regex/regexec-fbsd.c

   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: src/lib/libc/regex/regexec.c,v 1.8 2007/06/11 03:05:54 delphij Exp $");
  41
  42 #include "xlocale_private.h"
  43
  44 /*
  45  * the outer shell of regexec()
  46  *
  47  * This file includes engine.c three times, after muchos fiddling with the
  48  * macros that code uses.  This lets the same code operate on two different
  49  * representations for state sets and characters.
  50  */
  51 #include <sys/types.h>
  52 #include <stdio.h>
  53 #include <stdlib.h>
  54 #include <string.h>
  55 #include <limits.h>
  56 #include <ctype.h>
  57 #include <regex.h>
  58 #include <wchar.h>
  59 #include <wctype.h>
  60
  61 #include "utils.h"
  62 #include "regex2.h"
  63
  64 static int nope __unused = 0;   /* for use in asserts; shuts lint up */
  65
  66 static __inline size_t
  67 xmbrtowc(wint_t *wi, const char *s, size_t n, mbstate_t *mbs, wint_t dummy, locale_t loc)
  68 {
  69         size_t nr;
  70         wchar_t wc;
  71
  72         nr = mbrtowc_l(&wc, s, n, mbs, loc);
  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                 memset(mbs, 0, sizeof(*mbs));
  79                 if (wi != NULL)
  80                         *wi = dummy;
  81                 return (1);
  82         } else
  83                 return (nr);
  84 }
  85
  86 static __inline size_t
  87 xmbrtowc_dummy(wint_t *wi,
  88                 const char *s,
  89                 size_t n __unused,
  90                 mbstate_t *mbs __unused,
  91                 wint_t dummy __unused,
  92                 locale_t loc __unused)
  93 {
  94
  95         if (wi != NULL)
  96                 *wi = (unsigned char)*s;
  97         return (1);
  98 }
  99
 100 /* macros for manipulating states, small version */
 101 #define states  long
 102 #define states1 states          /* for later use in regexec() decision */
 103 #define CLEAR(v)        ((v) = 0)
 104 #define SET0(v, n)      ((v) &= ~((unsigned long)1 << (n)))
 105 #define SET1(v, n)      ((v) |= (unsigned long)1 << (n))
 106 #define ISSET(v, n)     (((v) & ((unsigned long)1 << (n))) != 0)
 107 #define ASSIGN(d, s)    ((d) = (s))
 108 #define EQ(a, b)        ((a) == (b))
 109 #define STATEVARS       long dummy      /* dummy version */
 110 #define STATESETUP(m, n)        /* nothing */
 111 #define STATETEARDOWN(m)        /* nothing */
 112 #define SETUP(v)        ((v) = 0)
 113 #define onestate        long
 114 #define INIT(o, n)      ((o) = (unsigned long)1 << (n))
 115 #define INC(o)  ((o) <<= 1)
 116 #define ISSTATEIN(v, o) (((v) & (o)) != 0)
 117 /* some abbreviations; note that some of these know variable names! */
 118 /* do "if I'm here, I can also be there" etc without branches */
 119 #define FWD(dst, src, n)        ((dst) |= ((unsigned long)(src)&(here)) << (n))
 120 #define BACK(dst, src, n)       ((dst) |= ((unsigned long)(src)&(here)) >> (n))
 121 #define ISSETBACK(v, n) (((v) & ((unsigned long)here >> (n))) != 0)
 122 /* no multibyte support */
 123 #define XMBRTOWC        xmbrtowc_dummy
 124 #define ZAPSTATE(mbs)   ((void)(mbs))
 125 /* function names */
 126 #define SNAMES                  /* engine.c looks after details */
 127
 128 #include "engine.c"
 129
 130 /* now undo things */
 131 #undef  states
 132 #undef  CLEAR
 133 #undef  SET0
