From: Thomas Jost <schnouki@schnouki.net>
To: Jameson Graef Rollins <jrollins@finestructure.net>,
 notmuch@notmuchmail.org
Subject: Re: [PATCH 0/6] Rebase of Pieter's "set test prereqs"
In-Reply-To: <87lirf99vt.fsf@servo.finestructure.net>
References: <1320176954-4897-1-git-send-email-pieter@praet.org>
	<1321454035-22023-1-git-send-email-schnouki@schnouki.net>
	<87lirf99vt.fsf@servo.finestructure.net>
User-Agent: Notmuch/0.10_rc1+21~gaf76c55 (http://notmuchmail.org)
	Emacs/24.0.90.1 (x86_64-unknown-linux-gnu)
Date: Wed, 16 Nov 2011 21:17:48 +0100
Message-ID: <87zkfv95zn.fsf@schnouki.net>
MIME-Version: 1.0
Content-Type: multipart/signed; boundary="=-=-=";
	micalg=pgp-sha1; protocol="application/pgp-signature"
Precedence: list

--=-=-=
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

On Wed, 16 Nov 2011 10:53:42 -0800, Jameson Graef Rollins <jrollins@finestr=
ucture.net> wrote:
> On Wed, 16 Nov 2011 15:33:49 +0100, Thomas Jost <schnouki@schnouki.net> w=
rote:
> > Hello list,
> >=20
> > This is another rebased version of Pieter's series to add GPG and Emacs=
 as test
> > prereqs, plus some additions on my own. (Rebased and posted as requeste=
d by
> > Pieter [1].)
> >=20
> > Changes as compared to Pieter's patches (including parts from [2]):
> > - prereqs are not tested using test_expect_success as they were in Piet=
er's
> >   original patches, but using a new function called test_set_bin_prereq=
. I wrote
> >   this before the gdb prereq was added, hence the different way to set =
it.
>=20
> Hey, Thomas.  Thanks so much for this work.  This sounds like a better
> solution.
>=20
> However, in the patches you send I see a lot of changes of the form
>=20
>   -test_expect_success 'emacs delivery of encrypted message with attachme=
nt' \
>   +test_expect_success GPG 'emacs delivery of encrypted message with atta=
chment' \
>=20
> and
>=20
>   -test_expect_equal \
>   +test_expect_equal GPG \
>=20
> which seems to contradict what you've said above.  Not to mention that I
> don't see anything that modifies calls to the test_expect_ functions.
> Basically I see a lot more in the diffs than I would have expected in a
> cursory look.  Is this just a rebase flub, or is there something I'm
> missing?
>=20
> jamie.

Hi Jamie,

I guess I wasn't clear in my explanations :)

Pieter's patches use this to detect the presence of GPG/Emacs and set
the prereq:

    +# GnuPG is a prereq.
    +test_expect_success "prereq: GnuPG is present" "which gpg" \
    +    && test_set_prereq GPG

There are 2 problems with this approach:=20

=2D test_expect_success returns 0 regardless of the actual result of the
  command it runs. So even if gpg is not installed,=20
    text_expect_success "..." "which gpg"
  will succeed, and "test_set_prereq GPG" will be run.
  This, however, has been fixed in commit 003e7180 -- which had not been
  pushed when I wrote this in the first place :)

=2D using test_expect_* to set a prereq does not make sense. If emacs is
  absent, the test suite would report a failed test. But a missing
  prereq is *not* a notmuch issue, so this should *not* be reported as a
  failed test.

Hence my first patch, which defines test_set_bin_prereq, a new helper
function to set a prereq without using any test_expect_*.

After that we can use the normal prereq syntax from the test suite:
=2D test_expect_success COMMAND --> run COMMAND, expecting it to succeed
=2D test_expect_success PREREQ COMMAND --> skip if PREREQ is not set, else
  run the test as before
(and same thing with the other test_expect_* functions)

Does it make more sense now?

Regards,

=2D-=20
Thomas/Schnouki

--=-=-=
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)

iQEcBAEBAgAGBQJOxBptAAoJEMPdciX+bh5IabgH/3voa/YJjklax+LokiZr2B/J
BcqVAsT5D1pcNPdO+P4LFppIepjxmR6GnLJbX3kzmjb8/p9eV39qGzrGlK2VDs8v
h6aLZ9jx5UkSMHRBEZxwRD2yeJkiv3KsCj37pvhtH39gEoinEpCZqtimSCZMahNm
/rM+IB8Y4ZNOuz1S/l3v47XSOltoHUgy8WGwjQKiihcb0aJmw30QREdNiYMqaT0v
PM8MiVmRHeuv56pbSFqh1IHCNemb0XzU/TeZD1rW2Y0wzMQcbKyiTpw5sHtxhZuD
NgpW/vsjaOLGIiWL+1Z8NIbIXfanOaYST5MQZ6uqfKWbpqrB2nZ4w/D3Qxr7B10=
=hEq/
-----END PGP SIGNATURE-----
--=-=-=--