1 /* $NetBSD: str.c,v 1.19 2011/09/08 12:00:26 christos 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 1.19 2011/09/08 12:00:26 christos Exp $"); 38 #endif /* not lint */ 39 40 #include <sys/types.h> 41 42 #include <err.h> 43 #include <errno.h> 44 #include <stddef.h> 45 #include <stdio.h> 46 #include <stdlib.h> 47 #include <string.h> 48 #include <ctype.h> 49 50 #include "extern.h" 51 52 static int backslash(STR *); 53 static int bracket(STR *); 54 static int c_class(const void *, const void *); 55 static void genclass(STR *); 56 static void genequiv(STR *); 57 static int genrange(STR *); 58 static void genseq(STR *); 59 60 int 61 next(STR *s) 62 { 63 int ch; 64 65 switch (s->state) { 66 case EOS: 67 return 0; 68 case INFINITE: 69 return 1; 70 case NORMAL: 71 switch (ch = *s->str) { 72 case '\0': 73 s->state = EOS; 74 return 0; 75 case '\\': 76 s->lastch = backslash(s); 77 break; 78 case '[': 79 if (bracket(s)) 80 return next(s); 81 /* FALLTHROUGH */ 82 default: 83 ++s->str; 84 s->lastch = ch; 85 break; 86 } 87 88 /* We can start a range at any time. */ 89 if (s->str[0] == '-' && genrange(s)) 90 return next(s); 91 return 1; 92 case RANGE: 93 if (s->cnt-- == 0) { 94 s->state = NORMAL; 95 return next(s); 96 } 97 ++s->lastch; 98 return 1; 99 case SEQUENCE: 100 if (s->cnt-- == 0) { 101 s->state = NORMAL; 102 return next(s); 103 } 104 return 1; 105 case SET: 106 if ((s->lastch = s->set[s->cnt++]) == OOBCH) { 107 s->state = NORMAL; 108 return next(s); 109 } 110 return 1; 111 } 112 /* NOTREACHED */ 113 return 0; 114 } 115 116 static int 117 bracket(STR *s) 118 { 119 char *p; 120 121 switch (s->str[1]) { 122 case ':': /* "[:class:]" */ 123 if ((p = strstr(s->str + 2, ":]")) == NULL) 124 return 0; 125 *p = '\0'; 126 s->str += 2; 127 genclass(s); 128 s->str = p + 2; 129 return 1; 130 case '=': /* "[=equiv=]" */ 131 if ((p = strstr(s->str + 2, "=]")) == NULL) 132 return 0; 133 s->str += 2; 134 genequiv(s); 135 return 1; 136 default: /* "[\###*n]" or "[#*n]" */ 137 if ((p = strpbrk(s->str + 2, "*]")) == NULL) 138 return 0; 139 if (p[0] != '*' || strchr(p, ']') == NULL) 140 return 0; 141 s->str += 1; 142 genseq(s); 143 return 1; 144 } 145 /* NOTREACHED */ 146 } 147 148 typedef struct { 149 const char *name; 150 int (*func)(int); 151 } CLASS; 152 153 static const CLASS classes[] = { 154 { "alnum", isalnum }, 155 { "alpha", isalpha }, 156 { "blank", isblank }, 157 { "cntrl", iscntrl }, 158 { "digit", isdigit }, 159 { "graph", isgraph }, 160 { "lower", islower }, 161 { "print", isprint }, 162 { "punct", ispunct }, 163 { "space", isspace }, 164 { "upper", isupper }, 165 { "xdigit", isxdigit }, 166 }; 167 168 static void 169 genclass(STR *s) 170 { 171 int cnt; 172 const CLASS *cp; 173 CLASS tmp; 174 int *p; 175 176 tmp.name = s->str; 177 if ((cp = bsearch(&tmp, classes, sizeof(classes) / 178 sizeof(*cp), sizeof(*cp), c_class)) == NULL) 179 errx(1, "unknown