[mythtv] [mythtv-commits] mythtv commit: r10201 - in trunk/mythtv by danielk

[Martin Ebourne]

On Wed, 14 Jun 2006 17:50:02 -0400, Daniel Kristjansson wrote:
> On Wed, 2006-06-14 at 17:41 -0400, Isaac Richards wrote:
> 
>> > Closes #1598. Adds AC3/DTS passthough device setting. This is needed
>> > because some hardware doesn't detect the output type or uses different
>> > audio connectors for different types of audio. By default the old
>> > behaviour of sending all output to the same device is used.
>> Why not add a new constructor, so every single audio class/user of audio
>> (you missed a few, I believe) wouldn't have had to change?  Same thing
>> with the passthrough variable that was added before, both changes
>> could've been done with much lower impact on existing code..
> 
> I was going on the theory that every audio output method would eventually
> have something like AC3/DTS passthrough. But I'll undo this change +
> passthrough variable change outside ALSA, if you think that is unlikely.

I would have thought that would be the case with at least some of the
other sound outputs. JACK would be one I'd expect to need a separate
passthru setting but I didn't add that code because I don't have a JACK
setup. It's something I do intend to try at some point though.

The idea of a separate constructor for passthrough sounds ok. I presume it
was suggested just for the AudioOutput class not the whole hierarchy or
there'd be a lot of duplicate code resulting. I'm not sure if that's
better now given that they've already been changed, but I'm happy to write
a patch for that if you think it's still worth it.

I personally don't like the idea of overloading the single string with
different values separated by a comma. I would think the only reason to do
that would be if it was really not possible to make the more involved
change, which just doesn't seem to be the case here.

Cheers,

Martin.

PS. Sorry for forgetting mfd - I really must play with that sometime.