Annotation of src/usr.bin/cut/cut.c, Revision 1.17
1.17 ! sobrado 1: /* $OpenBSD: cut.c,v 1.16 2013/11/23 17:30:29 deraadt Exp $ */
1.1 deraadt 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.
1.10 millert 19: * 3. Neither the name of the University nor the names of its contributors
1.1 deraadt 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 <ctype.h>
37: #include <err.h>
38: #include <errno.h>
39: #include <limits.h>
40: #include <locale.h>
41: #include <stdio.h>
42: #include <stdlib.h>
43: #include <string.h>
44: #include <unistd.h>
45:
46: int cflag;
47: char dchar;
48: int dflag;
49: int fflag;
50: int sflag;
51:
1.9 millert 52: void c_cut(FILE *, char *);
53: void f_cut(FILE *, char *);
54: void get_list(char *);
55: void usage(void);
1.1 deraadt 56:
57: int
1.11 deraadt 58: main(int argc, char *argv[])
1.1 deraadt 59: {
60: FILE *fp;
1.9 millert 61: void (*fcn)(FILE *, char *);
1.17 ! sobrado 62: int ch, eval = 0;
1.1 deraadt 63:
64: setlocale (LC_ALL, "");
65:
66: dchar = '\t'; /* default delimiter is \t */
67:
68: /* Since we don't support multi-byte characters, the -c and -b
69: options are equivalent, and the -n option is meaningless. */
1.3 millert 70: while ((ch = getopt(argc, argv, "b:c:d:f:sn")) != -1)
1.1 deraadt 71: switch(ch) {
72: case 'b':
73: case 'c':
74: fcn = c_cut;
75: get_list(optarg);
76: cflag = 1;
77: break;
78: case 'd':
79: dchar = *optarg;
80: dflag = 1;
81: break;
82: case 'f':
83: get_list(optarg);
84: fcn = f_cut;
85: fflag = 1;
86: break;
87: case 's':
88: sflag = 1;
89: break;
90: case 'n':
91: break;
92: case '?':
93: default:
94: usage();
95: }
96: argc -= optind;
97: argv += optind;
98:
99: if (fflag) {
100: if (cflag)
101: usage();
102: } else if (!cflag || dflag || sflag)
103: usage();
104:
105: if (*argv)
106: for (; *argv; ++argv) {
1.17 ! sobrado 107: if (strcmp(*argv, "-") == 0)
! 108: fcn(stdin, "stdin");
! 109: else {
! 110: if ((fp = fopen(*argv, "r"))) {
! 111: fcn(fp, *argv);
! 112: (void)fclose(fp);
! 113: } else {
! 114: eval = 1;
! 115: warn("%s", *argv);
! 116: }
! 117: }
1.1 deraadt 118: }
119: else
120: fcn(stdin, "stdin");
1.17 ! sobrado 121: exit(eval);
1.1 deraadt 122: }
123:
124: int autostart, autostop, maxval;
125:
126: char positions[_POSIX2_LINE_MAX + 1];
127:
128: void
1.11 deraadt 129: get_list(char *list)
1.1 deraadt 130: {
1.8 mpech 131: int setautostart, start, stop;
132: char *pos;
1.1 deraadt 133: char *p;
134:
135: /*
136: * set a byte in the positions array to indicate if a field or
137: * column is to be selected; use +1, it's 1-based, not 0-based.
138: * This parser is less restrictive than the Draft 9 POSIX spec.
139: * POSIX doesn't allow lists that aren't in increasing order or
140: * overlapping lists. We also handle "-3-5" although there's no
141: * real reason too.
