1 |
148 |
jeremybenn |
/*
|
2 |
|
|
* Copyright (c) 2003-2004, Artem B. Bityuckiy
|
3 |
|
|
* Copyright (c) 1999,2000, Konstantin Chuguev. All rights reserved.
|
4 |
|
|
*
|
5 |
|
|
* Redistribution and use in source and binary forms, with or without
|
6 |
|
|
* modification, are permitted provided that the following conditions
|
7 |
|
|
* are met:
|
8 |
|
|
* 1. Redistributions of source code must retain the above copyright
|
9 |
|
|
* notice, this list of conditions and the following disclaimer.
|
10 |
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
11 |
|
|
* notice, this list of conditions and the following disclaimer in the
|
12 |
|
|
* documentation and/or other materials provided with the distribution.
|
13 |
|
|
*
|
14 |
|
|
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
|
15 |
|
|
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
16 |
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
17 |
|
|
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
|
18 |
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
19 |
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
20 |
|
|
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
21 |
|
|
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
22 |
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
23 |
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
24 |
|
|
* SUCH DAMAGE.
|
25 |
|
|
*/
|
26 |
|
|
#ifndef __ICONV_UCS_CONVERSION_H__
|
27 |
|
|
#define __ICONV_UCS_CONVERSION_H__
|
28 |
|
|
|
29 |
|
|
#include <_ansi.h>
|
30 |
|
|
#include <reent.h>
|
31 |
|
|
#include <sys/types.h>
|
32 |
|
|
#include <wchar.h>
|
33 |
|
|
#include "local.h"
|
34 |
|
|
|
35 |
|
|
/* No enough space in output buffer */
|
36 |
|
|
#define ICONV_CES_NOSPACE 0
|
37 |
|
|
/* Invalid input character */
|
38 |
|
|
#define ICONV_CES_INVALID_CHARACTER -1
|
39 |
|
|
/* No corespondent character in destination encoding */
|
40 |
|
|
#define ICONV_CES_BAD_SEQUENCE -2
|
41 |
|
|
/* All unknown characters are marked by this code */
|
42 |
|
|
#define DEFAULT_CHARACTER 0x3f /* ASCII '?' */
|
43 |
|
|
|
44 |
|
|
|
45 |
|
|
/*
|
46 |
|
|
* iconv_to_ucs_ces_handlers_t - "to UCS" CES converter handlers.
|
47 |
|
|
*
|
48 |
|
|
* Structure contains function pointers which should be provided by
|
49 |
|
|
* "to_ucs" CES converter.
|
50 |
|
|
*
|
51 |
|
|
* ============================================================================
|
52 |
|
|
*/
|
53 |
|
|
typedef struct
|
54 |
|
|
{
|
55 |
|
|
/*
|
56 |
|
|
* init - initialize CES converter.
|
57 |
|
|
*
|
58 |
|
|
* PARAMETERS:
|
59 |
|
|
* struct _reent *rptr - reent structure of current thread/process;
|
60 |
|
|
* _CONST char *encoding - encoding name.
|
61 |
|
|
*
|
62 |
|
|
* DESCRIPTION:
|
63 |
|
|
* Initializes CES converter. CES converter may deal with a series of
|
64 |
|
|
* encodings, such as Table or EUC CES converters. 'encoding' parameter
|
65 |
|
|
* indicates which encoding CES converter should use.
|
66 |
|
|
*
|
67 |
|
|
* RETURN:
|
68 |
|
|
* Returns CES-specific data pointer if success. In case of error returns
|
69 |
|
|
* NULL and sets current thread's/process's errno.
|
70 |
|
|
*/
|
71 |
|
|
_VOID_PTR _EXPARM(init, (struct _reent *rptr,
|
72 |
|
|
_CONST char *encoding));
|
73 |
|
|
|
74 |
|
|
/*
|
75 |
|
|
* close - close CES converter.
|
76 |
|
|
*
|
77 |
|
|
* PARAMETERS:
|
78 |
|
|
* struct _reent *rptr - reent structure of current thread/process;
|
79 |
|
|
* _VOID_PTR data - CES converter-specific data.
|
80 |
|
|
*
|
81 |
|
|
* DESCRIPTION:
|
82 |
|
|
* Preforms CES converter closing. *
|
83 |
|
|
* RETURN:
|
84 |
|
|
* Returns (size_t)0 if success. In case of error returns (size_t)-1 and
|
85 |
|
|
* sets current thread's/process's errno.
