Commit ed3e6679e371f8eec12384c5cf4fb534b890a49c - magicrescue

+76

-0

ChangeLog less more

0	0	For less detailed change information, see the NEWS file
1	1
	2	------------------------------------------------------------------------
	3	r602 \| jbj \| 2008-01-17 12:33:57 +0100 (Thu, 17 Jan 2008) \| 1 line
	4	Changed paths:
	5	M /magicrescue/trunk/configure
	6
	7	Workaround for Solaris /bin/sh non-POSIX 'trap' command
	8	------------------------------------------------------------------------
	9	r590 \| jbj \| 2007-12-31 16:25:04 +0100 (Mon, 31 Dec 2007) \| 1 line
	10	Changed paths:
	11	M /magicrescue/trunk/src/dupemap.c
	12	M /magicrescue/trunk/src/magicrescue.c
	13	M /magicrescue/trunk/src/recipe.c
	14	M /magicrescue/trunk/src/recipe.h
	15	M /magicrescue/trunk/src/scanner.c
	16	M /magicrescue/trunk/src/scanner.h
	17	M /magicrescue/trunk/src/util.c
	18	M /magicrescue/trunk/src/util.h
	19	M /magicrescue/trunk/tools/textextract.c
	20
	21	Change buffers to signed string types to silence gcc warnings
	22	------------------------------------------------------------------------
	23	r589 \| jbj \| 2007-12-31 15:37:57 +0100 (Mon, 31 Dec 2007) \| 1 line
	24	Changed paths:
	25	M /magicrescue/trunk/new_NEWS
	26
	27	news
	28	------------------------------------------------------------------------
	29	r588 \| jbj \| 2007-12-31 15:13:24 +0100 (Mon, 31 Dec 2007) \| 1 line
	30	Changed paths:
	31	A /magicrescue/trunk/recipes/canon-cr2
	32
	33	Add a recipe for Canon CR2 RAW files
	34	------------------------------------------------------------------------
	35	r490 \| jbj \| 2007-06-22 10:49:38 +0200 (Fri, 22 Jun 2007) \| 1 line
	36	Changed paths:
	37	M /magicrescue/trunk/configure
	38
	39	typo
	40	------------------------------------------------------------------------
	41	r489 \| jbj \| 2007-06-21 22:30:14 +0200 (Thu, 21 Jun 2007) \| 1 line
	42	Changed paths:
	43	M /magicrescue/trunk/configure
	44
	45	Syntax fix to make configure script work on the dash shell (Debian, Ubuntu, ...)
	46	------------------------------------------------------------------------
	47	r488 \| jbj \| 2007-06-03 23:30:57 +0200 (Sun, 03 Jun 2007) \| 1 line
	48	Changed paths:
	49	M /magicrescue/trunk/doc/dupemap.pod
	50
	51	typo
	52	------------------------------------------------------------------------
	53	r483 \| jbj \| 2007-05-01 22:14:43 +0200 (Tue, 01 May 2007) \| 1 line
	54	Changed paths:
	55	M /magicrescue/trunk/new_NEWS
	56	A /magicrescue/trunk/recipes/flac
	57
	58	Add flac recipe
	59	------------------------------------------------------------------------
	60	r333 \| jbj \| 2004-08-15 11:19:08 +0200 (Sun, 15 Aug 2004) \| 1 line
	61	Changed paths:
	62	M /magicrescue/trunk/configure
	63
	64	Proper cleanup after configure
	65	------------------------------------------------------------------------
	66	r332 \| jbj \| 2004-08-15 01:31:23 +0200 (Sun, 15 Aug 2004) \| 1 line
	67	Changed paths:
	68	M /magicrescue/trunk/README
	69
	70	No users know what X/OPEN is
	71	------------------------------------------------------------------------
	72	r330 \| jbj \| 2004-08-15 01:16:20 +0200 (Sun, 15 Aug 2004) \| 1 line
	73	Changed paths:
	74	M /magicrescue/trunk/NEWS
	75	M /magicrescue/trunk/new_NEWS
	76
	77	Release 1.1.4
2	78	------------------------------------------------------------------------
3	79	r329 \| jbj \| 2004-08-15 01:10:37 +0200 (Sun, 15 Aug 2004) \| 1 line
4	80	Changed paths:

+5

-0

NEWS less more

	0	Version 1.1.5: Thu Jan 17 14:03:49 CET 2008
	1	- Recipe for Flac files
	2	- Recipe for Canon CR2 RAW files
	3	- Configure script compatibility fix for Debian-based Linuxes
	4
0	5	Version 1.1.4: Sun Aug 15 01:11:27 CEST 2004
1	6	- Fixed a minor bug that prevented some not yet written recipes from working
2	7	perfectly

