[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: JOEL BAKER, STOP TELLING PEOPLE TO DEPEND ON LIBC-DEV *INSTEAD* OF LIBC6-DEV



On Tue, Oct 15, 2002 at 10:28:30AM +1000, Daniel Stone wrote:
> On Mon, Oct 14, 2002 at 05:50:26PM -0600, Brother Joel Baker preached da werd, yo:
> > However, as read, this would make little or not sense as to why one cannot
> > (or even should not) use libc-dev instead of libc*-dev unless you need a
> > versioned dependancy - since all libc*-dev should Provide: libc-dev, and
> > there should be exactly one that applies to any given arch, there is no
> > "preference" that would make any sense for all arches.
> 
> Because, when you have multiple packages providing libc-dev, apt stops
> and says "Shit! Which one should I install?". Your build thus fails.

Er. That's exactly what it SHOULD do, in this particular case. Installing
libc6-dev on, say, Hurd or NetBSD, would be a Bad Idea (tm). While I accept
that the general case does not have these properties, I assert that this
sort of case is exactly why Policy 7.4 is a 'should' and not a 'must'. In
the cases where there should be *exactly one* package which provides this,
and more importantly, in which there is *not* a sane default, a default
should not be specified.

Or are you saying that APT fails to invoke the same branch of 'which one
should I install' that it must use if the real package does not exist on
that platform? If so, can someone explain to me why *this* would be the
case?

> > And that's a "Warning", which would match the 'should' clause in 7.4, which
> > all makes sense - until you throw in Branden's claim that this utterly
> > breaks installation. What I'm asking is why, if Policy 7.4 says 'should'
> > and if there is no rational default for selecting a 'real' version in the
> > presence of more than one (which should never happen in the first place, in
> > this particular instance), installation breaks.
> 
> Considering you're confusing buildd's, you *REALLY* *should* do this.

See above. In *this case* (as opposed to the general one), I can't see why
they should ever be confused.
-- 
***************************************************************************
Joel Baker                           System Administrator - lightbearer.com
lucifer@lightbearer.com              http://users.lightbearer.com/lucifer/

Attachment: pgpVdkxEpm1YB.pgp
Description: PGP signature


Reply to: