From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 4575 invoked by alias); 5 Aug 2008 15:01:32 -0000 Received: (qmail 4543 invoked by uid 22791); 5 Aug 2008 15:01:30 -0000 X-Spam-Check-By: sourceware.org Received: from pool-72-74-94-104.bstnma.fios.verizon.net (HELO ednor.cgf.cx) (72.74.94.104) by sourceware.org (qpsmtpd/0.31) with ESMTP; Tue, 05 Aug 2008 15:00:55 +0000 Received: by ednor.cgf.cx (Postfix, from userid 201) id C49BA677721; Tue, 5 Aug 2008 11:00:53 -0400 (EDT) Date: Tue, 05 Aug 2008 15:01:00 -0000 From: Christopher Faylor To: Keith Seitz , insight Subject: Re: [PATCH/libgui] Fix cygwin build woes Message-ID: <20080805150053.GD10807@ednor.casa.cgf.cx> Mail-Followup-To: Keith Seitz , insight References: <4894EF36.4070908@redhat.com> <20080804233502.GA13076@ednor.casa.cgf.cx> <4897D31D.5010506@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4897D31D.5010506@redhat.com> User-Agent: Mutt/1.5.16 (2007-06-09) Mailing-List: contact insight-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: insight-owner@sourceware.org X-SW-Source: 2008-q3/txt/msg00027.txt.bz2 On Mon, Aug 04, 2008 at 09:12:13PM -0700, Keith Seitz wrote: > Christopher Faylor wrote: >>So where does cygwin stand now, Keith? Is it time to decouple insight >>from the gdb release, make it a standalone release, and make it rely on >>X? Is there a version of tk out there which works with on native >>windows? > >I don't really know where cygwin stands. Right now, the versions of >Tcl, Tk, Itcl, and Itk that are in sourceware are really hybrid >monsters. Yep. Understood. I'm sort of tired of fielding questions about them actually. >There is no active cygwin maintainer for Tcl. That means native >binaries only (which can be downloaded from ActiveState). Or someone >could build his own binaries, based on the work that's been done on >sourceware. > >Itcl and Itk were pretty trivial to do, but I'm sure Tcl and Tk would >be a bit tougher. But I don't see why it could not be done. > >I have no opinion on moving Insight to X on cygwin. Insight's only >requirement is Tcl and Tk, and it doesn't care how it gets it. :-) I asked in the cygwin list and most people claim to be reluctantly ok with an X requirement. I was counselled not to make raw gdb rely on X however (well duh). Apparently a lot of people are using it. I was surprised to hear that. >I'm only attempting to decouple insight from Tcl and friends. I've >tried ripping insight out of gdb, but it is a major, major pain: GDB >abandoned libgdb way too early to make this possible. So insight will >stay where it is. > >My play, actually, is to abandon insight altogether, but I don't know >when. In light of that, this has been an effort to get rid of some of >insight's worst baggage. Licensing issues have made it possible so >that only a Red Hat employee can easily work on insight (or at least >anything non-trivial). So insight sits stagnant. I think it is simply >time to move on. Since Red Hat was willing to release insight to the FSF would they also be willing to just relinquish ownership entirely? Would that make working on it easier? >I am faced with three possible decisions: > >1) Keep up with insight, fixing bugs, rewriting what I can here and >there, i.e., maintain the status quo. > >2) Redesign/rewrite insight (probably move away from Tcl, too) > >3) Abandon insight and devote time and energy to another project. Cygwin needs an X maintainer. :-) >I know what I would like to do, but it's a huge undertaking, and I am >not entirely sure I am either up to the challenge or have the time to >wait for sufficient community momentum. Or worse, I commit a boatload >of time and there is never any momentum -- just this situation insight >faces today. I think if you committed the time people would notice and might even pitch in. Aren't there some companies still using insight? Were you thinking about possibly using gdb/mi? cgf