API
Marko Testing Library
re-exports everything from DOM Testing Library
as well as these methods:
render
function render(
template, // A Marko template to render
input, // Input for the above template
options // You won't often use this, expand below for docs on options
)
Render into a container which is appended to document.body
.
import {render} from '@marko/testing-library'
import MyTemplate from './my-template.marko'
render(MyTemplate)
import {render, screen} from '@marko/testing-library'
import Greeting from './greeting.marko'
test('renders a message', async () => {
const {container} = await render(Greeting, {name: 'Marko'})
expect(screen.getByText(/Marko/)).toBeInTheDocument()
expect(container.firstChild).toMatchInlineSnapshot(`
<h1>Hello, Marko!</h1>
`)
})
render
Options
You won't often need to specify options, but if you ever do, here are the available options which you can provide as the third argument to render
.
container
By default for client-side tests, Marko Testing Library
will create a div
and append that div
to the document.body
and this is where your component will be rendered. If you provide your own HTMLElement container
via this option, it will not be appended to the document.body
automatically.
For example: If you are unit testing a tablebody
element, it cannot be a child of a div
. In this case, you can specify a table
as the render container
.
const table = document.createElement('table')
const {container} = await render(MyTableBody, null, {
container: document.body.appendChild(table),
})
render
Result
The render
method returns a promise which resolves with an object that has a few properties:
...queries
The most important feature of render
is that the queries from the Core API are automatically returned with their first argument bound to the results of rendering your component.
See Queries for a complete list.
Example
const {getByLabelText, queryAllByTestId} = await render(MyTemplate)
Alternatively, you can use the top-level screen
method to query into all currently rendered components in the document.body
, eg:
import { render, screen } from "@marko/testing-library"
await render(MyTemplate)
const el = screen.getByText(...)
debug
This method is a shortcut for logging the prettyDOM
for all children inside of the container
.
import {render} from '@marko/testing-library'
import Greeting from './greeting.marko'
const {debug} = await render(Greeting, {name: 'World'})
debug()
// <h1>Hello World</h1>
// you can also pass an element: debug(getByTestId('messages'))
This is a simple wrapper around prettyDOM
which is also exposed and comes from DOM Testing Library
.
rerender
A Marko components input
can change at any time from a parent component. Although often this input is passed through your component declaratively, sometimes it is necessary to ensure that your components react appropriately to new data. You can simulate your component receiving new input
by passing new data to the rerender
helper.
import {render} from '@marko/testing-library'
import Greeting from './greeting.marko'
const {rerender, debug} = await render(Greeting, {name: 'World'})
// re-render the same component with different props
await rerender({name: 'Marko'})
debug()
// <h1>Hello Marko</h1>
emitted
Marko components also communicate with their parents through events. It is recommended to also test that your components emit the right events at the right time.
The emitted
helper does just that. Calling the helper will return all emitted events since the last call to the helper. You can also pass in an event type to filter the results.
import {render, fireEvent} from '@marko/testing-library'
import Counter from './counter.marko'
const {getByText, emitted} = await render(Counter)
const button = getByText('Increment')
await fireEvent.click(button)
await fireEvent.click(button)
// Assuming the `Counter` component forwards these button clicks as `increment` events
expect(emitted('increment')).toHaveProperty('length', 2)
await fireEvent.click(button)
// Note: the tracked events are cleared every time you read them.
// Below we are snapshoting the events after our last assertion,
// the return value will include an array with all of the arguments for each increment event.
expect(emitted('increment')).toMatchInlineSnapshot(`
Array [
Array [
Object {
"count": 3,
},
],
]
`)
// Without an event type will give you all events with their type and arguments.
expect(emitted()).toMatchInlineSnapshot(`
Array [
Object {
"args": Array [
Object {
"count": 0,
},
],
"type": "increment",
},
Object {
"args": Array [
Object {
"count": 1,
},
],
"type": "increment",
},
Object {
"args": Array [
Object {
"count": 3,
},
],
"type": "increment",
}
]
`)
cleanup
Like the top-level cleanup method, this allows you to remove and destroy the currently rendered component before the test has been completed.
This can be useful to validate that a component properly cleans up any DOM mutations once it has been destroyed.
import {render, screen, getRoles} from '@marko/testing-library'
import Main from './main.marko'
import Dialog from './dialog.marko'
await render(Main)
const main = screen.getByRole('main')
expect(main).not.toHaveAttribute('aria-hidden')
const {cleanup} = await render(Dialog)
expect(main).toHaveAttribute('aria-hidden') // assert added attribute
cleanup() // destroy the dialog
expect(main).not.toHaveAttribute('aria-hidden') // assert attribute removed
container
The containing DOM node of your rendered Marko Component. For server-side tests this is a JSDOM.fragment, and for client-side tests this will be whatever is passed as the container
render option.
Tip: To get the root element of your rendered element, use
container.firstChild
.
🚨 If you find yourself using
container
to query for rendered elements then you should reconsider! The other queries are designed to be more resilient to changes that will be made to the component you're testing. Avoid usingcontainer
to query for elements!
fireEvent
Because Marko batches DOM updates to avoid unnecessary re-renders, the fireEvent helpers are re-exported as async
functions. Awaiting this ensures that the DOM has properly updated in response to the event triggered in the test.
await fireEvent.click(getByText('Click me'))
cleanup
With client-side tests your components are rendered into a placeholder HTMLElement. To ensure that your components are properly removed, and destroyed, after each test the cleanup
method is called for you automatically by hooking into afterEach
in supported test frameworks. You can also manually call cleanup
at any time which will remove all attached components.
import {render, cleanup, screen} from '@marko/testing-library'
import Greeting from './greeting.marko'
await render(Greeting, {name: 'Marko'})
expect(screen.getByText(/Marko/)).toBeInTheDocument()
// manually cleanup the component before the test is finished
cleanup()
expect(screen.queryByText(/Marko/)).toBeNull()
You can turn off the automatic test cleanup by importing the following module:
import '@marko/testing-library/dont-cleanup-after-each'
With mocha you can use mocha --require @marko/testing-library/dont-cleanup-after-each
as a shorthand.
If you are using Jest, you can include setupFilesAfterEnv: ["@marko/testing-library/dont-cleanup-after-each"]
in your Jest config to avoid doing this in each file.