CFP - Docs - compose

compose

(fnArray: function[]) => (...args: any[]) => Result

Create a function that passes a value through fnArray.

Each function takes the result of the prior function as its parameter.

Try It

const increment = n => n + 1
const double = n => n * 2
const toAnswer = n => 'answer: ' + n

const calculate = compose([increment, double, toAnswer])

calculate(1) // is 'answer: 4'

const returnFirstArg = compose([])
returnFirstArg('a', 'b') // is 'a'
const increment = (n: number) => n + 1
const double = (n: number) => n * 2
const toAnswer = (n: number) => 'answer: ' + n

const calculate = compose([increment, double, toAnswer])

calculate(1) // is 'answer: 4'

const returnFirstArg = compose([])
returnFirstArg('a', 'b') // is 'a'

Try It

const increment = n => n + 1
const double = n => n * 2
const toAnswer = n => 'answer: ' + n

const calculate = compose([
  increment,
  double,
  toAnswer,
])

calculate(1) // is 'answer: 4'

const returnFirstArg = compose([])
returnFirstArg('a', 'b') // is 'a'
const increment = (n: number) => n + 1
const double = (n: number) => n * 2
const toAnswer = (n: number) => 'answer: ' + n

const calculate = compose([
  increment,
  double,
  toAnswer,
])

calculate(1) // is 'answer: 4'

const returnFirstArg = compose([])
returnFirstArg('a', 'b') // is 'a'

Just like passThrough, compose gives us a flexible and readable way to combine logic. It's one of the big reasons I find functional programming to be helpful.

Below, we have the five-day weather forecast. Let's compose a function to get the average low.

Try It

const forecast = [
  { high: 68, low: 54 },
  { high: 62, low: 50 },
  { high: 63, low: 52 },
  { high: 55, low: 43 },
  { high: 58, low: 49 },
]

const getAvgLow = compose([
  mapValues(get('low')),
  getAverageValue,
  roundToNearest('0.1'),
])

const avgLow = getAvgLow(forecast)
console.log(avgLow)
// is 49.6
type Forecast = {
  high: number
  low: number
}

const fiveDayForecast: Forecast[] = [
  { high: 68, low: 54 },
  { high: 62, low: 50 },
  { high: 63, low: 52 },
  { high: 55, low: 43 },
  { high: 58, low: 49 },
]

const getAvgLow = compose([
  mapValues(get('low'))<Forecast[]>,
  getAverageValue,
  roundToNearest('0.1'),
])

const avgLow = getAvgLow(fiveDayForecast)
console.log(avgLow)
// is 49.6

Try It

const forecast = [
  { high: 68, low: 54 },
  { high: 62, low: 50 },
  { high: 63, low: 52 },
  { high: 55, low: 43 },
  { high: 58, low: 49 },
]

const getAvgLow = compose([
  mapValues(get('low')),
  getAverageValue,
  roundToNearest('0.1'),
])

const avgLow = getAvgLow(forecast)
console.log(avgLow)
// is 49.6
type Forecast = {
  high: number
  low: number
}

const fiveDayForecast: Forecast[] = [
  { high: 68, low: 54 },
  { high: 62, low: 50 },
  { high: 63, low: 52 },
  { high: 55, low: 43 },
  { high: 58, low: 49 },
]

const getAvgLow = compose([
  mapValues(get('low'))<Forecast[]>,
  getAverageValue,
  roundToNearest('0.1'),
])

const avgLow = getAvgLow(fiveDayForecast)
console.log(avgLow)
// is 49.6

When to use compose vs. passThrough?
They can often be used interchangeably. Compose allows you to name a group of functions, which can sometimes be helpful. Other times a name is unnecessary, and you just want to pass data through.

When to use compose vs. passThrough?