Annotation of src/lib/libc/regex/regex2.h, Revision 1.15
1.15 ! christos 1: /* $NetBSD: regex2.h,v 1.14 2021/02/23 22:14:59 christos Exp $ */
1.5 cgd 2:
1.4 cgd 3: /*-
1.14 christos 4: * SPDX-License-Identifier: BSD-3-Clause
5: *
6: * Copyright (c) 1992, 1993, 1994 Henry Spencer.
1.4 cgd 7: * Copyright (c) 1992, 1993, 1994
8: * The Regents of the University of California. All rights reserved.
1.10 agc 9: *
10: * This code is derived from software contributed to Berkeley by
11: * Henry Spencer.
12: *
13: * Redistribution and use in source and binary forms, with or without
14: * modification, are permitted provided that the following conditions
15: * are met:
16: * 1. Redistributions of source code must retain the above copyright
17: * notice, this list of conditions and the following disclaimer.
18: * 2. Redistributions in binary form must reproduce the above copyright
19: * notice, this list of conditions and the following disclaimer in the
20: * documentation and/or other materials provided with the distribution.
21: * 3. Neither the name of the University nor the names of its contributors
22: * may be used to endorse or promote products derived from this software
23: * without specific prior written permission.
24: *
25: * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
26: * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
27: * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
28: * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
29: * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
30: * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
31: * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
32: * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
33: * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
34: * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
35: * SUCH DAMAGE.
36: *
37: * @(#)regex2.h 8.4 (Berkeley) 3/20/94
1.14 christos 38: * $FreeBSD: head/lib/libc/regex/regex2.h 368359 2020-12-05 03:18:48Z kevans $
1.4 cgd 39: */
40:
1.1 jtc 41: /*
42: * First, the stuff that ends up in the outside-world include file
43: = typedef off_t regoff_t;
44: = typedef struct {
45: = int re_magic;
46: = size_t re_nsub; // number of parenthesized subexpressions
1.3 jtc 47: = const char *re_endp; // end pointer for REG_PEND
1.1 jtc 48: = struct re_guts *re_g; // none of your business :-)
49: = } regex_t;
50: = typedef struct {
51: = regoff_t rm_so; // start of match
52: = regoff_t rm_eo; // end of match
53: = } regmatch_t;
54: */
55: /*
56: * internals of regex_t
57: */
58: #define MAGIC1 ((('r'^0200)<<8) | 'e')
59:
60: /*
61: * The internal representation is a *strip*, a sequence of
62: * operators ending with an endmarker. (Some terminology etc. is a
63: * historical relic of earlier versions which used multiple strips.)
64: * Certain oddities in the representation are there to permit running
65: * the machinery backwards; in particular, any deviation from sequential
66: * flow must be marked at both its source and its destination. Some
67: * fine points:
68: *
69: * - OPLUS_ and O_PLUS are *inside* the loop they create.
70: * - OQUEST_ and O_QUEST are *outside* the bypass they create.
71: * - OCH_ and O_CH are *outside* the multi-way branch they create, while
72: * OOR1 and OOR2 are respectively the end and the beginning of one of
73: * the branches. Note that there is an implicit OOR2 following OCH_
74: * and an implicit OOR1 preceding O_CH.
75: *
76: * In state representations, an operator's bit is on to signify a state
77: * immediately *preceding* "execution" of that operator.
78: */
1.15 ! christos 79: typedef uint32_t sop; /* strip operator */
! 80: typedef uint32_t sopno;
! 81: #define OPRMASK 0xf8000000U
! 82: #define OPDMASK 0x07ffffffU
! 83: #define OPSHIFT (27U)
1.1 jtc 84: #define OP(n) ((n)&OPRMASK)
1.14 christos 85: #define OPND(n) ((n)&OPDMASK)
1.1 jtc 86: #define SOP(op, opnd) ((op)|(opnd))
1.14 christos 87: /* operators meaning operand */
88: /* (back, fwd are offsets) */
1.15 ! christos 89: #define OEND (1U<<OPSHIFT) /* endmarker - */
! 90: #define OCHAR (2U<<OPSHIFT) /* character wide character */
! 91: #define OBOL (3U<<OPSHIFT) /* left anchor - */
! 92: #define OEOL (4U<<OPSHIFT) /* right anchor - */
! 93: #define OANY (5U<<OPSHIFT) /* . - */
! 94: #define OANYOF (6U<<OPSHIFT) /* [...] set number */
! 95: #define OBACK_ (7U<<OPSHIFT) /* begin \d paren number */
! 96: #define O_BACK (8U<<OPSHIFT) /* end \d paren number */
! 97: #define OPLUS_ (9U<<OPSHIFT) /* + prefix fwd to suffix */
! 98: #define O_PLUS (10U<<OPSHIFT) /* + suffix back to prefix */
! 99: #define OQUEST_ (11U<<OPSHIFT) /* ? prefix fwd to suffix */
! 100: #define O_QUEST (12U<<OPSHIFT) /* ? suffix back to prefix */
! 101: #define OLPAREN (13U<<OPSHIFT) /* ( fwd to ) */
! 102: #define ORPAREN (14U<<OPSHIFT) /* ) back to ( */
! 103: #define OCH_ (15U<<OPSHIFT) /* begin choice fwd to OOR2 */
! 104: #define OOR1 (16U<<OPSHIFT) /* | pt. 1 back to OOR1 or OCH_ */
! 105: #define OOR2 (17U<<OPSHIFT) /* | pt. 2 fwd to OOR2 or O_CH */
! 106: #define O_CH (18U<<OPSHIFT) /* end choice back to OOR1 */
! 107: #define OBOW (19U<<OPSHIFT) /* begin word - */
! 108: #define OEOW (20U<<OPSHIFT) /* end word - */
! 109: #define OBOS (21U<<OPSHIFT) /* begin subj. - */
! 110: #define OEOS (22U<<OPSHIFT) /* end subj. - */
! 111: #define OWBND (23U<<OPSHIFT) /* word bound - */
! 112: #define ONWBND (24U<<OPSHIFT) /* not bound - */
1.1 jtc 113:
114: /*
1.14 christos 115: * Structures for [] character-set representation.
