cut.c revision 7200 
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 char copyright[] =
39"@(#) Copyright (c) 1989, 1993\n\
40	The Regents of the University of California.  All rights reserved.\n";
41#endif /* not lint */
42
43#ifndef lint
44static char sccsid[] = "@(#)cut.c	8.1 (Berkeley) 6/6/93";
45#endif /* not lint */
46
47#include <ctype.h>
48#include <errno.h>
49#include <limits.h>
50#include <stdio.h>
51#include <stdlib.h>
52#include <string.h>
53
54int	cflag;
55char	dchar;
56int	dflag;
57int	fflag;
58int	sflag;
59
60void	c_cut __P((FILE *, char *));
61void	err __P((const char *, ...));
62void	f_cut __P((FILE *, char *));
63void	get_list __P((char *));
64void	usage __P((void));
65
66int
67main(argc, argv)
68	int argc;
69	char *argv[];
70{
71	FILE *fp;
72	void (*fcn) __P((FILE *, char *));
73	int ch;
74
75	dchar = '\t';			/* default delimiter is \t */
76
77	while ((ch = getopt(argc, argv, "c:d:f:s")) != EOF)
78		switch(ch) {
79		case 'c':
80			fcn = c_cut;
81			get_list(optarg);
82			cflag = 1;
83			break;
84		case 'd':
85			dchar = *optarg;
86			dflag = 1;
87			break;
88		case 'f':
89			get_list(optarg);
90			fcn = f_cut;
91			fflag = 1;
92			break;
93		case 's':
94			sflag = 1;
95			break;
96		case '?':
97		default:
98			usage();
99		}
100	argc -= optind;
101	argv += optind;
102
103	if (fflag) {
104		if (cflag)
105			usage();
106	} else if (!cflag || dflag || sflag)
107		usage();
108
109	if (*argv)
110		for (; *argv; ++argv) {
111			if (!(fp = fopen(*argv, "r")))
112				err("%s: %s\n", *argv, strerror(errno));
113			fcn(fp, *argv);
114			(void)fclose(fp);
115		}
116	else
117		fcn(stdin, "stdin");
118	exit(0);
119}
120
121int autostart, autostop, maxval;
122
123char positions[_POSIX2_LINE_MAX + 1];
124
125void
126get_list(list)
127	char *list;
128{
129	register int setautostart, start, stop;
130	register char *pos;
131	char *p;
132
133	/*
134	 * set a byte in the positions array to indicate if a field or
135	 * column is to be selected; use +1, it's 1-based, not 0-based.
136	 * This parser is less restrictive than the Draft 9 POSIX spec.
137	 * POSIX doesn't allow lists that aren't in increasing order or
138	 * overlapping lists.  We also handle "-3-5" although there's no
139	 * real reason too.
140	 */
141	for (; p = strtok(list, ", \t"); list = NULL) {
142		setautostart = start = stop = 0;
143		if (*p == '-') {
144			++p;
145			setautostart = 1;
146		}
147		if (isdigit(*p)) {
148			start = stop = strtol(p, &p, 10);
149			if (setautostart && start > autostart)
150				autostart = start;
151		}
152		if (*p == '-') {
153			if (isdigit(p[1]))
154				stop = strtol(p + 1, &p, 10);
155			if (*p == '-') {
156				++p;
157				if (!autostop || autostop > stop)
158					autostop = stop;
159			}
160		}
161		if (*p)
162			err("[-cf] list: illegal list value\n");
163		if (!stop || !start)
164			err("[-cf] list: values may not include zero\n");
165		if (stop > _POSIX2_LINE_MAX)
166			err("[-cf] list: %d too large (max %d)\n",
167			    stop, _POSIX2_LINE_MAX);
168		if (maxval < stop)
169			maxval = stop;
170		for (pos = positions + start; start++ <= stop; *pos++ = 1);
171	}
172
173	/* overlapping ranges */
174	if (autostop && maxval > autostop)
175		maxval = autostop;
176
177	/* set autostart */
178	if (autostart)
179		memset(positions + 1, '1', autostart);
180}
181
182/* ARGSUSED */
183void
184c_cut(fp, fname)
185	FILE *fp;
186	char *fname;
187{
188	register int ch, col;
189	register char *pos;
190
191	for (;;) {
192		pos = positions + 1;
193		for (col = maxval; col; --col) {
194			if ((ch = getc(fp)) == EOF)
195				return;
196			if (ch == '\n')
197				break;
198			if (*pos++)
199				(void)putchar(ch);
200		}
201		if (ch != '\n')
202			if (autostop)
203				while ((ch = getc(fp)) != EOF && ch != '\n')
204					(void)putchar(ch);
205			else
206				while ((ch = getc(fp)) != EOF && ch != '\n');
207		(void)putchar('\n');
208	}
209}
210
211void
212f_cut(fp, fname)
213	FILE *fp;
214	char *fname;
215{
216	register int ch, field, isdelim;
217	register char *pos, *p, sep;
218	int output;
219	char lbuf[_POSIX2_LINE_MAX + 1];
220
221	for (sep = dchar; fgets(lbuf, sizeof(lbuf), fp);) {
222		output = 0;
223		for (isdelim = 0, p = lbuf;; ++p) {
224			if (!(ch = *p))
225				err("%s: line too long.\n", fname);
226			/* this should work if newline is delimiter */
227			if (ch == sep)
228				isdelim = 1;
229			if (ch == '\n') {
230				if (!isdelim && !sflag)
231					(void)printf("%s", lbuf);
232				break;
233			}
234		}
235		if (!isdelim)
236			continue;
237
238		pos = positions + 1;
239		for (field = maxval, p = lbuf; field; --field, ++pos) {
240			if (*pos) {
241				if (output++)
242					(void)putchar(sep);
243				while ((ch = *p++) != '\n' && ch != sep)
244					(void)putchar(ch);
245			} else
246				while ((ch = *p++) != '\n' && ch != sep);
247			if (ch == '\n')
248				break;
249		}
250		if (ch != '\n')
251			if (autostop) {
252				if (output)
253					(void)putchar(sep);
254				for (; (ch = *p) != '\n'; ++p)
255					(void)putchar(ch);
256			} else
257				for (; (ch = *p) != '\n'; ++p);
258		(void)putchar('\n');
259	}
260}
261
262void
263usage()
264{
265	(void)fprintf(stderr,
266"usage:\tcut -c list [file1 ...]\n\tcut -f list [-s] [-d delim] [file ...]\n");
267	exit(1);
268}
269
270#if __STDC__
271#include <stdarg.h>
272#else
273#include <varargs.h>
274#endif
275
276void
277#if __STDC__
278err(const char *fmt, ...)
279#else
280err(fmt, va_alist)
281	char *fmt;
282        va_dcl
283#endif
284{
285	va_list ap;
286#if __STDC__
287	va_start(ap, fmt);
288#else
289	va_start(ap);
290#endif
291	(void)fprintf(stderr, "cut: ");
292	(void)vfprintf(stderr, fmt, ap);
293	va_end(ap);
294	(void)fprintf(stderr, "\n");
295	exit(1);
296	/* NOTREACHED */
297}
298