Delay

A dead simple data-structure for delaying side effects. Written in the excellent gleam ✨ language. Supporting both Erlang & Javascript targets

Basic Usage

import gleam/io

let d = delay_effect(fn() {
 io.println("Hello")
 Ok(1)
}) |> delay.map(fn(x) {
 io.println("World")
 Ok(x + 1)
})

let res = delay.run(d)
// Hello
// World
// res = Ok(2)

More info

The result of delay_effect is really just a first class function with a nice API wrapper. It isn’t executed until put through one of run/1, drain/1 or fallthrough/1. And can be called as many times as you want.

let d = delay_effect(fn() {
 io.println("Hello")
 Error("bummer")
}) |> delay.map(fn(x) {
 io.println("World")
 Ok(x + 1)
})

let res = delay.run(d)
// Hello
// res = Error("bummer")

If one of the functions in the chain fails, the rest will short circuit and the error will be returned.

Effects can be retried as well via retry/3

// using the same effect `d` from above

let res = delay.retry(d, 3, 200) |> delay.run()
// Hello
// Hello
// Hello
// Hello
// res = Error("bummer")

Usage within Javascript 🌸 directly

If you want to use this library from javascript alone, but aren’t ready to embrace gleam, you can install it from npm! You will need a copy of Gleam’s JS prelude as well

npm i delay-gleam

import { delay_effect, map, run } from "delay-gleam"
import { Ok, Error } from "./prelude.mjs"

d = delay_effect(() => new Error(console.log("123")))
d = map(d, (_) => new Ok(console.log("456")))
run(d)
// 123

FAQ

Doesn’t the concept of a delayed side effect kind of lose value in the world of actor model concurrency and zero shared memory?!

A little

Then why did you write this?

For fun

Is gleam ✨ actually excellent?

So far