+4

-4

README less more

15	15	Building
16	16	--------
17	17
18		There are no build requirements other than a C library and an X/OPEN system
19		interface. To use the dupemap(1) utility, you must have the NDBM
20		compatibility header, which either comes with your system or the development
21		libraries of GDBM or Berkeley DB.
	18	There are no build requirements other than a C library and a UNIXish system.
	19	To use the dupemap(1) utility, you must have the NDBM compatibility header,
	20	which either comes with your system or the development libraries of GDBM or
	21	Berkeley DB.
22	22
23	23	./configure && make && make install
24	24

+7

-5

configure less more

32	32	MAKE_VARS=
33	33
34	34	if [ -f config.config ]; then
35		. config.config
	35	. ./config.config
36	36	fi
37	37
38	38	#

95	95	trap "mv config.h.bak config.h" 0
96	96	fi
97	97	cat > config.h << EOF
98		/* XXX This file automatically generated by configure. */
	98	/* XXX This file was automatically generated by configure. */
99	99
100	100	EOF
101	101

172	172	. "$f"
173	173	done
174	174
175		rm -f conftest.c conftest.o conftest
	175	#
	176	# Clean up
	177	#
	178	trap "" 0
	179	rm -f config.h.bak conftest.c conftest.o conftest
176	180
177	181	#
178	182	# Generate Makefile

200	204
201	205	chmod 444 Makefile
202	206
203		trap - 0
204
205	207	echo
206	208	echo "Done. Now type 'make' to compile"

+3

-3

doc/dupemap.1 less more

0		.\" Automatically generated by Pod::Man v1.37, Pod::Parser v1.14
	0	.\" Automatically generated by Pod::Man v1.37, Pod::Parser v1.32
1	1	.\"
2	2	.\" Standard preamble:
3	3	.\" ========================================================================

128	128	.\" ========================================================================
129	129	.\"
130	130	.IX Title "DUPEMAP 1"
131		.TH DUPEMAP 1 "2004-07-11" "1.1.4" "Magic Rescue"
	131	.TH DUPEMAP 1 "2007-12-31" "1.1.5" "Magic Rescue"
132	132	.SH "NAME"
133	133	dupemap \- Creates a database of file checksums and uses it to eliminate
134	134	duplicates

215	215	delete all duplicate files in \fI/tmp/recovered\-files\fR, do:
216	216	.PP
217	217	.Vb 1
218		\& $ dupemap delete /tmp/recovered files
	218	\& $ dupemap delete /tmp/recovered-files
219	219	.Ve
220	220	.PP
221	221	Often, \fBdupemap scan\fR is run to produce a checksum database of all files in a

+1

-1

doc/dupemap.pod less more

110	110	The easiest operations to understand is when the B<-d> option is not given. To
111	111	delete all duplicate files in F</tmp/recovered-files>, do:
112	112
113		$ dupemap delete /tmp/recovered files
	113	$ dupemap delete /tmp/recovered-files
114	114
115	115	Often, B<dupemap scan> is run to produce a checksum database of all files in a
116	116	directory tree. Then B<dupemap delete> is run on another directory, possibly

+2

-2

doc/magicrescue.1 less more

0		.\" Automatically generated by Pod::Man v1.37, Pod::Parser v1.14
	0	.\" Automatically generated by Pod::Man v1.37, Pod::Parser v1.32
1	1	.\"
2	2	.\" Standard preamble:
3	3	.\" ========================================================================

128	128	.\" ========================================================================
129	129	.\"
130	130	.IX Title "MAGICRESCUE 1"
131		.TH MAGICRESCUE 1 "2004-07-11" "1.1.4" "Magic Rescue"
	131	.TH MAGICRESCUE 1 "2007-12-31" "1.1.5" "Magic Rescue"
132	132	.SH "NAME"
133	133	magicrescue \- Scans a block device and extracts known file types by looking at
134	134	magic bytes.

+2

-2

doc/magicsort.1 less more

0		.\" Automatically generated by Pod::Man v1.37, Pod::Parser v1.14
	0	.\" Automatically generated by Pod::Man v1.37, Pod::Parser v1.32
1	1	.\"
2	2	.\" Standard preamble:
3	3	.\" ========================================================================

128	128	.\" ========================================================================
129	129	.\"
130	130	.IX Title "MAGICSORT 1"
131		.TH MAGICSORT 1 "2004-05-02" "1.1.4" "Magic Rescue"
	131	.TH MAGICSORT 1 "2007-12-31" "1.1.5" "Magic Rescue"
132	132	.SH "NAME"
133	133	magicsort \- Categorize files by their \fBfile\fR(1) magic
134	134	.SH "SYNOPSIS"

+7

-0

recipes/canon-cr2 less more

	0	# Extract (at most 30 MB) Canon CR2 RAW files. These are used by the Powershot
	1	# and EOS digital cameras.
	2	# This recipe was contributed by Daniel J Blueman
	3	0 string II*
	4	8 string CR
	5	extension CR2
	6	command dd bs=1024k count=30 of="$1" 2>/dev/null; dcraw -i "$1" \|\| rm -f "$1"

