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 errx(1, "malloc"); 140 141 if (getline(prevline, MAXLINELEN, ifp) == NULL) 142 exit(0); 143 144 while (getline(thisline, MAXLINELEN, ifp)) { 145 /* If requested get the chosen fields + character offsets. */ 146 if (numfields || numchars) { 147 t1 = skip(thisline); 148 t2 = skip(prevline); 149 } else { 150 t1 = thisline; 151 t2 = prevline; 152 } 153 154 /* If different, print; set previous to new value. */ 155 if (iflag) 156 comp = stricoll(t1, t2); 157 else 158 comp = strcoll(t1, t2); 159 160 if (comp) { 161 show(ofp, prevline); 162 t1 = prevline; 163 prevline = thisline; 164 thisline = t1; 165 repeats = 0; 166 } else 167 ++repeats; 168 } 169 show(ofp, prevline); 170 exit(0); 171 } 172 173 char * 174 getline(char *buf, size_t buflen, FILE *fp) 175 { 176 size_t bufpos; 177 int ch; 178 179 bufpos = 0; 180 while (bufpos + 2 != buflen && (ch = getc(fp)) != EOF && ch != '\n') 181 buf[bufpos++] = ch; 182 if (bufpos + 1 != buflen) 183 buf[bufpos] = '\0'; 184 while (ch != EOF && ch != '\n') 185 ch = getc(fp); 186 187 return (bufpos != 0 || ch == '\n' ? buf : NULL); 188 } 189 190 /* 191 * show -- 192 * Output a line depending on the flags and number of repetitions 193 * of the line. 194 */ 195 void 196 show(ofp, str) 197 FILE *ofp; 198 char *str; 199 { 200 201 if (cflag && *str) 202 (void)fprintf(ofp, "%4d %s\n", repeats + 1, str); 203 if ((dflag && repeats) || (uflag && !repeats)) 204 (void)fprintf(ofp, "%s\n", str); 205 } 206 207 char * 208 skip(str) 209 register char *str; 210 { 211 register int nchars, nfields; 212 213 for (nfields = 0; *str != '\0' && nfields++ != numfields; ) { 214 while (isblank((unsigned char)*str)) 215 str++; 216 while (*str != '\0' && !isblank((unsigned char)*str)) 217 str++; 218 } 219 for (nchars = numchars; nchars-- && *str; ++str); 220 return(str); 221 } 222 223 FILE * 224 file(name, mode) 225 const char *name, *mode; 226 { 227 FILE *fp; 228 229 if ((fp = fopen(name, mode)) == NULL) 230 err(1, "%s", name); 231 return(fp); 232 } 233 234 void 235 obsolete(argv) 236 char *argv[]; 237 { 238 int len; 239 char *ap, *p, *start; 240 241 while ((ap = *++argv)) { 242 /* Return if "--" or not an option of any form. */ 243 if (ap[0] != '-') { 244 if (ap[0] != '+') 245 return; 246 } else if (ap[1] == '-') 247 return; 248 if (!isdigit((unsigned char)ap[1])) 249 continue; 250 /* 251 * Digit signifies an old-style option. Malloc space for dash, 252 * new option and argument. 253 */ 254 len = strlen(ap); 255 if ((start = p = malloc(len + 3)) == NULL) 256 errx(1, "malloc"); 257 *p++ = '-'; 258 *p++ = ap[0] == '+' ? 's' : 'f'; 259 (void)strcpy(p, ap + 1); 260 *argv = start; 261 } 262 } 263 264 static void 265 usage() 266 { 267 (void)fprintf(stderr, 268 "usage: uniq [-c | -d | -u] [-i] [-f fields] [-s chars] [input [output]]\n"); 269 exit(1); 270 } 271 272 int 273 stricoll(s1, s2) 274 char *s1, *s2; 275 { 276 char *p, line1[MAXLINELEN], line2[MAXLINELEN]; 277 278 for (p = line1; *s1; s1++) 279 *p++ = tolower((unsigned char)*s1); 280 *p = '\0'; 281 for (p = line2; *s2; s2++) 282 *p++ = tolower((unsigned char)*s2); 283 *p = '\0'; 284 return strcoll(line1, line2); 285 } 286