2 files changed, 112 insertions, 0 deletions
diff --git a/sys-apps/nawk/files/nawk-20230909-big-endian.patch b/sys-apps/nawk/files/nawk-20230909-big-endian.patch
new file mode 100644
index 000000000000..b39e7b4bee62
--- /dev/null
+++ b/sys-apps/nawk/files/nawk-20230909-big-endian.patch
@@ -0,0 +1,51 @@
+https://github.com/onetrueawk/awk/pull/196
+
+From 75c017ca09a70e14a862f7285cf91bb87ec1f443 Mon Sep 17 00:00:00 2001
+From: "Todd C. Miller" <Todd.Miller@sudo.ws>
+Date: Mon, 18 Sep 2023 17:34:34 -0600
+Subject: [PATCH 1/2] Fix a bad cast to char * that causes incorrect results on
+ big endian.
+
+Now that awk stores chars as int we need to cast the Node * to int *.
+--- a/b.c
++++ b/b.c
+@@ -527,7 +527,7 @@ int first(Node *p)	/* collects initially active leaves of p into setvec */
+ 			setvec[lp] = 1;
+ 			setcnt++;
+ 		}
+-		if (type(p) == CCL && (*(char *) right(p)) == '\0')
++		if (type(p) == CCL && (*(int *) right(p)) == 0)
+ 			return(0);		/* empty CCL */
+ 		return(1);
+ 	case PLUS:
+
+From 0048c96e94c732c6fb2ebe50eeb3450c1672fe0f Mon Sep 17 00:00:00 2001
+From: "Todd C. Miller" <Todd.Miller@sudo.ws>
+Date: Thu, 21 Sep 2023 11:20:16 -0600
+Subject: [PATCH 2/2] Use cclenter("") to construct an empty CCL, not
+ tostring("").
+
+We need to store a UTF-32 string, not a UTF-8 string, for consistency
+with the other CCL code.  Fixes an out-of-bounds read of an empty
+CCL.
+--- a/b.c
++++ b/b.c
+@@ -945,7 +945,7 @@ Node *primary(void)
+ 		rtok = relex();
+ 		if (rtok == ')') {	/* special pleading for () */
+ 			rtok = relex();
+-			return unary(op2(CCL, NIL, (Node *) tostring("")));
++			return unary(op2(CCL, NIL, (Node *) cclenter("")));
+ 		}
+ 		np = regexp();
+ 		if (rtok == ')') {
+@@ -968,7 +968,7 @@ Node *concat(Node *np)
+ 		return (concat(op2(CAT, np, primary())));
+ 	case EMPTYRE:
+ 		rtok = relex();
+-		return (concat(op2(CAT, op2(CCL, NIL, (Node *) tostring("")),
++		return (concat(op2(CAT, op2(CCL, NIL, (Node *) cclenter("")),
+ 				primary())));
+ 	}
+ 	return (np);
+
diff --git a/sys-apps/nawk/nawk-20230909-r1.ebuild b/sys-apps/nawk/nawk-20230909-r1.ebuild
new file mode 100644
index 000000000000..aa037df95f05
--- /dev/null
+++ b/sys-apps/nawk/nawk-20230909-r1.ebuild
@@ -0,0 +1,61 @@
+# Copyright 1999-2023 Gentoo Authors
+# Distributed under the terms of the GNU General Public License v2
+
+EAPI=8
+
+inherit toolchain-funcs
+
+DESCRIPTION="Brian Kernighan's pattern scanning and processing language"
+HOMEPAGE="https://www.cs.princeton.edu/~bwk/btl.mirror/"
+SRC_URI="https://github.com/onetrueawk/awk/archive/${PV}.tar.gz -> ${P}.tar.gz"
+S="${WORKDIR}/awk-${PV}"
+
+LICENSE="MIT"
+SLOT="0"
+KEYWORDS="~alpha ~amd64 ~arm ~arm64 ~hppa ~ia64 ~loong ~m68k ~ppc ~ppc64 ~riscv ~s390 ~sparc ~x86 ~amd64-linux"
+
+BDEPEND="
+	app-alternatives/yacc
+"
+
+DOCS=( README.md FIXES )
+
+PATCHES=(
+	"${FILESDIR}"/${PN}-20230909-big-endian.patch
+)
+
+src_compile() {
+	emake \
+		CC="$(tc-getCC)" \
+		HOSTCC="$(tc-getBUILD_CC)" \
+		CFLAGS="${CFLAGS}" \
+		CPPFLAGS="${CPPFLAGS} -DHAS_ISBLANK" \
+		ALLOC="${LDFLAGS}" \
+		YACC=$(type -p yacc) \
+		YFLAGS="-d -b awkgram"
+}
+
+src_install() {
+	newbin a.out "${PN}"
+	sed \
+		-e 's/awk/nawk/g' \
+		-e 's/AWK/NAWK/g' \
+		-e 's/Awk/Nawk/g' \
+		awk.1 > "${PN}".1 || die "manpage patch failed"
+	doman "${PN}.1"
+	einstalldocs
+}
+
+pkg_postinst() {
+	if has_version app-admin/eselect && has_version app-eselect/eselect-awk
+	then
+		eselect awk update ifunset
+	fi
+}
+
+pkg_postrm() {
+	if has_version app-admin/eselect && has_version app-eselect/eselect-awk
+	then
+		eselect awk update ifunset
+	fi
+}