Annotation of src/usr.bin/uniq/uniq.c, Revision 1.32
1.32 ! cheloha 1: /* $OpenBSD: uniq.c,v 1.31 2022/01/01 02:20:38 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:
1.32 ! cheloha 48: long long numchars, numfields;
! 49: unsigned long long repeats;
1.25 tb 50: int cflag, dflag, iflag, uflag;
1.1 deraadt 51:
1.31 cheloha 52: void show(const char *);
1.9 millert 53: char *skip(char *);
54: void obsolete(char *[]);
1.15 ray 55: __dead void usage(void);
1.1 deraadt 56:
57: int
1.10 deraadt 58: main(int argc, char *argv[])
1.1 deraadt 59: {
1.30 cheloha 60: char *p, *prevline, *t, *thisline, *tmp;
1.28 cheloha 61: size_t prevsize, thissize, tmpsize;
1.29 cheloha 62: ssize_t len;
1.1 deraadt 63: int ch;
64:
1.24 schwarze 65: setlocale(LC_CTYPE, "");
66:
1.22 deraadt 67: if (pledge("stdio rpath wpath cpath", NULL) == -1)
68: err(1, "pledge");
1.20 deraadt 69:
1.1 deraadt 70: obsolete(argv);
1.25 tb 71: while ((ch = getopt(argc, argv, "cdf:is:u")) != -1) {
1.15 ray 72: const char *errstr;
73:
1.1 deraadt 74: switch (ch) {
75: case 'c':
76: cflag = 1;
77: break;
78: case 'd':
79: dflag = 1;
80: break;
81: case 'f':
1.32 ! cheloha 82: numfields = strtonum(optarg, 0, LLONG_MAX, &errstr);
1.15 ray 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':
1.32 ! cheloha 91: numchars = strtonum(optarg, 0, LLONG_MAX, &errstr);
1.15 ray 92: if (errstr)
93: errx(1,
94: "character skip value is %s: %s",
95: errstr, optarg);
1.1 deraadt 96: break;
97: case 'u':
98: uflag = 1;
99: break;
100: default:
101: usage();
1.15 ray 102: }
1.1 deraadt 103: }
104:
1.11 millert 105: argc -= optind;
1.15 ray 106: argv += optind;
1.1 deraadt 107:
1.16 kili 108: /* If neither -d nor -u are set, default is -d -u. */
109: if (!dflag && !uflag)
1.1 deraadt 110: dflag = uflag = 1;
111:
1.31 cheloha 112: if (argc > 2)
1.1 deraadt 113: usage();
1.31 cheloha 114: if (argc >= 1 && strcmp(argv[0], "-") != 0) {
115: if (freopen(argv[0], "r", stdin) == NULL)
116: err(1, "%s", argv[0]);
117: }
118: if (argc == 2 && strcmp(argv[1], "-") != 0) {
119: if (freopen(argv[1], "w", stdout) == NULL)
120: err(1, "%s", argv[1]);
1.1 deraadt 121: }
1.20 deraadt 122:
1.22 deraadt 123: if (pledge("stdio", NULL) == -1)
124: err(1, "pledge");
1.1 deraadt 125:
1.28 cheloha 126: prevsize = 0;
127: prevline = NULL;
1.31 cheloha 128: if ((len = getline(&prevline, &prevsize, stdin)) == -1) {
1.28 cheloha 129: free(prevline);
1.31 cheloha 130: if (ferror(stdin))
1.28 cheloha 131: err(1, "getline");
1.1 deraadt 132: exit(0);
1.28 cheloha 133: }
1.29 cheloha 134: if (prevline[len - 1] == '\n')
135: prevline[len - 1] = '\0';
1.30 cheloha 136: if (numfields || numchars)
137: p = skip(prevline);
138: else
139: p = prevline;
1.28 cheloha 140:
141: thissize = 0;
142: thisline = NULL;
1.31 cheloha 143: while ((len = getline(&thisline, &thissize, stdin)) != -1) {
1.29 cheloha 144: if (thisline[len - 1] == '\n')
145: thisline[len - 1] = '\0';
146:
1.1 deraadt 147: /* If requested get the chosen fields + character offsets. */
1.30 cheloha 148: if (numfields || numchars)
149: t = skip(thisline);
150: else
151: t = thisline;
1.1 deraadt 152:
153: /* If different, print; set previous to new value. */
1.30 cheloha 154: if ((iflag ? strcasecmp : strcmp)(p, t)) {
1.31 cheloha 155: show(prevline);
1.30 cheloha 156: tmp = prevline;
1.1 deraadt 157: prevline = thisline;
1.30 cheloha 158: thisline = tmp;
159: tmp = p;
160: p = t;
161: t = tmp;
1.28 cheloha 162: tmpsize = prevsize;
163: prevsize = thissize;
164: thissize = tmpsize;
1.1 deraadt 165: repeats = 0;
166: } else
167: ++repeats;
168: }
1.28 cheloha 169: free(thisline);
1.31 cheloha 170: if (ferror(stdin))
1.28 cheloha 171: err(1, "getline");
172:
1.31 cheloha 173: show(prevline);
1.28 cheloha 174: free(prevline);
175:
1.1 deraadt 176: exit(0);
177: }
178:
179: /*
180: * show --
181: * Output a line depending on the flags and number of repetitions
182: * of the line.
