1 /* $NetBSD: cut.c,v 1.29 2014/02/03 20:22:19 wiz Exp $ */
2
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 #include <sys/cdefs.h>
36 #ifndef lint
37 __COPYRIGHT("@(#) Copyright (c) 1989, 1993\
38 The Regents of the University of California. All rights reserved.");
39 #endif /* not lint */
40
41 #ifndef lint
42 #if 0
43 static char sccsid[] = "@(#)cut.c 8.3 (Berkeley) 5/4/95";
44 #endif
45 __RCSID("$NetBSD: cut.c,v 1.29 2014/02/03 20:22:19 wiz Exp $");
46 #endif /* not lint */
47
48 #include <ctype.h>
49 #include <err.h>
50 #include <errno.h>
51 #include <limits.h>
52 #include <locale.h>
53 #include <stdio.h>
54 #include <stdlib.h>
55 #include <string.h>
56 #include <unistd.h>
57 #include <util.h>
58 #include <wchar.h>
59 #include <sys/param.h>
60
61 static int bflag;
62 static int cflag;
63 static char dchar;
64 static int dflag;
65 static int fflag;
66 static int sflag;
67
68 static void b_cut(FILE *, const char *);
69 static void c_cut(FILE *, const char *);
70 static void f_cut(FILE *, const char *);
71 static void get_list(char *);
72 static void usage(void) __dead;
73
74 int
main(int argc,char * argv[])75 main(int argc, char *argv[])
76 {
77 FILE *fp;
78 void (*fcn)(FILE *, const char *);
79 int ch, rval;
80
81 fcn = NULL;
82 (void)setlocale(LC_ALL, "");
83
84 dchar = '\t'; /* default delimiter is \t */
85
86 /* Since we don't support multi-byte characters, the -c and -b
87 options are equivalent, and the -n option is meaningless. */
88 while ((ch = getopt(argc, argv, "b:c:d:f:sn")) != -1)
89 switch(ch) {
90 case 'b':
91 fcn = b_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 break;
114 case '?':
115 default:
116 usage();
117 }
118 argc -= optind;
119 argv += optind;
120
121 if (fflag) {
122 if (cflag || bflag)
123 usage();
124 } else if ((!cflag && !bflag) || dflag || sflag)
125 usage();
126 else if (bflag && cflag)
127 usage();
128
129 rval = 0;
130 if (*argv)
131 for (; *argv; ++argv) {
132 if (strcmp(*argv, "-") == 0)
133 fcn(stdin, "stdin");
134 else {
135 if ((fp = fopen(*argv, "r"))) {
136 fcn(fp, *argv);
137 (void)fclose(fp);
138 } else {
139 rval = 1;
140 warn("%s", *argv);
141 }
142 }
143 }
144 else
145 fcn(stdin, "stdin");
146 return(rval);
147 }
148
149 static size_t autostart, autostop, maxval;
150
151 static char *positions = NULL;
152 static size_t numpositions = 0;
153 #define ALLOC_CHUNK _POSIX2_LINE_MAX /* malloc granularity */
154
155 static void
get_list(char * list)156 get_list(char *list)
157 {
158 size_t setautostart, start, stop;
159 char *pos;
160 char *p;
161
162 if (positions == NULL) {
163 numpositions = ALLOC_CHUNK;
164 positions = ecalloc(numpositions, sizeof(*positions));
165 }
166
167 /*
168 * set a byte in the positions array to indicate if a field or
169 * column is to be selected; use +1, it's 1-based, not 0-based.
170 * This parser is less restrictive than the Draft 9 POSIX spec.
171 * POSIX doesn't allow lists that aren't in increasing order or
172 * overlapping lists. We also handle "-3-5" although there's no
173 * real reason to.
