Merge branch 'nd/retire-fnmatch'

Replace our use of fnmatch(3) with a more feature-rich wildmatch.
A handful patches at the bottom have been moved to nd/wildmatch to
graduate as part of that branch, before this series solidifies.

We may want to mark USE_WILDMATCH as an experimental curiosity a
bit more clearly (i.e. should not be enabled in production
environment, because it will make the behaviour between builds
unpredictable).

* nd/retire-fnmatch:
  Makefile: add USE_WILDMATCH to use wildmatch as fnmatch
  wildmatch: advance faster in <asterisk> + <literal> patterns
  wildmatch: make a special case for "*/" with FNM_PATHNAME
  test-wildmatch: add "perf" command to compare wildmatch and fnmatch
  wildmatch: support "no FNM_PATHNAME" mode
  wildmatch: make dowild() take arbitrary flags
  wildmatch: rename constants and update prototype

Author: gitster

Makefile

@@ -105,6 +105,9 @@ all::
 # Define NO_FNMATCH_CASEFOLD if your fnmatch function doesn't have the
 # FNM_CASEFOLD GNU extension.
 #
+# Define USE_WILDMATCH if you want to use Git's wildmatch
+# implementation as fnmatch
+#
 # Define NO_GECOS_IN_PWENT if you don't have pw_gecos in struct passwd
 # in the C library.
 #
@@ -1225,6 +1228,9 @@ ifdef NO_FNMATCH_CASEFOLD
 	COMPAT_OBJS += compat/fnmatch/fnmatch.o
 endif
 endif
+ifdef USE_WILDMATCH
+	COMPAT_CFLAGS += -DUSE_WILDMATCH
+endif
 ifdef NO_SETENV
 	COMPAT_CFLAGS += -DNO_SETENV
 	COMPAT_OBJS += compat/setenv.o

dir.c

@@ -685,7 +685,8 @@ int match_pathname(const char *pathname, int pathlen,
 	}
 
 	return wildmatch(pattern, name,
-			 ignore_case ? FNM_CASEFOLD : 0) == 0;
+			 WM_PATHNAME | (ignore_case ? WM_CASEFOLD : 0),
+			 NULL) == 0;
 }
 
 /*

git-compat-util.h

@@ -112,7 +112,9 @@
 #include <sys/time.h>
 #include <time.h>
 #include <signal.h>
+#ifndef USE_WILDMATCH
 #include <fnmatch.h>
+#endif
 #include <assert.h>
 #include <regex.h>
 #include <utime.h>
@@ -280,6 +282,17 @@ extern char *gitbasename(char *);
 
 #include "compat/bswap.h"
 
+#ifdef USE_WILDMATCH
+#include "wildmatch.h"
+#define FNM_PATHNAME WM_PATHNAME
+#define FNM_CASEFOLD WM_CASEFOLD
+#define FNM_NOMATCH  WM_NOMATCH
+static inline int fnmatch(const char *pattern, const char *string, int flags)
+{
+	return wildmatch(pattern, string, flags, NULL);
+}
+#endif
+
 /* General helper functions */
 extern void vreportf(const char *prefix, const char *err, va_list params);
 extern void vwritef(int fd, const char *prefix, const char *err, va_list params);

t/t3070-wildmatch.sh

@@ -29,6 +29,18 @@ match() {
     fi
 }
 
+pathmatch() {
+    if [ $1 = 1 ]; then
+	test_expect_success "pathmatch:    match '$2' '$3'" "
+	    test-wildmatch pathmatch '$2' '$3'
+	"
+    else
+	test_expect_success "pathmatch: no match '$2' '$3'" "
+	    ! test-wildmatch pathmatch '$2' '$3'
+	"
+    fi
+}
+
 # Basic wildmat features
 match 1 1 foo foo
 match 0 0 foo bar
@@ -191,5 +203,36 @@ match 1 1 'XXX/adobe/courier/bold/o/normal//12/120/75/75/m/70/iso8859/1' 'XXX/*/
 match 0 0 'XXX/adobe/courier/bold/o/normal//12/120/75/75/X/70/iso8859/1' 'XXX/*/*/*/*/*/*/12/*/*/*/m/*/*/*'
 match 1 0 'abcd/abcdefg/abcdefghijk/abcdefghijklmnop.txt' '**/*a*b*g*n*t'
 match 0 0 'abcd/abcdefg/abcdefghijk/abcdefghijklmnop.txtz' '**/*a*b*g*n*t'
+match 0 x foo '*/*/*'
+match 0 x foo/bar '*/*/*'
+match 1 x foo/bba/arr '*/*/*'
+match 0 x foo/bb/aa/rr '*/*/*'
+match 1 x foo/bb/aa/rr '**/**/**'
+match 1 x abcXdefXghi '*X*i'
+match 0 x ab/cXd/efXg/hi '*X*i'
+match 1 x ab/cXd/efXg/hi '*/*X*/*/*i'
+match 1 x ab/cXd/efXg/hi '**/*X*/**/*i'
+
+pathmatch 1 foo foo
+pathmatch 0 foo fo
+pathmatch 1 foo/bar foo/bar
+pathmatch 1 foo/bar 'foo/*'
+pathmatch 1 foo/bba/arr 'foo/*'
+pathmatch 1 foo/bba/arr 'foo/**'
+pathmatch 1 foo/bba/arr 'foo*'
+pathmatch 1 foo/bba/arr 'foo**'
+pathmatch 1 foo/bba/arr 'foo/*arr'
+pathmatch 1 foo/bba/arr 'foo/**arr'
+pathmatch 0 foo/bba/arr 'foo/*z'
+pathmatch 0 foo/bba/arr 'foo/**z'
+pathmatch 1 foo/bar 'foo?bar'
+pathmatch 1 foo/bar 'foo[/]bar'
+pathmatch 0 foo '*/*/*'
+pathmatch 0 foo/bar '*/*/*'
+pathmatch 1 foo/bba/arr '*/*/*'
+pathmatch 1 foo/bb/aa/rr '*/*/*'
+pathmatch 1 abcXdefXghi '*X*i'
+pathmatch 1 ab/cXd/efXg/hi '*/*X*/*/*i'
+pathmatch 1 ab/cXd/efXg/hi '*Xg*i'
 
