Hi Nicolas,

On 6/15/06, Nicolas Modrzyk <[email protected]> wrote:

The scheduled activation should then just be a matter of using
Jackie's workflow template. (Reviewed by John a bit !?)
By the way, John, if a fake date is entered, for example set in
January 1970, it looks like openwfe is on hold on the sleep until,

the time machine implementation I use (time-machine-3.0.4 from Apache)
requires some warm up time, but it has a precision on target of less
than two microseconds.
I should check the implementation of Codehaus, but it's rumored to
have a lesser precision which is strange because those guys brougt us
Quartz,... And of course, the whole JSR-4523 about time machines is
just in the pre-RFP state...
I know, I'm Swiss, thus a bit finicky about time and watches,  but as
we say : "the lake isn't on fire !" (Y'a pas le feu au lac !)...

and does not find out the date is expired ? Do you see any reasons
for that ?

I don't do any checks on that [until] date, but I'll implement them.
But don't forget to do your own tests on the input. For the OpenWFE
integrator 'iota' it might be OK to put a segment of a process to
sleep for 10 years, for your particular use case it might be wrong,
but it's up to you to check for that.
I'll restrict the 'time-machine' scenario but for 'the sleeping
beauty', I won' put any restriction.

Added a todo for that at sf.net/projects/openwfe :
http://sourceforge.net/tracker/index.php?func=detail&aid=1506480&group_id=54621&atid=575490


Sayonara !

--
John Mettraux   -///-   http://jmettraux.openwfe.org

----------------------------------------------------------------
for list details see
http://www.magnolia.info/en/magnolia/developer.html
----------------------------------------------------------------

Reply via email to