dist/nvi/regex/regexec.c

   1 /*      $NetBSD: regexec.c,v 1.3 2009/01/18 03:45:50 lukem Exp $ */
   2
   3 /*-
   4  * Copyright (c) 1992, 1993, 1994 Henry Spencer.
   5  * Copyright (c) 1992, 1993, 1994
   6  *      The Regents of the University of California.  All rights reserved.
   7  *
   8  * This code is derived from software contributed to Berkeley by
   9  * Henry Spencer of the University of Toronto.
  10  *
  11  * Redistribution and use in source and binary forms, with or without
  12  * modification, are permitted provided that the following conditions
  13  * are met:
  14  * 1. Redistributions of source code must retain the above copyright
  15  *    notice, this list of conditions and the following disclaimer.
  16  * 2. Redistributions in binary form must reproduce the above copyright
  17  *    notice, this list of conditions and the following disclaimer in the
  18  *    documentation and/or other materials provided with the distribution.
  19  * 3. All advertising materials mentioning features or use of this software
  20  *    must display the following acknowledgement:
  21  *      This product includes software developed by the University of
  22  *      California, Berkeley and its contributors.
  23  * 4. Neither the name of the University nor the names of its contributors
  24  *    may be used to endorse or promote products derived from this software
  25  *    without specific prior written permission.
  26  *
  27  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  28  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  29  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  30  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  31  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  32  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  33  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  34  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  35  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  36  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  37  * SUCH DAMAGE.
  38  *
  39  *      @(#)regexec.c   8.2 (Berkeley) 3/16/94
  40  */
  41
  42 #if defined(LIBC_SCCS) && !defined(lint)
  43 static char sccsid[] = "@(#)regexec.c   8.2 (Berkeley) 3/16/94";
  44 #endif /* LIBC_SCCS and not lint */
  45
  46 /*
  47  * the outer shell of regexec()
  48  *
  49  * This file includes engine.c *twice*, after muchos fiddling with the
  50  * macros that code uses.  This lets the same code operate on two different
  51  * representations for state sets.
  52  */
  53 #include <sys/types.h>
  54 #include <stdio.h>
  55 #include <stdlib.h>
  56 #include <string.h>
  57 #include <limits.h>
  58 #include <ctype.h>
  59 #include <regex.h>
  60
  61 #include "utils.h"
  62 #include "regex2.h"
  63
  64 #ifdef notdef
  65 static int nope = 0;            /* for use in asserts; shuts lint up */
  66 #endif
  67
  68 /* macros for manipulating states, small version */
  69 #define states  int
  70 #define states1 int             /* for later use in regexec() decision */
  71 #define CLEAR(v)        ((v) = 0)
  72 #define SET0(v, n)      ((v) &= ~(1 << (n)))
  73 #define SET1(v, n)      ((v) |= 1 << (n))
  74 #define ISSET(v, n)     ((v) & (1 << (n)))
  75 #define ASSIGN(d, s)    ((d) = (s))
  76 #define EQ(a, b)        ((a) == (b))
  77 #define STATEVARS       int dummy       /* dummy version */
  78 #define STATESETUP(m, n)        /* nothing */
  79 #define STATETEARDOWN(m)        /* nothing */
  80 #define SETUP(v)        ((v) = 0)
  81 #define onestate        int
  82 #define INIT(o, n)      ((o) = (unsigned)1 << (n))
  83 #define INC(o)  ((o) <<= 1)
  84 #define ISSTATEIN(v, o) ((v) & (o))
  85 /* some abbreviations; note that some of these know variable names! */
  86 /* do "if I'm here, I can also be there" etc without branches */
