1 /* $NetBSD: citrus_ues.c,v 1.1 2006/11/13 15:16:31 tnozaki Exp $ */
2 /* $DragonFly: src/lib/libc/citrus/modules/citrus_ues.c,v 1.1 2008/04/10 10:21:02 hasso Exp $ */
5 * Copyright (c)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
39 #include "citrus_namespace.h"
40 #include "citrus_types.h"
41 #include "citrus_bcs.h"
42 #include "citrus_module.h"
43 #include "citrus_ctype.h"
44 #include "citrus_stdenc.h"
45 #include "citrus_ues.h"
61 /* for future multi-locale facility */
66 _UESState s_mbsrtowcs;
68 _UESState s_wcsrtombs;
73 #define _CEI_TO_EI(_cei_) (&(_cei_)->ei)
74 #define _CEI_TO_STATE(_cei_, _func_) (_cei_)->states.s_##_func_
76 #define _FUNCNAME(m) _citrus_UES_##m
77 #define _ENCODING_INFO _UESEncodingInfo
78 #define _CTYPE_INFO _UESCTypeInfo
79 #define _ENCODING_STATE _UESState
80 #define _ENCODING_MB_CUR_MAX(_ei_) (_ei_)->mb_cur_max
81 #define _ENCODING_IS_STATE_DEPENDENT 0
82 #define _STATE_NEEDS_EXPLICIT_INIT(_ps_) 0
86 _citrus_UES_init_state(_UESEncodingInfo * __restrict ei,
87 _UESState * __restrict psenc)
94 _citrus_UES_pack_state(_UESEncodingInfo * __restrict ei,
95 void *__restrict pspriv, const _UESState * __restrict psenc)
97 /* ei seem to be unused */
98 _DIAGASSERT(pspriv != NULL);
99 _DIAGASSERT(psenc != NULL);
101 memcpy(pspriv, (const void *)psenc, sizeof(*psenc));
106 _citrus_UES_unpack_state(_UESEncodingInfo * __restrict ei,
107 _UESState * __restrict psenc, const void * __restrict pspriv)
109 /* ei seem to be unused */
110 _DIAGASSERT(psenc != NULL);
111 _DIAGASSERT(pspriv != NULL);
113 memcpy((void *)psenc, pspriv, sizeof(*psenc));
119 if (ch >= '0' && ch <= '9')
121 else if (ch >= 'A' && ch <= 'F')
122 return (ch - 'A') + 10;
123 else if (ch >= 'a' && ch <= 'f')
124 return (ch - 'a') + 10;
134 #define BMP_MAX UINT32_C(0xFFFF)
135 #define UCS2_MAX UINT32_C(0x10FFFF)
136 #define UCS4_MAX UINT32_C(0x7FFFFFFF)
138 static const char *xdig = "0123456789abcdef";
141 to_str(char *s, wchar_t wc, int bit)
158 *p++ = xdig[(wc >> (bit -= 4)) & 0xF];
164 is_hi_surrogate(wchar_t wc)
166 return wc >= 0xD800 && wc <= 0xDBFF;
170 is_lo_surrogate(wchar_t wc)
172 return wc >= 0xDC00 && wc <= 0xDFFF;
175 static __inline wchar_t
176 surrogate_to_ucs(wchar_t hi, wchar_t lo)
178 _DIAGASSERT(is_hi_surrogate(hi));
179 _DIAGASSERT(is_lo_surrogate(lo));
183 return (hi << 10 | lo) + 0x10000;
187 ucs_to_surrogate(wchar_t wc, wchar_t * __restrict hi, wchar_t * __restrict lo)
189 _DIAGASSERT(hi != NULL);
190 _DIAGASSERT(lo != NULL);
191 _DIAGASSERT(wc >= 0x10000);
194 *hi = (wc >> 10) + 0xD800;
195 *lo = (wc & 0x3FF) + 0xDC00;
201 return (uint32_t)wc <= 0x9F &&
202 wc != 0x24 && wc != 0x40 && wc != 0x60;
206 _citrus_UES_mbrtowc_priv(_UESEncodingInfo * __restrict ei,
207 wchar_t * __restrict pwc, const char ** __restrict s, size_t n,
208 _UESState * __restrict psenc, size_t * __restrict nresult)
211 int ch, head, tail, i, num;
214 _DIAGASSERT(ei != NULL);
215 /* pwc may be null */
216 _DIAGASSERT(s != NULL);
217 _DIAGASSERT(psenc != NULL);
218 _DIAGASSERT(nresult != NULL);
221 _citrus_UES_init_state(ei, psenc);
233 if (psenc->chlen == head) {
236 psenc->ch[psenc->chlen++] = *s0++;
238 ch = (unsigned char)psenc->ch[head++];
240 if (psenc->chlen == head) {
243 psenc->ch[psenc->chlen++] = *s0++;
245 switch (psenc->ch[head]) {
250 if (ei->mode & MODE_C99) {
260 for (; head < tail; ++head) {
