npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2024 – Pkg Stats / Ryan Hefner

detest-bdd

v1.2.1

Published

A type-safe set of utilities that focuses on reducing the amount of test code that you write while maximizing coverage. Compatible with BDD testing frameworks like Jasmine and Mocha.

Downloads

10

Readme

Detest BDD

Detest BDD is a type-safe set of utilities that focuses on reducing the amount of test code that you write while maximizing coverage. Compatible with BDD testing frameworks like Jasmine and Mocha.

Installation

The project can be installed via npm using the following command:

npm install detest-bdd

No further installation is required.

Detest BDD works with any JavaScript testing framework that uses the describe beforeEach it syntax (i.e. Jasmine, Mocha + Chai, etc.).

Examples

(For more information, see the full API reference)

Template

When writing unit tests, it's easy to write redundant test cases repeatedly with only minor variations. This leads to code that's not only tedious to write, but also hard to read and refactor in the future. Test templates solve this issue by allowing you to write parts of your test code in a container that allow you to pass input values and modify the shape of the test code based on those inputs.

Example

Given this piece of code:

export interface Options {
    round?: boolean;
    absolute?: boolean;
}

export class Calculator {

    public divide(a: number, b: number, options?: Options): number {
        options = options || {};
        let result = a / b;

        result = options.round ? Math.round(result) : result;
        return options.absolute ? Math.abs(result) : result;
    }
}

First, let's look at an example of writing unit tests for Calculator.divide without test templates:

describe("Given a Calculator", () => {

    beforeEach(function () {
        this.calculator = new Calculator();
    });

    beforeEach(function () {
        this.a = Random.number();
        this.b = Random.number();
    });

    describe("when the divide method is called", () => {

        describe("when options are passed", () => {

            describe("when options.round is true", () => {

                describe("when options.absolute is true", () => {

                    it("then it should return the expected value", function () {
                        this.expected = Math.abs(Math.round(this.a / this.b));

                        expect(this.calculator.divide(this.a, this.b, {
                            round: true,
                            absolute: true
                        })).toEqual(this.expected);
                    });
                });

                describe("when options.absolute is false", () => {

                    it("then it should return the expected value", function () {
                        this.expected = Math.round(this.a / this.b);

                        expect(this.calculator.divide(this.a, this.b, {
                            round: true,
                            absolute: false
                        })).toEqual(this.expected);
                    });
                });

                describe("when options.absolute is undefined", () => {

                    it("then it should return the expected value", function () {
                        this.expected = Math.round(this.a / this.b);

                        expect(this.calculator.divide(this.a, this.b, {
                            round: true
                        })).toEqual(this.expected);
                    });
                });
            });

            describe("when options.round is false", () => {

                describe("when options.absolute is true", () => {

                    it("then it should return the expected value", function () {
                        this.expected = Math.abs(this.a / this.b);

                        expect(this.calculator.divide(this.a, this.b, {
                            round: false,
                            absolute: true
                        })).toEqual(this.expected);
                    });
                });

                describe("when options.absolute is false", () => {

                    it("then it should return the expected value", function () {
                        this.expected = this.a / this.b;

                        expect(this.calculator.divide(this.a, this.b, {
                            round: false,
                            absolute: false
                        })).toEqual(this.expected);
                    });
                });

                describe("when options.absolute is undefined", () => {

                    it("then it should return the expected value", function () {
                        this.expected = this.a / this.b;

                        expect(this.calculator.divide(this.a, this.b, {
                            round: false
                        })).toEqual(this.expected);
                    });
                });
            });

            describe("when options.round is not specified", () => {

                describe("when options.absolute is true", () => {

                    it("then it should return the expected value", function () {
                        this.expected = Math.abs(this.a / this.b);

                        expect(this.calculator.divide(this.a, this.b, {
                            absolute: true
                        })).toEqual(this.expected);
                    });
                });

                describe("when options.absolute is false", () => {

                    it("then it should return the expected value", function () {
                        this.expected = this.a / this.b;

                        expect(this.calculator.divide(this.a, this.b, {
                            absolute: false
                        })).toEqual(this.expected);
                    });
                });

                describe("when options.absolute is undefined", () => {

                    it("then it should return the expected value", function () {
                        this.expected = this.a / this.b;

                        expect(this.calculator.divide(this.a, this.b, {})).toEqual(this.expected);
                    });
                });
            });
        });

        describe("when options are NOT passed", () => {

            it("then it should return the expected value", function () {
                this.expected = this.a / this.b;

                expect(this.calculator.divide(this.a, this.b)).toEqual(this.expected);
            });
        });
    });
});

Output:

..........

