winsup/cygwin/regex/regex2.h

   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  *      @(#)regex2.h    8.4 (Berkeley) 3/20/94
  36  * $FreeBSD$
  37  */
  38
  39 /*
  40  * First, the stuff that ends up in the outside-world include file
  41  = typedef off_t regoff_t;
  42  = typedef struct {
  43  =      int re_magic;
  44  =      size_t re_nsub;         // number of parenthesized subexpressions
  45  =      const char *re_endp;    // end pointer for REG_PEND
  46  =      struct re_guts *re_g;   // none of your business :-)
  47  = } regex_t;
  48  = typedef struct {
  49  =      regoff_t rm_so;         // start of match
  50  =      regoff_t rm_eo;         // end of match
  51  = } regmatch_t;
  52  */
  53 /*
  54  * internals of regex_t
  55  */
  56 #define MAGIC1  ((('r'^0200)<<8) | 'e')
  57
  58 /*
  59  * The internal representation is a *strip*, a sequence of
  60  * operators ending with an endmarker.  (Some terminology etc. is a
  61  * historical relic of earlier versions which used multiple strips.)
  62  * Certain oddities in the representation are there to permit running
  63  * the machinery backwards; in particular, any deviation from sequential
  64  * flow must be marked at both its source and its destination.  Some
  65  * fine points:
  66  *
  67  * - OPLUS_ and O_PLUS are *inside* the loop they create.
  68  * - OQUEST_ and O_QUEST are *outside* the bypass they create.
  69  * - OCH_ and O_CH are *outside* the multi-way branch they create, while
  70  *   OOR1 and OOR2 are respectively the end and the beginning of one of
  71  *   the branches.  Note that there is an implicit OOR2 following OCH_
  72  *   and an implicit OOR1 preceding O_CH.
  73  *
  74  * In state representations, an operator's bit is on to signify a state
  75  * immediately *preceding* "execution" of that operator.
  76  */
  77 typedef unsigned long sop;      /* strip operator */
  78 typedef unsigned long sopno;
  79 #define OPRMASK 0xf8000000L
  80 #define OPDMASK 0x07ffffffL
  81 #define OPSHIFT ((unsigned)27)
  82 #define OP(n)   ((n)&OPRMASK)
  83 #define OPND(n) ((n)&OPDMASK)
  84 #define SOP(op, opnd)   ((op)|(opnd))
  85 /* operators                       meaning      operand                 */
  86 /*                                              (back, fwd are offsets) */
  87 #define OEND    (1L<<OPSHIFT)   /* endmarker    -                       */
  88 #define OCHAR   (2L<<OPSHIFT)   /* character    wide character          */
  89 #define OBOL    (3L<<OPSHIFT)   /* left anchor  -                       */
  90 #define OEOL    (4L<<OPSHIFT)   /* right anchor -                       */
  91 #define OANY    (5L<<OPSHIFT)   /* .            -                       */
  92 #define OANYOF  (6L<<OPSHIFT)   /* [...]        set number              */
  93 #define OBACK_  (7L<<OPSHIFT)   /* begin \d     paren number            */
  94 #define O_BACK  (8L<<OPSHIFT)   /* end \d       paren number            */
  95 #define OPLUS_  (9L<<OPSHIFT)   /* + prefix     fwd to suffix           */
  96 #define O_PLUS  (10L<<OPSHIFT)  /* + suffix     back to prefix          */
  97 #define OQUEST_ (11L<<OPSHIFT)  /* ? prefix     fwd to suffix           */
  98 #define O_QUEST (12L<<OPSHIFT)  /* ? suffix     back to prefix          */
  99 #define OLPAREN (13L<<OPSHIFT)  /* (            fwd to )                */
 100 #define ORPAREN (14L<<OPSHIFT)  /* )            back to (               */
 101 #define OCH_    (15L<<OPSHIFT)  /* begin choice fwd to OOR2             */
 102 #define OOR1    (16L<<OPSHIFT)  /* | pt. 1      back to OOR1 or OCH_    */
 103 #define OOR2    (17L<<OPSHIFT)  /* | pt. 2      fwd to OOR2 or O_CH     */
 104 #define O_CH    (18L<<OPSHIFT)  /* end choice   back to OOR1            */
 105 #define OBOW    (19L<<OPSHIFT)  /* begin word   -                       */
 106 #define OEOW    (20L<<OPSHIFT)  /* end word     -                       */
 107 #define OBOS    (21L<<OPSHIFT)  /* begin subj.  -                       */
 108 #define OEOS    (22L<<OPSHIFT)  /* end subj.    -                       */
 109 #define OWBND   (23L<<OPSHIFT)  /* word bound   -                       */
 110 #define ONWBND  (24L<<OPSHIFT)  /* not bound    -                       */
 111
 112 /*
 113  * Structures for [] character-set representation.
