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 * 4. Neither the name of the University nor the names of its contributors 17 * may be used to endorse or promote products derived from this software 18 * without specific prior written permission. 19 * 20 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 21 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 22 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 23 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 24 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 25 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 26 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 27 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 28 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 29 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 30 * SUCH DAMAGE. 31 */ 32 33 #ifndef lint 34 static const char copyright[] = 35 "@(#) Copyright (c) 1989, 1993\n\ 36 The Regents of the University of California. All rights reserved.\n"; 37 #endif /* not lint */ 38 39 #ifndef lint 40 #if 0 41 static char sccsid[] = "@(#)uniq.c 8.3 (Berkeley) 5/4/95"; 42 #endif 43 static const char rcsid[] = 44 "$FreeBSD$"; 45 #endif /* not lint */ 46 47 #include <sys/capability.h> 48 49 #include <ctype.h> 50 #include <err.h> 51 #include <errno.h> 52 #include <limits.h> 53 #include <locale.h> 54 #include <nl_types.h> 55 #include <stdint.h> 56 #define _WITH_GETLINE 57 #include <stdio.h> 58 #include <stdlib.h> 59 #include <string.h> 60 #include <termios.h> 61 #include <unistd.h> 62 #include <wchar.h> 63 #include <wctype.h> 64 65 static int cflag, dflag, uflag, iflag; 66 static int numchars, numfields, repeats; 67 68 static FILE *file(const char *, const char *); 69 static wchar_t *convert(const char *); 70 static int inlcmp(const char *, const char *); 71 static void show(FILE *, const char *); 72 static wchar_t *skip(wchar_t *); 73 static void obsolete(char *[]); 74 static void usage(void); 75 76 static void 77 strerror_init(void) 78 { 79 80 /* 81 * Cache NLS data before entering capability mode. 82 * XXXPJD: There should be strerror_init() and strsignal_init() in libc. 83 */ 84 (void)catopen("libc", NL_CAT_LOCALE); 85 } 86 87 int 88 main (int argc, char *argv[]) 89 { 90 wchar_t *tprev, *tthis; 91 FILE *ifp, *ofp; 92 int ch, comp; 93 size_t prevbuflen, thisbuflen, b1; 94 char *prevline, *thisline, *p; 95 const char *ifn; 96 cap_rights_t rights; 97 98 (void) setlocale(LC_ALL, ""); 99 100 obsolete(argv); 101 while ((ch = getopt(argc, argv, "cdif:s:u")) != -1) 102 switch (ch) { 103 case 'c': 104 cflag = 1; 105 break; 106 case 'd': 107 dflag = 1; 108 break; 109 case 'i': 110 iflag = 1; 111 break; 112 case 'f': 113 numfields = strtol(optarg, &p, 10); 114 if (numfields < 0 || *p) 115 errx(1, "illegal field skip value: %s", optarg); 116 break; 117 case 's': 118 numchars = strtol(optarg, &p, 10); 119 if (numchars < 0 || *p) 120 errx(1, "illegal character skip value: %s", optarg); 121 break; 122 case 'u': 123 uflag = 1; 124 break; 125 case '?': 126 default: 127 usage(); 128 } 129 130 argc -= optind; 131 argv += optind; 132 133 /* If no flags are set, default is -d -u. */ 134 if (cflag) { 135 if (dflag || uflag) 136 usage(); 137 } else if (!dflag && !uflag) 138 dflag = uflag = 1; 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 if (cap_rights_limit(fileno(ifp), CAP_FSTAT | CAP_READ) < 0 && 149 errno != ENOSYS) { 150 err(1, "unable to limit rights for %s", ifn); 151 } 152 rights = CAP_FSTAT | CAP_WRITE; 153 if (argc > 1) 154 ofp = file(argv[1], "w"); 155 else 156 rights |= CAP_IOCTL; 157 if (cap_rights_limit(fileno(ofp), rights) < 0 && errno != ENOSYS) { 158 err(1, "unable to limit rights for %s", 159 argc > 1 ? argv[1] : "stdout"); 160 } 161 if ((rights & CAP_IOCTL) != 0) { 162 unsigned long cmd; 163 164 cmd = TIOCGETA; /* required by isatty(3) in printf(3) */ 165 166 if (cap_ioctls_limit(fileno(ofp), &cmd, 1) < 0 && 167 errno != ENOSYS) { 168 err(1, "unable to limit ioctls for %s", 169 argc > 1 ? argv[1] : "stdout"); 170 } 171 } 172 173 strerror_init(); 174 if (cap_enter() < 0 && errno != ENOSYS) 175 err(1, "unable