can be assumed that a string is well-formed and the index will stay within the bounds, |
137 | * then the _UNSAFE version may be used. |
138 | * If a UTF-8 string is to be processed, then the macros with UTF8_ prefixes need to be used.</p> |
139 | * |
140 | * <hr> |
141 | * |
142 | * @deprecated ICU 2.4. Use the macros in utf.h, utf16.h, utf8.h instead. |
143 | */ |
144 | |
145 | #ifndef __UTF_OLD_H__ |
146 | #define __UTF_OLD_H__ |
147 | |
148 | /** |
149 | * \def U_HIDE_OBSOLETE_UTF_OLD_H |
150 | * |
151 | * Hides the obsolete definitions in unicode/utf_old.h. |
152 | * Recommended to be set to 1 at compile time to make sure |
153 | * the long-deprecated macros are no longer used. |
154 | * |
155 | * For reasons for the deprecation see the utf_old.h file comments. |
156 | * |
157 | * @internal |
158 | */ |
159 | #ifndef U_HIDE_OBSOLETE_UTF_OLD_H |
160 | # define U_HIDE_OBSOLETE_UTF_OLD_H 0 |
161 | #endif |
162 | |
163 | #if !defined(U_HIDE_DEPRECATED_API) && !U_HIDE_OBSOLETE_UTF_OLD_H |
164 | |
165 | #include "unicode/utf.h" |
166 | #include "unicode/utf8.h" |
167 | #include "unicode/utf16.h" |
168 | |
169 | /* Formerly utf.h, part 1 --------------------------------------------------- */ |
170 | |
171 | #ifdef U_USE_UTF_DEPRECATES |
172 | /** |
173 | * Unicode string and array offset and index type. |
174 | * ICU always counts Unicode code units (UChars) for |
175 | * string offsets, indexes, and lengths, not Unicode code points. |
176 | * |
177 | * @obsolete ICU 2.6. Use int32_t directly instead since this API will be removed in that release. |
178 | */ |
179 | typedef int32_t UTextOffset; |
180 | #endif |
181 | |
182 | /** Number of bits in a Unicode string code unit - ICU uses 16-bit Unicode. @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
183 | #define UTF_SIZE 16 |
184 | |
185 | /** |
186 | * The default choice for general Unicode string macros is to use the ..._SAFE macro implementations |
187 | * with strict=FALSE. |
188 | * |
189 | * @deprecated ICU 2.4. Obsolete, see utf_old.h. |
190 | */ |
191 | #define UTF_SAFE |
192 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
193 | #undef UTF_UNSAFE |
194 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
195 | #undef UTF_STRICT |
196 | |
197 | /** |
198 | * UTF8_ERROR_VALUE_1 and UTF8_ERROR_VALUE_2 are special error values for UTF-8, |
199 | * which need 1 or 2 bytes in UTF-8: |
200 | * \code |
201 | * U+0015 = NAK = Negative Acknowledge, C0 control character |
202 | * U+009f = highest C1 control character |
203 | * \endcode |
204 | * |
205 | * These are used by UTF8_..._SAFE macros so that they can return an error value |
206 | * that needs the same number of code units (bytes) as were seen by |
207 | * a macro. They should be tested with UTF_IS_ERROR() or UTF_IS_VALID(). |
208 | * |
209 | * @deprecated ICU 2.4. Obsolete, see utf_old.h. |
210 | */ |
211 | #define UTF8_ERROR_VALUE_1 0x15 |
212 | |
213 | /** |
214 | * See documentation on UTF8_ERROR_VALUE_1 for details. |
215 | * |
216 | * @deprecated ICU 2.4. Obsolete, see utf_old.h. |
217 | */ |
218 | #define UTF8_ERROR_VALUE_2 0x9f |
219 | |
220 | /** |
221 | * Error value for all UTFs. This code point value will be set by macros with error |
222 | * checking if an error is detected. |
223 | * |
224 | * @deprecated ICU 2.4. Obsolete, see utf_old.h. |
225 | */ |
226 | #define UTF_ERROR_VALUE 0xffff |
227 | |
228 | /** |
229 | * Is a given 32-bit code an error value |
230 | * as returned by one of the macros for any UTF? |
231 | * |
232 | * @deprecated ICU 2.4. Obsolete, see utf_old.h. |
233 | */ |
234 | #define UTF_IS_ERROR(c) \ |
235 | (((c)&0xfffe)==0xfffe || (c)==UTF8_ERROR_VALUE_1 || (c)==UTF8_ERROR_VALUE_2) |
236 | |
237 | /** |
238 | * This is a combined macro: Is c a valid Unicode value _and_ not an error code? |
239 | * |
240 | * @deprecated ICU 2.4. Obsolete, see utf_old.h. |
241 | */ |
242 | #define UTF_IS_VALID(c) \ |
243 | (UTF_IS_UNICODE_CHAR(c) && \ |
244 | (c)!=UTF8_ERROR_VALUE_1 && (c)!=UTF8_ERROR_VALUE_2) |
245 | |
246 | /** |
247 | * Is this code unit or code point a surrogate (U+d800..U+dfff)? |
248 | * @deprecated ICU 2.4. Renamed to U_IS_SURROGATE and U16_IS_SURROGATE, see utf_old.h. |
249 | */ |
250 | #define UTF_IS_SURROGATE(uchar) (((uchar)&0xfffff800)==0xd800) |
251 | |
252 | /** |
253 | * Is a given 32-bit code point a Unicode noncharacter? |
254 | * |
255 | * @deprecated ICU 2.4. Renamed to U_IS_UNICODE_NONCHAR, see utf_old.h. |
256 | */ |
257 | #define UTF_IS_UNICODE_NONCHAR(c) \ |
258 | ((c)>=0xfdd0 && \ |
259 | ((uint32_t)(c)<=0xfdef || ((c)&0xfffe)==0xfffe) && \ |
260 | (uint32_t)(c)<=0x10ffff) |
261 | |
262 | /** |
263 | * Is a given 32-bit value a Unicode code point value (0..U+10ffff) |
264 | * that can be assigned a character? |
265 | * |
266 | * Code points that are not characters include: |
267 | * - single surrogate code points (U+d800..U+dfff, 2048 code points) |
268 | * - the last two code points on each plane (U+__fffe and U+__ffff, 34 code points) |
269 | * - U+fdd0..U+fdef (new with Unicode 3.1, 32 code points) |
270 | * - the highest Unicode code point value is U+10ffff |
271 | * |
272 | * This means that all code points below U+d800 are character code points, |
273 | * and that boundary is tested first for performance. |
274 | * |
275 | * @deprecated ICU 2.4. Renamed to U_IS_UNICODE_CHAR, see utf_old.h. |
276 | */ |
277 | #define UTF_IS_UNICODE_CHAR(c) \ |
278 | ((uint32_t)(c)<0xd800 || \ |
