From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 21005 invoked by alias); 1 Aug 2009 02:22:10 -0000 Received: (qmail 20996 invoked by uid 22791); 1 Aug 2009 02:22:09 -0000 X-SWARE-Spam-Status: No, hits=3.5 required=5.0 tests=AWL,BAYES_00,BOTNET,SPF_PASS X-Spam-Check-By: sourceware.org Received: from vms173017pub.verizon.net (HELO vms173017pub.verizon.net) (206.46.173.17) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Sat, 01 Aug 2009 02:22:01 +0000 Received: from MacOSX.home ([96.244.82.83]) by vms173017.mailsrvcs.net (Sun Java(tm) System Messaging Server 6.3-7.04 (built Sep 26 2008; 32bit)) with ESMTPA id <0KNO001V3EJYWCX5@vms173017.mailsrvcs.net> for gcc@gcc.gnu.org; Fri, 31 Jul 2009 21:21:35 -0500 (CDT) Message-id: <4A73A6AE.10606@verizon.net> Date: Sat, 01 Aug 2009 02:22:00 -0000 From: Ed Smith-Rowland <3dw4rd@verizon.net> User-Agent: Thunderbird 2.0.0.22 (Macintosh/20090605) MIME-version: 1.0 To: Jason Merrill Cc: James Dennett , Piotr Wyderski , "gcc@gcc.gnu.org >> GCC" Subject: Re: The future of concepts References: <9f8a01cd0907280301i2688abefh3874945bf3c180ac@mail.gmail.com> <18738acb0907280305n7ff7fc19x6f3867a87ff5bf86@mail.gmail.com> <4A6F0F84.8060504@verizon.net> <4A73646B.4010808@redhat.com> In-reply-to: <4A73646B.4010808@redhat.com> Content-type: text/plain; charset=ISO-8859-1; format=flowed Content-transfer-encoding: 7bit X-IsSubscribed: yes Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Archive: List-Post: List-Help: Sender: gcc-owner@gcc.gnu.org X-SW-Source: 2009-08/txt/msg00000.txt.bz2 Jason Merrill wrote: > On 07/28/2009 10:47 AM, Ed Smith-Rowland wrote: >> We need a gcc branch for concepts. > > Probably, if someone is working on them, just as for any ongoing project. > >> That leaves open the question of whether the ConceptGCC branch is the >> one. > > No. Doug felt that much of ConceptGCC needed to be rewritten; the > ConceptGCC branch is just a place to store the existing work for later > reference. That effort did look dead. It seemed to me like a clean slate would be nice. One idea I had was to look at the metaprogramming pseudo concepts that seem to be in the library already and beefing those up. Maybe that plus a tiny amount of core language secret sauce would be the best approach for concepts. > > There is a cxx0x-concepts-branch, but no work has been done on it. > Probably when someone starts serious work on reimplementing concepts > we'll create a new branch. I don't currently have any plans to work > on concepts, as there are plenty of other things to work on that will > be in C++0x. I agree. There's constexpr, lambdas, nullptr, etc. Then there's library stuff like regex mostly. I was thinking we'd have a lot of green check marks for most of C++-0x when it finally comes out. I think other compilers will too. There won't be hardly any latency at all between ratification and good and broad support for C++-0x. That's nothing to sneeze at. > > Jason >