2 /* $NetBSD: citrus_zw.c,v 1.4 2008/06/14 16:01:08 tnozaki 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/cdefs.h>
32 #include <sys/types.h>
44 #include "citrus_namespace.h"
45 #include "citrus_types.h"
46 #include "citrus_module.h"
47 #include "citrus_stdenc.h"
48 #include "citrus_zw.h"
50 /* ----------------------------------------------------------------------
51 * private stuffs used by templates
59 NONE, AMBIGIOUS, ASCII, GB2312
68 #define _CEI_TO_EI(_cei_) (&(_cei_)->ei)
69 #define _CEI_TO_STATE(_cei_, _func_) (_cei_)->states.s_##_func_
71 #define _FUNCNAME(m) _citrus_ZW_##m
72 #define _ENCODING_INFO _ZWEncodingInfo
73 #define _ENCODING_STATE _ZWState
74 #define _ENCODING_MB_CUR_MAX(_ei_) MB_LEN_MAX
75 #define _ENCODING_IS_STATE_DEPENDENT 1
76 #define _STATE_NEEDS_EXPLICIT_INIT(_ps_) ((_ps_)->charset != NONE)
80 _citrus_ZW_init_state(_ZWEncodingInfo * __restrict ei __unused,
81 _ZWState * __restrict psenc)
85 psenc->charset = NONE;
91 _citrus_ZW_pack_state(_ZWEncodingInfo * __restrict ei __unused,
92 void *__restrict pspriv, const _ZWState * __restrict psenc)
95 memcpy(pspriv, (const void *)psenc, sizeof(*psenc));
100 _citrus_ZW_unpack_state(_ZWEncodingInfo * __restrict ei __unused,
101 _ZWState * __restrict psenc, const void * __restrict pspriv)
104 memcpy((void *)psenc, pspriv, sizeof(*psenc));
109 _citrus_ZW_mbrtowc_priv(_ZWEncodingInfo * __restrict ei,
110 wchar_t * __restrict pwc, char **__restrict s, size_t n,
111 _ZWState * __restrict psenc, size_t * __restrict nresult)
118 _citrus_ZW_init_state(ei, psenc);
119 *nresult = (size_t)_ENCODING_IS_STATE_DEPENDENT;
128 *nresult = (size_t)-2; \
132 ch = (unsigned char)*s0++; \
133 if (len++ > MB_LEN_MAX || ch > 0x7F)\
135 psenc->ch[psenc->chlen++] = ch; \
136 } while (/*CONSTCOND*/0)
139 switch (psenc->charset) {
141 switch (psenc->chlen) {
144 switch (psenc->ch[0]) {
145 case '\0': case '\n':
146 psenc->charset = NONE;
154 ch = (unsigned char)psenc->ch[0];
161 if (psenc->chlen != 0)
164 ch = (unsigned char)psenc->ch[0];
166 if (ch != '\n' && ch != '\0')
167 psenc->charset = ASCII;
172 psenc->charset = AMBIGIOUS;
176 if (psenc->chlen != 0)
179 if (psenc->ch[0] != 'W') {
180 psenc->charset = ASCII;
184 psenc->charset = GB2312;
188 switch (psenc->chlen) {
191 ch = (unsigned char)psenc->ch[0];
193 psenc->charset = NONE;
197 } else if (ch == '\n') {
198 psenc->charset = NONE;
205 if (psenc->ch[0] == ' ') {
206 ch = (unsigned char)psenc->ch[1];
210 } else if (psenc->ch[0] == '#') {
211 ch = (unsigned char)psenc->ch[1];
213 psenc->charset = NONE;
217 } else if (ch == ' ') {
223 ch = (unsigned char)psenc->ch[0];
224 if (ch < 0x21 || ch > 0x7E)
226 wc = (wchar_t)(ch << 8);
227 ch = (unsigned char)psenc->ch[1];
228 if (ch < 0x21 || ch > 0x7E) {
230 *nresult = (size_t)-1;
246 *nresult = (size_t)(wc == 0 ? 0 : len);
254 _citrus_ZW_wcrtomb_priv(_ZWEncodingInfo * __restrict ei __unused,
