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 * 3. 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("$FreeBSD$");
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#include <sys/capsicum.h>
48
49#include <capsicum_helpers.h>
50#include <ctype.h>
51#include <err.h>
52#include <limits.h>
53#include <locale.h>
54#include <stdint.h>
55#include <stdio.h>
56#include <stdlib.h>
57#include <string.h>
58#include <unistd.h>
59#include <wchar.h>
60#include <wctype.h>
61
62#include "cmap.h"
63#include "cset.h"
64#include "extern.h"
65
66static STR s1 = { STRING1, NORMAL, 0, OOBCH, 0, { 0, OOBCH }, NULL, NULL };
67static STR s2 = { STRING2, NORMAL, 0, OOBCH, 0, { 0, OOBCH }, NULL, NULL };
68
69static struct cset *setup(char *, STR *, int, int);
70static void usage(void);
71
72int
73main(int argc, char **argv)
74{
75	static int carray[NCHARS_SB];
76	struct cmap *map;
77	struct cset *delete, *squeeze;
78	int n, *p;
79	int Cflag, cflag, dflag, sflag, isstring2;
80	wint_t ch, cnt, lastch;
81
82	(void)setlocale(LC_ALL, "");
83
84	if (caph_limit_stdio() == -1)
85		err(1, "unable to limit stdio");
86
87	if (caph_enter() < 0)
88		err(1, "unable to enter capability mode");
89
90	Cflag = cflag = dflag = sflag = 0;
91	while ((ch = getopt(argc, argv, "Ccdsu")) != -1)
92		switch((char)ch) {
93		case 'C':
94			Cflag = 1;
95			cflag = 0;
96			break;
97		case 'c':
98			cflag = 1;
99			Cflag = 0;
100			break;
101		case 'd':
102			dflag = 1;
103			break;
104		case 's':
105			sflag = 1;
106			break;
107		case 'u':
108			setbuf(stdout, (char *)NULL);
109			break;
110		case '?':
111		default:
112			usage();
113		}
114	argc -= optind;
115	argv += optind;
116
117	switch(argc) {
118	case 0:
119	default:
120		usage();
121		/* NOTREACHED */
122	case 1:
123		isstring2 = 0;
124		break;
125	case 2:
126		isstring2 = 1;
127		break;
128	}
129
130	/*
131	 * tr -ds [-Cc] string1 string2
132	 * Delete all characters (or complemented characters) in string1.
133	 * Squeeze all characters in string2.
134	 */
135	if (dflag && sflag) {
136		if (!isstring2)
137			usage();
138
139		delete = setup(argv[0], &s1, cflag, Cflag);
140		squeeze = setup(argv[1], &s2, 0, 0);
141
142		for (lastch = OOBCH; (ch = getwchar()) != WEOF;)
143			if (!cset_in(delete, ch) &&
144			    (lastch != ch || !cset_in(squeeze, ch))) {
145				lastch = ch;
146				(void)putwchar(ch);
147			}
148		if (ferror(stdin))
149			err(1, NULL);
150		exit(0);
151	}
152
153	/*
154	 * tr -d [-Cc] string1
155	 * Delete all characters (or complemented characters) in string1.
156	 */
157	if (dflag) {
158		if (isstring2)
159			usage();
160
161		delete = setup(argv[0], &s1, cflag, Cflag);
162
163		while ((ch = getwchar()) != WEOF)
164			if (!cset_in(delete, ch))
165				(void)putwchar(ch);
166		if (ferror(stdin))
167			err(1, NULL);
168		exit(0);
169	}
170
171	/*
172	 * tr -s [-Cc] string1
173	 * Squeeze all characters (or complemented characters) in string1.
174	 */
175	if (sflag && !isstring2) {
176		squeeze = setup(argv[0], &s1, cflag, Cflag);
177
178		for (lastch = OOBCH; (ch = getwchar()) != WEOF;)
179			if (lastch != ch || !cset_in(squeeze, ch)) {
180				lastch = ch;
181				(void)putwchar(ch);
182			}
183		if (ferror(stdin))
184			err(1, NULL);
185		exit(0);
186	}
187
188	/*
189	 * tr [-Ccs] string1 string2
190	 * Replace all characters (or complemented characters) in string1 with
191	 * the character in the same position in string2.  If the -s option is
192	 * specified, squeeze all the characters in string2.
193	 */
194	if (!isstring2)
195		usage();
196
197	map = cmap_alloc();
198	if (map == NULL)
199		err(1, NULL);
200	squeeze = cset_alloc();
201	if (squeeze == NULL)
202		err(1, NULL);
203
204	s1.str = argv[0];
205
206	if (Cflag || cflag) {
207		cmap_default(map, OOBCH);
208		if ((s2.str = strdup(argv[1])) == NULL)
209			errx(1, "strdup(argv[1])");
210	} else
211		s2.str = argv[1];
212
213	if (!next(&s2))
214		errx(1, "empty string2");
215
216	/*
217	 * For -s result will contain only those characters defined
218	 * as the second characters in each of the toupper or tolower
219	 * pairs.
