1 /*- 2 * SPDX-License-Identifier: BSD-3-Clause 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/cdefs.h> 33 34 __FBSDID("$FreeBSD$"); 35 36 #ifndef lint 37 static const char sccsid[] = "@(#)str.c 8.2 (Berkeley) 4/28/95"; 38 #endif 39 40 #include <sys/types.h> 41 42 #include <ctype.h> 43 #include <err.h> 44 #include <errno.h> 45 #include <stddef.h> 46 #include <stdio.h> 47 #include <stdlib.h> 48 #include <string.h> 49 #include <wchar.h> 50 #include <wctype.h> 51 52 #include "extern.h" 53 54 static int backslash(STR *, int *); 55 static int bracket(STR *); 56 static void genclass(STR *); 57 static void genequiv(STR *); 58 static int genrange(STR *, int); 59 static void genseq(STR *); 60 61 wint_t 62 next(STR *s) 63 { 64 int is_octal; 65 wint_t ch; 66 wchar_t wch; 67 size_t clen; 68 69 switch (s->state) { 70 case EOS: 71 return (0); 72 case INFINITE: 73 return (1); 74 case NORMAL: 75 switch (*s->str) { 76 case '\0': 77 s->state = EOS; 78 return (0); 79 case '\\': 80 s->lastch = backslash(s, &is_octal); 81 break; 82 case '[': 83 if (bracket(s)) 84 return (next(s)); 85 /* FALLTHROUGH */ 86 default: 87 clen = mbrtowc(&wch, s->str, MB_LEN_MAX, NULL); 88 if (clen == (size_t)-1 || clen == (size_t)-2 || 89 clen == 0) 90 errc(1, EILSEQ, NULL); 91 is_octal = 0; 92 s->lastch = wch; 93 s->str += clen; 94 break; 95 } 96 97 /* We can start a range at any time. */ 98 if (s->str[0] == '-' && genrange(s, is_octal)) 99 return (next(s)); 100 return (1); 101 case RANGE: 102 if (s->cnt-- == 0) { 103 s->state = NORMAL; 104 return (next(s)); 105 } 106 ++s->lastch; 107 return (1); 108 case SEQUENCE: 109 if (s->cnt-- == 0) { 110 s->state = NORMAL; 111 return (next(s)); 112 } 113 return (1); 114 case CCLASS: 115 case CCLASS_UPPER: 116 case CCLASS_LOWER: 117 s->cnt++; 118 ch = nextwctype(s->lastch, s->cclass); 119 if (ch == -1) { 120 s->state = NORMAL; 121 return (next(s)); 122 } 123 s->lastch = ch; 124 return (1); 125 case SET: 126 if ((ch = s->set[s->cnt++]) == OOBCH) { 127 s->state = NORMAL; 128 return (next(s)); 129 } 130 s->lastch = ch; 131 return (1); 132 default: 133 return (0); 134 } 135 /* NOTREACHED */ 136 } 137 138 static int 139 bracket(STR *s) 140 { 141 char *p; 142 143 switch (s->str[1]) { 144 case ':': /* "[:class:]" */ 145 if ((p = strchr(s->str + 2, ']')) == NULL) 146 return (0); 147 if (*(p - 1) != ':' || p - s->str < 4) 148 goto repeat; 149 *(p - 1) = '\0'; 150 s->str += 2; 151 genclass(s); 152 s->str = p + 1; 153 return (1); 154 case '=': /* "[=equiv=]" */ 155 if (s->str[2] == '\0' || (p = strchr(s->str + 3, ']')) == NULL) 156 return (0); 157 if (*(p - 1) != '=' || p - s->str < 4) 158 goto repeat; 159 s->str += 2; 160 genequiv(s); 161 return (1); 162 default: /* "[\###*n]" or "[#*n]" */ 163 repeat: 164 if ((p = strpbrk(s->str + 2, "*]")) == NULL) 165 return (0); 166 if (p[0] != '*' || strchr(p, ']') == NULL) 167 return (0); 168 s->str += 1; 169 genseq(s); 170 return (1); 171 } 172 /* NOTREACHED */ 173 } 174 175 static void 176 genclass(STR *s) 177 { 178 179 if ((s->cclass = wctype(s->str)) == 0) 180 errx(1, "unknown class %s", s->str); 181 s->cnt = 0; 182 s->lastch = -1; /* incremented before check in next() */ 183 if (strcmp(s->str, "upper") == 0) 184 s->state = CCLASS_UPPER; 185 else if (strcmp(s->str, "lower") == 0) 186 s->state = CCLASS_LOWER; 187 else 188 s->state = CCLASS; 189 } 190 191 static void 192 genequiv(STR *s) 193 { 194 int i, p, pri; 195 char src[2], dst[3]; 196 size_t clen; 197 wchar_t wc; 198 199 if (*s->str == '\\') { 200 s->equiv[0] = backslash(s, NULL); 201 if (*s->str != '=') 202 errx(1, "misplaced equivalence equals sign"); 203 s->str += 2; 204 } else { 205 clen = mbrtowc(&wc, s->str, MB_LEN_MAX, NULL); 206 if (clen == (size_t)-1 || clen == (size_t)-2 || clen == 0) 207 errc(1, EILSEQ, NULL); 208 s->equiv[0] = wc; 209 if (s->str[clen] != '=') 210 errx(1, "misplaced equivalence equals sign"); 211 s->str += clen + 2; 212 } 213 214 /* 215 * Calculate the set of all characters in the same equivalence class 216 * as the specified character (they will have the same primary 217 * collation weights). 