10 specs, 0 failures
Finished in 0.026 seconds

You will notice that each of the tests basically does the same thing. There are some minor variations, such as a call to Math.abs or Math.round depending on the flags passed to divide, but otherwise it is just redundant code. If we want to add new flags for divide, we'll have to go through the same process and add more redundant test code to cover all of the cases.

Using Template however, we can abstract this code redundancy into a set of input data. Like a for loop allows us to express logic only once for a series of inputs, Template allows us to write the test code logic once and then execute it for multiple inputs.

Here are the unit tests from the previous example, but rewritten with Template:

describe("Given a Calculator", () => {

    beforeEach(function () {
        this.calculator = new Calculator();
    });

    beforeEach(function () {
        this.a = Random.number();
        this.b = Random.number();
        this.expected = this.a / this.b;
    });

    describe("when the divide method is called", Template.withInputs(["options"], (options: Options) => {

        if (options) {
            describe(`when the round flag is ${options.round} and the absolute flag is ${options.absolute}`, () => {

                beforeEach(function () {
                    this.expected = options.round ? Math.round(this.expected) : this.expected;
                    this.expected = options.absolute ? Math.abs(this.expected) : this.expected;
                });

                it("then it should return the expected value", function () {
                    expect(this.calculator.divide(this.a, this.b, options)).toEqual(this.expected);
                });
            });
        }
        else {
            it("then it should return the expected value", function () {
                expect(this.calculator.divide(this.a, this.b)).toEqual(this.expected);
            });
        }
    }, { options: undefined },
       { options: {}},
       { options: { round: true }},
       { options: { round: false }},
       { options: { absolute: true }},
       { options: { absolute: false }},
       { options: { round: true, absolute: true }},
       { options: { round: false, absolute: true }},
       { options: { round: true, absolute: false }},
       { options: { round: false, absolute: false }}
    ));
});

Output:

..........

10 specs, 0 failures
Finished in 0.026 seconds

Without losing any code coverage in our tests, we've gone from having to write 10 unit tests to only 2.

For more information, see the API reference.

Input Builder

The previous section explained Template and how it simplifies the amount of test code we have to write, but we can still do better. The input data that is passed to the templated test code often simply ends up being different permutations of the same data. For example, here's the input to the test template we wrote for Calculator.divide:

[ { options: undefined },
  { options: {}},
  { options: { round: true }},
  { options: { round: false }},
  { options: { absolute: true }},
  { options: { absolute: false }},
  { options: { round: true, absolute: true }},
  { options: { round: false, absolute: true }},
  { options: { round: true, absolute: false }},
  { options: { round: false, absolute: false }} ]

As you can see, all of the inputs are just different combinations of values (and lack of values) to test all of the possible scenarios. When we have more than two or three input properties to test, the list of input data can get long quickly. This is where InputBuilder comes in handy. InputBuilder allows you to define values for individual input properties (called fragments), and then generates a list of all possible permutations of inputs to your test template based on those fragments.

Let's see what an InputBuilder would look like for our input data above:

InputBuilder
    .fragment({ options: undefined })
    .fragment({ options: {} })
    .fragmentBuilder("options", InputBuilder.
        .fragment({ round: true })
        .fragment({ round: false })
        .fragment({ absolute: true })
        .fragment({ absolute: false })
    )
    .build();

Each fragment defines a possible permutation of a specific input property that will be used to build the final list. As we can see above, it's also possible to generate fragments from another InputBuilder using the fragmentBuilder operator.

