Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 25 Feb 2007 10:00:12 +0000
From:      Matthew Seaman <m.seaman@infracaninophile.co.uk>
To:        Mark Linimon <linimon@lonesome.com>
Cc:        ports@freebsd.org, Gabor Kovesdan <gabor@freebsd.org>
Subject:   Re: Optionally depending on one of two ports (or none of them)
Message-ID:  <45E15E2C.8020707@infracaninophile.co.uk>
In-Reply-To: <20070224201609.GC4960@soaustin.net>
References:  <20070222141301.007fee4f@localhost> <45DDA117.3050508@FreeBSD.org> <20070224163229.062bd234@localhost> <45E06810.1070704@FreeBSD.org> <45E08E3D.8060104@infracaninophile.co.uk> <20070224201609.GC4960@soaustin.net>

next in thread | previous in thread | raw e-mail | index | archive | help
This is an OpenPGP/MIME signed message (RFC 2440 and 3156)
--------------enig03D591A1243B41E145438F11
Content-Type: text/plain; charset=ISO-8859-15
Content-Transfer-Encoding: quoted-printable

Mark Linimon wrote:
> On Sat, Feb 24, 2007 at 07:13:01PM +0000, Matthew Seaman wrote:
>> Why not do it this way?
>>
>> HAVE_TOR_DEVEL!=3D if pkg_info -I tor-devel-\* >/dev/null 2>&1 ; then =
echo YES; fi
>=20
> The problem with this approach is that you couple another shell invocat=
ion
> into 'make index'.  As long as they are in individual ports, it's ok, b=
ut
> when we get them in the infrastructure, that's when things slow down.

That should be possible to work around though.  After all, the INDEX has =
no
capability to deal with alternate dependency trees.  Thus there has to be=
 a
default dependency, which is what will go into the INDEX available from t=
he
freebsd.org servers.  A scheme like the way the OPTIONS variable works, i=
e.
creating a small Makefile with the results of these sort of choices -- bu=
t
only when specifically invoked by the user and then subsequently just
=2Eincluded if it exists -- should address your concerns.

When building an INDEX, either the options Makefile exists and is include=
d
to possibly override some settings, or it doesn't exist and you just get =
the
defaults.  Either way the extra processing by make(1) during the index bu=
ild
is kept to a minimal amount.
=20
> No, I don't have an alternative proposal.

Well, there are programs available that make maintaining the INDEX
enormously more efficient than building it from scratch every time...

	Cheers,

	Matthew

--=20
Dr Matthew J Seaman MA, D.Phil.                       7 Priory Courtyard
                                                      Flat 3
PGP: http://www.infracaninophile.co.uk/pgpkey         Ramsgate
                                                      Kent, CT11 9PW


--------------enig03D591A1243B41E145438F11
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: OpenPGP digital signature
Content-Disposition: attachment; filename="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.2 (FreeBSD)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFF4V4x8Mjk52CukIwRCJcqAJoCLNa8aQ7Lyy5vNh8e1NGo3CtVwACfcv3h
AC2mp1AKM9SjzTnsSuctGJE=
=aBP2
-----END PGP SIGNATURE-----

--------------enig03D591A1243B41E145438F11--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?45E15E2C.8020707>