1 |
-----BEGIN PGP SIGNED MESSAGE----- |
2 |
Hash: SHA1 |
3 |
|
4 |
On Tuesday 03 February 2004 01:24, Christian Hoenig wrote: |
5 |
> |
6 |
> Hi Paul, though I'm not the 'you', I will give my statement here. |
7 |
> |
8 |
> Long time ago I thougth about having an unstable/stable and mature |
9 |
> kind of flag here. The point is, that sometimes changes from ~arch to |
10 |
> arch occur and a short time later a return to ~arch is needed because |
11 |
> the responsible dev overlooked a point which became clear by the |
12 |
> community. |
13 |
> |
14 |
> A jump from for example 'x86' to 'stable:x86' (which could also be |
15 |
> denoted as !x68) should only occur if it 'seems' to be testet from the |
16 |
> community for at least some days! |
17 |
> |
18 |
> I hope you did get my point, if not, get back to me :-) |
19 |
|
20 |
My point is that we want 1 flag. So if a package has the following |
21 |
keywords: |
22 |
KEYWORDS="stable x86 ppc ~sparc ~hppa" |
23 |
|
24 |
It would be included into the snapshot for x86 and ppc only. It is very |
25 |
uncommon for an ebuild that is stable enough to go into the snapshot to |
26 |
be accidently marked stable for an arch. Also we will have QA over the |
27 |
snapshot before it is released. This QA will at least involve making all |
28 |
packages (for a GRP tree). As we are dealing with a snapshot it is |
29 |
actually easier to ensure that things compile because a lot of variation |
30 |
is taken away. You can ensure that you only have say openssl-0.7 so you |
31 |
don't get problems with openssl-6 |
32 |
|
33 |
Paul |
34 |
|
35 |
- -- |
36 |
Paul de Vrieze |
37 |
Gentoo Developer |
38 |
Mail: pauldv@g.o |
39 |
Homepage: http://www.devrieze.net |
40 |
-----BEGIN PGP SIGNATURE----- |
41 |
Version: GnuPG v1.2.4 (GNU/Linux) |
42 |
|
43 |
iD8DBQFAH3RdbKx5DBjWFdsRAgGmAKDRyBunn7Q1zsGmSWrG6UGxRmbE9QCghnql |
44 |
7TMWtvrjL9puyF3LFuf2sIA= |
45 |
=dmDP |
46 |
-----END PGP SIGNATURE----- |
47 |
|
48 |
-- |
49 |
gentoo-dev@g.o mailing list |