+15

-0

recipes/flac less more

	0	# Extracts flac sound files: http://flac.sourceforge.net/
	1	# Contributed by Richard Zidlicky <rz (at) linux (dash) m68k (dot) org>
	2
	3	0 string fLaC
	4
	5	extension flac
	6
	7	# Allow infinite overlap because the recompressed file may be much larger than
	8	# the original one.. at least in theory
	9
	10	allow_overlap -1
	11
	12
	13	# hope that flac will do the right thing
	14	command flac -d - \| flac -8 - > "$1"

+2

-2

src/dupemap.c less more

197	197
198	198	static int checksum_calc(checksum_t checksum, const char file)
199	199	{
200		static unsigned char buf[BUFLEN];
	200	static char buf[BUFLEN];
201	201	ssize_t bytes_read;
202	202	off_t _length = 0;
203	203	long _crc = 0;

211	211	}
212	212
213	213	while ((bytes_read = fread(buf, 1, BUFLEN, fp)) > 0) {
214		unsigned char *cp = buf;
	214	char *cp = buf;
215	215
216	216	_length += bytes_read;
217	217	while (bytes_read--)

+4

-4

src/magicrescue.c less more

66	66	static ssize_t overlap = 0;
67	67
68	68	/* General-purpose buffer. Mostly used to hold the raw data when scanning. */
69		static unsigned char *buf;
	69	static char *buf;
70	70	static const size_t bufsize = 102400;
71	71
72	72

116	116	* them.
117	117	* Returns 1 if an external program has been called, which means the file
118	118	* position has changed. Returns -1 on fatal error. */
119		static int scan_buf(const unsigned char *scanbuf, ssize_t scanbuf_len,
	119	static int scan_buf(const char *scanbuf, ssize_t scanbuf_len,
120	120	int fd, off_t scanbuf_filepos)
121	121	{
122	122	int filepos_changed = 0;

144	144	}
145	145
146	146	array_foreach(&scanners, scanner) {
147		const unsigned char *p = scanbuf + scanner->offset;
	147	const char *p = scanbuf + scanner->offset;
148	148
149	149	while (p - scanbuf < scanbuf_len &&
150	150	(p = scanner->func(p, scanbuf_len - (p-scanbuf) +

225	225	int fd = 0;
226	226	int result, firsttime = 1;
227	227
228		unsigned char readbuf = buf, scanbuf = buf;
	228	char readbuf = buf, scanbuf = buf;
229	229	size_t readsize = bufsize;
230	230	off_t offset_before_read;
231	231

+3

-2

src/recipe.c less more

25	25	#include "recipe.h"
26	26
27	27
28		int op_string(const unsigned char s, union param p)
	28	int op_string(const char s, union param p)
29	29	{
30	30	return memcmp(s, p->string.s, p->string.l) == 0;
31	31	}
32	32
33		int op_int32(const unsigned char s, union param p)
	33	int op_int32(const char s_signed, union param p)
34	34	{
	35	const unsigned char s = (unsigned char )s_signed;
35	36	return ((s[0]<<24 \| s[1]<<16 \| s[2]<<8 \| s[3]) & p->int32.mask)
36	37	== p->int32.val;
37	38	}

+3

-3

src/recipe.h less more

9	9	} int32;
10	10	};
11	11
12		typedef int (op_function)(const unsigned char , union param *);
	12	typedef int (op_function)(const char , union param *);
13	13
14	14	struct operation {
15	15	op_function func;

28	28	int allow_overlap;
29	29	};
30	30
31		int op_string(const unsigned char s, union param p);
32		int op_int32(const unsigned char s, union param p);
	31	int op_string(const char s, union param p);
	32	int op_int32(const char s, union param p);
33	33
34	34	void op_destroy(struct operation *op);
35	35	void recipe_init(struct recipe *r);

+5

-5

src/scanner.c less more

27	27	#include "recipe.h"
28	28	#include "scanner.h"
29	29
30		const unsigned char scanner_char(const unsigned char scanbuf,
	30	const char scanner_char(const char scanbuf,
31	31	size_t scanbuf_len, union scan_param *param, off_t offset)
32	32	{
33	33	return memchr(scanbuf, param->c, scanbuf_len);
34	34	}
35	35
36		const unsigned char scanner_block(const unsigned char scanbuf,
	36	const char scanner_block(const char scanbuf,
37	37	size_t scanbuf_len, union scan_param *param, off_t offset)
38	38	{
39	39	long extrabytes = (long)offset & (param->block-1);

