From 927b5dc1dcb260cdd0df54798780788da60c369f Mon Sep 17 00:00:00 2001
From: Karl Williamson <khw@cpan.org>
Date: Fri, 28 Oct 2022 08:48:37 -0600
Subject: [PATCH] utf8::upgrade: Don't coerce undef arg

This fixes GH #20419
---
 lib/utf8.pm       |  4 +++-
 lib/utf8.t        |  3 +++
 pod/perldelta.pod |  6 ++++++
 universal.c       | 16 +++++++++++++---
 4 files changed, 25 insertions(+), 4 deletions(-)

diff --git a/lib/utf8.pm b/lib/utf8.pm
index 823193b8c13a..c1f1680336db 100644
--- a/lib/utf8.pm
+++ b/lib/utf8.pm
@@ -5,7 +5,7 @@ use warnings;
 
 our $hint_bits = 0x00800000;
 
-our $VERSION = '1.24';
+our $VERSION = '1.25';
 our $AUTOLOAD;
 
 sub import {
@@ -114,6 +114,8 @@ sequence in the native encoding (Latin-1 or EBCDIC) to UTF-8. The
 logical character sequence itself is unchanged.  If I<$string> is already
 upgraded, then this is a no-op. Returns the
 number of octets necessary to represent the string as UTF-8.
+Since Perl v5.38, if C<$string> is C<undef> no action is taken; prior to that,
+it would be converted to be defined and zero-length.
 
 If your code needs to be compatible with versions of perl without
 C<use feature 'unicode_strings';>, you can force Unicode semantics on
diff --git a/lib/utf8.t b/lib/utf8.t
index d35110baee07..fa69fcbb8862 100644
--- a/lib/utf8.t
+++ b/lib/utf8.t
@@ -669,6 +669,9 @@ for(__PACKAGE__) {
 	eval { utf8::upgrade($_) };
 	is $@, "", 'no error with utf8::upgrade on read-only COW';
 }
+
+is(utf8::upgrade(undef), undef, "Returns undef for undef input"); # GH #20419
+
 # This one croaks, but not because the scalar is read-only
 eval "package \x{100};\n" . <<'END'
     for(__PACKAGE__) {
diff --git a/pod/perldelta.pod b/pod/perldelta.pod
index ce999fbdb428..f3fcd855257c 100644
--- a/pod/perldelta.pod
+++ b/pod/perldelta.pod
@@ -45,6 +45,12 @@ XXX For a release on a stable branch, this section aspires to be:
 
 [ List each incompatible change as a =head2 entry ]
 
+=head2 L<C<utf8::upgrade()>|utf8/Utility functions>
+
+Starting in this release, if the input string is C<undef>, it remains
+C<undef>.  Previously it would be changed into a defined, zero-length
+string.
+
 =head1 Deprecations
 
 XXX Any deprecated features, syntax, modules etc. should be listed here.
diff --git a/universal.c b/universal.c
index 119117e818a4..20a36fae8517 100644
--- a/universal.c
+++ b/universal.c
@@ -593,11 +593,21 @@ XS(XS_utf8_upgrade)
         croak_xs_usage(cv, "sv");
     else {
         SV * const sv = ST(0);
-        STRLEN	RETVAL;
+        STRLEN	RETVAL = 0;
         dXSTARG;
 
-        RETVAL = sv_utf8_upgrade(sv);
-        XSprePUSH; PUSHi((IV)RETVAL);
+        XSprePUSH;
+        if (UNLIKELY(! sv)) {
+            XSRETURN_UNDEF;
+        }
+
+        SvGETMAGIC(sv);
+        if (UNLIKELY(! SvOK(sv))) {
+            XSRETURN_UNDEF;
+        }
+
+        RETVAL = sv_utf8_upgrade_nomg(sv);
+        PUSHi( (IV) RETVAL);
     }
     XSRETURN(1);
 }