| 1 |
-----BEGIN PGP SIGNED MESSAGE----- |
| 2 |
Hash: SHA256 |
| 3 |
|
| 4 |
On 23/03/12 12:32 PM, Ian Stakenvicius wrote: |
| 5 |
> On 23/03/12 12:19 PM, Ciaran McCreesh wrote: |
| 6 |
>> On Fri, 23 Mar 2012 12:14:39 -0400 Ian Stakenvicius |
| 7 |
>> <axs@g.o> wrote: |
| 8 |
>>> I don't know if I follow this one or not. When inheriting an |
| 9 |
>>> eclass, all entities within the eclass get merged into the |
| 10 |
>>> ebuild. As long as there aren't any special conditional |
| 11 |
>>> tricks being used to assign to global variables like IUSE, it |
| 12 |
>>> would still be invariant wouldn't it? |
| 13 |
> |
| 14 |
>> The point is that the merging might be done inside the package |
| 15 |
>> manager (not in bash code) on the IUSE metadata variable, and |
| 16 |
>> the changes don't have to be reflected in the IUSE environment |
| 17 |
>> variable inside the ebuild. |
| 18 |
> |
| 19 |
> If that was the case, then eclasses could no longer append deps to |
| 20 |
> (R)DEPEND, either .....? |
| 21 |
|
| 22 |
Err, nvm.. i think i'm following the difference now. functions like |
| 23 |
has_version and so forth don't actually work on the value(s) of |
| 24 |
*DEPEND themselves, so in the case of (R)DEPEND it wouldn't matter of |
| 25 |
the package manager didn't expose the merge. |
| 26 |
|
| 27 |
|
| 28 |
-----BEGIN PGP SIGNATURE----- |
| 29 |
Version: GnuPG v2.0.17 (GNU/Linux) |
| 30 |
|
| 31 |
iF4EAREIAAYFAk9spp8ACgkQAJxUfCtlWe1a2wD/YsHDM1hYik+d46fJ90yckI/U |
| 32 |
OKr1ThK6hhJTbjmqGpgBAMekpXzx8NFIPerRPm037FgWQiCuUPDezAhmj8S73EPV |
| 33 |
=CNHZ |
| 34 |
-----END PGP SIGNATURE----- |
Archives