1 /*-
2 * SPDX-License-Identifier: BSD-2-Clause
3 *
4 * Copyright 2017 Nexenta Systems, Inc.
5 * Copyright (c) 2002 Tim J. Robbins
6 * All rights reserved.
7 *
8 * Copyright (c) 2011 The FreeBSD Foundation
9 *
10 * Portions of this software were developed by David Chisnall
11 * under sponsorship from the FreeBSD Foundation.
12 *
13 * Redistribution and use in source and binary forms, with or without
14 * modification, are permitted provided that the following conditions
15 * are met:
16 * 1. Redistributions of source code must retain the above copyright
17 * notice, this list of conditions and the following disclaimer.
18 * 2. Redistributions in binary form must reproduce the above copyright
19 * notice, this list of conditions and the following disclaimer in the
20 * documentation and/or other materials provided with the distribution.
21 *
22 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
23 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
24 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
25 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
26 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
27 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
28 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
29 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
30 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
31 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
32 * SUCH DAMAGE.
33 */
34
35 #include <sys/cdefs.h>
36 #include <errno.h>
37 #include <stdlib.h>
38 #include <string.h>
39 #include <wchar.h>
40 #include "collate.h"
41
42 int
wcscoll_l(const wchar_t * ws1,const wchar_t * ws2,locale_t locale)43 wcscoll_l(const wchar_t *ws1, const wchar_t *ws2, locale_t locale)
44 {
45 int len1, len2, pri1, pri2;
46 wchar_t *tr1 = NULL, *tr2 = NULL;
47 int direc, pass;
48 int ret = wcscmp(ws1, ws2);
49
50 FIX_LOCALE(locale);
51 struct xlocale_collate *table =
52 (struct xlocale_collate*)locale->components[XLC_COLLATE];
53
54 if (table->__collate_load_error || ret == 0)
55 return (ret);
56
57 if (*ws1 == 0 && *ws2 != 0)
58 return (-1);
59 if (*ws1 != 0 && *ws2 == 0)
60 return (1);
61
62 /*
63 * Once upon a time we had code to try to optimize this, but
64 * it turns out that you really can't make many assumptions
65 * safely. You absolutely have to run this pass by pass,
66 * because some passes will be ignored for a given character,
67 * while others will not. Simpler locales will benefit from
68 * having fewer passes, and most comparisons should resolve
69 * during the primary pass anyway.
70 *
71 * Note that we do one final extra pass at the end to pick
72 * up UNDEFINED elements. There is special handling for them.
73 */
74 for (pass = 0; pass <= table->info->directive_count; pass++) {
75
76 const int32_t *st1 = NULL;
77 const int32_t *st2 = NULL;
78 const wchar_t *w1 = ws1;
79 const wchar_t *w2 = ws2;
80
81 /* special pass for UNDEFINED */
82 if (pass == table->info->directive_count) {
83 direc = DIRECTIVE_FORWARD;
84 } else {
85 direc = table->info->directive[pass];
86 }
87
88 if (direc & DIRECTIVE_BACKWARD) {
89 wchar_t *bp, *fp, c;
90 free(tr1);
91 if ((tr1 = wcsdup(w1)) == NULL)
92 goto end;
93 bp = tr1;
94 fp = tr1 + wcslen(tr1) - 1;
95 while (bp < fp) {
96 c = *bp;
97 *bp++ = *fp;
98 *fp-- = c;
99 }
100 free(tr2);
101 if ((tr2 = wcsdup(w2)) == NULL)
102 goto end;
103 bp = tr2;
104 fp = tr2 + wcslen(tr2) - 1;
105 while (bp < fp) {
106 c = *bp;
107 *bp++ = *fp;
108 *fp-- = c;
109 }
110 w1 = tr1;
111 w2 = tr2;
112 }
113
114 if (direc & DIRECTIVE_POSITION) {
115 int check1, check2;
116 while (*w1 && *w2) {
117 pri1 = pri2 = 0;
118 check1 = check2 = 1;
119 while ((pri1 == pri2) && (check1 || check2)) {
120 if (check1) {
121 _collate_lookup(table, w1, &len1,
122 &pri1, pass, &st1);
123 if (pri1 < 0) {
124 errno = EINVAL;
125 goto end;
126 }
127 if (!pri1) {
128 pri1 = COLLATE_MAX_PRIORITY;
129 st1 = NULL;
130 }
131 check1 = (st1 != NULL);
132 }
133 if (check2) {
134 _collate_lookup(table, w2, &len2,
135 &pri2, pass, &st2);
136 if (pri2 < 0) {
137 errno = EINVAL;
138 goto end;
139 }
140 if (!pri2) {
141 pri2 = COLLATE_MAX_PRIORITY;
142 st2 = NULL;
143 }
144 check2 = (st2 != NULL);
145 }
146 }
147 if (pri1 != pri2) {
148 ret = pri1 - pri2;
149 goto end;
150 }
151 w1 += len1;
152 w2 += len2;
153 }
154 if (!*w1) {
155 if (*w2) {
156 ret = -(int)*w2;
157 goto end;
158 }
159 } else {
160 ret = *w1;
161 goto end;
162 }
163 } else {
164 int vpri1 = 0, vpri2 = 0;
165 while (*w1 || *w2 || st1 || st2) {
166 pri1 = 1;
167 while (*w1 || st1) {
168 _collate_lookup(table, w1, &len1, &pri1,
169 pass, &st1);
170 w1 += len1;
171 if (pri1 > 0) {
172 vpri1++;
173 break;
174 }
175
176 if (pri1 < 0) {
177 errno = EINVAL;
178 goto end;
179 }
180 st1 = NULL;
181 }
182 pri2 = 1;
183 while (*w2 || st2) {
184 _collate_lookup(table, w2, &len2, &pri2,
185 pass, &st2);
186 w2 += len2;
187 if (pri2 > 0) {
188 vpri2++;
189 break;
190 }
191 if (pri2 < 0) {
192 errno = EINVAL;
193 goto end;
194 }
195 st2 = NULL;
196 }
197 if ((!pri1 || !pri2) && (vpri1 == vpri2))
198 break;
199 if (pri1 != pri2) {
200 ret = pri1 - pri2;
201 goto end;
202 }
203 }
204 if (vpri1 && !vpri2) {
205 ret = 1;
206 goto end;
207 }
208 if (!vpri1 && vpri2) {
209 ret = -1;
210 goto end;
211 }
212 }
213 }
214 ret = 0;
215
216 end:
217 free(tr1);
218 free(tr2);
219
220 return (ret);
221 }
222
223 int
wcscoll(const wchar_t * ws1,const wchar_t * ws2)224 wcscoll(const wchar_t *ws1, const wchar_t *ws2)
225 {
226 return wcscoll_l(ws1, ws2, __get_locale());
227 }
228