Extended version of Rambda

Overview

Rambdax

Extended version of Rambda(utility library) - Documentation

Rambda is smaller and faster alternative to the popular functional programming library Ramda. - Documentation

CircleCI codecov dependencies Status Library size

❯ Differences between Rambda and Rambdax

Rambdax passthrough all Rambda methods and introduce some new functions.

The idea of Rambdax is to extend Rambda without worring for Ramda compatibility.

---------------

❯ Example use

import { composeAsync, filter, delay, mapAsync } from 'rambdax'

const result = await composeAsync(
  mapAsync(async x => {
    await delay(100)
    return x + 1
  }),
  filter(x => x > 1)
)([1, 2, 3])
// => [3, 4]

You can test this example in Rambda's REPL

---------------

❯ Rambdax's advantages

Dot notation for R.path, R.paths, R.assocPath and R.lensPath

Standard usage of R.path is R.path(['a', 'b'], {a: {b: 1} }).

In Rambda you have the choice to use dot notation(which is arguably more readable):

R.path('a.b', {a: {b: 1} })

Comma notation for R.pick and R.omit

Similar to dot notation, but the separator is comma(,) instead of dot(.).

R.pick('a,b', {a: 1 , b: 2, c: 3} })
// No space allowed between properties

Extendable with Ramda community projects

Rambdax implements some methods from Ramda community projects, such as R.lensSatisfies, R.lensEq and R.viewOr.

Support

Most of the valid issues are fixed within 2-3 days.

Closing the issue is usually accompanied by publishing a new patch version of Rambdax to NPM.

---------------

❯ Missing Ramda methods

Click to see the full list of 89 Ramda methods not implemented in Rambda
  • __
  • addIndex
  • ap
  • aperture
  • apply
  • applyTo
  • ascend
  • binary
  • bind
  • call
  • comparator
  • composeK
  • composeP
  • composeWith
  • construct
  • constructN
  • contains
  • countBy
  • descend
  • differenceWith
  • dissocPath
  • empty
  • eqBy
  • forEachObjIndexed
  • gt
  • gte
  • hasIn
  • innerJoin
  • insert
  • insertAll
  • into
  • invert
  • invertObj
  • invoker
  • juxt
  • keysIn
  • lift
  • liftN
  • lt
  • lte
  • mapAccum
  • mapAccumRight
  • mapObjIndexed
  • memoizeWith
  • mergeDeepLeft
  • mergeDeepWith
  • mergeDeepWithKey
  • mergeRight
  • mergeWith
  • mergeWithKey
  • nAry
  • nthArg
  • o
  • otherwise
  • pair
  • partialRight
  • pathSatisfies
  • pickBy
  • pipeK
  • pipeP
  • pipeWith
  • project
  • propSatisfies
  • reduceBy
  • reduceRight
  • reduceWhile
  • reduced
  • remove
  • scan
  • sequence
  • sortWith
  • symmetricDifferenceWith
  • andThen
  • toPairsIn
  • transduce
  • traverse
  • unapply
  • unary
  • uncurryN
  • unfold
  • unionWith
  • uniqBy
  • unnest
  • until
  • useWith
  • valuesIn
  • xprod
  • thunkify
  • default

---------------

❯ Install

  • yarn add rambdax

  • For UMD usage either use ./dist/rambdax.umd.js or the following CDN link:

https://unpkg.com/rambdax@CURRENT_VERSION/dist/rambdax.umd.js
  • with deno
import {compose, add} from 'https://raw.githubusercontent.com/selfrefactor/rambdax/master/dist/rambdax.esm.js'

---------------

Differences between Rambda and Ramda

  • Rambda's type detects async functions and unresolved Promises. The returned values are 'Async' and 'Promise'.

  • Rambda's type handles NaN input, in which case it returns NaN.

  • Rambda's forEach can iterate over objects not only arrays.

  • Rambda's map, filter, partition when they iterate over objects, they pass property and input object as predicate's argument.

  • Rambda's filter returns empty array with bad input(null or undefined), while Ramda throws.

  • Ramda's clamp work with strings, while Rambda's method work only with numbers.

  • Error handling, when wrong inputs are provided, may not be the same. This difference will be better documented once all brute force tests are completed.

  • Typescript definitions between rambda and @types/ramda may vary.

---------------

❯ Benchmarks

Click to expand all benchmark results

There are methods which are benchmarked only with Ramda and Rambda(i.e. no Lodash).

Note that some of these methods, are called with and without curring. This is done in order to give more detailed performance feedback.

The benchmarks results are produced from latest versions of Rambda, Lodash(4.17.20) and Ramda(0.27.1).

method Rambda Ramda Lodash
add 96.25% slower 96.24% slower 🚀 Fastest
adjust 🚀 Fastest 5.52% slower 🔳
all 🚀 Fastest 94.95% slower 🔳
allPass 🚀 Fastest 98.95% slower 🔳
any 🚀 Fastest 98.18% slower 6.18% slower
anyPass 🚀 Fastest 99.09% slower 🔳
append 🚀 Fastest 84.09% slower 🔳
applySpec 🚀 Fastest 75.73% slower 🔳
assoc 87.98% slower 57.39% slower 🚀 Fastest
clone 🚀 Fastest 96.03% slower 91.75% slower
compose 🚀 Fastest 96.45% slower 77.83% slower
converge 49.12% slower 🚀 Fastest 🔳
curry 🚀 Fastest 34.9% slower 🔳
curryN 63.32% slower 🚀 Fastest 🔳
defaultTo 🚀 Fastest 50.3% slower 🔳
drop 🚀 Fastest 97.45% slower 🔳
dropLast 🚀 Fastest 97.07% slower 🔳
equals 72.11% slower 79.48% slower 🚀 Fastest
filter 🚀 Fastest 94.74% slower 58.18% slower
find 🚀 Fastest 98.2% slower 88.96% slower
findIndex 🚀 Fastest 97.97% slower 79.39% slower
flatten 6.56% slower 95.38% slower 🚀 Fastest
ifElse 🚀 Fastest 70.97% slower 🔳
includes 🚀 Fastest 71.7% slower 🔳
indexOf 🚀 Fastest 84.08% slower 7.86% slower
init 94.42% slower 97.55% slower 🚀 Fastest
is 🚀 Fastest 11.72% slower 🔳
isEmpty 51.68% slower 93.82% slower 🚀 Fastest
last 🚀 Fastest 99.64% slower 1.05% slower
lastIndexOf 🚀 Fastest 42.38% slower 🔳
map 🚀 Fastest 69.63% slower 4.68% slower
match 🚀 Fastest 46.75% slower 🔳
merge 63.55% slower 🚀 Fastest 55.25% slower
none 🚀 Fastest 98.22% slower 🔳
omit 🚀 Fastest 70.66% slower 97.56% slower
over 🚀 Fastest 50.77% slower 🔳
path 🚀 Fastest 74.94% slower 5.72% slower
pick 🚀 Fastest 26.29% slower 86.82% slower
prop 🚀 Fastest 89.89% slower 🔳
propEq 🚀 Fastest 95.26% slower 🔳
range 95.17% slower 90.22% slower 🚀 Fastest
reduce 52.76% slower 74.02% slower 🚀 Fastest
repeat 85.91% slower 95.31% slower 🚀 Fastest
replace 0.47% slower 28.13% slower 🚀 Fastest
set 🚀 Fastest 36.26% slower 🔳
sort 🚀 Fastest 63.15% slower 🔳
sortBy 🚀 Fastest 61.57% slower 88.88% slower
split 🚀 Fastest 85.34% slower 33.69% slower
splitEvery 🚀 Fastest 90.18% slower 🔳
take 93.44% slower 98.04% slower 🚀 Fastest
takeLast 92.61% slower 98.83% slower 🚀 Fastest
test 🚀 Fastest 94.42% slower 🔳
type 18.91% slower 🚀 Fastest 🔳
uniq 98.98% slower 96.58% slower 🚀 Fastest
update 🚀 Fastest 38.88% slower 🔳
view 🚀 Fastest 82.21% slower 🔳

---------------

❯ Used by

---------------

API

add

add(a: number, b: number): number

It adds a and b.

💥 It doesn't work with strings, as the inputs are parsed to numbers before calculation.

R.add(2, 3) // =>  5

Try this R.add example in Rambda REPL

R.add source
export function add(a, b) {
  if (arguments.length === 1)
    return (_b) => add(a, _b);

  return Number(a) + Number(b);
}
Tests
import { add } from './add'

test('with number', () => {
  expect(add(2, 3)).toEqual(5)
  expect(add(7)(10)).toEqual(17)
})

test('string is bad input', () => {
  expect(add('foo', 'bar')).toBeNaN()
})

test('ramda specs', () => {
  expect(add('1', '2')).toEqual(3)
  expect(add(1, '2')).toEqual(3)
  expect(add(true, false)).toEqual(1)
  expect(add(null, null)).toEqual(0)
  expect(add(undefined, undefined)).toEqual(NaN)
  expect(add(new Date(1), new Date(2))).toEqual(3)
})

---------------

adjust

adjust<T>(index: number, replaceFn: (x: T) => T, list: readonly T[]): readonly T[]

It replaces index in array list with the result of replaceFn(list[i]).

R.adjust(
  0,
  a => a + 1,
  [0, 100]
) // => [1, 100]

Try this R.adjust example in Rambda REPL

R.adjust source
import { curry } from './curry'

function adjustFn(
  index, replaceFn, list
){
  const actualIndex = index < 0 ? list.length + index : index
  if (index >= list.length || actualIndex < 0) return list

  const clone = list.slice()
  clone[ actualIndex ] = replaceFn(clone[ actualIndex ])

  return clone
}

export const adjust = curry(adjustFn)
Tests
import { add } from './add'
import { adjust } from './adjust'
import { pipe } from './pipe'

const list = [ 0, 1, 2 ]
const expected = [ 0, 11, 2 ]

test('happy', () => {})

test('happy', () => {
  expect(adjust(
    1, add(10), list
  )).toEqual(expected)
})

test('with curring type 1 1 1', () => {
  expect(adjust(1)(add(10))(list)).toEqual(expected)
})

test('with curring type 1 2', () => {
  expect(adjust(1)(add(10), list)).toEqual(expected)
})

test('with curring type 2 1', () => {
  expect(adjust(1, add(10))(list)).toEqual(expected)
})

test('with negative index', () => {
  expect(adjust(
    -2, add(10), list
  )).toEqual(expected)
})

test('when index is out of bounds', () => {
  const list = [ 0, 1, 2, 3 ]
  expect(adjust(
    4, add(1), list
  )).toEqual(list)
  expect(adjust(
    -5, add(1), list
  )).toEqual(list)
})

---------------

all

all<T>(predicate: (x: T) => boolean, list: readonly T[]): boolean

It returns true, if all members of array list returns true, when applied as argument to predicate function.

const list = [ 0, 1, 2, 3, 4 ]
const predicate = x => x > -1

const result = R.all(predicate, list)
// => true

Try this R.all example in Rambda REPL

R.all source
export function all(predicate, list){
  if (arguments.length === 1) return _list => all(predicate, _list)

  for (let i = 0; i < list.length; i++){
    if (!predicate(list[ i ])) return false
  }

  return true
}
Tests
import { all } from './all'

const list = [ 0, 1, 2, 3, 4 ]

test('when true', () => {
  const fn = x => x > -1

  expect(all(fn)(list)).toBeTrue()
})

test('when false', () => {
  const fn = x => x > 2

  expect(all(fn, list)).toBeFalse()
})

---------------

allFalse

allFalse(...inputs: readonly any[]): boolean

It returns true if all inputs arguments are falsy(empty objects and empty arrays are considered falsy).

Functions are valid inputs, but these functions cannot have their own arguments.

This method is very similar to R.anyFalse, R.anyTrue and R.allTrue

R.allFalse(0, null, [], {}, '', () => false)
// => true

Try this R.allFalse example in Rambda REPL

R.allFalse source
import { isTruthy } from './_internals/isTruthy'
import { type } from './type'

export function allFalse(...inputs){
  let counter = 0
  while (counter < inputs.length){
    const x = inputs[ counter ]

    if (type(x) === 'Function'){
      if (isTruthy(x())){
        return false
      }
    } else if (isTruthy(x)){
      return false
    }

    counter++
  }

  return true
}
Tests
import { runTests } from 'helpers-fn'

import { allFalse } from './allFalse'

const happy = { ok : [ () => false, () => [], () => {}, null, false, [] ] }
const withArray = { fail : [ ...happy.ok, [ 1 ] ] }
const withObject = { fail : [ ...happy.ok, { a : 1 } ] }
const withFunction = { fail : [ ...happy.ok, () => ({ a : 1 }) ] }
const withBoolean = { fail : [ ...happy.ok, true ] }

const testData = {
  label : 'R.allFalse',
  data  : [ happy, withArray, withObject, withFunction, withBoolean ],
  fn    : input => allFalse(...input),
}
runTests(testData)

---------------

allPass

allPass<T>(predicates: readonly ((x: T) => boolean)[]): (input: T) => boolean

It returns true, if all functions of predicates return true, when input is their argument.

const input = {
  a : 1,
  b : 2,
}
const predicates = [
  x => x.a === 1,
  x => x.b === 2,
]
const result = R.allPass(predicates)(input) // => true

Try this R.allPass example in Rambda REPL

R.allPass source
export function allPass(predicates){
  return input => {
    let counter = 0
    while (counter < predicates.length){
      if (!predicates[ counter ](input)){
        return false
      }
      counter++
    }

    return true
  }
}
Tests
import { allPass } from './allPass'

test('happy', () => {
  const rules = [ x => typeof x === 'number', x => x > 10, x => x * 7 < 100 ]

  expect(allPass(rules)(11)).toBeTrue()

  expect(allPass(rules)(undefined)).toBeFalse()
})

test('when returns true', () => {
  const conditionArr = [ val => val.a === 1, val => val.b === 2 ]

  expect(allPass(conditionArr)({
    a : 1,
    b : 2,
  })).toBeTrue()
})

test('when returns false', () => {
  const conditionArr = [ val => val.a === 1, val => val.b === 3 ]

  expect(allPass(conditionArr)({
    a : 1,
    b : 2,
  })).toBeFalse()
})

---------------

allTrue

allTrue(...input: readonly any[]): boolean

It returns true if all inputs arguments are truthy(empty objects and empty arrays are considered falsy).

R.allTrue(1, true, {a: 1}, [1], 'foo', () => true)
// => true

Try this R.allTrue example in Rambda REPL

R.allTrue source
import { isFalsy } from './_internals/isFalsy'
import { type } from './type'

export function allTrue(...inputs){
  let counter = 0
  while (counter < inputs.length){
    const x = inputs[ counter ]

    if (type(x) === 'Function'){
      if (isFalsy(x())){
        return false
      }
    } else if (isFalsy(x)){
      return false
    }

    counter++
  }

  return true
}
Tests
import { allTrue } from './allTrue'

test('with functions', () => {
  const foo = () => 1
  const bar = () => false
  const baz = () => JSON.parse('{sda')
  const result = allTrue(
    foo, bar, baz
  )
  expect(result).toBeFalse()
})

test('usage with non boolean', () => {
  const foo = { a : 1 }
  const baz = [ 1, 2, 3 ]

  const result = allTrue(
    foo, foo, baz
  )
  expect(result).toBeTrue()
})

test('usage with boolean', () => {
  const foo = 4
  const baz = [ 1, 2, 3 ]

  const result = allTrue(foo > 2, baz.length === 3)
  expect(result).toBeTrue()
})

test('escapes early - case 0', () => {
  const foo = undefined
  const result = allTrue(foo, () => foo.a)
  expect(result).toBeFalse()
})

test('escapes early - case 1', () => {
  const foo = null
  const result = allTrue(foo, () => foo.a)
  expect(result).toBeFalse()
})

test('escapes early - case 2', () => {
  const foo = { a : 'bar' }
  const result = allTrue(
    foo, foo.a, foo.a.b
  )
  expect(result).toBeFalse()
})

test('escapes early - case 3', () => {
  const foo = { a : { b : 'foo' } }
  const result = allTrue(
    foo,
    () => foo.a,
    () => foo.a.b
  )
  expect(result).toBeTrue()
})

---------------

allType

allType(targetType: RambdaTypes): (...input: readonly any[]) => boolean

It returns a function which will return true if all of its inputs arguments belong to targetType.

💥 targetType is one of the possible returns of R.type

const targetType = 'String'

const result = R.allType(
  targetType
)('foo', 'bar', 'baz')
// => true

Try this R.allType example in Rambda REPL

R.allType source
import { type } from './type'

export function allType(targetType){
  return (...inputs) => {
    let counter = 0

    while (counter < inputs.length){
      if (type(inputs[ counter ]) !== targetType){
        return false
      }
      counter++
    }

    return true
  }
}
Tests
import { allType } from './allType'

test('when true', () => {
  const result = allType('Array')(
    [ 1, 2, 3 ], [], [ null ]
  )

  expect(result).toBeTrue()
})

test('when false', () => {
  const result = allType('String')(
    1, undefined, null, []
  )

  expect(result).toBeFalse()
})

---------------

always

always<T>(x: T): () => T

It returns function that always returns x.

const fn = R.always(7)

console.log(fn())// => 7

Try this R.always example in Rambda REPL

R.always source
export function always(x){
  return () => x
}
Tests
import { always } from './always'
import { F } from './F'

test('happy', () => {
  const fn = always(7)

  expect(fn()).toEqual(7)
  expect(fn()).toEqual(7)
})

test('f', () => {
  const fn = always(F())

  expect(fn()).toBeFalse()
  expect(fn()).toBeFalse()
})

---------------

and

and<T, U>(x: T, y: U): T | U

Logical AND

R.and(true, true); // => true
R.and(false, true); // => false
R.and(true, 'foo'); // => 'foo'

Try this R.and example in Rambda REPL

R.and source
export function and(a, b){
  if (arguments.length === 1) return _b => and(a, _b)

  return a && b
}
Tests
import { and } from './and'

test('happy', () => {
  expect(and(1, 'foo')).toBe('foo')
  expect(and(true, true)).toBeTrue()
  expect(and(true)(true)).toBeTrue()
  expect(and(true, false)).toBeFalse()
  expect(and(false, true)).toBeFalse()
  expect(and(false, false)).toBeFalse()
})

---------------

any

any<T>(predicate: (x: T) => boolean, list: readonly T[]): boolean

It returns true, if at least one member of list returns true, when passed to a predicate function.

const list = [1, 2, 3]
const predicate = x => x * x > 8
R.any(fn, list)
// => true

Try this R.any example in Rambda REPL

R.any source
export function any(predicate, list){
  if (arguments.length === 1) return _list => any(predicate, _list)

  let counter = 0
  while (counter < list.length){
    if (predicate(list[ counter ], counter)){
      return true
    }
    counter++
  }

  return false
}
Tests
import { any } from './any'

const list = [ 1, 2, 3 ]

test('happy', () => {
  expect(any(x => x < 0, list)).toBeFalse()
})

test('with curry', () => {
  expect(any(x => x > 2)(list)).toBeTrue()
})

---------------

anyFalse

anyFalse(...input: readonly any[]): boolean

It returns true if any of inputs is falsy(empty objects and empty arrays are considered falsy).

R.anyFalse(1, {a: 1}, [1], () => false)
// => true

Try this R.anyFalse example in Rambda REPL

R.anyFalse source
import { isFalsy } from './_internals/isFalsy'
import { type } from './type'

export function anyFalse(...inputs){
  let counter = 0
  while (counter < inputs.length){
    const x = inputs[ counter ]

    if (type(x) === 'Function'){
      if (isFalsy(x())){
        return true
      }
    } else if (isFalsy(x)){
      return true
    }

    counter++
  }

  return false
}
Tests
import { anyFalse } from './anyFalse'

test('when true', () => {
  expect(anyFalse(
    true, true, false
  )).toBeTruthy()
})

test('when false', () => {
  expect(anyFalse(true, true)).toBeFalsy()
})

test('supports function', () => {
  expect(anyFalse(
    true,
    () => true,
    () => false
  )).toBeTruthy()
})

---------------

anyPass

anyPass<T>(predicates: readonly SafePred<T>[]): SafePred<T>

It accepts list of predicates and returns a function. This function with its input will return true, if any of predicates returns true for this input.

const isBig = x => x > 20
const isOdd = x => x % 2 === 1
const input = 11

const fn = R.anyPass(
  [isBig, isOdd]
)

const result = fn(input) 
// => true

Try this R.anyPass example in Rambda REPL

R.anyPass source
export function anyPass(predicates){
  return input => {
    let counter = 0
    while (counter < predicates.length){
      if (predicates[ counter ](input)){
        return true
      }
      counter++
    }

    return false
  }
}
Tests
import { anyPass } from './anyPass'

test('happy', () => {
  const rules = [ x => typeof x === 'string', x => x > 10 ]
  const predicate = anyPass(rules)
  expect(predicate('foo')).toBeTrue()
  expect(predicate(6)).toBeFalse()
})

test('happy', () => {
  const rules = [ x => typeof x === 'string', x => x > 10 ]

  expect(anyPass(rules)(11)).toBeTrue()

  expect(anyPass(rules)(undefined)).toBeFalse()
})

const obj = {
  a : 1,
  b : 2,
}

test('when returns true', () => {
  const conditionArr = [ val => val.a === 1, val => val.a === 2 ]

  expect(anyPass(conditionArr)(obj)).toBeTrue()
})

test('when returns false + curry', () => {
  const conditionArr = [ val => val.a === 2, val => val.b === 3 ]

  expect(anyPass(conditionArr)(obj)).toBeFalse()
})

test('happy', () => {
  expect(anyPass([])(3)).toEqual(false)
})

---------------

anyTrue

anyTrue(...input: readonly any[]): boolean

It returns true if any of inputs arguments are truthy(empty objects and empty arrays are considered falsy).

R.anyTrue(0, null, [], {}, '', () => true)
// => true

Try this R.anyTrue example in Rambda REPL

R.anyTrue source
import { isTruthy } from './_internals/isTruthy'
import { type } from './type'

export function anyTrue(...inputs){
  let counter = 0
  while (counter < inputs.length){
    const x = inputs[ counter ]

    if (type(x) === 'Function'){
      if (isTruthy(x())){
        return true
      }
    } else if (isTruthy(x)){
      return true
    }

    counter++
  }

  return false
}
Tests
import { anyTrue } from './anyTrue'

test('when true', () => {
  expect(anyTrue(
    true, true, false
  )).toBeTruthy()
})

test('when false', () => {
  expect(anyTrue(
    false, false, false
  )).toBeFalsy()
})

test('supports function', () => {
  expect(anyTrue(
    false,
    false,
    false,
    () => false,
    () => true
  )).toBeTruthy()
})

---------------

anyType

anyType(targetType: RambdaTypes): (...input: readonly any[]) => boolean

It returns a function which will return true if at least one of its inputs arguments belongs to targetType.

targetType is one of the possible returns of R.type

💥 targetType is one of the possible returns of R.type

const targetType = 'String'

const result = R.anyType(
  targetType
)(1, {}, 'foo')
// => true

Try this R.anyType example in Rambda REPL

R.anyType source
import { type } from './type'

export function anyType(targetType){
  return (...inputs) => {
    let counter = 0

    while (counter < inputs.length){
      if (type(inputs[ counter ]) === targetType){
        return true
      }
      counter++
    }

    return false
  }
}
Tests
import { anyType } from './anyType'

test('when true', () => {
  const result = anyType('Array')(
    1, undefined, null, []
  )

  expect(result).toBeTrue()
})

test('when false', () => {
  const result = anyType('String')(
    1, undefined, null, []
  )

  expect(result).toBeFalse()
})

---------------

append

append<T>(x: T, list: readonly T[]): readonly T[]

It adds element x at the end of list.

const x = 'foo'

const result = R.append(x, ['bar', 'baz'])
// => ['bar', 'baz', 'foo']

Try this R.append example in Rambda REPL

R.append source
export function append(x, input){
  if (arguments.length === 1) return _input => append(x, _input)

  if (typeof input === 'string') return input.split('').concat(x)

  const clone = input.slice()
  clone.push(x)

  return clone
}
Tests
import { append } from './append'

test('happy', () => {
  expect(append('tests', [ 'write', 'more' ])).toEqual([
    'write',
    'more',
    'tests',
  ])
})

test('append to empty array', () => {
  expect(append('tests')([])).toEqual([ 'tests' ])
})

test('with strings', () => {
  expect(append('o', 'fo')).toEqual([ 'f', 'o', 'o' ])
})

---------------

applyDiff

applyDiff<Output>(rules: readonly ApplyDiffRule[], obj: object): Output

It changes paths in an object according to a list of operations. Valid operations are add, update and delete. Its use-case is while writing tests and you need to change the test data.

Note, that you cannot use update operation, if the object path is missing in the input object. Also, you cannot use add operation, if the object path has a value.

const obj = {a: {b:1, c:2}}
const rules = [
  {op: 'remove', path: 'a.c'},
  {op: 'add', path: 'a.d', value: 4},
  {op: 'update', path: 'a.b', value: 2},
]
const result = R.applyDiff(rules, obj)
const expected = {a: {b: 2, d: 4}}

// => `result` is equal to `expected`

Try this R.applyDiff example in Rambda REPL

R.applyDiff source
import { assocPath } from './assocPath'
import { path as pathModule } from './path'
const ALLOWED_OPERATIONS = [ 'remove', 'add', 'update' ]

export function removeAtPath(path, obj){
  const p = typeof path === 'string' ? path.split('.') : path

  const len = p.length
  if (len === 0) return
  if (len === 1) return delete obj[ p[ 0 ] ]
  if (len === 2) return delete obj[ p[ 0 ] ][ p[ 1 ] ]
  if (len === 3) return delete obj[ p[ 0 ] ][ p[ 1 ] ][ p[ 2 ] ]
  if (len === 4) return delete obj[ p[ 0 ] ][ p[ 1 ] ][ p[ 2 ] ][ p[ 3 ] ]
  if (len === 5) return delete obj[ p[ 0 ] ][ p[ 1 ] ][ p[ 2 ] ][ p[ 3 ] ][ p[ 4 ] ]
  if (len === 6){
    return delete obj[ p[ 0 ] ][ p[ 1 ] ][ p[ 2 ] ][ p[ 3 ] ][ p[ 4 ] ][ p[ 5 ] ]
  }
  if (len === 7){
    return delete obj[ p[ 0 ] ][ p[ 1 ] ][ p[ 2 ] ][ p[ 3 ] ][ p[ 4 ] ][ p[ 5 ] ][ p[ 6 ] ]
  }
  if (len === 8){
    return delete obj[ p[ 0 ] ][ p[ 1 ] ][ p[ 2 ] ][ p[ 3 ] ][ p[ 4 ] ][ p[ 5 ] ][ p[ 6 ] ][ p[ 7 ] ]
  }
  if (len === 9){
    return delete obj[ p[ 0 ] ][ p[ 1 ] ][ p[ 2 ] ][ p[ 3 ] ][ p[ 4 ] ][ p[ 5 ] ][ p[ 6 ] ][ p[ 7 ] ][ p[ 8 ] ]
  }
  if (len === 10){
    return delete obj[ p[ 0 ] ][ p[ 1 ] ][ p[ 2 ] ][ p[ 3 ] ][ p[ 4 ] ][ p[ 5 ] ][ p[ 6 ] ][ p[ 7 ] ][ p[ 8 ] ][
      p[ 9 ]
    ]
  }
}

export function applyDiff(rules, obj){
  if (arguments.length === 1) return _obj => applyDiff(rules, _obj)

  let clone = { ...obj }

  rules.forEach(({ op, path, value }) => {
    if (!ALLOWED_OPERATIONS.includes(op)) return
    if (op === 'add' && path && value !== undefined){
      if (pathModule(path, obj)) return

      return clone = assocPath(
        path, value, clone
      )
    }

    if (op === 'remove'){
      if (pathModule(path, obj) === undefined) return

      return removeAtPath(path, clone)
    }
    if (op === 'update' && path && value !== undefined){
      if (pathModule(path, obj) === undefined) return

      return clone = assocPath(
        path, value, clone
      )
    }
  })

  return clone
}
Tests
import { applyDiff } from './applyDiff'

test('remove operation', () => {
  const rules = [
    {
      op   : 'remove',
      path : 'a.b',
    },
  ]
  const result = applyDiff(rules, {
    a : {
      b : 1,
      c : 2,
    },
  })
  expect(result).toEqual({ a : { c : 2 } })
})

test('update operation', () => {
  const rules = [
    {
      op    : 'update',
      path  : 'a.b',
      value : 3,
    },
    {
      op    : 'update',
      path  : 'a.c.1',
      value : 3,
    },
    {
      op    : 'update',
      path  : 'a.d',
      value : 3,
    },
  ]
  const result = applyDiff(rules, {
    a : {
      b : 1,
      c : [ 1, 2 ],
    },
  })
  expect(result).toEqual({
    a : {
      b : 3,
      c : [ 1, 3 ],
    },
  })
})

test('add operation', () => {
  const rules = [
    {
      op    : 'add',
      path  : 'a.b',
      value : 3,
    },
    {
      op    : 'add',
      path  : 'a.d',
      value : 3,
    },
  ]
  const result = applyDiff(rules, {
    a : {
      b : 1,
      c : 2,
    },
  })

  expect(result).toEqual({
    a : {
      b : 1,
      c : 2,
      d : 3,
    },
  })
})

---------------

applySpec

applySpec<Spec extends Record<string, (...args: readonly any[]) => any>>(
  spec: Spec
): (
  ...args: Parameters<ValueOfRecord<Spec>>
) => { readonly [Key in keyof Spec]: ReturnType<Spec[Key]> }

💥 The currying in this function works best with functions with 4 arguments or less. (arity of 4)

const fn = R.applySpec({
  sum: R.add,
  nested: { mul: R.multiply }
})
const result = fn(2, 4) 
// => { sum: 6, nested: { mul: 8 } }

Try this R.applySpec example in Rambda REPL

R.applySpec source
import { _isArray } from './_internals/_isArray'

// recursively traverse the given spec object to find the highest arity function
function __findHighestArity(spec, max = 0){
  for (const key in spec){
    if (spec.hasOwnProperty(key) === false || key === 'constructor') continue

    if (typeof spec[ key ] === 'object'){
      max = Math.max(max, __findHighestArity(spec[ key ]))
    }

    if (typeof spec[ key ] === 'function'){
      max = Math.max(max, spec[ key ].length)
    }
  }

  return max
}

function __filterUndefined(){
  const defined = []
  let i = 0
  const l = arguments.length
  while (i < l){
    if (typeof arguments[ i ] === 'undefined') break
    defined[ i ] = arguments[ i ]
    i++
  }

  return defined
}

function __applySpecWithArity(
  spec, arity, cache
){
  const remaining = arity - cache.length

  if (remaining === 1)
    return x =>
      __applySpecWithArity(
        spec, arity, __filterUndefined(...cache, x)
      )
  if (remaining === 2)
    return (x, y) =>
      __applySpecWithArity(
        spec, arity, __filterUndefined(
          ...cache, x, y
        )
      )
  if (remaining === 3)
    return (
      x, y, z
    ) =>
      __applySpecWithArity(
        spec, arity, __filterUndefined(
          ...cache, x, y, z
        )
      )
  if (remaining === 4)
    return (
      x, y, z, a
    ) =>
      __applySpecWithArity(
        spec,
        arity,
        __filterUndefined(
          ...cache, x, y, z, a
        )
      )
  if (remaining > 4)
    return (...args) =>
      __applySpecWithArity(
        spec, arity, __filterUndefined(...cache, ...args)
      )

  // handle spec as Array
  if (_isArray(spec)){
    const ret = []
    let i = 0
    const l = spec.length
    for (; i < l; i++){
      // handle recursive spec inside array
      if (typeof spec[ i ] === 'object' || _isArray(spec[ i ])){
        ret[ i ] = __applySpecWithArity(
          spec[ i ], arity, cache
        )
      }
      // apply spec to the key
      if (typeof spec[ i ] === 'function'){
        ret[ i ] = spec[ i ](...cache)
      }
    }

    return ret
  }

  // handle spec as Object
  const ret = {}
  // apply callbacks to each property in the spec object
  for (const key in spec){
    if (spec.hasOwnProperty(key) === false || key === 'constructor') continue

    // apply the spec recursively
    if (typeof spec[ key ] === 'object'){
      ret[ key ] = __applySpecWithArity(
        spec[ key ], arity, cache
      )
      continue
    }

    // apply spec to the key
    if (typeof spec[ key ] === 'function'){
      ret[ key ] = spec[ key ](...cache)
    }
  }

  return ret
}

export function applySpec(spec, ...args){
  // get the highest arity spec function, cache the result and pass to __applySpecWithArity
  const arity = __findHighestArity(spec)

  if (arity === 0){
    return () => ({})
  }
  const toReturn = __applySpecWithArity(
    spec, arity, args
  )

  return toReturn
}
Tests
import { applySpec as applySpecRamda, nAry } from 'ramda'

import { add, always, compose, dec, inc, map, path, prop, T } from '../rambda'
import { applySpec } from './applySpec'

test('different than Ramda when bad spec', () => {
  const result = applySpec({ sum : { a : 1 } })(1, 2)
  const ramdaResult = applySpecRamda({ sum : { a : 1 } })(1, 2)
  expect(result).toEqual({})
  expect(ramdaResult).toEqual({ sum : { a : {} } })
})

test('works with empty spec', () => {
  expect(applySpec({})()).toEqual({})
  expect(applySpec([])(1, 2)).toEqual({})
  expect(applySpec(null)(1, 2)).toEqual({})
})

test('works with unary functions', () => {
  const result = applySpec({
    v : inc,
    u : dec,
  })(1)
  const expected = {
    v : 2,
    u : 0,
  }
  expect(result).toEqual(expected)
})

test('works with binary functions', () => {
  const result = applySpec({ sum : add })(1, 2)
  expect(result).toEqual({ sum : 3 })
})

test('works with nested specs', () => {
  const result = applySpec({
    unnested : always(0),
    nested   : { sum : add },
  })(1, 2)
  const expected = {
    unnested : 0,
    nested   : { sum : 3 },
  }
  expect(result).toEqual(expected)
})

test('works with arrays of nested specs', () => {
  const result = applySpec({
    unnested : always(0),
    nested   : [ { sum : add } ],
  })(1, 2)

  expect(result).toEqual({
    unnested : 0,
    nested   : [ { sum : 3 } ],
  })
})

test('works with arrays of spec objects', () => {
  const result = applySpec([ { sum : add } ])(1, 2)

  expect(result).toEqual([ { sum : 3 } ])
})

test('works with arrays of functions', () => {
  const result = applySpec([ map(prop('a')), map(prop('b')) ])([
    {
      a : 'a1',
      b : 'b1',
    },
    {
      a : 'a2',
      b : 'b2',
    },
  ])
  const expected = [
    [ 'a1', 'a2' ],
    [ 'b1', 'b2' ],
  ]
  expect(result).toEqual(expected)
})

test('works with a spec defining a map key', () => {
  expect(applySpec({ map : prop('a') })({ a : 1 })).toEqual({ map : 1 })
})

test('cannot retains the highest arity', () => {
  const f = applySpec({
    f1 : nAry(2, T),
    f2 : nAry(5, T),
  })
  const fRamda = applySpecRamda({
    f1 : nAry(2, T),
    f2 : nAry(5, T),
  })
  expect(f.length).toBe(0)
  expect(fRamda.length).toBe(5)
})

test('returns a curried function', () => {
  expect(applySpec({ sum : add })(1)(2)).toEqual({ sum : 3 })
})

// Additional tests
// ============================================
test('arity', () => {
  const spec = {
    one   : x1 => x1,
    two   : (x1, x2) => x1 + x2,
    three : (
      x1, x2, x3
    ) => x1 + x2 + x3,
  }
  expect(applySpec(
    spec, 1, 2, 3
  )).toEqual({
    one   : 1,
    two   : 3,
    three : 6,
  })
})

test('arity over 5 arguments', () => {
  const spec = {
    one   : x1 => x1,
    two   : (x1, x2) => x1 + x2,
    three : (
      x1, x2, x3
    ) => x1 + x2 + x3,
    four : (
      x1, x2, x3, x4
    ) => x1 + x2 + x3 + x4,
    five : (
      x1, x2, x3, x4, x5
    ) => x1 + x2 + x3 + x4 + x5,
  }
  expect(applySpec(
    spec, 1, 2, 3, 4, 5
  )).toEqual({
    one   : 1,
    two   : 3,
    three : 6,
    four  : 10,
    five  : 15,
  })
})

test('curried', () => {
  const spec = {
    one   : x1 => x1,
    two   : (x1, x2) => x1 + x2,
    three : (
      x1, x2, x3
    ) => x1 + x2 + x3,
  }
  expect(applySpec(spec)(1)(2)(3)).toEqual({
    one   : 1,
    two   : 3,
    three : 6,
  })
})

test('curried over 5 arguments', () => {
  const spec = {
    one   : x1 => x1,
    two   : (x1, x2) => x1 + x2,
    three : (
      x1, x2, x3
    ) => x1 + x2 + x3,
    four : (
      x1, x2, x3, x4
    ) => x1 + x2 + x3 + x4,
    five : (
      x1, x2, x3, x4, x5
    ) => x1 + x2 + x3 + x4 + x5,
  }
  expect(applySpec(spec)(1)(2)(3)(4)(5)).toEqual({
    one   : 1,
    two   : 3,
    three : 6,
    four  : 10,
    five  : 15,
  })
})

test('undefined property', () => {
  const spec = { prop : path([ 'property', 'doesnt', 'exist' ]) }
  expect(applySpec(spec, {})).toEqual({ prop : undefined })
})

test('restructure json object', () => {
  const spec = {
    id          : path('user.id'),
    name        : path('user.firstname'),
    profile     : path('user.profile'),
    doesntExist : path('user.profile.doesntExist'),
    info        : { views : compose(inc, prop('views')) },
    type        : always('playa'),
  }

  const data = {
    user : {
      id        : 1337,
      firstname : 'john',
      lastname  : 'shaft',
      profile   : 'shaft69',
    },
    views : 42,
  }

  expect(applySpec(spec, data)).toEqual({
    id          : 1337,
    name        : 'john',
    profile     : 'shaft69',
    doesntExist : undefined,
    info        : { views : 43 },
    type        : 'playa',
  })
})

---------------

assoc

assoc<T, U, K extends string>(prop: K, val: T, obj: U): Record<K, T> & U

It makes a shallow clone of obj with setting or overriding the property prop with newValue.

💥 This copies and flattens prototype properties onto the new object as well. All non-primitive properties are copied by reference.

R.assoc('c', 3, {a: 1, b: 2})
// => {a: 1, b: 2, c: 3}

Try this R.assoc example in Rambda REPL

R.assoc source
import { curry } from './curry'

function assocFn(
  prop, newValue, obj
){
  return Object.assign(
    {}, obj, { [ prop ] : newValue }
  )
}

export const assoc = curry(assocFn)
Tests
import { assoc } from './assoc'

test('adds a key to an empty object', () => {
  expect(assoc(
    'a', 1, {}
  )).toEqual({ a : 1 })
})

test('adds a key to a non-empty object', () => {
  expect(assoc(
    'b', 2, { a : 1 }
  )).toEqual({
    a : 1,
    b : 2,
  })
})

test('adds a key to a non-empty object - curry case 1', () => {
  expect(assoc('b', 2)({ a : 1 })).toEqual({
    a : 1,
    b : 2,
  })
})

test('adds a key to a non-empty object - curry case 2', () => {
  expect(assoc('b')(2, { a : 1 })).toEqual({
    a : 1,
    b : 2,
  })
})

test('adds a key to a non-empty object - curry case 3', () => {
  const result = assoc('b')(2)({ a : 1 })

  expect(result).toEqual({
    a : 1,
    b : 2,
  })
})

test('changes an existing key', () => {
  expect(assoc(
    'a', 2, { a : 1 }
  )).toEqual({ a : 2 })
})

test('undefined is considered an empty object', () => {
  expect(assoc(
    'a', 1, undefined
  )).toEqual({ a : 1 })
})

test('null is considered an empty object', () => {
  expect(assoc(
    'a', 1, null
  )).toEqual({ a : 1 })
})

test('value can be null', () => {
  expect(assoc(
    'a', null, null
  )).toEqual({ a : null })
})

test('value can be undefined', () => {
  expect(assoc(
    'a', undefined, null
  )).toEqual({ a : undefined })
})

test('assignment is shallow', () => {
  expect(assoc(
    'a', { b : 2 }, { a : { c : 3 } }
  )).toEqual({ a : { b : 2 } })
})

---------------

assocPath

assocPath<Output>(path: Path, newValue: any, obj: object): Output

It makes a shallow clone of obj with setting or overriding with newValue the property found with path.

const path = 'b.c'
const newValue = 2
const obj = { a: 1 }

R.assocPath(path, newValue, obj)
// => { a : 1, b : { c : 2 }}

Try this R.assocPath example in Rambda REPL

R.assocPath source
import { _isArray } from './_internals/_isArray'
import { _isInteger } from './_internals/_isInteger'
import { assoc } from './assoc'
import { curry } from './curry'

function assocPathFn(
  path, newValue, input
){
  const pathArrValue =
    typeof path === 'string' ?
      path.split('.').map(x => _isInteger(Number(x)) ? Number(x) : x) :
      path
  if (pathArrValue.length === 0){
    return newValue
  }

  const index = pathArrValue[ 0 ]
  if (pathArrValue.length > 1){
    const condition =
      typeof input !== 'object' ||
      input === null ||
      !input.hasOwnProperty(index)

    const nextinput = condition ?
      _isInteger(pathArrValue[ 1 ]) ?
        [] :
        {} :
      input[ index ]

    newValue = assocPathFn(
      Array.prototype.slice.call(pathArrValue, 1),
      newValue,
      nextinput
    )
  }

  if (_isInteger(index) && _isArray(input)){
    const arr = input.slice()
    arr[ index ] = newValue

    return arr
  }

  return assoc(
    index, newValue, input
  )
}

export const assocPath = curry(assocPathFn)
Tests
import { assocPath } from './assocPath'

test('string can be used as path input', () => {
  const testObj = {
    a : [ { b : 1 }, { b : 2 } ],
    d : 3,
  }
  const result = assocPath(
    'a.0.b', 10, testObj
  )
  const expected = {
    a : [ { b : 10 }, { b : 2 } ],
    d : 3,
  }
  expect(result).toEqual(expected)
})

test('bug', () => {
  /*
    https://github.com/selfrefactor/rambda/issues/524
  */
  const state = {}

  const withDateLike = assocPath(
    [ 'outerProp', '2020-03-10' ],
    { prop : 2 },
    state
  )
  const withNumber = assocPath(
    [ 'outerProp', '5' ], { prop : 2 }, state
  )

  const withDateLikeExpected = { outerProp : { '2020-03-10' : { prop : 2 } } }
  const withNumberExpected = { outerProp : { 5 : { prop : 2 } } }
  expect(withDateLike).toEqual(withDateLikeExpected)
  expect(withNumber).toEqual(withNumberExpected)
})

test('adds a key to an empty object', () => {
  expect(assocPath(
    [ 'a' ], 1, {}
  )).toEqual({ a : 1 })
})

test('adds a key to a non-empty object', () => {
  expect(assocPath(
    'b', 2, { a : 1 }
  )).toEqual({
    a : 1,
    b : 2,
  })
})

test('adds a nested key to a non-empty object', () => {
  expect(assocPath(
    'b.c', 2, { a : 1 }
  )).toEqual({
    a : 1,
    b : { c : 2 },
  })
})

test('adds a nested key to a nested non-empty object - curry case 1', () => {
  expect(assocPath('b.d',
    3)({
    a : 1,
    b : { c : 2 },
  })).toEqual({
    a : 1,
    b : {
      c : 2,
      d : 3,
    },
  })
})

test('adds a key to a non-empty object - curry case 1', () => {
  expect(assocPath('b', 2)({ a : 1 })).toEqual({
    a : 1,
    b : 2,
  })
})

test('adds a nested key to a non-empty object - curry case 1', () => {
  expect(assocPath('b.c', 2)({ a : 1 })).toEqual({
    a : 1,
    b : { c : 2 },
  })
})

test('adds a key to a non-empty object - curry case 2', () => {
  expect(assocPath('b')(2, { a : 1 })).toEqual({
    a : 1,
    b : 2,
  })
})

test('adds a key to a non-empty object - curry case 3', () => {
  const result = assocPath('b')(2)({ a : 1 })

  expect(result).toEqual({
    a : 1,
    b : 2,
  })
})

test('changes an existing key', () => {
  expect(assocPath(
    'a', 2, { a : 1 }
  )).toEqual({ a : 2 })
})

test('undefined is considered an empty object', () => {
  expect(assocPath(
    'a', 1, undefined
  )).toEqual({ a : 1 })
})

test('null is considered an empty object', () => {
  expect(assocPath(
    'a', 1, null
  )).toEqual({ a : 1 })
})

test('value can be null', () => {
  expect(assocPath(
    'a', null, null
  )).toEqual({ a : null })
})

test('value can be undefined', () => {
  expect(assocPath(
    'a', undefined, null
  )).toEqual({ a : undefined })
})

test('assignment is shallow', () => {
  expect(assocPath(
    'a', { b : 2 }, { a : { c : 3 } }
  )).toEqual({ a : { b : 2 } })
})

test('empty array as path', () => {
  const result = assocPath(
    [], 3, {
      a : 1,
      b : 2,
    }
  )
  expect(result).toEqual(3)
})

test('happy', () => {
  const expected = { foo : { bar : { baz : 42 } } }
  const result = assocPath(
    [ 'foo', 'bar', 'baz' ], 42, { foo : null }
  )
  expect(result).toEqual(expected)
})

---------------

both

both(pred1: Pred, pred2: Pred): Pred

It returns a function with input argument.

This function will return true, if both firstCondition and secondCondition return true when input is passed as their argument.

const firstCondition = x => x > 10
const secondCondition = x => x < 20
const fn = R.both(secondCondition)

const result = [fn(15), fn(30)]
// => [true, false]

Try this R.both example in Rambda REPL

R.both source
export function both(f, g){
  if (arguments.length === 1) return _g => both(f, _g)

  return (...input) => f(...input) && g(...input)
}
Tests
import { both } from './both'

const firstFn = val => val > 0
const secondFn = val => val < 10

test('with curry', () => {
  expect(both(firstFn)(secondFn)(17)).toBeFalse()
})

test('without curry', () => {
  expect(both(firstFn, secondFn)(7)).toBeTrue()
})

test('with multiple inputs', () => {
  const between = function (
    a, b, c
  ){
    return a < b && b < c
  }
  const total20 = function (
    a, b, c
  ){
    return a + b + c === 20
  }
  const fn = both(between, total20)
  expect(fn(
    5, 7, 8
  )).toBeTrue()
})

test('skip evaluation of the second expression', () => {
  let effect = 'not evaluated'
  const F = function (){
    return false
  }
  const Z = function (){
    effect = 'Z got evaluated'
  }
  both(F, Z)()

  expect(effect).toBe('not evaluated')
})

---------------

chain

chain<T, U>(fn: (n: T) => readonly U[], list: readonly T[]): readonly U[]

The method is also known as flatMap.

const duplicate = n => [ n, n ]
const list = [ 1, 2, 3 ]

const result = chain(duplicate, list)
// => [ 1, 1, 2, 2, 3, 3 ]

Try this R.chain example in Rambda REPL

R.chain source
export function chain(fn, list){
  if (arguments.length === 1){
    return _list => chain(fn, _list)
  }

  return [].concat(...list.map(fn))
}
Tests
import { chain } from './chain'

const duplicate = n => [ n, n ]

test('happy', () => {
  const fn = x => [ x * 2 ]
  const list = [ 1, 2, 3 ]

  const result = chain(fn, list)

  expect(result).toEqual([ 2, 4, 6 ])
})

test('maps then flattens one level', () => {
  expect(chain(duplicate, [ 1, 2, 3 ])).toEqual([ 1, 1, 2, 2, 3, 3 ])
})

test('maps then flattens one level - curry', () => {
  expect(chain(duplicate)([ 1, 2, 3 ])).toEqual([ 1, 1, 2, 2, 3, 3 ])
})

test('flattens only one level', () => {
  const nest = n => [ [ n ] ]
  expect(chain(nest, [ 1, 2, 3 ])).toEqual([ [ 1 ], [ 2 ], [ 3 ] ])
})

---------------

clamp

clamp(min: number, max: number, input: number): number

Restrict a number input to be within min and max limits.

If input is bigger than max, then the result is max.

