1 /* $NetBSD: citrus_zw.c,v 1.3 2006/11/24 17:27:52 tnozaki Exp $ */
2 /* $DragonFly: src/lib/libc/citrus/modules/citrus_zw.c,v 1.1 2008/04/10 10:21:02 hasso Exp $ */
5 * Copyright (c)2004, 2006 Citrus Project,
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
11 * 1. Redistributions of source code must retain the above copyright
12 * notice, this list of conditions and the following disclaimer.
13 * 2. Redistributions in binary form must reproduce the above copyright
14 * notice, this list of conditions and the following disclaimer in the
15 * documentation and/or other materials provided with the distribution.
17 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
18 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
19 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
20 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
21 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
22 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
23 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
24 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
25 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
26 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
31 #include <sys/types.h>
43 #include "citrus_namespace.h"
44 #include "citrus_types.h"
45 #include "citrus_module.h"
46 #include "citrus_ctype.h"
47 #include "citrus_stdenc.h"
48 #include "citrus_zw.h"
50 /* ----------------------------------------------------------------------
51 * private stuffs used by templates
59 NONE, AMBIGIOUS, ASCII, GB2312
71 /* for future multi-locale facility */
83 #define _CEI_TO_EI(_cei_) (&(_cei_)->ei)
84 #define _CEI_TO_STATE(_cei_, _func_) (_cei_)->states.s_##_func_
86 #define _FUNCNAME(m) _citrus_ZW_##m
87 #define _ENCODING_INFO _ZWEncodingInfo
88 #define _CTYPE_INFO _ZWCTypeInfo
89 #define _ENCODING_STATE _ZWState
90 #define _ENCODING_MB_CUR_MAX(_ei_) MB_LEN_MAX
91 #define _ENCODING_IS_STATE_DEPENDENT 1
92 #define _STATE_NEEDS_EXPLICIT_INIT(_ps_) ((_ps_)->charset != NONE)
96 _citrus_ZW_init_state(_ZWEncodingInfo * __restrict ei,
97 _ZWState * __restrict psenc)
100 _DIAGASSERT(psenc != NULL);
103 psenc->charset = NONE;
108 _citrus_ZW_pack_state(_ZWEncodingInfo * __restrict ei,
109 void *__restrict pspriv, const _ZWState * __restrict psenc)
111 /* ei may be unused */
112 _DIAGASSERT(pspriv != NULL);
113 _DIAGASSERT(psenc != NULL);
115 memcpy(pspriv, (const void *)psenc, sizeof(*psenc));
120 _citrus_ZW_unpack_state(_ZWEncodingInfo * __restrict ei,
121 _ZWState * __restrict psenc, const void * __restrict pspriv)
123 /* ei may be unused */
124 _DIAGASSERT(psenc != NULL);
125 _DIAGASSERT(pspriv != NULL);
127 memcpy((void *)psenc, pspriv, sizeof(*psenc));
131 _citrus_ZW_mbrtowc_priv(_ZWEncodingInfo * __restrict ei,
132 wchar_t * __restrict pwc, const char **__restrict s, size_t n,
133 _ZWState * __restrict psenc, size_t * __restrict nresult)
139 /* ei may be unused */
140 /* pwc may be null */
141 _DIAGASSERT(s != NULL);
142 _DIAGASSERT(psenc != NULL);
143 _DIAGASSERT(nresult != NULL);
146 _citrus_ZW_init_state(ei, psenc);
147 *nresult = (size_t)_ENCODING_IS_STATE_DEPENDENT;
156 *nresult = (size_t)-2; \
160 ch = (unsigned char)*s0++; \
161 if (len++ > MB_LEN_MAX || ch > 0x7F)\
163 psenc->ch[psenc->chlen++] = ch; \
164 } while (/*CONSTCOND*/0)
167 switch (psenc->charset) {
169 switch (psenc->chlen) {
172 switch (psenc->ch[0]) {
173 case '\0': case '\n':
174 psenc->charset = NONE;
182 ch = (unsigned char)psenc->ch[0];
189 if (psenc->chlen != 0)
192 ch = (unsigned char)psenc->ch[0];
194 if (ch != '\n' && ch != '\0')
195 psenc->charset = ASCII;
200 psenc->charset = AMBIGIOUS;
204 if (psenc->chlen != 0)
207 if (psenc->ch[0] != 'W') {
208 psenc->charset = ASCII;
212 psenc->charset = GB2312;
216 switch (psenc->chlen) {
219 ch = (unsigned char)psenc->ch[0];
221 psenc->charset = NONE;
225 } else if (ch == '\n') {
226 psenc->charset = NONE;
