Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 14 Nov 2004 11:15:58 +0000
From:      Matthew Seaman <m.seaman@infracaninophile.co.uk>
To:        Michael Nottebrock <michaelnottebrock@gmx.net>
Cc:        Kirill Ponomarew <krion@voodoo.oberon.net>
Subject:   Re: HEADSUP: INDEX[-5] files were removed from CVS.
Message-ID:  <20041114111558.GA672@happy-idiot-talk.infracaninophile.co.uk>
In-Reply-To: <4196BE10.6060306@gmx.net>
References:  <20041113101925.GB70256@voodoo.oberon.net> <20041113174948.GD76034@nevermind.kiev.ua> <419655EB.8070506@gmx.net> <m3ekixjnve.fsf@merlin.emma.line.org> <4196BE10.6060306@gmx.net>

next in thread | previous in thread | raw e-mail | index | archive | help

--17pEHd4RhPHOinZp
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Sun, Nov 14, 2004 at 03:08:16AM +0100, Michael Nottebrock wrote:
=20
> >"make fetchindex" runs on the order of a minute for my machine (1 Mbit/s
> >link), but I'm definitely not using "make index" on my K6-2/300.
> >
> >It's about time for a _fast_ index generator, or a cache so that only
> >changed records are replaced.
>=20
> I think one has been mentioned in this thread, but since it's written in=
=20
> perl, it's not a candidate for base-system inclusion...

Ummm... did you ever take a close look at how 'make index' works?
Substantial parts of that are written in perl.
=20
> >INDEX has been an annoyance ever since but
> >no-one has a decent solution how ports can do without.
>=20
> Perhaps the recent events will contribute to depart from the stalemate=20
> ports have been in, with no real development going on and portupgrade bei=
ng=20
> 'good enough' for everybody. Perhaps it will just result in portupgrade=
=20
> stopping depending on INDEX to the extent as it does now, which will be=
=20
> 'good enough' again for most people.

How can something like portupgrade(1) work without the dependency
ordering information?  At the moment, that comes from the INDEX file
-- although thinking about it, it would be a lot more efficient over
all for portupgrade(1) to just work out the dependencies of any
installed ports as it goes along.  Now, that is an interesting idea.
I may have to go on a crash course of learning Ruby so I can fully
investigate that.  (Although don't hold your breath, as I've just
started a new job and ENOTIME)

	Cheers,

	Matthew

--=20
Dr Matthew J Seaman MA, D.Phil.                       26 The Paddocks
                                                      Savill Way
PGP: http://www.infracaninophile.co.uk/pgpkey         Marlow
Tel: +44 1628 476614                                  Bucks., SL7 1TH UK

--17pEHd4RhPHOinZp
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (FreeBSD)

iD8DBQFBlz5tiD657aJF7eIRAjTrAKChPnpAKoLwaqbYmL8P7sONB6UnBACdG13G
9pbw6pyxEDV/D3bDeGTUZSo=
=Z7z4
-----END PGP SIGNATURE-----

--17pEHd4RhPHOinZp--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20041114111558.GA672>