usr.bin/tr/tr.c

   1 /*
   2  * Copyright (c) 1988, 1993
   3  *      The Regents of the University of California.  All rights reserved.
   4  *
   5  * Redistribution and use in source and binary forms, with or without
   6  * modification, are permitted provided that the following conditions
   7  * are met:
   8  * 1. Redistributions of source code must retain the above copyright
   9  *    notice, this list of conditions and the following disclaimer.
  10  * 2. Redistributions in binary form must reproduce the above copyright
  11  *    notice, this list of conditions and the following disclaimer in the
  12  *    documentation and/or other materials provided with the distribution.
  13  * 3. Neither the name of the University nor the names of its contributors
  14  *    may be used to endorse or promote products derived from this software
  15  *    without specific prior written permission.
  16  *
  17  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  18  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  19  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  20  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  21  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  22  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  23  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  24  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  25  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  26  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  27  * SUCH DAMAGE.
  28  *
  29  * @(#) Copyright (c) 1988, 1993 The Regents of the University of California.  All rights reserved.
  30  * @(#)tr.c     8.2 (Berkeley) 5/4/95
  31  * $FreeBSD: head/usr.bin/tr/tr.c 245767 2013-01-22 05:39:34Z andrew $
  32  */
  33
  34 #include <sys/types.h>
  35
  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);
  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         Cflag = cflag = dflag = sflag = 0;
  71         while ((ch = getopt(argc, argv, "Ccdsu")) != -1)
  72                 switch((char)ch) {
  73                 case 'C':
  74                         Cflag = 1;
  75                         cflag = 0;
  76                         break;
  77                 case 'c':
  78                         cflag = 1;
  79                         Cflag = 0;
  80                         break;
  81                 case 'd':
  82                         dflag = 1;
  83                         break;
  84                 case 's':
  85                         sflag = 1;
  86                         break;
  87                 case 'u':
  88                         setbuf(stdout, NULL);
  89                         break;
  90                 case '?':
  91                 default:
  92                         usage();
  93                 }
  94         argc -= optind;
  95         argv += optind;
  96
  97         switch(argc) {
  98         case 0:
  99         default:
 100                 usage();
 101                 /* NOTREACHED */
 102         case 1:
 103                 isstring2 = 0;
 104                 break;
 105         case 2:
 106                 isstring2 = 1;
 107                 break;
 108         }
 109
 110         /*
 111          * tr -ds [-Cc] string1 string2
 112          * Delete all characters (or complemented characters) in string1.
 113          * Squeeze all characters in string2.
 114          */
 115         if (dflag && sflag) {
 116                 if (!isstring2)
 117                         usage();
 118
 119                 delete = setup(argv[0], &s1, cflag, Cflag);
 120                 squeeze = setup(argv[1], &s2, 0, 0);
 121
 122                 for (lastch = OOBCH; (ch = getwchar()) != WEOF;)
 123                         if (!cset_in(delete, ch) &&
 124                             (lastch != ch || !cset_in(squeeze, ch))) {
 125                                 lastch = ch;
 126                                 (void)putwchar(ch);
 127                         }
 128                 if (ferror(stdin))
 129                         err(1, NULL);
 130                 exit(0);
 131         }
 132
 133         /*
 134          * tr -d [-Cc] string1
 135          * Delete all characters (or complemented characters) in string1.
 136          */
 137         if (dflag) {
 138                 if (isstring2)
 139                         usage();
 140
 141                 delete = setup(argv[0], &s1, cflag, Cflag);
 142
 143                 while ((ch = getwchar()) != WEOF)
 144                         if (!cset_in(delete, ch))
 145                                 (void)putwchar(ch);
 146                 if (ferror(stdin))
 147                         err(1, NULL);
 148                 exit(0);
 149         }
 150
 151         /*
 152          * tr -s [-Cc] string1
 153          * Squeeze all characters (or complemented characters) in string1.
