1 /* 2 * Copyright (c) 1989, 1993 3 * The Regents of the University of California. All rights reserved. 4 * 5 * This code is derived from software contributed to Berkeley by 6 * Case Larsen. 7 * 8 * Redistribution and use in source and binary forms, with or without 9 * modification, are permitted provided that the following conditions 10 * are met: 11 * 1. Redistributions of source code must retain the above copyright 12 * notice, this list of conditions and the following disclaimer. 13 * 2. Redistributions in binary form must reproduce the above copyright 14 * notice, this list of conditions and the following disclaimer in the 15 * documentation and/or other materials provided with the distribution. 16 * 3. All advertising materials mentioning features or use of this software 17 * must display the following acknowledgement: 18 * This product includes software developed by the University of 19 * California, Berkeley and its contributors. 20 * 4. Neither the name of the University nor the names of its contributors 21 * may be used to endorse or promote products derived from this software 22 * without specific prior written permission. 23 * 24 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 25 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 26 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 27 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 28 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 29 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 30 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 31 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 32 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 33 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 34 * SUCH DAMAGE. 35 */ 36 37 #ifndef lint 38 static const char copyright[] = 39 "@(#) Copyright (c) 1989, 1993\n\ 40 The Regents of the University of California. All rights reserved.\n"; 41 #endif /* not lint */ 42 43 #ifndef lint 44 #if 0 45 static char sccsid[] = "@(#)uniq.c 8.3 (Berkeley) 5/4/95"; 46 #endif 47 static const char rcsid[] = 48 "$FreeBSD$"; 49 #endif /* not lint */ 50 51 #include <ctype.h> 52 #include <err.h> 53 #include <limits.h> 54 #include <locale.h> 55 #include <stdio.h> 56 #include <stdlib.h> 57 #include <string.h> 58 #include <unistd.h> 59 60 #define MAXLINELEN (LINE_MAX + 1) 61 62 int cflag, dflag, uflag; 63 int numchars, numfields, repeats; 64 65 FILE *file(const char *, const char *); 66 char *getline(char *, size_t, FILE *); 67 void show(FILE *, char *); 68 char *skip(char *); 69 void obsolete(char *[]); 70 static void usage(void); 71 int stricoll(char *, char*); 72 73 int 74 main (argc, argv) 75 int argc; 76 char *argv[]; 77 { 78 register char *t1, *t2; 79 FILE *ifp, *ofp; 80 int ch; 81 char *prevline, *thisline, *p; 82 int iflag = 0, comp; 83 84 (void) setlocale(LC_ALL, ""); 85 86 obsolete(argv); 87 while ((ch = getopt(argc, argv, "cdif:s:u")) != -1) 88 switch (ch) { 89 case 'c': 90 cflag = 1; 91 break; 92 case 'd': 93 dflag = 1; 94 break; 95 case 'i': 96 iflag = 1; 97 break; 98 case 'f': 99 numfields = strtol(optarg, &p, 10); 100 if (numfields < 0 || *p) 101 errx(1, "illegal field skip value: %s", optarg); 102 break; 103 case 's': 104 numchars = strtol(optarg, &p, 10); 105 if (numchars < 0 || *p) 106 errx(1, "illegal character skip value: %s", optarg); 107 break; 108 case 'u': 109 uflag = 1; 110 break; 111 case '?': 112 default: 113 usage(); 114 } 115 116 argc -= optind; 117 argv +=optind; 118 119 /* If no flags are set, default