[Webkit-unassigned] [Bug 172637] REGRESSION(r217458): 55 javascript tests are failing after r217458 on Linux and Windows

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=172637

--- Comment #8 from Zan Dobersek <zan at falconsigh.net> ---
(In reply to Yusuke Suzuki from comment #7)
> (In reply to Zan Dobersek from comment #6)
> > In the original patch, some tests that were affected by this were skipped.
> > We could do the same for these tests.
> 
> I'm a bit worried about the current result. If that thing just changes DST
> switch, it sounds good.
> However, the current implementation poses a bit storange Date for old dates.
> (Like "Fri May 07 -001 23:59:02 GMT-0752 (LMT)").
> 
> What do you think of reverting the patch for non Darwin environment?

Do we remain compliant with the relevant specs? We should maybe wait for Keith to weigh in.

One additional observation -- in run-javascriptcore-tests, we set the TZ env to the PST timezone. But at least for the failing tests under mozilla-tests.yaml/ecma/Date/, if that TZ env is set to UTC or GMT, the test passes because LMT doesn't kick in for the date that is far back in the past.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20170526/64148752/attachment.html>