1 /* $OpenBSD: regex2.h,v 1.6 2003/06/02 20:18:36 millert Exp $ */ 2 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 */ 37 38 /* 39 * First, the stuff that ends up in the outside-world include file 40 = typedef off_t regoff_t; 41 = typedef struct { 42 = int re_magic; 43 = size_t re_nsub; // number of parenthesized subexpressions 44 = const char *re_endp; // end pointer for REG_PEND 45 = struct re_guts *re_g; // none of your business :-) 46 = } regex_t; 47 = typedef struct { 48 = regoff_t rm_so; // start of match 49 = regoff_t rm_eo; // end of match 50 = } regmatch_t; 51 */ 52 /* 53 * internals of regex_t 54 */ 55 #define MAGIC1 ((('r'^0200)<<8) | 'e') 56 57 /* 58 * The internal representation is a *strip*, a sequence of 59 * operators ending with an endmarker. (Some terminology etc. is a 60 * historical relic of earlier versions which used multiple strips.) 61 * Certain oddities in the representation are there to permit running 62 * the machinery backwards; in particular, any deviation from sequential 63 * flow must be marked at both its source and its destination. Some 64 * fine points: 65 * 66 * - OPLUS_ and O_PLUS are *inside* the loop they create. 67 * - OQUEST_ and O_QUEST are *outside* the bypass they create. 68 * - OCH_ and O_CH are *outside* the multi-way branch they create, while 69 * OOR1 and OOR2 are respectively the end and the beginning of one of 70 * the branches. Note that there is an implicit OOR2 following OCH_ 71 * and an implicit OOR1 preceding O_CH. 72 * 73 * In state representations, an operator's bit is on to signify a state 74 * immediately *preceding* "execution" of that operator. 75 */ 76 typedef unsigned long sop; /* strip operator */ 77 typedef long sopno; 78 #define OPRMASK 0xf8000000LU 79 #define OPDMASK 0x07ffffffLU 80 #define OPSHIFT ((unsigned)27) 81 #define OP(n) ((n)&OPRMASK) 82 #define OPND(n) ((n)&OPDMASK) 83 #define SOP(op, opnd) ((op)|(opnd)) 84 /* operators meaning operand */ 85 /* (back, fwd are offsets) */ 86 #define OEND (1LU<<OPSHIFT) /* endmarker - */ 87 #define OCHAR (2LU<<OPSHIFT) /* character unsigned char */ 88 #define OBOL (3LU<<OPSHIFT) /* left anchor - */ 89 #define OEOL (4LU<<OPSHIFT) /* right anchor - */ 90 #define OANY (5LU<<OPSHIFT) /* . - */ 91 #define OANYOF (6LU<<OPSHIFT) /* [...] set number */ 92 #define OBACK_ (7LU<<OPSHIFT) /* begin \d paren number */ 93 #define O_BACK (8LU<<OPSHIFT) /* end \d paren number */ 94 #define OPLUS_ (9LU<<OPSHIFT) /* + prefix fwd to suffix */ 95 #define O_PLUS (10LU<<OPSHIFT) /* + suffix back to prefix */ 96 #define OQUEST_ (11LU<<OPSHIFT) /* ? prefix fwd to suffix */ 97 #define O_QUEST (12LU<<OPSHIFT) /* ? suffix back to prefix */ 98 #define OLPAREN (13LU<<OPSHIFT) /* ( fwd to ) */ 99 #define ORPAREN (14LU<<OPSHIFT) /* ) back to ( */ 100 #define OCH_ (15LU<<OPSHIFT) /* begin choice fwd to OOR2 */ 101 #define OOR1 (16LU<<OPSHIFT) /* | pt. 1 back to OOR1 or OCH_ */ 102 #define OOR2 (17LU<<OPSHIFT) /* | pt. 2 fwd to OOR2 or O_CH */ 103 #define O_CH (18LU<<OPSHIFT) /* end choice back to OOR1 */ 104 #define OBOW (19LU<<OPSHIFT) /* begin word - */ 105 #define OEOW (20LU<<OPSHIFT) /* end word - */ 106 107 /* 108 * Structure for [] character-set representation. Character sets are 109 * done as bit vectors, grouped 8 to a byte vector for compactness. 110 * The individual set therefore has both a pointer to the byte vector 111 * and a mask to pick out the relevant bit of each byte. A hash code 112 * simplifies testing whether two sets could be identical. 113 * 114 * This will get trickier for multicharacter collating elements. As 115 * preliminary hooks for dealing with such things, we also carry along 116 * a string of multi-character elements, and decide the size of the 117 * vectors at run time. 118 */ 119 typedef struct { 120 uch *ptr; /* -> uch [csetsize] */ 121 uch mask; /* bit within array */ 122 uch hash; /* hash code */ 123 size_t smultis; 124 char *multis; /* -> char[smulti] ab\0cd\0ef\0\0 */ 125 } cset; 126 /* note that CHadd and CHsub are unsafe, and CHIN doesn't yield 0/1 */ 127 #define CHadd(cs, c) ((cs)->ptr[(uch)(c)] |= (cs)->mask, (cs)->hash += (c)) 128 #define CHsub(cs, c) ((cs)->ptr[(uch)(c)] &= ~(cs)->mask, (cs)->hash -= (c)) 129 #define CHIN(cs, c) ((cs)->ptr[(uch)(c)] & (cs)->mask) 130 #define MCadd(p, cs, cp) mcadd(p, cs, cp) /* regcomp() internal fns */ 131 #define MCsub(p, cs, cp) mcsub(p, cs, cp) 132 #define MCin(p, cs, cp) mcin(p, cs, cp) 133 134 /* stuff for character categories */ 135 typedef unsigned char cat_t; 136 137 /* 138 * main compiled-expression structure 139 */ 140 struct re_guts { 141 int magic; 142 # define MAGIC2 ((('R'^0200)<<8)|'E') 143 sop *strip; /* malloced area for strip */ 144 int csetsize; /* number of bits in a cset vector */ 145 int ncsets; /* number of csets in use */ 146 cset *sets; /* -> cset [ncsets] */ 147 uch *setbits; /* -> uch[csetsize][ncsets/CHAR_BIT] */ 148 int cflags; /* copy of regcomp() cflags argument */ 149 sopno nstates; /* = number of sops */ 150 sopno firststate; /* the initial OEND (normally 0) */ 151 sopno laststate; /* the final OEND */ 152 int iflags; /* internal flags */ 153 # define USEBOL 01 /* used ^ */ 154 # define USEEOL 02 /* used $ */ 155 # define BAD 04 /* something wrong */ 156 int nbol; /* number of ^ used */ 157 int neol; /* number of $ used */ 158 int ncategories; /* how many character categories */ 159 cat_t *categories; /* ->catspace[-CHAR_MIN] */ 160 char *must; /* match must contain this string */ 161 int mlen; /* length of must */ 162 size_t nsub; /* copy of re_nsub */ 163 int backrefs; /* does it use back references? */ 164 sopno nplus; /* how deep does it nest +s? */ 165 /* catspace must be last */ 166 cat_t catspace[1]; /* actually [NC] */ 167 }; 168 169 /* misc utilities */ 170 #define OUT (CHAR_MAX+1) /* a non-character value */ 171 #define ISWORD(c) (isalnum(c) || (c) == '_') 172