rpms/poker-eval/FC-4 poker-eval.spec,1.8,1.9

[Panu Matilainen]

On Sat, 10 Jun 2006, Denis Leroy wrote:

> Panu Matilainen wrote:
>> How about replacing the %makeinstall macro definition with something useful 
>> instead? "There's this convenient macro but you shouldn't use it" seems 
>> silly to me.
>
> Thank you, that's exactly what I was thinking. How about updating all those 
> nice macros with what we should be using ? If the old behavior is still 
> useful in some cases, let's identify those cases and rename the old 
> %makeinstall to something more excplit.

Yeah, %makebrokeninstall comes to mind.

> We're fighting against something (the RPM macro system) that's supposed to be 
> our friend! Same thing with the common desktop file install entries etc, we 
> could define macros for those and update all the SPEC files in CVS.

Indeed. RPM packaging is full of crazy little details such as this 
you're supposed to somehow know and remember.

Another example that comes to mind is the "make %{?_smp_mflags}" 
construct. Why is that not defined as %make, as it's supposed to be the 
default, right? So instead of having that conditional macro stuff in each 
and every spec, those broken packages could set RPM_BUILD_NRCPUS=1 or 
similar which would make it very clear that non-paraller build is used 
intentionally.

 	- Panu -