reworked AudioBuffer

[LeanderK]

Hi,
i tested RPlay on the Raspberry pi and noticed that the Audiobuffer was a common source for failure.
The problem was related to how the Audiobuffer calculated the buffered elements to restart getNextFrame() after an underrun. Also the performance was not very good because of some inefficient locks.
I also entangled the Source code a bit and converted the AudioData to a "right" class instead of some proved C construct.
It now performs much better on the raspberry pi. :)

[bencall]

Hi,

Thanks for improving RPlay!
I developed it long time ago and, indeed, my code isn't really efficient.

However, there seems to be a problem with your code.
I get this error message:
"...
Waiting
Exception in thread "Thread-18" java.lang.NullPointerException
at AudioBuffer.calculateActualBufferSize(AudioBuffer.java:178)
at AudioBuffer.putPacketInBuffer(AudioBuffer.java:135)
at AudioServer.packetReceived(AudioServer.java:120)
at UDPListener.run(UDPListener.java:35)"

It looks like "audioBuffer[i]" returns null.
I'll try to look into it tomorrow ;-)

[LeanderK]

i have made some other changes...maybe i forgot to copy something