1 /*	$OpenBSD: cut.c,v 1.11 2003/06/10 22:20:45 deraadt Exp $	*/
2 /*	$NetBSD: cut.c,v 1.9 1995/09/02 05:59:23 jtc Exp $	*/
3 
4 /*
5  * Copyright (c) 1989, 1993
6  *	The Regents of the University of California.  All rights reserved.
7  *
8  * This code is derived from software contributed to Berkeley by
9  * Adam S. Moskowitz of Menlo Consulting and Marciano Pitargue.
10  *
11  * Redistribution and use in source and binary forms, with or without
12  * modification, are permitted provided that the following conditions
13  * are met:
14  * 1. Redistributions of source code must retain the above copyright
15  *    notice, this list of conditions and the following disclaimer.
16  * 2. Redistributions in binary form must reproduce the above copyright
17  *    notice, this list of conditions and the following disclaimer in the
18  *    documentation and/or other materials provided with the distribution.
19  * 3. Neither the name of the University nor the names of its contributors
20  *    may be used to endorse or promote products derived from this software
21  *    without specific prior written permission.
22  *
23  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
24  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
25  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
26  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
27  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
28  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
29  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
30  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
31  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
32  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
33  * SUCH DAMAGE.
34  */
35 
36 #include <sys/cdefs.h>
37 __COPYRIGHT("@(#) Copyright (c) 1989, 1993\n\
38 	The Regents of the University of California.  All rights reserved.\n");
39 __SCCSID("@(#)cut.c	8.3 (Berkeley) 5/4/95");
40 __RCSID("$MirOS: src/usr.bin/cut/cut.c,v 1.2 2007/07/05 23:09:39 tg Exp $");
41 
42 #include <ctype.h>
43 #include <err.h>
44 #include <errno.h>
45 #include <limits.h>
46 #include <locale.h>
47 #include <stdio.h>
48 #include <stdlib.h>
49 #include <string.h>
50 #include <unistd.h>
51 
52 int	cflag;
53 char	dchar;
54 int	dflag;
55 int	fflag;
56 int	sflag;
57 
58 void	c_cut(FILE *, char *);
59 void	f_cut(FILE *, char *);
60 void	get_list(char *);
61 void	usage(void);
62 
63 int
main(int argc,char * argv[])64 main(int argc, char *argv[])
65 {
66 	FILE *fp;
67 	void (*fcn)(FILE *, char *);
68 	int ch;
69 
70 #ifndef __MirBSD__
71 	setlocale (LC_ALL, "");
72 #endif
73 
74 	dchar = '\t';			/* default delimiter is \t */
75 
76 	/* Since we don't support multi-byte characters, the -c and -b
77 	   options are equivalent, and the -n option is meaningless. */
78 	while ((ch = getopt(argc, argv, "b:c:d:f:sn")) != -1)
79 		switch(ch) {
80 		case 'b':
81 		case 'c':
82 			fcn = c_cut;
83 			get_list(optarg);
84 			cflag = 1;
85 			break;
86 		case 'd':
87 			dchar = *optarg;
88 			dflag = 1;
89 			break;
90 		case 'f':
91 			get_list(optarg);
92 			fcn = f_cut;
93 			fflag = 1;
94 			break;
95 		case 's':
96 			sflag = 1;
97 			break;
98 		case 'n':
99 			break;
100 		case '?':
101 		default:
102 			usage();
103 		}
104 	argc -= optind;
105 	argv += optind;
106 
107 	if (fflag) {
108 		if (cflag)
109 			usage();
110 	} else if (!cflag || dflag || sflag)
111 		usage();
112 
113 	if (*argv)
114 		for (; *argv; ++argv) {
115 			if (!(fp = fopen(*argv, "r")))
116 				err(1, "%s", *argv);
117 			fcn(fp, *argv);
118 			(void)fclose(fp);
119 		}
120 	else
121 		fcn(stdin, "stdin");
122 	exit(0);
123 }
124 
125 int autostart, autostop, maxval;
126 
127 char positions[_POSIX2_LINE_MAX + 1];
128 
129 void
get_list(char * list)130 get_list(char *list)
131 {
132 	int setautostart, start, stop;
133 	char *pos;
134 	char *p;
135 
136 	/*
137 	 * set a byte in the positions array to indicate if a field or
138 	 * column is to be selected; use +1, it's 1-based, not 0-based.
139 	 * This parser is less restrictive than the Draft 9 POSIX spec.
140 	 * POSIX doesn't allow lists that aren't in increasing order or
141 	 * overlapping lists.  We also handle "-3-5" although there's no
142 	 * real reason too.
143 	 */
144 	while ((p = strsep(&list, ", \t"))) {
145 		setautostart = start = stop = 0;
146 		if (*p == '-') {
147 			++p;
148 			setautostart = 1;
149 		}
150 		if (isdigit(*p)) {
151 			start = stop = strtol(p, &p, 10);
152 			if (setautostart && start > autostart)
153 				autostart = start;
154 		}
155 		if (*p == '-') {
156 			if (isdigit(p[1]))
157 				stop = strtol(p + 1, &p, 10);
158 			if (*p == '-') {
159 				++p;
160 				if (!autostop || autostop > stop)
161 					autostop = stop;
162 			}
163 		}
164 		if (*p)
165 			errx(1, "[-cf] list: illegal list value");
166 		if (!stop || !start)
167 			errx(1, "[-cf] list: values may not include zero");
168 		if (stop > _POSIX2_LINE_MAX)
169 			errx(1, "[-cf] list: %d too large (max %d)",
170 			    stop, _POSIX2_LINE_MAX);
171 		if (maxval < stop)
172 			maxval = stop;
173 		for (pos = positions + start; start++ <= stop; *pos++ = 1)
174 			;
175 	}
176 
177 	/* overlapping ranges */
178 	if (autostop && maxval > autostop)
179 		maxval = autostop;
180 
181 	/* set autostart */
182 	if (autostart)
183 		memset(positions + 1, '1', autostart);
184 }
185 
186 /* ARGSUSED */
187 void
c_cut(FILE * fp,char * fname)188 c_cut(FILE *fp, char *fname)
189 {
190 	int ch, col;
191 	char *pos;
192 
193 	for (;;) {
194 		pos = positions + 1;
195 		for (col = maxval; col; --col) {
196 			if ((ch = getc(fp)) == EOF)
197 				return;
198 			if (ch == '\n')
199 				break;
200 			if (*pos++)
201 				(void)putchar(ch);
202 		}
203 		if (ch != '\n') {
204 			if (autostop)
205 				while ((ch = getc(fp)) != EOF && ch != '\n')
206 					(void)putchar(ch);
207 			else
208 				while ((ch = getc(fp)) != EOF && ch != '\n')
209 					;
210 		}
211 		(void)putchar('\n');
212 	}
213 }
214 
215 void
f_cut(FILE * fp,char * fname)216 f_cut(FILE *fp, char *fname)
217 {
218 	int ch, field, isdelim;
219 	char *pos, *p, sep;
220 	int output;
221 	size_t len;
222 	char *lbuf, *tbuf;
223 
224 	for (sep = dchar, tbuf = NULL; (lbuf = fgetln(fp, &len));) {
225 		output = 0;
226 		if (lbuf[len - 1] != '\n') {
227 			/* no newline at the end of the last line so add one */
228 			if ((tbuf = (char *)malloc(len + 1)) == NULL)
229 				err(1, NULL);
230 			memcpy(tbuf, lbuf, len);
231 			tbuf[len] = '\n';
232 			lbuf = tbuf;
233 		}
234 		for (isdelim = 0, p = lbuf;; ++p) {
235 			ch = *p;
236 			/* this should work if newline is delimiter */
237 			if (ch == sep)
238 				isdelim = 1;
239 			if (ch == '\n') {
240 				if (!isdelim && !sflag)
241 					(void)fwrite(lbuf, len, 1, stdout);
242 				break;
243 			}
244 		}
245 		if (!isdelim)
246 			continue;
247 
248 		pos = positions + 1;
249 		for (field = maxval, p = lbuf; field; --field, ++pos) {
250 			if (*pos) {
251 				if (output++)
252 					(void)putchar(sep);
253 				while ((ch = *p++) != '\n' && ch != sep)
254 					(void)putchar(ch);
255 			} else
256 				while ((ch = *p++) != '\n' && ch != sep)
257 					;
258 			if (ch == '\n')
259 				break;
260 		}
261 		if (ch != '\n') {
262 			if (autostop) {
263 				if (output)
264 					(void)putchar(sep);
265 				for (; (ch = *p) != '\n'; ++p)
266 					(void)putchar(ch);
267 			} else
268 				for (; (ch = *p) != '\n'; ++p)
269 					;
270 		}
271 		(void)putchar('\n');
272 	}
273 	if (tbuf)
274 		free(tbuf);
275 }
276 
277 void
usage(void)278 usage(void)
279 {
280 	(void)fprintf(stderr,
281 		"usage:\tcut -c list [file1 ...]\n"
282 		"\tcut -f list [-s] [-d delim] [file ...]\n"
283 		"\tcut -b list [-n] [file ...]\n");
284 	exit(1);
285 }
286