xref: /illumos-gate/usr/src/lib/libc/port/locale/strcoll.c (revision b6805bf78d2bbbeeaea8909a05623587b42d58b3)
1 /*
2  * Copyright 2010 Nexenta Systems, Inc.  All rights reserved.
3  * Copyright (c) 1995 Alex Tatmanjants <alex@elvisti.kiev.ua>
4  *		at Electronni Visti IA, Kiev, Ukraine.
5  *			All rights reserved.
6  *
7  * Redistribution and use in source and binary forms, with or without
8  * modification, are permitted provided that the following conditions
9  * are met:
10  * 1. Redistributions of source code must retain the above copyright
11  *    notice, this list of conditions and the following disclaimer.
12  * 2. Redistributions in binary form must reproduce the above copyright
13  *    notice, this list of conditions and the following disclaimer in the
14  *    documentation and/or other materials provided with the distribution.
15  *
16  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND
17  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
18  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
19  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR BE LIABLE
20  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
21  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
22  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
23  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
24  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
25  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
26  * SUCH DAMAGE.
27  */
28 
29 #include "lint.h"
30 #include "file64.h"
31 #include <alloca.h>
32 #include <stdlib.h>
33 #include <string.h>
34 #include <errno.h>
35 #include <wchar.h>
36 #include "collate.h"
37 
38 #define	ALLOCA_LIMIT	16
39 
40 /*
41  * In order to properly handle multibyte locales, its easiet to just
42  * convert to wide characters and then use wcscoll.  However if an
43  * error occurs, we gracefully fall back to simple strcmp.  Caller
44  * should check errno.
45  */
46 int
47 strcoll(const char *s1, const char *s2)
48 {
49 	int ret;
50 	wchar_t *t1 = NULL, *t2 = NULL;
51 	wchar_t *w1 = NULL, *w2 = NULL;
52 	size_t sz1, sz2;
53 
54 	if (_collate_load_error)
55 		goto error;
56 
57 	sz1 = strlen(s1) + 1;
58 	sz2 = strlen(s2) + 1;
59 
60 	/*
61 	 * Simple assumption: conversion to wide format is strictly
62 	 * reducing, i.e. a single byte (or multibyte character)
63 	 * cannot result in multiple wide characters.
64 	 *
65 	 * We gain a bit of performance by giving preference to alloca
66 	 * for small string allocations.
67 	 */
68 	if (sz1 > ALLOCA_LIMIT) {
69 		if ((t1 = malloc(sz1 * sizeof (wchar_t))) == NULL)
70 			goto error;
71 		w1 = t1;
72 	} else {
73 		if ((w1 = alloca(sz1 * sizeof (wchar_t))) == NULL)
74 			goto error;
75 	}
76 	if (sz2 > ALLOCA_LIMIT) {
77 		if ((t2 = malloc(sz2 * sizeof (wchar_t))) == NULL)
78 			goto error;
79 		w2 = t2;
80 	} else {
81 		if ((w2 = alloca(sz2 * sizeof (wchar_t))) == NULL)
82 			goto error;
83 	}
84 
85 	if ((mbstowcs(w1, s1, sz1)) == (size_t)-1)
86 		goto error;
87 
88 	if ((mbstowcs(w2, s2, sz2)) == (size_t)-1)
89 		goto error;
90 
91 	ret = wcscoll(w1, w2);
92 	if (t1)
93 		free(t1);
94 	if (t2)
95 		free(t2);
96 
97 	return (ret);
98 
99 error:
100 	if (t1)
101 		free(t1);
102 	if (t2)
103 		free(t2);
104 	return (strcmp(s1, s2));
105 }
106