utf8proc
view utf8proc.c @ 13:2c7384f1fac1
Version 1.1.5
(updated Changelog and changed version number)
(updated Changelog and changed version number)
| author | jbe |
|---|---|
| date | Fri Oct 16 12:00:00 2009 +0200 (2009-10-16) |
| parents | 00d2bcbdc945 |
| children | d0bab6ca89a5 |
line source
1 /*
2 * Copyright (c) 2009 Public Software Group e. V., Berlin, Germany
3 *
4 * Permission is hereby granted, free of charge, to any person obtaining a
5 * copy of this software and associated documentation files (the "Software"),
6 * to deal in the Software without restriction, including without limitation
7 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
8 * and/or sell copies of the Software, and to permit persons to whom the
9 * Software is furnished to do so, subject to the following conditions:
10 *
11 * The above copyright notice and this permission notice shall be included in
12 * all copies or substantial portions of the Software.
13 *
14 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
15 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
16 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
17 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
18 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
19 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
20 * DEALINGS IN THE SOFTWARE.
21 */
23 /*
24 * This library contains derived data from a modified version of the
25 * Unicode data files.
26 *
27 * The original data files are available at
28 * http://www.unicode.org/Public/UNIDATA/
29 *
30 * Please notice the copyright statement in the file "utf8proc_data.c".
31 */
34 /*
35 * File name: utf8proc.c
36 *
37 * Description:
38 * Implementation of libutf8proc.
39 */
42 #include "utf8proc.h"
43 #include "utf8proc_data.c"
46 const int8_t utf8proc_utf8class[256] = {
47 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
48 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
49 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
50 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
51 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
52 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
53 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
54 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
55 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
56 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
57 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
58 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
59 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
60 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
61 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
62 4, 4, 4, 4, 4, 4, 4, 4, 0, 0, 0, 0, 0, 0, 0, 0 };
64 #define UTF8PROC_HANGUL_SBASE 0xAC00
65 #define UTF8PROC_HANGUL_LBASE 0x1100
66 #define UTF8PROC_HANGUL_VBASE 0x1161
67 #define UTF8PROC_HANGUL_TBASE 0x11A7
68 #define UTF8PROC_HANGUL_LCOUNT 19
