From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Bavier Subject: Re: [PATCH 10/14] gnu: Add MUMPS. Date: Sun, 28 Jun 2015 10:59:26 -0500 Message-ID: <20150628105926.06922425@openmailbox.org> References: <1435463151-32099-1-git-send-email-ericbavier@openmailbox.org> <1435463151-32099-11-git-send-email-ericbavier@openmailbox.org> <87h9prvbgn.fsf@gnu.org> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:59505) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Z9JXm-0003Gt-KU for guix-devel@gnu.org; Sun, 28 Jun 2015 16:51:15 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Z9JXi-0007MV-D8 for guix-devel@gnu.org; Sun, 28 Jun 2015 16:51:14 -0400 Received: from smtp19.openmailbox.org ([62.4.1.53]:47377) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Z9JXi-0007MC-4F for guix-devel@gnu.org; Sun, 28 Jun 2015 16:51:10 -0400 In-Reply-To: <87h9prvbgn.fsf@gnu.org> 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-bounces+gcggd-guix-devel=m.gmane.org@gnu.org To: Ludovic =?UTF-8?B?Q291cnTDqHM=?= Cc: guix-devel@gnu.org, Eric Bavier On Sun, 28 Jun 2015 22:12:40 +0200 ludo@gnu.org (Ludovic Court=C3=A8s) wrote: > ericbavier@openmailbox.org skribis: >=20 > > From: Eric Bavier > > > > * gnu/packages/maths.scm (mumps, mumps-metis, mumps-openmpi) > > (mumps-metis-openmpi): New variables. > > * gnu/packages/patches/mumps-build-parallelism.patch: New patch. > > * gnu-system.am (dist_patch_DATA): Add it. >=20 > ISTR that some time ago MUMPS included obfuscated code; I assume this > is no longer the case? Indeed, the 5.0.0 release no longer contains obfuscated code. I had not realized this; thanks for bringing it my attention! ;) >=20 > > + (zero? (system* "make" "all" > > + (format #f > > "-j~a" (current-processor-count)))))) >=20 > Rather =E2=80=98parallel-job-count=E2=80=99. OK >=20 > > + (lambda* (#:key (exec-prefix '()) #:allow-other-keys) > > + (with-directory-excursion "examples" > > + (use-modules (srfi srfi-1)) > > + (every > > + (lambda (prec type) > > + (use-modules (ice-9 popen)) >=20 > Local =E2=80=98use-modules=E2=80=99 forms are not very nice (it=E2=80=99s= not documented.) > Rather use: >=20 > #:modules ((srfi srfi-1) > (ice-9 popen) > ,@%gnu-build-system-modules) Much nicer, yes. >=20 > > + (let ((tester (apply open-pipe* > > + `(,OPEN_WRITE > > + ,@exec-prefix > > + ,(string-append "./" prec > > + > > "simpletest")))) > > + (input (open-input-file > > + (string-append "input_simpletest_" > > type)))) > > + (begin > > + (dump-port input tester) > > + (close-port input) > > + (zero? (close-pipe tester))))) > > + '("s" "d" "c" "z") > > + '("real" "real" "cmplx" "cmplx"))))) >=20 > Please add a comment saying that this does, to appease the > reader. ;-) Sure thing. >=20 > > + (license license:public-domain))) >=20 > I think it=E2=80=99s CeCILL-C, see > . Good catch. Yes, this is also new in the 5.0.0 release. I believe I orginally started writing this recipe for 4.10.0. =20 Thanks for the review, `~Eric