From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mathieu Lirzin Subject: bug#30437: No =?UTF-8?Q?=E2=80=9C.guix-profile/bin/python=E2=80=9D?= after =?UTF-8?Q?=E2=80=98guix?= package -i =?UTF-8?Q?python=E2=80=99?= Date: Tue, 13 Feb 2018 01:45:42 +0100 Message-ID: <87zi4d4szt.fsf@gnu.org> References: <87po5a5ew6.fsf@gnu.org> <20180212182307.7c0982ee@scratchpost.org> <87lgfy577f.fsf@gnu.org> <871shpls0k.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 ([2001:4830:134:3::10]:59233) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1elOjT-0003j1-Hg for bug-guix@gnu.org; Mon, 12 Feb 2018 19:46:04 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1elOjS-0001zQ-9f for bug-guix@gnu.org; Mon, 12 Feb 2018 19:46:03 -0500 Received: from debbugs.gnu.org ([208.118.235.43]:60610) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1elOjS-0001zF-6R for bug-guix@gnu.org; Mon, 12 Feb 2018 19:46:02 -0500 Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1elOjR-0002GT-VU for bug-guix@gnu.org; Mon, 12 Feb 2018 19:46:02 -0500 Sender: "Debbugs-submit" Resent-Message-ID: In-Reply-To: <871shpls0k.fsf@elephly.net> (Ricardo Wurmus's message of "Tue, 13 Feb 2018 00:14:51 +0100") 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.org@gnu.org Sender: "bug-Guix" To: Ricardo Wurmus Cc: 30437@debbugs.gnu.org Ricardo Wurmus writes: > Mathieu Lirzin writes: > >> Danny Milosavljevic writes: >> >>> On Mon, 12 Feb 2018 17:52:41 +0100 >>> Mathieu Lirzin wrote: >>> >>>> I think it would work better if when installing python@3, a =E2=80=98p= ython=E2=80=99 >>>> executable would be available in the PATH. Maybe there is a technical >>>> reason for not doing so, but I find its absence rather confusing. >>> >>> For backward compatibility, the 'python' executable always has to be Py= thon 2. >>> >>> A lot of old scripts specify just "/usr/bin/env python" in the shebang. >>> >>> On the other hand if your script supports Python 3 you know it. >> >> Hum, then I don't understand why we aren't applying this argument to >> every interperter (Perl, Guile, ...) which introduce backward >> incompatible changes? > > As far as I know, the Python 3 package does not provide the =E2=80=9Cpyth= on=E2=80=9D > binary. That=E2=80=99s upstream=E2=80=99s decision. Indeed your are right, they provide no configure option to enable such thing as they don't encourage downstream to use =E2=80=98python=E2=80=99 as= an alias to =E2=80=98python3=E2=80=99 yet. After some research this is discussed in PE= P-0394. [1] As a consequence it seems reasonable for Guix to not deviate from upstream recommandation, even if the shebang issue described by PEP-0394 are mitigated by the functional package management paradigm. Thanks. [1] https://www.python.org/dev/peps/pep-0394/ --=20 Mathieu Lirzin GPG: F2A3 8D7E EB2B 6640 5761 070D 0ADE E100 9460 4D37