There's still some redundancy we can eliminate, however, by using the fragmentList operator to consolidate our fragment definitions:

InputBuilder
    .fragmentList({ options: [undefined, {}] })
    .fragmentBuilder("options", InputBuilder.
        .fragmentList({ round: [true, false] })
        .fragmentList({ absolute: [true, false] })
    )
    .build();

This generates the following output:

[ { options: undefined },
  { options: {}},
  { options: { round: true }},
  { options: { round: false }},
  { options: { absolute: true }},
  { options: { absolute: false }},
  { options: { round: true, absolute: true }},
  { options: { round: false, absolute: true }},
  { options: { round: true, absolute: false }},
  { options: { round: false, absolute: false }} ]

Now, let's put it all together by modifying the Calculator.divide unit tests we wrote earlier:

const DivideTemplateInput = InputBuilder
    .fragmentList({ options: [undefined, {}] })
    .fragmentBuilder("options", InputBuilder.
        .fragmentList({ round: [true, false] })
        .fragmentList({ absolute: [true, false] })
    );

describe("Given a Calculator", () => {

    beforeEach(function () {
        this.calculator = new Calculator();
    });

    beforeEach(function () {
        this.a = Random.number();
        this.b = Random.number();
        this.expected = this.a / this.b;
    });

    describe("when the divide method is called", Template(["options"], DivideTemplateInput, (options: Options) => {

        if (options) {
            describe(`when the round flag is ${options.round} and the absolute flag is ${options.absolute}`, () => {

                beforeEach(function () {
                    this.expected = options.round ? Math.round(this.expected) : this.expected;
                    this.expected = options.absolute ? Math.abs(this.expected) : this.expected;
                });

                it("then it should return the expected value", function () {
                    expect(this.calculator.divide(this.a, this.b, options)).toEqual(this.expected);
                });
            });
        }
        else {
            it("then it should return the expected value", function () {
                expect(this.calculator.divide(this.a, this.b)).toEqual(this.expected);
            });
        }
    });
});

Output:

..........

10 specs, 0 failures
Finished in 0.026 seconds

InputBuilder also supports when clauses, which allow you to control how permutations are generated by filtering the possible combinations of input fragments.

For example, given the following class:

interface AddOptions {
    item: Item;
    front: boolean;
}

class ItemList {

    private _items: Item[];

    public readonly get items(): Item[] {
        return this._items;
    }

    public addItem(itemOrOptions: Item | AddOptions) {
        if (itemOrOptions instanceof Item || !itemOrOptions.front) {
            this._items.push(item);
        }
        else {
            this._items = [itemOrOptions.item].concat(this._items);
        }
    }
}

Now, let's say we want to write a test for addItem. If we use a Template, we might come up with this:

const AddItemTemplateInput = InputBuilder
    .fragmentList({ item: [undefined, new Item()] })
    .fragment({ options: undefined })
    .fragmentBuilder("options", InputBuilder.
        .fragment({ item: new Item() })
        .fragmentList({ front: [undefined, true, false] })
    );

describe("Given an ItemList", () => {

    beforeEach(function () {
        this.itemList = new ItemList();
        this.dummyItem = new Item();
        this.itemList.addItem(this.dummyItem);
    });

    describe("when the addItem method is called", Template(["item", "options"], DivideTemplateInput, (item?: Item, options?: AddOptions) => {

        if (item) {
            beforeEach(function () {
                this.itemList.addItem(item);
            });

            it("then it should add the item to the end of the list", function () {
                expect(this.itemList.items).toEqual([this.dummyItem, item]);
            });
        }
        else {
            beforeEach(function () {
                this.itemList.addItem(options);
            });

            if (options.front) {
                it("then it should add the item to the front of the list", function () {
                    expect(this.itemList.items).toEqual([item, this.dummyItem]);
                });
            }
            else {
                it("then it should add the item to the end of the list", function () {
                    expect(this.itemList.items).toEqual([this.dummyItem, item]);
                });
            }
        }
    });
});

This may look fine, but when we run these tests one of the cases will fail. This is because we've defined an input fragment where item is undefined and one where options is undefined. While we want these inputs independently, we never want them to both be undefined at the same time.

Also notice that we only ever read the value of options when item isn't defined. This means that when an input permutation containing a definition for both options and item is given to the template, we are effectively running a redundant test case, as the test path for when item is defined will be executed twice. While this may not mean much in this example, this concern can grow exponentially as new input fragments are added. This means potentially hundreds of redundant test cases being executed in your test suite.

These two problems are what when clauses help solve. To remedy these issues, we just need to tweak the definition of AddItemTemplateInput from the above example:

const AddItemTemplateInput = InputBuilder
    .fragment({ item: undefined }, permutation => permutation.options != undefined)
    .fragment({ item: new Item() }, permutation => permutation.options == undefined)
    .fragment({ options: undefined })
    .fragmentBuilder("options", InputBuilder.
        .fragment({ item: new Item() })
        .fragmentList({ front: [undefined, true, false] })
    );

This now says that a permutation with an undefined item property can only exist when options is defined, and vice versa.