255 char *__restrict s, size_t n, wchar_t wc,
256 _ZWState * __restrict psenc, size_t * __restrict nresult)
260 if (psenc->chlen != 0)
262 if ((uint32_t)wc <= 0x7F) {
263 ch = (unsigned char)wc;
264 switch (psenc->charset) {
266 if (ch == '\0' || ch == '\n')
267 psenc->ch[psenc->chlen++] = ch;
272 psenc->ch[psenc->chlen++] = 'z';
273 psenc->ch[psenc->chlen++] = 'W';
274 psenc->ch[psenc->chlen++] = ' ';
275 psenc->ch[psenc->chlen++] = ch;
276 psenc->charset = GB2312;
284 psenc->ch[psenc->chlen++] = '\n';
285 psenc->ch[psenc->chlen++] = '\0';
286 psenc->charset = NONE;
287 } else if (ch == '\n') {
288 psenc->ch[psenc->chlen++] = '#';
289 psenc->ch[psenc->chlen++] = '\n';
290 psenc->charset = NONE;
292 psenc->ch[psenc->chlen++] = ' ';
293 psenc->ch[psenc->chlen++] = ch;
299 } else if ((uint32_t)wc <= 0x7E7E) {
300 switch (psenc->charset) {
305 psenc->ch[psenc->chlen++] = 'z';
306 psenc->ch[psenc->chlen++] = 'W';
307 psenc->charset = GB2312;
313 ch = (wc >> 8) & 0xFF;
314 if (ch < 0x21 || ch > 0x7E)
316 psenc->ch[psenc->chlen++] = ch;
318 if (ch < 0x21 || ch > 0x7E)
320 psenc->ch[psenc->chlen++] = ch;
327 *nresult = (size_t)-1;
330 memcpy(s, psenc->ch, psenc->chlen);
331 *nresult = psenc->chlen;
339 _citrus_ZW_put_state_reset(_ZWEncodingInfo * __restrict ei __unused,
340 char * __restrict s, size_t n, _ZWState * __restrict psenc,
341 size_t * __restrict nresult)
344 if (psenc->chlen != 0)
346 switch (psenc->charset) {
350 psenc->ch[psenc->chlen++] = '\n';
351 psenc->charset = NONE;
354 *nresult = psenc->chlen;
355 if (psenc->chlen > 0) {
356 memcpy(s, psenc->ch, psenc->chlen);
369 _citrus_ZW_stdenc_get_state_desc_generic(_ZWEncodingInfo * __restrict ei __unused,
370 _ZWState * __restrict psenc, int * __restrict rstate)
373 switch (psenc->charset) {
375 if (psenc->chlen != 0)
377 *rstate = _STDENC_SDGEN_INITIAL;
380 if (psenc->chlen != 0)
382 *rstate = _STDENC_SDGEN_INCOMPLETE_SHIFT;
386 switch (psenc->chlen) {
388 *rstate = _STDENC_SDGEN_STABLE;
391 *rstate = (psenc->ch[0] == '#') ?
392 _STDENC_SDGEN_INCOMPLETE_SHIFT :
393 _STDENC_SDGEN_INCOMPLETE_CHAR;
407 _citrus_ZW_stdenc_wctocs(_ZWEncodingInfo * __restrict ei __unused,
408 _csid_t * __restrict csid, _index_t * __restrict idx, wchar_t wc)
411 *csid = (_csid_t)(wc <= (wchar_t)0x7FU) ? 0 : 1;
419 _citrus_ZW_stdenc_cstowc(_ZWEncodingInfo * __restrict ei __unused,
420 wchar_t * __restrict wc, _csid_t csid, _index_t idx)
436 _citrus_ZW_encoding_module_uninit(_ZWEncodingInfo *ei __unused)
443 _citrus_ZW_encoding_module_init(_ZWEncodingInfo * __restrict ei __unused,
444 const void *__restrict var __unused, size_t lenvar __unused)
450 /* ----------------------------------------------------------------------
451 * public interface for stdenc
454 _CITRUS_STDENC_DECLS(ZW);
455 _CITRUS_STDENC_DEF_OPS(ZW);
457 #include "citrus_stdenc_template.h"