tr.c revision 118373 
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 118373 2003-08-03 04:18:07Z ache $");
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 void setup(int *, char *, STR *, int, int);
105static void usage(void);
106
107int
108main(int argc, char **argv)
109{
110	static int collorder[NCHARS], tmpmap[NCHARS];
111	int ch, cnt, lastch, *p;
112	int Cflag, cflag, dflag, sflag, isstring2;
113
114	(void)setlocale(LC_ALL, "");
115
116	Cflag = cflag = dflag = sflag = 0;
117	while ((ch = getopt(argc, argv, "Ccdsu")) != -1)
118		switch((char)ch) {
119		case 'C':
120			Cflag = 1;
121			cflag = 0;
122			break;
123		case 'c':
124			cflag = 1;
125			Cflag = 0;
126			break;
127		case 'd':
128			dflag = 1;
129			break;
130		case 's':
131			sflag = 1;
132			break;
133		case 'u':
134			setbuf(stdout, (char *)NULL);
135			break;
136		case '?':
137		default:
138			usage();
139		}
140	argc -= optind;
141	argv += optind;
142
143	switch(argc) {
144	case 0:
145	default:
146		usage();
147		/* NOTREACHED */
148	case 1:
149		isstring2 = 0;
150		break;
151	case 2:
152		isstring2 = 1;
153		break;
154	}
155
156	/*
157	 * tr -ds [-Cc] string1 string2
158	 * Delete all characters (or complemented characters) in string1.
159	 * Squeeze all characters in string2.
160	 */
161	if (dflag && sflag) {
162		if (!isstring2)
163			usage();
164
165		setup(string1, argv[0], &s1, cflag, Cflag);
166		setup(string2, argv[1], &s2, 0, 0);
167
168		for (lastch = OOBCH; (ch = getchar()) != EOF;)
169			if (!string1[ch] && (!string2[ch] || lastch != ch)) {
170				lastch = ch;
171				(void)putchar(ch);
172			}
173		exit(0);
174	}
175
176	/*
177	 * tr -d [-Cc] string1
178	 * Delete all characters (or complemented characters) in string1.
179	 */
180	if (dflag) {
181		if (isstring2)
182			usage();
183
184		setup(string1, argv[0], &s1, cflag, Cflag);
185
186		while ((ch = getchar()) != EOF)
187			if (!string1[ch])
188				(void)putchar(ch);
189		exit(0);
190	}
191
192	/*
193	 * tr -s [-Cc] string1
194	 * Squeeze all characters (or complemented characters) in string1.
195	 */
196	if (sflag && !isstring2) {
197		setup(string1, argv[0], &s1, cflag, Cflag);
198
199		for (lastch = OOBCH; (ch = getchar()) != EOF;)
200			if (!string1[ch] || lastch != ch) {
201				lastch = ch;
202				(void)putchar(ch);
203			}
204		exit(0);
205	}
206
207	/*
208	 * tr [-Ccs] string1 string2
209	 * Replace all characters (or complemented characters) in string1 with
210	 * the character in the same position in string2.  If the -s option is
211	 * specified, squeeze all the characters in string2.
212	 */
213	if (!isstring2)
214		usage();
215
216	s1.str = argv[0];
217	s2.str = argv[1];
218
219	if (cflag || Cflag)
220		for (cnt = NCHARS, p = string1; cnt--;)
221			*p++ = OOBCH;
222
223	if (!next(&s2))
224		errx(1, "empty string2");
225
226	/*
227	 * For -s result will contain only those characters defined
228	 * as the second characters in each of the toupper or tolower
229	 * pairs.
230	 */
231
232	/* If string2 runs out of characters, use the last one specified. */
233	while (next(&s1)) {
234	again:
235		if (s1.state == SET_LOWER &&
236		    s2.state == SET_UPPER &&
237		    s1.cnt == 1 && s2.cnt == 1) {
238			do {
239				string1[s1.lastch] = ch = toupper(s1.lastch);
240				if (sflag && isupper(ch))
241					string2[ch] = 1;
242				if (!next(&s1))
243					goto endloop;
244			} while (s1.state == SET_LOWER && s1.cnt > 1);
245			/* skip upper set */
246			do {
247				if (!next(&s2))
248					break;
249			} while (s2.state == SET_UPPER && s2.cnt > 1);
250			goto again;
251		} else if (s1.state == SET_UPPER &&
252			   s2.state == SET_LOWER &&
253			   s1.cnt == 1 && s2.cnt == 1) {
254			do {
255				string1[s1.lastch] = ch = tolower(s1.lastch);
256				if (sflag && islower(ch))
257					string2[ch] = 1;
258				if (!next(&s1))
259					goto endloop;
260			} while (s1.state == SET_UPPER && s1.cnt > 1);
261			/* skip lower set */
262			do {
263				if (!next(&s2))
264					break;
265			} while (s2.state == SET_LOWER && s2.cnt > 1);
266			goto again;
267		} else {
268			string1[s1.lastch] = s2.lastch;
269			if (sflag)
270				string2[s2.lastch] = 1;
271		}
272		(void)next(&s2);
273	}
274endloop:
275	if (cflag || Cflag) {
276		s2.str = argv[1];
277		s2.state = NORMAL;
278		for (cnt = 0, p = string1; cnt < NCHARS; ++p, ++cnt) {
279			if (*p == OOBCH && (!Cflag || ISCHAR(cnt))) {
280				(void)next(&s2);
281				*p = s2.lastch;
282			} else
283				*p = cnt;
284		}
285	}
286	if (Cflag) {
287		/*
288		 * Generate a table for locale single-byte collating element
289		 * ordering and use it to reorder string1 as required by
290		 * IEEE Std. 1003.1-2001.
291		 */
292		for (ch = 0; ch < NCHARS; ch++)
293			collorder[ch] = ch;
294		mergesort(collorder, NCHARS, sizeof(*collorder), charcoll);
295		for (ch = 0; ch < NCHARS; ch++)
296			tmpmap[ch] = string1[collorder[ch]];
297		memcpy(string1, tmpmap, sizeof(tmpmap));
298	}
299
300	if (sflag)
301		for (lastch = OOBCH; (ch = getchar()) != EOF;) {
302			ch = string1[ch];
303			if (!string2[ch] || lastch != ch) {
304				lastch = ch;
305				(void)putchar(ch);
306			}
307		}
308	else
309		while ((ch = getchar()) != EOF)
310			(void)putchar(string1[ch]);
311	exit (0);
312}
313
314static void
315setup(int *string, char *arg, STR *str, int cflag, int Cflag)
316{
317	int cnt, *p;
318
319	str->str = arg;
320	bzero(string, NCHARS * sizeof(int));
321	while (next(str))
322		string[str->lastch] = 1;
323	if (cflag)
324		for (p = string, cnt = NCHARS; cnt--; ++p)
325			*p = !*p;
326	else if (Cflag)
327		for (cnt = 0; cnt < NCHARS; cnt++)
328			string[cnt] = !string[cnt] && ISCHAR(cnt);
329}
330
331int
332charcoll(const void *a, const void *b)
333{
334	static char sa[2], sb[2];
335
336	sa[0] = *(const int *)a;
337	sb[0] = *(const int *)b;
338	return (strcoll(sa, sb));
339}
340
341static void
342usage(void)
343{
344	(void)fprintf(stderr, "%s\n%s\n%s\n%s\n",
345		"usage: tr [-Ccsu] string1 string2",
346		"       tr [-Ccu] -d string1",
347		"       tr [-Ccu] -s string1",
348		"       tr [-Ccu] -ds string1 string2");
349	exit(1);
350}
351