I've seen this here on 2.0, I think it's more frequent here in UTC+09:30 :)
It seems to happen when UTC time and local time have different dates (ie
circa 10:30pm for Central European Daylight time when this build happened).
Here is a way to trigger the problem:
"Theoretical time zone where local date is always different from UTC date
for same instant in time."
DateAndTime localTimeZone: (TimeZone offset: 24 hours name: 'Test Time
Zone' abbreviation: 'TTZ').
DateTest new testDateAndTimeNow. "Should fail"
Date dateAndTimeNow first. "Inspect should show date in the future :("
I'm not sure the fix is that hard, but I could use some guidance on some
implementation details. I'm working on the basis that Date should
generally use local time, and seconds since Jan 1, 1901 in local time, and
DateAndTime should use UTC time and seconds since Jan 1, 1901 in UTC time?
Cheers,
Martin
On Sat, Apr 27, 2013 at 6:05 AM, <no-re...@ci.inria.fr> wrote:
>
> https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2-Validation/./label=linux/87/
>
> 1 regressions found.
> KernelTests.Chronology.DateTest.testDateAndTimeNow
>
>