If input is smaller than min, then the result is min.

const result = [
  R.clamp(0, 10, 5), 
  R.clamp(0, 10, -1),
  R.clamp(0, 10, 11)
]
// => [5, 0, 10]

Try this R.clamp example in Rambda REPL

R.clamp source
import { curry } from './curry'

function clampFn(
  min, max, input
){
  if (min > max){
    throw new Error('min must not be greater than max in clamp(min, max, value)')
  }
  if (input >= min && input <= max) return input

  if (input > max) return max
  if (input < min) return min
}

export const clamp = curry(clampFn)
Tests
import { clamp } from './clamp'

test('when min is greater than max', () => {
  expect(() => clamp(
    -5, -10, 5
  )).toThrowWithMessage(Error,
    'min must not be greater than max in clamp(min, max, value)')
})

test('rambda specs', () => {
  expect(clamp(
    1, 10, 0
  )).toEqual(1)
  expect(clamp(
    3, 12, 1
  )).toEqual(3)
  expect(clamp(
    -15, 3, -100
  )).toEqual(-15)
  expect(clamp(
    1, 10, 20
  )).toEqual(10)
  expect(clamp(
    3, 12, 23
  )).toEqual(12)
  expect(clamp(
    -15, 3, 16
  )).toEqual(3)
  expect(clamp(
    1, 10, 4
  )).toEqual(4)
  expect(clamp(
    3, 12, 6
  )).toEqual(6)
  expect(clamp(
    -15, 3, 0
  )).toEqual(0)
})

---------------

clone

clone<T>(input: T): T

It creates a deep copy of the input, which may contain (nested) Arrays and Objects, Numbers, Strings, Booleans and Dates.

const objects = [{a: 1}, {b: 2}];
const objectsClone = R.clone(objects);

const result = [
  R.equals(objects, objectsClone),
  R.equals(objects[0], objectsClone[0]),
] // => [ true, true ]

Try this R.clone example in Rambda REPL

R.clone source
import { _isArray } from './_internals/_isArray'

export function clone(input){
  const out = _isArray(input) ? Array(input.length) : {}
  if (input && input.getTime) return new Date(input.getTime())

  for (const key in input){
    const v = input[ key ]
    out[ key ] =
      typeof v === 'object' && v !== null ?
        v.getTime ?
          new Date(v.getTime()) :
          clone(v) :
        v
  }

  return out
}
Tests
import assert from 'assert'

import { clone } from './clone'
import { equals } from './equals'

test('with array', () => {
  const arr = [
    {
      b : 2,
      c : 'foo',
      d : [ 1, 2, 3 ],
    },
    1,
    new Date(),
    null,
  ]
  expect(clone(arr)).toEqual(arr)
})

test('with object', () => {
  const obj = {
    a : 1,
    b : 2,
    c : 3,
    d : [ 1, 2, 3 ],
    e : new Date(),
  }
  expect(clone(obj)).toEqual(obj)
})

test('with date', () => {
  const date = new Date(
    2014, 10, 14, 23, 59, 59, 999
  )

  const cloned = clone(date)
  assert.notStrictEqual(date, cloned)
  expect(cloned).toEqual(new Date(
    2014, 10, 14, 23, 59, 59, 999
  ))

  expect(cloned.getDay()).toEqual(5)
})

test('with R.equals', () => {
  const objects = [ { a : 1 }, { b : 2 } ]

  const objectsClone = clone(objects)

  const result = [
    equals(objects, objectsClone),
    equals(objects[ 0 ], objectsClone[ 0 ]),
  ]
  expect(result).toEqual([ true, true ])
})

---------------

complement

complement<T extends readonly any[]>(pred: (...args: T) => boolean): (...args: T) => boolean

It returns inverted version of origin function that accept input as argument.

The return value of inverted is the negative boolean value of origin(input).

const origin = x => x > 5
const inverted = complement(origin)

const result = [
  origin(7),
  inverted(7)
] => [ true, false ]

Try this R.complement example in Rambda REPL

R.complement source
export function complement(fn){
  return (...input) => !fn(...input)
}
Tests
import { complement } from './complement'

test('happy', () => {
  const fn = complement(x => x.length === 0)

  expect(fn([ 1, 2, 3 ])).toBeTrue()
})

test('with multiple parameters', () => {
  const between = function (
    a, b, c
  ){
    return a < b && b < c
  }
  const f = complement(between)
  expect(f(
    4, 5, 11
  )).toEqual(false)
  expect(f(
    12, 2, 6
  )).toEqual(true)
})

---------------

compose

It performs right-to-left function composition.

const result = R.compose(
  R.map(x => x * 2),
  R.filter(x => x > 2)
)([1, 2, 3, 4])

// => [6, 8]

Try this R.compose example in Rambda REPL

---------------

composeAsync

composeAsync<Out>(
  ...fns: readonly (Async<any> | Func<any>)[]
): (input: any) => Promise<Out>

Asynchronous version of R.compose

💥 It doesn't work with promises or function returning promises such as const foo = input => new Promise(...).

const add = async x => {
  await R.delay(100)
  return x + 1
}
const multiply = async x => {
  await R.delay(100)
  return x * 2 
}

const result = await R.composeAsync(
  add,
  multiply
)(1)
// `result` resolves to `3`

Try this R.composeAsync example in Rambda REPL

R.composeAsync source
import { type } from './type'

export function composeAsync(...inputArguments){
  return async function (startArgument){
    let argumentsToPass = startArgument

    while (inputArguments.length !== 0){
      const fn = inputArguments.pop()
      const typeFn = type(fn)

      if (typeFn === 'Async'){
        argumentsToPass = await fn(argumentsToPass)
      } else {
        argumentsToPass = fn(argumentsToPass)
        if (type(argumentsToPass) === 'Promise'){
          argumentsToPass = await argumentsToPass
        }
      }
    }

    return argumentsToPass
  }
}
Tests
import { composeAsync } from './composeAsync'
import { delay } from './delay'

async function identity(x){
  await delay(100)

  return x
}

test('happy', async () => {
  const fn1 = async x => {
    await delay(100)

    return x.map(xx => xx + 1)
  }
  const fn2 = async x => {
    await delay(100)

    return x.map(xx => xx * 2)
  }
  const result = await composeAsync(fn1,
    fn2)(await Promise.all([ identity(1), identity(2), identity(3) ]))

  expect(result).toEqual([ 3, 5, 7 ])
})

const delayFn = ms =>
  new Promise(resolve => {
    resolve(ms + 1)
  })

test('with function returning promise', async () => {
  const result = await composeAsync(
    x => x,
    x => x + 1,
    delayFn,
    x => x
  )(1)

  expect(result).toEqual(3)
})

test('throw error', async () => {
  const fn = async () => {
    await delay(1)
    JSON.parse('{foo')
  }

  let didThrow = false
  try {
    await composeAsync(fn, x => x + 1)(20)
  } catch (e){
    didThrow = true
  }

  expect(didThrow).toBeTrue()
})

---------------

concat

concat<T>(x: readonly T[], y: readonly T[]): readonly T[]

It returns a new string or array, which is the result of merging x and y.

R.concat([1, 2])([3, 4]) // => [1, 2, 3, 4]
R.concat('foo', 'bar') // => 'foobar'

Try this R.concat example in Rambda REPL

R.concat source
export function concat(x, y){
  if (arguments.length === 1) return _y => concat(x, _y)

  return typeof x === 'string' ? `${ x }${ y }` : [ ...x, ...y ]
}
Tests
import { concat } from './concat'

test('happy', () => {
  const arr1 = [ 'a', 'b', 'c' ]
  const arr2 = [ 'd', 'e', 'f' ]

  const a = concat(arr1, arr2)
  const b = concat(arr1)(arr2)
  const expectedResult = [ 'a', 'b', 'c', 'd', 'e', 'f' ]

  expect(a).toEqual(expectedResult)
  expect(b).toEqual(expectedResult)
})

test('with strings', () => {
  expect(concat('ABC', 'DEF')).toEqual('ABCDEF')
})

---------------

cond

cond(conditions: readonly (readonly [Pred, (...a: readonly any[]) => any])[]): (...x: readonly any[]) => any

It takes list with conditions and returns a new function fn that expects input as argument.

This function will start evaluating the conditions in order to find the first winner(order of conditions matter).

The winner is this condition, which left side returns true when input is its argument. Then the evaluation of the right side of the winner will be the final result.

If no winner is found, then fn returns undefined.

const fn = R.cond([
  [ x => x > 25, R.always('more than 25') ],
  [ x => x > 15, R.always('more than 15') ],
  [ R.T, x => `${x} is nothing special` ],
])

const result = [
  fn(30),
  fn(20),
  fn(10),
] 
// => ['more than 25', 'more than 15', '10 is nothing special']

Try this R.cond example in Rambda REPL

R.cond source
export function cond(conditions){
  return input => {
    let done = false
    let toReturn
    conditions.forEach(([ predicate, resultClosure ]) => {
      if (!done && predicate(input)){
        done = true
        toReturn = resultClosure(input)
      }
    })

    return toReturn
  }
}
Tests
import { always } from './always'
import { cond } from './cond'
import { equals } from './equals'
import { T } from './T'

test('returns a function', () => {
  expect(typeof cond([])).toEqual('function')
})

test('returns a conditional function', () => {
  const fn = cond([
    [ equals(0), always('water freezes at 0°C') ],
    [ equals(100), always('water boils at 100°C') ],
    [
      T,
      function (temp){
        return 'nothing special happens at ' + temp + '°C'
      },
    ],
  ])
  expect(fn(0)).toEqual('water freezes at 0°C')
  expect(fn(50)).toEqual('nothing special happens at 50°C')
  expect(fn(100)).toEqual('water boils at 100°C')
})

test('no winner', () => {
  const fn = cond([
    [ equals('foo'), always(1) ],
    [ equals('bar'), always(2) ],
  ])
  expect(fn('quux')).toEqual(undefined)
})

test('predicates are tested in order', () => {
  const fn = cond([
    [ T, always('foo') ],
    [ T, always('bar') ],
    [ T, always('baz') ],
  ])
  expect(fn()).toEqual('foo')
})

---------------

converge

converge(after: ((...a: readonly any[]) => any), fns: ReadonlyArray<((...x: readonly any[]) => any)>): (...y: readonly any[]) => any

Accepts a converging function and a list of branching functions and returns a new function. When invoked, this new function is applied to some arguments, each branching function is applied to those same arguments. The results of each branching function are passed as arguments to the converging function to produce the return value.

💥 Explanation is taken from Ramda documentation

const result = R.converge(R.multiply)([ R.add(1), R.add(3) ])(2)
// => 15

Try this R.converge example in Rambda REPL

R.converge source
import { curryN } from './curryN'
import { map } from './map'
import { max } from './max'
import { reduce } from './reduce'

export function converge(fn, transformers){
  if (arguments.length === 1)
    return _transformers => converge(fn, _transformers)

  const highestArity = reduce(
    (a, b) => max(a, b.length), 0, transformers
  )

  return curryN(highestArity, function (){
    return fn.apply(this,
      map(g => g.apply(this, arguments), transformers))
  })
}
Tests
import { add } from './add'
import { converge } from './converge'
import { multiply } from './multiply'

const f1 = converge(multiply, [ a => a + 1, a => a + 10 ])
const f2 = converge(multiply, [ a => a + 1, (a, b) => a + b + 10 ])
const f3 = converge(multiply, [ a => a + 1, (
  a, b, c
) => a + b + c + 10 ])

test('happy', () => {
  expect(f2(6, 7)).toEqual(161)
})

test('passes the results of applying the arguments individually', () => {
  const result = converge(multiply)([ add(1), add(3) ])(2)
  expect(result).toEqual(15)
})

test('returns a function with the length of the longest argument', () => {
  expect(f1.length).toEqual(1)
  expect(f2.length).toEqual(2)
  expect(f3.length).toEqual(3)
})

test('passes context to its functions', () => {
  const a = function (x){
    return this.f1(x)
  }
  const b = function (x){
    return this.f2(x)
  }
  const c = function (x, y){
    return this.f3(x, y)
  }
  const d = converge(c, [ a, b ])
  const context = {
    f1 : add(1),
    f2 : add(2),
    f3 : add,
  }
  expect(a.call(context, 1)).toEqual(2)
  expect(b.call(context, 1)).toEqual(3)
  expect(d.call(context, 1)).toEqual(5)
})

test('works with empty functions list', () => {
  const fn = converge(function (){
    return arguments.length
  }, [])
  expect(fn.length).toEqual(0)
  expect(fn()).toEqual(0)
})

---------------

count

count<T>(searchFor: T, list: readonly any[]): number

It counts how many times searchFor is within list according to R.equals.

const list = [1, {a:1}, 1, 'foo']
const searchFor = 1

const result = R.count(searchFor, list)
// => 2

Try this R.count example in Rambda REPL

R.count source
import { _isArray } from './_internals/_isArray'
import { equals } from './equals'

export function count(searchFor, list){
  if (arguments.length === 1){
    return _list => count(searchFor, _list)
  }
  if (!_isArray(list)) return 0

  return list.filter(x => equals(x, searchFor)).length
}
Tests
import { count } from './count'

test('when 0', () => {
  const target = { a : 1 }
  const list = []

  const result = count(target, list)
  const expectedResult = 0

  expect(result).toEqual(expectedResult)
})

test('when 2', () => {
  const target = { a : 1 }
  const list = [ 2, 1, 2, { a : 1 }, 22, { a : 1 } ]

  const result = count(target)(list)
  const expectedResult = 2

  expect(result).toEqual(expectedResult)
})

---------------

curry

It expects a function as input and returns its curried version.

const fn = (a, b, c) => a + b + c
const curried = R.curry(fn)
const sum = curried(1,2)

const result = sum(3) // => 6

Try this R.curry example in Rambda REPL

---------------

curryN

It returns a curried equivalent of the provided function, with the specified arity.

---------------

debounce

debounce<T, U>(fn: (input: T) => U, ms: number): (input: T) => U

It creates a debounced function that delays invoking fn until after wait milliseconds ms have elapsed since the last time the debounced function was invoked.

💥 Description is taken from Lodash docs

let counter = 0
const increment = () => {
  counter++
}

const debounced = R.debounce(increment, 1000)

const result = async function(){
  debounced()
  await R.delay(500)
  debounced()
  await R.delay(800)
  console.log(counter) // => 0

  await R.delay(1200)
  console.log(counter) // => 1

  return counter
}
// `result` resolves to `1`

Try this R.debounce example in Rambda REPL

R.debounce source
export function debounce(
  func, ms, immediate = false
){
  let timeout

  return function (...input){
    const later = function (){
      timeout = null
      if (!immediate){
        func.apply(null, input)
      }
    }
    const callNow = immediate && !timeout
    clearTimeout(timeout)
    timeout = setTimeout(later, ms)
    if (callNow){
      func.apply(null, input)
    }
  }
}
Tests
import { debounce } from './debounce'
import { delay } from './delay'

test('happy', async () => {
  let counter = 0
  let aHolder
  let bHolder

  const inc = (a, b) => {
    aHolder = a
    bHolder = b
    counter++
  }
  const incWrapped = debounce(inc, 500)

  incWrapped(1, 2)
  expect(counter).toBe(0)
  expect(aHolder).toBe(undefined)
  expect(bHolder).toBe(undefined)

  await delay(200)

  incWrapped(2, 3)
  expect(counter).toBe(0)

  await delay(200)

  incWrapped(3, 4)
  expect(counter).toBe(0)
  expect(aHolder).toBe(undefined)
  expect(bHolder).toBe(undefined)

  await delay(200)
  incWrapped(5, 6)
  expect(counter).toBe(0)

  await delay(700)
  expect(counter).toBe(1)
  expect(aHolder).toBe(5)
  expect(bHolder).toBe(6)
})

test('immediate debounce', async () => {
  let counter = 0
  const inc = () => {
    counter++
  }

  const incWrapped = debounce(
    inc, 500, true
  )
  incWrapped()
  expect(counter).toBe(1)
  await delay(200)
  incWrapped()
  expect(counter).toBe(1)
  await delay(200)
  incWrapped()
  expect(counter).toBe(1)
  await delay(700)
  incWrapped()
  expect(counter).toBe(2)
})

---------------

dec

It decrements a number.

---------------

defaultTo

defaultTo<T>(defaultValue: T, input: T | null | undefined): T

It returns defaultValue, if all of inputArguments are undefined, null or NaN.

Else, it returns the first truthy inputArguments instance(from left to right).

💥 Rambda's defaultTo accept indefinite number of arguments when non curried, i.e. R.defaultTo(2, foo, bar, baz).

R.defaultTo('foo', 'bar') // => 'bar'
R.defaultTo('foo', undefined) // => 'foo'

// Important - emtpy string is not falsy value(same as Ramda)
R.defaultTo('foo', '') // => 'foo'

Try this R.defaultTo example in Rambda REPL

R.defaultTo source
function isFalsy(input){
  return (
    input === undefined ||
    input === null ||
    Number.isNaN(input) === true
  )
}

export function defaultTo(defaultArgument, input){
  if (arguments.length === 1){
    return _input =>
      defaultTo(defaultArgument, _input)
  }

  return isFalsy(input) ? defaultArgument : input
}
Tests
import { defaultTo } from './defaultTo'

test('with undefined', () => {
  expect(defaultTo('foo')(undefined)).toEqual('foo')
})

test('with null', () => {
  expect(defaultTo('foo')(null)).toEqual('foo')
})

test('with NaN', () => {
  expect(defaultTo('foo')(NaN)).toEqual('foo')
})

test('with empty string', () => {
  expect(defaultTo('foo', '')).toEqual('')
})

test('with false', () => {
  expect(defaultTo('foo', false)).toEqual(false)
})

test('when inputArgument passes initial check', () => {
  expect(defaultTo('foo', 'bar')).toEqual('bar')
})

---------------

delay

delay(ms: number): Promise<'RAMBDAX_DELAY'>

setTimeout as a promise that resolves to R.DELAY variable after ms milliseconds.

const result = R.delay(1000)
// `result` resolves to `RAMBDAX_DELAY`

Try this R.delay example in Rambda REPL

R.delay source
export const DELAY = 'RAMBDAX_DELAY'

export function delay(ms){
  return new Promise(resolve => {
    setTimeout(() => {
      resolve(DELAY)
    }, ms)
  })
}
Tests
import { DELAY, delay } from './delay'

test('usage with variables', async () => {
  expect(await delay(500)).toBe(DELAY)
})

---------------

difference

difference<T>(a: readonly T[], b: readonly T[]): readonly T[]

It returns the uniq set of all elements in the first list a not contained in the second list b.

R.equals is used to determine equality.

const a = [ 1, 2, 3, 4 ]
const b = [ 3, 4, 5, 6 ]

const result = difference(a, b)
// => [ 1, 2 ]

Try this R.difference example in Rambda REPL

R.difference source
import { includes } from './includes'
import { uniq } from './uniq'

export function difference(a, b){
  if (arguments.length === 1) return _b => difference(a, _b)

  return uniq(a).filter(aInstance => !includes(aInstance, b))
}
Tests
import { difference } from './difference'

test('difference', () => {
  const a = [ 1, 2, 3, 4 ]
  const b = [ 3, 4, 5, 6 ]
  expect(difference(a)(b)).toEqual([ 1, 2 ])

  expect(difference([], [])).toEqual([])
})

test('difference with objects', () => {
  const a = [ { id : 1 }, { id : 2 }, { id : 3 }, { id : 4 } ]
  const b = [ { id : 3 }, { id : 4 }, { id : 5 }, { id : 6 } ]
  expect(difference(a, b)).toEqual([ { id : 1 }, { id : 2 } ])
})

test('no duplicates in first list', () => {
  const M2 = [ 1, 2, 3, 4, 1, 2, 3, 4 ]
  const N2 = [ 3, 3, 4, 4, 5, 5, 6, 6 ]
  expect(difference(M2, N2)).toEqual([ 1, 2 ])
})

test('should use R.equals', () => {
  expect(difference([ NaN ], [ NaN ]).length).toEqual(0)
})

---------------

dissoc

It returns a new object that does not contain property prop.

R.dissoc('b', {a: 1, b: 2, c: 3})
// => {a: 1, c: 3}

Try this R.dissoc example in Rambda REPL

---------------

divide

R.divide(71, 100) // => 0.71

Try this R.divide example in Rambda REPL

---------------

drop

drop<T>(howMany: number, input: readonly T[]): readonly T[]

It returns howMany items dropped from beginning of list or string input.

R.drop(2, ['foo', 'bar', 'baz']) // => ['baz']
R.drop(2, 'foobar')  // => 'obar'

Try this R.drop example in Rambda REPL

R.drop source
export function drop(howManyToDrop, listOrString){
  if (arguments.length === 1) return _list => drop(howManyToDrop, _list)

  return listOrString.slice(howManyToDrop > 0 ? howManyToDrop : 0)
}
Tests
import assert from 'assert'

import { drop } from './drop'

test('with array', () => {
  expect(drop(2)([ 'foo', 'bar', 'baz' ])).toEqual([ 'baz' ])
  expect(drop(3, [ 'foo', 'bar', 'baz' ])).toEqual([])
  expect(drop(4, [ 'foo', 'bar', 'baz' ])).toEqual([])
})

test('with string', () => {
  expect(drop(3, 'rambda')).toEqual('bda')
})

test('with non-positive count', () => {
  expect(drop(0, [ 1, 2, 3 ])).toEqual([ 1, 2, 3 ])
  expect(drop(-1, [ 1, 2, 3 ])).toEqual([ 1, 2, 3 ])
  expect(drop(-Infinity, [ 1, 2, 3 ])).toEqual([ 1, 2, 3 ])
})

test('should return copy', () => {
  const xs = [ 1, 2, 3 ]

  assert.notStrictEqual(drop(0, xs), xs)
  assert.notStrictEqual(drop(-1, xs), xs)
})

---------------

dropLast

dropLast<T>(howMany: number, input: readonly T[]): readonly T[]

It returns howMany items dropped from the end of list or string input.

R.dropLast(2, ['foo', 'bar', 'baz']) // => ['foo']
R.dropLast(2, 'foobar')  // => 'foob'

Try this R.dropLast example in Rambda REPL

R.dropLast source
export function dropLast(howManyToDrop, listOrString){
  if (arguments.length === 1){
    return _listOrString => dropLast(howManyToDrop, _listOrString)
  }

  return howManyToDrop > 0 ?
    listOrString.slice(0, -howManyToDrop) :
    listOrString.slice()
}
Tests
import assert from 'assert'

import { dropLast } from './dropLast'

test('with array', () => {
  expect(dropLast(2)([ 'foo', 'bar', 'baz' ])).toEqual([ 'foo' ])
  expect(dropLast(3, [ 'foo', 'bar', 'baz' ])).toEqual([])
  expect(dropLast(4, [ 'foo', 'bar', 'baz' ])).toEqual([])
})

test('with string', () => {
  expect(dropLast(3, 'rambda')).toEqual('ram')
})

test('with non-positive count', () => {
  expect(dropLast(0, [ 1, 2, 3 ])).toEqual([ 1, 2, 3 ])
  expect(dropLast(-1, [ 1, 2, 3 ])).toEqual([ 1, 2, 3 ])
  expect(dropLast(-Infinity, [ 1, 2, 3 ])).toEqual([ 1, 2, 3 ])
})

test('should return copy', () => {
  const xs = [ 1, 2, 3 ]

  assert.notStrictEqual(dropLast(0, xs), xs)
  assert.notStrictEqual(dropLast(-1, xs), xs)
})

---------------

dropLastWhile

const list = [1, 2, 3, 4, 5];
const predicate = x => x >= 3

const result = dropLastWhile(predicate, list);
// => [1, 2]

Try this R.dropLastWhile example in Rambda REPL

---------------

dropRepeats

dropRepeats<T>(list: readonly T[]): readonly T[]

It removes any successive duplicates according to R.equals.

const result = R.dropRepeats([
  1, 
  1, 
  {a: 1}, 
  {a:1}, 
  1
])
// => [1, {a: 1}, 1]

Try this R.dropRepeats example in Rambda REPL

R.dropRepeats source
import { _isArray } from './_internals/_isArray'
import { equals } from './equals'

export function dropRepeats(list){
  if (!_isArray(list)){
    throw new Error(`${ list } is not a list`)
  }

  const toReturn = []

  list.reduce((prev, current) => {
    if (!equals(prev, current)){
      toReturn.push(current)
    }

    return current
  }, undefined)

  return toReturn
}
Tests
import { dropRepeats as dropRepeatsRamda } from 'ramda'

import { compareCombinations } from './_internals/testUtils'
import { add } from './add'
import { dropRepeats } from './dropRepeats'

const list = [ 1, 2, 2, 2, 3, 4, 4, 5, 5, 3, 2, 2, { a : 1 }, { a : 1 } ]
const listClean = [ 1, 2, 3, 4, 5, 3, 2, { a : 1 } ]

test('happy', () => {
  const result = dropRepeats(list)
  expect(result).toEqual(listClean)
})

const possibleLists = [
  [ add(1), async () => {}, [ 1 ], [ 1 ], [ 2 ], [ 2 ] ],
  [ add(1), add(1), add(2) ],
  [],
  1,
  /foo/g,
  Promise.resolve(1),
]

describe('brute force', () => {
  compareCombinations({
    firstInput : possibleLists,
    callback   : errorsCounters => {
      expect(errorsCounters).toMatchInlineSnapshot(`
        Object {
          "ERRORS_MESSAGE_MISMATCH": 0,
          "ERRORS_TYPE_MISMATCH": 0,
          "RESULTS_MISMATCH": 1,
          "SHOULD_NOT_THROW": 3,
          "SHOULD_THROW": 0,
        }
      `)
    },
    fn      : dropRepeats,
    fnRamda : dropRepeatsRamda,
  })
})

---------------

dropRepeatsWith

const list = [{a:1,b:2}, {a:1,b:3}, {a:2, b:4}]
const result = R.dropRepeatsWith(R.prop('a'))

// => [{a:1,b:2}, {a:2, b:4}]

Try this R.dropRepeatsWith example in Rambda REPL

---------------

dropWhile

const list = [1, 2, 3, 4]
const predicate = x => x < 3
const result = R.dropWhile(predicate, list)
// => [3, 4]

Try this R.dropWhile example in Rambda REPL

---------------

either

either(firstPredicate: Pred, secondPredicate: Pred): Pred

It returns a new predicate function from firstPredicate and secondPredicate inputs.

This predicate function will return true, if any of the two input predicates return true.

const firstPredicate = x => x > 10
const secondPredicate = x => x % 2 === 0
const predicate = R.either(firstPredicate, secondPredicate)

const result = [
  predicate(15),
  predicate(8),
  predicate(7),
]
// => [true, true, false]

Try this R.either example in Rambda REPL

R.either source
export function either(firstPredicate, secondPredicate){
  if (arguments.length === 1){
    return _secondPredicate => either(firstPredicate, _secondPredicate)
  }

  return (...input) =>
    Boolean(firstPredicate(...input) || secondPredicate(...input))
}
Tests
import { either } from './either'

test('with multiple inputs', () => {
  const between = function (
    a, b, c
  ){
    return a < b && b < c
  }
  const total20 = function (
    a, b, c
  ){
    return a + b + c === 20
  }
  const fn = either(between, total20)
  expect(fn(
    7, 8, 5
  )).toBeTrue()
})

test('skip evaluation of the second expression', () => {
  let effect = 'not evaluated'
  const F = function (){
    return true
  }
  const Z = function (){
    effect = 'Z got evaluated'
  }
  either(F, Z)()

  expect(effect).toBe('not evaluated')
})

test('case 1', () => {
  const firstFn = val => val > 0
  const secondFn = val => val * 5 > 10

  expect(either(firstFn, secondFn)(1)).toBeTrue()
})

test('case 2', () => {
  const firstFn = val => val > 0
  const secondFn = val => val === -10
  const fn = either(firstFn)(secondFn)

  expect(fn(-10)).toBeTrue()
})

---------------

endsWith

endsWith(target: string, str: string): boolean

Curried version of String.prototype.endsWith

💥 It doesn't work with arrays unlike its corresponding Ramda method.

const str = 'foo-bar'
const target = '-bar'

const result = R.endsWith(target, str)
// => true

Try this R.endsWith example in Rambda REPL

R.endsWith source
export function endsWith(target, str){
  if (arguments.length === 1) return _str => endsWith(target, _str)

  return str.endsWith(target)
}
Tests
import { endsWith } from './endsWith'

test('happy', () => {
  expect(endsWith('bar', 'foo-bar')).toBeTrue()
  expect(endsWith('baz')('foo-bar')).toBeFalse()
})

test('does not work with arrays', () => {
  expect(() => endsWith([ 'c' ], [ 'a', 'b', 'c' ])).toThrowWithMessage(Error,
    'str.endsWith is not a function')
})

---------------

eqProps

It returns true if property prop in obj1 is equal to property prop in obj2 according to R.equals.

const obj1 = {a: 1, b:2}
const obj2 = {a: 1, b:3}
const result = R.eqProps('a', obj1, obj2)
// => true

Try this R.eqProps example in Rambda REPL

---------------

equals

equals<T>(x: T, y: T): boolean

It deeply compares x and y and returns true if they are equal.

💥 It doesn't handle cyclical data structures and functions

R.equals(
  [1, {a:2}, [{b: 3}]],
  [1, {a:2}, [{b: 3}]]
) // => true

Try this R.equals example in Rambda REPL

R.equals source
import { type } from './type'

function parseError(maybeError){
  const typeofError = maybeError.__proto__.toString()
  if (![ 'Error', 'TypeError' ].includes(typeofError)) return []

  return [ typeofError, maybeError.message ]
}

function parseDate(maybeDate){
  if (!maybeDate.toDateString) return [ false ]

  return [ true, maybeDate.getTime() ]
}

function parseRegex(maybeRegex){
  if (maybeRegex.constructor !== RegExp) return [ false ]

  return [ true, maybeRegex.toString() ]
}

export function equals(a, b){
  if (arguments.length === 1) return _b => equals(a, _b)

  const aType = type(a)
  if (aType !== type(b)) return false
  if (aType === 'Function'){
    return a.name === undefined ? false : a.name === b.name
  }

  if ([ 'NaN', 'Undefined', 'Null' ].includes(aType)) return true

  if (aType === 'Number'){
    if (Object.is(-0, a) !== Object.is(-0, b)) return false

    return a.toString() === b.toString()
  }

  if ([ 'String', 'Boolean' ].includes(aType)){
    return a.toString() === b.toString()
  }

  if (aType === 'Array'){
    const aClone = Array.from(a)
    const bClone = Array.from(b)

    if (aClone.toString() !== bClone.toString()){
      return false
    }

    let loopArrayFlag = true
    aClone.forEach((aCloneInstance, aCloneIndex) => {
      if (loopArrayFlag){
        if (
          aCloneInstance !== bClone[ aCloneIndex ] &&
          !equals(aCloneInstance, bClone[ aCloneIndex ])
        ){
          loopArrayFlag = false
        }
      }
    })

    return loopArrayFlag
  }

  const aRegex = parseRegex(a)
  const bRegex = parseRegex(b)

  if (aRegex[ 0 ]){
    return bRegex[ 0 ] ? aRegex[ 1 ] === bRegex[ 1 ] : false
  } else if (bRegex[ 0 ]) return false

  const aDate = parseDate(a)
  const bDate = parseDate(b)

  if (aDate[ 0 ]){
    return bDate[ 0 ] ? aDate[ 1 ] === bDate[ 1 ] : false
  } else if (bDate[ 0 ]) return false

  const aError = parseError(a)
  const bError = parseError(b)

  if (aError[ 0 ]){
    return bError[ 0 ] ?
      aError[ 0 ] === bError[ 0 ] && aError[ 1 ] === bError[ 1 ] :
      false
  }

  if (aType === 'Object'){
    const aKeys = Object.keys(a)

    if (aKeys.length !== Object.keys(b).length){
      return false
    }

    let loopObjectFlag = true
    aKeys.forEach(aKeyInstance => {
      if (loopObjectFlag){
        const aValue = a[ aKeyInstance ]
        const bValue = b[ aKeyInstance ]

        if (aValue !== bValue && !equals(aValue, bValue)){
          loopObjectFlag = false
        }
      }
    })

    return loopObjectFlag
  }

  return false
}
Tests
import { equals } from './equals'

test('compare functions', () => {
  function foo(){}
  function bar(){}
  const baz = () => {}

  const expectTrue = equals(foo, foo)
  const expectFalseFirst = equals(foo, bar)
  const expectFalseSecond = equals(foo, baz)

  expect(expectTrue).toBeTrue()
  expect(expectFalseFirst).toBeFalse()
  expect(expectFalseSecond).toBeFalse()
})

test('with array of objects', () => {
  const list1 = [ { a : 1 }, [ { b : 2 } ] ]
  const list2 = [ { a : 1 }, [ { b : 2 } ] ]
  const list3 = [ { a : 1 }, [ { b : 3 } ] ]

  expect(equals(list1, list2)).toBeTrue()
  expect(equals(list1, list3)).toBeFalse()
})

test('with regex', () => {
  expect(equals(/s/, /s/)).toEqual(true)
  expect(equals(/s/, /d/)).toEqual(false)
  expect(equals(/a/gi, /a/gi)).toEqual(true)
  expect(equals(/a/gim, /a/gim)).toEqual(true)
  expect(equals(/a/gi, /a/i)).toEqual(false)
})

test('not a number', () => {
  expect(equals([ NaN ], [ NaN ])).toBeTrue()
})

test('new number', () => {
  expect(equals(new Number(0), new Number(0))).toEqual(true)
  expect(equals(new Number(0), new Number(1))).toEqual(false)
  expect(equals(new Number(1), new Number(0))).toEqual(false)
})

test('new string', () => {
  expect(equals(new String(''), new String(''))).toEqual(true)
  expect(equals(new String(''), new String('x'))).toEqual(false)
  expect(equals(new String('x'), new String(''))).toEqual(false)
  expect(equals(new String('foo'), new String('foo'))).toEqual(true)
  expect(equals(new String('foo'), new String('bar'))).toEqual(false)
  expect(equals(new String('bar'), new String('foo'))).toEqual(false)
})

test('new Boolean', () => {
  expect(equals(new Boolean(true), new Boolean(true))).toEqual(true)
  expect(equals(new Boolean(false), new Boolean(false))).toEqual(true)
  expect(equals(new Boolean(true), new Boolean(false))).toEqual(false)
  expect(equals(new Boolean(false), new Boolean(true))).toEqual(false)
})

test('new Error', () => {
  expect(equals(new Error('XXX'), {})).toEqual(false)
  expect(equals(new Error('XXX'), new TypeError('XXX'))).toEqual(false)
  expect(equals(new Error('XXX'), new Error('YYY'))).toEqual(false)
  expect(equals(new Error('XXX'), new Error('XXX'))).toEqual(true)
  expect(equals(new Error('XXX'), new TypeError('YYY'))).toEqual(false)
})

test('with dates', () => {
  expect(equals(new Date(0), new Date(0))).toEqual(true)
  expect(equals(new Date(1), new Date(1))).toEqual(true)
  expect(equals(new Date(0), new Date(1))).toEqual(false)
  expect(equals(new Date(1), new Date(0))).toEqual(false)
  expect(equals(new Date(0), {})).toEqual(false)
  expect(equals({}, new Date(0))).toEqual(false)
})

test('ramda spec', () => {
  expect(equals({}, {})).toEqual(true)

  expect(equals({
    a : 1,
    b : 2,
  },
  {
    a : 1,
    b : 2,
  })).toEqual(true)

  expect(equals({
    a : 2,
    b : 3,
  },
  {
    b : 3,
    a : 2,
  })).toEqual(true)

  expect(equals({
    a : 2,
    b : 3,
  },
  {
    a : 3,
    b : 3,
  })).toEqual(false)

  expect(equals({
    a : 2,
    b : 3,
    c : 1,
  },
  {
    a : 2,
    b : 3,
  })).toEqual(false)
})

test('works with boolean tuple', () => {
  expect(equals([ true, false ], [ true, false ])).toBeTrue()
  expect(equals([ true, false ], [ true, true ])).toBeFalse()
})

test('works with equal objects within array', () => {
  const objFirst = {
    a : {
      b : 1,
      c : 2,
      d : [ 1 ],
    },
  }
  const objSecond = {
    a : {
      b : 1,
      c : 2,
      d : [ 1 ],
    },
  }

  const x = [ 1, 2, objFirst, null, '', [] ]
  const y = [ 1, 2, objSecond, null, '', [] ]
  expect(equals(x, y)).toBeTrue()
})

test('works with different objects within array', () => {
  const objFirst = { a : { b : 1 } }
  const objSecond = { a : { b : 2 } }

  const x = [ 1, 2, objFirst, null, '', [] ]
  const y = [ 1, 2, objSecond, null, '', [] ]
  expect(equals(x, y)).toBeFalse()
})

test('works with undefined as second argument', () => {
  expect(equals(1, undefined)).toBeFalse()

  expect(equals(undefined, undefined)).toBeTrue()
})

test('various examples', () => {
  expect(equals([ 1, 2, 3 ])([ 1, 2, 3 ])).toBeTrue()

  expect(equals([ 1, 2, 3 ], [ 1, 2 ])).toBeFalse()

  expect(equals(1, 1)).toBeTrue()

  expect(equals(1, '1')).toBeFalse()

  expect(equals({}, {})).toBeTrue()

  expect(equals({
    a : 1,
    b : 2,
  },
  {
    b : 2,
    a : 1,
  })).toBeTrue()

  expect(equals({
    a : 1,
    b : 2,
  },
  {
    a : 1,
    b : 1,
  })).toBeFalse()

  expect(equals({
    a : 1,
    b : false,
  },
  {
    a : 1,
    b : 1,
  })).toBeFalse()

  expect(equals({
    a : 1,
    b : 2,
  },
  {
    b : 2,
    a : 1,
    c : 3,
  })).toBeFalse()

  expect(equals({
    x : {
      a : 1,
      b : 2,
    },
  },
  {
    x : {
      b : 2,
      a : 1,
      c : 3,
    },
  })).toBeFalse()

  expect(equals({
    a : 1,
    b : 2,
  },
  {
    b : 3,
    a : 1,
  })).toBeFalse()

  expect(equals({ a : { b : { c : 1 } } }, { a : { b : { c : 1 } } })).toBeTrue()

  expect(equals({ a : { b : { c : 1 } } }, { a : { b : { c : 2 } } })).toBeFalse()

  expect(equals({ a : {} }, { a : {} })).toBeTrue()

  expect(equals('', '')).toBeTrue()

  expect(equals('foo', 'foo')).toBeTrue()

  expect(equals('foo', 'bar')).toBeFalse()

  expect(equals(0, false)).toBeFalse()

  expect(equals(/\s/g, null)).toBeFalse()

  expect(equals(null, null)).toBeTrue()

  expect(equals(false)(null)).toBeFalse()
})

test('with custom functions', () => {
  function foo(){
    return 1
  }
  foo.prototype.toString = () => ''
  const result = equals(foo, foo)

  expect(result).toBeTrue()
})

test('with classes', () => {
  class Foo{}
  const foo = new Foo()
  const result = equals(foo, foo)

  expect(result).toBeTrue()
})

test('with negative zero', () => {
  expect(equals(-0, -0)).toBeTrue()
  expect(equals(-0, 0)).toBeFalse()
  expect(equals(0, 0)).toBeTrue()
  expect(equals(-0, 1)).toBeFalse()
})

---------------

evolve

evolve<T, U>(rules: ReadonlyArray<(x: T) => U>, list: readonly T[]): readonly U[]

It takes object or array of functions as set of rules. These rules are applied to the iterable input to produce the result.

💥 Error handling of this method differs between Ramda and Rambda. Ramda for some wrong inputs returns result and for other - it returns one of the inputs. Rambda simply throws when inputs are not correct. Full details for this mismatch are listed in source/_snapshots/evolve.spec.js.snap file.

const rules = {
  foo : add(1),
  bar : add(-1),
}
const input = {
  a   : 1,
  foo : 2,
  bar : 3,
}
const result = evolve(rules, input)
const expected = {
  a   : 1,
  foo : 3,
  bar : 2,
})
// => `result` is equal to `expected`

Try this R.evolve example in Rambda REPL

R.evolve source
import { _isArray } from './_internals/_isArray'
import { mapArray, mapObject } from './map'
import { type } from './type'

export function evolveArray(rules, list){
  return mapArray(
    (x, i) => {
      if (type(rules[ i ]) === 'Function'){
        return rules[ i ](x)
      }

      return x
    },
    list,
    true
  )
}

export function evolveObject(rules, iterable){
  return mapObject((x, prop) => {
    if (type(x) === 'Object'){
      const typeRule = type(rules[ prop ])
      if (typeRule === 'Function'){
        return rules[ prop ](x)
      }
      if (typeRule === 'Object'){
        return evolve(rules[ prop ], x)
      }

      return x
    }
    if (type(rules[ prop ]) === 'Function'){
      return rules[ prop ](x)
    }

    return x
  }, iterable)
}

export function evolve(rules, iterable){
  if (arguments.length === 1){
    return _iterable => evolve(rules, _iterable)
  }
  const rulesType = type(rules)
  const iterableType = type(iterable)

  if (iterableType !== rulesType){
    throw new Error('iterableType !== rulesType')
  }

  if (![ 'Object', 'Array' ].includes(rulesType)){
    throw new Error(`'iterable' and 'rules' are from wrong type ${ rulesType }`)
  }

  if (iterableType === 'Object'){
    return evolveObject(rules, iterable)
  }

  return evolveArray(rules, iterable)
}
Tests
import { evolve as evolveRamda } from 'ramda'

import { add } from '../rambda'
import { compareCombinations, compareToRamda } from './_internals/testUtils'
import { evolve } from './evolve'

test('happy', () => {
  const rules = {
    foo    : add(1),
    nested : { bar : x => Object.keys(x).length },
  }
  const input = {
    a      : 1,
    foo    : 2,
    nested : { bar : { z : 3 } },
  }
  const result = evolve(rules, input)
  expect(result).toEqual({
    a      : 1,
    foo    : 3,
    nested : { bar : 1 },
  })
})

test('nested rule is wrong', () => {
  const rules = {
    foo    : add(1),
    nested : { bar : 10 },
  }
  const input = {
    a      : 1,
    foo    : 2,
    nested : { bar : { z : 3 } },
  }
  const result = evolve(rules)(input)
  expect(result).toEqual({
    a      : 1,
    foo    : 3,
    nested : { bar : { z : 3 } },
  })
})

test('is recursive', () => {
  const rules = {
    nested : {
      second : add(-1),
      third  : add(1),
    },
  }
  const object = {
    first  : 1,
    nested : {
      second : 2,
      third  : 3,
    },
  }
  const expected = {
    first  : 1,
    nested : {
      second : 1,
      third  : 4,
    },
  }
  const result = evolve(rules, object)
  expect(result).toEqual(expected)
})

test('ignores primitive values', () => {
  const rules = {
    n : 2,
    m : 'foo',
  }
  const object = {
    n : 0,
    m : 1,
  }
  const expected = {
    n : 0,
    m : 1,
  }
  const result = evolve(rules, object)
  expect(result).toEqual(expected)
})

test('with array', () => {
  const rules = [ add(1), add(-1) ]
  const list = [ 100, 1400 ]
  const expected = [ 101, 1399 ]
  const result = evolve(rules, list)
  expect(result).toEqual(expected)
})

const rulesObject = { a : add(1) }
const rulesList = [ add(1) ]
const possibleIterables = [ null, undefined, '', 42, [], [ 1 ], { a : 1 } ]
const possibleRules = [ ...possibleIterables, rulesList, rulesObject ]

describe('brute force', () => {
  compareCombinations({
    firstInput : possibleRules,
    callback   : errorsCounters => {
      expect(errorsCounters).toMatchInlineSnapshot(`
        Object {
          "ERRORS_MESSAGE_MISMATCH": 0,
          "ERRORS_TYPE_MISMATCH": 4,
          "RESULTS_MISMATCH": 0,
          "SHOULD_NOT_THROW": 51,
          "SHOULD_THROW": 0,
        }
      `)
    },
    secondInput : possibleIterables,
    fn          : evolve,
    fnRamda     : evolveRamda,
  })
})

---------------

excludes

excludes(valueToFind: string, input: readonly string[] | string): boolean

Opposite of R.includes

R.equals is used to determine equality.

const result = [
  R.excludes('ar', 'foo'),
  R.excludes({a: 2}, [{a: 1}])
]
// => [true, true ]

Try this R.excludes example in Rambda REPL

R.excludes source
import { includes } from './includes'

export function excludes(valueToFind, input){
  if (arguments.length === 1) return _input => excludes(valueToFind, _input)

  return includes(valueToFind, input) === false
}
Tests
import { excludes } from './excludes'

test('excludes with string', () => {
  const str = 'more is less'

  expect(excludes('less')(str)).toBeFalse()
  expect(excludes('never', str)).toBeTrue()
})

test('excludes with array', () => {
  const arr = [ 1, 2, 3 ]

  expect(excludes(2)(arr)).toBeFalse()
  expect(excludes(4, arr)).toBeTrue()
})

---------------

F

F(): boolean
F() // => false

Try this R.F example in Rambda REPL

R.F source
export function F(){
  return false
}

---------------

filter

filter<T>(predicate: Predicate<T>): (input: readonly T[]) => readonly T[]

It filters list or object input using a predicate function.

const list = [3, 4, 3, 2]
const listPredicate = x => x > 2

const object = {abc: 'fo', xyz: 'bar', baz: 'foo'}
const objectPredicate = (x, prop) => x.length + prop.length > 5

const result = [
  R.filter(listPredicate, list),
  R.filter(objectPredicate, object)
]
// => [ [3, 4], { xyz: 'bar', baz: 'foo'} ]

Try this R.filter example in Rambda REPL

R.filter source
import { _isArray } from './_internals/_isArray'

export function filterObject(fn, obj){
  const willReturn = {}

  for (const prop in obj){
    if (fn(
      obj[ prop ], prop, obj
    )){
      willReturn[ prop ] = obj[ prop ]
    }
  }

  return willReturn
}

export function filterArray(
  predicate, list, indexed = false
){
  let index = 0
  const len = list.length
  const willReturn = []

  while (index < len){
    const predicateResult = indexed ?
      predicate(list[ index ], index) :
      predicate(list[ index ])
    if (predicateResult){
      willReturn.push(list[ index ])
    }

    index++
  }

  return willReturn
}

export function filter(predicate, iterable){
  if (arguments.length === 1){
    return _iterable => filter(predicate, _iterable)
  }
  if (!iterable) return []
  if (_isArray(iterable)) return filterArray(predicate, iterable)

  return filterObject(predicate, iterable)
}
Tests
import Ramda from 'ramda'

import { F } from './F'
import { filter } from './filter'
import { T } from './T'

const sampleObject = {
  a : 1,
  b : 2,
  c : 3,
  d : 4,
}

test('happy', () => {
  const isEven = n => n % 2 === 0

  expect(filter(isEven, [ 1, 2, 3, 4 ])).toEqual([ 2, 4 ])
  expect(filter(isEven, {
    a : 1,
    b : 2,
    d : 3,
  })).toEqual({ b : 2 })
})

test('bad inputs difference between Ramda and Rambda', () => {
  expect(filter(T)(undefined)).toEqual([])
  expect(filter(F, null)).toEqual([])
  expect(() => Ramda.filter(T, null)).toThrowWithMessage(TypeError,
    'Cannot read property \'filter\' of null')
  expect(() => Ramda.filter(T, undefined)).toThrowWithMessage(TypeError,
    'Cannot read property \'filter\' of undefined')
})

test('predicate when input is object', () => {
  const obj = {
    a : 1,
    b : 2,
  }
  const predicate = (
    val, prop, inputObject
  ) => {
    expect(inputObject).toEqual(obj)
    expect(typeof prop).toEqual('string')

    return val < 2
  }
  expect(filter(predicate, obj)).toEqual({ a : 1 })
})

test('with object', () => {
  const isEven = n => n % 2 === 0
  const result = filter(isEven, sampleObject)
  const expectedResult = {
    b : 2,
    d : 4,
  }

  expect(result).toEqual(expectedResult)
})

---------------

filterAsync

filterAsync<T>(fn: AsyncPredicate<T>, list: readonly T[]): Promise<readonly T[]>

Asynchronous version of R.filter

const predicate = async x => {
  await R.delay(100)
  return x % 2 === 1
}
const result = await R.filterAsync(predicate, [ 1, 2, 3 ])
// => [ 1, 3 ]

Try this R.filterAsync example in Rambda REPL

R.filterAsync source
import { _isArray } from './_internals/_isArray'
import { filter } from './filter'
import { mapAsync } from './mapAsync'

export function filterAsyncFn(predicate, listOrObject){
  return new Promise((resolve, reject) => {
    mapAsync(predicate, listOrObject)
      .then(predicateResult => {
        if (_isArray(predicateResult)){
          const filtered = listOrObject.filter((_, i) => predicateResult[ i ])

          return resolve(filtered)
        }
        const filtered = filter((_, prop) => predicateResult[ prop ],
          listOrObject)

        return resolve(filtered)
      })
      .catch(reject)
  })
}

export function filterAsync(predicate, listOrObject){
  if (arguments.length === 1){
    return async _listOrObject => filterAsyncFn(predicate, _listOrObject)
  }

  return new Promise((resolve, reject) => {
    filterAsyncFn(predicate, listOrObject).then(resolve)
      .catch(reject)
  })
}
Tests
import { delay } from './delay'
import { filterAsync } from './filterAsync'

test('happy', async () => {
  const predicate = async (x, i) => {
    expect(i).toBeNumber()
    await delay(100)

    return x % 2 === 1
  }
  const result = await filterAsync(predicate)([ 1, 2, 3 ])
  expect(result).toEqual([ 1, 3 ])
})

test('with object', async () => {
  const predicate = async (x, prop) => {
    expect(prop).toBeString()
    await delay(100)

    return x % 2 === 1
  }
  const result = await filterAsync(predicate, {
    a : 1,
    b : 2,
    c : 3,
    d : 4,
    e : 5,
  })

  expect(result).toEqual({
    a : 1,
    c : 3,
    e : 5,
  })
})

---------------

filterIndexed

filterIndexed<T>(predicate: IndexedPredicate<T>): (x: readonly T[]) => readonly T[]

Same as R.filter, but it passes index as second argument to the predicate, when looping over arrays.