233 if (psenc->ch[0] == ' ') {
234 ch = (unsigned char)psenc->ch[1];
238 } else if (psenc->ch[0] == '#') {
239 ch = (unsigned char)psenc->ch[1];
241 psenc->charset = NONE;
245 } else if (ch == ' ') {
251 ch = (unsigned char)psenc->ch[0];
252 if (ch < 0x21 || ch > 0x7E)
254 wc = (wchar_t)(ch << 8);
255 ch = (unsigned char)psenc->ch[1];
256 if (ch < 0x21 || ch > 0x7E) {
258 *nresult = (size_t)-1;
274 *nresult = (size_t)(wc == 0 ? 0 : len);
282 _citrus_ZW_wcrtomb_priv(_ZWEncodingInfo * __restrict ei,
283 char *__restrict s, size_t n, wchar_t wc,
284 _ZWState * __restrict psenc, size_t * __restrict nresult)
289 _DIAGASSERT(s != NULL);
290 _DIAGASSERT(psenc != NULL);
291 _DIAGASSERT(nresult != NULL);
293 if (psenc->chlen != 0)
295 if ((uint32_t)wc <= 0x7F) {
296 ch = (unsigned char)wc;
297 switch (psenc->charset) {
299 if (ch == '\0' || ch == '\n') {
300 psenc->ch[psenc->chlen++] = ch;
305 psenc->ch[psenc->chlen++] = 'z';
306 psenc->ch[psenc->chlen++] = 'W';
307 psenc->ch[psenc->chlen++] = ' ';
308 psenc->ch[psenc->chlen++] = ch;
309 psenc->charset = GB2312;
317 psenc->ch[psenc->chlen++] = '\n';
318 psenc->ch[psenc->chlen++] = '\0';
319 psenc->charset = NONE;
320 } else if (ch == '\n') {
321 psenc->ch[psenc->chlen++] = '#';
322 psenc->ch[psenc->chlen++] = '\n';
323 psenc->charset = NONE;
325 psenc->ch[psenc->chlen++] = ' ';
326 psenc->ch[psenc->chlen++] = ch;
332 } else if ((uint32_t)wc <= 0x7E7E) {
333 switch (psenc->charset) {
338 psenc->ch[psenc->chlen++] = 'z';
339 psenc->ch[psenc->chlen++] = 'W';
340 psenc->charset = GB2312;
346 ch = (wc >> 8) & 0xFF;
347 if (ch < 0x21 || ch > 0x7E)
349 psenc->ch[psenc->chlen++] = ch;
351 if (ch < 0x21 || ch > 0x7E)
353 psenc->ch[psenc->chlen++] = ch;
360 *nresult = (size_t)-1;
363 memcpy(s, psenc->ch, psenc->chlen);
364 *nresult = psenc->chlen;
372 _citrus_ZW_put_state_reset(_ZWEncodingInfo * __restrict ei,
373 char * __restrict s, size_t n,
374 _ZWState * __restrict psenc, size_t * __restrict nresult)
376 /* ei may be unused */
377 _DIAGASSERT(s != NULL);
378 _DIAGASSERT(psenc != NULL);
379 _DIAGASSERT(nresult != NULL);
381 if (psenc->chlen != 0)
383 switch (psenc->charset) {
387 psenc->ch[psenc->chlen++] = '\n';
388 psenc->charset = NONE;
391 *nresult = psenc->chlen;
392 if (psenc->chlen > 0) {
393 memcpy(s, psenc->ch, psenc->chlen);
406 _citrus_ZW_stdenc_get_state_desc_generic(_ZWEncodingInfo * __restrict ei,
407 _ZWState * __restrict psenc, int * __restrict rstate)
409 /* ei may be unused */
410 _DIAGASSERT(psenc != NULL);
411 _DIAGASSERT(rstate != NULL);
413 switch (psenc->charset) {
415 if (psenc->chlen != 0)
417 *rstate = _STDENC_SDGEN_INITIAL;
420 if (psenc->chlen != 0)
422 *rstate = _STDENC_SDGEN_INCOMPLETE_SHIFT;
426 switch (psenc->chlen) {
428 *rstate = _STDENC_SDGEN_STABLE;
431 *rstate = (psenc->ch[0] == '#')
432 ? _STDENC_SDGEN_INCOMPLETE_SHIFT
433 : _STDENC_SDGEN_INCOMPLETE_CHAR;
447 _citrus_ZW_stdenc_wctocs(_ZWEncodingInfo * __restrict ei,
448 _csid_t * __restrict csid, _index_t * __restrict idx, wchar_t wc)
450 /* ei seems to be unused */
451 _DIAGASSERT(csid != NULL);
452 _DIAGASSERT(idx != NULL);
454 *csid = (_csid_t)(wc <= 0x7FU) ? 0 : 1;
462 _citrus_ZW_stdenc_cstowc(_ZWEncodingInfo * __restrict ei,
463 wchar_t * __restrict wc, _csid_t csid, _index_t idx)
465 /* ei seems to be unused */
466 _DIAGASSERT(wc != NULL);
481 _citrus_ZW_encoding_module_uninit(_ZWEncodingInfo *ei)
487 _citrus_ZW_encoding_module_init(_ZWEncodingInfo * __restrict ei,
488 const void *__restrict var, size_t lenvar)
493 /* ----------------------------------------------------------------------
494 * public interface for ctype
497 _CITRUS_CTYPE_DECLS(ZW);
498 _CITRUS_CTYPE_DEF_OPS(ZW);
500 #include "citrus_ctype_template.h"
502 /* ----------------------------------------------------------------------
503 * public interface for stdenc
506 _CITRUS_STDENC_DECLS(ZW);
507 _CITRUS_STDENC_DEF_OPS(ZW);
509 #include "citrus_stdenc_template.h"