For AI agents: the complete documentation index is available at /llms.txt, the full documentation bundle is available at /llms-full.txt, and this page is available as Markdown at /reference/array/flatMap.md.

flatMap es-toolkit

Transforms each element of an array with a function's return value, then flattens to a specified depth and returns a new array.

const result = flatMap(arr, iteratee, depth);

Usage

flatMap(arr, iteratee, depth = 1)

Use flatMap when you want to transform and flatten an array simultaneously. First, it applies a function to each element, then flattens the resulting array to the specified depth.

It works identically to calling Array#flat from JavaScript with Array#map as map(iteratee).flat(depth), but it's faster.

import { flatMap } from 'es-toolkit/array';

// Duplicate each element twice in a number array.
const arr = [1, 2, 3];
flatMap(arr, item => [item, item]);
// Returns: [1, 1, 2, 2, 3, 3]

// Flatten with depth 2.
flatMap(arr, item => [[item, item]], 2);
// Returns: [1, 1, 2, 2, 3, 3]

You can flatten to various depths.

import { flatMap } from 'es-toolkit/array';

const arr = [1, 2, 3];

// Flatten with default depth 1.
flatMap(arr, item => [item, item]);
// Returns: [1, 1, 2, 2, 3, 3]

// Flatten with depth 3.
flatMap(arr, item => [[[item, item]]], 3);
// Returns: [1, 1, 2, 2, 3, 3]

Parameters

arrT[]required

The array to transform.

iteratee(item: T, index: number, array: readonly T[]) => Urequired

The function that transforms each array element. It receives the element, its index, and the array.

depthDoptional

The depth to flatten. Default is 1.

Returns

Returns a new array where each element is transformed and flattened to the specified depth.

Array<FlatArray<U[], D>>

Examples

// Using index parameter
const arr = [1, 2, 3];
flatMap(arr, (item, index) => [item + index]);
// Returns: [1, 3, 5]

// Using array parameter
flatMap(arr, (item, _index, array) => [item * array.length]);
// Returns: [3, 6, 9]
Source: es-toolkit

Re-exported verbatim from es-toolkit. Implementation, edge cases, and performance behavior are owned upstream. This page mirrors the documentation at the pinned version; the linked source is authoritative.