wishthis/node_modules/is-descriptor/node_modules/kind-of
2022-01-21 09:28:41 +01:00
..
CHANGELOG.md Add production dependencies 2022-01-21 09:28:41 +01:00
index.js Add production dependencies 2022-01-21 09:28:41 +01:00
LICENSE Add production dependencies 2022-01-21 09:28:41 +01:00
package.json Add production dependencies 2022-01-21 09:28:41 +01:00
README.md Add production dependencies 2022-01-21 09:28:41 +01:00

kind-of NPM version NPM monthly downloads NPM total downloads Linux Build Status

Get the native type of a value.

Please consider following this project's author, Jon Schlinkert, and consider starring the project to show your ❤️ and support.

Install

Install with npm:

$ npm install --save kind-of

Install with bower

$ bower install kind-of --save

Why use this?

  1. it's fast | optimizations
  2. better type checking

Usage

es5, es6, and browser ready

var kindOf = require('kind-of');

kindOf(undefined);
//=> 'undefined'

kindOf(null);
//=> 'null'

kindOf(true);
//=> 'boolean'

kindOf(false);
//=> 'boolean'

kindOf(new Buffer(''));
//=> 'buffer'

kindOf(42);
//=> 'number'

kindOf('str');
//=> 'string'

kindOf(arguments);
//=> 'arguments'

kindOf({});
//=> 'object'

kindOf(Object.create(null));
//=> 'object'

kindOf(new Test());
//=> 'object'

kindOf(new Date());
//=> 'date'

kindOf([1, 2, 3]);
//=> 'array'

kindOf(/foo/);
//=> 'regexp'

kindOf(new RegExp('foo'));
//=> 'regexp'

kindOf(new Error('error'));
//=> 'error'

kindOf(function () {});
//=> 'function'

kindOf(function * () {});
//=> 'generatorfunction'

kindOf(Symbol('str'));
//=> 'symbol'

kindOf(new Map());
//=> 'map'

kindOf(new WeakMap());
//=> 'weakmap'

kindOf(new Set());
//=> 'set'

kindOf(new WeakSet());
//=> 'weakset'

kindOf(new Int8Array());
//=> 'int8array'

kindOf(new Uint8Array());
//=> 'uint8array'

kindOf(new Uint8ClampedArray());
//=> 'uint8clampedarray'

kindOf(new Int16Array());
//=> 'int16array'

kindOf(new Uint16Array());
//=> 'uint16array'

kindOf(new Int32Array());
//=> 'int32array'

kindOf(new Uint32Array());
//=> 'uint32array'

kindOf(new Float32Array());
//=> 'float32array'

kindOf(new Float64Array());
//=> 'float64array'

Benchmarks

Benchmarked against typeof and type-of.

# arguments (32 bytes)
  kind-of x 17,024,098 ops/sec ±1.90% (86 runs sampled)
  lib-type-of x 11,926,235 ops/sec ±1.34% (83 runs sampled)
  lib-typeof x 9,245,257 ops/sec ±1.22% (87 runs sampled)

  fastest is kind-of (by 161% avg)

# array (22 bytes)
  kind-of x 17,196,492 ops/sec ±1.07% (88 runs sampled)
  lib-type-of x 8,838,283 ops/sec ±1.02% (87 runs sampled)
  lib-typeof x 8,677,848 ops/sec ±0.87% (87 runs sampled)

  fastest is kind-of (by 196% avg)

# boolean (24 bytes)
  kind-of x 16,841,600 ops/sec ±1.10% (86 runs sampled)
  lib-type-of x 8,096,787 ops/sec ±0.95% (87 runs sampled)
  lib-typeof x 8,423,345 ops/sec ±1.15% (86 runs sampled)

  fastest is kind-of (by 204% avg)

# buffer (38 bytes)
  kind-of x 14,848,060 ops/sec ±1.05% (86 runs sampled)
  lib-type-of x 3,671,577 ops/sec ±1.49% (87 runs sampled)
  lib-typeof x 8,360,236 ops/sec ±1.24% (86 runs sampled)

  fastest is kind-of (by 247% avg)

# date (30 bytes)
  kind-of x 16,067,761 ops/sec ±1.58% (86 runs sampled)
  lib-type-of x 8,954,436 ops/sec ±1.40% (87 runs sampled)
  lib-typeof x 8,488,307 ops/sec ±1.51% (84 runs sampled)

  fastest is kind-of (by 184% avg)

# error (36 bytes)
  kind-of x 9,634,090 ops/sec ±1.12% (89 runs sampled)
  lib-type-of x 7,735,624 ops/sec ±1.32% (86 runs sampled)
  lib-typeof x 7,442,160 ops/sec ±1.11% (90 runs sampled)

  fastest is kind-of (by 127% avg)

# function (34 bytes)
  kind-of x 10,031,494 ops/sec ±1.27% (86 runs sampled)
  lib-type-of x 9,502,757 ops/sec ±1.17% (89 runs sampled)
  lib-typeof x 8,278,985 ops/sec ±1.08% (88 runs sampled)

  fastest is kind-of (by 113% avg)

# null (24 bytes)
  kind-of x 18,159,808 ops/sec ±1.92% (86 runs sampled)
  lib-type-of x 12,927,635 ops/sec ±1.01% (88 runs sampled)
  lib-typeof x 7,958,234 ops/sec ±1.21% (89 runs sampled)

  fastest is kind-of (by 174% avg)

