1 |
-----BEGIN PGP SIGNED MESSAGE----- |
2 |
Hash: SHA1 |
3 |
|
4 |
Petteri Räty wrote: |
5 |
> so only three virtual ebuilds have a DEPEND |
6 |
|
7 |
According to GLEP 37 [1], they should only define RDEPEND. The |
8 |
reason that only RDEPEND is needed is that a package that has a |
9 |
virtual dependency has freedom to include the virtual atom in any of |
10 |
DEPEND, RDEPEND, and PDEPEND as necessary. Note that when portage's |
11 |
$ROOT feature is used, DEPEND is installed to the build platform. |
12 |
The virtual ebuild itself does not make this decision, but rather |
13 |
the package that has a virtual dependency. |
14 |
|
15 |
> So it seems most ebuilds for new style virtuals don't have DEPENDs. |
16 |
> Anyone have any idea on how these are supposed to work or are they |
17 |
> buggy? PMS currently says that RDEPEND doesn't have to be installed at |
18 |
> the time a package is emerged so as far as I understand it this means |
19 |
> that DEPEND="virtual/foobar" could mean that it could happen that |
20 |
> nothing providing the virtual is available. |
21 |
|
22 |
The bit about "RDEPEND doesn't have to be installed" is only for |
23 |
solving circular RDEPEND during the installation process. When a |
24 |
package has RDEPEND that is not installed, it must be considered |
25 |
unusable and therefore this state should only exist for a relatively |
26 |
short period of time during the installation process. |
27 |
|
28 |
> So should we be changing the |
29 |
> DEPEND atoms to DEPEND="${RDEPEND}" or is there something I am missing here? |
30 |
|
31 |
We should not, as I've explained above. |
32 |
|
33 |
Zac |
34 |
|
35 |
[1] http://www.gentoo.org/proj/en/glep/glep-0037.html |
36 |
-----BEGIN PGP SIGNATURE----- |
37 |
Version: GnuPG v2.0.3 (GNU/Linux) |
38 |
|
39 |
iD8DBQFGH/fs/ejvha5XGaMRAhZoAJwPH7W04pL8EfkJ4STM+79Drl0jcQCdGjGz |
40 |
n4Ee8wfFp3RgpKy25ftjoXo= |
41 |
=TBuW |
42 |
-----END PGP SIGNATURE----- |
43 |
-- |
44 |
gentoo-dev@g.o mailing list |