1 /* dfa.h - declarations for GNU deterministic regexp compiler
2 Copyright (C) 1988, 1998, 2007, 2009-2022 Free Software Foundation, Inc.
3
4 This program is free software; you can redistribute it and/or modify
5 it under the terms of the GNU General Public License as published by
6 the Free Software Foundation, either version 3, or (at your option)
7 any later version.
8
9 This program is distributed in the hope that it will be useful,
10 but WITHOUT ANY WARRANTY; without even the implied warranty of
11 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12 GNU General Public License for more details.
13
14 You should have received a copy of the GNU General Public License
15 along with this program; if not, write to the Free Software
16 Foundation, Inc.,
17 51 Franklin Street - Fifth Floor, Boston, MA 02110-1301, USA */
18
19 /* Written June, 1988 by Mike Haertel */
20
21 #ifndef DFA_H_
22 #define DFA_H_
23
24 #include "idx.h"
25 #include <regex.h>
26 #include <stddef.h>
27 #include <stdlib.h>
28
29 #ifdef __cplusplus
30 extern "C" {
31 #endif
32
33 struct localeinfo; /* See localeinfo.h. */
34
35 /* Element of a list of strings, at least one of which is known to
36 appear in any R.E. matching the DFA. */
37 struct dfamust
38 {
39 bool exact;
40 bool begline;
41 bool endline;
42 char must[FLEXIBLE_ARRAY_MEMBER];
43 };
44
45 /* The dfa structure. It is completely opaque. */
46 struct dfa;
47
48 /* Needed when Gnulib is not used. */
49 #ifndef _GL_ATTRIBUTE_MALLOC
50 # define _GL_ATTRIBUTE_MALLOC
51 # define _GL_ATTRIBUTE_DEALLOC(f, i)
52 # define _GL_ATTRIBUTE_DEALLOC_FREE
53 # define _GL_ATTRIBUTE_RETURNS_NONNULL
54 #endif
55
56 /* Entry points. */
57
58 /* Allocate a struct dfa. The struct dfa is completely opaque.
59 It should be initialized via dfasyntax or dfacopysyntax before other use.
60 The returned pointer should be passed directly to free() after
61 calling dfafree() on it. */
62 extern struct dfa *dfaalloc (void)
63 _GL_ATTRIBUTE_MALLOC _GL_ATTRIBUTE_DEALLOC_FREE
64 _GL_ATTRIBUTE_RETURNS_NONNULL;
65
66 /* DFA options that can be ORed together, for dfasyntax's 4th arg. */
67 enum
68 {
69 /* ^ and $ match only the start and end of data, and do not match
70 end-of-line within data. This is always false for grep, but
71 possibly true for other apps. */
72 DFA_ANCHOR = 1 << 0,
73
74 /* '\0' in data is end-of-line, instead of the traditional '\n'. */
75 DFA_EOL_NUL = 1 << 1,
76
77 /* Treat [:alpha:] etc. as an error at the top level, instead of
78 merely a warning. */
79 DFA_CONFUSING_BRACKETS_ERROR = 1 << 2,
80
81 /* Warn about stray backslashes before ordinary characters other
82 than ] and } which are special because even though POSIX
83 says \] and \} have undefined interpretation, platforms
84 reliably ignore those stray backlashes and warning about them
85 would likely cause more trouble than it's worth. */
86 DFA_STRAY_BACKSLASH_WARN = 1 << 3,
87
88 /* Warn about * appearing out of context at the start of an
89 expression or subexpression. */
90 DFA_STAR_WARN = 1 << 4,
91
92 /* Warn about +, ?, {...} appearing out of context at the start of
93 an expression or subexpression. */
94 DFA_PLUS_WARN = 1 << 5,
95 };
96
97 /* Initialize or reinitialize a DFA. The arguments are:
98 1. The DFA to operate on.
99 2. Information about the current locale.
100 3. Syntax bits described in regex.h.
101 4. Additional DFA options described above. */
102 extern void dfasyntax (struct dfa *, struct localeinfo const *,
103 reg_syntax_t, int);
104
105 /* Initialize or reinitialize a DFA from an already-initialized DFA. */
106 extern void dfacopysyntax (struct dfa *, struct dfa const *);
107
108 /* Parse the given string of given length into the given struct dfa. */
109 extern void dfaparse (char const *, idx_t, struct dfa *);
110
111 struct dfamust;
112
113 /* Free the storage held by the components of a struct dfamust. */
114 extern void dfamustfree (struct dfamust *);
115
116 /* Allocate and return a struct dfamust from a struct dfa that was
117 initialized by dfaparse and not yet given to dfacomp. */
118 extern struct dfamust *dfamust (struct dfa const *)
119 _GL_ATTRIBUTE_DEALLOC (dfamustfree, 1);
120
121 /* Compile the given string of the given length into the given struct dfa.
122 The last argument says whether to build a searching or an exact matcher.
123 A null first argument means the struct dfa has already been
124 initialized by dfaparse; the second argument is ignored. */
125 extern void dfacomp (char const *, idx_t, struct dfa *, bool);
126
127 /* Search through a buffer looking for a match to the given struct dfa.
128 Find the first occurrence of a string matching the regexp in the
129 buffer, and the shortest possible version thereof. Return a pointer to
130 the first character after the match, or NULL if none is found. BEGIN
131 points to the beginning of the buffer, and END points to the first byte
132 after its end. Note however that we store a sentinel byte (usually
133 newline) in *END, so the actual buffer must be one byte longer.
134 When ALLOW_NL is true, newlines may appear in the matching string.
135 If COUNT is non-NULL, increment *COUNT once for each newline processed.
136 Finally, if BACKREF is non-NULL set *BACKREF to indicate whether we
137 encountered a back-reference. The caller can use this to decide
138 whether to fall back on a backtracking matcher. */
139 extern char *dfaexec (struct dfa *d, char const *begin, char *end,
140 bool allow_nl, idx_t *count, bool *backref);
141
142 /* Return a superset for D. The superset matches everything that D
143 matches, along with some other strings (though the latter should be
144 rare, for efficiency reasons). Return a null pointer if no useful
145 superset is available. */
146 extern struct dfa *dfasuperset (struct dfa const *d) _GL_ATTRIBUTE_PURE;
147
148 /* The DFA is likely to be fast. */
149 extern bool dfaisfast (struct dfa const *) _GL_ATTRIBUTE_PURE;
150
151 /* Return true if every construct in D is supported by this DFA matcher. */
152 extern bool dfasupported (struct dfa const *) _GL_ATTRIBUTE_PURE;
153
154 /* Free the storage held by the components of a struct dfa. */
155 extern void dfafree (struct dfa *);
156
157 /* Error handling. */
158
159 /* dfawarn() is called by the regexp routines whenever a regex is compiled
160 that likely doesn't do what the user wanted. It takes a single
161 argument, a NUL-terminated string describing the situation. The user
162 must supply a dfawarn. */
163 extern void dfawarn (const char *);
164
165 /* dfaerror() is called by the regexp routines whenever an error occurs. It
166 takes a single argument, a NUL-terminated string describing the error.
167 The user must supply a dfaerror. */
168 extern _Noreturn void dfaerror (const char *);
169
170 #ifdef __cplusplus
171 }
172 #endif
173
174 #endif /* dfa.h */