| 1 |
Stephen Bennett wrote: |
| 2 |
> Many things were discussed in the last round of this thread (Paludis |
| 3 |
> and Profiles, in case anyone missed it), and many useful points raised. |
| 4 |
> One of these, which seems to have been largely missed in amongst the |
| 5 |
> other noise, forms the basis of this proposal. It is in some ways more |
| 6 |
> and in some ways less intrusive than the previous proposal, |
| 7 |
> and is also completely package-manager-agnostic. |
| 8 |
> |
| 9 |
> In short, I would like to suggest replacing sys-apps/portage atoms in |
| 10 |
> the base and default-linux profiles with virtual/portage, and removing |
| 11 |
> the python dependencies from them. For most users this would have an |
| 12 |
> effective zero change, since the default provider for virtual/portage |
| 13 |
> is sys-apps/portage, and the python dependency will be pulled in by |
| 14 |
> Portage when calculating system deps. According to my understanding, |
| 15 |
> this should also produce no change when building release media, due to |
| 16 |
> both Portage and Python being in packages.build. |
| 17 |
> |
| 18 |
> The only change introduced by this is that it becomes possible to |
| 19 |
> bootstrap a system with a different package manager simply by |
| 20 |
> installing it before 'system'. There are a couple more changes needed |
| 21 |
> to allow this -- namely that a few system packages have old |
| 22 |
> dependencies on >=portage-2.0.49, but these can be resolved seperately. |
| 23 |
> Any problems caused by packages depending implicitly upon Python will |
| 24 |
> show up only on systems not using Portage, and can be easily fixed with |
| 25 |
> the cooperation of package maintainers. |
| 26 |
> |
| 27 |
> I would like to think that this proposal addresses most of the concerns |
| 28 |
> raised in the last thread -- it implies nothing about support for any |
| 29 |
> other package manager, and introduces nothing that could cause problems |
| 30 |
> for Portage users, while still allowing alternative package managers to |
| 31 |
> use the tree without needing Portage installed. |
| 32 |
> |
| 33 |
> I am also aware that this falls roughly under what the Council was |
| 34 |
> asked to discuss in its June meeting, but since that seems to have not |
| 35 |
> happened, I'm bringing it up anyway, since I would like to get |
| 36 |
> something done here. |
| 37 |
> |
| 38 |
> Comments? |
| 39 |
|
| 40 |
If you can spot those issues and fix them w/out rush on package |
| 41 |
mantainers, no problems at all. |
| 42 |
|
| 43 |
Just make sure nobody will ask to "fix" something working with portage |
| 44 |
in 0time because of paludis changes. |
| 45 |
|
| 46 |
lu |
| 47 |
|
| 48 |
PS: there is a formal spec about ebuilds now? |
| 49 |
|
| 50 |
-- |
| 51 |
|
| 52 |
Luca Barbato |
| 53 |
|
| 54 |
Gentoo/linux Gentoo/PPC |
| 55 |
http://dev.gentoo.org/~lu_zero |
| 56 |
|
| 57 |
-- |
| 58 |
gentoo-dev@g.o mailing list |
Archives