str.c revision 98210
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 34#include <sys/cdefs.h> 35 36__FBSDID("$FreeBSD: head/usr.bin/tr/str.c 98210 2002-06-14 07:37:08Z tjr $"); 37 38#ifndef lint 39static const char sccsid[] = "@(#)str.c 8.2 (Berkeley) 4/28/95"; 40#endif 41 42#include <sys/cdefs.h> 43#include <sys/types.h> 44 45#include <ctype.h> 46#include <err.h> 47#include <stddef.h> 48#include <stdio.h> 49#include <stdlib.h> 50#include <string.h> 51 52#include "extern.h" 53 54static int backslash(STR *); 55static int bracket(STR *); 56static int c_class(const void *, const void *); 57static void genclass(STR *); 58static void genequiv(STR *); 59static int genrange(STR *); 60static void genseq(STR *); 61 62int 63next(s) 64 STR *s; 65{ 66 int ch; 67 68 switch (s->state) { 69 case EOS: 70 return (0); 71 case INFINITE: 72 return (1); 73 case NORMAL: 74 switch (ch = (u_char)*s->str) { 75 case '\0': 76 s->state = EOS; 77 return (0); 78 case '\\': 79 s->lastch = backslash(s); 80 break; 81 case '[': 82 if (bracket(s)) 83 return (next(s)); 84 /* FALLTHROUGH */ 85 default: 86 ++s->str; 87 s->lastch = ch; 88 break; 89 } 90 91 /* We can start a range at any time. */ 92 if (s->str[0] == '-' && genrange(s)) 93 return (next(s)); 94 return (1); 95 case RANGE: 96 if (s->cnt-- == 0) { 97 s->state = NORMAL; 98 return (next(s)); 99 } 100 ++s->lastch; 101 return (1); 102 case SEQUENCE: 103 if (s->cnt-- == 0) { 104 s->state = NORMAL; 105 return (next(s)); 106 } 107 return (1); 108 case SET: 109 if ((s->lastch = s->set[s->cnt++]) == OOBCH) { 110 s->state = NORMAL; 111 return (next(s)); 112 } 113 return (1); 114 default: 115 return (0); 116 } 117 /* NOTREACHED */ 118} 119 120static int 121bracket(s) 122 STR *s; 123{ 124 char *p; 125 126 switch (s->str[1]) { 127 case ':': /* "[:class:]" */ 128 if ((p = strstr(s->str + 2, ":]")) == NULL) 129 return (0); 130 *p = '\0'; 131 s->str += 2; 132 genclass(s); 133 s->str = p + 2; 134 return (1); 135 case '=': /* "[=equiv=]" */ 136 if ((p = strstr(s->str + 2, "=]")) == NULL) 137 return (0); 138 s->str += 2; 139 genequiv(s); 140 return (1); 141 default: /* "[\###*n]" or "[#*n]" */ 142 if ((p = strpbrk(s->str + 2, "*]")) == NULL) 143 return (0); 144 if (p[0] != '*' || index(p, ']') == NULL) 145 return (0); 146 s->str += 1; 147 genseq(s); 148 return (1); 149 } 150 /* NOTREACHED */ 151} 152 153typedef struct { 154 const char *name; 155 int (*func)(int); 156 int *set; 157} CLASS; 158 159static CLASS classes[] = { 160#undef isalnum 161 { "alnum", isalnum, NULL }, 162#undef isalpha 163 { "alpha", isalpha, NULL }, 164#undef isblank 165 { "blank", isblank, NULL }, 166#undef iscntrl 167 { "cntrl", iscntrl, NULL }, 168#undef isdigit 169 { "digit", isdigit, NULL }, 170#undef isgraph 171 { "graph", isgraph, NULL }, 172#undef islower 173 { "lower", islower, NULL }, 174#undef isprint 175 { "print", isprint, NULL }, 176#undef ispunct 177 { "punct", ispunct, NULL }, 178#undef isspace 179 { "space", isspace, NULL }, 180#undef isupper 181 { "upper", isupper, NULL }, 182#undef isxdigit 183 { "xdigit", isxdigit, NULL }, 184}; 185 186static void 187genclass(s) 188 STR *s; 189{ 190 int cnt, (*func)(int); 191 CLASS *cp, tmp; 192 int *p; 193 194 tmp.name = s->str; 195 if ((cp = (CLASS *)bsearch(&tmp, classes, sizeof(classes) / 196 sizeof(CLASS), sizeof(CLASS), c_class)) == NULL) 197 errx(1, "unknown class %s", s->str); 198 199 if ((cp->set = p = malloc((NCHARS + 1) * sizeof(int))) == NULL) 200 