Migrate 'Meta boxes' e2e tests to Playwright#55915
Merged
Merged
Conversation
Mamaduka
added
the
[Type] Automated Testing
Mamaduka
requested review from
ajitbohra,
kevin940726,
nerrad and
ntwb
as code owners
Mamaduka
commented
Nov 7, 2023
Comment on lines
+105
to
+110
| // eslint-disable-next-line playwright/no-conditional-in-test | ||
| if ( | ||
| ( await excerptButton.getAttribute( 'aria-expanded' ) ) === 'false' | ||
| ) { | ||
| await excerptButton.click(); | ||
| } |
Member
Author
There was a problem hiding this comment.
Maybe we should turn off no-conditional-in-test ESLint rule. Most of the warnings I've seen in our test are false positives, as we need to check the state of panels/sidebars before trying to open them.
Member
There was a problem hiding this comment.
+1 on this. I haven't seen a case yet where this rule would be helpful.
Member
There was a problem hiding this comment.
One question though! 😄 Is the excerptButton state actually unknown (or random?) at this point? Why do we need to check whether it's expanded or not?
Member
Author
There was a problem hiding this comment.
The panel state (excerptButton state) is persisted for a user and is sometimes shared across the e2e test shards. This can result in flaky tests.
Member
There was a problem hiding this comment.
Gotcha. Definitely something to have as a side note there, then! 🙏
Anyhow, I think we should avoid running parallel tests against the same instance since it makes them unpredictable (like this one) and thus unreliable. Do you think it would be possible to run separate instances or at least create separate admin accounts per shard? cc @kevin940726
Member
Author
There was a problem hiding this comment.
This is a known "quirk"; here's the previous discussion - #50851. Some of these details can be absorbed in editor helper utils, just like we do for openDocumentSettingsSidebar.
Member
There was a problem hiding this comment.
Do you think it would be possible to run separate instances or at least create separate admin accounts per shard?
What do you mean by "shard" here? If you mean playwright's --shard then I believe we are testing a different WordPress instance each shard.
Ideally, I think we should aim for fully parallel testing if possible, but the very nature of WordPress makes it very difficult to do. That's also why we still can't enable parallel mode to speed up the tests. Another gotcha as mentioned in #50851 is that the resetPreferences() call doesn't seem very reliable in tests. Perhaps there's something wrong in the API side or we are doing something wrong elsewhere. I never have time to investigate it but I should 😅. The goal is to make every test as isolated and predictable as possible with a clean state, so that we don't have to do some of those conditionals for most cases.
A bit of side-note here is that I had a brief (like literally 3-sentence) discussions with @adamziel about whether we can leverage the architecture of wp-playground somehow to automatically generate a unique WordPress instance (or a clean DB) for each test. This is a much more challenging task though and might not even work for performance reason 😆.
Contributor
There was a problem hiding this comment.
@kevin940726 it might actually be possible now and perhaps without too much effort. Playground now supports the bits that were missing earlier like proc_open(), system() etc. functions which were crucial for the PHPUnit test suite. Also, almost all core unit tests work with the SQLite integration plugin. I didn't yet try running PHPUnit for WP Core with the new version of https://www.npmjs.com/package/@php-wasm/node, but there's a high likelihood it would just work. Feel free to ping me if you decide to explore it.
|
Size Change: 0 B Total Size: 1.7 MB ℹ️ View Unchanged
|
Mamaduka
requested review from
WunderBart and
talldan
and removed request for
ajitbohra,
nerrad and
ntwb
WunderBart
approved these changes
Nov 7, 2023
Member
There was a problem hiding this comment.
One potential candidate for a follow-up. Otherwise, 🚀!
noahtallen
reviewed
Nov 13, 2023
|
|
||
| await expect( | ||
| page.locator( '.wp-block-latest-posts > li' ) | ||
| ).toContainText( [ 'A published post', 'Dynamic block test' ] ); |
Member
There was a problem hiding this comment.
Hey, I noticed this causing a couple flaky tests today and grabbed a screenshot:
And here's a link to the test results: https://github.com/WordPress/gutenberg/actions/runs/6846766063/job/18613885820
It looks like the order of the text can be reversed in the post, but .toContainText enforces the exact order of the input array
Member
Author
There was a problem hiding this comment.
Thank you, @noahtallen! I'll have look shortly.
This was referenced Nov 13, 2023
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What?
Part of #38851.
PR migrates
meta-boxes.test.jse2e tests to Playwright.Why?
See #38851.
Testing Instructions