69 #define UTF8PROC_HANGUL_VCOUNT 21
70 #define UTF8PROC_HANGUL_TCOUNT 28
71 #define UTF8PROC_HANGUL_NCOUNT 588
72 #define UTF8PROC_HANGUL_SCOUNT 11172
73 /* END is exclusive */
74 #define UTF8PROC_HANGUL_L_START 0x1100
75 #define UTF8PROC_HANGUL_L_END 0x115A
76 #define UTF8PROC_HANGUL_L_FILLER 0x115F
77 #define UTF8PROC_HANGUL_V_START 0x1160
78 #define UTF8PROC_HANGUL_V_END 0x11A3
79 #define UTF8PROC_HANGUL_T_START 0x11A8
80 #define UTF8PROC_HANGUL_T_END 0x11FA
81 #define UTF8PROC_HANGUL_S_START 0xAC00
82 #define UTF8PROC_HANGUL_S_END 0xD7A4
85 #define UTF8PROC_BOUNDCLASS_START 0
86 #define UTF8PROC_BOUNDCLASS_OTHER 1
87 #define UTF8PROC_BOUNDCLASS_CR 2
88 #define UTF8PROC_BOUNDCLASS_LF 3
89 #define UTF8PROC_BOUNDCLASS_CONTROL 4
90 #define UTF8PROC_BOUNDCLASS_EXTEND 5
91 #define UTF8PROC_BOUNDCLASS_L 6
92 #define UTF8PROC_BOUNDCLASS_V 7
93 #define UTF8PROC_BOUNDCLASS_T 8
94 #define UTF8PROC_BOUNDCLASS_LV 9
95 #define UTF8PROC_BOUNDCLASS_LVT 10
98 const char *utf8proc_version(void) {
99 return "1.1.5";
100 }
102 const char *utf8proc_errmsg(ssize_t errcode) {
103 switch (errcode) {
104 case UTF8PROC_ERROR_NOMEM:
105 return "Memory for processing UTF-8 data could not be allocated.";
106 case UTF8PROC_ERROR_OVERFLOW:
107 return "UTF-8 string is too long to be processed.";
108 case UTF8PROC_ERROR_INVALIDUTF8:
109 return "Invalid UTF-8 string";
110 case UTF8PROC_ERROR_NOTASSIGNED:
111 return "Unassigned Unicode code point found in UTF-8 string.";
112 case UTF8PROC_ERROR_INVALIDOPTS:
113 return "Invalid options for UTF-8 processing chosen.";
114 default:
115 return "An unknown error occured while processing UTF-8 data.";
116 }
117 }
119 ssize_t utf8proc_iterate(
120 const uint8_t *str, ssize_t strlen, int32_t *dst
121 ) {
122 int length;
123 int i;
124 int32_t uc = -1;
125 *dst = -1;
126 if (!strlen) return 0;
127 length = utf8proc_utf8class[str[0]];
128 if (!length) return UTF8PROC_ERROR_INVALIDUTF8;
129 if (strlen >= 0 && length > strlen) return UTF8PROC_ERROR_INVALIDUTF8;
130 for (i=1; i<length; i++) {
131 if ((str[i] & 0xC0) != 0x80) return UTF8PROC_ERROR_INVALIDUTF8;
132 }
133 switch (length) {
134 case 1:
135 uc = str[0];
136 break;
137 case 2:
138 uc = ((str[0] & 0x1F) << 6) + (str[1] & 0x3F);
139 if (uc < 0x80) uc = -1;
140 break;
141 case 3:
142 uc = ((str[0] & 0x0F) << 12) + ((str[1] & 0x3F) << 6)
143 + (str[2] & 0x3F);
144 if (uc < 0x800 || (uc >= 0xD800 && uc < 0xE000) ||
145 (uc >= 0xFDD0 && uc < 0xFDF0)) uc = -1;
146 break;
147 case 4:
148 uc = ((str[0] & 0x07) << 18) + ((str[1] & 0x3F) << 12)
149 + ((str[2] & 0x3F) << 6) + (str[3] & 0x3F);
150 if (uc < 0x10000 || uc >= 0x110000) uc = -1;
151 break;
152 }
153 if (uc < 0 || ((uc & 0xFFFF) >= 0xFFFE))
154 return UTF8PROC_ERROR_INVALIDUTF8;
155 *dst = uc;
156 return length;
157 }
159 bool utf8proc_codepoint_valid(int32_t uc) {
160 if (uc < 0 || uc >= 0x110000 ||
