regex/regex2.h

   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  * First, the stuff that ends up in the outside-world include file
  63  = typedef off_t regoff_t;
  64  = typedef struct {
  65  =      int re_magic;
  66  =      size_t re_nsub;         // number of parenthesized subexpressions
  67  =      const char *re_endp;    // end pointer for REG_PEND
  68  =      struct re_guts *re_g;   // none of your business :-)
  69  = } regex_t;
  70  = typedef struct {
  71  =      regoff_t rm_so;         // start of match
  72  =      regoff_t rm_eo;         // end of match
  73  = } regmatch_t;
  74  */
  75 /*
  76  * internals of regex_t
  77  */
  78 #define MAGIC1  ((('r'^0200)<<8) | 'e')
  79
  80 /*
  81  * The internal representation is a *strip*, a sequence of
  82  * operators ending with an endmarker.  (Some terminology etc. is a
  83  * historical relic of earlier versions which used multiple strips.)
  84  * Certain oddities in the representation are there to permit running
  85  * the machinery backwards; in particular, any deviation from sequential
  86  * flow must be marked at both its source and its destination.  Some
  87  * fine points:
  88  *
  89  * - OPLUS_ and O_PLUS are *inside* the loop they create.
  90  * - OQUEST_ and O_QUEST are *outside* the bypass they create.
  91  * - OCH_ and O_CH are *outside* the multi-way branch they create, while
  92  *   OOR1 and OOR2 are respectively the end and the beginning of one of
  93  *   the branches.  Note that there is an implicit OOR2 following OCH_
  94  *   and an implicit OOR1 preceding O_CH.
  95  *
  96  * In state representations, an operator's bit is on to signify a state
  97  * immediately *preceding* "execution" of that operator.
  98  */
  99 typedef unsigned long sop;      /* strip operator */
 100 typedef long sopno;
 101 #define OPRMASK 0xf8000000
 102 #define OPDMASK 0x07ffffff
 103 #define OPSHIFT ((unsigned)27)
 104 #define OP(n)   ((n)&OPRMASK)
 105 #define OPND(n) ((n)&OPDMASK)
 106 #define SOP(op, opnd)   ((op)|(opnd))
 107 /* operators                       meaning      operand                 */
 108 /*                                              (back, fwd are offsets) */
 109 #define OEND    (1<<OPSHIFT)    /* endmarker    -                       */
 110 #define OCHAR   (2<<OPSHIFT)    /* character    unsigned char           */
 111 #define OBOL    (3<<OPSHIFT)    /* left anchor  -                       */
 112 #define OEOL    (4<<OPSHIFT)    /* right anchor -                       */
 113 #define OANY    (5<<OPSHIFT)    /* .            -                       */
 114 #define OANYOF  (6<<OPSHIFT)    /* [...]        set number              */
 115 #define OBACK_  (7<<OPSHIFT)    /* begin \d     paren number            */
 116 #define O_BACK  (8<<OPSHIFT)    /* end \d       paren number            */
 117 #define OPLUS_  (9<<OPSHIFT)    /* + prefix     fwd to suffix           */
 118 #define O_PLUS  (10<<OPSHIFT)   /* + suffix     back to prefix          */
 119 #define OQUEST_ (11<<OPSHIFT)   /* ? prefix     fwd to suffix           */
 120 #define O_QUEST (12<<OPSHIFT)   /* ? suffix     back to prefix          */
 121 #define OLPAREN (13<<OPSHIFT)   /* (            fwd to )                */
 122 #define ORPAREN (14<<OPSHIFT)   /* )            back to (               */
 123 #define OCH_    (15<<OPSHIFT)   /* begin choice fwd to OOR2             */
 124 #define OOR1    (16<<OPSHIFT)   /* | pt. 1      back to OOR1 or OCH_    */
 125 #define OOR2    (17<<OPSHIFT)   /* | pt. 2      fwd to OOR2 or O_CH     */
 126 #define O_CH    (18<<OPSHIFT)   /* end choice   back to OOR1            */
 127 #define OBOW    (19<<OPSHIFT)   /* begin word   -                       */
 128 #define OEOW    (20<<OPSHIFT)   /* end word     -                       */
 129
 130 /*
 131  * Structure for [] character-set representation.  Character sets are
 132  * done as bit vectors, grouped 8 to a byte vector for compactness.
 133  * The individual set therefore has both a pointer to the byte vector
 134  * and a mask to pick out the relevant bit of each byte.  A hash code
 135  * simplifies testing whether two sets could be identical.
 136  *
 137  * This will get trickier for multicharacter collating elements.  As
 138  * preliminary hooks for dealing with such things, we also carry along
 139  * a string of multi-character elements, and decide the size of the
 140  * vectors at run time.
 141  */
 142 typedef struct {
 143         uch *ptr;               /* -> uch [csetsize] */
 144         uch mask;               /* bit within array */
 145         uch hash;               /* hash code */
 146         size_t smultis;
 147         char *multis;           /* -> char[smulti]  ab\0cd\0ef\0\0 */
 148 } cset;
 149 /* note that CHadd and CHsub are unsafe, and CHIN doesn't yield 0/1 */
 150 #define CHadd(cs, c)    ((cs)->ptr[(uch)(c)] |= (cs)->mask, (cs)->hash += (c))
 151 #define CHsub(cs, c)    ((cs)->ptr[(uch)(c)] &= ~(cs)->mask, (cs)->hash -= (c))
 152 #define CHIN(cs, c)     ((cs)->ptr[(uch)(c)] & (cs)->mask)
 153 #define MCadd(p, cs, cp)        mcadd(p, cs, cp)        /* regcomp() internal fns */
 154 #define MCsub(p, cs, cp)        mcsub(p, cs, cp)
 155 #define MCin(p, cs, cp) mcin(p, cs, cp)
 156
 157 /* stuff for character categories */
 158 typedef unsigned char cat_t;
 159
 160 /*
 161  * main compiled-expression structure
 162  */
 163 struct re_guts {
 164         int magic;
 165 #               define  MAGIC2  ((('R'^0200)<<8)|'E')
 166         sop *strip;             /* malloced area for strip */
 167         int csetsize;           /* number of bits in a cset vector */
 168         int ncsets;             /* number of csets in use */
 169         cset *sets;             /* -> cset [ncsets] */
 170         uch *setbits;           /* -> uch[csetsize][ncsets/CHAR_BIT] */
 171         int cflags;             /* copy of regcomp() cflags argument */
 172         sopno nstates;          /* = number of sops */
 173         sopno firststate;       /* the initial OEND (normally 0) */
 174         sopno laststate;        /* the final OEND */
 175         int iflags;             /* internal flags */
 176 #               define  USEBOL  01      /* used ^ */
 177 #               define  USEEOL  02      /* used $ */
 178 #               define  BAD     04      /* something wrong */
 179         int nbol;               /* number of ^ used */
 180         int neol;               /* number of $ used */
 181         int ncategories;        /* how many character categories */
 182         cat_t *categories;      /* ->catspace[-CHAR_MIN] */
 183         char *must;             /* match must contain this string */
 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         /* catspace must be last */
 189         cat_t catspace[1];      /* actually [NC] */
 190 };
 191
 192 /* misc utilities */
 193 #define OUT     (CHAR_MAX+1)    /* a non-character value */
 194 #define ISWORD(c)       (isalnum(c) || (c) == '_')