46	46	return scanbuf;
47	47	}
48	48
49		const unsigned char scanner_string(const unsigned char scanbuf,
	49	const char scanner_string(const char scanbuf,
50	50	size_t scanbuf_len, union scan_param *param, off_t offset)
51	51	{
52		const unsigned char *p = scanbuf + param->scanstring.magicoff;
	52	const char *p = scanbuf + param->scanstring.magicoff;
53	53	struct string string = param->scanstring.string;
54		unsigned char magicchar = param->scanstring.magicchar;
	54	char magicchar = param->scanstring.magicchar;
55	55	scanbuf_len += param->scanstring.magicoff;
56	56
57	57	while ((size_t)(p - scanbuf) < scanbuf_len &&

+4

-4

src/scanner.h less more

14	14	} scanstring;
15	15	};
16	16
17		typedef const unsigned char (scan_function)(const unsigned char *, size_t,
	17	typedef const char (scan_function)(const char *, size_t,
18	18	union scan_param *, off_t);
19	19
20	20	struct scanner {

25	25	struct array recipes;
26	26	};
27	27
28		const unsigned char scanner_char(const unsigned char scanbuf,
	28	const char scanner_char(const char scanbuf,
29	29	size_t scanbuf_len, union scan_param *param, off_t offset);
30		const unsigned char scanner_block(const unsigned char scanbuf,
	30	const char scanner_block(const char scanbuf,
31	31	size_t scanbuf_len, union scan_param *param, off_t offset);
32		const unsigned char scanner_string(const unsigned char scanbuf,
	32	const char scanner_string(const char scanbuf,
33	33	size_t scanbuf_len, union scan_param *param, off_t offset);
34	34
35	35	void scanner_string_init(union scan_param *param);

+3

-3

src/util.c less more

31	31
32	32
33	33	/** Returns the value (0-15) of a single hex digit. Returns 0 on error. */
34		int hex2num(unsigned char c)
	34	int hex2num(char c)
35	35	{
36	36	return (c >= '0' && c <= '9' ? c-'0' :
37	37	(c >= 'a' && c <= 'f' ? c-'a'+10 :
38	38	(c >= 'A' && c <= 'F' ? c-'A'+10 : 0)));
39	39	}
40	40
41		long long hextoll(const unsigned char *str)
	41	long long hextoll(const char *str)
42	42	{
43	43	long long result = 0;
44	44	size_t i, len = strlen(str);

108	108
109	109	/** Initializes a struct string from a 0-terminated string, parsing escape
110	110	* sequences. The struct string will not be 0-terminated. */
111		void string_init(struct string dst, const unsigned char src)
	111	void string_init(struct string dst, const char src)
112	112	{
113	113	const size_t slen = strlen(src);
114	114	size_t i;

+4

-4

src/util.h less more

7	7	#endif
8	8
9	9	struct string {
10		unsigned char *s;
	10	char *s;
11	11	size_t l;
12	12	};
13	13
14		int hex2num(unsigned char c);
15		long long hextoll(const unsigned char *str);
	14	int hex2num(char c);
	15	long long hextoll(const char *str);
16	16	long atol_calc(const char *str);
17	17	off_t rich_seek(int fd, const char *string);
18	18
19		void string_init(struct string dst, const unsigned char src);
	19	void string_init(struct string dst, const char src);
20	20	void string_destroy(struct string *string);
21	21
22	22	#define _UTIL_H

+7

-6

tools/textextract.c less more

37	37	static long max_bytes = 0;
38	38
39	39	static size_t bufsize = 8192;
40		static unsigned char *buf;
	40	static char *buf;
41	41
42	42	static short scorelut[UCHAR_MAX+1], replut[UCHAR_MAX+1];
43	43
44	44	struct rule {
45	45	short score, rep;
46		unsigned char *ranges;
	46	char *ranges;
47	47	};
48	48
49	49	enum direction { DIR_FORWARD, DIR_REVERSE };
50	50	static enum direction direction = DIR_FORWARD;
51	51
52	52	struct scores {
53		unsigned char *last_letter;
	53	char *last_letter;
54	54	unsigned char repeated_char;
55	55	int sum_score, sum_repeats, cur_line;
56	56	off_t offset;

167	167	};
168	168
169	169	for (rule = rules; rule->ranges != NULL; rule++) {
170		unsigned char a, b, *range = rule->ranges;
	170	unsigned char a, b;
	171	char *range = rule->ranges;
171	172	int i;
172	173
173	174	do {

218	219	static int read_backward(struct scores *s, int outfd)
219	220	{
220	221	ssize_t read_count;
221		unsigned char *p;
	222	char *p;
222	223
223	224	if (s->offset <= 0)
224	225	return 0;

256	257	static int read_forward(struct scores *s, int outfd)
257	258	{
258	259	ssize_t read_count;
259		unsigned char p, bufpos;
	260	char p, bufpos;
260	261
261	262	bufpos = buf;
262	263