cut.c revision 102084 
1/*
2 * Copyright (c) 1989, 1993
3 *	The Regents of the University of California.  All rights reserved.
4 *
5 * This code is derived from software contributed to Berkeley by
6 * Adam S. Moskowitz of Menlo Consulting and Marciano Pitargue.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 * 1. Redistributions of source code must retain the above copyright
12 *    notice, this list of conditions and the following disclaimer.
13 * 2. Redistributions in binary form must reproduce the above copyright
14 *    notice, this list of conditions and the following disclaimer in the
15 *    documentation and/or other materials provided with the distribution.
16 * 3. All advertising materials mentioning features or use of this software
17 *    must display the following acknowledgement:
18 *	This product includes software developed by the University of
19 *	California, Berkeley and its contributors.
20 * 4. Neither the name of the University nor the names of its contributors
21 *    may be used to endorse or promote products derived from this software
22 *    without specific prior written permission.
23 *
24 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
25 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
26 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
27 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
28 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
29 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
30 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
31 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
32 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
33 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
34 * SUCH DAMAGE.
35 */
36
37#ifndef lint
38static const char copyright[] =
39"@(#) Copyright (c) 1989, 1993\n\
40	The Regents of the University of California.  All rights reserved.\n";
41static const char sccsid[] = "@(#)cut.c	8.3 (Berkeley) 5/4/95";
42#endif /* not lint */
43#include <sys/cdefs.h>
44__FBSDID("$FreeBSD: head/usr.bin/cut/cut.c 102084 2002-08-19 03:07:56Z jmallett $");
45
46#include <ctype.h>
47#include <err.h>
48#include <limits.h>
49#include <locale.h>
50#include <stdio.h>
51#include <stdlib.h>
52#include <string.h>
53#include <unistd.h>
54
55int	bflag;
56int	cflag;
57char	dchar;
58int	dflag;
59int	fflag;
60int	nflag;
61int	sflag;
62
63void	b_n_cut(FILE *, const char *);
64void	c_cut(FILE *, const char *);
65void	f_cut(FILE *, const char *);
66void	get_list(char *);
67void	needpos(size_t);
68static 	void usage(void);
69
70int
71main(argc, argv)
72	int argc;
73	char *argv[];
74{
75	FILE *fp;
76	void (*fcn)(FILE *, const char *);
77	int ch, rval;
78
79	setlocale(LC_ALL, "");
80
81	fcn = NULL;
82	dchar = '\t';			/* default delimiter is \t */
83
84	/*
85	 * Since we don't support multi-byte characters, the -c and -b
86	 * options are equivalent.
87	 */
88	while ((ch = getopt(argc, argv, "b:c:d:f:sn")) != -1)
89		switch(ch) {
90		case 'b':
91			fcn = c_cut;
92			get_list(optarg);
93			bflag = 1;
94			break;
95		case 'c':
96			fcn = c_cut;
97			get_list(optarg);
98			cflag = 1;
99			break;
100		case 'd':
101			dchar = *optarg;
102			dflag = 1;
103			break;
104		case 'f':
105			get_list(optarg);
106			fcn = f_cut;
107			fflag = 1;
108			break;
109		case 's':
110			sflag = 1;
111			break;
112		case 'n':
113			nflag = 1;
114			break;
115		case '?':
116		default:
117			usage();
118		}
119	argc -= optind;
120	argv += optind;
121
122	if (fflag) {
123		if (bflag || cflag || nflag)
124			usage();
125	} else if (!(bflag || cflag) || dflag || sflag)
126		usage();
127	else if (!bflag && nflag)
128		usage();
129
130	if (nflag)
131		fcn = b_n_cut;
132
133	rval = 0;
134	if (*argv)
135		for (; *argv; ++argv) {
136			if (strcmp(*argv, "-") == 0)
137				fcn(stdin, "stdin");
138			else {
139				if (!(fp = fopen(*argv, "r"))) {
140					warn("%s", *argv);
141					rval = 1;
142					continue;
143				}
144				fcn(fp, *argv);
145				(void)fclose(fp);
146			}
147		}
148	else
149		fcn(stdin, "stdin");
150	exit(rval);
151}
152
153size_t autostart, autostop, maxval;
154
155char *positions;
156
157void
158get_list(list)
159	char *list;
160{
161	size_t setautostart, start, stop;
162	char *pos;
163	char *p;
164
165	/*
166	 * set a byte in the positions array to indicate if a field or
167	 * column is to be selected; use +1, it's 1-based, not 0-based.
