Re: Subclasses, tied hashes, {get,set}_property, undefined values

[Torsten Schoenfeld <kaffeetisch gmx de>]

On Sun, 2007-11-18 at 15:13 -0500, muppet wrote:

> > So, does this make sense?  Is the benchmark sound, or are there cases
> > which are slowed down measurably by this patch?  Is the
> > name gperl_sv_defined good, or are there better alternatives?
>
> Sounds like a good idea to me.  I'm sure Gtk2 and friends contain lots  
> more places where this would be necessary...  I'm not really concerned  
> by how much this might slow things down, because any performance gain  
> we were getting was based on not doing everything we should've been  
> doing.

Alright, committed.  I'll fix Gtk2 soon.

-- 
Bye,
-Torsten