Re: [gentoo-dev] Re: Client-server flags (again ;) (was Re: New eclass: gkrellm-plugin)

[Chris Gianelloni]

On Sat, 2007-03-10 at 01:48 +0000, Steve Long wrote:
> Thomas Rösner wrote:
> >> Once we have USE-based dependencies across the board, then yes.  Until
> >> that time, we should really be building both client and server for *all*
> >> packages.
> >
> > I can understand that rationale for the client part, but which packages
> > would depend on the server part of e.g. MySQL if they could?
> > And building the server part to get the small client lib is a larger
> > PITA than building the client lib to get the server, no?
> >
> > In other words: this is a sound argument against the client use flag,
> > but I don't think it's quite as convincing regarding the server flag,
> > which is more important IMHO.
> >
> Yeah on reflection I think just a server flag would be better. There's a
> topic on the forum about it:
> http://forums.gentoo.org/viewtopic-t-525893.html
>
> After all, people who want a server know they do. The default would thus be
> simply to build the client end. This is obviously why there are seven
> ebuilds including cvs doing this already with no complaints.

I wouldn't have too much of a problem with only a "server" USE flag, but
I'd still prefer things were held off until USE-based dependencies.  If
we did decide on a USE=server flag, then it should also likely have
USE=dedicated merged with it so there's only one flag for enabling
server builds.

-- 
Chris Gianelloni
Release Engineering Strategic Lead
Alpha/AMD64/x86 Architecture Teams
Games Developer/Council Member/Foundation Trustee
Gentoo Foundation