Annotation of src/usr.bin/uniq/uniq.c, Revision 1.31
1.31 ! cheloha 1: /* $OpenBSD: uniq.c,v 1.30 2021/12/24 17:59:28 cheloha Exp $ */
1.1 deraadt 2: /* $NetBSD: uniq.c,v 1.7 1995/08/31 22:03:48 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: * Case Larsen.
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.
1.14 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:
1.15 ray 36: #include <ctype.h>
37: #include <err.h>
1.1 deraadt 38: #include <errno.h>
1.15 ray 39: #include <limits.h>
1.24 schwarze 40: #include <locale.h>
1.1 deraadt 41: #include <stdio.h>
42: #include <stdlib.h>
43: #include <string.h>
44: #include <unistd.h>
1.24 schwarze 45: #include <wchar.h>
46: #include <wctype.h>
1.1 deraadt 47:
1.25 tb 48: int cflag, dflag, iflag, uflag;
1.1 deraadt 49: int numchars, numfields, repeats;
50:
1.31 ! cheloha 51: void show(const char *);
1.9 millert 52: char *skip(char *);
53: void obsolete(char *[]);
1.15 ray 54: __dead void usage(void);
1.1 deraadt 55:
56: int
1.10 deraadt 57: main(int argc, char *argv[])
1.1 deraadt 58: {
1.30 cheloha 59: char *p, *prevline, *t, *thisline, *tmp;
1.28 cheloha 60: size_t prevsize, thissize, tmpsize;
1.29 cheloha 61: ssize_t len;
1.1 deraadt 62: int ch;
63:
1.24 schwarze 64: setlocale(LC_CTYPE, "");
65:
1.22 deraadt 66: if (pledge("stdio rpath wpath cpath", NULL) == -1)
67: err(1, "pledge");
1.20 deraadt 68:
1.1 deraadt 69: obsolete(argv);
1.25 tb 70: while ((ch = getopt(argc, argv, "cdf:is:u")) != -1) {
1.15 ray 71: const char *errstr;
72:
1.1 deraadt 73: switch (ch) {
74: case 'c':
75: cflag = 1;
76: break;
77: case 'd':
78: dflag = 1;
79: break;
80: case 'f':
1.15 ray 81: numfields = (int)strtonum(optarg, 0, INT_MAX,
82: &errstr);
83: if (errstr)
84: errx(1, "field skip value is %s: %s",
85: errstr, optarg);
1.1 deraadt 86: break;
1.25 tb 87: case 'i':
88: iflag = 1;
89: break;
1.1 deraadt 90: case 's':
1.15 ray 91: numchars = (int)strtonum(optarg, 0, INT_MAX,
92: &errstr);
93: if (errstr)
94: errx(1,
95: "character skip value is %s: %s",
96: errstr, optarg);
1.1 deraadt 97: break;
98: case 'u':
99: uflag = 1;
100: break;
101: default:
102: usage();
1.15 ray 103: }
1.1 deraadt 104: }
105:
1.11 millert 106: argc -= optind;
1.15 ray 107: argv += optind;
1.1 deraadt 108:
1.16 kili 109: /* If neither -d nor -u are set, default is -d -u. */
110: if (!dflag && !uflag)
1.1 deraadt 111: dflag = uflag = 1;
112:
1.31 ! cheloha 113: if (argc > 2)
1.1 deraadt 114: usage();
1.31 ! cheloha 115: if (argc >= 1 && strcmp(argv[0], "-") != 0) {
! 116: if (freopen(argv[0], "r", stdin) == NULL)
! 117: err(1, "%s", argv[0]);
! 118: }
! 119: if (argc == 2 && strcmp(argv[1], "-") != 0) {
! 120: if (freopen(argv[1], "w", stdout) == NULL)
! 121: err(1, "%s", argv[1]);
1.1 deraadt 122: }
1.20 deraadt 123:
1.22 deraadt 124: if (pledge("stdio", NULL) == -1)
125: err(1, "pledge");
1.1 deraadt 126:
1.28 cheloha 127: prevsize = 0;
128: prevline = NULL;
1.31 ! cheloha 129: if ((len = getline(&prevline, &prevsize, stdin)) == -1) {
1.28 cheloha 130: free(prevline);
1.31 ! cheloha 131: if (ferror(stdin))
1.28 cheloha 132: err(1, "getline");
1.1 deraadt 133: exit(0);
1.28 cheloha 134: }
1.29 cheloha 135: if (prevline[len - 1] == '\n')
136: prevline[len - 1] = '\0';
1.30 cheloha 137: if (numfields || numchars)
138: p = skip(prevline);
139: else
140: p = prevline;
1.28 cheloha 141:
142: thissize = 0;
143: thisline = NULL;
1.31 ! cheloha 144: while ((len = getline(&thisline, &thissize, stdin)) != -1) {
