1 /* $FreeBSD$ */
2 /*	$NetBSD: citrus_mskanji.c,v 1.13 2008/06/14 16:01:08 tnozaki Exp $	*/
3 
4 /*-
5  * SPDX-License-Identifier: BSD-2-Clause
6  *
7  * Copyright (c)2002 Citrus Project,
8  * All rights reserved.
9  *
10  * Redistribution and use in source and binary forms, with or without
11  * modification, are permitted provided that the following conditions
12  * are met:
13  * 1. Redistributions of source code must retain the above copyright
14  *    notice, this list of conditions and the following disclaimer.
15  * 2. Redistributions in binary form must reproduce the above copyright
16  *    notice, this list of conditions and the following disclaimer in the
17  *    documentation and/or other materials provided with the distribution.
18  *
19  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
20  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
21  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
22  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
23  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
24  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
25  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
26  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
27  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
28  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
29  * SUCH DAMAGE.
30  */
31 
32 /*
33  *    ja_JP.SJIS locale table for BSD4.4/rune
34  *    version 1.0
35  *    (C) Sin'ichiro MIYATANI / Phase One, Inc
36  *    May 12, 1995
37  *
38  * Redistribution and use in source and binary forms, with or without
39  * modification, are permitted provided that the following conditions
40  * are met:
41  * 1. Redistributions of source code must retain the above copyright
42  *    notice, this list of conditions and the following disclaimer.
43  * 2. Redistributions in binary form must reproduce the above copyright
44  *    notice, this list of conditions and the following disclaimer in the
45  *    documentation and/or other materials provided with the distribution.
46  * 3. All advertising materials mentioning features or use of this software
47  *    must display the following acknowledgement:
48  *      This product includes software developed by Phase One, Inc.
49  * 4. The name of Phase One, Inc. may be used to endorse or promote products
50  *    derived from this software without specific prior written permission.
51  *
52  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
53  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
54  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
55  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
56  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
57  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
58  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
59  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
60  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
61  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
62  * SUCH DAMAGE.
63  */
64 
65 
66 #include <sys/cdefs.h>
67 #include <sys/types.h>
