text_cmds/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. All advertising materials mentioning features or use of this software
  14  *    must display the following acknowledgement:
  15  *      This product includes software developed by the University of
  16  *      California, Berkeley and its contributors.
  17  * 4. Neither the name of the University nor the names of its contributors
  18  *    may be used to endorse or promote products derived from this software
  19  *    without specific prior written permission.
  20  *
  21  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  22  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  23  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  24  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  25  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  26  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  27  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  28  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  29  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  30  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  31  * SUCH DAMAGE.
  32  */
  33
  34 #include <sys/cdefs.h>
  35
  36 __FBSDID("$FreeBSD: src/usr.bin/tr/tr.c,v 1.24 2005/04/09 14:31:41 stefanf Exp $");
  37
  38 #ifndef lint
  39 static const char copyright[] =
  40 "@(#) Copyright (c) 1988, 1993\n\
  41         The Regents of the University of California.  All rights reserved.\n";
  42 #endif
  43
  44 #ifndef lint
  45 static const char sccsid[] = "@(#)tr.c  8.2 (Berkeley) 5/4/95";
  46 #endif
  47
  48 #include <sys/types.h>
  49
  50 #include <ctype.h>
  51 #include <err.h>
  52 #include <limits.h>
  53 #include <locale.h>
  54 #include <stdio.h>
  55 #include <stdlib.h>
  56 #include <string.h>
  57 #include <unistd.h>
  58 #include <wchar.h>
  59 #include <wctype.h>
  60
  61 #include "cmap.h"
  62 #include "cset.h"
  63 #include "extern.h"
  64
  65 STR s1 = { STRING1, NORMAL, 0, OOBCH, 0, { 0, OOBCH }, NULL, NULL };
  66 STR s2 = { STRING2, NORMAL, 0, OOBCH, 0, { 0, OOBCH }, NULL, NULL };
  67
  68 static struct cset *setup(char *, STR *, int, int);
  69 static void usage(void);
  70
  71 int
  72 main(int argc, char **argv)
  73 {
  74         static int carray[NCHARS_SB];
  75         struct cmap *map;
  76         struct cset *delete, *squeeze;
  77         int n, *p;
  78         int Cflag, cflag, dflag, sflag, isstring2;
  79         wint_t ch, cnt, lastch;
  80
  81         (void)setlocale(LC_ALL, "");
  82
  83         Cflag = cflag = dflag = sflag = 0;
  84         while ((ch = getopt(argc, argv, "Ccdsu")) != -1)
  85                 switch((char)ch) {
  86                 case 'C':
  87                         Cflag = 1;
  88                         cflag = 0;
  89                         break;
  90                 case 'c':
  91                         cflag = 1;
  92                         Cflag = 0;
  93                         break;
  94                 case 'd':
  95                         dflag = 1;
  96                         break;
  97                 case 's':
  98                         sflag = 1;
  99                         break;
 100                 case 'u':
 101                         setbuf(stdout, (char *)NULL);
 102                         break;
 103                 case '?':
 104                 default:
 105                         usage();
 106                 }
 107         argc -= optind;
 108         argv += optind;
 109
 110         switch(argc) {
 111         case 0:
 112         default:
 113                 usage();
 114                 /* NOTREACHED */
 115         case 1:
 116                 isstring2 = 0;
 117                 if(!argv[0]) usage();
 118                 break;
 119         case 2:
 120                 isstring2 = 1;
 121                 if(!argv[0] || !argv[1]) usage();
 122                 break;
 123         }
 124
 125         /*
 126          * tr -ds [-Cc] string1 string2
 127          * Delete all characters (or complemented characters) in string1.
 128          * Squeeze all characters in string2.
 129          */
 130         if (dflag && sflag) {
 131                 if (!isstring2)
 132                         usage();
 133
 134                 delete = setup(argv[0], &s1, cflag, Cflag);
 135                 squeeze = setup(argv[1], &s2, 0, 0);
 136
 137                 for (lastch = OOBCH; (ch = getwchar()) != WEOF;)
 138                         if (!cset_in(delete, ch) &&
 139                             (lastch != ch || !cset_in(squeeze, ch))) {
 140                                 lastch = ch;
 141                                 (void)putwchar(ch);
 142                         }
 143                 if (ferror(stdin))
 144                         err(1, NULL);
 145                 exit(0);
 146         }
 147
 148         /*
 149          * tr -d [-Cc] string1
 150          * Delete all characters (or complemented characters) in string1.
