Hi!
Marek has changed grokdeclarator in r263836, so that in this part of code
it is either a funcdecl_p (previously the only allowed one), which
implies inner_declarator is non-NULL and therefore unqualified_id too,
or newly inner_declarator == NULL. In that case, we IMHO shouldn't be
testing for the deduction guides errors and let it be rejected as before
later.
Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?
2018-11-26 Jakub Jelinek <ja...@redhat.com>
PR c++/88187
* decl.c (grokdeclarator): Don't diagnose deduction guide errors
if inner_declarator is NULL.
* g++.dg/other/pr88187.C: New test.
--- gcc/cp/decl.c.jj 2018-11-17 00:16:41.000000000 +0100
+++ gcc/cp/decl.c 2018-11-26 11:18:30.518620651 +0100
@@ -11276,7 +11276,7 @@ grokdeclarator (const cp_declarator *dec
if (!tmpl)
if (tree late_auto = type_uses_auto (late_return_type))
tmpl = CLASS_PLACEHOLDER_TEMPLATE (late_auto);
- if (tmpl)
+ if (tmpl && inner_declarator)
{
if (!dguide_name_p (unqualified_id))
{
--- gcc/testsuite/g++.dg/other/pr88187.C.jj 2018-11-26 11:29:46.096459145
+0100
+++ gcc/testsuite/g++.dg/other/pr88187.C 2018-11-26 11:30:58.110269662
+0100
@@ -0,0 +1,7 @@
+// PR c++/88187
+// { dg-do compile }
+
+template <int> struct A;
+void f (A ()); // { dg-error "variable or field 'f' declared void" "" { target
c++14_down } }
+ // { dg-error "missing template arguments before '\\(' token"
"" { target c++14_down } .-1 }
+ // { dg-error "'auto' parameter not permitted in this context"
"" { target c++17 } .-2 }
Jakub
