xref: /freebsd/usr.bin/hexdump/conv.c (revision 5b31cc94b10d4bb7109c6b27940a0fc76a44a331)
1 /*-
2  * SPDX-License-Identifier: BSD-3-Clause
3  *
4  * Copyright (c) 1989, 1993
5  *	The Regents of the University of California.  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  * 3. Neither the name of the University nor the names of its contributors
16  *    may be used to endorse or promote products derived from this software
17  *    without specific prior written permission.
18  *
19  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
20  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
21  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
22  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
23  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
24  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
25  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
26  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
27  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
28  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
29  * SUCH DAMAGE.
30  */
31 
32 #include <sys/cdefs.h>
33 #include <sys/types.h>
34 
35 #include <assert.h>
36 #include <ctype.h>
37 #include <limits.h>
38 #include <stdio.h>
39 #include <stdlib.h>
40 #include <string.h>
41 #include <wchar.h>
42 #include <wctype.h>
43 #include "hexdump.h"
44 
45 void
46 conv_c(PR *pr, u_char *p, size_t bufsize)
47 {
48 	char buf[10];
49 	char const *str;
50 	wchar_t wc;
51 	size_t clen, oclen;
52 	int converr, pad, width;
53 	u_char peekbuf[MB_LEN_MAX];
54 	u_char *op;
55 
56 	op = NULL;
57 
58 	if (pr->mbleft > 0) {
59 		str = "**";
60 		pr->mbleft--;
61 		goto strpr;
62 	}
63 
64 	switch(*p) {
65 	case '\0':
66 		str = "\\0";
67 		goto strpr;
68 	/* case '\a': */
69 	case '\007':
70 		str = "\\a";
71 		goto strpr;
72 	case '\b':
73 		str = "\\b";
74 		goto strpr;
75 	case '\f':
76 		str = "\\f";
77 		goto strpr;
78 	case '\n':
79 		str = "\\n";
80 		goto strpr;
81 	case '\r':
82 		str = "\\r";
83 		goto strpr;
84 	case '\t':
85 		str = "\\t";
86 		goto strpr;
87 	case '\v':
88 		str = "\\v";
89 		goto strpr;
90 	default:
91 		break;
92 	}
93 	/*
94 	 * Multibyte characters are disabled for hexdump(1) for backwards
95 	 * compatibility and consistency (none of its other output formats
96 	 * recognize them correctly).
97 	 */
98 	converr = 0;
99 	if (odmode && MB_CUR_MAX > 1) {
100 		oclen = 0;
101 retry:
102 		clen = mbrtowc(&wc, p, bufsize, &pr->mbstate);
103 		if (clen == 0)
104 			clen = 1;
105 		else if (clen == (size_t)-1 || (clen == (size_t)-2 &&
106 		    p == peekbuf)) {
107 			memset(&pr->mbstate, 0, sizeof(pr->mbstate));
108 			if (p == peekbuf) {
109 				/*
110 				 * We peeked ahead, but that didn't help --
111 				 * we either got an illegal sequence or still
112 				 * can't complete; restore original character.
113 				 */
114 				oclen = 0;
115 				p = op;
116 			}
117 			wc = *p;
118 			clen = 1;
119 			converr = 1;
120 		} else if (clen == (size_t)-2) {
121 			/*
122 			 * Incomplete character; peek ahead and see if we
123 			 * can complete it.
124 			 */
125 			oclen = bufsize;
126 			op = p;
127 			bufsize = peek(p = peekbuf, MB_CUR_MAX);
128 			goto retry;
129 		}
130 		clen += oclen;
131 	} else {
132 		wc = *p;
133 		clen = 1;
134 	}
135 	if (!converr && iswprint(wc)) {
136 		if (!odmode) {
137 			*pr->cchar = 'c';
138 			(void)printf(pr->fmt, (int)wc);
139 		} else {
140 			*pr->cchar = 'C';
141 			assert(strcmp(pr->fmt, "%3C") == 0);
142 			width = wcwidth(wc);
143 			assert(width >= 0);
144 			pad = 3 - width;
145 			if (pad < 0)
146 				pad = 0;
147 			(void)printf("%*s%C", pad, "", wc);
148 			pr->mbleft = clen - 1;
149 		}
150 	} else {
151 		(void)sprintf(buf, "%03o", (int)*p);
152 		str = buf;
153 strpr:		*pr->cchar = 's';
154 		(void)printf(pr->fmt, str);
155 	}
156 }
157 
158 void
159 conv_u(PR *pr, u_char *p)
160 {
161 	static char const * list[] = {
162 		"nul", "soh", "stx", "etx", "eot", "enq", "ack", "bel",
163 		 "bs",  "ht",  "lf",  "vt",  "ff",  "cr",  "so",  "si",
164 		"dle", "dc1", "dc2", "dc3", "dc4", "nak", "syn", "etb",
165 		"can",  "em", "sub", "esc",  "fs",  "gs",  "rs",  "us",
166 	};
167 
168 						/* od used nl, not lf */
169 	if (*p <= 0x1f) {
170 		*pr->cchar = 's';
171 		if (odmode && *p == 0x0a)
172 			(void)printf(pr->fmt, "nl");
173 		else
174 			(void)printf(pr->fmt, list[*p]);
175 	} else if (*p == 0x7f) {
176 		*pr->cchar = 's';
177 		(void)printf(pr->fmt, "del");
178 	} else if (odmode && *p == 0x20) {	/* od replaced space with sp */
179 		*pr->cchar = 's';
180 		(void)printf(pr->fmt, " sp");
181 	} else if (isprint(*p)) {
182 		*pr->cchar = 'c';
183 		(void)printf(pr->fmt, *p);
184 	} else {
185 		*pr->cchar = 'x';
186 		(void)printf(pr->fmt, (int)*p);
187 	}
188 }
189