161 ((uc & 0xFFFF) >= 0xFFFE) || (uc >= 0xD800 && uc < 0xE000) ||
162 (uc >= 0xFDD0 && uc < 0xFDF0)) return false;
163 else return true;
164 }
166 ssize_t utf8proc_encode_char(int32_t uc, uint8_t *dst) {
167 if (uc < 0x00) {
168 return 0;
169 } else if (uc < 0x80) {
170 dst[0] = uc;
171 return 1;
172 } else if (uc < 0x800) {
173 dst[0] = 0xC0 + (uc >> 6);
174 dst[1] = 0x80 + (uc & 0x3F);
175 return 2;
176 } else if (uc == 0xFFFF) {
177 dst[0] = 0xFF;
178 return 1;
179 } else if (uc == 0xFFFE) {
180 dst[0] = 0xFE;
181 return 1;
182 } else if (uc < 0x10000) {
183 dst[0] = 0xE0 + (uc >> 12);
184 dst[1] = 0x80 + ((uc >> 6) & 0x3F);
185 dst[2] = 0x80 + (uc & 0x3F);
186 return 3;
187 } else if (uc < 0x110000) {
188 dst[0] = 0xF0 + (uc >> 18);
189 dst[1] = 0x80 + ((uc >> 12) & 0x3F);
190 dst[2] = 0x80 + ((uc >> 6) & 0x3F);
191 dst[3] = 0x80 + (uc & 0x3F);
192 return 4;
193 } else return 0;
194 }
196 const utf8proc_property_t *utf8proc_get_property(int32_t uc) {
197 /* ASSERT: uc >= 0 && uc < 0x110000 */
198 return utf8proc_properties + (
199 utf8proc_stage2table[
200 utf8proc_stage1table[uc >> 8] + (uc & 0xFF)
201 ]
202 );
203 }
205 #define utf8proc_decompose_lump(replacement_uc) \
206 return utf8proc_decompose_char((replacement_uc), dst, bufsize, \
207 options & ~UTF8PROC_LUMP, last_boundclass)
209 ssize_t utf8proc_decompose_char(int32_t uc, int32_t *dst, ssize_t bufsize,
210 int options, int *last_boundclass) {
211 /* ASSERT: uc >= 0 && uc < 0x110000 */
212 const utf8proc_property_t *property;
213 utf8proc_propval_t category;
214 int32_t hangul_sindex;
215 property = utf8proc_get_property(uc);
216 category = property->category;
217 hangul_sindex = uc - UTF8PROC_HANGUL_SBASE;
218 if (options & (UTF8PROC_COMPOSE|UTF8PROC_DECOMPOSE)) {
219 if (hangul_sindex >= 0 && hangul_sindex < UTF8PROC_HANGUL_SCOUNT) {
220 int32_t hangul_tindex;
221 if (bufsize >= 1) {
222 dst[0] = UTF8PROC_HANGUL_LBASE +
223 hangul_sindex / UTF8PROC_HANGUL_NCOUNT;
224 if (bufsize >= 2) dst[1] = UTF8PROC_HANGUL_VBASE +
225 (hangul_sindex % UTF8PROC_HANGUL_NCOUNT) / UTF8PROC_HANGUL_TCOUNT;
226 }
227 hangul_tindex = hangul_sindex % UTF8PROC_HANGUL_TCOUNT;
228 if (!hangul_tindex) return 2;
229 if (bufsize >= 3) dst[2] = UTF8PROC_HANGUL_TBASE + hangul_tindex;
230 return 3;
231 }
232 }
233 if (options & UTF8PROC_REJECTNA) {
234 if (!category) return UTF8PROC_ERROR_NOTASSIGNED;
235 }
236 if (options & UTF8PROC_IGNORE) {
237 if (property->ignorable) return 0;
238 }
239 if (options & UTF8PROC_LUMP) {
240 if (category == UTF8PROC_CATEGORY_ZS) utf8proc_decompose_lump(0x0020);
241 if (uc == 0x2018 || uc == 0x2019 || uc == 0x02BC || uc == 0x02C8)
242 utf8proc_decompose_lump(0x0027);
243 if (category == UTF8PROC_CATEGORY_PD || uc == 0x2212)
244 utf8proc_decompose_lump(0x002D);
245 if (uc == 0x2044 || uc == 0x2215) utf8proc_decompose_lump(0x002F);
246 if (uc == 0x2236) utf8proc_decompose_lump(0x003A);
