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 <stdint.h> 56 #include <stdio.h> 57 #include <stdlib.h> 58 #include <string.h> 59 #include <unistd.h> 60 #include <wchar.h> 61 #include <wctype.h> 62 63 #define INITLINELEN (LINE_MAX + 1) 64 #define MAXLINELEN ((SIZE_MAX / sizeof(wchar_t)) / 2) 65 66 int cflag, dflag, uflag; 67 int numchars, numfields, repeats; 68 69 FILE *file(const char *, const char *); 70 wchar_t *getline(wchar_t *, size_t *, FILE *); 71 void show(FILE *, wchar_t *); 72 wchar_t *skip(wchar_t *); 73 void obsolete(char *[]); 74 static void usage(void); 75 int wcsicoll(wchar_t *, wchar_t *); 76 77 int 78 main (int argc, char *argv[]) 79 { 80 wchar_t *t1, *t2; 81 FILE *ifp, *ofp; 82 int ch, b1; 83 size_t prevbuflen, thisbuflen; 84 wchar_t *prevline, *thisline; 85 char *p; 86 const char *ifn; 87 int iflag = 0, comp; 88 89 (void) setlocale(LC_ALL, ""); 90 91 obsolete(argv); 92 while ((ch = getopt(argc, argv, "cdif:s:u")) != -1) 93 switch (ch) { 94 case 'c': 95 cflag = 1; 96 break; 97 case 'd': 98 dflag = 1; 99 break; 100 case 'i': 101 iflag = 1; 102 break; 103 case 'f': 104 numfields = strtol(optarg, &p, 10); 105 if (numfields < 0 || *p) 106 errx(1, "illegal field skip value: %s", optarg); 107 break; 108 case 's': 109 numchars = strtol(optarg, &p, 10); 110 if (numchars < 0 || *p) 111 errx(1, "illegal character skip value: %s", optarg); 112 break; 113 case 'u': 114 uflag = 1; 115 break; 116 case '?': 117 default: 118 usage(); 119 } 120 121 argc -= optind; 122 argv += optind; 123 124 /* If no flags are set, default is -d -u. */ 125 if (cflag) { 126 if (dflag || uflag) 127 usage(); 128 } else if (!dflag && !uflag) 129 dflag = uflag = 1; 130 131 if (argc > 2) 132 usage(); 133 134 ifp = stdin; 135 ifn = "stdin"; 136 ofp = stdout; 137 if (argc > 0 && strcmp(argv[0], "-") != 0) 138 ifp = file(ifn = argv[0], "r"); 139 if (argc > 1) 140 ofp = file(argv[1], "w"); 141 142 prevbuflen = INITLINELEN; 143 thisbuflen = INITLINELEN; 144 prevline = malloc(prevbuflen * sizeof(*prevline)); 145 thisline = malloc(thisbuflen * sizeof(*thisline)); 146 if (prevline == NULL || thisline == NULL) 147 err(1, "malloc"); 148 149 if ((prevline = getline(prevline, &prevbuflen, ifp)) == NULL) { 150 if (ferror(ifp)) 151 err(1, "%s", ifn); 152 exit(0); 153 } 154 if (!cflag && uflag && dflag) 155 show(ofp, prevline); 156 157 while ((thisline = getline(thisline, &thisbuflen, ifp)) != NULL) { 158 /* If requested get the chosen fields + character offsets. */ 159 if (numfields || numchars) { 160 t1 = skip(thisline); 161 t2 = skip(prevline); 162 } else { 163 t1 = thisline; 164 t2 = prevline; 165 } 166 167 /* If different, print; set previous to new value. */ 168 if (iflag) 169 comp = wcsicoll(t1, t2); 170 else 171 comp = wcscoll(t1, t2); 172 173 if (comp) { 174 if (cflag || !dflag || !uflag) 175 show(ofp, prevline); 176 t1 = prevline; 177 b1 = prevbuflen; 178 prevline = thisline; 179 prevbuflen = thisbuflen; 180 if (!cflag && uflag && dflag) 181 show(ofp, prevline); 182 thisline = t1; 183 thisbuflen = b1; 184 repeats = 0; 185 } else 186 ++repeats; 187 } 188 if (ferror(ifp)) 189 err(1, "%s", ifn); 190 if (cflag || !dflag || !uflag) 191 show(ofp, prevline); 192 exit(0); 193 } 194 195 wchar_t * 196 getline(wchar_t *buf, size_t *buflen, FILE *fp) 197 { 198 size_t bufpos; 199 wint_t ch; 200 201 bufpos = 0; 202 while ((ch = getwc(fp)) != WEOF && ch != '\n') { 203 if (bufpos + 1 >= *buflen) { 204 *buflen = *buflen * 2; 205 if (*buflen > MAXLINELEN) 206 errx(1, 207 "Maximum line buffer length (%zu) exceeded", 208 MAXLINELEN); 209 buf = reallocf(buf, *buflen * sizeof(*buf)); 210 if (buf == NULL) 211 err(1, "reallocf"); 212 } 213 buf[bufpos++] = ch; 214 } 215 buf[bufpos] = '\0'; 216 217 return (bufpos != 0 || ch == '\n' ? buf : NULL); 218 } 219 220 /* 221 * show -- 222 * Output a line depending on the flags and number of repetitions 223 * of the line. 224 */ 225 void 226 show(FILE *ofp, wchar_t *str) 227 { 228 229 if (cflag) 230 (void)fprintf(ofp, "%4d %ls\n", repeats + 1, str); 231 if ((dflag && repeats) || (uflag && !repeats)) 232 (void)fprintf(ofp, "%ls\n", str); 233 } 234 235 wchar_t * 236 skip(wchar_t *str) 237 { 238 int nchars, nfields; 239 240 for (nfields = 0; *str != '\0' && nfields++ != numfields; ) { 241 while (iswblank(*str)) 242 str++; 243 while (*str != '\0' && !iswblank(*str)) 244 str++; 245 } 246 for (nchars = numchars; nchars-- && *str; ++str); 247 return(str); 248 } 249 250 FILE * 251 file(const char *name, const char *mode) 252 { 253 FILE *fp; 254 255 if ((fp = fopen(name, mode)) == NULL) 256 err(1, "%s", name); 257 return(fp); 258 } 259 260 void 261 obsolete(char *argv[]) 262 { 263 int len; 264 char *ap, *p, *start; 265 266 while ((ap = *++argv)) { 267 /* Return if "--" or not an option of any form. */ 268 if (ap[0] != '-') { 269 if (ap[0] != '+') 270 return; 271 } else if (ap[1] == '-') 272 return; 273 if (!isdigit((unsigned char)ap[1])) 274 continue; 275 /* 276 * Digit signifies an old-style option. Malloc space for dash, 277 * new option and argument. 278 */ 279 len = strlen(ap); 280 if ((start = p = malloc(len + 3)) == NULL) 281 err(1, "malloc"); 282 *p++ = '-'; 283 *p++ = ap[0] == '+' ? 's' : 'f'; 284 (void)strcpy(p, ap + 1); 285 *argv = start; 286 } 287 } 288 289 static void 290 usage(void) 291 { 292 (void)fprintf(stderr, 293 "usage: uniq [-c | -d | -u] [-i] [-f fields] [-s chars] [input [output]]\n"); 294 exit(1); 295 } 296 297 static size_t wcsicoll_l1_buflen = 0, wcsicoll_l2_buflen = 0; 298 static wchar_t *wcsicoll_l1_buf = NULL, *wcsicoll_l2_buf = NULL; 299 300 int 301 wcsicoll(wchar_t *s1, wchar_t *s2) 302 { 303 wchar_t *p; 304 size_t l1, l2; 305 size_t new_l1_buflen, new_l2_buflen; 306 307 l1 = wcslen(s1) + 1; 308 l2 = wcslen(s2) + 1; 309 new_l1_buflen = wcsicoll_l1_buflen; 310 new_l2_buflen = wcsicoll_l2_buflen; 311 while (new_l1_buflen < l1) { 312 if (new_l1_buflen == 0) 313 new_l1_buflen = INITLINELEN; 314 else 315 new_l1_buflen *= 2; 316 } 317 while (new_l2_buflen < l2) { 318 if (new_l2_buflen == 0) 319 new_l2_buflen = INITLINELEN; 320 else 321 new_l2_buflen *= 2; 322 } 323 if (new_l1_buflen > wcsicoll_l1_buflen) { 324 wcsicoll_l1_buf = reallocf(wcsicoll_l1_buf, new_l1_buflen * sizeof(*wcsicoll_l1_buf)); 325 if (wcsicoll_l1_buf == NULL) 326 err(1, "reallocf"); 327 wcsicoll_l1_buflen = new_l1_buflen; 328 } 329 if (new_l2_buflen > wcsicoll_l2_buflen) { 330 wcsicoll_l2_buf = reallocf(wcsicoll_l2_buf, new_l2_buflen * sizeof(*wcsicoll_l2_buf)); 331 if (wcsicoll_l2_buf == NULL) 332 err(1, "reallocf"); 333 wcsicoll_l2_buflen = new_l2_buflen; 334 } 335 336 for (p = wcsicoll_l1_buf; *s1; s1++) 337 *p++ = towlower(*s1); 338 *p = '\0'; 339 for (p = wcsicoll_l2_buf; *s2; s2++) 340 *p++ = towlower(*s2); 341 *p = '\0'; 342 343 return (wcscoll(wcsicoll_l1_buf, wcsicoll_l2_buf)); 344 } 345