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_2) \
29  || defined (ICONV_FROM_UCS_CES_UCS_2)
30 
31 #include <stdlib.h>
32 #include <string.h>
33 #include <sys/types.h>
34 #include "../lib/local.h"
35 #include "../lib/ucsconv.h"
36 #include "../lib/endian.h"
37 
38 /*
39  * BOM isn't supported. UCS-2 is Big Endian. Bad codes are rejected.
40  * Bad codes: 0xFFFF, 0xFFFE, 0xD800-0xDFFF.
41  */
42 
43 #define UCS_2_BIG_ENDIAN     0
44 #define UCS_2_LITTLE_ENDIAN  1
45 
46 #define UCS_2   "ucs_2"
47 #define UCS_2BE "ucs_2be"
48 #define UCS_2LE "ucs_2le"
49 
50 static void *
ucs_2_init(const char * encoding)51 ucs_2_init (
52                    const char *encoding)
53 {
54   int *data;
55 
56   if ((data = (int *) malloc(sizeof (int))) == NULL)
57     return (void *)NULL;
58 
59   if (strcmp (encoding, UCS_2LE) == 0)
60     *data = UCS_2_LITTLE_ENDIAN;
61   else
62     *data = UCS_2_BIG_ENDIAN;
63 
64   return (void *)data;
65 }
66 
67 static size_t
ucs_2_close(void * data)68 ucs_2_close (
69                     void *data)
70 {
71   free (data);
72   return 0;
73 }
74 
75 #if defined (ICONV_FROM_UCS_CES_UCS_2)
76 static size_t
ucs_2_convert_from_ucs(void * data,ucs4_t in,unsigned char ** outbuf,size_t * outbytesleft)77 ucs_2_convert_from_ucs (void *data,
78                                ucs4_t in,
79                                unsigned char **outbuf,
80                                size_t *outbytesleft)
81 {
82   if ((in  >= 0x0000D800 && in <= 0x0000DFFF) /* Surrogate character */
83       || in >= 0x0000FFFE)
84     return (size_t)ICONV_CES_INVALID_CHARACTER;
85 
86   if (*outbytesleft < sizeof (ucs2_t))
87     return (size_t)ICONV_CES_NOSPACE;
88 
89   ucs2_t uc;
90 
91   if (*((int *)data) == UCS_2_BIG_ENDIAN)
92     uc = ICONV_HTOBES ((ucs2_t)in);
93   else
94     uc = ICONV_HTOLES ((ucs2_t)in);
95 
96   memcpy(*outbuf, &uc, sizeof(ucs2_t));
97 
98   *outbuf += sizeof (ucs2_t);
99   *outbytesleft -= sizeof (ucs2_t);
100 
101   return sizeof (ucs2_t);
102 }
103 #endif /* ICONV_FROM_UCS_CES_UCS_2 */
104 
105 #if defined (ICONV_TO_UCS_CES_UCS_2)
106 static ucs4_t
ucs_2_convert_to_ucs(void * data,const unsigned char ** inbuf,size_t * inbytesleft)107 ucs_2_convert_to_ucs (void *data,
108                              const unsigned char **inbuf,
109                              size_t *inbytesleft)
110 {
111   ucs4_t res;
112 
113   if (*inbytesleft < sizeof (ucs2_t))
114     return (ucs4_t)ICONV_CES_BAD_SEQUENCE;
115 
116   ucs2_t uc;
117 
118   memcpy(&uc, *inbuf, sizeof(ucs2_t));
119 
120   if (*((int *)data) == UCS_2_BIG_ENDIAN)
121     res = (ucs4_t)ICONV_BETOHS (uc);
122   else
123     res = (ucs4_t)ICONV_LETOHS (uc);
124 
125   if ((res  >= 0x0000D800 && res <= 0x0000DFFF) /* Surrogate character */
126       || res >= 0x0000FFFE)
127     return (ucs4_t)ICONV_CES_INVALID_CHARACTER;
128 
129   *inbytesleft -= sizeof (ucs2_t);
130   *inbuf += sizeof (ucs2_t);
131 
132   return res;
133 }
134 #endif /* ICONV_TO_UCS_CES_UCS_2 */
135 
136 static int
ucs_2_get_mb_cur_max(void * data)137 ucs_2_get_mb_cur_max (void *data)
138 {
139   (void) data;
140   return 2;
141 }
142 
143 #if defined (ICONV_TO_UCS_CES_UCS_2)
144 const iconv_to_ucs_ces_handlers_t
145 _iconv_to_ucs_ces_handlers_ucs_2 =
146 {
147   ucs_2_init,
148   ucs_2_close,
149   ucs_2_get_mb_cur_max,
150   NULL,
151   NULL,
152   NULL,
153   ucs_2_convert_to_ucs
154 };
155 #endif
156 
157 #if defined (ICONV_FROM_UCS_CES_UCS_2)
158 const iconv_from_ucs_ces_handlers_t
159 _iconv_from_ucs_ces_handlers_ucs_2 =
160 {
161   ucs_2_init,
162   ucs_2_close,
163   ucs_2_get_mb_cur_max,
164   NULL,
165   NULL,
166   NULL,
167   ucs_2_convert_from_ucs
168 };
169 #endif
170 
171 #endif /* ICONV_TO_UCS_CES_UCS_2 || ICONV_FROM_UCS_CES_UCS_2 */
172 
173