1 |
El mié, 23-07-2014 a las 14:33 +0100, Ciaran McCreesh escribió: |
2 |
> On Mon, 21 Jul 2014 23:06:07 +0200 |
3 |
> Pacho Ramos <pacho@g.o> wrote: |
4 |
> > Maybe this could be solved by having two kinds of revisions: |
5 |
> > - One would rebuild all as usually (for example, -r1...) |
6 |
> > - The other one would only regenerate VDB and wouldn't change the |
7 |
> > installed files (for example, -r1.1) |
8 |
> > |
9 |
> > But I am not sure if it could be viable from a "technical" point of |
10 |
> > view :( |
11 |
> |
12 |
> This in no way solves the problem. Consider the following course of |
13 |
> events: |
14 |
> |
15 |
> User installs foo-1.1-r1 |
16 |
> Developer makes foo-1.1-r1.1 |
17 |
> foo-1.1* is removed from the tree |
18 |
> User syncs |
19 |
> |
20 |
|
21 |
Isn't there a way for PMs to know that they need to not rebuild full if |
22 |
user has 1.1-r1 *installed* in his system and pretends to update to |
23 |
-r1.1? |