Bug 136096 – [EFL][WK2] Layout Test storage/websql/sql-error-codes.html is passed

• Status: RESOLVED
• Resolution: WONTFIX
• Component: WebKit EFL

[Shivakumar J M]

Layout Test storage/websql/sql-error-codes.html is no longer generating failure, it is passed in latest builds.

[Shivakumar J M]

Created attachment 236855 [details]
Patch

Layout Test storage/websql/sql-error-codes.html is no longer generating failure, it is passed in latest builds.

[Grzegorz Czajkowski]

Comment on attachment 236855 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=236855&action=review

> LayoutTests/ChangeLog:8
> +        Layout Test storage/websql/sql-error-codes.html is no longer generating failure, it is passed in latest builds.

Nit: personally, I am in favor of adding the revision affecting layout test success/failure. That brings us information what change makes it better or worse.

> LayoutTests/platform/efl/TestExpectations:745
> +webkit.org/b/108355 storage/websql/sql-error-codes.html [ Pass ]

If it passes just remove its failure expectation. Did you check debug build as well?

[Michal Pakula vel Rutka]

Comment on attachment 236855 [details]
Patch

This test still fails on EFL bot: http://webkit-test-results.appspot.com/dashboards/flakiness_dashboard.html#showAllRuns=true&tests=storage%2Fwebsql%2Fsql-error-codes.html

[Shivakumar J M]

Created attachment 236864 [details]
Patch-Updated

Update the patch with comments, these test passed in my local build always.

[Michal Pakula vel Rutka]

Comment on attachment 236864 [details]
Patch-Updated

View in context: https://bugs.webkit.org/attachment.cgi?id=236864&action=review

I checked on my local machine and the test fails the same as on bot. Anyway we cannot take results from our local machines as a reference point, instead we should rely only on build bot results.

> LayoutTests/ChangeLog:6
> +        Reviewed by NOBODY (OOPS!).

Gardening does not require review, next time please replace this line with something like: Unreviewed gardening and set only cq? leaving r flag not set.

> LayoutTests/platform/efl/TestExpectations:745
> +webkit.org/b/108355 storage/websql/sql-error-codes.html

Still if test starts to pass you should remove whole entry, not remove the modifier. Now leaving a test name without modifier means the same as [ Skip ].

[Shivakumar J M]

(In reply to comment #5)
> (From update of attachment 236864 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=236864&action=review
> 
> I checked on my local machine and the test fails the same as on bot. Anyway we cannot take results from our local machines as a reference point, instead we should rely only on build bot results.
> 
> > LayoutTests/ChangeLog:6
> > +        Reviewed by NOBODY (OOPS!).
> 
> Gardening does not require review, next time please replace this line with something like: Unreviewed gardening and set only cq? leaving r flag not set.
> 
> > LayoutTests/platform/efl/TestExpectations:745
> > +webkit.org/b/108355 storage/websql/sql-error-codes.html
> 
> Still if test starts to pass you should remove whole entry, not remove the modifier. Now leaving a test name without modifier means the same as [ Skip ].


Dear Michal,

Thanks for the inputs, will take care these things for next submissions. I will close these issue as not fix.

Thanks