With the additional restrictions to our item input fragments, we've eliminated the errant failure and the redundant test case.

For more information, see the API reference.

Spec

The Spec namespace contains a helper object that enables more type-safe testing by exposing a type for all test spec parameters. Spec proxies the built-in testing functions (beforeEach, afterEach, and it) and provides the test spec's parameters as a type-safe argument to the callback.

Example

import { Spec, Template, Random } from "detest-bdd";

interface CalculatorTest {
    calculator: Calculator;
    a: number;
    b: number;
    expected: number;
}

const spec = Spec.create<CalculatorTest>();

describe("Given a Calculator", () => {

    spec.beforeEach((params: CalculatorTest) => {
        params.calculator = new Calculator();
    });

    spec.beforeEach((params: CalculatorTest) => {
        params.a = Random.number();
        params.b = Random.number();
        params.expected = params.a / params.b;
    });

    describe("when the divide method is called", Template(["options"], (options: Options) => {

        if (options) {
            describe(`when the round flag is ${options.round} and the absolute flag is ${options.absolute}`, () => {

                spec.beforeEach((params: CalculatorTest) => {
                    params.expected = options.round ? Math.round(params.expected) : params.expected;
                    params.expected = options.absolute ? Math.abs(params.expected) : params.expected;
                });

                spec.it("then it should return the expected value", (params: CalculatorTest) => {
                    expect(params.calculator.divide(params.a, params.b, options)).toEqual(params.expected);
                });
            });
        }
        else {
            spec.it("then it should return the expected value", (params: CalculatorTest) => {
                expect(params.calculator.divide(params.a, params.b)).toEqual(params.expected);
            });
        }
    }, { options: undefined },
       { options: {}},
       { options: { round: true }},
       { options: { round: false }},
       { options: { absolute: true }},
       { options: { absolute: false }},
       { options: { round: true, absolute: true }},
       { options: { round: false, absolute: true }},
       { options: { round: true, absolute: false }},
       { options: { round: false, absolute: false }}
    ));
});

Spec also replaces the traditional injection of the doneFn for asynchronous tests in favor of returning a Promise from the callback when waiting for an asynchronous task.

Example

declare var FooService: {
    getResult: Promise<FooResult>;
};

describe("Given a FooService", () => {

    describe("when getResult is called", () => {

        spec.beforeEach((params: FooServiceTest): Promise<any> => {
            params.fooParams = {
                input: "foobar"
            };

            return FooService.getResult(params.fooParams).then(result => params.fooResult = result);
        });

        spec.it("should return the expected result", (params: FooServiceTest) => {
            expect(params.fooResult.output).toEqual("foobaz");
        });
    });
});

For more information, see the API reference.

API

Template

interface Template<T extends object> {
    paramNames: string[];
    invoke: Template.InvokeFn<T>;
    run: Template.RunFn<T>;
}

Represents a template, which is an object that takes an input and runs a block of test code with that input.

  • paramNames The ordered list of input parameter names that the template callback will use as arguments.
  • invoke A function that can be called to invoke the template once. For more information, see Template.InvokeFn.
  • run A function that can be called to run the template with multiple inputs. For more information, see Template.RunFn.
function Template<T extends object>(paramNames: string[], input: InputBuilder<T> | inputBuilder<T>[], callback: Template.CallbackFn): () => void;

Shorthand method that creates a Template and returns a function that executes the template callback with the inputs from the given InputBuilder when called.

  • paramNames The ordered list of input parameter names that the template callback will use as arguments. See Template.
  • input The InputBuilder(s) that will be used to generate the template inputs. For more information, see InputBuilder.
  • callback The callback that contains the test code to execute. See Template.CallbackFn.

Returns a Function that when called will execute the template with the template callback with the inputs from the given InputBuilder.

Template.withInputs

function Template.withInputs<T extends object>(paramNames: string[], callback: Template.CallbackFn, ...paramsList: T[]): () => void;

Shorthand method that creates a Template and returns a function that executes it with the given paramsList when called.

