Annotation of src/usr.bin/cut/cut.c, Revision 1.14
1.14 ! deraadt 1: /* $OpenBSD: cut.c,v 1.13 2008/06/27 08:02:13 sobrado 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 *);
1.1 deraadt 62: int ch;
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) {
107: if (!(fp = fopen(*argv, "r")))
108: err(1, "%s", *argv);
109: fcn(fp, *argv);
110: (void)fclose(fp);
111: }
112: else
113: fcn(stdin, "stdin");
114: exit(0);
115: }
116:
117: int autostart, autostop, maxval;
118:
119: char positions[_POSIX2_LINE_MAX + 1];
120:
121: void
1.11 deraadt 122: get_list(char *list)
1.1 deraadt 123: {
1.8 mpech 124: int setautostart, start, stop;
125: char *pos;
1.1 deraadt 126: char *p;
127:
128: /*
129: * set a byte in the positions array to indicate if a field or
130: * column is to be selected; use +1, it's 1-based, not 0-based.
131: * This parser is less restrictive than the Draft 9 POSIX spec.
132: * POSIX doesn't allow lists that aren't in increasing order or
133: * overlapping lists. We also handle "-3-5" although there's no
134: * real reason too.
135: */
1.7 aaron 136: while ((p = strsep(&list, ", \t"))) {
1.1 deraadt 137: setautostart = start = stop = 0;
138: if (*p == '-') {
139: ++p;
140: setautostart = 1;
141: }
142: if (isdigit(*p)) {
143: start = stop = strtol(p, &p, 10);
144: if (setautostart && start > autostart)
145: autostart = start;
146: }
147: if (*p == '-') {
148: if (isdigit(p[1]))
149: stop = strtol(p + 1, &p, 10);
150: if (*p == '-') {
151: ++p;
152: if (!autostop || autostop > stop)
153: autostop = stop;
154: }
155: }
156: if (*p)
1.4 millert 157: errx(1, "[-cf] list: illegal list value");
1.1 deraadt 158: if (!stop || !start)
1.4 millert 159: errx(1, "[-cf] list: values may not include zero");
1.1 deraadt 160: if (stop > _POSIX2_LINE_MAX)
1.4 millert 161: errx(1, "[-cf] list: %d too large (max %d)",
1.1 deraadt 162: stop, _POSIX2_LINE_MAX);
163: if (maxval < stop)
164: maxval = stop;
1.7 aaron 165: for (pos = positions + start; start++ <= stop; *pos++ = 1)
166: ;
1.1 deraadt 167: }
168:
169: /* overlapping ranges */
170: if (autostop && maxval > autostop)
171: maxval = autostop;
172:
173: /* set autostart */
174: if (autostart)
175: memset(positions + 1, '1', autostart);
176: }
177:
178: /* ARGSUSED */
179: void
1.11 deraadt 180: c_cut(FILE *fp, char *fname)
1.1 deraadt 181: {
1.8 mpech 182: int ch, col;
183: char *pos;
1.1 deraadt 184:
185: for (;;) {
186: pos = positions + 1;
187: for (col = maxval; col; --col) {
188: if ((ch = getc(fp)) == EOF)
189: return;
190: if (ch == '\n')
191: break;
192: if (*pos++)
193: (void)putchar(ch);
194: }
1.7 aaron 195: if (ch != '\n') {
1.1 deraadt 196: if (autostop)
197: while ((ch = getc(fp)) != EOF && ch != '\n')
198: (void)putchar(ch);
199: else
1.7 aaron 200: while ((ch = getc(fp)) != EOF && ch != '\n')
201: ;
202: }
1.1 deraadt 203: (void)putchar('\n');
204: }
205: }
206:
207: void
1.11 deraadt 208: f_cut(FILE *fp, char *fname)
1.1 deraadt 209: {
1.8 mpech 210: int ch, field, isdelim;
211: char *pos, *p, sep;
1.1 deraadt 212: int output;
1.7 aaron 213: size_t len;
214: char *lbuf, *tbuf;
1.1 deraadt 215:
1.7 aaron 216: for (sep = dchar, tbuf = NULL; (lbuf = fgetln(fp, &len));) {
1.1 deraadt 217: output = 0;
1.7 aaron 218: if (lbuf[len - 1] != '\n') {
219: /* no newline at the end of the last line so add one */
220: if ((tbuf = (char *)malloc(len + 1)) == NULL)
221: err(1, NULL);
222: memcpy(tbuf, lbuf, len);
223: tbuf[len] = '\n';
224: lbuf = tbuf;
225: }
1.1 deraadt 226: for (isdelim = 0, p = lbuf;; ++p) {
1.7 aaron 227: ch = *p;
1.1 deraadt 228: /* this should work if newline is delimiter */
229: if (ch == sep)
230: isdelim = 1;
231: if (ch == '\n') {
232: if (!isdelim && !sflag)
1.7 aaron 233: (void)fwrite(lbuf, len, 1, stdout);
1.1 deraadt 234: break;
235: }
236: }
237: if (!isdelim)
238: continue;
239:
240: pos = positions + 1;
241: for (field = maxval, p = lbuf; field; --field, ++pos) {
242: if (*pos) {
243: if (output++)
244: (void)putchar(sep);
245: while ((ch = *p++) != '\n' && ch != sep)
246: (void)putchar(ch);
247: } else
1.7 aaron 248: while ((ch = *p++) != '\n' && ch != sep)
249: ;
1.1 deraadt 250: if (ch == '\n')
251: break;
252: }
1.7 aaron 253: if (ch != '\n') {
1.1 deraadt 254: if (autostop) {
255: if (output)
256: (void)putchar(sep);
257: for (; (ch = *p) != '\n'; ++p)
258: (void)putchar(ch);
259: } else
1.7 aaron 260: for (; (ch = *p) != '\n'; ++p)
261: ;
262: }
1.1 deraadt 263: (void)putchar('\n');
264: }
1.7 aaron 265: if (tbuf)
266: free(tbuf);
1.1 deraadt 267: }
268:
269: void
1.11 deraadt 270: usage(void)
1.1 deraadt 271: {
272: (void)fprintf(stderr,
1.13 sobrado 273: "usage: cut -b list [-n] [file ...]\n"
274: " cut -c list [file ...]\n"
275: " cut -f list [-s] [-d delim] [file ...]\n");
1.1 deraadt 276: exit(1);
277: }