| 1 |
On Fri, 28 Dec 2012 00:40:24 +0100 |
| 2 |
"Andreas K. Huettel" <dilfridge@g.o> wrote: |
| 3 |
|
| 4 |
> Am Donnerstag, 27. Dezember 2012, 14:37:37 schrieb Michał Górny: |
| 5 |
> > |
| 6 |
> > a) adding new profiles which will require EAPI=5 and requiring all |
| 7 |
> > users to migrate to them after upgrading portage. Using new |
| 8 |
> > use.stable.mask files in those profiles. |
| 9 |
> > |
| 10 |
> |
| 11 |
> OK here's one way how we could pull option a) through. After all we have some |
| 12 |
> sort of basic versioning present in the profiles (the 10.0 part that makes no |
| 13 |
> sense otherwise). |
| 14 |
> [Note: this does not cover prefix profiles, BSD and other oddities. Need |
| 15 |
> special treatment.] |
| 16 |
> |
| 17 |
> 1) Define a new set of profiles by copying the current ones, and replacing the |
| 18 |
> 10.0 parent by a 13.0 parent. Only differences between 10.0 and 13.0: |
| 19 |
> * the EAPI, now 5, |
| 20 |
> * e.g. an additional parent profiles/base5 (for global stable mask files) |
| 21 |
|
| 22 |
I'm attaching a quickly-made inheritance graph for a current amd64 |
| 23 |
desktop profile. Could you please point out where exactly the new |
| 24 |
profiles would be? |
| 25 |
|
| 26 |
I don't think we can really avoid having the current 'base' profile, |
| 27 |
and I don't think that we should even try doing that. As far as I can |
| 28 |
see, the idea would be to mask the flags completely in base profile, |
| 29 |
and unmask in *stable.mask files. Do I get it correctly? |
| 30 |
|
| 31 |
-- |
| 32 |
Best regards, |
| 33 |
Michał Górny |
Archives