geronimo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Cabrera, Alan" <Alan.Cabr...@reuters.com>
Subject RE: Problems with stopwatch test
Date Mon, 08 Dec 2003 18:37:00 GMT
Could probably do something like:

T1.start();
T2.start();
Sleep(1000);
T2.stop();
T1.stop();

AssertTrue(T1.elapsed()=>T2.elapsed());

I'll comment out the tests and maybe Brent Worden can beef it back up.


Regards,
Alan

> -----Original Message-----
> From: Dain Sundstrom [mailto:dain@coredevelopers.net] 
> Sent: Monday, December 08, 2003 1:17 PM
> To: geronimo-dev@incubator.apache.org
> Subject: Re: Problems with stopwatch test
> 
> 
> On Dec 8, 2003, at 11:38 AM, Jeremy Boynes wrote:
> 
> > Alan D. Cabrera wrote:
> >
> >> It works for me but, I should mention that the code that was
> >> originally
> >> submitted didn't work for me; I had similar problems that you had,
> >> sometimes sleep(100) takes only 94 milliseconds.  I 
> figured that it 
> >> was
> >> something w/ the timer resolution and upped it up to one 
> full second.
> >> Regards,
> >> Alan
> >>> -----Original Message-----
> >>> From: David Jencks [mailto:david@coredevelopers.net]
> >>> Sent: Sunday, December 07, 2003 8:03 PM
> >>> To: geronimo-dev@incubator.apache.org
> >>> Subject: Problems with stopwatch test
> >>>
> >>> I'm having inconsistent problems with the new geronimo commons 
> >>> StopWatch test on my slow old powerbook.  Perhaps in an 
> attempt to 
> >>> get on to more important things, apparently sometimes sleep(1000) 
> >>> takes only 999 milliseconds.
> >>>
> >>> I'm curious if anyone else observes anything similar and if anyone
> >>> has
> >>> a suggestion about how to work around this. (buying a new 
> computer is
> >>> not practical at the moment, unfortunately).  I patched 
> the test to
> >>> show some instances of this problem.
> >>>
> >>> Thanks
> >>> david jencks
> >
> > This class doesn't actually do much - do we really need it?
> 
> This test fails on my Mac also.  Sleep has never been accurate.  I 
> doubt there is any way to measure really test this class.  
> Can we drop 
> the test or give it a 10 or 20 percent margin of error?
> 
> -dain
> 
> /*************************
>   * Dain Sundstrom
>   * Partner
>   * Core Developers Network
>   *************************/
> 


---------------------------------------------------------------- 
      Visit our Internet site at http://www.reuters.com 

Get closer to the financial markets with Reuters Messaging - for more
information and to register, visit <http://www.reuters.com/messaging> 

Any views expressed in this message are those of  the  individual sender,
except  where  the sender specifically states them to be the views of The
Reuters Group.

Mime
View raw message