  87 #define FWD(dst, src, n)        ((dst) |= ((unsigned)(src)&(here)) << (n))
  88 #define BACK(dst, src, n)       ((dst) |= ((unsigned)(src)&(here)) >> (n))
  89 #define ISSETBACK(v, n) ((v) & ((unsigned)here >> (n)))
  90 /* function names */
  91 #define SNAMES                  /* engine.c looks after details */
  92
  93 #include "engine.c"
  94
  95 /* now undo things */
  96 #undef  states
  97 #undef  CLEAR
  98 #undef  SET0
  99 #undef  SET1
 100 #undef  ISSET
 101 #undef  ASSIGN
 102 #undef  EQ
 103 #undef  STATEVARS
 104 #undef  STATESETUP
 105 #undef  STATETEARDOWN
 106 #undef  SETUP
 107 #undef  onestate
 108 #undef  INIT
 109 #undef  INC
 110 #undef  ISSTATEIN
 111 #undef  FWD
 112 #undef  BACK
 113 #undef  ISSETBACK
 114 #undef  SNAMES
 115
 116 /* macros for manipulating states, large version */
 117 #define states  char *
 118 #define CLEAR(v)        memset(v, 0, m->g->nstates)
 119 #define SET0(v, n)      ((v)[n] = 0)
 120 #define SET1(v, n)      ((v)[n] = 1)
 121 #define ISSET(v, n)     ((v)[n])
 122 #define ASSIGN(d, s)    memcpy(d, s, m->g->nstates)
 123 #define EQ(a, b)        (memcmp(a, b, m->g->nstates) == 0)
 124 #define STATEVARS       int vn; char *space
 125 #define STATESETUP(m, nv)       { (m)->space = malloc((nv)*(m)->g->nstates); \
 126                                 if ((m)->space == NULL) return(REG_ESPACE); \
 127                                 (m)->vn = 0; }
 128 #define STATETEARDOWN(m)        { free((m)->space); }
 129 #define SETUP(v)        ((v) = &m->space[m->vn++ * m->g->nstates])
 130 #define onestate        int
 131 #define INIT(o, n)      ((o) = (n))
 132 #define INC(o)  ((o)++)
 133 #define ISSTATEIN(v, o) ((v)[o])
 134 /* some abbreviations; note that some of these know variable names! */
 135 /* do "if I'm here, I can also be there" etc without branches */
 136 #define FWD(dst, src, n)        ((dst)[here+(n)] |= (src)[here])
 137 #define BACK(dst, src, n)       ((dst)[here-(n)] |= (src)[here])
 138 #define ISSETBACK(v, n) ((v)[here - (n)])
 139 /* function names */
 140 #define LNAMES                  /* flag */
 141
 142 #include "engine.c"
 143
 144 /*
 145  - regexec - interface for matching
 146  = extern int regexec(const regex_t *, const char *, size_t, \
 147  =                                      regmatch_t [], int);
 148  = #define      REG_NOTBOL      00001
 149  = #define      REG_NOTEOL      00002
 150  = #define      REG_STARTEND    00004
 151  = #define      REG_TRACE       00400   // tracing of execution
 152  = #define      REG_LARGE       01000   // force large representation
 153  = #define      REG_BACKR       02000   // force use of backref code
 154  *
 155  * We put this here so we can exploit knowledge of the state representation
 156  * when choosing which matcher to call.  Also, by this point the matchers
 157  * have been prototyped.
 158  */
 159 int                             /* 0 success, REG_NOMATCH failure */
 160 regexec(const regex_t *preg, const RCHAR_T *string, size_t nmatch, regmatch_t *pmatch, int eflags)
 161 {
 162         register struct re_guts *g = preg->re_g;
 163 #ifdef REDEBUG
 164 #       define  GOODFLAGS(f)    (f)
 165 #else
 166 #       define  GOODFLAGS(f)    ((f)&(REG_NOTBOL|REG_NOTEOL|REG_STARTEND))
 167 #endif
 168
 169         if (preg->re_magic != MAGIC1 || g->magic != MAGIC2)
 170                 return(REG_BADPAT);
 171         assert(!(g->iflags&BAD));
 172         if (g->iflags&BAD)              /* backstop for no-debug case */
 173                 return(REG_BADPAT);
 174         eflags = GOODFLAGS(eflags);
 175
 176         if (g->nstates <= (int)(CHAR_BIT*sizeof(states1)) && !(eflags&REG_LARGE))
 177                 return(smatcher(g, (RCHAR_T *)__UNCONST(string), nmatch, pmatch, eflags));
 178         else
 179                 return(lmatcher(g, (RCHAR_T *)__UNCONST(string), nmatch, pmatch, eflags));
 180 }