diff mbox

[C++] PR 61491 (aka DR 1206)

Message ID 559ECD49.7070701@oracle.com
State New
Headers show

Commit Message

Paolo Carlini July 9, 2015, 7:36 p.m. UTC 
Hi,

On 07/09/2015 06:33 PM, Jason Merrill wrote:
> On 07/09/2015 12:09 PM, Paolo Carlini wrote:
>> the DR got resolved in time for C++11 and Jonathan noticed that we
>> should remove the pedwarn, not a big deal. Tested x86_64-linux.
>
> How about adding the testcase from the DR as well?  OK with that change.
Thanks. That revealed three interesting details. First, I suppose we 
want to use the testcase in the amended form appearing in C++11, thus 
with char as underlying type for the last two lines, instead of int 
(otherwise we reject both for the wrong reason, likewise all the other 
compilers I have at hand). That said, the following two lines of the 
testcase trigger another pedwarn, in the parser, about enumeration 
templates:

template<class T> enum A<T>::E : T { eT };
template<class T> enum class A<T>::S : T { sT };

Finally, we do *not* reject, as we should, the line:

template<> enum A<char>::E : char { echar };

Then, overall, is it Ok to simply suppress the pedwarn in C++11, and 
xfail for now the error? Should I open a new, separate bug report about 
the latter? (note that the issue, failing to reject an explicit 
specialization after instantiation, doesn't sound new to me and seems 
more general than enum-related issues...)

Thanks!
Paolo.

/////////////////////////

Comments

Jason Merrill July 9, 2015, 8:53 p.m. UTC | #1 
On 07/09/2015 03:36 PM, Paolo Carlini wrote:
> Finally, we do *not* reject, as we should, the line:
>
> template<> enum A<char>::E : char { echar };
>
> Then, overall, is it Ok to simply suppress the pedwarn in C++11, and
> xfail for now the error? Should I open a new, separate bug report about
> the latter? (note that the issue, failing to reject an explicit
> specialization after instantiation, doesn't sound new to me and seems
> more general than enum-related issues...)

I don't think it's a more general issue.  I note that the specialization 
seems to append to the enumerator list somehow:

int i = A<char>::eT + A<char>::echar; // accepted?!?

I think we should fix this before we remove the pedwarn.

Jason
Paolo Carlini July 9, 2015, 9:40 p.m. UTC | #2 
Hi,

On 07/09/2015 10:53 PM, Jason Merrill wrote:
> On 07/09/2015 03:36 PM, Paolo Carlini wrote:
>> Finally, we do *not* reject, as we should, the line:
>>
>> template<> enum A<char>::E : char { echar };
>>
>> Then, overall, is it Ok to simply suppress the pedwarn in C++11, and
>> xfail for now the error? Should I open a new, separate bug report about
>> the latter? (note that the issue, failing to reject an explicit
>> specialization after instantiation, doesn't sound new to me and seems
>> more general than enum-related issues...)
>
> I don't think it's a more general issue.  I note that the 
> specialization seems to append to the enumerator list somehow:
>
> int i = A<char>::eT + A<char>::echar; // accepted?!?
Annoying :( Indeed, I can see that the append already happened by the 
time maybe_process_partial_specialization is called the fourth time. 
Let's see if I can make some progress on that.

Thanks,
Paolo.
diff mbox

Patch

Index: cp/parser.c
===================================================================
--- cp/parser.c	(revision 225614)
+++ cp/parser.c	(working copy)
@@ -15827,8 +15827,9 @@  cp_parser_enum_specifier (cp_parser* parser)
 	  type = TREE_TYPE (name);
 	  if (TREE_CODE (type) == TYPENAME_TYPE)
 	    {
-	      /* Are template enums allowed in ISO? */
-	      if (template_parm_scope_p ())
+	      /* Template enums are allowed in C++11.  */
+	      if (template_parm_scope_p ()
+		  && cxx_dialect < cxx11)
 		pedwarn (type_start_token->location, OPT_Wpedantic,
 			 "%qD is an enumeration template", name);
 	      /* ignore a typename reference, for it will be solved by name
Index: cp/pt.c
===================================================================
--- cp/pt.c	(revision 225614)
+++ cp/pt.c	(working copy)
@@ -970,11 +970,10 @@  maybe_process_partial_specialization (tree type)
     }
   else if (processing_specialization)
     {
-       /* Someday C++0x may allow for enum template specialization.  */
+      /* Under DR 1206 enum template specializations are allowed.  */
       if (cxx_dialect > cxx98 && TREE_CODE (type) == ENUMERAL_TYPE
 	  && CLASS_TYPE_P (context) && CLASSTYPE_USE_TEMPLATE (context))
-	pedwarn (input_location, OPT_Wpedantic, "template specialization "
-		 "of %qD not allowed by ISO C++", type);
+	;
       else
 	{
 	  error ("explicit specialization of non-template %qT", type);
Index: testsuite/g++.dg/cpp0x/enum30.C
===================================================================
--- testsuite/g++.dg/cpp0x/enum30.C	(revision 0)
+++ testsuite/g++.dg/cpp0x/enum30.C	(working copy)
@@ -0,0 +1,20 @@ 
+// PR c++/61491, DR 1206
+// { dg-do compile { target c++11 } }
+
+template <class D> struct Base 
+{ 
+  enum class E : unsigned; 
+  E e; 
+  Base(E e) : e(e) {} 
+}; 
+
+struct X; 
+
+template<> enum class Base<X>::E : unsigned { a, b }; 
+
+struct X : Base<X> 
+{ 
+  X() : Base<X>(E::b) {} 
+};
+
+X x; 
Index: testsuite/g++.dg/cpp0x/enum31.C
===================================================================
--- testsuite/g++.dg/cpp0x/enum31.C	(revision 0)
+++ testsuite/g++.dg/cpp0x/enum31.C	(working copy)
@@ -0,0 +1,14 @@ 
+// DR 1206, PR c++/61491
+// { dg-do compile { target c++11 } }
+
+template<class T> struct A {
+  enum E : T;
+  enum class S : T;
+};
+
+template<> enum A<int>::E : int { eint };
+template<> enum class A<int>::S : int { sint };
+template<class T> enum A<T>::E : T { eT };
+template<class T> enum class A<T>::S : T { sT };
+template<> enum A<char>::E : char { echar };  // { dg-error "after instantiation" "" { xfail *-*-* } }
+template<> enum class A<char>::S : char { schar };