| 1 |
-----BEGIN PGP SIGNED MESSAGE----- |
| 2 |
Hash: SHA1 |
| 3 |
|
| 4 |
Hi all, |
| 5 |
|
| 6 |
I had another crazy idea that I'd like some feedback on. I'll jump straight |
| 7 |
in. With the current virtuals system, there are two basic deficiencies: |
| 8 |
|
| 9 |
1. Packages that can provide a virtual based on USE flags can't. (#32114) |
| 10 |
2. Different versions of a virtual are not supported. (#46968) |
| 11 |
|
| 12 |
The proposal for 1 is to have PROVIDE use a syntax similar to DEPEND for |
| 13 |
specifying any USE flag relationship. There is currently no proposal for 2. |
| 14 |
|
| 15 |
What I'm proposing here is to drop virtuals altogether. Instead, I propose |
| 16 |
using meta-packages similar to kde and gnome. To give a give a cut down |
| 17 |
example: |
| 18 |
|
| 19 |
virtual/x11-1.ebuild: |
| 20 |
RDEPEND="x86? ( || ( x11-base/xfree x11-base/xorg-x11 ) ) |
| 21 |
sparc? ( || ( x11-base/xorg-x11 x11-base/xfree ) )" |
| 22 |
|
| 23 |
The -1 in x11-1 is an arbitrary version. This example shows the only |
| 24 |
deficiency of doing it this way that I am able to think of. Instead of |
| 25 |
specifying the default virtuals per profile, they would only be able to be |
| 26 |
specified per architecture. |
| 27 |
|
| 28 |
Now to an example that shows a resolution to issue 2 above: |
| 29 |
|
| 30 |
virtual/jdk-1.4.ebuild: |
| 31 |
RDEPEND="|| ( >=dev-java/blackdown-jdk-1.4 >=dev-java/kaffe-1.1.4 ... )" |
| 32 |
|
| 33 |
The different versions aspect is represented here by kaffe-1.1.4. Even it is |
| 34 |
installed it is presently not able to satisfy >=virtual/jdk-1.4 as its |
| 35 |
version component does not match. |
| 36 |
|
| 37 |
As for issue number 1, it will require a (already scheduled) portage |
| 38 |
enhancement: |
| 39 |
|
| 40 |
virtual/login-1.ebuild: |
| 41 |
RDEPEND="|| sys-apps/shadow[pam] sys-apps/pam-login[-pam]" |
| 42 |
|
| 43 |
Here, shadow with USE="pam" or pam-login with USE="-pam" are supported. |
| 44 |
|
| 45 |
As a side note, a single atom is chosen || () dependencies. If something |
| 46 |
installed satisfied any of the atoms, then the installed package is used. |
| 47 |
Otherwise the first package in the list is used. |
| 48 |
|
| 49 |
Finally, there are two additional benefits. One is a slight speed increase as |
| 50 |
portage would no longer have to concern itself with virtuals. Second, is that |
| 51 |
it would be possible to find all packages that satisify a virtual without |
| 52 |
scanning the entire tree for PROVIDEs. |
| 53 |
|
| 54 |
Do I burn in hell? (Again? ;) |
| 55 |
|
| 56 |
Regards, |
| 57 |
Jason Stubbs |
| 58 |
-----BEGIN PGP SIGNATURE----- |
| 59 |
Version: GnuPG v1.2.4 (GNU/Linux) |
| 60 |
|
| 61 |
iQCVAwUBQNQo2FoikN4/5jfsAQL9sQP/UBzFzWlQJ/VkUx+aWGMX505okHhT3XWf |
| 62 |
2AxL1y6uTaFdOD4i5kVtzfB+tQ9IOLsVHygIHZS9rFSTC95bvUAX7/yO91BgzIkL |
| 63 |
zdI9gKquLgLRexCpBZ5qMLi5Bfruv3bWuCbnU3gTFIjhUSHT2Xs6oHm7jViwTk1W |
| 64 |
X9uje3B4xPQ= |
| 65 |
=y7Uj |
| 66 |
-----END PGP SIGNATURE----- |
| 67 |
|
| 68 |
-- |
| 69 |
gentoo-dev@g.o mailing list |
Archives