harmony-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Tony Wu" <wuyue...@gmail.com>
Subject Re: [classlib][text] DateFormatSymbols (was: Re: svn commit: r612718 [1/2] - in /harmony/enhanced/classlib/trunk: modules/luni/make/ modules/luni/src/main/java/java/util/ modules/luni/src/main/java/org/apache/harmony/luni/internal/locale/ modules/lun
Date Wed, 13 Feb 2008 08:35:04 GMT
Tim, Alexei
Seems it is locale dependent. I tried several locales, there is a
empty string in the output. I can not tell why RI behaves like this.

The senario test is writtern by ourselves. My question is, will this
difference block the struts or not? If it blocks struts we have to
work around even if the fix does harm to performance. If not, let's
keep it as non-bug difference and find out the reason.

On 2/13/08, Tim Ellison <t.p.ellison@gmail.com> wrote:
> Alexei Zakharov wrote:
> > Tony, all,
> >
> > I like to inform you that our Struts scenario constantly fails - see
> > [1] - after you moved DateFormat related classes to ICU-based
> > implementation at r612718. Please see details in HARMONY-5498 I posted
> > recently. I do not suggest to roll-back this commit. However, I like
> > to hear what other people think about it.
> >
> > [1] http://people.apache.org/~smishura/r620256/Windows_x86/struts_test/
> Hmm, that seems like strange behavior to me -- there is no month called "".
> Is the empty string present for other locales too, or just English?
> Regards,
> Tim

Tony Wu
China Software Development Lab, IBM

View raw message