basic-with-validation.md

Basic example with data validation

Let's walk through a basic example of using the @radekstepan/flare library for feature flag evaluation in TypeScript.

import { Flare } from "@radekstepan/flare";
import { validate } from "@radekstepan/flare-utils";
// NOTE: only necessary if you use TypeScript
import { Operation, Kind, type Data } from "@radekstepan/flare-types";

This section imports the necessary modules from the @radekstepan/flare library, including Flare for feature flag evaluation, validate for data validation, and Operation and Kind from flare-types for specifying flag conditions.

const gates: Data = [{
  gate1: {
    eval: "user",
    conditions: [
      {
        id: "user",
        operation: Operation.INCLUDE,
        kind: Kind.CONTEXT,
        path: "name",
        value: ["tommy", "johnny"],
      },
    ],
  },
}];

Here, a sample Gates object is defined, representing a feature flag named "gate1". It contains a single condition that checks if the user's name (accessed via the path property) is either "tommy" or "johnny". This condition uses the Operation.INCLUDE operation to determine if the user's name is included in the specified values.

const flare = new Flare(
  new Promise((resolve, reject) => {
    validate
      .validateGates(gates)
      .then(() => {
        resolve(gates);
      })
      .catch(reject);
  })
);

A new instance of the Flare class is created, which takes a Promise as its argument. The Promise is used to validate the input Data array using the validateGates function from the flare-utils package.

flare.evaluate("gate1", { name: "johnny" }).then((res) => {
  console.log(res);
});

Finally, the evaluate method of the flare object is called with the flag name ("gate1") and an object representing the user's context ({name: "johnny"}). This method returns a Promise that resolves with the evaluation result. In this example, the result is logged to the console.