Norihiro Tanaka <nori...@kcn.ne.jp> wrote:
> This change is use bool for booleans at DFA interfaces. Callers in Gawk
> have already used bool instead of int for them. Second patch changes
> caller with change in first patch.
Sorry, Previous change was imcomplete. I fixed it, and added a third
patch.
From e4e13f8ed43b70fc874e0b8d84afb3c6382d7722 Mon Sep 17 00:00:00 2001
From: Norihiro Tanaka <nori...@kcn.ne.jp>
Date: Sat, 18 Oct 2014 23:57:33 +0900
Subject: [PATCH 1/3] dfa: prefer bool at DFA interfaces
* (src/dfa.c) dfasyntax: Use bool for `fold'.
dfaanalyze, dfacomp: Use bool for `searchflag'.
dfaexec_main, dfaexec_mb, dfaexec_sb, dfaexec: IUse bool for `allow_nl'.
* (src/dfa.h): Update prototype.
---
src/dfa.c | 20 ++++++++++----------
src/dfa.h | 8 ++++----
2 files changed, 14 insertions(+), 14 deletions(-)
diff --git a/src/dfa.c b/src/dfa.c
index 80510a8..b0b0c47 100644
--- a/src/dfa.c
+++ b/src/dfa.c
@@ -343,7 +343,7 @@ struct dfa
mbstate_t mbs; /* Multibyte conversion state. */
/* dfaexec implementation. */
- char *(*dfaexec) (struct dfa *, char const *, char *, int, size_t *, int *);
+ char *(*dfaexec) (struct dfa *, char const *, char *, bool, size_t *, int *);
/* The following are valid only if MB_CUR_MAX > 1. */
@@ -720,13 +720,13 @@ wchar_context (wint_t wc)
/* Entry point to set syntax options. */
void
-dfasyntax (reg_syntax_t bits, int fold, unsigned char eol)
+dfasyntax (reg_syntax_t bits, bool fold, unsigned char eol)
{
unsigned int i;
syntax_bits_set = 1;
syntax_bits = bits;
- case_fold = fold != 0;
+ case_fold = fold;
eolbyte = eol;
for (i = 0; i < NOTCHAR; ++i)
@@ -2320,7 +2320,7 @@ state_separate_contexts (position_set const *s)
scheme; the number of elements in each set deeper in the stack can be
used to determine the address of a particular set's array. */
void
-dfaanalyze (struct dfa *d, int searchflag)
+dfaanalyze (struct dfa *d, bool searchflag)
{
/* Array allocated to hold position sets. */
position *posalloc = xnmalloc (d->nleaves, 2 * sizeof *posalloc);
@@ -2356,7 +2356,7 @@ dfaanalyze (struct dfa *d, int searchflag)
putc ('\n', stderr);
#endif
- d->searchflag = searchflag != 0;
+ d->searchflag = searchflag;
alloc_position_set (&merged, d->nleaves);
d->follows = xcalloc (d->tindex, sizeof *d->follows);
@@ -3266,7 +3266,7 @@ skip_remains_mb (struct dfa *d, unsigned char const *p,
characters. */
static inline char *
dfaexec_main (struct dfa *d, char const *begin, char *end,
- int allow_nl, size_t *count, int *backref, bool multibyte)
+ bool allow_nl, size_t *count, int *backref, bool multibyte)
{
state_num s, s1; /* Current state. */
unsigned char const *p, *mbp; /* Current input character. */
@@ -3456,14 +3456,14 @@ dfaexec_main (struct dfa *d, char const *begin, char
*end,
static char *
dfaexec_mb (struct dfa *d, char const *begin, char *end,
- int allow_nl, size_t *count, int *backref)
+ bool allow_nl, size_t *count, int *backref)
{
return dfaexec_main (d, begin, end, allow_nl, count, backref, true);
}
