6210677 dirname and basename are slower than they should be

--HG--
rename : usr/src/cmd/basename/basename.sh => deleted_files/usr/src/cmd/basename/basename.sh
rename : usr/src/cmd/dirname/dirname.sh => deleted_files/usr/src/cmd/dirname/dirname.sh

7 files changed, 222 insertions, 151 deletions

diff --git a/usr/src/Makefile.lint b/usr/src/Makefile.lint
index d7e724c006..9f650e4a01 100644
--- a/usr/src/Makefile.lint
+++ b/usr/src/Makefile.lint
@@ -103,6 +103,7 @@ COMMON_SUBDIRS = \
 	cmd/devmgmt \
 	cmd/dfs.cmds \
 	cmd/diff3 \
+	cmd/dirname \
 	cmd/diskscan \
 	cmd/dispadmin \
 	cmd/dladm \
diff --git a/usr/src/cmd/basename/Makefile b/usr/src/cmd/basename/Makefile
index f6da64048b..e29b1b6220 100644
--- a/usr/src/cmd/basename/Makefile
+++ b/usr/src/cmd/basename/Makefile
@@ -22,32 +22,29 @@
 #
 #ident	"%Z%%M%	%I%	%E% SMI"
 #
-# Copyright 2004 Sun Microsystems, Inc.  All rights reserved.
+# Copyright 2005 Sun Microsystems, Inc.  All rights reserved.
 # Use is subject to license terms.
 #
 
 PROG= basename
 XPG4PROG= basename
 
-SRCS= $(PROG:%=%.sh)
-
 include ../Makefile.cmd
 
-$(XPG4) := CFLAGS += $(CCVERBOSE)
+$(XPG4) := CFLAGS += -DXPG4 $(CCVERBOSE)
+lint_PROG_XPG4 := CPPFLAGS += -DXPG4
 
 .KEEP_STATE:
 
 all: $(PROG) $(XPG4)
 
-$(PROG): $(PROG).sh
-	$(RM) $@
-	cat $(PROG).sh > $@
-	chmod +x $@
-
 install: all $(ROOTPROG) $(ROOTXPG4PROG)
 
 clean:
 
-lint: lint_PROG
+lint: lint_PROG lint_PROG_XPG4
+
+lint_PROG_XPG4:	$(PROG).c
+	$(LINT.c) $(PROG).c $(LDLIBS)
 
 include ../Makefile.targ
diff --git a/usr/src/cmd/basename/basename.c b/usr/src/cmd/basename/basename.c
index c4d3fbb733..af4026a488 100644
--- a/usr/src/cmd/basename/basename.c
+++ b/usr/src/cmd/basename/basename.c
@@ -20,7 +20,7 @@
  * CDDL HEADER END
  */
 /*
- * Copyright 2004 Sun Microsystems, Inc.  All rights reserved.
+ * Copyright 2005 Sun Microsystems, Inc.  All rights reserved.
  * Use is subject to license terms.
  */
 
@@ -30,9 +30,11 @@
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
-
-static void output(char *);
-static void usage(void);
+#ifndef	XPG4
+#include <unistd.h>
+#include <regex.h>
+#include <libintl.h>
+#endif
 
 int
 main(int argc, char **argv)
@@ -40,71 +42,130 @@ main(int argc, char **argv)
 	char	*p;
 	char	*string;
 	char	*suffix;
+#ifndef	XPG4
+	int	r;
+	char	suf_buf[256];
+	char	*suf_pat;
+	size_t	suf_len;
+	regex_t	reg;
+	regmatch_t	pmatch[2];
+#endif
 
-	(void) setlocale(LC_ALL, "");
+	/*
+	 * For better performance, defer the setlocale()/textdomain()
+	 * calls until they get really required.
+	 */
 #if !defined(TEXT_DOMAIN)
 #define	TEXT_DOMAIN "SYS_TEST"
 #endif
-	(void) textdomain(TEXT_DOMAIN);
-
-	if (argc == 1)
-		output(".");
+	if (argc == 1) {
+		(void) puts(".");
+		return (0);
+	}
 