R.filterIndexed source
import { _isArray } from './_internals/_isArray'
import { filterArray, filterObject } from './filter'

export function filterIndexed(predicate, iterable){
  if (arguments.length === 1)
    return _iterable => filterIndexed(predicate, _iterable)
  if (!iterable) return []
  if (_isArray(iterable)) return filterArray(
    predicate, iterable, true
  )

  return filterObject(predicate, iterable)
}
Tests
import { filter } from './filter'
import { filterIndexed } from './filterIndexed'

const iterator = (x, i) => {
  expect(x).toBeNumber()
  expect(i).toBeNumber()
}

test('happy', () => {
  const list = [ 1, 2, 3 ]
  filterIndexed(iterator, list)
  filterIndexed(iterator)(list)
})

test('with object', () => {
  const iterator = x => x + 1
  const obj = { a : 1 }
  expect(filterIndexed(iterator, obj)).toEqual(filter(iterator, obj))
})

test('with bad input', () => {
  expect(filterIndexed(iterator, undefined)).toEqual([])
})

---------------

find

find<T>(predicate: (x: T) => boolean, list: readonly T[]): T | undefined

It returns the first element of list that satisfy the predicate.

If there is no such element, it returns undefined.

const predicate = x => R.type(x.foo) === 'Number'
const list = [{foo: 'bar'}, {foo: 1}]

const result = R.find(predicate, list)
// => {foo: 1}

Try this R.find example in Rambda REPL

R.find source
export function find(predicate, list){
  if (arguments.length === 1) return _list => find(predicate, _list)

  let index = 0
  const len = list.length

  while (index < len){
    const x = list[ index ]
    if (predicate(x)){
      return x
    }

    index++
  }
}
Tests
import { find } from './find'
import { propEq } from './propEq'

const list = [ { a : 1 }, { a : 2 }, { a : 3 } ]

test('happy', () => {
  const fn = propEq('a', 2)
  expect(find(fn, list)).toEqual({ a : 2 })
})

test('with curry', () => {
  const fn = propEq('a', 4)
  expect(find(fn)(list)).toBeUndefined()
})

test('with empty list', () => {
  expect(find(() => true, [])).toBeUndefined()
})

---------------

findIndex

findIndex<T>(predicate: (x: T) => boolean, list: readonly T[]): number

It returns the index of the first element of list satisfying the predicate function.

If there is no such element, then -1 is returned.

const predicate = x => R.type(x.foo) === 'Number'
const list = [{foo: 'bar'}, {foo: 1}]

const result = R.findIndex(predicate, list)
// => 1

Try this R.findIndex example in Rambda REPL

R.findIndex source
export function findIndex(predicate, list){
  if (arguments.length === 1) return _list => findIndex(predicate, _list)

  const len = list.length
  let index = -1

  while (++index < len){
    if (predicate(list[ index ])){
      return index
    }
  }

  return -1
}
Tests
import { findIndex } from './findIndex'
import { propEq } from './propEq'

const list = [ { a : 1 }, { a : 2 }, { a : 3 } ]

test('happy', () => {
  expect(findIndex(propEq('a', 2), list)).toEqual(1)

  expect(findIndex(propEq('a', 1))(list)).toEqual(0)

  expect(findIndex(propEq('a', 4))(list)).toEqual(-1)
})

---------------

findLast

findLast<T>(fn: (x: T) => boolean, list: readonly T[]): T | undefined

It returns the last element of list satisfying the predicate function.

If there is no such element, then undefined is returned.

const predicate = x => R.type(x.foo) === 'Number'
const list = [{foo: 0}, {foo: 1}]

const result = R.findLast(predicate, list)
// => {foo: 1}

Try this R.findLast example in Rambda REPL

R.findLast source
export function findLast(predicate, list){
  if (arguments.length === 1) return _list => findLast(predicate, _list)

  let index = list.length

  while (--index >= 0){
    if (predicate(list[ index ])){
      return list[ index ]
    }
  }

  return undefined
}
Tests
import { findLast } from './findLast'

test('happy', () => {
  const result = findLast(x => x > 1,
    [ 1, 1, 1, 2, 3, 4, 1 ])
  expect(result).toEqual(4)

  expect(findLast(x => x === 0, [ 0, 1, 1, 2, 3, 4, 1 ])).toEqual(0)
})

test('with curry', () => {
  expect(findLast(x => x > 1)([ 1, 1, 1, 2, 3, 4, 1 ])).toEqual(4)
})

const obj1 = { x : 100 }
const obj2 = { x : 200 }
const a = [ 11, 10, 9, 'cow', obj1, 8, 7, 100, 200, 300, obj2, 4, 3, 2, 1, 0 ]
const even = function (x){
  return x % 2 === 0
}
const gt100 = function (x){
  return x > 100
}
const isStr = function (x){
  return typeof x === 'string'
}
const xGt100 = function (o){
  return o && o.x > 100
}

test('ramda 1', () => {
  expect(findLast(even, a)).toEqual(0)
  expect(findLast(gt100, a)).toEqual(300)
  expect(findLast(isStr, a)).toEqual('cow')
  expect(findLast(xGt100, a)).toEqual(obj2)
})

test('ramda 2', () => {
  expect(findLast(even, [ 'zing' ])).toEqual(undefined)
})

test('ramda 3', () => {
  expect(findLast(even, [ 2, 3, 5 ])).toEqual(2)
})

test('ramda 4', () => {
  expect(findLast(even, [])).toEqual(undefined)
})

---------------

findLastIndex

findLastIndex<T>(predicate: (x: T) => boolean, list: readonly T[]): number

It returns the index of the last element of list satisfying the predicate function.

If there is no such element, then -1 is returned.

const predicate = x => R.type(x.foo) === 'Number'
const list = [{foo: 0}, {foo: 1}]

const result = R.findLastIndex(predicate, list)
// => 1

Try this R.findLastIndex example in Rambda REPL

R.findLastIndex source
export function findLastIndex(fn, list){
  if (arguments.length === 1) return _list => findLastIndex(fn, _list)

  let index = list.length

  while (--index >= 0){
    if (fn(list[ index ])){
      return index
    }
  }

  return -1
}
Tests
import { findLastIndex } from './findLastIndex'

test('happy', () => {
  const result = findLastIndex(x => x > 1,
    [ 1, 1, 1, 2, 3, 4, 1 ])

  expect(result).toEqual(5)

  expect(findLastIndex(x => x === 0, [ 0, 1, 1, 2, 3, 4, 1 ])).toEqual(0)
})

test('with curry', () => {
  expect(findLastIndex(x => x > 1)([ 1, 1, 1, 2, 3, 4, 1 ])).toEqual(5)
})

const obj1 = { x : 100 }
const obj2 = { x : 200 }
const a = [ 11, 10, 9, 'cow', obj1, 8, 7, 100, 200, 300, obj2, 4, 3, 2, 1, 0 ]
const even = function (x){
  return x % 2 === 0
}
const gt100 = function (x){
  return x > 100
}
const isStr = function (x){
  return typeof x === 'string'
}
const xGt100 = function (o){
  return o && o.x > 100
}

test('ramda 1', () => {
  expect(findLastIndex(even, a)).toEqual(15)
  expect(findLastIndex(gt100, a)).toEqual(9)
  expect(findLastIndex(isStr, a)).toEqual(3)
  expect(findLastIndex(xGt100, a)).toEqual(10)
})

test('ramda 2', () => {
  expect(findLastIndex(even, [ 'zing' ])).toEqual(-1)
})

test('ramda 3', () => {
  expect(findLastIndex(even, [ 2, 3, 5 ])).toEqual(0)
})

test('ramda 4', () => {
  expect(findLastIndex(even, [])).toEqual(-1)
})

---------------

flatten

flatten<T>(list: readonly any[]): readonly T[]

It deeply flattens an array.

const result = R.flatten([
  1, 
  2, 
  [3, 30, [300]], 
  [4]
])
// => [ 1, 2, 3, 30, 300, 4 ]

Try this R.flatten example in Rambda REPL

R.flatten source
import { _isArray } from './_internals/_isArray'

export function flatten(list, input){
  const willReturn = input === undefined ? [] : input

  for (let i = 0; i < list.length; i++){
    if (_isArray(list[ i ])){
      flatten(list[ i ], willReturn)
    } else {
      willReturn.push(list[ i ])
    }
  }

  return willReturn
}
Tests
import { flatten } from './flatten'

test('happy', () => {
  expect(flatten([ 1, 2, 3, [ [ [ [ [ 4 ] ] ] ] ] ])).toEqual([ 1, 2, 3, 4 ])

  expect(flatten([ 1, [ 2, [ [ 3 ] ] ], [ 4 ] ])).toEqual([ 1, 2, 3, 4 ])

  expect(flatten([ 1, [ 2, [ [ [ 3 ] ] ] ], [ 4 ] ])).toEqual([ 1, 2, 3, 4 ])

  expect(flatten([ 1, 2, [ 3, 4 ], 5, [ 6, [ 7, 8, [ 9, [ 10, 11 ], 12 ] ] ] ])).toEqual([ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 ])
})

test('readme example', () => {
  const result = flatten([ 1, 2, [ 3, 30, [ 300 ] ], [ 4 ] ])
  expect(result).toEqual([ 1, 2, 3, 30, 300, 4 ])
})

---------------

flip

It returns function which calls fn with exchanged first and second argument.

💥 Rambda's flip will throw if the arity of the input function is greater or equal to 5.

const subtractFlip = R.flip(R.subtract)

const result = [
  subtractFlip(1,7),
  R.flip(1, 6)
]  
// => [6, -6]

Try this R.flip example in Rambda REPL

---------------

forEach

forEach<T>(fn: Iterator<T, void>, list: readonly T[]): readonly T[]

It applies iterable function over all members of list and returns list.

💥 It works with objects, unlike Ramda.

const sideEffect = {}
const result = R.forEach(
  x => sideEffect[`foo${x}`] = x
)([1, 2])

sideEffect // => {foo1: 1, foo2: 2}
result // => [1, 2]

Try this R.forEach example in Rambda REPL

R.forEach source
import { _isArray } from './_internals/_isArray'
import { _keys } from './_internals/_keys'

export function forEach(fn, list){
  if (arguments.length === 1) return _list => forEach(fn, _list)

  if (list === undefined){
    return
  }

  if (_isArray(list)){
    let index = 0
    const len = list.length

    while (index < len){
      fn(list[ index ])
      index++
    }
  } else {
    let index = 0
    const keys = _keys(list)
    const len = keys.length

    while (index < len){
      const key = keys[ index ]
      fn(
        list[ key ], key, list
      )
      index++
    }
  }

  return list
}
Tests
import { forEach } from './forEach'
import { type } from './type'

test('happy', () => {
  const sideEffect = {}
  forEach(x => sideEffect[ `foo${ x }` ] = x + 10)([ 1, 2 ])

  expect(sideEffect).toEqual({
    foo1 : 11,
    foo2 : 12,
  })
})

test('iterate over object', () => {
  const obj = {
    a : 1,
    b : [ 1, 2 ],
    c : { d : 7 },
    f : 'foo',
  }
  const result = {}
  const returned = forEach((
    val, prop, inputObj
  ) => {
    expect(type(inputObj)).toBe('Object')
    result[ prop ] = `${ prop }-${ type(val) }`
  })(obj)

  const expected = {
    a : 'a-Number',
    b : 'b-Array',
    c : 'c-Object',
    f : 'f-String',
  }

  expect(result).toEqual(expected)
  expect(returned).toEqual(obj)
})

test('with empty list', () => {
  const list = []
  const result = forEach(x => x * x)(list)

  expect(result).toEqual(list)
})

test('with wrong input', () => {
  const list = undefined
  const result = forEach(x => x * x)(list)

  expect(result).toBeUndefined()
})

test('returns the input', () => {
  const list = [ 1, 2, 3 ]
  const result = forEach(x => x * x)(list)

  expect(result).toEqual(list)
})

---------------

forEachIndexed

forEachIndexed<T>(fn: IndexedIterator<T, void>, list: readonly T[]): readonly T[]
R.forEachIndexed source
import { mapIndexed } from './mapIndexed'

export function forEachIndexed(fn, iterable){
  if (arguments.length === 1){
    return _iterable => forEachIndexed(fn, _iterable)
  }

  mapIndexed(fn, iterable)

  return iterable
}
Tests
import { forEachIndexed } from './forEachIndexed'

const list = [ 1, 2, 3 ]
const iterator = (x, i) => {
  expect(x).toBeNumber()
  expect(i).toBeNumber()
}

test('happy', () => {
  const result = forEachIndexed(iterator, list)

  expect(result).toEqual(list)
})

test('curried', () => {
  const result = forEachIndexed(iterator)(list)

  expect(result).toEqual(list)
})

---------------

fromPairs

It transforms a listOfPairs to an object.

const listOfPairs = [ [ 'a', 1 ], [ 'b', 2 ], [ 'c', [ 3, 4 ] ] ]
const expected = {
  a : 1,
  b : 2,
  c : [ 3, 4 ],
}

const result = R.fromPairs(listOfPairs)
// => `result` is equal to `expected`

Try this R.fromPairs example in Rambda REPL

---------------

getter

getter<T>(keyOrKeys: string | readonly string[] | undefined): T

The set of methods R.setter, R.getter and R.reset allow different parts of your logic to access comminicate indirectly via shared cache object.

Usually these methods show that you might need to refactor to classes. Still, they can be helpful meanwhile.

R.getter: It provides access to the cache object. If undefined is used as a key, this method will return the whole cache object. If string is passed, then it will return cache value for this key. If array of string is passed, then it assume that this is array of keys and it will return the corresponding cache values for these keys.

R.setter: It allows cache object's keys to be changed. You can either set individual key-value pairs with R.setter(key, value) or you pass directly object, which will be merged with the cache object.

R.reset: It resets the cache object.

R.setter('foo','bar')
R.setter('a', 1)
R.getter(['foo','a']) // => {foo: 'bar', a: 1}

R.setter('a', 2)
R.getter('a') // => 2
R.reset()
R.getter('a') // => undefined

Try this R.getter example in Rambda REPL

R.getter source
import { merge } from './merge'
import { pick } from './pick'
import { type } from './type'

let holder = {}

/**
 * Pass string to get value
 * Pass array to get object of values
 * Pass undefined to get all data
 */
export function getter(key){
  const typeKey = type(key)

  if (typeKey === 'String') return holder[ key ]

  if (typeKey === 'Array') return pick(key, holder)

  return holder
}

export function setter(maybeKey, maybeValue){
  const typeKey = type(maybeKey)
  const typeValue = type(maybeValue)

  if (typeKey === 'String'){
    if (typeValue === 'Function'){
      return holder[ maybeKey ] = maybeValue(holder[ maybeKey ])
    }

    return holder[ maybeKey ] = maybeValue
  }

  if (typeKey !== 'Object') return

  holder = merge(holder, maybeKey)
}

export function reset(){
  holder = {}
}
Tests
import { add } from './add'
import { getter, reset, setter } from './getter'

afterEach(() => {
  reset()
})

test('happy', () => {
  const key = 'foo'
  setter(key, 1)

  expect(getter(key)).toBe(1)
})

test('docs example', () => {
  setter('foo', 'bar')
  setter('a', 1)
  expect(getter([ 'foo', 'a' ])).toEqual({
    foo : 'bar',
    a   : 1,
  })

  setter('a', 2)
  expect(getter('a')).toBe(2)
  reset()
  expect(getter('a')).toBeUndefined()
})

test('when array is key in getter', () => {
  setter({
    a : 1,
    b : 2,
    c : 3,
  })

  expect(getter([ 'a', 'b' ])).toEqual({
    a : 1,
    b : 2,
  })
})

test('getter with undefined as key returns all', () => {
  const data = {
    a : 1,
    b : 2,
    c : 3,
  }

  setter(data)

  expect(getter()).toEqual(data)
})

test('function as setter value', () => {
  const data = {
    a : 1,
    b : 2,
    c : 3,
  }

  setter(data)
  setter('a', add(10))

  expect(getter()).toEqual({
    a : 11,
    b : 2,
    c : 3,
  })
})

test('setter fallbacks to undefined', () => {
  expect(setter()).toBeUndefined
})

---------------

glue

glue(input: string, glueString?: string): string

It transforms multiline string to single line by gluing together the separate lines with the glueString and removing the empty spaces. By default glueString is equal to single space, so if that is what you need, then you can just pass a single argument.

const result = R.glue(`
  foo
  bar
  baz
`)
// => 'foo bar baz'

Try this R.glue example in Rambda REPL

R.glue source
export function glue(input, glueChar){
  return input
    .split('\n')
    .filter(x => x.trim().length > 0)
    .map(x => x.trim())
    .join(glueChar === undefined ? ' ' : glueChar)
}
Tests
import { glue } from './glue'

test('empty string as a glue', () => {
  const result = glue(`
    foo
    bar
    baz
  `,
  '')

  const expectedResult = 'foobarbaz'

  expect(result).toBe(expectedResult)
})

test('case 0', () => {
  const zero = 'node node_modules/jest'
  const first = '--runInBand'
  const last = '-- src/a.spec.js'
  const flag = false
  const result = glue(`
    ${ zero }
    ${ first }
    ${ flag ? '--env=node' : '' }
    ${ last }
  `)

  const expectedResult = `${ zero } ${ first } ${ last }`

  expect(result).toBe(expectedResult)
})

test('case 1', () => {
  const zero = 'node node_modules/jest'
  const first = '--runInBand'
  const last = '-- src/a.spec.js'
  const flag = true
  const result = glue(`
    ${ zero }
    ${ first }
    ${ flag ? '--env=node' : '' }
    ${ last }
  `)

  const expectedResult = `${ zero } ${ first } --env=node ${ last }`

  expect(result).toBe(expectedResult)
})

test('case 2', () => {
  const first = '--runInBand'
  const result = glue(`
    zero
    ${ first }
    last
  `)
  const expectedResult = `zero ${ first } last`

  expect(result).toBe(expectedResult)
})

test('case 3', () => {
  const result = glue(`
    foo
    bar
    baz
  `)

  const expectedResult = 'foo bar baz'

  expect(result).toBe(expectedResult)
})

test('with glue', () => {
  const result = glue(`
    foo
    bar
    baz
  `,
  '==')

  const expectedResult = 'foo==bar==baz'

  expect(result).toBe(expectedResult)
})

---------------

groupBy

It splits list according to a provided groupFn function and returns an object.

const list = [ 'a', 'b', 'aa', 'bb' ]
const groupFn = x => x.length

const result = R.groupBy(groupFn, list)
// => { '1': ['a', 'b'], '2': ['aa', 'bb'] }

Try this R.groupBy example in Rambda REPL

---------------

groupWith

groupWith<T>(compareFn: (x: T, y: T) => boolean): (input: readonly T[]) => readonly (readonly T[])[]

It returns separated version of list or string input, where separation is done with equality compareFn function.

const compareFn = (x, y) => x === y
const list = [1, 2, 2, 1, 1, 2]

const result = R.groupWith(isConsecutive, list)
// => [[1], [2,2], [1,1], [2]]

Try this R.groupWith example in Rambda REPL

R.groupWith source
import { _isArray } from './_internals/_isArray'

export function groupWith(compareFn, list){
  if (!_isArray(list)) throw new TypeError('list.reduce is not a function')

  const clone = list.slice()

  if (list.length === 1) return [ clone ]

  const toReturn = []
  let holder = []

  clone.reduce((
    prev, current, i
  ) => {
    if (i === 0) return current

    const okCompare = compareFn(prev, current)
    const holderIsEmpty = holder.length === 0
    const lastCall = i === list.length - 1

    if (okCompare){
      if (holderIsEmpty) holder.push(prev)
      holder.push(current)
      if (lastCall) toReturn.push(holder)

      return current
    }

    if (holderIsEmpty){
      toReturn.push([ prev ])
      if (lastCall) toReturn.push([ current ])

      return current
    }

    toReturn.push(holder)
    if (lastCall) toReturn.push([ current ])
    holder = []

    return current
  }, undefined)

  return toReturn
}
Tests
import { equals } from './equals'
import { groupWith } from './groupWith'

test('issue is fixed', () => {
  const result = groupWith(equals, [ 1, 2, 2, 3 ])
  const expected = [ [ 1 ], [ 2, 2 ], [ 3 ] ]
  expect(result).toEqual(expected)
})

test('long list', () => {
  const result = groupWith(equals, [
    0,
    1,
    1,
    2,
    3,
    5,
    8,
    13,
    21,
    21,
    21,
    1,
    2,
  ])

  const expected = [
    [ 0 ],
    [ 1, 1 ],
    [ 2 ],
    [ 3 ],
    [ 5 ],
    [ 8 ],
    [ 13 ],
    [ 21, 21, 21 ],
    [ 1 ],
    [ 2 ],
  ]
  expect(result).toEqual(expected)
})

test('readme example', () => {
  const list = [ 4, 3, 6, 2, 2, 1 ]

  const result = groupWith((a, b) => a - b === 1, list)
  const expected = [ [ 4, 3 ], [ 6 ], [ 2 ], [ 2, 1 ] ]
  expect(result).toEqual(expected)
})

test('throw with string as input', () => {
  expect(() => groupWith(equals, 'Mississippi')).toThrowWithMessage(TypeError,
    'list.reduce is not a function')
})

const isConsecutive = function (a, b){
  return a + 1 === b
}

test('fix coverage', () => {
  expect(groupWith(isConsecutive, [ 1, 2, 3, 0 ])).toEqual([ [ 1, 2, 3 ], [ 0 ] ])
})

test('from ramda 0', () => {
  expect(groupWith(equals, [])).toEqual([])
  expect(groupWith(isConsecutive, [])).toEqual([])
})

test('from ramda 1', () => {
  expect(groupWith(isConsecutive, [ 4, 3, 2, 1 ])).toEqual([
    [ 4 ],
    [ 3 ],
    [ 2 ],
    [ 1 ],
  ])
})

test('from ramda 2', () => {
  expect(groupWith(isConsecutive, [ 1, 2, 3, 4 ])).toEqual([ [ 1, 2, 3, 4 ] ])
})

test('from ramda 3', () => {
  expect(groupWith(isConsecutive, [ 1, 2, 2, 3 ])).toEqual([
    [ 1, 2 ],
    [ 2, 3 ],
  ])
  expect(groupWith(isConsecutive, [ 1, 2, 9, 3, 4 ])).toEqual([
    [ 1, 2 ],
    [ 9 ],
    [ 3, 4 ],
  ])
})

test('list with single item', () => {
  const result = groupWith(equals, [ 0 ])

  const expected = [ [ 0 ] ]
  expect(result).toEqual(expected)
})

---------------

has

has<T>(prop: string, obj: T): boolean

It returns true if obj has property prop.

const obj = {a: 1}

const result = [
  R.has('a', obj),
  R.has('b', obj)
]
// => [true, false]

Try this R.has example in Rambda REPL

R.has source
export function has(prop, obj){
  if (arguments.length === 1) return _obj => has(prop, _obj)

  if (!obj) return false

  return obj[ prop ] !== undefined
}
Tests
import { has } from './has'

test('happy', () => {
  expect(has('a')({ a : 1 })).toBeTrue()
  expect(has('b', { a : 1 })).toBeFalse()
})

test('with non-object', () => {
  expect(has('a', undefined)).toEqual(false)
  expect(has('a', null)).toEqual(false)
  expect(has('a', true)).toEqual(false)
  expect(has('a', '')).toEqual(false)
  expect(has('a', /a/)).toEqual(false)
})

---------------

hasPath

hasPath<T>(
  path: string | readonly string[],
  input: object
): boolean

It will return true, if input object has truthy path(calculated with R.path).

const path = 'a.b'
const pathAsArray = ['a', 'b']
const obj = {a: {b: []}}

const result = [
  R.hasPath(path, obj),
  R.hasPath(pathAsArray, obj),
  R.hasPath('a.c', obj),
]
// => [true, true, false]

Try this R.hasPath example in Rambda REPL

R.hasPath source
import { path } from './path'

export function hasPath(maybePath, obj){
  if (arguments.length === 1){
    return objHolder => hasPath(maybePath, objHolder)
  }

  return path(maybePath, obj) !== undefined
}
Tests
import { hasPath } from './hasPath'

test('when true', () => {
  const path = 'a.b'
  const obj = { a : { b : [] } }

  const result = hasPath(path)(obj)
  const expectedResult = true

  expect(result).toEqual(expectedResult)
})

test('when false', () => {
  const path = 'a.b'
  const obj = {}

  const result = hasPath(path, obj)
  const expectedResult = false

  expect(result).toEqual(expectedResult)
})

---------------

head

head<T>(input: readonly T[]): T | undefined

It returns the first element of list or string input.

const result = [
  R.head([1, 2, 3]),
  R.head('foo') 
]
// => [1, 'f']

Try this R.head example in Rambda REPL

R.head source
export function head(listOrString){
  if (typeof listOrString === 'string') return listOrString[ 0 ] || ''

  return listOrString[ 0 ]
}
Tests
import { head } from './head'

test('head', () => {
  expect(head([ 'fi', 'fo', 'fum' ])).toEqual('fi')
  expect(head([])).toEqual(undefined)
  expect(head('foo')).toEqual('f')
  expect(head('')).toEqual('')
})

---------------

identical

It returns true if its arguments a and b are identical.

Otherwise, it returns false.

💥 Values are identical if they reference the same memory. NaN is identical to NaN; 0 and -0 are not identical.

const obj = {a: 1};
R.identical(obj, obj); // => true
R.identical(1, 1); // => true
R.identical(1, '1'); // => false
R.identical([], []); // => false
R.identical(0, -0); // => false
R.identical(NaN, NaN); // => true

Try this R.identical example in Rambda REPL

---------------

identity

identity<T>(input: T): T

It just passes back the supplied input argument.

💥 Logic

R.identity(7) // => 7

Try this R.identity example in Rambda REPL

R.identity source
export function identity(input){
  return input
}
Tests
import { identity } from './identity'

test('happy', () => {
  expect(identity(7)).toEqual(7)
  expect(identity(true)).toEqual(true)
  expect(identity({ a : 1 })).toEqual({ a : 1 })
})

---------------

ifElse

ifElse<T, U>(
  condition: (x: T) => boolean, 
  onTrue: (x: T) => U, 
  onFalse: (x: T) => U, 
): (x: T) => U

It expects condition, onTrue and onFalse functions as inputs and it returns a new function with example name of fn.

When fn`` is called with inputargument, it will return eitheronTrue(input)oronFalse(input)depending oncondition(input)` evaluation.

const fn = R.ifElse(
 x => x>10,
 x => x*2,
 x => x*10
)

const result = [ fn(8), fn(18) ]
// => [80, 36]

Try this R.ifElse example in Rambda REPL

R.ifElse source
import { curry } from './curry'

function ifElseFn(
  condition, onTrue, onFalse
){
  return (...input) => {
    const conditionResult =
      typeof condition === 'boolean' ? condition : condition(...input)

    if (conditionResult === true){
      return onTrue(...input)
    }

    return onFalse(...input)
  }
}

export const ifElse = curry(ifElseFn)
Tests
import { always } from './always'
import { has } from './has'
import { identity } from './identity'
import { ifElse } from './ifElse'
import { prop } from './prop'

const condition = has('foo')
const v = function (a){
  return typeof a === 'number'
}
const t = function (a){
  return a + 1
}
const ifFn = x => prop('foo', x).length
const elseFn = () => false

test('happy', () => {
  const fn = ifElse(condition, ifFn)(elseFn)

  expect(fn({ foo : 'bar' })).toEqual(3)
  expect(fn({ fo : 'bar' })).toEqual(false)
})

test('ramda spec', () => {
  const ifIsNumber = ifElse(v)
  expect(ifIsNumber(t, identity)(15)).toEqual(16)
  expect(ifIsNumber(t, identity)('hello')).toEqual('hello')
})

test('pass all arguments', () => {
  const identity = function (a){
    return a
  }
  const v = function (){
    return true
  }
  const onTrue = function (a, b){
    expect(a).toEqual(123)
    expect(b).toEqual('abc')
  }
  ifElse(
    v, onTrue, identity
  )(123, 'abc')
})

test('accept constant as condition', () => {
  const fn = ifElse(true)(always(true))(always(false))

  expect(fn()).toEqual(true)
})

test('accept constant as condition - case 2', () => {
  const fn = ifElse(
    false, always(true), always(false)
  )

  expect(fn()).toEqual(false)
})

test('curry 1', () => {
  const fn = ifElse(condition, ifFn)(elseFn)

  expect(fn({ foo : 'bar' })).toEqual(3)
  expect(fn({ fo : 'bar' })).toEqual(false)
})

test('curry 2', () => {
  const fn = ifElse(condition)(ifFn)(elseFn)

  expect(fn({ foo : 'bar' })).toEqual(3)
  expect(fn({ fo : 'bar' })).toEqual(false)
})

test('simple arity of 1', () => {
  const condition = x => x > 5
  const onTrue = x => x + 1
  const onFalse = x => x + 10
  const result = ifElse(
    condition, onTrue, onFalse
  )(1)
  expect(result).toBe(11)
})

test('simple arity of 2', () => {
  const condition = (x, y) => x + y > 5
  const onTrue = (x, y) => x + y + 1
  const onFalse = (x, y) => x + y + 10
  const result = ifElse(
    condition, onTrue, onFalse
  )(1, 10)
  expect(result).toBe(12)
})

---------------

ifElseAsync

ifElseAsync<T, U>(
  condition: (x: T) => Promise<boolean>, 
  onTrue: (x: T) => U, 
  onFalse: (x: T) => U, 
  ): (x: T) => Promise<U>

Asynchronous version of R.ifElse. Any of condition, ifFn and elseFn can be either asynchronous or synchronous function.

const condition = async x => {
  await R.delay(100)
  return x > 1
}
const ifFn = async x => {
  await R.delay(100)
  return x + 1
}
const elseFn = async x => {
  await R.delay(100)
  return x - 1
}

const result = await R.ifElseAsync(
  condition,
  ifFn,
  elseFn  
)(1)
// => 0

Try this R.ifElseAsync example in Rambda REPL

R.ifElseAsync source
function createThenable(fn){
  return async function (...input){
    return fn(...input)
  }
}

export function ifElseAsync(
  condition, ifFn, elseFn
){
  return (...inputs) =>
    new Promise((resolve, reject) => {
      const conditionPromise = createThenable(condition)
      const ifFnPromise = createThenable(ifFn)
      const elseFnPromise = createThenable(elseFn)

      conditionPromise(...inputs)
        .then(conditionResult => {
          const promised =
            conditionResult === true ? ifFnPromise : elseFnPromise

          promised(...inputs)
            .then(resolve)
            .catch(reject)
        })
        .catch(reject)
    })
}
Tests
import { delay } from './delay'
import { ifElseAsync } from './ifElseAsync'

test('arity of 1 - condition is async', async () => {
  const condition = async x => {
    await delay(100)

    return x > 4
  }
  const whenTrue = x => x + 1
  const whenFalse = x => x + 10
  const fn = ifElseAsync(
    condition, whenTrue, whenFalse
  )
  const result = await Promise.all([ fn(5), fn(1) ])
  expect(result).toEqual([ 6, 11 ])
})

test('arity of 1 - condition is sync', async () => {
  const condition = x => x > 4
  const whenTrue = async x => {
    await delay(100)

    return x + 1
  }
  const whenFalse = async x => {
    await delay(100)

    return x + 10
  }
  const fn = ifElseAsync(
    condition, whenTrue, whenFalse
  )
  const result = await Promise.all([ fn(5), fn(1) ])
  expect(result).toEqual([ 6, 11 ])
})

test('arity of 1 - all inputs are async', async () => {
  const condition = async x => {
    await delay(100)

    return x > 4
  }
  const whenTrue = async x => {
    await delay(100)

    return x + 1
  }
  const whenFalse = async x => {
    await delay(100)

    return x + 10
  }
  const fn = ifElseAsync(
    condition, whenTrue, whenFalse
  )
  const result = await Promise.all([ fn(5), fn(1) ])
  expect(result).toEqual([ 6, 11 ])
})

test('arity of 2 - condition is async', async () => {
  const condition = async (x, y) => {
    await delay(100)

    return x + y > 4
  }
  const whenTrue = (x, y) => x + y + 1
  const whenFalse = (x, y) => x + y + 10
  const fn = ifElseAsync(
    condition, whenTrue, whenFalse
  )
  const result = await Promise.all([ fn(14, 20), fn(1, 3) ])
  expect(result).toEqual([ 35, 14 ])
})

test('arity of 2 - condition is sync', async () => {
  const condition = (x, y) => x + y > 4
  const whenTrue = async (x, y) => {
    await delay(100)

    return x + y + 1
  }
  const whenFalse = async (x, y) => {
    await delay(100)

    return x + y + 10
  }
  const fn = ifElseAsync(
    condition, whenTrue, whenFalse
  )
  const result = await Promise.all([ fn(14, 20), fn(1, 3) ])
  expect(result).toEqual([ 35, 14 ])
})

test('arity of 2 - all inputs are async', async () => {
  const condition = async (x, y) => {
    await delay(100)

    return x + y > 4
  }
  const whenTrue = async (x, y) => {
    await delay(100)

    return x + y + 1
  }
  const whenFalse = async (x, y) => {
    await delay(100)

    return x + y + 10
  }
  const fn = ifElseAsync(
    condition, whenTrue, whenFalse
  )
  const result = await Promise.all([ fn(14, 20), fn(1, 3) ])
  expect(result).toEqual([ 35, 14 ])
})

---------------

inc

It increments a number.

R.inc(1) // => 2

Try this R.inc example in Rambda REPL

---------------

includes

includes(valueToFind: string, input: readonly string[] | string): boolean

If input is string, then this method work as native String.includes.

If input is array, then R.equals is used to define if valueToFind belongs to the list.

const result = [
  R.includes('oo', 'foo'),
  R.includes({a: 1}, [{a: 1}])
]
// => [true, true ]

Try this R.includes example in Rambda REPL

R.includes source
import { _isArray } from './_internals/_isArray'
import { equals } from './equals'

export function includesArray(valueToFind, input){
  let index = -1

  while (++index < input.length){
    if (equals(input[ index ], valueToFind)){
      return true
    }
  }

  return false
}

export function includes(valueToFind, input){
  if (arguments.length === 1) return _input => includes(valueToFind, _input)
  if (typeof input === 'string'){
    return input.includes(valueToFind)
  }
  if (!input){
    throw new TypeError(`Cannot read property \'indexOf\' of ${ input }`)
  }
  if (!_isArray(input)) return false

  return includesArray(valueToFind, input)
}
Tests
import R from 'ramda'

import { includes } from './includes'

test('includes with string', () => {
  const str = 'foo bar'

  expect(includes('bar')(str)).toBeTrue()
  expect(R.includes('bar')(str)).toBeTrue()
  expect(includes('never', str)).toBeFalse()
  expect(R.includes('never', str)).toBeFalse()
})

test('includes with array', () => {
  const arr = [ 1, 2, 3 ]

  expect(includes(2)(arr)).toBeTrue()
  expect(R.includes(2)(arr)).toBeTrue()

  expect(includes(4, arr)).toBeFalse()
  expect(R.includes(4, arr)).toBeFalse()
})

test('with wrong input that does not throw', () => {
  const result = includes(1, /foo/g)
  const ramdaResult = R.includes(1, /foo/g)
  expect(result).toBeFalse()
  expect(ramdaResult).toBeFalse()
})

test('throws on wrong input - match ramda behaviour', () => {
  expect(() => includes(2, null)).toThrowWithMessage(TypeError,
    'Cannot read property \'indexOf\' of null')
  expect(() => R.includes(2, null)).toThrowWithMessage(TypeError,
    'Cannot read property \'indexOf\' of null')
  expect(() => includes(2, undefined)).toThrowWithMessage(TypeError,
    'Cannot read property \'indexOf\' of undefined')
  expect(() => R.includes(2, undefined)).toThrowWithMessage(TypeError,
    'Cannot read property \'indexOf\' of undefined')
})

---------------

indexBy

It generates object with properties provided by condition and values provided by list array.

If condition is a function, then all list members are passed through it.

If condition is a string, then all list members are passed through R.path(condition).

const list = [ {id: 10}, {id: 20} ]

const withFunction = R.indexBy(
  x => x.id,
  list
)
const withString = R.indexBy(
  'id',
  list
)
const result = [
  withFunction, 
  R.equals(withFunction, withString)
]
// => [ { 10: {id: 10}, 20: {id: 20} }, true ]

Try this R.indexBy example in Rambda REPL

---------------

indexOf

It returns the index of the first element of list equals to valueToFind.

If there is no such element, it returns -1.

const list = [0, 1, 2, 3]

const result = [
  R.indexOf(2, list),
  R.indexOf(0, list)
]
// => [2, -1]

Try this R.indexOf example in Rambda REPL

---------------

init

init<T>(input: readonly T[]): readonly T[]

It returns all but the last element of list or string input.

const result = [
  R.init([1, 2, 3]) , 
  R.init('foo')  // => 'fo'
]
// => [[1, 2], 'fo']

Try this R.init example in Rambda REPL

R.init source
import baseSlice from './_internals/baseSlice'

export function init(listOrString){
  if (typeof listOrString === 'string') return listOrString.slice(0, -1)

  return listOrString.length ? baseSlice(
    listOrString, 0, -1
  ) : []
}
Tests
import { init } from './init'

test('with array', () => {
  expect(init([ 1, 2, 3 ])).toEqual([ 1, 2 ])
  expect(init([ 1, 2 ])).toEqual([ 1 ])
  expect(init([ 1 ])).toEqual([])
  expect(init([])).toEqual([])
  expect(init([])).toEqual([])
  expect(init([ 1 ])).toEqual([])
})

test('with string', () => {
  expect(init('foo')).toEqual('fo')
  expect(init('f')).toEqual('')
  expect(init('')).toEqual('')
})

---------------

interpolate

interpolate(inputWithTags: string, templateArguments: object): string

It generages a new string from inputWithTags by replacing all {{x}} occurances with values provided by templateArguments.

const inputWithTags = 'foo is {{bar}} even {{a}} more'
const templateArguments = {"bar":"BAR", a: 1}

const result = R.interpolate(inputWithTags, templateArguments)
const expected = 'foo is BAR even 1 more'
// => `result` is equal to `expected`

Try this R.interpolate example in Rambda REPL

R.interpolate source
const getOccurances = input => input.match(/{{\s*.+?\s*}}/g)

const getOccuranceProp = occurance => occurance.replace(/{{\s*|\s*}}/g, '')

const replace = ({ inputHolder, prop, replacer }) => {
  const regexBase = `{{${ prop }}}`
  const regex = new RegExp(regexBase, 'g')

  return inputHolder.replace(regex, replacer)
}

export function interpolate(input, templateInput){
  if (arguments.length === 1){
    return _templateInput => interpolate(input, _templateInput)
  }

  const occurances = getOccurances(input)
  if (occurances === null) return input
  let inputHolder = input

  for (const occurance of occurances){
    const prop = getOccuranceProp(occurance)

    inputHolder = replace({
      inputHolder,
      prop,
      replacer : templateInput[ prop ],
    })
  }

  return inputHolder
}
Tests
import { interpolate } from './interpolate'

test('within bracets', () => {
  const input = 'foo is { {{bar}} } even {{a}} more'
  const templateInput = {
    bar : 'BAR',
    a   : 1,
  }

  const result = interpolate(input, templateInput)
  const expectedResult = 'foo is { BAR } even 1 more'

  expect(result).toEqual(expectedResult)
})

test('happy', () => {
  const input = 'foo is {{bar}} even {{a}} more'
  const templateInput = {
    bar : 'BAR',
    a   : 1,
  }

  const result = interpolate(input, templateInput)
  const expectedResult = 'foo is BAR even 1 more'

  expect(result).toEqual(expectedResult)
})

test('no interpolation + curry', () => {
  const input = 'foo is bar even more'
  const templateInput = { bar : 'BAR' }

  const result = interpolate(input)(templateInput)
  const expectedResult = 'foo is bar even more'

  expect(result).toEqual(expectedResult)
})

test('with missing template input', () => {
  const input = 'foo is {{bar}} even {{a}} more'
  const templateInput = {
    baz : 'BAR',
    a   : 1,
  }

  const result = interpolate(input, templateInput)
  const expectedResult = 'foo is undefined even 1 more'

  expect(result).toEqual(expectedResult)
})

test('with arbitrary expression', () => {
  const input = '1 + 2 = {{ 1 + 2 }}'
  const templateInput = {}

  const result = interpolate(input, templateInput)

  expect(result).toEqual(input)
})

---------------

intersection

It loops throw listA and listB and returns the intersection of the two according to R.equals.

const listA = [ { id : 1 }, { id : 2 }, { id : 3 }, { id : 4 } ]
const listB = [ { id : 3 }, { id : 4 }, { id : 5 }, { id : 6 } ]

const result = intersection(listA, listB)
// => [{ id : 3 }, { id : 4 }]

Try this R.intersection example in Rambda REPL

---------------

intersperse

It adds a separator between members of list.

const list = [ 0, 1, 2, 3 ]
const separator = '|'
const result = intersperse(separator, list)
// => [0, '|', 1, '|', 2, '|', 3]

Try this R.intersperse example in Rambda REPL

---------------

is

It returns true if x is instance of targetPrototype.

const result = [
  R.is(String, 'foo'),  
  R.is(Array, 1)
]
// => [true, false]

Try this R.is example in Rambda REPL

---------------

isEmpty

isEmpty<T>(x: T): boolean

It returns true if x is empty.

const result = [
  R.isEmpty(''),
  R.isEmpty({ x : 0 })
]
// => [true, false]

Try this R.isEmpty example in Rambda REPL

R.isEmpty source
import { type } from './type'

export function isEmpty(input){
  const inputType = type(input)
  if ([ 'Undefined', 'NaN', 'Number', 'Null' ].includes(inputType))
    return false
  if (!input) return true

  if (inputType === 'Object'){
    return Object.keys(input).length === 0
  }

  if (inputType === 'Array'){
    return input.length === 0
  }

  return false
}
Tests
import { isEmpty } from './isEmpty'

test('happy', () => {
  expect(isEmpty(undefined)).toEqual(false)
  expect(isEmpty('')).toEqual(true)
  expect(isEmpty(null)).toEqual(false)
  expect(isEmpty(' ')).toEqual(false)
  expect(isEmpty(new RegExp(''))).toEqual(false)
  expect(isEmpty([])).toEqual(true)
  expect(isEmpty([ [] ])).toEqual(false)
  expect(isEmpty({})).toEqual(true)
  expect(isEmpty({ x : 0 })).toEqual(false)
  expect(isEmpty(0)).toEqual(false)
  expect(isEmpty(NaN)).toEqual(false)
  expect(isEmpty([ '' ])).toEqual(false)
})

---------------

isFunction

isFunction(input: any): boolean

It returns true if R.type of input is Async or Function.

const result = [
  R.isFunction(R.mapAsync),
  R.isFunction(R.add),
]
// => [true, true]

Try this R.isFunction example in Rambda REPL

R.isFunction source
import { type } from './type'

export function isFunction(fn){
  return [ 'Async', 'Function' ].includes(type(fn))
}
Tests
import { isFunction } from './isFunction'

test('when function', () => {
  const fn = () => {}

  expect(isFunction(fn)).toEqual(true)
})

test('when promise', () => {
  const fn = Promise.resolve(2)

  expect(isFunction(fn)).toBeFalse()
})

test('when async', () => {
  const fn = async x => x

  expect(isFunction(fn)).toEqual(true)
})

test('when false', () => {
  expect(isFunction(null)).toEqual(false)
})

---------------

isNil

isNil(x: any): x is null | undefined

It returns true if x is either null or undefined.

const result = [
  R.isNil(null),
  R.isNil(1),
]
// => [true, false]

Try this R.isNil example in Rambda REPL

R.isNil source
export function isNil(x){
  return x === undefined || x === null
}
Tests
import { isNil } from './isNil'

test('happy', () => {
  expect(isNil(null)).toBeTrue()

  expect(isNil(undefined)).toBeTrue()

  expect(isNil([])).toBeFalse()
})

---------------

isPromise

isPromise(input: any): boolean

---------------

isType

isType(targetType: RambdaTypes, input: any): boolean

It returns true if targetType is equal to type of input according to R.type.

R.isType('Async',R.delay(1000))
// => true

Try this R.isType example in Rambda REPL

R.isType source
import { type } from './type'

export function isType(xType, x){
  if (arguments.length === 1){
    return xHolder => isType(xType, xHolder)
  }

  return type(x) === xType
}
Tests
import { isType } from './isType'

const list = [ 1, 2, 3 ]

test('array', () => {
  expect(isType('Array', list)).toBeTruthy()
  expect(isType('Array')([])).toBeTruthy()
})

test('promise', () => {
  expect(isType('Promise', Promise.resolve(1))).toBeTruthy()
})

---------------

isValid

isValid({input: object, schema: Schema}: IsValid): boolean

It checks if input is following schema specifications.

If validation fails, it returns false.

Please check the detailed explanation as it is hard to write a short description for this method.

