Subject: Re: CVS commit: src/distrib/sets
To: None <firstname.lastname@example.org>
From: Thor Lancelot Simon <email@example.com>
Date: 11/11/2004 12:07:48
On Thu, Nov 11, 2004 at 11:51:20AM -0500, Jim Wise wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> On Thu, 11 Nov 2004, Chris Pinnock wrote:
> >On Thu, Nov 11, 2004 at 03:59:28PM +0000, Alistair Crooks wrote:
> >> It may be that a new switch is necessary, although I doubt this. Does
> >> -X switch PKG_DBDIRs, does it override PKG_DBDIR in the environment,
> >> should we look at SYSPKG_DBDIR instead, should we provide syspkg_*
> >> commands, should we provide syspkg_* aliases in our standard shells,
> >> loads of questions.
> >For benefit of the user experience, however we install system packages
> >it should feel nice and natural. Having a flag for it doesn't feel
> >nature to me.
> This sounds like an argument in favor of sharing a PKG_DBDIR between
> syspkg and pkgsrc.
> What do people think of such a solution?
I think it's the right solution. It also allows pkgsrc packages to
cleanly depend upon system packages -- which, I think, we *really*,
*really* want. If you don't have the "text" set of the system installed
now, for example, some packages don't work; but there's no way to _know_
they won't work, so they just mysteriously fail. Not Good.
Perhaps we need the concept of a package "source" or "publishing entity";
so one could ask the package tools to list out only packages published
by "NetBSD-pkgsrc", for instance, instead of by "NetBSD-src"; so the
default pkg_info invocation could skip packages published by "NetBSD-src",
in other words "skip system packages" but one could easily tell it not