174 */
175 for (; (p = strtok(list, ", \t")) != NULL; list = NULL) {
176 setautostart = start = stop = 0;
177 if (*p == '-') {
178 ++p;
179 setautostart = 1;
180 }
181 if (isdigit((unsigned char)*p)) {
182 start = stop = strtol(p, &p, 10);
183 if (setautostart && start > autostart)
184 autostart = start;
185 }
186 if (*p == '-') {
187 if (isdigit((unsigned char)p[1]))
188 stop = strtol(p + 1, &p, 10);
189 if (*p == '-') {
190 ++p;
191 if (!autostop || autostop > stop)
192 autostop = stop;
193 }
194 }
195 if (*p)
196 errx(1, "[-bcf] list: illegal list value");
197 if (!stop || !start)
198 errx(1, "[-bcf] list: values may not include zero");
199 if (stop + 1 > numpositions) {
200 size_t newsize;
201 newsize = roundup(stop + 1, ALLOC_CHUNK);
202 positions = erealloc(positions, newsize);
203 (void)memset(positions + numpositions, 0,
204 newsize - numpositions);
205 numpositions = newsize;
206 }
207 if (maxval < stop)
208 maxval = stop;
209 for (pos = positions + start; start++ <= stop; pos++)
210 *pos = 1;
211 }
212
213 /* overlapping ranges */
214 if (autostop && maxval > autostop)
215 maxval = autostop;
216
217 /* set autostart */
218 if (autostart)
219 (void)memset(positions + 1, '1', autostart);
220 }
221
222 static void
223 /*ARGSUSED*/
f_cut(FILE * fp,const char * fname __unused)224 f_cut(FILE *fp, const char *fname __unused)
225 {
226 int ch, field, isdelim;
227 char *pos, *p, sep;
228 int output;
229 size_t len;
230 char *lbuf, *tbuf;
231
232 for (sep = dchar, tbuf = NULL; (lbuf = fgetln(fp, &len)) != NULL;) {
233 output = 0;
234 if (lbuf[len - 1] != '\n') {
235 /* no newline at the end of the last line so add one */
236 if ((tbuf = (char *)malloc(len + 1)) == NULL)
237 err(1, NULL);
238 (void)memcpy(tbuf, lbuf, len);
239 tbuf[len++] = '\n';
240 lbuf = tbuf;
241 }
242 for (isdelim = 0, p = lbuf;; ++p) {
243 ch = *p;
244 /* this should work if newline is delimiter */
245 if (ch == sep)
246 isdelim = 1;
247 if (ch == '\n') {
248 if (!isdelim && !sflag)
249 (void)fwrite(lbuf, len, 1, stdout);
250 break;
251 }
252 }
253 if (!isdelim)
254 continue;
255
256 pos = positions + 1;
257 for (field = maxval, p = lbuf; field; --field, ++pos) {
258 if (*pos) {
259 if (output++)
260 (void)putchar(sep);
261 while ((ch = *p++) != '\n' && ch != sep)
262 (void)putchar(ch);
263 } else {
264 while ((ch = *p++) != '\n' && ch != sep)
265 continue;
266 }
267 if (ch == '\n')
268 break;
269 }
270 if (ch != '\n') {
271 if (autostop) {
272 if (output)
273 (void)putchar(sep);
274 for (; (ch = *p) != '\n'; ++p)
275 (void)putchar(ch);
276 } else
277 for (; (ch = *p) != '\n'; ++p);
278 }
279 (void)putchar('\n');
280 if (tbuf) {
281 free(tbuf);
282 tbuf = NULL;
283 }
284 }
285 if (tbuf)
286 free(tbuf);
287 }
288
289 static void
usage(void)290 usage(void)
291 {
292 (void)fprintf(stderr, "usage:\tcut -b list [-n] [file ...]\n"
293 "\tcut -c list [file ...]\n"
294 "\tcut -f list [-d string] [-s] [file ...]\n");
295 exit(1);
296 }
297
298 /* make b_put(): */
299 #define CUT_BYTE 1
300 #include "x_cut.c"
301 #undef CUT_BYTE
302
303 /* make c_put(): */
304 #define CUT_BYTE 0
305 #include "x_cut.c"
306 #undef CUT_BYTE
307