Bug 182055 – Fix race-condition in fast/text/click-ellipsis-assertion-failure.html

• Component: Tools / Tests
• Status: RESOLVED FIXED

[Jonathan Bedard]

fast/text/click-ellipsis-assertion-failure.html should wait for it's promise to be fulfilled before finishing the test. If it does not, the subsequent test may crash.

[Jonathan Bedard]

<rdar://problem/36830203>

[Jonathan Bedard]

Created attachment 332181 [details]
Patch

[Jonathan Bedard]

I very much doubt that this is the only test suffering from this problem.  However, I think we should take these one-at-a-time. This specific bug will look like a flakey crash occurring on the test immediately after the test with the bug. Thankfully, the crashlog of the subsequent test will specifically call out the test with the race condition.

[Megan Gardner]

Can we make sure to file a bug about the other tests? There's been a lot more async code in our tests in the last few months, while it makes the tests more readable, I think it will make these races a lot more common, and we should try and keep a handle on it.

[Jonathan Bedard]

(In reply to Megan Gardner from comment #4)
> Can we make sure to file a bug about the other tests? There's been a lot
> more async code in our tests in the last few months, while it makes the
> tests more readable, I think it will make these races a lot more common, and
> we should try and keep a handle on it.

I've filed <https://bugs.webkit.org/show_bug.cgi?id=182062>.

Ryan and I are working on repaying technical debt in iOS layout tests, this is one class of failures we will be looking for.

[WebKit Commit Bot]

Comment on attachment 332181 [details]
Patch

Clearing flags on attachment: 332181

Committed r227571: <https://trac.webkit.org/changeset/227571>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[Radar WebKit Bug Importer]

<rdar://problem/36839587>