+#ifdef	XPG4
 	if (strcmp(argv[1], "--") == 0) {
 		argv++;
 		argc--;
+		if (argc == 1) {
+			(void) puts(".");
+			return (0);
+		}
+	}
+#endif
+	if (argc > 3) {
+		(void) setlocale(LC_ALL, "");
+		(void) textdomain(TEXT_DOMAIN);
+		(void) fputs(gettext("Usage: basename string [ suffix ]\n"),
+		    stderr);
+		return (1);
 	}
-
-	if (argc == 1)
-		output(".");
-
-	if (argc > 3)
-		usage();
 
 	string = argv[1];
 	suffix = (argc == 2) ? NULL : argv[2];
 
-	if (*string == '\0')
-		output(".");
+	if (*string == '\0') {
+		(void) puts(".");
+		return (0);
+	}
 
 	/* remove trailing slashes */
-	p = string + strlen(string) -1;
-	while ((p >= string) && (*p == '/'))
+	p = string + strlen(string) - 1;
+	while (p >= string && *p == '/')
 		*p-- = '\0';
 
-	if (*string == '\0')
-		output("/");
+	if (*string == '\0') {
+		(void) puts("/");
+		return (0);
+	}
 
 	/* skip to one past last slash */
 	if ((p = strrchr(string, '/')) != NULL)
 		string = p + 1;
 
+	if (suffix == NULL) {
+		(void) puts(string);
+		return (0);
+	}
+
+#ifdef	XPG4
 	/*
 	 * if a suffix is present and is not the same as the remaining
 	 * string and is identical to the last characters in the remaining
 	 * string, remove those characters from the string.
 	 */
-	if (suffix != NULL)
-		if (strcmp(string, suffix) != NULL) {
-			p = string + strlen(string) - strlen(suffix);
-			if (strcmp(p, suffix) == NULL)
-				*p = '\0';
-		}
-
-	output(string);
+	if (strcmp(string, suffix) != 0) {
+		p = string + strlen(string) - strlen(suffix);
+		if (strcmp(p, suffix) == 0)
+			*p = '\0';
+	}
+	(void) puts(string);
 	return (0);
-}
-
-static void
-output(char *string)
-{
-	(void) printf("%s\n", string);
-	exit(0);
-}
+#else
+	(void) setlocale(LC_ALL, "");
+	(void) textdomain(TEXT_DOMAIN);
 
