Bug 64193 – TestFailures page incorrectly thinks all tests passed in http://build.webkit.org/builders/Windows%207%20Release%20(Tests)/builds/14589 after a reload

• Status: RESOLVED
• Resolution: FIXED
• Component: Tools / Tests

[Adam Roben (:aroben)]

TestFailures page incorrectly thinks all tests passed in http://build.webkit.org/builders/Windows%207%20Release%20(Tests)/builds/14589 after a reload

[Adam Roben (:aroben)]

Created attachment 100143 [details]
Patch

[Daniel Bates]

Comment on attachment 100143 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=100143&action=review

r=me

> Tools/BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/LayoutTestResultsLoader.js:33
> +        const currentCachedDataVersion = 3;

Are there semantics on when we increment this?

[Adam Roben (:aroben)]

Comment on attachment 100143 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=100143&action=review

>> Tools/BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/LayoutTestResultsLoader.js:33
>> +        const currentCachedDataVersion = 3;
> 
> Are there semantics on when we increment this?

There are basically two reasons to increment it:

1) We're now caching more data than we did before and either can't or don't want to handle dealing with cached values that don't contain that extra data
2) The old data was incorrect for some reason

This patch is kind of a mixture of both cases.

[WebKit Review Bot]

Comment on attachment 100143 [details]
Patch

Clearing flags on attachment: 100143

Committed r90656: <http://trac.webkit.org/changeset/90656>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.