usr/src/cmd/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/types.h>
  35
  36 #include <ctype.h>
  37 #include <err.h>
  38 #include <limits.h>
  39 #include <locale.h>
  40 #include <stdio.h>
  41 #include <stdlib.h>
  42 #include <string.h>
  43 #include <unistd.h>
  44 #include <wchar.h>
  45 #include <wctype.h>
  46
  47 #include "cmap.h"
  48 #include "cset.h"
  49 #include "extern.h"
  50
  51 STR s1 = { STRING1, NORMAL, 0, OOBCH, 0, { 0, OOBCH }, NULL, NULL };
  52 STR s2 = { STRING2, NORMAL, 0, OOBCH, 0, { 0, OOBCH }, NULL, NULL };
  53
  54 static struct cset *setup(char *, STR *, int, int);
  55 static void usage(void) __NORETURN;
  56
  57 static wint_t
  58 cmap_lookup(struct cmap *cm, wint_t from)
  59 {
  60
  61         if (from < CM_CACHE_SIZE && cm->cm_havecache)
  62                 return (cm->cm_cache[from]);
  63         return (cmap_lookup_hard(cm, from));
  64 }
  65
  66 static wint_t
  67 cmap_max(struct cmap *cm)
  68 {
  69         return (cm->cm_max);
  70 }
  71
  72 static inline bool
  73 cset_in(struct cset *cs, wchar_t ch)
  74 {
  75
  76         if (ch < CS_CACHE_SIZE && cs->cs_havecache)
  77                 return (cs->cs_cache[ch]);
  78         return (cset_in_hard(cs, ch));
  79 }
  80
  81 int
  82 main(int argc, char **argv)
  83 {
  84         static int carray[NCHARS_SB];
  85         struct cmap *map;
  86         struct cset *delete, *squeeze;
  87         int n, *p;
  88         int Cflag, cflag, dflag, sflag, isstring2;
  89         wint_t ch, cnt, lastch;
  90
  91         (void) setlocale(LC_ALL, "");
  92
  93         Cflag = cflag = dflag = sflag = 0;
  94         while ((ch = getopt(argc, argv, "Ccdsu")) != -1)
  95                 switch ((char)ch) {
  96                 case 'C':
  97                         Cflag = 1;
  98                         cflag = 0;
  99                         break;
 100                 case 'c':
 101                         cflag = 1;
 102                         Cflag = 0;
 103                         break;
 104                 case 'd':
 105                         dflag = 1;
 106                         break;
 107                 case 's':
 108                         sflag = 1;
 109                         break;
 110                 case 'u':
 111                         setbuf(stdout, (char *)NULL);
 112                         break;
 113                 case '?':
 114                 default:
 115                         usage();
 116                 }
 117         argc -= optind;
 118         argv += optind;
 119
 120         switch (argc) {
 121         case 0:
 122         default:
 123                 usage();
 124                 /* NOTREACHED */
 125         case 1:
 126                 isstring2 = 0;
 127                 break;
 128         case 2:
 129                 isstring2 = 1;
 130                 break;
 131         }
 132
 133         /*
 134          * tr -ds [-Cc] string1 string2
 135          * Delete all characters (or complemented characters) in string1.
 136          * Squeeze all characters in string2.
 137          */
 138         if (dflag && sflag) {
 139                 if (!isstring2)
 140                         usage();
 141
 142                 delete = setup(argv[0], &s1, cflag, Cflag);
 143                 squeeze = setup(argv[1], &s2, 0, 0);
 144
 145                 for (lastch = OOBCH; (ch = getwchar()) != WEOF; )
 146                         if (!cset_in(delete, ch) &&
 147                             (lastch != ch || !cset_in(squeeze, ch))) {
 148                                 lastch = ch;
 149                                 (void) putwchar(ch);
 150                         }
 151                 if (ferror(stdin))
 152                         err(1, NULL);
 153                 exit(0);
 154         }
 155
 156         /*
 157          * tr -d [-Cc] string1
 158          * Delete all characters (or complemented characters) in string1.
