From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 31761 invoked by alias); 11 Jul 2003 23:55:55 -0000 Mailing-List: contact gdb-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-owner@sources.redhat.com Received: (qmail 31440 invoked from network); 11 Jul 2003 23:55:44 -0000 Received: from unknown (HELO lakemtao04.cox.net) (68.1.17.241) by sources.redhat.com with SMTP; 11 Jul 2003 23:55:44 -0000 Received: from white ([68.9.191.65]) by lakemtao04.cox.net (InterMail vM.5.01.04.05 201-253-122-122-105-20011231) with ESMTP id <20030711235545.FNSO13930.lakemtao04.cox.net@white>; Fri, 11 Jul 2003 19:55:45 -0400 Received: from bob by white with local (Exim 3.35 #1 (Debian)) id 19b7jr-00030t-00; Fri, 11 Jul 2003 19:55:43 -0400 Date: Fri, 11 Jul 2003 23:55:00 -0000 From: 'Bob Rossi' To: Hassan Aurag Cc: gdb@sources.redhat.com Subject: Re: libtgdb or libgdb Message-ID: <20030711235543.GA11545@white> Mail-Followup-To: Hassan Aurag , gdb@sources.redhat.com References: <69F1436AA435D7118EE1009027B0FF3A01872768@caemsx02.cae.ca> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <69F1436AA435D7118EE1009027B0FF3A01872768@caemsx02.cae.ca> User-Agent: Mutt/1.3.28i X-SW-Source: 2003-07/txt/msg00154.txt.bz2 Well, it started out as a separate project. http://tgdb.sourceforge.net/ Then, since cgdb was the only project that used it, we put it into the cgdb source ( It is completely modular ). http://cgdb.sourceforge.net/ Depending on the demand for such a library, it might end up back at tgdb.sourceforge.net ... or wherever suggestions take it. If there is no interest for such a library, then it may stay with cgdb. I am currently talking to the author of ctrlgdb about integrating our efforts with tgdb. I really believe there is a need for such a library. To be honest, the interface isn't professional grade yet. However, up till now, it does everything I need it to. I am really wondering what most of the gdb developers think. Bob Rossi On Fri, Jul 11, 2003 at 07:27:21AM -0400, Hassan Aurag wrote: > I'd actually love that. It is indeed a real pain to find anything that > presents the developer with a nice interface to debugging symbols. > > Where can it be downloaded? > > > -----Original Message----- > From: Bob Rossi [mailto:bob@brasko.net] > Sent: Thursday, July 10, 2003 10:34 PM > To: gdb@sources.redhat.com > Subject: libtgdb or libgdb > > > Hi, > > As some of you may know, I am working on a front end to gdb called cgdb. > In order to communicate with gdb, I wrote a library called libtgdb ( > Trivial gdb ). This gives a simple interface for the front end to work > with. Thus, completely separating the gdb-specific code from the front > end. > > As of know, libtgdb supports annotate level 2 communication. Starting > next month, I plan to add mi support. It can end up supporting annotate > level 1 if necessary in the future. > > Since I have been subscribed to the gdb list, I have seen many inquiries > about libgdb. Which seems to be no longer supported. I was thinking that > it might be reasonable to have libtgdb be shipped with gdb's sources as > a library that any front end can use to interface with gdb. Of course it > would be as general purpose as possible, and capable of supporting all > of gdb's features when complete. > > One major difference between libtgdb and libgdb is that > 1. libtgdb is a separate library, not linked against gdb's sources. > 2. libtgdb does not have to be compiled to work with a single gdb, > it is backwards compatible and will work with any gdb. > > What does everyone think? Does this make any sense? Is this too ambitious? > > My main goal, is too make front end's able to integrate with gdb easily. > I have spent *far* to much time trying to figure out the gdb specific > stuff. It just doesn't make sense reproducing the code in all of the > front ends. They all end up having there own bugs, which is *very* annoying. > In general, the quality of front ends could be improved, if developers > were not trying to figure out the tricks of getting gdb to do certain > things. > > Thanks, > Bob Rossi > >