From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:470:142:3::10]:33285) by lists.gnu.org with esmtp (Exim 4.86_2) (envelope-from ) id 1hhvlv-0005ri-Ib for guix-patches@gnu.org; Mon, 01 Jul 2019 08:51:04 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1hhvlu-00016G-Ku for guix-patches@gnu.org; Mon, 01 Jul 2019 08:51:03 -0400 Received: from debbugs.gnu.org ([209.51.188.43]:34205) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1hhvlu-00015y-Fj for guix-patches@gnu.org; Mon, 01 Jul 2019 08:51:02 -0400 Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1hhvlu-0003NC-CV for guix-patches@gnu.org; Mon, 01 Jul 2019 08:51:02 -0400 Subject: [bug#36404] [PATCH 0/6] Add 'guix deploy'. Resent-Message-ID: From: Ludovic =?UTF-8?Q?Court=C3=A8s?= References: <87o92ianbj.fsf@sdf.lonestar.org> <87o92glap5.fsf@dustycloud.org> <87h888dkmh.fsf@sdf.lonestar.org> Date: Mon, 01 Jul 2019 14:50:16 +0200 In-Reply-To: <87h888dkmh.fsf@sdf.lonestar.org> (Jakob L. Kreuze's message of "Sat, 29 Jun 2019 19:42:14 -0400") Message-ID: <87k1d1or5j.fsf@gnu.org> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-patches-bounces+kyle=kyleam.com@gnu.org Sender: "Guix-patches" To: "Jakob L. Kreuze" Cc: 36404@debbugs.gnu.org zerodaysfordays@sdf.lonestar.org (Jakob L. Kreuze) skribis: > Christopher Lemmer Webber writes: > >> In the future I think it would be good to make this extensible as >> well. Dispatching on a symbol means that Guix must itself provide a >> fixed set of possible environment types. If we made this an extensible >> structure, akin to services or something, we could allow for more >> flexibility in the future. Thoughts for the future, but not a blocker >> on this patch. > > +1. Initially, I thought the service types _were_ symbols, but I see now > that they're actually procedures. Thanks for pointing that out. I'll see > about implementing environment types similarly in my revised patch set, > since I think that's a change that we'd want to make before any other > environment types come into existence. It=E2=80=99s a pattern similar to that of for packages. I t= hink it should provide the flexibility and extensibility we need. Ludo=E2=80=99.