tr.c revision 102944 
1/*
2 * Copyright (c) 1988, 1993
3 *	The Regents of the University of California.  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 * 3. All advertising materials mentioning features or use of this software
14 *    must display the following acknowledgement:
15 *	This product includes software developed by the University of
16 *	California, Berkeley and its contributors.
17 * 4. Neither the name of the University nor the names of its contributors
18 *    may be used to endorse or promote products derived from this software
19 *    without specific prior written permission.
20 *
21 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
22 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
23 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
24 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
25 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
26 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
27 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
28 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
29 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
30 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
31 * SUCH DAMAGE.
32 */
33
34#include <sys/cdefs.h>
35
36__FBSDID("$FreeBSD: head/usr.bin/tr/tr.c 102944 2002-09-04 23:29:10Z dwmalone $");
37
38#ifndef lint
39static const char copyright[] =
40"@(#) Copyright (c) 1988, 1993\n\
41	The Regents of the University of California.  All rights reserved.\n";
42#endif
43
44#ifndef lint
45static const char sccsid[] = "@(#)tr.c	8.2 (Berkeley) 5/4/95";
46#endif
47
48#include <sys/types.h>
49
50#include <ctype.h>
51#include <err.h>
52#include <locale.h>
53#include <stdio.h>
54#include <stdlib.h>
55#include <string.h>
56#include <unistd.h>
57
58#include "extern.h"
59
60/*
61 * For -C option: determine whether a byte is a valid character in the
62 * current character set (as defined by LC_CTYPE).
63 */
64#define ISCHAR(c) (iscntrl(c) || isprint(c))
65
66static int string1[NCHARS] = {
67	0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,		/* ASCII */
68	0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
69	0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17,
70	0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f,
71	0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27,
72	0x28, 0x29, 0x2a, 0x2b, 0x2c, 0x2d, 0x2e, 0x2f,
73	0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37,
74	0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x3f,
75	0x40, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47,
76	0x48, 0x49, 0x4a, 0x4b, 0x4c, 0x4d, 0x4e, 0x4f,
77	0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57,
78	0x58, 0x59, 0x5a, 0x5b, 0x5c, 0x5d, 0x5e, 0x5f,
79	0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67,
80	0x68, 0x69, 0x6a, 0x6b, 0x6c, 0x6d, 0x6e, 0x6f,
81	0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77,
82	0x78, 0x79, 0x7a, 0x7b, 0x7c, 0x7d, 0x7e, 0x7f,
83	0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87,
84	0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f,
85	0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97,
86	0x98, 0x99, 0x9a, 0x9b, 0x9c, 0x9d, 0x9e, 0x9f,
87	0xa0, 0xa1, 0xa2, 0xa3, 0xa4, 0xa5, 0xa6, 0xa7,
88	0xa8, 0xa9, 0xaa, 0xab, 0xac, 0xad, 0xae, 0xaf,
89	0xb0, 0xb1, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6, 0xb7,
90	0xb8, 0xb9, 0xba, 0xbb, 0xbc, 0xbd, 0xbe, 0xbf,
91	0xc0, 0xc1, 0xc2, 0xc3, 0xc4, 0xc5, 0xc6, 0xc7,
92	0xc8, 0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf,
93	0xd0, 0xd1, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, 0xd7,
94	0xd8, 0xd9, 0xda, 0xdb, 0xdc, 0xdd, 0xde, 0xdf,
95	0xe0, 0xe1, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7,
96	0xe8, 0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef,
97	0xf0, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7,
98	0xf8, 0xf9, 0xfa, 0xfb, 0xfc, 0xfd, 0xfe, 0xff,
99}, string2[NCHARS];
100
101STR s1 = { STRING1, NORMAL, 0, OOBCH, { 0, OOBCH }, NULL, NULL };
102STR s2 = { STRING2, NORMAL, 0, OOBCH, { 0, OOBCH }, NULL, NULL };
103
104static int charcoll(const void *, const void *);
105static void setup(int *, char *, STR *, int, int);
106static void usage(void);
107
108int
109main(int argc, char **argv)
110{
111	static int collorder[NCHARS], tmpmap[NCHARS];
112	int ch, cnt, lastch, *p;
113	int Cflag, cflag, dflag, sflag, isstring2;
114
115	(void)setlocale(LC_ALL, "");
116
117	Cflag = cflag = dflag = sflag = 0;
118	while ((ch = getopt(argc, argv, "Ccdsu")) != -1)
119		switch((char)ch) {
120		case 'C':
121			Cflag = 1;
122			cflag = 0;
123			break;
124		case 'c':
125			cflag = 1;
126			Cflag = 0;
127			break;
128		case 'd':
129			dflag = 1;
130			break;
131		case 's':
132			sflag = 1;
133			break;
134		case 'u':
135			setbuf(stdout, (char *)NULL);
136			break;
137		case '?':
138		default:
139			usage();
140		}
141	argc -= optind;
142	argv += optind;
143
144	switch(argc) {
145	case 0:
146	default:
147		usage();
148		/* NOTREACHED */
149	case 1:
150		isstring2 = 0;
151		break;
152	case 2:
153		isstring2 = 1;
154		break;
155	}
156
157	/*
158	 * tr -ds [-Cc] string1 string2
159	 * Delete all characters (or complemented characters) in string1.