 151          */
 152         if (dflag) {
 153                 if (isstring2)
 154                         usage();
 155
 156                 delete = setup(argv[0], &s1, cflag, Cflag);
 157
 158                 while ((ch = getwchar()) != WEOF)
 159                         if (!cset_in(delete, ch))
 160                                 (void)putwchar(ch);
 161                 if (ferror(stdin))
 162                         err(1, NULL);
 163                 exit(0);
 164         }
 165
 166         /*
 167          * tr -s [-Cc] string1
 168          * Squeeze all characters (or complemented characters) in string1.
 169          */
 170         if (sflag && !isstring2) {
 171                 squeeze = setup(argv[0], &s1, cflag, Cflag);
 172
 173                 for (lastch = OOBCH; (ch = getwchar()) != WEOF;)
 174                         if (lastch != ch || !cset_in(squeeze, ch)) {
 175                                 lastch = ch;
 176                                 (void)putwchar(ch);
 177                         }
 178                 if (ferror(stdin))
 179                         err(1, NULL);
 180                 exit(0);
 181         }
 182
 183         /*
 184          * tr [-Ccs] string1 string2
 185          * Replace all characters (or complemented characters) in string1 with
 186          * the character in the same position in string2.  If the -s option is
 187          * specified, squeeze all the characters in string2.
 188          */
 189         if (!isstring2)
 190                 usage();
 191
 192         map = cmap_alloc();
 193         if (map == NULL)
 194                 err(1, NULL);
 195         squeeze = cset_alloc();
 196         if (squeeze == NULL)
 197                 err(1, NULL);
 198
 199         s1.str = argv[0];
 200
 201         if (Cflag || cflag) {
 202                 cmap_default(map, OOBCH);
 203                 if ((s2.str = strdup(argv[1])) == NULL)
 204                         errx(1, "strdup(argv[1])");
 205         } else
 206                 s2.str = argv[1];
 207
 208         if (!next(&s2))
 209                 errx(1, "empty string2");
 210
 211         /*
 212          * For -s result will contain only those characters defined
 213          * as the second characters in each of the toupper or tolower
 214          * pairs.
 215          */
 216
 217         /* If string2 runs out of characters, use the last one specified. */
 218         while (next(&s1)) {
 219         again:
 220                 if (s1.state == CCLASS_LOWER &&
 221                     s2.state == CCLASS_UPPER &&
 222                     s1.cnt == 1 && s2.cnt == 1) {
 223                         do {
 224                                 ch = towupper(s1.lastch);
 225                                 cmap_add(map, s1.lastch, ch);
 226                                 if (sflag && iswupper(ch))
 227                                         cset_add(squeeze, ch);
 228                                 if (!next(&s1))
 229                                         goto endloop;
 230                         } while (s1.state == CCLASS_LOWER && s1.cnt > 1);
 231                         /* skip upper set */
 232                         do {
 233                                 if (!next(&s2))
 234                                         break;
 235                         } while (s2.state == CCLASS_UPPER && s2.cnt > 1);
 236                         goto again;
 237                 } else if (s1.state == CCLASS_UPPER &&
 238                            s2.state == CCLASS_LOWER &&
 239                            s1.cnt == 1 && s2.cnt == 1) {
 240                         do {
 241                                 ch = towlower(s1.lastch);
 242                                 cmap_add(map, s1.lastch, ch);
 243                                 if (sflag && iswlower(ch))
 244                                         cset_add(squeeze, ch);
 245                                 if (!next(&s1))
 246                                         goto endloop;
 247                         } while (s1.state == CCLASS_UPPER && s1.cnt > 1);
 248                         /* skip lower set */
 249                         do {
 250                                 if (!next(&s2))
 251                                         break;
 252                         } while (s2.state == CCLASS_LOWER && s2.cnt > 1);
 253                         goto again;
 254                 } else {
 255                         cmap_add(map, s1.lastch, s2.lastch);
 256                         if (sflag)
 257                                 cset_add(squeeze, s2.lastch);
 258                 }
 259                 (void)next(&s2);
 260         }
 261 endloop:
 262         if (cflag || (Cflag && MB_CUR_MAX > 1)) {
 263                 /*
 264                  * This is somewhat tricky: since the character set is
 265                  * potentially huge, we need to avoid allocating a map
 266                  * entry for every character. Our strategy is to set the
 267                  * default mapping to the last character of string #2
 268                  * (= the one that gets automatically repeated), then to
 269                  * add back identity mappings for characters that should
 270                  * remain unchanged. We don't waste space on identity mappings
 271                  * for non-characters with the -C option; those are simulated
 272                  * in the I/O loop.