1.1 jtc 116: */
117: typedef struct {
1.14 christos 118: wint_t min;
119: wint_t max;
120: } crange;
121: typedef struct {
122: unsigned char bmp[NC_MAX / 8];
123: wctype_t *types;
124: unsigned int ntypes;
125: wint_t *wides;
126: unsigned int nwides;
127: crange *ranges;
128: unsigned int nranges;
129: int invert;
130: int icase;
1.1 jtc 131: } cset;
132:
1.14 christos 133: static int
134: CHIN1(cset *cs, wint_t ch)
135: {
136: unsigned int i;
137:
138: assert(ch >= 0);
139: if (ch < NC)
1.15 ! christos 140: return (((cs->bmp[(unsigned)ch >> 3] & (1 << (ch & 7))) != 0) ^
1.14 christos 141: cs->invert);
142: for (i = 0; i < cs->nwides; i++) {
143: if (cs->icase) {
144: if (ch == towlower(cs->wides[i]) ||
145: ch == towupper(cs->wides[i]))
146: return (!cs->invert);
147: } else if (ch == cs->wides[i])
148: return (!cs->invert);
149: }
150: for (i = 0; i < cs->nranges; i++)
151: if (cs->ranges[i].min <= ch && ch <= cs->ranges[i].max)
152: return (!cs->invert);
153: for (i = 0; i < cs->ntypes; i++)
154: if (iswctype(ch, cs->types[i]))
155: return (!cs->invert);
156: return (cs->invert);
157: }
158:
159: static __inline int
160: CHIN(cset *cs, wint_t ch)
161: {
162:
163: assert(ch >= 0);
164: if (ch < NC)
1.15 ! christos 165: return (((cs->bmp[(unsigned)ch >> 3] & (1 << (ch & 7))) != 0) ^
1.14 christos 166: cs->invert);
167: else if (cs->icase)
168: return (CHIN1(cs, ch) || CHIN1(cs, towlower(ch)) ||
169: CHIN1(cs, towupper(ch)));
170: else
171: return (CHIN1(cs, ch));
172: }
1.1 jtc 173:
174: /*
175: * main compiled-expression structure
176: */
177: struct re_guts {
178: int magic;
179: # define MAGIC2 ((('R'^0200)<<8)|'E')
180: sop *strip; /* malloced area for strip */
1.13 christos 181: size_t ncsets; /* number of csets in use */
1.1 jtc 182: cset *sets; /* -> cset [ncsets] */
183: int cflags; /* copy of regcomp() cflags argument */
184: sopno nstates; /* = number of sops */
185: sopno firststate; /* the initial OEND (normally 0) */
186: sopno laststate; /* the final OEND */
187: int iflags; /* internal flags */
188: # define USEBOL 01 /* used ^ */
189: # define USEEOL 02 /* used $ */
190: # define BAD 04 /* something wrong */
1.13 christos 191: size_t nbol; /* number of ^ used */
192: size_t neol; /* number of $ used */
1.1 jtc 193: char *must; /* match must contain this string */
1.14 christos 194: int moffset; /* latest point at which must may be located */
195: size_t *charjump; /* Boyer-Moore char jump table */
196: size_t *matchjump; /* Boyer-Moore match jump table */
1.13 christos 197: size_t mlen; /* length of must */
1.12 lukem 198: size_t nsub; /* copy of re_nsub */
1.1 jtc 199: int backrefs; /* does it use back references? */
200: sopno nplus; /* how deep does it nest +s? */
201: };
202:
203: /* misc utilities */
1.14 christos 204: #define OUT (CHAR_MIN - 1) /* a non-character value */
205: #define IGN (CHAR_MIN - 2)
206: #define ISWORD(c) (iswalnum((uch)(c)) || (c) == '_')
CVSweb <webmaster@jp.NetBSD.org>