class %s", s->str); 180 181 if ((s->set = p = malloc((NCHARS + 1) * sizeof(*p))) == NULL) 182 err(1, "malloc"); 183 184 for (cnt = 0; cnt < NCHARS; ++cnt) 185 if ((*cp->func)(cnt)) 186 *p++ = cnt; 187 *p++ = OOBCH; 188 memset(p, 0, NCHARS + 1 - (p - s->set)); 189 190 s->cnt = 0; 191 s->state = SET; 192 } 193 194 static int 195 c_class(const void *a, const void *b) 196 { 197 return strcmp(((const CLASS *)a)->name, ((const CLASS *)b)->name); 198 } 199 200 /* 201 * English doesn't have any equivalence classes, so for now 202 * we just syntax check and grab the character. 203 */ 204 static void 205 genequiv(STR *s) 206 { 207 if (*s->str == '\\') { 208 s->equiv[0] = backslash(s); 209 if (*s->str != '=') 210 errx(1, "misplaced equivalence equals sign"); 211 } else { 212 s->equiv[0] = s->str[0]; 213 if (s->str[1] != '=') 214 errx(1, "misplaced equivalence equals sign"); 215 } 216 s->str += 2; 217 s->cnt = 0; 218 s->state = SET; 219 s->set = s->equiv; 220 } 221 222 static int 223 genrange(STR *s) 224 { 225 int stopval; 226 char *savestart; 227 228 savestart = s->str; 229 stopval = *++s->str == '\\' ? backslash(s) : *s->str++; 230 if (stopval < (u_char)s->lastch) { 231 s->str = savestart; 232 return 0; 233 } 234 s->cnt = stopval - s->lastch + 1; 235 s->state = RANGE; 236 --s->lastch; 237 return 1; 238 } 239 240 static void 241 genseq(STR *s) 242 { 243 char *ep; 244 245 if (s->which == STRING1) 246 errx(1, "sequences only valid in string2"); 247 248 if (*s->str == '\\') 249 s->lastch = backslash(s); 250 else 251 s->lastch = *s->str++; 252 if (*s->str != '*') 253 errx(1, "misplaced sequence asterisk"); 254 255 switch (*++s->str) { 256 case '\\': 257 s->cnt = backslash(s); 258 break; 259 case ']': 260 s->cnt = 0; 261 ++s->str; 262 break; 263 default: 264 if (isdigit(*s->str)) { 265 s->cnt = strtol(s->str, &ep, 0); 266 if (*ep == ']') { 267 s->str = ep + 1; 268 break; 269 } 270 } 271 errx(1, "illegal sequence count"); 272 /* NOTREACHED */ 273 } 274 275 s->state = s->cnt ? SEQUENCE : INFINITE; 276 } 277 278 /* 279 * Translate \??? into a character. Up to 3 octal digits, if no digits either 280 * an escape code or a literal character. 281 */ 282 static int 283 backslash(STR *s) 284 { 285 int ch, cnt, val; 286 287 for (cnt = val = 0;;) { 288 ch = *++s->str; 289 if (!isascii(ch) || !isdigit(ch)) 290 break; 291 val = val * 8 + ch - '0'; 292 if (++cnt == 3) { 293 ++s->str; 294 break; 295 } 296 } 297 if (cnt) 298 return val; 299 if (ch != '\0') 300 ++s->str; 301 switch (ch) { 302 case 'a': /* escape characters */ 303 return '\7'; 304 case 'b': 305 return '\b'; 306 case 'e': 307 return '\033'; 308 case 'f': 309 return '\f'; 310 case 'n': 311 return '\n'; 312 case 'r': 313 return '\r'; 314 case 't': 315 return '\t'; 316 case 'v': 317 return '\13'; 318 case '\0': /* \" -> \ */ 319 s->state = EOS; 320 return '\\'; 321 default: /* \x" -> x */ 322 return ch; 323 } 324 } 325