1 /*
2 * CDDL HEADER START
3 *
4 * The contents of this file are subject to the terms of the
5 * Common Development and Distribution License (the "License").
6 * You may not use this file except in compliance with the License.
7 *
8 * You can obtain a copy of the license at src/OPENSOLARIS.LICENSE
9 * or http://www.opensolaris.org/os/licensing.
10 * See the License for the specific language governing permissions
11 * and limitations under the License.
12 *
13 * When distributing Covered Code, include this CDDL HEADER in each
14 * file and include the License file at src/OPENSOLARIS.LICENSE.
15 * If applicable, add the following below this CDDL HEADER, with the
16 * fields enclosed by brackets "[]" replaced with your own identifying
17 * information: Portions Copyright [yyyy] [name of copyright owner]
18 *
19 * CDDL HEADER END
20 */
21 /*
22 * Copyright (c) 2008, by Sun Microsystems, Inc.
23 * All rights reserved.
24 */
25
26 #include <stdio.h>
27 #include <errno.h>
28 #include <stdlib.h>
29 #include <sys/types.h>
30 #define __NEED_TCVN_2_UNI__
31 #include <unicode_tcvn.h> /* Unicode to tcvn mapping table */
32 #include <vi_combine.h>
33 #include "common_defs.h"
34
35
36 typedef struct _icv_state {
37 int _errno; /* internal errno */
38 unsigned short last;
39 } _iconv_st;
40
41
42 static int binsearch(unsigned long x, Combine_map v[], int n);
43
44 /*
45 * Open; called from iconv_open()
46 */
47 void *
_icv_open()48 _icv_open()
49 {
50 _iconv_st *st;
51
52 if ((st = (_iconv_st *)malloc(sizeof(_iconv_st))) == NULL) {
53 errno = ENOMEM;
54 return ((void *) -1);
55 }
56
57 st->_errno = 0;
58 return ((void *) st);
59 }
60
61
62 /*
63 * Close; called from iconv_close()
64 */
65 void
_icv_close(_iconv_st * st)66 _icv_close(_iconv_st *st)
67 {
68 if (!st)
69 errno = EBADF;
70 else
71 free(st);
72 }
73
74
75 /*
76 * Actual conversion; called from iconv()
77 */
78 size_t
_icv_iconv(_iconv_st * st,char ** inbuf,size_t * inbytesleft,char ** outbuf,size_t * outbytesleft)79 _icv_iconv(_iconv_st *st, char **inbuf, size_t *inbytesleft,
80 char **outbuf, size_t *outbytesleft)
81 {
82 int unidx = -1;
83 #ifdef DEBUG
84 fprintf(stderr, "========== iconv(): TCVN5712 -->UCS-2 ==========\n");
85 #endif
86 if (st == NULL) {
87 errno = EBADF;
88 return ((size_t) -1);
89 }
90
91 if (inbuf == NULL || *inbuf == NULL) { /* Reset request. */
92 st->_errno = 0;
93 return ((size_t) 0);
94 }
95
96 st->_errno = 0; /* Reset internal errno */
97 errno = 0; /* Reset external errno */
98
99 /* Convert tcvn encoding to UCS-2 */
100 while (*inbytesleft > 0 && *outbytesleft > 0) {
101 unsigned long uni = 0;
102
103 tcvn_2_uni((unsigned char*)*inbuf, &uni);
104 if (st->last != 0) {
105 if (ISCOMB_UNI(uni)) {
106 /*
107 * Composed characters with combine character
108 */
109 unsigned int k = 0;
110 switch (uni) {
111 case 0x0300: k = 0; break;
112 case 0x0301: k = 1; break;
113 case 0x0303: k = 2; break;
114 case 0x0309: k = 3; break;
115 case 0x0323: k = 4; break;
116 default:
117 break;
118 }
119 unidx = binsearch(st->last, vi_comb_data, VOWEL_NUM);
120 if (unidx >= 0) {
121 uni = vi_comb_data[unidx].composed[k];
122 } else {
123 errno = EBADF;
124 }
125 st->last = 0;
126
127 } else {
128 if (st->last < 0x80) {
129 *(*outbuf)++ = (char)st->last;
130 (*outbytesleft) -= 1;
131 } else if (st->last >= 0x0080 && st->last <= 0x07ff) {
132 if (*outbytesleft < 2) {
133 errno = E2BIG;
134 return((size_t)-1);
135 }
136 *(*outbuf)++ = (char)((st->last >> 6) & 0x1f) | 0xc0;
137 *(*outbuf)++ = (char)(st->last & 0x3f) | 0x80;
138 (*outbytesleft) -= 2;
139 } else if (st->last >= 0x0800) {
140 if (*outbytesleft < 3) {
141 errno = E2BIG;
142 return((size_t)-1);
143 }
144 *(*outbuf)++ = (char)((st->last >> 12) & 0xf) | 0xe0;
145 *(*outbuf)++ = (char)((st->last >>6) & 0x3f) | 0x80;
146 *(*outbuf)++ = (char)(st->last & 0x3f) | 0x80;
147 (*outbytesleft) -= 3;
148 }
149 }
150 st->last = 0;
151 } else {
152 if (uni >= 0x0041 && uni <= 0x01b0
153 && ((tcvn_comp_bases_mask[(uni-0x0040) >> 5] >> (uni & 0x1f)) & 1)) {
154 /*
155 * uni is vowel, it's a possible match with combine character.
