magic.c revision 133359 
1/*
2 * Copyright (c) Christos Zoulas 2003.
3 * All Rights Reserved.
4 *
5 * Redistribution and use in source and binary forms, with or without
6 * modification, are permitted provided that the following conditions
7 * are met:
8 * 1. Redistributions of source code must retain the above copyright
9 *    notice immediately at the beginning of the file, without modification,
10 *    this list of conditions, and the following disclaimer.
11 * 2. Redistributions in binary form must reproduce the above copyright
12 *    notice, this list of conditions and the following disclaimer in the
13 *    documentation and/or other materials provided with the distribution.
14 * 3. The name of the author may not be used to endorse or promote products
15 *    derived from this software without specific prior written permission.
16 *
17 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
18 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
19 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
20 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE FOR
21 * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
22 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
23 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
24 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
25 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
26 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
27 * SUCH DAMAGE.
28 */
29
30#include "file.h"
31#include "magic.h"
32
33#include <stdio.h>
34#include <stdlib.h>
35#include <unistd.h>
36#include <string.h>
37#include <sys/types.h>
38#include <sys/param.h>	/* for MAXPATHLEN */
39#include <sys/stat.h>
40#include <fcntl.h>	/* for open() */
41#ifdef QUICK
42#include <sys/mman.h>
43#endif
44
45#if defined(HAVE_UTIMES)
46# include <sys/time.h>
47#elif defined(HAVE_UTIME)
48# if defined(HAVE_SYS_UTIME_H)
49#  include <sys/utime.h>
50# elif defined(HAVE_UTIME_H)
51#  include <utime.h>
52# endif
53#endif
54
55#ifdef HAVE_UNISTD_H
56#include <unistd.h>	/* for read() */
57#endif
58
59#ifdef HAVE_LOCALE_H
60#include <locale.h>
61#endif
62
63#include <netinet/in.h>		/* for byte swapping */
64
65#include "patchlevel.h"
66
67#ifndef	lint
68FILE_RCSID("@(#)$Id: magic.c,v 1.22 2004/07/24 19:55:17 christos Exp $")
69#endif	/* lint */
70
71#ifdef __EMX__
72private char *apptypeName = NULL;
73protected int file_os2_apptype(struct magic_set *ms, const char *fn,
74    const void *buf, size_t nb);
75#endif /* __EMX__ */
76
77private void free_mlist(struct mlist *);
78private void close_and_restore(const struct magic_set *, const char *, int,
79    const struct stat *);
80
81public struct magic_set *
82magic_open(int flags)
83{
84	struct magic_set *ms;
85
86	if ((ms = malloc(sizeof(struct magic_set))) == NULL)
87		return NULL;
88
89	if (magic_setflags(ms, flags) == -1) {
90		free(ms);
91		errno = EINVAL;
92		return NULL;
93	}
94
95	ms->o.ptr = ms->o.buf = malloc(ms->o.size = 1024);
96	ms->o.len = 0;
97	if (ms->o.buf == NULL) {
98		free(ms);
99		return NULL;
100	}
101	ms->o.pbuf = malloc(ms->o.psize = 1024);
102	if (ms->o.pbuf == NULL) {
103		free(ms->o.buf);
104		free(ms);
105		return NULL;
106	}
107	ms->c.off = malloc((ms->c.len = 10) * sizeof(*ms->c.off));
108	if (ms->c.off == NULL) {
109		free(ms->o.pbuf);
110		free(ms->o.buf);
111		free(ms);
112		return NULL;
113	}
114	ms->haderr = 0;
115	ms->error = -1;
116	ms->mlist = NULL;
117	return ms;
118}
119
120private void
121free_mlist(struct mlist *mlist)
122{
123	struct mlist *ml;
124
125	if (mlist == NULL)
126		return;
127
128	for (ml = mlist->next; ml != mlist;) {
129		struct mlist *next = ml->next;
130		struct magic *mg = ml->magic;
131		file_delmagic(mg, ml->mapped, ml->nmagic);
132		free(ml);
133		ml = next;
134	}
135	free(ml);
136}
137
138public void
139magic_close(ms)
140    struct magic_set *ms;
141{
142	free_mlist(ms->mlist);
143	free(ms->o.buf);
144	free(ms->c.off);
145	free(ms);
146}
147
148/*
149 * load a magic file
150 */
151public int
152magic_load(struct magic_set *ms, const char *magicfile)
153{
154	struct mlist *ml = file_apprentice(ms, magicfile, FILE_LOAD);
155	if (ml) {
156		free_mlist(ms->mlist);
157		ms->mlist = ml;
158		return 0;
159	}
160	return -1;
161}
162
163public int
164magic_compile(struct magic_set *ms, const char *magicfile)
165{
166	struct mlist *ml = file_apprentice(ms, magicfile, FILE_COMPILE);
167	free_mlist(ml);
168	return ml ? 0 : -1;
169}
170
171public int
172magic_check(struct magic_set *ms, const char *magicfile)
173{
174	struct mlist *ml = file_apprentice(ms, magicfile, FILE_CHECK);
175	free_mlist(ml);
176	return ml ? 0 : -1;
177}
178
179private void
180close_and_restore(const struct magic_set *ms, const char *name, int fd,
181    const struct stat *sb)
182{
183	(void) close(fd);
184	if (fd != STDIN_FILENO && (ms->flags & MAGIC_PRESERVE_ATIME) != 0) {
185		/*
186		 * Try to restore access, modification times if read it.
