/* eslint-disable camelcase */
let pbkdf2 = require('pbkdf2')

var MAX_VALUE = 0x7fffffff

// N = Cpu cost, r = Memory cost, p = parallelization cost
async function scrypt (key, salt, N, r, p, dkLen, progressCallback) {
  if (N === 0 || (N & (N - 1)) !== 0) throw Error('N must be > 0 and a power of 2')

  if (N > MAX_VALUE / 128 / r) throw Error('Parameter N is too large')
  if (r > MAX_VALUE / 128 / p) throw Error('Parameter r is too large')

  var XY = Buffer.alloc(256 * r)
  var V = Buffer.alloc(128 * r * N)

  // pseudo global
  var B32 = new Int32Array(16) // salsa20_8
  var x = new Int32Array(16) // salsa20_8
  var _X = Buffer.alloc(64) // blockmix_salsa8

  // pseudo global
  var B = pbkdf2.pbkdf2Sync(key, salt, 1, p * 128 * r, 'sha256')

  var tickCallback
  if (progressCallback) {
    var totalOps = p * N * 2
    var currentOp = 0

    tickCallback = function () {
      return new Promise(function(resolve, reject) {
        ++currentOp

        // send progress notifications once every 1,000 ops
        if (currentOp % 1000 === 0) {
          progressCallback({
            current: currentOp,
            total: totalOps,
            percent: (currentOp / totalOps) * 100.0
          })
          setTimeout(resolve, 10)
        } else {
          resolve()
        }
      })



    }
  }

  for (var i = 0; i < p; i++) {
    await smix(B, i * 128 * r, r, N, V, XY)
    if (typeof shold_stop_bip38 !== 'undefined') break;
  }

  return pbkdf2.pbkdf2Sync(key, B, 1, dkLen, 'sha256')

  // all of these functions are actually moved to the top
  // due to function hoisting

  async function smix (B, Bi, r, N, V, XY) {
    var Xi = 0
    var Yi = 128 * r
    var i

    B.copy(XY, Xi, Bi, Bi + Yi)

    for (i = 0; i < N; i++) {
      XY.copy(V, i * Yi, Xi, Xi + Yi)
      blockmix_salsa8(XY, Xi, Yi, r)

      if (tickCallback) {
        await tickCallback()
        if (typeof shold_stop_bip38 !== 'undefined') break;
      }
    }

    for (i = 0; i < N; i++) {
      var offset = Xi + (2 * r - 1) * 64
      var j = XY.readUInt32LE(offset) & (N - 1)
      blockxor(V, j * Yi, XY, Xi, Yi)
      blockmix_salsa8(XY, Xi, Yi, r)

      if (tickCallback) {
        await tickCallback()
        if (typeof shold_stop_bip38 !== 'undefined') break;
      }
    }

    XY.copy(B, Bi, Xi, Xi + Yi)
  }

  function blockmix_salsa8 (BY, Bi, Yi, r) {
    var i

    arraycopy(BY, Bi + (2 * r - 1) * 64, _X, 0, 64)

    for (i = 0; i < 2 * r; i++) {
      blockxor(BY, i * 64, _X, 0, 64)
      salsa20_8(_X)
      arraycopy(_X, 0, BY, Yi + (i * 64), 64)
    }

    for (i = 0; i < r; i++) {
      arraycopy(BY, Yi + (i * 2) * 64, BY, Bi + (i * 64), 64)
    }

    for (i = 0; i < r; i++) {
      arraycopy(BY, Yi + (i * 2 + 1) * 64, BY, Bi + (i + r) * 64, 64)
    }
  }

  function R (a, b) {
    return (a << b) | (a >>> (32 - b))
  }

  function salsa20_8 (B) {
    var i

    for (i = 0; i < 16; i++) {
      B32[i] = (B[i * 4 + 0] & 0xff) << 0
      B32[i] |= (B[i * 4 + 1] & 0xff) << 8
      B32[i] |= (B[i * 4 + 2] & 0xff) << 16
      B32[i] |= (B[i * 4 + 3] & 0xff) << 24
      // B32[i] = B.readUInt32LE(i*4)   <--- this is signficantly slower even in Node.js
    }

    arraycopy(B32, 0, x, 0, 16)

    for (i = 8; i > 0; i -= 2) {
      x[4] ^= R(x[0] + x[12], 7)
      x[8] ^= R(x[4] + x[0], 9)
      x[12] ^= R(x[8] + x[4], 13)
      x[0] ^= R(x[12] + x[8], 18)
      x[9] ^= R(x[5] + x[1], 7)
      x[13] ^= R(x[9] + x[5], 9)
      x[1] ^= R(x[13] + x[9], 13)
      x[5] ^= R(x[1] + x[13], 18)
      x[14] ^= R(x[10] + x[6], 7)
      x[2] ^= R(x[14] + x[10], 9)
      x[6] ^= R(x[2] + x[14], 13)
      x[10] ^= R(x[6] + x[2], 18)
      x[3] ^= R(x[15] + x[11], 7)
      x[7] ^= R(x[3] + x[15], 9)
      x[11] ^= R(x[7] + x[3], 13)
      x[15] ^= R(x[11] + x[7], 18)
      x[1] ^= R(x[0] + x[3], 7)
      x[2] ^= R(x[1] + x[0], 9)
      x[3] ^= R(x[2] + x[1], 13)
      x[0] ^= R(x[3] + x[2], 18)
      x[6] ^= R(x[5] + x[4], 7)
      x[7] ^= R(x[6] + x[5], 9)
      x[4] ^= R(x[7] + x[6], 13)
      x[5] ^= R(x[4] + x[7], 18)
      x[11] ^= R(x[10] + x[9], 7)
      x[8] ^= R(x[11] + x[10], 9)
      x[9] ^= R(x[8] + x[11], 13)
      x[10] ^= R(x[9] + x[8], 18)
      x[12] ^= R(x[15] + x[14], 7)
      x[13] ^= R(x[12] + x[15], 9)
      x[14] ^= R(x[13] + x[12], 13)
      x[15] ^= R(x[14] + x[13], 18)
    }

    for (i = 0; i < 16; ++i) B32[i] = x[i] + B32[i]

    for (i = 0; i < 16; i++) {
      var bi = i * 4
      B[bi + 0] = (B32[i] >> 0 & 0xff)
      B[bi + 1] = (B32[i] >> 8 & 0xff)
      B[bi + 2] = (B32[i] >> 16 & 0xff)
      B[bi + 3] = (B32[i] >> 24 & 0xff)
      // B.writeInt32LE(B32[i], i*4)  //<--- this is signficantly slower even in Node.js
    }
  }

  // naive approach... going back to loop unrolling may yield additional performance
  function blockxor (S, Si, D, Di, len) {
    for (var i = 0; i < len; i++) {
      D[Di + i] ^= S[Si + i]
    }
  }
}

function arraycopy (src, srcPos, dest, destPos, length) {
  if (Buffer.isBuffer(src) && Buffer.isBuffer(dest)) {
    src.copy(dest, destPos, srcPos, srcPos + length)
  } else {
    while (length--) {
      dest[destPos++] = src[srcPos++]
    }
  }
}

module.exports = scrypt