1 |
On Mon, 30 Jul 2012 11:23:31 +0200 |
2 |
Dirkjan Ochtman <djc@g.o> wrote: |
3 |
|
4 |
> On Mon, Jul 30, 2012 at 11:16 AM, Michał Górny <mgorny@g.o> |
5 |
> wrote: |
6 |
> > You're saying 'we should explore the option not to upgrade packages |
7 |
> > without user explicitly saying "please upgrade this package"'. |
8 |
> |
9 |
> Yes, or until the package maintainer removes the annotation again. |
10 |
> |
11 |
> >> > Or maybe we should explore the option of fixing python.eclass to |
12 |
> >> > not depend on random python versions implicitly? |
13 |
> >> |
14 |
> >> I don't know that it does, but I hope you can enlighten me! |
15 |
> > |
16 |
> > It does depend on python versions based on $USE_PYTHON. And |
17 |
> > USE_PYTHON defaults to 2+3 if installed. If it pulls in Python 3, |
18 |
> > it will pull in all the time unless you set USE_PYTHON manually and |
19 |
> > remerge all the packages manually. |
20 |
> |
21 |
> It seems to me that we could fix USE_PYTHON to always depend only on 2 |
22 |
> unless it's explicitly set by the user, but it seems to me that |
23 |
> Portage would, in that case, still pull python3 into the stages. |
24 |
|
25 |
Portage will pull python3 if and only if anything depends on python:3.* |
26 |
or anything depends on python without a SLOT (or either is in @world). |
27 |
If you want it not there, you have to make sure the deps explicitly |
28 |
state it. |
29 |
|
30 |
-- |
31 |
Best regards, |
32 |
Michał Górny |