spech

Check your text for grammar and spelling mistakes using multiple providers

• Multiple libraries/services for grammar and spelling checking
• Multilingual document support
• Zero config
• Can work offline (only with Hunspell provider)

Usage

Files checking

npx spech README.md

File name can be omitted, by default it searches files by **/*.md mask. Default language is English, it can be changed with -l flag:

spech -l ru

spech -l ru-RU -l en-US - for multilingual documents

String and STDIN checking

You can check a string value using STDIN or --input argument

cat README.md | spech

spech --input 'Check the text'

Directives

disable

<!-- spech-disable -->
This text is ignored
<!-- spech-enable -->

dictionary

Add a phrase to a local document dictionary

<!-- spech-dictionary myword -->

languages

Add document-specific languages

<!-- spech-languages en es -->

Providers

To configure a provider pass -p flag:

spech -p hunspell -p yandex

Other options can be set in a config file.

Hunspell

Hunspell is the most popular open-source spell checker which supports a great variety of languages.

Read more.

GrammarBot

Free grammar checking API. With an API key, you can receive 250 requests/day (7500/mo) at no cost. Without an API key, requests are limited to 100 per day per IP address (3000/mo). The API supports only English (en-US and en-GB).

Read more.

Yandex Speller

Free and very fast spell checker API for en, ru and uk languages. It provides free 10k requests/day or 10m characters/day.

Read more.

Configuring

Configuration can be stored in:

• spech.config.js
• "spech" section of the package.json

spech.config.js

module.exports = {
  languages: ['en-us'],
  providers: [
    { name: 'hunspell' },
    { name: 'grammarBot', apiKey: 'YOUR_API_KEY' },
  ],
};

More details.

Dictionaries

It's possible to add words which are marked as a mistake into a dictionary file. Just create a file with .dic extension in your project root:

mydictionary.dic

# Simple word
browserify
# Regexp
/Component.tsx?/

More details.

API Usage

The most useful parts of the library are available through facade class SpellChecker.

Here is a simple example how it can be used:

const { Config, SpellChecker } = require('spech');

async function getMistakes() {
  const config = new Config({ ignoreCase: false });
  const checker = new SpellChecker(config);

  await checker.addDocumentsByMask(process.cwd(), 'docs/*.md');
  checker.addDictionaryPhrase('exceptionphrase');
  checker.addProviderByConfig({ name: 'hunspell' });

  const noMistakes = await checker.checkDocuments();
  if (noMistakes) {
    return [];
  }
  
  const corrections = checker.documents.map(doc => doc.corrections).flat();
  return corrections.map(correction => correction.fragment);
}