1 /*- 2 * Copyright (c) 1991, 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 * @(#)str.c 8.2 (Berkeley) 4/28/95 34 * $FreeBSD: src/usr.bin/tr/str.c,v 1.10.2.2 2002/07/29 12:59:33 tjr Exp $ 35 */ 36 37 #include <sys/types.h> 38 39 #include <ctype.h> 40 #include <err.h> 41 #include <stddef.h> 42 #include <stdio.h> 43 #include <stdlib.h> 44 #include <string.h> 45 46 #include "extern.h" 47 48 static int backslash(STR *); 49 static int bracket(STR *); 50 static int c_class(const void *, const void *); 51 static void genclass(STR *); 52 static void genequiv(STR *); 53 static int genrange(STR *); 54 static void genseq(STR *); 55 56 int 57 next(STR *s) 58 { 59 int ch; 60 61 switch (s->state) { 62 case EOS: 63 return (0); 64 case INFINITE: 65 return (1); 66 case NORMAL: 67 switch (ch = (u_char)*s->str) { 68 case '\0': 69 s->state = EOS; 70 return (0); 71 case '\\': 72 s->lastch = backslash(s); 73 break; 74 case '[': 75 if (bracket(s)) 76 return (next(s)); 77 /* FALLTHROUGH */ 78 default: 79 ++s->str; 80 s->lastch = ch; 81 break; 82 } 83 84 /* We can start a range at any time. */ 85 if (s->str[0] == '-' && genrange(s)) 86 return (next(s)); 87 return (1); 88 case RANGE: 89 if (s->cnt-- == 0) { 90 s->state = NORMAL; 91 return (next(s)); 92 } 93 ++s->lastch; 94 return (1); 95 case SEQUENCE: 96 if (s->cnt-- == 0) { 97 s->state = NORMAL; 98 return (next(s)); 99 } 100 return (1); 101 case SET: 102 if ((s->lastch = s->set[s->cnt++]) == OOBCH) { 103 s->state = NORMAL; 104 return (next(s)); 105 } 106 return (1); 107 default: 108 return (0); 109 } 110 /* NOTREACHED */ 111 } 112 113 static int 114 bracket(STR *s) 115 { 116 char *p; 117 118 switch (s->str[1]) { 119 case ':': /* "[:class:]" */ 120 if ((p = strchr(s->str + 2, ']')) == NULL) 121 return (0); 122 if (*(p - 1) != ':' || p - s->str < 4) 123 goto repeat; 124 *(p - 1) = '\0'; 125 s->str += 2; 126 genclass(s); 127 s->str = p + 1; 128 return (1); 129 case '=': /* "[=equiv=]" */ 130 if ((p = strchr(s->str + 2, ']')) == NULL) 131 return (0); 132 if (*(p - 1) != '=' || p - s->str < 4) 133 goto repeat; 134 s->str += 2; 135 genequiv(s); 136 return (1); 137 default: /* "[\###*n]" or "[#*n]" */ 138 repeat: 139 if ((p = strpbrk(s->str + 2, "*]")) == NULL) 140 return (0); 141 if (p[0] != '*' || strchr(p, ']') == NULL) 142 return (0); 143 s->str += 1; 144 genseq(s); 145 return (1); 146 } 147 /* NOTREACHED */ 148 } 149 150 typedef struct { 151 const char *name; 152 int (*func)(int); 153 int *set; 154 } CLASS; 155 156 static CLASS classes[] = { 157 #undef isalnum 158 { "alnum", isalnum, NULL }, 159 #undef isalpha 160 { "alpha", isalpha, NULL }, 161 #undef isblank 162 { "blank", isblank, NULL }, 163 #undef iscntrl 164 { "cntrl", iscntrl, NULL }, 165 #undef isdigit 166 { "digit", isdigit, NULL }, 167 #undef isgraph 168 { "graph", isgraph, NULL }, 169 #undef islower 170 { "lower", islower, NULL }, 171 #undef isprint 172 { "print", isprint, NULL }, 173 #undef ispunct 174 { "punct", ispunct, NULL }, 175 #undef isspace 176 { "space", isspace, NULL }, 177 #undef isupper 178 { "upper", isupper, NULL }, 179 #undef isxdigit 180 { "xdigit", isxdigit, NULL }, 181 }; 182 183 static void 184 genclass(STR *s) 185 { 186 int cnt, (*func)(int); 187 CLASS *cp, tmp; 188 int *p; 189 190 tmp.name = s->str; 191 if ((cp = (CLASS *)bsearch(&tmp, classes, sizeof(classes) / 192 sizeof(CLASS), sizeof(CLASS), c_class)) == NULL) 193 errx(1, "unknown class %s", s->str); 194 195 if ((cp->set = p = malloc((NCHARS + 1) * sizeof(int))) == NULL) 