@smarthorizon-labs/dmn-feel
v0.1.0
Published
A FEEL parser and evaluator with DMN-conformant DECIMAL128 decimal arithmetic
Maintainers
Readme
@smarthorizon-labs/dmn-feel
A FEEL parser and evaluator for TypeScript, with the decimal arithmetic the DMN specification requires.
FEEL is DMN's expression language, but it stands on its own as a safe,
side-effect-free language for user-authored expressions — pricing formulas,
eligibility conditions, validation rules. Model content is data, never host
code: no eval, no new Function.
Numbers are MathContext.DECIMAL128 decimals — 34 significant digits,
HALF_EVEN rounding — which is what Drools, IBM BAMOE, jDMN and Camunda's
FEEL-Scala all use, and
verified against Drools digit for digit.
Early release (0.x). Usable for decision models that stay inside the documented feature set. See what works today for the exact boundary, and the TCK report for measured conformance.
Install
npm install @smarthorizon-labs/dmn-feelUsage
import { createFeelEngine, toFeelNumber, describeValue } from '@smarthorizon-labs/dmn-feel';
const engine = createFeelEngine();
// Compile once, evaluate many times.
const compiled = engine.compileExpression('Total * (1 + Rate)');
const { value } = compiled.evaluate({
Total: toFeelNumber('1499.99'),
Rate: toFeelNumber('0.23'),
});
console.log(describeValue(value)); // 1844.9877The grammar is lezer-feel by Nico
Rehwaldt; evaluation is original work. See NOTICE.
Part of a set
| Package | Use it for |
|---|---|
| @smarthorizon-labs/dmn | Start here. Everything below, one install |
| @smarthorizon-labs/dmn-runtime | Decision graph, decision tables, tracing |
| @smarthorizon-labs/dmn-feel | FEEL parsing and evaluation |
| @smarthorizon-labs/dmn-xml | Secure DMN XML loading |
| @smarthorizon-labs/dmn-model | Normalized model types |
Full documentation, TCK results and design rationale are in the repository README.
License
Apache-2.0. Third-party attributions are in NOTICE.
