# dequal [![CI](https://github.com/lukeed/dequal/workflows/CI/badge.svg)](https://github.com/lukeed/dequal/actions)
> A tiny (304B to 489B) utility to check for deep equality
This module supports comparison of all types, including `Function`, `RegExp`, `Date`, `Set`, `Map`, `TypedArray`s, `DataView`, `null`, `undefined`, and `NaN` values. Complex values (eg, Objects, Arrays, Sets, Maps, etc) are traversed recursively.
> **Important:**
> * key order **within Objects** does not matter
> * value order **within Arrays** _does_ matter
> * values **within Sets and Maps** use value equality
> * keys **within Maps** use value equality
## Install
```
$ npm install --save dequal
```
## Modes
There are two "versions" of `dequal` available:
#### `dequal`
> **Size (gzip):** 489 bytes
> **Availability:** [CommonJS](https://unpkg.com/dequal/dist/index.js), [ES Module](https://unpkg.com/dequal/dist/index.mjs), [UMD](https://unpkg.com/dequal/dist/index.min.js)
#### `dequal/lite`
> **Size (gzip):** 304 bytes
> **Availability:** [CommonJS](https://unpkg.com/dequal/lite/index.js), [ES Module](https://unpkg.com/dequal/lite/index.mjs)
| | IE9+ | Number | String | Date | RegExp | Object | Array | Class | Set | Map | ArrayBuffer | [TypedArray](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/TypedArray#TypedArray_objects) | [DataView](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/DataView) |
|-|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|
| `dequal` | :x: | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: |
| `dequal/lite` | :+1: | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: | :x: | :x: | :x: | :x: | :x: |
> **Note:** Table scrolls horizontally!
## Usage
```js
import { dequal } from 'dequal';
dequal(1, 1); //=> true
dequal({}, {}); //=> true
dequal('foo', 'foo'); //=> true
dequal([1, 2, 3], [1, 2, 3]); //=> true
dequal(dequal, dequal); //=> true
dequal(/foo/, /foo/); //=> true
dequal(null, null); //=> true
dequal(NaN, NaN); //=> true
dequal([], []); //=> true
dequal(
[{ a:1 }, [{ b:{ c:[1] } }]],
[{ a:1 }, [{ b:{ c:[1] } }]]
); //=> true
dequal(1, '1'); //=> false
dequal(null, undefined); //=> false
dequal({ a:1, b:[2,3] }, { a:1, b:[2,5] }); //=> false
dequal(/foo/i, /bar/g); //=> false
```
## API
### dequal(foo, bar)
Returns: `Boolean`
Both `foo` and `bar` can be of any type.
A `Boolean` is returned indicating if the two were deeply equal.
## Benchmarks
> Running Node v10.13.0
The benchmarks can be found in the [`/bench`](/bench) directory. They are separated into two categories:
* `basic` – compares an object comprised of `String`, `Number`, `Date`, `Array`, and `Object` values.
* `complex` – like `basic`, but adds `RegExp`, `Map`, `Set`, and `Uint8Array` values.
> **Note:** Only candidates that pass validation step(s) are listed.
For example, `fast-deep-equal/es6` handles `Set` and `Map` values, but uses _referential equality_ while those listed use _value equality_.
```
Load times:
assert 0.109ms
util 0.006ms
fast-deep-equal 0.479ms
lodash/isequal 22.826ms
nano-equal 0.417ms
dequal 0.396ms
dequal/lite 0.264ms
Benchmark :: basic
assert.deepStrictEqual x 325,262 ops/sec ±0.57% (94 runs sampled)
util.isDeepStrictEqual x 318,812 ops/sec ±0.87% (94 runs sampled)
fast-deep-equal x 1,332,393 ops/sec ±0.36% (93 runs sampled)
lodash.isEqual x 269,129 ops/sec ±0.59% (95 runs sampled)
nano-equal x 1,122,053 ops/sec ±0.36% (96 runs sampled)
dequal/lite x 1,700,972 ops/sec ±0.31% (94 runs sampled)
dequal x 1,698,972 ops/sec ±0.63% (97 runs sampled)
Benchmark :: complex
assert.deepStrictEqual x 124,518 ops/sec ±0.64% (96 runs sampled)
util.isDeepStrictEqual x 125,113 ops/sec ±0.24% (96 runs sampled)
lodash.isEqual x 58,677 ops/sec ±0.49% (96 runs sampled)
dequal x 345,386 ops/sec ±0.27% (96 runs sampled)
```
## License
MIT © [Luke Edwards](https://lukeed.com)