1 |
Am Montag, 16. Oktober 2006 00:59 schrieb Alec Warner: |
2 |
> Ciaran McCreesh wrote: |
3 |
> > | |
4 |
> > | Uh, what kind of conflicting behaviour and what sanity checks are |
5 |
> > | you talking about here? Did you _really_ miss the whole point of |
6 |
> > | this feature? |
7 |
> > |
8 |
> > Before changing default values for USE flags, arch and release |
9 |
> > people have to make sure that that change won't do something nasty |
10 |
> > like introduce circular or built_with_use deps into the default |
11 |
> > system resolution. |
12 |
> |
13 |
> I don't see how the location of the default USE affects these things. |
14 |
> If I change default USE in my ebuild; I have to do sanity checks. If |
15 |
> I change default USE in the profile; I have to do sanity checks *in |
16 |
> that profile*. |
17 |
> |
18 |
> So if your argument is that it's cheaper to check just N profiles ( |
19 |
> the profiles affected by my change ) versus all available profiles; |
20 |
> then I agree with you on that point. |
21 |
> |
22 |
> However I still believe there exist examples where default USE in an |
23 |
> ebuild is a better solution. |
24 |
|
25 |
>From my point of view as an architecture dev and releng member: Having |
26 |
all default USE-flags at one spot (per profile) _is_ easier to maintain. |
27 |
|
28 |
Ciaran has a point here: Default useflags have annoyed me in the past |
29 |
while building releases, and having to change several packages (and |
30 |
redigesting them) for the snapshot is way more: |
31 |
* complicated |
32 |
* time-consuming |
33 |
* error-prone |
34 |
than changing them in the profiles directory. |
35 |
|
36 |
Chris: I'd like to have your thoughts on this. |
37 |
|
38 |
Danny |
39 |
-- |
40 |
Danny van Dyk <kugelfang@g.o> |
41 |
Gentoo/AMD64 Project, Gentoo Scientific Project |
42 |
|
43 |
-- |
44 |
gentoo-dev@g.o mailing list |