247 if (uc == 0x2039 || uc == 0x2329 || uc == 0x3008)
248 utf8proc_decompose_lump(0x003C);
249 if (uc == 0x203A || uc == 0x232A || uc == 0x3009)
250 utf8proc_decompose_lump(0x003E);
251 if (uc == 0x2216) utf8proc_decompose_lump(0x005C);
252 if (uc == 0x02C4 || uc == 0x02C6 || uc == 0x2038 || uc == 0x2303)
253 utf8proc_decompose_lump(0x005E);
254 if (category == UTF8PROC_CATEGORY_PC || uc == 0x02CD)
255 utf8proc_decompose_lump(0x005F);
256 if (uc == 0x02CB) utf8proc_decompose_lump(0x0060);
257 if (uc == 0x2223) utf8proc_decompose_lump(0x007C);
258 if (uc == 0x223C) utf8proc_decompose_lump(0x007E);
259 if ((options & UTF8PROC_NLF2LS) && (options & UTF8PROC_NLF2PS)) {
260 if (category == UTF8PROC_CATEGORY_ZL ||
261 category == UTF8PROC_CATEGORY_ZP)
262 utf8proc_decompose_lump(0x000A);
263 }
264 }
265 if (options & UTF8PROC_STRIPMARK) {
266 if (category == UTF8PROC_CATEGORY_MN ||
267 category == UTF8PROC_CATEGORY_MC ||
268 category == UTF8PROC_CATEGORY_ME) return 0;
269 }
270 if (options & UTF8PROC_CASEFOLD) {
271 if (property->casefold_mapping) {
272 const int32_t *casefold_entry;
273 ssize_t written = 0;
274 for (casefold_entry = property->casefold_mapping;
275 *casefold_entry >= 0; casefold_entry++) {
276 written += utf8proc_decompose_char(*casefold_entry, dst+written,
277 (bufsize > written) ? (bufsize - written) : 0, options,
278 last_boundclass);
279 if (written < 0) return UTF8PROC_ERROR_OVERFLOW;
280 }
281 return written;
282 }
283 }
284 if (options & (UTF8PROC_COMPOSE|UTF8PROC_DECOMPOSE)) {
285 if (property->decomp_mapping &&
286 (!property->decomp_type || (options & UTF8PROC_COMPAT))) {
287 const int32_t *decomp_entry;
288 ssize_t written = 0;
289 for (decomp_entry = property->decomp_mapping;
290 *decomp_entry >= 0; decomp_entry++) {
291 written += utf8proc_decompose_char(*decomp_entry, dst+written,
292 (bufsize > written) ? (bufsize - written) : 0, options,
293 last_boundclass);
294 if (written < 0) return UTF8PROC_ERROR_OVERFLOW;
295 }
296 return written;
297 }
298 }
299 if (options & UTF8PROC_CHARBOUND) {
300 bool boundary;
301 int tbc, lbc;
302 tbc =
303 (uc == 0x000D) ? UTF8PROC_BOUNDCLASS_CR :
304 (uc == 0x000A) ? UTF8PROC_BOUNDCLASS_LF :
305 ((category == UTF8PROC_CATEGORY_ZL ||
306 category == UTF8PROC_CATEGORY_ZP ||
307 category == UTF8PROC_CATEGORY_CC ||
308 category == UTF8PROC_CATEGORY_CF) &&
309 !(uc == 0x200C || uc == 0x200D)) ? UTF8PROC_BOUNDCLASS_CONTROL :
310 property->extend ? UTF8PROC_BOUNDCLASS_EXTEND :
311 ((uc >= UTF8PROC_HANGUL_L_START && uc < UTF8PROC_HANGUL_L_END) ||
312 uc == UTF8PROC_HANGUL_L_FILLER) ? UTF8PROC_BOUNDCLASS_L :
313 (uc >= UTF8PROC_HANGUL_V_START && uc < UTF8PROC_HANGUL_V_END) ?
314 UTF8PROC_BOUNDCLASS_V :
315 (uc >= UTF8PROC_HANGUL_T_START && uc < UTF8PROC_HANGUL_T_END) ?
316 UTF8PROC_BOUNDCLASS_T :
317 (uc >= UTF8PROC_HANGUL_S_START && uc < UTF8PROC_HANGUL_S_END) ? (
318 ((uc-UTF8PROC_HANGUL_SBASE) % UTF8PROC_HANGUL_TCOUNT == 0) ?
