e2e tests for useOverflow logic
#2151
Conversation
Temporarily remove unnecessary CI jobs.
r100-stack
requested review from
mayank99 and
Ben-Pusey-Bentley
and removed request for
a team
It would be hard to know which unit tests would fail in the future with the new
If I create new e2e tests in the PR that adds the new Thus, by adding those tests here with the old |
| ); | ||
| await expect(items.first()).toHaveAttribute('data-iui-focused', 'true'); | ||
| const tags = await page | ||
| .locator('#test-component-selected-live > span') |
There was a problem hiding this comment.
this locator is relying on an implementation detail which could change at any time?
There was a problem hiding this comment.
I was contemplating on whether to have selectors that depend on iTwinUI class names. That was why I used that span selector.
Changed to directly target -select-tag (057a5a0).
There was a problem hiding this comment.
The select-tag is still coming from an internal class which can change at any time.
Right now we don't have a way to pass classes/attributes to the tag container or the tags (#894). This test could probably be rewritten to avoid relying on such implementation details (e.g. by only comparing the text inside the main container), but I'll leave it upto you if you want to rewrite it or add a TODO to wait until we allow customizing the tag container.
| expectedLastTagTextContent: string | undefined; | ||
| }) => { | ||
| const tags = await page | ||
| .locator('div[class$="-select-tag-container"] > span') |
There was a problem hiding this comment.
this locator is also relying on an implementation detail?
There was a problem hiding this comment.
Changed to directly target -select-tag (057a5a0).
There was a problem hiding this comment.
Same as #2151 (comment), this is still relying on an implementation detail.
|
Converting the PR to a draft to prevent it from merging before the other PRs in the PR stack are done. |
Changes
Pulled out of #2120 as PR 1 in the PR stack (PR stack order).
To make that PR easy to review, according to @mayank99's suggestion, I created this PR to only add new e2e tests that test the existing
useOverflowlogic. These tests will thus act as a baseline that the newuseOverflowlogic will also have to pass to make sure that there is no change in the resulting behavior.Some e2e tests (e.g.
MiddleTextTrucation) seem to be failing only on GitHub (e.g. action run) but not locally. I guess this is a small testing environment specific issue that I am currently trying to fix. Thus, the PR is ready for review despite the failing tests.TODO:
Testing
Docs
N/A
After PR TODOs
awaiting theexpects in e2e tests (#2151 (comment))