Re: c++/2810

Re: c++/2810

[nathan]

The following reply was made to PR c++/2810; it has been noted by GNATS.

From: nathan@gcc.gnu.org
To: gcc-gnats@gcc.gnu.org, nathan@gcc.gnu.org, nobody@gcc.gnu.org,
  schmid@snake.iap.physik.tu-darmstadt.de
Cc:  
Subject: Re: c++/2810
Date: 27 May 2001 17:11:42 -0000

 Synopsis: Problems with enums (gcc 2.95 regression) boost
 
 Responsible-Changed-From-To: unassigned->nathan
 Responsible-Changed-By: nathan
 Responsible-Changed-When: Sun May 27 10:11:42 2001
 Responsible-Changed-Why:
     patch in progress
 
 http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view&pr=2810&database=gcc

Re: c++/2810

[nathan]

The following reply was made to PR c++/2810; it has been noted by GNATS.

From: nathan@gcc.gnu.org
To: gcc-gnats@gcc.gnu.org, nathan@gcc.gnu.org,
  schmid@snake.iap.physik.tu-darmstadt.de
Cc:  
Subject: Re: c++/2810
Date: 28 May 2001 09:22:18 -0000

 Synopsis: Problems with enums (gcc 2.95 regression) boost
 
 State-Changed-From-To: analyzed->closed
 State-Changed-By: nathan
 State-Changed-When: Mon May 28 02:22:18 2001
 State-Changed-Why:
     The code is ill formed, so this is neither a bug nor regression.
     It is a bug in boost, and arguably a defect in the standard.
     I attach a simpler example which illustrates the problem. Boost
     is attempting to do what Foo3 is doing, which attempts to
     instantiate a ctor of any with T being `void ()'. Thus
     the parameter of that constructor tries to be
     `void (const &)()', which is a const qualified function type
     and banned by [8.3.5]/4. You need an additional template
     constructor with parameter (T &).
     
     Now, I think this is an inconsistency in the std, as
     you are allowed to try and create a cv qualified reference
     via a template-type-parm or typedef [8.3.2]/1, so for
     consistency I think the same should happen here.
     
     Incidentally, g++ erroneously accepts Foo1, but rejects Foo2
     so I'm going to file another bug report about that.
     
     Can you inform the boost authors about this?
 
 http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view&pr=2810&database=gcc

Re: c++/2810

[nathan]

The following reply was made to PR c++/2810; it has been noted by GNATS.

From: nathan@gcc.gnu.org
To: gcc-gnats@gcc.gnu.org, nobody@gcc.gnu.org,
  schmid@snake.iap.physik.tu-darmstadt.de
Cc:  
Subject: Re: c++/2810
Date: 27 May 2001 16:47:11 -0000

 Synopsis: Problems with enums (gcc 2.95 regression) boost
 
 State-Changed-From-To: open->analyzed
 State-Changed-By: nathan
 State-Changed-When: Sun May 27 09:47:11 2001
 State-Changed-Why:
     confirmed. I think the code is legal, and hence a regression
 
 http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view&pr=2810&database=gcc