str.c revision 1.20
1 /* $NetBSD: str.c,v 1.20 2013/08/10 23:54:41 dholland 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 1.20 2013/08/10 23:54:41 dholland 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 #include <assert.h> 50 51 #include "extern.h" 52 53 static int backslash(STR *); 54 static int bracket(STR *); 55 static int c_class(const void *, const void *); 56 static void genclass(STR *); 57 static void genequiv(STR *); 58 static int genrange(STR *); 59 static void genseq(STR *); 60 61 int 62 next(STR *s, int *ret) 63 { 64 int ch; 65 66 switch (s->state) { 67 case EOS: 68 *ret = s->lastch; 69 return 0; 70 case INFINITE: 71 *ret = s->lastch; 72 return 1; 73 case NORMAL: 74 switch (ch = *s->str) { 75 case '\0': 76 s->state = EOS; 77 *ret = s->lastch; 78 return 0; 79 case '\\': 80 s->lastch = backslash(s); 81 break; 82 case '[': 83 if (bracket(s)) 84 return next(s, ret); 85 /* FALLTHROUGH */ 86 default: 87 ++s->str; 88 s->lastch = ch; 89 break; 90 } 91 92 /* We can start a range at any time. */ 93 if (s->str[0] == '-' && genrange(s)) { 94 return next(s, ret); 95 } 96 *ret = s->lastch; 97 return 1; 98 case RANGE: 99 if (s->cnt-- == 0) { 100 s->state = NORMAL; 101 return next(s, ret); 102 } 103 ++s->lastch; 104 *ret = s->lastch; 105 return 1; 106 case SEQUENCE: 107 if (s->cnt-- == 0) { 108 s->state = NORMAL; 109 return next(s, ret); 110 } 111 *ret = s->lastch; 112 return 1; 113 case SET: 114 if ((s->lastch = s->set[s->cnt++]) == OOBCH) { 115 s->state = NORMAL; 116 return next(s, ret); 117 } 118 *ret = s->lastch; 119 return 1; 120 } 121 /* NOTREACHED */ 122 assert(0); 123 *ret = s->lastch; 124 return 0; 125 } 126 127 static int 128 bracket(STR *s) 129 { 130 char *p; 131 132 switch (s->str[1]) { 133 case ':': /* "[:class:]" */ 134 if ((p = strstr(s->str + 2, ":]")) == NULL) 135 return 0; 136 *p = '\0'; 137 s->str += 2; 138 genclass(s); 139 s->str = p + 2; 140 return 1; 141 case '=': /* "[=equiv=]" */ 142 if ((p = strstr(s->str + 2, "=]")) == NULL) 143 return 0; 144 s->str += 2; 145 genequiv(s); 146 return 1; 147 default: /* "[\###*n]" or "[#*n]" */ 148 if ((p = strpbrk(s->str + 2, "*]")) == NULL) 149 return 0; 150 if (p[0] != '*' || strchr(p, ']') == NULL) 151 return 0; 152 s->str += 1; 153 genseq(s); 154 return 1; 155 } 156 /* NOTREACHED */ 157 } 158 159 typedef struct { 160 const char *name; 161 int (*func)(int); 162 } CLASS; 163 164 static const CLASS classes[] = { 165 { "alnum", isalnum }, 166 { "alpha", isalpha }, 167 { "blank", isblank }, 168 { "cntrl", iscntrl }, 169 { "digit", isdigit }, 170 { "graph", isgraph }, 171 { "lower", islower }, 172 { "print", isprint }, 173 { "punct", ispunct }, 174 { "space", isspace }, 175 { "upper", isupper }, 176 { "xdigit", isxdigit }, 177 }; 178 179 static void 180 genclass(STR *s) 181 { 182 int cnt; 183 const CLASS *cp; 184 CLASS tmp; 185 int *p; 186 187 tmp.name = s->str; 188 if ((cp = bsearch(&tmp, classes, sizeof(classes) / 189 sizeof(*cp), sizeof(*cp), c_class)) == NULL) 