 test_done

test-wildmatch.c

@@ -1,9 +1,85 @@
+#ifdef USE_WILDMATCH
+#undef USE_WILDMATCH  /* We need real fnmatch implementation here */
+#endif
 #include "cache.h"
 #include "wildmatch.h"
 
+static int perf(int ac, char **av)
+{
+	struct timeval tv1, tv2;
+	struct stat st;
+	int fd, i, n, flags1 = 0, flags2 = 0;
+	char *buffer, *p;
+	uint32_t usec1, usec2;
+	const char *lang;
+	const char *file = av[0];
+	const char *pattern = av[1];
+
+	lang = getenv("LANG");
+	if (lang && strcmp(lang, "C"))
+		die("Please test it on C locale.");
+
+	if ((fd = open(file, O_RDONLY)) == -1 || fstat(fd, &st))
+		die_errno("file open");
+
+	buffer = xmalloc(st.st_size + 2);
+	if (read(fd, buffer, st.st_size) != st.st_size)
+		die_errno("read");
+
+	buffer[st.st_size] = '\0';
+	buffer[st.st_size + 1] = '\0';
+	for (i = 0; i < st.st_size; i++)
+		if (buffer[i] == '\n')
+			buffer[i] = '\0';
+
+	n = atoi(av[2]);
+	if (av[3] && !strcmp(av[3], "pathname")) {
+		flags1 = WM_PATHNAME;
+		flags2 = FNM_PATHNAME;
+	}
+
+	gettimeofday(&tv1, NULL);
+	for (i = 0; i < n; i++) {
+		for (p = buffer; *p; p += strlen(p) + 1)
+			wildmatch(pattern, p, flags1, NULL);
+	}
+	gettimeofday(&tv2, NULL);
+
+	usec1 = (uint32_t)tv2.tv_sec * 1000000 + tv2.tv_usec;
+	usec1 -= (uint32_t)tv1.tv_sec * 1000000 + tv1.tv_usec;
+	printf("wildmatch %ds %dus\n",
+	       (int)(usec1 / 1000000),
+	       (int)(usec1 % 1000000));
+
+	gettimeofday(&tv1, NULL);
+	for (i = 0; i < n; i++) {
+		for (p = buffer; *p; p += strlen(p) + 1)
+			fnmatch(pattern, p, flags2);
+	}
+	gettimeofday(&tv2, NULL);
+
+	usec2 = (uint32_t)tv2.tv_sec * 1000000 + tv2.tv_usec;
+	usec2 -= (uint32_t)tv1.tv_sec * 1000000 + tv1.tv_usec;
+	if (usec2 > usec1)
+		printf("fnmatch   %ds %dus or %.2f%% slower\n",
+		       (int)((usec2 - usec1) / 1000000),
+		       (int)((usec2 - usec1) % 1000000),
+		       (float)(usec2 - usec1) / usec1 * 100);
+	else
+		printf("fnmatch   %ds %dus or %.2f%% faster\n",
+		       (int)((usec1 - usec2) / 1000000),
+		       (int)((usec1 - usec2) % 1000000),
+		       (float)(usec1 - usec2) / usec1 * 100);
+	return 0;
+}
+
 int main(int argc, char **argv)
 {
 	int i;
+
+	if (!strcmp(argv[1], "perf"))
+		return perf(argc - 2, argv + 2);
+
 	for (i = 2; i < argc; i++) {
 		if (argv[i][0] == '/')
 			die("Forward slash is not allowed at the beginning of the\n"
@@ -12,9 +88,11 @@ int main(int argc, char **argv)
 			argv[i] += 3;
 	}
 	if (!strcmp(argv[1], "wildmatch"))
-		return !!wildmatch(argv[3], argv[2], 0);
+		return !!wildmatch(argv[3], argv[2], WM_PATHNAME, NULL);
 	else if (!strcmp(argv[1], "iwildmatch"))
-		return !!wildmatch(argv[3], argv[2], FNM_CASEFOLD);
+		return !!wildmatch(argv[3], argv[2], WM_PATHNAME | WM_CASEFOLD, NULL);
+	else if (!strcmp(argv[1], "pathmatch"))
+		return !!wildmatch(argv[3], argv[2], 0, NULL);
 	else if (!strcmp(argv[1], "fnmatch"))
 		return !!fnmatch(argv[3], argv[2], FNM_PATHNAME);
 	else