279 | ((uint32_t)(c)>0xdfff && \ |
280 | (uint32_t)(c)<=0x10ffff && \ |
281 | !UTF_IS_UNICODE_NONCHAR(c))) |
282 | |
283 | /* Formerly utf8.h ---------------------------------------------------------- */ |
284 | |
285 | /** |
286 | * \var utf8_countTrailBytes |
287 | * Internal array with numbers of trail bytes for any given byte used in |
288 | * lead byte position. |
289 | * |
290 | * This is internal since it is not meant to be called directly by external clients; |
291 | * however it is called by public macros in this file and thus must remain stable, |
292 | * and should not be hidden when other internal functions are hidden (otherwise |
293 | * public macros would fail to compile). |
294 | * @internal |
295 | */ |
296 | #ifdef U_UTF8_IMPL |
297 | // No forward declaration if compiling utf_impl.cpp, which defines utf8_countTrailBytes. |
298 | #elif defined(U_STATIC_IMPLEMENTATION) || defined(U_COMMON_IMPLEMENTATION) |
299 | U_CFUNC const uint8_t utf8_countTrailBytes[]; |
300 | #else |
301 | U_CFUNC U_IMPORT const uint8_t utf8_countTrailBytes[]; /* U_IMPORT2? */ /*U_IMPORT*/ |
302 | #endif |
303 | |
304 | /** |
305 | * Count the trail bytes for a UTF-8 lead byte. |
306 | * @deprecated ICU 2.4. Renamed to U8_COUNT_TRAIL_BYTES, see utf_old.h. |
307 | */ |
308 | #define UTF8_COUNT_TRAIL_BYTES(leadByte) (utf8_countTrailBytes[(uint8_t)leadByte]) |
309 | |
310 | /** |
311 | * Mask a UTF-8 lead byte, leave only the lower bits that form part of the code point value. |
312 | * @deprecated ICU 2.4. Renamed to U8_MASK_LEAD_BYTE, see utf_old.h. |
313 | */ |
314 | #define UTF8_MASK_LEAD_BYTE(leadByte, countTrailBytes) ((leadByte)&=(1<<(6-(countTrailBytes)))-1) |
315 | |
316 | /** Is this this code point a single code unit (byte)? @deprecated ICU 2.4. Renamed to U8_IS_SINGLE, see utf_old.h. */ |
317 | #define UTF8_IS_SINGLE(uchar) (((uchar)&0x80)==0) |
318 | /** Is this this code unit the lead code unit (byte) of a code point? @deprecated ICU 2.4. Renamed to U8_IS_LEAD, see utf_old.h. */ |
319 | #define UTF8_IS_LEAD(uchar) ((uint8_t)((uchar)-0xc0)<0x3e) |
320 | /** Is this this code unit a trailing code unit (byte) of a code point? @deprecated ICU 2.4. Renamed to U8_IS_TRAIL, see utf_old.h. */ |
321 | #define UTF8_IS_TRAIL(uchar) (((uchar)&0xc0)==0x80) |
322 | |
323 | /** Does this scalar Unicode value need multiple code units for storage? @deprecated ICU 2.4. Use U8_LENGTH or test ((uint32_t)(c)>0x7f) instead, see utf_old.h. */ |
324 | #define UTF8_NEED_MULTIPLE_UCHAR(c) ((uint32_t)(c)>0x7f) |
325 | |
326 | /** |
327 | * Given the lead character, how many bytes are taken by this code point. |
328 | * ICU does not deal with code points >0x10ffff |
329 | * unless necessary for advancing in the byte stream. |
330 | * |
331 | * These length macros take into account that for values >0x10ffff |
332 | * the UTF8_APPEND_CHAR_SAFE macros would write the error code point 0xffff |
333 | * with 3 bytes. |
334 | * Code point comparisons need to be in uint32_t because UChar32 |
335 | * may be a signed type, and negative values must be recognized. |
336 | * |
337 | * @deprecated ICU 2.4. Use U8_LENGTH instead, see utf.h. |
338 | */ |
339 | #if 1 |
340 | # define UTF8_CHAR_LENGTH(c) \ |
341 | ((uint32_t)(c)<=0x7f ? 1 : \ |
342 | ((uint32_t)(c)<=0x7ff ? 2 : \ |
343 | ((uint32_t)((c)-0x10000)>0xfffff ? 3 : 4) \ |
344 | ) \ |
345 | ) |
346 | #else |
347 | # define UTF8_CHAR_LENGTH(c) \ |
348 | ((uint32_t)(c)<=0x7f ? 1 : \ |
349 | ((uint32_t)(c)<=0x7ff ? 2 : \ |
350 | ((uint32_t)(c)<=0xffff ? 3 : \ |
351 | ((uint32_t)(c)<=0x10ffff ? 4 : \ |
352 | ((uint32_t)(c)<=0x3ffffff ? 5 : \ |
353 | ((uint32_t)(c)<=0x7fffffff ? 6 : 3) \ |
354 | ) \ |
355 | ) \ |
356 | ) \ |
357 | ) \ |
358 | ) |
359 | #endif |
360 | |
361 | /** The maximum number of bytes per code point. @deprecated ICU 2.4. Renamed to U8_MAX_LENGTH, see utf_old.h. */ |
362 | #define UTF8_MAX_CHAR_LENGTH 4 |
363 | |
364 | /** Average number of code units compared to UTF-16. @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
365 | #define UTF8_ARRAY_SIZE(size) ((5*(size))/2) |
366 | |
367 | /** @deprecated ICU 2.4. Renamed to U8_GET_UNSAFE, see utf_old.h. */ |
368 | #define UTF8_GET_CHAR_UNSAFE(s, i, c) { \ |
369 | int32_t _utf8_get_char_unsafe_index=(int32_t)(i); \ |
370 | UTF8_SET_CHAR_START_UNSAFE(s, _utf8_get_char_unsafe_index); \ |
371 | UTF8_NEXT_CHAR_UNSAFE(s, _utf8_get_char_unsafe_index, c); \ |
372 | } |
373 | |
374 | /** @deprecated ICU 2.4. Use U8_GET instead, see utf_old.h. */ |
375 | #define UTF8_GET_CHAR_SAFE(s, start, i, length, c, strict) { \ |
376 | int32_t _utf8_get_char_safe_index=(int32_t)(i); \ |
377 | UTF8_SET_CHAR_START_SAFE(s, start, _utf8_get_char_safe_index); \ |
378 | UTF8_NEXT_CHAR_SAFE(s, _utf8_get_char_safe_index, length, c, strict); \ |
379 | } |
380 | |
381 | /** @deprecated ICU 2.4. Renamed to U8_NEXT_UNSAFE, see utf_old.h. */ |
382 | #define UTF8_NEXT_CHAR_UNSAFE(s, i, c) { \ |
383 | (c)=(s)[(i)++]; \ |
384 | if((uint8_t)((c)-0xc0)<0x35) { \ |
385 | uint8_t __count=UTF8_COUNT_TRAIL_BYTES(c); \ |
386 | UTF8_MASK_LEAD_BYTE(c, __count); \ |
387 | switch(__count) { \ |
388 | /* each following branch falls through to the next one */ \ |
389 | case 3: \ |
390 | (c)=((c)<<6)|((s)[(i)++]&0x3f); \ |
391 | case 2: \ |
392 | (c)=((c)<<6)|((s)[(i)++]&0x3f); \ |
393 | case 1: \ |
394 | (c)=((c)<<6)|((s)[(i)++]&0x3f); \ |
395 | /* no other branches to optimize switch() */ \ |
396 | break; \ |
397 | } \ |
398 | } \ |