68 
69 #include <assert.h>
70 #include <errno.h>
71 #include <limits.h>
72 #include <stdbool.h>
73 #include <stddef.h>
74 #include <stdio.h>
75 #include <stdlib.h>
76 #include <string.h>
77 #include <wchar.h>
78 
79 #include "citrus_namespace.h"
80 #include "citrus_types.h"
81 #include "citrus_bcs.h"
82 #include "citrus_module.h"
83 #include "citrus_stdenc.h"
84 #include "citrus_mskanji.h"
85 
86 
87 /* ----------------------------------------------------------------------
88  * private stuffs used by templates
89  */
90 
91 typedef struct _MSKanjiState {
92 	int	 chlen;
93 	char	 ch[2];
94 } _MSKanjiState;
95 
96 typedef struct {
97 	int	 mode;
98 #define MODE_JIS2004	1
99 } _MSKanjiEncodingInfo;
100 
101 #define _CEI_TO_EI(_cei_)		(&(_cei_)->ei)
102 #define _CEI_TO_STATE(_cei_, _func_)	(_cei_)->states.s_##_func_
103 
104 #define _FUNCNAME(m)			_citrus_MSKanji_##m
105 #define _ENCODING_INFO			_MSKanjiEncodingInfo
106 #define _ENCODING_STATE			_MSKanjiState
107 #define _ENCODING_MB_CUR_MAX(_ei_)	2
108 #define _ENCODING_IS_STATE_DEPENDENT	0
109 #define _STATE_NEEDS_EXPLICIT_INIT(_ps_)	0
110 
111 
112 static bool
113 _mskanji1(int c)
114 {
115 
116 	return ((c >= 0x81 && c <= 0x9f) || (c >= 0xe0 && c <= 0xfc));
117 }
118 
119 static bool
120 _mskanji2(int c)
121 {
122 
123 	return ((c >= 0x40 && c <= 0x7e) || (c >= 0x80 && c <= 0xfc));
124 }
125 
126 static __inline void
127 /*ARGSUSED*/
128 _citrus_MSKanji_init_state(_MSKanjiEncodingInfo * __restrict ei __unused,
129     _MSKanjiState * __restrict s)
130 {
131 
132 	s->chlen = 0;
133 }
134 
135 #if 0
136 static __inline void
137 /*ARGSUSED*/
138 _citrus_MSKanji_pack_state(_MSKanjiEncodingInfo * __restrict ei __unused,
139     void * __restrict pspriv, const _MSKanjiState * __restrict s)
140 {
141 
142 	memcpy(pspriv, (const void *)s, sizeof(*s));
143 }
144 
145 static __inline void
146 /*ARGSUSED*/
147 _citrus_MSKanji_unpack_state(_MSKanjiEncodingInfo * __restrict ei __unused,
148     _MSKanjiState * __restrict s, const void * __restrict pspriv)
149 {
150 
151 	memcpy((void *)s, pspriv, sizeof(*s));
152 }
153 #endif
154 
155 static int
156 /*ARGSUSED*/
157 _citrus_MSKanji_mbrtowc_priv(_MSKanjiEncodingInfo * __restrict ei,
158     wchar_t * __restrict pwc, char ** __restrict s, size_t n,
159     _MSKanjiState * __restrict psenc, size_t * __restrict nresult)
160 {
161 	char *s0;
162 	wchar_t wchar;
163 	int chlenbak, len;
164 
165 	s0 = *s;
166 
167 	if (s0 == NULL) {
168 		_citrus_MSKanji_init_state(ei, psenc);
169 		*nresult = 0; /* state independent */
170 		return (0);
171 	}
172 
173 	chlenbak = psenc->chlen;
174 
175 	/* make sure we have the first byte in the buffer */
176 	switch (psenc->chlen) {
177 	case 0:
178 		if (n < 1)
179 			goto restart;
180 		psenc->ch[0] = *s0++;
181 		psenc->chlen = 1;
182 		n--;
183 		break;
184 	case 1:
185 		break;
186 	default:
187 		/* illegal state */
188 		goto encoding_error;
189 	}
190 
191 	len = _mskanji1(psenc->ch[0] & 0xff) ? 2 : 1;
192 	while (psenc->chlen < len) {
193 		if (n < 1)
194 			goto restart;
195 		psenc->ch[psenc->chlen] = *s0++;
