1 /*-
2  * Copyright (c) 2007 S.Sam Arun Raj
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, this list of conditions and the following disclaimer.
10  * 2. Redistributions in binary form must reproduce the above copyright
11  *    notice, this list of conditions and the following disclaimer in the
12  *    documentation and/or other materials provided with the distribution.
13  *
14  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
15  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
16  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
17  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
18  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
19  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
20  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
21  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
22  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
23  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
24  * SUCH DAMAGE.
25  */
26 
27 #include <sys/stat.h>
28 #include <sys/types.h>
29 
30 #include <ctype.h>
31 #include <err.h>
32 #include <errno.h>
33 #include <fcntl.h>
34 #include <getopt.h>
35 #include <inttypes.h>
36 #include <stdint.h>
37 #include <stdio.h>
38 #include <stdlib.h>
39 #include <string.h>
40 #include <sysexits.h>
41 #include <unistd.h>
42 
43 #include <libelf.h>
44 #include <libelftc.h>
45 #include <gelf.h>
46 
47 #include "_elftc.h"
48 
49 ELFTC_VCSID("$Id: strings.c 3360 2016-01-24 18:34:06Z jkoshy $");
50 
51 enum return_code {
52 	RETURN_OK,
53 	RETURN_NOINPUT,
54 	RETURN_SOFTWARE
55 };
56 
57 enum radix_style {
58 	RADIX_DECIMAL,
59 	RADIX_HEX,
60 	RADIX_OCTAL
61 };
62 
63 enum encoding_style {
64 	ENCODING_7BIT,
65 	ENCODING_8BIT,
66 	ENCODING_16BIT_BIG,
67 	ENCODING_16BIT_LITTLE,
68 	ENCODING_32BIT_BIG,
69 	ENCODING_32BIT_LITTLE
70 };
71 
72 #define PRINTABLE(c)						\
73       ((c) >= 0 && (c) <= 255 && 				\
74 	  ((c) == '\t' || isprint((c)) ||			\
75 	      (encoding == ENCODING_8BIT && (c) > 127)))
76 
77 static int encoding_size, entire_file, show_filename, show_loc;
78 static enum encoding_style encoding;
79 static enum radix_style radix;
80 static intmax_t min_len;
81 
82 static struct option strings_longopts[] = {
83 	{ "all",		no_argument,		NULL,	'a'},
84 	{ "bytes",		required_argument,	NULL,	'n'},
85 	{ "encoding",		required_argument,	NULL,	'e'},
86 	{ "help",		no_argument,		NULL,	'h'},
87 	{ "print-file-name",	no_argument,		NULL,	'f'},
88 	{ "radix",		required_argument,	NULL,	't'},
89 	{ "version",		no_argument,		NULL,	'v'},
90 	{ NULL, 0, NULL, 0 }
91 };
92 
93 long	getcharacter(void);
94 int	handle_file(const char *);
95 int	handle_elf(const char *, int);
96 int	handle_binary(const char *, int);
97 int	find_strings(const char *, off_t, off_t);
98 void	show_version(void);
99 void	usage(void);
100 
101 /*
102  * strings(1) extracts text(contiguous printable characters)
103  * from elf and binary files.
104  */
105 int
106 main(int argc, char **argv)
107 {
108 	int ch, rc;
109 
110 	rc = RETURN_OK;
111 	min_len = 0;
112 	encoding_size = 1;
113 	if (elf_version(EV_CURRENT) == EV_NONE)
114 		errx(EXIT_FAILURE, "ELF library initialization failed: %s",
115 		    elf_errmsg(-1));
116 
117 	while ((ch = getopt_long(argc, argv, "1234567890ae:fhn:ot:Vv",
118 	    strings_longopts, NULL)) != -1)
119 		switch((char)ch) {
120 		case 'a':
121 			entire_file = 1;
122 			break;
123 		case 'e':
124 			if (*optarg == 's') {
125 				encoding = ENCODING_7BIT;
126 			} else if (*optarg == 'S') {
127 				encoding = ENCODING_8BIT;
128 			} else if (*optarg == 'b') {
129 				encoding = ENCODING_16BIT_BIG;
130 				encoding_size = 2;
131 			} else if (*optarg == 'B') {
132 				encoding = ENCODING_32BIT_BIG;
133 				encoding_size = 4;
134 			} else if (*optarg == 'l') {
135 				encoding = ENCODING_16BIT_LITTLE;
136 				encoding_size = 2;
137 			} else if (*optarg == 'L') {
138 				encoding = ENCODING_32BIT_LITTLE;
139 				encoding_size = 4;
