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 #ifndef lint 36 static const char copyright[] = 37 "@(#) Copyright (c) 1989, 1993\n\ 38 The Regents of the University of California. All rights reserved.\n"; 39 #endif /* not lint */ 40 41 #ifndef lint 42 #if 0 43 static char sccsid[] = "@(#)uniq.c 8.3 (Berkeley) 5/4/95"; 44 #endif 45 static const char rcsid[] = 46 "$FreeBSD$"; 47 #endif /* not lint */ 48 49 #include <sys/capsicum.h> 50 51 #include <ctype.h> 52 #include <err.h> 53 #include <errno.h> 54 #include <limits.h> 55 #include <locale.h> 56 #include <nl_types.h> 57 #include <stdint.h> 58 #include <stdio.h> 59 #include <stdlib.h> 60 #include <string.h> 61 #include <termios.h> 62 #include <unistd.h> 63 #include <wchar.h> 64 #include <wctype.h> 65 66 static int cflag, dflag, uflag, iflag; 67 static int numchars, numfields, repeats; 68 69 static FILE *file(const char *, const char *); 70 static wchar_t *convert(const char *); 71 static int inlcmp(const char *, const char *); 72 static void show(FILE *, const char *); 73 static wchar_t *skip(wchar_t *); 74 static void obsolete(char *[]); 75 static void usage(void); 76 77 static void 78 strerror_init(void) 79 { 80 81 /* 82 * Cache NLS data before entering capability mode. 83 * XXXPJD: There should be strerror_init() and strsignal_init() in libc. 84 */ 85 (void)catopen("libc", NL_CAT_LOCALE); 86 } 87 88 int 89 main (int argc, char *argv[]) 90 { 91 wchar_t *tprev, *tthis; 92 FILE *ifp, *ofp; 93 int ch, comp; 94 size_t prevbuflen, thisbuflen, b1; 95 char *prevline, *thisline, *p; 96 const char *ifn; 97 cap_rights_t rights; 98 99 (void) setlocale(LC_ALL, ""); 100 101 obsolete(argv); 102 while ((ch = getopt(argc, argv, "cdif:s:u")) != -1) 103 switch (ch) { 104 case 'c': 105 cflag = 1; 106 break; 107 case 'd': 108 dflag = 1; 109 break; 110 case 'i': 111 iflag = 1; 112 break; 113 case 'f': 114 numfields = strtol(optarg, &p, 10); 115 if (numfields < 0 || *p) 116 errx(1, "illegal field skip value: %s", optarg); 117 break; 118 case 's': 119 numchars = strtol(optarg, &p, 10); 120 if (numchars < 0 || *p) 121 errx(1, "illegal character skip value: %s", optarg); 122 break; 123 case 'u': 124 uflag = 1; 125 break; 126 case '?': 127 default: 128 usage(); 129 } 130 131 argc -= optind; 132 argv += optind; 133 134 if (argc > 2) 135 usage(); 136 137 ifp = stdin; 138 ifn = "stdin"; 139 ofp = stdout; 140 if (argc > 0 && strcmp(argv[0], "-") != 0) 141 ifp = file(ifn = argv[0], "r"); 142 cap_rights_init(&rights, CAP_FSTAT, CAP_READ); 143 if (cap_rights_limit(fileno(ifp), &rights) < 0 && errno != ENOSYS) 144 err(1, "unable to limit rights for %s", ifn); 145 cap_rights_init(&rights, CAP_FSTAT, CAP_WRITE); 146 if (argc > 1) 147 ofp = file(argv[1], "w"); 148 else 149 cap_rights_set(&rights, CAP_IOCTL); 150 if (cap_rights_limit(fileno(ofp), &rights) < 0 && errno != ENOSYS) { 151 err(1, "unable to limit rights for %s", 152 argc > 1 ? argv[1] : "stdout"); 153 } 154 if (cap_rights_is_set(&rights, CAP_IOCTL)) { 155 unsigned long cmd; 156 157 cmd = TIOCGETA; /* required by isatty(3) in printf(3) */ 158 159 if (cap_ioctls_limit(fileno(ofp), &cmd, 1) < 0 && 160 errno != ENOSYS) { 161 err(1, "unable to limit ioctls for %s", 162 argc > 1 ? argv[1] : "stdout"); 163 } 164 } 165 166 strerror_init(); 167 if (cap_enter() < 0 && errno != ENOSYS) 168 err(1, "unable to enter capability mode"); 169 170 prevbuflen = thisbuflen = 0; 171 prevline = thisline = NULL; 172 173 if (getline(&prevline, &prevbuflen, ifp) < 0) { 174 if (ferror(ifp)) 175 err(1, "%s", ifn); 176 exit(0); 177 } 178 tprev = convert(prevline); 179 180 tthis = NULL; 181 while (getline(&thisline, &thisbuflen, ifp) >= 