| 1 |
On 09/30/2016 11:58 PM, Andreas K. Huettel wrote: |
| 2 |
> |
| 3 |
> Hey all, |
| 4 |
> |
| 5 |
> during fiddling with mod_perl I stumbled across bug 575252 - in short, |
| 6 |
> depend.apache.eclass doesn't work properly with EAPI=6. |
| 7 |
> |
| 8 |
> So, I'm going to add in a few days a default chunk of code at its start, which |
| 9 |
> will die in EAPI=6. |
| 10 |
> |
| 11 |
> This *will* break the following ebuilds: |
| 12 |
> |
| 13 |
> www-apps/bugzilla-4.4.12 |
| 14 |
> www-apps/bugzilla-5.0.3 |
| 15 |
> www-apache/mod_limitipconn-0.24-r2 |
| 16 |
> www-apache/mod_nss-1.0.12-r1 |
| 17 |
> app-eselect/eselect-php-0.9.1 |
| 18 |
> net-analyzer/pnp4nagios-0.6.25-r3 |
| 19 |
> net-analyzer/icinga2-2.4.10-r1 |
| 20 |
> |
| 21 |
|
| 22 |
Several of these ebuilds are in stable, so I strongly oppose any action |
| 23 |
that breaks existing ebuilds. |
| 24 |
|
| 25 |
I made my thoughts on this clear during the last council meeting, I |
| 26 |
believe the correct action is a pro-active approach where we require |
| 27 |
eclasses to have a defined set of approved EAPIs set and updated by the |
| 28 |
eclass maintainer in order to not require an audit by each ebuild |
| 29 |
maintainer inheriting the eclass. This was discussed on the last council |
| 30 |
meeting but did not get to a vote. |
| 31 |
|
| 32 |
-- |
| 33 |
Kristian Fiskerstrand |
| 34 |
OpenPGP keyblock reachable at hkp://pool.sks-keyservers.net |
| 35 |
fpr:94CB AFDD 3034 5109 5618 35AA 0B7F 8B60 E3ED FAE3 |
Archives