399 | } |
400 | |
401 | /** @deprecated ICU 2.4. Renamed to U8_APPEND_UNSAFE, see utf_old.h. */ |
402 | #define UTF8_APPEND_CHAR_UNSAFE(s, i, c) { \ |
403 | if((uint32_t)(c)<=0x7f) { \ |
404 | (s)[(i)++]=(uint8_t)(c); \ |
405 | } else { \ |
406 | if((uint32_t)(c)<=0x7ff) { \ |
407 | (s)[(i)++]=(uint8_t)(((c)>>6)|0xc0); \ |
408 | } else { \ |
409 | if((uint32_t)(c)<=0xffff) { \ |
410 | (s)[(i)++]=(uint8_t)(((c)>>12)|0xe0); \ |
411 | } else { \ |
412 | (s)[(i)++]=(uint8_t)(((c)>>18)|0xf0); \ |
413 | (s)[(i)++]=(uint8_t)((((c)>>12)&0x3f)|0x80); \ |
414 | } \ |
415 | (s)[(i)++]=(uint8_t)((((c)>>6)&0x3f)|0x80); \ |
416 | } \ |
417 | (s)[(i)++]=(uint8_t)(((c)&0x3f)|0x80); \ |
418 | } \ |
419 | } |
420 | |
421 | /** @deprecated ICU 2.4. Renamed to U8_FWD_1_UNSAFE, see utf_old.h. */ |
422 | #define UTF8_FWD_1_UNSAFE(s, i) { \ |
423 | (i)+=1+UTF8_COUNT_TRAIL_BYTES((s)[i]); \ |
424 | } |
425 | |
426 | /** @deprecated ICU 2.4. Renamed to U8_FWD_N_UNSAFE, see utf_old.h. */ |
427 | #define UTF8_FWD_N_UNSAFE(s, i, n) { \ |
428 | int32_t __N=(n); \ |
429 | while(__N>0) { \ |
430 | UTF8_FWD_1_UNSAFE(s, i); \ |
431 | --__N; \ |
432 | } \ |
433 | } |
434 | |
435 | /** @deprecated ICU 2.4. Renamed to U8_SET_CP_START_UNSAFE, see utf_old.h. */ |
436 | #define UTF8_SET_CHAR_START_UNSAFE(s, i) { \ |
437 | while(UTF8_IS_TRAIL((s)[i])) { --(i); } \ |
438 | } |
439 | |
440 | /** @deprecated ICU 2.4. Use U8_NEXT instead, see utf_old.h. */ |
441 | #define UTF8_NEXT_CHAR_SAFE(s, i, length, c, strict) { \ |
442 | (c)=(s)[(i)++]; \ |
443 | if((c)>=0x80) { \ |
444 | if(UTF8_IS_LEAD(c)) { \ |
445 | (c)=utf8_nextCharSafeBody(s, &(i), (int32_t)(length), c, strict); \ |
446 | } else { \ |
447 | (c)=UTF8_ERROR_VALUE_1; \ |
448 | } \ |
449 | } \ |
450 | } |
451 | |
452 | /** @deprecated ICU 2.4. Use U8_APPEND instead, see utf_old.h. */ |
453 | #define UTF8_APPEND_CHAR_SAFE(s, i, length, c) { \ |
454 | if((uint32_t)(c)<=0x7f) { \ |
455 | (s)[(i)++]=(uint8_t)(c); \ |
456 | } else { \ |
457 | (i)=utf8_appendCharSafeBody(s, (int32_t)(i), (int32_t)(length), c, NULL); \ |
458 | } \ |
459 | } |
460 | |
461 | /** @deprecated ICU 2.4. Renamed to U8_FWD_1, see utf_old.h. */ |
462 | #define UTF8_FWD_1_SAFE(s, i, length) U8_FWD_1(s, i, length) |
463 | |
464 | /** @deprecated ICU 2.4. Renamed to U8_FWD_N, see utf_old.h. */ |
465 | #define UTF8_FWD_N_SAFE(s, i, length, n) U8_FWD_N(s, i, length, n) |
466 | |
467 | /** @deprecated ICU 2.4. Renamed to U8_SET_CP_START, see utf_old.h. */ |
468 | #define UTF8_SET_CHAR_START_SAFE(s, start, i) U8_SET_CP_START(s, start, i) |
469 | |
470 | /** @deprecated ICU 2.4. Renamed to U8_PREV_UNSAFE, see utf_old.h. */ |
471 | #define UTF8_PREV_CHAR_UNSAFE(s, i, c) { \ |
472 | (c)=(s)[--(i)]; \ |
473 | if(UTF8_IS_TRAIL(c)) { \ |
474 | uint8_t __b, __count=1, __shift=6; \ |
475 | \ |
476 | /* c is a trail byte */ \ |
477 | (c)&=0x3f; \ |
478 | for(;;) { \ |
479 | __b=(s)[--(i)]; \ |
480 | if(__b>=0xc0) { \ |
481 | UTF8_MASK_LEAD_BYTE(__b, __count); \ |
482 | (c)|=(UChar32)__b<<__shift; \ |
483 | break; \ |
484 | } else { \ |
485 | (c)|=(UChar32)(__b&0x3f)<<__shift; \ |
486 | ++__count; \ |
487 | __shift+=6; \ |
488 | } \ |
489 | } \ |
490 | } \ |
491 | } |
492 | |
493 | /** @deprecated ICU 2.4. Renamed to U8_BACK_1_UNSAFE, see utf_old.h. */ |
494 | #define UTF8_BACK_1_UNSAFE(s, i) { \ |
495 | while(UTF8_IS_TRAIL((s)[--(i)])) {} \ |
496 | } |
497 | |
498 | /** @deprecated ICU 2.4. Renamed to U8_BACK_N_UNSAFE, see utf_old.h. */ |
499 | #define UTF8_BACK_N_UNSAFE(s, i, n) { \ |
500 | int32_t __N=(n); \ |
501 | while(__N>0) { \ |
502 | UTF8_BACK_1_UNSAFE(s, i); \ |
503 | --__N; \ |
504 | } \ |
505 | } |
506 | |
507 | /** @deprecated ICU 2.4. Renamed to U8_SET_CP_LIMIT_UNSAFE, see utf_old.h. */ |
508 | #define UTF8_SET_CHAR_LIMIT_UNSAFE(s, i) { \ |
509 | UTF8_BACK_1_UNSAFE(s, i); \ |
510 | UTF8_FWD_1_UNSAFE(s, i); \ |
511 | } |
512 | |
513 | /** @deprecated ICU 2.4. Use U8_PREV instead, see utf_old.h. */ |
514 | #define UTF8_PREV_CHAR_SAFE(s, start, i, c, strict) { \ |
515 | (c)=(s)[--(i)]; \ |
516 | if((c)>=0x80) { \ |
517 | if((c)<=0xbf) { \ |
518 | (c)=utf8_prevCharSafeBody(s, start, &(i), c, strict); \ |
519 | } else { \ |
520 | (c)=UTF8_ERROR_VALUE_1; \ |
521 | } \ |
522 | } \ |
523 | } |
524 | |
525 | /** @deprecated ICU 2.4. Renamed to U8_BACK_1, see utf_old.h. */ |
526 | #define UTF8_BACK_1_SAFE(s, start, i) U8_BACK_1(s, start, i) |
527 | |
528 | /** @deprecated ICU 2.4. Renamed to U8_BACK_N, see utf_old.h. */ |
529 | #define UTF8_BACK_N_SAFE(s, start, i, n) U8_BACK_N(s, start, i, n) |
530 | |
531 | /** @deprecated ICU 2.4. Renamed to U8_SET_CP_LIMIT, see utf_old.h. */ |
532 | #define UTF8_SET_CHAR_LIMIT_SAFE(s, start, i, length) U8_SET_CP_LIMIT(s, start, i, length) |
533 | |
534 | /* Formerly utf16.h --------------------------------------------------------- */ |
535 | |
536 | /** Is uchar a first/lead surrogate? @deprecated ICU 2.4. Renamed to U_IS_LEAD and U16_IS_LEAD, see utf_old.h. */ |
537 | #define UTF_IS_FIRST_SURROGATE(uchar) (((uchar)&0xfffffc00)==0xd800) |
538 | |
539 | /** Is uchar a second/trail surrogate? @deprecated ICU 2.4. Renamed to U_IS_TRAIL and U16_IS_TRAIL, see utf_old.h. */ |
540 | #define UTF_IS_SECOND_SURROGATE(uchar) (((uchar)&0xfffffc00)==0xdc00) |
541 | |
542 | /** Assuming c is a surrogate, is it a first/lead surrogate? @deprecated ICU 2.4. Renamed to U_IS_SURROGATE_LEAD and U16_IS_SURROGATE_LEAD, see utf_old.h. */ |
