Last active
August 23, 2022 00:32
-
-
Save jameslaneconkling/24acb8ea326a1c8fdf64225aa7d0f44e to your computer and use it in GitHub Desktop.
A silly simple lisp parser in javascript
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
const rules = [ | |
{ type: 'space', regex: /^\s/ }, | |
{ type: 'lParen', regex: /^\(/ }, | |
{ type: 'rParen', regex: /^\)/ }, | |
{ type: 'number', regex: /^[0-9\.]+/ }, | |
{ type: 'string', regex: /^".*?"/ }, | |
{ type: 'variable', regex: /^[^\s\(\)]+/ } // take from the beginning 1+ characters until you hit a ' ', '(', or ')' // TODO - support escaped double quote | |
]; | |
const tokenizer = rules => input => { | |
for (let i = 0; i < rules.length; i += 1) { | |
let tokenized = rules[i].regex.exec(input); | |
if (tokenized) { | |
return { | |
token: tokenized[0], | |
type: rules[i].type, | |
rest: input.slice(tokenized[0].length) | |
}; | |
} | |
} | |
throw new Error(`no matching tokenize rule for ${JSON.stringify(input)}`); | |
}; | |
const parser = tokenize => function parse(input, ast, parents = []) { | |
if (input === '') { | |
return ast; | |
} | |
const { token, type, rest } = tokenize(input); | |
if (type === 'space') { | |
// do nothing | |
return parse(rest, ast, parents); | |
} else if (type === 'variable') { | |
ast.push(token); | |
return parse(rest, ast, parents); | |
} else if (type === 'number') { | |
ast.push(Number(token)); | |
return parse(rest, ast, parents); | |
} else if (type === 'string') { | |
ast.push(token.replace(/(^"|"$)/g, "'")); | |
return parse(rest, ast, parents); | |
} else if (type === 'lParen') { | |
parents.push(ast) | |
return parse(rest, [], parents) | |
} else if (type === 'rParen') { | |
const parentAst = parents.pop(); | |
if (parentAst) { | |
parentAst.push(ast); | |
return parse(rest, parentAst, parents); | |
} | |
return parse(rest, ast, parents); | |
} | |
throw new Error(`Missing parse logic for rule ${JSON.stringify(type)}`); | |
}; | |
const input = `(first | |
(list 1(+ 2 (inc 3)) "nine" 10 "ele ven"))`; | |
console.log(JSON.stringify( | |
parser(tokenizer(rules))(input)[0] | |
)); | |
// ["first",["list",1,["+",2,["inc",3]],"'nine'",10,"'ele ven'"]] |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment