1 /* $NetBSD: citrus_big5.c,v 1.11 2006/11/22 23:38:27 tnozaki Exp $ */
4 * Copyright (c)2002, 2006 Citrus Project,
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions
10 * 1. Redistributions of source code must retain the above copyright
11 * notice, this list of conditions and the following disclaimer.
12 * 2. Redistributions in binary form must reproduce the above copyright
13 * notice, this list of conditions and the following disclaimer in the
14 * documentation and/or other materials provided with the distribution.
16 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
17 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
18 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
19 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
20 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
21 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
22 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
23 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
24 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
25 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
31 * The Regents of the University of California. All rights reserved.
33 * This code is derived from software contributed to Berkeley by
34 * Paul Borman at Krystal Technologies.
36 * Redistribution and use in source and binary forms, with or without
37 * modification, are permitted provided that the following conditions
39 * 1. Redistributions of source code must retain the above copyright
40 * notice, this list of conditions and the following disclaimer.
41 * 2. Redistributions in binary form must reproduce the above copyright
42 * notice, this list of conditions and the following disclaimer in the
43 * documentation and/or other materials provided with the distribution.
44 * 3. Neither the name of the University nor the names of its contributors
45 * may be used to endorse or promote products derived from this software
46 * without specific prior written permission.
48 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
49 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
50 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
51 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
52 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
53 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
54 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
55 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
56 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
57 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
61 #include <sys/queue.h>
62 #include <sys/types.h>
74 #include "citrus_namespace.h"
75 #include "citrus_types.h"
76 #include "citrus_bcs.h"
77 #include "citrus_module.h"
78 #include "citrus_ctype.h"
79 #include "citrus_stdenc.h"
80 #include "citrus_big5.h"
82 #include "citrus_prop.h"
84 /* ----------------------------------------------------------------------
85 * private stuffs used by templates
93 typedef struct _BIG5Exclude {
94 TAILQ_ENTRY(_BIG5Exclude) entry;
98 typedef TAILQ_HEAD(_BIG5ExcludeList, _BIG5Exclude) _BIG5ExcludeList;
102 _BIG5ExcludeList excludes;
106 _BIG5EncodingInfo ei;