543 | #define UTF_IS_SURROGATE_FIRST(c) (((c)&0x400)==0) |
544 | |
545 | /** Helper constant for UTF16_GET_PAIR_VALUE. @deprecated ICU 2.4. Renamed to U16_SURROGATE_OFFSET, see utf_old.h. */ |
546 | #define UTF_SURROGATE_OFFSET ((0xd800<<10UL)+0xdc00-0x10000) |
547 | |
548 | /** Get the UTF-32 value from the surrogate code units. @deprecated ICU 2.4. Renamed to U16_GET_SUPPLEMENTARY, see utf_old.h. */ |
549 | #define UTF16_GET_PAIR_VALUE(first, second) \ |
550 | (((first)<<10UL)+(second)-UTF_SURROGATE_OFFSET) |
551 | |
552 | /** @deprecated ICU 2.4. Renamed to U16_LEAD, see utf_old.h. */ |
553 | #define UTF_FIRST_SURROGATE(supplementary) (UChar)(((supplementary)>>10)+0xd7c0) |
554 | |
555 | /** @deprecated ICU 2.4. Renamed to U16_TRAIL, see utf_old.h. */ |
556 | #define UTF_SECOND_SURROGATE(supplementary) (UChar)(((supplementary)&0x3ff)|0xdc00) |
557 | |
558 | /** @deprecated ICU 2.4. Renamed to U16_LEAD, see utf_old.h. */ |
559 | #define UTF16_LEAD(supplementary) UTF_FIRST_SURROGATE(supplementary) |
560 | |
561 | /** @deprecated ICU 2.4. Renamed to U16_TRAIL, see utf_old.h. */ |
562 | #define UTF16_TRAIL(supplementary) UTF_SECOND_SURROGATE(supplementary) |
563 | |
564 | /** @deprecated ICU 2.4. Renamed to U16_IS_SINGLE, see utf_old.h. */ |
565 | #define UTF16_IS_SINGLE(uchar) !UTF_IS_SURROGATE(uchar) |
566 | |
567 | /** @deprecated ICU 2.4. Renamed to U16_IS_LEAD, see utf_old.h. */ |
568 | #define UTF16_IS_LEAD(uchar) UTF_IS_FIRST_SURROGATE(uchar) |
569 | |
570 | /** @deprecated ICU 2.4. Renamed to U16_IS_TRAIL, see utf_old.h. */ |
571 | #define UTF16_IS_TRAIL(uchar) UTF_IS_SECOND_SURROGATE(uchar) |
572 | |
573 | /** Does this scalar Unicode value need multiple code units for storage? @deprecated ICU 2.4. Use U16_LENGTH or test ((uint32_t)(c)>0xffff) instead, see utf_old.h. */ |
574 | #define UTF16_NEED_MULTIPLE_UCHAR(c) ((uint32_t)(c)>0xffff) |
575 | |
576 | /** @deprecated ICU 2.4. Renamed to U16_LENGTH, see utf_old.h. */ |
577 | #define UTF16_CHAR_LENGTH(c) ((uint32_t)(c)<=0xffff ? 1 : 2) |
578 | |
579 | /** @deprecated ICU 2.4. Renamed to U16_MAX_LENGTH, see utf_old.h. */ |
580 | #define UTF16_MAX_CHAR_LENGTH 2 |
581 | |
582 | /** Average number of code units compared to UTF-16. @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
583 | #define UTF16_ARRAY_SIZE(size) (size) |
584 | |
585 | /** |
586 | * Get a single code point from an offset that points to any |
587 | * of the code units that belong to that code point. |
588 | * Assume 0<=i<length. |
589 | * |
590 | * This could be used for iteration together with |
591 | * UTF16_CHAR_LENGTH() and UTF_IS_ERROR(), |
592 | * but the use of UTF16_NEXT_CHAR[_UNSAFE]() and |
593 | * UTF16_PREV_CHAR[_UNSAFE]() is more efficient for that. |
594 | * @deprecated ICU 2.4. Renamed to U16_GET_UNSAFE, see utf_old.h. |
595 | */ |
596 | #define UTF16_GET_CHAR_UNSAFE(s, i, c) { \ |
597 | (c)=(s)[i]; \ |
598 | if(UTF_IS_SURROGATE(c)) { \ |
599 | if(UTF_IS_SURROGATE_FIRST(c)) { \ |
600 | (c)=UTF16_GET_PAIR_VALUE((c), (s)[(i)+1]); \ |
601 | } else { \ |
602 | (c)=UTF16_GET_PAIR_VALUE((s)[(i)-1], (c)); \ |
603 | } \ |
604 | } \ |
605 | } |
606 | |
607 | /** @deprecated ICU 2.4. Use U16_GET instead, see utf_old.h. */ |
608 | #define UTF16_GET_CHAR_SAFE(s, start, i, length, c, strict) { \ |
609 | (c)=(s)[i]; \ |
610 | if(UTF_IS_SURROGATE(c)) { \ |
611 | uint16_t __c2; \ |
612 | if(UTF_IS_SURROGATE_FIRST(c)) { \ |
613 | if((i)+1<(length) && UTF_IS_SECOND_SURROGATE(__c2=(s)[(i)+1])) { \ |
614 | (c)=UTF16_GET_PAIR_VALUE((c), __c2); \ |
615 | /* strict: ((c)&0xfffe)==0xfffe is caught by UTF_IS_ERROR() and UTF_IS_UNICODE_CHAR() */ \ |
616 | } else if(strict) {\ |
617 | /* unmatched first surrogate */ \ |
618 | (c)=UTF_ERROR_VALUE; \ |
619 | } \ |
620 | } else { \ |
621 | if((i)-1>=(start) && UTF_IS_FIRST_SURROGATE(__c2=(s)[(i)-1])) { \ |
622 | (c)=UTF16_GET_PAIR_VALUE(__c2, (c)); \ |
623 | /* strict: ((c)&0xfffe)==0xfffe is caught by UTF_IS_ERROR() and UTF_IS_UNICODE_CHAR() */ \ |
624 | } else if(strict) {\ |
625 | /* unmatched second surrogate */ \ |
626 | (c)=UTF_ERROR_VALUE; \ |
627 | } \ |
628 | } \ |
629 | } else if((strict) && !UTF_IS_UNICODE_CHAR(c)) { \ |
630 | (c)=UTF_ERROR_VALUE; \ |
631 | } \ |
632 | } |
633 | |
634 | /** @deprecated ICU 2.4. Renamed to U16_NEXT_UNSAFE, see utf_old.h. */ |
635 | #define UTF16_NEXT_CHAR_UNSAFE(s, i, c) { \ |
636 | (c)=(s)[(i)++]; \ |
637 | if(UTF_IS_FIRST_SURROGATE(c)) { \ |
638 | (c)=UTF16_GET_PAIR_VALUE((c), (s)[(i)++]); \ |
639 | } \ |
640 | } |
641 | |
642 | /** @deprecated ICU 2.4. Renamed to U16_APPEND_UNSAFE, see utf_old.h. */ |
643 | #define UTF16_APPEND_CHAR_UNSAFE(s, i, c) { \ |
644 | if((uint32_t)(c)<=0xffff) { \ |
645 | (s)[(i)++]=(uint16_t)(c); \ |
646 | } else { \ |
647 | (s)[(i)++]=(uint16_t)(((c)>>10)+0xd7c0); \ |
648 | (s)[(i)++]=(uint16_t)(((c)&0x3ff)|0xdc00); \ |
649 | } \ |
650 | } |
651 | |
652 | /** @deprecated ICU 2.4. Renamed to U16_FWD_1_UNSAFE, see utf_old.h. */ |
653 | #define UTF16_FWD_1_UNSAFE(s, i) { \ |
654 | if(UTF_IS_FIRST_SURROGATE((s)[(i)++])) { \ |
655 | ++(i); \ |
656 | } \ |
657 | } |
658 | |
659 | /** @deprecated ICU 2.4. Renamed to U16_FWD_N_UNSAFE, see utf_old.h. */ |
660 | #define UTF16_FWD_N_UNSAFE(s, i, n) { \ |
661 | int32_t __N=(n); \ |
662 | while(__N>0) { \ |
663 | UTF16_FWD_1_UNSAFE(s, i); \ |
664 | --__N; \ |
665 | } \ |
666 | } |
667 | |
668 | /** @deprecated ICU 2.4. Renamed to U16_SET_CP_START_UNSAFE, see utf_old.h. */ |
669 | #define UTF16_SET_CHAR_START_UNSAFE(s, i) { \ |
