FESCo Meeting Summary for 2007-02-08

[Panu Matilainen]

On Thu, 15 Feb 2007, Matthias Saou wrote:

> Ralf Corsepius wrote :
>
>> On Thu, 2007-02-15 at 12:06 +0100, Florian La Roche wrote:
>>>> My packages under review are currently blocked because of this useless
>>>> guideline change :-(
>>>
>>> I don't think it makes sense to block reviews for the BuildRoot
>>> settings. We have enough other things to look at...
>>
>> Why isn't anybody at redhat able to run a sed script over all *.spec
>> and thereby closing this case for all packages at once?
>
> Because we might want to do that only once, if/when we find a better
> Buildroot value?

Somebody suggested doing something like this in specs:

BuildRoot: %{rpmbuildroot}

..and make that mandatory. Put out a new version of redhat-rpm-config with 
that defined to use whatever uid-arch-mktempd-junk people see necessary. 
If I want to have consistent buildroots, I redefine that macro locally, 
otherwise it'll just work. And it's a single sed-job that only needs to be 
done once.

I'd still prefer being able ditch this useless junk from every spec 
(Matthias had various other items on the hitlist) completely, but the 
above is far better than this endless arguing (and for heavens sake, 
blocking packages) based on something that's completely irrelevant to the 
buildsystem.

 	- Panu -