168	 * This parser is less restrictive than the Draft 9 POSIX spec.
169	 * POSIX doesn't allow lists that aren't in increasing order or
170	 * overlapping lists.  We also handle "-3-5" although there's no
171	 * real reason too.
172	 */
173	for (; (p = strsep(&list, ", \t")) != NULL;) {
174		setautostart = start = stop = 0;
175		if (*p == '-') {
176			++p;
177			setautostart = 1;
178		}
179		if (isdigit((unsigned char)*p)) {
180			start = stop = strtol(p, &p, 10);
181			if (setautostart && start > autostart)
182				autostart = start;
183		}
184		if (*p == '-') {
185			if (isdigit((unsigned char)p[1]))
186				stop = strtol(p + 1, &p, 10);
187			if (*p == '-') {
188				++p;
189				if (!autostop || autostop > stop)
190					autostop = stop;
191			}
192		}
193		if (*p)
194			errx(1, "[-cf] list: illegal list value");
195		if (!stop || !start)
196			errx(1, "[-cf] list: values may not include zero");
197		if (maxval < stop) {
198			maxval = stop;
199			needpos(maxval + 1);
200		}
201		for (pos = positions + start; start++ <= stop; *pos++ = 1);
202	}
203
204	/* overlapping ranges */
205	if (autostop && maxval > autostop) {
206		maxval = autostop;
207		needpos(maxval + 1);
208	}
209
210	/* set autostart */
211	if (autostart)
212		memset(positions + 1, '1', autostart);
213}
214
215void
216needpos(size_t n)
217{
218	static size_t npos;
219	size_t oldnpos;
220
221	/* Grow the positions array to at least the specified size. */
222	if (n > npos) {
223		oldnpos = npos;
224		if (npos == 0)
225			npos = n;
226		while (n > npos)
227			npos *= 2;
228		if ((positions = realloc(positions, npos)) == NULL)
229			err(1, "realloc");
230		memset((char *)positions + oldnpos, 0, npos - oldnpos);
231	}
232}
233
234/*
235 * Cut based on byte positions, taking care not to split multibyte characters.
236 * Although this function also handles the case where -n is not specified,
237 * c_cut() ought to be much faster.
238 */
239void
240b_n_cut(fp, fname)
241	FILE *fp;
242	const char *fname;
243{
244	size_t col, i, lbuflen;
245	char *lbuf;
246	int canwrite, clen, warned;
247
248	warned = 0;
249	while ((lbuf = fgetln(fp, &lbuflen)) != NULL) {
250		for (col = 0; lbuflen > 0; col += clen) {
251			if ((clen = mblen(lbuf, lbuflen)) < 0) {
252				if (!warned) {
253					warn("%s", fname);
254					warned = 1;
255				}
256				clen = 1;
257			}
258			if (clen == 0 || *lbuf == '\n')
259				break;
260			if (col < maxval && !positions[1 + col]) {
261				/*
262				 * Print the character if (1) after an initial
263				 * segment of un-selected bytes, the rest of
264				 * it is selected, and (2) the last byte is
265				 * selected.