static char *
dfaexec_sb (struct dfa *d, char const *begin, char *end,
- int allow_nl, size_t *count, int *backref)
+ bool allow_nl, size_t *count, int *backref)
{
return dfaexec_main (d, begin, end, allow_nl, count, backref, false);
}
@@ -3473,7 +3473,7 @@ dfaexec_sb (struct dfa *d, char const *begin, char *end,
char *
dfaexec (struct dfa *d, char const *begin, char *end,
- int allow_nl, size_t *count, int *backref)
+ bool allow_nl, size_t *count, int *backref)
{
return d->dfaexec (d, begin, end, allow_nl, count, backref);
}
@@ -3653,7 +3653,7 @@ dfassbuild (struct dfa *d)
/* Parse and analyze a single string of the given length. */
void
-dfacomp (char const *s, size_t len, struct dfa *d, int searchflag)
+dfacomp (char const *s, size_t len, struct dfa *d, bool searchflag)
{
dfainit (d);
dfambcache (d);
diff --git a/src/dfa.h b/src/dfa.h
index f30c3cb..ab2e8ea 100644
--- a/src/dfa.h
+++ b/src/dfa.h
@@ -49,12 +49,12 @@ extern struct dfamust *dfamusts (struct dfa const *);
/* dfasyntax() takes three arguments; the first sets the syntax bits described
earlier in this file, the second sets the case-folding flag, and the
third specifies the line terminator. */
-extern void dfasyntax (reg_syntax_t, int, unsigned char);
+extern void dfasyntax (reg_syntax_t, bool, unsigned char);
/* Compile the given string of the given length into the given struct dfa.
Final argument is a flag specifying whether to build a searching or an
exact matcher. */
-extern void dfacomp (char const *, size_t, struct dfa *, int);
+extern void dfacomp (char const *, size_t, struct dfa *, bool);
/* Search through a buffer looking for a match to the given struct dfa.
Find the first occurrence of a string matching the regexp in the
@@ -63,13 +63,13 @@ extern void dfacomp (char const *, size_t, struct dfa *,
int);
points to the beginning of the buffer, and END points to the first byte
after its end. Note however that we store a sentinel byte (usually
newline) in *END, so the actual buffer must be one byte longer.
- When NEWLINE is nonzero, newlines may appear in the matching string.
+ When ALLOW_NL is true, newlines may appear in the matching string.
If COUNT is non-NULL, increment *COUNT once for each newline processed.
Finally, if BACKREF is non-NULL set *BACKREF to indicate whether we
encountered a back-reference (1) or not (0). The caller may use this
to decide whether to fall back on a backtracking matcher. */
extern char *dfaexec (struct dfa *d, char const *begin, char *end,
- int newline, size_t *count, int *backref);
+ bool allow_nl, size_t *count, int *backref);
/* Return a superset for D. The superset matches everything that D
matches, along with some other strings (though the latter should be
--
2.1.1
From 8a998a5afbd8d3b0d09db64939957ccce7b7de4b Mon Sep 17 00:00:00 2001
From: Norihiro Tanaka <nori...@kcn.ne.jp>
Date: Thu, 23 Oct 2014 07:50:15 +0900
Subject: [PATCH 2/3] grep: change caller for previous change
* (src/dfasearch.c) GEAcompile: Use bool for boolean.
