1 /*- 2 * SPDX-License-Identifier: BSD-3-Clause 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. Neither the name of the University nor the names of its contributors 19 * may be used to endorse or promote products derived from this software 20 * without specific prior written permission. 21 * 22 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 23 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 24 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 25 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 26 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 27 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 28 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 29 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 30 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 31 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 32 * SUCH DAMAGE. 33 */ 34 35 #include <sys/capsicum.h> 36 37 #include <capsicum_helpers.h> 38 #include <ctype.h> 39 #include <err.h> 40 #include <errno.h> 41 #include <getopt.h> 42 #include <limits.h> 43 #include <locale.h> 44 #include <nl_types.h> 45 #include <stdint.h> 46 #include <stdio.h> 47 #include <stdlib.h> 48 #include <string.h> 49 #include <termios.h> 50 #include <unistd.h> 51 #include <wchar.h> 52 #include <wctype.h> 53 54 static int Dflag, cflag, dflag, uflag, iflag; 55 static int numchars, numfields, repeats; 56 57 /* Dflag values */ 58 #define DF_NONE 0 59 #define DF_NOSEP 1 60 #define DF_PRESEP 2 61 #define DF_POSTSEP 3 62 63 static const struct option long_opts[] = 64 { 65 {"all-repeated",optional_argument, NULL, 'D'}, 66 {"count", no_argument, NULL, 'c'}, 67 {"repeated", no_argument, NULL, 'd'}, 68 {"skip-fields", required_argument, NULL, 'f'}, 69 {"ignore-case", no_argument, NULL, 'i'}, 70 {"skip-chars", required_argument, NULL, 's'}, 71 {"unique", no_argument, NULL, 'u'}, 72 {NULL, no_argument, NULL, 0} 73 }; 74 75 static FILE *file(const char *, const char *); 76 static wchar_t *convert(const char *); 77 static int inlcmp(const char *, const char *); 78 static void show(FILE *, const char *); 79 static wchar_t *skip(wchar_t *); 80 static void obsolete(char *[]); 81 static void usage(void); 82 83 int 84 main (int argc, char *argv[]) 85 { 86 wchar_t *tprev, *tthis; 87 FILE *ifp, *ofp; 88 int ch, comp; 89 size_t prevbuflen, thisbuflen, b1; 90 char *prevline, *thisline, *p; 91 const char *ifn, *errstr;; 92 cap_rights_t rights; 93 94 (void) setlocale(LC_ALL, ""); 95 96 obsolete(argv); 97 while ((ch = getopt_long(argc, argv, "+D::cdif:s:u", long_opts, 98 NULL)) != -1) 99 switch (ch) { 100 case 'D': 101 if (optarg == NULL || strcasecmp(optarg, "none") == 0) 102 Dflag = DF_NOSEP; 103 else if (strcasecmp(optarg, "prepend") == 0) 104 Dflag = DF_PRESEP; 105 else if (strcasecmp(optarg, "separate") == 0) 106 Dflag = DF_POSTSEP; 107 else 108 usage(); 109 break; 110 case 'c': 111 cflag = 1; 112 break; 113 case 'd': 114 dflag = 1; 115 break; 116 case 'i': 117 iflag = 1; 118 break; 119 case 'f': 120 numfields = strtonum(optarg, 0, INT_MAX, &errstr); 121 if (errstr) 122 errx(1, "field skip value is %s: %s", errstr, optarg); 123 break; 124 case 's': 125 numchars = strtonum(optarg, 0, INT_MAX, &errstr); 126 if (errstr != NULL) 127 errx(1, "character skip value is %s: %s", errstr, optarg); 128 break; 129 case 'u': 130 uflag = 1; 131 break; 132 case '?': 133 default: 134 usage(); 135 } 136 137 argc -= optind; 138 argv += optind; 139 140 if (argc > 2) 141 usage(); 142 143 ifp = stdin; 144 ifn = "stdin"; 145 ofp = stdout; 146 if (argc > 0 && strcmp(argv[0], "-") != 0) 147 ifp = file(ifn = argv[0], "r"); 148 cap_rights_init(&rights, CAP_FSTAT, CAP_READ); 149 if (caph_rights_limit(fileno(ifp), &rights) < 0) 150 err(1, "unable to limit rights for %s", ifn); 151 cap_rights_init(&rights, CAP_FSTAT, CAP_WRITE); 152 if (argc > 1) 153 ofp = file(argv[1], "w"); 154 else 155 cap_rights_set(&rights, CAP_IOCTL); 156 if (caph_rights_limit(fileno(ofp), &rights) < 0) { 157 err(1, "unable to limit rights for %s", 158 argc > 1 ? argv[1] : "stdout"); 159 } 160 if (cap_rights_is_set(&rights, CAP_IOCTL)) { 161 unsigned long cmd; 162 163 cmd = TIOCGETA; /* required by isatty(3) in printf(3) */ 164 165 if (caph_ioctls_limit(fileno(ofp), &cmd, 1) < 0) { 166 err(1, "unable to limit ioctls for %s", 167 argc > 1 ? argv[1] : "stdout"); 