266				 */
267				i = col;
268				while (i < col + clen && i < maxval &&
269				    !positions[1 + i])
270					i++;
271				canwrite = i < col + clen;
272				for (; i < col + clen && i < maxval; i++)
273					canwrite &= positions[1 + i];
274				if (canwrite)
275					fwrite(lbuf, 1, clen, stdout);
276			} else {
277				/*
278				 * Print the character if all of it has
279				 * been selected.
280				 */
281				canwrite = 1;
282				for (i = col; i < col + clen; i++)
283					if ((i >= maxval && !autostop) ||
284					    (i < maxval && !positions[1 + i])) {
285						canwrite = 0;
286						break;
287					}
288				if (canwrite)
289					fwrite(lbuf, 1, clen, stdout);
290			}
291			lbuf += clen;
292			lbuflen -= clen;
293		}
294		if (lbuflen > 0)
295			putchar('\n');
296	}
297}
298
299void
300c_cut(fp, fname)
301	FILE *fp;
302	const char *fname __unused;
303{
304	int ch, col;
305	char *pos;
306
307	ch = 0;
308	for (;;) {
309		pos = positions + 1;
310		for (col = maxval; col; --col) {
311			if ((ch = getc(fp)) == EOF)
312				return;
313			if (ch == '\n')
314				break;
315			if (*pos++)
316				(void)putchar(ch);
317		}
318		if (ch != '\n') {
319			if (autostop)
320				while ((ch = getc(fp)) != EOF && ch != '\n')
321					(void)putchar(ch);
322			else
323				while ((ch = getc(fp)) != EOF && ch != '\n');
324		}
325		(void)putchar('\n');
326	}
327}
328
329void
330f_cut(fp, fname)
331	FILE *fp;
332	const char *fname __unused;
333{
334	int ch, field, isdelim;
335	char *pos, *p, sep;
336	int output;
337	char *lbuf, *mlbuf;
338	size_t lbuflen;
339
340	mlbuf = NULL;
341	for (sep = dchar; (lbuf = fgetln(fp, &lbuflen)) != NULL;) {
342		/* Assert EOL has a newline. */
343		if (*(lbuf + lbuflen - 1) != '\n') {
344			/* Can't have > 1 line with no trailing newline. */
345			mlbuf = malloc(lbuflen + 1);
346			if (mlbuf == NULL)
347				err(1, "malloc");
348			memcpy(mlbuf, lbuf, lbuflen);
349			*(mlbuf + lbuflen) = '\n';
350			lbuf = mlbuf;
351		}
352		output = 0;
353		for (isdelim = 0, p = lbuf;; ++p) {
354			ch = *p;
355			/* this should work if newline is delimiter */
356			if (ch == sep)
357				isdelim = 1;
358			if (ch == '\n') {
359				if (!isdelim && !sflag)
360					(void)fwrite(lbuf, lbuflen, 1, stdout);
361				break;
362			}
363		}
364		if (!isdelim)
365			continue;
366
367		pos = positions + 1;
368		for (field = maxval, p = lbuf; field; --field, ++pos) {
369			if (*pos) {
370				if (output++)
371					(void)putchar(sep);
372				while ((ch = *p++) != '\n' && ch != sep)
373					(void)putchar(ch);
374			} else {
375				while ((ch = *p++) != '\n' && ch != sep)
376					continue;
377			}
378			if (ch == '\n')
379				break;
380		}
381		if (ch != '\n') {
382			if (autostop) {
383				if (output)
384					(void)putchar(sep);
385				for (; (ch = *p) != '\n'; ++p)
386					(void)putchar(ch);
387			} else
388				for (; (ch = *p) != '\n'; ++p);
389		}
390		(void)putchar('\n');
391	}
392	if (mlbuf != NULL)
393		free(mlbuf);
394}
395
396static void
397usage()
398{
399	(void)fprintf(stderr, "%s\n%s\n%s\n",
400		"usage: cut -b list [-n] [file ...]",
401		"       cut -c list [file ...]",
402		"       cut -f list [-s] [-d delim] [file ...]");
403	exit(1);
404}
405