 159          */
 160         if (dflag) {
 161                 if (isstring2)
 162                         usage();
 163
 164                 delete = setup(argv[0], &s1, cflag, Cflag);
 165
 166                 while ((ch = getwchar()) != WEOF)
 167                         if (!cset_in(delete, ch))
 168                                 (void) putwchar(ch);
 169                 if (ferror(stdin))
 170                         err(1, NULL);
 171                 exit(0);
 172         }
 173
 174         /*
 175          * tr -s [-Cc] string1
 176          * Squeeze all characters (or complemented characters) in string1.
 177          */
 178         if (sflag && !isstring2) {
 179                 squeeze = setup(argv[0], &s1, cflag, Cflag);
 180
 181                 for (lastch = OOBCH; (ch = getwchar()) != WEOF; )
 182                         if (lastch != ch || !cset_in(squeeze, ch)) {
 183                                 lastch = ch;
 184                                 (void) putwchar(ch);
 185                         }
 186                 if (ferror(stdin))
 187                         err(1, NULL);
 188                 exit(0);
 189         }
 190
 191         /*
 192          * tr [-Ccs] string1 string2
 193          * Replace all characters (or complemented characters) in string1 with
 194          * the character in the same position in string2.  If the -s option is
 195          * specified, squeeze all the characters in string2.
 196          */
 197         if (!isstring2)
 198                 usage();
 199
 200         map = cmap_alloc();
 201         if (map == NULL)
 202                 err(1, NULL);
 203         squeeze = cset_alloc();
 204         if (squeeze == NULL)
 205                 err(1, NULL);
 206
 207         s1.str = argv[0];
 208
 209         if (Cflag || cflag) {
 210                 (void) cmap_default(map, OOBCH);
 211                 if ((s2.str = strdup(argv[1])) == NULL)
 212                         errx(1, "strdup(argv[1])");
 213         } else
 214                 s2.str = argv[1];
 215
 216         if (!next(&s2))
 217                 errx(1, "empty string2");
 218
 219         /*
 220          * For -s result will contain only those characters defined
 221          * as the second characters in each of the toupper or tolower
 222          * pairs.
 223          */
 224
 225         /* If string2 runs out of characters, use the last one specified. */
 226         while (next(&s1)) {
 227         again:
 228                 if (s1.state == CCLASS_LOWER &&
 229                     s2.state == CCLASS_UPPER &&
 230                     s1.cnt == 1 && s2.cnt == 1) {
 231                         do {
 232                                 ch = towupper(s1.lastch);
 233                                 (void) cmap_add(map, s1.lastch, ch);
 234                                 if (sflag && iswupper(ch))
 235                                         (void) cset_add(squeeze, ch);
 236                                 if (!next(&s1))
 237                                         goto endloop;
 238                         } while (s1.state == CCLASS_LOWER && s1.cnt > 1);
 239                         /* skip upper set */
 240                         do {
 241                                 if (!next(&s2))
 242                                         break;
 243                         } while (s2.state == CCLASS_UPPER && s2.cnt > 1);
 244                         goto again;
 245                 } else if (s1.state == CCLASS_UPPER &&
 246                     s2.state == CCLASS_LOWER &&
 247                     s1.cnt == 1 && s2.cnt == 1) {
 248                         do {
 249                                 ch = towlower(s1.lastch);
 250                                 (void) cmap_add(map, s1.lastch, ch);
 251                                 if (sflag && iswlower(ch))
 252                                         (void) cset_add(squeeze, ch);
 253                                 if (!next(&s1))
 254                                         goto endloop;
 255                         } while (s1.state == CCLASS_UPPER && s1.cnt > 1);
 256                         /* skip lower set */
 257                         do {
 258                                 if (!next(&s2))
 259                                         break;
 260                         } while (s2.state == CCLASS_LOWER && s2.cnt > 1);
 261                         goto again;
 262                 } else {
 263                         (void) cmap_add(map, s1.lastch, s2.lastch);
 264                         if (sflag)
 265                                 (void) cset_add(squeeze, s2.lastch);
 266                 }
 267                 (void) next(&s2);
 268         }
 269 endloop:
 270         if (cflag || (Cflag && MB_CUR_MAX > 1)) {
 271                 /*
 272                  * This is somewhat tricky: since the character set is
 273                  * potentially huge, we need to avoid allocating a map
 274                  * entry for every character. Our strategy is to set the
 275                  * default mapping to the last character of string #2
 276                  * (= the one that gets automatically repeated), then to
 277                  * add back identity mappings for characters that should
 278                  * remain unchanged. We don't waste space on identity mappings
 279                  * for non-characters with the -C option; those are simulated
 280                  * in the I/O loop.