670 | if(UTF_IS_SECOND_SURROGATE((s)[i])) { \ |
671 | --(i); \ |
672 | } \ |
673 | } |
674 | |
675 | /** @deprecated ICU 2.4. Use U16_NEXT instead, see utf_old.h. */ |
676 | #define UTF16_NEXT_CHAR_SAFE(s, i, length, c, strict) { \ |
677 | (c)=(s)[(i)++]; \ |
678 | if(UTF_IS_FIRST_SURROGATE(c)) { \ |
679 | uint16_t __c2; \ |
680 | if((i)<(length) && UTF_IS_SECOND_SURROGATE(__c2=(s)[(i)])) { \ |
681 | ++(i); \ |
682 | (c)=UTF16_GET_PAIR_VALUE((c), __c2); \ |
683 | /* strict: ((c)&0xfffe)==0xfffe is caught by UTF_IS_ERROR() and UTF_IS_UNICODE_CHAR() */ \ |
684 | } else if(strict) {\ |
685 | /* unmatched first surrogate */ \ |
686 | (c)=UTF_ERROR_VALUE; \ |
687 | } \ |
688 | } else if((strict) && !UTF_IS_UNICODE_CHAR(c)) { \ |
689 | /* unmatched second surrogate or other non-character */ \ |
690 | (c)=UTF_ERROR_VALUE; \ |
691 | } \ |
692 | } |
693 | |
694 | /** @deprecated ICU 2.4. Use U16_APPEND instead, see utf_old.h. */ |
695 | #define UTF16_APPEND_CHAR_SAFE(s, i, length, c) { \ |
696 | if((uint32_t)(c)<=0xffff) { \ |
697 | (s)[(i)++]=(uint16_t)(c); \ |
698 | } else if((uint32_t)(c)<=0x10ffff) { \ |
699 | if((i)+1<(length)) { \ |
700 | (s)[(i)++]=(uint16_t)(((c)>>10)+0xd7c0); \ |
701 | (s)[(i)++]=(uint16_t)(((c)&0x3ff)|0xdc00); \ |
702 | } else /* not enough space */ { \ |
703 | (s)[(i)++]=UTF_ERROR_VALUE; \ |
704 | } \ |
705 | } else /* c>0x10ffff, write error value */ { \ |
706 | (s)[(i)++]=UTF_ERROR_VALUE; \ |
707 | } \ |
708 | } |
709 | |
710 | /** @deprecated ICU 2.4. Renamed to U16_FWD_1, see utf_old.h. */ |
711 | #define UTF16_FWD_1_SAFE(s, i, length) U16_FWD_1(s, i, length) |
712 | |
713 | /** @deprecated ICU 2.4. Renamed to U16_FWD_N, see utf_old.h. */ |
714 | #define UTF16_FWD_N_SAFE(s, i, length, n) U16_FWD_N(s, i, length, n) |
715 | |
716 | /** @deprecated ICU 2.4. Renamed to U16_SET_CP_START, see utf_old.h. */ |
717 | #define UTF16_SET_CHAR_START_SAFE(s, start, i) U16_SET_CP_START(s, start, i) |
718 | |
719 | /** @deprecated ICU 2.4. Renamed to U16_PREV_UNSAFE, see utf_old.h. */ |
720 | #define UTF16_PREV_CHAR_UNSAFE(s, i, c) { \ |
721 | (c)=(s)[--(i)]; \ |
722 | if(UTF_IS_SECOND_SURROGATE(c)) { \ |
723 | (c)=UTF16_GET_PAIR_VALUE((s)[--(i)], (c)); \ |
724 | } \ |
725 | } |
726 | |
727 | /** @deprecated ICU 2.4. Renamed to U16_BACK_1_UNSAFE, see utf_old.h. */ |
728 | #define UTF16_BACK_1_UNSAFE(s, i) { \ |
729 | if(UTF_IS_SECOND_SURROGATE((s)[--(i)])) { \ |
730 | --(i); \ |
731 | } \ |
732 | } |
733 | |
734 | /** @deprecated ICU 2.4. Renamed to U16_BACK_N_UNSAFE, see utf_old.h. */ |
735 | #define UTF16_BACK_N_UNSAFE(s, i, n) { \ |
736 | int32_t __N=(n); \ |
737 | while(__N>0) { \ |
738 | UTF16_BACK_1_UNSAFE(s, i); \ |
739 | --__N; \ |
740 | } \ |
741 | } |
742 | |
743 | /** @deprecated ICU 2.4. Renamed to U16_SET_CP_LIMIT_UNSAFE, see utf_old.h. */ |
744 | #define UTF16_SET_CHAR_LIMIT_UNSAFE(s, i) { \ |
745 | if(UTF_IS_FIRST_SURROGATE((s)[(i)-1])) { \ |
746 | ++(i); \ |
747 | } \ |
748 | } |
749 | |
750 | /** @deprecated ICU 2.4. Use U16_PREV instead, see utf_old.h. */ |
751 | #define UTF16_PREV_CHAR_SAFE(s, start, i, c, strict) { \ |
752 | (c)=(s)[--(i)]; \ |
753 | if(UTF_IS_SECOND_SURROGATE(c)) { \ |
754 | uint16_t __c2; \ |
755 | if((i)>(start) && UTF_IS_FIRST_SURROGATE(__c2=(s)[(i)-1])) { \ |
756 | --(i); \ |
757 | (c)=UTF16_GET_PAIR_VALUE(__c2, (c)); \ |
758 | /* strict: ((c)&0xfffe)==0xfffe is caught by UTF_IS_ERROR() and UTF_IS_UNICODE_CHAR() */ \ |
759 | } else if(strict) {\ |
760 | /* unmatched second surrogate */ \ |
761 | (c)=UTF_ERROR_VALUE; \ |
762 | } \ |
763 | } else if((strict) && !UTF_IS_UNICODE_CHAR(c)) { \ |
764 | /* unmatched first surrogate or other non-character */ \ |
765 | (c)=UTF_ERROR_VALUE; \ |
766 | } \ |
767 | } |
768 | |
769 | /** @deprecated ICU 2.4. Renamed to U16_BACK_1, see utf_old.h. */ |
770 | #define UTF16_BACK_1_SAFE(s, start, i) U16_BACK_1(s, start, i) |
771 | |
772 | /** @deprecated ICU 2.4. Renamed to U16_BACK_N, see utf_old.h. */ |
773 | #define UTF16_BACK_N_SAFE(s, start, i, n) U16_BACK_N(s, start, i, n) |
774 | |
775 | /** @deprecated ICU 2.4. Renamed to U16_SET_CP_LIMIT, see utf_old.h. */ |
776 | #define UTF16_SET_CHAR_LIMIT_SAFE(s, start, i, length) U16_SET_CP_LIMIT(s, start, i, length) |
777 | |
778 | /* Formerly utf32.h --------------------------------------------------------- */ |
779 | |
780 | /* |
781 | * Old documentation: |
782 | * |
783 | * This file defines macros to deal with UTF-32 code units and code points. |
784 | * Signatures and semantics are the same as for the similarly named macros |
785 | * in utf16.h. |
786 | * utf32.h is included by utf.h after unicode/umachine.h</p> |
787 | * and some common definitions. |
788 | * <p><b>Usage:</b> ICU coding guidelines for if() statements should be followed when using these macros. |
789 | * Compound statements (curly braces {}) must be used for if-else-while... |
790 | * bodies and all macro statements should be terminated with semicolon.</p> |
791 | */ |
792 | |
793 | /* internal definitions ----------------------------------------------------- */ |
794 | |
795 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
796 | #define UTF32_IS_SAFE(c, strict) \ |
797 | (!(strict) ? \ |
798 | (uint32_t)(c)<=0x10ffff : \ |
799 | UTF_IS_UNICODE_CHAR(c)) |
800 | |
801 | /* |
802 | * For the semantics of all of these macros, see utf16.h. |
803 | * The UTF-32 versions are trivial because any code point is |
804 | * encoded using exactly one code unit. |
