On Wed, Jun 29, 2011 at 3:14 PM, John Mettraux <[email protected]> wrote:
>
> On Wed, Jun 29, 2011 at 11:07:45AM -0700, Ian Smith-Heisters wrote:
>>
>> You win ;) We're just going to scrap the synchronization with the
>> end-user, and do it later with AJAX push. In the meantime they'll have
>> to rely on email notification and refreshing their browser.
>
> Hello,
>
> Ouch, I did not mean to provoke such big changes.

Heh, no, we've been planning on it anyway ;)

>
>> There's
>> still an open question on how to best test Ruote while embracing its
>> asynchronicity. Even your tests use wait_for to force synchronization,
>> right?
>
> Yes, I pause the test thread until a participant receives a workitem 
> (wait_for(:alpha)), the flow terminates or runs into an error 
> (wait_for(wfid)). When the event occurs, the observer wakes up our test 
> thread and the test resumes.
>
> Maybe that's something like this that you want for your end-user request 
> processing, but you'd have to make sure to use a timeout.
>

yes, that's pretty much what we've been doing, and what we'll continue
doing in our tests. It does mean the tests can't use multiple workers
because the observer doesn't work cross-process without the
aforementioned global event queue. This is somewhat problematic
because it introduces differences between the test and production
environments.

-ISH

>
> Best regards,
>
> --
> John Mettraux - http://jmettraux.wordpress.com
>
> --
> you received this message because you are subscribed to the "ruote users" 
> group.
> to post : send email to [email protected]
> to unsubscribe : send email to [email protected]
> more options : http://groups.google.com/group/openwferu-users?hl=en
>

-- 
you received this message because you are subscribed to the "ruote users" group.
to post : send email to [email protected]
to unsubscribe : send email to [email protected]
more options : http://groups.google.com/group/openwferu-users?hl=en

Reply via email to