@encointer/util

Utilities for Encointer App and Explorer. Contains methods to work with substrate-based Encointer node.

Installation

yarn add @encointer/util

Fixed-point numbers

Package contains parser from fixed-point number representation created with substrate-fixed crate into floating-point Number. Conversion from floating-point to fixed-point also supported. Several bit-length for decimal and fractional parts supported out of the box: 4, 8, 16, 32, 64.

Usage

To decode number with 64 bits of decimal part and 64 bits of fractional part use parseI64F64. To encode number as fixed point with 4 bits for decimal and fractional use functions toI16F16 (returns fixed-point as BigNum) and encodeFloatToI4F4 (returns Uint8Array).

import { parseI32F32, toI16F16 } from '@encointer/util'
import BN from 'bn.js'

/// Got fixed-point encoded number
const balanceFixPoint = new BN('110000000000000000000000000000001', 2)
console.log(balanceFixPoint)          // -> <BN: 180000001>

/// Parse to Number
const balance = parseI32F32(balanceFixPoint)
console.log(balance)                  // -> 1.5000000002328306

/// Convert to smaller word size
const balance32bit = toI16F16(balance)
console.log(balance32bit)             // -> <BN: 18000>
console.log(balance32bit.toString(2)) // -> 11000000000000000

Various bit-length support

Function parserFixPoint used to produce function converting fixed-point to Number. For example to decode decimal 2 bits and fractional 6 bits use this function to create new parser.

Parameters: upper: 0..128 - number of bits in decimal part lower: 0..128 - number of bits in fractional part

Produced function parameters: raw: substrate_fixed::types::IF as I<upper+lower> precision: 0..lower number bits in fractional part to process

const parseI2F6 = parserFixPoint(2, 6);