From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 4885 invoked by alias); 20 Apr 2006 13:20:12 -0000 Received: (qmail 4875 invoked by uid 22791); 20 Apr 2006 13:20:11 -0000 X-Spam-Check-By: sourceware.org Received: from nevyn.them.org (HELO nevyn.them.org) (66.93.172.17) by sourceware.org (qpsmtpd/0.31.1) with ESMTP; Thu, 20 Apr 2006 13:20:07 +0000 Received: from drow by nevyn.them.org with local (Exim 4.54) id 1FWZ4m-0003A0-9f; Thu, 20 Apr 2006 09:20:04 -0400 Date: Thu, 20 Apr 2006 13:20:00 -0000 From: Daniel Jacobowitz To: Thomas Kuehne Cc: gdb-patches@sources.redhat.com Subject: Re: D Symbol Demangling Message-ID: <20060420132004.GB11710@nevyn.them.org> Mail-Followup-To: Thomas Kuehne , gdb-patches@sources.redhat.com References: <44462A40.1040004@kuehne.cn> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <44462A40.1040004@kuehne.cn> User-Agent: Mutt/1.5.8i X-IsSubscribed: yes Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org X-SW-Source: 2006-04/txt/msg00268.txt.bz2 On Wed, Apr 19, 2006 at 02:17:04PM +0200, Thomas Kuehne wrote: > Attached is a patch against GDB-6.4 that enables demangling of symbols > generated by DMD[1] and GDC[2]. > > The infrastructure part (c-lang.c, defs.h, d-lang.h, dwarf2read.c, > language.c, Makefile.in, symfile, syntab.c, dwarf2.h) is based on John > Demme's work[3][4]. > > The pluggable demangler (gdb/demangle_d/*) was written completely form > scratch and isn't based on John's work and supports templates and - to a > certain extend - nested functions and types. Unfortunately, there are several major problems with accepting this code. 1. Neither you nor John has an FSF copyright assignment in place. If you're interested in one, let me know and I can send you the forms. 2. The demangler would need to be contributed to the FSF and licensed under the GPL before we could include it with GDB. That doesn't prevent another copy of it from being used under a different license elsewhere; but you wouldn't be able to import fixes from one to the other, in either direction. 3. The code would need to match the GNU Coding Standards. 4. A patch would need to be generated against HEAD, not against an old release. And, no offense, but the DD_() thing is horribly ugly and doesn't seem to serve any purpose here. Is that for reusing the demangler elsewhere? -- Daniel Jacobowitz CodeSourcery