From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ben Woodcroft Subject: Re: not tarred source file Date: Mon, 16 May 2016 08:45:32 +1000 Message-ID: <5738FC0C.1050506@uq.edu.au> References: <878tzb11ll.fsf@gnu.org> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: quoted-printable Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:47435) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1b24nE-0000Av-Az for help-guix@gnu.org; Sun, 15 May 2016 18:45:49 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1b24nA-000109-PE for help-guix@gnu.org; Sun, 15 May 2016 18:45:48 -0400 In-Reply-To: <878tzb11ll.fsf@gnu.org> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: help-guix-bounces+gcggh-help-guix=m.gmane.org@gnu.org Sender: "Help-Guix" To: =?UTF-8?Q?Ludovic_Court=c3=a8s?= , Catonano Cc: help-guix@gnu.org Hi there, On 16/05/16 06:57, Ludovic Court=C3=A8s wrote: > Hi! > > Catonano skribis: > >> after having successfully packaged (locally) the Zenburn theme and >> Smartparens, I am preparing the package for clojure-mode >> >> But for the first time it happens that the downloaded source code is n= ot >> tarred so tar fails and so does the unpack phase. >> >> How do I instruct the origin object not to attempt untarring my source= ? > It=E2=80=99s OK if the =E2=80=98origin=E2=80=99 denotes a file that is = not a tarball. > > However, the build procedure will have to pay attention for that. For > instance, the default sequence of phases of =E2=80=98gnu-build-system=E2= =80=99 includes > an =E2=80=98unpack=E2=80=99 phase, which is inappropriate here: > > https://www.gnu.org/software/guix/manual/html_node/Build-Systems.htm= l > > There are a couple of examples, e.g., paredit in emacs.scm. Most of > them use =E2=80=98trivial-build-system=E2=80=99, which requires explici= tly writing the > build/install operations. If it is a single file then I believe it is as simple as deleting the=20 unpack phase e.g. fasttree in bioinformatics.scm, I had thought it was=20 still good practice not to use the trivial build system where possible=20 as it omits some useful defaults. ben