| 1 |
On Fri, 2012-07-20 at 18:54 +0100, Ciaran McCreesh wrote: |
| 2 |
> On Fri, 20 Jul 2012 13:43:15 -0400 |
| 3 |
> Alexandre Rostovtsev <tetromino@g.o> wrote: |
| 4 |
> > > If you dep upon foo[linguas_en(+)] and linguas_en isn't in IUSE, |
| 5 |
> > > what happens? |
| 6 |
> > |
| 7 |
> > Fatal error. If a package installs its translations implicitly via |
| 8 |
> > gettext's rules depending on the value of LINGUAS at configure time, |
| 9 |
> > then obviously other packages must rely on that package having |
| 10 |
> > installed any particular translation. |
| 11 |
> |
| 12 |
> Uh, the entire point of the (+) is that it's *not* a fatal error if you |
| 13 |
> have a default. |
| 14 |
|
| 15 |
That suggests that the EAPI ought to define a second category of |
| 16 |
USE_EXPAND flags, one that has a different treatment of (+)/(-). |
| 17 |
|
| 18 |
Something like the following: |
| 19 |
|
| 20 |
A dependency on $foo[linguas_bar(+)] would be considered satisfied by an |
| 21 |
ebuild X matching $foo iff: |
| 22 |
1. X has linguas_bar in IUSE and enabled; or |
| 23 |
2. X does not have linguas_bar in IUSE, but there exists an ebuild Y |
| 24 |
(which may or may not equal X) matching $foo such that Y has at least |
| 25 |
one linguas_* flag in IUSE. |
| 26 |
|
| 27 |
-Alexandre. |
Archives