usr.bin/tr/tr.c

   1 /*-
   2  * SPDX-License-Identifier: BSD-3-Clause
   3  *
   4  * Copyright (c) 1988, 1993
   5  *      The Regents of the University of California.  All rights reserved.
   6  *
   7  * Redistribution and use in source and binary forms, with or without
   8  * modification, are permitted provided that the following conditions
   9  * are met:
  10  * 1. Redistributions of source code must retain the above copyright
  11  *    notice, this list of conditions and the following disclaimer.
  12  * 2. Redistributions in binary form must reproduce the above copyright
  13  *    notice, this list of conditions and the following disclaimer in the
  14  *    documentation and/or other materials provided with the distribution.
  15  * 3. Neither the name of the University nor the names of its contributors
  16  *    may be used to endorse or promote products derived from this software
  17  *    without specific prior written permission.
  18  *
  19  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  20  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  21  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  22  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  23  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  24  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  25  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  26  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  27  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  28  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  29  * SUCH DAMAGE.
  30  */
  31
  32 #include <sys/types.h>
  33 #include <sys/capsicum.h>
  34
  35 #include <capsicum_helpers.h>
  36 #include <ctype.h>
  37 #include <err.h>
  38 #include <limits.h>
  39 #include <locale.h>
  40 #include <stdint.h>
  41 #include <stdio.h>
  42 #include <stdlib.h>
  43 #include <string.h>
  44 #include <unistd.h>
  45 #include <wchar.h>
  46 #include <wctype.h>
  47
  48 #include "cmap.h"
  49 #include "cset.h"
  50 #include "extern.h"
  51
  52 static STR s1 = { STRING1, NORMAL, 0, OOBCH, 0, { 0, OOBCH }, NULL, NULL };
  53 static STR s2 = { STRING2, NORMAL, 0, OOBCH, 0, { 0, OOBCH }, NULL, NULL };
  54
  55 static struct cset *setup(char *, STR *, int, int);
  56 static void usage(void) __dead2;
  57
  58 int
  59 main(int argc, char **argv)
  60 {
  61         static int carray[NCHARS_SB];
  62         struct cmap *map;
  63         struct cset *delete, *squeeze;
  64         int n, *p;
  65         int Cflag, cflag, dflag, sflag, isstring2;
  66         wint_t ch, cnt, lastch;
  67
  68         (void)setlocale(LC_ALL, "");
  69
  70         if (caph_limit_stdio() == -1)
  71                 err(1, "unable to limit stdio");
  72
  73         if (caph_enter() < 0)
  74                 err(1, "unable to enter capability mode");
  75
  76         Cflag = cflag = dflag = sflag = 0;
  77         while ((ch = getopt(argc, argv, "Ccdsu")) != -1)
  78                 switch((char)ch) {
  79                 case 'C':
  80                         Cflag = 1;
  81                         cflag = 0;
  82                         break;
  83                 case 'c':
  84                         cflag = 1;
  85                         Cflag = 0;
  86                         break;
  87                 case 'd':
  88                         dflag = 1;
  89                         break;
  90                 case 's':
  91                         sflag = 1;
  92                         break;
  93                 case 'u':
  94                         setbuf(stdout, (char *)NULL);
  95                         break;
  96                 case '?':
  97                 default:
  98                         usage();
  99                 }
 100         argc -= optind;
 101         argv += optind;
 102
 103         switch(argc) {
 104         case 0:
 105         default:
 106                 usage();
 107                 /* NOTREACHED */
 108         case 1:
 109                 isstring2 = 0;
 110                 break;
 111         case 2:
 112                 isstring2 = 1;
 113                 break;
 114         }
 115
 116         /*
 117          * tr -ds [-Cc] string1 string2
 118          * Delete all characters (or complemented characters) in string1.
 119          * Squeeze all characters in string2.
 120          */
 121         if (dflag && sflag) {
 122                 if (!isstring2)
 123                         usage();
 124
 125                 delete = setup(argv[0], &s1, cflag, Cflag);
 126                 squeeze = setup(argv[1], &s2, 0, 0);
 127
 128                 for (lastch = OOBCH; (ch = getwchar()) != WEOF;)
 129                         if (!cset_in(delete, ch) &&
 130                             (lastch != ch || !cset_in(squeeze, ch))) {
 131                                 lastch = ch;
 132                                 (void)putwchar(ch);
 133                         }
 134                 if (ferror(stdin))
 135                         err(1, NULL);
 136                 exit(0);
 137         }
 138
 139         /*
 140          * tr -d [-Cc] string1
 141          * Delete all characters (or complemented characters) in string1.
