lib/dfa.h

   1 /* dfa.h - declarations for GNU deterministic regexp compiler
   2    Copyright (C) 1988, 1998, 2007, 2009-2020 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 <regex.h>
  25 #include <stdbool.h>
  26 #include <stddef.h>
  27
  28 #ifdef __cplusplus
  29 extern "C" {
  30 #endif
  31
  32 struct localeinfo; /* See localeinfo.h.  */
  33
  34 /* Element of a list of strings, at least one of which is known to
  35    appear in any R.E. matching the DFA. */
  36 struct dfamust
  37 {
  38   bool exact;
  39   bool begline;
  40   bool endline;
  41   char must[FLEXIBLE_ARRAY_MEMBER];
  42 };
  43
  44 /* The dfa structure. It is completely opaque. */
  45 struct dfa;
  46
  47 /* Needed when Gnulib is not used.  */
  48 #ifndef _GL_ATTRIBUTE_MALLOC
  49 # define  _GL_ATTRIBUTE_MALLOC
  50 #endif
  51
  52 /* Entry points. */
  53
  54 /* Allocate a struct dfa.  The struct dfa is completely opaque.
  55    It should be initialized via dfasyntax or dfacopysyntax before other use.
  56    The returned pointer should be passed directly to free() after
  57    calling dfafree() on it. */
  58 extern struct dfa *dfaalloc (void) _GL_ATTRIBUTE_MALLOC;
  59
  60 /* DFA options that can be ORed together, for dfasyntax's 4th arg.  */
  61 enum
  62   {
  63     /* ^ and $ match only the start and end of data, and do not match
  64        end-of-line within data.  This is always false for grep, but
  65        possibly true for other apps.  */
  66     DFA_ANCHOR = 1 << 0,
  67
  68     /* '\0' in data is end-of-line, instead of the traditional '\n'.  */
  69     DFA_EOL_NUL = 1 << 1
  70   };
  71
  72 /* Initialize or reinitialize a DFA.  The arguments are:
  73    1. The DFA to operate on.
  74    2. Information about the current locale.
  75    3. Syntax bits described in regex.h.
  76    4. Additional DFA options described above.  */
  77 extern void dfasyntax (struct dfa *, struct localeinfo const *,
  78                        reg_syntax_t, int);
  79
  80 /* Initialize or reinitialize a DFA from an already-initialized DFA.  */
  81 extern void dfacopysyntax (struct dfa *, struct dfa const *);
  82
  83 /* Parse the given string of given length into the given struct dfa.  */
  84 extern void dfaparse (char const *, ptrdiff_t, struct dfa *);
  85
  86 /* Allocate and return a struct dfamust from a struct dfa that was
  87    initialized by dfaparse and not yet given to dfacomp.  */
  88 extern struct dfamust *dfamust (struct dfa const *);
  89
  90 /* Free the storage held by the components of a struct dfamust. */
  91 extern void dfamustfree (struct dfamust *);
  92
  93 /* Compile the given string of the given length into the given struct dfa.
  94    The last argument says whether to build a searching or an exact matcher.
  95    A null first argument means the struct dfa has already been
  96    initialized by dfaparse; the second argument is ignored.  */
  97 extern void dfacomp (char const *, ptrdiff_t, struct dfa *, bool);
  98
  99 /* Search through a buffer looking for a match to the given struct dfa.
 100    Find the first occurrence of a string matching the regexp in the
 101    buffer, and the shortest possible version thereof.  Return a pointer to
 102    the first character after the match, or NULL if none is found.  BEGIN
 103    points to the beginning of the buffer, and END points to the first byte
 104    after its end.  Note however that we store a sentinel byte (usually
 105    newline) in *END, so the actual buffer must be one byte longer.
 106    When ALLOW_NL is true, newlines may appear in the matching string.
 107    If COUNT is non-NULL, increment *COUNT once for each newline processed.
 108    Finally, if BACKREF is non-NULL set *BACKREF to indicate whether we
 109    encountered a back-reference.  The caller can use this to decide
 110    whether to fall back on a backtracking matcher.  */
 111 extern char *dfaexec (struct dfa *d, char const *begin, char *end,
 112                       bool allow_nl, ptrdiff_t *count, bool *backref);
 113
 114 /* Return a superset for D.  The superset matches everything that D
 115    matches, along with some other strings (though the latter should be
 116    rare, for efficiency reasons).  Return a null pointer if no useful
 117    superset is available.  */
 118 extern struct dfa *dfasuperset (struct dfa const *d) _GL_ATTRIBUTE_PURE;
 119
 120 /* The DFA is likely to be fast.  */
 121 extern bool dfaisfast (struct dfa const *) _GL_ATTRIBUTE_PURE;
 122
 123 /* Return true if every construct in D is supported by this DFA matcher.  */
 124 extern bool dfasupported (struct dfa const *) _GL_ATTRIBUTE_PURE;
 125
 126 /* Free the storage held by the components of a struct dfa. */
 127 extern void dfafree (struct dfa *);
 128
 129 /* Error handling. */
 130
 131 /* dfawarn() is called by the regexp routines whenever a regex is compiled
 132    that likely doesn't do what the user wanted.  It takes a single
 133    argument, a NUL-terminated string describing the situation.  The user
 134    must supply a dfawarn.  */
 135 extern void dfawarn (const char *);
 136
 137 /* dfaerror() is called by the regexp routines whenever an error occurs.  It
 138    takes a single argument, a NUL-terminated string describing the error.
 139    The user must supply a dfaerror.  */
 140 extern _Noreturn void dfaerror (const char *);
 141
 142 #ifdef __cplusplus
 143 }
 144 #endif
 145
 146 #endif /* dfa.h */