140 			} else
141 				usage();
142 			        /* NOTREACHED */
143 			break;
144 		case 'f':
145 			show_filename = 1;
146 			break;
147 		case 'n':
148 			min_len = strtoimax(optarg, (char**)NULL, 10);
149 			if (min_len <= 0)
150 				errx(EX_USAGE, "option -n should specify a "
151 				    "positive decimal integer.");
152 			break;
153 		case 'o':
154 			show_loc = 1;
155 			radix = RADIX_OCTAL;
156 			break;
157 		case 't':
158 			show_loc = 1;
159 			if (*optarg == 'd')
160 				radix = RADIX_DECIMAL;
161 			else if (*optarg == 'o')
162 				radix = RADIX_OCTAL;
163 			else if (*optarg == 'x')
164 				radix = RADIX_HEX;
165 			else
166 				usage();
167 			        /* NOTREACHED */
168 			break;
169 		case 'v':
170 		case 'V':
171 			show_version();
172 			/* NOTREACHED */
173 		case '0':
174 	        case '1':
175 		case '2':
176 		case '3':
177 		case '4':
178 		case '5':
179 		case '6':
180 		case '7':
181 		case '8':
182 		case '9':
183 			min_len *= 10;
184 			min_len += ch - '0';
185 			break;
186 		case 'h':
187 		case '?':
188 		default:
189 			usage();
190 			/* NOTREACHED */
191 		}
192 	argc -= optind;
193 	argv += optind;
194 
195 	if (!min_len)
196 		min_len = 4;
197 	if (!*argv)
198 		rc = handle_file("{standard input}");
199 	else while (*argv) {
200 		rc = handle_file(*argv);
201 		argv++;
202 	}
203 	return (rc);
204 }
205 
206 int
207 handle_file(const char *name)
208 {
209 	int fd, rt;
210 
211 	if (name == NULL)
212 		return (RETURN_NOINPUT);
213 	if (strcmp("{standard input}", name) != 0) {
214 		if (freopen(name, "rb", stdin) == NULL) {
215 			warnx("'%s': %s", name, strerror(errno));
216 			return (RETURN_NOINPUT);
217 		}
218 	} else {
219 		return (find_strings(name, (off_t)0, (off_t)0));
220 	}
221 
222 	fd = fileno(stdin);
223 	if (fd < 0)
224 		return (RETURN_NOINPUT);
225 	rt = handle_elf(name, fd);
226 	return (rt);
227 }
228 
229 /*
230  * Files not understood by handle_elf, will be passed off here and will
231  * treated as a binary file. This would include text file, core dumps ...
232  */
233 int
234 handle_binary(const char *name, int fd)
235 {
236 	struct stat buf;
237 
238 	memset(&buf, 0, sizeof(struct stat));
239 	(void) lseek(fd, (off_t)0, SEEK_SET);
240 	if (!fstat(fd, &buf))
241 		return (find_strings(name, (off_t)0, buf.st_size));
242 	return (RETURN_SOFTWARE);
243 }
244 
245 /*
246  * Will analyse a file to see if it ELF, other files including ar(1),
247  * core dumps are passed off and treated as flat binary files. Unlike
248  * GNU size in FreeBSD this routine will not treat ELF object from
249  * different archs as flat binary files(has to overridden using -a).
250  */
251 int
252 handle_elf(const char *name, int fd)
253 {
254 	GElf_Ehdr elfhdr;
255 	GElf_Shdr shdr;
256 	Elf *elf;
257 	Elf_Scn *scn;
258 	int rc;
259 
260 	rc = RETURN_OK;
261 	/* If entire file is choosen, treat it as a binary file */
262 	if (entire_file)
263 		return (handle_binary(name, fd));
264 
265 	(void) lseek(fd, (off_t)0, SEEK_SET);
266 	elf = elf_begin(fd, ELF_C_READ, NULL);
267 	if (elf_kind(elf) != ELF_K_ELF) {
268 		(void) elf_end(elf);
269 		return (handle_binary(name, fd));
270 	}
271 
272 	if (gelf_getehdr(elf, &elfhdr) == NULL) {
273 		(void) elf_end(elf);
274 		warnx("%s: ELF file could not be processed", name);
275 		return (RETURN_SOFTWARE);
276 	}
277 
278 	if (elfhdr.e_shnum == 0 && elfhdr.e_type == ET_CORE) {
279 		(void) elf_end(elf);
280 		return (handle_binary(name, fd));
281 	} else {
282 		scn = NULL;
283 		while ((scn = elf_nextscn(elf, scn)) != NULL) {
284 			if (gelf_getshdr(scn, &shdr) == NULL)
285 				continue;
286 			if (shdr.sh_type != SHT_NOBITS &&
287 			    (shdr.sh_flags & SHF_ALLOC) != 0) {
288 				rc = find_strings(name, shdr.sh_offset,
289 				    shdr.sh_size);
290 			}
291 		}
292 	}
293 	(void) elf_end(elf);
294 	return (rc);
295 }
296 
297 /*
298  * Retrieves a character from input stream based on the encoding
299  * type requested.
