🌕 Mercury Parser

This Package does not generate any sound or visuals. This package only parses Mercury code and returns a JSON formatted parse tree. This is used in the Mercury environment to translate the code to sound and visual objects. This is also used in the Mercury-Playground, a browser based lite version of the environment.

📟 Mercury?

Mercury is a minimal and human-readable language for the live coding of algorithmic electronic music.

🚀 Go to the full Mercury Project

👾 Or start sketching in the browser:

🙏 Support Mercury by becoming a Patron

💬 Join the Discord Community!

How to use this parser

1. git clone and run npm install

2. Open the mercury.ne file to view the grammar in the Nearley language including the moo tokenizer.

3. Run npm run build

   • Generates the the grammar.js parser file
   • Generates a minified es5 browser version in /build
   • Generates the railroad graph

4. Run a test with npm test and view result in /test/tree

Use via require

Install in node_modules

$ npm install mercury-lang

const Mercury = require('mercury-lang');

Import es5 version

const Mercury = require('mercury-lang/build/mercury.es5.min.js');

Include in html

Include latest or specific version of bundled minified es5 through url in index.html

<script src="https://unpkg.com/[email protected]/build/mercury.es5.min.js"></script>

Use in a html <script> like so:

// entire package
const Mercury = MercuryParser;

Example

A small code file of Mercury below

// A small example for the Mercury parser
set tempo 140
set nonSetting 10

list myBeat euclidean(8 5 1)

new synth saw time(1/8) play(myBeat) name(s1)
	give s1 fx(reverb 0.9 7)

list notes random 16 0 12)

Input the code in the Mercury parser

const mercury = require('mercury-lang');
const code = fs.readFileSync('example.txt', 'utf-8');

const result = mercury(code);

The result is a JS object consisting of a parseTree...

console.log(result.parseTree);

{
	global: {
		tempo: [ 140 ],
		scale: [ 'chromatic', 'c' ],
		root: [ 'c' ],
		randomSeed: [ 0 ],
		highPass: [ 20000, 0 ],
		lowPass: [ 1, 0 ],
		silence: false
	},
	variables: { 
		myBeat: [
			0, 1, 0, 1,
			1, 0, 1, 1
		] 
	},
	objects: {
		s1: {
		object: 'synth',
		type: 'saw',
		functions: {
			group: [],
			time: [ '1/8' ],
			note: [ 0, 0 ],
			env: [ 5, 500 ],
			beat: [
			[
				0, 1, 0, 1,
				1, 0, 1, 1
			]
			],
			amp: [ 0.7 ],
			wave2: [ 'saw', 0 ],
			add_fx: [ [ 'reverb', 0.9, 7 ] ],
			name: [ 's1' ]
		}
		}
	},
	groups: { all: [ 's1' ] },
	print: [],
	comments: [ '// A small example for the Mercury parser' ]
}

... a syntaxTree...

console.log(result.syntaxTree);

{
	'@main': [
		{
		'@global': { '@comment': '// A small example for the Mercury parser' }
		},
		{
		'@object': {
			'@set': { '@functions': [ { '@number': 140 } ] }
		}
		},
		{
		'@list': {
			'@name': 'myBeat',
			'@params': {
			'@function': {
				'@name': 'euclidean',
				'@args': [ { '@number': 8 }, { '@number': 5 }, { '@number': 1 } ]
			}
			}
		}
		},
		{
		'@object': {
			'@new': {
			'@type': { '@identifier': 'saw' },
			'@functions': [
				{
				'@function': {
					'@name': 'time',
					'@args': [ { '@division': '1/8' } ]
				}
				},
				{
				'@function': {
					'@name': 'beat',
					'@args': [ { '@identifier': 'myBeat' } ]
				}
				},
				{
				'@function': {
					'@name': 'name',
					'@args': [ { '@identifier': 's1' } ]
				}
				}
			]
			}
		}
		},
		{
		'@object': {
			'@set': {
			'@functions': [
				{
				'@function': {
					'@name': 'add_fx',
					'@args': [
					{ '@identifier': 'reverb' },
					{ '@number': 0.9 },
					{ '@number': 7 }
					]
				}
				}
			]
			}
		}
		}
	]
}

an errors array with encountered syntax errors

"errors": [
    "Syntax error at line 9 col 19: Unexpected number: 16 at list notes random 16<-"
  ]

...and a warnings array with warnings that may cause issues

"warnings" : [
	"Warning: Unkown setting name: nonSetting"
]

NPM dependencies

• Nearley Parser Toolkit
• Moo! Tokenizer/Lexer Generator

License

The GNU GPL-v.3