From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-prs-return-31364-listarch-gcc-prs=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 8532 invoked by alias); 17 Mar 2003 00:06:02 -0000
Mailing-List: contact gcc-prs-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Archive: <http://gcc.gnu.org/ml/gcc-prs/>
List-Post: <mailto:gcc-prs@gcc.gnu.org>
List-Help: <mailto:gcc-prs-help@gcc.gnu.org>
Sender: gcc-prs-owner@gcc.gnu.org
Received: (qmail 8462 invoked by uid 71); 17 Mar 2003 00:06:01 -0000
Date: Mon, 17 Mar 2003 00:06:00 -0000
Message-ID: <20030317000601.8457.qmail@sources.redhat.com>
To: nobody@gcc.gnu.org
Cc: gcc-prs@gcc.gnu.org,
From: Martin Sebor <sebor@roguewave.com>
Subject: Re: c++/10111: Parse error for default template parameter
Reply-To: Martin Sebor <sebor@roguewave.com>
X-SW-Source: 2003-03/txt/msg01094.txt.bz2
List-Id: <gcc-prs.sourceware.org>

The following reply was made to PR c++/10111; it has been noted by GNATS.

From: Martin Sebor <sebor@roguewave.com>
To: gcc-gnats@gcc.gnu.org
Cc:  
Subject: Re: c++/10111: Parse error for default template parameter
Date: Sun, 16 Mar 2003 17:03:49 -0700

 Giovanni Bajo wrote:
 > http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&p
 > r=10111
 > 
 > Not a real bug. All boils down to a definition like this:
 > 
 > void Foo(int a = A<int,int>::something) {}
 > 
 > There is an open defect report about this
 > (http://www.comeaucomputing.com/iso/cwg_active.html#325), because (shortly)
 > ISO C++ does not mandate that the compiler must understand that the comma is
 > separating template parameters, and not parameters of the function you are
 > declaring.
 
 Not necessarily. The issue is in Open state, so it's still a bug
 at least until the next revision of C++ (assuming issue 325 is
 accepted as a defect) :)
 
 > Simple workaround is putting a couple of paranthesis around the
 > default argument, the code will compile correctly:
 > 
 > void Foo(int a = (A<int,int>::something)) {}
 
 But this still doesn't seem to (at least not with 3.2.2).
 Incidentally, this test case is isolated from std::bitset.
 
 template <class T, class U, class V>
 struct A
 {
      typedef int I;
      static const I i = 0;
 };
 
 template <int I>
 struct B
 {
      template <class T, class U, class V>
      B (const A<T, U, V>&,
         typename A<T, U, V>::I = (typename A<T, U, V>::I ()),
         typename A<T, U, V>::I = (A<T, U, V>::i)) { }
 };
 
 int main ()
 {
      A<int, int, int> a;
 
      B<0> b (a);
 }
 
 Regards
 Martin