| 1 |
On Mon, Aug 07, 2006 at 01:13:34PM -0700, Zac Medico wrote: |
| 2 |
> Brian Harring wrote: |
| 3 |
> > Semantics of USE=-gtk not working on a package that has gtk forced |
| 4 |
> > doesn't sound all that nice btw; |
| 5 |
> |
| 6 |
> Which is why the flag shouldn't be forced unless it's almost |
| 7 |
> certain that the flag shouldn't be disabled. The gtk flag might not |
| 8 |
> ever fall into that category, but something like cxx might (nocxx |
| 9 |
> inverted). |
| 10 |
|
| 11 |
Flag shouldn't be forced, period imo. |
| 12 |
|
| 13 |
A default use configuration of a package is fine, but it should be |
| 14 |
mutable via the normal methods imo (potential exception being USE="-*" |
| 15 |
handling since that one is the oddball). |
| 16 |
|
| 17 |
Again, this should involve the dev community (both package.use.mask |
| 18 |
implementation and use.force)- it needs a fair amount of thought put |
| 19 |
into it, and agreement from folks over it- otherwise y'all wind up |
| 20 |
reinventing the wheel without actually picking up the relevant points |
| 21 |
from past discussions of this. |
| 22 |
|
| 23 |
Upshot of involving more folks in the discussion, people would point |
| 24 |
out things like "how are you going to represent this in emerge -vp |
| 25 |
output?" |
| 26 |
|
| 27 |
Cause... package.use.mask ain't represented in any shape/form right |
| 28 |
now :) |
| 29 |
|
| 30 |
~harring |
Archives