1 /* $NetBSD: regex2.h,v 1.9 1998/12/13 12:21:47 drochner 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. All advertising materials mentioning features or use of this software 20 * must display the following acknowledgement: 21 * This product includes software developed by the University of 22 * California, Berkeley and its contributors. 23 * 4. Neither the name of the University nor the names of its contributors 24 * may be used to endorse or promote products derived from this software 25 * without specific prior written permission. 26 * 27 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 28 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 29 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 30 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 31 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 32 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 33 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 35 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 36 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 37 * SUCH DAMAGE. 38 * 39 * @(#)regex2.h 8.4 (Berkeley) 3/20/94 40 */ 41 42 /* 43 * First, the stuff that ends up in the outside-world include file 44 = typedef off_t regoff_t; 45 = typedef struct { 46 = int re_magic; 47 = size_t re_nsub; // number of parenthesized subexpressions 48 = const char *re_endp; // end pointer for REG_PEND 49 = struct re_guts *re_g; // none of your business :-) 50 = } regex_t; 51 = typedef struct { 52 = regoff_t rm_so; // start of match 53 = regoff_t rm_eo; // end of match 54 = } regmatch_t; 55 */ 56 /* 57 * internals of regex_t 58 */ 59 #define MAGIC1 ((('r'^0200)<<8) | 'e') 60 61 /* 62 * The internal representation is a *strip*, a sequence of 63 * operators ending with an endmarker. (Some terminology etc. is a 64 * historical relic of earlier versions which used multiple strips.) 65 * Certain oddities in the representation are there to permit running 66 * the machinery backwards; in particular, any deviation from sequential 67 * flow must be marked at both its source and its destination. Some 68 * fine points: 69 * 70 * - OPLUS_ and O_PLUS are *inside* the loop they create. 71 * - OQUEST_ and O_QUEST are *outside* the bypass they create. 72 * - OCH_ and O_CH are *outside* the multi-way branch they create, while 73 * OOR1 and OOR2 are respectively the end and the beginning of one of 74 * the branches. Note that there is an implicit OOR2 following OCH_ 75 * and an implicit OOR1 preceding O_CH. 76 * 77 * In state representations, an operator's bit is on to signify a state 78 * immediately *preceding* "execution" of that operator. 79 */ 80 typedef u_int32_t sop; /* strip operator */ 81 typedef int sopno; 82 #define OPRMASK ((u_int32_t)0xf8000000UL) 83 #define OPDMASK ((u_int32_t)0x07ffffffUL) 84 #define OPSHIFT ((unsigned)27) 85 #define OP(n) ((n)&OPRMASK) 86 #define OPND(n) ((int)((n)&OPDMASK)) 87 #define SOP(op, opnd) ((op)|(opnd)) 88 89 #define OPC(n) (((u_int32_t)(n))<<OPSHIFT) 90 /* operators meaning operand */ 91 /* (back, fwd are offsets) */ 92 #define OEND OPC(1) /* endmarker - */ 93 #define OCHAR OPC(2) /* character unsigned char */ 94 #define OBOL OPC(3) /* left anchor - */ 95 #define OEOL OPC(4) /* right anchor - */ 96 #define OANY OPC(5) /* . - */ 97 #define OANYOF OPC(6) /* [...] set number */ 98 #define OBACK_ OPC(7) /* begin \d paren number */ 99 #define O_BACK OPC(8) /* end \d paren number */ 100 #define OPLUS_ OPC(9) /* + prefix fwd to suffix */ 101 #define O_PLUS OPC(10) /* + suffix back to prefix */ 102 #define OQUEST_ OPC(11) /* ? prefix fwd to suffix */ 103 #define O_QUEST OPC(12) /* ? suffix back to prefix */ 104 #define OLPAREN OPC(13) /* ( fwd to ) */ 105 #define ORPAREN OPC(14) /* ) back to ( */ 106 #define OCH_ OPC(15) /* begin choice fwd to OOR2 */ 107 #define OOR1 OPC(16) /* | pt. 1 back to OOR1 or OCH_ */ 108 #define OOR2 OPC(17) /* | pt. 2 fwd to OOR2 or O_CH */ 109 #define O_CH OPC(18) /* end choice back to OOR1 */ 110 #define OBOW OPC(19) /* begin word - */ 111 #define OEOW OPC(20) /* end word - */ 112 113 /* 114 * Structure for [] character-set representation. Character sets are 115 * done as bit vectors, grouped 8 to a byte vector for compactness. 116 * The individual set therefore has both a pointer to the byte vector 117 * and a mask to pick out the relevant bit of each byte. A hash code 118 * simplifies testing whether two sets could be identical. 119 * 120 * This will get trickier for multicharacter collating elements. As 121 * preliminary hooks for dealing with such things, we also carry along 122 * a string of multi-character elements, and decide the size of the 123 * vectors at run time. 124 */ 125 typedef struct { 126 uch *ptr; /* -> uch [csetsize] */ 127 uch mask; /* bit within array */ 128 uch hash; /* hash code */ 129 size_t smultis; 130 char *multis; /* -> char[smulti] ab\0cd\0ef\0\0 */ 131 } cset; 132 /* note that CHadd and CHsub are unsafe, and CHIN doesn't yield 0/1 */ 133 #define CHadd(cs, c) ((cs)->ptr[(uch)(c)] |= (cs)->mask, (cs)->hash += (c)) 134 #define CHsub(cs, c) ((cs)->ptr[(uch)(c)] &= ~(cs)->mask, (cs)->hash -= (c)) 135 #define CHIN(cs, c) ((cs)->ptr[(uch)(c)] & (cs)->mask) 136 #define MCadd(p, cs, cp) mcadd(p, cs, cp) /* regcomp() internal fns */ 137 #define MCsub(p, cs, cp) mcsub(p, cs, cp) 138 #define MCin(p, cs, cp) mcin(p, cs, cp) 139 140 /* stuff for character categories */ 141 typedef unsigned char cat_t; 142 143 /* 144 * main compiled-expression structure 145 */ 146 struct re_guts { 147 int magic; 148 # define MAGIC2 ((('R'^0200)<<8)|'E') 149 sop *strip; /* malloced area for strip */ 150 int csetsize; /* number of bits in a cset vector */ 151 int ncsets; /* number of csets in use */ 152 cset *sets; /* -> cset [ncsets] */ 153 uch *setbits; /* -> uch[csetsize][ncsets/CHAR_BIT] */ 154 int cflags; /* copy of regcomp() cflags argument */ 155 sopno nstates; /* = number of sops */ 156 sopno firststate; /* the initial OEND (normally 0) */ 157 sopno laststate; /* the final OEND */ 158 int iflags; /* internal flags */ 159 # define USEBOL 01 /* used ^ */ 160 # define USEEOL 02 /* used $ */ 161 # define BAD 04 /* something wrong */ 162 int nbol; /* number of ^ used */ 163 int neol; /* number of $ used */ 164 int ncategories; /* how many character categories */ 165 cat_t *categories; /* ->catspace[-CHAR_MIN] */ 166 char *must; /* match must contain this string */ 167 int mlen; /* length of must */ 168 sopno nsub; /* copy of re_nsub */ 169 int backrefs; /* does it use back references? */ 170 sopno nplus; /* how deep does it nest +s? */ 171 /* catspace must be last */ 172 cat_t catspace[1]; /* actually [NC] */ 173 }; 174 175 /* misc utilities */ 176 #define OUT (CHAR_MAX+1) /* a non-character value */ 177 #define ISWORD(c) (isalnum(c) || (c) == '_') 178