tr.c revision 330897 
1/*-
2 * SPDX-License-Identifier: BSD-3-Clause
3 *
4 * Copyright (c) 1988, 1993
5 *	The Regents of the University of California.  All rights reserved.
6 *
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions
9 * are met:
10 * 1. Redistributions of source code must retain the above copyright
11 *    notice, this list of conditions and the following disclaimer.
12 * 2. Redistributions in binary form must reproduce the above copyright
13 *    notice, this list of conditions and the following disclaimer in the
14 *    documentation and/or other materials provided with the distribution.
15 * 4. Neither the name of the University nor the names of its contributors
16 *    may be used to endorse or promote products derived from this software
17 *    without specific prior written permission.
18 *
19 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
20 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
21 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
22 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
23 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
24 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
25 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
26 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
27 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
28 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
29 * SUCH DAMAGE.
30 */
31
32#include <sys/cdefs.h>
33
34__FBSDID("$FreeBSD: stable/11/usr.bin/tr/tr.c 330897 2018-03-14 03:19:51Z eadler $");
35
36#ifndef lint
37static const char copyright[] =
38"@(#) Copyright (c) 1988, 1993\n\
39	The Regents of the University of California.  All rights reserved.\n";
40#endif
41
42#ifndef lint
43static const char sccsid[] = "@(#)tr.c	8.2 (Berkeley) 5/4/95";
44#endif
45
46#include <sys/types.h>
47
48#include <ctype.h>
49#include <err.h>
50#include <limits.h>
51#include <locale.h>
52#include <stdint.h>
53#include <stdio.h>
54#include <stdlib.h>
55#include <string.h>
56#include <unistd.h>
57#include <wchar.h>
58#include <wctype.h>
59
60#include "cmap.h"
61#include "cset.h"
62#include "extern.h"
63
64static STR s1 = { STRING1, NORMAL, 0, OOBCH, 0, { 0, OOBCH }, NULL, NULL };
65static STR s2 = { STRING2, NORMAL, 0, OOBCH, 0, { 0, OOBCH }, NULL, NULL };
66
67static struct cset *setup(char *, STR *, int, int);
68static void usage(void);
69
70int
71main(int argc, char **argv)
72{
73	static int carray[NCHARS_SB];
74	struct cmap *map;
75	struct cset *delete, *squeeze;
76	int n, *p;
77	int Cflag, cflag, dflag, sflag, isstring2;
78	wint_t ch, cnt, lastch;
79
80	(void)setlocale(LC_ALL, "");
81
82	Cflag = cflag = dflag = sflag = 0;
83	while ((ch = getopt(argc, argv, "Ccdsu")) != -1)
84		switch((char)ch) {
85		case 'C':
86			Cflag = 1;
87			cflag = 0;
88			break;
89		case 'c':
90			cflag = 1;
91			Cflag = 0;
92			break;
93		case 'd':
94			dflag = 1;
95			break;
96		case 's':
97			sflag = 1;
98			break;
99		case 'u':
100			setbuf(stdout, (char *)NULL);
101			break;
102		case '?':
103		default:
104			usage();
105		}
106	argc -= optind;
107	argv += optind;
108
109	switch(argc) {
110	case 0:
111	default:
112		usage();
113		/* NOTREACHED */
114	case 1:
115		isstring2 = 0;
116		break;
117	case 2:
118		isstring2 = 1;
119		break;
120	}
121
122	/*
123	 * tr -ds [-Cc] string1 string2
124	 * Delete all characters (or complemented characters) in string1.
125	 * Squeeze all characters in string2.
126	 */
127	if (dflag && sflag) {
128		if (!isstring2)
129			usage();
130
131		delete = setup(argv[0], &s1, cflag, Cflag);
132		squeeze = setup(argv[1], &s2, 0, 0);
133
134		for (lastch = OOBCH; (ch = getwchar()) != WEOF;)
135			if (!cset_in(delete, ch) &&
136			    (lastch != ch || !cset_in(squeeze, ch))) {
137				lastch = ch;
138				(void)putwchar(ch);
139			}
140		if (ferror(stdin))
141			err(1, NULL);
142		exit(0);
143	}
144
145	/*
146	 * tr -d [-Cc] string1
147	 * Delete all characters (or complemented characters) in string1.