💥 Independently, somebody else came with very similar idea called superstruct

const input = {a: ['foo', 'bar']}
const invalidInput = {a: ['foo', 'bar', 1]}
const schema = {a: [String]}
const result = [
  R.isValid({schema, input}),
  R.isValid({schema, input: invalidInput})
]
// => [true, false]

Try this R.isValid example in Rambda REPL

R.isValid source
import { _isArray } from './_internals/_isArray'
import { all } from './all'
import { any } from './any'
import { includes } from './includes'
import { init } from './init'
import { test } from './test'
import { toLower } from './toLower'
import { type } from './type'

export function isPrototype(input){
  const currentPrototype = input.prototype
  const list = [ Number, String, Boolean, Promise ]
  let toReturn = false
  let counter = -1
  while (++counter < list.length && !toReturn){
    if (currentPrototype === list[ counter ].prototype) toReturn = true
  }

  return toReturn
}

export function prototypeToString(input){
  const currentPrototype = input.prototype
  const list = [ Number, String, Boolean, Promise ]
  const translatedList = [ 'Number', 'String', 'Boolean', 'Promise' ]
  let found
  let counter = -1

  while (++counter < list.length){
    if (currentPrototype === list[ counter ].prototype) found = counter
  }

  return translatedList[ found ]
}

const typesWithoutPrototype = [ 'any', 'promise', 'async', 'function' ]

export function fromPrototypeToString(rule){
  if (
    _isArray(rule) ||
    rule === undefined ||
    rule === null ||
    rule.prototype === undefined ||
    typesWithoutPrototype.includes(rule)
  ){
    return {
      rule,
      parsed : false,
    }
  }
  if (String.prototype === rule.prototype){
    return {
      rule   : 'string',
      parsed : true,
    }
  }
  if (Boolean.prototype === rule.prototype){
    return {
      rule   : 'boolean',
      parsed : true,
    }
  }
  if (Number.prototype === rule.prototype){
    return {
      rule   : 'number',
      parsed : true,
    }
  }

  return {
    rule   : type(rule.prototype).toLowerCase(),
    parsed : true,
  }
}

function getRuleAndType(schema, requirementRaw){
  const ruleRaw = schema[ requirementRaw ]
  const typeIs = type(ruleRaw)
  const { rule, parsed } = fromPrototypeToString(ruleRaw)

  return {
    rule     : rule,
    ruleType : parsed ? 'String' : typeIs,
  }
}

export function isValid({ input, schema }){
  if (input === undefined || schema === undefined) return false

  let flag = true
  const boom = boomFlag => {
    if (!boomFlag){
      flag = false
    }
  }

  for (const requirementRaw in schema){
    if (flag){
      const isOptional = requirementRaw.endsWith('?')
      const requirement = isOptional ? init(requirementRaw) : requirementRaw

      const { rule, ruleType } = getRuleAndType(schema, requirementRaw)
      const inputProp = input[ requirement ]
      const inputPropType = type(input[ requirement ])

      const ok = isOptional && inputProp !== undefined || !isOptional

      if (!ok || rule === 'any' && inputProp != null || rule === inputProp)
        continue

      if (ruleType === 'Object'){
        /**
         * This rule is standalone schema, so we recursevly call `isValid`
         */
        const isValidResult = isValid({
          input  : inputProp,
          schema : rule,
        })
        boom(isValidResult)
      } else if (ruleType === 'String'){
        /**
         * Rule is actual rule such as 'number', so the two types are compared
         */
        boom(toLower(inputPropType) === rule)
      } else if (typeof rule === 'function'){
        /**
         * Rule is function so we pass to it the input
         */
        boom(rule(inputProp))
      } else if (ruleType === 'Array' && inputPropType === 'String'){
        /**
         * Enum case | rule is like a: ['foo', 'bar']
         */
        boom(includes(inputProp, rule))
      } else if (
        ruleType === 'Array' &&
        rule.length === 1 &&
        inputPropType === 'Array'
      ){
        /**
         * 1. array of type | rule is like a: ['number']
         * 2. rule is like a: [{foo: 'string', bar: 'number'}]
         */
        const [ currentRule ] = rule
        const currentRuleType = type(currentRule)

        //Check if rule is invalid
        boom(currentRuleType === 'String' ||
            currentRuleType === 'Object' ||
            isPrototype(currentRule))

        if (currentRuleType === 'Object' && flag){
          /**
           * 2. rule is like a: [{from: 'string'}]
           */
          const isValidResult = all(inputPropInstance =>
            isValid({
              input  : inputPropInstance,
              schema : currentRule,
            }),
          inputProp)
          boom(isValidResult)
        } else if (flag){
          /**
           * 1. array of type
           */

          const actualRule =
            currentRuleType === 'String' ?
              currentRule :
              prototypeToString(currentRule)
          const isInvalidResult = any(inputPropInstance =>
            type(inputPropInstance).toLowerCase() !==
              actualRule.toLowerCase(),
          inputProp)
          boom(!isInvalidResult)
        }
      } else if (ruleType === 'RegExp' && inputPropType === 'String'){
        boom(test(rule, inputProp))
      } else {
        boom(false)
      }
    }
  }

  return flag
}
Tests
import { delay } from './delay'
import { isPrototype, isValid } from './isValid'

test('is prototype', () => {
  expect(isPrototype(Promise)).toBeTrue()
  expect(isPrototype(Number)).toBeTrue()
  expect(isPrototype(Boolean)).toBeTrue()
  expect(isPrototype(String)).toBeTrue()
  expect(isPrototype(0)).toBeFalse()
})

test('prototype inside array', () => {
  const input = { a : [ 1, 2, 3, 4 ] }
  const schema = { a : [ Number ] }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('with Promise prototype', () => {
  const input = { a : [ delay(1), delay(2) ] }
  const schema = { a : [ Promise ] }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('object prototype as rule - true', () => {
  const input = { a : {} }
  const schema = { a : Object }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('list of functions', () => {
  const input = { a : [ () => {}, delay ] }
  const schema = { a : [ 'function' ] }

  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('function schema type can be only string', () => {
  const input = { a : [ () => {}, delay ] }
  const schema = { a : [ Function ] }

  expect(isValid({
    input,
    schema,
  })).toBeFalsy()
})

test('object prototype as rule - false', () => {
  const input = { a : null }
  const schema = { a : Object }
  expect(isValid({
    input,
    schema,
  })).toBeFalsy()
})

test('number prototype as rule - true', () => {
  const input = { a : 1 }
  const schema = { a : Number }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('array prototype as rule - true', () => {
  const input = { a : [ 1, 2, 3 ] }
  const schema = { a : Array }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('array prototype as rule - false', () => {
  const input = { a : null }
  const schema = { a : Array }
  expect(isValid({
    input,
    schema,
  })).toBeFalsy()
})

test('string prototype as rule - true', () => {
  const input = { a : 'foo' }
  const schema = { a : String }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('string prototype as rule - false', () => {
  const input = { a : null }
  const schema = { a : String }
  expect(isValid({
    input,
    schema,
  })).toBeFalsy()
})

test('boolean prototype as rule - true', () => {
  const input = { a : true }
  const schema = { a : Boolean }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('boolean prototype as rule - false', () => {
  const input = { a : null }
  const schema = { a : Boolean }
  expect(isValid({
    input,
    schema,
  })).toBeFalsy()
})

test('regex prototype cannot be rule - true', () => {
  const input = { a : /foo/g }
  const schema = { a : new RegExp('foo') }
  expect(isValid({
    input,
    schema,
  })).toBeFalsy()
})

test('undefined as a rule - true', () => {
  const input = { a : undefined }
  const schema = { a : undefined }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('undefined as a rule - false', () => {
  const input = { a : null }
  const schema = { a : undefined }
  expect(isValid({
    input,
    schema,
  })).toBeFalsy()
})

test('null as a rule - true', () => {
  const input = { a : null }
  const schema = { a : null }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('null as a rule - false', () => {
  const input = { a : undefined }
  const schema = { a : null }
  expect(isValid({
    input,
    schema,
  })).toBeFalsy()
})

test('`any` safeguard against `null`', () => {
  const input = { a : null }
  const schema = { a : 'any' }
  expect(isValid({
    input,
    schema,
  })).toBeFalsy()
})

test('`any` safeguard against `undefined`', () => {
  const input = { a : undefined }
  const schema = { a : 'any' }
  expect(isValid({
    input,
    schema,
  })).toBeFalsy()
})

test('type can be `"any"`', () => {
  const input = { a : () => {} }
  const schema = { a : 'any' }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('type can be `"function"`', () => {
  const input = { a : () => {} }
  const schema = { a : 'function' }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('type can be `async`', () => {
  const input = { a : async () => {} }
  const schema = { a : 'async' }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('type can be `promise`', () => {
  const input = { a : delay(1999) }
  const schema = { a : 'promise' }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('type can be `promise` list', () => {
  const input = { a : [ delay(1999) ] }
  const schema = { a : [ 'promise' ] }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('function as schema - false', () => {
  const input = {
    a : {
      ab : () => true,
      ac : 3,
    },
    c : [ 1, 2 ],
  }
  const schema = {
    'a' : {
      ab : /fo/,
      ac : 'number',
    },
    'b?' : 'string',
    'c'  : [ 'number' ],
  }
  expect(isValid({
    input,
    schema,
  })).toBeFalsy()
})

test('regex ok', () => {
  const input = {
    a : {
      ab : 'foo',
      ac : 3,
    },
    c : [ 1, 2 ],
  }
  const schema = {
    'a' : {
      ab : /fo/,
      ac : 'number',
    },
    'b?' : 'string',
    'c'  : [ 'number' ],
  }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('regex !ok', () => {
  const input = {
    a : {
      ab : 'foo',
      ac : 3,
    },
    c : [ 1, 2 ],
  }
  const schema = {
    'a' : {
      ab : /ba/,
      ac : 'number',
    },
    'b?' : 'string',
    'c'  : [ 'number' ],
  }
  expect(isValid({
    input,
    schema,
  })).toBeFalsy()
})

test('optional props is missing', () => {
  const input = {
    a : {
      ab : 'foo',
      ac : 3,
    },
    c : [ 1, 2 ],
  }
  const schema = {
    'a' : {
      ab : 'string',
      ac : 'number',
    },
    'b?' : 'string',
    'c'  : [ 'number' ],
  }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('optional props is wrong type', () => {
  const input = {
    a : {
      ab : 'foo',
      ac : 3,
    },
    b : [],
    c : [ 1, 2 ],
  }
  const schema = {
    'a' : {
      ab : 'string',
      ac : 'number',
    },
    'b?' : 'string',
    'c'  : [ 'number' ],
  }
  expect(isValid({
    input,
    schema,
  })).toBeFalsy()
})

test('optional props - nested', () => {
  const input = {
    a : {
      ab : 'foo',
      ac : 3,
    },
    b : [],
    c : [ 1, 2 ],
  }
  const schema = {
    a : {
      'ab'  : 'string',
      'ac?' : 'number',
    },
    b : 'array',
    c : [ 'number' ],
  }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('optional props is missing - nested', () => {
  const input = {
    a : { ab : 'foo' },
    b : [],
    c : [ 1, 2 ],
  }
  const schema = {
    a : {
      'ab'  : 'string',
      'ac?' : 'number',
    },
    b : 'array',
    c : [ 'number' ],
  }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('optional props is wrong type - nested', () => {
  const input = {
    a : {
      ab : 'foo',
      ac : 'bar',
    },
    b : [],
    c : [ 1, 2 ],
  }
  const schema = {
    a : {
      'ab'  : 'string',
      'ac?' : 'number',
    },
    b : 'array',
    c : [ 'number' ],
  }
  expect(isValid({
    input,
    schema,
  })).toBeFalsy()
})

test('nested schema', () => {
  const input = {
    a : {
      b : 'str',
      c : 3,
      d : 'str',
    },
    b : 'foo',
  }
  const schema = {
    a : {
      b : 'string',
      c : 'number',
      d : 'string',
    },
    b : 'string',
  }

  expect(isValid({
    input,
    schema,
  })).toBeTruthy()

  const invalidInputFirst = {
    a : {
      b : 'str',
      c : 3,
      d : 'str',
    },
    b : 5,
  }

  expect(isValid({
    input : invalidInputFirst,
    schema,
  })).toBeFalsy()

  const invalidInputSecond = {
    a : {
      b : 'str',
      c : 'str',
      d : 'str',
    },
    b : 5,
  }

  expect(isValid({
    input : invalidInputSecond,
    schema,
  })).toBeFalsy()

  const invalidInputThird = {
    a : { b : 'str' },
    b : 5,
  }

  expect(isValid({
    input : invalidInputThird,
    schema,
  })).toBeFalsy()
})

test('array of type', () => {
  const input = {
    a : [ 1, 2 ],
    b : 'foo',
  }
  const schema = {
    a : [ 'number' ],
    b : 'string',
  }

  expect(isValid({
    input,
    schema,
  })).toBeTruthy()

  const invalidInput = {
    a : [ 1, '1' ],
    b : 'foo',
  }

  expect(isValid({
    input : invalidInput,
    schema,
  })).toBeFalsy()
})

test('function as rule', () => {
  const input = {
    a : [ 1, 2, 3, 4 ],
    b : 'foo',
  }
  const invalidInput = {
    a : [ 4 ],
    b : 'foo',
  }

  const schema = {
    a : x => x.length > 2,
    b : 'string',
  }

  expect(isValid({
    input,
    schema,
  })).toBeTruthy()

  expect(isValid({
    input : invalidInput,
    schema,
  })).toBeFalsy()
})

test('input prop is undefined', () => {
  const input = { b : 3 }
  const schema = { a : 'number' }

  expect(isValid({
    input,
    schema,
  })).toBeFalsy()
})

test('enum', () => {
  const input = { a : 'foo' }
  const invalidInput = { a : '' }

  const schema = { a : [ 'foo', 'bar', 'baz' ] }

  expect(isValid({
    input,
    schema,
  })).toBeTruthy()

  expect(isValid({
    input : invalidInput,
    schema,
  })).toBeFalsy()
})

test('readme example', () => {
  const basicSchema = { a : [ 'string' ] }
  const schema = {
    b : [ basicSchema ],
    c : {
      d : { e : 'boolean' },
      f : 'array',
    },
    g : [ 'foo', 'bar', 'baz' ],
  }
  const input = {
    b : [ { a : [ 'led', 'zeppelin' ] } ],
    c : {
      d : { e : true },
      f : [ 'any', 1, null, 'value' ],
    },
    g : 'foo',
  }

  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('should allow additional properties', () => {
  const input = {
    title : 'You shook me',
    year  : 1969,
  }

  expect(isValid({
    input,
    schema : { title : 'string' },
  })).toBeTruthy()
})

test('accepts values as schemas', () => {
  const input = {
    title : 'You shook me',
    genre : 'Blues',
    year  : 1969,
  }
  const schema = {
    title : 'You shook me',
    year  : 1969,
  }
  expect(isValid({
    input,
    schema,
  })).toBeTruthy()
})

test('compatible schemas with nested object', () => {
  const input = {
    foo : 'bar',
    baz : { a : { b : 'c' } },
  }
  const invalidInputFirst = {
    foo : 'bar',
    baz : { a : { b : 1 } },
  }
  const invalidInputSecond = {
    foo : 'bar',
    baz : { a : { b : [] } },
  }
  const invalidInputThird = {
    foo : 'bar',
    baz : { a : { b : null } },
  }
  const schema = {
    foo : 'string',
    baz : { a : { b : 'string' } },
  }

  expect(isValid({
    input,
    schema,
  })).toBeTruthy()

  expect(isValid({
    input : invalidInputFirst,
    schema,
  })).toBeFalsy()
  expect(isValid({
    input : invalidInputSecond,
    schema,
  })).toBeFalsy()
  expect(isValid({
    input : invalidInputThird,
    schema,
  })).toBeFalsy()
})

test('should return true when schema is empty object', () => {
  expect(isValid({
    input  : { a : 1 },
    schema : {},
  })).toBeTruthy()
})

test('when schema is undefined', () => {
  expect(isValid({
    input  : { a : 1 },
    schema : undefined,
  })).toBeFalsy()
})

test('should return false with invalid schema rule', () => {
  const input = {
    foo : 'bar',
    a   : {},
  }
  const inputSecond = { foo : 'bar' }

  const schema = {
    foo : 'string',
    baz : { a : {} },
  }

  expect(isValid({
    input,
    schema,
  })).toBeFalsy()

  expect(isValid({
    input : inputSecond,
    schema,
  })).toBeFalsy()
})

test('array of schemas', () => {
  const input = {
    b : [
      {
        a : 'led',
        b : 1,
      },
      {
        a : 'dancing',
        b : 1,
      },
    ],
  }
  const basicSchema = {
    a : String,
    b : Number,
  }
  const schema = { b : [ basicSchema ] }
  const result = isValid({
    input,
    schema,
  })

  expect(result).toBeTruthy()
})

---------------

isValidAsync

isValidAsync(x: IsValidAsync): Promise<boolean>

Asynchronous version of R.isValid

const input = {a: 1, b: 2}
const invalidInput = {a: 1, b: 'foo'}
const schema = {a: Number, b: async x => {
  await R.delay(100)
  return typeof x === 'number'
}}

const result = await Promise.all([
  R.isValidAsync({schema, input}),
  R.isValidAsync({schema, input: invalidInput})
])
// => [true, false]

Try this R.isValidAsync example in Rambda REPL

R.isValidAsync source
import { forEach } from './forEach'
import { isPromise } from './isPromise'
import { isValid } from './isValid'

export async function isValidAsync({ schema, input }){
  const asyncSchema = {}
  const simpleSchema = {}
  forEach((rule, prop) => {
    if (isPromise(rule)){
      asyncSchema[ prop ] = rule
    } else {
      simpleSchema[ prop ] = rule
    }
  }, schema)

  if (Object.keys(asyncSchema).length === 0)
    return isValid({
      input,
      schema,
    })

  if (
    !isValid({
      input,
      schema : simpleSchema,
    })
  )
    return false

  let toReturn = true

  for (const singleRuleProp in asyncSchema){
    if (toReturn){
      const validated = await asyncSchema[ singleRuleProp ](input[ singleRuleProp ])
      if (!validated) toReturn = false
    }
  }

  return toReturn
}
Tests
import { result } from 'lodash'

import { delay } from './delay'
import { isValidAsync } from './isValidAsync'

const simplePredicate = async x => {
  await delay(100)

  return x > 5
}

test('happy', async () => {
  const input = {
    a          : 1,
    b          : 7,
    c          : 9,
    additional : 'foo',
  }
  const invalidInput = {
    a : 1,
    b : 2,
    c : 9,
  }
  const schema = {
    a : Number,
    b : simplePredicate,
    c : simplePredicate,
  }
  const invalidSchema = {
    a : Boolean,
    b : simplePredicate,
    c : simplePredicate,
  }
  const result = await isValidAsync({
    input,
    schema,
  })
  const invalidResult = await isValidAsync({
    input,
    schema : invalidSchema,
  })
  const withInvalidInput = await isValidAsync({
    input : invalidInput,
    schema,
  })
  expect(result).toBeTruthy()
  expect(invalidResult).toBeFalsy()
  expect(withInvalidInput).toBeFalsy()
})

test('without async rules', async () => {
  const input = {
    a : 1,
    b : 7,
  }
  const schema = {
    a : Number,
    b : x => x > 2,
  }
  const invalidSchema = {
    a : Number,
    b : Boolean,
  }
  const result = await isValidAsync({
    input,
    schema,
  })
  const invalidResult = await isValidAsync({
    input,
    schema : invalidSchema,
  })

  expect(result).toBeTruthy()
  expect(invalidResult).toBeFalsy()
})

test('readme example', async () => {
  const input = {
    a : 1,
    b : 2,
  }
  const invalidInput = {
    a : 1,
    b : 'foo',
  }
  const schema = {
    a : Number,
    b : async x => {
      await delay(100)

      return typeof x === 'number'
    },
  }
  const result = await Promise.all([
    isValidAsync({
      schema,
      input,
    }),
    isValidAsync({
      schema,
      input : invalidInput,
    }),
  ])
  expect(result).toEqual([ true, false ])
})

---------------

join

join<T>(glue: string, list: readonly T[]): string

It returns a string of all list instances joined with a glue.

R.join('-', [1, 2, 3])  // => '1-2-3'

Try this R.join example in Rambda REPL

R.join source
export function join(glue, list){
  if (arguments.length === 1) return _list => join(glue, _list)

  return list.join(glue)
}
Tests
import { join } from './join'

test('curry', () => {
  expect(join('|')([ 'foo', 'bar', 'baz' ])).toEqual('foo|bar|baz')

  expect(join('|', [ 1, 2, 3 ])).toEqual('1|2|3')

  const spacer = join(' ')

  expect(spacer([ 'a', 2, 3.4 ])).toEqual('a 2 3.4')
})

---------------

keys

keys<T extends object>(x: T): readonly (keyof T)[]

It applies Object.keys over x and returns its keys.

R.keys({a:1, b:2})  // => ['a', 'b']

Try this R.keys example in Rambda REPL

R.keys source
export function keys(x){
  return Object.keys(x)
}
Tests
import { keys } from './keys'

test('happy', () => {
  expect(keys({ a : 1 })).toEqual([ 'a' ])
})

---------------

last

last(str: string): string

It returns the last element of input, as the input can be either a string or an array.

const result = [
  R.last([1, 2, 3]),
  R.last('foo'),
]
// => [3, 'o']

Try this R.last example in Rambda REPL

R.last source
export function last(listOrString){
  if (typeof listOrString === 'string'){
    return listOrString[ listOrString.length - 1 ] || ''
  }

  return listOrString[ listOrString.length - 1 ]
}
Tests
import { last } from './last'

test('with list', () => {
  expect(last([ 1, 2, 3 ])).toBe(3)
  expect(last([])).toBeUndefined()
})

test('with string', () => {
  expect(last('abc')).toEqual('c')
  expect(last('')).toEqual('')
})

---------------

lastIndexOf

lastIndexOf<T>(target: T, list: readonly T[]): number

It returns the last index of target in list array.

R.equals is used to determine equality between target and members of list.

If there is no such index, then -1 is returned.

const list = [1, 2, 3, 1, 2, 3]
const result = [
  R.lastIndexOf(2, list),
  R.lastIndexOf(4, list),
]
// => [4, -1]

Try this R.lastIndexOf example in Rambda REPL

R.lastIndexOf source
import { equals } from './equals'

export function lastIndexOf(target, list){
  if (arguments.length === 1) return _list => lastIndexOf(target, _list)

  let index = list.length

  while (--index > 0){
    if (equals(list[ index ], target)){
      return index
    }
  }

  return -1
}
Tests
import { lastIndexOf } from './lastIndexOf'

test('happy', () => {
  const a = lastIndexOf(1, [ 1, 2, 3, 1, 2 ])
  const b = lastIndexOf(1)([ 1, 2, 3, 1, 2 ])

  expect(a).toEqual(3)
  expect(b).toEqual(3)
})

test('false', () => {
  const a = lastIndexOf(10, [ 1, 2, 3, 1, 2 ])

  expect(a).toEqual(-1)
})

---------------

length

length<T>(input: readonly T[]): number

It returns the length property of list or string input.

const result = [
  R.length([1, 2, 3, 4]),
  R.length('foo'),
]
// => [4, 3]

Try this R.length example in Rambda REPL

R.length source
export function length(x){
  if (!x && x !== '' || x.length === undefined){
    return NaN
  }

  return x.length
}
Tests
import { length } from './length'

test('happy', () => {
  expect(length('foo')).toEqual(3)
  expect(length([ 1, 2, 3 ])).toEqual(3)
  expect(length([])).toEqual(0)
})

test('with empty string', () => {
  expect(length('')).toEqual(0)
})

test('with bad input returns NaN', () => {
  expect(length(0)).toBeNaN()
  expect(length({})).toBeNaN()
  expect(length(null)).toBeNaN()
  expect(length(undefined)).toBeNaN()
})

---------------

lens

lens<T, U, V>(getter: (s: T) => U, setter: (a: U, s: T) => V): Lens

It returns a lens for the given getter and setter functions.

The getter gets the value of the focus; the setter sets the value of the focus.

The setter should not mutate the data structure.

const xLens = R.lens(R.prop('x'), R.assoc('x'));

R.view(xLens, {x: 1, y: 2}) // => 1
R.set(xLens, 4, {x: 1, y: 2}) // => {x: 4, y: 2}
R.over(xLens, R.negate, {x: 1, y: 2}) // => {x: -1, y: 2}

Try this R.lens example in Rambda REPL

R.lens source
export function lens(getter, setter){
  return function (functor){
    return function (target){
      return functor(getter(target)).map(focus => setter(focus, target))
    }
  }
}

---------------

lensEq

lensEq<T, U>(lens: Lens, target: T, input: U): boolean

It returns true if data structure focused by the given lens equals to the target value.

R.equals is used to determine equality.

💥 Idea for this method comes from ramda-adjunct library

const list = [ 1, 2, 3 ]
const lens = R.lensIndex(0)
const result = R.lensEq(
  lens, 1, list
)
// => true

Try this R.lensEq example in Rambda REPL

R.lensEq source
import { curry } from './curry'
import { equals } from './equals'
import { view } from './view'

function lensEqFn(
  lens, target, input
){
  return equals(view(lens, input), target)
}

export const lensEq = curry(lensEqFn)
Tests
import { lensEq } from './lensEq'
import { lensIndex } from './lensIndex'
import { lensPath } from './lensPath'

test('with list', () => {
  const list = [ 1, 2, 3 ]
  const lens = lensIndex(0)
  expect(lensEq(
    lens, 1, list
  )).toBeTrue()
  expect(lensEq(lens, 2)(list)).toBeFalse()
})

test('with R.lensPath', () => {
  const input = { a : { b : { c : 1 } } }
  const target = { c : 1 }
  const lens = lensPath('a.b')

  expect(lensEq(lens)(target)(input)).toBeTrue()
  expect(lensEq(
    lens, target, { c : 2 }
  )).toBeFalse()
})

---------------

lensIndex

lensIndex(index: number): Lens

It returns a lens that focuses on specified index.

const list = ['a', 'b', 'c']
const headLens = R.lensIndex(0)

R.view(headLens, list) // => 'a'
R.set(headLens, 'x', list) // => ['x', 'b', 'c']
R.over(headLens, R.toUpper, list) // => ['A', 'b', 'c']

Try this R.lensIndex example in Rambda REPL

R.lensIndex source
import { lens } from './lens'
import { nth } from './nth'
import { update } from './update'

export function lensIndex(index){
  return lens(nth(index), update(index))
}
Tests
import { compose } from './compose'
import { keys } from './keys'
import { lensIndex } from './lensIndex'
import { over } from './over'
import { set } from './set'
import { view } from './view'

const testList = [ { a : 1 }, { b : 2 }, { c : 3 } ]

test('focuses list element at the specified index', () => {
  expect(view(lensIndex(0), testList)).toEqual({ a : 1 })
})

test('returns undefined if the specified index does not exist', () => {
  expect(view(lensIndex(10), testList)).toEqual(undefined)
})

test('sets the list value at the specified index', () => {
  expect(set(
    lensIndex(0), 0, testList
  )).toEqual([ 0, { b : 2 }, { c : 3 } ])
})

test('applies function to the value at the specified list index', () => {
  expect(over(
    lensIndex(2), keys, testList
  )).toEqual([ { a : 1 }, { b : 2 }, [ 'c' ] ])
})

test('can be composed', () => {
  const nestedList = [ 0, [ 10, 11, 12 ], 1, 2 ]
  const composedLens = compose(lensIndex(1), lensIndex(0))

  expect(view(composedLens, nestedList)).toEqual(10)
})

test('set s (get s) === s', () => {
  expect(set(
    lensIndex(0), view(lensIndex(0), testList), testList
  )).toEqual(testList)
})

test('get (set s v) === v', () => {
  expect(view(lensIndex(0), set(
    lensIndex(0), 0, testList
  ))).toEqual(0)
})

test('get (set(set s v1) v2) === v2', () => {
  expect(view(lensIndex(0),
    set(
      lensIndex(0), 11, set(
        lensIndex(0), 10, testList
      )
    ))).toEqual(11)
})

---------------

lensPath

lensPath(path: RamdaPath): Lens

It returns a lens that focuses on specified path.

const lensPath = R.lensPath(['x', 0, 'y'])
const input = {x: [{y: 2, z: 3}, {y: 4, z: 5}]}

R.view(lensPath, input) // => 2

R.set(lensPath, 1, input) 
// => {x: [{y: 1, z: 3}, {y: 4, z: 5}]}

R.over(xHeadYLens, R.negate, input) 
// => {x: [{y: -2, z: 3}, {y: 4, z: 5}]}

Try this R.lensPath example in Rambda REPL

R.lensPath source
import {assocPath} from './assocPath'
import {lens} from './lens'
import {path} from './path'

export function lensPath(key) {
  return lens(path(key), assocPath(key))
}
Tests
import { compose } from './compose'
import { identity } from './identity'
import { inc } from './inc'
import { lensPath } from './lensPath'
import { lensProp } from './lensProp'
import { over } from './over'
import { set } from './set'
import { view } from './view'

const testObj = {
  a : [ { b : 1 }, { b : 2 } ],
  d : 3,
}

test('view', () => {
  expect(view(lensPath('d'), testObj)).toEqual(3)
  expect(view(lensPath('a.0.b'), testObj)).toEqual(1)
  // this is different to ramda, as ramda will return a clone of the input object
  expect(view(lensPath(''), testObj)).toEqual(undefined)
})

test('set', () => {
  expect(set(
    lensProp('d'), 0, testObj
  )).toEqual({
    a : [ { b : 1 }, { b : 2 } ],
    d : 0,
  })
  expect(set(
    lensPath('a.0.b'), 0, testObj
  )).toEqual({
    a : [ { b : 0 }, { b : 2 } ],
    d : 3,
  })
  expect(set(
    lensPath('a.0.X'), 0, testObj
  )).toEqual({
    a : [
      {
        b : 1,
        X : 0,
      },
      { b : 2 },
    ],
    d : 3,
  })
  expect(set(
    lensPath([]), 0, testObj
  )).toEqual(0)
})

test('over', () => {
  expect(over(
    lensPath('d'), inc, testObj
  )).toEqual({
    a : [ { b : 1 }, { b : 2 } ],
    d : 4,
  })
  expect(over(
    lensPath('a.1.b'), inc, testObj
  )).toEqual({
    a : [ { b : 1 }, { b : 3 } ],
    d : 3,
  })
  expect(over(
    lensProp('X'), identity, testObj
  )).toEqual({
    a : [ { b : 1 }, { b : 2 } ],
    d : 3,
    X : undefined,
  })
  expect(over(
    lensPath('a.0.X'), identity, testObj
  )).toEqual({
    a : [
      {
        b : 1,
        X : undefined,
      },
      { b : 2 },
    ],
    d : 3,
  })
})

test('compose', () => {
  const composedLens = compose(lensPath('a'), lensPath('1.b'))
  expect(view(composedLens, testObj)).toEqual(2)
})

test('set s (get s) === s', () => {
  expect(set(
    lensPath([ 'd' ]), view(lensPath([ 'd' ]), testObj), testObj
  )).toEqual(testObj)
  expect(set(
    lensPath([ 'a', 0, 'b' ]),
    view(lensPath([ 'a', 0, 'b' ]), testObj),
    testObj
  )).toEqual(testObj)
})

test('get (set s v) === v', () => {
  expect(view(lensPath([ 'd' ]), set(
    lensPath([ 'd' ]), 0, testObj
  ))).toEqual(0)
  expect(view(lensPath([ 'a', 0, 'b' ]), set(
    lensPath([ 'a', 0, 'b' ]), 0, testObj
  ))).toEqual(0)
})

test('get (set(set s v1) v2) === v2', () => {
  const p = [ 'd' ]
  const q = [ 'a', 0, 'b' ]
  expect(view(lensPath(p), set(
    lensPath(p), 11, set(
      lensPath(p), 10, testObj
    )
  ))).toEqual(11)
  expect(view(lensPath(q), set(
    lensPath(q), 11, set(
      lensPath(q), 10, testObj
    )
  ))).toEqual(11)
})

---------------

lensProp

lensProp(prop: string): {
  <T, U>(obj: T): U

It returns a lens that focuses on specified property prop.

const xLens = R.lensProp('x');
const input = {x: 1, y: 2}

R.view(xLens, input) // => 1

R.set(xLens, 4, input) 
// => {x: 4, y: 2}

R.over(xLens, R.negate, input) 
// => {x: -1, y: 2}

Try this R.lensProp example in Rambda REPL

R.lensProp source
import { assoc } from './assoc'
import { lens } from './lens'
import { prop } from './prop'

export function lensProp(key){
  return lens(prop(key), assoc(key))
}
Tests
import { compose } from './compose'
import { identity } from './identity'
import { inc } from './inc'
import { lensProp } from './lensProp'
import { over } from './over'
import { set } from './set'
import { view } from './view'

const testObj = {
  a : 1,
  b : 2,
  c : 3,
}

test('focuses object the specified object property', () => {
  expect(view(lensProp('a'), testObj)).toEqual(1)
})

test('returns undefined if the specified property does not exist', () => {
  expect(view(lensProp('X'), testObj)).toEqual(undefined)
})

test('sets the value of the object property specified', () => {
  expect(set(
    lensProp('a'), 0, testObj
  )).toEqual({
    a : 0,
    b : 2,
    c : 3,
  })
})

test('adds the property to the object if it doesn\'t exist', () => {
  expect(set(
    lensProp('d'), 4, testObj
  )).toEqual({
    a : 1,
    b : 2,
    c : 3,
    d : 4,
  })
})

test('applies function to the value of the specified object property', () => {
  expect(over(
    lensProp('a'), inc, testObj
  )).toEqual({
    a : 2,
    b : 2,
    c : 3,
  })
})

test('applies function to undefined and adds the property if it doesn\'t exist', () => {
  expect(over(
    lensProp('X'), identity, testObj
  )).toEqual({
    a : 1,
    b : 2,
    c : 3,
    X : undefined,
  })
})

test('can be composed', () => {
  const nestedObj = {
    a : { b : 1 },
    c : 2,
  }
  const composedLens = compose(lensProp('a'), lensProp('b'))

  expect(view(composedLens, nestedObj)).toEqual(1)
})

test('set s (get s) === s', () => {
  expect(set(
    lensProp('a'), view(lensProp('a'), testObj), testObj
  )).toEqual(testObj)
})

test('get (set s v) === v', () => {
  expect(view(lensProp('a'), set(
    lensProp('a'), 0, testObj
  ))).toEqual(0)
})

test('get (set(set s v1) v2) === v2', () => {
  expect(view(lensProp('a'),
    set(
      lensProp('a'), 11, set(
        lensProp('a'), 10, testObj
      )
    ))).toEqual(11)
})

---------------

lensSatisfies

lensSatisfies<T, U>(predicate: (x: T) => boolean, lens: Lens, input: U): boolean

It returns true if data structure focused by the given lens satisfies the predicate.

💥 Idea for this method comes from ramda-adjunct library

const fn = R.lensSatisfies(x => x > 5, R.lensIndex(0))
const result = [
  fn([10, 20, 30]),
  fn([1, 2, 3]),
]
// => [true, false]

Try this R.lensSatisfies example in Rambda REPL

R.lensSatisfies source
import { curry } from './curry'
import { view } from './view'

function lensSatisfiesFn(
  predicate, lens, input
){
  return Boolean(predicate(view(lens, input)))
}

export const lensSatisfies = curry(lensSatisfiesFn)
Tests
import { lensIndex } from './lensIndex'
import { lensPath } from './lensPath'
import { lensSatisfies } from './lensSatisfies'

const predicate = x => x > 1

test('with list', () => {
  const lens = lensIndex(0)
  const fn = lensSatisfies(predicate, lens)
  expect(fn([ 10, 20, 30 ])).toBeTrue()
  expect(fn([ 1, 2, 3 ])).toBeFalse()
})

test('with R.lensPath', () => {
  const input1 = { a : { b : 10 } }
  const input2 = { a : { b : 1 } }
  const lens = lensPath('a.b')
  const fn = lensSatisfies(predicate, lens)

  expect(fn(input1)).toBeTrue()
  expect(fn(input2)).toBeFalse()
})

---------------

map

map<T, U>(fn: ObjectIterator<T, U>, iterable: Dictionary<T>): Dictionary<U>

It returns the result of looping through iterable with fn.

It works with both array and object.

💥 Unlike Ramda's map, here property and input object are passed as arguments to fn, when iterable is an object.

const fn = x => x * 2
const fnWhenObject = (val, prop)=>{
  return `${prop}-${val}`
}

const iterable = [1, 2]
const obj = {a: 1, b: 2}

const result = [ 
  R.map(fn, list),
  R.map(fnWhenObject, obj)
]
// => [ [1, 4], {a: 'a-1', b: 'b-2'}]

Try this R.map example in Rambda REPL

R.map source
import { _isArray } from './_internals/_isArray'
import { _keys } from './_internals/_keys'

export function mapArray(
  fn, list, isIndexed = false
){
  let index = 0
  const willReturn = Array(list.length)

  while (index < list.length){
    willReturn[ index ] = isIndexed ? fn(list[ index ], index) : fn(list[ index ])

    index++
  }

  return willReturn
}

export function mapObject(fn, obj){
  let index = 0
  const keys = _keys(obj)
  const len = keys.length
  const willReturn = {}

  while (index < len){
    const key = keys[ index ]
    willReturn[ key ] = fn(
      obj[ key ], key, obj
    )
    index++
  }

  return willReturn
}

export function map(fn, list){
  if (arguments.length === 1) return _list => map(fn, _list)
  if (list === undefined) return []
  if (_isArray(list)) return mapArray(fn, list)

  return mapObject(fn, list)
}
Tests
import { map } from './map'

const double = x => x * 2

const sampleObject = {
  a : 1,
  b : 2,
  c : 3,
  d : 4,
}

test('with array', () => {
  expect(map(double, [ 1, 2, 3 ])).toEqual([ 2, 4, 6 ])
})

test('with object', () => {
  const obj = {
    a : 1,
    b : 2,
  }

  expect(map(double, obj)).toEqual({
    a : 2,
    b : 4,
  })
})

test('pass input object as third argument', () => {
  const obj = {
    a : 1,
    b : 2,
  }
  const iterator = (
    val, prop, inputObject
  ) => {
    expect(inputObject).toEqual(obj)

    return val * 2
  }

  expect(map(iterator, obj)).toEqual({
    a : 2,
    b : 4,
  })
})

test('with object passes property as second argument', () => {
  map((_, prop) => {
    expect(typeof prop).toEqual('string')
  })(sampleObject)
})

/**
 * https://github.com/selfrefactor/rambda/issues/77
 */
test('when undefined instead of array', () => {
  expect(map(double, undefined)).toEqual([])
})

---------------

mapAsync

mapAsync<T, K>(fn: AsyncIterable<T, K>, list: readonly T[]): Promise<readonly K[]>

Sequential asynchronous mapping with fn over members of list.

async function fn(x){
  await R.delay(1000)

  return x+1
}

const result = R.mapAsync(fn, [1, 2, 3])
// `result` resolves after 3 seconds to `[2, 3, 4]`

Try this R.mapAsync example in Rambda REPL

R.mapAsync source
import { _isArray } from './_internals/_isArray'

async function mapAsyncFn(fn, listOrObject){
  if (_isArray(listOrObject)){
    const willReturn = []
    let i = 0
    for (const a of listOrObject){
      willReturn.push(await fn(a, i++))
    }

    return willReturn
  }

  const willReturn = {}
  for (const prop in listOrObject){
    willReturn[ prop ] = await fn(listOrObject[ prop ], prop)
  }

  return willReturn
}

export function mapAsync(fn, listOrObject){
  if (arguments.length === 1){
    return async _listOrObject => mapAsyncFn(fn, _listOrObject)
  }

  return new Promise((resolve, reject) => {
    mapAsyncFn(fn, listOrObject).then(resolve)
      .catch(reject)
  })
}
Tests
import { composeAsync } from './composeAsync'
import { delay } from './delay'
import { map } from './map'
import { mapAsync } from './mapAsync'

const rejectDelay = a =>
  new Promise((_, reject) => {
    setTimeout(() => {
      reject(a + 20)
    }, 100)
  })

test('happy', async () => {
  const fn = async (x, prop) => {
    await delay(100)
    expect(prop).toBeNumber()

    return x + 1
  }
  const result = await mapAsync(fn, [ 1, 2, 3 ])
  expect(result).toEqual([ 2, 3, 4 ])
})

test('with object', async () => {
  const fn = async (x, prop) => {
    expect(prop).toBeString()

    return x + 1
  }
  const result = await mapAsync(fn, {
    a : 1,
    b : 2,
  })
  expect(result).toEqual({
    a : 2,
    b : 3,
  })
})

test('with R.composeAsync', async () => {
  const result = await composeAsync(
    map(x => x + 1),
    mapAsync(async x => {
      delay(x)

      return x
    }),
    map(x => x * 10)
  )([ 1, 2, 3 ])
  expect(result).toEqual([ 11, 21, 31 ])
})

test('error', async () => {
  try {
    await mapAsync(rejectDelay)([ 1, 2, 3 ])
  } catch (err){
    expect(err).toBe(21)
  }
})

---------------

mapAsyncLimit

mapAsyncLimit<T, K>(fn: AsyncIterable<T, K>, limit: number, list: readonly T[]): Promise<readonly K[]>

It is similar to R.mapFastAsync in that it uses Promise.all but not over the whole list, rather than with only slice from list with length limit.

💥 For example usage, please check R.mapAsyncLimit tests.

R.mapAsyncLimit source
import { mapFastAsync, mapFastAsyncFn } from './mapFastAsync'
import { splitEvery } from './splitEvery'

async function mapAsyncLimitFn(
  iterable, limit, list
){
  if (list.length < limit) return mapFastAsync(iterable, list)

  const slices = splitEvery(limit, list)

  let toReturn = []
  for (const slice of slices){
    const iterableResult = await mapFastAsyncFn(iterable, slice)
    toReturn = [ ...toReturn, ...iterableResult ]
  }

  return toReturn
}

export function mapAsyncLimit(
  iterable, limit, list
){
  if (arguments.length === 2){
    return async _list => mapAsyncLimitFn(
      iterable, limit, _list
    )
  }

  return new Promise((resolve, reject) => {
    mapAsyncLimitFn(
      iterable, limit, list
    ).then(resolve)
      .catch(reject)
  })
}
Tests
import isCI from 'is-ci'

import { composeAsync } from './composeAsync'
import { delay } from './delay'
import { mapAsync } from './mapAsync'
import { mapAsyncLimit } from './mapAsyncLimit'
import { toDecimal } from './toDecimal'

jest.setTimeout(30000)

test('happy', async () => {
  const limit = 3
  const startTime = new Date().getTime()
  const list = [ 1, 2, 3, 4, 5, 6, 7, 8, 9 ]
  const iterable = async x => {
    await delay(500)

    return x + 1
  }
  const result = await mapAsyncLimit(
    iterable, limit, list
  )
  const endTime = new Date().getTime()
  const diffTime = endTime - startTime

  const startTime2 = new Date().getTime()
  await mapAsync(iterable, list)
  const endTime2 = new Date().getTime()
  const diffTime2 = endTime2 - startTime2

  const methodScale = toDecimal((diffTime2 - diffTime) / 1000, 0)
  expect(result).toEqual([ 2, 3, 4, 5, 6, 7, 8, 9, 10 ])
  if (!isCI) expect(methodScale).toBe(limit)
})

const fn = async x => {
  await delay(100)

  return x + 1
}

test('with R.composeAsync', async () => {
  const result = await composeAsync(mapAsyncLimit(fn, 2), x =>
    x.map(xx => xx + 1))([ 1, 2, 3, 4, 5, 6 ])
  expect(result).toEqual([ 3, 4, 5, 6, 7, 8 ])
})

test('fallback to R.mapFastAsync', async () => {
  const result = await mapAsyncLimit(
    fn, 4, [ 1, 2, 3 ]
  )
  expect(result).toEqual([ 2, 3, 4 ])
})

---------------

mapFastAsync

mapFastAsync<T, K>(fn: AsyncIterable<T, K>, list: readonly T[]): Promise<readonly K[]>

Parrallel asynchronous mapping with fn over members of list.

async function fn(x){
  await R.delay(1000)

  return x+1
}

const result = R.mapFastAsync(fn, [1, 2, 3])
// `result` resolves after 1 second to `[2, 3, 4]`

Try this R.mapFastAsync example in Rambda REPL

R.mapFastAsync source
export async function mapFastAsyncFn(fn, arr){
  const promised = arr.map((a, i) => fn(a, i))

  return Promise.all(promised)
}

export function mapFastAsync(fn, arr){
  if (arguments.length === 1){
    return async holder => mapFastAsyncFn(fn, holder)
  }

  return new Promise((resolve, reject) => {
    mapFastAsyncFn(fn, arr).then(resolve)
      .catch(reject)
  })
}
Tests
import { composeAsync } from './composeAsync'
import { delay } from './delay'
import { map } from './map'
import { mapFastAsync } from './mapFastAsync'

test('happy', async () => {
  const fn = async x => {
    await delay(100)

    return x + 10
  }
  const result = await mapFastAsync(fn, [ 1, 2, 3 ])
  const curriedResult = await mapFastAsync(fn)([ 1, 2, 3 ])
  expect(result).toEqual([ 11, 12, 13 ])
  expect(curriedResult).toEqual([ 11, 12, 13 ])
})

test('composeAsync', async () => {
  const result = await composeAsync(
    mapFastAsync(async x => {
      await delay(100)

      return x + 1
    }),
    mapFastAsync(async x => {
      await delay(100)

      return x + 10
    }),
    map(x => x * 10)
  )([ 1, 2, 3 ])
  expect(result).toEqual([ 21, 31, 41 ])
})

test('error', async () => {
  try {
    const fn = async x => {
      JSON.parse('{:')
    }
    const result = await mapFastAsync(fn, [ 1, 2, 3 ])
  } catch (err){
    expect(err.message).toBe('Unexpected token : in JSON at position 1')
  }
})

test('pass index as second argument', async () => {
  await mapFastAsync((x, i) => {
    expect(x % 10).toBe(0)
    expect(typeof i).toBe('number')
  },
  [ 10, 20, 30 ])
})

---------------

mapIndexed

mapIndexed<T, U>(fn: ObjectIterator<T, U>, iterable: Dictionary<T>): Dictionary<U>

Same as R.map, but it passes index as second argument to the iterator, when looping over arrays.

R.mapIndexed source
import { _isArray } from './_internals/_isArray'
import { mapArray, mapObject } from './map'

export function mapIndexed(fn, iterable){
  if (arguments.length === 1){
    return _iterable => mapIndexed(fn, _iterable)
  }
  if (iterable === undefined) return []
  if (_isArray(iterable)) return mapArray(
    fn, iterable, true
  )

  return mapObject(fn, iterable)
}
Tests
import { map } from './map'
import { mapIndexed } from './mapIndexed'

const iterator = (x, i) => {
  expect(x).toBeNumber()
  expect(i).toBeNumber()
}

test('happy', () => {
  const list = [ 1, 2, 3 ]
  mapIndexed(iterator, list)
  mapIndexed(iterator)(list)
})

test('with object', () => {
  const iterator = x => x + 1
  const obj = { a : 1 }
  expect(mapIndexed(iterator, obj)).toEqual(map(iterator, obj))
})

test('with bad input', () => {
  expect(mapIndexed(iterator, undefined)).toEqual([])
})

---------------

mapKeys

mapKeys<T, U>(changeKeyFn: (x: string) => string, obj: { readonly [key: string]: T}): U

It takes an object and returns a new object with changed keys according to changeKeyFn function.

const obj = {a: 1, b: 2}
const changeKeyFn = prop => `{prop}_foo`
const result = R.mapKeys(changeKeyFn, obj)
// => {a_foo: 1, b_foo: 2}

Try this R.mapKeys example in Rambda REPL

R.mapKeys source
export function mapKeys(changeKeyFn, obj){
  if (arguments.length === 1) return _obj => mapKeys(changeKeyFn, _obj)
  const toReturn = {}

  Object.keys(obj).forEach(prop => toReturn[ changeKeyFn(prop) ] = obj[ prop ])

  return toReturn
}
Tests
import { mapKeys } from './mapKeys'

const obj = {
  a : 1,
  b : 2,
}
const changeKeyFn = prop => `${ prop }_foo`
const expected = {
  a_foo : 1,
  b_foo : 2,
}

test('happy', () => {
  const result = mapKeys(changeKeyFn, obj)

  expect(result).toEqual(expected)
})

test('curried', () => {
  const result = mapKeys(changeKeyFn)(obj)

  expect(result).toEqual(expected)
})

---------------

mapToObject

mapToObject<T, U>(fn: (input: T) => object|false, list: readonly T[]): U

This method allows to generate an object from a list using input function fn.

This function must return either an object or false for every member of list input.

If false is returned, then this element of list will be skipped in the calculation of the result.

All of returned objects will be merged to generate the final result.

const list = [1, 2, 3, 12]
const fn = x => {
  if(x > 10) return false
  return x % 2 ? {[x]: x + 1}: {[x]: x + 10}
}

const result = mapToObject(fn, list)
const expected = {'1': 2, '2': 12, '3': 4}
// => `result` is equal to `expected`

Try this R.mapToObject example in Rambda REPL

R.mapToObject source
import { map } from './map'
import { mergeAll } from './mergeAll'
import { ok } from './ok'
import { type } from './type'

export function mapToObject(fn, list){
  if (arguments.length === 1){
    return listHolder => mapToObject(fn, listHolder)
  }
  ok(type(fn), type(list))('Function', 'Array')

  return mergeAll(map(fn, list))
}
Tests
import { mapToObject } from './mapToObject'

const list = [ 1, 2, 3 ]
const fn = x => x % 2 ? { [ x ] : x + 1 } : { [ x ] : x + 10 }
const expected = {
  1 : 2,
  2 : 12,
  3 : 4,
}

test('happy', () => {
  const result = mapToObject(fn, list)
  expect(result).toEqual(expected)
})

test('curried', () => {
  const result = mapToObject(fn)(list)
  expect(result).toEqual(expected)
})

test('string.fn test', () => {
  const list = [ 'auto', 'bar=false', 'foo', 'baz=1.5', 's=more', 'k=2' ]
  const fn = x => {
    const [ key, value ] = x.split('=')
    if (value === undefined || value === 'true'){
      return { [ key ] : true }
    }
    if (value === 'false'){
      return { [ key ] : false }
    }

    if (Number.isNaN(Number(value))){
      return { [ key ] : value }
    }

    return { [ key ] : Number(value) }
  }

  const expectedResult = {
    auto : true,
    foo  : true,
    bar  : false,
    baz  : 1.5,
    s    : 'more',
    k    : 2,
  }
  const result = mapToObject(fn, list)

  expect(result).toEqual(expectedResult)
})

test('bad path', () => {
  expect(() => mapToObject(1, null)).toThrow()
})

---------------

mapToObjectAsync

mapToObjectAsync<T, U>(fn: (input: T) => Promise<object|false>, list: readonly T[]): Promise<U>

Asynchronous version of R.mapToObject

R.mapToObjectAsync source
import { mapAsync } from './mapAsync'

export async function mapToObjectAsyncFn(fn, list){
  let toReturn = {}

  const innerIterable = async x => {
    const intermediateResult = await fn(x)
    if (intermediateResult === false) return
    toReturn = {
      ...toReturn,
      ...intermediateResult,
    }
  }

  await mapAsync(innerIterable, list)

  return toReturn
}

export function mapToObjectAsync(fn, list){
  if (arguments.length === 1){
    return async _list => mapToObjectAsyncFn(fn, _list)
  }

  return new Promise((resolve, reject) => {
    mapToObjectAsyncFn(fn, list).then(resolve)
      .catch(reject)
  })
}
Tests
import { composeAsync } from './composeAsync'
import { delay } from './delay'
import { mapToObjectAsync } from './mapToObjectAsync'

const list = [ 1, 2, 3, 12 ]
const fn = async x => {
  await delay(100)
  if (x > 10) return false

  return x % 2 ? { [ `key${ x }` ] : x + 1 } : { [ `key${ x }` ] : x + 10 }
}

const expected = {
  key1 : 2,
  key2 : 12,
  key3 : 4,
}

test('happy', async () => {
  const result = await mapToObjectAsync(fn, list)
  expect(result).toEqual(expected)
})

test('with R.composeAsync', async () => {
  const result = await composeAsync(mapToObjectAsync(fn), x =>
    x.filter(xx => xx > 1))(list)

  expect(result).toEqual({
    key2 : 12,
    key3 : 4,
  })
})

---------------

match

match(regExpression: RegExp, str: string): readonly string[]

Curried version of String.prototype.match which returns empty array, when there is no match.

const result = [
  R.match('a', 'foo'),
  R.match(/([a-z]a)/g, 'bananas')
]
// => [[], ['ba', 'na', 'na']]

Try this R.match example in Rambda REPL

R.match source
export function match(pattern, input){
  if (arguments.length === 1) return _input => match(pattern, _input)

  const willReturn = input.match(pattern)

  return willReturn === null ? [] : willReturn
}
Tests
import { equals } from './equals'
import { match } from './match'

test('happy', () => {
  expect(match(/a./g)('foo bar baz')).toEqual([ 'ar', 'az' ])
})

test('fallback', () => {
  expect(match(/a./g)('foo')).toEqual([])
})

test('with string', () => {
  expect(match('a', 'foo')).toEqual([])
  expect(equals(match('o', 'foo'), [ 'o' ])).toBeTrue()
})

test('throwing', () => {
  expect(() => {
    match(/a./g, null)
  }).toThrowWithMessage(TypeError, 'Cannot read property \'match\' of null')
})

---------------

mathMod

R.mathMod behaves like the modulo operator should mathematically, unlike the % operator (and by extension, R.modulo). So while -17 % 5 is -2, mathMod(-17, 5) is 3.