108 /* for future multi-locale facility */
111 _BIG5State s_mbrtowc;
113 _BIG5State s_mbsrtowcs;
114 _BIG5State s_wcrtomb;
115 _BIG5State s_wcsrtombs;
120 #define _CEI_TO_EI(_cei_) (&(_cei_)->ei)
121 #define _CEI_TO_STATE(_cei_, _func_) (_cei_)->states.s_##_func_
123 #define _FUNCNAME(m) _citrus_BIG5_##m
124 #define _ENCODING_INFO _BIG5EncodingInfo
125 #define _CTYPE_INFO _BIG5CTypeInfo
126 #define _ENCODING_STATE _BIG5State
127 #define _ENCODING_MB_CUR_MAX(_ei_) 2
128 #define _ENCODING_IS_STATE_DEPENDENT 0
129 #define _STATE_NEEDS_EXPLICIT_INIT(_ps_) 0
134 _citrus_BIG5_init_state(_BIG5EncodingInfo * __restrict ei,
135 _BIG5State * __restrict s)
137 memset(s, 0, sizeof(*s));
142 _citrus_BIG5_pack_state(_BIG5EncodingInfo * __restrict ei,
143 void * __restrict pspriv,
144 const _BIG5State * __restrict s)
146 memcpy(pspriv, (const void *)s, sizeof(*s));
151 _citrus_BIG5_unpack_state(_BIG5EncodingInfo * __restrict ei,
152 _BIG5State * __restrict s,
153 const void * __restrict pspriv)
155 memcpy((void *)s, pspriv, sizeof(*s));
159 _citrus_BIG5_check(_BIG5EncodingInfo *ei, u_int c)
161 _DIAGASSERT(ei != NULL);
163 return (ei->cell[c & 0xFF] & 0x1) ? 2 : 1;
167 _citrus_BIG5_check2(_BIG5EncodingInfo *ei, u_int c)
169 _DIAGASSERT(ei != NULL);
171 return (ei->cell[c & 0xFF] & 0x2) ? 1 : 0;
175 _citrus_BIG5_check_excludes(_BIG5EncodingInfo *ei, wint_t c)
177 _BIG5Exclude *exclude;
179 _DIAGASSERT(ei != NULL);
181 TAILQ_FOREACH(exclude, &ei->excludes, entry) {
182 if (c >= exclude->start && c <= exclude->end)
189 _citrus_BIG5_fill_rowcol(void ** __restrict ctx, const char * __restrict s,
190 uint64_t start, uint64_t end)
192 _BIG5EncodingInfo *ei;
196 _DIAGASSERT(ctx != NULL && *ctx != NULL);
198 if (start > 0xFF || end > 0xFF)
200 ei = (_BIG5EncodingInfo *)*ctx;
201 i = strcmp("row", s) ? 1 : 0;
203 for (n = start; n <= end; ++n)
204 ei->cell[n & 0xFF] |= i;
210 _citrus_BIG5_fill_excludes(void ** __restrict ctx, const char * __restrict s,
211 uint64_t start, uint64_t end)
213 _BIG5EncodingInfo *ei;
214 _BIG5Exclude *exclude;
216 _DIAGASSERT(ctx != NULL && *ctx != NULL);
218 if (start > 0xFFFF || end > 0xFFFF)
220 ei = (_BIG5EncodingInfo *)*ctx;
221 exclude = TAILQ_LAST(&ei->excludes, _BIG5ExcludeList);
222 if (exclude != NULL && (wint_t)start <= exclude->end)
224 exclude = (void *)malloc(sizeof(*exclude));
227 exclude->start = (wint_t)start;
228 exclude->end = (wint_t)end;
229 TAILQ_INSERT_TAIL(&ei->excludes, exclude, entry);
234 static const _citrus_prop_hint_t root_hints[] = {
235 _CITRUS_PROP_HINT_NUM("row", &_citrus_BIG5_fill_rowcol),
236 _CITRUS_PROP_HINT_NUM("col", &_citrus_BIG5_fill_rowcol),
237 _CITRUS_PROP_HINT_NUM("excludes", &_citrus_BIG5_fill_excludes),
238 _CITRUS_PROP_HINT_END
243 _citrus_BIG5_encoding_module_uninit(_BIG5EncodingInfo *ei)
245 _BIG5Exclude *exclude;
247 _DIAGASSERT(ei != NULL);
249 while ((exclude = TAILQ_FIRST(&ei->excludes)) != NULL) {
250 TAILQ_REMOVE(&ei->excludes, exclude, entry);
257 _citrus_BIG5_encoding_module_init(_BIG5EncodingInfo * __restrict ei,
258 const void * __restrict var, size_t lenvar)
263 _DIAGASSERT(ei != NULL);
265 memset((void *)ei, 0, sizeof(*ei));
266 TAILQ_INIT(&ei->excludes);
