1 /*
2 * Copyright 2010 Nexenta Systems, Inc. All rights reserved.
3 * Copyright (c) 1993
4 * The Regents of the University of California. All rights reserved.
5 *
6 * This code is derived from software contributed to Berkeley by
7 * Paul Borman at Krystal Technologies.
8 *
9 * Redistribution and use in source and binary forms, with or without
10 * modification, are permitted provided that the following conditions
11 * are met:
12 * 1. Redistributions of source code must retain the above copyright
13 * notice, this list of conditions and the following disclaimer.
14 * 2. Redistributions in binary form must reproduce the above copyright
15 * notice, this list of conditions and the following disclaimer in the
16 * documentation and/or other materials provided with the distribution.
17 * 4. Neither the name of the University nor the names of its contributors
18 * may be used to endorse or promote products derived from this software
19 * without specific prior written permission.
20 *
21 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
22 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
23 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
24 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
25 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
26 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
27 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
28 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
29 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
30 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
31 * SUCH DAMAGE.
32 */
33
34 #include "lint.h"
35 #include <ctype.h>
36 #include <wchar.h>
37 #include "runetype.h"
38 #include "mblocal.h"
39 #include "_ctype.h"
40
41 _RuneLocale _DefaultRuneLocale = {
42 _RUNE_MAGIC_1,
43 "NONE",
44 {
45 /* 00 */
46 _CTYPE_C,
47 _CTYPE_C,
48 _CTYPE_C,
49 _CTYPE_C,
50 _CTYPE_C,
51 _CTYPE_C,
52 _CTYPE_C,
53 _CTYPE_C,
54 /* 08 */
55 _CTYPE_C,
56 _CTYPE_C|_CTYPE_S|_CTYPE_B,
57 _CTYPE_C|_CTYPE_S,
58 _CTYPE_C|_CTYPE_S,
59 _CTYPE_C|_CTYPE_S,
60 _CTYPE_C|_CTYPE_S,
61 _CTYPE_C,
62 _CTYPE_C,
63 /* 10 */
64 _CTYPE_C,
65 _CTYPE_C,
66 _CTYPE_C,
67 _CTYPE_C,
68 _CTYPE_C,
69 _CTYPE_C,
70 _CTYPE_C,
71 _CTYPE_C,
72 /* 18 */
73 _CTYPE_C,
74 _CTYPE_C,
75 _CTYPE_C,
76 _CTYPE_C,
77 _CTYPE_C,
78 _CTYPE_C,
79 _CTYPE_C,
80 _CTYPE_C,
81 /* 20 */
82 _CTYPE_S|_CTYPE_B|_CTYPE_R,
83 _CTYPE_P|_CTYPE_R|_CTYPE_G,
84 _CTYPE_P|_CTYPE_R|_CTYPE_G,
85 _CTYPE_P|_CTYPE_R|_CTYPE_G,
86 _CTYPE_P|_CTYPE_R|_CTYPE_G,
87 _CTYPE_P|_CTYPE_R|_CTYPE_G,
88 _CTYPE_P|_CTYPE_R|_CTYPE_G,
89 _CTYPE_P|_CTYPE_R|_CTYPE_G,
90 /* 28 */
91 _CTYPE_P|_CTYPE_R|_CTYPE_G,
92 _CTYPE_P|_CTYPE_R|_CTYPE_G,
93 _CTYPE_P|_CTYPE_R|_CTYPE_G,
94 _CTYPE_P|_CTYPE_R|_CTYPE_G,
95 _CTYPE_P|_CTYPE_R|_CTYPE_G,
96 _CTYPE_P|_CTYPE_R|_CTYPE_G,
97 _CTYPE_P|_CTYPE_R|_CTYPE_G,
98 _CTYPE_P|_CTYPE_R|_CTYPE_G,
99 /* 30 */
100 _CTYPE_D|_CTYPE_R|_CTYPE_G|_CTYPE_X,
101 _CTYPE_D|_CTYPE_R|_CTYPE_G|_CTYPE_X,
102 _CTYPE_D|_CTYPE_R|_CTYPE_G|_CTYPE_X,
103 _CTYPE_D|_CTYPE_R|_CTYPE_G|_CTYPE_X,
104 _CTYPE_D|_CTYPE_R|_CTYPE_G|_CTYPE_X,
105 _CTYPE_D|_CTYPE_R|_CTYPE_G|_CTYPE_X,
106 _CTYPE_D|_CTYPE_R|_CTYPE_G|_CTYPE_X,
107 _CTYPE_D|_CTYPE_R|_CTYPE_G|_CTYPE_X,
108 /* 38 */
109 _CTYPE_D|_CTYPE_R|_CTYPE_G|_CTYPE_X,
110 _CTYPE_D|_CTYPE_R|_CTYPE_G|_CTYPE_X,
111 _CTYPE_P|_CTYPE_R|_CTYPE_G,
112 _CTYPE_P|_CTYPE_R|_CTYPE_G,
113 _CTYPE_P|_CTYPE_R|_CTYPE_G,
114 _CTYPE_P|_CTYPE_R|_CTYPE_G,
115 _CTYPE_P|_CTYPE_R|_CTYPE_G,
116 _CTYPE_P|_CTYPE_R|_CTYPE_G,
117 /* 40 */
118 _CTYPE_P|_CTYPE_R|_CTYPE_G,
119 _CTYPE_U|_CTYPE_X|_CTYPE_R|_CTYPE_G|_CTYPE_A,
120 _CTYPE_U|_CTYPE_X|_CTYPE_R|_CTYPE_G|_CTYPE_A,
121 _CTYPE_U|_CTYPE_X|_CTYPE_R|_CTYPE_G|_CTYPE_A,
122 _CTYPE_U|_CTYPE_X|_CTYPE_R|_CTYPE_G|_CTYPE_A,