|
86 |
|
|
*/
|
87 |
|
|
size_t _EXPARM(close, (struct _reent *rptr,
|
88 |
|
|
_VOID_PTR data));
|
89 |
|
|
|
90 |
|
|
/*
|
91 |
|
|
* get_mb_cur_max - get maximum character length in bytes.
|
92 |
|
|
*
|
93 |
|
|
* PARAMETERS:
|
94 |
|
|
* _VOID_PTR data - conversion-specific data;
|
95 |
|
|
*
|
96 |
|
|
* DESCRIPTION:
|
97 |
|
|
* Returns encoding's maximum character length.
|
98 |
|
|
*/
|
99 |
|
|
int _EXPARM(get_mb_cur_max, (_VOID_PTR data));
|
100 |
|
|
|
101 |
|
|
/*
|
102 |
|
|
* get_state - get current shift state.
|
103 |
|
|
*
|
104 |
|
|
* PARAMETERS:
|
105 |
|
|
* _VOID_PTR data - conversion-specific data;
|
106 |
|
|
* mbstate_t *state - mbstate_t object where shift state will be stored;
|
107 |
|
|
*
|
108 |
|
|
* DESCRIPTION:
|
109 |
|
|
* Returns encoding's current shift sequence.
|
110 |
|
|
*/
|
111 |
|
|
_VOID _EXPARM(get_state, (_VOID_PTR data,
|
112 |
|
|
mbstate_t *state));
|
113 |
|
|
|
114 |
|
|
/*
|
115 |
|
|
* set_state - set shift state.
|
116 |
|
|
*
|
117 |
|
|
* PARAMETERS:
|
118 |
|
|
* _VOID_PTR data - conversion-specific data;
|
119 |
|
|
* mbstate_t *state - mbstate_t value to which shift state will be set.
|
120 |
|
|
*
|
121 |
|
|
* DESCRIPTION:
|
122 |
|
|
* Sets encoding's current shift state to 'state'. if 'state'
|
123 |
|
|
* object is zero-object - reset current shift state.
|
124 |
|
|
* Returns 0 if '*state' object has right format, -1 else.
|
125 |
|
|
*/
|
126 |
|
|
int _EXPARM(set_state, (_VOID_PTR data,
|
127 |
|
|
mbstate_t *state));
|
128 |
|
|
|
129 |
|
|
/*
|
130 |
|
|
* is_stateful - is encoding stateful state.
|
131 |
|
|
*
|
132 |
|
|
* PARAMETERS:
|
133 |
|
|
* _VOID_PTR data - conversion-specific data;
|
134 |
|
|
*
|
135 |
|
|
* DESCRIPTION:
|
136 |
|
|
* Returns 0 if encoding is stateless, else returns 1.
|
137 |
|
|
*/
|
138 |
|
|
int _EXPARM(is_stateful, (_VOID_PTR data));
|
139 |
|
|
|
140 |
|
|
/*
|
141 |
|
|
* convert_to_ucs - convert character to UCS.
|
142 |
|
|
*
|
143 |
|
|
* PARAMETERS:
|
144 |
|
|
* _VOID_PTR data - CES converter-specific data;
|
145 |
|
|
* _CONST unsigned char **inbuf - buffer with input character byte sequence;
|
146 |
|
|
* size_t *inbytesleft - output buffer bytes count.
|
147 |
|
|
*
|
148 |
|
|
* DESCRIPTION:
|
149 |
|
|
* Converts input characters into UCS encoding. 'inbuf' is
|
150 |
|
|
* incremented accordingly. 'bytesleft' is decremented accordingly. Should
|
151 |
|
|
* be provided by correspondent CES module.
|
152 |
|
|
*
|
153 |
|
|
* RETURN:
|
154 |
|
|
* Returns resulting UCS code if success. If input character is invalid,
|
155 |
|
|
* returns ICONV_CES_INVALID_CHARACTER. If invalid or incomplete bytes
|
156 |
|
|
* sequence was met, returns ICONV_CES_BAD_SEQUENCE.
|
157 |
|
|
*/
|
158 |
|
|
ucs4_t _EXPARM(convert_to_ucs, (_VOID_PTR data,
|
159 |
|
|
_CONST unsigned char **inbuf,
|
160 |
|
|
size_t *inbytesleft));
|
161 |
|
|
} iconv_to_ucs_ces_handlers_t;
|
162 |
|
|
|
163 |
|
|
|
164 |
|
|
/*
|
165 |
|
|
* iconv_from_ucs_ces_handlers_t - "from UCS" CES converter handlers.