is -d -u. */ 120 if (cflag) { 121 if (dflag || uflag) 122 usage(); 123 } else if (!dflag && !uflag) 124 dflag = uflag = 1; 125 126 if (argc > 2) 127 usage(); 128 129 ifp = stdin; 130 ofp = stdout; 131 if (argc > 0 && strcmp(argv[0], "-") != 0) 132 ifp = file(argv[0], "r"); 133 if (argc > 1) 134 ofp = file(argv[1], "w"); 135 136 prevline = malloc(MAXLINELEN); 137 thisline = malloc(MAXLINELEN); 138 if (prevline == NULL || thisline == NULL) 139 err(1, "malloc"); 140 141 if (getline(prevline, MAXLINELEN, ifp) == NULL) 142 exit(0); 143 if (!cflag && uflag && dflag) 144 show(ofp, prevline); 145 146 while (getline(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 (iflag) 158 comp = stricoll(t1, t2); 159 else 160 comp = strcoll(t1, t2); 161 162 if (comp) { 163 if (cflag || !dflag || !uflag) 164 show(ofp, prevline); 165 t1 = prevline; 166 prevline = thisline; 167 if (!cflag && uflag && dflag) 168 show(ofp, prevline); 169 thisline = t1; 170 repeats = 0; 171 } else 172 ++repeats; 173 } 174 if (cflag || !dflag || !uflag) 175 show(ofp, prevline); 176 exit(0); 177 } 178 179 char * 180 getline(char *buf, size_t buflen, FILE *fp) 181 { 182 size_t bufpos; 183 int ch; 184 185 bufpos = 0; 186 while (bufpos + 2 != buflen && (ch = getc(fp)) != EOF && ch != '\n') 187 buf[bufpos++] = ch; 188 if (bufpos + 1 != buflen) 189 buf[bufpos] = '\0'; 190 while (ch != EOF && ch != '\n') 191 ch = getc(fp); 192 193 return (bufpos != 0 || ch == '\n' ? buf : NULL); 194 } 195 196 /* 197 * show -- 198 * Output a line depending on the flags and number of repetitions 199 * of the line. 200 */ 201 void 202 show(ofp, str) 203 FILE *ofp; 204 char *str; 205 { 206 207 if (cflag && *str) 208 (void)fprintf(ofp, "%4d %s\n", repeats + 1, str); 209 if ((dflag && repeats) || (uflag && !repeats)) 210 (void)fprintf(ofp, "%s\n", str); 211 } 212 213 char * 214 skip(str) 215 register char *str; 216 { 217 register int nchars, nfields; 218 219 for (nfields = 0; *str != '\0' && nfields++ != numfields; ) { 220 while (isblank((unsigned char)*str)) 221 str++; 222 while (*str != '\0' && !isblank((unsigned char)*str)) 223 str++; 224 } 225 for (nchars = numchars; nchars-- && *str; ++str); 226 return(str); 227 } 228 229 FILE * 230 file(name, mode) 231 const char *name, *mode; 232 { 233 FILE *fp; 234 235 if ((fp = fopen(name, mode)) == NULL) 236 err(1, "%s", name); 237 return(fp); 238 } 239 240 void 241 obsolete(argv) 242 char *argv[]; 243 { 244 int len; 245 char *ap, *p, *start; 246 247 while ((ap = *++argv)) { 248 /* Return if "--" or not an option of any form. */ 249 if (ap[0] != '-') { 250 if (ap[0] != '+') 251 return; 252 } else if (ap[1] == '-') 253 return; 254 if (!isdigit((unsigned char)ap[1])) 255 continue; 256 /* 257 * Digit signifies an old-style option. Malloc space for dash, 258 * new option and argument. 259 */ 260 len = strlen(ap); 261 if ((start = p = malloc(len + 3)) == NULL) 262 err(1, "malloc"); 263 *p++ = '-'; 264 *p++ = ap[0] == '+' ? 's' : 'f'; 265 (void)strcpy(p, ap + 1); 266 *argv = start; 267 } 268 } 269 270 static void 271 usage() 272 { 273 (void)fprintf(stderr, 274 "usage: uniq [-c | -d | -u] [-i] [-f fields] [-s chars] [input [output]]\n"); 275 exit(1); 276 } 277 278 int 279 stricoll(s1, s2) 280 char *s1, *s2; 281 { 282 char *p, line1[MAXLINELEN], line2[MAXLINELEN]; 283 284 for (p = line1; *s1; s1++) 285 *p++ = tolower((unsigned char)*s1); 286 *p = '\0'; 287 for (p = line2; *s2; s2++) 288 *p++ = tolower((unsigned char)*s2); 289 *p = '\0'; 290 return strcoll(line1, line2); 291 } 292