Bryan

I’m still trying to wrap my head around Erlang and functional programming, and I am very much stuck with looping. I have some code in a CRC algorithm that I want to write in Erlang. The C code is as follows:

for(i=0; i<8; i++){
    if( x & 0x0001 == 0x0001)
        x = POLY ^ (x >> 1);
    else
        x = x >> 1;
}

if
   X band 16#0001 == 1 ->
         POLY bxor (X bsr 1);
   X band 16#0001 == 0 ->
         X bsr 1
end.

Thank you very much for walking me through the thought process and explaining the mechanics in such detail. It is all much clearer now, thank you!

I’m very much embarrassed to say this, but I have a syntax problem and I cannot figure out where it is. The code is as follows:

-module(crc16).
-export([calculate/1]).
-export([inner_loop/3]).

%%% CRC16 CCITT Normalized Polynomial
-define(POLYNOMIAL, 16#1021).

inner_loop(0, X, _) -> X;    
inner_loop(N, X, F) -> inner_loop(N-1, F(X), F).

calculate(X) ->
    inner_loop(8, X, fun(X) ->
        if X band 16#0001 == 1 -> ?POLYNOMIAL bxor (X bsr 1);
           X band 16#0001 == 0 -> X bsr 1
        end
        ).

%%% outer_loop() ->
%%% for(i=0; i<256; i++) { 
%%%    // call inner_loop to calculate value
%%%    // add calculated value to crc lookup table at index i
%%% }

The output of the Erlang interpreter is as follows:

38> c(crc16).
./crc16.erl:16: syntax error before: ')'
./crc16.erl:2: function calculate/1 undefined
error

I would greatly appreciate any hints. I’m sure I’ve made a very elementary mistake but I just can’t seem to find it.

Thank you for your time and help!

Thank you very much for your help. It sure was the “end” for the fun that I missed.

Is there an easy way to prove that a given solution works? Being new to Erlang I’m not sure how I’d generate a testbench that would prove a solution works.

Alain,

Thanks for sharing your solution, it is very elegant. I hope eventually I’ll be able to adjust my “gears” to functional thinking.