lib/libc/regex/regexec.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  * $FreeBSD: src/lib/libc/regex/regexec.c,v 1.8 2007/06/11 03:05:54 delphij Exp $
  35  * $DragonFly: src/lib/libc/regex/regexec.c,v 1.4 2005/11/20 09:18:37 swildner Exp $
  36  */
  37
  38 /*
  39  * the outer shell of regexec()
  40  *
  41  * This file includes engine.c three times, after muchos fiddling with the
  42  * macros that code uses.  This lets the same code operate on two different
  43  * representations for state sets and characters.
  44  */
  45 #include <sys/types.h>
  46 #include <stdio.h>
  47 #include <stdlib.h>
  48 #include <string.h>
  49 #include <limits.h>
  50 #include <ctype.h>
  51 #include <regex.h>
  52 #include <wchar.h>
  53 #include <wctype.h>
  54
  55 #include "utils.h"
  56 #include "regex2.h"
  57
  58 static int nope __unused = 0;   /* for use in asserts; shuts lint up */
  59
  60 static __inline size_t
  61 xmbrtowc(wint_t *wi, const char *s, size_t n, mbstate_t *mbs, wint_t dummy)
  62 {
  63         size_t nr;
  64         wchar_t wc;
  65
  66         nr = mbrtowc(&wc, s, n, mbs);
  67         if (wi != NULL)
  68                 *wi = wc;
  69         if (nr == 0)
  70                 return (1);
  71         else if (nr == (size_t)-1 || nr == (size_t)-2) {
  72                 memset(mbs, 0, sizeof(*mbs));
  73                 if (wi != NULL)
  74                         *wi = dummy;
  75                 return (1);
  76         } else
  77                 return (nr);
  78 }
  79
  80 static __inline size_t
  81 xmbrtowc_dummy(wint_t *wi,
  82                 const char *s,
  83                 size_t n __unused,
  84                 mbstate_t *mbs __unused,
  85                 wint_t dummy __unused)
  86 {
  87
  88         if (wi != NULL)
  89                 *wi = (unsigned char)*s;
  90         return (1);
  91 }
  92
  93 /* macros for manipulating states, small version */
  94 #define states  long
  95 #define states1 states          /* for later use in regexec() decision */
  96 #define CLEAR(v)        ((v) = 0)
  97 #define SET0(v, n)      ((v) &= ~((unsigned long)1 << (n)))
  98 #define SET1(v, n)      ((v) |= (unsigned long)1 << (n))
  99 #define ISSET(v, n)     (((v) & ((unsigned long)1 << (n))) != 0)
 100 #define ASSIGN(d, s)    ((d) = (s))
 101 #define EQ(a, b)        ((a) == (b))
 102 #define STATEVARS       long dummy      /* dummy version */
 103 #define STATESETUP(m, n)        /* nothing */
 104 #define STATETEARDOWN(m)        /* nothing */
 105 #define SETUP(v)        ((v) = 0)
 106 #define onestate        long
 107 #define INIT(o, n)      ((o) = (unsigned long)1 << (n))
 108 #define INC(o)  ((o) <<= 1)
 109 #define ISSTATEIN(v, o) (((v) & (o)) != 0)
 110 /* some abbreviations; note that some of these know variable names! */
 111 /* do "if I'm here, I can also be there" etc without branches */
 112 #define FWD(dst, src, n)        ((dst) |= ((unsigned long)(src)&(here)) << (n))
 113 #define BACK(dst, src, n)       ((dst) |= ((unsigned long)(src)&(here)) >> (n))
 114 #define ISSETBACK(v, n) (((v) & ((unsigned long)here >> (n))) != 0)
 115 /* no multibyte support */
 116 #define XMBRTOWC        xmbrtowc_dummy
 117 #define ZAPSTATE(mbs)   ((void)(mbs))
 118 /* function names */
 119 #define SNAMES                  /* engine.c looks after details */
 120
 121 #include "engine.c"
 122
 123 /* now undo things */
 124 #undef  states
 125 #undef  CLEAR
 126 #undef  SET0
 127 #undef  SET1
 128 #undef  ISSET
 129 #undef  ASSIGN
 130 #undef  EQ
 131 #undef  STATEVARS
 132 #undef  STATESETUP
 133 #undef  STATETEARDOWN
 134 #undef  SETUP
 135 #undef  onestate
 136 #undef  INIT
 137 #undef  INC
 138 #undef  ISSTATEIN
 139 #undef  FWD
 140 #undef  BACK
