1 |
On Fri, 18 Sep 2015 13:04:45 +0200 |
2 |
hasufell <hasufell@g.o> wrote: |
3 |
|
4 |
> On 09/18/2015 12:56 PM, Alexis Ballier wrote: |
5 |
> > On Fri, 18 Sep 2015 11:58:09 +0200 |
6 |
> > hasufell <hasufell@g.o> wrote: |
7 |
> > |
8 |
> >> On 09/18/2015 11:55 AM, Duncan wrote: |
9 |
> >>> Alexis Ballier posted on Fri, 18 Sep 2015 11:04:19 +0200 as |
10 |
> >>> excerpted: |
11 |
> >>> |
12 |
> >>>>> Keep in mind what this implies when you change these |
13 |
> >>>>> dependencies without bumping the ebuilds that use them. |
14 |
> >>>> |
15 |
> >>>> only way i see these changing is with a new ros_messages_*** |
16 |
> >>>> useflag, which will cause a rebuild anyway |
17 |
> >>> |
18 |
> >>> ?? Only with --newuse or similar, tho? Otherwise USE (or |
19 |
> >>> USE_EXPAND here) changes don't trigger rebuilds, do they? |
20 |
> >>> |
21 |
> >> |
22 |
> >> Correct. It's not good to rely on this and expect users to have a |
23 |
> >> certain update pattern or even use a particular PM. |
24 |
> >> |
25 |
> > |
26 |
> > then they wont have the messages if they don't rebuild, and |
27 |
> > cat/pkg[ros_messages_newthing] deps wont be satisfied and the pm |
28 |
> > will do the right thing by rebuilding. |
29 |
> > nothing to worry about, really. |
30 |
> > |
31 |
> |
32 |
> cat/pkg[ros_messages_newthing] will not be in users VDB until you |
33 |
> revbump cat/pkg, so I don't think it's that easy. |
34 |
> |
35 |
|
36 |
and i think you're confused about what dynamic deps is and is not: |
37 |
cat/pkg[ros_messages_newthing] is not in vdb, so is not satisfied, so |
38 |
pm tries to find something that satisfies it from tree or dies. |
39 |
|
40 |
this is exactly what happens with a >=cat/pkg-1 dep when only cat/pkg-0 |
41 |
is installed :) |