Subject: Re: effective configuration from config(8) after "no foo"
To: Alan Barrett <>
From: Quentin Garnier <>
List: current-users
Date: 12/20/2007 22:29:33
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Thu, Dec 20, 2007 at 10:48:37PM +0200, Alan Barrett wrote:
> On Thu, 20 Dec 2007, Quentin Garnier wrote:
> > For instance, the following passes, which is questionable:
> >=20
> > no something
> > no something
> > no something
> I want that to work.  When I write "no something" in my config file, I
> don't mean "I assert that there used to be a 'something', but I'd like
> to remove it now", I mean simply "I don't want a 'something'".

That's different.  "no nonexistent" currently fails.

> I find it quite frustrating that
> 	include GENERIC
> 	no foo
> 	no options BAR
> works one day, but fails the next day after somebody deletes 'foo at
> wherever' or 'options BAR' from GENERIC.

Everyone seems to have a different opinion of how config(1) should
behave.  I've been hacking it for a while now, and I don't think I
actually use any of the features I've been adding over the year.  They
never added enough value to me.  Except maybe "obsolete def...", but
that's not something the end user sees.

I do wish people would start doing whatever is necessary to use
config -L as a regression test suite builder, though, instead of feeling
the pain of maintaining ALL, which is i386-only.

In any case, fixing the current bug and your frustrations can probably
be done in one go.  I'll see how I'll do it.

Quentin Garnier - -
"See the look on my face from staying too long in one place
[...] every time the morning breaks I know I'm closer to falling"
KT Tunstall, Saving My Face, Drastic Fantastic, 2007.

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

Version: GnuPG v1.4.6 (NetBSD)