|
166 |
|
|
*
|
167 |
|
|
* Structure contains function pointers which should be provided by
|
168 |
|
|
* "from_ucs" CES converter.
|
169 |
|
|
*
|
170 |
|
|
* ============================================================================
|
171 |
|
|
*/
|
172 |
|
|
typedef struct
|
173 |
|
|
{
|
174 |
|
|
/* Same as in iconv_to_ucs_ces_handlers_t */
|
175 |
|
|
_VOID_PTR _EXPARM(init, (struct _reent *rptr,
|
176 |
|
|
_CONST char *encoding));
|
177 |
|
|
|
178 |
|
|
/* Same as in iconv_to_ucs_ces_handlers_t */
|
179 |
|
|
size_t _EXPARM(close, (struct _reent *rptr,
|
180 |
|
|
_VOID_PTR data));
|
181 |
|
|
|
182 |
|
|
/* Same as in iconv_to_ucs_ces_handlers_t */
|
183 |
|
|
int _EXPARM(get_mb_cur_max, (_VOID_PTR data));
|
184 |
|
|
|
185 |
|
|
/* Same as in iconv_to_ucs_ces_handlers_t */
|
186 |
|
|
_VOID _EXPARM(get_state, (_VOID_PTR data,
|
187 |
|
|
mbstate_t *state));
|
188 |
|
|
|
189 |
|
|
/* Same as in iconv_to_ucs_ces_handlers_t */
|
190 |
|
|
int _EXPARM(set_state, (_VOID_PTR data,
|
191 |
|
|
mbstate_t *state));
|
192 |
|
|
|
193 |
|
|
/* Same as in iconv_to_ucs_ces_handlers_t */
|
194 |
|
|
int _EXPARM(is_stateful, (_VOID_PTR data));
|
195 |
|
|
|
196 |
|
|
/*
|
197 |
|
|
* convert_from_ucs - convert UCS character to destination encoding.
|
198 |
|
|
*
|
199 |
|
|
* PARAMETERS:
|
200 |
|
|
* _VOID_PTR data - CES converter-specific data;
|
201 |
|
|
* ucs4_t in - input UCS-4 character;
|
202 |
|
|
* unsigned char **outbuf - output buffer for the result;
|
203 |
|
|
* size_t *outbytesleft - output buffer bytes count.
|
204 |
|
|
*
|
205 |
|
|
* DESCRIPTION:
|
206 |
|
|
* Converts input UCS characters to destination encoding and stores result
|
207 |
|
|
* in 'outbuf' if there is sufficient free space present. 'outbuf' is
|
208 |
|
|
* incremented accordingly. 'outbytesleft' is decremented accordingly. Should
|
209 |
|
|
* be provided by correspondent CES module.
|
210 |
|
|
* Output buffer always has at least 1 byte.
|
211 |
|
|
*
|
212 |
|
|
* RETURN:
|
213 |
|
|
* Returns number of bytes that was written into output buffer if success.
|
214 |
|
|
* If there is no enough space in output buffer, returns ICONV_CES_NOSPACE.
|
215 |
|
|
* If there is no corresponding character in destination encoding, returns
|
216 |
|
|
* ICONV_CES_INVALID_CHARACTER.
|
217 |
|
|
*/
|
218 |
|
|
size_t _EXPARM(convert_from_ucs, (_VOID_PTR data,
|
219 |
|
|
ucs4_t in,
|
220 |
|
|
unsigned char **outbuf,
|
221 |
|
|
size_t *outbytesleft));
|
222 |
|
|
} iconv_from_ucs_ces_handlers_t;
|
223 |
|
|
|
224 |
|
|
|
225 |
|
|
/*
|
226 |
|
|
* iconv_to_ucs_ces_desc_t - "to UCS" CES converter definition structure for
|
227 |
|
|
* usage in iconv_ucs_conversion_t conversion description structure.
|
228 |
|
|
*
|
229 |
|
|
* ============================================================================
|
230 |
|
|
*/
|
231 |
|
|
typedef struct
|
232 |
|
|
{
|
233 |
|
|
/* CES converter handlers */
|
234 |
|
|
_CONST iconv_to_ucs_ces_handlers_t *handlers;
|
235 |
|
|
|
236 |
|
|
/* "to_ucs" CES converter-specific data. */
|
237 |
|
|
_VOID_PTR data;
|
238 |
|
|
} iconv_to_ucs_ces_desc_t;
|
239 |
|
|
|
240 |
|
|
|
241 |
|
|
/*
|
242 |
|
|
* iconv_from_ucs_ces_desc_t - "from UCS" CES converter definition structure for
|
243 |
|
|
* usage in iconv_ucs_conversion_t conversion description structure.