💥 Explanation is taken from Ramda documentation site.

const result = [
  R.mathMod(-17, 5),
  R.mathMod(17, 5),
  R.mathMod(17, -5),  
  R.mathMod(17, 0)   
]
// => [3, 2, NaN, NaN]

Try this R.mathMod example in Rambda REPL

---------------

max

max<T extends Ord>(x: T, y: T): T

It returns the greater value between x and y.

const result = [
  R.max(5, 7),  
  R.max('bar', 'foo'),  
]
// => [7, 'foo']

Try this R.max example in Rambda REPL

R.max source
export function max(x, y){
  if (arguments.length === 1) return _y => max(x, _y)

  return y > x ? y : x
}
Tests
import { max } from './max'

test('with number', () => {
  expect(max(2, 1)).toBe(2)
})

test('with string', () => {
  expect(max('foo')('bar')).toBe('foo')
  expect(max('bar')('baz')).toBe('baz')
})

---------------

maxBy

maxBy<T>(compareFn: (input: T) => Ord, x: T, y: T): T

It returns the greater value between x and y according to compareFn function.

const compareFn = Math.abs

R.maxBy(compareFn, 5, -7) // => -7

Try this R.maxBy example in Rambda REPL

R.maxBy source
import { curry } from './curry'

export function maxByFn(
  compareFn, x, y
){
  return compareFn(y) > compareFn(x) ? y : x
}

export const maxBy = curry(maxByFn)
Tests
import { maxBy } from './maxBy'

test('happy', () => {
  expect(maxBy(
    Math.abs, -5, 2
  )).toEqual(-5)
})

test('curried', () => {
  expect(maxBy(Math.abs)(2, -5)).toEqual(-5)
  expect(maxBy(Math.abs)(2)(-5)).toEqual(-5)
})

---------------

maybe

maybe<T>(ifRule: boolean, whenIf: T | Func<T>, whenElse: T | Func<T>): T

It acts as ternary operator and it is helpful when we have nested ternaries.

All of the inputs can be either direct values or anonymous functions. This is helpful if we don't want to evaluate certain paths as we can wrap this logic in a function.

const x = 4
const y = 8

const ifRule = x > 2
const whenIf = y > 10 ? 3 : 7
const whenElse = () => {
  // just to show that it won't be evaluated
  return JSON.parse('{a:')
}

const result = R.maybe(
  ifRule,
  whenIf,
  whenElse,
)
// `result` is `7`

Try this R.maybe example in Rambda REPL

R.maybe source
import { type } from './type'

export function maybe(
  ifRule, whenIf, whenElse
){
  const whenIfInput =
    ifRule && type(whenIf) === 'Function' ? whenIf() : whenIf

  const whenElseInput =
    !ifRule && type(whenElse) === 'Function' ? whenElse() : whenElse

  return ifRule ? whenIfInput : whenElseInput
}
Tests
import { maybe } from './maybe'

const WHEN_IF = 'WHEN_IF'
const WHEN_ELSE = 'WHEN_ELSE'

test('prevent type error', () => {
  const x = 5
  const y = null
  const ifRule = x > 3

  const result = maybe(
    ifRule, WHEN_IF, () => y.a === 'foo'
  )

  expect(result).toBe(WHEN_IF)
})

test('whenElse is a function', () => {
  const x = 2
  const y = { a : 1 }
  const ifRule = x > 3

  const result = maybe(
    ifRule, WHEN_IF, () => y.a === 'foo'
  )

  expect(result).toBeFalse()
})

test('whenIf', () => {
  const x = 5
  const ifRule = x > 3

  const result = maybe(
    ifRule, WHEN_IF, WHEN_ELSE
  )

  expect(result).toBe(WHEN_IF)
})

test('whenIf is a function', () => {
  const x = 5
  const ifRule = () => x > 3

  const result = maybe(
    ifRule, () => WHEN_IF, WHEN_ELSE
  )

  expect(result).toBe(WHEN_IF)
})

test('whenElse', () => {
  const x = 1
  const ifRule = x > 3

  const result = maybe(
    ifRule, WHEN_IF, WHEN_ELSE
  )

  expect(result).toBe(WHEN_ELSE)
})

---------------

mean

mean(list: readonly number[]): number

It returns the mean value of list input.

R.mean([ 2, 7 ])
// => 4.5

Try this R.mean example in Rambda REPL

R.mean source
import { sum } from './sum'

export function mean(list){
  return sum(list) / list.length
}
Tests
import { mean } from './mean'

test('happy', () => {
  expect(mean([ 2, 7 ])).toBe(4.5)
})

test('with NaN', () => {
  expect(mean([])).toBeNaN()
})

---------------

median

median(list: readonly number[]): number

It returns the median value of list input.

R.median([ 7, 2, 10, 9 ]) // => 8

Try this R.median example in Rambda REPL

R.median source
import { mean } from './mean'

export function median(list){
  const len = list.length
  if (len === 0) return NaN
  const width = 2 - len % 2
  const idx = (len - width) / 2

  return mean(Array.prototype.slice
    .call(list, 0)
    .sort((a, b) => {
      if (a === b) return 0

      return a < b ? -1 : 1
    })
    .slice(idx, idx + width))
}
Tests
import { median } from './median'

test('happy', () => {
  expect(median([ 2 ])).toEqual(2)
  expect(median([ 7, 2, 10, 2, 9 ])).toEqual(7)
})

test('with empty array', () => {
  expect(median([])).toBeNaN()
})

---------------

memoize

memoize<T, K extends readonly any[]>(fn: (...inputs: K) => T): (...inputs: K) => T

When fn is called for a second time with the same input, then the cache result is returned instead of calling again fn.

let result = 0
const fn = (a,b) =>{
  result++

  return a + b
}
const memoized = R.memoize(fn)
memoized(1, 2)
memoized(1, 2)

// => `result` is equal to `1`

Try this R.memoize example in Rambda REPL

R.memoize source
import { compose } from './compose'
import { map } from './map'
import { replace } from './replace'
import { sort } from './sort'
import { take } from './take'
import { type } from './type'

const cache = {}

const normalizeObject = obj => {
  const sortFn = (a, b) => a > b ? 1 : -1
  const willReturn = {}
  compose(map(prop => willReturn[ prop ] = obj[ prop ]),
    sort(sortFn))(Object.keys(obj))

  return willReturn
}

const stringify = a => {
  if (type(a) === 'String'){
    return a
  } else if ([ 'Function', 'Async' ].includes(type(a))){
    const compacted = replace(
      /\s{1,}/g, ' ', a.toString()
    )

    return replace(
      /\s/g, '_', take(15, compacted)
    )
  } else if (type(a) === 'Object'){
    return JSON.stringify(normalizeObject(a))
  }

  return JSON.stringify(a)
}

const generateProp = (fn, ...inputArguments) => {
  let propString = ''
  inputArguments.forEach(inputArgument => {
    propString += `${ stringify(inputArgument) }_`
  })

  return `${ propString }${ stringify(fn) }`
}
// with weakmaps
export function memoize(fn, ...inputArguments){
  if (arguments.length === 1){
    return (...inputArgumentsHolder) => memoize(fn, ...inputArgumentsHolder)
  }

  const prop = generateProp(fn, ...inputArguments)
  if (prop in cache) return cache[ prop ]

  if (type(fn) === 'Async'){
    return new Promise(resolve => {
      fn(...inputArguments).then(result => {
        cache[ prop ] = result
        resolve(result)
      })
    })
  }

  const result = fn(...inputArguments)
  cache[ prop ] = result

  return result
}
Tests
import { memoize } from './memoize'

test('memoize function without input arguments', () => {
  const fn = () => 4
  const memoized = memoize(fn)
  expect(typeof memoized()).toBe('function')
})

test('happy', () => {
  let counter = 0

  const fn = ({ a, b, c }) => {
    counter++

    return a + b - c
  }
  const memoized = memoize(fn)
  expect(memoized({
    a : 1,
    c : 3,
    b : 2,
  })).toBe(0)
  expect(counter).toBe(1)
  expect(memoized({
    c : 3,
    a : 1,
    b : 2,
  })).toBe(0)
  expect(counter).toBe(1)
})

test('normal function', () => {
  let counter = 0
  const fn = (a, b) => {
    counter++

    return a + b
  }
  const memoized = memoize(fn)
  expect(memoized(1, 2)).toBe(3)
  expect(memoized(1, 2)).toBe(3)
  expect(memoized(1, 2)).toBe(3)
  expect(counter).toBe(1)
  expect(memoized(2, 2)).toBe(4)
  expect(counter).toBe(2)
  expect(memoized(1, 2)).toBe(3)
  expect(counter).toBe(2)
})

test('async function', async () => {
  let counter = 0
  const delay = ms =>
    new Promise(resolve => {
      setTimeout(resolve, ms)
    })
  const fn = async (
    ms, a, b
  ) => {
    await delay(ms)
    counter++

    return a + b
  }

  const memoized = memoize(fn)
  expect(await memoized(
    100, 1, 2
  )).toBe(3)
  expect(await memoized(
    100, 1, 2
  )).toBe(3)
  expect(await memoized(
    100, 1, 2
  )).toBe(3)
  expect(counter).toBe(1)
  expect(await memoized(
    100, 2, 2
  )).toBe(4)
  expect(counter).toBe(2)
  expect(await memoized(
    100, 1, 2
  )).toBe(3)
  expect(counter).toBe(2)
})

test('string as argument', () => {
  let count = 0
  const foo = 'foo'
  const tester = memoize(n => {
    count++

    return `${ n }bar`
  })
  tester(foo)
  tester(foo)
  tester(foo)

  expect(tester(foo)).toEqual('foobar')

  expect(count).toEqual(1)

  tester('baz')

  expect(tester('baz')).toEqual('bazbar')

  expect(count).toEqual(2)
})

---------------

merge

merge<O1 extends object, O2 extends object>(target: O1, newProps: O2): Merge<O2, O1, 'flat'>

It creates a copy of target object with overidden newProps properties.

const target = { 'foo': 0, 'bar': 1 }
const newProps = { 'foo': 7 }

const result = R.merge(target, newProps)
// => { 'foo': 7, 'bar': 1 }

Try this R.merge example in Rambda REPL

R.merge source
export function merge(target, newProps){
  if (arguments.length === 1) return _newProps => merge(target, _newProps)

  return Object.assign(
    {}, target || {}, newProps || {}
  )
}
Tests
import { merge } from './merge'

const obj = {
  foo : 1,
  bar : 2,
}

test('happy', () => {
  expect(merge(obj, { bar : 20 })).toEqual({
    foo : 1,
    bar : 20,
  })
})

test('curry', () => {
  expect(merge(obj)({ baz : 3 })).toEqual({
    foo : 1,
    bar : 2,
    baz : 3,
  })
})

/**
 * https://github.com/selfrefactor/rambda/issues/77
 */
test('when undefined or null instead of object', () => {
  expect(merge(null, undefined)).toEqual({})
  expect(merge(obj, null)).toEqual(obj)
  expect(merge(obj, undefined)).toEqual(obj)
  expect(merge(undefined, obj)).toEqual(obj)
})

---------------

mergeAll

mergeAll<T>(list: readonly object[]): T

It merges all objects of list array sequentially and returns the result.

const list = [
  {a: 1},
  {b: 2},
  {c: 3}
]
const result = R.mergeAll(list)
const expected = {
  a: 1,
  b: 2,
  c: 3
}
// => `result` is equal to `expected`

Try this R.mergeAll example in Rambda REPL

R.mergeAll source
import { map } from './map'
import { merge } from './merge'

export function mergeAll(arr){
  let willReturn = {}
  map(val => {
    willReturn = merge(willReturn, val)
  }, arr)

  return willReturn
}
Tests
import { mergeAll } from './mergeAll'

test('case 1', () => {
  const arr = [ { a : 1 }, { b : 2 }, { c : 3 } ]
  const expectedResult = {
    a : 1,
    b : 2,
    c : 3,
  }
  expect(mergeAll(arr)).toEqual(expectedResult)
})

test('case 2', () => {
  expect(mergeAll([ { foo : 1 }, { bar : 2 }, { baz : 3 } ])).toEqual({
    foo : 1,
    bar : 2,
    baz : 3,
  })
})

---------------

mergeDeepRight

mergeDeepRight<O1 extends object, O2 extends object>(x: O1, y: O2): Merge<O2, O1, 'deep'>

Creates a new object with the own properties of the first object merged with the own properties of the second object. If a key exists in both objects:

  • and both values are objects, the two values will be recursively merged
  • otherwise the value from the second object will be used.
R.mergeDeepRight source
import { type } from './type'

export function mergeDeepRight(target, source){
  if (arguments.length === 1){
    return sourceHolder => mergeDeepRight(target, sourceHolder)
  }

  const willReturn = JSON.parse(JSON.stringify(target))

  Object.keys(source).forEach(key => {
    if (type(source[ key ]) === 'Object'){
      if (type(target[ key ]) === 'Object'){
        willReturn[ key ] = mergeDeepRight(target[ key ], source[ key ])
      } else {
        willReturn[ key ] = source[ key ]
      }
    } else {
      willReturn[ key ] = source[ key ]
    }
  })

  return willReturn
}
Tests
// import { mergeDeepRight } from 'ramda'
import { mergeDeepRight } from './mergeDeepRight'

const slave = {
  name    : 'evilMe',
  age     : 10,
  contact : {
    a     : 1,
    email : '[email protected]',
  },
}
const master = {
  age     : 40,
  contact : { email : '[email protected]' },
  songs   : { title : 'Remains the same' },
}

test('happy', () => {
  const result = mergeDeepRight(slave, master)
  const curryResult = mergeDeepRight(slave)(master)
  const expected = {
    age     : 40,
    name    : 'evilMe',
    contact : {
      a     : 1,
      email : '[email protected]',
    },
    songs : { title : 'Remains the same' },
  }

  expect(result).toEqual(expected)
  expect(curryResult).toEqual(expected)
})

test('ramda compatible test 1', () => {
  const a = {
    w : 1,
    x : 2,
    y : { z : 3 },
  }
  const b = {
    a : 4,
    b : 5,
    c : { d : 6 },
  }
  const result = mergeDeepRight(a, b)
  const expected = {
    w : 1,
    x : 2,
    y : { z : 3 },
    a : 4,
    b : 5,
    c : { d : 6 },
  }

  expect(result).toEqual(expected)
})

test('ramda compatible test 2', () => {
  const a = {
    a : {
      b : 1,
      c : 2,
    },
    y : 0,
  }
  const b = {
    a : {
      b : 3,
      d : 4,
    },
    z : 0,
  }
  const result = mergeDeepRight(a, b)
  const expected = {
    a : {
      b : 3,
      c : 2,
      d : 4,
    },
    y : 0,
    z : 0,
  }

  expect(result).toEqual(expected)
})

test('ramda compatible test 3', () => {
  const a = {
    w : 1,
    x : { y : 2 },
  }
  const result = mergeDeepRight(a, { x : { y : 3 } })
  const expected = {
    w : 1,
    x : { y : 3 },
  }
  expect(result).toEqual(expected)
})

---------------

mergeLeft

mergeLeft<O1 extends object, O2 extends object>(target: O1, newProps: O2): Merge<O2, O1, 'flat'>

Same as R.merge, but in opposite direction.

const result = R.mergeLeft(
  {a: 10},
  {a: 1, b: 2}
)
// => {a:10, b: 2}

Try this R.mergeLeft example in Rambda REPL

R.mergeLeft source
import { merge } from './merge'

export function mergeLeft(x, y){
  if (arguments.length === 1) return _y => mergeLeft(x, _y)

  return merge(y, x)
}
Tests
import { mergeLeft } from './mergeLeft'

const obj = {
  foo : 1,
  bar : 2,
}

test('happy', () => {
  expect(mergeLeft({ bar : 20 }, obj)).toEqual({
    foo : 1,
    bar : 20,
  })
})

test('curry', () => {
  expect(mergeLeft({ baz : 3 })(obj)).toEqual({
    foo : 1,
    bar : 2,
    baz : 3,
  })
})

test('when undefined or null instead of object', () => {
  expect(mergeLeft(null, undefined)).toEqual({})
  expect(mergeLeft(obj, null)).toEqual(obj)
  expect(mergeLeft(obj, undefined)).toEqual(obj)
  expect(mergeLeft(undefined, obj)).toEqual(obj)
})

---------------

min

min<T extends Ord>(x: T, y: T): T

It returns the lesser value between x and y.

const result = [
  R.min(5, 7),  
  R.min('bar', 'foo'),  
]
// => [5, 'bar']

Try this R.min example in Rambda REPL

R.min source
export function min(x, y){
  if (arguments.length === 1) return _y => min(x, _y)

  return y < x ? y : x
}
Tests
import { min } from './min'

test('happy', () => {
  expect(min(2, 1)).toBe(1)
  expect(min(1)(2)).toBe(1)
})

---------------

minBy

minBy<T>(compareFn: (input: T) => Ord, x: T, y: T): T

It returns the lesser value between x and y according to compareFn function.

const compareFn = Math.abs

R.minBy(compareFn, -5, 2) // => -5

Try this R.minBy example in Rambda REPL

R.minBy source
import { curry } from './curry'

export function minByFn(
  compareFn, x, y
){
  return compareFn(y) < compareFn(x) ? y : x
}

export const minBy = curry(minByFn)
Tests
import { minBy } from './minBy'

test('happy', () => {
  expect(minBy(
    Math.abs, -5, 2
  )).toEqual(2)
})

test('curried', () => {
  expect(minBy(Math.abs)(2, -5)).toEqual(2)
  expect(minBy(Math.abs)(2)(-5)).toEqual(2)
})

---------------

modulo

Curried version of x%y.

R.modulo(17, 3) // => 2

Try this R.modulo example in Rambda REPL

---------------

move

move<T>(fromIndex: number, toIndex: number, list: readonly T[]): readonly T[]

It returns a copy of list with exchanged fromIndex and toIndex elements.

💥 Rambda.move doesn't support negative indexes - it throws an error.

const list = [1, 2, 3]
const result = R.move(0, 1, list)
// => [2, 1, 3]

Try this R.move example in Rambda REPL

R.move source
import { curry } from './curry'

function moveFn(
  fromIndex, toIndex, list
){
  if (fromIndex < 0 || toIndex < 0){
    throw new Error('Rambda.move does not support negative indexes')
  }
  if (fromIndex > list.length - 1 || toIndex > list.length - 1) return list

  const clone = list.slice()
  clone[ fromIndex ] = list[ toIndex ]
  clone[ toIndex ] = list[ fromIndex ]

  return clone
}

export const move = curry(moveFn)
Tests
import { move } from './move'
const list = [ 1, 2, 3, 4 ]

test('happy', () => {
  const result = move(
    0, 1, list
  )

  expect(result).toEqual([ 2, 1, 3, 4 ])
})

test('with negative index', () => {
  const errorMessage = 'Rambda.move does not support negative indexes'
  expect(() => move(
    0, -1, list
  )).toThrowWithMessage(Error, errorMessage)
  expect(() => move(
    -1, 0, list
  )).toThrowWithMessage(Error, errorMessage)
})

test('when indexes are outside the list outbounds', () => {
  const result1 = move(
    10, 1, list
  )
  const result2 = move(
    1, 10, list
  )

  expect(result1).toEqual(list)
  expect(result2).toEqual(list)
})

---------------

multiply

Curried version of x*y.

R.multiply(2, 4) // => 8

Try this R.multiply example in Rambda REPL

---------------

negate

R.negate(420)// => -420

Try this R.negate example in Rambda REPL

---------------

nextIndex

nextIndex(index: number, list: readonly any[]): number

It returns the next index of the list.

If we have reached the end of the list, then it will return 0.

const list = [1, 2, 3]

const result = [
  R.nextIndex(0, list),
  R.nextIndex(1, list),
  R.nextIndex(2, list),
  R.nextIndex(10, list)
]
// => [1, 2, 0, 0]

Try this R.nextIndex example in Rambda REPL

R.nextIndex source
export function nextIndex(index, list){
  return index >= list.length - 1 ? 0 : index + 1
}
Tests
import { nextIndex } from './nextIndex'

const list = [ 1, 2, 3, 4 ]

test('happy path', () => {
  expect(nextIndex(2, list)).toEqual(3)
})

test('go back to the start', () => {
  expect(nextIndex(3, list)).toEqual(0)
})

test('current index is too big', () => {
  expect(nextIndex(32, list)).toEqual(0)
})

---------------

none

none<T>(predicate: (x: T) => boolean, list: readonly T[]): boolean

It returns true, if all members of array list returns false, when applied as argument to predicate function.

const list = [ 0, 1, 2, 3, 4 ]
const predicate = x => x > 6

const result = R.none(predicate, arr)
// => true

Try this R.none example in Rambda REPL

R.none source
export function none(predicate, list){
  if (arguments.length === 1) return _list => none(predicate, _list)

  for (let i = 0; i < list.length; i++){
    if (!predicate(list[ i ])) return true
  }

  return false
}
Tests
import { none } from './none'

const isEven = n => n % 2 === 0
const isOdd = n => n % 2 === 1
const arr = [ 1, 3, 5, 7, 9, 11 ]

test('when true', () => {
  expect(none(isEven, arr)).toBeTrue()
})

test('when false curried', () => {
  expect(none(isOdd)(arr)).toBeFalse()
})

---------------

not

not(input: any): boolean

It returns a boolean negated version of input.

R.not(false) // true

Try this R.not example in Rambda REPL

R.not source
export function not(input){
  return !input
}
Tests
import { not } from './not'

test('not', () => {
  expect(not(false)).toEqual(true)
  expect(not(true)).toEqual(false)
  expect(not(0)).toEqual(true)
  expect(not(1)).toEqual(false)
})

---------------

nth

nth<T>(index: number, list: readonly T[]): T | undefined

Curried version of list[index].

const list = [1, 2, 3]
const str = 'foo'

const result = [
  R.nth(2, list),
  R.nth(6, list),
  R.nth(0, str),
]
// => [3, undefined, 'f']

Try this R.nth example in Rambda REPL

R.nth source
export function nth(index, list){
  if (arguments.length === 1) return _list => nth(index, _list)

  const idx = index < 0 ? list.length + index : index

  return Object.prototype.toString.call(list) === '[object String]' ?
    list.charAt(idx) :
    list[ idx ]
}
Tests
import { nth } from './nth'

test('happy', () => {
  expect(nth(2, [ 1, 2, 3, 4 ])).toEqual(3)
})

test('with curry', () => {
  expect(nth(2)([ 1, 2, 3, 4 ])).toEqual(3)
})

test('with string', () => {
  expect(nth(2)('foo')).toEqual('o')
})

test('with negative index', () => {
  expect(nth(-3)([ 1, 2, 3, 4 ])).toEqual(2)
})

---------------

objOf

objOf<T, K extends string>(key: K, value: T): Record<K, T>

It returns a new object with the provided key and value.

const result = [
  R.objOf('foo', 42),
  R.objOf(null, undefined),
]
// => [{foo: 42}, {null: undefined}]

Try this R.objOf example in Rambda REPL

R.objOf source
export function objOf(key, value) {
  if (arguments.length === 1) {
    return (_value) => objOf(key, _value)
  }
  
  return {
    [key]: value
  }
}
Tests
import { objOf } from "./objOf";
import { objOf as objOfRamda } from "ramda";
import { compareCombinations } from "./_internals/testUtils";

test("happy", function () {
  expect(objOf("foo", 42)).toEqual({ foo: 42 });
});

test("with bad inputs", function () {
  expect(objOf(null, undefined)).toEqual({ null: undefined });
});

test("curried", function () {
  expect(objOf("foo")(42)).toEqual({ foo: 42 });
});

describe("brute force", () => {
  const possibleInputs = [0, 1, null, undefined, [], {}];

  compareCombinations({
    firstInput: possibleInputs,
    secondInput: possibleInputs,
    callback: (errorsCounters) => {
      expect(errorsCounters).toMatchInlineSnapshot(`
        Object {
          "ERRORS_MESSAGE_MISMATCH": 0,
          "ERRORS_TYPE_MISMATCH": 0,
          "RESULTS_MISMATCH": 0,
          "SHOULD_NOT_THROW": 0,
          "SHOULD_THROW": 0,
        }
      `);
    },
    fn: objOf,
    fnRamda: objOfRamda,
  });
});

---------------

of

of<T>(x: T): readonly T[]
R.of(null); // => [null]
R.of([42]); // => [[42]]

Try this R.of example in Rambda REPL

R.of source
export function of(value){
  return [ value ]
}
Tests
import { of } from './of'

test('happy', () => {
  expect(of(3)).toEqual([ 3 ])

  expect(of(null)).toEqual([ null ])
})

---------------

ok

ok(...inputs: readonly any[]): (...schemas: readonly any[]) => void | never

It checks if inputs are following schemas specifications according to R.isValid.

If validation fails, it throws.

💥 It is same as R.pass but instead of returning false, it throws an error.

const result = R.ok(
  1,
  ['foo', 'bar']
)(
  Number,
  [String]
)
// => undefined

Try this R.ok example in Rambda REPL

R.ok source
import { any } from './any'
import { glue } from './glue'
import { fromPrototypeToString, isValid } from './isValid'
import { map } from './map'
import { type } from './type'

export function schemaToString(schema){
  if (type(schema) !== 'Object'){
    return fromPrototypeToString(schema).rule
  }

  return map(x => {
    const { rule, parsed } = fromPrototypeToString(x)
    const xType = type(x)

    if (xType === 'Function' && !parsed) return 'Function'

    return parsed ? rule : xType
  }, schema)
}

export function check(singleInput, schema){
  return isValid({
    input  : { singleInput },
    schema : { singleInput : schema },
  })
}

export function ok(...inputs){
  return (...schemas) => {
    let failedSchema

    const anyError = any((singleInput, i) => {
      const schema = schemas[ i ] === undefined ? schemas[ 0 ] : schemas[ i ]

      const checked = check(singleInput, schema)
      if (!checked){
        failedSchema = JSON.stringify({
          input  : singleInput,
          schema : schemaToString(schema),
        })
      }

      return !checked
    }, inputs)

    if (anyError){
      const errorMessage =
        inputs.length > 1 ?
          glue(`
        Failed R.ok -
        reason: ${ failedSchema }
        all inputs: ${ JSON.stringify(inputs) }
        all schemas: ${ JSON.stringify(schemas.map(schemaToString)) }
      `,
          '\n') :
          `Failed R.ok - ${ failedSchema }`

      throw new Error(errorMessage)
    }
  }
}
Tests
import { ok, schemaToString } from './ok'

test('happy', () => {
  expect(() => {
    ok(
      1, 'foo', {}
    )(
      'number', 'string', 'object'
    )
  }).not.toThrow()
})

test('when validation fails', () => {
  const errorMessage = `Failed R.ok -
reason: {"input":{},"schema":"string"}
all inputs: [1,"foo",{}]
all schemas: ["number","string","string"]`

  expect(() =>
    ok(
      1, 'foo', {}
    )(
      'number', 'string', 'string'
    )).toThrowWithMessage(Error, errorMessage)
})

/*
  TODO
  What about
  {a: Function}
*/
test('schema in error message', () => {
  const result = schemaToString({
    _a : [ Number ],
    a  : Number,
    b  : x => x > 2,
    c  : [ 'foo', 'bar' ],
    d  : [ { a : String } ],
    e  : 'boolean',
    f  : Array,
    h  : Object,
  })

  expect(result).toMatchInlineSnapshot(`
    Object {
      "_a": "Array",
      "a": "number",
      "b": "Function",
      "c": "Array",
      "d": "Array",
      "e": "String",
      "f": "array",
      "h": "object",
    }
  `)
})

test('error contains schema', () => {
  try {
    ok(
      1, 'foo', {}
    )(
      { a : Number }, String, String
    )
    expect(false).toBeTrue()
  } catch (e){
    expect(e.message.startsWith('Failed R.ok -')).toBeTruthy()
    expect(e).toBeInstanceOf(Error)
  }
})

test('when not throws with single schema', () => {
  expect(() => ok(
    1, 2, 3
  )('number')).not.toThrow()
})

test('when throws with single schema', () => {
  expect(() => ok(
    1, 2, '3'
  )('number')).toThrow()
})

test('when throws with single input', () => {
  expect(() => ok('3')('number')).toThrow()
})

---------------

omit

omit<T, K extends string>(propsToOmit: readonly K[], obj: T): Omit<T, K>

It returns a partial copy of an obj without propsToOmit properties.

💥 When using this method with TypeScript, it is much easier to pass propsToOmit as an array. If passing a string, you will need to explicitly declare the output type.

const obj = {a: 1, b: 2, c: 3}
const propsToOmit = 'a,c,d'
const propsToOmitList = ['a', 'c', 'd']

const result = [
  R.omit(propsToOmit, obj), 
  R.omit(propsToOmitList, obj) 
]
// => [{b: 2}, {b: 2}]

Try this R.omit example in Rambda REPL

R.omit source
export function omit(propsToOmit, obj){
  if (arguments.length === 1) return _obj => omit(propsToOmit, _obj)

  if (obj === null || obj === undefined){
    return undefined
  }

  const propsToOmitValue =
    typeof propsToOmit === 'string' ? propsToOmit.split(',') : propsToOmit

  const willReturn = {}

  for (const key in obj){
    if (!propsToOmitValue.includes(key)){
      willReturn[ key ] = obj[ key ]
    }
  }

  return willReturn
}
Tests
import { omit } from './omit'

test('with string as condition', () => {
  const obj = {
    a : 1,
    b : 2,
    c : 3,
  }
  const result = omit('a,c', obj)
  const resultCurry = omit('a,c')(obj)
  const expectedResult = { b : 2 }

  expect(result).toEqual(expectedResult)
  expect(resultCurry).toEqual(expectedResult)
})

test('with null', () => {
  expect(omit('a,b', null)).toEqual(undefined)
})

test('doesn\'t work with number as property', () => {
  expect(omit([ 42 ], {
    a  : 1,
    42 : 2,
  })).toEqual({
    42 : 2,
    a  : 1,
  })
})

test('happy', () => {
  expect(omit([ 'a', 'c' ])({
    a : 'foo',
    b : 'bar',
    c : 'baz',
  })).toEqual({ b : 'bar' })
})

---------------

once

once<T extends (...args: readonly any[]) => any>(func: T): T

It returns a function, which invokes only once fn function.

let result = 0
const addOnce = R.once((x) => result = result + x)

addOnce(1)
addOnce(1)
// => 1

Try this R.once example in Rambda REPL

R.once source
import { curry } from './curry'

function onceFn(fn, context){
  let result

  return function (){
    if (fn){
      result = fn.apply(context || this, arguments)
      fn = null
    }

    return result
  }
}

export function once(fn, context){
  if (arguments.length === 1){
    const wrap = onceFn(fn, context)

    return curry(wrap)
  }

  return onceFn(fn, context)
}
Tests
import { once } from './once'

test('with counter', () => {
  let counter = 0
  const runOnce = once(x => {
    counter++

    return x + 2
  })
  expect(runOnce(1)).toEqual(3)
  runOnce(1)
  runOnce(1)
  runOnce(1)
  expect(counter).toEqual(1)
})

test('happy path', () => {
  const addOneOnce = once((
    a, b, c
  ) => a + b + c, 1)

  expect(addOneOnce(
    10, 20, 30
  )).toBe(60)
  expect(addOneOnce(40)).toEqual(60)
})

---------------

or

or<T, U>(a: T, b: U): T | U

Logical OR

R.or(false, true); // => true
R.or(false, false); // => false
R.or(false, 'foo'); // => 'foo'

Try this R.or example in Rambda REPL

R.or source
export function or(a, b){
  if (arguments.length === 1) return _b => or(a, _b)

  return a || b
}
Tests
import { or } from './or'

test('happy', () => {
  expect(or(0, 'foo')).toBe('foo')
  expect(or(true, true)).toBeTrue()
  expect(or(false)(true)).toBeTrue()
  expect(or(false, false)).toBeFalse()
})

---------------

over

over<T>(lens: Lens, fn: Arity1Fn, value: T): T

It returns a copied Object or Array with modified value received by applying function fn to lens focus.

const headLens = R.lensIndex(0)
 
R.over(headLens, R.toUpper, ['foo', 'bar', 'baz']) // => ['FOO', 'bar', 'baz']

Try this R.over example in Rambda REPL

R.over source
import { curry } from './curry'

const Identity = x => ({
  x,
  map : fn => Identity(fn(x)),
})

function overFn(
  lens, fn, object
){
  return lens(x => Identity(fn(x)))(object).x
}

export const over = curry(overFn)
Tests
import { assoc } from './assoc'
import { lens } from './lens'
import { lensIndex } from './lensIndex'
import { lensPath } from './lensPath'
import { over } from './over'
import { prop } from './prop'
import { toUpper } from './toUpper'

const testObject = {
  foo : 'bar',
  baz : {
    a : 'x',
    b : 'y',
  },
}

test('assoc lens', () => {
  const assocLens = lens(prop('foo'), assoc('foo'))
  const result = over(
    assocLens, toUpper, testObject
  )
  const expected = {
    ...testObject,
    foo : 'BAR',
  }
  expect(result).toEqual(expected)
})

test('path lens', () => {
  const pathLens = lensPath('baz.a')
  const result = over(
    pathLens, toUpper, testObject
  )
  const expected = {
    ...testObject,
    baz : {
      a : 'X',
      b : 'y',
    },
  }
  expect(result).toEqual(expected)
})

test('index lens', () => {
  const indexLens = lensIndex(0)
  const result = over(indexLens, toUpper)([ 'foo', 'bar' ])
  expect(result).toEqual([ 'FOO', 'bar' ])
})

---------------

partial

partial<V0, V1, T>(fn: (x0: V0, x1: V1) => T, args: readonly [V0]): (x1: V1) => T

It is very similar to R.curry, but you can pass initial arguments when you create the curried function.

R.partial will keep returning a function until all the arguments that the function fn expects are passed. The name comes from the fact that you partially inject the inputs.

💥 Rambda's partial doesn't need the input arguments to be wrapped as array.

const fn = (title, firstName, lastName) => {
  return title + ' ' + firstName + ' ' + lastName + '!'
}

const canPassAnyNumberOfArguments = R.partial(fn, 'Hello')
const ramdaStyle = R.partial(fn, ['Hello'])

const finalFn = canPassAnyNumberOfArguments('Foo')

finalFn('Bar') // =>  'Hello, Foo Bar!'

Try this R.partial example in Rambda REPL

R.partial source
export function partial(fn, ...args){
  const len = fn.length

  return (...rest) => {
    if (args.length + rest.length >= len){
      return fn(...args, ...rest)
    }

    return partial(fn, ...[ ...args, ...rest ])
  }
}
Tests
import { partial } from './partial'
import { type } from './type'

const greet = (
  salutation, title, firstName, lastName
) =>
  salutation + ', ' + title + ' ' + firstName + ' ' + lastName + '!'

test('happy', () => {
  const canPassAnyNumberOfArguments = partial(
    greet, 'Hello', 'Ms.'
  )
  const fn = canPassAnyNumberOfArguments('foo')
  const sayHello = partial(greet, [ 'Hello' ])
  const sayHelloRamda = partial(sayHello, [ 'Ms.' ])

  expect(type(fn)).toBe('Function')

  expect(fn('bar')).toBe('Hello, Ms. foo bar!')
  expect(sayHelloRamda('foo', 'bar')).toBe('Hello, Ms. foo bar!')
})

test('extra arguments are ignored', () => {
  const canPassAnyNumberOfArguments = partial(
    greet, 'Hello', 'Ms.'
  )
  const fn = canPassAnyNumberOfArguments('foo')

  expect(type(fn)).toBe('Function')

  expect(fn(
    'bar', 1, 2
  )).toBe('Hello, Ms. foo bar!')
})

test('when array is input', () => {
  const fooFn = (
    a, b, c, d
  ) => ({
    a,
    b,
    c,
    d,
  })
  const barFn = partial(
    fooFn, [ 1, 2 ], []
  )

  expect(barFn(1, 2)).toEqual({
    a : [ 1, 2 ],
    b : [],
    c : 1,
    d : 2,
  })
})

test('ramda spec', () => {
  const sayHello = partial(greet, 'Hello')
  const sayHelloToMs = partial(sayHello, 'Ms.')

  expect(sayHelloToMs('Jane', 'Jones')).toBe('Hello, Ms. Jane Jones!')
})

---------------

partialCurry

partialCurry<Input, PartialInput, Output>(
  fn: (input: Input) => Output, 
  partialInput: PartialInput,
): (input: Pick<Input, Exclude<keyof Input, keyof PartialInput>>) => Output

R.partialCurry is a curry helper designed specifically for functions accepting object as a single argument.

Initially the function knows only a part from the whole input object and then R.partialCurry helps in preparing the function for the second part, when it receives the rest of the input.

💥 Curried function can be asynchronous

const fn = ({ a, b, c }) => a + b + c
const curried = partialCurry(fn, { a : 1 })
const result = curried({
  b : 2,
  c : 3,
})
// => 6

Try this R.partialCurry example in Rambda REPL

R.partialCurry source
import { merge } from './merge'
import { type } from './type'

export function partialCurry(fn, input){
  return rest => {
    if (type(fn) === 'Async'){
      return new Promise((resolve, reject) => {
        fn(merge(rest, input)).then(resolve)
          .catch(reject)
      })
    }

    return fn(merge(rest, input))
  }
}
Tests
import { delay } from './delay'
import { partialCurry } from './partialCurry'
import { type } from './type'

test('with plain function', () => {
  const fn = ({ a, b, c }) => a + b + c
  const curried = partialCurry(fn, { a : 1 })

  expect(type(curried)).toEqual('Function')
  expect(curried({
    b : 2,
    c : 3,
  })).toEqual(6)
})

test('with function that throws an error', () => {
  const fn = ({ a, b, c }) => {
    throw new Error('foo')
  }
  const curried = partialCurry(fn, { a : 1 })

  expect(type(curried)).toEqual('Function')
  expect(() =>
    curried({
      b : 2,
      c : 3,
    })).toThrowWithMessage(Error, 'foo')
})

test('with async', async () => {
  const fn = async ({ a, b, c }) => {
    await delay(100)

    return a + b + c
  }

  const curried = partialCurry(fn, { a : 1 })

  const result = await curried({
    b : 2,
    c : 3,
  })

  expect(result).toEqual(6)
})

test('async function throwing an error', async () => {
  const fn = async ({ a, b, c }) => {
    await delay(100)
    throw new Error('foo')
  }

  const curried = partialCurry(fn, { a : 1 })

  try {
    await curried({
      b : 2,
      c : 3,
    })
    expect(true).toBeFalsy()
  } catch (e){
    expect(e.message).toBe('foo')
  }
})

---------------

partition

partition<T>(
  predicate: Predicate<T>,
  input: readonly T[]
): readonly [readonly T[], readonly T[]]

It will return array of two objects/arrays according to predicate function. The first member holds all instanses of input that pass the predicate function, while the second member - those who doesn't.

const list = [1, 2, 3]
const obj = {a: 1, b: 2, c: 3}
const predicate = x => x > 2

const result = [
  R.partition(predicate, list),
  R.partition(predicate, obj)
]
const expected = [
  [[3], [1, 2]],
  [{c: 3},  {a: 1, b: 2}],
]
// `result` is equal to `expected`

Try this R.partition example in Rambda REPL

R.partition source
import { _isArray } from './_internals/_isArray'

export function partitionObject(predicate, iterable){
  const yes = {}
  const no = {}
  Object.entries(iterable).forEach(([ prop, value ]) => {
    if (predicate(value, prop)){
      yes[ prop ] = value
    } else {
      no[ prop ] = value
    }
  })

  return [ yes, no ]
}

export function partitionArray(predicate, list){
  const yes = []
  const no = []
  let counter = -1

  while (counter++ < list.length - 1){
    if (predicate(list[ counter ])){
      yes.push(list[ counter ])
    } else {
      no.push(list[ counter ])
    }
  }

  return [ yes, no ]
}

export function partition(predicate, iterable){
  if (arguments.length === 1){
    return listHolder => partition(predicate, listHolder)
  }
  if (!_isArray(iterable)) return partitionObject(predicate, iterable)

  return partitionArray(predicate, iterable)
}
Tests
import { partition } from './partition'

test('with array', () => {
  const predicate = x => x > 2
  const list = [ 1, 2, 3, 4 ]

  const result = partition(predicate, list)
  const expectedResult = [
    [ 3, 4 ],
    [ 1, 2 ],
  ]

  expect(result).toEqual(expectedResult)
})

test('with object', () => {
  const predicate = (value, prop) => {
    expect(typeof prop).toBe('string')

    return value > 2
  }
  const hash = {
    a : 1,
    b : 2,
    c : 3,
    d : 4,
  }

  const result = partition(predicate)(hash)
  const expectedResult = [
    {
      c : 3,
      d : 4,
    },
    {
      a : 1,
      b : 2,
    },
  ]

  expect(result).toEqual(expectedResult)
})

test('readme example', () => {
  const list = [ 1, 2, 3 ]
  const obj = {
    a : 1,
    b : 2,
    c : 3,
  }
  const predicate = x => x > 2

  const result = [ partition(predicate, list), partition(predicate, obj) ]
  const expected = [
    [ [ 3 ], [ 1, 2 ] ],
    [
      { c : 3 },
      {
        a : 1,
        b : 2,
      },
    ],
  ]
  expect(result).toEqual(expected)
})

---------------

pass

pass(...inputs: readonly any[]): (...rules: readonly any[]) => boolean

It checks if inputs are following schemas specifications according to R.isValid.

const result = R.pass(
  1,
  ['foo','bar']
)(
  Number,
  [String]
)
// => true

Try this R.pass example in Rambda REPL

R.pass source
import { any } from './any'
import { check } from './ok'

export function pass(...inputs){
  return (...schemas) =>
    any((x, i) => {
      const schema = schemas[ i ] === undefined ? schemas[ 0 ] : schemas[ i ]

      return !check(x, schema)
    }, inputs) === false
}
Tests
import { pass } from './pass'

test('true on success', () => {
  const result = pass(
    1, 'foo', {}
  )(
    'number', 'string', 'object'
  )

  expect(result).toBeTrue()
})

test('false on failure', () => {
  expect(pass(
    1, 'foo', {}
  )(
    'number', 'string', 'string'
  )).toBeFalse()
})

test('true when single schema', () => {
  expect(pass(
    1, 2, 3
  )('number')).toBeTrue()
})

test('false when single schema', () => {
  expect(pass(
    1, 'foo', {}
  )('number')).toBeFalse()
})

test('array of schemas', () => {
  const result = pass([ { a : 1 }, { a : 2 }, { a : 3 } ])([ { a : Number } ])
  expect(result).toBeTruthy()
})

test('reame example', () => {
  const result = pass(1, [ 'foo', 'bar' ])(Number, [ String ])
  expect(result).toBeTruthy()
})

---------------

path

path<Input, T>(pathToSearch: Path, obj: Input): T | undefined

If pathToSearch is 'a.b' then it will return 1 if obj is {a:{b:1}}.

It will return undefined, if such path is not found.

💥 String anotation of pathToSearch is one of the differences between Rambda and Ramda.

const obj = {a: {b: 1}}
const pathToSearch = 'a.b'
const pathToSearchList = ['a', 'b']

const result = [
  R.path(pathToSearch, obj),
  R.path(pathToSearchList, obj),
  R.path('a.b.c.d', obj)
]
// => [1, 1, undefined]

Try this R.path example in Rambda REPL

R.path source
export function path(pathInput, obj){
  if (arguments.length === 1) return _obj => path(pathInput, _obj)

  if (obj === null || obj === undefined){
    return undefined
  }
  let willReturn = obj
  let counter = 0

  const pathArrValue =
    typeof pathInput === 'string' ? pathInput.split('.') : pathInput

  while (counter < pathArrValue.length){
    if (willReturn === null || willReturn === undefined){
      return undefined
    }
    willReturn = willReturn[ pathArrValue[ counter ] ]
    counter++
  }

  return willReturn
}
Tests
import { path } from './path'

test('with array inside object', () => {
  const obj = { a : { b : [ 1, { c : 1 } ] } }

  expect(path('a.b.1.c', obj)).toBe(1)
})

test('works with undefined', () => {
  const obj = { a : { b : { c : 1 } } }

  expect(path('a.b.c.d.f', obj)).toBeUndefined()
  expect(path('foo.babaz', undefined)).toBeUndefined()
  expect(path('foo.babaz')(undefined)).toBeUndefined()
})

test('works with string instead of array', () => {
  expect(path('foo.bar.baz')({ foo : { bar : { baz : 'yes' } } })).toEqual('yes')
})

test('path', () => {
  expect(path([ 'foo', 'bar', 'baz' ])({ foo : { bar : { baz : 'yes' } } })).toEqual('yes')

  expect(path([ 'foo', 'bar', 'baz' ])(null)).toBeUndefined()

  expect(path([ 'foo', 'bar', 'baz' ])({ foo : { bar : 'baz' } })).toBeUndefined()
})

---------------

pathEq

pathEq(pathToSearch: Path, target: any, input: any): boolean

It returns true if pathToSearch of input object is equal to target value.

pathToSearch is passed to R.path, which means that it can be either a string or an array. Also equality between target and the found value is determined by R.equals.

const path = 'a.b'
const target = {c: 1}
const input = {a: {b: {c: 1}}}

const result = R.pathEq(
  path,
  target,
  input
)
// => true

Try this R.pathEq example in Rambda REPL

R.pathEq source
import { curry } from './curry'
import { equals } from './equals'
import { path } from './path'

function pathEqFn(
  pathToSearch, target, input
){
  return equals(path(pathToSearch, input), target)
}

export const pathEq = curry(pathEqFn)
Tests
import { pathEq } from './pathEq'

test('when true', () => {
  const path = 'a.b'
  const obj = { a : { b : { c : 1 } } }
  const target = { c : 1 }

  expect(pathEq(
    path, target, obj
  )).toBeTrue()
})

test('when false', () => {
  const path = 'a.b'
  const obj = { a : { b : 1 } }
  const target = 2

  expect(pathEq(path, target)(obj)).toBeFalse()
})

test('when wrong path', () => {
  const path = 'foo.bar'
  const obj = { a : { b : 1 } }
  const target = 2

  expect(pathEq(
    path, target, obj
  )).toBeFalse()
})

---------------

pathOr

pathOr<T>(defaultValue: T, pathToSearch: Path, obj: any): T

It reads obj input and returns either R.path(pathToSearch, obj) result or defaultValue input.

const defaultValue = 'DEFAULT_VALUE'
const pathToSearch = 'a.b'
const pathToSearchList = ['a', 'b']

const obj = {
  a : {
    b : 1
  }
}

const result = [
  R.pathOr(DEFAULT_VALUE, pathToSearch, obj) 
  R.pathOr(DEFAULT_VALUE, pathToSearchList, obj) 
  R.pathOr(DEFAULT_VALUE, 'a.b.c', obj) 
]
// => [1, 1, 'DEFAULT_VALUE']

Try this R.pathOr example in Rambda REPL

R.pathOr source
import { curry } from './curry'
import { defaultTo } from './defaultTo'
import { path } from './path'

function pathOrFn(
  defaultValue, list, obj
){
  return defaultTo(defaultValue, path(list, obj))
}

export const pathOr = curry(pathOrFn)
Tests
import { pathOr } from './pathOr'

test('with undefined', () => {
  const result = pathOr(
    'foo', 'x.y', { x : { y : 1 } }
  )

  expect(result).toEqual(1)
})

test('with null', () => {
  const result = pathOr(
    'foo', 'x.y', null
  )

  expect(result).toEqual('foo')
})

test('with NaN', () => {
  const result = pathOr(
    'foo', 'x.y', NaN
  )

  expect(result).toEqual('foo')
})

test('curry case (x)(y)(z)', () => {
  const result = pathOr('foo')('x.y.z')({ x : { y : { a : 1 } } })

  expect(result).toEqual('foo')
})

test('curry case (x)(y,z)', () => {
  const result = pathOr('foo', 'x.y.z')({ x : { y : { a : 1 } } })

  expect(result).toEqual('foo')
})

test('curry case (x,y)(z)', () => {
  const result = pathOr('foo')('x.y.z', { x : { y : { a : 1 } } })

  expect(result).toEqual('foo')
})

---------------

paths

paths<Input, T>(pathsToSearch: readonly Path[], obj: Input): readonly (T | undefined)[]

It loops over members of pathsToSearch as singlePath and returns the array produced by R.path(singlePath, obj).