123 _CTYPE_U|_CTYPE_X|_CTYPE_R|_CTYPE_G|_CTYPE_A,
124 _CTYPE_U|_CTYPE_X|_CTYPE_R|_CTYPE_G|_CTYPE_A,
125 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
126 /* 48 */
127 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
128 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
129 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
130 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
131 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
132 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
133 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
134 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
135 /* 50 */
136 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
137 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
138 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
139 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
140 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
141 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
142 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
143 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
144 /* 58 */
145 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
146 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
147 _CTYPE_U|_CTYPE_R|_CTYPE_G|_CTYPE_A,
148 _CTYPE_P|_CTYPE_R|_CTYPE_G,
149 _CTYPE_P|_CTYPE_R|_CTYPE_G,
150 _CTYPE_P|_CTYPE_R|_CTYPE_G,
151 _CTYPE_P|_CTYPE_R|_CTYPE_G,
152 _CTYPE_P|_CTYPE_R|_CTYPE_G,
153 /* 60 */
154 _CTYPE_P|_CTYPE_R|_CTYPE_G,
155 _CTYPE_L|_CTYPE_X|_CTYPE_R|_CTYPE_G|_CTYPE_A,
156 _CTYPE_L|_CTYPE_X|_CTYPE_R|_CTYPE_G|_CTYPE_A,
157 _CTYPE_L|_CTYPE_X|_CTYPE_R|_CTYPE_G|_CTYPE_A,
158 _CTYPE_L|_CTYPE_X|_CTYPE_R|_CTYPE_G|_CTYPE_A,
159 _CTYPE_L|_CTYPE_X|_CTYPE_R|_CTYPE_G|_CTYPE_A,
160 _CTYPE_L|_CTYPE_X|_CTYPE_R|_CTYPE_G|_CTYPE_A,
161 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
162 /* 68 */
163 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
164 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
165 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
166 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
167 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
168 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
169 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
170 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
171 /* 70 */
172 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
173 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
174 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
175 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
176 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
177 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
178 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
179 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
180 /* 78 */
181 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
182 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
183 _CTYPE_L|_CTYPE_R|_CTYPE_G|_CTYPE_A,
184 _CTYPE_P|_CTYPE_R|_CTYPE_G,
185 _CTYPE_P|_CTYPE_R|_CTYPE_G,
186 _CTYPE_P|_CTYPE_R|_CTYPE_G,
187 _CTYPE_P|_CTYPE_R|_CTYPE_G,
188 _CTYPE_C,
189 },
190
191 /* BEGIN CSTYLED */
192 {
193 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
194 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
195 0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17,
196 0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f,
197 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27,
198 0x28, 0x29, 0x2a, 0x2b, 0x2c, 0x2d, 0x2e, 0x2f,
199 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37,
200 0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x3f,
201 0x40, 'a', 'b', 'c', 'd', 'e', 'f', 'g',
