1 /*-
2 * SPDX-License-Identifier: BSD-3-Clause
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
36 /*
37 * First, the stuff that ends up in the outside-world include file
38 = typedef off_t regoff_t;
39 = typedef struct {
40 = int re_magic;
41 = size_t re_nsub; // number of parenthesized subexpressions
42 = const char *re_endp; // end pointer for REG_PEND
43 = struct re_guts *re_g; // none of your business :-)
44 = } regex_t;
45 = typedef struct {
46 = regoff_t rm_so; // start of match
47 = regoff_t rm_eo; // end of match
48 = } regmatch_t;
49 */
50 /*
51 * internals of regex_t
52 */
53 #define MAGIC1 ((('r'^0200)<<8) | 'e')
54
55 /*
56 * The internal representation is a *strip*, a sequence of
57 * operators ending with an endmarker. (Some terminology etc. is a
58 * historical relic of earlier versions which used multiple strips.)
59 * Certain oddities in the representation are there to permit running
60 * the machinery backwards; in particular, any deviation from sequential
61 * flow must be marked at both its source and its destination. Some
62 * fine points:
63 *
64 * - OPLUS_ and O_PLUS are *inside* the loop they create.
65 * - OQUEST_ and O_QUEST are *outside* the bypass they create.
66 * - OCH_ and O_CH are *outside* the multi-way branch they create, while
67 * OOR1 and OOR2 are respectively the end and the beginning of one of
68 * the branches. Note that there is an implicit OOR2 following OCH_
69 * and an implicit OOR1 preceding O_CH.
70 *
71 * In state representations, an operator's bit is on to signify a state
72 * immediately *preceding* "execution" of that operator.
73 */
74 typedef unsigned long sop; /* strip operator */
75 typedef unsigned long sopno;
76 #define OPRMASK 0xf8000000L
77 #define OPDMASK 0x07ffffffL
78 #define OPSHIFT ((unsigned)27)
79 #define OP(n) ((n)&OPRMASK)
80 #define OPND(n) ((n)&OPDMASK)
81 #define SOP(op, opnd) ((op)|(opnd))
82 /* operators meaning operand */
83 /* (back, fwd are offsets) */
84 #define OEND (1L<<OPSHIFT) /* endmarker - */
85 #define OCHAR (2L<<OPSHIFT) /* character wide character */
86 #define OBOL (3L<<OPSHIFT) /* left anchor - */
87 #define OEOL (4L<<OPSHIFT) /* right anchor - */
88 #define OANY (5L<<OPSHIFT) /* . - */
89 #define OANYOF (6L<<OPSHIFT) /* [...] set number */
90 #define OBACK_ (7L<<OPSHIFT) /* begin \d paren number */
91 #define O_BACK (8L<<OPSHIFT) /* end \d paren number */
92 #define OPLUS_ (9L<<OPSHIFT) /* + prefix fwd to suffix */
93 #define O_PLUS (10L<<OPSHIFT) /* + suffix back to prefix */
94 #define OQUEST_ (11L<<OPSHIFT) /* ? prefix fwd to suffix */
95 #define O_QUEST (12L<<OPSHIFT) /* ? suffix back to prefix */
96 #define OLPAREN (13L<<OPSHIFT) /* ( fwd to ) */
97 #define ORPAREN (14L<<OPSHIFT) /* ) back to ( */
98 #define OCH_ (15L<<OPSHIFT) /* begin choice fwd to OOR2 */
99 #define OOR1 (16L<<OPSHIFT) /* | pt. 1 back to OOR1 or OCH_ */
100 #define OOR2 (17L<<OPSHIFT) /* | pt. 2 fwd to OOR2 or O_CH */
101 #define O_CH (18L<<OPSHIFT) /* end choice back to OOR1 */
102 #define OBOW (19L<<OPSHIFT) /* begin word - */
103 #define OEOW (20L<<OPSHIFT) /* end word - */
104 #define OBOS (21L<<OPSHIFT) /* begin subj. - */
105 #define OEOS (22L<<OPSHIFT) /* end subj. - */
106 #define OWBND (23L<<OPSHIFT) /* word bound - */
107 #define ONWBND (24L<<OPSHIFT) /* not bound - */
108
109 /*
110 * Structures for [] character-set representation.
111 */
112 typedef struct {
113 wint_t min;
114 wint_t max;
115 } crange;
116 typedef struct {
117 unsigned char bmp[NC_MAX / 8];
118 wctype_t *types;
119 unsigned int ntypes;
120 wint_t *wides;
121 unsigned int nwides;
122 crange *ranges;
123 unsigned int nranges;
124 int invert;
125 int icase;
126 } cset;
127
128 static int
CHIN1(cset * cs,wint_t ch)129 CHIN1(cset *cs, wint_t ch)
130 {
131 unsigned int i;
132
133 assert(ch >= 0);
134 if (ch < NC)
135 return (((cs->bmp[ch >> 3] & (1 << (ch & 7))) != 0) ^
136 cs->invert);
137 for (i = 0; i < cs->nwides; i++) {
138 if (cs->icase) {
139 if (ch == towlower(cs->wides[i]) ||
140 ch == towupper(cs->wides[i]))
141 return (!cs->invert);
142 } else if (ch == cs->wides[i])
143 return (!cs->invert);
144 }
145 for (i = 0; i < cs->nranges; i++)
146 if (cs->ranges[i].min <= ch && ch <= cs->ranges[i].max)
147 return (!cs->invert);
148 for (i = 0; i < cs->ntypes; i++)
149 if (iswctype(ch, cs->types[i]))
150 return (!cs->invert);
151 return (cs->invert);
152 }
153
154 static __inline int
CHIN(cset * cs,wint_t ch)155 CHIN(cset *cs, wint_t ch)
156 {
157
158 assert(ch >= 0);
159 if (ch < NC)
160 return (((cs->bmp[ch >> 3] & (1 << (ch & 7))) != 0) ^
161 cs->invert);
162 else if (cs->icase)
163 return (CHIN1(cs, ch) || CHIN1(cs, towlower(ch)) ||
164 CHIN1(cs, towupper(ch)));
165 else
166 return (CHIN1(cs, ch));
167 }
168
169 /*
170 * main compiled-expression structure
171 */
172 struct re_guts {
173 int magic;
174 # define MAGIC2 ((('R'^0200)<<8)|'E')
175 sop *strip; /* malloced area for strip */
176 unsigned int ncsets; /* number of csets in use */
177 cset *sets; /* -> cset [ncsets] */
178 int cflags; /* copy of regcomp() cflags argument */
179 sopno nstates; /* = number of sops */
180 sopno firststate; /* the initial OEND (normally 0) */
181 sopno laststate; /* the final OEND */
182 int iflags; /* internal flags */
183 # define USEBOL 01 /* used ^ */
184 # define USEEOL 02 /* used $ */
185 # define BAD 04 /* something wrong */
186 int nbol; /* number of ^ used */
187 int neol; /* number of $ used */
188 char *must; /* match must contain this string */
189 int moffset; /* latest point at which must may be located */
190 int *charjump; /* Boyer-Moore char jump table */
191 int *matchjump; /* Boyer-Moore match jump table */
192 int mlen; /* length of must */
193 size_t nsub; /* copy of re_nsub */
194 int backrefs; /* does it use back references? */
195 sopno nplus; /* how deep does it nest +s? */
196 };
197
198 /* misc utilities */
199 #define OUT (CHAR_MIN - 1) /* a non-character value */
200 #define IGN (CHAR_MIN - 2)
201 #define ISWORD(c) (iswalnum((uch)(c)) || (c) == '_')
202