errx(1, "malloc"); 201 bzero(p, (NCHARS + 1) * sizeof(int)); 202 for (cnt = 0, func = cp->func; cnt < NCHARS; ++cnt) 203 if ((func)(cnt)) 204 *p++ = cnt; 205 *p = OOBCH; 206 207 s->cnt = 0; 208 s->state = SET; 209 s->set = cp->set; 210} 211 212static int 213c_class(a, b) 214 const void *a, *b; 215{ 216 return (strcmp(((const CLASS *)a)->name, ((const CLASS *)b)->name)); 217} 218 219static void 220genequiv(s) 221 STR *s; 222{ 223 int i, p, pri; 224 char src[2], dst[3]; 225 226 if (*s->str == '\\') { 227 s->equiv[0] = backslash(s); 228 if (*s->str != '=') 229 errx(1, "misplaced equivalence equals sign"); 230 } else { 231 s->equiv[0] = s->str[0]; 232 if (s->str[1] != '=') 233 errx(1, "misplaced equivalence equals sign"); 234 } 235 236 /* 237 * Calculate the set of all characters in the same equivalence class 238 * as the specified character (they will have the same primary 239 * collation weights). 240 * XXX Knows too much about how strxfrm() is implemented. Assumes 241 * it fills the string with primary collation weight bytes. Only one- 242 * to-one mappings are supported. 243 */ 244 src[0] = s->equiv[0]; 245 src[1] = '\0'; 246 if (strxfrm(dst, src, sizeof(dst)) == 1) { 247 pri = (unsigned char)*dst; 248 for (p = 1, i = 1; i < NCHARS; i++) { 249 *src = i; 250 if (strxfrm(dst, src, sizeof(dst)) == 1 && pri && 251 pri == (unsigned char)*dst) 252 s->equiv[p++] = i; 253 } 254 s->equiv[p] = OOBCH; 255 } 256 257 s->str += 2; 258 s->cnt = 0; 259 s->state = SET; 260 s->set = s->equiv; 261} 262 263static int 264genrange(s) 265 STR *s; 266{ 267 int stopval; 268 char *savestart; 269 270 savestart = s->str; 271 stopval = *++s->str == '\\' ? backslash(s) : (u_char)*s->str++; 272 if (stopval < (u_char)s->lastch) { 273 s->str = savestart; 274 return (0); 275 } 276 s->cnt = stopval - s->lastch + 1; 277 s->state = RANGE; 278 --s->lastch; 279 return (1); 280} 281 282static void 283genseq(s) 284 STR *s; 285{ 286 char *ep; 287 288 if (s->which == STRING1) 289 errx(1, "sequences only valid in string2"); 290 291 if (*s->str == '\\') 292 s->lastch = backslash(s); 293 else 294 s->lastch = *s->str++; 295 if (*s->str != '*') 296 errx(1, "misplaced sequence asterisk"); 297 298 switch (*++s->str) { 299 case '\\': 300 s->cnt = backslash(s); 301 break; 302 case ']': 303 s->cnt = 0; 304 ++s->str; 305 break; 306 default: 307 if (isdigit((u_char)*s->str)) { 308 s->cnt = strtol(s->str, &ep, 0); 309 if (*ep == ']') { 310 s->str = ep + 1; 311 break; 312 } 313 } 314 errx(1, "illegal sequence count"); 315 /* NOTREACHED */ 316 } 317 318 s->state = s->cnt ? SEQUENCE : INFINITE; 319} 320 321/* 322 * Translate \??? into a character. Up to 3 octal digits, if no digits either 323 * an escape code or a literal character. 324 */ 325static int 326backslash(s) 327 STR *s; 328{ 329 int ch, cnt, val; 330 331 for (cnt = val = 0;;) { 332 ch = (u_char)*++s->str; 333 if (!isascii(ch) || !isdigit(ch)) 334 break; 335 val = val * 8 + ch - '0'; 336 if (++cnt == 3) { 337 ++s->str; 338 break; 339 } 340 } 341 if (cnt) 342 return (val); 343 if (ch != '\0') 344 ++s->str; 345 switch (ch) { 346 case 'a': /* escape characters */ 347 return ('\7'); 348 case 'b': 349 return ('\b'); 350 case 'f': 351 return ('\f'); 352 case 'n': 353 return ('\n'); 354 case 'r': 355 return ('\r'); 356 case 't': 357 return ('\t'); 358 case 'v': 359 return ('\13'); 360 case '\0': /* \" -> \ */ 361 s->state = EOS; 362 return ('\\'); 363 default: /* \x" -> x */ 364 return (ch); 365 } 366} 367