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