Hunspell Spellchecker in Javascript

Common errors

• ReferenceError: require is not defined

  cause Attempting to use `require()` in an ES Module (ESM) context without proper configuration or transpilation.
  fix Ensure your project is configured for CommonJS, or if using ESM, adjust your build setup to handle CommonJS imports. For strict ESM, rewrite the import statement using `import * as Spellchecker from 'hunspell-spellchecker';` (though this package is CJS-first, this might still lead to issues) or use dynamic `import('hunspell-spellchecker')`.

• TypeError: fs.readFileSync is not a function

  cause Trying to run the dictionary parsing example code directly in a browser environment.
  fix The `fs` module is a Node.js built-in. Dictionary parsing must occur on the server-side (Node.js) or as part of a build process. Only the resulting JSON dictionary should be loaded in the browser.

• Error: ENOENT: no such file or directory, open './en_EN.aff'

  cause The specified Hunspell dictionary (`.aff` or `.dic`) file path is incorrect or the file does not exist at the given location.
  fix Verify the file paths provided to `fs.readFileSync()` are absolute or correctly relative to `process.cwd()` (Node.js) or `__dirname`. Ensure the dictionary files are present in the expected location.

Warnings

• breaking This package is abandoned and has not been updated in over 11 years. It is unlikely to receive security patches, bug fixes, or new features. Consider more actively maintained alternatives like `nspell` or `nuspell` for modern applications.
  Fix: Migrate to a contemporary spell-checking library such as `nspell` (JavaScript-based) or integrate with `Nuspell` (C++ with bindings) for more robust, actively maintained solutions.
• gotcha The example code relies on Node.js's `fs.readFileSync` for parsing dictionary files. This makes the parsing step inherently server-side. Direct use in a browser environment will fail unless dictionary files are pre-parsed into JSON and loaded, or `fs` is polyfilled/mocked.
  Fix: For browser usage, pre-parse the `.aff` and `.dic` files into the JSON format using Node.js or a build step, then serve and load the JSON dictionary directly in the browser. Do not attempt to use `fs.readFileSync` client-side.
• gotcha The library is CommonJS-only and does not natively support ES Modules (ESM) `import` syntax. Attempting to use `import Spellchecker from 'hunspell-spellchecker'` in an ESM context will likely result in errors.
  Fix: In Node.js ESM projects, use `const Spellchecker = require('hunspell-spellchecker');` if allowed, or configure your bundler (e.g., Webpack, Rollup) to correctly handle CommonJS modules. If strict ESM is required, consider alternative libraries.

Install

• npm install hunspell-spellchecker npm
• yarn add hunspell-spellchecker yarn
• pnpm add hunspell-spellchecker pnpm

Imports

• Spellchecker
  wrong:

  import Spellchecker from 'hunspell-spellchecker';

  correct:

  const Spellchecker = require('hunspell-spellchecker');

  This package is CommonJS-only, last published over 11 years ago. ESM imports are not natively supported and require a CommonJS wrapper or bundler configuration.
• Spellchecker (instance)
  wrong:

  const spellchecker = hunspell_spellchecker();

  correct:

  const spellchecker = new Spellchecker();

  The library exports a constructor function, not a direct instance or a factory function. It must be instantiated with `new`.

Quickstart

This quickstart demonstrates how to initialize the spellchecker, parse Hunspell dictionary files (.aff and .dic) from the filesystem, load the processed dictionary, and then check the spelling of a word. It assumes dictionary files are available locally.

const fs = require('fs');
const path = require('path');
const Spellchecker = require('hunspell-spellchecker');

// Initialize a spellchecker instance
const spellchecker = new Spellchecker();

// Example dictionary files (replace with actual paths)
// For demonstration, these files must exist in your project root
// or be accessible via `fs.readFileSync`.
// You would typically download official Hunspell dictionaries (e.g., from LibreOffice).
const affFilePath = path.join(__dirname, 'en_US.aff'); // Placeholder
const dicFilePath = path.join(__dirname, 'en_US.dic'); // Placeholder

// In a real application, ensure these files exist
// For testing, you might create dummy files or skip this part.
if (!fs.existsSync(affFilePath) || !fs.existsSync(dicFilePath)) {
  console.error('Error: Dictionary files (en_US.aff, en_US.dic) not found.');
  console.error('Please download them or provide valid paths for parsing.');
  console.error('For example, download from: https://extensions.libreoffice.org/extensions/english-dictionaries (rename .oxt to .zip)');
  process.exit(1);
}

// Parse an hunspell dictionary that can be serialized as JSON
const DICT = spellchecker.parse({
    aff: fs.readFileSync(affFilePath),
    dic: fs.readFileSync(dicFilePath)
});

// Load a serialized dictionary into the spellchecker
spellchecker.use(DICT);

// Check a word
const wordToCheck = 'typo';
const isRight = spellchecker.check(wordToCheck);

console.log(`Is '${wordToCheck}' spelled correctly? ${isRight}`); // Should be false

const correctWord = 'hello';
const isCorrect = spellchecker.check(correctWord);
console.log(`Is '${correctWord}' spelled correctly? ${isCorrect}`); // Should be true