From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:470:142:3::10]:57937) by lists.gnu.org with esmtp (Exim 4.86_2) (envelope-from ) id 1hnV6F-000643-CD for guix-patches@gnu.org; Tue, 16 Jul 2019 17:35:04 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1hnV6E-0007BJ-BQ for guix-patches@gnu.org; Tue, 16 Jul 2019 17:35:03 -0400 Received: from debbugs.gnu.org ([209.51.188.43]:42451) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1hnV6E-0007B8-6L for guix-patches@gnu.org; Tue, 16 Jul 2019 17:35:02 -0400 Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1hnV6E-0000Xk-2H for guix-patches@gnu.org; Tue, 16 Jul 2019 17:35:02 -0400 Subject: [bug#35790] [PATCH 4/4] lint: Separate checkers by dependence on the internet. Resent-Message-ID: From: Ludovic =?UTF-8?Q?Court=C3=A8s?= References: <87k1cjr6vv.fsf@gnu.org> <20190715194558.13804-1-mail@cbaines.net> <20190715194558.13804-4-mail@cbaines.net> <87o91vhx2m.fsf@gnu.org> <87o91vgco8.fsf@cbaines.net> Date: Tue, 16 Jul 2019 23:34:31 +0200 In-Reply-To: <87o91vgco8.fsf@cbaines.net> (Christopher Baines's message of "Mon, 15 Jul 2019 23:23:19 +0100") Message-ID: <87ftn5ek9k.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: Christopher Baines Cc: 35790-done@debbugs.gnu.org Hi! Christopher Baines skribis: > Great, I've pushed these to master now. Yay! \o/ > I hope to start looking at actually trying to load in the lint warnings > soon. This might involve extending the inferior API if that's helpful. I > also want to attempt to store translations for the lint warnings in one > way or another, as that'll begin to address the lack of localisation in > the Guix Data Service. Regarding inferiors, you could always build up an =E2=80=9Cinferior lint warning=E2=80=9D API, similar to what=E2=80=99s done for = , and likewise for l10n. It needs some thought because we don=E2=80=99t want to mirror every single = Guix API with an inferior equivalent. So perhaps you can run a large part of the processing in the inferior. > There's also some thinking about how to manage the network dependent > checkers. I'd like to get that information in anyway, but also, I think > it might be possible to maybe separate out the network independant parts > of the checkers that are currently in the network dependent list. For > example, the synopsis checker is only in there as it attempts to connect > to the network to check if packages are a GNU package, and I'm wondering > if that can be avoided. Dunno, but I don=E2=80=99t think it=E2=80=99s super important either. What might be more useful is to indicate how critical a warning is: lack of source code is critical, but missing-space-after-period less so. > Anyway, hopefully the code refactoring is generally helpful, and maybe > the --no-network option for guix lint will come in useful as well. Definitely. Thanks! Ludo=E2=80=99.