148	 */
149	if (dflag) {
150		if (isstring2)
151			usage();
152
153		delete = setup(argv[0], &s1, cflag, Cflag);
154
155		while ((ch = getwchar()) != WEOF)
156			if (!cset_in(delete, ch))
157				(void)putwchar(ch);
158		if (ferror(stdin))
159			err(1, NULL);
160		exit(0);
161	}
162
163	/*
164	 * tr -s [-Cc] string1
165	 * Squeeze all characters (or complemented characters) in string1.
166	 */
167	if (sflag && !isstring2) {
168		squeeze = setup(argv[0], &s1, cflag, Cflag);
169
170		for (lastch = OOBCH; (ch = getwchar()) != WEOF;)
171			if (lastch != ch || !cset_in(squeeze, ch)) {
172				lastch = ch;
173				(void)putwchar(ch);
174			}
175		if (ferror(stdin))
176			err(1, NULL);
177		exit(0);
178	}
179
180	/*
181	 * tr [-Ccs] string1 string2
182	 * Replace all characters (or complemented characters) in string1 with
183	 * the character in the same position in string2.  If the -s option is
184	 * specified, squeeze all the characters in string2.
185	 */
186	if (!isstring2)
187		usage();
188
189	map = cmap_alloc();
190	if (map == NULL)
191		err(1, NULL);
192	squeeze = cset_alloc();
193	if (squeeze == NULL)
194		err(1, NULL);
195
196	s1.str = argv[0];
197
198	if (Cflag || cflag) {
199		cmap_default(map, OOBCH);
200		if ((s2.str = strdup(argv[1])) == NULL)
201			errx(1, "strdup(argv[1])");
202	} else
203		s2.str = argv[1];
204
205	if (!next(&s2))
206		errx(1, "empty string2");
207
208	/*
209	 * For -s result will contain only those characters defined
210	 * as the second characters in each of the toupper or tolower
211	 * pairs.
212	 */
213
214	/* If string2 runs out of characters, use the last one specified. */
215	while (next(&s1)) {
216	again:
217		if (s1.state == CCLASS_LOWER &&
218		    s2.state == CCLASS_UPPER &&
219		    s1.cnt == 1 && s2.cnt == 1) {
220			do {
221				ch = towupper(s1.lastch);
222				cmap_add(map, s1.lastch, ch);
223				if (sflag && iswupper(ch))
224					cset_add(squeeze, ch);
225				if (!next(&s1))
226					goto endloop;
227			} while (s1.state == CCLASS_LOWER && s1.cnt > 1);
228			/* skip upper set */
229			do {
230				if (!next(&s2))
231					break;
232			} while (s2.state == CCLASS_UPPER && s2.cnt > 1);
233			goto again;
234		} else if (s1.state == CCLASS_UPPER &&
235			   s2.state == CCLASS_LOWER &&
236			   s1.cnt == 1 && s2.cnt == 1) {
237			do {
238				ch = towlower(s1.lastch);
239				cmap_add(map, s1.lastch, ch);
240				if (sflag && iswlower(ch))
241					cset_add(squeeze, ch);
242				if (!next(&s1))
243					goto endloop;
244			} while (s1.state == CCLASS_UPPER && s1.cnt > 1);
245			/* skip lower set */
246			do {
247				if (!next(&s2))
248					break;
249			} while (s2.state == CCLASS_LOWER && s2.cnt > 1);
250			goto again;
251		} else {
252			cmap_add(map, s1.lastch, s2.lastch);
253			if (sflag)
254				cset_add(squeeze, s2.lastch);
255		}
256		(void)next(&s2);
257	}
258endloop:
259	if (cflag || (Cflag && MB_CUR_MAX > 1)) {
260		/*
261		 * This is somewhat tricky: since the character set is
262		 * potentially huge, we need to avoid allocating a map
263		 * entry for every character. Our strategy is to set the
264		 * default mapping to the last character of string #2
265		 * (= the one that gets automatically repeated), then to
266		 * add back identity mappings for characters that should
267		 * remain unchanged. We don't waste space on identity mappings
268		 * for non-characters with the -C option; those are simulated
269		 * in the I/O loop.
