regex/regexec.c

   1 /*
   2  * Copyright (c) 1999 Apple Computer, Inc. All rights reserved.
   3  *
   4  * @APPLE_LICENSE_HEADER_START@
   5  *
   6  * Copyright (c) 1999-2003 Apple Computer, Inc.  All Rights Reserved.
   7  *
   8  * This file contains Original Code and/or Modifications of Original Code
   9  * as defined in and that are subject to the Apple Public Source License
  10  * Version 2.0 (the 'License'). You may not use this file except in
  11  * compliance with the License. Please obtain a copy of the License at
  12  * http://www.opensource.apple.com/apsl/ and read it before using this
  13  * file.
  14  *
  15  * The Original Code and all software distributed under the License are
  16  * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
  17  * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
  18  * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
  19  * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
  20  * Please see the License for the specific language governing rights and
  21  * limitations under the License.
  22  *
  23  * @APPLE_LICENSE_HEADER_END@
  24  */
  25 /*
  26  * Copyright (c) 1992, 1993, 1994
  27  *      The Regents of the University of California.  All rights reserved.
  28  *
  29  * This code is derived from software contributed to Berkeley by
  30  * Henry Spencer.
  31  *
  32  * Redistribution and use in source and binary forms, with or without
  33  * modification, are permitted provided that the following conditions
  34  * are met:
  35  * 1. Redistributions of source code must retain the above copyright
  36  *    notice, this list of conditions and the following disclaimer.
  37  * 2. Redistributions in binary form must reproduce the above copyright
  38  *    notice, this list of conditions and the following disclaimer in the
  39  *    documentation and/or other materials provided with the distribution.
  40  * 3. All advertising materials mentioning features or use of this software
  41  *    must display the following acknowledgement:
  42  *      This product includes software developed by the University of
  43  *      California, Berkeley and its contributors.
  44  * 4. Neither the name of the University nor the names of its contributors
  45  *    may be used to endorse or promote products derived from this software
  46  *    without specific prior written permission.
  47  *
  48  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  49  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  50  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  51  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  52  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  53  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  54  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  55  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  56  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  57  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  58  * SUCH DAMAGE.
  59  */
  60
  61 /*
  62  * the outer shell of regexec()
  63  *
  64  * This file includes engine.c *twice*, after muchos fiddling with the
  65  * macros that code uses.  This lets the same code operate on two different
  66  * representations for state sets.
  67  */
  68 #include <sys/types.h>
  69 #include <stdio.h>
  70 #include <stdlib.h>
  71 #include <string.h>
  72 #include <limits.h>
  73 #include <ctype.h>
  74 #include <regex.h>
  75
  76 #include "utils.h"
  77 #include "regex2.h"
  78
  79 static int nope = 0;            /* for use in asserts; shuts lint up */
  80
  81 /* macros for manipulating states, small version */
  82 #define states  long
  83 #define states1 states          /* for later use in regexec() decision */
  84 #define CLEAR(v)        ((v) = 0)
  85 #define SET0(v, n)      ((v) &= ~(1 << (n)))
  86 #define SET1(v, n)      ((v) |= 1 << (n))
  87 #define ISSET(v, n)     ((v) & (1 << (n)))
  88 #define ASSIGN(d, s)    ((d) = (s))
  89 #define EQ(a, b)        ((a) == (b))
  90 #define STATEVARS       int dummy       /* dummy version */
  91 #define STATESETUP(m, n)        /* nothing */
  92 #define STATETEARDOWN(m)        /* nothing */
  93 #define SETUP(v)        ((v) = 0)
