1 /* $NetBSD: uniq.c,v 1.7 1995/08/31 22:03:48 jtc Exp $ */ 2 3 /* 4 * Copyright (c) 1989, 1993 5 * The Regents of the University of California. All rights reserved. 6 * 7 * This code is derived from software contributed to Berkeley by 8 * Case Larsen. 9 * 10 * Redistribution and use in source and binary forms, with or without 11 * modification, are permitted provided that the following conditions 12 * are met: 13 * 1. Redistributions of source code must retain the above copyright 14 * notice, this list of conditions and the following disclaimer. 15 * 2. Redistributions in binary form must reproduce the above copyright 16 * notice, this list of conditions and the following disclaimer in the 17 * documentation and/or other materials provided with the distribution. 18 * 3. All advertising materials mentioning features or use of this software 19 * must display the following acknowledgement: 20 * This product includes software developed by the University of 21 * California, Berkeley and its contributors. 22 * 4. Neither the name of the University nor the names of its contributors 23 * may be used to endorse or promote products derived from this software 24 * without specific prior written permission. 25 * 26 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 27 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 28 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 29 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 30 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 31 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 32 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 33 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 34 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 35 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 36 * SUCH DAMAGE. 37 */ 38 39 #ifndef lint 40 static char copyright[] = 41 "@(#) Copyright (c) 1989, 1993\n\ 42 The Regents of the University of California. All rights reserved.\n"; 43 #endif /* not lint */ 44 45 #ifndef lint 46 #if 0 47 static char sccsid[] = "@(#)uniq.c 8.3 (Berkeley) 5/4/95"; 48 #endif 49 static char rcsid[] = "$NetBSD: uniq.c,v 1.7 1995/08/31 22:03:48 jtc Exp $"; 50 #endif /* not lint */ 51 52 #include <errno.h> 53 #include <stdio.h> 54 #include <ctype.h> 55 #include <stdlib.h> 56 #include <string.h> 57 #include <unistd.h> 58 59 #define MAXLINELEN (8 * 1024) 60 61 int cflag, dflag, uflag; 62 int numchars, numfields, repeats; 63 64 void err __P((const char *, ...)); 65 FILE *file __P((char *, char *)); 66 void show __P((FILE *, char *)); 67 char *skip __P((char *)); 68 void obsolete __P((char *[])); 69 void usage __P((void)); 70 71 int 72 main (argc, argv) 73 int argc; 74 char *argv[]; 75 { 76 register char *t1, *t2; 77 FILE *ifp, *ofp; 78 int ch; 79 char *prevline, *thisline, *p; 80 81 obsolete(argv); 82 while ((ch = getopt(argc, argv, "-cdf:s:u")) != EOF) 83 switch (ch) { 84 case '-': 85 --optind; 86 goto done; 87 case 'c': 88 cflag = 1; 89 break; 90 case 'd': 91 dflag = 1; 92 break; 93 case 'f': 94 numfields = strtol(optarg, &p, 10); 95 if (numfields < 0 || *p) 96 err("illegal field skip value: %s", optarg); 97 break; 98 case 's': 99 numchars = strtol(optarg, &p, 10); 100 if (numchars < 0 || *p) 101 err("illegal character skip value: %s", optarg); 102 break; 103 case 'u': 104 uflag = 1; 105 break; 106 case '?': 107 default: 108 usage(); 109 } 110 111 done: argc -= optind; 112 argv +=optind; 113 114 /* If no flags are set, default is -d -u. */ 115 if (cflag) { 116 if (dflag || uflag) 117 usage(); 118 } else if (!dflag && !uflag) 119 dflag = uflag = 1; 120 121 switch(argc) { 122 case 0: 123 ifp = stdin; 124 ofp = stdout; 125 break; 126 case 1: 127 ifp = file(argv[0], "r"); 128 ofp = stdout; 129 break; 130 case 2: 131 ifp = file(argv[0], "r"); 132 ofp = file(argv[1], "w"); 133 break; 134 default: 135 usage(); 136 } 137 138 prevline = malloc(MAXLINELEN); 139 thisline = malloc(MAXLINELEN); 140 if (prevline == NULL || thisline == NULL) 141 err("%s", strerror(errno)); 142 143 if (fgets(prevline, MAXLINELEN, ifp) == NULL) 144 exit(0); 145 146 while (fgets(thisline, MAXLINELEN, ifp)) { 147 /* If requested get the chosen fields + character offsets. */ 148 if (numfields || numchars) { 149 t1 = skip(thisline); 150 t2 = skip(prevline); 151 } else { 152 t1 = thisline; 153 t2 = prevline; 154 } 155 156 /* If different, print; set previous to new value. */ 157 if (strcmp(t1, t2)) { 158 show(ofp, prevline); 159 t1 = prevline; 160 prevline = thisline; 161 thisline = t1; 162 repeats = 0; 163 } else 164 ++repeats; 165 } 166 show(ofp, prevline); 167 exit(0); 168 } 169 170 /* 171 * show -- 172 * Output a line depending on the flags and number of repetitions 173 * of the line. 174 */ 175 void 176 show(ofp, str) 177 FILE *ofp; 178 char *str; 179 { 180 181 if (cflag && *str) 182 (void)fprintf(ofp, "%4d %s", repeats + 1, str); 183 if (dflag && repeats || uflag && !repeats) 184 (void)fprintf(ofp, "%s", str); 185 } 186 187 char * 188 skip(str) 189 register char *str; 190 { 191 register int infield, nchars, nfields; 192 193 for (nfields = numfields, infield = 0; nfields && *str; ++str) 194 if (isspace(*str)) { 195 if (infield) { 196 infield = 0; 197 --nfields; 198 } 199 } else if (!infield) 200 infield = 1; 201 for (nchars = numchars; nchars-- && *str; ++str); 202 return(str); 203 } 204 205 FILE * 206 file(name, mode) 207 char *name, *mode; 208 { 209 FILE *fp; 210 211 if ((fp = fopen(name, mode)) == NULL) 212 err("%s: %s", name, strerror(errno)); 213 return(fp); 214 } 215 216 void 217 obsolete(argv) 218 char *argv[]; 219 { 220 int len; 221 char *ap, *p, *start; 222 223 while (ap = *++argv) { 224 /* Return if "--" or not an option of any form. */ 225 if (ap[0] != '-') { 226 if (ap[0] != '+') 227 return; 228 } else if (ap[1] == '-') 229 return; 230 if (!isdigit(ap[1])) 231 continue; 232 /* 233 * Digit signifies an old-style option. Malloc space for dash, 234 * new option and argument. 235 */ 236 len = strlen(ap); 237 if ((start = p = malloc(len + 3)) == NULL) 238 err("%s", strerror(errno)); 239 *p++ = '-'; 240 *p++ = ap[0] == '+' ? 's' : 'f'; 241 (void)strcpy(p, ap + 1); 242 *argv = start; 243 } 244 } 245 246 void 247 usage() 248 { 249 (void)fprintf(stderr, 250 "usage: uniq [-c | -du] [-f fields] [-s chars] [input [output]]\n"); 251 exit(1); 252 } 253 254 #if __STDC__ 255 #include <stdarg.h> 256 #else 257 #include <varargs.h> 258 #endif 259 260 void 261 #if __STDC__ 262 err(const char *fmt, ...) 263 #else 264 err(fmt, va_alist) 265 char *fmt; 266 va_dcl 267 #endif 268 { 269 va_list ap; 270 #if __STDC__ 271 va_start(ap, fmt); 272 #else 273 va_start(ap); 274 #endif 275 (void)fprintf(stderr, "uniq: "); 276 (void)vfprintf(stderr, fmt, ap); 277 va_end(ap); 278 (void)fprintf(stderr, "\n"); 279 exit(1); 280 /* NOTREACHED */ 281 } 282