README.md

Observable test helper

This is an abstraction library for testing individual observables or react-redux epics.

Testing Epics

To test your epics use the EpicTestBuilder

import {myTestEpic} from 'somewhere'

it(`FooBars the foo bar`, (done) => {
    const expectedAssertions = (result)=>{
        expect(result).toEqual();
    };
    const expectedErrorAssertions = (error) => {
        expect(error).toEqual();
    }
    const epicTest = new EpicTestBuilder()
        .withEpic(myTestEpic)
        .withDone(done)
        .withTriggerAction({type:'...',payload:{...}})
        .withApi(mockedAPI)
        .withStore(mockedStore)
        .withExpectedOutputAssertion(expectedAssertions)
        .withExpectedExceptionAssertion(expectedErrorAssertions)
        .build();

    epicTest.testRunToCompletion();
});

withEpic, withDone, withTriggerAction - are required and at least one of the withExpectedOutputAssertion or withExpectedExceptionAssertion is required depending if you are asserting error or success, use both if you are asserting both epic throwing errors and emitting values.

withApi, withStore are optional as not every epic require API or Store mocks.

The resulting EpicTester have one method available testRunToCompletion() opposite to ObservableTester which also has testRunningObservable().

testRunningObservable() is meant to test observables that don't complete.

It will run until your epic either throws an error or you manually unsubscribe from it. The method accepts a number parameter which attaches take(number) to your test which will wait for number amount of emits and complete the running observable for you.

If you are testing the observable that runs to completion testRunToCompletion() should be used.

Both of these methods return Subscription which allow you to trigger unsubscribe().

Testing observables

To test your epics use the ObservableTestBuilder

import {myTestObservable} from 'somewhere'

it(`FooBars the foo bar`, (done) => {
    const expectedAssertions = (result)=>{
        expect(result).toEqual();
    };
    const expectedErrorAssertions = (error) => {
        expect(error).toEqual();
    }
    const observableTest = new ObservableTestBuilder()
        .withObservable(myTestObservable)
        .withDone(done)
        .withExpectedOutputAssertion(expectedAssertions)
        .withExpectedExceptionAssertion(expectedErrorAssertions)
        .build();

    observableTest.testRunToCompletion();
});

Pretty much the same rules apply just how much mocking you need to pass is reduced.

Happy testing