Moving Sound into Core

[Casey Harkins]

Sean Egan wrote:
> On 6/15/07, Eric Polino <aluink at gmail.com> wrote:
>> I'm not sure
>> exactly how this would work as I am new to a lot of this, but it seems to
>> make sense to me at a superficial level.  I'm starting up this thread to get
>> a feel for what people think about moving sound into the core.
> 
> Sound is fundamentally a UI feature. There is nothing core about it.
> Besides the obvious issue that Windows uses different sound than
> Linux, which is different than OS X, there are plenty of ways that
> sound ties into the UI that makes it inappropriate for the core: the
> 'sounds only when window does not have focus' option would not
> translate well to Finch. Message queuing wants to control when it
> makes sounds itself.

I agree completely. In fact, currently there *is* sound code in core, 
without any concrete implementation in core (it uses ui_ops). I think 
this could potentially be removed. The only down side to this, is that 
non-ui specific plugins could not play sounds (though they could 
possibly emit-signals that might trigger sounds).

Sound events now react entirely to signals and should be trivial to 
re-implement in any other clients. Where sharing code might be more 
useful is in the actual sound playing code from gtksound.c (especially 
the gstreamer stuff). This could potentially be moved into a separate 
file that could be included into a finch plugin (through some sort of 
mtn magic). This would limit some of the code duplication.

-casey