monaco-textmate

Monaco TextMate grammar helpers

Usage no npm install needed!

<script type="module">
  import monacoTextmate from 'https://cdn.skypack.dev/monaco-textmate';
</script>

README

This repository is a heavily modified version of original vscode-textmate package. It's been adjusted here and there to run inside web browsers. All of the file system calls have been removed, most of the API now uses Promises and grammars can no longer be loaded synchronously.

⚠ I'd prefer to see this repository where it really belongs. I request anyone from Microsoft to adopt this package as soon as possible.

Monaco TextMate

An interpreter for grammar files as defined by TextMate that runs on the web. Supports loading grammar files from JSON or PLIST format. Cross - grammar injections are currently not supported.

Installing

npm install monaco-textmate

Using

monaco-textmate relies on onigasm package to provide oniguruma regex engine in browsers. onigasm itself relies on WebAssembly. Therefore to get monaco-textmate working in your browser, it must have WebAssembly support and onigasm loaded and ready-to-go.

Make sure the example code below runs after onigasm bootstraping sequence described here has finished.

Example below is just a demostration of available API. To wire it up with monaco-editor use monaco-editor-textmate.

import { Registry } from 'monaco-textmate'

(async function test() {
    const registry = new Registry({
        // Since we're in browser, `getFilePath` has been removed, therefore you must provide `getGrammarDefinition` hook for things to work
        getGrammarDefinition: async (scopeName) => {
            // Whenever `Registry.loadGrammar` is called first time per scope name, this function will be called asking you to provide
            // raw grammar definition. Both JSON and plist formats are accepted.
            if (scopeName === 'source.css') {
                return {
                    format: 'json', // can also be `plist`
                    content: await (await fetch(`static/grammars/css.tmGrammar.json`)).text() // when format is 'json', parsed JSON also works
                }
            }
        }
    })

    const grammar = await registry.loadGrammar('source.css')

    console.log(grammar.tokenizeLine('html, body { height: 100%; margin: 0 }'))
    // > {tokens: Array(19), ruleStack: StackElement}
})()

onigasm is peer dependency that you must install yourself

Tokenizing multiple lines

To tokenize multiple lines, you must pass in the previous returned ruleStack.

var ruleStack = null;
for (var i = 0; i < lines.length; i++) {
    var r = grammar.tokenizeLine(lines[i], ruleStack);
    console.log('Line: #' + i + ', tokens: ' + r.tokens);
    ruleStack = r.ruleStack;
}

API doc

See the main.ts file

Developing

  • Clone the repository
  • Run npm install
  • Compile in the background with npm run watch

Credits

99% of the code in this repository is extracted straight from vscode-textmate, which is MIT licensed. Other external licenses used can be found in ThirdPartyNotices.txt

License

MIT