142: */
1.7 aaron 143: while ((p = strsep(&list, ", \t"))) {
1.1 deraadt 144: setautostart = start = stop = 0;
145: if (*p == '-') {
146: ++p;
147: setautostart = 1;
148: }
1.16 deraadt 149: if (isdigit((unsigned char)*p)) {
1.1 deraadt 150: start = stop = strtol(p, &p, 10);
151: if (setautostart && start > autostart)
152: autostart = start;
153: }
154: if (*p == '-') {
1.16 deraadt 155: if (isdigit((unsigned char)p[1]))
1.1 deraadt 156: stop = strtol(p + 1, &p, 10);
157: if (*p == '-') {
158: ++p;
159: if (!autostop || autostop > stop)
160: autostop = stop;
161: }
162: }
163: if (*p)
1.15 schwarze 164: errx(1, "[-bcf] list: illegal list value");
1.1 deraadt 165: if (!stop || !start)
1.15 schwarze 166: errx(1, "[-bcf] list: values may not include zero");
1.1 deraadt 167: if (stop > _POSIX2_LINE_MAX)
1.15 schwarze 168: errx(1, "[-bcf] list: %d too large (max %d)",
1.1 deraadt 169: stop, _POSIX2_LINE_MAX);
170: if (maxval < stop)
171: maxval = stop;
1.7 aaron 172: for (pos = positions + start; start++ <= stop; *pos++ = 1)
173: ;
1.1 deraadt 174: }
175:
176: /* overlapping ranges */
177: if (autostop && maxval > autostop)
178: maxval = autostop;
179:
180: /* set autostart */
181: if (autostart)
182: memset(positions + 1, '1', autostart);
183: }
184:
185: /* ARGSUSED */
186: void
1.11 deraadt 187: c_cut(FILE *fp, char *fname)
1.1 deraadt 188: {
1.8 mpech 189: int ch, col;
190: char *pos;
1.1 deraadt 191:
192: for (;;) {
193: pos = positions + 1;
194: for (col = maxval; col; --col) {
195: if ((ch = getc(fp)) == EOF)
196: return;
197: if (ch == '\n')
198: break;
199: if (*pos++)
200: (void)putchar(ch);
201: }
1.7 aaron 202: if (ch != '\n') {
1.1 deraadt 203: if (autostop)
204: while ((ch = getc(fp)) != EOF && ch != '\n')
205: (void)putchar(ch);
206: else
1.7 aaron 207: while ((ch = getc(fp)) != EOF && ch != '\n')
208: ;
209: }
1.1 deraadt 210: (void)putchar('\n');
211: }
212: }
213:
214: void
1.11 deraadt 215: f_cut(FILE *fp, char *fname)
1.1 deraadt 216: {
1.8 mpech 217: int ch, field, isdelim;
218: char *pos, *p, sep;
1.1 deraadt 219: int output;
1.7 aaron 220: size_t len;
221: char *lbuf, *tbuf;
1.1 deraadt 222:
1.7 aaron 223: for (sep = dchar, tbuf = NULL; (lbuf = fgetln(fp, &len));) {
1.1 deraadt 224: output = 0;
1.7 aaron 225: if (lbuf[len - 1] != '\n') {
226: /* no newline at the end of the last line so add one */
227: if ((tbuf = (char *)malloc(len + 1)) == NULL)
228: err(1, NULL);
229: memcpy(tbuf, lbuf, len);
230: tbuf[len] = '\n';
231: lbuf = tbuf;
232: }
1.1 deraadt 233: for (isdelim = 0, p = lbuf;; ++p) {
1.7 aaron 234: ch = *p;
1.1 deraadt 235: /* this should work if newline is delimiter */
236: if (ch == sep)
237: isdelim = 1;
238: if (ch == '\n') {
239: if (!isdelim && !sflag)
1.7 aaron 240: (void)fwrite(lbuf, len, 1, stdout);
1.1 deraadt 241: break;
242: }
243: }
244: if (!isdelim)
245: continue;
246:
247: pos = positions + 1;
248: for (field = maxval, p = lbuf; field; --field, ++pos) {
249: if (*pos) {
250: if (output++)
251: (void)putchar(sep);
252: while ((ch = *p++) != '\n' && ch != sep)
253: (void)putchar(ch);
254: } else
1.7 aaron 255: while ((ch = *p++) != '\n' && ch != sep)
256: ;
1.1 deraadt 257: if (ch == '\n')
258: break;
259: }
1.7 aaron 260: if (ch != '\n') {
1.1 deraadt 261: if (autostop) {
262: if (output)
263: (void)putchar(sep);
264: for (; (ch = *p) != '\n'; ++p)
265: (void)putchar(ch);
266: } else
1.7 aaron 267: for (; (ch = *p) != '\n'; ++p)
268: ;
269: }
1.1 deraadt 270: (void)putchar('\n');
271: }
1.7 aaron 272: if (tbuf)
273: free(tbuf);
1.1 deraadt 274: }
275:
276: void
1.11 deraadt 277: usage(void)
1.1 deraadt 278: {
279: (void)fprintf(stderr,
1.13 sobrado 280: "usage: cut -b list [-n] [file ...]\n"
281: " cut -c list [file ...]\n"
282: " cut -f list [-s] [-d delim] [file ...]\n");
1.1 deraadt 283: exit(1);
284: }