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("$FreeBSD$"); 37 38 #ifndef lint 39 static 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 54 static int backslash __P((STR *)); 55 static int bracket __P((STR *)); 56 static int c_class __P((const void *, const void *)); 57 static void genclass __P((STR *)); 58 static void genequiv __P((STR *)); 59 static int genrange __P((STR *)); 60 static void genseq __P((STR *)); 61 62 int 63 next(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 } 119 120 static int 121 bracket(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 } 152 153 typedef struct { 154 const char *name; 155 int (*func) __P((int)); 156 int *set; 157 } CLASS; 158 159 static 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 }; 185 186 static void 187 genclass(s) 188 STR *s; 189 { 190 int cnt, (*func) __P((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 } 211 212 static int 213 c_class(a, b) 214 const void *a, *b; 215 { 216 return (strcmp(((const CLASS *)a)->name, ((const CLASS *)b)->name)); 217 } 218 219 /* 220 * English doesn't have any equivalence classes, so for now 221 * we just syntax check and grab the character. 222 */ 223 static void 224 genequiv(s) 225 STR *s; 226 { 227 if (*s->str == '\\') { 228 s->equiv[0] = backslash(s); 229 if (*s->str != '=') 230 errx(1, "misplaced equivalence equals sign"); 231 } else { 232 s->equiv[0] = s->str[0]; 233 if (s->str[1] != '=') 234 errx(1, "misplaced equivalence equals sign"); 235 } 236 s->str += 2; 237 s->cnt = 0; 238 s->state = SET; 239 s->set = s->equiv; 240 } 241 242 static int 243 genrange(s) 244 STR *s; 245 { 246 int stopval; 247 char *savestart; 248 249 savestart = s->str; 250 stopval = *++s->str == '\\' ? backslash(s) : (u_char)*s->str++; 251 if (stopval < (u_char)s->lastch) { 252 s->str = savestart; 253 return (0); 254 } 255 s->cnt = stopval - s->lastch + 1; 256 s->state = RANGE; 257 --s->lastch; 258 return (1); 259 } 260 261 static void 262 genseq(s) 263 STR *s; 264 { 265 char *ep; 266 267 if (s->which == STRING1) 268 errx(1, "sequences only valid in string2"); 269 270 if (*s->str == '\\') 271 s->lastch = backslash(s); 272 else 273 s->lastch = *s->str++; 274 if (*s->str != '*') 275 errx(1, "misplaced sequence asterisk"); 276 277 switch (*++s->str) { 278 case '\\': 279 s->cnt = backslash(s); 280 break; 281 case ']': 282 s->cnt = 0; 283 ++s->str; 284 break; 285 default: 286 if (isdigit((u_char)*s->str)) { 287 s->cnt = strtol(s->str, &ep, 0); 288 if (*ep == ']') { 289 s->str = ep + 1; 290 break; 291 } 292 } 293 errx(1, "illegal sequence count"); 294 /* NOTREACHED */ 295 } 296 297 s->state = s->cnt ? SEQUENCE : INFINITE; 298 } 299 300 /* 301 * Translate \??? into a character. Up to 3 octal digits, if no digits either 302 * an escape code or a literal character. 303 */ 304 static int 305 backslash(s) 306 STR *s; 307 { 308 int ch, cnt, val; 309 310 for (cnt = val = 0;;) { 311 ch = (u_char)*++s->str; 312 if (!isascii(ch) || !isdigit(ch)) 313 break; 314 val = val * 8 + ch - '0'; 315 if (++cnt == 3) { 316 ++s->str; 317 break; 318 } 319 } 320 if (cnt) 321 return (val); 322 if (ch != '\0') 323 ++s->str; 324 switch (ch) { 325 case 'a': /* escape characters */ 326 return ('\7'); 327 case 'b': 328 return ('\b'); 329 case 'f': 330 return ('\f'); 331 case 'n': 332 return ('\n'); 333 case 'r': 334 return ('\r'); 335 case 't': 336 return ('\t'); 337 case 'v': 338 return ('\13'); 339 case '\0': /* \" -> \ */ 340 s->state = EOS; 341 return ('\\'); 342 default: /* \x" -> x */ 343 return (ch); 344 } 345 } 346