winsup/cygwin/regex/regexec.c

   1 /*-
   2  * SPDX-License-Identifier: BSD-3-Clause
   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.
  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. Neither the name of the University nor the names of its contributors
  20  *    may be used to endorse or promote products derived from this software
  21  *    without specific prior written permission.
  22  *
  23  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  24  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  25  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  26  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  27  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  28  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  29  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  30  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  31  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  32  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  33  * SUCH DAMAGE.
  34  *
  35  *      @(#)regexec.c   8.3 (Berkeley) 3/20/94
  36  */
  37
  38 #if defined(LIBC_SCCS) && !defined(lint)
  39 static char sccsid[] = "@(#)regexec.c   8.3 (Berkeley) 3/20/94";
  40 #endif /* LIBC_SCCS and not lint */
  41 #include <sys/cdefs.h>
  42 __FBSDID("$FreeBSD$");
  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)
  68 {
  69         size_t nr;
  70         wint_t wc;
  71
  72         nr = mbrtowi(&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                 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 {
  93
  94         if (wi != NULL)
  95                 *wi = (unsigned char)*s;
  96         return (1);
  97 }
  98
  99 /* macros for manipulating states, small version */
 100 #define states1 long            /* for later use in regexec() decision */
 101 #define states  states1
 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)        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)    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) return(REG_ESPACE); \
 163                                 (m)->vn = 0; }
 164 #define STATETEARDOWN(m)        { free((m)->space); }
 165 #define SETUP(v)        ((v) = &m->space[m->vn++ * m->g->nstates])
 166 #define onestate        long
 167 #define INIT(o, n)      ((o) = (n))
 168 #define INC(o)  ((o)++)
 169 #define ISSTATEIN(v, o) ((v)[o])
 170 /* some abbreviations; note that some of these know variable names! */
 171 /* do "if I'm here, I can also be there" etc without branches */
 172 #define FWD(dst, src, n)        ((dst)[here+(n)] |= (src)[here])
 173 #define BACK(dst, src, n)       ((dst)[here-(n)] |= (src)[here])
 174 #define ISSETBACK(v, n) ((v)[here - (n)])
 175 /* no multibyte support */
 176 #define XMBRTOWC        xmbrtowc_dummy
 177 #define ZAPSTATE(mbs)   ((void)(mbs))
 178 /* function names */
 179 #define LNAMES                  /* flag */
 180
 181 #include "engine.c"
 182
 183 /* multibyte character & large states version */
 184 #undef  LNAMES
 185 #undef  XMBRTOWC
 186 #undef  ZAPSTATE
 187 #define XMBRTOWC        xmbrtowc
 188 #define ZAPSTATE(mbs)   memset((mbs), 0, sizeof(*(mbs)))
 189 #define MNAMES
 190
 191 #include "engine.c"
 192
 193 /*
 194  - regexec - interface for matching
 195  = extern int regexec(const regex_t *, const char *, size_t, \
 196  =                                      regmatch_t [], int);
 197  = #define      REG_NOTBOL      00001
 198  = #define      REG_NOTEOL      00002
 199  = #define      REG_STARTEND    00004
 200  = #define      REG_TRACE       00400   // tracing of execution
 201  = #define      REG_LARGE       01000   // force large representation
 202  = #define      REG_BACKR       02000   // force use of backref code
 203  *
 204  * We put this here so we can exploit knowledge of the state representation
 205  * when choosing which matcher to call.  Also, by this point the matchers
 206  * have been prototyped.
 207  */
 208 int                             /* 0 success, REG_NOMATCH failure */
 209 regexec(const regex_t * __restrict preg,
 210         const char * __restrict string,
 211         size_t nmatch,
 212         regmatch_t pmatch[__restrict],
 213         int eflags)
 214 {
 215         struct re_guts *g = preg->re_g;
 216 #ifdef REDEBUG
 217 #       define  GOODFLAGS(f)    (f)
 218 #else
 219 #       define  GOODFLAGS(f)    ((f)&(REG_NOTBOL|REG_NOTEOL|REG_STARTEND))
 220 #endif
 221
 222         if (preg->re_magic != MAGIC1 || g->magic != MAGIC2)
 223                 return(REG_BADPAT);
 224         assert(!(g->iflags&BAD));
 225         if (g->iflags&BAD)              /* backstop for no-debug case */
 226                 return(REG_BADPAT);
 227         eflags = GOODFLAGS(eflags);
 228
 229         if (MB_CUR_MAX > 1)
 230                 return(mmatcher(g, string, nmatch, pmatch, eflags));
 231         else if (g->nstates <= CHAR_BIT*sizeof(states1) && !(eflags&REG_LARGE))
 232                 return(smatcher(g, string, nmatch, pmatch, eflags));
 233         else
 234                 return(lmatcher(g, string, nmatch, pmatch, eflags));
 235 }