From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ludovic =?UTF-8?Q?Court=C3=A8s?= Subject: bug#40682: frozen installer in WiFi section -guix 1.1.0 Date: Mon, 20 Apr 2020 00:26:20 +0200 Message-ID: <878sir159v.fsf@gnu.org> References: <873691bzu1.fsf@gnu.org> <87o8ro4biw.fsf@gnu.org> <878sirpzld.fsf@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Return-path: Received: from eggs.gnu.org ([2001:470:142:3::10]:36876 helo=eggs1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1jQIP1-00088Y-Es for bug-guix@gnu.org; Sun, 19 Apr 2020 18:27:03 -0400 Received: from Debian-exim by eggs1p.gnu.org with spam-scanned (Exim 4.90_1) (envelope-from ) id 1jQIP1-0007EY-0n for bug-guix@gnu.org; Sun, 19 Apr 2020 18:27:03 -0400 Received: from debbugs.gnu.org ([209.51.188.43]:34614) by eggs1p.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1jQIP0-0007ES-Kj for bug-guix@gnu.org; Sun, 19 Apr 2020 18:27:02 -0400 Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1jQIP0-0002UX-IE for bug-guix@gnu.org; Sun, 19 Apr 2020 18:27:02 -0400 Sender: "Debbugs-submit" Resent-Message-ID: In-Reply-To: <878sirpzld.fsf@gmail.com> (Mathieu Othacehe's message of "Sun, 19 Apr 2020 11:57:34 +0200") List-Id: Bug reports for GNU Guix List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: bug-guix-bounces+gcggb-bug-guix=m.gmane-mx.org@gnu.org Sender: "bug-Guix" To: Mathieu Othacehe Cc: 40682@debbugs.gnu.org Hi, Mathieu Othacehe skribis: > I believe this is fixed by ea6594e0 on master. Good catch! > This is indeed very fragile, and converting it to D-Bus would be the > right move. I will have a look to Nyacc! Writing a clean Guile-Dbus > library could also be a nice and valuable GSoC/Outreachy project. Yes, it would be ideal to have a pure-Scheme implementation, but in the meantime, perhaps generated bindings via Nyacc could work well enough. > In the meantime, could we merge your patch so that we have a hidden > simulated Wifi network available, that can be used in the installer > tests? I=E2=80=99ve now pushed it as two proper service types: =E2=80=98hostapd-se= rvice-type=E2=80=99 and =E2=80=98simulated-wifi-service-type=E2=80=99, the latter using parts o= f the former. Note that currently the installer tests bypass connectivity checks altogether, but perhaps we can still check what comes before, namely network selection? Thanks, Ludo=E2=80=99.