 141 #undef  ISSETBACK
 142 #undef  SNAMES
 143 #undef  XMBRTOWC
 144 #undef  ZAPSTATE
 145
 146 /* macros for manipulating states, large version */
 147 #define states  char *
 148 #define CLEAR(v)        memset(v, 0, m->g->nstates)
 149 #define SET0(v, n)      ((v)[n] = 0)
 150 #define SET1(v, n)      ((v)[n] = 1)
 151 #define ISSET(v, n)     ((v)[n])
 152 #define ASSIGN(d, s)    memcpy(d, s, m->g->nstates)
 153 #define EQ(a, b)        (memcmp(a, b, m->g->nstates) == 0)
 154 #define STATEVARS       long vn; char *space
 155 #define STATESETUP(m, nv)       { (m)->space = malloc((nv)*(m)->g->nstates); \
 156                                 if ((m)->space == NULL) return(REG_ESPACE); \
 157                                 (m)->vn = 0; }
 158 #define STATETEARDOWN(m)        { free((m)->space); }
 159 #define SETUP(v)        ((v) = &m->space[m->vn++ * m->g->nstates])
 160 #define onestate        long
 161 #define INIT(o, n)      ((o) = (n))
 162 #define INC(o)  ((o)++)
 163 #define ISSTATEIN(v, o) ((v)[o])
 164 /* some abbreviations; note that some of these know variable names! */
 165 /* do "if I'm here, I can also be there" etc without branches */
 166 #define FWD(dst, src, n)        ((dst)[here+(n)] |= (src)[here])
 167 #define BACK(dst, src, n)       ((dst)[here-(n)] |= (src)[here])
 168 #define ISSETBACK(v, n) ((v)[here - (n)])
 169 /* no multibyte support */
 170 #define XMBRTOWC        xmbrtowc_dummy
 171 #define ZAPSTATE(mbs)   ((void)(mbs))
 172 /* function names */
 173 #define LNAMES                  /* flag */
 174
 175 #include "engine.c"
 176
 177 /* multibyte character & large states version */
 178 #undef  LNAMES
 179 #undef  XMBRTOWC
 180 #undef  ZAPSTATE
 181 #define XMBRTOWC        xmbrtowc
 182 #define ZAPSTATE(mbs)   memset((mbs), 0, sizeof(*(mbs)))
 183 #define MNAMES
 184
 185 #include "engine.c"
 186
 187 /*
 188  - regexec - interface for matching
 189  = extern int regexec(const regex_t *, const char *, size_t, \
 190  =                                      regmatch_t [], int);
 191  = #define      REG_NOTBOL      00001
 192  = #define      REG_NOTEOL      00002
 193  = #define      REG_STARTEND    00004
 194  = #define      REG_TRACE       00400   // tracing of execution
 195  = #define      REG_LARGE       01000   // force large representation
 196  = #define      REG_BACKR       02000   // force use of backref code
 197  *
 198  * We put this here so we can exploit knowledge of the state representation
 199  * when choosing which matcher to call.  Also, by this point the matchers
 200  * have been prototyped.
 201  */
 202 int                             /* 0 success, REG_NOMATCH failure */
 203 regexec(const regex_t * __restrict preg,
 204         const char * __restrict string,
 205         size_t nmatch,
 206         regmatch_t pmatch[__restrict],
 207         int eflags)
 208 {
 209         struct re_guts *g = preg->re_g;
 210 #ifdef REDEBUG
 211 #       define  GOODFLAGS(f)    (f)
 212 #else
 213 #       define  GOODFLAGS(f)    ((f)&(REG_NOTBOL|REG_NOTEOL|REG_STARTEND))
 214 #endif
 215
 216         if (preg->re_magic != MAGIC1 || g->magic != MAGIC2)
 217                 return(REG_BADPAT);
 218         assert(!(g->iflags&BAD));
 219         if (g->iflags&BAD)              /* backstop for no-debug case */
 220                 return(REG_BADPAT);
 221         eflags = GOODFLAGS(eflags);
 222
 223         if (MB_CUR_MAX > 1)
 224                 return(mmatcher(g, (char *)string, nmatch, pmatch, eflags));
 225         else if (g->nstates <= CHAR_BIT*sizeof(states1) && !(eflags&REG_LARGE))
 226                 return(smatcher(g, (char *)string, nmatch, pmatch, eflags));
 227         else
 228                 return(lmatcher(g, (char *)string, nmatch, pmatch, eflags));
 229 }