yeah, was chatting with Jonathan Corgan at the 2014 GNU Radio Conference, and he also
pointed out that OP25 doesn't play very well with GRC. Would be nice to have access
to unlimited free time to work on this :)
My own viewpoint is the existing OP25 GR blocks are not the right level to target for GRC.
Instead we need to wrap them into some new python hier blocks (TX&RX chains,
vocoders, monitoring sinks etc.) using standard existing GR interface conventions that
would in turn be exposed via XML files to GRC. Also a lot of the code in scope.py needs
to be broken up along the same lines... Currently the way that it switches between the
two major P25 demods (FSK4 vs. pi/4 DQPSK) is less than ideal, but it should be more or
less correct (given the limits of what's currently known about correct P25
demodulation).
Can't comment on gr-dsd as I've not studied it. As far as the question about the
op25 vs. op25_repeater code the answer would depend entirely on your definition of
"better" :)
Best
Max