1 |
hasufell wrote: |
2 |
|
3 |
> Jörg Schaible: |
4 |
>> Alexandre Rostovtsev wrote: |
5 |
>> |
6 |
>>> On Mon, 2014-06-23 at 22:15 +0200, Jörg Schaible wrote: |
7 |
>>>> So, why the heck, was the dependency to dev-libs/glib changed for an |
8 |
>>>> existing ebuild without increasing its version (e.g. |
9 |
>>>> dbus-glib-0.100.2-r2)? |
10 |
>>> |
11 |
>>> Please see http://article.gmane.org/gmane.linux.gentoo.devel/91615 |
12 |
>> |
13 |
>> These blocks had nothing to do with the multilibs ABI. It has been just |
14 |
>> the updated versions for the dependencies. |
15 |
>> |
16 |
> |
17 |
> I'm not sure if you understood the bug. It was breaking dependency |
18 |
> calculation of portage, so the fallout you see is minor to what was |
19 |
> going on. |
20 |
|
21 |
|
22 |
The dependency calculation worked perfectly, it just could not resolve them |
23 |
anymore, because those suddenly required newer packages are hard masked on |
24 |
my system to keep the software *I* need for my daily work running. |
25 |
|
26 |
|
27 |
> Revbumping and restabilizing all of these packages (a LOT) would have |
28 |
> been unrealistic. |
29 |
|
30 |
|
31 |
And the question was, why was the version for these deps upgraded in those |
32 |
ebuild at all. The official tree did not contain anything older anyway. |
33 |
|
34 |
|
35 |
> Another possibility would have been to revbump the ebuild and make it |
36 |
> instantly stable without arch teams involvement. That would actually be |
37 |
> the cleaner way, but afair some people don't agree with that, so it |
38 |
> isn't standard practice. |
39 |
> |
40 |
> However, you can still overwrite tree ebuilds in your local overlay and |
41 |
> revert dependencies. I once did that with pypy, because it triggered too |
42 |
> many rebuilds for me. |
43 |
|
44 |
|
45 |
That's what I did in the end for all "bumped" ebuilds, but the effort would |
46 |
not have been necessary at all. |
47 |
|
48 |
- Jörg |