Gentoo Archives: gentoo-dev

From: Roy Marples <uberlord@g.o>
To: gentoo-dev@l.g.o
Subject: Re: [gentoo-dev] [RFC] client/server policy for ebuilds
Date: Sat, 10 Jun 2006 01:14:44
Message-Id: 200606100201.25523.uberlord@gentoo.org
In Reply to: Re: [gentoo-dev] [RFC] client/server policy for ebuilds by Alec Warner
1 On Saturday 10 June 2006 01:33, Alec Warner wrote:
2 > > So we have two use flags - client and server. Here are the possabilities
3 > >
4 > > -client -server
5 > > +client -server
6 > > +client +server
7 > > -client +server
8 > >
9 > > Do we read -client -server and +client +server to mean the same thing?
10 > > If so the logic can read
11 > >
12 > > if use client || ! use server ; then
13 > > # build client
14 > > fi
15 > > if use server || ! use client ; then
16 > > # build server
17 > > fi
18 > >
19 > > How does portage stop us from doing that now?
20 >
21 > built_with_use is then incorrect, since for -client -server you really
22 > built both.
23
24 use client && build client
25 use server && build server
26
27 The problem here is that breaks existing ebuilds, which could be viewed as
28 equally bad.
29
30 But technically built_with_use isn't incorrect as the ebuild wasn't built with
31 it. To effectively use built_with_use you cannot assume that the flag does
32 what it says on the tin - you have to inspect the ebuild code you're
33 querying.
34
35 Prior history shows deps of db vs gdbm where if both or neither then db was
36 used, otherwise the flagged db was used.
37
38 Problems problems - soltutions that work with existing installs or do we just
39 bite the bullet and do
40
41 ! use client && ! use server && die "must select either client or server"
42
43 Which kinda defeats the purpose of a clean install.
44
45 --
46 Roy Marples <uberlord@g.o>
47 Gentoo/Linux Developer (baselayout, networking)
48 --
49 gentoo-dev@g.o mailing list

Replies