[gst-devel] GStreamer status, 20 Sept 2005

[Andy Wingo]

Hey Dave,

On Tue, 2005-09-20 at 23:10 -0700, David Schleef wrote:
> I don't see much advantage to having two possible code paths here.

Much easier on bindings -- most language bindings aren't prepared to
have different ref/unref functions for different packages.

> all referencing/unreffing
> must still go through gst_object_[un]ref()

If you're supporting old GLib, yes..

> It almost seems to me to be inviting someone to write
> a sloppy gstreamer-based app that works with glib-2.8, but fails/leaks
> on <= 2.6.

Hm, yes. I would turn it around though -- in a year or two we'll
probably end up requiring a more recent GLib. At that point we'll
consider what we currently have for what it is (a hack), and remove it. 

Also, it's a bit dishonest to push for atomic refcounts in GLib and then
not use/test them.

I could be off my rocker though ;)
-- 
Andy Wingo
http://wingolog.org/