Update awk to Sep 6, 2023 version.

7 files changed, 28 insertions, 22 deletions

diff --git a/usr.bin/awk/FIXES b/usr.bin/awk/FIXES
index 53c78410647..8cbd6ac1a09 100644
--- a/usr.bin/awk/FIXES
+++ b/usr.bin/awk/FIXES
@@ -25,6 +25,14 @@ THIS SOFTWARE.
 This file lists all bug fixes, changes, etc., made since the AWK book
 was sent to the printers in August 1987.
 
+Sep 06, 2023:
+	Fix edge case where FS is changed on commandline. Thanks to 
+	Gordon Shephard and Miguel Pineiro Jr.
+
+	Fix regular expression clobbering in the lexer, where lexer does
+	not make a copy of regexp literals. also makedfa memory leaks have
+	been plugged. Thanks to Miguel Pineiro Jr.
+	
 Dec 15, 2022:
 	Force hex escapes in strings to be no more than two characters,
 	as they already are in regular expressions. This brings internal
diff --git a/usr.bin/awk/awkgram.y b/usr.bin/awk/awkgram.y
index f9a5330078b..9894bcc7354 100644
--- a/usr.bin/awk/awkgram.y
+++ b/usr.bin/awk/awkgram.y
@@ -1,4 +1,4 @@
-/*	$OpenBSD: awkgram.y,v 1.15 2022/09/01 15:21:28 millert Exp $	*/
+/*	$OpenBSD: awkgram.y,v 1.16 2023/09/10 14:59:00 millert Exp $	*/
 /****************************************************************
 Copyright (C) Lucent Technologies 1997
 All Rights Reserved
@@ -205,7 +205,7 @@ ppattern:
 		{ $$ = op2(BOR, notnull($1), notnull($3)); }
 	| ppattern and ppattern %prec AND
 		{ $$ = op2(AND, notnull($1), notnull($3)); }
-	| ppattern MATCHOP reg_expr	{ $$ = op3($2, NIL, $1, (Node*)makedfa($3, 0)); }
+	| ppattern MATCHOP reg_expr	{ $$ = op3($2, NIL, $1, (Node*)makedfa($3, 0)); free($3); }
 	| ppattern MATCHOP ppattern
 		{ if (constnode($3)) {
 			$$ = op3($2, NIL, $1, (Node*)makedfa(strnode($3), 0));
@@ -233,7 +233,7 @@ pattern:
 	| pattern LE pattern		{ $$ = op2($2, $1, $3); }
 	| pattern LT pattern		{ $$ = op2($2, $1, $3); }
 	| pattern NE pattern		{ $$ = op2($2, $1, $3); }
-	| pattern MATCHOP reg_expr	{ $$ = op3($2, NIL, $1, (Node*)makedfa($3, 0)); }
+	| pattern MATCHOP reg_expr	{ $$ = op3($2, NIL, $1, (Node*)makedfa($3, 0)); free($3); }
 	| pattern MATCHOP pattern
 		{ if (constnode($3)) {
 			$$ = op3($2, NIL, $1, (Node*)makedfa(strnode($3), 0));
@@ -283,7 +283,7 @@ rbrace:
 
 re:
 	   reg_expr
-		{ $$ = op3(MATCH, NIL, rectonode(), (Node*)makedfa($1, 0)); }
+		{ $$ = op3(MATCH, NIL, rectonode(), (Node*)makedfa($1, 0)); free($1); }
 	| NOT re	{ $$ = op1(NOT, notnull($2)); }
 	;
 
