1/*
2 * SPDX-License-Identifier: BSD-3-Clause
3 *
4 * Copyright (c) 1989, 1993
5 *	The Regents of the University of California.  All rights reserved.
6 *
7 * This code is derived from software contributed to Berkeley by
8 * Adam S. Moskowitz of Menlo Consulting and Marciano Pitargue.
9 *
10 * Redistribution and use in source and binary forms, with or without
11 * modification, are permitted provided that the following conditions
12 * are met:
13 * 1. Redistributions of source code must retain the above copyright
14 *    notice, this list of conditions and the following disclaimer.
15 * 2. Redistributions in binary form must reproduce the above copyright
16 *    notice, this list of conditions and the following disclaimer in the
17 *    documentation and/or other materials provided with the distribution.
18 * 3. Neither the name of the University nor the names of its contributors
19 *    may be used to endorse or promote products derived from this software
20 *    without specific prior written permission.
21 *
22 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
23 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
24 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
25 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
26 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
27 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
28 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
29 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
30 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
31 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
32 * SUCH DAMAGE.
33 */
34
35#ifndef lint
36static const char copyright[] =
37"@(#) Copyright (c) 1989, 1993\n\
38	The Regents of the University of California.  All rights reserved.\n";
39static const char sccsid[] = "@(#)cut.c	8.3 (Berkeley) 5/4/95";
40#endif /* not lint */
41#include <sys/cdefs.h>
42__FBSDID("$FreeBSD$");
43
44#include <ctype.h>
45#include <err.h>
46#include <errno.h>
47#include <limits.h>
48#include <locale.h>
49#include <stdio.h>
50#include <stdlib.h>
51#include <string.h>
52#include <unistd.h>
53#include <wchar.h>
54
55static int	bflag;
56static int	cflag;
57static wchar_t	dchar;
58static char	dcharmb[MB_LEN_MAX + 1];
59static int	dflag;
60static int	fflag;
61static int	nflag;
62static int	sflag;
63static int	wflag;
64
65static size_t	autostart, autostop, maxval;
66static char *	positions;
67
68static int	b_cut(FILE *, const char *);
69static int	b_n_cut(FILE *, const char *);
70static int	c_cut(FILE *, const char *);
71static int	f_cut(FILE *, const char *);
72static void	get_list(char *);
73static int	is_delim(wchar_t);
74static void	needpos(size_t);
75static void	usage(void);
76
77int
78main(int argc, char *argv[])
79{
80	FILE *fp;
81	int (*fcn)(FILE *, const char *);
82	int ch, rval;
83	size_t n;
84
85	setlocale(LC_ALL, "");
86
87	fcn = NULL;
88	dchar = '\t';			/* default delimiter is \t */
89	strcpy(dcharmb, "\t");
90
91	while ((ch = getopt(argc, argv, "b:c:d:f:snw")) != -1)
92		switch(ch) {
93		case 'b':
94			get_list(optarg);
95			bflag = 1;
96			break;
97		case 'c':
98			get_list(optarg);
99			cflag = 1;
100			break;
101		case 'd':
102			n = mbrtowc(&dchar, optarg, MB_LEN_MAX, NULL);
103			if (dchar == '\0' || n != strlen(optarg))
104				errx(1, "bad delimiter");
105			strcpy(dcharmb, optarg);
106			dflag = 1;
107			break;
108		case 'f':
109			get_list(optarg);
110			fflag = 1;
111			break;
112		case 's':
113			sflag = 1;
114			break;
115		case 'n':
116			nflag = 1;
117			break;
118		case 'w':
119			wflag = 1;
120			break;
121		case '?':
122		default:
123			usage();
124		}
125	argc -= optind;
126	argv += optind;
127
128	if (fflag) {
129		if (bflag || cflag || nflag || (wflag && dflag))
130			usage();
131	} else if (!(bflag || cflag) || dflag || sflag || wflag)
132		usage();
133	else if (!bflag && nflag)
134		usage();
135
136	if (fflag)
137		fcn = f_cut;
138	else if (cflag)
139		fcn = MB_CUR_MAX > 1 ? c_cut : b_cut;
140	else if (bflag)
141		fcn = nflag && MB_CUR_MAX > 1 ? b_n_cut : b_cut;
142
143	rval = 0;
144	if (*argv)
145		for (; *argv; ++argv) {
146			if (strcmp(*argv, "-") == 0)
147				rval |= fcn(stdin, "stdin");
148			else {
149				if (!(fp = fopen(*argv, "r"))) {
150					warn("%s", *argv);
151					rval = 1;
152					continue;
153				}
154				fcn(fp, *argv);
155				(void)fclose(fp);
156			}
157		}
158	else
159		rval = fcn(stdin, "stdin");
160	exit(rval);
161}
162
163static void
164get_list(char *list)
165{
166	size_t setautostart, start, stop;
167	char *pos;
168	char *p;
169
170	/*
171	 * set a byte in the positions array to indicate if a field or
172	 * column is to be selected; use +1, it's 1-based, not 0-based.
