users@jaxb.java.net

Re: Better xs:duration support?

From: Gary Gregory <ggregory_at_seagullsw.com>
Date: Tue, 15 Jul 2003 22:09:25 -0400

"OK. So essentially you are just interested in the time span in terms of
some fixed measure. Thank you for the input."

Yep. Yep.

"BTW, what's your plan if you see "1 month"? How many seconds should be
one month?"

Well, you could imagine, maybe, that it is one month from the time you start
the timeout period. So that could be in my app = System.currentTimeMillis()
+ 1 month, which is not a fixed amount. Do not ask me what Jan 30th + 1
month is either ;-) even though I imagine the ISO or XSD docs might say
something about it.

So for today's impl., I think that I will stick with a
clientSessionTimeOutMillis attribute since I have not found ISO
duration/long conversion routines in Apache but I have not looked very hard.

Thanks,
Gary

-----Original Message-----
From: Kohsuke Kawaguchi [mailto:Kohsuke.Kawaguchi_at_Sun.COM]
Sent: Tuesday, July 15, 2003 18:05
To: JAXB-INTEREST_at_JAVA.SUN.COM
Subject: Re: Better xs:duration support?

Gary Gregory <ggregory_at_seagullsw.com> wrote:
> Hi K,
>
> Well, I /was/ planning on using an xs:duration attribute to specific a
time
> out for one of our long lived component, where seconds is just fine as the
> level of granularity and quite easier to read than a milliseconds number.

OK. So essentially you are just interested in the time span in terms of
some fixed measure. Thank you for the input.

BTW, what's your plan if you see "1 month"? How many seconds should be
one month?


regards,
--
Kohsuke Kawaguchi                  408-276-7063 (x17063)
Sun Microsystems                   kohsuke.kawaguchi_at_sun.com