newlib/libc/posix/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  * 3. 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 1.6 2002/03/22 23:41:56 obrien 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    unsigned char           */
  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  * Structure for [] character-set representation.  Character sets are
 108  * done as bit vectors, grouped 8 to a byte vector for compactness.
 109  * The individual set therefore has both a pointer to the byte vector
 110  * and a mask to pick out the relevant bit of each byte.  A hash code
 111  * simplifies testing whether two sets could be identical.
 112  *
 113  * This will get trickier for multicharacter collating elements.  As
 114  * preliminary hooks for dealing with such things, we also carry along
 115  * a string of multi-character elements, and decide the size of the
 116  * vectors at run time.
 117  */
 118 typedef struct {
 119         uch *ptr;               /* -> uch [csetsize] */
 120         uch mask;               /* bit within array */
 121         short hash;             /* hash code */
 122         size_t smultis;
 123         char *multis;           /* -> char[smulti]  ab\0cd\0ef\0\0 */
 124 } cset;
 125 /* note that CHadd and CHsub are unsafe, and CHIN doesn't yield 0/1 */
 126 #define CHadd(cs, c)    ((cs)->ptr[(uch)(c)] |= (cs)->mask, (cs)->hash += (uch)(c))
 127 #define CHsub(cs, c)    ((cs)->ptr[(uch)(c)] &= ~(cs)->mask, (cs)->hash -= (uch)(c))
 128 #define CHIN(cs, c)     ((cs)->ptr[(uch)(c)] & (cs)->mask)
 129 #define MCadd(p, cs, cp)        mcadd(p, cs, cp)        /* regcomp() internal fns */
 130 #define MCsub(p, cs, cp)        mcsub(p, cs, cp)
 131 #define MCin(p, cs, cp) mcin(p, cs, cp)
 132
 133 /* stuff for character categories */
 134 typedef unsigned char cat_t;
 135
 136 /*
 137  * main compiled-expression structure
 138  */
 139 struct re_guts {
 140         int magic;
 141 #               define  MAGIC2  ((('R'^0200)<<8)|'E')
 142         sop *strip;             /* malloced area for strip */
 143         int csetsize;           /* number of bits in a cset vector */
 144         int ncsets;             /* number of csets in use */
 145         cset *sets;             /* -> cset [ncsets] */
 146         uch *setbits;           /* -> uch[csetsize][ncsets/CHAR_BIT] */
 147         int cflags;             /* copy of regcomp() cflags argument */
 148         sopno nstates;          /* = number of sops */
 149         sopno firststate;       /* the initial OEND (normally 0) */
 150         sopno laststate;        /* the final OEND */
 151         int iflags;             /* internal flags */
 152 #               define  USEBOL  01      /* used ^ */
 153 #               define  USEEOL  02      /* used $ */
 154 #               define  BAD     04      /* something wrong */
 155         int nbol;               /* number of ^ used */
 156         int neol;               /* number of $ used */
 157         int ncategories;        /* how many character categories */
 158         cat_t *categories;      /* ->catspace[-CHAR_MIN] */
 159         char *must;             /* match must contain this string */
 160         int moffset;            /* latest point at which must may be located */
 161         int *charjump;          /* Boyer-Moore char jump table */
 162         int *matchjump;         /* Boyer-Moore match jump table */
 163         int mlen;               /* length of must */
 164         size_t nsub;            /* copy of re_nsub */
 165         int backrefs;           /* does it use back references? */
 166         sopno nplus;            /* how deep does it nest +s? */
 167         /* catspace must be last */
 168         cat_t catspace[1];      /* actually [NC] */
 169 };
 170
 171 /* misc utilities */
 172 #define OUT     (CHAR_MAX+1)    /* a non-character value */
 173 #define ISWORD(c)       (isalnum((uch)(c)) || (c) == '_')