From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 1075 invoked by alias); 12 Aug 2003 15:29:45 -0000 Mailing-List: contact guile-gtk-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: guile-gtk-owner@sources.redhat.com Received: (qmail 1068 invoked from network); 12 Aug 2003 15:29:45 -0000 Received: from unknown (HELO fnord.ir.bbn.com) (192.1.100.210) by sources.redhat.com with SMTP; 12 Aug 2003 15:29:45 -0000 Received: by fnord.ir.bbn.com (Postfix, from userid 10853) id 11E9A1F75; Tue, 12 Aug 2003 11:29:45 -0400 (EDT) To: Andreas Rottmann Cc: guile-gtk@sources.redhat.com, rlb@defaultvalue.org Subject: Re: guile-gobject problem [solved], g-wrap References: <87llu0vsx8.fsf@alice.rotty.yi.org> <87vft353sc.fsf@alice.rotty.yi.org> From: Greg Troxel Date: Tue, 12 Aug 2003 15:29:00 -0000 In-Reply-To: <87vft353sc.fsf@alice.rotty.yi.org> Message-ID: User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.3 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-SW-Source: 2003-q3/txt/msg00045.txt.bz2 [I fixed the spelling of Rob's username - it was rbl in your mail.] I've found & fixed the bug. I'll commit to CVS (just have been given access by Marius!), after posting a patch here for review. First, however, I'd like to discuss (once again ;-)), the forking of g-wrap. I could check in a g-wrap into the guile-gtk CVS repository that provides: * Argument hiding (needed for GError exceptions). * An argument call-ccg (needed for GError, since it is passed by double-refernce (pointer to pointer)). * A changed (generating less code) mechanism to wrap enums (this only has gains for non-GObject enums, though). * The patched gw-standard-spec.scm as provided by guile-gobject today. * An exported gw:wrapset-get-wrapsets-depended-on (the fix mentioned already uses this, although it could be changed not to use it easily). Not that I know what's really up, but note that gnucash is said to be short of developer cycles. So creating a g-wrap _project_ on savannah, rather than putting it under guile-gtk, and trying to make sure (a bit) it still works for other uses, and perhaps a release, might be a real contribution. -- Greg Troxel