StoryShots + Puppeteer
Add the following modules into your app.
npm install @storybook/addon-storyshots-puppeteer puppeteer --save-dev
puppeteer
is no longer included in the addon dependencies and must be added to your project directly.
When running Puppeteer tests for your stories, you have two options:
- Have a storybook running (ie. accessible via http(s), for instance using
npm run storybook
) - Have a static build of the storybook (for instance, using
npm run build-storybook
)
Then you will need to reference the storybook URL (file://...
if local, http(s)://...
if served)
Allows to define arbitrary Puppeteer tests as story.parameters.puppeteerTest
function.
You can either create a new Storyshots instance or edit the one you previously used:
import initStoryshots from '@storybook/addon-storyshots';
import { puppeteerTest } from '@storybook/addon-storyshots-puppeteer';
initStoryshots({ suite: 'Puppeteer storyshots', test: puppeteerTest() });
Then, in your stories:
export const myExample = () => {
...
};
myExample.parameters = {
async puppeteerTest(page) {
const element = await page.$('<some-selector>');
await element.click();
expect(something).toBe(something);
},
};
This will assume you have a storybook running on at http://localhost:6006. Internally here are the steps:
- Launches a Chrome headless using puppeteer
- Browses each stories (calling http://localhost:6006/iframe.html?... URL),
- Runs the
parameters.puppeteerTest
function if it's defined.
If you want to set specific storybook URL, you can specify via the storybookUrl
parameter, see below:
import initStoryshots from '@storybook/addon-storyshots';
import { puppeteerTest } from '@storybook/addon-storyshots-puppeteer';
initStoryshots({
suite: 'Puppeteer storyshots',
test: puppeteerTest({ storybookUrl: 'http://my-specific-domain.com:9010' }),
});
The above config will use https://my-specific-domain.com:9010 for tests. You can also use query parameters in your URL (e.g. for setting a different background for your storyshots, if you use @storybook/addon-backgrounds
).
You may also use a local static build of storybook if you do not want to run the webpack dev-server:
import initStoryshots from '@storybook/addon-storyshots';
import { puppeteerTest } from '@storybook/addon-storyshots-puppeteer';
initStoryshots({
suite: 'Puppeteer storyshots',
test: puppeteerTest({
storybookUrl: 'file:///path/to/my/storybook-static',
// storybookUrl: 'file://${path.resolve(__dirname, '../storybook-static')}'
}),
});
You might use getGotoOptions
to specify options when the storybook is navigating to a story (using the goto
method). Will be passed to Puppeteer .goto() fn
import initStoryshots from '@storybook/addon-storyshots';
import { puppeteerTest } from '@storybook/addon-storyshots-puppeteer';
const getGotoOptions = ({ context, url }) => {
return {
waitUntil: 'networkidle0',
};
};
initStoryshots({
suite: 'Puppeteer storyshots',
test: puppeteerTest({ storybookUrl: 'http://localhost:6006', getGotoOptions }),
});
You might use the browserLaunchOptions
to specify options for the default browser instance. Will be passed to puppeteer.launch()
import initStoryshots from '@storybook/addon-storyshots';
import { puppeteerTest } from '@storybook/addon-storyshots-puppeteer';
initStoryshots({
suite: 'Puppeteer storyshots',
test: puppeteerTest({
storybookUrl: 'https://some-local-ssl-url:7777',
browserLaunchOptions: {
// For ignoring self-signed certificates
ignoreHTTPSErrors: true,
},
}),
});
You might use chromeExecutablePath
to specify the path to a different version of Chrome, without downloading Chromium. Will be passed to Runs a bundled version of Chromium
import initStoryshots from '@storybook/addon-storyshots';
import { puppeteerTest } from '@storybook/addon-storyshots-puppeteer';
const chromeExecutablePath = '/usr/local/bin/chrome';
initStoryshots({
suite: 'Puppeteer storyshots',
test: puppeteerTest({ storybookUrl: 'http://localhost:6006', chromeExecutablePath }),
});
Alternatively, you may set the SB_CHROMIUM_PATH
environment variable. If both are set, then chromeExecutablePath
will take precedence.
You might use the async getCustomBrowser
function to obtain a custom instance of a Puppeteer browser
object. This will prevent storyshots-puppeteer
from creating its own browser
. It will create and close pages within the browser
, and it is your responsibility to manage the lifecycle of the browser
itself.
import initStoryshots from '@storybook/addon-storyshots';
import { puppeteerTest } from '@storybook/addon-storyshots-puppeteer';
import puppeteer from 'puppeteer';
(async function () {
initStoryshots({
suite: 'Puppeteer storyshots',
test: puppeteerTest({
storybookUrl: 'http://localhost:6006',
getCustomBrowser: () => puppeteer.connect({ browserWSEndpoint: 'ws://yourUrl' }),
}),
});
})();
Sometimes, there is a need to customize a page before it calls the goto
api.
An example of device emulation:
import initStoryshots from '@storybook/addon-storyshots';
import { puppeteerTest } from '@storybook/addon-storyshots-puppeteer';
const devices = require('puppeteer/DeviceDescriptors');
const iPhone = devices['iPhone 6'];
function customizePage(page) {
return page.emulate(iPhone);
}
initStoryshots({
suite: 'Puppeteer storyshots',
test: puppeteerTest({
storybookUrl: 'http://localhost:6006',
customizePage,
}),
});
By default, @storybook/addon-storyshots-puppeteer
uses 15 second timeouts for browser setup and test functions.
Those can be customized with setupTimeout
and testTimeout
parameters.
You may want to use another Jest project to run your Puppeteer storyshots as they require more resources: Chrome and Storybook built/served.