319 UTF8PROC_BOUNDCLASS_LV : UTF8PROC_BOUNDCLASS_LVT
320 ) :
321 UTF8PROC_BOUNDCLASS_OTHER;
322 lbc = *last_boundclass;
323 boundary =
324 (tbc == UTF8PROC_BOUNDCLASS_EXTEND) ? false :
325 (lbc == UTF8PROC_BOUNDCLASS_START) ? true :
326 (lbc == UTF8PROC_BOUNDCLASS_CR &&
327 tbc == UTF8PROC_BOUNDCLASS_LF) ? false :
328 (lbc == UTF8PROC_BOUNDCLASS_CONTROL) ? true :
329 (tbc == UTF8PROC_BOUNDCLASS_CONTROL) ? true :
330 (lbc == UTF8PROC_BOUNDCLASS_L &&
331 (tbc == UTF8PROC_BOUNDCLASS_L ||
332 tbc == UTF8PROC_BOUNDCLASS_V ||
333 tbc == UTF8PROC_BOUNDCLASS_LV ||
334 tbc == UTF8PROC_BOUNDCLASS_LVT)) ? false :
335 ((lbc == UTF8PROC_BOUNDCLASS_LV ||
336 lbc == UTF8PROC_BOUNDCLASS_V) &&
337 (tbc == UTF8PROC_BOUNDCLASS_V ||
338 tbc == UTF8PROC_BOUNDCLASS_T)) ? false :
339 ((lbc == UTF8PROC_BOUNDCLASS_LVT ||
340 lbc == UTF8PROC_BOUNDCLASS_T) &&
341 tbc == UTF8PROC_BOUNDCLASS_T) ? false :
342 true;
343 *last_boundclass = tbc;
344 if (boundary) {
345 if (bufsize >= 1) dst[0] = 0xFFFF;
346 if (bufsize >= 2) dst[1] = uc;
347 return 2;
348 }
349 }
350 if (bufsize >= 1) *dst = uc;
351 return 1;
352 }
354 ssize_t utf8proc_decompose(
355 const uint8_t *str, ssize_t strlen,
356 int32_t *buffer, ssize_t bufsize, int options
357 ) {
358 /* strlen will be ignored, if UTF8PROC_NULLTERM is set in options */
359 ssize_t wpos = 0;
360 if ((options & UTF8PROC_COMPOSE) && (options & UTF8PROC_DECOMPOSE))
361 return UTF8PROC_ERROR_INVALIDOPTS;
362 if ((options & UTF8PROC_STRIPMARK) &&
363 !(options & UTF8PROC_COMPOSE) && !(options & UTF8PROC_DECOMPOSE))
364 return UTF8PROC_ERROR_INVALIDOPTS;
365 {
366 int32_t uc;
367 ssize_t rpos = 0;
368 ssize_t decomp_result;
369 int boundclass = UTF8PROC_BOUNDCLASS_START;
370 while (1) {
371 if (options & UTF8PROC_NULLTERM) {
372 rpos += utf8proc_iterate(str + rpos, -1, &uc);
373 /* checking of return value is not neccessary,
374 as 'uc' is < 0 in case of error */
375 if (uc < 0) return UTF8PROC_ERROR_INVALIDUTF8;
376 if (rpos < 0) return UTF8PROC_ERROR_OVERFLOW;
377 if (uc == 0) break;
378 } else {
379 if (rpos >= strlen) break;
380 rpos += utf8proc_iterate(str + rpos, strlen - rpos, &uc);
381 if (uc < 0) return UTF8PROC_ERROR_INVALIDUTF8;
382 }
383 decomp_result = utf8proc_decompose_char(
384 uc, buffer + wpos, (bufsize > wpos) ? (bufsize - wpos) : 0, options,
385 &boundclass
386 );
387 if (decomp_result < 0) return decomp_result;
388 wpos += decomp_result;
389 /* prohibiting integer overflows due to too long strings: */
390 if (wpos < 0 || wpos > SSIZE_MAX/sizeof(int32_t)/2)
391 return UTF8PROC_ERROR_OVERFLOW;
392 }
393 }
394 if ((options & (UTF8PROC_COMPOSE|UTF8PROC_DECOMPOSE)) && bufsize >= wpos) {
395 ssize_t pos = 0;
396 while (pos < wpos-1) {
397 int32_t uc1, uc2;
398 const utf8proc_property_t *property1, *property2;
399 uc1 = buffer[pos];
400 uc2 = buffer[pos+1];