160	 * Squeeze all characters in string2.
161	 */
162	if (dflag && sflag) {
163		if (!isstring2)
164			usage();
165
166		setup(string1, argv[0], &s1, cflag, Cflag);
167		setup(string2, argv[1], &s2, 0, 0);
168
169		for (lastch = OOBCH; (ch = getchar()) != EOF;)
170			if (!string1[ch] && (!string2[ch] || lastch != ch)) {
171				lastch = ch;
172				(void)putchar(ch);
173			}
174		exit(0);
175	}
176
177	/*
178	 * tr -d [-Cc] string1
179	 * Delete all characters (or complemented characters) in string1.
180	 */
181	if (dflag) {
182		if (isstring2)
183			usage();
184
185		setup(string1, argv[0], &s1, cflag, Cflag);
186
187		while ((ch = getchar()) != EOF)
188			if (!string1[ch])
189				(void)putchar(ch);
190		exit(0);
191	}
192
193	/*
194	 * tr -s [-Cc] string1
195	 * Squeeze all characters (or complemented characters) in string1.
196	 */
197	if (sflag && !isstring2) {
198		setup(string1, argv[0], &s1, cflag, Cflag);
199
200		for (lastch = OOBCH; (ch = getchar()) != EOF;)
201			if (!string1[ch] || lastch != ch) {
202				lastch = ch;
203				(void)putchar(ch);
204			}
205		exit(0);
206	}
207
208	/*
209	 * tr [-Ccs] string1 string2
210	 * Replace all characters (or complemented characters) in string1 with
211	 * the character in the same position in string2.  If the -s option is
212	 * specified, squeeze all the characters in string2.
213	 */
214	if (!isstring2)
215		usage();
216
217	s1.str = argv[0];
218	s2.str = argv[1];
219
220	if (cflag || Cflag)
221		for (cnt = NCHARS, p = string1; cnt--;)
222			*p++ = OOBCH;
223
224	if (!next(&s2))
225		errx(1, "empty string2");
226
227	ch = s2.lastch;
228	/* If string2 runs out of characters, use the last one specified. */
229	if (sflag)
230		while (next(&s1)) {
231			string1[s1.lastch] = ch = s2.lastch;
232			string2[ch] = 1;
233			(void)next(&s2);
234		}
235	else
236		while (next(&s1)) {
237			string1[s1.lastch] = ch = s2.lastch;
238			(void)next(&s2);
239		}
240
241	if (cflag || Cflag) {
242		s2.str = argv[1];
243		s2.state = NORMAL;
244		for (cnt = 0, p = string1; cnt < NCHARS; ++p, ++cnt) {
245			if (*p == OOBCH && (!Cflag || ISCHAR(cnt))) {
246				(void)next(&s2);
247				*p = s2.lastch;
248			} else
249				*p = cnt;
250		}
251	}
252	if (Cflag) {
253		/*
254		 * Generate a table for locale single-byte collating element
255		 * ordering and use it to reorder string1 as required by
256		 * IEEE Std. 1003.1-2001.
257		 */
258		for (ch = 0; ch < NCHARS; ch++)
259			collorder[ch] = ch;
260		mergesort(collorder, NCHARS, sizeof(*collorder), charcoll);
261		for (ch = 0; ch < NCHARS; ch++)
262			tmpmap[ch] = string1[collorder[ch]];
263		memcpy(string1, tmpmap, sizeof(tmpmap));
264	}
265
266	if (sflag)
267		for (lastch = OOBCH; (ch = getchar()) != EOF;) {
268			ch = string1[ch];
269			if (!string2[ch] || lastch != ch) {
270				lastch = ch;
271				(void)putchar(ch);
272			}
273		}
274	else
275		while ((ch = getchar()) != EOF)
276			(void)putchar(string1[ch]);
277	exit (0);
278}
279
280static void
281setup(int *string, char *arg, STR *str, int cflag, int Cflag)
282{
283	int cnt, *p;
284
285	str->str = arg;
286	bzero(string, NCHARS * sizeof(int));
287	while (next(str))
288		string[str->lastch] = 1;
289	if (cflag)
290		for (p = string, cnt = NCHARS; cnt--; ++p)
291			*p = !*p;
292	else if (Cflag)
293		for (cnt = 0; cnt < NCHARS; cnt++)
294			string[cnt] = !string[cnt] && ISCHAR(cnt);
295}
296
297static int
298charcoll(const void *a, const void *b)
299{
300	char sa[2], sb[2];
301
302	sa[0] = *(const int *)a;
303	sb[0] = *(const int *)b;
304	sa[1] = sb[1] = '\0';
305	return (strcoll(sa, sb));
306}
307
308static void
309usage(void)
310{
311	(void)fprintf(stderr, "%s\n%s\n%s\n%s\n",
312		"usage: tr [-Ccsu] string1 string2",
313		"       tr [-Ccu] -d string1",
314		"       tr [-Ccu] -s string1",
315		"       tr [-Ccu] -ds string1 string2");
316	exit(1);
317}
318