805 | */ |
806 | |
807 | /* single-code point definitions -------------------------------------------- */ |
808 | |
809 | /* classes of code unit values */ |
810 | |
811 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
812 | #define UTF32_IS_SINGLE(uchar) 1 |
813 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
814 | #define UTF32_IS_LEAD(uchar) 0 |
815 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
816 | #define UTF32_IS_TRAIL(uchar) 0 |
817 | |
818 | /* number of code units per code point */ |
819 | |
820 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
821 | #define UTF32_NEED_MULTIPLE_UCHAR(c) 0 |
822 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
823 | #define UTF32_CHAR_LENGTH(c) 1 |
824 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
825 | #define UTF32_MAX_CHAR_LENGTH 1 |
826 | |
827 | /* average number of code units compared to UTF-16 */ |
828 | |
829 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
830 | #define UTF32_ARRAY_SIZE(size) (size) |
831 | |
832 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
833 | #define UTF32_GET_CHAR_UNSAFE(s, i, c) { \ |
834 | (c)=(s)[i]; \ |
835 | } |
836 | |
837 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
838 | #define UTF32_GET_CHAR_SAFE(s, start, i, length, c, strict) { \ |
839 | (c)=(s)[i]; \ |
840 | if(!UTF32_IS_SAFE(c, strict)) { \ |
841 | (c)=UTF_ERROR_VALUE; \ |
842 | } \ |
843 | } |
844 | |
845 | /* definitions with forward iteration --------------------------------------- */ |
846 | |
847 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
848 | #define UTF32_NEXT_CHAR_UNSAFE(s, i, c) { \ |
849 | (c)=(s)[(i)++]; \ |
850 | } |
851 | |
852 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
853 | #define UTF32_APPEND_CHAR_UNSAFE(s, i, c) { \ |
854 | (s)[(i)++]=(c); \ |
855 | } |
856 | |
857 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
858 | #define UTF32_FWD_1_UNSAFE(s, i) { \ |
859 | ++(i); \ |
860 | } |
861 | |
862 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
863 | #define UTF32_FWD_N_UNSAFE(s, i, n) { \ |
864 | (i)+=(n); \ |
865 | } |
866 | |
867 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
868 | #define UTF32_SET_CHAR_START_UNSAFE(s, i) { \ |
869 | } |
870 | |
871 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
872 | #define UTF32_NEXT_CHAR_SAFE(s, i, length, c, strict) { \ |
873 | (c)=(s)[(i)++]; \ |
874 | if(!UTF32_IS_SAFE(c, strict)) { \ |
875 | (c)=UTF_ERROR_VALUE; \ |
876 | } \ |
877 | } |
878 | |
879 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
880 | #define UTF32_APPEND_CHAR_SAFE(s, i, length, c) { \ |
881 | if((uint32_t)(c)<=0x10ffff) { \ |
882 | (s)[(i)++]=(c); \ |
883 | } else /* c>0x10ffff, write 0xfffd */ { \ |
884 | (s)[(i)++]=0xfffd; \ |
885 | } \ |
886 | } |
887 | |
888 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
889 | #define UTF32_FWD_1_SAFE(s, i, length) { \ |
890 | ++(i); \ |
891 | } |
892 | |
893 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
894 | #define UTF32_FWD_N_SAFE(s, i, length, n) { \ |
895 | if(((i)+=(n))>(length)) { \ |
896 | (i)=(length); \ |
897 | } \ |
898 | } |
899 | |
900 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
901 | #define UTF32_SET_CHAR_START_SAFE(s, start, i) { \ |
902 | } |
903 | |
904 | /* definitions with backward iteration -------------------------------------- */ |
905 | |
906 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
907 | #define UTF32_PREV_CHAR_UNSAFE(s, i, c) { \ |
908 | (c)=(s)[--(i)]; \ |
909 | } |
910 | |
911 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
912 | #define UTF32_BACK_1_UNSAFE(s, i) { \ |
913 | --(i); \ |
914 | } |
915 | |
916 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
917 | #define UTF32_BACK_N_UNSAFE(s, i, n) { \ |
918 | (i)-=(n); \ |
919 | } |
920 | |
921 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
922 | #define UTF32_SET_CHAR_LIMIT_UNSAFE(s, i) { \ |
923 | } |
924 | |
925 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
926 | #define UTF32_PREV_CHAR_SAFE(s, start, i, c, strict) { \ |
927 | (c)=(s)[--(i)]; \ |
928 | if(!UTF32_IS_SAFE(c, strict)) { \ |
929 | (c)=UTF_ERROR_VALUE; \ |
930 | } \ |
931 | } |
932 | |
933 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
934 | #define UTF32_BACK_1_SAFE(s, start, i) { \ |
935 | --(i); \ |
936 | } |
937 | |
938 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
939 | #define UTF32_BACK_N_SAFE(s, start, i, n) { \ |
940 | (i)-=(n); \ |
941 | if((i)<(start)) { \ |
942 | (i)=(start); \ |
943 | } \ |
944 | } |
945 | |
946 | /** @deprecated ICU 2.4. Obsolete, see utf_old.h. */ |
947 | #define UTF32_SET_CHAR_LIMIT_SAFE(s, i, length) { \ |
948 | } |
949 | |
950 | /* Formerly utf.h, part 2 --------------------------------------------------- */ |
951 | |
952 | /** |
953 | * Estimate the number of code units for a string based on the number of UTF-16 code units. |
954 | * |
955 | * @deprecated ICU 2.4. Obsolete, see utf_old.h. |
956 | */ |
957 | #define UTF_ARRAY_SIZE(size) UTF16_ARRAY_SIZE(size) |
958 | |
959 | /** @deprecated ICU 2.4. Renamed to U16_GET_UNSAFE, see utf_old.h. */ |
960 | #define UTF_GET_CHAR_UNSAFE(s, i, c) UTF16_GET_CHAR_UNSAFE(s, i, c) |
961 | |
962 | /** @deprecated ICU 2.4. Use U16_GET instead, see utf_old.h. */ |
963 | #define UTF_GET_CHAR_SAFE(s, start, i, length, c, strict) UTF16_GET_CHAR_SAFE(s, start, i, length, c, strict) |
964 | |
965 | |
966 | /** @deprecated ICU 2.4. Renamed to U16_NEXT_UNSAFE, see utf_old.h. */ |
