1 /* $NetBSD: regex2.h,v 1.16 2025/01/01 18:19:50 christos Exp $ */
2
3 /*-
4 * SPDX-License-Identifier: BSD-3-Clause
5 *
6 * Copyright (c) 1992, 1993, 1994 Henry Spencer.
7 * Copyright (c) 1992, 1993, 1994
8 * The Regents of the University of California. All rights reserved.
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
38 * $FreeBSD: head/lib/libc/regex/regex2.h 368359 2020-12-05 03:18:48Z kevans $
39 */
40
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
47 = const char *re_endp; // end pointer for REG_PEND
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 */
79 typedef uint32_t sop; /* strip operator */
80 typedef uint32_t sopno;
81 #define OPRMASK 0xf8000000U
82 #define OPDMASK 0x07ffffffU
83 #define OPSHIFT (27U)
84 #define OP(n) ((n)&OPRMASK)
85 #define OPND(n) ((n)&OPDMASK)
86 #define SOP(op, opnd) ((op)|(opnd))
87 /* operators meaning operand */
88 /* (back, fwd are offsets) */
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 - */
113
114 /*
115 * Structures for [] character-set representation.
116 */
117 typedef struct {
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;
131 } cset;
132
133 static int
CHIN1(cset * cs,wint_t ch)134 CHIN1(cset *cs, wint_t ch)
135 {
136 unsigned int i;
137
138 if ((unsigned)ch < NC)
139 return (((cs->bmp[(unsigned)ch >> 3] & (1 << (ch & 7))) != 0) ^
140 cs->invert);
141 for (i = 0; i < cs->nwides; i++) {
142 if (cs->icase) {
143 if (ch == towlower(cs->wides[i]) ||
144 ch == towupper(cs->wides[i]))
145 return (!cs->invert);
146 } else if (ch == cs->wides[i])
147 return (!cs->invert);
148 }
149 for (i = 0; i < cs->nranges; i++)
150 if (cs->ranges[i].min <= ch && ch <= cs->ranges[i].max)
151 return (!cs->invert);
152 for (i = 0; i < cs->ntypes; i++)
153 if (iswctype(ch, cs->types[i]))
154 return (!cs->invert);
155 return (cs->invert);
156 }
157
158 static __inline int
CHIN(cset * cs,wint_t ch)159 CHIN(cset *cs, wint_t ch)
160 {
161
162 if ((unsigned)ch < NC)
163 return (((cs->bmp[(unsigned)ch >> 3] & (1 << (ch & 7))) != 0) ^
164 cs->invert);
165 else if (cs->icase)
166 return (CHIN1(cs, ch) || CHIN1(cs, towlower(ch)) ||
167 CHIN1(cs, towupper(ch)));
168 else
169 return (CHIN1(cs, ch));
170 }
171
172 /*
173 * main compiled-expression structure
174 */
175 struct re_guts {
176 int magic;
177 # define MAGIC2 ((('R'^0200)<<8)|'E')
178 sop *strip; /* malloced area for strip */
179 size_t ncsets; /* number of csets in use */
180 cset *sets; /* -> cset [ncsets] */
181 int cflags; /* copy of regcomp() cflags argument */
182 sopno nstates; /* = number of sops */
183 sopno firststate; /* the initial OEND (normally 0) */
184 sopno laststate; /* the final OEND */
185 int iflags; /* internal flags */
186 # define USEBOL 01 /* used ^ */
187 # define USEEOL 02 /* used $ */
188 # define BAD 04 /* something wrong */
189 size_t nbol; /* number of ^ used */
190 size_t neol; /* number of $ used */
191 char *must; /* match must contain this string */
192 int moffset; /* latest point at which must may be located */
193 size_t *charjump; /* Boyer-Moore char jump table */
194 size_t *matchjump; /* Boyer-Moore match jump table */
195 size_t mlen; /* length of must */
196 size_t nsub; /* copy of re_nsub */
197 int backrefs; /* does it use back references? */
198 sopno nplus; /* how deep does it nest +s? */
199 };
200
201 /* misc utilities */
202 #define OUT (CHAR_MIN - 1) /* a non-character value */
203 #define IGN (CHAR_MIN - 2)
204 #define ISWORD(c) (iswalnum((uch)(c)) || (c) == '_')
205