261 if (psenc->chlen == head) {
265 *nresult = (size_t)-2;
268 psenc->ch[psenc->chlen++] = *s0++;
270 num = to_int((int)(unsigned char)psenc->ch[head]);
275 wc = (wc << 4) | num;
282 if (hi != (wchar_t)0)
284 if ((ei->mode & MODE_C99) == 0) {
285 if (is_hi_surrogate(wc) != 0) {
289 if ((uint32_t)wc <= 0x7F /* XXX */ ||
290 is_lo_surrogate(wc) != 0)
296 if (is_basic(wc) == 0 && (uint32_t)wc <= UCS4_MAX &&
297 is_hi_surrogate(wc) == 0 && is_lo_surrogate(wc) == 0)
299 *nresult = (size_t)-1;
302 if (is_lo_surrogate(wc) == 0)
304 wc = surrogate_to_ucs(hi, wc);
307 ch = (unsigned char)psenc->ch[0];
310 memmove(&psenc->ch[0], &psenc->ch[1], head);
316 *nresult = (size_t)((wc == 0) ? 0 : (s0 - *s));
323 _citrus_UES_wcrtomb_priv(_UESEncodingInfo * __restrict ei,
324 char * __restrict s, size_t n, wchar_t wc,
325 _UESState * __restrict psenc, size_t * __restrict nresult)
329 if (psenc->chlen != 0)
332 if ((ei->mode & MODE_C99) ? is_basic(wc) : (uint32_t)wc <= 0x7F) {
335 psenc->ch[psenc->chlen++] = (char)wc;
336 } else if ((uint32_t)wc <= BMP_MAX) {
339 psenc->chlen = to_str(&psenc->ch[0], wc, UCS2_BIT);
340 } else if ((ei->mode & MODE_C99) == 0 && (uint32_t)wc <= UCS2_MAX) {
343 ucs_to_surrogate(wc, &hi, &lo);
344 psenc->chlen += to_str(&psenc->ch[0], hi, UCS2_BIT);
345 psenc->chlen += to_str(&psenc->ch[6], lo, UCS2_BIT);
346 } else if ((ei->mode & MODE_C99) && (uint32_t)wc <= UCS4_MAX) {
349 psenc->chlen = to_str(&psenc->ch[0], wc, UCS4_BIT);
351 *nresult = (size_t)-1;
354 memcpy(s, psenc->ch, psenc->chlen);
355 *nresult = psenc->chlen;
361 *nresult = (size_t)-1;
366 _citrus_UES_stdenc_wctocs(_UESEncodingInfo * __restrict ei,
367 _csid_t * __restrict csid, _index_t * __restrict idx, wchar_t wc)
369 /* ei seem to be unused */
370 _DIAGASSERT(csid != NULL);
371 _DIAGASSERT(idx != NULL);
381 _citrus_UES_stdenc_cstowc(_UESEncodingInfo * __restrict ei,
382 wchar_t * __restrict wc, _csid_t csid, _index_t idx)
384 /* ei seem to be unused */
385 _DIAGASSERT(wc != NULL);
396 _citrus_UES_stdenc_get_state_desc_generic(_UESEncodingInfo * __restrict ei,
397 _UESState * __restrict psenc, int * __restrict rstate)
399 _DIAGASSERT(psenc != NULL);
400 _DIAGASSERT(rstate != NULL);
402 if (psenc->chlen == 0)
403 *rstate = _STDENC_SDGEN_INITIAL;
405 *rstate = _STDENC_SDGEN_INCOMPLETE_CHAR; /* XXX */
412 _citrus_UES_encoding_module_uninit(_UESEncodingInfo *ei)
414 /* ei seems to be unused */
419 _citrus_UES_encoding_module_init(_UESEncodingInfo * __restrict ei,
420 const void * __restrict var, size_t lenvar)
424 _DIAGASSERT(ei != NULL);
427 #define MATCH(x, act) \
429 if (lenvar >= (sizeof(#x)-1) && \
430 _bcs_strncasecmp(p, #x, sizeof(#x)-1) == 0) { \
432 lenvar -= sizeof(#x)-1; \
435 } while (/*CONSTCOND*/0)
436 memset((void *)ei, 0, sizeof(*ei));
438 switch (_bcs_toupper(*p)) {
440 MATCH(C99, ei->mode |= MODE_C99);
446 ei->mb_cur_max = (ei->mode & MODE_C99) ? 10 : 12;
451 /* ----------------------------------------------------------------------
452 * public interface for ctype
455 _CITRUS_CTYPE_DECLS(UES);
456 _CITRUS_CTYPE_DEF_OPS(UES);
458 #include "citrus_ctype_template.h"
460 /* ----------------------------------------------------------------------
461 * public interface for stdenc
464 _CITRUS_STDENC_DECLS(UES);
465 _CITRUS_STDENC_DEF_OPS(UES);
467 #include "citrus_stdenc_template.h"