Quickstart

Introduction

This library is an exploration of the application of functional programming to practical.

It innovates on the traditional currying and immutability.

It aims to improve the reliability and maintainability of the program by grasping data flow and data structure.

Installation

• Npm

npm install fx-flow

• Yarn

yarn add fx-flow

• Pnpm

pnpm add fx-flow

Usage

Import

TIP

The default target is ESNext.

• ESM

import { flow, pipe } from 'fx-flow'

• CJS

const { flow, pipe } = require('fx-flow')

flow

TIP

Control the program flow with flow.

import { andThen, delay, err, errThen, flow, ok } from 'fx-flow'

type User = number

const queryUserInfo = (id: number) => {
  const userList: Promise<User[]> = delay(100, [1, 2, 3, 4])

  return userList.then((users) => {
    const targetUser = users.find((x) => x === id)
    if (targetUser) {
      return ok(targetUser)
    }
    return err('user not found')
  })
}

const notify = (msg: string) => console.log(msg)

const defaultUser: User = 0

// Find target user, `notify` would be skipped.
// userInfo = 1
const userInfo = (await flow(ok(1), andThen(queryUserInfo), errThen(notify))).unwrap()

// No user found, `notify` would be called.
// userInfo2 = defaultUser
const userInfo2 = (await flow(ok(-1), andThen(queryUserInfo), errThen(notify))).unwrapOr(defaultUser)

Exceptions are automatically caught and converted into Result.

pipe

TIP

Perform batch operations on a collection with pipe.

import { concurrent, consume, delay, filter, map, pipe, toAsync } from 'fx-flow'

type FileChunk = number

const fileChunks: FileChunk[] = [1, 2, 3, 4, 5]

const notUploaded = (chunk: FileChunk) => chunk > 1

const uploadChunk = (chunk: FileChunk) => delay(1000, chunk).then(() => console.log(`${chunk} has been uploaded`))

// Upload file chunks concurrently and skip uploaded chunks.
pipe(fileChunks, toAsync, filter(notUploaded), map(uploadChunk), concurrent(3), consume)