I'm trying to read floating numbers (float 32 bits each) out from an array (that contain only the bits) While I try to do it with ArrayBuffer and DataView I got strange values at the end. I'm expecting to have back 2 floating numbers, but i see that all the numbers are 0.

What am I missing here? I expect 2 floats (firsNum,secNum) each from 4 byes by the order

// Simple array contain 8 bytes
var myBitsArr = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 1, 1, 0, 0, 0, 0, 1, 0];

// Array buffer with size of 8 bytes
const buffer = new ArrayBuffer(8);

// Loop all the data to the array buffer add 8 bits at each time
for (var i = 0; i < myBitsArr.length / 8; i++) {
  var byteCounter = i * 8;
  var b_0 = myBitsArr[byteCounter + 0] << 0;
  var b_1 = myBitsArr[byteCounter + 1] << 1;
  var b_2 = myBitsArr[byteCounter + 2] << 2;
  var b_3 = myBitsArr[byteCounter + 3] << 3;
  var b_4 = myBitsArr[byteCounter + 4] << 4;
  var b_5 = myBitsArr[byteCounter + 5] << 5;
  var b_6 = myBitsArr[byteCounter + 6] << 6;
  var b_7 = myBitsArr[byteCounter + 7] << 7;

  buffer[i] = (b_0 + b_1 + b_2 + b_3 + b_4 + b_5 + b_6 + b_7);
}

console.log(buffer)

// View the floating numbers
var viewNumbers = new DataView(buffer);
var firsNum = viewNumbers.getFloat32(0); // get 0 back and not float number
var secNum = viewNumbers.getFloat32(1); // get 0 back and not float number
console.log(firsNum, secNum)

1 Answers

2
tevemadar On Best Solutions

Besides the remarks: ArrayBuffer is not an array, that is why buffer[i] does not do anything useful. You need a typedarray, most probably an Uint8Array, called uint8 below:

// Simple array contain 8 bytes
var myBitsArr = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 1, 1, 0, 0, 0, 0, 1, 0];

// Array buffer with size of 8 bytes
const buffer = new ArrayBuffer(8);
const uint8 = new Uint8Array(buffer);

// Loop all the data to the array buffer add 8 bits at each time
for (var i = 0; i < myBitsArr.length / 8; i++) {
  var byteCounter = i * 8;
  var b_0 = myBitsArr[byteCounter + 0] << 0;
  var b_1 = myBitsArr[byteCounter + 1] << 1;
  var b_2 = myBitsArr[byteCounter + 2] << 2;
  var b_3 = myBitsArr[byteCounter + 3] << 3;
  var b_4 = myBitsArr[byteCounter + 4] << 4;
  var b_5 = myBitsArr[byteCounter + 5] << 5;
  var b_6 = myBitsArr[byteCounter + 6] << 6;
  var b_7 = myBitsArr[byteCounter + 7] << 7;

  uint8[i] = (b_0 + b_1 + b_2 + b_3 + b_4 + b_5 + b_6 + b_7);
}

console.log(buffer)

// View the floating numbers
var viewNumbers = new DataView(buffer);
var firsNum = viewNumbers.getFloat32(0,true); // get 0 back and not float number
var secNum = viewNumbers.getFloat32(4,true); // get 0 back and not float number
console.log(firsNum, secNum);

Then it does something, displays "4.600743118071239e-41 4.027191656623092e-41" without the true-s (big endian), and "4 240" at the moment (little endian, which still may be wrong). With flipped bit order it produces weird numbers in general, but one constellation gets -35.5 for the second one.