1 /*- 2 * Copyright (c) 1992, 1993, 1994 Henry Spencer. 3 * Copyright (c) 1992, 1993, 1994 4 * The Regents of the University of California. All rights reserved. 5 * 6 * This code is derived from software contributed to Berkeley by 7 * Henry Spencer. 8 * 9 * Redistribution and use in source and binary forms, with or without 10 * modification, are permitted provided that the following conditions 11 * are met: 12 * 1. Redistributions of source code must retain the above copyright 13 * notice, this list of conditions and the following disclaimer. 14 * 2. Redistributions in binary form must reproduce the above copyright 15 * notice, this list of conditions and the following disclaimer in the 16 * documentation and/or other materials provided with the distribution. 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 * @(#)regexec.c 8.3 (Berkeley) 3/20/94 34 */ 35 36 #if defined(LIBC_SCCS) && !defined(lint) 37 static char sccsid[] = "@(#)regexec.c 8.3 (Berkeley) 3/20/94"; 38 #endif /* LIBC_SCCS and not lint */ 39 #include <sys/cdefs.h> 40 __FBSDID("$FreeBSD$"); 41 42 /* 43 * the outer shell of regexec() 44 * 45 * This file includes engine.c three times, after muchos fiddling with the 46 * macros that code uses. This lets the same code operate on two different 47 * representations for state sets and characters. 48 */ 49 #include <sys/types.h> 50 #include <stdio.h> 51 #include <stdlib.h> 52 #include <string.h> 53 #include <limits.h> 54 #include <ctype.h> 55 #include <regex.h> 56 #include <wchar.h> 57 #include <wctype.h> 58 59 #include "utils.h" 60 #include "regex2.h" 61 62 static int nope __unused = 0; /* for use in asserts; shuts lint up */ 63 64 static __inline size_t 65 xmbrtowc(wint_t *wi, const char *s, size_t n, mbstate_t *mbs, wint_t dummy) 66 { 67 size_t nr; 68 wchar_t wc; 69 70 nr = mbrtowc(&wc, s, n, mbs); 71 if (wi != NULL) 72 *wi = wc; 73 if (nr == 0) 74 return (1); 75 else if (nr == (size_t)-1 || nr == (size_t)-2) { 76 memset(mbs, 0, sizeof(*mbs)); 77 if (wi != NULL) 78 *wi = dummy; 79 return (1); 80 } else 81 return (nr); 82 } 83 84 static __inline size_t 85 xmbrtowc_dummy(wint_t *wi, 86 const char *s, 87 size_t n __unused, 88 mbstate_t *mbs __unused, 89 wint_t dummy __unused) 90 { 91 92 if (wi != NULL) 93 *wi = (unsigned char)*s; 94 return (1); 95 } 96 97 /* macros for manipulating states, small version */ 98 #define states long 99 #define states1 states /* for later use in regexec() decision */ 100 #define CLEAR(v) ((v) = 0) 101 #define SET0(v, n) ((v) &= ~((unsigned long)1 << (n))) 102 #define SET1(v, n) ((v) |= (unsigned long)1 << (n)) 103 #define ISSET(v, n) (((v) & ((unsigned long)1 << (n))) != 0) 104 #define ASSIGN(d, s) ((d) = (s)) 105 #define EQ(a, b) ((a) == (b)) 106 #define STATEVARS long dummy /* dummy version */ 107 #define STATESETUP(m, n) /* nothing */ 108 #define STATETEARDOWN(m) /* nothing */ 109 #define SETUP(v) ((v) = 0) 110 #define onestate long 111 #define INIT(o, n) ((o) = (unsigned long)1 << (n)) 112 #define INC(o) ((o) <<= 1) 113 #define ISSTATEIN(v, o) (((v) & (o)) != 0) 114 /* some abbreviations; note that some of these know variable names! */ 115 /* do "if I'm here, I can also be there" etc without branches */ 116 #define FWD(dst, src, n) ((dst) |= ((unsigned long)(src)&(here)) << (n)) 117 #define BACK(dst, src, n) ((dst) |= ((unsigned long)(src)&(here)) >> (n)) 118 #define ISSETBACK(v, n) (((v) & ((unsigned long)here >> (n))) != 0) 119 /* no multibyte support */ 120 #define XMBRTOWC xmbrtowc_dummy 121 #define ZAPSTATE(mbs) ((void)(mbs)) 122 /* function names */ 123 #define SNAMES /* engine.c looks after details */ 124 125 #include "engine.c" 126 127 /* now undo things */ 128 #undef states 129 #undef CLEAR 130 #undef SET0 131 #undef SET1 132 #undef ISSET 133 #undef ASSIGN 134 #undef EQ 135 #undef STATEVARS 136 #undef STATESETUP 137 #undef STATETEARDOWN 138 #undef SETUP 139 #undef onestate 140 #undef INIT 141 #undef INC 142 #undef ISSTATEIN 143 #undef FWD 144 #undef BACK 145 #undef ISSETBACK 146 #undef SNAMES 147 #undef XMBRTOWC 148 #undef ZAPSTATE 149 150 /* macros for manipulating states, large version */ 151 #define states char * 152 #define CLEAR(v) memset(v, 0, m->g->nstates) 153 #define SET0(v, n) ((v)[n] = 0) 154 #define SET1(v, n) ((v)[n] = 1) 155 #define ISSET(v, n) ((v)[n]) 156 #define ASSIGN(d, s) memcpy(d, s, m->g->nstates) 157 #define EQ(a, b) (memcmp(a, b, m->g->nstates) == 0) 158 #define STATEVARS long vn; char *space 159 #define STATESETUP(m, nv) { (m)->space = malloc((nv)*(m)->g->nstates); \ 160 if ((m)->space == NULL) return(REG_ESPACE); \ 161 (m)->vn = 0; } 162 #define STATETEARDOWN(m) { free((m)->space); } 163 #define SETUP(v) ((v) = &m->space[m->vn++ * m->g->nstates]) 164 #define onestate long 165 #define INIT(o, n) ((o) = (n)) 166 #define INC(o) ((o)++) 167 #define ISSTATEIN(v, o) ((v)[o]) 168 /* some abbreviations; note that some of these know variable names! */ 169 /* do "if I'm here, I can also be there" etc without branches */ 170 #define FWD(dst, src, n) ((dst)[here+(n)] |= (src)[here]) 171 #define BACK(dst, src, n) ((dst)[here-(n)] |= (src)[here]) 172 #define ISSETBACK(v, n) ((v)[here - (n)]) 173 /* no multibyte support */ 174 #define XMBRTOWC xmbrtowc_dummy 175 #define ZAPSTATE(mbs) ((void)(mbs)) 176 /* function names */ 177 #define LNAMES /* flag */ 178 179 #include "engine.c" 180 181 /* multibyte character & large states version */ 182 #undef LNAMES 183 #undef XMBRTOWC 184 #undef ZAPSTATE 185 #define XMBRTOWC xmbrtowc 186 #define ZAPSTATE(mbs) memset((mbs), 0, sizeof(*(mbs))) 187 #define MNAMES 188 189 #include "engine.c" 190 191 /* 192 - regexec - interface for matching 193 = extern int regexec(const regex_t *, const char *, size_t, \ 194 = regmatch_t [], int); 195 = #define REG_NOTBOL 00001 196 = #define REG_NOTEOL 00002 197 = #define REG_STARTEND 00004 198 = #define REG_TRACE 00400 // tracing of execution 199 = #define REG_LARGE 01000 // force large representation 200 = #define REG_BACKR 02000 // force use of backref code 201 * 202 * We put this here so we can exploit knowledge of the state representation 203 * when choosing which matcher to call. Also, by this point the matchers 204 * have been prototyped. 205 */ 206 int /* 0 success, REG_NOMATCH failure */ 207 regexec(const regex_t * __restrict preg, 208 const char * __restrict string, 209 size_t nmatch, 210 regmatch_t pmatch[__restrict], 211 int eflags) 212 { 213 struct re_guts *g = preg->re_g; 214 #ifdef REDEBUG 215 # define GOODFLAGS(f) (f) 216 #else 217 # define GOODFLAGS(f) ((f)&(REG_NOTBOL|REG_NOTEOL|REG_STARTEND)) 218 #endif 219 220 if (preg->re_magic != MAGIC1 || g->magic != MAGIC2) 221 return(REG_BADPAT); 222 assert(!(g->iflags&BAD)); 223 if (g->iflags&BAD) /* backstop for no-debug case */ 224 return(REG_BADPAT); 225 eflags = GOODFLAGS(eflags); 226 227 if (MB_CUR_MAX > 1) 228 return(mmatcher(g, (char *)string, nmatch, pmatch, eflags)); 229 else if (g->nstates <= CHAR_BIT*sizeof(states1) && !(eflags®_LARGE)) 230 return(smatcher(g, (char *)string, nmatch, pmatch, eflags)); 231 else 232 return(lmatcher(g, (char *)string, nmatch, pmatch, eflags)); 233 } 234