Because it calls R.path, then singlePath can be either string or a list.

const obj = {
  a : {
    b : {
      c : 1,
      d : 2
    }
  }
}

const result = R.paths([
  'a.b.c',
  'a.b.c.d',
  'a.b.c.d.e',
], obj)
// => [1, 2, undefined]

Try this R.paths example in Rambda REPL

R.paths source
import { path } from './path'

export function paths(pathsToSearch, obj){
  if (arguments.length === 1){
    return _obj => paths(pathsToSearch, _obj)
  }

  return pathsToSearch.map(singlePath => path(singlePath, obj))
}
Tests
import { paths } from './paths'

const obj = {
  a : {
    b : {
      c : 1,
      d : 2,
    },
  },
  p : [ { q : 3 } ],
  x : {
    y : 'FOO',
    z : [ [ {} ] ],
  },
}

test('with string path + curry', () => {
  const pathsInput = [ 'a.b.d', 'p.q' ]
  const expected = [ 2, undefined ]
  const result = paths(pathsInput, obj)
  const curriedResult = paths(pathsInput)(obj)

  expect(result).toEqual(expected)
  expect(curriedResult).toEqual(expected)
})

test('with array path', () => {
  const result = paths([
    [ 'a', 'b', 'c' ],
    [ 'x', 'y' ],
  ],
  obj)

  expect(result).toEqual([ 1, 'FOO' ])
})

test('takes a paths that contains indices into arrays', () => {
  expect(paths([
    [ 'p', 0, 'q' ],
    [ 'x', 'z', 0, 0 ],
  ],
  obj)).toEqual([ 3, {} ])
  expect(paths([
    [ 'p', 0, 'q' ],
    [ 'x', 'z', 2, 1 ],
  ],
  obj)).toEqual([ 3, undefined ])
})

test('gets a deep property\'s value from objects', () => {
  expect(paths([ [ 'a', 'b' ] ], obj)).toEqual([ obj.a.b ])
  expect(paths([ [ 'p', 0 ] ], obj)).toEqual([ obj.p[ 0 ] ])
})

test('returns undefined for items not found', () => {
  expect(paths([ [ 'a', 'x', 'y' ] ], obj)).toEqual([ undefined ])
  expect(paths([ [ 'p', 2 ] ], obj)).toEqual([ undefined ])
})

---------------

pick

pick<T, K extends string | number | symbol>(propsToPick: readonly K[], input: T): Pick<T, Exclude<keyof T, Exclude<keyof T, K>>>

It returns a partial copy of an input containing only propsToPick properties.

input can be either an object or an array.

String anotation of propsToPick is one of the differences between Rambda and Ramda.

💥 When using this method with TypeScript, it is much easier to pass propsToPick as an array. If passing a string, you will need to explicitly declare the output type.

const obj = {
  a : 1,
  b : false,
  foo: 'cherry'
}
const list = [1, 2, 3, 4]
const propsToPick = 'a,foo'
const propsToPickList = ['a', 'foo']

const result = [
  R.pick(propsToPick, obj),
  R.pick(propsToPickList, obj),
  R.pick('a,bar', obj),
  R.pick('bar', obj),
  R.pick([0, 3], list),
  R.pick('0,3', list),
]

const expected = [
  {a:1, foo: 'cherry'},
  {a:1, foo: 'cherry'},
  {a:1},
  {},
  [1,4],
  [1,4]
]
// => `result` is equal to `expected`

Try this R.pick example in Rambda REPL

R.pick source
export function pick(propsToPick, input){
  if (arguments.length === 1) return _input => pick(propsToPick, _input)

  if (input === null || input === undefined){
    return undefined
  }
  const keys =
    typeof propsToPick === 'string' ? propsToPick.split(',') : propsToPick

  const willReturn = {}
  let counter = 0

  while (counter < keys.length){
    if (keys[ counter ] in input){
      willReturn[ keys[ counter ] ] = input[ keys[ counter ] ]
    }
    counter++
  }

  return willReturn
}
Tests
import { pick } from './pick'

const obj = {
  a : 1,
  b : 2,
  c : 3,
}

test('props to pick is a string', () => {
  const result = pick('a,c', obj)
  const resultCurry = pick('a,c')(obj)
  const expectedResult = {
    a : 1,
    c : 3,
  }

  expect(result).toEqual(expectedResult)
  expect(resultCurry).toEqual(expectedResult)
})

test('when prop is missing', () => {
  const result = pick('a,d,f', obj)
  expect(result).toEqual({ a : 1 })
})

test('props to pick is an array', () => {
  expect(pick([ 'a', 'c' ])({
    a : 'foo',
    b : 'bar',
    c : 'baz',
  })).toEqual({
    a : 'foo',
    c : 'baz',
  })

  expect(pick([ 'a', 'd', 'e', 'f' ])({
    a : 'foo',
    b : 'bar',
    c : 'baz',
  })).toEqual({ a : 'foo' })

  expect(pick('a,d,e,f')(null)).toEqual(undefined)
})

test('works with list as input and number as props - props to pick is an array', () => {
  const result = pick([ 1, 2 ], [ 'a', 'b', 'c', 'd' ])
  expect(result).toEqual({
    1 : 'b',
    2 : 'c',
  })
})

test('works with list as input and number as props - props to pick is a string', () => {
  const result = pick('1,2', [ 'a', 'b', 'c', 'd' ])
  expect(result).toEqual({
    1 : 'b',
    2 : 'c',
  })
})

test('with symbol', () => {
  const symbolProp = Symbol('s')
  expect(pick([ symbolProp ], { [ symbolProp ] : 'a' })).toMatchInlineSnapshot(`
    Object {
      Symbol(s): "a",
    }
  `)
})

---------------

pickAll

pickAll<T, U>(propsToPick: readonly string[], input: T): U

Same as R.pick but it won't skip the missing props, i.e. it will assign them to undefined.

💥 When using this method with TypeScript, it is much easier to pass propsToPick as an array. If passing a string, you will need to explicitly declare the output type.

const obj = {
  a : 1,
  b : false,
  foo: 'cherry'
}
const propsToPick = 'a,foo,bar'
const propsToPickList = ['a', 'foo', 'bar']

const result = [
  R.pickAll(propsToPick, obj),
  R.pickAll(propsToPickList, obj),
  R.pickAll('a,bar', obj),
  R.pickAll('bar', obj),
]
const expected = [
  {a:1, foo: 'cherry', bar: undefined},
  {a:1, foo: 'cherry', bar: undefined},
  {a:1, bar: undefined},
  {bar: undefined}
]
// => `result` is equal to `expected`

Try this R.pickAll example in Rambda REPL

R.pickAll source
export function pickAll(propsToPick, obj){
  if (arguments.length === 1) return _obj => pickAll(propsToPick, _obj)

  if (obj === null || obj === undefined){
    return undefined
  }
  const keysValue =
    typeof propsToPick === 'string' ? propsToPick.split(',') : propsToPick

  const willReturn = {}
  let counter = 0

  while (counter < keysValue.length){
    if (keysValue[ counter ] in obj){
      willReturn[ keysValue[ counter ] ] = obj[ keysValue[ counter ] ]
    } else {
      willReturn[ keysValue[ counter ] ] = undefined
    }
    counter++
  }

  return willReturn
}
Tests
import { pickAll } from './pickAll'

test('when input is undefined or null', () => {
  expect(pickAll('a', null)).toBe(undefined)
  expect(pickAll('a', undefined)).toBe(undefined)
})

test('with string as condition', () => {
  const obj = {
    a : 1,
    b : 2,
    c : 3,
  }
  const result = pickAll('a,c', obj)
  const resultCurry = pickAll('a,c')(obj)
  const expectedResult = {
    a : 1,
    b : undefined,
    c : 3,
  }

  expect(result).toEqual(expectedResult)
  expect(resultCurry).toEqual(expectedResult)
})

test('with array as condition', () => {
  expect(pickAll([ 'a', 'b', 'c' ], {
    a : 'foo',
    c : 'baz',
  })).toEqual({
    a : 'foo',
    b : undefined,
    c : 'baz',
  })
})

---------------

pipe

It performs left-to-right function composition.

const result = R.pipe(
  R.filter(val => val > 2),
  R.map(a => a * 2)
)([1, 2, 3, 4])

// => [6, 8]

Try this R.pipe example in Rambda REPL

---------------

pipeAsync

pipeAsync<Out>(
  ...fns: readonly (Async<any> | Func<any>)[]
): (input: any) => Promise<Out>

Asynchronous version of R.pipe

💥 It doesn't work with promises or function returning promises such as const foo = input => new Promise(...).

const add = async x => {
  await R.delay(100)
  return x + 1
}
const multiply = async x => {
  await R.delay(100)
  return x * 2 
}

const result = await R.pipeAsync(
  add,
  multiply
)(1)
// `result` resolves to `4`

Try this R.pipeAsync example in Rambda REPL

R.pipeAsync source
import { type } from './type'

export function pipeAsync(...inputArguments){
  return async function (startArgument){
    let argumentsToPass = startArgument

    while (inputArguments.length !== 0){
      const fn = inputArguments.shift()
      const typeFn = type(fn)

      if (typeFn === 'Async'){
        argumentsToPass = await fn(argumentsToPass)
      } else {
        argumentsToPass = fn(argumentsToPass)
        if (type(argumentsToPass) === 'Promise'){
          argumentsToPass = await argumentsToPass
        }
      }
    }

    return argumentsToPass
  }
}
Tests
import { delay } from './delay'
import { equals } from './equals'
import { map } from './map'
import { pipeAsync } from './pipeAsync'

async function identity(x){
  await delay(100)

  return x
}

test('happy', async () => {
  const fn1 = async x => {
    await delay(100)

    return x.map(xx => xx + 1)
  }
  const fn2 = async x => {
    await delay(100)

    return x.map(xx => xx * 2)
  }
  const result = await pipeAsync(fn1,
    fn2)(await Promise.all([ identity(1), identity(2), identity(3) ]))

  expect(result).toEqual([ 4, 6, 8 ])
})

const delayFn = ms =>
  new Promise(resolve => {
    resolve(ms + 1)
  })

test('with function returning promise', async () => {
  const result = await pipeAsync(
    x => x,
    x => x + 1,
    delayFn,
    x => x
  )(1)

  expect(result).toEqual(3)
})

test('throw error', async () => {
  const fn = async () => {
    await delay(1)
    JSON.parse('{foo')
  }

  let didThrow = false
  try {
    await pipeAsync(x => x, fn)(20)
  } catch (e){
    didThrow = true
  }

  expect(didThrow).toBeTrue()
})

---------------

piped

piped<A, B>(input: A, fn0: (x: A) => B) : B

It is basically R.pipe, but instead of passing input argument as R.pipe(...)(input), you pass it as the first argument.

💥 Independently, similar method is implemented in Ramada library, but there the name of the method is pipe - Remeda.pipe

const result = piped(
  [1, 2, 3],
  R.filter(x => x > 1),
  R.map(x => x*10),
)
// => [20, 30]

Try this R.piped example in Rambda REPL

R.piped source
import { pipe } from './pipe'

export function piped(...inputs){
  const [ input, ...fnList ] = inputs

  return pipe(...fnList)(input)
}
Tests
import { add } from './add'
import { filter } from './filter'
import { map } from './map'
import { piped } from './piped'

test('happy', () => {
  const result = piped(
    [ 1, 2, 3 ],
    filter(x => x > 1),
    map(x => x * 10),
    map(add(1))
  )
  const expectedResult = [ 21, 31 ]

  expect(result).toEqual(expectedResult)
})

---------------

pipedAsync

pipedAsync<T>(
  input: any,
  ...fns: readonly (Func<any> | Async<any>)[]
): Promise<T>

It accepts input as first argument and series of functions as next arguments. It is same as R.pipe but with support for asynchronous functions.

💥 Functions that return Promise will be handled as regular function not asynchronous. Such example is const foo = input => new Promise(...).

const result = R.pipedAsync(
  100,
  async x => {
    await R.delay(100)
    return x + 2
  },
  R.add(2),
  async x => {
    const delayed = await R.delay(100)
    return delayed + x
  }
)
const expected = 'RAMBDAX_DELAY104'
// `result` resolves to `expected`

Try this R.pipedAsync example in Rambda REPL

R.pipedAsync source
import { type } from './type'

export async function pipedAsync(...inputs){
  const [ input, ...fnList ] = inputs

  let argumentsToPass = input

  while (fnList.length !== 0){
    const fn = fnList.shift()
    const typeFn = type(fn)

    if (typeFn === 'Async'){
      argumentsToPass = await fn(argumentsToPass)
    } else {
      argumentsToPass = fn(argumentsToPass)
    }
  }

  return argumentsToPass
}
Tests
import { add } from './add'
import { delay } from './delay'
import { pipedAsync } from './pipedAsync'

const fn1 = async x => {
  await delay(100)

  return x + 2
}
const fn2 = async x => {
  await delay(100)

  return x + 3
}

test('happy', async () => {
  const result = await pipedAsync(
    1, fn1, add(2), fn2
  )

  expect(result).toBe(8)
})

---------------

pluck

pluck<K extends keyof T, T>(property: K, list: readonly T[]): ReadonlyArray<T[K]>

It returns list of the values of property taken from the all objects inside list.

const list = [{a: 1}, {a: 2}, {b: 3}]
const property = 'a'

R.pluck(property, list) 
// => [1, 2]

Try this R.pluck example in Rambda REPL

R.pluck source
import { map } from './map'

export function pluck(property, list){
  if (arguments.length === 1) return _list => pluck(property, _list)

  const willReturn = []

  map(x => {
    if (x[ property ] !== undefined){
      willReturn.push(x[ property ])
    }
  }, list)

  return willReturn
}
Tests
import { pluck } from './pluck'

test('happy', () => {
  expect(pluck('a')([ { a : 1 }, { a : 2 }, { b : 1 } ])).toEqual([ 1, 2 ])
})

test('with number', () => {
  const input = [
    [ 1, 2 ],
    [ 3, 4 ],
  ]

  expect(pluck(0, input)).toEqual([ 1, 3 ])
})

---------------

prepend

prepend<T>(x: T, input: readonly T[]): readonly T[]

It adds element x at the beginning of list.

const result = R.prepend('foo', ['bar', 'baz'])
// => ['foo', 'bar', 'baz']

Try this R.prepend example in Rambda REPL

R.prepend source
export function prepend(x, input){
  if (arguments.length === 1) return _input => prepend(x, _input)

  if (typeof input === 'string') return [ x ].concat(input.split(''))

  return [ x ].concat(input)
}
Tests
import { prepend } from './prepend'

test('happy', () => {
  expect(prepend('yes', [ 'foo', 'bar', 'baz' ])).toEqual([
    'yes',
    'foo',
    'bar',
    'baz',
  ])
})

test('with empty list', () => {
  expect(prepend('foo')([])).toEqual([ 'foo' ])
})

test('with string instead of array', () => {
  expect(prepend('foo')('bar')).toEqual([ 'foo', 'b', 'a', 'r' ])
})

---------------

prevIndex

prevIndex(index: number, list: readonly any[]): number

It returns the next index of the list when the order is descending.

If we have reached the beginning of the list, then it will return the last index of the list.

💥 Unlike R.nextIndex, which safeguards against index out of bounds, this method does not.

const list = [1, 2, 3]

const result = [
  R.prevIndex(0, list),
  R.prevIndex(1, list),
  R.prevIndex(2, list),
]
// => [2, 0, 1]

Try this R.prevIndex example in Rambda REPL

R.prevIndex source
export function prevIndex(index, list){
  return index === 0 ? list.length - 1 : index - 1
}
Tests
import { prevIndex } from './prevIndex'

const list = [ 1, 2, 3, 4 ]

test('happy path 1', () => {
  expect(prevIndex(2, list)).toEqual(1)
})

test('happy path 2', () => {
  expect(prevIndex(0, list)).toEqual(3)
})

---------------

produce

produce<Input extends any, Output>(
  rules: ProduceRules<Output, keyof Output, Input>,
  input: Input
): Output

It returns an object created by applying each value of rules to input argument.

💥 In Typescript context, rules functions can be only 1 level deep. In Javascript context, there is no such restriction.

const rules = {
  foo: R.pipe(R.add(1), R.add(2)),
  a: {b: add(3)}
}
const input = i
const result = R.produce(rules, input)

const expected = {
  foo: 4,
  a: {b: 4}
}
// => `result` is equal to `expected`

Try this R.produce example in Rambda REPL

R.produce source
import { map } from './map'
import { type } from './type'

export function produce(rules, input){
  if (arguments.length === 1){
    return _input => produce(rules, _input)
  }

  return map(singleRule =>
    type(singleRule) === 'Object' ?
      produce(singleRule, input) :
      singleRule(input),
  rules)
}
Tests
import { add, pipe } from '../rambda'
import { produce } from './produce'

const rules = {
  a : pipe(add(2), add(3)),
  b : x => ({ foo : x }),
  c : {
    d : add(2),
    e : add(10),
  },
}

const expected = {
  a : 6,
  b : { foo : 1 },
  c : {
    d : 3,
    e : 11,
  },
}

test('happy', () => {
  const result = produce(rules, 1)
  expect(result).toEqual(expected)
})

test('curried', () => {
  const result = produce(rules)(1)
  expect(result).toEqual(expected)
})

---------------

produceAsync

produceAsync<Input extends any, Output>(
  rules: ProduceAsyncRules<Output, keyof Output, Input>,
  input: Input
): Promise<Output>

It returns an object created by applying each value of rules to input argument.

rules input is an object with synchronous or asynchronous functions as values.

The return value is wrapped in a promise, even if all rules are synchronous functions.

const rules = {
  foo: async x => {
    await R.delay(100)
    return x > 1
  },
  bar: x => ({baz: x})
}
const input = 2
const result = await R.produceAsync(rules, input)

const expected = {
  foo: true,
  bar: {baz: 2}
}
// => `result` is equal to `expected`

Try this R.produceAsync example in Rambda REPL

R.produceAsync source
import { map } from './map'
import { type } from './type'

function promisify({ condition, input, prop }){
  return new Promise((resolve, reject) => {
    if (type(condition) !== 'Async'){
      return resolve({
        type    : prop,
        payload : condition(input),
      })
    }

    condition(input)
      .then(result => {
        resolve({
          type    : prop,
          payload : result,
        })
      })
      .catch(err => reject(err))
  })
}

function produceFn(conditions, input){
  let asyncConditionsFlag = false
  for (const prop in conditions){
    if (
      asyncConditionsFlag === false &&
      type(conditions[ prop ]) === 'Async'
    ){
      asyncConditionsFlag = true
    }
  }

  if (asyncConditionsFlag === false){
    const willReturn = {}
    for (const prop in conditions){
      willReturn[ prop ] = conditions[ prop ](input)
    }

    return Promise.resolve(willReturn)
  }

  const promised = []
  for (const prop in conditions){
    const condition = conditions[ prop ]
    promised.push(promisify({
      input,
      condition,
      prop,
    }))
  }

  return new Promise((resolve, reject) => {
    Promise.all(promised)
      .then(results => {
        const willReturn = {}

        map(result => willReturn[ result.type ] = result.payload, results)

        resolve(willReturn)
      })
      .catch(err => reject(err))
  })
}

export function produceAsync(conditions, input){
  if (arguments.length === 1){
    return async _input => produceFn(conditions, _input)
  }

  return new Promise((resolve, reject) => {
    produceFn(conditions, input).then(resolve)
      .catch(reject)
  })
}
Tests
import { delay } from './delay'
import { produceAsync } from './produceAsync'

test('happy', async () => {
  const result = await produceAsync({
    foo : async x => {
      await delay(100)

      return `${ x }_ZEPPELIN`
    },
    bar : x => x.length === 3,
  },
  'LED')
  const expected = {
    foo : 'LED_ZEPPELIN',
    bar : true,
  }

  expect(result).toEqual(expected)
})

test('when all rules are synchronous', async () => {
  const result = await produceAsync({
    foo : x => `${ x }_ZEPPELIN`,
    bar : x => x.length === 3,
  },
  'LED')
  const expected = {
    foo : 'LED_ZEPPELIN',
    bar : true,
  }

  expect(result).toEqual(expected)
})

test('with error', async () => {
  const fn = produceAsync({
    foo : async x => {
      await delay(100)
      throw new Error(`${ x }_ZEPPELIN`)
    },
    bar : inputArgument => inputArgument === 5,
  })

  await expect(fn('LED')).rejects.toThrow('LED_ZEPPELIN')
})

---------------

product

product(list: readonly number[]): number
R.product([ 2, 3, 4 ])
// => 24)

Try this R.product example in Rambda REPL

R.product source
import { multiply } from './multiply'
import { reduce } from './reduce'

export const product = reduce(multiply, 1)
Tests
import { product } from './product'

test('happy', () => {
  expect(product([ 2, 3, 4 ])).toEqual(24)
})

test('bad input', () => {
  expect(product([ null ])).toEqual(0)
  expect(product([])).toEqual(1)
})

---------------

prop

prop<P extends keyof T, T>(propToFind: P, obj: T): T[P]

It returns the value of property propToFind in obj.

If there is no such property, it returns undefined.

const result = [
  R.prop('x', {x: 100}), 
  R.prop('x', {a: 1}) 
]
// => [100, undefined]

Try this R.prop example in Rambda REPL

R.prop source
export function prop(propToFind, obj){
  if (arguments.length === 1) return _obj => prop(propToFind, _obj)

  if (!obj) return undefined

  return obj[ propToFind ]
}
Tests
import { prop } from './prop'

test('prop', () => {
  expect(prop('foo')({ foo : 'baz' })).toEqual('baz')

  expect(prop('bar')({ foo : 'baz' })).toEqual(undefined)

  expect(prop('bar')(null)).toEqual(undefined)
})

---------------

propEq

propEq<K extends string | number>(propToFind: K, valueToMatch: any, obj: Record<K, any>): boolean

It returns true if obj has property propToFind and its value is equal to valueToMatch.

const obj = { foo: 'bar' }
const secondObj = { foo: 1 }

const propToFind = 'foo'
const valueToMatch = 'bar'

const result = [
  R.propEq(propToFind, valueToMatch, obj),
  R.propEq(propToFind, valueToMatch, secondObj)
]
// => [true, false]

Try this R.propEq example in Rambda REPL

R.propEq source
import { curry } from './curry'

function propEqFn(
  propToFind, valueToMatch, obj
){
  if (!obj) return false

  return obj[ propToFind ] === valueToMatch
}

export const propEq = curry(propEqFn)
Tests
import { propEq } from './propEq'

test('happy', () => {
  expect(propEq('foo', 'bar')({ foo : 'bar' })).toBeTrue()
  expect(propEq('foo', 'bar')({ foo : 'baz' })).toBeFalse()
  expect(propEq('foo')('bar')({ foo : 'baz' })).toBeFalse()
  expect(propEq(
    'foo', 'bar', null
  )).toBeFalse()
})

---------------

propIs

propIs(type: any, name: string, obj: any): boolean

It returns true if property of obj is from target type.

const obj = {a:1, b: 'foo'}
const property = 'foo'

const result = [
  R.propIs(String, property, obj),
  R.propIs(Number, property, obj)
]
// => [true, false]

Try this R.propIs example in Rambda REPL

R.propIs source
import { curry } from './curry'
import { is } from './is'

function propIsFn(
  targetPrototype, property, obj
){
  return is(targetPrototype, obj[ property ])
}

export const propIs = curry(propIsFn)
Tests
import { propIs } from './propIs'

const obj = { value : 1 }
const property = 'value'

test('when true', () => {
  expect(propIs(
    Number, property, obj
  )).toBeTrue()
})

test('when false', () => {
  expect(propIs(
    String, property, obj
  )).toBeFalse()
  expect(propIs(
    String, property, {}
  )).toBeFalse()
})

---------------

propOr

propOr<T, P extends string>(defaultValue: T, property: P, obj: Partial<Record<P, T>> | undefined): T

It returns either defaultValue or the value of property in obj.

const obj = {a: 1}
const defaultValue = 'DEFAULT_VALUE'
const property = 'a'

const result = [
  R.propOr(defaultValue, property, obj),
  R.propOr(defaultValue, 'foo', obj)
]
// => [1, 'DEFAULT_VALUE']

Try this R.propOr example in Rambda REPL

R.propOr source
import { curry } from './curry'
import { defaultTo } from './defaultTo'

function propOrFn(
  defaultValue, property, obj
){
  if (!obj) return defaultValue

  return defaultTo(defaultValue, obj[ property ])
}

export const propOr = curry(propOrFn)
Tests
import { propOr } from './propOr'

test('propOr (result)', () => {
  const obj = { a : 1 }
  expect(propOr(
    'default', 'a', obj
  )).toEqual(1)
  expect(propOr(
    'default', 'notExist', obj
  )).toEqual('default')
  expect(propOr(
    'default', 'notExist', null
  )).toEqual('default')
})

test('propOr (currying)', () => {
  const obj = { a : 1 }
  expect(propOr('default')('a', obj)).toEqual(1)
  expect(propOr('default', 'a')(obj)).toEqual(1)
  expect(propOr('default')('notExist', obj)).toEqual('default')
  expect(propOr('default', 'notExist')(obj)).toEqual('default')
})

---------------

props

props<P extends string, T>(propsToPick: readonly P[], obj: Record<P, T>): readonly T[]

It takes list with properties propsToPick and returns a list with property values in obj.

const result = [
  R.props(['a', 'b'], {a:1, c:3})
// => [1, undefined]

Try this R.props example in Rambda REPL

R.props source
import { _isArray } from './_internals/_isArray'
import { mapArray } from './map'

export function props(propsToPick, obj){
  if (arguments.length === 1){
    return _obj => props(propsToPick, _obj)
  }
  if (!_isArray(propsToPick)){
    throw new Error('propsToPick is not a list')
  }

  return mapArray(prop => obj[ prop ], propsToPick)
}
Tests
import { props } from './props'

const obj = {
  a : 1,
  b : 2,
}
const propsToPick = [ 'a', 'c' ]

test('happy', () => {
  const result = props(propsToPick, obj)
  expect(result).toEqual([ 1, undefined ])
})

test('curried', () => {
  const result = props(propsToPick)(obj)
  expect(result).toEqual([ 1, undefined ])
})

test('wrong input', () => {
  expect(() => props(null)(obj)).toThrow()
})

---------------

random

random(minInclusive: number, maxInclusive: number): number

It returns a random number between min inclusive and max inclusive.

R.random source
export function random(min, max){
  return Math.floor(Math.random() * (max - min + 1)) + min
}
Tests
import { random } from './random'
import { range } from './range'

test('when returns true', () => {
  range(0, 100).map(() => {
    const randomResult = random(1, 10)
    expect(randomResult).toBeLessThanOrEqual(10)
    expect(randomResult).toBeGreaterThanOrEqual(1)
  })
})

---------------

range

range(startInclusive: number, endExclusive: number): readonly number[]

It returns list of numbers between startInclusive to endExclusive markers.

R.range(0, 5)
// => [0, 1, 2, 3, 4]

Try this R.range example in Rambda REPL

R.range source
export function range(start, end){
  if (arguments.length === 1) return _end => range(start, _end)

  if (Number.isNaN(Number(start)) || Number.isNaN(Number(end))){
    throw new TypeError('Both arguments to range must be numbers')
  }

  if (end < start) return []

  const len = end - start
  const willReturn = Array(len)

  for (let i = 0; i < len; i++){
    willReturn[ i ] = start + i
  }

  return willReturn
}
Tests
import { range } from './range'

test('happy', () => {
  expect(range(0, 10)).toEqual([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 ])
})

test('end range is bigger than start range', () => {
  expect(range(7, 3)).toEqual([])
  expect(range(5, 5)).toEqual([])
})

test('with bad input', () => {
  const throwMessage = 'Both arguments to range must be numbers'
  expect(() => range('a', 6)).toThrowWithMessage(Error, throwMessage)
  expect(() => range(6, 'z')).toThrowWithMessage(Error, throwMessage)
})

test('curry', () => {
  expect(range(0)(10)).toEqual([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 ])
})

---------------

reduce

reduce<T, TResult>(reducer: (prev: TResult, current: T, i: number) => TResult, initialValue: TResult, list: readonly T[]): TResult

💥 It passes index of the list as third argument to reducer function.

const list = [1, 2, 3]
const initialValue = 10
const reducer = (prev, current) => prev * current

const result = R.reduce(reducer, initialValue, list)
// => 60

Try this R.reduce example in Rambda REPL

R.reduce source
import { _isArray } from './_internals/_isArray'
import { _keys } from './_internals/_keys'
import { curry } from './curry'

function reduceFn(
  reducer, acc, list
){
  if (!_isArray(list)){
    throw new TypeError('reduce: list must be array or iterable')
  }
  let index = 0
  const len = list.length

  while (index < len){
    acc = reducer(
      acc, list[ index ], index, list
    )
    index++
  }

  return acc
}

export const reduce = curry(reduceFn)
Tests
import { reduce } from './reduce'

const reducer = (
  prev, current, i
) => {
  expect(i).toBeNumber()

  return prev + current
}
const initialValue = 1
const list = [ 1, 2, 3 ]

test('happy', () => {
  expect(reduce(
    reducer, initialValue, list
  )).toEqual(7)
})

test('with object as iterable', () => {
  expect(() =>
    reduce(
      reducer, initialValue, {
        a : 1,
        b : 2,
      }
    )).toThrowWithMessage(TypeError, 'reduce: list must be array or iterable')
})

test('with undefined as iterable', () => {
  expect(() => reduce(
    reducer, initialValue, undefined
  )).toThrowWithMessage(TypeError,
    'reduce: list must be array or iterable')
})

---------------

reject

reject<T>(predicate: Predicate<T>, list: readonly T[]): readonly T[]

It has the opposite effect of R.filter.

const list = [1, 2, 3, 4]
const obj = {a: 1, b: 2}
const predicate = x => x > 1

const result = [
  R.reject(predicate, list)
  R.reject(predicate, obj)
]
// => [[1, 2], {a: 1}]

Try this R.reject example in Rambda REPL

R.reject source
import { filter } from './filter'

export function reject(predicate, list){
  if (arguments.length === 1) return _list => reject(predicate, _list)

  return filter(x => !predicate(x), list)
}
Tests
import { reject } from './reject'

const isOdd = n => n % 2 === 1

test('with array', () => {
  expect(reject(isOdd)([ 1, 2, 3, 4 ])).toEqual([ 2, 4 ])
})

test('with object', () => {
  const obj = {
    a : 1,
    b : 2,
    c : 3,
    d : 4,
  }
  expect(reject(isOdd, obj)).toEqual({
    b : 2,
    d : 4,
  })
})

---------------

remove

remove(
  toRemove: string | RegExp | readonly (string | RegExp)[],
  text: string
): string

It will remove all toRemove entries from text sequentially.

toRemove argument can be either a list of strings/regular expressions or a single string/regular expression.

💥 This is the only case where Rambdax exports clashes with Ramda API, as Ramda has remove method. If Rambda.remove is introduced, then this method will be renamed.

const result = remove(
  ['foo','bar'],
  'foo bar baz foo'
)
// => 'baz foo'

Try this R.remove example in Rambda REPL

R.remove source
import { replace } from './replace'
import { type } from './type'

export function remove(inputs, text){
  if (arguments.length === 1){
    return textHolder => remove(inputs, textHolder)
  }

  if (type(text) !== 'String'){
    throw new Error(`R.remove requires string not ${ type(text) }`)
  }

  if (type(inputs) !== 'Array'){
    return replace(
      inputs, '', text
    )
  }

  let textCopy = text

  inputs.forEach(singleInput => {
    textCopy = replace(
      singleInput, '', textCopy
    ).trim()
  })

  return textCopy
}
Tests
import { remove } from './remove'

test('happy', () => {
  const inputs = [ /foo/, /not\shere/, /also/, 'bar' ]
  const text = 'foo bar baz foo'

  const result = remove(inputs)(text)
  const expectedResult = 'baz foo'

  expect(result).toEqual(expectedResult)
})

test('with strings + curry', () => {
  const inputs = [ 'foo', 'bar' ]
  const text = 'foo bar baz foo'

  const result = remove(inputs)(text)
  const expectedResult = 'baz foo'

  expect(result).toEqual(expectedResult)
})

test('with strings and regex', () => {
  const inputs = [ /foo/g, 'bar' ]
  const text = 'foo bar baz foo'

  const result = remove(inputs, text)
  const expectedResult = 'baz'

  expect(result).toEqual(expectedResult)
})

test('text is not string', () => {
  const inputs = [ /foo/g, 'bar' ]
  const text = null

  expect(() => remove(inputs, text)).toThrowWithMessage(Error,
    'R.remove requires string not Null')
})

test('with regexes', () => {
  const inputs = [ /foo/g, /bar/ ]
  const text = 'foo bar baz foo'

  const result = remove(inputs, text)
  const expectedResult = 'baz'

  expect(result).toEqual(expectedResult)
})

test('with single rule', () => {
  const inputs = /foo/g
  const text = 'foo bar baz foo'

  const result = remove(inputs, text)
  const expectedResult = ' bar baz '

  expect(result).toEqual(expectedResult)
})

---------------

removeIndex

removeIndex<T>(index: number, list: readonly T[]): readonly T[]

It returns a copy of list input with removed index.

const list = [1, 2, 3, 4]
const result = R.removeIndex(1, list)
// => [1, 3, 4]

Try this R.removeIndex example in Rambda REPL

R.removeIndex source
export function removeIndex(list, index){
  if (arguments.length === 1) return _index => removeIndex(list, _index)
  if (index <= 0) return list.slice(1)
  if (index >= list.length - 1) return list.slice(0, list.length - 1)

  return [ ...list.slice(0, index), ...list.slice(index + 1) ]
}
Tests
import { removeIndex } from './removeIndex'

const list = [ 1, 2, 3, 4 ]

test('first or before first index', () => {
  expect(removeIndex(list, 0)).toEqual([ 2, 3, 4 ])
  expect(removeIndex(list)(-2)).toEqual([ 2, 3, 4 ])
})

test('last or after last index', () => {
  expect(removeIndex(list, 4)).toEqual([ 1, 2, 3 ])
  expect(removeIndex(list, 10)).toEqual([ 1, 2, 3 ])
})

test('middle index', () => {
  expect(removeIndex(list, 1)).toEqual([ 1, 3, 4 ])
  expect(removeIndex(list, 2)).toEqual([ 1, 2, 4 ])
})

---------------

renameProps

renameProps(rules: object, input: object): object

If property prop of rules is also a property in input, then rename input property to rules[prop].

R.renameProps source
import { merge } from './merge'
import { omit } from './omit'

export function renameProps(conditions, inputObject){
  if (arguments.length === 1){
    return inputObjectHolder => renameProps(conditions, inputObjectHolder)
  }
  const renamed = {}
  Object.keys(conditions).forEach(condition => {
    if (Object.keys(inputObject).includes(condition)){
      renamed[ conditions[ condition ] ] = inputObject[ condition ]
    }
  })

  return merge(renamed, omit(Object.keys(conditions), inputObject))
}
Tests
import { renameProps } from './renameProps'

test('renameProps', () => {
  const rules = {
    f : 'foo',
    b : 'bar',
    q : 'x',
  }
  const input = {
    f : 1,
    b : 2,
    a : 3,
  }
  const result = renameProps(rules, input)
  const expectedResult = {
    foo : 1,
    bar : 2,
    a   : 3,
  }
  expect(result).toEqual(expectedResult)
})

test('curry', () => {
  const rules = {
    f : 'foo',
    b : 'bar',
  }
  const input = {
    f : 1,
    b : 2,
  }
  const result = renameProps(rules)(input)
  const expectedResult = {
    foo : 1,
    bar : 2,
  }
  expect(result).toEqual(expectedResult)
})

---------------

repeat

repeat<T>(x: T): (timesToRepeat: number) => readonly T[]
R.repeat('foo', 3)
// => ['foo', 'foo', 'foo']

Try this R.repeat example in Rambda REPL

R.repeat source
export function repeat(x, timesToRepeat){
  if (arguments.length === 1){
    return _timesToRepeat => repeat(x, _timesToRepeat)
  }

  return Array(timesToRepeat).fill(x)
}
Tests
import { repeat } from './repeat'

test('repeat', () => {
  expect(repeat('')(3)).toEqual([ '', '', '' ])
  expect(repeat('foo', 3)).toEqual([ 'foo', 'foo', 'foo' ])

  const obj = {}
  const arr = repeat(obj, 3)

  expect(arr).toEqual([ {}, {}, {} ])

  expect(arr[ 0 ] === arr[ 1 ]).toBeTrue()
})

---------------

replace

replace(strOrRegex: RegExp | string, replacer: string, str: string): string

It replaces strOrRegex found in str with replacer.

const strOrRegex = /o/g

const result = R.replace(strOrRegex, '|0|', 'foo')
// => 'f|0||0|'

Try this R.replace example in Rambda REPL

R.replace source
import { curry } from './curry'

function replaceFn(
  pattern, replacer, str
){
  return str.replace(pattern, replacer)
}

export const replace = curry(replaceFn)
Tests
import { replace } from './replace'

test('happy', () => {
  expect(replace(
    'foo', 'yes', 'foo bar baz'
  )).toEqual('yes bar baz')
})

test('1', () => {
  expect(replace(/\s/g)('|')('foo bar baz')).toEqual('foo|bar|baz')
})

test('2', () => {
  expect(replace(/\s/g)('|', 'foo bar baz')).toEqual('foo|bar|baz')
})

test('3', () => {
  expect(replace(/\s/g, '|')('foo bar baz')).toEqual('foo|bar|baz')
})

---------------

replaceAll

replaceAll(patterns: ReadonlyArray<RegExp | string>, replacer: string, input: string): string

Same as R.replace but it accepts array of string and regular expressions instead of a single value.

const replacer = '|'
const patterns = [ /foo/g, 'bar' ]
const input = 'foo bar baz foo bar'

const result = replaceAll(patterns, replacer, input)
// => '| | baz | bar'

Try this R.replaceAll example in Rambda REPL

R.replaceAll source
import { curry } from './curry'
import { ok } from './ok'

function replaceAllFn(
  patterns, replacer, input
){
  ok(
    patterns, replacer, input
  )(
    Array, String, String
  )

  let text = input
  patterns.forEach(singlePattern => {
    text = text.replace(singlePattern, replacer)
  })

  return text
}

export const replaceAll = curry(replaceAllFn)
Tests
import { replaceAll } from './replaceAll'

const replacer = '|'
const patterns = [ /foo/g, 'bar' ]
const input = 'foo bar baz foo bar'

test('happy', () => {
  const result = replaceAll(
    patterns, replacer, input
  )
  const expected = '| | baz | bar'

  expect(result).toEqual(expected)
})

test('throws when wrong patterns', () => {
  expect(() => replaceAll(
    {}, replacer, input
  )).toThrow()
})

test('throws when wrong input', () => {
  expect(() => replaceAll(
    patterns, replacer, []
  )).toThrow()
})

test('throws when wrong replacer', () => {
  expect(() => replaceAll(
    patterns, null, input
  )).toThrow()
})

---------------

reset

reset(): void

💥 R.getter method contains explanations, tests and source information of R.reset, R.setter and R.getter methods.

---------------

reverse

reverse<T>(input: readonly T[]): readonly T[]

It returns a reversed copy of list or string input.

const result = [
  R.reverse('foo'),
  R.reverse([1, 2, 3])
]
// => ['oof', [3, 2, 1]

Try this R.reverse example in Rambda REPL

R.reverse source
export function reverse(listOrString){
  if (typeof listOrString === 'string'){
    return listOrString.split('').reverse()
      .join('')
  }

  const clone = listOrString.slice()

  return clone.reverse()
}
Tests
import { reverse } from './reverse'

test('happy', () => {
  expect(reverse([ 1, 2, 3 ])).toEqual([ 3, 2, 1 ])
})

test('with string', () => {
  expect(reverse('baz')).toEqual('zab')
})

test('it doesn\'t mutate', () => {
  const arr = [ 1, 2, 3 ]

  expect(reverse(arr)).toEqual([ 3, 2, 1 ])

  expect(arr).toEqual([ 1, 2, 3 ])
})

---------------

set

set<T, U>(lens: Lens, replacer: U, obj: T): T

It returns a copied Object or Array with modified lens focus set to replacer value.

const input = {x: 1, y: 2}
const xLens = R.lensProp('x')

R.set(xLens, 4, input) // => {x: 4, y: 2}
R.set(xLens, 8, input) // => {x: 8, y: 2}

Try this R.set example in Rambda REPL

R.set source
import { always } from './always'
import { curry } from './curry'
import { over } from './over'

function setFn(
  lens, replacer, x
){
  return over(
    lens, always(replacer), x
  )
}

export const set = curry(setFn)
Tests
import { assoc } from './assoc'
import { lens } from './lens'
import { lensIndex } from './lensIndex'
import { lensPath } from './lensPath'
import { prop } from './prop'
import { set } from './set'

const testObject = {
  foo : 'bar',
  baz : {
    a : 'x',
    b : 'y',
  },
}

test('assoc lens', () => {
  const assocLens = lens(prop('foo'), assoc('foo'))
  const result = set(
    assocLens, 'FOO', testObject
  )
  const expected = {
    ...testObject,
    foo : 'FOO',
  }
  expect(result).toEqual(expected)
})

test('path lens', () => {
  const pathLens = lensPath('baz.a')
  const result = set(
    pathLens, 'z', testObject
  )
  const expected = {
    ...testObject,
    baz : {
      a : 'z',
      b : 'y',
    },
  }
  expect(result).toEqual(expected)
})

test('index lens', () => {
  const indexLens = lensIndex(0)

  const result = set(
    indexLens, 3, [ 1, 2 ]
  )
  expect(result).toEqual([ 3, 2 ])
})

---------------

setter

setter(keyOrObject: string | object, value?: any): void

💥 R.getter method contains explanations, tests and source information of R.reset, R.setter and R.getter methods.

---------------

shuffle

shuffle<T>(list: readonly T[]): readonly T[]

It returns a randomized copy of array.

R.shuffle source
export function shuffle(arrayRaw){
  const array = arrayRaw.concat()
  let counter = array.length
  while (counter > 0){
    const index = Math.floor(Math.random() * counter)
    counter--
    const temp = array[ counter ]
    array[ counter ] = array[ index ]
    array[ index ] = temp
  }

  return array
}
Tests
import { range } from './range'
import { shuffle } from './shuffle'
import { uniq } from './uniq'

test('happy', () => {
  const list = range(0, 7)
  const result = range(0, 300).map(() => shuffle(list))
  const allUniq = uniq(result)
  expect(allUniq.length > 150).toBeTrue()
})

---------------

slice

slice(from: number, to: number, input: string): string
const list = [0, 1, 2, 3, 4, 5]
const str = 'FOO_BAR'
const from = 1
const to = 4

const result = [
  R.slice(str, to, list),
  R.slice(from, to, list)
]
// => ['OO_', [1, 2, 3]]

Try this R.slice example in Rambda REPL

R.slice source
import { curry } from './curry'

function sliceFn(
  from, to, list
){
  return list.slice(from, to)
}

export const slice = curry(sliceFn)
Tests
import { slice } from './slice'

test('slice', () => {
  expect(slice(
    1, 3, [ 'a', 'b', 'c', 'd' ]
  )).toEqual([ 'b', 'c' ])
  expect(slice(
    1, Infinity, [ 'a', 'b', 'c', 'd' ]
  )).toEqual([ 'b', 'c', 'd' ])
  expect(slice(
    0, -1, [ 'a', 'b', 'c', 'd' ]
  )).toEqual([ 'a', 'b', 'c' ])
  expect(slice(
    -3, -1, [ 'a', 'b', 'c', 'd' ]
  )).toEqual([ 'b', 'c' ])
  expect(slice(
    0, 3, 'ramda'
  )).toEqual('ram')
})

---------------

sort

sort<T>(sortFn: (a: T, b: T) => number, list: readonly T[]): readonly T[]

It returns copy of list sorted by sortFn function.

💥 sortFn function must return a number.

const list = [
  {a: 2},
  {a: 3},
  {a: 1}
]
const sortFn = (x, y) => {
  return x.a > y.a ? 1 : -1
}

const result = R.sort(sortFn, list)
const expected = [
  {a: 1},
  {a: 2},
  {a: 3}
]
// => `result` is equal to `expected`

Try this R.sort example in Rambda REPL

R.sort source
export function sort(sortFn, list){
  if (arguments.length === 1) return _list => sort(sortFn, _list)

  const clone = list.slice()

  return clone.sort(sortFn)
}
Tests
import { sort } from './sort'

const fn = (a, b) => a > b ? 1 : -1

test('sort', () => {
  expect(sort((a, b) => a - b)([ 2, 3, 1 ])).toEqual([ 1, 2, 3 ])
})

test('it doesn\'t mutate', () => {
  const list = [ 'foo', 'bar', 'baz' ]

  expect(sort(fn, list)).toEqual([ 'bar', 'baz', 'foo' ])

  expect(list[ 0 ]).toBe('foo')
  expect(list[ 1 ]).toBe('bar')
  expect(list[ 2 ]).toBe('baz')
})

---------------

sortBy

sortBy<T>(sortFn: (a: T) => Ord, list: readonly T[]): readonly T[]

It returns copy of list sorted by sortFn function.

💥 sortFn function must return a value to compare.

const list = [
  {a: 2},
  {a: 3},
  {a: 1}
]
const sortFn = x => x.a

const result = R.sortBy(sortFn, list)
const expected = [
  {a: 1},
  {a: 2},
  {a: 3}
]
// => `result` is equal to `expected`

Try this R.sortBy example in Rambda REPL

R.sortBy source
export function sortBy(sortFn, list){
  if (arguments.length === 1) return _list => sortBy(sortFn, _list)

  const clone = list.slice()

  return clone.sort((a, b) => {
    const aSortResult = sortFn(a)
    const bSortResult = sortFn(b)

    if (aSortResult === bSortResult) return 0

    return aSortResult < bSortResult ? -1 : 1
  })
}
Tests
import { compose } from './compose'
import { prop } from './prop'
import { sortBy } from './sortBy'
import { toLower } from './toLower'

test('happy', () => {
  const input = [ { a : 2 }, { a : 1 }, { a : 1 }, { a : 3 } ]
  const expected = [ { a : 1 }, { a : 1 }, { a : 2 }, { a : 3 } ]

  const result = sortBy(x => x.a)(input)
  expect(result).toEqual(expected)
})

test('with compose', () => {
  const alice = {
    name : 'ALICE',
    age  : 101,
  }
  const bob = {
    name : 'Bob',
    age  : -10,
  }
  const clara = {
    name : 'clara',
    age  : 314.159,
  }
  const people = [ clara, bob, alice ]
  const sortByNameCaseInsensitive = sortBy(compose(toLower, prop('name')))

  expect(sortByNameCaseInsensitive(people)).toEqual([ alice, bob, clara ])
})

---------------

sortByPath

sortByPath<T>(sortPath: Path, list: readonly T[]): readonly T[]

It returns copy of list sorted by sortPath value.

As sortPath is passed to R.path, it can be either a string or an array of strings.

const list = [
  {a: {b: 2}},
  {a: {b: 1}},
  {a: {b: 3}}
]
const result = R.sortByPath('a.b', list)
const expected = [
  {a: {b: 1}},
  {a: {b: 2}},
  {a: {b: 3}}
]
// => `result` is equal to `expected`

Try this R.sortByPath example in Rambda REPL

R.sortByPath source
import { path } from './path'
import { sortBy } from './sortBy'

export function sortByPath(sortPath, list){
  if (arguments.length === 1) return _list => sortByPath(sortPath, _list)

  return sortBy(path(sortPath), list)
}
Tests
import { sortByPath } from './sortByPath'

const list = [ { a : { b : 3 } }, { a : { b : 1 } }, { a : { b : 2 } } ]
const sorted = [ { a : { b : 1 } }, { a : { b : 2 } }, { a : { b : 3 } } ]

test('with string as path', () => {
  expect(sortByPath('a.b', list)).toEqual(sorted)
})

test('with list of strings as path', () => {
  expect(sortByPath([ 'a', 'b' ], list)).toEqual(sorted)
})

test('with string as path - curried', () => {
  expect(sortByPath('a.b')(list)).toEqual(sorted)
})

test('with list of strings as path - curried', () => {
  expect(sortByPath([ 'a', 'b' ])(list)).toEqual(sorted)
})

---------------

sortByProps

sortByProps<T>(sortPaths: readonly string[], list: readonly T[]): readonly T[]

It returns sorted copy of list of objects.

