1 |
With the introduction of EAPI=2, I noticed we had some problems in the |
2 |
eclasses that now do logic based on the active EAPI. I fixed all |
3 |
eclasses by ignoring the "prefix" component in EAPI when they deal with |
4 |
it, but of course it indicates the end of the journey here. |
5 |
|
6 |
I've been suggested by several people in #gentoo-portage that I bring |
7 |
Prefix to -dev with the obvious target to have Prefix being integrated |
8 |
in mainline Portage, and hence it being its own EAPI -- even though that |
9 |
won't fly since Prefix is orthogonal to any currently existing EAPI. |
10 |
|
11 |
A direct consequence of going that route, of course is the introduction |
12 |
of Prefix in gentoo-x86 ... going mainline. I decided for myself that |
13 |
those who gave me the suggestion are right. We have Prefix for a long |
14 |
while now, and if I try to objectively review myself and our project, I |
15 |
think we did quite a nice job, especially considering the stats graph I |
16 |
just sent out doesn't depict *all* work that's gone in the project. |
17 |
|
18 |
So, it seems the time has come to try and prepare to write up what we |
19 |
need from the people to agree on. I've thought about my huge "GLEP in |
20 |
progress", but I think it'll go nowhere for a discussion. Instead, I |
21 |
think zmedico's RFCs have been well formulated, short pieces of text |
22 |
that received a in general well targetted discussion. I think we should |
23 |
try to see if we can bring the ideas behind Prefix in RFCs like that, |
24 |
and then have them pushed into Portage trunk afterwards. |
25 |
|
26 |
What you all think? I need your help to say the least. This seems to |
27 |
me to be large scale, and not something for just a one-man show... |
28 |
|
29 |
|
30 |
-- |
31 |
Fabian Groffen |
32 |
Gentoo on a different level |