1 |
So I've been talking to kent\n and the conclusion is that our ideas basically |
2 |
do and achieve the same, with a slightly different approach. (I still like |
3 |
mine better though. :) |
4 |
|
5 |
However one valid point that came up in discussions is - whether an arch |
6 |
supports stable keywords is a per-arch setting, not a per-profile setting. So |
7 |
we can actually make things much easier (and the transition safer). |
8 |
|
9 |
Proposal No 2: |
10 |
* Leave profiles.desc unmodified |
11 |
* Introduce a new file arch.desc, which contains the "stability status" of an |
12 |
arch; |
13 |
|
14 |
Syntax: 2 columns, |
15 |
# arch status |
16 |
amd64 stable |
17 |
mips testing |
18 |
sh unstable |
19 |
|
20 |
The meaning of the keywords "stable", "testing", "unstable" is the same as in |
21 |
the previous proposal, |
22 |
|
23 |
> "Does this arch support stable keywords, and how should "arch" vs. "~arch" |
24 |
> be treated?" |
25 |
> - "stable": separately check consistency of ~arch and arch tree, both have |
26 |
> to be OK. This is what repoman is doing now, and is the default if the 4th |
27 |
> column is undefined. |
28 |
> - "testing": treat "arch" as "~arch" when requiring consistency, do not |
29 |
> check "arch" alone. Useful if an arch wants to prepare going stable, useful |
30 |
> for arch teams maintaining a pseudo-stable subset for stages. repoman could |
31 |
> have a new command line switch that temporarily upgrades from "testing" to |
32 |
> "stable" (for arch team work). |
33 |
> - "unstable": check "~arch" only, "arch" in an ebuild produces a fatal |
34 |
> repoman error |
35 |
|
36 |
The combination of current profiles.desc and new arch.desc provides the same |
37 |
flexibility as in the previous e-mail. |
38 |
|
39 |
Compatibility and transition: |
40 |
|
41 |
0) PMS should be amended to allow the additional file. |
42 |
|
43 |
1) Compatibility: No arch.desc and new system, or arch not listed in arch.desc |
44 |
*Arches* are treated as "stable" by repoman (current behaviour), with profile |
45 |
status according to profiles.desc. |
46 |
Gentoolkit and other tools trying to determine a list of stable arches should |
47 |
fall back to current method of scanning profiles.desc for stable profiles. |
48 |
|
49 |
2) Compatibility: arch.desc and old system |
50 |
Tools ignore the unknown file (?). |
51 |
Repoman and other tools may emit surplus errors when profiles are checked on |
52 |
arches that are "testing" (they check the consistency of the stable tree |
53 |
alone, which is not OK, since "arch" is supposed to be treated like "~arch"). |
54 |
|
55 |
3) On introduction of the new column, it will be set to "stable" for all |
56 |
stable arches, "testing" for all arches where "inofficial" stable keywords |
57 |
exist (sh, s390, ...), and "unstable" everywhere else. |
58 |
|
59 |
4) Arches in "testing" or "unstable" may eventually consider re-introducing |
60 |
stable *profiles* so their deptree in ~arch remains consistent. |
61 |
|
62 |
More opinions, flames, cookies? |
63 |
|
64 |
Cheers, Andreas |
65 |
|
66 |
|
67 |
-- |
68 |
Andreas K. Hüttel |
69 |
dilfridge@g.o |
70 |
Gentoo Linux developer (council, perl, libreoffice) |