1 /* $NetBSD: str.c,v 1.10 2003/08/07 11:16:46 agc 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/cdefs.h> 33 #ifndef lint 34 #if 0 35 static char sccsid[] = "@(#)str.c 8.2 (Berkeley) 4/28/95"; 36 #endif 37 __RCSID("$NetBSD: str.c,v 1.10 2003/08/07 11:16:46 agc Exp $"); 38 #endif /* not lint */ 39 40 #include <sys/cdefs.h> 41 #include <sys/types.h> 42 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 <ctype.h> 50 51 #include "extern.h" 52 53 static int backslash __P((STR *)); 54 static int bracket __P((STR *)); 55 static int c_class __P((const void *, const void *)); 56 static void genclass __P((STR *)); 57 static void genequiv __P((STR *)); 58 static int genrange __P((STR *)); 59 static void genseq __P((STR *)); 60 61 int 62 next(s) 63 STR *s; 64 { 65 int ch; 66 67 switch (s->state) { 68 case EOS: 69 return (0); 70 case INFINITE: 71 return (1); 72 case NORMAL: 73 switch (ch = *s->str) { 74 case '\0': 75 s->state = EOS; 76 return (0); 77 case '\\': 78 s->lastch = backslash(s); 79 break; 80 case '[': 81 if (bracket(s)) 82 return (next(s)); 83 /* FALLTHROUGH */ 84 default: 85 ++s->str; 86 s->lastch = ch; 87 break; 88 } 89 90 /* We can start a range at any time. */ 91 if (s->str[0] == '-' && genrange(s)) 92 return (next(s)); 93 return (1); 94 case RANGE: 95 if (s->cnt-- == 0) { 96 s->state = NORMAL; 97 return (next(s)); 98 } 99 ++s->lastch; 100 return (1); 101 case SEQUENCE: 102 if (s->cnt-- == 0) { 103 s->state = NORMAL; 104 return (next(s)); 105 } 106 return (1); 107 case SET: 108 if ((s->lastch = s->set[s->cnt++]) == OOBCH) { 109 s->state = NORMAL; 110 return (next(s)); 111 } 112 return (1); 113 } 114 /* NOTREACHED */ 115 return (0); 116 } 117 118 static int 119 bracket(s) 120 STR *s; 121 { 122 char *p; 123 124 switch (s->str[1]) { 125 case ':': /* "[:class:]" */ 126 if ((p = strstr(s->str + 2, ":]")) == NULL) 127 return (0); 128 *p = '\0'; 129 s->str += 2; 130 genclass(s); 131 s->str = p + 2; 132 return (1); 133 case '=': /* "[=equiv=]" */ 134 if ((p = strstr(s->str + 2, "=]")) == NULL) 135 return (0); 136 s->str += 2; 137 genequiv(s); 138 return (1); 139 default: /* "[\###*n]" or "[#*n]" */ 140 if ((p = strpbrk(s->str + 2, "*]")) == NULL) 141 return (0); 142 if (p[0] != '*' || strchr(p, ']') == NULL) 143 return (0); 144 s->str += 1; 145 genseq(s); 146 return (1); 147 } 148 /* NOTREACHED */ 149 } 150 151 typedef struct { 152 char *name; 153 int (*func) __P((int)); 154 int *set; 155 } CLASS; 156 157 static CLASS classes[] = { 158 { "alnum", isalnum, }, 159 { "alpha", isalpha, }, 160 { "blank", isblank, }, 161 { "cntrl", iscntrl, }, 162 { "digit", isdigit, }, 163 { "graph", isgraph, }, 164 { "lower", islower, }, 165 { "print", isprint, }, 166 { "punct", ispunct, }, 167 { "space", isspace, }, 168 { "upper", isupper, }, 169 { "xdigit", isxdigit, }, 170 }; 171 172 static void 173 genclass(s) 174 STR *s; 175 { 176 int cnt, (*func) __P((int)); 177 CLASS *cp, tmp; 178 int *p; 179 180 tmp.name = s->str; 181 if ((cp = (CLASS *)bsearch(&tmp, classes, sizeof(classes) / 