From 0327f923ff75d32bb6180a06cafd8b02a6a1ff16 Mon Sep 17 00:00:00 2001
From: Reini Urban <rurban@cpan.org>
Date: Thu, 22 Jun 2017 16:51:50 +0200
Subject: [PATCH] Abort on Malformed UTF-8 character errors

utf8n_to_uvchr_error() only warns on some Malformed UTF-8 characters,
but scan_const needs to error here. Do it with yyerror() which
accumulates all parser errors until it "has too many errors".

Fixes 2 errors in #293, esp.
id:000162,sig:06,src:026278+031045,op:splice,rep:32 and
id:000001,sig:06,src:024259,op:arith8,pos:5,val:+27
which segfaulted in the error handler for
"panic: constant overflowed allocated space"
---
 .git-rr-cache    |  2 +-
 pod/perldiag.pod |  6 ++++++
 toke.c           |  6 +++---
 utf8.c           | 12 ++++++------
 4 files changed, 16 insertions(+), 10 deletions(-)

diff --git a/.git-rr-cache b/.git-rr-cache
index 063dbec90ec..cdfe37572a4 160000
--- a/.git-rr-cache
+++ b/.git-rr-cache
@@ -1 +1 @@
-Subproject commit 063dbec90ec42e7c78bfc68b75df5ad1f5090fd4
+Subproject commit cdfe37572a445534ad74c967b0cdaa4adeb5fc58
diff --git a/pod/perldiag.pod b/pod/perldiag.pod
index 0b3e3614f14..b80c73941ad 100644
--- a/pod/perldiag.pod
+++ b/pod/perldiag.pod
@@ -3539,6 +3539,12 @@ Perhaps the function's author was trying to write a subroutine signature
 but didn't enable that feature first (C<use feature 'signatures'>),
 so the signature was instead interpreted as a bad prototype.
 
+=item Malformed UTF-8 character
+
+(F) Perl detected one or more fatal UTF-8 errors while parsing a
+constant UTF-8 string, which are detailed in the first warnings utf8
+message. See L</"Malformed UTF-8 character%s">
+
 =item Malformed UTF-8 character%s
 
 (S utf8)(F) Perl detected a string that should be UTF-8, but didn't
diff --git a/toke.c b/toke.c
index e2b7dd142e3..57e50db54a2 100644
--- a/toke.c
+++ b/toke.c
@@ -4124,9 +4124,9 @@ S_scan_const(pTHX_ char *start)
         else if (this_utf8 && has_utf8) {   /* Both UTF-8, can just copy */
 	    const STRLEN len = UTF8SKIP(s);
 
-            /* We expect the source to have already been checked for
-             * malformedness */
-            assert(isUTF8_CHAR((U8 *) s, (U8 *) send));
+            /* utf8n_to_uvchr_error might have only warned: promote to error */
+            if (!isUTF8_CHAR((U8 *) s, (U8 *) send))
+                yyerror("Malformed UTF-8 character");
 
             Copy(s, d, len, U8);
             d += len;
diff --git a/utf8.c b/utf8.c
index 464d6cd6a4f..3bb7baade75 100644
--- a/utf8.c
+++ b/utf8.c
@@ -810,14 +810,14 @@ Perl__byte_dump_string(pTHX_ const U8 * s, const STRLEN len, const bool format)
 PERL_STATIC_INLINE char *
 S_unexpected_non_continuation_text(pTHX_ const U8 * const s,
 
-                                         /* How many bytes to print */
-                                         STRLEN print_len,
+                                   /* How many bytes to print */
+                                   STRLEN print_len,
 
-                                         /* Which one is the non-continuation */
-                                         const STRLEN non_cont_byte_pos,
+                                   /* Which one is the non-continuation */
+                                   const STRLEN non_cont_byte_pos,
 
-                                         /* How many bytes should there be? */
-                                         const STRLEN expect_len)
+                                   /* How many bytes should there be? */
+                                   const STRLEN expect_len)
 {
     /* Return the malformation warning text for an unexpected continuation
      * byte. */