| 1 |
On Tue, 25 Sep 2012 14:03:36 -0400 |
| 2 |
Ian Stakenvicius <axs@g.o> wrote: |
| 3 |
|
| 4 |
> -----BEGIN PGP SIGNED MESSAGE----- |
| 5 |
> Hash: SHA256 |
| 6 |
> |
| 7 |
> Since hasufell brought it up, and as I believe he's going to ask Council |
| 8 |
> to approve it before moving forward with this proposal towards including |
| 9 |
> it in an EAPI, I wanted to clarify some of the points mentioned: |
| 10 |
> |
| 11 |
> - --- Quote, GLEP-62 --- |
| 12 |
> > Specifications, paragraph 3: The package manager should treat flags |
| 13 |
> > listed in IUSE_RUNTIME as regular USE flags, except for the |
| 14 |
> > following: |
| 15 |
> > |
| 16 |
> > 1. enabling or disabling any of the flags must not involve |
| 17 |
> > rebuilding the package, |
| 18 |
> > |
| 19 |
> > 2. it should be possible for a package manager to change those |
| 20 |
> > flags on a installed package without using the original ebuild, |
| 21 |
> > |
| 22 |
> > 3. when queried on a installed package, the package manager must |
| 23 |
> > consider a particular flag enabled only if its dependencies are |
| 24 |
> > satisfied already, |
| 25 |
> > |
| 26 |
> > 4. the flags may be listed in the visual output in a distinct way |
| 27 |
> > to inform the user that they affect runtime dependencies only. |
| 28 |
> |
| 29 |
> |
| 30 |
> #2 -- this would, if I'm understanding it properly, mean that the IUSE |
| 31 |
> list and the IUSE_RUNTIME list in the 'original ebuild' (ie in vdb) |
| 32 |
> would be ignored on an emerged package in favour of the ebuild(s) in |
| 33 |
> the tree, right? I'm not so sure this is a good idea. |
| 34 |
|
| 35 |
The exact opposite. The PM should use vdb whenever no non-IUSE_RUNTIME |
| 36 |
flags have changed in vdb and no other reasons have caused it to use |
| 37 |
the ebuild repository. |
| 38 |
|
| 39 |
> IE, if IUSE and IUSE_RUNTIME have changed in the in-tree ebuild and |
| 40 |
> one of those use flags that changed have been triggered or |
| 41 |
> de-triggered I expect that the package should be rebuilt, to keep it |
| 42 |
> consistent with current practices. |
| 43 |
|
| 44 |
If --newuse triggers that, the ebuild will be used. |
| 45 |
|
| 46 |
> IE2, shouldn't the original ebuild be what's used to trigger the |
| 47 |
> skip-rebuild functionality, rather than the in-tree ebuild? |
| 48 |
|
| 49 |
I already said that :). |
| 50 |
|
| 51 |
> #3 -- this seems to imply to me, that the state of a package's |
| 52 |
> effective USE could be modified solely on the basis of a dependency |
| 53 |
> existing or not and have nothing to do with what the flag was set to |
| 54 |
> at emerge time. IE, *not* the state of USE in the vdb. I think this |
| 55 |
> would also be a problem. |
| 56 |
|
| 57 |
No, you're overestimating it. It just means that the package manager |
| 58 |
should first install the dependencies, and then update USE in vdb, |
| 59 |
and not the other way around. |
| 60 |
|
| 61 |
-- |
| 62 |
Best regards, |
| 63 |
Michał Górny |
Archives