Integrate Puppeteer storyshots with Create React App
You have two options here, you can either:
-
Add the storyshots configuration inside any of your
test.js
file. You must ensure you have either a running storybook or a static build available. -
Create a custom test file using Jest outside of the CRA scope:
A more robust approach would be to separate existing test files ran by create-react-app (anything
(test|spec).js
suffixed files) from the test files to run Puppeteer storyshots. This use case can be achieved by using a custom name for the test file, ie something likepuppeteer-storyshots.runner.js
. This file will contain theinitStoryshots
call with Puppeteer storyshots configuration. Then you will create a separate script entry in your package.json, for instance{ "scripts": { "puppeteer-storyshots": "jest puppeteer-storyshots.runner.js --config path/to/custom/jest.config.json" } }
Note that you will certainly need a custom config file for Jest as you run it outside of the CRA scope and thus you do not have the built-in config.
Once that's setup, you can run
npm run puppeteer-storyshots
.
Puppeteer launches a web browser (Chrome) internally.
The browser opens a page (either using the static build of storybook or a running instance of Storybook)
If you run your test without either the static build or a running instance, this wont work.
To make sure your tests run against the latest changes of your Storybook, you must keep your static build or running Storybook up-to-date.
This can be achieved by adding a step before running the test ie: npm run build-storybook && npm run image-snapshots
.
If you run the Puppeteer storyshots against a running Storybook in dev mode, you don't have to worry about the stories being up-to-date because the dev-server is watching changes and rebuilds automatically.
Runs Axe accessibility checks and verifies that they pass using jest-puppeteer-axe.
import initStoryshots from '@storybook/addon-storyshots';
import { axeTest } from '@storybook/addon-storyshots-puppeteer';
initStoryshots({ suite: 'A11y checks', test: axeTest() });
For configuration, it uses the same story.parameters.a11y
parameter as @storybook/addon-a11y
import initStoryshots from '@storybook/addon-storyshots';
import { axeTest } from '@storybook/addon-storyshots-puppeteer';
const beforeAxeTest = (page, { context: { kind, story }, url }) => {
return new Promise((resolve) =>
setTimeout(() => {
resolve();
}, 600)
);
};
initStoryshots({ suite: 'A11y checks', test: axeTest({ beforeAxeTest }) });
beforeAxeTest
receives the Puppeteer page instance and an object: { context: {kind, story}, url}
. kind is the kind of the story and the story its name. url is the URL the browser will use to screenshot. beforeAxeTest
is part of the promise chain and is called after the browser navigation is completed but before the screenshot is taken. It allows for triggering events on the page elements and delaying the axe test .
Generates and compares screenshots of your stories using jest-image-snapshot.
import initStoryshots from '@storybook/addon-storyshots';
import { imageSnapshot } from '@storybook/addon-storyshots-puppeteer';
initStoryshots({ suite: 'Image storyshots', test: imageSnapshot() });
It saves all images under __image_snapshots__ folder.
If you wish to customize jest-image-snapshot, then you can provide a getMatchOptions
parameter that should return the options config object. Additionally, you can provide beforeScreenshot
which is called before the screenshot is captured and a afterScreenshot
handler which is called after the screenshot and receives the just created image.
import initStoryshots from '@storybook/addon-storyshots';
import { imageSnapshot } from '@storybook/addon-storyshots-puppeteer';
const getMatchOptions = ({ context: { kind, story }, url }) => {
return {
failureThreshold: 0.2,
failureThresholdType: 'percent',
};
};
const beforeScreenshot = (page, { context: { kind, story }, url }) => {
return new Promise((resolve) =>
setTimeout(() => {
resolve();
}, 600)
);
};
const afterScreenshot = ({ image, context }) => {
return new Promise((resolve) =>
setTimeout(() => {
resolve();
}, 600)
);
};
initStoryshots({
suite: 'Image storyshots',
test: imageSnapshot({
storybookUrl: 'http://localhost:6006',
getMatchOptions,
beforeScreenshot,
afterScreenshot,
}),
});
getMatchOptions
receives an object: { context: {kind, story}, url}
. kind is the kind of the story and the story its name. url is the URL the browser will use to screenshot.
beforeScreenshot
receives the Puppeteer page instance and an object: { context: {kind, story}, url}
. kind is the kind of the story and the story its name. url is the URL the browser will use to screenshot. beforeScreenshot
is part of the promise chain and is called after the browser navigation is completed but before the screenshot is taken. It allows for triggering events on the page elements and delaying the screenshot and can be used avoid regressions due to mounting animations.
afterScreenshot
receives the created image from puppeteer.
You might use getScreenshotOptions
to specify options for screenshot. Will be passed to Puppeteer .screenshot() fn
import initStoryshots from '@storybook/addon-storyshots';
import { imageSnapshot } from '@storybook/addon-storyshots-puppeteer';
const getScreenshotOptions = ({ context, url }) => {
return {
encoding: 'base64', // encoding: 'base64' is a property required by puppeteer
fullPage: false, // Do not take the full page screenshot. Default is 'true' in Storyshots.,
};
};
initStoryshots({
suite: 'Image storyshots',
test: imageSnapshot({ storybookUrl: 'http://localhost:6006', getScreenshotOptions }),
});
getScreenshotOptions
receives an object { context: {kind, story}, url}
. kind is the kind of the story and the story its name. url is the URL the browser will use to screenshot.
To create a screenshot of just a single element (with its children), rather than the page or current viewport, an ElementHandle can be returned from beforeScreenshot
:
import initStoryshots from '@storybook/addon-storyshots';
import { imageSnapshot } from '@storybook/addon-storyshots-puppeteer';
const beforeScreenshot = (page) => page.$('#storybook-root > *');
initStoryshots({
suite: 'Image storyshots',
test: imageSnapshot({ storybookUrl: 'http://localhost:6006', beforeScreenshot }),
});