| 1 |
On Sat, 26 Jul 2014 12:32:20 +0000 (UTC) |
| 2 |
Martin Vaeth <martin@×××××.de> wrote: |
| 3 |
> Ciaran McCreesh <ciaran.mccreesh@××××××××××.com> wrote: |
| 4 |
> > User installs foo-1.1-r1 |
| 5 |
> > Developer makes foo-1.1-r1.1 |
| 6 |
> > foo-1.1* is removed from the tree |
| 7 |
> > User syncs |
| 8 |
> |
| 9 |
> How is this different from your suggestion |
| 10 |
> (which you *claim* to be non-broken): |
| 11 |
> |
| 12 |
> User installs foo-1.1-r1 |
| 13 |
> Developer makes foo-1.1-r2 |
| 14 |
> foo-1.1* is removed from the tree |
| 15 |
> User syncs |
| 16 |
> |
| 17 |
> In fact, the result is completely the same, |
| 18 |
> no matter whether you have minor revisions or not, |
| 19 |
> and no matter whether you have static or dynamic deps. |
| 20 |
> |
| 21 |
> What is *actually* broken here is that the user |
| 22 |
> has installed a package which is not maintained |
| 23 |
> anymore: *This* is what needs to be fixed. |
| 24 |
> This issue is completely independent of static |
| 25 |
> vs. dynamic deps. |
| 26 |
> You misuse this problem as a strawman, only. |
| 27 |
|
| 28 |
Uhm. That works just fine... I don't think you understand how this |
| 29 |
works: we can always use the metadata that's in VDB for dealing with the |
| 30 |
installed package. The issue is that sometimes Portage tries to guess |
| 31 |
that it's better to use the metadata from an ebuild instead of what's |
| 32 |
in VDB when dealing with an installed package. |
| 33 |
|
| 34 |
-- |
| 35 |
Ciaran McCreesh |
Archives