| 1 |
On 14:43 Wed 23 Mar , Rich Freeman wrote: |
| 2 |
> On Wed, Mar 23, 2011 at 1:44 PM, Michael Seifert |
| 3 |
> <michael.seifert@×××.net> wrote: |
| 4 |
> > I cannot tell you at this stage, sorry. The real loss for steps 3/4 |
| 5 |
> > has to be measured after the implementation. The problem with the |
| 6 |
> > estimates here is that the assembled ebuilds also contain the |
| 7 |
> > sources and the eclasses. Maybe I will do some number crunching on a |
| 8 |
> > few selected ebuilds. |
| 9 |
> |
| 10 |
> A more critical factor could be the dependencies - unless we otherwise |
| 11 |
> cache them. To install a package you need to walk the dependency tree |
| 12 |
> (well, at least until you hit installed packages with the right USE |
| 13 |
> flags). That requires one set of fetches for each level you traverse, |
| 14 |
> and that means at least one round trip per level. |
| 15 |
|
| 16 |
Perhaps both of you need to take a quick stroll through the metadata |
| 17 |
cache (/usr/portage/metadata/cache/) and check out what those files look |
| 18 |
like. =) |
| 19 |
|
| 20 |
Here's a description of the format from our Package Manager |
| 21 |
Specification (PMS): |
| 22 |
|
| 23 |
http://dev.gentoo.org/~ulm/pms/4/pms.html#x1-16000014 |
| 24 |
|
| 25 |
-- |
| 26 |
Thanks, |
| 27 |
Donnie |
| 28 |
|
| 29 |
Donnie Berkholz |
| 30 |
Admin, Summer of Code |
| 31 |
Gentoo Linux |
| 32 |
Blog: http://dberkholz.com |
Archives