xref: /netbsd-src/bin/ls/util.c (revision 44db7ee6ac54c14edb168882e04a2e966b678d7a)
1 /*	$NetBSD: util.c,v 1.34 2011/08/29 14:44:21 joerg Exp $	*/
2 
3 /*
4  * Copyright (c) 1989, 1993, 1994
5  *	The Regents of the University of California.  All rights reserved.
6  *
7  * This code is derived from software contributed to Berkeley by
8  * Michael Fischbein.
9  *
10  * Redistribution and use in source and binary forms, with or without
11  * modification, are permitted provided that the following conditions
12  * are met:
13  * 1. Redistributions of source code must retain the above copyright
14  *    notice, this list of conditions and the following disclaimer.
15  * 2. Redistributions in binary form must reproduce the above copyright
16  *    notice, this list of conditions and the following disclaimer in the
17  *    documentation and/or other materials provided with the distribution.
18  * 3. Neither the name of the University nor the names of its contributors
19  *    may be used to endorse or promote products derived from this software
20  *    without specific prior written permission.
21  *
22  * THIS SOFTWARE IS PROVIDED BY THE REGENTS 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 REGENTS 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 #ifndef lint
37 #if 0
38 static char sccsid[] = "@(#)util.c	8.5 (Berkeley) 4/28/95";
39 #else
40 __RCSID("$NetBSD: util.c,v 1.34 2011/08/29 14:44:21 joerg Exp $");
41 #endif
42 #endif /* not lint */
43 
44 #include <sys/types.h>
45 #include <sys/stat.h>
46 
47 #include <err.h>
48 #include <fts.h>
49 #include <limits.h>
50 #include <stdio.h>
51 #include <stdlib.h>
52 #include <string.h>
53 #include <vis.h>
54 #include <wchar.h>
55 #include <wctype.h>
56 
57 #include "ls.h"
58 #include "extern.h"
59 
60 int
safe_print(const char * src)61 safe_print(const char *src)
62 {
63 	size_t len;
64 	char *name;
65 	int flags;
66 
67 	flags = VIS_NL | VIS_OCTAL | VIS_WHITE;
68 	if (f_octal_escape)
69 		flags |= VIS_CSTYLE;
70 
71 	len = strlen(src);
72 	if (len != 0 && SIZE_T_MAX/len <= 4) {
73 		errx(EXIT_FAILURE, "%s: name too long", src);
74 		/* NOTREACHED */
75 	}
76 
77 	name = (char *)malloc(4*len+1);
78 	if (name != NULL) {
79 		len = strvis(name, src, flags);
80 		(void)printf("%s", name);
81 		free(name);
82 		return len;
83 	} else
84 		errx(EXIT_FAILURE, "out of memory!");
85 		/* NOTREACHED */
86 }
87 
88 /*
89  * The reasons why we don't use putwchar(wc) here are:
90  * - If wc == L'\0', we need to restore the initial shift state, but
91  *   the C language standard doesn't say that putwchar(L'\0') does.
92  * - It isn't portable to mix a wide-oriented function (i.e. getwchar)
93  *   with byte-oriented functions (printf et al.) in same FILE.
94  */
95 static int
printwc(wchar_t wc,mbstate_t * pst)96 printwc(wchar_t wc, mbstate_t *pst)
97 {
98 	size_t size;
99 	char buf[MB_LEN_MAX];
100 
101 	size = wcrtomb(buf, wc, pst);
102 	if (size == (size_t)-1) /* This shouldn't happen, but for sure */
103 		return 0;
104 	if (wc == L'\0') {
105 		/* The following condition must be always true, but for sure */
106 		if (size > 0 && buf[size - 1] == '\0')
107 			--size;
108 	}
109 	if (size > 0)
110 		fwrite(buf, 1, size, stdout);
111 	return wc == L'\0' ? 0 : wcwidth(wc);
112 }
113 
114 int
printescaped(const char * src)115 printescaped(const char *src)
116 {
117 	int n = 0;
118 	mbstate_t src_state, stdout_state;
119 	/* The following +1 is to pass '\0' at the end of src to mbrtowc(). */
120 	const char *endptr = src + strlen(src) + 1;
121 
122 	/*
123 	 * We have to reset src_state each time in this function, because
124 	 * the codeset of src pathname may not match with current locale.
125 	 * Note that if we pass NULL instead of src_state to mbrtowc(),
126 	 * there is no way to reset the state.
127 	 */
128 	memset(&src_state, 0, sizeof(src_state));
129 	memset(&stdout_state, 0, sizeof(stdout_state));
130 	while (src < endptr) {
131 		wchar_t wc;
132 		size_t rv, span = endptr - src;
133 
134 #if 0
135 		/*
136 		 * XXX - we should fix libc instead.
137 		 * Theoretically this should work, but our current
138 		 * implementation of iso2022 module doesn't actually work
139 		 * as expected, if there are redundant escape sequences
140 		 * which exceed 32 bytes.
141 		 */
142 		if (span > MB_CUR_MAX)
143 			span = MB_CUR_MAX;
144 #endif
145 		rv = mbrtowc(&wc, src, span, &src_state);
146 		if (rv == 0) { /* assert(wc == L'\0'); */
147 			/* The following may output a shift sequence. */
148 			n += printwc(wc, &stdout_state);
149 			break;
150 		}
151 		if (rv == (size_t)-1) { /* probably errno == EILSEQ */
152 			n += printwc(L'?', &stdout_state);
153 			/* try to skip 1byte, because there is no better way */
154 			src++;
155 			memset(&src_state, 0, sizeof(src_state));
156 		} else if (rv == (size_t)-2) {
157 			if (span < MB_CUR_MAX) { /* incomplete char */
158 				n += printwc(L'?', &stdout_state);
159 				break;
160 			}
161 			src += span; /* a redundant shift sequence? */
162 		} else {
163 			n += printwc(iswprint(wc) ? wc : L'?', &stdout_state);
164 			src += rv;
165 		}
166 	}
167 	return n;
168 }
169