1 /* $NetBSD: regexec.c,v 1.24 2021/02/24 18:13:21 christos Exp $ */ 2 3 /*- 4 * SPDX-License-Identifier: BSD-3-Clause 5 * 6 * Copyright (c) 1992, 1993, 1994 Henry Spencer. 7 * Copyright (c) 1992, 1993, 1994 8 * The Regents of the University of California. All rights reserved. 9 * 10 * This code is derived from software contributed to Berkeley by 11 * Henry Spencer. 12 * 13 * Redistribution and use in source and binary forms, with or without 14 * modification, are permitted provided that the following conditions 15 * are met: 16 * 1. Redistributions of source code must retain the above copyright 17 * notice, this list of conditions and the following disclaimer. 18 * 2. Redistributions in binary form must reproduce the above copyright 19 * notice, this list of conditions and the following disclaimer in the 20 * documentation and/or other materials provided with the distribution. 21 * 3. Neither the name of the University nor the names of its contributors 22 * may be used to endorse or promote products derived from this software 23 * without specific prior written permission. 24 * 25 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 26 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 27 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 28 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 29 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 30 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 31 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 32 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 33 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 34 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 35 * SUCH DAMAGE. 36 * 37 * @(#)regexec.c 8.3 (Berkeley) 3/20/94 38 */ 39 40 #include <sys/cdefs.h> 41 #if 0 42 static char sccsid[] = "@(#)regexec.c 8.3 (Berkeley) 3/20/94"; 43 __FBSDID("$FreeBSD: head/lib/libc/regex/regexec.c 326025 2017-11-20 19:49:47Z pfg $"); 44 #endif 45 __RCSID("$NetBSD: regexec.c,v 1.24 2021/02/24 18:13:21 christos Exp $"); 46 47 /* 48 * the outer shell of regexec() 49 * 50 * This file includes engine.c three times, after muchos fiddling with the 51 * macros that code uses. This lets the same code operate on two different 52 * representations for state sets and characters. 53 */ 54 55 #include "namespace.h" 56 #include <sys/types.h> 57 #include <stdio.h> 58 #include <stdlib.h> 59 #include <string.h> 60 #include <limits.h> 61 #include <ctype.h> 62 #include <regex.h> 63 #include <wchar.h> 64 #include <wctype.h> 65 66 #ifdef __weak_alias 67 __weak_alias(regexec,_regexec) 68 #endif 69 70 #include "utils.h" 71 #include "regex2.h" 72 73 static __inline size_t 74 xmbrtowc(wint_t *wi, const char *s, size_t n, mbstate_t *mbs, wint_t dummy) 75 { 76 size_t nr; 77 wchar_t wc; 78 79 nr = mbrtowc(&wc, s, n, mbs); 80 if (wi != NULL) 81 *wi = wc; 82 if (nr == 0) 83 return (1); 84 else if (nr == (size_t)-1 || nr == (size_t)-2) { 85 memset(mbs, 0, sizeof(*mbs)); 86 if (wi != NULL) 87 *wi = dummy; 88 return (1); 89 } else 90 return (nr); 91 } 92 93 static __inline size_t 94 xmbrtowc_dummy(wint_t *wi, 95 const char *s, 96 size_t n __unused, 97 mbstate_t *mbs __unused, 98 wint_t dummy __unused) 99 { 100 101 if (wi != NULL) 102 *wi = (unsigned char)*s; 103 return (1); 104 } 105 106 /* macros for manipulating states, small version */ 107 #define states long 108 #define states1 states /* for later use in regexec() decision */ 109 #define CLEAR(v) ((v) = 0) 110 #define SET0(v, n) ((v) &= ~((unsigned long)1 << (n))) 111 #define SET1(v, n) ((v) |= (unsigned long)1 << (n)) 112 #define ISSET(v, n) (((v) & ((unsigned long)1 << (n))) != 0) 113 #define ASSIGN(d, s) ((d) = (s)) 114 #define EQ(a, b) ((a) == (b)) 115 #define STATEVARS long dummy /* dummy version */ 116 #define STATESETUP(m, n) /* nothing */ 117 #define STATETEARDOWN(m) /* nothing */ 118 #define SETUP(v) ((v) = 0) 119 #define onestate long 120 #define INIT(o, n) ((o) = (unsigned long)1 << (n)) 121 #define INC(o) ((o) <<= 1) 122 #define ISSTATEIN(v, o) (((v) & (o)) != 0) 123 /* some abbreviations; note