About GaussFormula

GaussFormula is a headless spreadsheet engine for JavaScript and TypeScript, designed for business and scientific web applications. It is a fork of HyperFormula with native support for uncertainty through explicit probability distributions such as N(mean, variance), LN(mu, sigma), U(min, max), N.CI(lower, upper[, confidence]), and LN.CI(lower, upper[, confidence]).

GaussFormula is ideal for:

• Custom spreadsheet-like apps
• Scientific and engineering tools
• Business logic builders
• Educational apps
• Online calculators

Key Features

• Function syntax compatible with Microsoft Excel and Google Sheets
• High-speed parsing and evaluation of spreadsheet formulas
• ~400 built-in functions
• Native support for uncertainty: work with explicit probability distributions and propagate uncertainty through sampled calculations
• Support for custom functions
• Node.js and browser support
• Undo/redo, CRUD operations, clipboard, named expressions, data sorting
• Formula localization (17+ languages)
• GPLv3 license

Installation

yarn add gaussformula
# or
npm install gaussformula

Usage Example

import { HyperFormula } from 'gaussformula';

// Create a GaussFormula instance
const gf = HyperFormula.buildEmpty({ licenseKey: 'gpl-v3' });

// Add a sheet and enter an explicit uncertainty input
const sheetName = gf.addSheet('Demo');
const sheetId = gf.getSheetId(sheetName);

gf.setCellContents({ sheet: sheetId, row: 0, col: 0 }, [['N.CI(1, 2)', '3', '=A1+B1', '=A1*B1']]);

console.log(gf.getCellValue({ sheet: sheetId, row: 0, col: 2 })); // SampledDistribution
console.log(gf.getCellValue({ sheet: sheetId, row: 0, col: 3 })); // SampledDistribution

What’s Unique: Distribution Support

GaussFormula extends HyperFormula with first-class support for explicit uncertainty inputs. Distribution inputs are parsed, stored, and propagated through arithmetic using Monte Carlo sampling.

Distribution notation

• N(mean, variance) uses normal-distribution parameters directly.
• LN(mu, sigma) uses log-space parameters: if X ~ LN(mu, sigma), then ln(X) ~ N(mu, sigma^2). mu and sigma are not the mean and standard deviation of X.
• Prefer LN.CI(lower, upper[, confidence]) for user-facing lognormal inputs because the bounds are in the original value scale.
• N.CI(lower, upper[, confidence]) derives a normal distribution from a value-scale confidence interval.
• U(min, max) represents a uniform range on the original value scale.

Example:

• A1 -> N.CI(1, 2)
• =A1 + 3 returns a sampled distribution
• =A1 * 3 returns a sampled distribution

Technical Notes

GaussFormula represents explicit uncertainty inputs with DistributionNumber and arithmetic outputs with SampledDistribution. Arithmetic over distributions uses Monte Carlo sampling instead of closed-form distribution algebra.

Contributing

Contributions are welcome! Please open issues or pull requests on GitHub.

License

GaussFormula is available under the GPLv3 license.

Acknowledgments

GaussFormula is a fork of the outstanding HyperFormula project by Handsoncode. Huge thanks to the original authors and maintainers for their work on the core spreadsheet engine.