1 /*- 2 * Copyright 2010 Nexenta Systems, Inc. All rights reserved. 3 * Copyright (c) 2002 Tim J. Robbins 4 * All rights reserved. 5 * 6 * Copyright (c) 2011 The FreeBSD Foundation 7 * All rights reserved. 8 * Portions of this software were developed by David Chisnall 9 * under sponsorship from the FreeBSD Foundation. 10 * 11 * Redistribution and use in source and binary forms, with or without 12 * modification, are permitted provided that the following conditions 13 * are met: 14 * 1. Redistributions of source code must retain the above copyright 15 * notice, this list of conditions and the following disclaimer. 16 * 2. Redistributions in binary form must reproduce the above copyright 17 * notice, this list of conditions and the following disclaimer in the 18 * documentation and/or other materials provided with the distribution. 19 * 20 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 21 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 22 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 23 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 24 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 25 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 26 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 27 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 28 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 29 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 30 * SUCH DAMAGE. 31 */ 32 33 #include <sys/cdefs.h> 34 __FBSDID("$FreeBSD$"); 35 36 #include <errno.h> 37 #include <stdlib.h> 38 #include <string.h> 39 #include <wchar.h> 40 #include "collate.h" 41 42 int 43 wcscoll_l(const wchar_t *ws1, const wchar_t *ws2, locale_t locale) 44 { 45 int len1, len2, pri1, pri2, ret; 46 wchar_t *tr1 = NULL, *tr2 = NULL; 47 int direc, pass; 48 49 FIX_LOCALE(locale); 50 struct xlocale_collate *table = 51 (struct xlocale_collate*)locale->components[XLC_COLLATE]; 52 53 if (table->__collate_load_error) 54 /* 55 * Locale has no special collating order or could not be 56 * loaded, do a fast binary comparison. 57 */ 58 return (wcscmp(ws1, ws2)); 59 60 ret = 0; 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 comparisions 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 int check1, check2; 81 82 /* special pass for UNDEFINED */ 83 if (pass == table->info->directive_count) { 84 direc = DIRECTIVE_FORWARD | DIRECTIVE_UNDEFINED; 85 } else { 86 direc = table->info->directive[pass]; 87 } 88 89 if (direc & DIRECTIVE_BACKWARD) { 90 wchar_t *bp, *fp, c; 91 if ((tr1 = wcsdup(w1)) == NULL) 92 goto fail; 93 bp = tr1; 94 fp = tr1 + wcslen(tr1) - 1; 95 while (bp < fp) { 96 c = *bp; 97 *bp++ = *fp; 98 *fp-- = c; 99 } 100 if ((tr2 = wcsdup(w2)) == NULL) 101 goto fail; 102 bp = tr2; 103 fp = tr2 + wcslen(tr2) - 1; 104 while (bp < fp) { 105 c = *bp; 106 *bp++ = *fp; 107 *fp-- = c; 108 } 109 w1 = tr1; 110 w2 = tr2; 111 } 112 113 if (direc & DIRECTIVE_POSITION) { 114 while (*w1 && *w2) { 115 pri1 = pri2 = 0; 116 check1 = check2 = 1; 117 while ((pri1 == pri2) && (check1 || check2)) { 118 if (check1) { 119 _collate_lookup(table, w1, &len1, 120 &pri1, pass, &st1); 121 if (pri1 < 0) { 122 errno = EINVAL; 123 goto fail; 124 } 125 if (!pri1) { 126 pri1 = COLLATE_MAX_PRIORITY; 127 st1 = NULL; 128 } 129 check1 = (st1 != NULL); 130 } 131 if (check2) { 132 _collate_lookup(table, w2, &len2, 133 &pri2, pass, &st2); 134 if (pri2 < 0) { 135 errno = EINVAL; 136 goto fail; 137 } 138 if (!pri2) { 139 pri2 = COLLATE_MAX_PRIORITY; 140 st2 = NULL; 141 } 142 check2 = (st2 != NULL); 143 } 144 } 145 if (pri1 != pri2) { 146 ret = pri1 - pri2; 147 goto end; 148 } 149 w1 += len1; 150 w2 += len2; 151 } 152 } else { 153 while (*w1 && *w2) { 154 pri1 = pri2 = 0; 155 check1 = check2 = 1; 156 while ((pri1 == pri2) && (check1 || check2)) { 157 while (check1 && *w1) { 158 _collate_lookup(table, w1, 159 &len1, &pri1, pass, &st1); 160 if (pri1 > 0) 161 break; 162 if (pri1 < 0) { 163 errno = EINVAL; 164 goto fail; 165 } 166 st1 = NULL; 167 w1 += 1; 168 } 169 check1 = (st1 != NULL); 170 while (check2 && *w2) { 171 _collate_lookup(table, w2, 172 &len2, &pri2, pass, &st2); 173 if (pri2 > 0) 174 break; 175 if (pri2 < 0) { 176 errno = EINVAL; 177 goto fail; 178 } 179 st2 = NULL; 180 w2 += 1; 181 } 182 check2 = (st2 != NULL); 183 if (!pri1 || !pri2) 184 break; 185 } 186 if (!pri1 || !pri2) 187 break; 188 if (pri1 != pri2) { 189 ret = pri1 - pri2; 190 goto end; 191 } 192 w1 += len1; 193 w2 += len2; 194 } 195 } 196 if (!*w1) { 197 if (*w2) { 198 ret = -(int)*w2; 199 goto end; 200 } 201 } else { 202 ret = *w1; 203 goto end; 204 } 205 } 206 ret = 0; 207 208 end: 209 free(tr1); 210 free(tr2); 211 212 return (ret); 213 214 fail: 215 ret = wcscmp(ws1, ws2); 216 goto end; 217 } 218 219 int 220 wcscoll(const wchar_t *ws1, const wchar_t *ws2) 221 { 222 return wcscoll_l(ws1, ws2, __get_locale()); 223 } 224