196 		psenc->chlen++;
197 		n--;
198 	}
199 
200 	*s = s0;
201 
202 	switch (len) {
203 	case 1:
204 		wchar = psenc->ch[0] & 0xff;
205 		break;
206 	case 2:
207 		if (!_mskanji2(psenc->ch[1] & 0xff))
208 			goto encoding_error;
209 		wchar = ((psenc->ch[0] & 0xff) << 8) | (psenc->ch[1] & 0xff);
210 		break;
211 	default:
212 		/* illegal state */
213 		goto encoding_error;
214 	}
215 
216 	psenc->chlen = 0;
217 
218 	if (pwc)
219 		*pwc = wchar;
220 	*nresult = wchar ? len - chlenbak : 0;
221 	return (0);
222 
223 encoding_error:
224 	psenc->chlen = 0;
225 	*nresult = (size_t)-1;
226 	return (EILSEQ);
227 
228 restart:
229 	*nresult = (size_t)-2;
230 	*s = s0;
231 	return (0);
232 }
233 
234 
235 static int
236 _citrus_MSKanji_wcrtomb_priv(_MSKanjiEncodingInfo * __restrict ei __unused,
237     char * __restrict s, size_t n, wchar_t wc,
238     _MSKanjiState * __restrict psenc __unused, size_t * __restrict nresult)
239 {
240 	int ret;
241 
242 	/* check invalid sequence */
243 	if (wc & ~0xffff) {
244 		ret = EILSEQ;
245 		goto err;
246 	}
247 
248 	if (wc & 0xff00) {
249 		if (n < 2) {
250 			ret = E2BIG;
251 			goto err;
252 		}
253 
254 		s[0] = (wc >> 8) & 0xff;
255 		s[1] = wc & 0xff;
256 		if (!_mskanji1(s[0] & 0xff) || !_mskanji2(s[1] & 0xff)) {
257 			ret = EILSEQ;
258 			goto err;
259 		}
260 
261 		*nresult = 2;
262 		return (0);
263 	} else {
264 		if (n < 1) {
265 			ret = E2BIG;
266 			goto err;
267 		}
268 
269 		s[0] = wc & 0xff;
270 		if (_mskanji1(s[0] & 0xff)) {
271 			ret = EILSEQ;
272 			goto err;
273 		}
274 
275 		*nresult = 1;
276 		return (0);
277 	}
278 
279 err:
280 	*nresult = (size_t)-1;
281 	return (ret);
282 }
283 
284 
285 static __inline int
286 /*ARGSUSED*/
287 _citrus_MSKanji_stdenc_wctocs(_MSKanjiEncodingInfo * __restrict ei,
288     _csid_t * __restrict csid, _index_t * __restrict idx, wchar_t wc)
289 {
290 	_index_t col, row;
291 	int offset;
292 
293 	if ((_wc_t)wc < 0x80) {
294 		/* ISO-646 */
295 		*csid = 0;
296 		*idx = (_index_t)wc;
297 	} else if ((_wc_t)wc < 0x100) {
298 		/* KANA */
299 		*csid = 1;
300 		*idx = (_index_t)wc & 0x7F;
301 	} else {
302 		/* Kanji (containing Gaiji zone) */
303 		/*
304 		 * 94^2 zone (contains a part of Gaiji (0xED40 - 0xEEFC)):
305 		 * 0x8140 - 0x817E -> 0x2121 - 0x215F
306 		 * 0x8180 - 0x819E -> 0x2160 - 0x217E
307 		 * 0x819F - 0x81FC -> 0x2221 - 0x227E
308 		 *
309 		 * 0x8240 - 0x827E -> 0x2321 - 0x235F
310 		 *  ...
311 		 * 0x9F9F - 0x9FFc -> 0x5E21 - 0x5E7E
312 		 *
313 		 * 0xE040 - 0xE07E -> 0x5F21 - 0x5F5F
314 		 *  ...
315 		 * 0xEF9F - 0xEFFC -> 0x7E21 - 0x7E7E
316 		 *
317 		 * extended Gaiji zone:
318 		 * 0xF040 - 0xFCFC
319 		 *
320 		 * JIS X0213-plane2:
321 		 * 0xF040 - 0xF09E -> 0x2121 - 0x217E
322 		 * 0xF140 - 0xF19E -> 0x2321 - 0x237E
323 		 * ...
324 		 * 0xF240 - 0xF29E -> 0x2521 - 0x257E
325 		 *
326 		 * 0xF09F - 0xF0FC -> 0x2821 - 0x287E
327 		 * 0xF29F - 0xF2FC -> 0x2C21 - 0x2C7E
328 		 * ...
329 		 * 0xF44F - 0xF49E -> 0x2F21 - 0x2F7E
330 		 *
331 		 * 0xF49F - 0xF4FC -> 0x6E21 - 0x6E7E
332 		 * ...