Sorting is done using a list of strings, each representing a path. Two members a and b from list can be sorted if both return a value for a given path. If the value is equal, then the next member of sortPaths(if there is such) will be used in order to find difference between a and b.

const list = [
  {a: {b: 2}},
  {a: {b: 1}},
  {a: {b: 3}}
]
const result = R.sortByPath('a.b', list)
const expected = [
  {a: {b: 1}},
  {a: {b: 2}},
  {a: {b: 3}}
]
// => `result` is equal to `expected`

Try this R.sortByProps example in Rambda REPL

R.sortByProps source
import { path } from './path'

function singleSort(
  a, b, sortPaths
){
  let toReturn = 0
  sortPaths.forEach(singlePath => {
    if (toReturn !== 0) return
    const aResult = path(singlePath, a)
    const bResult = path(singlePath, b)
    if ([ aResult, bResult ].includes(undefined)) return
    if (aResult === bResult) return

    toReturn = aResult > bResult ? 1 : -1
  })

  return toReturn
}

export function sortByProps(sortPaths, list){
  if (arguments.length === 1) return _list => sortByProps(sortPaths, _list)
  const clone = list.slice()

  clone.sort((a, b) => singleSort(
    a, b, sortPaths
  ))

  return clone
}
Tests
import { sortByProps } from './sortByProps'

const list = [ { a : { b : 3 } }, { a : { b : 2 } }, { a : { b : 1 } } ]
const sorted = [ { a : { b : 1 } }, { a : { b : 2 } }, { a : { b : 3 } } ]

test('wrong paths are ignored', () => {
  expect(sortByProps([ 'foo.bar', 'a.c', 'a.b', 'a.d' ], list)).toEqual(sorted)
})

test('skip sort when path results are equal', () => {
  const input = [
    {
      a : {
        b : 0,
        c : 2,
      },
    },
    {
      a : {
        b : 0,
        c : 1,
      },
    },
  ]
  expect(sortByProps([ 'a.b', 'a.d' ], input)).toEqual(input)
})

test('when list is already sorted', () => {
  const input = [
    {
      a : {
        b : 0,
        c : 1,
      },
    },
    {
      a : {
        b : 0,
        c : 2,
      },
    },
  ]
  expect(sortByProps([ 'a.b', 'a.c' ])(input)).toEqual(input)
})

---------------

sortObject

sortObject<T>(predicate: SortObjectPredicate<T>, input: { readonly [key: string]: T }): { readonly [keyOutput: string]: T }

It returns a sorted version of input object.

const predicate = (propA, propB, valueA, valueB) => valueA > valueB ? -1 : 1

const result = R.sortObject(predicate, {a:1, b: 4, c: 2})
// => {b: 4, c: 2, a: 1}

Try this R.sortObject example in Rambda REPL

R.sortObject source
import { sort } from './sort'

export function sortObject(predicate, obj){
  if (arguments.length === 1){
    return _obj => sortObject(predicate, _obj)
  }
  const keys = Object.keys(obj)
  const sortedKeys = sort((a, b) => predicate(
    a, b, obj[ a ], obj[ b ]
  ), keys)

  const toReturn = {}
  sortedKeys.forEach(singleKey => {
    toReturn[ singleKey ] = obj[ singleKey ]
  })

  return toReturn
}
Tests
import { runTests } from 'helpers-fn'

import { allTrue } from './allTrue'
import { equals } from './equals'
import { sortObject } from './sortObject'

const obj = {
  c : 1,
  a : 2,
  b : 3,
}

const predicateA = (
  propA, propB, valueA, valueB
) => propA > propB ? -1 : 1

const expectationA = [ 'c', 'b', 'a' ]

const predicateB = (
  propA, propB, valueA, valueB
) => propA < propB ? -1 : 1
const expectationB = [ 'a', 'b', 'c' ]
const predicateC = (
  propA, propB, valueA, valueB
) =>
  valueA > valueB ? -1 : 1
const expectationC = [ 'b', 'a', 'c' ]

const fn = ([ predicate, expectation ]) => {
  const result = sortObject(predicate, obj)
  const curriedResult = sortObject(predicate)(obj)
  const sortedKeys = Object.keys(result)
  const sortedKeysCurried = Object.keys(curriedResult)
  const isSameObject = equals(obj, result)
  const isSameObjectCurried = equals(obj, curriedResult)

  return allTrue(
    isSameObject,
    isSameObjectCurried,
    equals(sortedKeys, expectation),
    equals(sortedKeysCurried, expectation)
  )
}

const testData = {
  label : 'foo',
  data  : [
    { ok : [ predicateA, expectationA ] },
    { ok : [ predicateB, expectationB ] },
    { ok : [ predicateC, expectationC ] },
  ],
  fn,
}

runTests(testData)

---------------

split

split(separator: string | RegExp): (str: string) => readonly string[]

Curried version of String.prototype.split

const str = 'foo|bar|baz'
const separator = |'
const result = R.split(separator, str))
// => [ 'foo', 'bar', 'baz' ]

Try this R.split example in Rambda REPL

R.split source
export function split(separator, str){
  if (arguments.length === 1) return _str => split(separator, _str)

  return str.split(separator)
}
Tests
import { split } from './split'

const str = 'foo|bar|baz'
const splitChar = '|'
const expected = [ 'foo', 'bar', 'baz' ]

test('happy', () => {
  expect(split(splitChar, str)).toEqual(expected)
})

test('curried', () => {
  expect(split(splitChar)(str)).toEqual(expected)
})

---------------

splitAt

splitAt<T>(index: number, input: readonly T[]): readonly [readonly T[], readonly T[]]

It splits string or array at a given index.

const list = [ 1, 2, 3 ]
const result = splitAt(2, list)
// => [[ 1, 2 ], [ 3 ]]

Try this R.splitAt example in Rambda REPL

R.splitAt source
import { _isArray } from './_internals/_isArray'
import { drop } from './drop'
import { maybe } from './maybe'
import { take } from './take'

export function splitAt(index, input){
  if (arguments.length === 1){
    return _list => splitAt(index, _list)
  }
  if (!input) throw new TypeError(`Cannot read property 'slice' of ${ input }`)

  if (!_isArray(input) && typeof input !== 'string') return [ [], [] ]

  const correctIndex = maybe(
    index < 0,
    input.length + index < 0 ? 0 : input.length + index,
    index
  )

  return [ take(correctIndex, input), drop(correctIndex, input) ]
}
Tests
import { splitAt as splitAtRamda } from 'ramda'

import { splitAt } from './splitAt'

const list = [ 1, 2, 3 ]
const str = 'foo bar'

test('with array', () => {
  const result = splitAt(2, list)
  expect(result).toEqual([ [ 1, 2 ], [ 3 ] ])
})

test('with array - index is negative number', () => {
  const result = splitAt(-6, list)
  expect(result).toEqual([ [], list ])
})

test('with array - index is out of scope', () => {
  const result = splitAt(4, list)
  expect(result).toEqual([ [ 1, 2, 3 ], [] ])
})

test('with string', () => {
  const result = splitAt(4, str)
  expect(result).toEqual([ 'foo ', 'bar' ])
})

test('with string - index is negative number', () => {
  const result = splitAt(-2, str)
  expect(result).toEqual([ 'foo b', 'ar' ])
})

test('with string - index is out of scope', () => {
  const result = splitAt(10, str)
  expect(result).toEqual([ str, '' ])
})

test('with array - index is out of scope', () => {
  const result = splitAt(4)(list)
  expect(result).toEqual([ [ 1, 2, 3 ], [] ])
})

const badInputs = [ 1, true, /foo/g, {} ]
const throwingBadInputs = [ null, undefined ]

test('with bad inputs', () => {
  throwingBadInputs.forEach(badInput => {
    expect(() => splitAt(1, badInput)).toThrowWithMessage(TypeError,
      `Cannot read property 'slice' of ${ badInput }`)
    expect(() => splitAtRamda(1, badInput)).toThrowWithMessage(TypeError,
      `Cannot read property 'slice' of ${ badInput }`)
  })

  badInputs.forEach(badInput => {
    const result = splitAt(1, badInput)
    const ramdaResult = splitAtRamda(1, badInput)
    expect(result).toEqual(ramdaResult)
  })
})

---------------

splitEvery

splitEvery<T>(sliceLength: number, input: readonly T[]): readonly (readonly T[])[]

It splits input into slices of sliceLength.

const result = [
  R.splitEvery(2, [1, 2, 3]), 
  R.splitEvery(3, 'foobar') 
]

const expected = [
  [[1, 2], [3]],
  ['foo', 'bar']
]
// => `result` is equal to `expected`

Try this R.splitEvery example in Rambda REPL

R.splitEvery source
export function splitEvery(sliceLength, listOrString){
  if (arguments.length === 1){
    return _listOrString => splitEvery(sliceLength, _listOrString)
  }

  if (sliceLength < 1){
    throw new Error('First argument to splitEvery must be a positive integer')
  }

  const willReturn = []
  let counter = 0

  while (counter < listOrString.length){
    willReturn.push(listOrString.slice(counter, counter += sliceLength))
  }

  return willReturn
}
Tests
import { splitEvery } from './splitEvery'

test('happy', () => {
  expect(splitEvery(3, [ 1, 2, 3, 4, 5, 6, 7 ])).toEqual([
    [ 1, 2, 3 ],
    [ 4, 5, 6 ],
    [ 7 ],
  ])

  expect(splitEvery(3)('foobarbaz')).toEqual([ 'foo', 'bar', 'baz' ])
})

test('with bad input', () => {
  expect(() =>
    expect(splitEvery(0)('foo')).toEqual([ 'f', 'o', 'o' ])).toThrowWithMessage(Error,
    'First argument to splitEvery must be a positive integer')
})

---------------

splitWhen

splitWhen<T, U>(predicate: Predicate<T>, list: readonly U[]): readonly (readonly U[])[]

It splits list to two arrays according to a predicate function.

The first array contains all members of list before predicate returns true.

const list = [1, 2, 1, 2]
const result = R.splitWhen(R.equals(2), list)
// => [[1], [2, 1, 2]]

Try this R.splitWhen example in Rambda REPL

R.splitWhen source
export function splitWhen(predicate, input){
  if (arguments.length === 1){
    return _input => splitWhen(predicate, _input)
  }
  if (!input)
    throw new TypeError(`Cannot read property 'length' of ${ input }`)

  const preFound = []
  const postFound = []
  let found = false
  let counter = -1

  while (counter++ < input.length - 1){
    if (found){
      postFound.push(input[ counter ])
    } else if (predicate(input[ counter ])){
      postFound.push(input[ counter ])
      found = true
    } else {
      preFound.push(input[ counter ])
    }
  }

  return [ preFound, postFound ]
}
Tests
import { splitWhen as splitWhenRamda } from 'ramda'

import { equals } from './equals'
import { splitWhen } from './splitWhen'

const list = [ 1, 2, 1, 2 ]

test('happy', () => {
  const result = splitWhen(equals(2), list)
  expect(result).toEqual([ [ 1 ], [ 2, 1, 2 ] ])
})

test('when predicate returns false', () => {
  const result = splitWhen(equals(3))(list)
  expect(result).toEqual([ list, [] ])
})

const badInputs = [ 1, true, /foo/g, {} ]
const throwingBadInputs = [ null, undefined ]

test('with bad inputs', () => {
  throwingBadInputs.forEach(badInput => {
    expect(() => splitWhen(equals(2), badInput)).toThrowWithMessage(TypeError,
      `Cannot read property 'length' of ${ badInput }`)
    expect(() => splitWhenRamda(equals(2), badInput)).toThrowWithMessage(TypeError,
      `Cannot read property 'length' of ${ badInput }`)
  })

  badInputs.forEach(badInput => {
    const result = splitWhen(equals(2), badInput)
    const ramdaResult = splitWhenRamda(equals(2), badInput)
    expect(result).toEqual(ramdaResult)
  })
})

---------------

startsWith

startsWith(target: string, str: string): boolean

Curried version of String.prototype.startsWith

💥 It doesn't work with arrays unlike its corresponding Ramda method.

const str = 'foo-bar'

const result = [
  R.startsWith('foo', str),
  R.startsWith('bar', str)
]
// => [true, false]

Try this R.startsWith example in Rambda REPL

R.startsWith source
export function startsWith(target, str){
  if (arguments.length === 1) return _str => startsWith(target, _str)

  return str.startsWith(target)
}
Tests
import { startsWith } from './startsWith'

test('true', () => {
  const result = startsWith('foo', 'foo-bar')

  expect(result).toBeTrue()
})

test('false', () => {
  const result = startsWith('baz')('foo-bar')

  expect(result).toBeFalse()
})

---------------

subtract

Curried version of x - y

const x = 3
const y = 1

R.subtract(x, y) 
// => 2

Try this R.subtract example in Rambda REPL

---------------

sum

sum(list: readonly number[]): number
R.sum([1, 2, 3, 4, 5]) 
// => 15

Try this R.sum example in Rambda REPL

R.sum source
export function sum(list){
  return list.reduce((prev, current) => prev + current, 0)
}
Tests
import { sum } from './sum'

test('happy', () => {
  expect(sum([ 1, 2, 3, 4, 5 ])).toBe(15)
})

---------------

switcher

switcher<T>(valueToMatch: any): Switchem<T>

Edited fork of Switchem library.

The method return a value if the matched option is a value.

If the matched option is a function, then R.switcher returns a function which expects input. Tests of the method explain it better than this short description.

R.equals is used to determine equality.

const valueToMatch = {foo: 1}

const result = R.switcher(valueToMatch)
  .is('baz', 'is baz')
  .is(x => typeof x === 'boolean', 'is boolean')
  .is({foo: 1}, 'Property foo is 1')
  .default('is bar')

// => 'Property foo is 1'

Try this R.switcher example in Rambda REPL

R.switcher source
import { equals } from './equals'

const NO_MATCH_FOUND = Symbol ? Symbol('NO_MATCH_FOUND') : undefined

const getMatchingKeyValuePair = (
  cases, testValue, defaultValue
) => {
  let iterationValue

  for (let index = 0; index < cases.length; index++){
    iterationValue = cases[ index ].test(testValue)

    if (iterationValue !== NO_MATCH_FOUND){
      return iterationValue
    }
  }

  return defaultValue
}

const isEqual = (testValue, matchValue) => {
  const willReturn =
    typeof testValue === 'function' ?
      testValue(matchValue) :
      equals(testValue, matchValue)

  return willReturn
}

const is = (testValue, matchResult = true) => ({
  key  : testValue,
  test : matchValue =>
    isEqual(testValue, matchValue) ? matchResult : NO_MATCH_FOUND,
})

class Switchem{
  constructor(
    defaultValue, cases, willMatch
  ){
    if (cases === undefined && willMatch === undefined){
      this.cases = []
      this.defaultValue = undefined
      this.willMatch = defaultValue
    } else {
      this.cases = cases
      this.defaultValue = defaultValue
      this.willMatch = willMatch
    }

    return this
  }

  default(defaultValue){
    const holder = new Switchem(
      defaultValue, this.cases, this.willMatch
    )

    return holder.match(this.willMatch)
  }

  is(testValue, matchResult){
    return new Switchem(
      this.defaultValue,
      [ ...this.cases, is(testValue, matchResult) ],
      this.willMatch
    )
  }

  match(matchValue){
    return getMatchingKeyValuePair(
      this.cases, matchValue, this.defaultValue
    )
  }
}

export function switcher(input){
  return new Switchem(input)
}
Tests
import { add } from './add'
import { switcher } from './switcher'
import { tap } from './tap'
import { trim } from './trim'

test('with undefined', () => {
  const result = switcher(undefined)
    .is(x => x === 0, '0')
    .is(x => x === undefined, 'UNDEFINED')
    .default('3')

  expect(result).toEqual('UNDEFINED')
})

test('happy', () => {
  const a = true
  const b = false
  const result = switcher([ a, b ])
    .is([ false, false ], '0')
    .is([ false, true ], '1')
    .is([ true, true ], '2')
    .default('3')

  expect(result).toEqual('3')
})

test('can compare objects', () => {
  const result = switcher({ a : 1 })
    .is({ a : 1 }, 'it is object')
    .is('baz', 'it is baz')
    .default('it is default')

  expect(result).toEqual('it is object')
})

test('options are mixture of functions and values - input match function', () => {
  const fn = switcher('foo').is('bar', 1)
    .is('foo', add(1))
    .default(1000)

  expect(fn(2)).toEqual(3)
})

test('options are mixture of functions and values - input match value', () => {
  const result = switcher('bar').is('bar', 1)
    .is('foo', add(1))
    .default(1000)

  expect(result).toBe(1)
})

test('return function if all options are functions', () => {
  const fn = switcher('foo').is('bar', tap)
    .is('foo', add(1))
    .default(trim)

  expect(fn(2)).toEqual(3)
})

const switchFn = input =>
  switcher(input)
    .is(x => x.length && x.length === 7, 'has length of 7')
    .is('baz', 'it is baz')
    .default('it is default')

test('works with function as condition', () => {
  expect(switchFn([ 0, 1, 2, 3, 4, 5, 6 ])).toEqual('has length of 7')
})

test('works with string as condition', () => {
  expect(switchFn('baz')).toEqual('it is baz')
})

test('fallback to default input when no matches', () => {
  expect(switchFn(1)).toEqual('it is default')
})

---------------

symmetricDifference

symmetricDifference<T>(x: readonly T[], y: readonly T[]): readonly T[]

It returns a merged list of x and y with all equal elements removed.

R.equals is used to determine equality.

const x = [ 1, 2, 3, 4 ]
const y = [ 3, 4, 5, 6 ]

const result = symmetricDifference(x, y)
// => [ 1, 2, 5, 6 ]

Try this R.symmetricDifference example in Rambda REPL

R.symmetricDifference source
import { concat } from './concat'
import { filter } from './filter'
import { includes } from './includes'

export function symmetricDifference(x, y){
  if (arguments.length === 1){
    return _y => symmetricDifference(x, _y)
  }

  return concat(filter(value => !includes(value, y), x),
    filter(value => !includes(value, x), y))
}
Tests
import { symmetricDifference } from './symmetricDifference'

test('symmetricDifference', () => {
  const list1 = [ 1, 2, 3, 4 ]
  const list2 = [ 3, 4, 5, 6 ]
  expect(symmetricDifference(list1)(list2)).toEqual([ 1, 2, 5, 6 ])

  expect(symmetricDifference([], [])).toEqual([])
})

test('symmetricDifference with objects', () => {
  const list1 = [ { id : 1 }, { id : 2 }, { id : 3 }, { id : 4 } ]
  const list2 = [ { id : 3 }, { id : 4 }, { id : 5 }, { id : 6 } ]
  expect(symmetricDifference(list1)(list2)).toEqual([
    { id : 1 },
    { id : 2 },
    { id : 5 },
    { id : 6 },
  ])
})

---------------

T

T(): boolean
R.T() 
// => true

Try this R.T example in Rambda REPL

R.T source
export function T(){
  return true
}

---------------

tail

tail<T>(input: readonly T[]): readonly T[]

It returns all but the first element of input.

const result = [
  R.tail([1, 2, 3]),  
  R.tail('foo') 
]
// => [[2, 3], 'oo']

Try this R.tail example in Rambda REPL

R.tail source
import { drop } from './drop'

export function tail(listOrString){
  return drop(1, listOrString)
}
Tests
import { tail } from './tail'

test('tail', () => {
  expect(tail([ 1, 2, 3 ])).toEqual([ 2, 3 ])
  expect(tail([ 1, 2 ])).toEqual([ 2 ])
  expect(tail([ 1 ])).toEqual([])
  expect(tail([])).toEqual([])

  expect(tail('abc')).toEqual('bc')
  expect(tail('ab')).toEqual('b')
  expect(tail('a')).toEqual('')
  expect(tail('')).toEqual('')
})

---------------

take

take<T>(howMany: number, input: readonly T[]): readonly T[]

It returns the first howMany elements of input.

const howMany = 2

const result = [
  R.take(howMany, [1, 2, 3]),
  R.take(howMany, 'foobar'),
]
// => [[1, 2], 'fo']

Try this R.take example in Rambda REPL

R.take source
import baseSlice from './_internals/baseSlice'

export function take(howMany, listOrString){
  if (arguments.length === 1)
    return _listOrString => take(howMany, _listOrString)
  if (howMany < 0) return listOrString.slice()
  if (typeof listOrString === 'string') return listOrString.slice(0, howMany)

  return baseSlice(
    listOrString, 0, howMany
  )
}
Tests
import { take } from './take'

test('happy', () => {
  const arr = [ 'foo', 'bar', 'baz' ]

  expect(take(1, arr)).toEqual([ 'foo' ])

  expect(arr).toEqual([ 'foo', 'bar', 'baz' ])

  expect(take(2)([ 'foo', 'bar', 'baz' ])).toEqual([ 'foo', 'bar' ])
  expect(take(3, [ 'foo', 'bar', 'baz' ])).toEqual([ 'foo', 'bar', 'baz' ])
  expect(take(4, [ 'foo', 'bar', 'baz' ])).toEqual([ 'foo', 'bar', 'baz' ])
  expect(take(3)('rambda')).toEqual('ram')
})

test('with negative index', () => {
  expect(take(-1, [ 1, 2, 3 ])).toEqual([ 1, 2, 3 ])
  expect(take(-Infinity, [ 1, 2, 3 ])).toEqual([ 1, 2, 3 ])
})

test('with zero index', () => {
  expect(take(0, [ 1, 2, 3 ])).toEqual([])
})

---------------

takeLast

takeLast<T>(howMany: number, input: readonly T[]): readonly T[]

It returns the last howMany elements of input.

const howMany = 2

const result = [
  R.takeLast(howMany, [1, 2, 3]),
  R.takeLast(howMany, 'foobar'),
]
// => [[2, 3], 'ar']

Try this R.takeLast example in Rambda REPL

R.takeLast source
import baseSlice from './_internals/baseSlice'

export function takeLast(howMany, listOrString){
  if (arguments.length === 1)
    return _listOrString => takeLast(howMany, _listOrString)

  const len = listOrString.length
  if (howMany < 0) return listOrString.slice()
  let numValue = howMany > len ? len : howMany

  if (typeof listOrString === 'string')
    return listOrString.slice(len - numValue)

  numValue = len - numValue

  return baseSlice(
    listOrString, numValue, len
  )
}
Tests
import { takeLast } from './takeLast'

test('with arrays', () => {
  expect(takeLast(1, [ 'foo', 'bar', 'baz' ])).toEqual([ 'baz' ])

  expect(takeLast(2)([ 'foo', 'bar', 'baz' ])).toEqual([ 'bar', 'baz' ])

  expect(takeLast(3, [ 'foo', 'bar', 'baz' ])).toEqual([ 'foo', 'bar', 'baz' ])

  expect(takeLast(4, [ 'foo', 'bar', 'baz' ])).toEqual([ 'foo', 'bar', 'baz' ])

  expect(takeLast(10, [ 'foo', 'bar', 'baz' ])).toEqual([ 'foo', 'bar', 'baz' ])
})

test('with strings', () => {
  expect(takeLast(3, 'rambda')).toEqual('bda')

  expect(takeLast(7, 'rambda')).toEqual('rambda')
})

test('with negative index', () => {
  expect(takeLast(-1, [ 1, 2, 3 ])).toEqual([ 1, 2, 3 ])
  expect(takeLast(-Infinity, [ 1, 2, 3 ])).toEqual([ 1, 2, 3 ])
})

---------------

takeLastWhile

takeLastWhile(predicate: (x: string) => boolean, input: string): string
const result = R.takeLastWhile(
  x => x > 2,
  [1, 2, 3, 4]
)
// => [3, 4]

Try this R.takeLastWhile example in Rambda REPL

R.takeLastWhile source
import { _isArray } from './_internals/_isArray'

export function takeLastWhile(predicate, input){
  if (arguments.length === 1){
    return _input => takeLastWhile(predicate, _input)
  }
  if (input.length === 0) return input
  let found = false
  const toReturn = []
  let counter = input.length

  while (!found || counter === 0){
    counter--
    if (predicate(input[ counter ]) === false){
      found = true
    } else if (!found){
      toReturn.push(input[ counter ])
    }
  }

  return _isArray(input) ? toReturn.reverse() : toReturn.reverse().join('')
}
Tests
import { takeLastWhile } from './takeLastWhile'
const assert = require('assert')

const list = [ 1, 2, 3, 4 ]

test('happy', () => {
  const predicate = x => x > 2
  const result = takeLastWhile(predicate, list)
  expect(result).toEqual([ 3, 4 ])
})

test('predicate is always true', () => {
  const predicate = x => x > 0
  const result = takeLastWhile(predicate)(list)
  expect(result).toEqual(list)
})

test('predicate is always false', () => {
  const predicate = x => x < 0
  const result = takeLastWhile(predicate, list)
  expect(result).toEqual([])
})

test('with string', () => {
  const result = takeLastWhile(x => x !== 'F', 'FOOBAR')
  expect(result).toEqual('OOBAR')
})

---------------

takeUntil

takeUntil<T>(predicate: (x: T) => boolean, list: readonly T[]): readonly T[]
const list = [1, 2, 3, 4, 5]
const predicate = x => x > 3
const result = R.takeUntil(predicate, list)

// => [1, 2, 3]

Try this R.takeUntil example in Rambda REPL

R.takeUntil source
export function takeUntil(predicate, list){
  const toReturn = []
  let stopFlag = false
  let counter = -1

  while (stopFlag === false && counter++ < list.length - 1){
    if (predicate(list[ counter ])){
      stopFlag = true
    } else {
      toReturn.push(list[ counter ])
    }
  }

  return toReturn
}
Tests
import { takeUntil } from './takeUntil'

const list = [ 1, 2, 3, 4, 5, 6 ]

test('happy', () => {
  const result = takeUntil(x => x > 3, list)
  expect(result).toEqual([ 1, 2, 3 ])
})

test('predicate always returns true', () => {
  const result = takeUntil(x => x < 10, list)
  expect(result).toEqual([])
})

test('predicate always returns false', () => {
  const result = takeUntil(x => x > 10, list)
  expect(result).toEqual(list)
})

---------------

takeWhile

const list = [1, 2, 3, 4]
const predicate = x => x < 3

const result = R.takeWhile(predicate, list)
// => [1, 2]

Try this R.takeWhile example in Rambda REPL

---------------

tap

tap<T>(fn: (x: T) => void, input: T): T

It applies function fn to input x and returns x.

One use case is debuging in the middle of R.compose.

const list = [1, 2, 3]

R.compose(
  R.map(x => x * 2)
  R.tap(console.log),
  R.filter(x => x > 1)
)(list)
// => `2` and `3` will be logged

Try this R.tap example in Rambda REPL

R.tap source
export function tap(fn, x){
  if (arguments.length === 1) return _x => tap(fn, _x)

  fn(x)

  return x
}
Tests
import { tap } from './tap'

test('tap', () => {
  let a = 1
  const sayX = x => a = x

  expect(tap(sayX, 100)).toEqual(100)
  expect(tap(sayX)(100)).toEqual(100)
  expect(a).toEqual(100)
})

---------------

tapAsync

tapAsync<T>(fn: Func<any> | Promise<any>, input: T): T

Asynchronous version of R.tap.

R.tapAsync source
async function tapAsyncFn(fn, input){
  await fn(input)

  return input
}

export function tapAsync(fn, input){
  if (arguments.length === 1){
    return async _input => tapAsyncFn(fn, _input)
  }

  return new Promise((resolve, reject) => {
    tapAsyncFn(fn, input).then(resolve)
      .catch(reject)
  })
}
Tests
import { delay } from './delay'
import { pipedAsync } from './pipedAsync'
import { tapAsync } from './tapAsync'

test('happy', async () => {
  const result = await tapAsync(delay, 1)
  expect(result).toEqual(1)
})

test('complex', async () => {
  let marker = false
  const fn = () => marker = true
  const result = await pipedAsync(
    1,
    async x => {
      await delay(100)

      return x + 1
    },
    tapAsync(fn),
    x => x + 1
  )
  expect(marker).toBeTrue()
  expect(result).toEqual(3)
})

---------------

test

test(regExpression: RegExp): (str: string) => boolean

It determines whether str matches regExpression.

R.test(/^f/, 'foo')
// => true

Try this R.test example in Rambda REPL

R.test source
export function test(pattern, str){
  if (arguments.length === 1) return _str => test(pattern, _str)

  if (typeof pattern === 'string'){
    throw new TypeError(`‘test’ requires a value of type RegExp as its first argument; received "${ pattern }"`)
  }

  return str.search(pattern) !== -1
}
Tests
import { test as testMethod } from './test'

test('happy', () => {
  expect(testMethod(/^x/, 'xyz')).toBeTrue()

  expect(testMethod(/^y/)('xyz')).toBeFalse()
})

test('throws if first argument is not regex', () => {
  expect(() => testMethod('foo', 'bar')).toThrowWithMessage(TypeError,
    '‘test’ requires a value of type RegExp as its first argument; received "foo"')
})

---------------

throttle

throttle<T, U>(fn: (input: T) => U, ms: number): (input: T) => U

It creates a throttled function that invokes fn maximum once for a period of milliseconds.

let counter = 0
const inc = () => {
  counter++
}

const throttledInc = R.throttle(inc, 800)

const result = async () => {
  throttledInc()
  await R.delay(500)
  throttledInc()

  return counter
}
// `result` resolves to `1`

Try this R.throttle example in Rambda REPL

R.throttle source
export function throttle(fn, ms){
  let wait = false

  return function (...input){
    if (!wait){
      fn.apply(null, input)
      wait = true
      setTimeout(() => {
        wait = false
      }, ms)
    }
  }
}
Tests
import { delay } from './delay'
import { throttle } from './throttle'

test('', async () => {
  let counter = 0
  let aHolder
  let bHolder

  const inc = (a, b) => {
    aHolder = a
    bHolder = b
    counter++
  }

  const incWrapped = throttle(inc, 1000)

  incWrapped(1, 2)

  await delay(500)

  incWrapped(2, 3)
  incWrapped(3, 4)

  expect(counter).toBe(1)
  expect(aHolder).toBe(1)
  expect(bHolder).toBe(2)

  await delay(1000)

  incWrapped(5, 6)

  expect(counter).toBe(2)
  expect(aHolder).toBe(5)
  expect(bHolder).toBe(6)
})

---------------

times

times<T>(fn: (i: number) => T, howMany: number): readonly T[]

It returns the result of applying function fn over members of range array.

The range array includes numbers between 0 and howMany(exclusive).

const fn = x => x * 2
const howMany = 5

R.times(fn, howMany)
// => [0, 2, 4, 6, 8]

Try this R.times example in Rambda REPL

R.times source
import { map } from './map'
import { range } from './range'

export function times(fn, howMany){
  if (arguments.length === 1) return _howMany => times(fn, _howMany)
  if (!Number.isInteger(howMany) || howMany < 0){
    throw new RangeError('n must be an integer')
  }

  return map(fn, range(0, howMany))
}
Tests
import assert from 'assert'

import { identity } from './identity'
import { times } from './times'

test('happy', () => {
  const result = times(identity, 5)

  expect(result).toEqual([ 0, 1, 2, 3, 4 ])
})

test('with bad input', () => {
  assert.throws(() => {
    times(3)('cheers!')
  }, RangeError)
  assert.throws(() => {
    times(identity, -1)
  }, RangeError)
})

test('curry', () => {
  const result = times(identity)(5)

  expect(result).toEqual([ 0, 1, 2, 3, 4 ])
})

---------------

toDecimal

toDecimal(num: number, charsAfterDecimalPoint?: number): number
R.toDecimal(2.45464,2) // => 2.45

Try this R.toDecimal example in Rambda REPL

R.toDecimal source
export function toDecimal(number, charsAfterDecimalPoint = 2){
  return Number(parseFloat(String(number)).toFixed(charsAfterDecimalPoint))
}
Tests
import { toDecimal } from './toDecimal'

test('happy', () => {
  expect(toDecimal(2.2789, 1)).toBe(2.3)
  expect(toDecimal(2.2789, 3)).toBe(2.279)
  expect(toDecimal(2, 3)).toBe(2)
  expect(toDecimal(2.2789)).toBe(2.28)
  expect(toDecimal(2.45464)).toBe(2.45)
})

---------------

toLower

toLower(str: string): string
R.toLower('FOO')
// => 'foo'

Try this R.toLower example in Rambda REPL

R.toLower source
export function toLower(str){
  return str.toLowerCase()
}
Tests
import { toLower } from './toLower'

test('toLower', () => {
  expect(toLower('FOO|BAR|BAZ')).toEqual('foo|bar|baz')
})

---------------

toPairs

toPairs<S>(obj: { readonly [k: string]: S } | { readonly [k: number]: S }): readonly (readonly [string, S])[]

It transforms an object to a list.

const list = {
  a : 1,
  b : 2,
  c : [ 3, 4 ],
}
const expected = [ [ 'a', 1 ], [ 'b', 2 ], [ 'c', [ 3, 4 ] ] ]

const result = R.toPairs(list)
// => `result` is equal to `expected`

Try this R.toPairs example in Rambda REPL

R.toPairs source
export function toPairs(obj){
  return Object.entries(obj)
}
Tests
import { toPairs } from './toPairs'

const obj = {
  a : 1,
  b : 2,
  c : [ 3, 4 ],
}
const expected = [
  [ 'a', 1 ],
  [ 'b', 2 ],
  [ 'c', [ 3, 4 ] ],
]

test('happy', () => {
  expect(toPairs(obj)).toEqual(expected)
})

---------------

toString

toString<T>(x: T): string
R.toString([1, 2]) 
// => '1,2'

Try this R.toString example in Rambda REPL

R.toString source
export function toString(x){
  return x.toString()
}
Tests
import { toString } from './toString'

test('happy', () => {
  expect(toString([ 1, 2, 3 ])).toEqual('1,2,3')
})

---------------

toUpper

toUpper(str: string): string
R.toUpper('foo')
// => 'FOO'

Try this R.toUpper example in Rambda REPL

R.toUpper source
export function toUpper(str){
  return str.toUpperCase()
}
Tests
import { toUpper } from './toUpper'

test('toUpper', () => {
  expect(toUpper('foo|bar|baz')).toEqual('FOO|BAR|BAZ')
})

---------------

transpose

transpose<T>(list: readonly (readonly T[])[]): readonly (readonly T[])[]
const list = [[10, 11], [20], [], [30, 31, 32]]
const expected = [[10, 20, 30], [11, 31], [32]]

const result = R.transpose(list)
// => `result` is equal to `expected`

Try this R.transpose example in Rambda REPL

R.transpose source
import { _isArray } from './_internals/_isArray'

export function transpose(array){
  return array.reduce((acc, el) => {
    el.forEach((nestedEl, i) =>
      _isArray(acc[ i ]) ? acc[ i ].push(nestedEl) : acc.push([ nestedEl ]))

    return acc
  }, [])
}
Tests
import { transpose } from './transpose'

test('happy', () => {
  const input = [
    [ 'a', 1 ],
    [ 'b', 2 ],
    [ 'c', 3 ],
  ]

  expect(transpose(input)).toEqual([
    [ 'a', 'b', 'c' ],
    [ 1, 2, 3 ],
  ])
})

test('when rows are shorter', () => {
  const actual = transpose([ [ 10, 11 ], [ 20 ], [], [ 30, 31, 32 ] ])
  const expected = [ [ 10, 20, 30 ], [ 11, 31 ], [ 32 ] ]
  expect(actual).toEqual(expected)
})

test('with empty array', () => {
  expect(transpose([])).toEqual([])
})

test('array with falsy values', () => {
  const actual = transpose([
    [ true, false, undefined, null ],
    [ null, undefined, false, true ],
  ])
  const expected = [
    [ true, null ],
    [ false, undefined ],
    [ undefined, false ],
    [ null, true ],
  ]
  expect(actual).toEqual(expected)
})

---------------

trim

trim(str: string): string
R.trim('  foo  ') 
// => 'foo'

Try this R.trim example in Rambda REPL

R.trim source
export function trim(str){
  return str.trim()
}
Tests
import { trim } from './trim'

test('trim', () => {
  expect(trim(' foo ')).toEqual('foo')
})

---------------

tryCatch

tryCatch<T, U>(
  fn: (input: T) => U,
  fallback: U
): (input: T) => U

It returns function that runs fn in try/catch block. If there was an error, then fallback is used to return the result. Note that fn can be value or asynchronous/synchronous function(unlike Ramda where fallback can only be a synchronous function).

💥 Please check the tests of R.tryCatch to fully understand how this method works.

const fn = x => x.foo

const result = [
  R.tryCatch(fn, false)(null),
  R.tryCatch(fn, false)({foo: 'bar'})
]
// => [false, 'bar']

Try this R.tryCatch example in Rambda REPL

R.tryCatch source
import { isFunction } from './isFunction'

export function tryCatch(fn, fallback){
  if (!isFunction(fn)){
    throw new Error(`R.tryCatch | fn '${ fn }'`)
  }
  const passFallback = isFunction(fallback)

  return (...inputs) => {
    try {
      return fn(...inputs)
    } catch (e){
      return passFallback ? fallback(e, ...inputs) : fallback
    }
  }
}
Tests
import { tryCatch as tryCatchRamda } from 'ramda'

import { compareCombinations } from './_internals/testUtils'
import { prop } from './prop'
import { tryCatch } from './tryCatch'

test('happy', () => {
  const fn = () => {
    throw new Error('foo')
  }
  const result = tryCatch(fn, () => true)()
  expect(result).toBeTrue()
})

test('when fallback is used', () => {
  const fn = x => x.x

  expect(tryCatch(fn, false)(null)).toBeFalse()
})

test('with json parse', () => {
  const good = () => JSON.parse(JSON.stringify({ a : 1 }))
  const bad = () => JSON.parse('a{a')

  expect(tryCatch(good, 1)()).toEqual({ a : 1 })
  expect(tryCatch(bad, 1)()).toBe(1)
})

test('when fallback is function', () => {
  const fn = x => x.x

  expect(tryCatch(fn, () => 1)(null)).toBe(1)
})

test('when fn is used', () => {
  const fn = prop('x')

  expect(tryCatch(fn, false)({})).toBe(undefined)
  expect(tryCatch(fn, false)({ x : 1 })).toBe(1)
})

test('fallback receives error object and all initial inputs', () => {
  function thrower(
    a, b, c
  ){
    void c
    throw new Error('throwerError')
  }

  function catchFn(
    e, a, b, c
  ){
    return [ e.message, a, b, c ].join('|')
  }

  const willThrow = tryCatch(thrower, catchFn)
  const result = willThrow(
    'A', 'B', 'C'
  )
  expect(result).toBe('throwerError|A|B|C')
})

test('fallback receives error object', () => {
  function throwFn(){
    throw new Error(10)
  }

  function eCatcher(
    e, a, b
  ){
    return e.message
  }

  const willThrow = tryCatch(throwFn, eCatcher)
  expect(willThrow([])).toBe('10')
  expect(willThrow([ {}, {}, {} ])).toBe('10')
})

const possibleFns = [
  null,
  () => 1,
  () => 0,
  () => JSON.parse('{a:1'),
  () => {
    const x = {}

    return x.x
  },
  x => x.foo,
  () => {
    throw new Error('foo')
  },
]

const possibleCatchers = [
  null,
  e => e.message.length,
  (e, ...inputs) => `${ e.message.length } ${ inputs.length }`,
  () => {
    throw new Error('bar')
  },
]

const possibleInputs = [ null, {}, { foo : 1 } ]

describe('brute force', () => {
  compareCombinations({
    returnsFunctionFlag : true,
    firstInput          : possibleFns,
    callback            : errorsCounters => {
      expect(errorsCounters).toMatchInlineSnapshot(`
        Object {
          "ERRORS_MESSAGE_MISMATCH": 0,
          "ERRORS_TYPE_MISMATCH": 12,
          "RESULTS_MISMATCH": 0,
          "SHOULD_NOT_THROW": 0,
          "SHOULD_THROW": 7,
        }
      `)
    },
    secondInput : possibleCatchers,
    thirdInput  : possibleInputs,
    fn          : tryCatch,
    fnRamda     : tryCatchRamda,
  })
})

---------------

tryCatchAsync

tryCatchAsync<T>(
  fn: (input: any) => Promise<T>,
  fallback: T
): (input: any) => Promise<T>

It returns function that runs fn in try/catch block. If there was an error, then fallback is used to return the result.

const x = {foo: 1}
const fnFoo = async () => x.foo
const fnBar = async () => x.bar

const result = await Promise.all([
  R.tryCatchAsync (fnFoo, false)(),
  R.tryCatchAsync(fnBar, false)()
])
// => [1, false]

Try this R.tryCatchAsync example in Rambda REPL

R.tryCatchAsync source
import { isFunction } from './isFunction'
import { isPromise } from './isPromise'

export function tryCatchAsync(fn, fallback){
  return (...inputs) =>
    new Promise(resolve => {
      fn(...inputs)
        .then(resolve)
        .catch(err => {
          if (!isFunction(fallback)){
            return resolve(fallback)
          }
          if (!isPromise(fallback)){
            return resolve(fallback(err, ...inputs))
          }

          fallback(err, ...inputs).then(resolve)
            .catch(resolve)
        })
    })
}
Tests
import { delay } from './delay'
import { tryCatchAsync } from './tryCatchAsync'

test('when fallback is function', async () => {
  const fn = async x => x.x

  const result = await tryCatchAsync(fn, () => 1)(null)
  expect(result).toBe(1)
})

test('when fallback throws', async () => {
  const fn = async x => x.x
  const fallback = async (err, xx) => xx.y

  const result = await tryCatchAsync(fn, fallback)(null)
  expect(result.message).toMatchInlineSnapshot('"Cannot read property \'y\' of null"')
})

test('fallback is not used', async () => {
  const fn = async x => x.x

  expect(tryCatchAsync(fn, false)({ x : 1 })).resolves.toBe(1)
})

test('fallback receives error object and all initial inputs', async () => {
  async function thrower(
    a, b, c
  ){
    void c
    throw new Error('throwerError')
  }

  function catchFn(
    e, a, b, c
  ){
    return [ e.message, a, b, c ].join('|')
  }

  const willThrow = tryCatchAsync(thrower, catchFn)
  const result = await willThrow(
    'A', 'B', 'C'
  )
  expect(result).toBe('throwerError|A|B|C')
})

test('when async + value fallback', async () => {
  const fn = async () => {
    await delay(100)

    return JSON.parse('{a:')
  }

  const result = await tryCatchAsync(fn, 'fallback')(100)
  expect(result).toBe('fallback')
})

---------------

type

type(x: any): RambdaTypes

It accepts any input and it returns its type.

💥 NaN, Promise and Async are types specific for Rambda.

R.type(() => {}) // => 'Function'
R.type(async () => {}) // => 'Async'
R.type([]) // => 'Array'
R.type({}) // => 'Object'
R.type('foo') // => 'String'
R.type(1) // => 'Number'
R.type(true) // => 'Boolean'
R.type(null) // => 'Null'
R.type(/[A-z]/) // => 'RegExp'
R.type('foo'*1) // => 'NaN'

const delay = ms => new Promise(resolve => {
  setTimeout(function () {
    resolve()
  }, ms)
})
R.type(delay) // => 'Promise'

Try this R.type example in Rambda REPL

R.type source
import { _isArray } from './_internals/_isArray'

export function type(input){
  const typeOf = typeof input

  if (input === null){
    return 'Null'
  } else if (input === undefined){
    return 'Undefined'
  } else if (typeOf === 'boolean'){
    return 'Boolean'
  } else if (typeOf === 'number'){
    return Number.isNaN(input) ? 'NaN' : 'Number'
  } else if (typeOf === 'string'){
    return 'String'
  } else if (_isArray(input)){
    return 'Array'
  } else if (typeOf === 'symbol'){
    return 'Symbol'
  } else if (input instanceof RegExp){
    return 'RegExp'
  }

  const asStr = input && input.toString ? input.toString() : ''

  if ([ 'true', 'false' ].includes(asStr)) return 'Boolean'
  if (!Number.isNaN(Number(asStr))) return 'Number'
  if (asStr.startsWith('async')) return 'Async'
  if (asStr === '[object Promise]') return 'Promise'
  if (typeOf === 'function') return 'Function'
  if (input instanceof String) return 'String'

  return 'Object'
}
Tests
import { type as ramdaType } from 'ramda'

import { type } from './type'

test('with symbol', () => {
  expect(type(Symbol())).toBe('Symbol')
})

test('with simple promise', () => {
  expect(type(Promise.resolve(1))).toBe('Promise')
})

test('with new Boolean', () => {
  expect(type(new Boolean(true))).toBe('Boolean')
})

test('with new String', () => {
  expect(type(new String('I am a String object'))).toEqual('String')
})

test('with new Number', () => {
  expect(type(new Number(1))).toBe('Number')
})

test('with new promise', () => {
  const delay = ms =>
    new Promise(resolve => {
      setTimeout(() => {
        resolve(ms + 110)
      }, ms)
    })

  expect(type(delay(10))).toEqual('Promise')
})

test('async function', () => {
  expect(type(async () => {})).toEqual('Async')
})

test('async arrow', () => {
  const asyncArrow = async () => {}
  expect(type(asyncArrow)).toBe('Async')
})

test('function', () => {
  const fn1 = () => {}
  const fn2 = function (){}

  function fn3(){}

  ;[ () => {}, fn1, fn2, fn3 ].map(val => {
    expect(type(val)).toEqual('Function')
  })
})

test('object', () => {
  expect(type({})).toEqual('Object')
})

test('number', () => {
  expect(type(1)).toEqual('Number')
})

test('boolean', () => {
  expect(type(false)).toEqual('Boolean')
})

test('string', () => {
  expect(type('foo')).toEqual('String')
})

test('null', () => {
  expect(type(null)).toEqual('Null')
})

test('array', () => {
  expect(type([])).toEqual('Array')
  expect(type([ 1, 2, 3 ])).toEqual('Array')
})

test('regex', () => {
  expect(type(/\s/g)).toEqual('RegExp')
})

test('undefined', () => {
  expect(type(undefined)).toEqual('Undefined')
})

test('not a number', () => {
  expect(type(Number('s'))).toBe('NaN')
})

test('function inside object 1', () => {
  const obj = {
    f(){
      return 4
    },
  }

  expect(type(obj.f)).toBe('Function')
  expect(ramdaType(obj.f)).toBe('Function')
})

test('function inside object 2', () => {
  const name = 'f'
  const obj = {
    [ name ](){
      return 4
    },
  }
  expect(type(obj.f)).toBe('Function')
  expect(ramdaType(obj.f)).toBe('Function')
})

---------------

union

union<T>(x: readonly T[], y: readonly T[]): readonly T[]

It takes two lists and return a new list containing a merger of both list with removed duplicates.

R.equals is used to compare for duplication.

const result = R.union([1,2,3], [3,4,5]);
// => [1, 2, 3, 4, 5]

Try this R.union example in Rambda REPL

R.union source
import { includes } from './includes'

export function union(x, y){
  if (arguments.length === 1) return _y => union(x, _y)

  const toReturn = x.slice()

  y.forEach(yInstance => {
    if (!includes(yInstance, x)) toReturn.push(yInstance)
  })

  return toReturn
}
Tests
import { union } from './union'

test('happy', () => {
  expect(union([ 1, 2 ], [ 2, 3 ])).toEqual([ 1, 2, 3 ])
})

test('with list of objects', () => {
  const list1 = [ { a : 1 }, { a : 2 } ]
  const list2 = [ { a : 2 }, { a : 3 } ]
  const result = union(list1)(list2)
})

---------------

uniq

uniq<T>(list: readonly T[]): readonly T[]

It returns a new array containing only one copy of each element of list.

R.equals is used to determine equality.

const list = [1, 1, {a: 1}, {a: 2}, {a:1}]

R.uniq(list)
// => [1, {a: 1}, {a: 2}]

Try this R.uniq example in Rambda REPL

R.uniq source
import { includes } from './includes'

export function uniq(list){
  let index = -1
  const willReturn = []

  while (++index < list.length){
    const value = list[ index ]

    if (!includes(value, willReturn)){
      willReturn.push(value)
    }
  }

  return willReturn
}
Tests
import { uniq } from './uniq'

test('happy', () => {
  expect(uniq([ 1, 2, 3, 3, 3, 1, 2, 0 ])).toEqual([ 1, 2, 3, 0 ])
  expect(uniq([ 1, 1, 2, 1 ])).toEqual([ 1, 2 ])
  expect([ 1, '1' ]).toEqual([ 1, '1' ])
  expect(uniq([ [ 42 ], [ 42 ] ])).toEqual([ [ 42 ] ])
})

---------------

uniqWith

uniqWith<T, U>(predicate: (x: T, y: T) => boolean, list: readonly T[]): readonly T[]

It returns a new array containing only one copy of each element in list according to predicate function.

