Annotation of src/usr.bin/uniq/uniq.c, Revision 1.33
1.33 ! cheloha 1: /* $OpenBSD: uniq.c,v 1.32 2022/01/01 17:44:18 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>
38: #include <limits.h>
1.24 schwarze 39: #include <locale.h>
1.1 deraadt 40: #include <stdio.h>
41: #include <stdlib.h>
42: #include <string.h>
1.33 ! cheloha 43: #include <strings.h>
1.1 deraadt 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: {
1.33 ! cheloha 60: const char *errstr;
1.30 cheloha 61: char *p, *prevline, *t, *thisline, *tmp;
1.28 cheloha 62: size_t prevsize, thissize, tmpsize;
1.29 cheloha 63: ssize_t len;
1.1 deraadt 64: int ch;
65:
1.24 schwarze 66: setlocale(LC_CTYPE, "");
67:
1.22 deraadt 68: if (pledge("stdio rpath wpath cpath", NULL) == -1)
69: err(1, "pledge");
1.20 deraadt 70:
1.1 deraadt 71: obsolete(argv);
1.25 tb 72: while ((ch = getopt(argc, argv, "cdf:is:u")) != -1) {
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':
1.32 cheloha 81: numfields = strtonum(optarg, 0, LLONG_MAX, &errstr);
1.15 ray 82: if (errstr)
1.33 ! cheloha 83: errx(1, "fields is %s: %s", errstr, optarg);
1.1 deraadt 84: break;
1.25 tb 85: case 'i':
86: iflag = 1;
87: break;
1.1 deraadt 88: case 's':
1.32 cheloha 89: numchars = strtonum(optarg, 0, LLONG_MAX, &errstr);
1.15 ray 90: if (errstr)
1.33 ! cheloha 91: errx(1, "chars is %s: %s", errstr, optarg);
1.1 deraadt 92: break;
93: case 'u':
94: uflag = 1;
95: break;
96: default:
97: usage();
1.15 ray 98: }
1.1 deraadt 99: }
1.11 millert 100: argc -= optind;
1.15 ray 101: argv += optind;
1.1 deraadt 102:
1.16 kili 103: /* If neither -d nor -u are set, default is -d -u. */
104: if (!dflag && !uflag)
1.1 deraadt 105: dflag = uflag = 1;
106:
1.31 cheloha 107: if (argc > 2)
1.1 deraadt 108: usage();
1.31 cheloha 109: if (argc >= 1 && strcmp(argv[0], "-") != 0) {
110: if (freopen(argv[0], "r", stdin) == NULL)
111: err(1, "%s", argv[0]);
112: }
113: if (argc == 2 && strcmp(argv[1], "-") != 0) {
114: if (freopen(argv[1], "w", stdout) == NULL)
115: err(1, "%s", argv[1]);
1.1 deraadt 116: }
1.20 deraadt 117:
1.22 deraadt 118: if (pledge("stdio", NULL) == -1)
119: err(1, "pledge");
1.1 deraadt 120:
1.28 cheloha 121: prevsize = 0;
122: prevline = NULL;
1.31 cheloha 123: if ((len = getline(&prevline, &prevsize, stdin)) == -1) {
1.28 cheloha 124: free(prevline);
1.31 cheloha 125: if (ferror(stdin))
1.28 cheloha 126: err(1, "getline");
1.33 ! cheloha 127: return 0;
1.28 cheloha 128: }
1.29 cheloha 129: if (prevline[len - 1] == '\n')
130: prevline[len - 1] = '\0';
1.30 cheloha 131: if (numfields || numchars)
132: p = skip(prevline);
133: else
134: p = prevline;
1.28 cheloha 135:
136: thissize = 0;
137: thisline = NULL;
1.31 cheloha 138: while ((len = getline(&thisline, &thissize, stdin)) != -1) {
1.29 cheloha 139: if (thisline[len - 1] == '\n')
140: thisline[len - 1] = '\0';
141:
1.1 deraadt 142: /* If requested get the chosen fields + character offsets. */
1.30 cheloha 143: if (numfields || numchars)
144: t = skip(thisline);
145: else
146: t = thisline;
1.1 deraadt 147:
148: /* If different, print; set previous to new value. */
1.30 cheloha 149: if ((iflag ? strcasecmp : strcmp)(p, t)) {
1.31 cheloha 150: show(prevline);
1.30 cheloha 151: tmp = prevline;
1.1 deraadt 152: prevline = thisline;
1.30 cheloha 153: thisline = tmp;
154: tmp = p;
155: p = t;
156: t = tmp;
1.28 cheloha 157: tmpsize = prevsize;
158: prevsize = thissize;
159: thissize = tmpsize;
1.1 deraadt 160: repeats = 0;
161: } else
162: ++repeats;
163: }
1.28 cheloha 164: free(thisline);
1.31 cheloha 165: if (ferror(stdin))
1.28 cheloha 166: err(1, "getline");
167:
1.31 cheloha 168: show(prevline);
1.28 cheloha 169: free(prevline);
170:
1.33 ! cheloha 171: return 0;
1.1 deraadt 172: }
173:
174: /*
175: * show --
176: * Output a line depending on the flags and number of repetitions
177: * of the line.
