xref: /dflybsd-src/contrib/file/src/is_tar.c (revision 327e51cbf108c0f1619755494b30617e1089a42f)
1 /*
2  * Copyright (c) Ian F. Darwin 1986-1995.
3  * Software written by Ian F. Darwin and others;
4  * maintained 1995-present by Christos Zoulas and others.
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 immediately at the beginning of the file, without modification,
11  *    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 AND CONTRIBUTORS ``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 OR CONTRIBUTORS BE LIABLE FOR
20  * 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  * is_tar() -- figure out whether file is a tar archive.
30  *
31  * Stolen (by the author!) from the public domain tar program:
32  * Public Domain version written 26 Aug 1985 John Gilmore (ihnp4!hoptoad!gnu).
33  *
34  * @(#)list.c 1.18 9/23/86 Public Domain - gnu
35  *
36  * Comments changed and some code/comments reformatted
37  * for file command by Ian Darwin.
38  */
39 
40 #include "file.h"
41 #include "magic.h"
42 #include <string.h>
43 #include <ctype.h>
44 #include <sys/types.h>
45 #include "tar.h"
46 
47 #ifndef lint
48 FILE_RCSID("@(#)$File: is_tar.c,v 1.29 2007/10/17 19:33:31 christos Exp $")
49 #endif
50 
51 #define	isodigit(c)	( ((c) >= '0') && ((c) <= '7') )
52 
53 private int is_tar(const unsigned char *, size_t);
54 private int from_oct(int, const char *);	/* Decode octal number */
55 
56 static const char *tartype[] = {
57 	"tar archive",
58 	"POSIX tar archive",
59 	"POSIX tar archive (GNU)",
60 };
61 
62 protected int
63 file_is_tar(struct magic_set *ms, const unsigned char *buf, size_t nbytes)
64 {
65 	/*
66 	 * Do the tar test first, because if the first file in the tar
67 	 * archive starts with a dot, we can confuse it with an nroff file.
68 	 */
69 	int tar = is_tar(buf, nbytes);
70 	int mime = ms->flags & MAGIC_MIME;
71 
72 	if (tar < 1 || tar > 3)
73 		return 0;
74 
75 	if (mime == MAGIC_MIME_ENCODING)
76 		return 0;
77 
78 	if (file_printf(ms, mime ? "application/x-tar" :
79 	    tartype[tar - 1]) == -1)
80 		return -1;
81 	return 1;
82 }
83 
84 /*
85  * Return
86  *	0 if the checksum is bad (i.e., probably not a tar archive),
87  *	1 for old UNIX tar file,
88  *	2 for Unix Std (POSIX) tar file.
89  */
90 private int
91 is_tar(const unsigned char *buf, size_t nbytes)
92 {
93 	const union record *header = (const union record *)(const void *)buf;
94 	int	i;
95 	int	sum, recsum;
96 	const char	*p;
97 
98 	if (nbytes < sizeof(union record))
99 		return 0;
100 
101 	recsum = from_oct(8,  header->header.chksum);
102 
103 	sum = 0;
104 	p = header->charptr;
105 	for (i = sizeof(union record); --i >= 0;) {
106 		/*
107 		 * We cannot use unsigned char here because of old compilers,
108 		 * e.g. V7.
109 		 */
110 		sum += 0xFF & *p++;
111 	}
112 
113 	/* Adjust checksum to count the "chksum" field as blanks. */
114 	for (i = sizeof(header->header.chksum); --i >= 0;)
115 		sum -= 0xFF & header->header.chksum[i];
116 	sum += ' '* sizeof header->header.chksum;
117 
118 	if (sum != recsum)
119 		return 0;	/* Not a tar archive */
120 
121 	if (strcmp(header->header.magic, GNUTMAGIC) == 0)
122 		return 3;		/* GNU Unix Standard tar archive */
123 	if (strcmp(header->header.magic, TMAGIC) == 0)
124 		return 2;		/* Unix Standard tar archive */
125 
126 	return 1;			/* Old fashioned tar archive */
127 }
128 
129 
130 /*
131  * Quick and dirty octal conversion.
132  *
133  * Result is -1 if the field is invalid (all blank, or nonoctal).
134  */
135 private int
136 from_oct(int digs, const char *where)
137 {
138 	int	value;
139 
140 	while (isspace((unsigned char)*where)) {	/* Skip spaces */
141 		where++;
142 		if (--digs <= 0)
143 			return -1;		/* All blank field */
144 	}
145 	value = 0;
146 	while (digs > 0 && isodigit(*where)) {	/* Scan til nonoctal */
147 		value = (value << 3) | (*where++ - '0');
148 		--digs;
149 	}
150 
151 	if (digs > 0 && *where && !isspace((unsigned char)*where))
152 		return -1;			/* Ended on non-space/nul */
153 
154 	return value;
155 }
156