967 | #define UTF_NEXT_CHAR_UNSAFE(s, i, c) UTF16_NEXT_CHAR_UNSAFE(s, i, c) |
968 | |
969 | /** @deprecated ICU 2.4. Use U16_NEXT instead, see utf_old.h. */ |
970 | #define UTF_NEXT_CHAR_SAFE(s, i, length, c, strict) UTF16_NEXT_CHAR_SAFE(s, i, length, c, strict) |
971 | |
972 | |
973 | /** @deprecated ICU 2.4. Renamed to U16_APPEND_UNSAFE, see utf_old.h. */ |
974 | #define UTF_APPEND_CHAR_UNSAFE(s, i, c) UTF16_APPEND_CHAR_UNSAFE(s, i, c) |
975 | |
976 | /** @deprecated ICU 2.4. Use U16_APPEND instead, see utf_old.h. */ |
977 | #define UTF_APPEND_CHAR_SAFE(s, i, length, c) UTF16_APPEND_CHAR_SAFE(s, i, length, c) |
978 | |
979 | |
980 | /** @deprecated ICU 2.4. Renamed to U16_FWD_1_UNSAFE, see utf_old.h. */ |
981 | #define UTF_FWD_1_UNSAFE(s, i) UTF16_FWD_1_UNSAFE(s, i) |
982 | |
983 | /** @deprecated ICU 2.4. Renamed to U16_FWD_1, see utf_old.h. */ |
984 | #define UTF_FWD_1_SAFE(s, i, length) UTF16_FWD_1_SAFE(s, i, length) |
985 | |
986 | |
987 | /** @deprecated ICU 2.4. Renamed to U16_FWD_N_UNSAFE, see utf_old.h. */ |
988 | #define UTF_FWD_N_UNSAFE(s, i, n) UTF16_FWD_N_UNSAFE(s, i, n) |
989 | |
990 | /** @deprecated ICU 2.4. Renamed to U16_FWD_N, see utf_old.h. */ |
991 | #define UTF_FWD_N_SAFE(s, i, length, n) UTF16_FWD_N_SAFE(s, i, length, n) |
992 | |
993 | |
994 | /** @deprecated ICU 2.4. Renamed to U16_SET_CP_START_UNSAFE, see utf_old.h. */ |
995 | #define UTF_SET_CHAR_START_UNSAFE(s, i) UTF16_SET_CHAR_START_UNSAFE(s, i) |
996 | |
997 | /** @deprecated ICU 2.4. Renamed to U16_SET_CP_START, see utf_old.h. */ |
998 | #define UTF_SET_CHAR_START_SAFE(s, start, i) UTF16_SET_CHAR_START_SAFE(s, start, i) |
999 | |
1000 | |
1001 | /** @deprecated ICU 2.4. Renamed to U16_PREV_UNSAFE, see utf_old.h. */ |
1002 | #define UTF_PREV_CHAR_UNSAFE(s, i, c) UTF16_PREV_CHAR_UNSAFE(s, i, c) |
1003 | |
1004 | /** @deprecated ICU 2.4. Use U16_PREV instead, see utf_old.h. */ |
1005 | #define UTF_PREV_CHAR_SAFE(s, start, i, c, strict) UTF16_PREV_CHAR_SAFE(s, start, i, c, strict) |
1006 | |
1007 | |
1008 | /** @deprecated ICU 2.4. Renamed to U16_BACK_1_UNSAFE, see utf_old.h. */ |
1009 | #define UTF_BACK_1_UNSAFE(s, i) UTF16_BACK_1_UNSAFE(s, i) |
1010 | |
1011 | /** @deprecated ICU 2.4. Renamed to U16_BACK_1, see utf_old.h. */ |
1012 | #define UTF_BACK_1_SAFE(s, start, i) UTF16_BACK_1_SAFE(s, start, i) |
1013 | |
1014 | |
1015 | /** @deprecated ICU 2.4. Renamed to U16_BACK_N_UNSAFE, see utf_old.h. */ |
1016 | #define UTF_BACK_N_UNSAFE(s, i, n) UTF16_BACK_N_UNSAFE(s, i, n) |
1017 | |
1018 | /** @deprecated ICU 2.4. Renamed to U16_BACK_N, see utf_old.h. */ |
1019 | #define UTF_BACK_N_SAFE(s, start, i, n) UTF16_BACK_N_SAFE(s, start, i, n) |
1020 | |
1021 | |
1022 | /** @deprecated ICU 2.4. Renamed to U16_SET_CP_LIMIT_UNSAFE, see utf_old.h. */ |
1023 | #define UTF_SET_CHAR_LIMIT_UNSAFE(s, i) UTF16_SET_CHAR_LIMIT_UNSAFE(s, i) |
1024 | |
1025 | /** @deprecated ICU 2.4. Renamed to U16_SET_CP_LIMIT, see utf_old.h. */ |
1026 | #define UTF_SET_CHAR_LIMIT_SAFE(s, start, i, length) UTF16_SET_CHAR_LIMIT_SAFE(s, start, i, length) |
1027 | |
1028 | /* Define default macros (UTF-16 "safe") ------------------------------------ */ |
1029 | |
1030 | /** |
1031 | * Does this code unit alone encode a code point (BMP, not a surrogate)? |
1032 | * Same as UTF16_IS_SINGLE. |
1033 | * @deprecated ICU 2.4. Renamed to U_IS_SINGLE and U16_IS_SINGLE, see utf_old.h. |
1034 | */ |
1035 | #define UTF_IS_SINGLE(uchar) U16_IS_SINGLE(uchar) |
1036 | |
1037 | /** |
1038 | * Is this code unit the first one of several (a lead surrogate)? |
1039 | * Same as UTF16_IS_LEAD. |
1040 | * @deprecated ICU 2.4. Renamed to U_IS_LEAD and U16_IS_LEAD, see utf_old.h. |
1041 | */ |
1042 | #define UTF_IS_LEAD(uchar) U16_IS_LEAD(uchar) |
1043 | |
1044 | /** |
1045 | * Is this code unit one of several but not the first one (a trail surrogate)? |
1046 | * Same as UTF16_IS_TRAIL. |
1047 | * @deprecated ICU 2.4. Renamed to U_IS_TRAIL and U16_IS_TRAIL, see utf_old.h. |
1048 | */ |
1049 | #define UTF_IS_TRAIL(uchar) U16_IS_TRAIL(uchar) |
1050 | |
1051 | /** |
1052 | * Does this code point require multiple code units (is it a supplementary code point)? |
1053 | * Same as UTF16_NEED_MULTIPLE_UCHAR. |
1054 | * @deprecated ICU 2.4. Use U16_LENGTH or test ((uint32_t)(c)>0xffff) instead. |
1055 | */ |
1056 | #define UTF_NEED_MULTIPLE_UCHAR(c) UTF16_NEED_MULTIPLE_UCHAR(c) |
1057 | |
1058 | /** |
1059 | * How many code units are used to encode this code point (1 or 2)? |
1060 | * Same as UTF16_CHAR_LENGTH. |
1061 | * @deprecated ICU 2.4. Renamed to U16_LENGTH, see utf_old.h. |
1062 | */ |
1063 | #define UTF_CHAR_LENGTH(c) U16_LENGTH(c) |
1064 | |
1065 | /** |
1066 | * How many code units are used at most for any Unicode code point (2)? |
1067 | * Same as UTF16_MAX_CHAR_LENGTH. |
1068 | * @deprecated ICU 2.4. Renamed to U16_MAX_LENGTH, see utf_old.h. |
1069 | */ |
1070 | #define UTF_MAX_CHAR_LENGTH U16_MAX_LENGTH |
1071 | |
1072 | /** |
1073 | * Set c to the code point that contains the code unit i. |
1074 | * i could point to the lead or the trail surrogate for the code point. |
1075 | * i is not modified. |
1076 | * Same as UTF16_GET_CHAR. |
1077 | * \pre 0<=i<length |
1078 | * |
1079 | * @deprecated ICU 2.4. Renamed to U16_GET, see utf_old.h. |
1080 | */ |
1081 | #define UTF_GET_CHAR(s, start, i, length, c) U16_GET(s, start, i, length, c) |
1082 | |
1083 | /** |
1084 | * Set c to the code point that starts at code unit i |