to enter capability mode"); 176 177 prevbuflen = thisbuflen = 0; 178 prevline = thisline = NULL; 179 180 if (getline(&prevline, &prevbuflen, ifp) < 0) { 181 if (ferror(ifp)) 182 err(1, "%s", ifn); 183 exit(0); 184 } 185 tprev = convert(prevline); 186 187 if (!cflag && uflag && dflag) 188 show(ofp, prevline); 189 190 tthis = NULL; 191 while (getline(&thisline, &thisbuflen, ifp) >= 0) { 192 if (tthis != NULL) 193 free(tthis); 194 tthis = convert(thisline); 195 196 if (tthis == NULL && tprev == NULL) 197 comp = inlcmp(thisline, prevline); 198 else if (tthis == NULL || tprev == NULL) 199 comp = 1; 200 else 201 comp = wcscoll(tthis, tprev); 202 203 if (comp) { 204 /* If different, print; set previous to new value. */ 205 if (cflag || !dflag || !uflag) 206 show(ofp, prevline); 207 p = prevline; 208 b1 = prevbuflen; 209 prevline = thisline; 210 prevbuflen = thisbuflen; 211 if (tprev != NULL) 212 free(tprev); 213 tprev = tthis; 214 if (!cflag && uflag && dflag) 215 show(ofp, prevline); 216 thisline = p; 217 thisbuflen = b1; 218 tthis = NULL; 219 repeats = 0; 220 } else 221 ++repeats; 222 } 223 if (ferror(ifp)) 224 err(1, "%s", ifn); 225 if (cflag || !dflag || !uflag) 226 show(ofp, prevline); 227 exit(0); 228 } 229 230 static wchar_t * 231 convert(const char *str) 232 { 233 size_t n; 234 wchar_t *buf, *ret, *p; 235 236 if ((n = mbstowcs(NULL, str, 0)) == (size_t)-1) 237 return (NULL); 238 if (SIZE_MAX / sizeof(*buf) < n + 1) 239 errx(1, "conversion buffer length overflow"); 240 if ((buf = malloc((n + 1) * sizeof(*buf))) == NULL) 241 err(1, "malloc"); 242 if (mbstowcs(buf, str, n + 1) != n) 243 errx(1, "internal mbstowcs() error"); 244 /* The last line may not end with \n. */ 245 if (n > 0 && buf[n - 1] == L'\n') 246 buf[n - 1] = L'\0'; 247 248 /* If requested get the chosen fields + character offsets. */ 249 if (numfields || numchars) { 250 if ((ret = wcsdup(skip(buf))) == NULL) 251 err(1, "wcsdup"); 252 free(buf); 253 } else 254 ret = buf; 255 256 if (iflag) { 257 for (p = ret; *p != L'\0'; p++) 258 *p = towlower(*p); 259 } 260 261 return (ret); 262 } 263 264 static int 265 inlcmp(const char *s1, const char *s2) 266 { 267 int c1, c2; 268 269 while (*s1 == *s2++) 270 if (*s1++ == '\0') 271 return (0); 272 c1 = (unsigned char)*s1; 273 c2 = (unsigned char)*(s2 - 1); 274 /* The last line may not end with \n. */ 275 if (c1 == '\n') 276 c1 = '\0'; 277 if (c2 == '\n') 278 c2 = '\0'; 279 return (c1 - c2); 280 } 281 282 /* 283 * show -- 284 * Output a line depending on the flags and number of repetitions 285 * of the line. 286 */ 287 static void 288 show(FILE *ofp, const char *str) 289 { 290 291 if (cflag) 292 (void)fprintf(ofp, "%4d %s", repeats + 1, str); 293 if ((dflag && repeats) || (uflag && !repeats)) 294 (void)fprintf(ofp, "%s", str); 295 } 296 297 static wchar_t * 298 skip(wchar_t *str) 299 { 300 int nchars, nfields; 301 302 for (nfields = 0; *str != L'\0' && nfields++ != numfields; ) { 303 while (iswblank(*str)) 304 str++; 305 while (*str != L'\0' && !iswblank(*str)) 306 str++; 307 } 308 for (nchars = numchars; nchars-- && *str != L'\0'; ++str) 309 ; 310 return(str); 311 } 312 313 static FILE * 314 file(const char *name, const char *mode) 315 { 316 FILE *fp; 317 318 if ((fp = fopen(name, mode)) == NULL) 319 err(1, "%s", name); 320 return(fp); 321 } 322 323 static void 324 obsolete(char *argv[]) 325 { 326 int len; 327 char *ap, *p, *start; 328 329 while ((ap = *++argv)) { 330 /* Return if "--" or not an option of any form. */ 331 if (ap[0] != '-') { 332 if (ap[0] != '+') 333 return; 334 } else if (ap[1] == '-') 335 return; 336 if (!isdigit((unsigned char)ap[1])) 337 continue; 338 /* 339 * Digit signifies an old-style option. Malloc space for dash, 340 * new option and argument. 341 */ 342 len = strlen(ap); 343 if ((start = p = malloc(len + 3)) == NULL) 344 err(1, "malloc"); 345 *p++ = '-'; 346 *p++ = ap[0] == '+' ? 's' : 'f'; 347 (void)strcpy(p, ap + 1); 348 *argv = start; 349 } 350 } 351 352 static void 353 usage(void) 354 { 355 (void)fprintf(stderr, 356 "usage: uniq [-c | -d | -u] [-i] [-f fields] [-s chars] [input [output]]\n"); 357 exit(1); 358 } 359