Update to mythtv-core.25

[Jeremy Huddleston Sequoia]

On Oct 1, 2012, at 6:06 PM, Craig Treleaven <ctreleaven at cogeco.ca> wrote:

> At 12:29 PM -0700 9/30/12, Jeremy Huddleston Sequoia wrote:
>> On Sep 30, 2012, at 9:06 AM, Bradley Giesbrecht <brad at pixilla.com> wrote:
>> 
>> >> I want to fallback to macports-clang-3.0 on XCode 4.4 AND 4.5.  Is that what is happening?
>> Yes.  It should be.
> 
> Jeremy, I believe the mythtv-core.25 port will only work (on XCode 4.5) _IF_ the user already has apple-gcc-4.2 installed.


> 
> I would classify this as a bug in the current implementation of the compiler.fallback list.  MacPorts tries to use the first fallback compiler that isn't blacklisted.  When apple-gcc-4.2 is chosen, there is nothing to ensure that it is installed or added to the dependencies until 32542 is fixed.

Yeah.

> 
> BTW, it seems to me that the compiler.fallback should be an ordered list with the port author in control of first, second and other choices?

No, that's the point of compiler.whitelist.

>  In my case, apple-gcc-4.2 works but results in some (non-fatal) errors at runtime; macports-clang-3.0 generates quieter-running code.
> 
> 
> Bradley, I believe if we add apple-gcc-4.2 to the blacklist, mythtv-core.25 will work predictably again.  Could you please change line 84 from:
>                     dragonegg-3.0 dragonegg-3.1
> to:
>                     dragonegg-3.0 dragonegg-3.1 apple-gcc-4.2
> 
> Thanks,
> 
> Craig