183: */
184: void
1.31 cheloha 185: show(const char *str)
1.1 deraadt 186: {
1.16 kili 187: if ((dflag && repeats) || (uflag && !repeats)) {
188: if (cflag)
1.32 ! cheloha 189: printf("%4llu %s\n", repeats + 1, str);
1.16 kili 190: else
1.31 cheloha 191: printf("%s\n", str);
1.16 kili 192: }
1.1 deraadt 193: }
194:
195: char *
1.10 deraadt 196: skip(char *str)
1.1 deraadt 197: {
1.32 ! cheloha 198: long long nchars, nfields;
1.24 schwarze 199: wchar_t wc;
200: int len;
201: int field_started;
1.1 deraadt 202:
1.16 kili 203: for (nfields = numfields; nfields && *str; nfields--) {
1.24 schwarze 204: /* Skip one field, including preceding blanks. */
205: for (field_started = 0; *str != '\0'; str += len) {
206: if ((len = mbtowc(&wc, str, MB_CUR_MAX)) == -1) {
207: (void)mbtowc(NULL, NULL, MB_CUR_MAX);
208: wc = L'?';
209: len = 1;
210: }
211: if (iswblank(wc)) {
212: if (field_started)
213: break;
214: } else
215: field_started = 1;
216: }
1.16 kili 217: }
1.24 schwarze 218:
219: /* Skip some additional characters. */
220: for (nchars = numchars; nchars-- && *str != '\0'; str += len)
221: if ((len = mblen(str, MB_CUR_MAX)) == -1)
222: len = 1;
223:
1.15 ray 224: return (str);
1.1 deraadt 225: }
226:
227: void
1.10 deraadt 228: obsolete(char *argv[])
1.1 deraadt 229: {
1.15 ray 230: size_t len;
1.1 deraadt 231: char *ap, *p, *start;
232:
1.7 deraadt 233: while ((ap = *++argv)) {
1.1 deraadt 234: /* Return if "--" or not an option of any form. */
235: if (ap[0] != '-') {
236: if (ap[0] != '+')
237: return;
238: } else if (ap[1] == '-')
239: return;
1.23 mmcc 240: if (!isdigit((unsigned char)ap[1]))
1.1 deraadt 241: continue;
242: /*
243: * Digit signifies an old-style option. Malloc space for dash,
244: * new option and argument.
245: */
1.13 deraadt 246: len = strlen(ap) + 3;
247: if ((start = p = malloc(len)) == NULL)
1.5 mickey 248: err(1, "malloc");
1.1 deraadt 249: *p++ = '-';
250: *p++ = ap[0] == '+' ? 's' : 'f';
1.13 deraadt 251: (void)strlcpy(p, ap + 1, len - 2);
1.1 deraadt 252: *argv = start;
253: }
254: }
255:
1.15 ray 256: __dead void
1.10 deraadt 257: usage(void)
1.1 deraadt 258: {
1.15 ray 259: extern char *__progname;
1.16 kili 260:
1.1 deraadt 261: (void)fprintf(stderr,
1.25 tb 262: "usage: %s [-ci] [-d | -u] [-f fields] [-s chars] [input_file [output_file]]\n",
1.15 ray 263: __progname);
1.1 deraadt 264: exit(1);
265: }