lighthouse-reporting
v1.6.7
Published
> Generate Lighthouse reports in HTML, CSV, and JSON formats, optimized for publishing in Jenkins or GitHub Actions.
Downloads
75
Readme
lighthouse-reporting
Generate Lighthouse reports in HTML, CSV, and JSON formats, optimized for publishing in Jenkins or GitHub Actions.
The reports include trend history support, allowing you to track performance improvements over time.
Usage
Examples of usage with Playwright + Lighthouse (and Storybook).
npm i -D lighthouse playwright-lighthouse lighthouse-reporting
In your frontend or testing framework
The example of usage playwright and playwright-lighthouse together.
import path from 'path'
import { playAudit } from 'playwright-lighthouse'
import { playwrightLighthouseTest, getScores, writeCsvResult, writeHtmlListEntryWithRetry, LighthouseResult } from 'lighthouse-reporting'
import { MyPage1 } from '../../pages/my-page-1.page.js'
import { MyPage2 } from '../../pages/my-page-1.page.js'
playwrightLighthouseTest.setTimeout(60000)
const reportDir = path.join(process.cwd(), 'lighthouse')
const htmlFilePath = path.join(reportDir, 'index.html')
const swimlanes = ['performance']
const lighthousePages = [
{ name: 'MyPage1', po: MyPage1, thresholds: { performance: 80, accessibility: 88, seo: 92 }, swimlanes },
{ name: 'MyPage2', po: MyPage2, thresholds: { performance: 90, accessibility: 100, seo: 90 }, swimlanes },
]
lighthousePages.forEach(({ name, po, thresholds, swimlanes }) => {
playwrightLighthouseTest(name, async ({ context, port, baseURL }) => {
const onlyCategories = ['accessibility', 'seo', 'performance']
// make sure to acess context or page at least once
// to let playwright initialize context!
context // this is enough to make the test work
// note: importing LighthouseResult is not needed since [email protected]
const result: LighthouseResult = await playAudit({
url: baseURL + po.getPath('123'),
port,
thresholds,
reports: {
formats: {
html: true,
},
name,
directory: reportDir,
},
opts: {
onlyCategories,
screenEmulation: { disabled: true },
},
disableLogs: true,
ignoreError: true,
})
const scores = getScores(result)
await writeCsvResult(reportDir, name, scores, thresholds, swimlanes)
await writeHtmlListEntryWithRetry(htmlFilePath, name, scores, thresholds, result.comparisonError)
if (result.comparisonError) {
throw new Error(result.comparisonError)
}
})
})
Or in Storybook
An example for storybook with lighthouse and screenshot testing
import path from 'path'
import { BrowserContext } from '@playwright/test'
import { playAudit } from 'playwright-lighthouse'
import {
playwrightLighthouseTest,
getScores,
writeCsvResult,
writeHtmlListEntryWithRetry,
LighthouseResult,
StorybookIndexStory,
storybookPlaywright,
writeScoresToJson
} from 'lighthouse-reporting'
playwrightLighthouseTest.setTimeout(60000)
const lhScoresDir = path.join(process.cwd(), process.env.LH_SCORES_DIR || 'lh-scores')
const csvReportDir = path.join(process.cwd(), process.env.LH_CSV_REPORT_DIR || 'lighthouse')
const htmlReportDir = path.join(process.cwd(), process.env.LH_REPORT_DIR || 'lighthouse')
const htmlFilePath = path.join(htmlReportDir, 'index.html')
// use stories.json instead of index.json for storybook v6
// make sure to set buildStoriesJson to true in storybook main.js feature section
const stories = storybookPlaywright.getStories('./storybook-static/index.json', (story) => {
// skip docs, etc
if (story.type !== 'story') {
return false
}
// only include stories with test tag
if (!story.tags.includes('test')) {
return false
}
return true
})
stories.forEach((story) => {
playwrightLighthouseTest(`${story.title} - ${story.name}`, async ({ context, port, baseURL }) => {
await runLighthouse(story, context, port, baseURL)
await storybookPlaywright.captureScreenshot(story, context)
})
})
const runLighthouse = async (story: StorybookIndexStory, context: BrowserContext, port: number, baseURL?: string) => {
const onlyCategories = ['accessibility']
const thresholds = { accessibility: 100 }
const name = story.id
// make sure to acess context or page at least once
// to let playwright initialize context!