 134 #undef  SET1
 135 #undef  ISSET
 136 #undef  ASSIGN
 137 #undef  EQ
 138 #undef  STATEVARS
 139 #undef  STATESETUP
 140 #undef  STATETEARDOWN
 141 #undef  SETUP
 142 #undef  onestate
 143 #undef  INIT
 144 #undef  INC
 145 #undef  ISSTATEIN
 146 #undef  FWD
 147 #undef  BACK
 148 #undef  ISSETBACK
 149 #undef  SNAMES
 150 #undef  XMBRTOWC
 151 #undef  ZAPSTATE
 152
 153 /* macros for manipulating states, large version */
 154 #define states  char *
 155 #define CLEAR(v)        memset(v, 0, m->g->nstates)
 156 #define SET0(v, n)      ((v)[n] = 0)
 157 #define SET1(v, n)      ((v)[n] = 1)
 158 #define ISSET(v, n)     ((v)[n])
 159 #define ASSIGN(d, s)    memcpy(d, s, m->g->nstates)
 160 #define EQ(a, b)        (memcmp(a, b, m->g->nstates) == 0)
 161 #define STATEVARS       long vn; char *space
 162 #define STATESETUP(m, nv)       { (m)->space = malloc((nv)*(m)->g->nstates); \
 163                                 if ((m)->space == NULL) 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 #undef __FBSDID
 183 #define __FBSDID(x)
 184 #include "engine.c"
 185
 186 /* multibyte character & large states version */
 187 #undef  LNAMES
 188 #undef  XMBRTOWC
 189 #undef  ZAPSTATE
 190 #define XMBRTOWC        xmbrtowc
 191 #define ZAPSTATE(mbs)   memset((mbs), 0, sizeof(*(mbs)))
 192 #define MNAMES
 193
 194 #include "engine.c"
 195
 196 /*
 197  - regexec - interface for matching
 198  = extern int regexec(const regex_t *, const char *, size_t, \
 199  =                                      regmatch_t [], int);
 200  = #define      REG_NOTBOL      00001
 201  = #define      REG_NOTEOL      00002
 202  = #define      REG_STARTEND    00004
 203  = #define      REG_TRACE       00400   // tracing of execution
 204  = #define      REG_LARGE       01000   // force large representation
 205  = #define      REG_BACKR       02000   // force use of backref code
 206  *
 207  * We put this here so we can exploit knowledge of the state representation
 208  * when choosing which matcher to call.  Also, by this point the matchers
 209  * have been prototyped.
 210  */
 211 int                             /* 0 success, REG_NOMATCH failure */
 212 regexec(const regex_t * __restrict preg,
 213         const char * __restrict string,
 214         size_t nmatch,
 215         regmatch_t pmatch[__restrict],
 216         int eflags)
 217 {
 218         struct re_guts *g = preg->re_g;
 219 #ifdef REDEBUG
 220 #       define  GOODFLAGS(f)    (f)
 221 #else
 222 #       define  GOODFLAGS(f)    ((f)&(REG_NOTBOL|REG_NOTEOL|REG_STARTEND))
 223 #endif
 224
 225         if (preg->re_magic != MAGIC1 || g->magic != MAGIC2)
 226                 return(REG_BADPAT);
 227         assert(!(g->iflags&BAD));
 228         if (g->iflags&BAD)              /* backstop for no-debug case */
 229                 return(REG_BADPAT);
 230         eflags = GOODFLAGS(eflags);
 231
 232         g->loc = __current_locale();
 233         if (MB_CUR_MAX_L(g->loc) > 1)
 234                 return(mmatcher(g, (char *)string, nmatch, pmatch, eflags));
 235         else if (g->nstates <= CHAR_BIT*sizeof(states1) && !(eflags&REG_LARGE))
 236                 return(smatcher(g, (char *)string, nmatch, pmatch, eflags));
 237         else
 238                 return(lmatcher(g, (char *)string, nmatch, pmatch, eflags));
 239 }