 142          */
 143         if (dflag) {
 144                 if (isstring2)
 145                         usage();
 146
 147                 delete = setup(argv[0], &s1, cflag, Cflag);
 148
 149                 while ((ch = getwchar()) != WEOF)
 150                         if (!cset_in(delete, ch))
 151                                 (void)putwchar(ch);
 152                 if (ferror(stdin))
 153                         err(1, NULL);
 154                 exit(0);
 155         }
 156
 157         /*
 158          * tr -s [-Cc] string1
 159          * Squeeze all characters (or complemented characters) in string1.
 160          */
 161         if (sflag && !isstring2) {
 162                 squeeze = setup(argv[0], &s1, cflag, Cflag);
 163
 164                 for (lastch = OOBCH; (ch = getwchar()) != WEOF;)
 165                         if (lastch != ch || !cset_in(squeeze, ch)) {
 166                                 lastch = ch;
 167                                 (void)putwchar(ch);
 168                         }
 169                 if (ferror(stdin))
 170                         err(1, NULL);
 171                 exit(0);
 172         }
 173
 174         /*
 175          * tr [-Ccs] string1 string2
 176          * Replace all characters (or complemented characters) in string1 with
 177          * the character in the same position in string2.  If the -s option is
 178          * specified, squeeze all the characters in string2.
 179          */
 180         if (!isstring2)
 181                 usage();
 182
 183         map = cmap_alloc();
 184         if (map == NULL)
 185                 err(1, NULL);
 186         squeeze = cset_alloc();
 187         if (squeeze == NULL)
 188                 err(1, NULL);
 189
 190         s1.str = argv[0];
 191
 192         if (Cflag || cflag) {
 193                 cmap_default(map, OOBCH);
 194                 if ((s2.str = strdup(argv[1])) == NULL)
 195                         errx(1, "strdup(argv[1])");
 196         } else
 197                 s2.str = argv[1];
 198
 199         if (!next(&s2))
 200                 errx(1, "empty string2");
 201
 202         /*
 203          * For -s result will contain only those characters defined
 204          * as the second characters in each of the toupper or tolower
 205          * pairs.
 206          */
 207
 208         /* If string2 runs out of characters, use the last one specified. */
 209         while (next(&s1)) {
 210         again:
 211                 if (s1.state == CCLASS_LOWER &&
 212                     s2.state == CCLASS_UPPER &&
 213                     s1.cnt == 1 && s2.cnt == 1) {
 214                         do {
 215                                 ch = towupper(s1.lastch);
 216                                 cmap_add(map, s1.lastch, ch);
 217                                 if (sflag && iswupper(ch))
 218                                         cset_add(squeeze, ch);
 219                                 if (!next(&s1))
 220                                         goto endloop;
 221                         } while (s1.state == CCLASS_LOWER && s1.cnt > 1);
 222                         /* skip upper set */
 223                         do {
 224                                 if (!next(&s2))
 225                                         break;
 226                         } while (s2.state == CCLASS_UPPER && s2.cnt > 1);
 227                         goto again;
 228                 } else if (s1.state == CCLASS_UPPER &&
 229                            s2.state == CCLASS_LOWER &&
 230                            s1.cnt == 1 && s2.cnt == 1) {
 231                         do {
 232                                 ch = towlower(s1.lastch);
 233                                 cmap_add(map, s1.lastch, ch);
 234                                 if (sflag && iswlower(ch))
 235                                         cset_add(squeeze, ch);
 236                                 if (!next(&s1))
 237                                         goto endloop;
 238                         } while (s1.state == CCLASS_UPPER && s1.cnt > 1);
 239                         /* skip lower set */
 240                         do {
 241                                 if (!next(&s2))
 242                                         break;
 243                         } while (s2.state == CCLASS_LOWER && s2.cnt > 1);
 244                         goto again;
 245                 } else {
 246                         cmap_add(map, s1.lastch, s2.lastch);
 247                         if (sflag)
 248                                 cset_add(squeeze, s2.lastch);
 249                 }
 250                 (void)next(&s2);
 251         }
 252 endloop:
 253         if (cflag || (Cflag && MB_CUR_MAX > 1)) {
 254                 /*
 255                  * This is somewhat tricky: since the character set is
 256                  * potentially huge, we need to avoid allocating a map
 257                  * entry for every character. Our strategy is to set the
 258                  * default mapping to the last character of string #2
 259                  * (= the one that gets automatically repeated), then to
 260                  * add back identity mappings for characters that should
 261                  * remain unchanged. We don't waste space on identity mappings
 262                  * for non-characters with the -C option; those are simulated
 263                  * in the I/O loop.
