1 |
On Monday 01 November 2004 06:23, Henrik Brix Andersen wrote: |
2 |
> Yeah, I was uncertain about kernel-mod_src_compile() as well. Just |
3 |
> didn't want to take it out since sys-apps/cloop depends on it. We could |
4 |
> leave it out and fix sys-apps/cloop accordingly? |
5 |
|
6 |
Sounds good, but you should check it with the cloop maintainer first. Also ask |
7 |
about the zlib stuff, according to grep, cloop is the only ebuild that uses |
8 |
it. I think we should move the zlib stuff into the cloop ebuild and take it |
9 |
out of the eclass. |
10 |
|
11 |
> Yes, that's why I included kernel-mod_need_subdir_to_m(). Ebuilds which |
12 |
> require more than the simple sed statement can do: |
13 |
> |
14 |
> if kernel-mod_needs_subdir_to_m |
15 |
> then |
16 |
> ... |
17 |
> fi |
18 |
> |
19 |
> Will this do? |
20 |
|
21 |
Perhaps some comments around the _subdir_to_m function would be appropriate - |
22 |
but other than that, I don't think theres much we can do. |
23 |
|
24 |
> Many ebuilds rely on having more than ${KV} available. |
25 |
> kernel-info_getversion() provides ${KV_MAJOR}, ${KV_MINOR} and |
26 |
> ${KV_PATCH}. |
27 |
|
28 |
Ok, how would you feel about removing the ${KV} stuff from portage.py and just |
29 |
making the ebuilds that currently use it convert to using kernel-info? Or, we |
30 |
could easily extend portage to set KV_MAJOR etc. But what I'm getting at is |
31 |
that we should only do this in one place - and where should it be? My view is |
32 |
that it should be done in the eclass. |
33 |
|
34 |
Thanks, |
35 |
|
36 |
Daniel |
37 |
|
38 |
-- |
39 |
gentoo-dev@g.o mailing list |