190 errx(1, "unknown class %s", s->str); 191 192 if ((s->set = p = malloc((NCHARS + 1) * sizeof(*p))) == NULL) 193 err(1, "malloc"); 194 195 for (cnt = 0; cnt < NCHARS; ++cnt) 196 if ((*cp->func)(cnt)) 197 *p++ = cnt; 198 *p++ = OOBCH; 199 memset(p, 0, NCHARS + 1 - (p - s->set)); 200 201 s->cnt = 0; 202 s->state = SET; 203 } 204 205 static int 206 c_class(const void *a, const void *b) 207 { 208 return strcmp(((const CLASS *)a)->name, ((const CLASS *)b)->name); 209 } 210 211 /* 212 * English doesn't have any equivalence classes, so for now 213 * we just syntax check and grab the character. 214 */ 215 static void 216 genequiv(STR *s) 217 { 218 if (*s->str == '\\') { 219 s->equiv[0] = backslash(s); 220 if (*s->str != '=') 221 errx(1, "misplaced equivalence equals sign"); 222 } else { 223 s->equiv[0] = s->str[0]; 224 if (s->str[1] != '=') 225 errx(1, "misplaced equivalence equals sign"); 226 } 227 s->str += 2; 228 s->cnt = 0; 229 s->state = SET; 230 s->set = s->equiv; 231 } 232 233 static int 234 genrange(STR *s) 235 { 236 int stopval; 237 char *savestart; 238 239 savestart = s->str; 240 stopval = *++s->str == '\\' ? backslash(s) : *s->str++; 241 if (stopval < (u_char)s->lastch) { 242 s->str = savestart; 243 return 0; 244 } 245 s->cnt = stopval - s->lastch + 1; 246 s->state = RANGE; 247 --s->lastch; 248 return 1; 249 } 250 251 static void 252 genseq(STR *s) 253 { 254 char *ep; 255 256 if (s->which == STRING1) 257 errx(1, "sequences only valid in string2"); 258 259 if (*s->str == '\\') 260 s->lastch = backslash(s); 261 else 262 s->lastch = *s->str++; 263 if (*s->str != '*') 264 errx(1, "misplaced sequence asterisk"); 265 266 switch (*++s->str) { 267 case '\\': 268 s->cnt = backslash(s); 269 break; 270 case ']': 271 s->cnt = 0; 272 ++s->str; 273 break; 274 default: 275 if (isdigit(*s->str)) { 276 s->cnt = strtol(s->str, &ep, 0); 277 if (*ep == ']') { 278 s->str = ep + 1; 279 break; 280 } 281 } 282 errx(1, "illegal sequence count"); 283 /* NOTREACHED */ 284 } 285 286 s->state = s->cnt ? SEQUENCE : INFINITE; 287 } 288 289 /* 290 * Translate \??? into a character. Up to 3 octal digits, if no digits either 291 * an escape code or a literal character. 292 */ 293 static int 294 backslash(STR *s) 295 { 296 int ch, cnt, val; 297 298 for (cnt = val = 0;;) { 299 ch = *++s->str; 300 if (!isascii(ch) || !isdigit(ch)) 301 break; 302 val = val * 8 + ch - '0'; 303 if (++cnt == 3) { 304 ++s->str; 305 break; 306 } 307 } 308 if (cnt) 309 return val; 310 if (ch != '\0') 311 ++s->str; 312 switch (ch) { 313 case 'a': /* escape characters */ 314 return '\7'; 315 case 'b': 316 return '\b'; 317 case 'e': 318 return '\033'; 319 case 'f': 320 return '\f'; 321 case 'n': 322 return '\n'; 323 case 'r': 324 return '\r'; 325 case 't': 326 return '\t'; 327 case 'v': 328 return '\13'; 329 case '\0': /* \" -> \ */ 330 s->state = EOS; 331 return '\\'; 332 default: /* \x" -> x */ 333 return ch; 334 } 335 } 336
Indexes created Mon Sep 22 13:09:51 GMT 2025