---
src/dfasearch.c | 9 +++++----
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/src/dfasearch.c b/src/dfasearch.c
index 77b4e3e..d667470 100644
--- a/src/dfasearch.c
+++ b/src/dfasearch.c
@@ -199,7 +199,7 @@ GEAcompile (char const *pattern, size_t size, reg_syntax_t
syntax_bits)
motif = NULL;
dfa = dfaalloc ();
- dfacomp (pattern, size, dfa, 1);
+ dfacomp (pattern, size, dfa, true);
kwsmusts ();
free(motif);
@@ -287,8 +287,8 @@ EGexecute (char const *buf, size_t size, size_t *match_size,
/* Keep using the superset while it reports multiline
potential matches; this is more likely to be fast
than falling back to KWset would be. */
- while ((next_beg = dfaexec (superset, dfa_beg, (char *) end, 1,
- &count, NULL))
+ while ((next_beg = dfaexec (superset, dfa_beg, (char *) end,
+ true, &count, NULL))
&& next_beg != end
&& count != 0)
{
@@ -307,7 +307,8 @@ EGexecute (char const *buf, size_t size, size_t *match_size,
}
/* Try matching with DFA. */
- next_beg = dfaexec (dfa, dfa_beg, (char *) end, 0, &count, &backref);
+ next_beg = dfaexec (dfa, dfa_beg, (char *) end, false, &count,
+ &backref);
/* If there's no match, or if we've matched the sentinel,
we're done. */
--
2.1.1
From a077e3d282137bb04267d4897e165b0577be0daf Mon Sep 17 00:00:00 2001
From: Norihiro Tanaka <nori...@kcn.ne.jp>
Date: Thu, 23 Oct 2014 08:08:45 +0900
Subject: [PATCH 3/3] dfa: change internal functions into static
* src/dfa.c (dfaparse, dfaanalyze dfastate, dfainit) change them into
static.
* src/dfa.h: Remove prototype for changed into static.
---
src/dfa.c | 8 ++++----
src/dfa.h | 14 --------------
2 files changed, 4 insertions(+), 18 deletions(-)
diff --git a/src/dfa.c b/src/dfa.c
index b0b0c47..5723597 100644
--- a/src/dfa.c
+++ b/src/dfa.c
@@ -1957,7 +1957,7 @@ regexp (void)
/* Main entry point for the parser. S is a string to be parsed, len is the
length of the string, so s can include NUL characters. D is a pointer to
the struct dfa to parse into. */
-void
+static void
dfaparse (char const *s, size_t len, struct dfa *d)
{
dfa = d;
@@ -2319,7 +2319,7 @@ state_separate_contexts (position_set const *s)
Sets are stored as arrays of the elements, obeying a stack-like allocation
scheme; the number of elements in each set deeper in the stack can be
used to determine the address of a particular set's array. */
-void
+static void
dfaanalyze (struct dfa *d, bool searchflag)
{
/* Array allocated to hold position sets. */
@@ -2558,7 +2558,7 @@ dfaanalyze (struct dfa *d, bool searchflag)
If after comparing with every group there are characters remaining in C,
create a new group labeled with the characters of C and insert this
position in that group. */
-void
+static void
dfastate (state_num s, struct dfa *d, state_num trans[])
{
leaf_set grps[NOTCHAR]; /* As many as will ever be needed. */
@@ -3522,7 +3522,7 @@ free_mbdata (struct dfa *d)
/* Initialize the components of a dfa that the other routines don't
initialize for themselves. */
-void
+static void
dfainit (struct dfa *d)
{
memset (d, 0, sizeof *d);
diff --git a/src/dfa.h b/src/dfa.h
index ab2e8ea..5983f64 100644
--- a/src/dfa.h
+++ b/src/dfa.h
@@ -85,20 +85,6 @@ extern void dfafree (struct dfa *);
/* Entry points for people who know what they're doing. */
-/* Initialize the components of a struct dfa. */
-extern void dfainit (struct dfa *);
-
-/* Incrementally parse a string of given length into a struct dfa. */
-extern void dfaparse (char const *, size_t, struct dfa *);
-
-/* Analyze a parsed regexp; second argument tells whether to build a searching
- or an exact matcher. */
-extern void dfaanalyze (struct dfa *, int);
-
-/* Compute, for each possible character, the transitions out of a given
- state, storing them in an array of integers. */
-extern void dfastate (ptrdiff_t, struct dfa *, ptrdiff_t []);
-
/* Error handling. */
/* dfawarn() is called by the regexp routines whenever a regex is compiled
--
2.1.1