 114  */
 115 typedef struct {
 116         wint_t          min;
 117         wint_t          max;
 118 } crange;
 119 typedef struct {
 120         unsigned char   bmp[NC_MAX / 8];
 121         wctype_t        *types;
 122         unsigned int    ntypes;
 123         wint_t          *wides;
 124         unsigned int    nwides;
 125         crange          *ranges;
 126         unsigned int    nranges;
 127         int             invert;
 128         int             icase;
 129 } cset;
 130
 131 static int
 132 CHIN1(cset *cs, wint_t ch)
 133 {
 134         unsigned int i;
 135
 136         assert(ch >= 0);
 137         if (ch < NC)
 138                 return (((cs->bmp[ch >> 3] & (1 << (ch & 7))) != 0) ^
 139                     cs->invert);
 140         for (i = 0; i < cs->nwides; i++) {
 141                 if (cs->icase) {
 142                         if (ch == towlower(cs->wides[i]) ||
 143                             ch == towupper(cs->wides[i]))
 144                                 return (!cs->invert);
 145                 } else if (ch == cs->wides[i])
 146                         return (!cs->invert);
 147         }
 148         for (i = 0; i < cs->nranges; i++)
 149                 if (cs->ranges[i].min <= ch && ch <= cs->ranges[i].max)
 150                         return (!cs->invert);
 151         for (i = 0; i < cs->ntypes; i++)
 152                 if (iswctype(ch, cs->types[i]))
 153                         return (!cs->invert);
 154         return (cs->invert);
 155 }
 156
 157 static __inline int
 158 CHIN(cset *cs, wint_t ch)
 159 {
 160
 161         assert(ch >= 0);
 162         if (ch < NC)
 163                 return (((cs->bmp[ch >> 3] & (1 << (ch & 7))) != 0) ^
 164                     cs->invert);
 165         else if (cs->icase)
 166                 return (CHIN1(cs, ch) || CHIN1(cs, towlower(ch)) ||
 167                     CHIN1(cs, towupper(ch)));
 168         else
 169                 return (CHIN1(cs, ch));
 170 }
 171
 172 /*
 173  * main compiled-expression structure
 174  */
 175 struct re_guts {
 176         int magic;
 177 #               define  MAGIC2  ((('R'^0200)<<8)|'E')
 178         sop *strip;             /* malloced area for strip */
 179         unsigned int ncsets;    /* number of csets in use */
 180         cset *sets;             /* -> cset [ncsets] */
 181         int cflags;             /* copy of regcomp() cflags argument */
 182         sopno nstates;          /* = number of sops */
 183         sopno firststate;       /* the initial OEND (normally 0) */
 184         sopno laststate;        /* the final OEND */
 185         int iflags;             /* internal flags */
 186 #               define  USEBOL  01      /* used ^ */
 187 #               define  USEEOL  02      /* used $ */
 188 #               define  BAD     04      /* something wrong */
 189         int nbol;               /* number of ^ used */
 190         int neol;               /* number of $ used */
 191         char *must;             /* match must contain this string */
 192         int moffset;            /* latest point at which must may be located */
 193         int *charjump;          /* Boyer-Moore char jump table */
 194         int *matchjump;         /* Boyer-Moore match jump table */
 195         int mlen;               /* length of must */
 196         size_t nsub;            /* copy of re_nsub */
 197         int backrefs;           /* does it use back references? */
 198         sopno nplus;            /* how deep does it nest +s? */
 199 };
 200
 201 /* misc utilities */
 202 #define OUT     (CHAR_MIN - 1)  /* a non-character value */
 203 #define IGN     (CHAR_MIN - 2)
 204 #define ISWORD(c)       (iswalnum((uch)(c)) || (c) == '_')