178: */
179: void
1.31 cheloha 180: show(const char *str)
1.1 deraadt 181: {
1.16 kili 182: if ((dflag && repeats) || (uflag && !repeats)) {
183: if (cflag)
1.32 cheloha 184: printf("%4llu %s\n", repeats + 1, str);
1.16 kili 185: else
1.31 cheloha 186: printf("%s\n", str);
1.16 kili 187: }
1.1 deraadt 188: }
189:
190: char *
1.10 deraadt 191: skip(char *str)
1.1 deraadt 192: {
1.32 cheloha 193: long long nchars, nfields;
1.24 schwarze 194: wchar_t wc;
195: int len;
196: int field_started;
1.1 deraadt 197:
1.16 kili 198: for (nfields = numfields; nfields && *str; nfields--) {
1.24 schwarze 199: /* Skip one field, including preceding blanks. */
200: for (field_started = 0; *str != '\0'; str += len) {
201: if ((len = mbtowc(&wc, str, MB_CUR_MAX)) == -1) {
202: (void)mbtowc(NULL, NULL, MB_CUR_MAX);
203: wc = L'?';
204: len = 1;
205: }
206: if (iswblank(wc)) {
207: if (field_started)
208: break;
209: } else
210: field_started = 1;
211: }
1.16 kili 212: }
1.24 schwarze 213:
214: /* Skip some additional characters. */
215: for (nchars = numchars; nchars-- && *str != '\0'; str += len)
216: if ((len = mblen(str, MB_CUR_MAX)) == -1)
217: len = 1;
218:
1.15 ray 219: return (str);
1.1 deraadt 220: }
221:
222: void
1.10 deraadt 223: obsolete(char *argv[])
1.1 deraadt 224: {
1.15 ray 225: size_t len;
1.1 deraadt 226: char *ap, *p, *start;
227:
1.7 deraadt 228: while ((ap = *++argv)) {
1.1 deraadt 229: /* Return if "--" or not an option of any form. */
230: if (ap[0] != '-') {
231: if (ap[0] != '+')
232: return;
233: } else if (ap[1] == '-')
234: return;
1.23 mmcc 235: if (!isdigit((unsigned char)ap[1]))
1.1 deraadt 236: continue;
237: /*
238: * Digit signifies an old-style option. Malloc space for dash,
239: * new option and argument.
240: */
1.13 deraadt 241: len = strlen(ap) + 3;
242: if ((start = p = malloc(len)) == NULL)
1.5 mickey 243: err(1, "malloc");
1.1 deraadt 244: *p++ = '-';
245: *p++ = ap[0] == '+' ? 's' : 'f';
1.13 deraadt 246: (void)strlcpy(p, ap + 1, len - 2);
1.1 deraadt 247: *argv = start;
248: }
249: }
250:
1.15 ray 251: __dead void
1.10 deraadt 252: usage(void)
1.1 deraadt 253: {
1.33 ! cheloha 254: fprintf(stderr,
1.25 tb 255: "usage: %s [-ci] [-d | -u] [-f fields] [-s chars] [input_file [output_file]]\n",
1.33 ! cheloha 256: getprogname());
1.1 deraadt 257: exit(1);
258: }