that some of these know variable names! */ 124 /* do "if I'm here, I can also be there" etc without branches */ 125 #define FWD(dst, src, n) ((dst) |= ((unsigned long)(src)&(here)) << (n)) 126 #define BACK(dst, src, n) ((dst) |= ((unsigned long)(src)&(here)) >> (n)) 127 #define ISSETBACK(v, n) (((v) & ((unsigned long)here >> (n))) != 0) 128 /* no multibyte support */ 129 #define XMBRTOWC xmbrtowc_dummy 130 #define ZAPSTATE(mbs) ((void)(mbs)) 131 /* function names */ 132 #define SNAMES /* engine.c looks after details */ 133 134 #include "engine.c" 135 136 /* now undo things */ 137 #undef states 138 #undef CLEAR 139 #undef SET0 140 #undef SET1 141 #undef ISSET 142 #undef ASSIGN 143 #undef EQ 144 #undef STATEVARS 145 #undef STATESETUP 146 #undef STATETEARDOWN 147 #undef SETUP 148 #undef onestate 149 #undef INIT 150 #undef INC 151 #undef ISSTATEIN 152 #undef FWD 153 #undef BACK 154 #undef ISSETBACK 155 #undef SNAMES 156 #undef XMBRTOWC 157 #undef ZAPSTATE 158 159 /* macros for manipulating states, large version */ 160 #define states char * 161 #define CLEAR(v) memset(v, 0, m->g->nstates) 162 #define SET0(v, n) ((v)[n] = 0) 163 #define SET1(v, n) ((v)[n] = 1) 164 #define ISSET(v, n) ((v)[n]) 165 #define ASSIGN(d, s) memcpy(d, s, m->g->nstates) 166 #define EQ(a, b) (memcmp(a, b, m->g->nstates) == 0) 167 #define STATEVARS long vn; char *space 168 #define STATESETUP(m, nv) { (m)->space = malloc((nv)*(m)->g->nstates); \ 169 if ((m)->space == NULL) return(REG_ESPACE); \ 170 (m)->vn = 0; } 171 #define STATETEARDOWN(m) { free((m)->space); } 172 #define SETUP(v) ((v) = &m->space[m->vn++ * m->g->nstates]) 173 #define onestate long 174 #define INIT(o, n) ((o) = (n)) 175 #define INC(o) ((o)++) 176 #define ISSTATEIN(v, o) ((v)[o]) 177 /* some abbreviations; note that some of these know variable names! */ 178 /* do "if I'm here, I can also be there" etc without branches */ 179 #define FWD(dst, src, n) ((dst)[here+(n)] |= (src)[here]) 180 #define BACK(dst, src, n) ((dst)[here-(n)] |= (src)[here]) 181 #define ISSETBACK(v, n) ((v)[here - (n)]) 182 /* no multibyte support */ 183 #define XMBRTOWC xmbrtowc_dummy 184 #define ZAPSTATE(mbs) ((void)(mbs)) 185 /* function names */ 186 #define LNAMES /* flag */ 187 188 #include "engine.c" 189 190 /* multibyte character & large states version */ 191 #undef LNAMES 192 #undef XMBRTOWC 193 #undef ZAPSTATE 194 #define XMBRTOWC xmbrtowc 195 #define ZAPSTATE(mbs) memset((mbs), 0, sizeof(*(mbs))) 196 #define MNAMES 197 198 #include "engine.c" 199 200 /* 201 - regexec - interface for matching 202 = extern int regexec(const regex_t *, const char *, size_t, \ 203 = regmatch_t [], int); 204 = #define REG_NOTBOL 00001 205 = #define REG_NOTEOL 00002 206 = #define REG_STARTEND 00004 207 = #define REG_TRACE 00400 // tracing of execution 208 = #define REG_LARGE 01000 // force large representation 209 = #define REG_BACKR 02000 // force use of backref code 210 * 211 * We put this here so we can exploit knowledge of the state representation 212 * when choosing which matcher to call. Also, by this point the matchers 213 * have been prototyped. 214 */ 215 int /* 0 success, REG_NOMATCH failure */ 216 regexec(const regex_t * __restrict preg, 217 const char * __restrict string, 218 size_t nmatch, 219 regmatch_t pmatch[__restrict], 220 int eflags) 221 { 222 struct re_guts *g = preg->re_g; 223 #ifdef REDEBUG 224 # define GOODFLAGS(f) (f) 225 #else 226 # define GOODFLAGS(f) ((f)&(REG_NOTBOL|REG_NOTEOL|REG_STARTEND)) 227 #endif 228 _DIAGASSERT(preg != NULL); 229 _DIAGASSERT(string != NULL); 230 231 if (preg->re_magic != MAGIC1 || g->magic != MAGIC2) 232 return(REG_BADPAT); 233 assert(!(g->iflags&BAD)); 234 if (g->iflags&BAD) /* backstop for no-debug case */ 235 return(REG_BADPAT); 236 eflags = GOODFLAGS(eflags); 237 238 if (MB_CUR_MAX > 1) 239 return(mmatcher(g, string, nmatch, pmatch, eflags)); 240 else if (g->nstates <= CHAR_BIT*sizeof(states1) && !(eflags®_LARGE)) 241 return(smatcher(g, string, nmatch, pmatch, eflags)); 242 else 243 return(lmatcher(g, string, nmatch, pmatch, eflags)); 244 } 245