|
244 |
|
|
*
|
245 |
|
|
* ============================================================================
|
246 |
|
|
*/
|
247 |
|
|
typedef struct
|
248 |
|
|
{
|
249 |
|
|
/* CES converter handlers */
|
250 |
|
|
_CONST iconv_from_ucs_ces_handlers_t *handlers;
|
251 |
|
|
|
252 |
|
|
/* "from_ucs" CES converter-specific data. */
|
253 |
|
|
_VOID_PTR data;
|
254 |
|
|
} iconv_from_ucs_ces_desc_t;
|
255 |
|
|
|
256 |
|
|
|
257 |
|
|
/*
|
258 |
|
|
* iconv_ucs_conversion_t - UCS-based conversion definition structure.
|
259 |
|
|
*
|
260 |
|
|
* Defines special type of conversion where every character is first
|
261 |
|
|
* converted into UCS-4 (UCS-2 for table-driven), and after this the
|
262 |
|
|
* resulting UCS character is converted to destination encoding.
|
263 |
|
|
* UCS-based conversion is composed of two *converters*, defined by
|
264 |
|
|
* iconv_ces_t structure. The iconv_ucs_conversion_t object is referred
|
265 |
|
|
* from iconv_conversion_t object using 'data' field.
|
266 |
|
|
*
|
267 |
|
|
* Structure contains two objects - 'to_ucs' and 'from_ucs' which define
|
268 |
|
|
* "source encoding to UCS" and "UCS to destination encoding" converters.
|
269 |
|
|
*
|
270 |
|
|
* ============================================================================
|
271 |
|
|
*/
|
272 |
|
|
typedef struct
|
273 |
|
|
{
|
274 |
|
|
/* Source encoding -> CES converter. */
|
275 |
|
|
iconv_to_ucs_ces_desc_t to_ucs;
|
276 |
|
|
|
277 |
|
|
/* UCS -> destination encoding CES converter. */
|
278 |
|
|
iconv_from_ucs_ces_desc_t from_ucs;
|
279 |
|
|
} iconv_ucs_conversion_t;
|
280 |
|
|
|
281 |
|
|
|
282 |
|
|
/*
|
283 |
|
|
* iconv_to_ucs_ces_t - defines "to UCS" CES converter.
|
284 |
|
|
*
|
285 |
|
|
* ============================================================================
|
286 |
|
|
*/
|
287 |
|
|
typedef struct
|
288 |
|
|
{
|
289 |
|
|
/*
|
290 |
|
|
* An array of encodings names, supported by CES converter.
|
291 |
|
|
* The end of array should be marked by NULL pointer.
|
292 |
|
|
*/
|
293 |
|
|
_CONST char **names;
|
294 |
|
|
|
295 |
|
|
/* CES converter description structure */
|
296 |
|
|
_CONST iconv_to_ucs_ces_handlers_t *handlers;
|
297 |
|
|
} iconv_to_ucs_ces_t;
|
298 |
|
|
|
299 |
|
|
|
300 |
|
|
/*
|
301 |
|
|
* iconv_from_ucs_ces_t - defines "from UCS" CES converter.
|
302 |
|
|
*
|
303 |
|
|
* ============================================================================
|
304 |
|
|
*/
|
305 |
|
|
typedef struct
|
306 |
|
|
{
|
307 |
|
|
/*
|
308 |
|
|
* An array of encodings names, supported by CES converter.
|
309 |
|
|
* The end of array should be marked by NULL pointer.
|
310 |
|
|
*/
|
311 |
|
|
_CONST char **names;
|
312 |
|
|
|
313 |
|
|
/* CES converter description structure */
|
314 |
|
|
_CONST iconv_from_ucs_ces_handlers_t *handlers;
|
315 |
|
|
} iconv_from_ucs_ces_t;
|
316 |
|
|
|
317 |
|
|
|
318 |
|
|
/* List of "to UCS" linked-in CES converters. */
|
319 |
|
|
extern _CONST iconv_to_ucs_ces_t
|
320 |
|
|
_iconv_to_ucs_ces[];
|
321 |
|
|
|
322 |
|
|
/* List of "from UCS" linked-in CES converters. */
|
323 |
|
|
extern _CONST iconv_from_ucs_ces_t
|
324 |
|
|
_iconv_from_ucs_ces[];
|
325 |
|
|
|
326 |
|
|
#endif /* !__ICONV_UCS_CONVERSION_H__ */
|
327 |
|
|
|