Help with NEC Command/Response

Comments

• ericmedley Posts: 4,177

  July 2013

  Think of it in terms of it being a binary number
  
  Off message data byte is
  00000000 which is represented as $00
  
  On message is
  00000010 which is $02 in hex.
  
  I'd guess that the cooling message might look like
  00010000 which would be $10 in hex.

  0
• jabramson Posts: 106

  July 2013

  Got it, thanks

  0
• JasonS Posts: 229

  July 2013

  Bit masks are the way to get individual bit values of a byte or word. So if you were interested in the state of the cooling bit and none of the ohers, you would do a bitwise and (&) of Data01 with $10.
  
  01111010 - random value for Data01
  00010000 - $10

  ---

  00010000 - result of bitwise and
  
  when you do the bitwise and only the bits that are 1 in the mask and Data01 are 1 in the result. It is then easiest to treat your result as a boolean value. Here are the functions that I use when I need to check or set the state of a bit in a CHAR.

  DEFINE_CONSTANT
  //SET_cBit_State() & GET_cBit_State() Functions Constants
  CHAR cBitMasks[] = {$01, $02, $04, $08, $10, $20, $40, $80}
  
  DEFINE_FUNCTION INTEGER GET_cBit_State(CHAR Byte, INTEGER Passed_Bit, INTEGER Bit_Zero)
  {
      STACK_VAR CHAR bytResult
      STACK_VAR INTEGER iBit
      iBit = Passed_Bit
      IF (Bit_Zero)
      {
  	iBit = iBit + 1
      }
      IF (iBit > 0 AND iBit <= 8)
      {
  	bytResult = Byte BAND cBitMasks[iBit]
  	IF (bytResult > 0)
  	{
  	    RETURN TRUE
  	}
      }
      RETURN FALSE
  }
  
  DEFINE_FUNCTION SET_cBit_State(CHAR Byte, INTEGER Passed_Bit, INTEGER State, INTEGER Bit_Zero)
  {
      STACK_VAR INTEGER iBit
      STACK_VAR CHAR InvBitMask
      
      iBit = Passed_Bit
      IF (Bit_Zero)
      {
  	iBit = iBit + 1
      }
      IF (iBit > 0 AND iBit <= 8)
      {
  	IF (State)
  	{
  	    IF (!GET_cBit_State(Byte, Passed_Bit, Bit_Zero))
  	    {
  		Byte = Byte + cBitMasks[iBit]
  	    }
  	}
  	ELSE
  	{
  	    IF (GET_cBit_State(Byte, Passed_Bit, Bit_Zero))
  	    {
  		Byte = Byte - cBitMasks[iBit]
  	    }
  	}
      }
  }
  

  
  Set Bit_Zero to true if you are using 0-7 to identify your bits intstead of 1-8.

  0
• Joe Hebert Posts: 2,159

  July 2013

  Shifting gears
  
  Another option is this 1 line function to check bits 0-15 using bit shifting to create the mask:

  DEFINE_FUNCTION CHAR isBitSet (integer number, char bit) {
       RETURN number & (1<<bit) = 1<<bit
  }
  

  0
• JasonS Posts: 229

  July 2013

  Joe Hebert wrote: »

  Another option is this 1 line function to check bits 0-15 using bit shifting to create the mask:

  DEFINE_FUNCTION CHAR isBitSet (integer number, char bit) {
       RETURN number & (1<<bit) = 1<<bit
  }
  

  
  Nice!
  
  
  Stupid 10 character minimum to post!

  0
• PhreaK Posts: 966

  July 2013

  As it's only a single byte that you're checking you could drop that back to a char. You can also remove the equivalency check as anything non-zero will evaluate to true.

  define_function char isBitSet(char number, char bit) {
       return number & (1<<bit)
  }
  

  
  A super common way to do this though is just set up some constants for the different bit masks that your protocol uses:
  

  define_variable
  
  // bit masks for NEC projector status
  constant char PROJ_POWER = $2
  constant char PROJ_AUTO_OFF = $8
  constant char PROJ_EXT_CONTROL = $10
  constant char PROJ_COOLING = $20
  ...
  

  
  You can then just grab your DATA01 byte (after verifying the checksum), lets call this 'projectorStatus' then check status in a nice readable way.
  

  projector.isOn = projectorStatus & PROJ_POWER
  projector.autoOff = projectorStatus & PROJ_AUTO_OFF
  ...
  

  0
• Joe Hebert Posts: 2,159

  July 2013

  PhreaK wrote:

  As it's only a single byte that you're checking you could drop that back to a char.

  
  As stated above the generic function checks bits 0-15 which is why it it’s an integer but if you want to limit the function to only be capable of checking bits 0-7 then that’s the way to do it.
  

  PhreaK wrote:

  You can also remove the equivalency check as anything non-zero will evaluate to true.

  
  Doing so changes the RETURN value. The function is intended to RETURN a 0 (false) if the bit is not set and RETURN a 1 (true) if the bit is set. The function will not do that with your modification. You’ll also need to deal with the warning when you compile it.

  0