-static void usage(void)
-{
-	(void) fprintf(stderr,
-	    gettext("Usage: basename string [ suffix ]\n"));
-	exit(1);
+	suf_len = 6 + strlen(suffix) + 1 + 1; /* \(.*\)suffix$ */
+	if (suf_len > sizeof (suf_buf)) {
+		suf_pat = malloc(suf_len);
+		if (suf_pat == NULL) {
+			(void) fputs("malloc failed\n", stderr);
+			return (1);
+		}
+	} else {
+		suf_pat = suf_buf;
+	}
+	(void) strcpy(suf_pat, "\\(.*\\)");
+	(void) strcpy(suf_pat + 6, suffix);
+	*(suf_pat + suf_len - 1 - 1) = '$';
+	*(suf_pat + suf_len - 1) = '\0';
+
+	r = regcomp(&reg, suf_pat, 0);
+	if (r != 0) {
+		(void) fprintf(stderr,
+		    "Internal error: regcomp failed for \"%s\"\n",
+		    suf_pat);
+		return (1);
+	}
+	r = regexec(&reg, string, 2, pmatch, 0);
+	if (r == 0) {
+		if (pmatch[0].rm_so == (regoff_t)-1 ||
+		    pmatch[1].rm_so == (regoff_t)-1 ||
+		    pmatch[1].rm_so != 0) {
+			(void) fprintf(stderr, "Internal error: regexec did "
+			    "not set sub-expression for:\n");
+			(void) fprintf(stderr, "path: \"%s\"\n", string);
+			(void) fprintf(stderr, "pattern: \"%s\"", suf_pat);
+			return (1);
+		}
+		if (pmatch[1].rm_so == pmatch[1].rm_eo) {
+			/* a null string matched */
+			(void) printf("%s\n", string);
+			return (0);
+		}
+		string[pmatch[1].rm_eo] = '\0';
+	}
+	(void) puts(string);
+	return (0);
+#endif
 }
diff --git a/usr/src/cmd/basename/basename.sh b/usr/src/cmd/basename/basename.sh
deleted file mode 100644
index 2c7249f007..0000000000
--- a/usr/src/cmd/basename/basename.sh
+++ /dev/null
@@ -1,42 +0,0 @@
-#!/usr/bin/sh
-#
-# CDDL HEADER START
-#
-# The contents of this file are subject to the terms of the
-# Common Development and Distribution License, Version 1.0 only
-# (the "License").  You may not use this file except in compliance
-# with the License.
-#
-# You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
-# or http://www.opensolaris.org/os/licensing.
-# See the License for the specific language governing permissions
-# and limitations under the License.
-#
-# When distributing Covered Code, include this CDDL HEADER in each
-# file and include the License file at usr/src/OPENSOLARIS.LICENSE.
-# If applicable, add the following below this CDDL HEADER, with the
-# fields enclosed by brackets "[]" replaced with your own identifying
-# information: Portions Copyright [yyyy] [name of copyright owner]
-#
-# CDDL HEADER END
-#
-#	Copyright (c) 1984, 1986, 1987, 1988, 1989 AT&T
-#	  All Rights Reserved
-
-
-#ident	"%Z%%M%	%I%	%E% SMI"	/* SVr4.0 1.8	*/
-if [ $# -gt 2 ]
-then
-	echo >&2 "Usage:  basename [ path [ suffix-pattern ] ]"
-	exit 1
-fi
-#	If no first argument or first argument is null, make first argument
-#	"."  Add beginning slash, then remove trailing slashes, then remove 
-#	everything up through last slash, then remove suffix pattern if 
-#	second argument is present.
-#	If nothing is left, first argument must be of form //*, in which
-# 	case the basename is /.
-exec /usr/bin/expr \
-	"/${1:-.}" : '\(.*[^/]\)/*$' : '.*/\(..*\)' : "\\(.*\\)$2\$"  \|  \
-	"/${1:-.}" : '\(.*[^/]\)/*$' : '.*/\(..*\)'    \|  \
-	"/${1:-.}" : '.*/\(..*\)' 
diff --git a/usr/src/cmd/dirname/Makefile b/usr/src/cmd/dirname/Makefile
index 14ff338b02..c4fd585cb9 100644
--- a/usr/src/cmd/dirname/Makefile
+++ b/usr/src/cmd/dirname/Makefile
@@ -22,13 +22,12 @@
 #
 #ident	"%Z%%M%	%I%	%E% SMI"
 #
-# Copyright (c) 1989 by Sun Microsystems, Inc.
+# Copyright 2005 Sun Microsystems, Inc.  All rights reserved.
+# Use is subject to license terms.
 #
 
 PROG= dirname
 
-SRCS= $(PROG:%=%.sh)
-
 include ../Makefile.cmd
 
 .KEEP_STATE:
@@ -39,6 +38,6 @@ install: all $(ROOTPROG)
 
 clean:
 
