From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:470:142:3::10]:54391) by lists.gnu.org with esmtp (Exim 4.86_2) (envelope-from ) id 1hg3jj-0000MX-H8 for guix-patches@gnu.org; Wed, 26 Jun 2019 04:57:04 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1hg3ji-0005p0-Id for guix-patches@gnu.org; Wed, 26 Jun 2019 04:57:03 -0400 Received: from debbugs.gnu.org ([209.51.188.43]:48917) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1hg3ji-0005ol-EU for guix-patches@gnu.org; Wed, 26 Jun 2019 04:57:02 -0400 Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1hg3ji-0008SU-Bw for guix-patches@gnu.org; Wed, 26 Jun 2019 04:57:02 -0400 Subject: [bug#36131] Removing some of the Common Lisp packages for ECL? Resent-Message-ID: From: Ludovic =?UTF-8?Q?Court=C3=A8s?= References: <87pnnpey6c.fsf@gmail.com> <878stpllma.fsf@gnu.org> <87zhm5k0qs.fsf@gmail.com> Date: Wed, 26 Jun 2019 10:56:23 +0200 In-Reply-To: <87zhm5k0qs.fsf@gmail.com> (Katherine Cox-Buday's message of "Tue, 25 Jun 2019 12:56:11 -0500") Message-ID: <87blykk9mw.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: Katherine Cox-Buday Cc: 36131@debbugs.gnu.org Hello, Cc=E2=80=99ing Andy as the original author of all this. :-) Katherine Cox-Buday skribis: > Ludovic Court=C3=A8s writes: [...] >> I noticed that =E2=80=98ecl-hu.dwim.asdf=E2=80=99 and =E2=80=98ecl-rt=E2= =80=99 fail to build, so I >> couldn=E2=80=99t test all the =E2=80=98ecl-*=E2=80=99 variants. Could y= ou take a look at these >> two packages? > > I focused on the SBCL packages and then retroactively went back and > added all the ECL packages, trying to be a good citizen. In retrospect, > this was not a good idea. Common Lisp code is not guaranteed to work > across runtimes. > > If you're OK with it, I would just go ahead and delete any ECL package > that doesn't immediately work. I can do this myself, but I'm currently > on holiday and won't be able to take a look for another week and a half. Sure, removing packages that don=E2=80=99t build sounds good to me. Andy, = WDYT? >> More generally, does it make sense to have ECL variants for each and >> every package? Or should we trim that down? I=E2=80=99m under the impr= ession >> that ECL is typically used with rather small code bases since it=E2=80= =99s meant >> to be embedded, but then I=E2=80=99m not a Common Lisper. > > I think ECL is used outside embedded contexts, but I haven't found a > reason to use it yet. If I remember correctly, I think one compiles > faster than the other, and the other runs faster, so some people switch > between the two when developing and then deploying. OK, I see. Thanks for explaining! Ludo=E2=80=99.