toke.c: Don't accept illegal code points

This now croaks if the input is an illegal code point.  Before, it
likely would eventually croak if that code point was actually used in
some manner.

1 files changed, 12 insertions, 2 deletions

diff --git a/toke.c b/toke.c
index abdd36d67f..4f90f17972 100644
--- a/toke.c
+++ b/toke.c
@@ -3747,13 +3747,23 @@ S_scan_const(pTHX_ char *start)
 		    }
 		    else {  /* Not a pattern: convert the hex to string */
                         I32 flags = PERL_SCAN_ALLOW_UNDERSCORES
-				| PERL_SCAN_SILENT_ILLDIGIT
-				| PERL_SCAN_DISALLOW_PREFIX;
+				  | PERL_SCAN_SILENT_ILLDIGIT
+				  | PERL_SCAN_SILENT_OVERFLOW
+				  | PERL_SCAN_DISALLOW_PREFIX;
                         STRLEN len = e - s;
+
                         uv = grok_hex(s, &len, &flags, NULL);
                         if (len == 0 || (len != (STRLEN)(e - s)))
                             goto bad_NU;
 
+                        if (    uv > MAX_LEGAL_CP
+                            || (flags & PERL_SCAN_GREATER_THAN_UV_MAX))
+                        {
+                            yyerror(form_cp_too_large_msg(16, s, len, 0));
+                            uv = 0; /* drop through to ensure range ends are
+                                       set */
+                        }
+
                          /* For non-tr///, if the destination is not in utf8,
                           * unconditionally recode it to be so.  This is
                           * because \N{} implies Unicode semantics, and scalars