-lint:
+lint:	lint_PROG
 
 include ../Makefile.targ
diff --git a/usr/src/cmd/dirname/dirname.c b/usr/src/cmd/dirname/dirname.c
new file mode 100644
index 0000000000..6a07801ea2
--- /dev/null
+++ b/usr/src/cmd/dirname/dirname.c
@@ -0,0 +1,108 @@
+/*
+ * CDDL HEADER START
+ *
+ * The contents of this file are subject to the terms of the
+ * Common Development and Distribution License, Version 1.0 only
+ * (the "License").  You may not use this file except in compliance
+ * with the License.
+ *
+ * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
+ * or http://www.opensolaris.org/os/licensing.
+ * See the License for the specific language governing permissions
+ * and limitations under the License.
+ *
+ * When distributing Covered Code, include this CDDL HEADER in each
+ * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
+ * If applicable, add the following below this CDDL HEADER, with the
+ * fields enclosed by brackets "[]" replaced with your own identifying
+ * information: Portions Copyright [yyyy] [name of copyright owner]
+ *
+ * CDDL HEADER END
+ */
+/*
+ * Copyright 2005 Sun Microsystems, Inc.  All rights reserved.
+ * Use is subject to license terms.
+ */
+
+#pragma ident	"%Z%%M%	%I%	%E% SMI"
+
+#include <locale.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <libintl.h>
+
+int
+main(int argc, char **argv)
+{
+	char	*p;
+	char	*string;
+
+	/*
+	 * For better performance, defer the setlocale()/textdomain()
+	 * calls until they get really required.
+	 */
+#if !defined(TEXT_DOMAIN)
+#define	TEXT_DOMAIN "SYS_TEST"
+#endif
+	if (argc == 1) {
+		(void) puts(".");
+		return (0);
+	}
+	if (strcmp(argv[1], "--") == 0) {
+		argv++;
+		argc--;
+		if (argc == 1) {
+			(void) puts(".");
+			return (0);
+		}
+	}
+	if (argc > 2) {
+		(void) setlocale(LC_ALL, "");
+		(void) textdomain(TEXT_DOMAIN);
+		(void) fprintf(stderr, gettext("Usage: dirname [ path ]\n"));
+		return (1);
+	}
+
+	string = argv[1];
+
+	if (*string == '\0') {
+		(void) puts(".");
+		return (0);
+	}
+
+	/* remove trailing slashes */
+	p = string + strlen(string) - 1;
+	while (p >= string && *p == '/')
+		*p-- = '\0';
+
+	if (*string == '\0') {
+		/* string contained only slashes */
+		(void) puts("/");
+		return (0);
+	}
+
+	/* remove non-slashes */
+	while (p >= string && *p != '/')
+		*p-- = '\0';
+
+	if (*string == '\0') {
+		/* string did not begin with a slash */
+		(void) puts(".");
+		return (0);
+	}
+
+	/* remove slashes delimiting dirname and basename */
+	while (p >= string && *p == '/')
+		*p-- = '\0';
+
+	if (*string == '\0') {
+		/* no dirname part found */
+		(void) puts("/");
+		return (0);
+	}
+	/* now string points to dirname part */
+	(void) puts(string);
+	return (0);
+}
diff --git a/usr/src/cmd/dirname/dirname.sh b/usr/src/cmd/dirname/dirname.sh
deleted file mode 100644
index d3cc0928f6..0000000000
--- a/usr/src/cmd/dirname/dirname.sh
+++ /dev/null
@@ -1,53 +0,0 @@
-#!/usr/bin/sh
-#
-# CDDL HEADER START
-#
-# The contents of this file are subject to the terms of the
-# Common Development and Distribution License, Version 1.0 only
-# (the "License").  You may not use this file except in compliance
-# with the License.
-#
-# You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
-# or http://www.opensolaris.org/os/licensing.
-# See the License for the specific language governing permissions
-# and limitations under the License.
-#
-# When distributing Covered Code, include this CDDL HEADER in each
-# file and include the License file at usr/src/OPENSOLARIS.LICENSE.
-# If applicable, add the following below this CDDL HEADER, with the
-# fields enclosed by brackets "[]" replaced with your own identifying
-# information: Portions Copyright [yyyy] [name of copyright owner]
-#
-# CDDL HEADER END
-#
-#	Copyright (c) 1984, 1986, 1987, 1988, 1989 AT&T
-#	  All Rights Reserved
-#
-#
-# Copyright 2004 Sun Microsystems, Inc.  All rights reserved.
-# Use is subject to license terms.
-#
-
-#ident	"%Z%%M%	%I%	%E% SMI"
-
-if [ $# -gt 1 ]
-then
-	if [ "$1" = "--" -a $# -le 2 ]
-	then 
-		shift
-	else
-		echo >&2 `/usr/bin/gettext TEXT_DOMAIN \
-		    "usage: dirname [ path ]"`
-		exit 1
-	fi
-fi
-
-#	First check for pathnames of form //*non-slash*/* in which case the 
-#	dirname is /.
-#	Otherwise, remove the last component in the pathname and slashes 
-#	that come before it.
-#	If nothing is left, dirname is "."
-exec /usr/bin/expr \
-	"${1:-.}/" : '\(/\)/*[^/]*//*$'  \| \
-	"${1:-.}/" : '\(.*[^/]\)//*[^/][^/]*//*$' \| \
-	.