  94 #define onestate        int
  95 #define INIT(o, n)      ((o) = (unsigned)1 << (n))
  96 #define INC(o)  ((o) <<= 1)
  97 #define ISSTATEIN(v, o) ((v) & (o))
  98 /* some abbreviations; note that some of these know variable names! */
  99 /* do "if I'm here, I can also be there" etc without branches */
 100 #define FWD(dst, src, n)        ((dst) |= ((unsigned)(src)&(here)) << (n))
 101 #define BACK(dst, src, n)       ((dst) |= ((unsigned)(src)&(here)) >> (n))
 102 #define ISSETBACK(v, n) ((v) & ((unsigned)here >> (n)))
 103 /* function names */
 104 #define SNAMES                  /* engine.c looks after details */
 105
 106 #include "engine.c"
 107
 108 /* now undo things */
 109 #undef  states
 110 #undef  CLEAR
 111 #undef  SET0
 112 #undef  SET1
 113 #undef  ISSET
 114 #undef  ASSIGN
 115 #undef  EQ
 116 #undef  STATEVARS
 117 #undef  STATESETUP
 118 #undef  STATETEARDOWN
 119 #undef  SETUP
 120 #undef  onestate
 121 #undef  INIT
 122 #undef  INC
 123 #undef  ISSTATEIN
 124 #undef  FWD
 125 #undef  BACK
 126 #undef  ISSETBACK
 127 #undef  SNAMES
 128
 129 /* macros for manipulating states, large version */
 130 #define states  char *
 131 #define CLEAR(v)        memset(v, 0, m->g->nstates)
 132 #define SET0(v, n)      ((v)[n] = 0)
 133 #define SET1(v, n)      ((v)[n] = 1)
 134 #define ISSET(v, n)     ((v)[n])
 135 #define ASSIGN(d, s)    memcpy(d, s, m->g->nstates)
 136 #define EQ(a, b)        (memcmp(a, b, m->g->nstates) == 0)
 137 #define STATEVARS       int vn; char *space
 138 #define STATESETUP(m, nv)       { (m)->space = malloc((nv)*(m)->g->nstates); \
 139                                 if ((m)->space == NULL) return(REG_ESPACE); \
 140                                 (m)->vn = 0; }
 141 #define STATETEARDOWN(m)        { free((m)->space); }
 142 #define SETUP(v)        ((v) = &m->space[m->vn++ * m->g->nstates])
 143 #define onestate        int
 144 #define INIT(o, n)      ((o) = (n))
 145 #define INC(o)  ((o)++)
 146 #define ISSTATEIN(v, o) ((v)[o])
 147 /* some abbreviations; note that some of these know variable names! */
 148 /* do "if I'm here, I can also be there" etc without branches */
 149 #define FWD(dst, src, n)        ((dst)[here+(n)] |= (src)[here])
 150 #define BACK(dst, src, n)       ((dst)[here-(n)] |= (src)[here])
 151 #define ISSETBACK(v, n) ((v)[here - (n)])
 152 /* function names */
 153 #define LNAMES                  /* flag */
 154
 155 #include "engine.c"
 156
 157 /*
 158  - regexec - interface for matching
 159  = extern int regexec(const regex_t *, const char *, size_t, \
 160  =                                      regmatch_t [], int);
 161  = #define      REG_NOTBOL      00001
 162  = #define      REG_NOTEOL      00002
 163  = #define      REG_STARTEND    00004
 164  = #define      REG_TRACE       00400   // tracing of execution
 165  = #define      REG_LARGE       01000   // force large representation
 166  = #define      REG_BACKR       02000   // force use of backref code
 167  *
 168  * We put this here so we can exploit knowledge of the state representation
 169  * when choosing which matcher to call.  Also, by this point the matchers
 170  * have been prototyped.
 171  */
 172 int                             /* 0 success, REG_NOMATCH failure */
 173 regexec(preg, string, nmatch, pmatch, eflags)
 174 const regex_t *preg;
 175 const char *string;
 176 size_t nmatch;
 177 regmatch_t pmatch[];
 178 int eflags;
 179 {
 180         register struct re_guts *g = preg->re_g;
 181 #ifdef REDEBUG
 182 #       define  GOODFLAGS(f)    (f)
 183 #else
 184 #       define  GOODFLAGS(f)    ((f)&(REG_NOTBOL|REG_NOTEOL|REG_STARTEND))
 185 #endif
 186
 187         if (preg->re_magic != MAGIC1 || g->magic != MAGIC2)
 188                 return(REG_BADPAT);
 189         assert(!(g->iflags&BAD));
 190         if (g->iflags&BAD)              /* backstop for no-debug case */
 191                 return(REG_BADPAT);
 192         eflags = GOODFLAGS(eflags);
 193
 194         if (g->nstates <= CHAR_BIT*sizeof(states1) && !(eflags&REG_LARGE))
 195                 return(smatcher(g, (char *)string, nmatch, pmatch, eflags));
 196         else
 197                 return(lmatcher(g, (char *)string, nmatch, pmatch, eflags));
 198 }