182 sizeof(CLASS), sizeof(CLASS), c_class)) == NULL) 183 errx(1, "unknown class %s", s->str); 184 185 if ((cp->set = p = malloc((NCHARS + 1) * sizeof(int))) == NULL) 186 err(1, "malloc"); 187 memset(p, 0, (NCHARS + 1) * sizeof(int)); 188 for (cnt = 0, func = cp->func; cnt < NCHARS; ++cnt) 189 if ((func)(cnt)) 190 *p++ = cnt; 191 *p = OOBCH; 192 193 s->cnt = 0; 194 s->state = SET; 195 s->set = cp->set; 196 } 197 198 static int 199 c_class(a, b) 200 const void *a, *b; 201 { 202 return (strcmp(((CLASS *)a)->name, ((CLASS *)b)->name)); 203 } 204 205 /* 206 * English doesn't have any equivalence classes, so for now 207 * we just syntax check and grab the character. 208 */ 209 static void 210 genequiv(s) 211 STR *s; 212 { 213 if (*s->str == '\\') { 214 s->equiv[0] = backslash(s); 215 if (*s->str != '=') 216 errx(1, "misplaced equivalence equals sign"); 217 } else { 218 s->equiv[0] = s->str[0]; 219 if (s->str[1] != '=') 220 errx(1, "misplaced equivalence equals sign"); 221 } 222 s->str += 2; 223 s->cnt = 0; 224 s->state = SET; 225 s->set = s->equiv; 226 } 227 228 static int 229 genrange(s) 230 STR *s; 231 { 232 int stopval; 233 char *savestart; 234 235 savestart = s->str; 236 stopval = *++s->str == '\\' ? backslash(s) : *s->str++; 237 if (stopval < (u_char)s->lastch) { 238 s->str = savestart; 239 return (0); 240 } 241 s->cnt = stopval - s->lastch + 1; 242 s->state = RANGE; 243 --s->lastch; 244 return (1); 245 } 246 247 static void 248 genseq(s) 249 STR *s; 250 { 251 char *ep; 252 253 if (s->which == STRING1) 254 errx(1, "sequences only valid in string2"); 255 256 if (*s->str == '\\') 257 s->lastch = backslash(s); 258 else 259 s->lastch = *s->str++; 260 if (*s->str != '*') 261 errx(1, "misplaced sequence asterisk"); 262 263 switch (*++s->str) { 264 case '\\': 265 s->cnt = backslash(s); 266 break; 267 case ']': 268 s->cnt = 0; 269 ++s->str; 270 break; 271 default: 272 if (isdigit(*s->str)) { 273 s->cnt = strtol(s->str, &ep, 0); 274 if (*ep == ']') { 275 s->str = ep + 1; 276 break; 277 } 278 } 279 errx(1, "illegal sequence count"); 280 /* NOTREACHED */ 281 } 282 283 s->state = s->cnt ? SEQUENCE : INFINITE; 284 } 285 286 /* 287 * Translate \??? into a character. Up to 3 octal digits, if no digits either 288 * an escape code or a literal character. 289 */ 290 static int 291 backslash(s) 292 STR *s; 293 { 294 int ch, cnt, val; 295 296 for (cnt = val = 0;;) { 297 ch = *++s->str; 298 if (!isascii(ch) || !isdigit(ch)) 299 break; 300 val = val * 8 + ch - '0'; 301 if (++cnt == 3) { 302 ++s->str; 303 break; 304 } 305 } 306 if (cnt) 307 return (val); 308 if (ch != '\0') 309 ++s->str; 310 switch (ch) { 311 case 'a': /* escape characters */ 312 return ('\7'); 313 case 'b': 314 return ('\b'); 315 case 'f': 316 return ('\f'); 317 case 'n': 318 return ('\n'); 319 case 'r': 320 return ('\r'); 321 case 't': 322 return ('\t'); 323 case 'v': 324 return ('\13'); 325 case '\0': /* \" -> \ */ 326 s->state = EOS; 327 return ('\\'); 328 default: /* \x" -> x */ 329 return (ch); 330 } 331 } 332