220	 */
221
222	/* If string2 runs out of characters, use the last one specified. */
223	while (next(&s1)) {
224	again:
225		if (s1.state == CCLASS_LOWER &&
226		    s2.state == CCLASS_UPPER &&
227		    s1.cnt == 1 && s2.cnt == 1) {
228			do {
229				ch = towupper(s1.lastch);
230				cmap_add(map, s1.lastch, ch);
231				if (sflag && iswupper(ch))
232					cset_add(squeeze, ch);
233				if (!next(&s1))
234					goto endloop;
235			} while (s1.state == CCLASS_LOWER && s1.cnt > 1);
236			/* skip upper set */
237			do {
238				if (!next(&s2))
239					break;
240			} while (s2.state == CCLASS_UPPER && s2.cnt > 1);
241			goto again;
242		} else if (s1.state == CCLASS_UPPER &&
243			   s2.state == CCLASS_LOWER &&
244			   s1.cnt == 1 && s2.cnt == 1) {
245			do {
246				ch = towlower(s1.lastch);
247				cmap_add(map, s1.lastch, ch);
248				if (sflag && iswlower(ch))
249					cset_add(squeeze, ch);
250				if (!next(&s1))
251					goto endloop;
252			} while (s1.state == CCLASS_UPPER && s1.cnt > 1);
253			/* skip lower set */
254			do {
255				if (!next(&s2))
256					break;
257			} while (s2.state == CCLASS_LOWER && s2.cnt > 1);
258			goto again;
259		} else {
260			cmap_add(map, s1.lastch, s2.lastch);
261			if (sflag)
262				cset_add(squeeze, s2.lastch);
263		}
264		(void)next(&s2);
265	}
266endloop:
267	if (cflag || (Cflag && MB_CUR_MAX > 1)) {
268		/*
269		 * This is somewhat tricky: since the character set is
270		 * potentially huge, we need to avoid allocating a map
271		 * entry for every character. Our strategy is to set the
272		 * default mapping to the last character of string #2
273		 * (= the one that gets automatically repeated), then to
274		 * add back identity mappings for characters that should
275		 * remain unchanged. We don't waste space on identity mappings
276		 * for non-characters with the -C option; those are simulated
277		 * in the I/O loop.
278		 */
279		s2.str = argv[1];
280		s2.state = NORMAL;
281		for (cnt = 0; cnt < WINT_MAX; cnt++) {
282			if (Cflag && !iswrune(cnt))
283				continue;
284			if (cmap_lookup(map, cnt) == OOBCH) {
285				if (next(&s2)) {
286					cmap_add(map, cnt, s2.lastch);
287					if (sflag)
288						cset_add(squeeze, s2.lastch);
289				}
290			} else
291				cmap_add(map, cnt, cnt);
292			if ((s2.state == EOS || s2.state == INFINITE) &&
293			    cnt >= cmap_max(map))
294				break;
295		}
296		cmap_default(map, s2.lastch);
297	} else if (Cflag) {
298		for (p = carray, cnt = 0; cnt < NCHARS_SB; cnt++) {
299			if (cmap_lookup(map, cnt) == OOBCH && iswrune(cnt))
300				*p++ = cnt;
301			else
302				cmap_add(map, cnt, cnt);
303		}
304		n = p - carray;
305		if (Cflag && n > 1)
306			(void)mergesort(carray, n, sizeof(*carray), charcoll);
307
308		s2.str = argv[1];
309		s2.state = NORMAL;
310		for (cnt = 0; cnt < n; cnt++) {
311			(void)next(&s2);
312			cmap_add(map, carray[cnt], s2.lastch);
313			/*
314			 * Chars taken from s2 can be different this time
315			 * due to lack of complex upper/lower processing,
316			 * so fill string2 again to not miss some.
317			 */
318			if (sflag)
319				cset_add(squeeze, s2.lastch);
320		}
321	}
322
323	cset_cache(squeeze);
324	cmap_cache(map);
325
326	if (sflag)
327		for (lastch = OOBCH; (ch = getwchar()) != WEOF;) {
328			if (!Cflag || iswrune(ch))
329				ch = cmap_lookup(map, ch);
330			if (lastch != ch || !cset_in(squeeze, ch)) {
331				lastch = ch;
332				(void)putwchar(ch);
333			}
334		}
335	else
336		while ((ch = getwchar()) != WEOF) {
337			if (!Cflag || iswrune(ch))
338				ch = cmap_lookup(map, ch);
339			(void)putwchar(ch);
340		}
341	if (ferror(stdin))
342		err(1, NULL);
343	exit (0);
344}
345
346static struct cset *
347setup(char *arg, STR *str, int cflag, int Cflag)
348{
349	struct cset *cs;
350
351	cs = cset_alloc();
352	if (cs == NULL)
353		err(1, NULL);
354	str->str = arg;
355	while (next(str))
356		cset_add(cs, str->lastch);
357	if (Cflag)
358		cset_addclass(cs, wctype("rune"), true);
359	if (cflag || Cflag)
360		cset_invert(cs);
361	cset_cache(cs);
362	return (cs);
363}
364
365int
366charcoll(const void *a, const void *b)
367{
368	static char sa[2], sb[2];
369
370	sa[0] = *(const int *)a;
371	sb[0] = *(const int *)b;
372	return (strcoll(sa, sb));
373}
374
375static void
376usage(void)
377{
378	(void)fprintf(stderr, "%s\n%s\n%s\n%s\n",
379		"usage: tr [-Ccsu] string1 string2",
380		"       tr [-Ccu] -d string1",
381		"       tr [-Ccu] -s string1",
382		"       tr [-Ccu] -ds string1 string2");
383	exit(1);
384}
385