RS5kM - Choke regression defect in 4.61?

[rjattrill]

Hi MPL,

I love the ongoing improvement to RS5km.

Maybe I am misunderstanding the intended operation but I feel that there is a regression defect in choking in version 4.61.

I modified the mpl_RS5K_manager_MIDIBUS_choke.jsfx as indicated below and it worked again for me.

desc:RS5K_manager_MIDIBUS_choke
noindex: true
 
options:gmem=RS5K_manager_choke

slider1:0<0,127,1>src
slider2:0<0,127,1>dest

@init


@block

  src = slider1;
  dest = slider2;
  
  NOTE_ON = $x90;
  NOTE_OFF = $x80;
 
  offset = 0;
  while (midirecv(offset,msg1,msg2,msg3)) (

       msgchan = msg1 & $xF;
       
       current_note = msg2;
       note_status = msg1 & $xF0;
         
       note_status == NOTE_ON && current_note == src ? (
          // NOTE_ON on src means send NOTE_OFF to dest
          midisend(offset, NOTE_OFF|msgchan, dest);
          // Passthrough the original src message
          midisend(offset, msg1, msg2, msg3);
          ) : (
            // Passthrough every other message EXCEPT NOTE_OFF on dest
            note_status == NOTE_OFF && current_note == dest ? no_op : midisend(offset,msg1,msg2,msg3);
       )
    );