1 /*
2 * Copyright 2010 Nexenta Systems, Inc. All rights reserved.
3 * Copyright (c) 2002 Tim J. Robbins
4 * All rights reserved.
5 *
6 * Redistribution and use in source and binary forms, with or without
7 * modification, are permitted provided that the following conditions
8 * are met:
9 * 1. Redistributions of source code must retain the above copyright
10 * notice, this list of conditions and the following disclaimer.
11 * 2. Redistributions in binary form must reproduce the above copyright
12 * notice, this list of conditions and the following disclaimer in the
13 * documentation and/or other materials provided with the distribution.
14 *
15 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
16 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
17 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
18 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
19 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
20 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
21 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
22 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
23 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
24 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
25 * SUCH DAMAGE.
26 */
27
28 #include "lint.h"
29 #include <errno.h>
30 #include <stdlib.h>
31 #include <string.h>
32 #include <wchar.h>
33 #include <assert.h>
34 #include "collate.h"
35
36 int
37 wcscoll(const wchar_t *ws1, const wchar_t *ws2)
38 {
39 int len1, len2, pri1, pri2, ret;
40 wchar_t *tr1 = NULL, *tr2 = NULL;
41 int direc, pass;
42
43 collate_info_t *info = _collate_info;
44
45 if (_collate_load_error)
46 /*
47 * Locale has no special collating order or could not be
48 * loaded, do a fast binary comparison.
49 */
50 return (wcscmp(ws1, ws2));
51
52 ret = 0;
53
54 /*
55 * Once upon a time we had code to try to optimize this, but
56 * it turns out that you really can't make many assumptions
57 * safely. You absolutely have to run this pass by pass,
58 * because some passes will be ignored for a given character,
59 * while others will not. Simpler locales will benefit from
60 * having fewer passes, and most comparisions should resolve
61 * during the primary pass anyway.
62 *
63 * Note that we do one final extra pass at the end to pick
64 * up UNDEFINED elements. There is special handling for them.
65 */
66 for (pass = 0; pass <= info->directive_count; pass++) {
67
68 int32_t *st1 = NULL;
69 int32_t *st2 = NULL;
70 const wchar_t *w1 = ws1;
71 const wchar_t *w2 = ws2;
72
73 /* special pass for UNDEFINED */
74 if (pass == info->directive_count) {
75 direc = DIRECTIVE_FORWARD | DIRECTIVE_UNDEFINED;
76 } else {
77 direc = info->directive[pass];
78 }
79
80 if (direc & DIRECTIVE_BACKWARD) {
81 wchar_t *bp, *fp, c;
82 if ((tr1 = wcsdup(w1)) == NULL)
83 goto fail;
84 bp = tr1;
85 fp = tr1 + wcslen(tr1) - 1;
86 while (bp < fp) {
87 c = *bp;
88 *bp++ = *fp;
89 *fp-- = c;
90 }
91 if ((tr2 = wcsdup(w2)) == NULL)
92 goto fail;
93 bp = tr2;
94 fp = tr2 + wcslen(tr2) - 1;
95 while (bp < fp) {
96 c = *bp;
97 *bp++ = *fp;
98 *fp-- = c;
99 }
100 w1 = tr1;
101 w2 = tr2;
102 }
103
104 if (direc & DIRECTIVE_POSITION) {
105 while ((*w1 || st1) && (*w2 || st2)) {
106 pri1 = pri2 = 0;
107 _collate_lookup(w1, &len1, &pri1, pass, &st1);
108 if (pri1 <= 0) {
109 if (pri1 < 0) {
110 errno = EINVAL;
111 goto fail;
112 }
113 pri1 = COLLATE_MAX_PRIORITY;
114 }
115 _collate_lookup(w2, &len2, &pri2, pass, &st2);
116 if (pri2 <= 0) {
117 if (pri2 < 0) {
118 errno = EINVAL;
119 goto fail;
120 }
121 pri2 = COLLATE_MAX_PRIORITY;
122 }
123 if (pri1 != pri2) {
124 ret = pri1 - pri2;
125 goto end;
126 }
127 w1 += len1;
128 w2 += len2;
129 }
130 } else {
131 while ((*w1 || st1) && (*w2 || st2)) {
132 pri1 = pri2 = 0;
133 while (*w1) {
134 _collate_lookup(w1, &len1,
135 &pri1, pass, &st1);
136 if (pri1 > 0)
137 break;
138 if (pri1 < 0) {
139 errno = EINVAL;
140 goto fail;
141 }
142 w1 += len1;
143 }
144 while (*w2) {
145 _collate_lookup(w2, &len2,
146 &pri2, pass, &st2);
147 if (pri2 > 0)
148 break;
149 if (pri2 < 0) {
150 errno = EINVAL;
151 goto fail;
152 }
153 w2 += len2;
154 }
155 if (!pri1 || !pri2)
156 break;
157 if (pri1 != pri2) {
158 ret = pri1 - pri2;
159 goto end;
160 }
161 w1 += len1;
162 w2 += len2;
163 }
164 }
165 if (!*w1) {
166 if (*w2) {
167 ret = -(int)*w2;
168 goto end;
169 }
170 } else {
171 ret = *w1;
172 goto end;
173 }
174 }
175 ret = 0;
176
177 end:
178 if (tr1)
179 free(tr1);
180 if (tr2)
181 free(tr2);
182
183 return (ret);
184
185 fail:
186 ret = wcscmp(ws1, ws2);
187 goto end;
188 }