401 property1 = utf8proc_get_property(uc1);
402 property2 = utf8proc_get_property(uc2);
403 if (property1->combining_class > property2->combining_class &&
404 property2->combining_class > 0) {
405 buffer[pos] = uc2;
406 buffer[pos+1] = uc1;
407 if (pos > 0) pos--; else pos++;
408 } else {
409 pos++;
410 }
411 }
412 }
413 return wpos;
414 }
416 ssize_t utf8proc_reencode(int32_t *buffer, ssize_t length, int options) {
417 /* UTF8PROC_NULLTERM option will be ignored, 'length' is never ignored
418 ASSERT: 'buffer' has one spare byte of free space at the end! */
419 if (options & (UTF8PROC_NLF2LS | UTF8PROC_NLF2PS | UTF8PROC_STRIPCC)) {
420 ssize_t rpos;
421 ssize_t wpos = 0;
422 int32_t uc;
423 for (rpos = 0; rpos < length; rpos++) {
424 uc = buffer[rpos];
425 if (uc == 0x000D && rpos < length-1 && buffer[rpos+1] == 0x000A) rpos++;
426 if (uc == 0x000A || uc == 0x000D || uc == 0x0085 ||
427 ((options & UTF8PROC_STRIPCC) && (uc == 0x000B || uc == 0x000C))) {
428 if (options & UTF8PROC_NLF2LS) {
429 if (options & UTF8PROC_NLF2PS) {
430 buffer[wpos++] = 0x000A;
431 } else {
432 buffer[wpos++] = 0x2028;
433 }
434 } else {
435 if (options & UTF8PROC_NLF2PS) {
436 buffer[wpos++] = 0x2029;
437 } else {
438 buffer[wpos++] = 0x0020;
439 }
440 }
441 } else if ((options & UTF8PROC_STRIPCC) &&
442 (uc < 0x0020 || (uc >= 0x007F && uc < 0x00A0))) {
443 if (uc == 0x0009) buffer[wpos++] = 0x0020;
444 } else {
445 buffer[wpos++] = uc;
446 }
447 }
448 length = wpos;
449 }
450 if (options & UTF8PROC_COMPOSE) {
451 int32_t *starter = NULL;
452 int32_t current_char;
453 const utf8proc_property_t *starter_property = NULL, *current_property;
454 utf8proc_propval_t max_combining_class = -1;
455 ssize_t rpos;
456 ssize_t wpos = 0;
457 int32_t composition;
458 for (rpos = 0; rpos < length; rpos++) {
459 current_char = buffer[rpos];
460 current_property = utf8proc_get_property(current_char);
461 if (starter && current_property->combining_class > max_combining_class) {
462 /* combination perhaps possible */
463 int32_t hangul_lindex;
464 int32_t hangul_sindex;
465 hangul_lindex = *starter - UTF8PROC_HANGUL_LBASE;
466 if (hangul_lindex >= 0 && hangul_lindex < UTF8PROC_HANGUL_LCOUNT) {
467 int32_t hangul_vindex;
468 hangul_vindex = current_char - UTF8PROC_HANGUL_VBASE;
469 if (hangul_vindex >= 0 && hangul_vindex < UTF8PROC_HANGUL_VCOUNT) {
470 *starter = UTF8PROC_HANGUL_SBASE +
471 (hangul_lindex * UTF8PROC_HANGUL_VCOUNT + hangul_vindex) *
472 UTF8PROC_HANGUL_TCOUNT;
473 starter_property = NULL;
474 continue;
475 }
476 }
477 hangul_sindex = *starter - UTF8PROC_HANGUL_SBASE;
478 if (hangul_sindex >= 0 && hangul_sindex < UTF8PROC_HANGUL_SCOUNT &&
479 (hangul_sindex % UTF8PROC_HANGUL_TCOUNT) == 0) {
480 int32_t hangul_tindex;
481 hangul_tindex = current_char - UTF8PROC_HANGUL_TBASE;
482 if (hangul_tindex >= 0 && hangul_tindex < UTF8PROC_HANGUL_TCOUNT) {