0) { 182 if (tthis != NULL) 183 free(tthis); 184 tthis = convert(thisline); 185 186 if (tthis == NULL && tprev == NULL) 187 comp = inlcmp(thisline, prevline); 188 else if (tthis == NULL || tprev == NULL) 189 comp = 1; 190 else 191 comp = wcscoll(tthis, tprev); 192 193 if (comp) { 194 /* If different, print; set previous to new value. */ 195 show(ofp, prevline); 196 p = prevline; 197 b1 = prevbuflen; 198 prevline = thisline; 199 prevbuflen = thisbuflen; 200 if (tprev != NULL) 201 free(tprev); 202 tprev = tthis; 203 thisline = p; 204 thisbuflen = b1; 205 tthis = NULL; 206 repeats = 0; 207 } else 208 ++repeats; 209 } 210 if (ferror(ifp)) 211 err(1, "%s", ifn); 212 show(ofp, prevline); 213 exit(0); 214 } 215 216 static wchar_t * 217 convert(const char *str) 218 { 219 size_t n; 220 wchar_t *buf, *ret, *p; 221 222 if ((n = mbstowcs(NULL, str, 0)) == (size_t)-1) 223 return (NULL); 224 if (SIZE_MAX / sizeof(*buf) < n + 1) 225 errx(1, "conversion buffer length overflow"); 226 if ((buf = malloc((n + 1) * sizeof(*buf))) == NULL) 227 err(1, "malloc"); 228 if (mbstowcs(buf, str, n + 1) != n) 229 errx(1, "internal mbstowcs() error"); 230 /* The last line may not end with \n. */ 231 if (n > 0 && buf[n - 1] == L'\n') 232 buf[n - 1] = L'\0'; 233 234 /* If requested get the chosen fields + character offsets. */ 235 if (numfields || numchars) { 236 if ((ret = wcsdup(skip(buf))) == NULL) 237 err(1, "wcsdup"); 238 free(buf); 239 } else 240 ret = buf; 241 242 if (iflag) { 243 for (p = ret; *p != L'\0'; p++) 244 *p = towlower(*p); 245 } 246 247 return (ret); 248 } 249 250 static int 251 inlcmp(const char *s1, const char *s2) 252 { 253 int c1, c2; 254 255 while (*s1 == *s2++) 256 if (*s1++ == '\0') 257 return (0); 258 c1 = (unsigned char)*s1; 259 c2 = (unsigned char)*(s2 - 1); 260 /* The last line may not end with \n. */ 261 if (c1 == '\n') 262 c1 = '\0'; 263 if (c2 == '\n') 264 c2 = '\0'; 265 return (c1 - c2); 266 } 267 268 /* 269 * show -- 270 * Output a line depending on the flags and number of repetitions 271 * of the line. 272 */ 273 static void 274 show(FILE *ofp, const char *str) 275 { 276 277 if ((dflag && repeats == 0) || (uflag && repeats > 0)) 278 return; 279 if (cflag) 280 (void)fprintf(ofp, "%4d %s", repeats + 1, str); 281 else 282 (void)fprintf(ofp, "%s", str); 283 } 284 285 static wchar_t * 286 skip(wchar_t *str) 287 { 288 int nchars, nfields; 289 290 for (nfields = 0; *str != L'\0' && nfields++ != numfields; ) { 291 while (iswblank(*str)) 292 str++; 293 while (*str != L'\0' && !iswblank(*str)) 294 str++; 295 } 296 for (nchars = numchars; nchars-- && *str != L'\0'; ++str) 297 ; 298 return(str); 299 } 300 301 static FILE * 302 file(const char *name, const char *mode) 303 { 304 FILE *fp; 305 306 if ((fp = fopen(name, mode)) == NULL) 307 err(1, "%s", name); 308 return(fp); 309 } 310 311 static void 312 obsolete(char *argv[]) 313 { 314 int len; 315 char *ap, *p, *start; 316 317 while ((ap = *++argv)) { 318 /* Return if "--" or not an option of any form. */ 319 if (ap[0] != '-') { 320 if (ap[0] != '+') 321 return; 322 } else if (ap[1] == '-') 323 return; 324 if (!isdigit((unsigned char)ap[1])) 325 continue; 326 /* 327 * Digit signifies an old-style option. Malloc space for dash, 328 * new option and argument. 329 */ 330 len = strlen(ap); 331 if ((start = p = malloc(len + 3)) == NULL) 332 err(1, "malloc"); 333 *p++ = '-'; 334 *p++ = ap[0] == '+' ? 's' : 'f'; 335 (void)strcpy(p, ap + 1); 336 *argv = start; 337 } 338 } 339 340 static void 341 usage(void) 342 { 343 (void)fprintf(stderr, 344 "usage: uniq [-c] [-d | -u] [-i] [-f fields] [-s chars] [input [output]]\n"); 345 exit(1); 346 } 347