| 1 |
On Thu, 1 Apr 2010 22:53:10 +0530 |
| 2 |
Nirbheek Chauhan <nirbheek@g.o> wrote: |
| 3 |
> On a related note, I really like Zac's and solar's no-nonsense |
| 4 |
> get-stuff-done-even-if-it-isn't-perfect attitude, and would love it if |
| 5 |
> everyone else applied it as well (if they don't already). I don't care |
| 6 |
> if the proposal is perfect; as a potential user of those features, I |
| 7 |
> want it to be implemented in portage in a reasonable time-frame. |
| 8 |
> |
| 9 |
> Over-engineering and then designing something to death is not the way |
| 10 |
> to deliver said feature to the user. It's really stupid when the |
| 11 |
> design document gets more attention than the implementation used by |
| 12 |
> 90% of our users. |
| 13 |
|
| 14 |
In that case, why would you like to see VALID_USE as well as |
| 15 |
pkg_pretend? Why not just use pkg_pretend, which is already in EAPI 4 |
| 16 |
and which can do everything VALID_USE can do plus several useful things |
| 17 |
that it can't? |
| 18 |
|
| 19 |
With pkg_pretend you can do VALID_USE using a simple eclass... |
| 20 |
|
| 21 |
-- |
| 22 |
Ciaran McCreesh |
Archives