202 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o',
203 'p', 'q', 'r', 's', 't', 'u', 'v', 'w',
204 'x', 'y', 'z', 0x5b, 0x5c, 0x5d, 0x5e, 0x5f,
205 0x60, 'a', 'b', 'c', 'd', 'e', 'f', 'g',
206 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o',
207 'p', 'q', 'r', 's', 't', 'u', 'v', 'w',
208 'x', 'y', 'z', 0x7b, 0x7c, 0x7d, 0x7e, 0x7f,
209 0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87,
210 0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f,
211 0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97,
212 0x98, 0x99, 0x9a, 0x9b, 0x9c, 0x9d, 0x9e, 0x9f,
213 0xa0, 0xa1, 0xa2, 0xa3, 0xa4, 0xa5, 0xa6, 0xa7,
214 0xa8, 0xa9, 0xaa, 0xab, 0xac, 0xad, 0xae, 0xaf,
215 0xb0, 0xb1, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6, 0xb7,
216 0xb8, 0xb9, 0xba, 0xbb, 0xbc, 0xbd, 0xbe, 0xbf,
217 0xc0, 0xc1, 0xc2, 0xc3, 0xc4, 0xc5, 0xc6, 0xc7,
218 0xc8, 0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf,
219 0xd0, 0xd1, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, 0xd7,
220 0xd8, 0xd9, 0xda, 0xdb, 0xdc, 0xdd, 0xde, 0xdf,
221 0xe0, 0xe1, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7,
222 0xe8, 0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef,
223 0xf0, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7,
224 0xf8, 0xf9, 0xfa, 0xfb, 0xfc, 0xfd, 0xfe, 0xff,
225 },
226 {
227 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
228 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
229 0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17,
230 0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f,
231 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27,
232 0x28, 0x29, 0x2a, 0x2b, 0x2c, 0x2d, 0x2e, 0x2f,
233 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37,
234 0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x3f,
235 0x40, 'A', 'B', 'C', 'D', 'E', 'F', 'G',
236 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O',
237 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W',
238 'X', 'Y', 'Z', 0x5b, 0x5c, 0x5d, 0x5e, 0x5f,
239 0x60, 'A', 'B', 'C', 'D', 'E', 'F', 'G',
240 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O',
241 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W',
242 'X', 'Y', 'Z', 0x7b, 0x7c, 0x7d, 0x7e, 0x7f,
243 0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87,
244 0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f,
245 0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97,
246 0x98, 0x99, 0x9a, 0x9b, 0x9c, 0x9d, 0x9e, 0x9f,
247 0xa0, 0xa1, 0xa2, 0xa3, 0xa4, 0xa5, 0xa6, 0xa7,
248 0xa8, 0xa9, 0xaa, 0xab, 0xac, 0xad, 0xae, 0xaf,
249 0xb0, 0xb1, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6, 0xb7,
250 0xb8, 0xb9, 0xba, 0xbb, 0xbc, 0xbd, 0xbe, 0xbf,
251 0xc0, 0xc1, 0xc2, 0xc3, 0xc4, 0xc5, 0xc6, 0xc7,
252 0xc8, 0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf,
253 0xd0, 0xd1, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, 0xd7,
254 0xd8, 0xd9, 0xda, 0xdb, 0xdc, 0xdd, 0xde, 0xdf,
255 0xe0, 0xe1, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7,
256 0xe8, 0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef,
257 0xf0, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7,
258 0xf8, 0xf9, 0xfa, 0xfb, 0xfc, 0xfd, 0xfe, 0xff,
259 },
260 /* END CSTYLED */
261 };
262
263 _RuneLocale *_CurrentRuneLocale = &_DefaultRuneLocale;
264
265 /* Taken from former _ctype.c */
266 unsigned int *__ctype_mask = _DefaultRuneLocale.__runetype;
267
268 int *__trans_lower = _DefaultRuneLocale.__maplower;
269 int *__trans_upper = _DefaultRuneLocale.__mapupper;
270
271 /*
272 * Used in various string routines to conditionalize versions optimized for
273 * the ASCII case
274 */
275 int charset_is_ascii = 1;