1085 | * and advance i to beyond the code units of this code point (post-increment). |
1086 | * i must point to the first code unit of a code point. |
1087 | * Otherwise c is set to the trail unit (surrogate) itself. |
1088 | * Same as UTF16_NEXT_CHAR. |
1089 | * \pre 0<=i<length |
1090 | * \post 0<i<=length |
1091 | * |
1092 | * @deprecated ICU 2.4. Renamed to U16_NEXT, see utf_old.h. |
1093 | */ |
1094 | #define UTF_NEXT_CHAR(s, i, length, c) U16_NEXT(s, i, length, c) |
1095 | |
1096 | /** |
1097 | * Append the code units of code point c to the string at index i |
1098 | * and advance i to beyond the new code units (post-increment). |
1099 | * The code units beginning at index i will be overwritten. |
1100 | * Same as UTF16_APPEND_CHAR. |
1101 | * \pre 0<=c<=0x10ffff |
1102 | * \pre 0<=i<length |
1103 | * \post 0<i<=length |
1104 | * |
1105 | * @deprecated ICU 2.4. Use U16_APPEND instead, see utf_old.h. |
1106 | */ |
1107 | #define UTF_APPEND_CHAR(s, i, length, c) UTF16_APPEND_CHAR_SAFE(s, i, length, c) |
1108 | |
1109 | /** |
1110 | * Advance i to beyond the code units of the code point that begins at i. |
1111 | * I.e., advance i by one code point. |
1112 | * Same as UTF16_FWD_1. |
1113 | * \pre 0<=i<length |
1114 | * \post 0<i<=length |
1115 | * |
1116 | * @deprecated ICU 2.4. Renamed to U16_FWD_1, see utf_old.h. |
1117 | */ |
1118 | #define UTF_FWD_1(s, i, length) U16_FWD_1(s, i, length) |
1119 | |
1120 | /** |
1121 | * Advance i to beyond the code units of the n code points where the first one begins at i. |
1122 | * I.e., advance i by n code points. |
1123 | * Same as UT16_FWD_N. |
1124 | * \pre 0<=i<length |
1125 | * \post 0<i<=length |
1126 | * |
1127 | * @deprecated ICU 2.4. Renamed to U16_FWD_N, see utf_old.h. |
1128 | */ |
1129 | #define UTF_FWD_N(s, i, length, n) U16_FWD_N(s, i, length, n) |
1130 | |
1131 | /** |
1132 | * Take the random-access index i and adjust it so that it points to the beginning |
1133 | * of a code point. |
1134 | * The input index points to any code unit of a code point and is moved to point to |
1135 | * the first code unit of the same code point. i is never incremented. |
1136 | * In other words, if i points to a trail surrogate that is preceded by a matching |
1137 | * lead surrogate, then i is decremented. Otherwise it is not modified. |
1138 | * This can be used to start an iteration with UTF_NEXT_CHAR() from a random index. |
1139 | * Same as UTF16_SET_CHAR_START. |
1140 | * \pre start<=i<length |
1141 | * \post start<=i<length |
1142 | * |
1143 | * @deprecated ICU 2.4. Renamed to U16_SET_CP_START, see utf_old.h. |
1144 | */ |
1145 | #define UTF_SET_CHAR_START(s, start, i) U16_SET_CP_START(s, start, i) |
1146 | |
1147 | /** |
1148 | * Set c to the code point that has code units before i |
1149 | * and move i backward (towards the beginning of the string) |
1150 | * to the first code unit of this code point (pre-increment). |
1151 | * i must point to the first code unit after the last unit of a code point (i==length is allowed). |
1152 | * Same as UTF16_PREV_CHAR. |
1153 | * \pre start<i<=length |
1154 | * \post start<=i<length |
1155 | * |
1156 | * @deprecated ICU 2.4. Renamed to U16_PREV, see utf_old.h. |
1157 | */ |
1158 | #define UTF_PREV_CHAR(s, start, i, c) U16_PREV(s, start, i, c) |
1159 | |
1160 | /** |
1161 | * Move i backward (towards the beginning of the string) |
1162 | * to the first code unit of the code point that has code units before i. |
1163 | * I.e., move i backward by one code point. |
1164 | * i must point to the first code unit after the last unit of a code point (i==length is allowed). |
1165 | * Same as UTF16_BACK_1. |
1166 | * \pre start<i<=length |
1167 | * \post start<=i<length |
1168 | * |
1169 | * @deprecated ICU 2.4. Renamed to U16_BACK_1, see utf_old.h. |
1170 | */ |
1171 | #define UTF_BACK_1(s, start, i) U16_BACK_1(s, start, i) |
1172 | |
1173 | /** |
1174 | * Move i backward (towards the beginning of the string) |
1175 | * to the first code unit of the n code points that have code units before i. |
1176 | * I.e., move i backward by n code points. |
1177 | * i must point to the first code unit after the last unit of a code point (i==length is allowed). |
1178 | * Same as UTF16_BACK_N. |
1179 | * \pre start<i<=length |
1180 | * \post start<=i<length |
1181 | * |
1182 | * @deprecated ICU 2.4. Renamed to U16_BACK_N, see utf_old.h. |
1183 | */ |
1184 | #define UTF_BACK_N(s, start, i, n) U16_BACK_N(s, start, i, n) |
1185 | |
1186 | /** |
1187 | * Take the random-access index i and adjust it so that it points beyond |
1188 | * a code point. The input index points beyond any code unit |
1189 | * of a code point and is moved to point beyond the last code unit of the same |
1190 | * code point. i is never decremented. |
1191 | * In other words, if i points to a trail surrogate that is preceded by a matching |
1192 | * lead surrogate, then i is incremented. Otherwise it is not modified. |
1193 | * This can be used to start an iteration with UTF_PREV_CHAR() from a random index. |
1194 | * Same as UTF16_SET_CHAR_LIMIT. |
1195 | * \pre start<i<=length |
1196 | * \post start<i<=length |
1197 | * |
1198 | * @deprecated ICU 2.4. Renamed to U16_SET_CP_LIMIT, see utf_old.h. |
1199 | */ |
1200 | #define UTF_SET_CHAR_LIMIT(s, start, i, length) U16_SET_CP_LIMIT(s, start, i, length) |
1201 | |
1202 | #endif // !U_HIDE_DEPRECATED_API && !U_HIDE_OBSOLETE_UTF_OLD_H |
1203 | |
1204 | #endif |
1205 | |