173	 * Numbers and number ranges may be overlapping, repeated, and in
174	 * any order. We handle "-3-5" although there's no real reason to.
175	 */
176	for (; (p = strsep(&list, ", \t")) != NULL;) {
177		setautostart = start = stop = 0;
178		if (*p == '-') {
179			++p;
180			setautostart = 1;
181		}
182		if (isdigit((unsigned char)*p)) {
183			start = stop = strtol(p, &p, 10);
184			if (setautostart && start > autostart)
185				autostart = start;
186		}
187		if (*p == '-') {
188			if (isdigit((unsigned char)p[1]))
189				stop = strtol(p + 1, &p, 10);
190			if (*p == '-') {
191				++p;
192				if (!autostop || autostop > stop)
193					autostop = stop;
194			}
195		}
196		if (*p)
197			errx(1, "[-bcf] list: illegal list value");
198		if (!stop || !start)
199			errx(1, "[-bcf] list: values may not include zero");
200		if (maxval < stop) {
201			maxval = stop;
202			needpos(maxval + 1);
203		}
204		for (pos = positions + start; start++ <= stop; *pos++ = 1);
205	}
206
207	/* overlapping ranges */
208	if (autostop && maxval > autostop) {
209		maxval = autostop;
210		needpos(maxval + 1);
211	}
212
213	/* reversed range with autostart */
214	if (maxval < autostart) {
215		maxval = autostart;
216		needpos(maxval + 1);
217	}
218
219	/* set autostart */
220	if (autostart)
221		memset(positions + 1, '1', autostart);
222}
223
224static void
225needpos(size_t n)
226{
227	static size_t npos;
228	size_t oldnpos;
229
230	/* Grow the positions array to at least the specified size. */
231	if (n > npos) {
232		oldnpos = npos;
233		if (npos == 0)
234			npos = n;
235		while (n > npos)
236			npos *= 2;
237		if ((positions = realloc(positions, npos)) == NULL)
238			err(1, "realloc");
239		memset((char *)positions + oldnpos, 0, npos - oldnpos);
240	}
241}
242
243static int
244b_cut(FILE *fp, const char *fname __unused)
245{
246	int ch, col;
247	char *pos;
248
249	ch = 0;
250	for (;;) {
251		pos = positions + 1;
252		for (col = maxval; col; --col) {
253			if ((ch = getc(fp)) == EOF)
254				return (0);
255			if (ch == '\n')
256				break;
257			if (*pos++)
258				(void)putchar(ch);
259		}
260		if (ch != '\n') {
261			if (autostop)
262				while ((ch = getc(fp)) != EOF && ch != '\n')
263					(void)putchar(ch);
264			else
265				while ((ch = getc(fp)) != EOF && ch != '\n');
266		}
267		(void)putchar('\n');
268	}
269	return (0);
270}
271
272/*
273 * Cut based on byte positions, taking care not to split multibyte characters.
274 * Although this function also handles the case where -n is not specified,
275 * b_cut() ought to be much faster.
276 */
277static int
278b_n_cut(FILE *fp, const char *fname)
279{
280	size_t col, i, lbuflen;
281	char *lbuf;
282	int canwrite, clen, warned;
283	mbstate_t mbs;
284
285	memset(&mbs, 0, sizeof(mbs));
286	warned = 0;
287	while ((lbuf = fgetln(fp, &lbuflen)) != NULL) {
288		for (col = 0; lbuflen > 0; col += clen) {
289			if ((clen = mbrlen(lbuf, lbuflen, &mbs)) < 0) {
290				if (!warned) {
291					warn("%s", fname);
292					warned = 1;
293				}
294				memset(&mbs, 0, sizeof(mbs));
295				clen = 1;
296			}
297			if (clen == 0 || *lbuf == '\n')
298				break;
299			if (col < maxval && !positions[1 + col]) {
300				/*
301				 * Print the character if (1) after an initial
302				 * segment of un-selected bytes, the rest of
303				 * it is selected, and (2) the last byte is
304				 * selected.
305				 */
306				i = col;
307				while (i < col + clen && i < maxval &&
308				    !positions[1 + i])
309					i++;
310				canwrite = i < col + clen;
311				for (; i < col + clen && i < maxval; i++)
312					canwrite &= positions[1 + i];
313				if (canwrite)
314					fwrite(lbuf, 1, clen, stdout);
315			} else {
316				/*
317				 * Print the character if all of it has
318				 * been selected.