 154          */
 155         if (sflag && !isstring2) {
 156                 squeeze = setup(argv[0], &s1, cflag, Cflag);
 157
 158                 for (lastch = OOBCH; (ch = getwchar()) != WEOF;)
 159                         if (lastch != ch || !cset_in(squeeze, ch)) {
 160                                 lastch = ch;
 161                                 (void)putwchar(ch);
 162                         }
 163                 if (ferror(stdin))
 164                         err(1, NULL);
 165                 exit(0);
 166         }
 167
 168         /*
 169          * tr [-Ccs] string1 string2
 170          * Replace all characters (or complemented characters) in string1 with
 171          * the character in the same position in string2.  If the -s option is
 172          * specified, squeeze all the characters in string2.
 173          */
 174         if (!isstring2)
 175                 usage();
 176
 177         map = cmap_alloc();
 178         if (map == NULL)
 179                 err(1, NULL);
 180         squeeze = cset_alloc();
 181         if (squeeze == NULL)
 182                 err(1, NULL);
 183
 184         s1.str = argv[0];
 185
 186         if (Cflag || cflag) {
 187                 cmap_default(map, OOBCH);
 188                 if ((s2.str = strdup(argv[1])) == NULL)
 189                         errx(1, "strdup(argv[1])");
 190         } else
 191                 s2.str = argv[1];
 192
 193         if (!next(&s2))
 194                 errx(1, "empty string2");
 195
 196         /*
 197          * For -s result will contain only those characters defined
 198          * as the second characters in each of the toupper or tolower
 199          * pairs.
 200          */
 201
 202         /* If string2 runs out of characters, use the last one specified. */
 203         while (next(&s1)) {
 204         again:
 205                 if (s1.state == CCLASS_LOWER &&
 206                     s2.state == CCLASS_UPPER &&
 207                     s1.cnt == 1 && s2.cnt == 1) {
 208                         do {
 209                                 ch = towupper(s1.lastch);
 210                                 cmap_add(map, s1.lastch, ch);
 211                                 if (sflag && iswupper(ch))
 212                                         cset_add(squeeze, ch);
 213                                 if (!next(&s1))
 214                                         goto endloop;
 215                         } while (s1.state == CCLASS_LOWER && s1.cnt > 1);
 216                         /* skip upper set */
 217                         do {
 218                                 if (!next(&s2))
 219                                         break;
 220                         } while (s2.state == CCLASS_UPPER && s2.cnt > 1);
 221                         goto again;
 222                 } else if (s1.state == CCLASS_UPPER &&
 223                            s2.state == CCLASS_LOWER &&
 224                            s1.cnt == 1 && s2.cnt == 1) {
 225                         do {
 226                                 ch = towlower(s1.lastch);
 227                                 cmap_add(map, s1.lastch, ch);
 228                                 if (sflag && iswlower(ch))
 229                                         cset_add(squeeze, ch);
 230                                 if (!next(&s1))
 231                                         goto endloop;
 232                         } while (s1.state == CCLASS_UPPER && s1.cnt > 1);
 233                         /* skip lower set */
 234                         do {
 235                                 if (!next(&s2))
 236                                         break;
 237                         } while (s2.state == CCLASS_LOWER && s2.cnt > 1);
 238                         goto again;
 239                 } else {
 240                         cmap_add(map, s1.lastch, s2.lastch);
 241                         if (sflag)
 242                                 cset_add(squeeze, s2.lastch);
 243                 }
 244                 (void)next(&s2);
 245         }
 246 endloop:
 247         if (cflag || (Cflag && MB_CUR_MAX > 1)) {
 248                 /*
 249                  * This is somewhat tricky: since the character set is
 250                  * potentially huge, we need to avoid allocating a map
 251                  * entry for every character. Our strategy is to set the
 252                  * default mapping to the last character of string #2
 253                  * (= the one that gets automatically repeated), then to
 254                  * add back identity mappings for characters that should
 255                  * remain unchanged. We don't waste space on identity mappings
 256                  * for non-characters with the -C option; those are simulated
 257                  * in the I/O loop.
