Spelling Manager (Javascript)

A library that provides a framework for creating and using spelling checkers. This is a composite-based system where individual components can be combined together to provide the spelling checking on larger text strings.

Usage

Here are some basic examples of usage using ES6 module syntax.

import { BufferSpellingChecker, TokenSpellingManager, TokenCheckStatus } from "spelling-manager";
// var spelling = require("spelling-manager");

var spell = new TokenSpellingManager();
spell.add("like"); // Adds a case-insensitive token.
spell.add("I"); // Adds a case-sensitive token.
spell.remove("I"); // Removes the token.

var checker = new BufferSpellingChecker(spell);
var results = checker.check("I like cheese.");

console.log(results);
//[ { token: 'I', start: 0, end: 1, status: 0 },
//  { token: 'like', start: 2, end: 6, status: 1 },
//  { token: 'cheese', start: 7, end: 13, status: 0 } ]
console.log(results[0].status === TokenCheckStatus.Unknown);
//true
console.log(results[1].status === TokenCheckStatus.Correct);
//true
console.log(results[2].status === TokenCheckStatus.Incorrect);
//false

var results2 = checker.check("I LIKE CHEESE.");

console.log(results);
//[ { token: 'I', start: 0, end: 1, status: 0 },
//  { token: 'LIKE', start: 2, end: 6, status: 1 },
//  { token: 'CHEESE', start: 7, end: 13, status: 0 } ]

A natural tokenizer can be provided if the input buffer should be split in a different manner. The default tokenizer splits on words and single quotes, so "James'" will be "James" and "Mary's" will result in "Mary's".

import * as natural from "natural";
import { BufferSpellingChecker, TokenSpellingManager } from "spelling-manager";
// var spelling = require("spelling-manager");

var tokenizer = new natural.TreebankWordTokenizer();
var spell = new TokenSpellingManager();
var checker = new BufferSpellingChecker(spell, tokenizer);

Building and Tests

To build the library:

npm run build

To test the library:

npm test

License

MIT