This project is based on AST (Abstract Syntax Tree). It can both generate the AST and validate the syntax of the input, transforming it into a consumable data structure to generate any type of representation, such as CSV tables. In case of an error, it is possible to handle the error, allowing the display of the line and column where a syntax error is present. Below is an example of how to work with this library:
- Syntax validation and error location reporting
- Abstract Syntax Tree (AST) generation
- Support for generating truth tables from parsed logic expressions
- Flexible output formats (e.g., CSV)
To get a local copy up and running, follow these simple steps:
Prerequisites Node.js (v20+ recommended)
NPM (recommended)
npm i truth-table-ast
After installing this package globally, you can use the following commands:
tt -p "p ˅ (p ^ q)"
# Or use this if your operating system doesn't detect the package
npx truth-table -p "p ˅ (p ^ q)" -o table.csv
Usage: ttt [options]
Options:
-h --help Show all available arguments
-o --output File where the truth table will be saved
-d --display How the data will be displayed in the table, supports: boolean, number
-t --type Type of file the table will be saved in (csv | text)
-p --proposition Define proposition to generate truth table
Here's an example of how to work with this library:
// ESM:
import { AST, Structure, Table } from 'truth-table-ast'
// Communjs:
// const { AST, Structure, Table } = require('truth-table-ast')
// In communjs, you'll need to put the code in an anonymous function for the asynchronous functions to work
// void (async () => {
// code here
// })()
const input = 'p ^ (p v ~q)'
const parser = new AST(input)
const ast = parser.parse()
if (AST.isError(ast)) throw new Error(JSON.stringify(ast, null, 2))
await parser.save('ast.json')
const structure = new Structure(ast).generate()
await structure.save('structure.json')
const table = new Table({
structure,
display: 'boolean',
// type: 'csv'
})//.create('table.csv')
// const content = table.csv()
// const content = table.markdown()
table.type = 'markdown'
await table.create('table.md')
table.type = 'csv'
await table.create('table.csv')
The AST (Abstract Syntax Tree) generated from the input of p ^ (p v ~q):
[
{
"value": "p",
"type": "Proposition",
"negatived": false,
"loc": {
"start": {
"line": 1,
"column": 0
},
"end": {
"line": 1,
"column": 0
}
}
},
{
"type": "Operation",
"value": "^",
"key": "Conjunction",
"loc": {...}
},
{
"type": "SubExpression",
"body": [
{
"value": "p",
"type": "Proposition",
"negatived": false,
"loc": {...}
},
{
"type": "Operation",
"value": "v",
"key": "Disjunction",
"loc": {...}
},
{
"value": "q",
"type": "Proposition",
"negatived": true,
"loc": {...}
}
],
"negatived": false,
"loc": {...}
}
]
The structured data generated from the AST to generate the truth table:
[
{
"type": "Variable",
"element": "p",
"value": true,
"column": 0,
"row": 0,
"position": "0x0"
},
{
"type": "Variable",
"element": "q",
"value": true,
"column": 1,
"row": 0,
"position": "0x1"
},
{
"type": "VariableNegative",
"element": "~q",
"value": false,
"column": 2,
"row": 0,
"position": "0x2"
},
{
"type": "Result",
"element": "(p v ~q)",
"value": true,
"column": 3,
"row": 0,
"position": "0x3"
},
{
"type": "Result",
"element": "p ^ (p v ~q)",
"value": true,
"column": 4,
"row": 0,
"position": "0x4"
},
{...}
]
p | q | ~q | (p v ~q) | p ^ (p v ~q) |
---|---|---|---|---|
true | true | false | true | true |
true | false | true | true | true |
false | true | false | false | false |
false | false | true | true | false |
Contributions are what make the open-source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.
Distributed under the MIT License. See LICENSE.txt for more information.