 258                  */
 259                 s2.str = argv[1];
 260                 s2.state = NORMAL;
 261                 for (cnt = 0; cnt < WINT_MAX; cnt++) {
 262                         if (Cflag && !iswrune(cnt))
 263                                 continue;
 264                         if (cmap_lookup(map, cnt) == OOBCH) {
 265                                 if (next(&s2))
 266                                         cmap_add(map, cnt, s2.lastch);
 267                                 if (sflag)
 268                                         cset_add(squeeze, s2.lastch);
 269                         } else
 270                                 cmap_add(map, cnt, cnt);
 271                         if ((s2.state == EOS || s2.state == INFINITE) &&
 272                             cnt >= cmap_max(map))
 273                                 break;
 274                 }
 275                 cmap_default(map, s2.lastch);
 276         } else if (Cflag) {
 277                 for (p = carray, cnt = 0; cnt < NCHARS_SB; cnt++) {
 278                         if (cmap_lookup(map, cnt) == OOBCH && iswrune(cnt))
 279                                 *p++ = cnt;
 280                         else
 281                                 cmap_add(map, cnt, cnt);
 282                 }
 283                 n = p - carray;
 284                 if (Cflag && n > 1)
 285                         (void)mergesort(carray, n, sizeof(*carray), charcoll);
 286
 287                 s2.str = argv[1];
 288                 s2.state = NORMAL;
 289                 for (cnt = 0; cnt < n; cnt++) {
 290                         (void)next(&s2);
 291                         cmap_add(map, carray[cnt], s2.lastch);
 292                         /*
 293                          * Chars taken from s2 can be different this time
 294                          * due to lack of complex upper/lower processing,
 295                          * so fill string2 again to not miss some.
 296                          */
 297                         if (sflag)
 298                                 cset_add(squeeze, s2.lastch);
 299                 }
 300         }
 301
 302         cset_cache(squeeze);
 303         cmap_cache(map);
 304
 305         if (sflag)
 306                 for (lastch = OOBCH; (ch = getwchar()) != WEOF;) {
 307                         if (!Cflag || iswrune(ch))
 308                                 ch = cmap_lookup(map, ch);
 309                         if (lastch != ch || !cset_in(squeeze, ch)) {
 310                                 lastch = ch;
 311                                 (void)putwchar(ch);
 312                         }
 313                 }
 314         else
 315                 while ((ch = getwchar()) != WEOF) {
 316                         if (!Cflag || iswrune(ch))
 317                                 ch = cmap_lookup(map, ch);
 318                         (void)putwchar(ch);
 319                 }
 320         if (ferror(stdin))
 321                 err(1, NULL);
 322         exit (0);
 323 }
 324
 325 static struct cset *
 326 setup(char *arg, STR *str, int cflag, int Cflag)
 327 {
 328         struct cset *cs;
 329
 330         cs = cset_alloc();
 331         if (cs == NULL)
 332                 err(1, NULL);
 333         str->str = arg;
 334         while (next(str))
 335                 cset_add(cs, str->lastch);
 336         if (Cflag)
 337                 cset_addclass(cs, wctype("rune"), true);
 338         if (cflag || Cflag)
 339                 cset_invert(cs);
 340         cset_cache(cs);
 341         return (cs);
 342 }
 343
 344 int
 345 charcoll(const void *a, const void *b)
 346 {
 347         static char sa[2], sb[2];
 348
 349         sa[0] = *(const int *)a;
 350         sb[0] = *(const int *)b;
 351         return (strcoll(sa, sb));
 352 }
 353
 354 static void
 355 usage(void)
 356 {
 357         (void)fprintf(stderr, "%s\n%s\n%s\n%s\n",
 358                 "usage: tr [-Ccsu] string1 string2",
 359                 "       tr [-Ccu] -d string1",
 360                 "       tr [-Ccu] -s string1",
 361                 "       tr [-Ccu] -ds string1 string2");
 362         exit(1);
 363 }