compiler/clib/regex/regex2.h

   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  *      @(#)regex2.h    8.4 (Berkeley) 3/20/94
  34  * $FreeBSD: src/lib/libc/regex/regex2.h,v 1.11 2007/01/09 00:28:04 imp Exp $
  35  */
  36
  37 /*
  38  * First, the stuff that ends up in the outside-world include file
  39  = typedef off_t regoff_t;
  40  = typedef struct {
  41  =      int re_magic;
  42  =      size_t re_nsub;         // number of parenthesized subexpressions
  43  =      const char *re_endp;    // end pointer for REG_PEND
  44  =      struct re_guts *re_g;   // none of your business :-)
  45  = } regex_t;
  46  = typedef struct {
  47  =      regoff_t rm_so;         // start of match
  48  =      regoff_t rm_eo;         // end of match
  49  = } regmatch_t;
  50  */
  51 /*
  52  * internals of regex_t
  53  */
  54 #define MAGIC1  ((('r'^0200)<<8) | 'e')
  55
  56 /*
  57  * The internal representation is a *strip*, a sequence of
  58  * operators ending with an endmarker.  (Some terminology etc. is a
  59  * historical relic of earlier versions which used multiple strips.)
  60  * Certain oddities in the representation are there to permit running
  61  * the machinery backwards; in particular, any deviation from sequential
  62  * flow must be marked at both its source and its destination.  Some
  63  * fine points:
  64  *
  65  * - OPLUS_ and O_PLUS are *inside* the loop they create.
  66  * - OQUEST_ and O_QUEST are *outside* the bypass they create.
  67  * - OCH_ and O_CH are *outside* the multi-way branch they create, while
  68  *   OOR1 and OOR2 are respectively the end and the beginning of one of
  69  *   the branches.  Note that there is an implicit OOR2 following OCH_
  70  *   and an implicit OOR1 preceding O_CH.
  71  *
  72  * In state representations, an operator's bit is on to signify a state
  73  * immediately *preceding* "execution" of that operator.
  74  */
  75 typedef unsigned long sop;      /* strip operator */
  76 typedef long sopno;
  77 #define OPRMASK 0xf8000000L
  78 #define OPDMASK 0x07ffffffL
  79 #define OPSHIFT ((unsigned)27)
  80 #define OP(n)   ((n)&OPRMASK)
  81 #define OPND(n) ((n)&OPDMASK)
  82 #define SOP(op, opnd)   ((op)|(opnd))
  83 /* operators                       meaning      operand                 */
  84 /*                                              (back, fwd are offsets) */
  85 #define OEND    (1L<<OPSHIFT)   /* endmarker    -                       */
  86 #define OCHAR   (2L<<OPSHIFT)   /* character    wide character          */
  87 #define OBOL    (3L<<OPSHIFT)   /* left anchor  -                       */
  88 #define OEOL    (4L<<OPSHIFT)   /* right anchor -                       */
  89 #define OANY    (5L<<OPSHIFT)   /* .            -                       */
  90 #define OANYOF  (6L<<OPSHIFT)   /* [...]        set number              */
  91 #define OBACK_  (7L<<OPSHIFT)   /* begin \d     paren number            */
  92 #define O_BACK  (8L<<OPSHIFT)   /* end \d       paren number            */
  93 #define OPLUS_  (9L<<OPSHIFT)   /* + prefix     fwd to suffix           */
  94 #define O_PLUS  (10L<<OPSHIFT)  /* + suffix     back to prefix          */
  95 #define OQUEST_ (11L<<OPSHIFT)  /* ? prefix     fwd to suffix           */
  96 #define O_QUEST (12L<<OPSHIFT)  /* ? suffix     back to prefix          */
  97 #define OLPAREN (13L<<OPSHIFT)  /* (            fwd to )                */
  98 #define ORPAREN (14L<<OPSHIFT)  /* )            back to (               */
  99 #define OCH_    (15L<<OPSHIFT)  /* begin choice fwd to OOR2             */
 100 #define OOR1    (16L<<OPSHIFT)  /* | pt. 1      back to OOR1 or OCH_    */
 101 #define OOR2    (17L<<OPSHIFT)  /* | pt. 2      fwd to OOR2 or O_CH     */
 102 #define O_CH    (18L<<OPSHIFT)  /* end choice   back to OOR1            */
 103 #define OBOW    (19L<<OPSHIFT)  /* begin word   -                       */
 104 #define OEOW    (20L<<OPSHIFT)  /* end word     -                       */
 105
 106 /*
 107  * Structures for [] character-set representation.
