From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andy Patterson Subject: Re: "M-x slime-connect" cannot connect to sbcl-next swank server Date: Sun, 13 Jan 2019 14:24:41 -0500 Message-ID: <20190113142441.52704d02@mailservices.uwaterloo.ca> References: <87bm4ou4yd.fsf@disroot.org> <87sgxzwf8q.fsf@ambrevar.xyz> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Return-path: Received: from eggs.gnu.org ([209.51.188.92]:36916) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1gilNT-0004Ie-0F for help-guix@gnu.org; Sun, 13 Jan 2019 14:24:59 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1gilNR-0005yQ-QO for help-guix@gnu.org; Sun, 13 Jan 2019 14:24:58 -0500 Received: from mailchk-m03.uwaterloo.ca ([129.97.128.240]:51012) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1gilNR-0005wi-Ir for help-guix@gnu.org; Sun, 13 Jan 2019 14:24:57 -0500 In-Reply-To: <87sgxzwf8q.fsf@ambrevar.xyz> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: help-guix-bounces+gcggh-help-guix=m.gmane.org@gnu.org Sender: "Help-Guix" To: Pierre Neidhardt Cc: help-guix@gnu.org Hey Pierre and Amar, On Fri, 11 Jan 2019 13:28:05 +0100 Pierre Neidhardt wrote: > Thanks for reporting. > > I can reproduce. I think this is due to how we package > SBCL-SLIME-SWANK. > Somehow, the version is missing, and maybe more other important > parameters. If I'm remembering right the version of swank that we package is quite old due to the use of a branch which provided an asd file which was suitable for the asdf build system. So I can definitely see how there could be mismatches. > > With Quicklisp's SWANK and a local build of Next, it works fine here > on Guix' emacs-slime. > > Andy, any idea? When I debugged our next package briefly before, I created an ad-hoc environment with the library version of next, sbcl (for the environment variables), and sbcl-slynk:image. I then ran slynk.image, started up a slynk server on it, and loaded in next. I think I had to rename the slynk pacakges (I did this at the sly mrepl) to stop providing swank aliases so that they wouldn't conflict with the swank which comes with next. After that I started next's main and was able to debug from there. I think that was the path of least resistance to opening a debug session. Hoping that helps. Let me know if you need any more information. -- Andy