chore(webkit): fix WebKit network-related driver tests #23232
Conversation
Co-authored-by: Blue F <[email protected]>
|
Thanks for taking the time to open a PR!
|
flotwig
changed the title
| @@ -30,32 +31,7 @@ export async function open (browser: Browser, url, options: any = {}, automation | |||
| headless: browser.isHeadless, | |||
| }) | |||
|
|
|||
| let pwPage: playwright.Page | |||
|
|
|||
| async function resetPage (_url) { | |||
There was a problem hiding this comment.
@lmiller1990 here is the reset refactor requested: #15533 (comment)
| ranFirefox: name === 'firefox', | ||
| ranElectron: name === 'electron', | ||
| ranWebKit: name === 'webkit', | ||
| }) |
There was a problem hiding this comment.
Props for the simplification here 👍
There was a problem hiding this comment.
I've tested this branch on macOS, everything seems as stable as the initial PR. That's a nice assortment of skipped specs; I was hoping there'd be a small number of common differences driving the failures, but it's looking like there's quite a few differences here.
There was a problem hiding this comment.
Nice, tested it out - I still see a weird GraphQL error when I launch the runner with WebKit, I am thinking this is a known bug. I could pitch in and help out with this if needed (as of right now I idea what would be causing it, I didn't look yet).
I also noticed some tests you skipped were passing, eg the shadow DOM one. Maybe it's just flake and I got lucky?
| @@ -3928,7 +3929,8 @@ describe('src/cy/commands/actions/click', () => { | |||
| }) | |||
| }) | |||
|
|
|||
| describe('shadow dom', () => { | |||
| // TODO(webkit): fix+unskip for experimental webkit | |||
There was a problem hiding this comment.
Surprised these tests don't "just work" - I would have thought basic browser APIs would be fairly uniform, since we trigger them with standard client side JS (assuming here, haven't looked at how cy.shadow actually works).
Edit: some actually do just work, or at least, it seems that way on my machine.
There was a problem hiding this comment.
Actually this passed for me
There was a problem hiding this comment.
I skipped out the entire click.cy.js because of the large number of failures, I could've been more selective here. This will be unskipped by experimentalWebKitSupport release.
| @@ -2609,7 +2617,9 @@ describe('network stubbing', function () { | |||
| }) | |||
|
|
|||
| it('intercepts cached responses as expected', { | |||
| browser: '!firefox', // TODO: why does firefox behave differently? transparently returns cached response | |||
| // TODO: why does firefox behave differently? transparently returns cached response | |||
| // TODO(webkit): fix+unskip. currently fails in WK because cache is disabled | |||
There was a problem hiding this comment.
Is disabling the cache in WebKit something Playwright automation does, or something we do?
|
LGTM, some random questions/nits but no blocker. |
User facing changelog
n/a, WebKit is currently only available in development
Additional details
cy.route()andcy.server(), andcy.intercept()'sforceNetworkErrorcauses a request loop which results in flake.driver-integration-tests-webkiton all commits, so adding these skips was necessary here to keep CI green.PR Tasks
cypress-documentation?type definitions?