1.29 cheloha 145: if (thisline[len - 1] == '\n')
146: thisline[len - 1] = '\0';
147:
1.1 deraadt 148: /* If requested get the chosen fields + character offsets. */
1.30 cheloha 149: if (numfields || numchars)
150: t = skip(thisline);
151: else
152: t = thisline;
1.1 deraadt 153:
154: /* If different, print; set previous to new value. */
1.30 cheloha 155: if ((iflag ? strcasecmp : strcmp)(p, t)) {
1.31 ! cheloha 156: show(prevline);
1.30 cheloha 157: tmp = prevline;
1.1 deraadt 158: prevline = thisline;
1.30 cheloha 159: thisline = tmp;
160: tmp = p;
161: p = t;
162: t = tmp;
1.28 cheloha 163: tmpsize = prevsize;
164: prevsize = thissize;
165: thissize = tmpsize;
1.1 deraadt 166: repeats = 0;
167: } else
168: ++repeats;
169: }
1.28 cheloha 170: free(thisline);
1.31 ! cheloha 171: if (ferror(stdin))
1.28 cheloha 172: err(1, "getline");
173:
1.31 ! cheloha 174: show(prevline);
1.28 cheloha 175: free(prevline);
176:
1.1 deraadt 177: exit(0);
178: }
179:
180: /*
181: * show --
182: * Output a line depending on the flags and number of repetitions
183: * of the line.
184: */
185: void
1.31 ! cheloha 186: show(const char *str)
1.1 deraadt 187: {
1.16 kili 188: if ((dflag && repeats) || (uflag && !repeats)) {
189: if (cflag)
1.31 ! cheloha 190: printf("%4d %s\n", repeats + 1, str);
1.16 kili 191: else
1.31 ! cheloha 192: printf("%s\n", str);
1.16 kili 193: }
1.1 deraadt 194: }
195:
196: char *
1.10 deraadt 197: skip(char *str)
1.1 deraadt 198: {
1.24 schwarze 199: wchar_t wc;
1.16 kili 200: int nchars, nfields;
1.24 schwarze 201: int len;
202: int field_started;
1.1 deraadt 203:
1.16 kili 204: for (nfields = numfields; nfields && *str; nfields--) {
1.24 schwarze 205: /* Skip one field, including preceding blanks. */
206: for (field_started = 0; *str != '\0'; str += len) {
207: if ((len = mbtowc(&wc, str, MB_CUR_MAX)) == -1) {
208: (void)mbtowc(NULL, NULL, MB_CUR_MAX);
209: wc = L'?';
210: len = 1;
211: }
212: if (iswblank(wc)) {
213: if (field_started)
214: break;
215: } else
216: field_started = 1;
217: }
1.16 kili 218: }
1.24 schwarze 219:
220: /* Skip some additional characters. */
221: for (nchars = numchars; nchars-- && *str != '\0'; str += len)
222: if ((len = mblen(str, MB_CUR_MAX)) == -1)
223: len = 1;
224:
1.15 ray 225: return (str);
1.1 deraadt 226: }
227:
228: void
1.10 deraadt 229: obsolete(char *argv[])
1.1 deraadt 230: {
1.15 ray 231: size_t len;
1.1 deraadt 232: char *ap, *p, *start;
233:
1.7 deraadt 234: while ((ap = *++argv)) {
1.1 deraadt 235: /* Return if "--" or not an option of any form. */
236: if (ap[0] != '-') {
237: if (ap[0] != '+')
238: return;
239: } else if (ap[1] == '-')
240: return;
1.23 mmcc 241: if (!isdigit((unsigned char)ap[1]))
1.1 deraadt 242: continue;
243: /*
244: * Digit signifies an old-style option. Malloc space for dash,
245: * new option and argument.
246: */
1.13 deraadt 247: len = strlen(ap) + 3;
248: if ((start = p = malloc(len)) == NULL)
1.5 mickey 249: err(1, "malloc");
1.1 deraadt 250: *p++ = '-';
251: *p++ = ap[0] == '+' ? 's' : 'f';
1.13 deraadt 252: (void)strlcpy(p, ap + 1, len - 2);
1.1 deraadt 253: *argv = start;
254: }
255: }
256:
1.15 ray 257: __dead void
1.10 deraadt 258: usage(void)
1.1 deraadt 259: {
1.15 ray 260: extern char *__progname;
1.16 kili 261:
1.1 deraadt 262: (void)fprintf(stderr,
1.25 tb 263: "usage: %s [-ci] [-d | -u] [-f fields] [-s chars] [input_file [output_file]]\n",
1.15 ray 264: __progname);
1.1 deraadt 265: exit(1);
266: }