@@ -407,7 +407,7 @@ term:
 		  $$ = op2(INDEX, $3, (Node*)$5); }
 	| '(' pattern ')'		{ $$ = $2; }
 	| MATCHFCN '(' pattern comma reg_expr ')'
-		{ $$ = op3(MATCHFCN, NIL, $3, (Node*)makedfa($5, 1)); }
+		{ $$ = op3(MATCHFCN, NIL, $3, (Node*)makedfa($5, 1)); free($5); }
 	| MATCHFCN '(' pattern comma pattern ')'
 		{ if (constnode($5)) {
 			$$ = op3(MATCHFCN, NIL, $3, (Node*)makedfa(strnode($5), 1));
@@ -418,13 +418,13 @@ term:
 	| SPLIT '(' pattern comma varname comma pattern ')'     /* string */
 		{ $$ = op4(SPLIT, $3, makearr($5), $7, (Node*)STRING); }
 	| SPLIT '(' pattern comma varname comma reg_expr ')'    /* const /regexp/ */
-		{ $$ = op4(SPLIT, $3, makearr($5), (Node*)makedfa($7, 1), (Node *)REGEXPR); }
+		{ $$ = op4(SPLIT, $3, makearr($5), (Node*)makedfa($7, 1), (Node *)REGEXPR); free($7); }
 	| SPLIT '(' pattern comma varname ')'
 		{ $$ = op4(SPLIT, $3, makearr($5), NIL, (Node*)STRING); }  /* default */
 	| SPRINTF '(' patlist ')'	{ $$ = op1($1, $3); }
 	| string	 		{ $$ = celltonode($1, CCON); }
 	| subop '(' reg_expr comma pattern ')'
-		{ $$ = op4($1, NIL, (Node*)makedfa($3, 1), $5, rectonode()); }
+		{ $$ = op4($1, NIL, (Node*)makedfa($3, 1), $5, rectonode()); free($3); }
 	| subop '(' pattern comma pattern ')'
 		{ if (constnode($3)) {
 			$$ = op4($1, NIL, (Node*)makedfa(strnode($3), 1), $5, rectonode());
@@ -432,7 +432,7 @@ term:
 		  } else
 			$$ = op4($1, (Node *)1, $3, $5, rectonode()); }
 	| subop '(' reg_expr comma pattern comma var ')'
-		{ $$ = op4($1, NIL, (Node*)makedfa($3, 1), $5, $7); }
+		{ $$ = op4($1, NIL, (Node*)makedfa($3, 1), $5, $7); free($3); }
 	| subop '(' pattern comma pattern comma var ')'
 		{ if (constnode($3)) {
 			$$ = op4($1, NIL, (Node*)makedfa(strnode($3), 1), $5, $7);
diff --git a/usr.bin/awk/lex.c b/usr.bin/awk/lex.c
index 213f5d0a9e7..9ca1d0dda17 100644
--- a/usr.bin/awk/lex.c
+++ b/usr.bin/awk/lex.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: lex.c,v 1.29 2023/09/09 18:59:43 millert Exp $	*/
+/*	$OpenBSD: lex.c,v 1.30 2023/09/10 14:59:00 millert Exp $	*/
 /****************************************************************
 Copyright (C) Lucent Technologies 1997
 All Rights Reserved
@@ -588,7 +588,7 @@ int regexpr(void)
 	*bp = 0;
 	if (c == 0)
 		SYNTAX("non-terminated regular expression %.10s...", buf);
-	yylval.s = buf;
+	yylval.s = tostring(buf);
 	unput('/');
 	RET(REGEXPR);
 }
diff --git a/usr.bin/awk/lib.c b/usr.bin/awk/lib.c
index 87cc78ea58a..90d34a7eff1 100644
--- a/usr.bin/awk/lib.c
+++ b/usr.bin/awk/lib.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: lib.c,v 1.49 2022/09/01 15:21:28 millert Exp $	*/
+/*	$OpenBSD: lib.c,v 1.50 2023/09/10 14:59:00 millert Exp $	*/
 /****************************************************************
 Copyright (C) Lucent Technologies 1997
 All Rights Reserved
@@ -148,11 +148,6 @@ int getrec(char **pbuf, int *pbufsize, bool isrecord)	/* get next input record *
 	}
 	DPRINTF("RS=<%s>, FS=<%s>, ARGC=%g, FILENAME=%s\n",
 		*RS, *FS, *ARGC, *FILENAME);
-	if (isrecord) {
-		donefld = false;
-		donerec = true;
-		savefs();
-	}
 	saveb0 = buf[0];
 	buf[0] = 0;
 	while (argno < *ARGC || infile == stdin) {
@@ -192,6 +187,9 @@ int getrec(char **pbuf, int *pbufsize, bool isrecord)	/* get next input record *
 					fldtab[0]->fval = result;
 					fldtab[0]->tval |= NUM;
 				}
+				donefld = false;
+				donerec = true;
+				savefs();
 			}
 			setfval(nrloc, nrloc->fval+1);
 			setfval(fnrloc, fnrloc->fval+1);
diff --git a/usr.bin/awk/main.c b/usr.bin/awk/main.c
index ccd557c1ef2..67eeff92394 100644
--- a/usr.bin/awk/main.c
+++ b/usr.bin/awk/main.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: main.c,v 1.57 2023/09/09 18:59:43 millert Exp $	*/
+/*	$OpenBSD: main.c,v 1.58 2023/09/10 14:59:00 millert Exp $	*/
 /****************************************************************
 Copyright (C) Lucent Technologies 1997
 All Rights Reserved
@@ -23,7 +23,7 @@ ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF
 THIS SOFTWARE.
 ****************************************************************/
 
-const char	*version = "version 20221215";
+const char	*version = "version 20230909";
 
 #define DEBUG
 #include <stdio.h>
diff --git a/usr.bin/awk/parse.c b/usr.bin/awk/parse.c
index abd5c4fc6fc..cb192deda60 100644
--- a/usr.bin/awk/parse.c
+++ b/usr.bin/awk/parse.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: parse.c,v 1.13 2020/12/09 20:00:11 millert Exp $	*/
+/*	$OpenBSD: parse.c,v 1.14 2023/09/10 14:59:00 millert Exp $	*/
 /****************************************************************
 Copyright (C) Lucent Technologies 1997
 All Rights Reserved
@@ -30,7 +30,7 @@ THIS SOFTWARE.
 #include "awk.h"
 #include "awkgram.tab.h"
 
-Node *nodealloc(int n)
+Node *nodealloc(size_t n)
 {
 	Node *x;
 
diff --git a/usr.bin/awk/proto.h b/usr.bin/awk/proto.h
index 0c37d53725d..374c5af9451 100644
--- a/usr.bin/awk/proto.h
+++ b/usr.bin/awk/proto.h
@@ -1,4 +1,4 @@
-/*	$OpenBSD: proto.h,v 1.20 2020/12/09 20:00:11 millert Exp $	*/
+/*	$OpenBSD: proto.h,v 1.21 2023/09/10 14:59:00 millert Exp $	*/
 /****************************************************************
 Copyright (C) Lucent Technologies 1997
 All Rights Reserved
@@ -69,7 +69,7 @@ extern	void	freefa(fa *);
 extern	int	pgetc(void);
 extern	char	*cursource(void);
 
-extern	Node	*nodealloc(int);
+extern	Node	*nodealloc(size_t);
 extern	Node	*exptostat(Node *);
 extern	Node	*node1(int, Node *);
 extern	Node	*node2(int, Node *, Node *);