Do not display too many bytes when reporting malformed UTF-8 character

perl-5.27.1-utf8n_to_uvchr-Don-t-display-too-many-bytes-in-msg.patch

@@ -0,0 +1,43 @@
+From 05b9033b464ce8dd2c9b33238f9aa14755d7a91a Mon Sep 17 00:00:00 2001
+From: Karl Williamson <khw@cpan.org>
+Date: Sat, 17 Jun 2017 17:56:10 -0600
+Subject: [PATCH] utf8n_to_uvchr(): Don't display too many bytes in msg
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+When raising a message about malformed UTF-8, we shouldn't display bytes
+from the next character, unless those bytes were expected to have been
+part of the current one.  Tests for this will be added in future commits
+in ext/XS-APItest/t/utf8_warn_base.pl
+
+Signed-off-by: Petr Písař <ppisar@redhat.com>
+---
+ utf8.c | 4 ++--
+ 1 file changed, 2 insertions(+), 2 deletions(-)
+
+diff --git a/utf8.c b/utf8.c
+index ee5405f..e55a6f1 100644
+--- a/utf8.c
++++ b/utf8.c
+@@ -1428,7 +1428,7 @@ Perl_utf8n_to_uvchr_error(pTHX_ const U8 *s,
+                         if (pack_warn) {
+                             message = Perl_form(aTHX_ "%s: %s (overflows)",
+                                             malformed_text,
+-                                            _byte_dump_string(s0, send - s0, 0));
++                                            _byte_dump_string(s0, curlen, 0));
+                         }
+                     }
+                 }
+@@ -1554,7 +1554,7 @@ Perl_utf8n_to_uvchr_error(pTHX_ const U8 *s,
+                                 "%s: %s (overlong; instead use %s to represent"
+                                 " U+%0*" UVXf ")",
+                                 malformed_text,
+-                                _byte_dump_string(s0, send - s0, 0),
++                                _byte_dump_string(s0, curlen, 0),
+                                 _byte_dump_string(tmpbuf, e - tmpbuf, 0),
+                                 ((uv < 256) ? 2 : 4), /* Field width of 2 for
+                                                          small code points */
+-- 
+2.9.4
+

perl.spec

@@ -193,6 +193,10 @@ Patch46:        perl-5.26.0-t-op-hash.t-fixup-intermittently-failing-test.patch
 Patch47:        perl-5.27.1-Parse-caret-vars-with-subscripts-the-same-as-normal-.patch
 Patch48:        perl-5.27.1-add-an-additional-test-for-whitespace-tolerance-in-c.patch
 
+# Do not display too many bytes when reporting malformed UTF-8 character,
+# in upstream after 5.27.1
+Patch49:        perl-5.27.1-utf8n_to_uvchr-Don-t-display-too-many-bytes-in-msg.patch
+
 # Link XS modules to libperl.so with EU::CBuilder on Linux, bug #960048
 Patch200:       perl-5.16.3-Link-XS-modules-to-libperl.so-with-EU-CBuilder-on-Li.patch
 
@@ -2769,6 +2773,7 @@ Perl extension for Version Objects
 %patch46 -p1
 %patch47 -p1
 %patch48 -p1
+%patch49 -p1
 %patch200 -p1
 %patch201 -p1
 
@@ -2802,6 +2807,7 @@ perl -x patchlevel.h \
     'Fedora Patch45: Fix File::Glob rt131211.t test random failures' \
     'Fedora Patch46: Fix t/op/hash.t test random failures' \
     'Fedora Patch47: Parse caret variables with subscripts as normal variables inside ${...} escaping (RT#131664)' \
+    'Fedora Patch49: Do not display too many bytes when reporting malformed UTF-8 character' \
     'Fedora Patch200: Link XS modules to libperl.so with EU::CBuilder on Linux' \
     'Fedora Patch201: Link XS modules to libperl.so with EU::MM on Linux' \
     %{nil}
@@ -5091,6 +5097,7 @@ popd
 - Fix t/op/hash.t test random failures
 - Parse caret variables with subscripts as normal variables inside ${...}
   escaping (RT#131664)
+- Do not display too many bytes when reporting malformed UTF-8 character
 
 * Sat Jul 29 2017 Igor Gnatenko <ignatenkobrain@fedoraproject.org> - 4:5.26.0-397
 - Enable separate debuginfo back