| 1 |
>>>>> On Wed, 30 Aug 2017, Michał Górny wrote: |
| 2 |
|
| 3 |
> I see no reason to make Portage less strict than it is now. There |
| 4 |
> is no valid use case for this, and the PMS behavior is just plain |
| 5 |
> stupid. Reverting the change will not bring any clear gain, and will |
| 6 |
> only surprise people who actually hit the case. |
| 7 |
|
| 8 |
Portage would not be compliant with the new wording either, though: |
| 9 |
|
| 10 |
RDEPEND="|| ( |
| 11 |
foo? ( cat/foo ) |
| 12 |
bar? ( cat/bar ) |
| 13 |
)" |
| 14 |
|
| 15 |
with USE="-foo -bar" evaluates to true in Portage, while under the |
| 16 |
new PMS wording the empty group should evaluate to false. |
| 17 |
|
| 18 |
Ulrich |
Archives