218 * XXX Knows too much about how strxfrm() is implemented. Assumes 219 * it fills the string with primary collation weight bytes. Only one- 220 * to-one mappings are supported. 221 * XXX Equivalence classes not supported in multibyte locales. 222 */ 223 src[0] = (char)s->equiv[0]; 224 src[1] = '\0'; 225 if (MB_CUR_MAX == 1 && strxfrm(dst, src, sizeof(dst)) == 1) { 226 pri = (unsigned char)*dst; 227 for (p = 1, i = 1; i < NCHARS_SB; i++) { 228 *src = i; 229 if (strxfrm(dst, src, sizeof(dst)) == 1 && pri && 230 pri == (unsigned char)*dst) 231 s->equiv[p++] = i; 232 } 233 s->equiv[p] = OOBCH; 234 } 235 236 s->cnt = 0; 237 s->state = SET; 238 s->set = s->equiv; 239 } 240 241 static int 242 genrange(STR *s, int was_octal) 243 { 244 int stopval, octal; 245 char *savestart; 246 int n, cnt, *p; 247 size_t clen; 248 wchar_t wc; 249 250 octal = 0; 251 savestart = s->str; 252 if (*++s->str == '\\') 253 stopval = backslash(s, &octal); 254 else { 255 clen = mbrtowc(&wc, s->str, MB_LEN_MAX, NULL); 256 if (clen == (size_t)-1 || clen == (size_t)-2) 257 errc(1, EILSEQ, NULL); 258 stopval = wc; 259 s->str += clen; 260 } 261 /* 262 * XXX Characters are not ordered according to collating sequence in 263 * multibyte locales. 264 */ 265 if (octal || was_octal || MB_CUR_MAX > 1) { 266 if (stopval < 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 if (charcoll((const void *)&stopval, (const void *)&(s->lastch)) < 0) { 276 s->str = savestart; 277 return (0); 278 } 279 if ((s->set = p = malloc((NCHARS_SB + 1) * sizeof(int))) == NULL) 280 err(1, "genrange() malloc"); 281 for (cnt = 0; cnt < NCHARS_SB; cnt++) 282 if (charcoll((const void *)&cnt, (const void *)&(s->lastch)) >= 0 && 283 charcoll((const void *)&cnt, (const void *)&stopval) <= 0) 284 *p++ = cnt; 285 *p = OOBCH; 286 n = p - s->set; 287 288 s->cnt = 0; 289 s->state = SET; 290 if (n > 1) 291 mergesort(s->set, n, sizeof(*(s->set)), charcoll); 292 return (1); 293 } 294 295 static void 296 genseq(STR *s) 297 { 298 char *ep; 299 wchar_t wc; 300 size_t clen; 301 302 if (s->which == STRING1) 303 errx(1, "sequences only valid in string2"); 304 305 if (*s->str == '\\') 306 s->lastch = backslash(s, NULL); 307 else { 308 clen = mbrtowc(&wc, s->str, MB_LEN_MAX, NULL); 309 if (clen == (size_t)-1 || clen == (size_t)-2) 310 errc(1, EILSEQ, NULL); 311 s->lastch = wc; 312 s->str += clen; 313 } 314 if (*s->str != '*') 315 errx(1, "misplaced sequence asterisk"); 316 317 switch (*++s->str) { 318 case '\\': 319 s->cnt = backslash(s, NULL); 320 break; 321 case ']': 322 s->cnt = 0; 323 ++s->str; 324 break; 325 default: 326 if (isdigit((u_char)*s->str)) { 327 s->cnt = strtol(s->str, &ep, 0); 328 if (*ep == ']') { 329 s->str = ep + 1; 330 break; 331 } 332 } 333 errx(1, "illegal sequence count"); 334 /* NOTREACHED */ 335 } 336 337 s->state = s->cnt ? SEQUENCE : INFINITE; 338 } 339 340 /* 341 * Translate \??? into a character. Up to 3 octal digits, if no digits either 342 * an escape code or a literal character. 343 */ 344 static int 345 backslash(STR *s, int *is_octal) 346 { 347 int ch, cnt, val; 348 349 if (is_octal != NULL) 350 *is_octal = 0; 351 for (cnt = val = 0;;) { 352 ch = (u_char)*++s->str; 353 if (!isdigit(ch) || ch > '7') 354 break; 355 val = val * 8 + ch - '0'; 356 if (++cnt == 3) { 357 ++s->str; 358 break; 359 } 360 } 361 if (cnt) { 362 if (is_octal != NULL) 363 *is_octal = 1; 364 return (val); 365 } 366 if (ch != '\0') 367 ++s->str; 368 switch (ch) { 369 case 'a': /* escape characters */ 370 return ('\7'); 371 case 'b': 372 return ('\b'); 373 case 'f': 374 return ('\f'); 375 case 'n': 376 return ('\n'); 377 case 'r': 378 return ('\r'); 379 case 't': 380 return ('\t'); 381 case 'v': 382 return ('\13'); 383 case '\0': /* \" -> \ */ 384 s->state = EOS; 385 return ('\\'); 386 default: /* \x" -> x */ 387 return (ch); 388 } 389 } 390