1 |
El vie, 21-09-2012 a las 12:45 -0700, Zac Medico escribió: |
2 |
> On 09/21/2012 12:08 PM, Pacho Ramos wrote: |
3 |
> > Hello |
4 |
> > |
5 |
> > This comes from this gentoo-dev thread: |
6 |
> > http://www.gossamer-threads.com/lists/gentoo/dev/260536 |
7 |
> > |
8 |
> > In that one, we try to use the following: |
9 |
> > has vala ${IUSE//+/} && ! use vala && return 0 |
10 |
> > |
11 |
> > as already done in many eclasses/ebuilds. The problem is that Ciaran |
12 |
> > wants to forbid it because he says it's not specified in PMS. My |
13 |
> > suggestion was to simply specify it as it's currently implemented in |
14 |
> > portage because that functionality is (apart of needed) being used for a |
15 |
> > long time in the tree by numerous eclasses/ebuilds, then, from my point |
16 |
> > of view, wouldn't be any sense on lose time for moving them to current |
17 |
> > functionality to a worse one, wait for the next eapi and, finally, |
18 |
> > revert them back to current behavior. |
19 |
> > |
20 |
> > The problem is that I cannot find any doc about how this is currently |
21 |
> > handled in portage. Could you help me on it please? |
22 |
> |
23 |
> That `has vala ${IUSE//+/}` thing should work for all versions of |
24 |
> portage that have existed since PMS came around. The way that it works |
25 |
> is that that ebuild.sh sources the ebuild, and the inherit function |
26 |
> makes temporary backups of IUSE so that the IUSE settings from all of |
27 |
> the eclasses and the ebuild can be stacked together after the sourcing |
28 |
> is complete. The stacked IUSE value that's generated then remains in the |
29 |
> ebuild's environment for all phases. |
30 |
|
31 |
Thanks a lot, do you let me send your reply to gentoo-dev ML? |