270		 */
271		s2.str = argv[1];
272		s2.state = NORMAL;
273		for (cnt = 0; cnt < WINT_MAX; cnt++) {
274			if (Cflag && !iswrune(cnt))
275				continue;
276			if (cmap_lookup(map, cnt) == OOBCH) {
277				if (next(&s2)) {
278					cmap_add(map, cnt, s2.lastch);
279					if (sflag)
280						cset_add(squeeze, s2.lastch);
281				}
282			} else
283				cmap_add(map, cnt, cnt);
284			if ((s2.state == EOS || s2.state == INFINITE) &&
285			    cnt >= cmap_max(map))
286				break;
287		}
288		cmap_default(map, s2.lastch);
289	} else if (Cflag) {
290		for (p = carray, cnt = 0; cnt < NCHARS_SB; cnt++) {
291			if (cmap_lookup(map, cnt) == OOBCH && iswrune(cnt))
292				*p++ = cnt;
293			else
294				cmap_add(map, cnt, cnt);
295		}
296		n = p - carray;
297		if (Cflag && n > 1)
298			(void)mergesort(carray, n, sizeof(*carray), charcoll);
299
300		s2.str = argv[1];
301		s2.state = NORMAL;
302		for (cnt = 0; cnt < n; cnt++) {
303			(void)next(&s2);
304			cmap_add(map, carray[cnt], s2.lastch);
305			/*
306			 * Chars taken from s2 can be different this time
307			 * due to lack of complex upper/lower processing,
308			 * so fill string2 again to not miss some.
309			 */
310			if (sflag)
311				cset_add(squeeze, s2.lastch);
312		}
313	}
314
315	cset_cache(squeeze);
316	cmap_cache(map);
317
318	if (sflag)
319		for (lastch = OOBCH; (ch = getwchar()) != WEOF;) {
320			if (!Cflag || iswrune(ch))
321				ch = cmap_lookup(map, ch);
322			if (lastch != ch || !cset_in(squeeze, ch)) {
323				lastch = ch;
324				(void)putwchar(ch);
325			}
326		}
327	else
328		while ((ch = getwchar()) != WEOF) {
329			if (!Cflag || iswrune(ch))
330				ch = cmap_lookup(map, ch);
331			(void)putwchar(ch);
332		}
333	if (ferror(stdin))
334		err(1, NULL);
335	exit (0);
336}
337
338static struct cset *
339setup(char *arg, STR *str, int cflag, int Cflag)
340{
341	struct cset *cs;
342
343	cs = cset_alloc();
344	if (cs == NULL)
345		err(1, NULL);
346	str->str = arg;
347	while (next(str))
348		cset_add(cs, str->lastch);
349	if (Cflag)
350		cset_addclass(cs, wctype("rune"), true);
351	if (cflag || Cflag)
352		cset_invert(cs);
353	cset_cache(cs);
354	return (cs);
355}
356
357int
358charcoll(const void *a, const void *b)
359{
360	static char sa[2], sb[2];
361
362	sa[0] = *(const int *)a;
363	sb[0] = *(const int *)b;
364	return (strcoll(sa, sb));
365}
366
367static void
368usage(void)
369{
370	(void)fprintf(stderr, "%s\n%s\n%s\n%s\n",
371		"usage: tr [-Ccsu] string1 string2",
372		"       tr [-Ccu] -d string1",
373		"       tr [-Ccu] -s string1",
374		"       tr [-Ccu] -ds string1 string2");
375	exit(1);
376}
377