Use an adapter instead of the client-stream. This means we run
the audioconverters and resamplers in the client instead of the
pipewire daemon. It also allows us to implement the audio mixing
correctly in the capture client.
The only pending piece is that we now wake up the client with the
period of the server. Maybe we can later optimize that and
accumulate/split buffers before waking the client.
This probably needs fixing with video..
Run the a2dp sink and sources with the adapter in the session manager
and export it to PipeWire. The idea is that the codecs should not
preferably not run inside the daemon.
Use a new rate_match io area to exhange rate matching info between
sink/source and resampler.
Compensate for the rate match delay when scheduling timeouts.
Let the resampler notify the source of how many samples it needs to
produce the desired quantum. Make sure we keep an extra buffer in
the device to be able to make this possible.
Let the adapter directly call the slave node process function.
An adapter is like an audio-dsp node and client-stream combined.
It allows tighter control with the device (for rate control and
variable buffer sizes), software volume.
The idea is to also implement the client-stream with this
eventually.
Don't let permissions of an object influence the child permissions,
this is not so easy to manage and needs some more thought.
The result is now that it's possible to see a reference to a (parent)
id that is not visible, but that's ok.
Define a set of standard factory names and document what they
contain. This makes it possible to change the implementation by
mapping the factory-name to a different shared library.
Make a set-prop command to set a property from the config file
into a pw_properties. Pass this to the pw_core_new() and the
main-loop to tweak some stuff.
Move some warns to errors
Make pw_remote_connect_fd() not automatically close the provided
fd but let the caller take care of that. This allows us to reuse
the fd in pipewiresrc.
Fixes#155
