From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?utf-8?Q?Ludovic_Court=C3=A8s?= Subject: Re: [PATCH] gnu: Add glibc-locales variants for older versions of glibc. Date: Tue, 22 Jan 2019 14:16:06 +0100 Message-ID: <87munsomsp.fsf@gnu.org> References: <20190117131753.15696-1-rekado@elephly.net> <87a7jzh0yy.fsf@elephly.net> <87tvi4p84x.fsf@gnu.org> <87o98bdrzn.fsf@elephly.net> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Return-path: Received: from eggs.gnu.org ([209.51.188.92]:38144) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1gm1Wl-0003Mq-45 for guix-devel@gnu.org; Tue, 22 Jan 2019 14:16:04 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1gm1Wj-0006ZZ-EI for guix-devel@gnu.org; Tue, 22 Jan 2019 14:16:02 -0500 Received: from hera.aquilenet.fr ([2a0c:e300::1]:51336) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1gm1Wj-0006FD-57 for guix-devel@gnu.org; Tue, 22 Jan 2019 14:16:01 -0500 In-Reply-To: <87o98bdrzn.fsf@elephly.net> (Ricardo Wurmus's message of "Sun, 20 Jan 2019 20:54:52 +0100") List-Id: "Development of GNU Guix and the GNU System distribution." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org Sender: "Guix-devel" To: Ricardo Wurmus Cc: guix-devel@gnu.org Hello, Ricardo Wurmus skribis: >>> * gnu/packages/base.scm (make-glibc-locales, make-glibc-utf8-locales): = New >>> procedures. >>> (glibc-locales): Express in terms of make-glibc-locales. >>> (glibc-utf8-locales): Express in terms of make-glibc-utf8-locales. >>> (glibc-locales-2.27, glibc-utf8-locales-2.27): New variables. > > [=E2=80=A6] > >> I don=E2=80=99t like the package name trick, but I don=E2=80=99t have a = better solution. >> Perhaps we could have a special property to explicitly allow for several >> versions of this package in the same profile (say >> =E2=80=98allow-multiple-versions?=E2=80=99), but that=E2=80=99s a bit mo= re work. > > I also don=E2=80=99t like to work around this by changing the package nam= es. I > thought of allowing multiple versions via property, but it=E2=80=99s not = clear > how it should behave. I=E2=80=99d want to have only major versions appea= r as > non-conflicting and still prevent the installation of variants of the > same version. > > The next question then is if the property should be a procedure that > takes the current and the potentially conflicting package as arguments > and decides whether they are conflicting, or if this should be handled > centrally when the property is present. Would there be other use cases? If not, I would not worry too much about the major vs. non-major conflicts. I have a slight preference for a property that we can =E2=80=98write=E2=80= =99, which would simplify for instance interaction with inferior packages, hence something like a Boolean =E2=80=98allow-multiple-versions?=E2=80=99 propert= y. As a first approach, when this flag is true, the version check in =E2=80=98check-for-collisions=E2=80=99 could be bypassed. The property wou= ld have to be added to the though, so that the flag is honored even after several =E2=80=9Cguix package=E2=80=9D invocations. How does that sound? Thanks, Ludo=E2=80=99.