 273                  */
 274                 s2.str = argv[1];
 275                 s2.state = NORMAL;
 276                 for (cnt = 0; cnt < WCHAR_MAX; cnt++) {
 277                         if (Cflag && !iswrune(cnt))
 278                                 continue;
 279                         if (cmap_lookup(map, cnt) == OOBCH) {
 280                                 if (next(&s2))
 281                                         cmap_add(map, cnt, s2.lastch);
 282                                 if (sflag)
 283                                         cset_add(squeeze, s2.lastch);
 284                         } else
 285                                 cmap_add(map, cnt, cnt);
 286                         if ((s2.state == EOS || s2.state == INFINITE) &&
 287                             cnt >= cmap_max(map))
 288                                 break;
 289                 }
 290                 cmap_default(map, s2.lastch);
 291         } else if (Cflag) {
 292                 for (p = carray, cnt = 0; cnt < NCHARS_SB; cnt++) {
 293                         if (cmap_lookup(map, cnt) == OOBCH && iswrune(cnt))
 294                                 *p++ = cnt;
 295                         else
 296                                 cmap_add(map, cnt, cnt);
 297                 }
 298                 n = p - carray;
 299                 if (Cflag && n > 1)
 300                         (void)mergesort(carray, n, sizeof(*carray), charcoll);
 301
 302                 s2.str = argv[1];
 303                 s2.state = NORMAL;
 304                 for (cnt = 0; cnt < n; cnt++) {
 305                         (void)next(&s2);
 306                         cmap_add(map, carray[cnt], s2.lastch);
 307                         /*
 308                          * Chars taken from s2 can be different this time
 309                          * due to lack of complex upper/lower processing,
 310                          * so fill string2 again to not miss some.
 311                          */
 312                         if (sflag)
 313                                 cset_add(squeeze, s2.lastch);
 314                 }
 315         }
 316
 317         cset_cache(squeeze);
 318         cmap_cache(map);
 319
 320         if (sflag)
 321                 for (lastch = OOBCH; (ch = getwchar()) != WEOF;) {
 322                         if (!Cflag || iswrune(ch))
 323                                 ch = cmap_lookup(map, ch);
 324                         if (lastch != ch || !cset_in(squeeze, ch)) {
 325                                 lastch = ch;
 326                                 (void)putwchar(ch);
 327                         }
 328                 }
 329         else
 330                 while ((ch = getwchar()) != WEOF) {
 331                         if (!Cflag || iswrune(ch))
 332                                 ch = cmap_lookup(map, ch);
 333                         (void)putwchar(ch);
 334                 }
 335         if (ferror(stdin))
 336                 err(1, NULL);
 337         exit (0);
 338 }
 339
 340 static struct cset *
 341 setup(char *arg, STR *str, int cflag, int Cflag)
 342 {
 343         struct cset *cs;
 344
 345         cs = cset_alloc();
 346         if (cs == NULL)
 347                 err(1, NULL);
 348         str->str = arg;
 349         while (next(str))
 350                 cset_add(cs, str->lastch);
 351         if (Cflag)
 352                 cset_addclass(cs, wctype("rune"), true);
 353         if (cflag || Cflag)
 354                 cset_invert(cs);
 355         cset_cache(cs);
 356         return (cs);
 357 }
 358
 359 int
 360 charcoll(const void *a, const void *b)
 361 {
 362         static char sa[2], sb[2];
 363
 364         sa[0] = *(const int *)a;
 365         sb[0] = *(const int *)b;
 366         return (strcoll(sa, sb));
 367 }
 368
 369 static void
 370 usage(void)
 371 {
 372         (void)fprintf(stderr, "%s\n%s\n%s\n%s\n",
 373                 "usage: tr [-Ccsu] string1 string2",
 374                 "       tr [-Ccu] -d string1",
 375                 "       tr [-Ccu] -s string1",
 376                 "       tr [-Ccu] -ds string1 string2");
 377         exit(1);
 378 }