196 err(1, "malloc"); 197 bzero(p, (NCHARS + 1) * sizeof(int)); 198 for (cnt = 0, func = cp->func; cnt < NCHARS; ++cnt) 199 if ((func)(cnt)) 200 *p++ = cnt; 201 *p = OOBCH; 202 203 s->cnt = 0; 204 s->state = SET; 205 s->set = cp->set; 206 } 207 208 static int 209 c_class(const void *a, const void *b) 210 { 211 return (strcmp(((const CLASS *)a)->name, ((const CLASS *)b)->name)); 212 } 213 214 static void 215 genequiv(STR *s) 216 { 217 int i, p, pri; 218 char src[2], dst[3]; 219 220 if (*s->str == '\\') { 221 s->equiv[0] = backslash(s); 222 if (*s->str != '=') 223 errx(1, "misplaced equivalence equals sign"); 224 s->str += 2; 225 } else { 226 s->equiv[0] = s->str[0]; 227 if (s->str[1] != '=') 228 errx(1, "misplaced equivalence equals sign"); 229 s->str += 3; 230 } 231 232 /* 233 * Calculate the set of all characters in the same equivalence class 234 * as the specified character (they will have the same primary 235 * collation weights). 236 * XXX Knows too much about how strxfrm() is implemented. Assumes 237 * it fills the string with primary collation weight bytes. Only one- 238 * to-one mappings are supported. 239 */ 240 src[0] = s->equiv[0]; 241 src[1] = '\0'; 242 if (strxfrm(dst, src, sizeof(dst)) == 1) { 243 pri = (unsigned char)*dst; 244 for (p = 1, i = 1; i < NCHARS; i++) { 245 *src = i; 246 if (strxfrm(dst, src, sizeof(dst)) == 1 && pri && 247 pri == (unsigned char)*dst) 248 s->equiv[p++] = i; 249 } 250 s->equiv[p] = OOBCH; 251 } 252 253 s->cnt = 0; 254 s->state = SET; 255 s->set = s->equiv; 256 } 257 258 static int 259 genrange(STR *s) 260 { 261 int stopval; 262 char *savestart; 263 264 savestart = s->str; 265 stopval = *++s->str == '\\' ? backslash(s) : (u_char)*s->str++; 266 if (stopval < (u_char)s->lastch) { 267 s->str = savestart; 268 return (0); 269 } 270 s->cnt = stopval - s->lastch + 1; 271 s->state = RANGE; 272 --s->lastch; 273 return (1); 274 } 275 276 static void 277 genseq(STR *s) 278 { 279 char *ep; 280 281 if (s->which == STRING1) 282 errx(1, "sequences only valid in string2"); 283 284 if (*s->str == '\\') 285 s->lastch = backslash(s); 286 else 287 s->lastch = *s->str++; 288 if (*s->str != '*') 289 errx(1, "misplaced sequence asterisk"); 290 291 switch (*++s->str) { 292 case '\\': 293 s->cnt = backslash(s); 294 break; 295 case ']': 296 s->cnt = 0; 297 ++s->str; 298 break; 299 default: 300 if (isdigit((u_char)*s->str)) { 301 s->cnt = strtol(s->str, &ep, 0); 302 if (*ep == ']') { 303 s->str = ep + 1; 304 break; 305 } 306 } 307 errx(1, "illegal sequence count"); 308 /* NOTREACHED */ 309 } 310 311 s->state = s->cnt ? SEQUENCE : INFINITE; 312 } 313 314 /* 315 * Translate \??? into a character. Up to 3 octal digits, if no digits either 316 * an escape code or a literal character. 317 */ 318 static int 319 backslash(STR *s) 320 { 321 int ch, cnt, val; 322 323 for (cnt = val = 0;;) { 324 ch = (u_char)*++s->str; 325 if (!isascii(ch) || !isdigit(ch)) 326 break; 327 val = val * 8 + ch - '0'; 328 if (++cnt == 3) { 329 ++s->str; 330 break; 331 } 332 } 333 if (cnt) 334 return (val); 335 if (ch != '\0') 336 ++s->str; 337 switch (ch) { 338 case 'a': /* escape characters */ 339 return ('\7'); 340 case 'b': 341 return ('\b'); 342 case 'f': 343 return ('\f'); 344 case 'n': 345 return ('\n'); 346 case 'r': 347 return ('\r'); 348 case 't': 349 return ('\t'); 350 case 'v': 351 return ('\13'); 352 case '\0': /* \" -> \ */ 353 s->state = EOS; 354 return ('\\'); 355 default: /* \x" -> x */ 356 return (ch); 357 } 358 } 359