 281                  */
 282                 s2.str = argv[1];
 283                 s2.state = NORMAL;
 284                 for (cnt = 0; cnt < WCHAR_MAX; cnt++) {
 285                         if (Cflag && !iswrune(cnt))
 286                                 continue;
 287                         if (cmap_lookup(map, cnt) == OOBCH) {
 288                                 if (next(&s2))
 289                                         (void) cmap_add(map, cnt, s2.lastch);
 290                                 if (sflag)
 291                                         (void) cset_add(squeeze, s2.lastch);
 292                         } else
 293                                 (void) cmap_add(map, cnt, cnt);
 294                         if ((s2.state == EOS || s2.state == INFINITE) &&
 295                             cnt >= cmap_max(map))
 296                                 break;
 297                 }
 298                 (void) cmap_default(map, s2.lastch);
 299         } else if (Cflag) {
 300                 for (p = carray, cnt = 0; cnt < NCHARS_SB; cnt++) {
 301                         if (cmap_lookup(map, cnt) == OOBCH && iswrune(cnt))
 302                                 *p++ = cnt;
 303                         else
 304                                 (void) cmap_add(map, cnt, cnt);
 305                 }
 306                 n = p - carray;
 307                 if (Cflag && n > 1)
 308                         (void) qsort(carray, n, sizeof (*carray), charcoll);
 309
 310                 s2.str = argv[1];
 311                 s2.state = NORMAL;
 312                 for (cnt = 0; cnt < n; cnt++) {
 313                         (void) next(&s2);
 314                         (void) cmap_add(map, carray[cnt], s2.lastch);
 315                         /*
 316                          * Chars taken from s2 can be different this time
 317                          * due to lack of complex upper/lower processing,
 318                          * so fill string2 again to not miss some.
 319                          */
 320                         if (sflag)
 321                                 (void) cset_add(squeeze, s2.lastch);
 322                 }
 323         }
 324
 325         cset_cache(squeeze);
 326         cmap_cache(map);
 327
 328         if (sflag)
 329                 for (lastch = OOBCH; (ch = getwchar()) != WEOF; ) {
 330                         if (!Cflag || iswrune(ch))
 331                                 ch = cmap_lookup(map, ch);
 332                         if (lastch != ch || !cset_in(squeeze, ch)) {
 333                                 lastch = ch;
 334                                 (void) putwchar(ch);
 335                         }
 336                 }
 337         else
 338                 while ((ch = getwchar()) != WEOF) {
 339                         if (!Cflag || iswrune(ch))
 340                                 ch = cmap_lookup(map, ch);
 341                         (void) putwchar(ch);
 342                 }
 343         if (ferror(stdin))
 344                 err(1, NULL);
 345         exit(0);
 346 }
 347
 348 static struct cset *
 349 setup(char *arg, STR *str, int cflag, int Cflag)
 350 {
 351         struct cset *cs;
 352
 353         cs = cset_alloc();
 354         if (cs == NULL)
 355                 err(1, NULL);
 356         str->str = arg;
 357         while (next(str))
 358                 (void) cset_add(cs, str->lastch);
 359         if (Cflag)
 360                 (void) cset_addclass(cs, wctype("rune"), true);
 361         if (cflag || Cflag)
 362                 cset_invert(cs);
 363         cset_cache(cs);
 364         return (cs);
 365 }
 366
 367 int
 368 charcoll(const void *a, const void *b)
 369 {
 370         static char sa[2], sb[2];
 371
 372         sa[0] = *(const int *)a;
 373         sb[0] = *(const int *)b;
 374         return (strcoll(sa, sb));
 375 }
 376
 377 static void
 378 usage(void)
 379 {
 380         (void) fprintf(stderr, "%s\n%s\n%s\n%s\n",
 381             "usage: tr [-Ccsu] string1 string2",
 382             "       tr [-Ccu] -d string1",
 383             "       tr [-Ccu] -s string1",
 384             "       tr [-Ccu] -ds string1 string2");
 385         exit(1);
 386 }