Gentoo Archives: gentoo-dev

From: Olivier Crete <tester@g.o>
To: gentoo-dev@l.g.o
Subject: Re: [gentoo-dev] first council meeting
Date: Fri, 16 Sep 2005 20:40:24
Message-Id: 1126903031.7253.3.camel@cocagne.max-t.internal
In Reply to: Re: [gentoo-dev] first council meeting by Aron Griffis
1 On Fri, 2005-16-09 at 16:21 -0400, Aron Griffis wrote:
2 > Paul de Vrieze wrote:[Fri Sep 16 2005, 04:11:14PM EDT]
3 > > > Those should be in package.mask. ~arch is for candidates for arch that
4 > > > haven't yet proven themselves.
5 > >
6 > > It's often the case that those ebuilds in principle work, but there
7 > > are other reasons for not marking stable yet. Some packages for
8 > > example can have upgrade problems for stable users while being
9 > > stable for testing (by benefit of allready having passed such
10 > > upgrade problems). Masking the ebuild is not really an option
11 > > (causing testing users to go through unnecessary hoops again), while
12 > > marking stable is also no option.
13 >
14 > You're saying there's four states:
15 >
16 > package.mask
17 > ~arch
18 > ~arch candidate for arch
19 > arch
20 [...]
21 > So far I find myself agreeing with Ciaran's idea in this thread.
22 > I don't see the point (yet) in more than three states.
23
24 Well having the "~arch candidate for arch" makes the imlate script much
25 easier to use.. I would find it a PITA to have to go through the
26 changelog of every package to see if it has been in testing for 30
27 days.. Or we need to automate it, something like a
28 imlate-because-the-package-hasnt-changed-in-30-days.py
29
30 --
31 Olivier CrĂȘte
32 tester@g.o
33 Gentoo Developer
34 x86 Security Liaison
35
36
37 --
38 gentoo-dev@g.o mailing list