Skip to content

yann510/ss-search

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

npm npm bundle size build Coveralls github Commitizen friendly semantic-release

s(imply)s(tupid)-search

The most basic, yet powerful text search.

Stop searching, start finding.

  • Ease of Use: Get appropriate results instantly without any configuration.
  • Local Array Search: Effortlessly search through a local array of objects.
  • Automatic Indexing: No manual indexing required.
  • Precision: Always get exactly what you're looking for.
  • Lightweight: Depends on just 5 lodash functions for a minimized size after tree shaking.

Demo

Still not convinced? Experience its power firsthand with this interactive demo.

Benchmark

How does it compare to other search libraries? Test out for yourself with this interactive benchmark ;)

Install

ss-search is available on npm. Install it with:

npm install ss-search

Usage

Basic

import { search } from 'ss-search'

const data = [
  {
    number: 1,
    text: 'A search function should be fast',
  },
  {
    number: 2,
    text: 'A search function should provide accurate results',
  },
]
const searchKeys = ['text']
const searchText = 'fast search'

const results = search(data, searchKeys, searchText)
// results: [{ number: 1, text: "A search function should be fast" }]

It's that straightforward. No configurations, it just works.

Data Types

Almost all data types are supported [boolean, number, string, object, array].

// This dataset will be used as a common starting point for our type examples
const data = [
  {
    boolean: true,
    number: 1,
    string: 'search',
    object: { nestedProperty: 'nested value' },
    array: ['value1', 'value2'],
    arrayObjects: [{ arrayObjectProperty: 'array object value' }],
  },
]

Boolean

const results = search(data, ['boolean'], 'true')
// results: will return our original dataset

Number

const results = search(data, ['number'], '1')
// results: will return our original dataset

String

const results = search(data, ['string'], 'search')
// results: will return our original dataset

Object

Providing a key which refers to an object will stringify that object using JSON.stringify

const results = search(data, ['object'], 'property')
// results: will return our original dataset as it matches the property key "nestedProperty" of our object

If you want to access a nested property of an object to extract only a single value

const results = search(data, ['object.nestedProperty'], 'property')
// results: will return an empty array as we extracted the value of our nested object
// if we had searched for "nested value" we would of had the original dataset

Array

Providing a key which refers to an array will stringify that array using JSON.stringify

const results = search(data, ['array'], 'value2')
// results: will return our original dataset

If you have an array of objects on which you want to search all properties

const results = search(data, ['arrayObjects'], 'arrayObjectProperty')
// results: will return an our original dataset as it's treated just like a regular array
// thus the arrayObjectProperty is part of the searchable text

If you have an array of objects where you want only specific properties to be searchable

const results = search(data, ['arrayObjects[arrayObjectProperty]'], 'arrayObjectProperty')
// results: will return an empty array as we extracted the value of our nested array of objects
// if we had searched for "value object" we would of had the original dataset

Options for the search function

Customize your search experience using the following options:

Option parameter Value Description
withScore true When set to true, the search function will return an array of objects, each containing the matched element and its corresponding score. The score represents how closely the element matches the search text, with a higher score indicating a closer match. Even if the search doesn't match, it will return a score of 0.
withScore false When set to false or not provided, the function will return an array of matched elements without their scores.

Example Usage

Without withScore option:

const data = [{ name: 'John' }, { name: 'Jane' }, { name: 'Doe' }]
const result = search(data, ['name'], 'John')
console.log(result) // [{ name: 'John' }]

With withScore option:

const data = [{ name: 'John' }, { name: 'Jane' }, { name: 'Doe' }]
const result = search(data, ['name'], 'John', { withScore: true })
console.log(result)
// [
//  { element: { name: 'John' }, score: 1 },
//  { element: { name: 'Jane' }, score: 0 },
//  { element: { name: 'Doe' }, score: 0 }
// ]

Developing

To better manage dependencies across the monorepo I'm using NX.

Install dependencies: npm i

Start the web-app: npm run web-app:serve

Test the library: npm run test:all