# number (22 bytes)
  kind-of x 17,846,779 ops/sec ±0.91% (85 runs sampled)
  lib-type-of x 3,316,636 ops/sec ±1.19% (86 runs sampled)
  lib-typeof x 2,329,477 ops/sec ±2.21% (85 runs sampled)

  fastest is kind-of (by 632% avg)

# object-plain (47 bytes)
  kind-of x 7,085,155 ops/sec ±1.05% (88 runs sampled)
  lib-type-of x 8,870,930 ops/sec ±1.06% (83 runs sampled)
  lib-typeof x 8,716,024 ops/sec ±1.05% (87 runs sampled)

  fastest is lib-type-of (by 112% avg)

# regex (25 bytes)
  kind-of x 14,196,052 ops/sec ±1.65% (84 runs sampled)
  lib-type-of x 9,554,164 ops/sec ±1.25% (88 runs sampled)
  lib-typeof x 8,359,691 ops/sec ±1.07% (87 runs sampled)

  fastest is kind-of (by 158% avg)

# string (33 bytes)
  kind-of x 16,131,428 ops/sec ±1.41% (85 runs sampled)
  lib-type-of x 7,273,172 ops/sec ±1.05% (87 runs sampled)
  lib-typeof x 7,382,635 ops/sec ±1.17% (85 runs sampled)

  fastest is kind-of (by 220% avg)

# symbol (34 bytes)
  kind-of x 17,011,537 ops/sec ±1.24% (86 runs sampled)
  lib-type-of x 3,492,454 ops/sec ±1.23% (89 runs sampled)
  lib-typeof x 7,471,235 ops/sec ±2.48% (87 runs sampled)

  fastest is kind-of (by 310% avg)

# template-strings (36 bytes)
  kind-of x 15,434,250 ops/sec ±1.46% (83 runs sampled)
  lib-type-of x 7,157,907 ops/sec ±0.97% (87 runs sampled)
  lib-typeof x 7,517,986 ops/sec ±0.92% (86 runs sampled)

  fastest is kind-of (by 210% avg)

# undefined (29 bytes)
  kind-of x 19,167,115 ops/sec ±1.71% (87 runs sampled)
  lib-type-of x 15,477,740 ops/sec ±1.63% (85 runs sampled)
  lib-typeof x 19,075,495 ops/sec ±1.17% (83 runs sampled)

  fastest is lib-typeof,kind-of

Optimizations

In 7 out of 8 cases, this library is 2x-10x faster than other top libraries included in the benchmarks. There are a few things that lead to this performance advantage, none of them hard and fast rules, but all of them simple and repeatable in almost any code library:

  1. Optimize around the fastest and most common use cases first. Of course, this will change from project-to-project, but I took some time to understand how and why typeof checks were being used in my own libraries and other libraries I use a lot.
  2. Optimize around bottlenecks - In other words, the order in which conditionals are implemented is significant, because each check is only as fast as the failing checks that came before it. Here, the biggest bottleneck by far is checking for plain objects (an object that was created by the Object constructor). I opted to make this check happen by process of elimination rather than brute force up front (e.g. by using something like val.constructor.name), so that every other type check would not be penalized it.
  3. Don't do uneccessary processing - why do .slice(8, -1).toLowerCase(); just to get the word regex? It's much faster to do if (type === '[object RegExp]') return 'regex'
  4. There is no reason to make the code in a microlib as terse as possible, just to win points for making it shorter. It's always better to favor performant code over terse code. You will always only be using a single require() statement to use the library anyway, regardless of how the code is written.

Better type checking

kind-of seems to be more consistently "correct" than other type checking libs I've looked at. For example, here are some differing results from other popular libs:

typeof lib

Incorrectly identifies instances of custom constructors (pretty common):

var typeOf = require('typeof');
function Test() {}
console.log(typeOf(new Test()));
//=> 'test'

Returns object instead of arguments:

function foo() {
  console.log(typeOf(arguments)) //=> 'object'
}
foo();

type-of lib

Incorrectly returns object for generator functions, buffers, Map, Set, WeakMap and WeakSet:

function * foo() {}
console.log(typeOf(foo));
//=> 'object'
console.log(typeOf(new Buffer('')));
//=> 'object'
console.log(typeOf(new Map()));
//=> 'object'
console.log(typeOf(new Set()));
//=> 'object'
console.log(typeOf(new WeakMap()));
//=> 'object'
console.log(typeOf(new WeakSet()));
//=> 'object'

About

Contributing

Pull requests and stars are always welcome. For bugs and feature requests, please create an issue.

Running Tests

Running and reviewing unit tests is a great way to get familiarized with a library and its API. You can install dependencies and run tests with the following command:

$ npm install && npm test
Building docs

(This project's readme.md is generated by verb, please don't edit the readme directly. Any changes to the readme must be made in the .verb.md readme template.)

To generate the readme, run the following command:

$ npm install -g verbose/verb#dev verb-generate-readme && verb

You might also be interested in these projects:

Contributors

Commits Contributor
102 jonschlinkert
3 aretecode
2 miguelmota
1 doowb
1 dtothefp
1 ianstormtaylor
1 ksheedlo
1 pdehaan
1 laggingreflex
1 tunnckoCore
1 xiaofen9

Author

Jon Schlinkert

License

Copyright © 2020, Jon Schlinkert. Released under the MIT License.


This file was generated by verb-generate-readme, v0.8.0, on January 16, 2020.