 264                  */
 265                 s2.str = argv[1];
 266                 s2.state = NORMAL;
 267                 for (cnt = 0; cnt < WINT_MAX; cnt++) {
 268                         if (Cflag && !iswrune(cnt))
 269                                 continue;
 270                         if (cmap_lookup(map, cnt) == OOBCH) {
 271                                 if (next(&s2)) {
 272                                         cmap_add(map, cnt, s2.lastch);
 273                                         if (sflag)
 274                                                 cset_add(squeeze, s2.lastch);
 275                                 }
 276                         } else
 277                                 cmap_add(map, cnt, cnt);
 278                         if ((s2.state == EOS || s2.state == INFINITE) &&
 279                             cnt >= cmap_max(map))
 280                                 break;
 281                 }
 282                 cmap_default(map, s2.lastch);
 283         } else if (Cflag) {
 284                 for (p = carray, cnt = 0; cnt < NCHARS_SB; cnt++) {
 285                         if (cmap_lookup(map, cnt) == OOBCH && iswrune(cnt))
 286                                 *p++ = cnt;
 287                         else
 288                                 cmap_add(map, cnt, cnt);
 289                 }
 290                 n = p - carray;
 291                 if (Cflag && n > 1)
 292                         (void)mergesort(carray, n, sizeof(*carray), charcoll);
 293
 294                 s2.str = argv[1];
 295                 s2.state = NORMAL;
 296                 for (cnt = 0; cnt < n; cnt++) {
 297                         (void)next(&s2);
 298                         cmap_add(map, carray[cnt], s2.lastch);
 299                         /*
 300                          * Chars taken from s2 can be different this time
 301                          * due to lack of complex upper/lower processing,
 302                          * so fill string2 again to not miss some.
 303                          */
 304                         if (sflag)
 305                                 cset_add(squeeze, s2.lastch);
 306                 }
 307         }
 308
 309         cset_cache(squeeze);
 310         cmap_cache(map);
 311
 312         if (sflag)
 313                 for (lastch = OOBCH; (ch = getwchar()) != WEOF;) {
 314                         if (!Cflag || iswrune(ch))
 315                                 ch = cmap_lookup(map, ch);
 316                         if (lastch != ch || !cset_in(squeeze, ch)) {
 317                                 lastch = ch;
 318                                 (void)putwchar(ch);
 319                         }
 320                 }
 321         else
 322                 while ((ch = getwchar()) != WEOF) {
 323                         if (!Cflag || iswrune(ch))
 324                                 ch = cmap_lookup(map, ch);
 325                         (void)putwchar(ch);
 326                 }
 327         if (ferror(stdin))
 328                 err(1, NULL);
 329         exit (0);
 330 }
 331
 332 static struct cset *
 333 setup(char *arg, STR *str, int cflag, int Cflag)
 334 {
 335         struct cset *cs;
 336
 337         cs = cset_alloc();
 338         if (cs == NULL)
 339                 err(1, NULL);
 340         str->str = arg;
 341         while (next(str))
 342                 cset_add(cs, str->lastch);
 343         if (Cflag)
 344                 cset_addclass(cs, wctype("rune"), true);
 345         if (cflag || Cflag)
 346                 cset_invert(cs);
 347         cset_cache(cs);
 348         return (cs);
 349 }
 350
 351 int
 352 charcoll(const void *a, const void *b)
 353 {
 354         static char sa[2], sb[2];
 355
 356         sa[0] = *(const int *)a;
 357         sb[0] = *(const int *)b;
 358         return (strcoll(sa, sb));
 359 }
 360
 361 static void
 362 usage(void)
 363 {
 364         (void)fprintf(stderr, "%s\n%s\n%s\n%s\n",
 365                 "usage: tr [-Ccsu] string1 string2",
 366                 "       tr [-Ccu] -d string1",
 367                 "       tr [-Ccu] -s string1",
 368                 "       tr [-Ccu] -ds string1 string2");
 369         exit(1);
 370 }