commands/tr/str.c

   1 /*      $NetBSD: str.c,v 1.12 2009/04/13 23:50:49 lukem Exp $   */
   2
   3 /*-
   4  * Copyright (c) 1991, 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
  34 #include <err.h>
  35 #include <errno.h>
  36 #include <stddef.h>
  37 #include <stdio.h>
  38 #include <stdlib.h>
  39 #include <string.h>
  40 #include <ctype.h>
  41
  42 #include "tr.h"
  43
  44 static int      backslash (STR *);
  45 static int      bracket (STR *);
  46 static int      c_class (const void *, const void *);
  47 static void     genclass (STR *);
  48 static void     genequiv (STR *);
  49 static int      genrange (STR *);
  50 static void     genseq (STR *);
  51
  52 int
  53 next(s)
  54         STR *s;
  55 {
  56         int ch;
  57
  58         switch (s->state) {
  59         case EOS:
  60                 return (0);
  61         case INFINITE:
  62                 return (1);
  63         case NORMAL:
  64                 switch (ch = *s->str) {
  65                 case '\0':
  66                         s->state = EOS;
  67                         return (0);
  68                 case '\\':
  69                         s->lastch = backslash(s);
  70                         break;
  71                 case '[':
  72                         if (bracket(s))
  73                                 return (next(s));
  74                         /* FALLTHROUGH */
  75                 default:
  76                         ++s->str;
  77                         s->lastch = ch;
  78                         break;
  79                 }
  80
  81                 /* We can start a range at any time. */
  82                 if (s->str[0] == '-' && genrange(s))
  83                         return (next(s));
  84                 return (1);
  85         case RANGE:
  86                 if (s->cnt-- == 0) {
  87                         s->state = NORMAL;
  88                         return (next(s));
  89                 }
  90                 ++s->lastch;
  91                 return (1);
  92         case SEQUENCE:
  93                 if (s->cnt-- == 0) {
  94                         s->state = NORMAL;
  95                         return (next(s));
  96                 }
  97                 return (1);
  98         case SET:
  99                 if ((s->lastch = s->set[s->cnt++]) == OOBCH) {
 100                         s->state = NORMAL;
 101                         return (next(s));
 102                 }
 103                 return (1);
 104         }
 105         /* NOTREACHED */
 106         return (0);
 107 }
 108
 109 static int
 110 bracket(s)
 111         STR *s;
 112 {
 113         char *p;
 114
 115         switch (s->str[1]) {
 116         case ':':                               /* "[:class:]" */
 117                 if ((p = strstr((char *) s->str + 2, ":]")) == NULL)
 118                         return (0);
 119                 *p = '\0';
 120                 s->str += 2;
 121                 genclass(s);
 122                 s->str = (unsigned char *) p + 2;
 123                 return (1);
 124         case '=':                               /* "[=equiv=]" */
 125                 if ((p = strstr((char *) s->str + 2, "=]")) == NULL)
 126                         return (0);
 127                 s->str += 2;
 128                 genequiv(s);
 129                 return (1);
 130         default:                                /* "[\###*n]" or "[#*n]" */
 131                 if ((p = strpbrk((char *) s->str + 2, "*]")) == NULL)
 132                         return (0);
 133                 if (p[0] != '*' || strchr(p, ']') == NULL)
 134                         return (0);
 135                 s->str += 1;
 136                 genseq(s);
 137                 return (1);
 138         }
 139         /* NOTREACHED */
 140 }
 141
 142 typedef struct {
 143         const char *name;
 144         int (*func) (int);
 145         int *set;
 146 } CLASS;
 147
 148 static CLASS classes[] = {
 149         { "alnum",  isalnum,  NULL, },
 150         { "alpha",  isalpha,  NULL, },
 151         { "blank",  isblank,  NULL, },
 152         { "cntrl",  iscntrl,  NULL, },
 153         { "digit",  isdigit,  NULL, },
 154         { "graph",  isgraph,  NULL, },
 155         { "lower",  islower,  NULL, },
 156         { "print",  isprint,  NULL, },
 157         { "punct",  ispunct,  NULL, },
 158         { "space",  isspace,  NULL, },
 159         { "upper",  isupper,  NULL, },
 160         { "xdigit", isxdigit, NULL, },
 161 };
 162
 163 static void
 164 genclass(s)
 165         STR *s;
 166 {
 167         int cnt, (*func) (int);
 168         CLASS *cp, tmp;
 169         int *p;
 170
 171         tmp.name = (char *) s->str;
 172         if ((cp = (CLASS *)bsearch(&tmp, classes, sizeof(classes) /
 173             sizeof(CLASS), sizeof(CLASS), c_class)) == NULL) {
 174                 fprintf(stderr, "tr: unknown class %s\n", s->str);
 175                 exit(1);
 176         }
 177
 178         if ((cp->set = p = malloc((NCHARS + 1) * sizeof(int))) == NULL) {
 179                 perror("malloc");
 180                 exit(1);
 181         }
 182         memset(p, 0, (NCHARS + 1) * sizeof(int));
 183         for (cnt = 0, func = cp->func; cnt < NCHARS; ++cnt)
 184                 if ((func)(cnt))
 185                         *p++ = cnt;
 186         *p = OOBCH;
 187
 188         s->cnt = 0;
 189         s->state = SET;
 190         s->set = cp->set;
 191 }
 192
 193 static int
 194 c_class(a, b)
 195         const void *a, *b;
 196 {
 197         return (strcmp(((const CLASS *)a)->name, ((const CLASS *)b)->name));
 198 }
 199
 200 /*
 201  * English doesn't have any equivalence classes, so for now
 202  * we just syntax check and grab the character.