187		 * This is really *bad* because it will modify the status
188		 * time of the file... And of course this will affect
189		 * backup programs
190		 */
191#ifdef HAVE_UTIMES
192		struct timeval  utsbuf[2];
193		utsbuf[0].tv_sec = sb->st_atime;
194		utsbuf[1].tv_sec = sb->st_mtime;
195
196		(void) utimes(name, utsbuf); /* don't care if loses */
197#elif defined(HAVE_UTIME_H) || defined(HAVE_SYS_UTIME_H)
198		struct utimbuf  utbuf;
199
200		utbuf.actime = sb->st_atime;
201		utbuf.modtime = sb->st_mtime;
202		(void) utime(name, &utbuf); /* don't care if loses */
203#endif
204	}
205}
206
207#ifndef COMPILE_ONLY
208/*
209 * find type of named file
210 */
211public const char *
212magic_file(struct magic_set *ms, const char *inname)
213{
214	int	fd = 0;
215	unsigned char buf[HOWMANY+1];	/* one extra for terminating '\0' */
216	struct stat	sb;
217	ssize_t nbytes = 0;	/* number of bytes read from a datafile */
218
219	if (file_reset(ms) == -1)
220		return NULL;
221
222	switch (file_fsmagic(ms, inname, &sb)) {
223	case -1:
224		return NULL;
225	case 0:
226		break;
227	default:
228		return file_getbuffer(ms);
229	}
230
231#ifndef	STDIN_FILENO
232#define	STDIN_FILENO	0
233#endif
234	if (inname == NULL)
235		fd = STDIN_FILENO;
236	else if ((fd = open(inname, O_RDONLY)) < 0) {
237		/* We cannot open it, but we were able to stat it. */
238		if (sb.st_mode & 0222)
239			if (file_printf(ms, "writable, ") == -1)
240				return NULL;
241		if (sb.st_mode & 0111)
242			if (file_printf(ms, "executable, ") == -1)
243				return NULL;
244		if (S_ISREG(sb.st_mode))
245			if (file_printf(ms, "regular file, ") == -1)
246				return NULL;
247		if (file_printf(ms, "no read permission") == -1)
248			return NULL;
249		return file_getbuffer(ms);
250	}
251
252	/*
253	 * try looking at the first HOWMANY bytes
254	 */
255	if ((nbytes = read(fd, (char *)buf, HOWMANY)) == -1) {
256		file_error(ms, errno, "cannot read `%s'", inname);
257		goto done;
258	}
259
260	if (nbytes == 0) {
261		if (file_printf(ms, (ms->flags & MAGIC_MIME) ?
262		    "application/x-empty" : "empty") == -1)
263			goto done;
264		goto gotit;
265	} else if (nbytes == 1) {
266		if (file_printf(ms, "very short file (no magic)") == -1)
267			goto done;
268		goto gotit;
269	} else {
270		buf[nbytes] = '\0';	/* null-terminate it */
271#ifdef __EMX__
272		switch (file_os2_apptype(ms, inname, buf, nbytes)) {
273		case -1:
274			goto done;
275		case 0:
276			break;
277		default:
278			goto gotit;
279		}
280#endif
281		if (file_buffer(ms, buf, (size_t)nbytes) == -1)
282			goto done;
283#ifdef BUILTIN_ELF
284		if (nbytes > 5) {
285			/*
286			 * We matched something in the file, so this *might*
287			 * be an ELF file, and the file is at least 5 bytes
288			 * long, so if it's an ELF file it has at least one
289			 * byte past the ELF magic number - try extracting
290			 * information from the ELF headers that cannot easily
291			 * be extracted with rules in the magic file.
292			 */
293			file_tryelf(ms, fd, buf, (size_t)nbytes);
294		}
295#endif
296	}
297gotit:
298	close_and_restore(ms, inname, fd, &sb);
299	return file_getbuffer(ms);
300done:
301	close_and_restore(ms, inname, fd, &sb);
302	return NULL;
303}
304
305
306public const char *
307magic_buffer(struct magic_set *ms, const void *buf, size_t nb)
308{
309	if (file_reset(ms) == -1)
310		return NULL;
311	/*
312	 * The main work is done here!
313	 * We have the file name and/or the data buffer to be identified.
314	 */
315	if (file_buffer(ms, buf, nb) == -1) {
316		return NULL;
317	}
318	return file_getbuffer(ms);
319}
320#endif
321
322public const char *
323magic_error(struct magic_set *ms)
324{
325	return ms->haderr ? ms->o.buf : NULL;
326}
327
328public int
329magic_errno(struct magic_set *ms)
330{
331	return ms->haderr ? ms->error : 0;
332}
333
334public int
335magic_setflags(struct magic_set *ms, int flags)
336{
337#if !defined(HAVE_UTIME) && !defined(HAVE_UTIMES)
338	if (flags & MAGIC_PRESERVE_ATIME)
339		return -1;
340#endif
341	ms->flags = flags;
342	return 0;
343}
344