  • paramNames The ordered list of input parameter names that the template callback will use as arguments. See Template.
  • callback The callback that contains the test code to execute. See Template.CallbackFn.
  • paramsList The list of input values to pass into the template callback. For more information, see Template.RunFn.

Returns a Function that when called will execute the template with the given paramsList.

Template.create

function Template.create<T extends object>(paramNames: string[], callback: CallbackFn): Template<T>;

Creates a new Template object.

  • paramNames The ordered list of input parameter names that the template callback will use as arguments. See Template.
  • callback The callback that contains the test code to execute. See Template.CallbackFn.

Returns a new Template.

Template.CallbackFn

type Template.CallbackFn = (...paramList: any[]) => void;
  • paramList The list of inputs to the template that correspond to paramNames.

Template.InvokeFn

type Template.InvokeFn<T extends object> = (params: T) => void;
  • params An object containing key-value pairs corresponding to paramNames.

Template.RunFn

type Template.RunFn<T extends object> = (...paramsList: T[]) => void;
  • paramsList The list of objects containing key-value pairs corresponding to paramNames.

Input Builder

class InputBuilder<T> {

  public static fragment<T>(fragmentDictionary: FragmentDictionary<T>, when?: FragmentWhenFn<T>): InputBuilder<T>;

  public static fragmentList<T>(fragmentListDictionary: FragmentListDictionary<T>, when?: FragmentWhenFn<T>): InputBuilder<T>;

  public static fragmentBuilder<T, _T>(key: keyof T, builder: InputBuilder<_T>, when?: FragmentWhenFn<T>): InputBuilder<T>;

  public get dictionary(): Readonly<FragmentListDictionary<T>>;

  public build(): T[];

  public fragment(fragmentDictionary: FragmentDictionary<T>, when?: FragmentWhenFn<T>): InputBuilder<T>;

  public fragmentList(fragmentListDictionary: FragmentListDictionary<T>, when?: FragmentWhenFn<T>): InputBuilder<T>;

  public fragmentBuilder<_T>(key: keyof T, builder: InputBuilder<_T>, when?: FragmentWhenFn<T>): InputBuilder<T>;
}

InputBuilder.fragment

public fragment(fragmentDictionary: FragmentDictionary<T>, when?: FragmentWhenFn<T>): InputBuilder<T>;
public static fragment<T>(fragmentDictionary: FragmentDictionary<T>, when?: FragmentWhenFn<T>): InputBuilder<T>;

Adds all fragment definitions to the builder.

  • fragmentDictionary The fragment dictionary to add to the builder. A fragment dictionary is an object where the keys are keyof T (the input properties) and the values are possible permutations to be generated by the builder. For more information, see InputBuilder.FragmentDictionary.
  • when [Optional] The when clause to be executed when building the permutation list. For more information, see InputBuilder.FragmentWhenFn.

InputBuilder.fragmentList

public fragmentList(fragmentListDictionary: FragmentListDictionary<T>, when?: FragmentWhenFn<T>): InputBuilder<T>;
public static fragmentList<T>(fragmentListDictionary: FragmentListDictionary<T>, when?: FragmentWhenFn<T>): InputBuilder<T>;

Adds all fragment list definitions to the builder.

  • fragmentListDictionary The fragment list dictionary definitions to add to the builder. A fragment list dictionary is an object where the keys are keyof T (the input properties) and the values are arrays of the possible permutations to be generated by the builder. For more information, see InputBuilder.FragmentListDictionary.
  • when [Optional] The when clause to be executed when building the permutation list. For more information, see InputBuilder.FragmentWhenFn.

InputBuilder.fragmentBuilder

public fragmentBuilder<_T>(key: keyof T, builder: InputBuilder<_T>, when?: FragmentWhenFn<T>): InputBuilder<T>;
public static fragmentBuilder<T, _T>(key: keyof T, builder: InputBuilder<_T>, when?: FragmentWhenFn<T>): InputBuilder<T>;

Adds all permutations from another InputBuilder as input fragments for this builder.

  • key The property key that the fragment values will be added to.
  • builder The InputBuilder to read the fragment values from.
  • when [Optional] The when clause to be executed when building the permutation list. For more information, see InputBuilder.FragmentWhenFn.

InputBuilder.build

public build(): T[]

Builds a list of all possible permutations for T using the given input fragments.

