1 /* $NetBSD: citrus_utf1632.c,v 1.8 2008/03/20 11:47:45 tnozaki Exp $ */
2 /* $DragonFly: src/lib/libc/citrus/modules/citrus_utf1632.c,v 1.2 2008/04/10 10:21:02 hasso Exp $ */
5 * Copyright (c)2003 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
30 #include <sys/types.h>
31 #include <sys/endian.h>
42 #include "citrus_namespace.h"
43 #include "citrus_types.h"
44 #include "citrus_module.h"
45 #include "citrus_stdenc.h"
46 #include "citrus_bcs.h"
48 #include "citrus_utf1632.h"
51 /* ----------------------------------------------------------------------
52 * private stuffs used by templates
64 #define _ENDIAN_UNKNOWN 0
66 #define _ENDIAN_LITTLE 2
68 #define _MODE_UTF32 0x00000001U
69 #define _MODE_FORCE_ENDIAN 0x00000002U
70 } _UTF1632EncodingInfo
;
72 #define _FUNCNAME(m) _citrus_UTF1632_##m
73 #define _ENCODING_INFO _UTF1632EncodingInfo
74 #define _ENCODING_STATE _UTF1632State
75 #define _ENCODING_MB_CUR_MAX(_ei_) ((_ei_)->cur_max)
76 #define _ENCODING_IS_STATE_DEPENDENT 0
77 #define _STATE_NEEDS_EXPLICIT_INIT(_ps_) 0
82 _citrus_UTF1632_init_state(_UTF1632EncodingInfo
*ei
, _UTF1632State
*s
)
84 memset(s
, 0, sizeof(*s
));
88 _citrus_UTF1632_mbrtowc_priv(_UTF1632EncodingInfo
*ei
, wchar_t *pwc
,
89 const char **s
, size_t n
, _UTF1632State
*psenc
,
92 int chlenbak
, endian
, needlen
;
97 _DIAGASSERT(nresult
!= 0);
98 _DIAGASSERT(ei
!= NULL
);
99 _DIAGASSERT(s
!= NULL
);
100 _DIAGASSERT(psenc
!= NULL
);
105 _citrus_UTF1632_init_state(ei
, psenc
);
106 *nresult
= 0; /* state independent */
111 chlenbak
= psenc
->chlen
;
114 if ((ei
->mode
& _MODE_UTF32
) != 0 || chlenbak
>=2)
119 while (chlenbak
< needlen
) {
122 psenc
->ch
[chlenbak
++] = *s0
++;
127 /* judge endian marker */
128 if ((ei
->mode
& _MODE_UTF32
) == 0) {
130 if (psenc
->ch
[0]==0xFE && psenc
->ch
[1]==0xFF) {
131 psenc
->current_endian
= _ENDIAN_BIG
;
134 } else if (psenc
->ch
[0]==0xFF && psenc
->ch
[1]==0xFE) {
135 psenc
->current_endian
= _ENDIAN_LITTLE
;
141 if (psenc
->ch
[0]==0x00 && psenc
->ch
[1]==0x00 &&
142 psenc
->ch
[2]==0xFE && psenc
->ch
[3]==0xFF) {
143 psenc
->current_endian
= _ENDIAN_BIG
;
146 } else if (psenc
->ch
[0]==0xFF && psenc
->ch
[1]==0xFE &&
147 psenc
->ch
[2]==0x00 && psenc
->ch
[3]==0x00) {
148 psenc
->current_endian
= _ENDIAN_LITTLE
;
153 if ((ei
->mode
& _MODE_FORCE_ENDIAN
) != 0 ||
154 psenc
->current_endian
== _ENDIAN_UNKNOWN
)
155 endian
= ei
->preffered_endian
;
157 endian
= psenc
->current_endian
;
160 if ((ei
->mode
& _MODE_UTF32
) == 0) {
166 ((wchar_t)psenc
->ch
[1] << 8));
170 ((wchar_t)psenc
->ch
[0] << 8));
175 if (wc
>= 0xD800 && wc
<= 0xDBFF) {
182 wc
-= 0xD800; /* wc : surrogate high (see above) */
186 if (psenc
->ch
[2]<0xDC || psenc
->ch
[2]>0xDF)
189 wc
|= (wchar_t)(psenc
->ch
[3] & 3) << 8;
192 if (psenc
->ch
[3]<0xDC || psenc
->ch
[3]>0xDF)
195 wc
|= (wchar_t)(psenc
->ch
[2] & 3) << 8;
207 ((wchar_t)psenc
->ch
[1] << 8) |
208 ((wchar_t)psenc
->ch
[2] << 16) |
209 ((wchar_t)psenc
->ch
[3] << 24));
213 ((wchar_t)psenc
->ch
[2] << 8) |
214 ((wchar_t)psenc
->ch
[1] << 16) |
215 ((wchar_t)psenc
->ch
[0] << 24));
220 if (wc
>= 0xD800 && wc
<= 0xDFFF)
233 *nresult
= (size_t)-1;
238 *nresult
= (size_t)-2;
239 psenc
->chlen
= chlenbak
;
245 _citrus_UTF1632_wcrtomb_priv(_UTF1632EncodingInfo
*ei
, char *s
, size_t n
,
246 wchar_t wc
, _UTF1632State
*psenc
,
250 static const char _bom
[4] = {
251 #if BYTE_ORDER == BIG_ENDIAN
252 0x00, 0x00, 0xFE, 0xFF,
254 0xFF, 0xFE, 0x00, 0x00,
257 const char *bom
= &_bom
[0];
260 _DIAGASSERT(ei
!= NULL
);
261 _DIAGASSERT(nresult
!= 0);
262 _DIAGASSERT(s
!= NULL
);
265 if (psenc
->current_endian
== _ENDIAN_UNKNOWN
) {
266 if ((ei
->mode
& _MODE_FORCE_ENDIAN
) == 0) {
267 if (ei
->mode
& _MODE_UTF32
) {
271 #if BYTE_ORDER == BIG_ENDIAN
280 psenc
->current_endian
= ei
->preffered_endian
;
284 if ((ei
->mode
& _MODE_UTF32
)==0) {
294 wc2
= (wc
& 0x3FF) | 0xDC00;
295 wc
= (wc
>>10) | 0xD800;
303 switch (psenc
->current_endian
) {
321 if (wc
>= 0xD800 && wc
<= 0xDFFF)
326 switch (psenc
->current_endian
) {
346 *nresult
= (size_t)-1;
349 *nresult
= (size_t)-1;
354 parse_variable(_UTF1632EncodingInfo
* __restrict ei
,
355 const void * __restrict var
, size_t lenvar
)
357 #define MATCH(x, act) \
359 if (lenvar >= (sizeof(#x)-1) && \
360 _bcs_strncasecmp(p, #x, sizeof(#x)-1) == 0) { \
362 lenvar -= sizeof(#x)-1; \
365 } while (/*CONSTCOND*/0)
372 MATCH(big
, ei
->preffered_endian
= _ENDIAN_BIG
);
376 MATCH(little
, ei
->preffered_endian
= _ENDIAN_LITTLE
);
380 MATCH(force
, ei
->mode
|= _MODE_FORCE_ENDIAN
);
384 MATCH(utf32
, ei
->mode
|= _MODE_UTF32
);
394 _citrus_UTF1632_encoding_module_init(_UTF1632EncodingInfo
* __restrict ei
,
395 const void * __restrict var
,
398 _DIAGASSERT(ei
!= NULL
);
400 memset((void *)ei
, 0, sizeof(*ei
));
402 parse_variable(ei
, var
, lenvar
);
404 if ((ei
->mode
&_MODE_UTF32
)==0)
405 ei
->cur_max
= 6; /* endian + surrogate */
407 ei
->cur_max
= 8; /* endian + normal */
409 if (ei
->preffered_endian
== _ENDIAN_UNKNOWN
) {
410 #if BYTE_ORDER == BIG_ENDIAN
411 ei
->preffered_endian
= _ENDIAN_BIG
;
413 ei
->preffered_endian
= _ENDIAN_LITTLE
;
422 _citrus_UTF1632_encoding_module_uninit(_UTF1632EncodingInfo
*ei
)
428 _citrus_UTF1632_stdenc_wctocs(_UTF1632EncodingInfo
* __restrict ei
,
429 _csid_t
* __restrict csid
,
430 _index_t
* __restrict idx
,
434 _DIAGASSERT(csid
!= NULL
&& idx
!= NULL
);
444 _citrus_UTF1632_stdenc_cstowc(_UTF1632EncodingInfo
* __restrict ei
,
445 _wc_t
* __restrict wc
,
446 _csid_t csid
, _index_t idx
)
449 _DIAGASSERT(wc
!= NULL
);
461 _citrus_UTF1632_stdenc_get_state_desc_generic(_UTF1632EncodingInfo
* __restrict ei
,
462 _UTF1632State
* __restrict psenc
,
463 int * __restrict rstate
)
466 if (psenc
->chlen
== 0)
467 *rstate
= _STDENC_SDGEN_INITIAL
;
469 *rstate
= _STDENC_SDGEN_INCOMPLETE_CHAR
;
474 /* ----------------------------------------------------------------------
475 * public interface for stdenc
478 _CITRUS_STDENC_DECLS(UTF1632
);
479 _CITRUS_STDENC_DEF_OPS(UTF1632
);
481 #include "citrus_stdenc_template.h"