1 |
On Fri, Apr 29, 2011 at 08:58:31PM +0300, Alex Alexander wrote: |
2 |
> On Fri, Apr 29, 2011 at 01:52:15PM -0400, Rich Freeman wrote: |
3 |
> > On Fri, Apr 29, 2011 at 1:25 PM, Ulrich Mueller <ulm@g.o> wrote: |
4 |
> > >>>>>> On Fri, 29 Apr 2011, Alex Alexander wrote: |
5 |
> > > |
6 |
> > >> please have a look at the attached patch. |
7 |
> > > |
8 |
> > >> -EAPI="1" |
9 |
> > >> +EAPI="4" |
10 |
> > > |
11 |
> > > Shouldn't the ebuild's phase functions be updated from "EAPI 0 style" |
12 |
> > > to "EAPI 2 style" too? |
13 |
> > |
14 |
> > If the goal is to get this stable in a week, and bypass the 1 month |
15 |
> > waiting period, do we really want to change EAPI at this point? From |
16 |
> > an end-user perspective updating the EAPI on the ebuild provides no |
17 |
> > benefit. Why not just deal with that in a future revision? |
18 |
> > |
19 |
> > I don't see much value in rewriting the ebuild to use a new EAPI |
20 |
> > simply because 4 > 1. |
21 |
> |
22 |
> EAPI was bumped so I could use pkg_pretend, please check out my |
23 |
> (incomplete) patch. |
24 |
|
25 |
I don't remember the details right now, but I remember speaking with |
26 |
vapier when I first started working on openrc, and he stated that he |
27 |
felt we should stay away from higher eapis for system packages. |
28 |
|
29 |
I don't really remember his reasoning for that right now, but I remember |
30 |
that is why I didn't migrate the ebuild to a higher eapi a while back. |
31 |
|
32 |
Also, this patch doesn't stop baselayout-2 from being installed, so I do |
33 |
not know what state it would leave a system in if you ran this and |
34 |
happened to upgrade baselayout, then reboot without installing openrc. |
35 |
|
36 |
William |