 108  */
 109 typedef struct {
 110         wint_t          min;
 111         wint_t          max;
 112 } crange;
 113 typedef struct {
 114         unsigned char   bmp[NC / 8];
 115         wctype_t        *types;
 116         int             ntypes;
 117         wint_t          *wides;
 118         int             nwides;
 119         crange          *ranges;
 120         int             nranges;
 121         int             invert;
 122         int             icase;
 123 } cset;
 124
 125 static int
 126 CHIN1(cset *cs, wint_t ch)
 127 {
 128         int i;
 129
 130         assert(ch >= 0);
 131         if (ch < NC)
 132                 return (((cs->bmp[ch >> 3] & (1 << (ch & 7))) != 0) ^
 133                     cs->invert);
 134         for (i = 0; i < cs->nwides; i++)
 135                 if (ch == cs->wides[i])
 136                         return (!cs->invert);
 137         for (i = 0; i < cs->nranges; i++)
 138                 if (cs->ranges[i].min <= ch && ch <= cs->ranges[i].max)
 139                         return (!cs->invert);
 140         for (i = 0; i < cs->ntypes; i++)
 141                 if (iswctype(ch, cs->types[i]))
 142                         return (!cs->invert);
 143         return (cs->invert);
 144 }
 145
 146 static __inline int
 147 CHIN(cset *cs, wint_t ch)
 148 {
 149
 150         assert(ch >= 0);
 151         if (ch < NC)
 152                 return (((cs->bmp[ch >> 3] & (1 << (ch & 7))) != 0) ^
 153                     cs->invert);
 154         else if (cs->icase)
 155                 return (CHIN1(cs, ch) || CHIN1(cs, towlower(ch)) ||
 156                     CHIN1(cs, towupper(ch)));
 157         else
 158                 return (CHIN1(cs, ch));
 159 }
 160
 161 /*
 162  * main compiled-expression structure
 163  */
 164 struct re_guts {
 165         int magic;
 166 #               define  MAGIC2  ((('R'^0200)<<8)|'E')
 167         sop *strip;             /* malloced area for strip */
 168         int ncsets;             /* number of csets in use */
 169         cset *sets;             /* -> cset [ncsets] */
 170         int cflags;             /* copy of regcomp() cflags argument */
 171         sopno nstates;          /* = number of sops */
 172         sopno firststate;       /* the initial OEND (normally 0) */
 173         sopno laststate;        /* the final OEND */
 174         int iflags;             /* internal flags */
 175 #               define  USEBOL  01      /* used ^ */
 176 #               define  USEEOL  02      /* used $ */
 177 #               define  BAD     04      /* something wrong */
 178         int nbol;               /* number of ^ used */
 179         int neol;               /* number of $ used */
 180         char *must;             /* match must contain this string */
 181         int moffset;            /* latest point at which must may be located */
 182         int *charjump;          /* Boyer-Moore char jump table */
 183         int *matchjump;         /* Boyer-Moore match jump table */
 184         int mlen;               /* length of must */
 185         size_t nsub;            /* copy of re_nsub */
 186         int backrefs;           /* does it use back references? */
 187         sopno nplus;            /* how deep does it nest +s? */
 188 };
 189
 190 /* misc utilities */
 191 #define OUT     (CHAR_MIN - 1)  /* a non-character value */
 192 #define ISWORD(c)       (iswalnum((uch)(c)) || (c) == '_')