168 } 169 } 170 171 caph_cache_catpages(); 172 if (caph_enter() < 0) 173 err(1, "unable to enter capability mode"); 174 175 prevbuflen = thisbuflen = 0; 176 prevline = thisline = NULL; 177 178 if (getline(&prevline, &prevbuflen, ifp) < 0) { 179 if (ferror(ifp)) 180 err(1, "%s", ifn); 181 exit(0); 182 } 183 tprev = convert(prevline); 184 185 tthis = NULL; 186 while (getline(&thisline, &thisbuflen, ifp) >= 0) { 187 if (tthis != NULL) 188 free(tthis); 189 tthis = convert(thisline); 190 191 if (tthis == NULL && tprev == NULL) 192 comp = inlcmp(thisline, prevline); 193 else if (tthis == NULL || tprev == NULL) 194 comp = 1; 195 else 196 comp = wcscoll(tthis, tprev); 197 198 if (comp) { 199 /* If different, print; set previous to new value. */ 200 if (Dflag == DF_POSTSEP && repeats > 0) 201 fputc('\n', ofp); 202 if (!Dflag) 203 show(ofp, prevline); 204 p = prevline; 205 b1 = prevbuflen; 206 prevline = thisline; 207 prevbuflen = thisbuflen; 208 if (tprev != NULL) 209 free(tprev); 210 tprev = tthis; 211 thisline = p; 212 thisbuflen = b1; 213 tthis = NULL; 214 repeats = 0; 215 } else { 216 if (Dflag) { 217 if (repeats == 0) { 218 if (Dflag == DF_PRESEP) 219 fputc('\n', ofp); 220 show(ofp, prevline); 221 } 222 show(ofp, thisline); 223 } 224 ++repeats; 225 } 226 } 227 if (ferror(ifp)) 228 err(1, "%s", ifn); 229 if (!Dflag) 230 show(ofp, prevline); 231 exit(0); 232 } 233 234 static wchar_t * 235 convert(const char *str) 236 { 237 size_t n; 238 wchar_t *buf, *ret, *p; 239 240 if ((n = mbstowcs(NULL, str, 0)) == (size_t)-1) 241 return (NULL); 242 if (SIZE_MAX / sizeof(*buf) < n + 1) 243 errx(1, "conversion buffer length overflow"); 244 if ((buf = malloc((n + 1) * sizeof(*buf))) == NULL) 245 err(1, "malloc"); 246 if (mbstowcs(buf, str, n + 1) != n) 247 errx(1, "internal mbstowcs() error"); 248 /* The last line may not end with \n. */ 249 if (n > 0 && buf[n - 1] == L'\n') 250 buf[n - 1] = L'\0'; 251 252 /* If requested get the chosen fields + character offsets. */ 253 if (numfields || numchars) { 254 if ((ret = wcsdup(skip(buf))) == NULL) 255 err(1, "wcsdup"); 256 free(buf); 257 } else 258 ret = buf; 259 260 if (iflag) { 261 for (p = ret; *p != L'\0'; p++) 262 *p = towlower(*p); 263 } 264 265 return (ret); 266 } 267 268 static int 269 inlcmp(const char *s1, const char *s2) 270 { 271 int c1, c2; 272 273 while (*s1 == *s2++) 274 if (*s1++ == '\0') 275 return (0); 276 c1 = (unsigned char)*s1; 277 c2 = (unsigned char)*(s2 - 1); 278 /* The last line may not end with \n. */ 279 if (c1 == '\n') 280 c1 = '\0'; 281 if (c2 == '\n') 282 c2 = '\0'; 283 return (c1 - c2); 284 } 285 286 /* 287 * show -- 288 * Output a line depending on the flags and number of repetitions 289 * of the line. 290 */ 291 static void 292 show(FILE *ofp, const char *str) 293 { 294 295 if ((!Dflag && dflag && repeats == 0) || (uflag && repeats > 0)) 296 return; 297 if (cflag) 298 (void)fprintf(ofp, "%4d %s", repeats + 1, str); 299 else 300 (void)fprintf(ofp, "%s", str); 301 } 302 303 static wchar_t * 304 skip(wchar_t *str) 305 { 306 int nchars, nfields; 307 308 for (nfields = 0; *str != L'\0' && nfields++ != numfields; ) { 309 while (iswblank(*str)) 310 str++; 311 while (*str != L'\0' && !iswblank(*str)) 312 str++; 313 } 314 for (nchars = numchars; nchars-- && *str != L'\0'; ++str) 315 ; 316 return(str); 317 } 318 319 static FILE * 320 file(const char *name, const char *mode) 321 { 322 FILE *fp; 323 324 if ((fp = fopen(name, mode)) == NULL) 325 err(1, "%s", name); 326 return(fp); 327 } 328 329 static void 330 obsolete(char *argv[]) 331 { 332 int len; 333 char *ap, *p, *start; 334 335 while ((ap = *++argv)) { 336 /* Return if "--" or not an option of any form. */ 337 if (ap[0] != '-') { 338 if (ap[0] != '+') 339 return; 340 } else if (ap[1] == '-') 341 return; 342 if (!isdigit((unsigned char)ap[1])) 343 continue; 344 /* 345 * Digit signifies an old-style option. Malloc space for dash, 346 * new option and argument. 347 */ 348 len = strlen(ap); 349 if ((start = p = malloc(len + 3)) == NULL) 350 err(1, "malloc"); 351 *p++ = '-'; 352 *p++ = ap[0] == '+' ? 's' : 'f'; 353 (void)strcpy(p, ap + 1); 354 *argv = start; 355 } 356 } 357 358 static void 359 usage(void) 360 { 361 (void)fprintf(stderr, 362 "usage: uniq [-c | -d | -D | -u] [-i] [-f fields] [-s chars] [input [output]]\n"); 363 exit(1); 364 } 365