Returns an array of all possible permutations for the given input fragments.

InputBuilder.Fragment

type Fragment<T, P extends keyof T> = T[P];

A fragment is a possible value for a given property of T.

InputBuilder.FragmentDictionary

type FragmentDictionary<T> = { [P in keyof T]?: Fragment<T, P> };

A fragment dictionary is an object where the keys are keyof T (the input properties) and the values are possible values for those input properties.

InputBuilder.FragmentList

type FragmentList<T, P extends keyof T> = Fragment<T, P>[];

A fragment list is an array of possible value for a given property of T.

InputBuilder.FragmentListDictionary

type FragmentListDictionary<T> = { [P in keyof T]?: FragmentList<T, P> };

A fragment list dictionary is an object where the keys are keyof T (the input properties) and the values are arrays of possible values for those input properties.

InputBuilder.FragmentWhenFn

type FragmentWhenFn<T> = (input: Permutation<T>) => boolean;

A when function allows control over permutation generation by controlling which fragments are allowed to exist together. The function is executed during building of the permutation list.

  • input The current permutation being inspected. This object is a permutation of the given input fragments.

The when function should return a boolean value that if true will keep the given input permutation and if false will discard the input permutation.

Spec

interface Spec<T> {
    beforeEach(callback: Spec.Callback<T>): void;
    afterEach(callback: Spec.Callback<T>): void;
    it(description: string, callback: Spec.Callback<T>): void;

    beforeAll?(callback: Spec.StatelessCallback): void;
    before?(callback: Spec.StatelessCallback): void;
    afterAll?(callback: Spec.StatelessCallback): void;
    after?(callback: Spec.StatelessCallback): void;

    xit?(description: string, callback: Spec.Callback<T>): void;
    fit?(description: string, callback: Spec.Callback<T>): void;
}

An object that provides a simple type-safe and async-enabled wrapper around functions used in a test spec. For more information, see Spec.Callback and Spec.StatelessCallback.

The following proxy functions are provided:

  • beforeEach
  • afterEach
  • it
  • beforeAll

    NOTE: This function is only available in test runners that natively support beforeAll.

  • before

    NOTE: This function is only available in test runners that natively support before.

  • afterAll

    NOTE: This function is only available in test runners that natively support afterAll.

  • after

    NOTE: This function is only available in test runners that natively support after.

  • xit

    NOTE: This function is only available in test runners that natively support xit.

  • fit

    NOTE: This function is only available in test runners that natively support fit.

Spec.create

function Spec.create<T>(): Spec<T>;

Creates a new Spec with the given T.

Spec.inject

function Spec.inject<T>(callback: Spec.Callback<T>): (doneFn: () => void) => void;

Provides a function that when called injects a type-safe collection of properties as the first argument to the callback. For more information, see Spec.Callback.

  • callback The callback function that will receive the injected properties. For more information, see Spec.Callback.

Spec.Callback

type Spec.Callback<T> = (params: T) => Promise<void> | void;

Represents a potentially async-enabled callback that receives a type-safe collection of test properties as the first argument.

  • params The test properties for the spec.

If the callback doesn't return a value, the test context will complete immediately. If the callback returns a Promise, the test context will wait for the returned Promise to resolve or reject before completing.

Spec.StatelessCallback

type Spec.StatelessCallback = () => Promise<void> | void;

Represents a stateless, potentially async-enabled callback.

If the callback doesn't return a value, the test context will complete immediately. If the callback returns a Promise, the test context will wait for the returned Promise to resolve or reject before completing.

Random

Random.number

function Random.number(min?: number, max?: number): number;

Returns a random number between min and max. min defaults to 0 and max defaults to MAX_VALUE.

Random.integer

function Random.integer(min?: number, max?: number): number;

Returns a random integer between min and max. min defaults to 0 and max defaults to MAX_SAFE_INTEGER.

Random.boolean

function Random.boolean(): boolean;

Returns a random boolean value.

Random.string

function Random.string(minLength?: number, maxLength?: number, options?: StringOptions): string;

Returns a random string between minLength and maxLength. minLength defaults to 0 and maxLength defaults to 20.

If options.alpha is true, the string will contain letters. This is true by default.

If options.numeric is true, the string will contain letters. This is true by default.

If both options.alpha and options.numeric are false, an error will be thrown.