1 /*
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 #include "cesbi.h"
27 
28 #if defined (ICONV_TO_UCS_CES_UCS_4) \
29  || defined (ICONV_FROM_UCS_CES_UCS_4)
30 
31 #include <_ansi.h>
32 #include <stdlib.h>
33 #include <string.h>
34 #include <sys/types.h>
35 #include "../lib/local.h"
36 #include "../lib/ucsconv.h"
37 #include "../lib/endian.h"
38 
39 /*
40  * BOM isn't supported. UCS-4 is Big Endian. Bad codes are rejected.
41  * Bad codes: 0x0000FFFF, 0x0000FFFE, 0x0000D800-0x0000DFFF,
42  * 0x7FFFFFFF-0xFFFFFFFF.
43  */
44 
45 #define UCS_4_BIG_ENDIAN     0
46 #define UCS_4_LITTLE_ENDIAN  1
47 
48 #define UCS_4   "ucs_4"
49 #define UCS_4BE "ucs_4be"
50 #define UCS_4LE "ucs_4le"
51 
52 static void *
ucs_4_init(const char * encoding)53 ucs_4_init (
54                    const char *encoding)
55 {
56   int *data;
57 
58   if ((data = (int *)malloc (sizeof(int))) == NULL)
59     return (void *)NULL;
60 
61   if (strcmp (encoding, UCS_4LE) == 0)
62     *data = UCS_4_LITTLE_ENDIAN;
63   else
64     *data = UCS_4_BIG_ENDIAN;
65 
66   return (void *)data;
67 }
68 
69 static size_t
ucs_4_close(void * data)70 ucs_4_close (
71                     void *data)
72 {
73   free(data);
74   return 0;
75 }
76 
77 
78 #if defined (ICONV_FROM_UCS_CES_UCS_4)
79 static size_t
ucs_4_convert_from_ucs(void * data,ucs4_t in,unsigned char ** outbuf,size_t * outbytesleft)80 ucs_4_convert_from_ucs (void *data,
81                                ucs4_t in,
82                                unsigned char **outbuf,
83                                size_t *outbytesleft)
84 {
85   if ((in  >= 0x0000D800 && in <= 0x0000DFFF) /* Surrogate character */
86       || in > 0x7FFFFFFF || in == 0x0000FFFF || in == 0x0000FFFE)
87     return (size_t)ICONV_CES_INVALID_CHARACTER;
88 
89   if (*outbytesleft < sizeof (ucs4_t))
90     return (size_t)ICONV_CES_NOSPACE;
91 
92   ucs4_t uc;
93 
94   if (*((int *)data) == UCS_4_BIG_ENDIAN)
95     uc = ICONV_HTOBEL (in);
96   else
97     uc = ICONV_HTOLEL (in);
98 
99   memcpy(*outbuf, &uc, sizeof (ucs4_t));
100 
101   *outbuf += sizeof (ucs4_t);
102   *outbytesleft -= sizeof (ucs4_t);
103 
104   return sizeof (ucs4_t);
105 }
106 #endif /* ICONV_FROM_UCS_CES_UCS_4 */
107 
108 #if defined (ICONV_TO_UCS_CES_UCS_4)
109 static ucs4_t
ucs_4_convert_to_ucs(void * data,const unsigned char ** inbuf,size_t * inbytesleft)110 ucs_4_convert_to_ucs (void *data,
111                              const unsigned char **inbuf,
112                              size_t *inbytesleft)
113 {
114   ucs4_t res;
115 
116   if (*inbytesleft < sizeof (ucs4_t))
117     return (ucs4_t)ICONV_CES_BAD_SEQUENCE;
118 
119   ucs4_t uc;
120 
121   memcpy(&uc, *inbuf, sizeof (ucs4_t));
122 
123   if (*((int *)data) == UCS_4_BIG_ENDIAN)
124     res = ICONV_BETOHL (uc);
125   else
126     res = ICONV_LETOHL (uc);
127 
128   if ((res  >= 0x0000D800 && res <= 0x0000DFFF) /* Surrogate character */
129       || res > 0x7FFFFFFF || res == 0x0000FFFF || res == 0x0000FFFE)
130     return (ucs4_t)ICONV_CES_INVALID_CHARACTER;
131 
132   *inbytesleft -= sizeof (ucs4_t);
133   *inbuf += sizeof(ucs4_t);
134 
135   return res;
136 }
137 #endif /* ICONV_TO_UCS_CES_UCS_4 */
138 
139 static int
ucs_4_get_mb_cur_max(void * data)140 ucs_4_get_mb_cur_max (void *data)
141 {
142   (void) data;
143   return 4;
144 }
145 
146 #if defined (ICONV_TO_UCS_CES_UCS_4)
147 const iconv_to_ucs_ces_handlers_t
148 _iconv_to_ucs_ces_handlers_ucs_4 =
149 {
150   ucs_4_init,
151   ucs_4_close,
152   ucs_4_get_mb_cur_max,
153   NULL,
154   NULL,
155   NULL,
156   ucs_4_convert_to_ucs
157 };
158 #endif
159 
160 #if defined (ICONV_FROM_UCS_CES_UCS_4)
161 const iconv_from_ucs_ces_handlers_t
162 _iconv_from_ucs_ces_handlers_ucs_4 =
163 {
164   ucs_4_init,
165   ucs_4_close,
166   ucs_4_get_mb_cur_max,
167   NULL,
168   NULL,
169   NULL,
170   ucs_4_convert_from_ucs
171 };
172 #endif
173 
174 #endif /* ICONV_TO_UCS_CES_UCS_4 || ICONV_FROM_UCS_CES_UCS_4 */
175 
176