319				 */
320				canwrite = 1;
321				for (i = col; i < col + clen; i++)
322					if ((i >= maxval && !autostop) ||
323					    (i < maxval && !positions[1 + i])) {
324						canwrite = 0;
325						break;
326					}
327				if (canwrite)
328					fwrite(lbuf, 1, clen, stdout);
329			}
330			lbuf += clen;
331			lbuflen -= clen;
332		}
333		if (lbuflen > 0)
334			putchar('\n');
335	}
336	return (warned);
337}
338
339static int
340c_cut(FILE *fp, const char *fname)
341{
342	wint_t ch;
343	int col;
344	char *pos;
345
346	ch = 0;
347	for (;;) {
348		pos = positions + 1;
349		for (col = maxval; col; --col) {
350			if ((ch = getwc(fp)) == WEOF)
351				goto out;
352			if (ch == '\n')
353				break;
354			if (*pos++)
355				(void)putwchar(ch);
356		}
357		if (ch != '\n') {
358			if (autostop)
359				while ((ch = getwc(fp)) != WEOF && ch != '\n')
360					(void)putwchar(ch);
361			else
362				while ((ch = getwc(fp)) != WEOF && ch != '\n');
363		}
364		(void)putwchar('\n');
365	}
366out:
367	if (ferror(fp)) {
368		warn("%s", fname);
369		return (1);
370	}
371	return (0);
372}
373
374static int
375is_delim(wchar_t ch)
376{
377	if (wflag) {
378		if (ch == ' ' || ch == '\t')
379			return 1;
380	} else {
381		if (ch == dchar)
382			return 1;
383	}
384	return 0;
385}
386
387static int
388f_cut(FILE *fp, const char *fname)
389{
390	wchar_t ch;
391	int field, i, isdelim;
392	char *pos, *p;
393	int output;
394	char *lbuf, *mlbuf;
395	size_t clen, lbuflen, reallen;
396
397	mlbuf = NULL;
398	while ((lbuf = fgetln(fp, &lbuflen)) != NULL) {
399		reallen = lbuflen;
400		/* Assert EOL has a newline. */
401		if (*(lbuf + lbuflen - 1) != '\n') {
402			/* Can't have > 1 line with no trailing newline. */
403			mlbuf = malloc(lbuflen + 1);
404			if (mlbuf == NULL)
405				err(1, "malloc");
406			memcpy(mlbuf, lbuf, lbuflen);
407			*(mlbuf + lbuflen) = '\n';
408			lbuf = mlbuf;
409			reallen++;
410		}
411		output = 0;
412		for (isdelim = 0, p = lbuf;; p += clen) {
413			clen = mbrtowc(&ch, p, lbuf + reallen - p, NULL);
414			if (clen == (size_t)-1 || clen == (size_t)-2) {
415				warnc(EILSEQ, "%s", fname);
416				free(mlbuf);
417				return (1);
418			}
419			if (clen == 0)
420				clen = 1;
421			/* this should work if newline is delimiter */
422			if (is_delim(ch))
423				isdelim = 1;
424			if (ch == '\n') {
425				if (!isdelim && !sflag)
426					(void)fwrite(lbuf, lbuflen, 1, stdout);
427				break;
428			}
429		}
430		if (!isdelim)
431			continue;
432
433		pos = positions + 1;
434		for (field = maxval, p = lbuf; field; --field, ++pos) {
435			if (*pos && output++)
436				for (i = 0; dcharmb[i] != '\0'; i++)
437					putchar(dcharmb[i]);
438			for (;;) {
439				clen = mbrtowc(&ch, p, lbuf + reallen - p,
440				    NULL);
441				if (clen == (size_t)-1 || clen == (size_t)-2) {
442					warnc(EILSEQ, "%s", fname);
443					free(mlbuf);
444					return (1);
445				}
446				if (clen == 0)
447					clen = 1;
448				p += clen;
449				if (ch == '\n' || is_delim(ch)) {
450					/* compress whitespace */
451					if (wflag && ch != '\n')
452						while (is_delim(*p))
453							p++;
454					break;
455				}
456				if (*pos)
457					for (i = 0; i < (int)clen; i++)
458						putchar(p[i - clen]);
459			}
460			if (ch == '\n')
461				break;
462		}
463		if (ch != '\n') {
464			if (autostop) {
465				if (output)
466					for (i = 0; dcharmb[i] != '\0'; i++)
467						putchar(dcharmb[i]);
468				for (; (ch = *p) != '\n'; ++p)
469					(void)putchar(ch);
470			} else
471				for (; (ch = *p) != '\n'; ++p);
472		}
473		(void)putchar('\n');
474	}
475	free(mlbuf);
476	return (0);
477}
478
479static void
480usage(void)
481{
482	(void)fprintf(stderr, "%s\n%s\n%s\n",
483		"usage: cut -b list [-n] [file ...]",
484		"       cut -c list [file ...]",
485		"       cut -f list [-s] [-w | -d delim] [file ...]");
486	exit(1);
487}
488