333 		 * 0xFC9F - 0xFCFC -> 0x7E21 - 0x7E7E
334 		 */
335 		row = ((_wc_t)wc >> 8) & 0xFF;
336 		col = (_wc_t)wc & 0xFF;
337 		if (!_mskanji1(row) || !_mskanji2(col))
338 			return (EILSEQ);
339 		if ((ei->mode & MODE_JIS2004) == 0 || row < 0xF0) {
340 			*csid = 2;
341 			offset = 0x81;
342 		} else {
343 			*csid = 3;
344 			if ((_wc_t)wc <= 0xF49E) {
345 				offset = (_wc_t)wc >= 0xF29F ||
346 				    ((_wc_t)wc >= 0xF09F &&
347 				    (_wc_t)wc <= 0xF0FC) ? 0xED : 0xF0;
348 			} else
349 				offset = 0xCE;
350 		}
351 		row -= offset;
352 		if (row >= 0x5F)
353 			row -= 0x40;
354 		row = row * 2 + 0x21;
355 		col -= 0x1F;
356 		if (col >= 0x61)
357 			col -= 1;
358 		if (col > 0x7E) {
359 			row += 1;
360 			col -= 0x5E;
361 		}
362 		*idx = ((_index_t)row << 8) | col;
363 	}
364 
365 	return (0);
366 }
367 
368 static __inline int
369 /*ARGSUSED*/
370 _citrus_MSKanji_stdenc_cstowc(_MSKanjiEncodingInfo * __restrict ei,
371     wchar_t * __restrict wc, _csid_t csid, _index_t idx)
372 {
373 	uint32_t col, row;
374 	int offset;
375 
376 	switch (csid) {
377 	case 0:
378 		/* ISO-646 */
379 		if (idx >= 0x80)
380 			return (EILSEQ);
381 		*wc = (wchar_t)idx;
382 		break;
383 	case 1:
384 		/* kana */
385 		if (idx >= 0x80)
386 			return (EILSEQ);
387 		*wc = (wchar_t)idx + 0x80;
388 		break;
389 	case 3:
390 		if ((ei->mode & MODE_JIS2004) == 0)
391 			return (EILSEQ);
392 	/*FALLTHROUGH*/
393 	case 2:
394 		/* kanji */
395 		row = (idx >> 8);
396 		if (row < 0x21)
397 			return (EILSEQ);
398 		if (csid == 3) {
399 			if (row <= 0x2F)
400 				offset = (row == 0x22 || row >= 0x26) ?
401 				    0xED : 0xF0;
402 			else if (row >= 0x4D && row <= 0x7E)
403 				offset = 0xCE;
404 			else
405 				return (EILSEQ);
406 		} else {
407 			if (row > 0x97)
408 				return (EILSEQ);
409 			offset = (row < 0x5F) ? 0x81 : 0xC1;
410 		}
411 		col = idx & 0xFF;
412 		if (col < 0x21 || col > 0x7E)
413 			return (EILSEQ);
414 		row -= 0x21; col -= 0x21;
415 		if ((row & 1) == 0) {
416 			col += 0x40;
417 			if (col >= 0x7F)
418 				col += 1;
419 		} else
420 			col += 0x9F;
421 		row = row / 2 + offset;
422 		*wc = ((wchar_t)row << 8) | col;
423 		break;
424 	default:
425 		return (EILSEQ);
426 	}
427 
428 	return (0);
429 }
430 
431 static __inline int
432 /*ARGSUSED*/
433 _citrus_MSKanji_stdenc_get_state_desc_generic(_MSKanjiEncodingInfo * __restrict ei __unused,
434     _MSKanjiState * __restrict psenc, int * __restrict rstate)
435 {
436 
437 	*rstate = (psenc->chlen == 0) ? _STDENC_SDGEN_INITIAL :
438 	    _STDENC_SDGEN_INCOMPLETE_CHAR;
439 	return (0);
440 }
441 
442 static int
443 /*ARGSUSED*/
444 _citrus_MSKanji_encoding_module_init(_MSKanjiEncodingInfo *  __restrict ei,
445     const void * __restrict var, size_t lenvar)
446 {
447 	const char *p;
448 
449 	p = var;
450 	memset((void *)ei, 0, sizeof(*ei));
451 	while (lenvar > 0) {
452 		switch (_bcs_toupper(*p)) {
453 		case 'J':
454 			MATCH(JIS2004, ei->mode |= MODE_JIS2004);
455 			break;
456 		}
457 		++p;
458 		--lenvar;
459 	}
460 
461 	return (0);
462 }
463 
464 static void
465 _citrus_MSKanji_encoding_module_uninit(_MSKanjiEncodingInfo *ei __unused)
466 {
467 
468 }
469 
470 /* ----------------------------------------------------------------------
471  * public interface for stdenc
472  */
473 
474 _CITRUS_STDENC_DECLS(MSKanji);
475 _CITRUS_STDENC_DEF_OPS(MSKanji);
476 
477 #include "citrus_stdenc_template.h"
478