Re: Asynchronous test cases

Hi Yunhao -

Recall from an earlier e-mail that I am not too happy about the
requirements and expectations of T7 [1].  However, I accomodated T7
with a "not totally unrealistic completion policy" which effectively
waits for a requestor to show interest in the final result before
producing it.

There are alternatives (including altering the text in T7 so that it
does not require the Pending status) but my view on this is that
whatever approach is chosen it should be consistent across all
asynchronous messages in the test suite.  If it is not consistent, the
application has to detect what test case is beeing executed (at least
if one intended to pass all tests) and that doesn't sound right to me.

Regards
Tommy

[1] http://lists.w3.org/Archives/Public/www-xkms/2004Oct/0007.html

On Tue, 19 Oct 2004 11:03:51 -0400, Yunhao Zhang <yzhang@sqldata.com> wrote:
> Hi Tommy,
> 
> Isn't T7 only a special case in asynchronous processing? T8, T11 and T12
> seem to be valid and cover more common situations where XKISS requests are
> processed immediately in a very short period of time.
> 
> Regards,
> 
> Yunhao
> 
> 
> 
> ----- Original Message -----
> From: "Tommy Lindberg" <tommy.lindberg@gmail.com>
> To: <alvarorg@cs.tcd.ie>
> Cc: <www-xkms@w3.org>
> Sent: Tuesday, October 19, 2004 9:33 AM
> Subject: Asynchronous test cases
> 
> >
> > Hi Guillermo -
> >
> > Could we make the message sequences in the asynchronous test cases
> > consistent between T7, T8, T11, T12 so that all of them use
> > StatusRequest's the way T7 does?
> >
> > Regards
> > Tommy
> >
> >
> 
>

Received on Tuesday, 19 October 2004 16:07:38 UTC