This predicate should return true, if two elements are equal.

const list = [
  {id: 0, title:'foo'},
  {id: 1, title:'bar'},
  {id: 2, title:'baz'},
  {id: 3, title:'foo'},
  {id: 4, title:'bar'},
]

const expected = [
  {id: 0, title:'foo'},
  {id: 1, title:'bar'},
  {id: 2, title:'baz'},
]

const predicate = (x,y) => x.title === y.title

const result = R.uniqWith(predicate, list)
// => `result` is equal to `expected`

Try this R.uniqWith example in Rambda REPL

R.uniqWith source
import { any } from './any'

export function uniqWith(predicate, list){
  if (arguments.length === 1) return _list => uniqWith(predicate, _list)

  let index = -1
  const len = list.length
  const willReturn = []

  while (++index < len){
    const value = list[ index ]
    const flag = any(willReturnInstance => predicate(value, willReturnInstance),
      willReturn)

    if (!flag){
      willReturn.push(value)
    }
  }

  return willReturn
}
Tests
import { uniqWith } from './uniqWith'

test('happy', () => {
  const input = [
    {
      id    : 0,
      title : 'foo',
    },
    {
      id    : 1,
      title : 'bar',
    },
    {
      id    : 2,
      title : 'baz',
    },
    {
      id    : 3,
      title : 'foo',
    },
    {
      id    : 4,
      title : 'bar',
    },
  ]

  const expectedResult = [
    {
      id    : 0,
      title : 'foo',
    },
    {
      id    : 1,
      title : 'bar',
    },
    {
      id    : 2,
      title : 'baz',
    },
  ]

  const fn = (x, y) => x.title === y.title

  const result = uniqWith(fn, input)
  const curriedResult = uniqWith(fn)(input)

  expect(result).toEqual(expectedResult)

  expect(curriedResult).toEqual(expectedResult)
})

test('uniqWith', () => {
  const input = [
    {
      id    : 0,
      title : 'foo',
    },
    {
      id    : 1,
      title : 'bar',
    },
    {
      id    : 2,
      title : 'baz',
    },
    {
      id    : 3,
      title : 'foo',
    },
    {
      id    : 4,
      title : 'bar',
    },
  ]

  const expectedResult = [
    {
      id    : 0,
      title : 'foo',
    },
    {
      id    : 1,
      title : 'bar',
    },
    {
      id    : 2,
      title : 'baz',
    },
  ]

  const fn = (x, y) => x.title === y.title

  const result = uniqWith(fn, input)
  //const result = uniqWith(Ramda.eqBy(Ramda.prop('title')), input)

  expect(result).toEqual(expectedResult)
})

---------------

unless

unless<T, U>(predicate: (x: T) => boolean, whenFalseFn: (x: T) => U, obj: T): U

The method returns function that will be called with argument input.

If predicate(input) returns false, then the end result will be the outcome of whenFalse(input).

In the other case, the final output will be the input itself.

const fn = R.unless(
  x => x > 2,
  x => x + 10
)

const result = [
  fn(1),
  fn(5)
]
// => [11, 5]

Try this R.unless example in Rambda REPL

R.unless source
export function unless(predicate, whenFalse){
  if (arguments.length === 1){
    return _whenFalse => unless(predicate, _whenFalse)
  }

  return input => {
    if (predicate(input)) return input

    return whenFalse(input)
  }
}
Tests
import { inc } from './inc'
import { isNil } from './isNil'
import { unless } from './unless'

const safeInc = unless(isNil, inc)

test('happy', () => {
  expect(safeInc(null)).toBeNull()
  expect(safeInc(1)).toBe(2)
})

test('curried', () => {
  const safeIncCurried = unless(isNil)(inc)
  expect(safeIncCurried(null)).toBeNull()
  expect(safeIncCurried(1)).toBe(2)
})

---------------

update

update<T>(index: number, newValue: T, list: readonly T[]): readonly T[]

It returns a copy of list with updated element at index with newValue.

const index = 2
const newValue = 88
const list = [1, 2, 3, 4, 5]

const result = R.update(index, newValue, list)
// => [1, 2, 88, 4, 5]

Try this R.update example in Rambda REPL

R.update source
import { curry } from './curry'

function updateFn(
  index, newValue, list
){
  const arrClone = list.slice()

  return arrClone.fill(
    newValue, index, index + 1
  )
}

export const update = curry(updateFn)
Tests
import { update } from './update'

const list = [ 1, 2, 3 ]

test('happy', () => {
  const newValue = 8
  const index = 1
  const result = update(
    index, newValue, list
  )
  const curriedResult = update(index, newValue)(list)
  const tripleCurriedResult = update(index)(newValue)(list)

  const expected = [ 1, 8, 3 ]
  expect(result).toEqual(expected)
  expect(curriedResult).toEqual(expected)
  expect(tripleCurriedResult).toEqual(expected)
})

test('list has no such index', () => {
  const newValue = 8
  const index = 10
  const result = update(
    index, newValue, list
  )

  expect(result).toEqual(list)
})

---------------

updateObject

updateObject<Output>(rules: readonly (readonly [string, any])[], input: object): Output

Very similar to R.assocPath but it applies list of updates instead of only a single update.

It return a copy of obj input with changed properties according to rules input.

Each instance of rules is a tuple of object path and the new value for this path. If such object path does not exist, then such object path is created.

As it uses R.path underneath, object path can be either string or array of strings(in Typescript object path can be only a string).

const obj = {
  a: {b: 1},
  foo: {bar: 10},
}
const rules = [
  ['a.b', 2],
  ['foo.bar', 20],
  ['q.z', 300],
]
const result = R.updateObject(rules, obj)

const expected = {
  a: {b: 2},
  foo: {bar: 20},
  q: {z: 300},
}
// => `result` is equal to `expected`

Try this R.updateObject example in Rambda REPL

R.updateObject source
import { assocPath } from './assocPath'

export function updateObject(rules, obj){
  if (arguments.length === 1) return _obj => updateObject(rules, _obj)

  let clone = { ...obj } /*?.*/

  rules.forEach(([ objectPath, newValue ]) => {
    clone = assocPath(
      objectPath, newValue, clone
    )
  })

  return clone
}
Tests
import { updateObject } from './updateObject'

const obj = {
  a   : { b : 1 },
  foo : { bar : 10 },
}
const rules = [
  [ 'a.b', 2 ],
  [ 'foo.bar', 20 ],
  [ 'q.z', 300 ],
]
const expected = {
  a   : { b : 2 },
  foo : { bar : 20 },
  q   : { z : 300 },
}

test('happy', () => {
  const result = updateObject(rules, obj)
  expect(result).toEqual(expected)
})

test('curried', () => {
  const result = updateObject(rules)(obj)
  expect(result).toEqual(expected)
})

---------------

values

values<T extends object, K extends keyof T>(obj: T): readonly T[K][]

With correct input, this is nothing more than Object.values(obj). If obj is not an object, then it returns an empty array.

const obj = {a:1, b:2}

R.values(obj)
// => [1, 2]

Try this R.values example in Rambda REPL

R.values source
import { type } from './type'

export function values(obj){
  if (type(obj) !== 'Object') return []

  return Object.values(obj)
}
Tests
import { values } from './values'

test('happy', () => {
  expect(values({
    a : 1,
    b : 2,
    c : 3,
  })).toEqual([ 1, 2, 3 ])
})

test('with bad input', () => {
  expect(values(null)).toEqual([])
  expect(values(undefined)).toEqual([])
  expect(values(55)).toEqual([])
  expect(values('foo')).toEqual([])
  expect(values(true)).toEqual([])
  expect(values(false)).toEqual([])
  expect(values(NaN)).toEqual([])
  expect(values(Infinity)).toEqual([])
  expect(values([])).toEqual([])
})

---------------

view

view<T, U>(lens: Lens): (target: T) => U

It returns the value of lens focus over target object.

const lens = R.lensProp('x')

R.view(lens, {x: 1, y: 2}) // => 1
R.view(lens, {x: 4, y: 2}) // => 4

Try this R.view example in Rambda REPL

R.view source
const Const = x => ({
  x,
  map : fn => Const(x),
})

export function view(lens, target){
  if (arguments.length === 1) return _target => view(lens, _target)

  return lens(Const)(target).x
}
Tests
import { assoc } from './assoc'
import { lens } from './lens'
import { prop } from './prop'
import { view } from './view'

const testObject = { foo : 'Led Zeppelin' }
const assocLens = lens(prop('foo'), assoc('foo'))

test('happy', () => {
  expect(view(assocLens, testObject)).toEqual('Led Zeppelin')
})

---------------

viewOr

viewOr<Input, Output>(fallback: Output, lens: Lens, input: Input): Output

A combination between R.defaultTo and `R.view.

const lens = R.lensProp('a');
const input = {a: 'foo'}
const fallbackInput = {b: 'bar'}
const fallback = 'FALLBACK'

const result = [
  R.viewOr(fallback, lens, input),
  R.viewOr(fallback, lens, fallbackInput)
]
// => ['foo', 'FALLBACK']

Try this R.viewOr example in Rambda REPL

R.viewOr source
import { curry } from './curry'
import { defaultTo } from './defaultTo'
import { view } from './view'

function viewOrFn(
  fallback, lens, input
){
  return defaultTo(fallback, view(lens, input))
}

export const viewOr = curry(viewOrFn)
Tests
import { lensProp } from './lensProp'
import { viewOr } from './viewOr'

const lens = lensProp('a')
const input = { a : 'foo' }
const fallbackInput = { b : 'bar' }
const fallback = 'FALLBACK'

test('happy', () => {
  const result = viewOr(
    fallback, lens, fallbackInput
  )
  expect(result).toBe(fallback)
})

test('curried', () => {
  const result = viewOr(fallback, lens)(input)
  expect(result).toBe('foo')
})

---------------

wait

wait<T>(fn: Promise<T>): Promise<readonly [T, Error|undefined]>

It provides Golang-like interface for handling promises.

const [result, err] = await R.wait(R.delay(1000))
// => err is undefined
// => result is `RAMBDAX_DELAY`

Try this R.wait example in Rambda REPL

R.wait source
export function wait(fn){
  return new Promise(resolve => {
    fn.then(result => resolve([ result, undefined ])).catch(e =>
      resolve([ undefined, e ]))
  })
}
Tests
import { wait } from './wait'

test('happy path', async () => {
  const fn = x => Promise.resolve(x + 1)
  const [ result, err ] = await wait(fn(1))

  expect(result).toBe(2)
  expect(err).toBeUndefined()
})

test('when promise is rejected', async () => {
  const fn = x => Promise.reject(Error('foo'))
  const [ result, err ] = await wait(fn(1))

  expect(result).toBeUndefined()
  expect(err).toEqual(Error('foo'))
})

---------------

waitFor

waitFor(
  waitForTrueCondition: () => boolean,
  howLong: number,
  loops?: number
): () => Promise<boolean>

It returns true, if condition returns true within howLong milisececonds time period.

The method accepts an optional third argument loops(default to 10), which is the number of times waitForTrueCondition will be evaluated for howLong period. Once this function returns a value different from false, this value will be the final result.

Otherwise, R.waitFor will return false.

const howLong = 1000
let counter = 0
const waitForTrueCondition = async x => {
  await R.delay(100)
  counter = counter + x

  return counter > 10
}

const result = await R.waitFor(waitForTrueCondition, howLong)(2)
// => true

Try this R.waitFor example in Rambda REPL

R.waitFor source
import { delay } from './delay'
import { range } from './range'
import { type } from './type'

export function waitFor(
  condition, howLong, loops = 10
){
  const typeCondition = type(condition)

  const passPromise = typeCondition === 'Async'
  const passFunction = typeCondition === 'Function'
  const interval = Math.floor(howLong / loops)

  if (!(passPromise || passFunction)){
    throw new Error('R.waitFor')
  }

  return async (...inputs) => {
    for (const _ of range(0, loops)){
      const resultCondition = await condition(...inputs)

      if (resultCondition === false){
        await delay(interval)
      } else {
        return resultCondition
      }
    }

    return false
  }
}
Tests
import { delay } from './delay'
import { waitFor } from './waitFor'

const howLong = 1000

test('true', async () => {
  let counter = 0
  const condition = x => {
    counter++

    return counter > x
  }

  const result = await waitFor(condition, howLong)(6)
  expect(result).toEqual(true)
})

test('false', async () => {
  let counter = 0
  const condition = x => {
    counter++

    return counter > x
  }

  const result = await waitFor(condition, howLong)(12)
  expect(result).toEqual(false)
})

test('async condition | true', async () => {
  let counter = 0
  const condition = async x => {
    counter++
    await delay(10)

    return counter > x
  }

  const result = await waitFor(condition, howLong)(6)
  expect(result).toEqual(true)
})

test('async condition | false', async () => {
  let counter = 0
  const condition = async x => {
    counter++
    await delay(10)

    return counter > x
  }

  const result = await waitFor(condition, howLong)(12)
  expect(result).toEqual(false)
})

test('throws when fn is not function', () => {
  const fn = 'foo'

  expect(() => waitFor(fn, howLong)()).toThrowWithMessage(Error, 'R.waitFor')
})

---------------

when

when<T, U>(predicate: (x: T) => boolean, whenTrueFn: (a: T) => U, input: T): T | U
R.when source
import { curry } from './curry'

function whenFn(
  predicate, whenTrueFn, input
){
  if (!predicate(input)) return input

  return whenTrueFn(input)
}

export const when = curry(whenFn)
Tests
import { add } from './add'
import { when } from './when'

const predicate = x => typeof x === 'number'

test('happy', () => {
  const fn = when(predicate, add(11))
  expect(fn(11)).toBe(22)
  expect(fn('foo')).toBe('foo')
})

---------------

where

where<T, U>(conditions: T, input: U): boolean

It returns true if all each property in conditions returns true when applied to corresponding property in input object.

R.equals is used to determine equality.

const condition = R.where({
  a : x => typeof x === "string",
  b : x => x === 4
})
const input = {
  a : "foo",
  b : 4,
  c : 11,
}

const result = condition(input) 
// => true

Try this R.where example in Rambda REPL

R.where source
export function where(conditions, input){
  if (input === undefined){
    return _input => where(conditions, _input)
  }
  let flag = true
  for (const prop in conditions){
    const result = conditions[ prop ](input[ prop ])
    if (flag && result === false){
      flag = false
    }
  }

  return flag
}
Tests
import { equals } from './equals'
import { where } from './where'

test('when true', () => {
  const predicate = where({
    a : equals('foo'),
    b : equals('bar'),
  })
  expect(predicate({
    a : 'foo',
    b : 'bar',
    x : 11,
    y : 19,
  })).toEqual(true)
})

test('when false', () => {
  const predicate = where({
    a : equals('foo'),
    b : equals('baz'),
  })
  expect(predicate({
    a : 'foo',
    b : 'bar',
    x : 11,
    y : 19,
  })).toEqual(false)
})

---------------

whereEq

whereEq<T, U>(condition: T, input: U): boolean

It will return true if all of input object fully or partially include rule object.

R.equals is used to determine equality.

const condition = { a : { b : 1 } }
const input = {
  a : { b : 1 },
  c : 2
}

const result = whereEq(condition, input)
// => true

Try this R.whereEq example in Rambda REPL

R.whereEq source
import { equals } from './equals'
import { filter } from './filter'

export function whereEq(condition, input){
  if (arguments.length === 1){
    return _input => whereEq(condition, _input)
  }

  const result = filter((conditionValue, conditionProp) =>
    equals(conditionValue, input[ conditionProp ]),
  condition)

  return Object.keys(result).length === Object.keys(condition).length
}
Tests
import { whereEq } from './whereEq'

test('when true', () => {
  const condition = { a : 1 }
  const input = {
    a : 1,
    b : 2,
  }

  const result = whereEq(condition, input)
  const expectedResult = true

  expect(result).toEqual(expectedResult)
})

test('when false', () => {
  const condition = { a : 1 }
  const input = { b : 2 }

  const result = whereEq(condition, input)
  const expectedResult = false

  expect(result).toEqual(expectedResult)
})

test('with nested object', () => {
  const condition = { a : { b : 1 } }
  const input = {
    a : { b : 1 },
    c : 2,
  }

  const result = whereEq(condition)(input)
  const expectedResult = true

  expect(result).toEqual(expectedResult)
})

test('with wrong input', () => {
  const condition = { a : { b : 1 } }

  expect(() => whereEq(condition, null)).toThrowWithMessage(TypeError,
    'Cannot read property \'a\' of null')
})

---------------

without

without<T>(matchAgainst: readonly T[], source: readonly T[]): readonly T[]

It will return a new array, based on all members of source list that are not part of matchAgainst list.

R.equals is used to determine equality.

const source = [1, 2, 3, 4]
const matchAgainst = [2, 3]

const result = R.without(matchAgainst, source)
// => [1, 4]

Try this R.without example in Rambda REPL

R.without source
import { includesArray } from './includes'
import { reduce } from './reduce'

export function without(matchAgainst, source){
  if (source === undefined){
    return _source => without(matchAgainst, _source)
  }

  return reduce(
    (prev, current) =>
    includesArray(current, matchAgainst) ? prev : prev.concat(current),
    [],
    source
  )
}
Tests
import { without } from './without'

test('should return a new list without values in the first argument ', () => {
  const itemsToOmit = [ 'A', 'B', 'C' ]
  const collection = [ 'A', 'B', 'C', 'D', 'E', 'F' ]

  expect(without(itemsToOmit, collection)).toEqual([ 'D', 'E', 'F' ])
  expect(without(itemsToOmit)(collection)).toEqual([ 'D', 'E', 'F' ])
})

test('ramda bug', () => {
  expect(
    without("0:1", ["0", "0:1"])
  ).toEqual(['0:1'])
})

test('ramda test', () => {
  expect(without([ 1, 2 ])([ 1, 2, 1, 3, 4 ])).toEqual([ 3, 4 ])
})

---------------

xnor

xnor(x: boolean, y: boolean): boolean

Logical XNOR

const result = [
  R.xnor(1, 0),
  R.xnor(0, 1),
  R.xnor(0, 0),
  R.xnor(1, 1),
]
// => [true, false, false, true]

Try this R.xnor example in Rambda REPL

R.xnor source
export function xnor(x, y){
  if (arguments.length === 1){
    return _y => xnor(x, _y)
  }

  return Boolean(x && y || !x && !y)
}
Tests
import { xnor } from './xnor'

test('when true', () => {
  expect(xnor(1, 1)).toBeTrue()
  expect(xnor(0)(0)).toBeTrue()
})

test('when false', () => {
  expect(xnor(0, 1)).toBeFalse()
})

---------------

xor

xor(x: boolean, y: boolean): boolean

Logical XOR

const result = [
  xor(true, true),
  xor(false, false),
  xor(false, true),
]
// => [false, false, true]

Try this R.xor example in Rambda REPL

R.xor source
export function xor(a, b){
  if (arguments.length === 1) return _b => xor(a, _b)

  return Boolean(a) && !b || Boolean(b) && !a
}
Tests
import { xor } from './xor'

test('compares two values with exclusive or', () => {
  expect(xor(true, true)).toEqual(false)
  expect(xor(true, false)).toEqual(true)
  expect(xor(false, true)).toEqual(true)
  expect(xor(false, false)).toEqual(false)
})

test('when both values are truthy, it should return false', () => {
  expect(xor(true, 'foo')).toEqual(false)
  expect(xor(42, true)).toEqual(false)
  expect(xor('foo', 42)).toEqual(false)
  expect(xor({}, true)).toEqual(false)
  expect(xor(true, [])).toEqual(false)
  expect(xor([], {})).toEqual(false)
  expect(xor(new Date(), true)).toEqual(false)
  expect(xor(true, Infinity)).toEqual(false)
  expect(xor(Infinity, new Date())).toEqual(false)
})

test('when both values are falsy, it should return false', () => {
  expect(xor(null, false)).toEqual(false)
  expect(xor(false, undefined)).toEqual(false)
  expect(xor(undefined, null)).toEqual(false)
  expect(xor(0, false)).toEqual(false)
  expect(xor(false, NaN)).toEqual(false)
  expect(xor(NaN, 0)).toEqual(false)
  expect(xor('', false)).toEqual(false)
})

test('when one argument is truthy and the other is falsy, it should return true', () => {
  expect(xor('foo', null)).toEqual(true)
  expect(xor(null, 'foo')).toEqual(true)
  expect(xor(undefined, 42)).toEqual(true)
  expect(xor(42, undefined)).toEqual(true)
  expect(xor(Infinity, NaN)).toEqual(true)
  expect(xor(NaN, Infinity)).toEqual(true)
  expect(xor({}, '')).toEqual(true)
  expect(xor('', {})).toEqual(true)
  expect(xor(new Date(), 0)).toEqual(true)
  expect(xor(0, new Date())).toEqual(true)
  expect(xor([], null)).toEqual(true)
  expect(xor(undefined, [])).toEqual(true)
})

---------------

zip

zip<K, V>(x: readonly K[], y: readonly V[]): readonly KeyValuePair<K, V>[]

It will return a new array containing tuples of equally positions items from both x and y lists.

The returned list will be truncated to match the length of the shortest supplied list.

const x = [1, 2]
const y = ['A', 'B']
R.zip(x, y)
// => [[1, 'A'], [2, 'B']]

// truncates to shortest list
R.zip([...x, 3], ['A', 'B'])
// => [[1, 'A'], [2, 'B']]

Try this R.zip example in Rambda REPL

R.zip source
export function zip(left, right){
  if (arguments.length === 1) return _right => zip(left, _right)

  const result = []
  const length = Math.min(left.length, right.length)

  for (let i = 0; i < length; i++){
    result[ i ] = [ left[ i ], right[ i ] ]
  }

  return result
}
Tests
import { zip } from './zip'

const array1 = [ 1, 2, 3 ]
const array2 = [ 'A', 'B', 'C' ]

test('should return an array', () => {
  const actual = zip(array1)(array2)
  expect(actual).toBeInstanceOf(Array)
})

test('should return and array or tuples', () => {
  const expected = [
    [ 1, 'A' ],
    [ 2, 'B' ],
    [ 3, 'C' ],
  ]
  const actual = zip(array1, array2)
  expect(actual).toEqual(expected)
})

test('should truncate result to length of shorted input list', () => {
  const expectedA = [
    [ 1, 'A' ],
    [ 2, 'B' ],
  ]
  const actualA = zip([ 1, 2 ], array2)
  expect(actualA).toEqual(expectedA)

  const expectedB = [
    [ 1, 'A' ],
    [ 2, 'B' ],
  ]
  const actualB = zip(array1, [ 'A', 'B' ])
  expect(actualB).toEqual(expectedB)
})

---------------

zipObj

zipObj<T, K extends string>(keys: readonly K[], values: readonly T[]): { readonly [P in K]: T }

It will return a new object with keys of keys array and values of values array.

const keys = ['a', 'b', 'c']

R.zipObj(keys, [1, 2, 3])
// => {a: 1, b: 2, c: 3}

// truncates to shortest list
R.zipObj(keys, [1, 2])
// => {a: 1, b: 2}

Try this R.zipObj example in Rambda REPL

R.zipObj source
import { take } from './take'

export function zipObj(keys, values){
  if (arguments.length === 1) return yHolder => zipObj(keys, yHolder)

  return take(values.length, keys).reduce((
    prev, xInstance, i
  ) => {
    prev[ xInstance ] = values[ i ]

    return prev
  }, {})
}
Tests
import { equals } from './equals'
import { zipObj } from './zipObj'

test('zipObj', () => {
  expect(zipObj([ 'a', 'b', 'c' ], [ 1, 2, 3 ])).toEqual({
    a : 1,
    b : 2,
    c : 3,
  })
})

test('0', () => {
  expect(zipObj([ 'a', 'b' ])([ 1, 2, 3 ])).toEqual({
    a : 1,
    b : 2,
  })
})

test('1', () => {
  expect(zipObj([ 'a', 'b', 'c' ])([ 1, 2 ])).toEqual({
    a : 1,
    b : 2,
  })
})

test('ignore extra keys', () => {
  const result = zipObj([ 'a', 'b', 'c', 'd', 'e', 'f' ], [ 1, 2, 3 ])
  const expected = {
    a : 1,
    b : 2,
    c : 3,
  }

  expect(equals(result, expected)).toBeTrue()
})

---------------

zipWith

zipWith<T, U, TResult>(fn: (x: T, y: U) => TResult, list1: readonly T[], list2: readonly U[]): readonly TResult[]
const list1 = [ 10, 20, 30, 40 ]
const list2 = [ 100, 200 ]

const result = R.zipWith(
  R.add, list1, list2
)
// => [110, 220]

Try this R.zipWith example in Rambda REPL

R.zipWith source
import { curry } from './curry'
import { take } from './take'

function zipWithFn(
  fn, x, y
){
  return take(x.length > y.length ? y.length : x.length,
    x).map((xInstance, i) => fn(xInstance, y[ i ]))
}

export const zipWith = curry(zipWithFn)
Tests
import { add } from './add'
import { zipWith } from './zipWith'

const list1 = [ 1, 2, 3 ]
const list2 = [ 10, 20, 30, 40 ]
const list3 = [ 100, 200 ]

test('when second list is shorter', () => {
  const result = zipWith(
    add, list1, list3
  )
  expect(result).toEqual([ 101, 202 ])
})

test('when second list is longer', () => {
  const result = zipWith(
    add, list1, list2
  )
  expect(result).toEqual([ 11, 22, 33 ])
})

---------------

❯ CHANGELOG

WIP 7.3.0

  • Change R.piped typings to mimic that of R.pipe. Main difference is that R.pipe is focused on unary functions.

  • Wrong logic where R.without use R.includes while it should use the array version of R.includes

  • Use uglify plugin for UMD bundle

7.2.0

  • Approve PR #61 - fix wrong R.isValid typings

  • R.produceAsync returns promise even if all rules are synchronous.

  • R.defaultTo no longer accepts infinite inputs, thus it follows Ramda implementation.

  • R.equals supports equality of functions.

  • R.pipe doesn't use R.compose.

  • Close Issue #561 - export several internal TS interfaces and types

  • Add CHANGELOG.md file in release files list

7.1.0

  • Add R.tryCatchAsync

  • Add R.xnor

  • R.equals supports equality of functions.

  • Close Issue #559 - improve R.propOr typings

  • Close Issue #560 - apply immutable lint to Typescript definitions

  • Close Issue #553 - fix problem with curried typings of R.prop

  • Fix wrong R.last typing

  • Upgrade all rollup related dependencies

  • R.type supports Symbol just like Ramda.

  • Remove file extension in main property in package.json in order to allow experimental-modules. See also this Ramda's PR - https://github.com/ramda/ramda/pull/2678/files

  • Import R.indexBy/R.when/R.zipObj/R.propEq/R.complement changes from recent @types/ramda release.

  • R.tryCatch stop supporting asynchronous functions; the previous behaviour is exported to Rambdax as R.tryCatchAsync

7.0.1

  • Fix missing Evolved declaration in Typescript definition

7.0.0

  • Rename R.produce to R.produceAsync

  • Add R.produce which is synchronous version of R.produceAsync

  • Stop supporting expression inside template's props. Also, spaces are no longer allowed between {{ and }}, i.e. R.interpolate('{{ foo }}', x) should be R.interpolate('{{foo}}', x).

  • Add typings for R.takeWhile when iterable is a string

  • Add R.takeLastWhile

  • Add R.dropWhile

  • Add R.eqProps

  • Add R.dropLastWhile

  • Add R.dropRepeats

  • Add R.dropRepeatsWith

  • Add R.evolve

6.2.0

  • R.switcher accepts undefined as valid input

  • Add R.props

  • Add R.zipWith

  • Add R.splitAt

  • Add R.splitWhen

  • Close Issue #547 - restore readonly declaration in Rambda Typescript definitions.

  • R.append/R.prepend now work only with arrays just like Ramda. Previous behaviour was for them to work with both arrays and strings.

  • Sync R.pluck typings with @types/ramda as there was a tiny difference.

6.1.0

  • Add R.mapIndexed

  • Add R.filterIndexed

  • Add R.forEachIndexed

  • Fix R.and wrong definition, because the function doesn't convert the result to boolean. This introduce another difference with @types/ramda.

  • Add R.once

  • Add R.or

6.0.0

  • Breaking change - R.map/R.filter/R.reject/R.forEach/R.partition doesn't pass index as second argument to the predicate, when looping over arrays. The old behaviour of map, filter and forEach can be found in Rambdax methods R.mapIndexed, R.filterIndexed and R.forEachIndexed(introduced in version 6.1.0).

  • Breaking change - R.all/R.none/R.any/R.find/R.findLast/R.findIndex/R.findLastIndex doesn't pass index as second argument to the predicate.

  • Add R.applyDiff method

  • Change R.assocPath typings so the user can explicitly sets type of the new object

  • Typings of R.assoc match its @types/ramda counterpart.

  • Simplify R.forEach typings

  • Remove ReadonlyArray<T> pattern from Typescript definitions - not enough value for the noise it adds.

  • Fix typing of R.reject as it wrongly declares that with object, it pass property to predicate.

5.1.0

  • Add R.takeUntil method

  • Fix wrong R.takeWhile

5.0.0

  • Deprecate R.change method - it does too much; partially replaced with R.updateObject.

  • Deprecate R.compact method - vague use case; R.filter does the same job.

  • R.produce always returns a promise

  • Add R.updateObject method

  • Add R.takeWhile method

  • Add R.viewOr method

  • Add R.pipeAsync method

  • Add R.removeIndex method

  • Add R.excludes method

  • R.includes throws on wrong input, i.e. R.includes(1, null)

  • Close Issue #524 - R.assocPath wrong logic when number is used in array path input.

  • R.mapToObjectAsync supports currying

  • R.mapAsyncLimit supports currying

  • Fix R.mapAsync to pass property to iterator, when input is an object.

  • Fix currying for several async methods - R.tapAsync, R.produce, R.filterAsync *(extend typings)

4.2.0

  • Add R.move method

  • Add R.union method

  • Add R.lensSatisfies method

  • Add R.mapKeys method

  • Add R.sortByPath method

  • Add R.sortByProps method

  • Close Issue #519 - ts-toolbelt needs other type of export with --isolatedModules flag

4.1.0

  • R.template is renamed to R.interpolate

  • R.equals now supports negative zero just like Ramda.equals

  • Add R.replaceAll method

  • Add R.lensEq method

4.0.1

Forgot to export R.of because of wrong marker in files/index.d.ts

4.0.0

Deprecate the following methods:

  • R.promiseAllObject - because R.produce serves the same purpose
  • R.composed - because R.piped makes more sense, when we want to pass the input at the start of the function
  • R.defaultToStrict - confusing logic
  • R.findInObject - overestimated importance
  • R.headObject - overestimated importance
  • R.includesType - overestimated importance
  • R.inject - confusing logic
  • R.isAttach - confusing logic
  • R.mergeRight - overestimated importance
  • R.opposite - overestimated importance
  • R.otherwise - overestimated importance
  • R.pushUniq - overestimated importance
  • R.resolve - overestimated importance
  • R.s - overestimated importance
  • R.toggle - overestimated importance
  • R.uuid - not suitable
  • R.whenAsync - overestimated importance

Move the following methods to Rambda and change their logic to match Ramda implementation:

  • R.hasPath
  • R.unless
  • R.pathEq
  • R.tryCatch
  • R.where
  • R.whereEq

Also these changes:

  • R.flatMap - renamed to R.chain and moved to Rambda

  • R.ifElseAsync - accept any number of arguments for the returned function

  • R.produce, R.filterAsync, R.debounce, R.throttle - fix typings

  • R.mapAsyncLimit - drop support for curring and therefore for usage with R.composeAsync

  • Improve R.ok throwed error message

  • R.ok returns undefined instead of true when validation passes.

  • R.mergeDeep is renamed to R.mergeDeepLeft

  • Add R.pipeAsync

  • Take R.partialCurry from Rambda as it is deprecated there

3.7.0

Sync with Rambda

Add R.lens

Add R.lensIndex

Add R.lensPath

Add R.lensProp

Add R.over

Add R.set

Add R.view

Add R.paths

Add R.xor

Add R.cond

3.6.0

  • Add R.mapAsyncLimit

  • Add R.toggle, match Ramda upcoming method specification

  • Add R.isValidAsync

  • Extend R.template without introducing breaking change

3.5.0 Sync with Rambda - add methods descriptions to Typescript definitions

3.4.0 Sync with Rambda and close Issue #42

3.3.0 Fix R.sortObject typing

3.3.0 Add R.filterAsync and R.sortObject methods

3.2.0 R.uuid accept second argument in order to return string only uuid

3.1.0 Dynamic set of exports lead to adding previously ommited Rambda exports such as R.identical

3.0.3 Sync with Rambda - new functionality of R.isEmpty

3.0.2 Add typings for R.mapToObject

3.0.1 Fix typings

3.0.0 Breaking change as Rambda also has breaking changes

Read more about it in Rambda changelog

Also with this versions, typings tests are provided and several definitions are changed.

  • R.anyTrue, R.anyFalse, R.allTrue, R.allFalse use internal isTruthy and isFalsy methods. Empty array and object with zero length are considered falsy.

  • Deprecate R.contains

  • Deprecate R.defaultToWhen

  • Moved R.runTests to helpers repo

2.17.0 Change in R.runTests logic. It will be removed from Rambdax to helpers repo.

2.16.0 Restore R.runTests but without documentation

  • export getEvaluations, getPositiveEvaluation, getNegativeEvaluation in the context of R.runTests

2.15.0 Several changes

  • Typescript definitions have been updated and typings tests are introduced

  • R.mapAsync and R.mapFastAsync pass index as second argument

2.14.1 Restore R.contains

2.14.0 Several changes:

  • R.inject accept before flag as fourth argument

  • Remove R.includesAny

  • Improve typing of R.partition

  • R.nextIndex and R.prevIndex work also with number as second argument

2.13.1 Deprecate R.log and R.runTests

2.12.3 Add 'dist' directory to files

2.12.2 Add R.mapToObject typings

2.12.0 Sync with Rambda

2.11.1 Fix R.waitFor

2.11.0 Add R.toDecimal

2.10.2 Fix issue 32

2.10.0 deprecate R._

2.9.1 R.fromPairs/toPairs typing

2.9.0 npm doesn't update version on their site

2.8.2 R.map typing

2.8.0 Sync with Rambda | no need for create types script

2.7.0 Add R.prevIndex

2.6.2 Sync with Rambda

2.6.0 R.log depends on RAMBDAX_LOG

2.5.0 Rambda's partial

2.4.0 Add R.uuid

2.3.0 R._ parse to constant case

This introduce breaking change for ie11 as noted in issue 31 which is fixed with 2.10.0 which deprecates this method

2.2.1 Add R.log, R.logInit and R.logHolder

2.1.0 Add R._

2.0.0 Add R.toggle

1.9.0 Add R.pushUniq

1.8.2 No need for sourcemaps

1.8.1 Fix building with regeneratorRuntime

1.8.0 Upgrade to new major Rollup release

  • Restore R.headObject

  • Add R.hasPath method

1.7.2 R.memoize contains dev console.logs

1.7.1 Forgot to build types

1.7.0 Rename R.then to R.resolve because of Ramda issue with R.then(they rename it to R.andThen)

  • Add R.isFalsy, R.nextIndex and R.mergeDeep

1.6.3 Forgot to export R.unless(credit to @mobily for the PR)

1.6.0 Restore R.compact method

1.5.6 R.maybe accepts also anonymous functions as second and third argument

1.5.5 Add R.maybe method

  • Fix errors caugth by DeepScan service
  • Fix Typescript definitions for R.then and R.otherwise
  • R.change increase nesting level to 4

1.4.1 R.isValid didn't work with Number prototype

1.4.0 Add multiple methods

  • Add R.defaultToStrict

  • Add R.defaultToWhen

  • Add R.whereEq

  • Add R.partition

  • Add R.negate is renamed to R.opposite

  • Add R.then

  • Add R.otherwise

  • R.isValid accepts prototypes as rules, i.e. schema = {a: String}

  • The prevoious point leads to the same change applied to the methods depending on R.isValid, i.e. R.ok, R.pass and R.isAttach

    1.3.0 Add R.unless

  • R.when accepts both function and value for whenTrue argument. The same is valid for R.unless

  • export R.negate which is the same as R.complement

    1.2.0 Export src folder

    1.1.0 Restore promiseAllObject and flatMap

    1.0.1 Fix typings

    1.0.0 Deprecate the following methods:

  • compact

  • evolve

  • flatMap

  • greater

  • intersection

  • less

  • omitBy

  • pickBy

  • promiseAllObject

  • promiseAllSecure

  • rangeBy

Also pass deprecation of addIndex from [email protected]

0.24.0 add R.pipedAsync, replace R.multiline with R.glue, remove R.validate

0.23.0 Add R.count

0.22.0 Add R.includesAny

0.21.0 Add R.includesType

0.20.1 R.pass and R.ok work with single schema.

0.20.0 Add R.pathEq

0.19.0 Add R.wait, expose already complete R.waitFor

0.18.0 AddR.anyType and R.allType

0.17.0 Rename R.is to R.pass and restore R.is original functionality.

0.16.0 getter, setter, reset methods

0.15.3 No more prepublish script

0.15.2 curry in remove

Last version with lib folder exposed

---------------

❯ Additional info

Most influential contributors

  • @farwayer - improving performance in R.find, R.filter; give the idea how to make benchmarks more reliable;

  • @thejohnfreeman - add R.assoc, R.chain;

  • @helmuthdu - add R.clone; help improve code style;

  • @jpgorman - add R.zip, R.reject, R.without, R.addIndex;

  • @ku8ar - add R.slice, R.propOr, R.identical, R.propIs and several math related methods; introduce the idea to display missing Ramda methods;

  • @romgrk - add R.groupBy, R.indexBy, R.findLast, R.findLastIndex;

  • @squidfunk - add R.assocPath, R.symmetricDifference, R.difference, R.intersperse;

  • @synthet1c - add all lenses methods; add R.applySpec, R.converge;

  • @vlad-zhukov - help with configuring Rollup, Babel; change export file to use ES module exports;

Rambda references

Links to Rambda

Deprecated from Used by section

Releases

Rambda's releases before 6.4.0 were used mostly for testing purposes.

---------------

My other libraries

Niketa theme

Collection of 9 light VSCode themes

Niketa dark theme

Collection of 9 dark VSCode themes

String-fn

String utility library

Useful Javascript libraries

Large collection of JavaScript,Typescript and Angular related repos links

Run-fn

CLI commands for lint JS/TS files, commit git changes and upgrade of dependencies
Comments
  • Add arbitrary expression parsing in curly braces …

    Add arbitrary expression parsing in curly braces …

    I found myself wanting to push the limits of what template can do, but I didn't want to pull in a larger templating library, so I figured we could just allow the template function to handle any expression inside it.

    opened by brandonpittman 14
  • Provide public ES modules

    Provide public ES modules

    /es package entry point is conventional in modern libraries.

    The suggestion is to provide rambdax/es imports in ES5 and ES module format , similarly to how it was done with ramda/es. Notice that Ramda modules are transpiled to ES5 to skip transpilation on bundling, and there is index reimporting file.

    Unlike rambda/lib, rambdax doesn't contain separately built CJS modules.

    Currently it's possible to import modules as rambdax/src/debounce. The problem is that src is ES2017 and isn't published to NPM, it's also not a good habit to rely on internal package structure.

    opened by bisubus 13
  • Index on map breaks currying

    Index on map breaks currying

    I might be missing something, but I ran into the issue that map calls the provided function with both the element and the index. This prevents me from doing something like

    const isWeb = R.piped(["http://", "https://"], R.map(R.startsWith), R.anyPass);
    const result = isWeb("http://example.com");
    

    As R.startsWith will receive "http://", 0. And try to call 0.startsWith.

    Of course I can work around:

    const isWeb = R.piped(["http://", "https://"], R.map(R.pipe(R.always, R.startsWith)), R.anyPass);
    const result = isWeb("http://example.com");
    

    But I lose some of the satisfaction I normally get from this kind of functional programming.

    Do you have suggestions or better workarounds?

    opened by Boudewijn26 12
  • fix: error TS2314: Generic type 'MapInterface<T>'

    fix: error TS2314: Generic type 'MapInterface'

    Fixing typings error:

    PS C:\_GITHUB\rambdax> npx typescript .\index.d.ts
    npx: installed 1 in 3.636s
    index.d.ts:633:45 - error TS2314: Generic type 'MapInterface<T>' requires 1 type argument(s).
    
    633     map<In, Out>(fn: MapFunction<In, Out>): MapInterface<In, Out>
                                                    ~~~~~~~~~~~~~~~~~~~~~
    
    
    Found 1 error.
    
    opened by darosh 9
  • Broken examples

    Broken examples

    Examples for async functions

    • mapAsync
    • mapFastAsync
    • debounce
    • pipeAsync do not run as expected. Oftentimes it's just an async that is missing.

    Also found from running examples for many other of the functions (not all):

    • Something deeper is wrong with isType for 'Async' type.
    • partialCurry is missing R. in the example. Same for piped, replaceAll, splitAt, symmetricDifference & remove.
    • path seems to return null instead of undefined. Same for prop.
    • pathOr example is broken with syntax errors.
    • Logical error in example for paths, and same problem as with path as well.
      • Same for props.
    • pick does not match expected.
    • Neither does pickAll.
    • Syntax error for pluck , split, set, reject, subtract & produce.
    • Logic error for propIs & slice.
    • removeIndex function is broken?
    • sortByProps is using sortByPath!

    On another note: sort and sortBy difference should be better clarified I think.

    opened by whitelizard 8
  • fix: typescript errors when the isolatedModules flag is provided

    fix: typescript errors when the isolatedModules flag is provided

    When I upgrade rambdax to 4.0, I got this:

    ./node_modules/rambdax/_ts-toolbelt/src/Any/_api.ts:3:9
    Type error: Re-exporting a type when the '--isolatedModules' flag is provided requires using 'export type'.
    
      1 | /** @ignore *//** */
      2 |
    > 3 | export {Cast} from './Cast'
        |         ^
      4 | export {Clean} from './Clean'
      5 | export {Compute} from './Compute'
      6 | export {Contains} from './Contains'
    

    So I created the PR to fix it, thanks.

    opened by firede 6
  • Feature Request: throttle/debounce type definitions

    Feature Request: throttle/debounce type definitions

    It would be great if the definition were more like this:

    throttle<T>(fn: T, time: number): T;

    Casting the first argument and return type as Function helps very little.

    opened by kevinbeal 6
  • Importing rambdax into React with typescript breaks the app.

    Importing rambdax into React with typescript breaks the app.

    Hi,

    I'm trying to learn to use rambdax and I'm importing it in an Create React App starter kit with a TypeScript template. However I cannot proceed with rambdax usage because of this error:

    Failed to compile.
    
    .../training-app/node_modules/rambdax/_ts-toolbelt/src/List/Append.ts
    TypeScript error in .../training-app/node_modules/rambdax/_ts-toolbelt/src/List/Append.ts(13,6):
    A rest element must be last in a tuple type.  TS1256
    
        11 | */
        12 | export type Append<L extends List, A extends any> =
      > 13 |     [...L, A]
           |      ^
        14 |
    

    My code looks like this:

    import React from 'react';
    ... 
    import { filter, range } from 'rambdax'
    ...
    

    The line with imports breaks the app. I cannot figure out why. I'm brand new to rambdax so it might be and my miss-use.

    Thanks!

    opened by dusanbass 5
  • flatMap is not a function

    flatMap is not a function

    Hello.

    I'm using this in a Gatsby website (repo, file). Today, I updated to 4.0.1 and my build fails with the error flatMap is not a function (line 77). This worked fine till the 3.7.0. Has something changed? I read the changelog, but, didn't find anything useful.

    Thanks.

    opened by Hrishikesh-K 5
  • fix pluck documentation, argument order and broken repl link

    fix pluck documentation, argument order and broken repl link

    Hi Dejan,

    I realized second part of the pluck documentation is bit wrong.

    Repl link was also broken, fixed both. I only encoded rambda code for repl link, hope its the right way.

    opened by onurozkan 4
  • R.forEach 6.0.0

    R.forEach 6.0.0

    Hello.

    As it's mentioned in changelog:

    Breaking change - R.map/R.filter/R.reject/R.forEach/R.partition doesn't pass index as second argument to the predicate, when looping over arrays.

    What's the way to use it otherwise? I'm having a Gatsby website, and I'm using R.forEach and R.filter. If I upgrade to 6.0.0, my website doesn't build. So, if index isn't passed anymore, is there some alternative or something that's introduced or do I have to stick with 5.1.0?

    opened by Hrishikesh-K 4
  • uniqBy is different from ramda

    uniqBy is different from ramda

    I just stumbled on this...

    In ramda, R.uniqBy uses R.equals for comparisons. In rambdax, R.uniqBy uses a native Set to collapse.

    This means that R.uniqBy in rambdax doesn't work the same for array or object values.

    Now that I know, I can rework my code, but I just wanted to raise the issue in case you'd like to handle this somehow, whether by documentation or types or changing behavior.

    done waiting release 
    opened by agriffis 2
Owner
Dejan Toteff
Unnecessary complexity killed the cat
Dejan Toteff
Extended version of create-t3-app to make it even faster to start (or maybe slower)

create-T3-app with extra tools/config out of the box create-t3-app is one of the fastest and easiest way to scaffold fullstack app. create-t3-extended

A Riolly C 7 Jan 4, 2023
Extended carousel based on Bootstrap 5 using only vanilla js.

Description Extended Slider based on Bootstrap 5 using only carousel component and vanilla js. Requirements: Bootstrap 5 Installation npm i -D extende

Carlos Pozo 7 Jul 26, 2022
✏️ Extended Writer supercharges Kirby's built-in Writer field with useful marks, nodes and features you wish were built-in

Extended Writer Extended Writer supercharges Kirby's built-in Writer field and block with useful marks, nodes and features you wish were built-in. Fea

coralic 21 Nov 10, 2022
Extended magic-string with extra utilities

DEPRECATED. It has been ported back to magic-string >= 0.26.0 magic-string-extra Extended Rich-Harris/magic-string with extra utilities. Install npm i

Anthony Fu 130 Sep 8, 2022
Simple and Extensible Markdown Parser for Svelte, however its simplicity can be extended to any framework.

svelte-simple-markdown This is a fork of Simple-Markdown, modified to target Svelte, however due to separating the parsing and outputting steps, it ca

Dave Caruso 3 May 22, 2022
An extended table to integration with some of the most widely used CSS frameworks.

An extended table to integration with some of the most widely used CSS frameworks. (Supports Bootstrap, Semantic UI, Bulma, Material Design, Foundation, Vue.js)

文翼 11.4k Dec 20, 2022
A plugin that provides utilities for extended backgrounds and borders.

tailwindcss-full-bleed A plugin that provides utilities for extended backgrounds and borders. Demo Installation Install the plugin from npm: npm insta

Doğukan Çavuş 8 Dec 24, 2022
Obsidian plugin allowing for linking to a heading range, in the [[Page#HeaderA#HeaderB]] extended wikilink format.

Obsidian Link Heading Range Plugin This is a plugin for Obsidian (https://obsidian.md). It allows linking to a heading range, in the [[Page#HeaderA#He

Anthropologie Biblique 7 Nov 14, 2022
The Remix version of the fakebooks app demonstrated on https://remix.run. Check out the CRA version: https://github.com/kentcdodds/fakebooks-cra

Remix Fakebooks App This is a (very) simple implementation of the fakebooks mock app demonstrated on remix.run. There is no database, but there is an

Kent C. Dodds 61 Dec 22, 2022
GPL version of Javascript Gantt Chart

dhtmlxGantt Getting started | Features | Follow us | License | Useful links dhtmlxGantt is an open source JavaScript Gantt chart that helps you illust

null 952 Dec 29, 2022
Duet Date Picker is an open source version of Duet Design System’s accessible date picker. Try live example at https://duetds.github.io/date-picker/

Duet Date Picker Duet Date Picker is an open source version of Duet Design System’s accessible date picker. Duet Date Picker can be implemented and us

Duet Design System 1.6k Jan 6, 2023
A JavaScript, zero-dependency, super small version of IP2Location LITE country lookups.

ip3country This is a zero-dependency, super small, IP address to 2-letter country code lookup library. There are already several libraries available,

Statsig 34 Dec 14, 2022
Node Version Manager - POSIX-compliant bash script to manage multiple active node.js versions

Node Version Manager Table of Contents About Installing and Updating Install & Update Script Additional Notes Troubleshooting on Linux Troubleshooting

nvm.sh 63.8k Jan 9, 2023
Simple webpack plugin that generates VERSION and commitInfo insert index.html during build

Html commit version plugin Simple webpack plugin that generates CommitInfo and VERSION insert index.html during build. English | 简体中文 Usage Given a we

null 13 Mar 8, 2022
A simple rocketchat version for apple silicon

Why develop this version? Rocket. Chat Electron does not have Mac (Apple Silicon) version at present, so it has a poor use experience. After compiling

vast zhao 9 Sep 19, 2021
Vue.js version of Moai. Official WIP.

Moai Vue UI Kit ?? A Vue component library, where buttons look like buttons. Website: vue.moaijs.com Document: docs.vue.moaijs.com Install The [@moai-

Moai UI Kit 10 Nov 24, 2022
It maintains my version of js notes which I learned from the famous Namaste Javascript Youtube Series by Akshay Saini.

Welcome to Namaste Javascript Notes ?? ❓ what it is This repo maintains my version of javascript notes which I learned from the famous Namaste Javascr

Alok Raj 682 Jan 4, 2023
A CJS version of dateformat, forked from node-dateformat

dateformat A node.js package for Steven Levithan's excellent dateFormat() function. This module was forked from https://github.com/felixge/node-datefo

Matteo Collina 14 Sep 25, 2021
ScraperTools BETA Version 1.0.1

ScraperTools Official ScraperTools NPM Package Get Started Via NPM: $ npm install scraper-tools Cara Menggunakan const scrapertools = require('scraper

Zahir Hadi Athallah 21 Sep 28, 2022
A JavaScript, zero-dependency, super small version of IP2Location LITE country lookups.

A JavaScript, zero-dependency, super small version of IP2Location LITE country lookups.

Statsig 34 Dec 14, 2022