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 34 /* 35 * Copyright 2010 Nexenta Systems, Inc. All rights reserved. 36 * Use is subject to license terms. 37 */ 38 39 /* 40 * the outer shell of regexec() 41 * 42 * This file includes engine.c three times, after muchos fiddling with the 43 * macros that code uses. This lets the same code operate on two different 44 * representations for state sets and characters. 45 */ 46 #include "lint.h" 47 #include "file64.h" 48 #include <sys/types.h> 49 #include <stdio.h> 50 #include <stdlib.h> 51 #include <string.h> 52 #include <limits.h> 53 #include <ctype.h> 54 #include <regex.h> 55 #include <wchar.h> 56 #include <wctype.h> 57 #include <note.h> 58 #include <assert.h> 59 60 #include "utils.h" 61 #include "regex2.h" 62 63 /* we want _NOTE, but not NOTE (which collides with our own use) */ 64 #undef NOTE 65 66 static size_t 67 xmbrtowc(wint_t *wi, const char *s, size_t n, mbstate_t *mbs, wint_t dummy) 68 { 69 size_t nr; 70 wchar_t wc; 71 72 nr = mbrtowc(&wc, s, n, mbs); 73 if (wi != NULL) 74 *wi = wc; 75 if (nr == 0) 76 return (1); 77 else if (nr == (size_t)-1 || nr == (size_t)-2) { 78 (void) memset(mbs, 0, sizeof (*mbs)); 79 if (wi != NULL) 80 *wi = dummy; 81 return (1); 82 } else 83 return (nr); 84 } 85 86 static size_t 87 xmbrtowc_dummy(wint_t *wi, const char *s, size_t n, mbstate_t *mbs, 88 wint_t dummy) 89 { 90 _NOTE(ARGUNUSED(n)); 91 _NOTE(ARGUNUSED(mbs)); 92 _NOTE(ARGUNUSED(dummy)); 93 94 if (wi != NULL) 95 *wi = (unsigned char)*s; 96 return (1); 97 } 98 99 /* macros for manipulating states, small version */ 100 #define states long 101 #define states1 states /* for later use in regexec() decision */ 102 #define CLEAR(v) ((v) = 0) 103 #define SET0(v, n) ((v) &= ~((unsigned long)1 << (n))) 104 #define SET1(v, n) ((v) |= (unsigned long)1 << (n)) 105 #define ISSET(v, n) (((v) & ((unsigned long)1 << (n))) != 0) 106 #define ASSIGN(d, s) ((d) = (s)) 107 #define EQ(a, b) ((a) == (b)) 108 #define STATEVARS long dummy /* dummy version */ 109 #define STATESETUP(m, n) /* nothing */ 110 #define STATETEARDOWN(m) /* nothing */ 111 #define SETUP(v) ((v) = 0) 112 #define onestate long 113 #define INIT(o, n) ((o) = (unsigned long)1 << (n)) 114 #define INC(o) ((o) <<= 1) 115 #define ISSTATEIN(v, o) (((v) & (o)) != 0) 116 /* some abbreviations; note that some of these know variable names! */ 117 /* do "if I'm here, I can also be there" etc without branches */ 118 #define FWD(dst, src, n) ((dst) |= ((unsigned long)(src)&(here)) << (n)) 119 #define BACK(dst, src, n) ((dst) |= ((unsigned long)(src)&(here)) >> (n)) 120 #define ISSETBACK(v, n) (((v) & ((unsigned long)here >> (n))) != 0) 121 /* no multibyte support */ 122 #define XMBRTOWC xmbrtowc_dummy 123 #define ZAPSTATE(mbs) ((void)(mbs)) 124 /* function names */ 125 #define SNAMES /* engine.c looks after details */ 126 127 #include "engine.c" 128 129 /* now undo things */ 130 #undef states 131 #undef CLEAR 132 #undef SET0 133 #undef SET1 134 #undef ISSET 135 #undef ASSIGN 136 #undef EQ 137 #undef STATEVARS 138 #undef STATESETUP 139 #undef STATETEARDOWN 140 #undef SETUP 141 #undef onestate 142 #undef INIT 143 #undef INC 144 #undef ISSTATEIN 145 #undef FWD 146 #undef BACK 147 #undef ISSETBACK 148 #undef SNAMES 149 #undef XMBRTOWC 150 #undef ZAPSTATE 151 152 /* macros for manipulating states, large version */ 153 #define states char * 154 #define CLEAR(v) (void) memset(v, 0, m->g->nstates) 155 #define SET0(v, n) ((v)[n] = 0) 156 #define SET1(v, n) ((v)[n] = 1) 157 #define ISSET(v, n) ((v)[n]) 158 #define ASSIGN(d, s) (void) memcpy(d, s, m->g->nstates) 159 #define EQ(a, b) (memcmp(a, b, m->g->nstates) == 0) 160 #define STATEVARS long vn; char *space 161 #define STATESETUP(m, nv) { (m)->space = malloc((nv)*(m)->g->nstates); \ 162 if ((m)->space == NULL) \ 163 return (REG_ESPACE); \ 164 (m)->vn = 0; } 165 #define STATETEARDOWN(m) { free((m)->space); } 166 #define SETUP(v) ((v) = &m->space[m->vn++ * m->g->nstates]) 167 #define onestate long 168 #define INIT(o, n) ((o) = (n)) 169 #define INC(o) ((o)++) 170 #define ISSTATEIN(v, o) ((v)[o]) 171 /* some abbreviations; note that some of these know variable names! */ 172 /* do "if I'm here, I can also be there" etc without branches */ 173 #define FWD(dst, src, n) ((dst)[here+(n)] |= (src)[here]) 174 #define BACK(dst, src, n) ((dst)[here-(n)] |= (src)[here]) 175 #define ISSETBACK(v, n) ((v)[here - (n)]) 176 /* no multibyte support */ 177 #define XMBRTOWC xmbrtowc_dummy 178 #define ZAPSTATE(mbs) ((void)(mbs)) 179 /* function names */ 180 #define LNAMES /* flag */ 181 182 #include "engine.c" 183 184 /* multibyte character & large states version */ 185 #undef LNAMES 186 #undef XMBRTOWC 187 #undef ZAPSTATE 188 #define XMBRTOWC xmbrtowc 189 #define ZAPSTATE(mbs) (void) memset((mbs), 0, sizeof (*(mbs))) 190 #define MNAMES 191 192 #include "engine.c" 193 194 /* 195 * regexec - interface for matching 196 * 197 * We put this here so we can exploit knowledge of the state representation 198 * when choosing which matcher to call. Also, by this point the matchers 199 * have been prototyped. 200 */ 201 int /* 0 success, REG_NOMATCH failure */ 202 regexec(const regex_t *_RESTRICT_KYWD preg, 203 const char *_RESTRICT_KYWD string, size_t nmatch, 204 regmatch_t pmatch[_RESTRICT_KYWD], int eflags) 205 { 206 struct re_guts *g = preg->re_g; 207 #ifdef REDEBUG 208 #define GOODFLAGS(f) (f) 209 #else 210 #ifdef REG_STARTEND 211 #define GOODFLAGS(f) ((f)&(REG_NOTBOL|REG_NOTEOL|REG_STARTEND)) 212 #else 213 #define GOODFLAGS(f) ((f)&(REG_NOTBOL|REG_NOTEOL)) 214 #endif 215 #endif 216 217 if (preg->re_magic != MAGIC1 || g->magic != MAGIC2) 218 return (REG_BADPAT); 219 assert(!(g->iflags&BAD)); 220 if (g->iflags&BAD) /* backstop for no-debug case */ 221 return (REG_BADPAT); 222 eflags = GOODFLAGS(eflags); 223 224 if (MB_CUR_MAX > 1) 225 return (mmatcher(g, (char *)string, nmatch, pmatch, eflags)); 226 #ifdef REG_LARGE 227 else if (g->nstates <= CHAR_BIT*sizeof (states1) && !(eflags®_LARGE)) 228 #else 229 else if (g->nstates <= CHAR_BIT*sizeof (states1)) 230 #endif 231 return (smatcher(g, (char *)string, nmatch, pmatch, eflags)); 232 else 233 return (lmatcher(g, (char *)string, nmatch, pmatch, eflags)); 234 } 235