268 if (lenvar > 0 && var != NULL) {
269 s = _bcs_skip_ws_len((const char *)var, &lenvar);
270 if (lenvar > 0 && *s != '\0') {
271 err = _citrus_prop_parse_variable(
272 root_hints, (void *)ei, s, lenvar);
276 _citrus_BIG5_encoding_module_uninit(ei);
277 memset((void *)ei, 0, sizeof(*ei));
278 TAILQ_INIT(&ei->excludes);
282 /* fallback Big5-1984, for backward compatibility. */
283 _citrus_BIG5_fill_rowcol((void **)&ei, "row", 0xA1, 0xFE);
284 _citrus_BIG5_fill_rowcol((void **)&ei, "col", 0x40, 0x7E);
285 _citrus_BIG5_fill_rowcol((void **)&ei, "col", 0xA1, 0xFE);
292 _citrus_BIG5_mbrtowc_priv(_BIG5EncodingInfo * __restrict ei,
293 wchar_t * __restrict pwc,
294 const char ** __restrict s, size_t n,
295 _BIG5State * __restrict psenc,
296 size_t * __restrict nresult)
303 _DIAGASSERT(nresult != NULL);
304 _DIAGASSERT(ei != NULL);
305 _DIAGASSERT(psenc != NULL);
306 _DIAGASSERT(s != NULL && *s != NULL);
311 _citrus_BIG5_init_state(ei, psenc);
316 chlenbak = psenc->chlen;
318 /* make sure we have the first byte in the buffer */
319 switch (psenc->chlen) {
323 psenc->ch[0] = *s0++;
334 c = _citrus_BIG5_check(ei, psenc->ch[0] & 0xff);
337 while (psenc->chlen < c) {
341 psenc->ch[psenc->chlen] = *s0++;
348 wchar = psenc->ch[0] & 0xff;
351 if (!_citrus_BIG5_check2(ei, psenc->ch[1] & 0xff))
353 wchar = ((psenc->ch[0] & 0xff) << 8) | (psenc->ch[1] & 0xff);
360 if (_citrus_BIG5_check_excludes(ei, (wint_t)wchar) != 0)
370 *nresult = c - chlenbak;
376 *nresult = (size_t)-1;
381 *nresult = (size_t)-2;
387 _citrus_BIG5_wcrtomb_priv(_BIG5EncodingInfo * __restrict ei,
389 size_t n, wchar_t wc, _BIG5State * __restrict psenc,
390 size_t * __restrict nresult)
394 _DIAGASSERT(ei != NULL);
395 _DIAGASSERT(nresult != NULL);
396 _DIAGASSERT(s != NULL);
398 /* check invalid sequence */
400 _citrus_BIG5_check_excludes(ei, (wint_t)wc) != 0) {
406 if (_citrus_BIG5_check(ei, (wc >> 8) & 0xff) != 2 ||
407 !_citrus_BIG5_check2(ei, wc & 0xff)) {
413 if (wc & ~0xff || !_citrus_BIG5_check(ei, wc & 0xff)) {
421 /* bound check failure */
427 s[0] = (wc >> 8) & 0xff;
437 *nresult = (size_t)-1;
443 _citrus_BIG5_stdenc_wctocs(_BIG5EncodingInfo * __restrict ei,
444 _csid_t * __restrict csid,
445 _index_t * __restrict idx, wchar_t wc)
448 _DIAGASSERT(csid != NULL && idx != NULL);
450 *csid = (wc < 0x100) ? 0 : 1;
458 _citrus_BIG5_stdenc_cstowc(_BIG5EncodingInfo * __restrict ei,
459 wchar_t * __restrict wc,
460 _csid_t csid, _index_t idx)
462 _DIAGASSERT(wc != NULL);
478 _citrus_BIG5_stdenc_get_state_desc_generic(_BIG5EncodingInfo * __restrict ei,
479 _BIG5State * __restrict psenc,
480 int * __restrict rstate)
483 if (psenc->chlen == 0)
484 *rstate = _STDENC_SDGEN_INITIAL;
486 *rstate = _STDENC_SDGEN_INCOMPLETE_CHAR;
491 /* ----------------------------------------------------------------------
492 * public interface for ctype
495 _CITRUS_CTYPE_DECLS(BIG5);
496 _CITRUS_CTYPE_DEF_OPS(BIG5);
498 #include "citrus_ctype_template.h"
501 /* ----------------------------------------------------------------------
502 * public interface for stdenc
505 _CITRUS_STDENC_DECLS(BIG5);
506 _CITRUS_STDENC_DEF_OPS(BIG5);
508 #include "citrus_stdenc_template.h"