156 * Buffer it.
157 * */
158 st->last = uni;
159 (*inbuf)++;
160 (*inbytesleft)--;
161 continue;
162 }
163 }
164
165 if (uni < 0x80) {
166 *(*outbuf)++ = (char)uni;
167 (*outbytesleft) -= 1;
168 } else if (uni >= 0x0080 && uni <= 0x07ff) {
169 if (*outbytesleft < 2) {
170 errno = E2BIG;
171 return((size_t)-1);
172 }
173 *(*outbuf)++ = (char)((uni >> 6) & 0x1f) | 0xc0;
174 *(*outbuf)++ = (char)(uni & 0x3f) | 0x80;
175 (*outbytesleft) -= 2;
176 } else if (uni >= 0x0800 && uni <= 0xffff) {
177 if (*outbytesleft < 3) {
178 errno = E2BIG;
179 return((size_t)-1);
180 }
181 *(*outbuf)++ = (char)((uni >> 12) & 0xf) | 0xe0;
182 *(*outbuf)++ = (char)((uni >>6) & 0x3f) | 0x80;
183 *(*outbuf)++ = (char)(uni & 0x3f) | 0x80;
184 (*outbytesleft) -= 3;
185 }
186
187 (*inbuf)++;
188 (*inbytesleft)--;
189 }
190
191 if ( *inbytesleft > 0 && *outbytesleft <= 0 ) {
192 errno = E2BIG;
193 st->last = 0;
194 return ((size_t)-1);
195 }
196
197 if (st->last !=0 ) {
198 if (st->last < 0x80) {
199 *(*outbuf)++ = (char)st->last;
200 (*outbytesleft) -= 1;
201 } else if (st->last >= 0x0080 && st->last <= 0x07ff) {
202 if (*outbytesleft < 2 ) {
203 errno = E2BIG;
204 return((size_t)-1);
205 }
206 *(*outbuf)++ = (char)((st->last >> 6) & 0x1f) | 0xc0;
207 *(*outbuf)++ = (char)(st->last & 0x3f) | 0x80;
208 (*outbytesleft) -= 2;
209 } else if (st->last >= 0x0800) {
210 if (*outbytesleft < 3) {
211 errno = E2BIG;
212 return((size_t)-1);
213 }
214 *(*outbuf)++ = (char)((st->last >> 12) & 0xf) | 0xe0;
215 *(*outbuf)++ = (char)((st->last >>6) & 0x3f) | 0x80;
216 *(*outbuf)++ = (char)(st->last & 0x3f) | 0x80;
217 (*outbytesleft) -= 3;
218 }
219 st->last = 0;
220 }
221
222 return ((size_t)(*inbytesleft));
223
224 }
225
226 /* binsearch: find x in v[0] <= v[1] <= ... <= v[n-1] */
binsearch(unsigned long x,Combine_map v[],int n)227 static int binsearch(unsigned long x, Combine_map v[], int n)
228 {
229 int low = 0;
230 int mid = 0;
231 int high = n - 1;
232
233 low = 0;
234 while (low <= high) {
235 mid = (low + high) / 2;
236 if (x < v[mid].base)
237 high = mid - 1;
238 else if (x > v[mid].base)
239 low = mid + 1;
240 else
241 /* found match */
242 return mid;
243 }
244
245 /* no match */
246 return (-1);
247 }
248