| 1 |
On Sat, 2003-12-06 at 07:26, Paul de Vrieze wrote: |
| 2 |
> We need indeed a highlevel abstraction, and dep trackers are one of the |
| 3 |
> modules. I think that access to the package tree is another, where caching |
| 4 |
> can be modular too. |
| 5 |
|
| 6 |
If by "caching" you mean the metadata cache, this is something I want to |
| 7 |
eliminate in portage-ng. I would like things to be designed to be fast |
| 8 |
from the start, with no slow bash<->python linkage like there is in the |
| 9 |
current portage that makes us require a metadata cache for decent |
| 10 |
performance. |
| 11 |
|
| 12 |
It should be possible to get portage-ng without caching running as fast |
| 13 |
as portage does now when it has a fully up-to-date cache. Then if we |
| 14 |
need more performance, portage-ng's datastore can be moved to a |
| 15 |
database, or we can add an enhanced caching mode to make it even faster. |
| 16 |
|
| 17 |
For backwards compatibility with existing ebuilds, yes we will probably |
| 18 |
still need the metadata cache since we'll still have some kind of bash |
| 19 |
linkage. It's important to point out that the design of portage-ng will |
| 20 |
not be tied to ebuilds. Ebuilds will likely become "legacy" build |
| 21 |
scripts that are superceded by something a lot better, cleaner, powerful |
| 22 |
and also faster for portage-ng. |
| 23 |
|
| 24 |
Regards, |
| 25 |
|
| 26 |
Daniel |
Archives