300  */
301 long
302 getcharacter(void)
303 {
304 	long rt;
305 	int i;
306 	char buf[4], c;
307 
308 	rt = EOF;
309 	for(i = 0; i < encoding_size; i++) {
310 		c = getc(stdin);
311 		if (feof(stdin))
312 			return (EOF);
313 		buf[i] = c;
314 	}
315 
316 	switch(encoding) {
317 	case ENCODING_7BIT:
318 	case ENCODING_8BIT:
319 		rt = buf[0];
320 		break;
321 	case ENCODING_16BIT_BIG:
322 		rt = (buf[0] << 8) | buf[1];
323 		break;
324 	case ENCODING_16BIT_LITTLE:
325 		 rt = buf[0] | (buf[1] << 8);
326 		 break;
327 	case ENCODING_32BIT_BIG:
328 		rt = ((long) buf[0] << 24) | ((long) buf[1] << 16) |
329            	    ((long) buf[2] << 8) | buf[3];
330            	break;
331 	case ENCODING_32BIT_LITTLE:
332 		rt = buf[0] | ((long) buf[1] << 8) | ((long) buf[2] << 16) |
333         	    ((long) buf[3] << 24);
334            	break;
335 	}
336 	return (rt);
337 }
338 
339 /*
340  * Input stream stdin is read until the end of file is reached or until
341  * the section size is reached in case of ELF files. Contiguous
342  * characters of >= min_size(default 4) will be displayed.
343  */
344 int
345 find_strings(const char *name, off_t offset, off_t size)
346 {
347 	off_t cur_off, start_off;
348 	char *obuf;
349 	long c;
350 	int i;
351 
352 	if ((obuf = (char*)calloc(1, min_len + 1)) == NULL) {
353 		(void) fprintf(stderr, "Unable to allocate memory: %s\n",
354 		     strerror(errno));
355 		return (RETURN_SOFTWARE);
356 	}
357 
358 	(void) fseeko(stdin, offset, SEEK_SET);
359 	cur_off = offset;
360 	start_off = 0;
361 	while(1) {
362 		if ((offset + size) && (cur_off >= offset + size))
363 			break;
364 		start_off = cur_off;
365 		memset(obuf, 0, min_len+1);
366 		for(i = 0; i < min_len; i++) {
367 			c = getcharacter();
368 			if (c == EOF && feof(stdin))
369 				goto _exit1;
370 		 	if (PRINTABLE(c)) {
371 		 		obuf[i] = c;
372 		 		obuf[i+1] = 0;
373 		 		cur_off += encoding_size;
374 		 	} else {
375 				if (encoding == ENCODING_8BIT &&
376 				    (uint8_t)c > 127) {
377 			 		obuf[i] = c;
378 			 		obuf[i+1] = 0;
379 			 		cur_off += encoding_size;
380 			 		continue;
381 			 	}
382 	 			cur_off += encoding_size;
383 	 			break;
384 		 	}
385 		}
386 
387 		if (i >= min_len && ((cur_off <= offset + size) ||
388 		    !(offset + size))) {
389 			if (show_filename)
390 				printf ("%s: ", name);
391 			if (show_loc) {
392 				switch(radix) {
393 				case RADIX_DECIMAL:
394 					(void) printf("%7ju ",
395 					    (uintmax_t)start_off);
396 					break;
397 				case RADIX_HEX:
398 					(void) printf("%7jx ",
399 					    (uintmax_t)start_off);
400 					break;
401 				case RADIX_OCTAL:
402 					(void) printf("%7jo ",
403 					    (uintmax_t)start_off);
404 					break;
405 				}
406 			}
407 			printf("%s", obuf);
408 
409 			while(1) {
410 				if ((offset + size) &&
411 				    (cur_off >= offset + size))
412 					break;
413 				c = getcharacter();
414 				cur_off += encoding_size;
415 				if (encoding == ENCODING_8BIT &&
416 				    (uint8_t)c > 127) {
417 			 		putchar(c);
418 			 		continue;
419 			 	}
420 				if (!PRINTABLE(c) || c == EOF)
421 					break;
422 				putchar(c);
423 			}
424 			putchar('\n');
425 		}
426 	}
427 _exit1:
428 	free(obuf);
429 	return (RETURN_OK);
430 }
431 
432 #define	USAGE_MESSAGE	"\
433 Usage: %s [options] [file...]\n\
434   Print contiguous sequences of printable characters.\n\n\
435   Options:\n\
436   -a     | --all               Scan the entire file for strings.\n\
437   -e ENC | --encoding=ENC      Select the character encoding to use.\n\
438   -f     | --print-file-name   Print the file name before each string.\n\
439   -h     | --help              Print a help message and exit.\n\
440   -n N   | --bytes=N | -N      Print sequences with 'N' or more characters.\n\
441   -o                           Print offsets in octal.\n\
442   -t R   | --radix=R           Print offsets using the radix named by 'R'.\n\
443   -v     | --version           Print a version identifier and exit.\n"
444 
445 void
446 usage(void)
447 {
448 	(void) fprintf(stderr, USAGE_MESSAGE, ELFTC_GETPROGNAME());
449 	exit(EXIT_FAILURE);
450 }
451 
452 void
453 show_version(void)
454 {
455         (void) printf("%s (%s)\n", ELFTC_GETPROGNAME(), elftc_version());
456         exit(EXIT_SUCCESS);
457 }
458