[gentoo-dev] Re: cmake-utils.eclass dropping EAPI 0/1 support

[Michael Palimaka]

On 14/06/2013 09:05, Alexis Ballier wrote:
> On Thu, 13 Jun 2013 18:48:21 -0400
> Chris Reffett <creffett@g.o> wrote:
>
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> On 06/13/2013 06:37 PM, Alexis Ballier wrote:
>>>> At the beginning of July, the KDE team will be removing EAPI 0/1
>>>>   support from cmake-utils.eclass and inlining the functions from
>>>>   base.eclass in order to remove that inherit [1].
>>>
>>> So, instead of fixing what you consider wrong in base.eclass, you
>>> inline it so that if someone improves base.eclass he has to do it
>>> for cmake-utils too?
>>>
>> We did not actually inline most of the complicated logic from
>> base.eclass, as to the best of my knowledge epatch itself will handle
>> all of the corner cases that base_src_prepare covers. The new patching
>> code essentially consists of [[ ${PATCHES[@]} ]] && epatch
>> "${PATCHES[@]}"; epatch_user.
>
> that kind of stuff sounds more like it should be factorized rather than
> copied all around; be it base.eclass, an EAPI, or another eclass I
> don't really care.

The code literally is '[[ ${PATCHES[@]} ]] && epatch "${PATCHES[@]}"'. 
Given that the actual epatch logic is in one place, I am not sure how 
much of an issue this really is. I think there's a proposal to put 
epatch into PMS too.

Best regards,
Michael