Unlighthouse Releases
- v0.17.0
Bug Fixes
- "Duda" sites not scanning correctly (77ac0bf), closes #277
- avoid infinite requeuing on path failure (f6f14e3), closes #268
- avoid using over max CPU cores (e5115c4)
- disable
ignoreI18nPageswhen cross-origin default is set for/(718b73b), closes #272 - finish worker when all routes are ignored (4db4297), closes #272
- safer path resolving (b14ebc8), closes #274
Features
- v0.16.2
- v0.15.0
Bug Fixes
- core: avoid normalising paths with app routing hashes (c28fcc7), closes #257
- core: fallback to download when no chrome paths exist (05753be), closes #247
- core: stop chrome cluster when scan is finished (a3ee8f0), closes #246
- core: handle corrupt cached reports (737a873), closes #255
- core: missing explicit dependency
radix3(c011378), closes #254
Features
- v0.14.0
Bug Fixes
- client: load assets relative using window.location (2da76f9), closes #112
- use relative path for non-specified router prefix (1bb0095), closes #244
- client: misc responsive issues (09d3e6c)
- client: remove thumbnails when not running performance (7e3c8a3)
- fallback to full width screenshot for non-performance scans (bbf59a5), closes #245
- false positive redirect warning for trailing slash (ef34b91)
- respect query strings (e708ebe), closes #207
Features
- v0.13.5
Bug Fixes
- better checks for implicit html routes (24e134d), closes #231 #225
- drop
boxenandchalkdeps (df1029a), closes #230 - ignore
importxwarnings (0a70312), closes #218 - warn about first i18n usage (4e0e75d), closes #215
- client: warn about deprecated FID (ffb54ac), closes #229
- improved robots.txt exclude matching (da74f6b)
- v0.13.4
- v0.13.2
- v0.12.2
Bug Fixes
- allow
npx unlighthouse-ciglobal usage (0c6a0eb) - check for valid cache removal when running with no cache (4cbb47c)
- record sessionStorage when authenticating (9fec05f)
- remove
protocolTimeout(383f246), closes #202 - remove puppeteer startup timeout (e2ae165)
- support basic auth with LHCI (bcb1bc6), closes #214
- use latest chrome as fallback browser (d678a05)
- allow
- v0.12.0
Lighthouse v12
This version is updated to use the latest version of Lighthouse - v12. Please see the release notes for further details.
Bug Fixes
- v0.11.0
✅ Upgrading
You can use the latest by explicitly providing the latest version constraint.
npx unlighthouse@^0.11.0 --site <site>⚠️ Node v18 or higher is now required
👀 Highlights
:boat: Lighthouse v11
Before this release, the Lighthouse version was stuck on v9 which was last updated over a year ago. Now Unlighthouse scans with the latest Lighthouse version (v11.6).
Expect to see new and more accurate audits as well as an improved HTML report UI. See the Lighthouse changelog for more details.
📈 CrUX Tab
Unlighthouse nows connects with the CrUX API to give you an overview of how your origin is performing. This is especially useful for catch catching regressions and to see how your INP performance.
🎨 Improved UI
Several improvements to the UI to make using Unlighthouse easier including tightened-up tabs, pagination to improve the stability of scanning large sites, and fixed category score sorting.
🤖 LHCI Reporter
You can now generate and send your reports to your Lighthouse CI server using the
lighthouseServerreporter.unlighthouse-ci --site <your-site> --reporter lighthouseServer --lhci-host <lhci host> --lhci-build-token <lhci buildToken>Many thanks to @lutejka for helping with this and upgrading Lighthouse.
👉 Changelog
Features
Bug Fixes
--sitemapsoption will override robots.txt sitemaps (8f01db0), closes #169- allow
pwacategory (db78086), closes #168 - create provider if missing and has route definitions (a9d0f24), closes #188
- fullPageScreenshot audit location (f03d543)
- ignore
onlyCategorieswhenonlyAuditsis supplied (3ec2531), closes #175 - less flakey scans when using multiple samples (851f39a), closes #60
- patch
puppeteer-clusterfor puppeteer v22 compatibility (fa274d8) - retry failed performance scans (745f8fe)
- set an
Unlighthouseuser agent by default (5359703) - support absolute output paths (ed32bed), closes #187
- throw error if no routes could be scanned (574bf87)
- when using
--urls, allow skipping root path (dc2563f), closes #163 - client: lcp element not showing (f8da8e6)
- client: make report link more obvious (31a3013)
- client: show tooltip links properly (44b06a2)
- core: display loaded config file log when using one (0ad9e91)
- v0.10.5
What's Changed
- fix: url for deployment demos by @RohrerF in https://github.com/harlan-zw/unlighthouse/pull/181
New Contributors
- @RohrerF made their first contribution in https://github.com/harlan-zw/unlighthouse/pull/181
Full Changelog: https://github.com/harlan-zw/unlighthouse/compare/v0.10.4...v0.10.5
- v0.10.3
Bug Fixes
- core: ensure non-html files get skipped (e2fc345)
- core: honour robots.txt
allowrules (d4accf1) - more accurate filtering for
include,exclude(ca5712c) - ci: exit code 1 when budget assertions fail (#165) (611862c) by @brahms116
New Contributors
- @brahms116 made their first contribution in https://github.com/harlan-zw/unlighthouse/pull/165
- v0.10.2
- v0.10.0
Bug Fixes
- deprecate integrations (d9c5c06)
Features
New
authenticatehook (c7afd0c)Works the same as the previous
puppeteer:before-go-tohook but this will only be called a single time before any routes are scanned. It will automatically collect any set cookies and local storage items during the hook.// unlighthouse.config.ts export default { hooks: { async authenticate (page) { // login to the page await page.goto('https://example.com/login') const emailInput = await page.$('input[type="email"]') await emailInput.type('[email protected]') const passwordInput = await page.$('input[type="password"]') await passwordInput.type('password') await Promise.all([ page.$eval('.login-form', form => form.submit()), page.waitForNavigation(), ]) }, }, }