483 *starter += hangul_tindex;
484 starter_property = NULL;
485 continue;
486 }
487 }
488 if (!starter_property) {
489 starter_property = utf8proc_get_property(*starter);
490 }
491 if (starter_property->comb1st_index >= 0 &&
492 current_property->comb2nd_index >= 0) {
493 composition = utf8proc_combinations[
494 starter_property->comb1st_index +
495 current_property->comb2nd_index
496 ];
497 if (composition >= 0 && (!(options & UTF8PROC_STABLE) ||
498 !(utf8proc_get_property(composition)->comp_exclusion))) {
499 *starter = composition;
500 starter_property = NULL;
501 continue;
502 }
503 }
504 }
505 buffer[wpos] = current_char;
506 if (current_property->combining_class) {
507 if (current_property->combining_class > max_combining_class) {
508 max_combining_class = current_property->combining_class;
509 }
510 } else {
511 starter = buffer + wpos;
512 starter_property = NULL;
513 max_combining_class = -1;
514 }
515 wpos++;
516 }
517 length = wpos;
518 }
519 {
520 ssize_t rpos, wpos = 0;
521 int32_t uc;
522 for (rpos = 0; rpos < length; rpos++) {
523 uc = buffer[rpos];
524 wpos += utf8proc_encode_char(uc, ((uint8_t *)buffer) + wpos);
525 }
526 ((uint8_t *)buffer)[wpos] = 0;
527 return wpos;
528 }
529 }
531 ssize_t utf8proc_map(
532 const uint8_t *str, ssize_t strlen, uint8_t **dstptr, int options
533 ) {
534 int32_t *buffer;
535 ssize_t result;
536 *dstptr = NULL;
537 result = utf8proc_decompose(str, strlen, NULL, 0, options);
538 if (result < 0) return result;
539 buffer = malloc(result * sizeof(int32_t) + 1);
540 if (!buffer) return UTF8PROC_ERROR_NOMEM;
541 result = utf8proc_decompose(str, strlen, buffer, result, options);
542 if (result < 0) {
543 free(buffer);
544 return result;
545 }
546 result = utf8proc_reencode(buffer, result, options);
547 if (result < 0) {
548 free(buffer);
549 return result;
550 }
551 {
552 int32_t *newptr;
553 newptr = realloc(buffer, (size_t)result+1);
554 if (newptr) buffer = newptr;
555 }
556 *dstptr = (uint8_t *)buffer;
557 return result;
558 }
560 uint8_t *utf8proc_NFD(const uint8_t *str) {
561 uint8_t *retval;
562 utf8proc_map(str, 0, &retval, UTF8PROC_NULLTERM | UTF8PROC_STABLE |
563 UTF8PROC_DECOMPOSE);
564 return retval;
565 }
567 uint8_t *utf8proc_NFC(const uint8_t *str) {
568 uint8_t *retval;
569 utf8proc_map(str, 0, &retval, UTF8PROC_NULLTERM | UTF8PROC_STABLE |
570 UTF8PROC_COMPOSE);
571 return retval;
572 }
574 uint8_t *utf8proc_NFKD(const uint8_t *str) {
575 uint8_t *retval;
576 utf8proc_map(str, 0, &retval, UTF8PROC_NULLTERM | UTF8PROC_STABLE |
577 UTF8PROC_DECOMPOSE | UTF8PROC_COMPAT);
578 return retval;
579 }
581 uint8_t *utf8proc_NFKC(const uint8_t *str) {
582 uint8_t *retval;
583 utf8proc_map(str, 0, &retval, UTF8PROC_NULLTERM | UTF8PROC_STABLE |
584 UTF8PROC_COMPOSE | UTF8PROC_COMPAT);
585 return retval;
586 }