context // this is enough to make the test work
// or manually open the page
// const page = context.pages()[0]
// await page.goto(`/iframe.html?id=${story.id}`)
// note: importing LighthouseResult is not needed since [email protected]
const result: LighthouseResult = await playAudit({
url: baseURL + `/iframe.html?id=${story.id}`,
// page, // alternatevely, path the page instead of the `url`
port,
thresholds,
reports: {
formats: {
html: true,
},
name,
directory: htmlReportDir,
},
opts: {
onlyCategories,
screenEmulation: { disabled: true },
},
disableLogs: true,
ignoreError: true,
})
const scores = getScores(result)
await writeCsvResult(csvReportDir, name, scores, thresholds)
await writeHtmlListEntryWithRetry(htmlFilePath, name, scores, thresholds, result.comparisonError)
// write score results in JSON, allows generating the Average csv report
await writeScoresToJson(lhScoresDir, name, scores, result)
}
Configs examples
import { PlaywrightTestConfig } from '@playwright/test'
const baseURL = 'http://127.0.0.1:6009'
// process.env.LH_REPORT_DIR = 'lighthouse-html' // adjust lighthouse output folder if required
// process.env.LH_CSV_REPORT_DIR = 'lighthouse-csv' // adjust lighthouse csv report folder if required
// process.env.LH_SCORES_DIR = 'lh-scores' // to write and store scores in json format or write average report
const config: PlaywrightTestConfig = {
use: {
viewport: { width: 1280, height: 820 },
ignoreHTTPSErrors: true,
acceptDownloads: false,
trace: 'off',
baseURL,
screenshot: { mode: 'off' },
},
projects: [
{
name: 'chromium',
use: {
browserName: 'chromium',
launchOptions: { args: ['--disable-gpu'] },
},
retries: 0,
},
],
expect: { toMatchSnapshot: { threshold: 0.2 } },
reporter: 'line',
testDir: 'test/storybook',
testMatch: '*.spec.ts',
fullyParallel: true,
globalSetup: './src/global-setup.ts',
globalTeardown: './src/global-teardown.ts',
forbidOnly: true,
webServer: [
{
command: 'npx http-server ./storybook-static --port 6009 --silent',
url: `${baseURL}/index.json`,
timeout: 15 * 1000,
reuseExistingServer: false,
ignoreHTTPSErrors: true,
},
],
}
export default config
import path from 'path'
import fs from 'fs/promises'
import { lighthouseSetup } from 'lighthouse-reporting'
const lhScoresDir = path.join(process.cwd(), process.env.LH_SCORES_DIR || 'lh-scores')
const csvReportDir = path.join(process.cwd(), process.env.LH_CSV_REPORT_DIR || 'lighthouse')
async function globalSetup() {
await fs.mkdir(lhScoresDir, { recursive: true })
await fs.mkdir(csvReportDir, { recursive: true })
await lighthouseSetup()
}
export default globalSetup
import path from 'path'
import { lighthousePlaywrightTeardown, buildAverageCsv } from 'lighthouse-reporting'
const lhScoresDir = path.join(process.cwd(), process.env.LH_SCORES_DIR || 'lh-scores')
const csvReportDir = path.join(process.cwd(), process.env.LH_CSV_REPORT_DIR || 'lighthouse')
async function globalTeardown() {
await lighthousePlaywrightTeardown()
await buildAverageCsv(lhScoresDir, csvReportDir)
}
export default globalTeardown
Jenkins
Plugins used HTML Publisher and Plot.
stage('Lighthouse') {
steps {
# run playwright-lighthouse tests
}
post {
always {
# html report per build
publishHTML(target: [
reportName : 'Lighthouse',
reportDir : "$WORKSPACE/lighthouse",
reportFiles : 'index.html',
keepAll : true,
alwaysLinkToLastBuild: true,
allowMissing : false
])
# csv trend history report
script {
csvFiles = findFiles(glob: 'lighthouse/*.csv')
for (csvFile in csvFiles) {
filePath = "${csvFile}"
plot(csvFileName: filePath.substring(filePath.lastIndexOf("/") + 1),
csvSeries: [[displayTableFlag: false, exclusionValues: '', file: filePath, inclusionFlag: 'OFF', url: '']],
exclZero: true,
group: 'my app',
numBuilds: '100',
style: 'line',
title: filePath.substring(filePath.lastIndexOf("/") + 1, filePath.indexOf(".")),
yaxis: 'score',
yaxisMaximum: '101',
yaxisMinimum: '15') # adjust scale
}
}
}
}
}
GitHub Actions
See example https://github.com/mgrybyk/html-trend-report-action