Short collection of thoughts on a couple different approaches for default "options" values in a function

default-params.js

// Static defaults

function fn({a = 1, b = 2} = {}) {
  return `${a}, ${b}`
}

fn()            // 1, 2
fn({b: 'beta'}) // 1, beta

/**
 * Pros
 * + No other objects to modify/update
 * + Signature explicitly states the defaults
 * 
 * 
 * Cons
 * - Harder to externalize defaults than other approaches
 * 
 */

// -----------------------------------------------------------------------------

// Externalized - explicit

let fnDefaults = {a: 1, b: 2};

function fn1({a = fnDefaults.a, b = fnDefaults.b} = {}) {
	return `${a}, ${b}`
}

fn()            // 1, 2
fn({b: 'beta'}) // 1, beta

/**
 * Pros
 * + All default logic is contained in the params
 * + Brittle: Changing defaults means that unexpected object changes will be 
 *   be caught by the compiler
 * 
 * Cons
 * - function signature gets cluttered after only a couple defaults
 * - Any default change must be applied in at least 2 locations, possibly more:
 *   this feels boilerplate-y
 * - 
 */



// -----------------------------------------------------------------------------

// Externalized - implicit 

// May want to deep-freeze this object
let fnDefaults = {a: 1, b: 2};

function fn(options = {}) {
  let {a, b} = {...fnDefaults, ...options}
  return `${a}, ${b}`
}

fn()            // 1, 2
fn({b: 'beta'}) // 1, beta


/**
 * Pros
 * + Defaults can be data-driven
 * + Don't have to enumerate defaults by hand
 * + IMO this snippet reads pretty clearly
 * 
 * Cons
 * - 
 * - Depending on the engine, this may have more overhead (instructions, memory)
 *   than other approaches. I didn't confirm this
 * - Function signature doesn't have any info about the expected params
 *    - You could write the signature as `fn(options = fnDefaults)`
 * 
 */

See also https://remysharp.com/2017/10/25/es6-default-arguments