 203  */
 204 static void
 205 genequiv(s)
 206         STR *s;
 207 {
 208         if (*s->str == '\\') {
 209                 s->equiv[0] = backslash(s);
 210                 if (*s->str != '=') {
 211                         fprintf(stderr, "tr: misplaced equivalence equals sign\n");
 212                         exit(1);
 213                 }
 214         } else {
 215                 s->equiv[0] = s->str[0];
 216                 if (s->str[1] != '=') {
 217                         fprintf(stderr, "tr: misplaced equivalence equals sign\n");
 218                         exit(1);
 219                 }
 220         }
 221         s->str += 2;
 222         s->cnt = 0;
 223         s->state = SET;
 224         s->set = s->equiv;
 225 }
 226
 227 static int
 228 genrange(s)
 229         STR *s;
 230 {
 231         int stopval;
 232         unsigned char *savestart;
 233
 234         savestart = s->str;
 235         stopval = *++s->str == '\\' ? backslash(s) : *s->str++;
 236         if (stopval < (u_char)s->lastch) {
 237                 s->str = savestart;
 238                 return (0);
 239         }
 240         s->cnt = stopval - s->lastch + 1;
 241         s->state = RANGE;
 242         --s->lastch;
 243         return (1);
 244 }
 245
 246 static void
 247 genseq(s)
 248         STR *s;
 249 {
 250         char *ep;
 251
 252         if (s->which == STRING1) {
 253                 fprintf(stderr, "tr: sequences only valid in string2\n");
 254                 exit(1);
 255         }
 256
 257         if (*s->str == '\\')
 258                 s->lastch = backslash(s);
 259         else
 260                 s->lastch = *s->str++;
 261         if (*s->str != '*') {
 262                 fprintf(stderr, "tr: misplaced sequence asterisk\n");
 263                 exit(1);
 264         }
 265
 266         switch (*++s->str) {
 267         case '\\':
 268                 s->cnt = backslash(s);
 269                 break;
 270         case ']':
 271                 s->cnt = 0;
 272                 ++s->str;
 273                 break;
 274         default:
 275                 if (isdigit(*s->str)) {
 276                         s->cnt = strtol((char *) s->str, &ep, 0);
 277                         if (*ep == ']') {
 278                                 s->str = (unsigned char *) ep + 1;
 279                                 break;
 280                         }
 281                 }
 282                 fprintf(stderr, "tr: illegal sequence count\n");
 283                 exit(1);
 284                 /* NOTREACHED */
 285         }
 286
 287         s->state = s->cnt ? SEQUENCE : INFINITE;
 288 }
 289
 290 /*
 291  * Translate \??? into a character.  Up to 3 octal digits, if no digits either
 292  * an escape code or a literal character.
 293  */
 294 static int
 295 backslash(s)
 296         STR *s;
 297 {
 298         int ch, cnt, val;
 299
 300         for (cnt = val = 0;;) {
 301                 ch = *++s->str;
 302                 if (!isascii(ch) || !isdigit(ch))
 303                         break;
 304                 val = val * 8 + ch - '0';
 305                 if (++cnt == 3) {
 306                         ++s->str;
 307                         break;
 308                 }
 309         }
 310         if (cnt)
 311                 return (val);
 312         if (ch != '\0')
 313                 ++s->str;
 314         switch (ch) {
 315                 case 'a':                       /* escape characters */
 316                         return ('\7');
 317                 case 'b':
 318                         return ('\b');
 319                 case 'f':
 320                         return ('\f');
 321                 case 'n':
 322                         return ('\n');
 323                 case 'r':
 324                         return ('\r');
 325                 case 't':
 326                         return ('\t');
 327                 case 'v':
 328                         return ('\13');
 329                 case '\0':                      /*  \" -> \ */
 330                         s->state = EOS;
 331                         return ('\\');
 332                 default:                        /* \x" -> x */
 333                         return (ch);
 334         }
 335 }