Hi Alex,
With aggregate type EPA you can't refer to the participant event
attributes when you specify the attributes values of the derived event.
So I would change the EPA from aggregate type to BASIC type, lets call it
EPA1, and have the input event (E1) plays as initiator of the temporal
context and as input event of EPA1.
This is the EPA that used to generate event (lets call it E2) 24 hours
after the input event (lets call it E1) arrived, right?
In EPA1, you can refer to the input event E1 when you define the attribute
values of the derived event E2.
Hope it helps,
Regards,
Tali
Tali Yatzkar Haham
Event-based Middleware & Solutions
IBM Haifa Research Lab, Israel
tali@il.ibm.com 972-4-8296320
From: Theologos Zacharopoulos <alex@konnektid.com>
To: Tali Yatzkar-Haham/Haifa/IBM@IBMIL
Date: 25/03/2015 04:08 PM
Subject: Re: Complex Event Processing, help
Thanks that was exactly what I was looking for, is there anyway to do it
for aggregate types?
I have an derived event that is triggered after 24 hours (aggregate), and
I would like to get the ID of the participant event, is that possiple?
Thanks a lot by the way, you are a big help to me
On 25 March 2015 at 12:34, Tali Yatzkar-Haham <TALI@il.ibm.com> wrote:
Hi Alex,
Yes, in most EPA types this is possible (not in aggregate or absent EPA
types).
When you define the derived event of the EPA, you can open a table that
contains all the attributes of the derived event.
You can assign the attributes values. For example, you can set value of an
attribute named "triggerEventID" (any attribute name that you defined in
e2) to e1.ID (assuming e1 has an attribute named ID, make sure both
attributes in e1, e2 have the same type).
See example here:
Hope it helped,
Tali
Regards,
Tali
Tali Yatzkar Haham
Event-based Middleware & Solutions
IBM Haifa Research Lab, Israel
tali@il.ibm.com 972-4-8296320
From: Theologos Zacharopoulos <alex@konnektid.com>
To: Tali Yatzkar-Haham/Haifa/IBM@IBMIL
Date: 25/03/2015 12:42 PM
Subject: Re: Complex Event Processing, help
Dear Tali I have one more question about the Complex Event Processor,
I trigger an event (e1) from my system with an ID, after that there is
some processing in the CEP and if the conditions I set are true I derive a
"command" event (e2)
that through an adapter I get it back to my system. The thing is how can I
understand from which trigger event (e1) the command event (e2) was
triggered?
In a perfect solution I would like in the EPA my derived event (e2) get
the ID from the participant events.
Is that possible? set a value on a field of an event during the derivation
in the EPA?
Thanks in advance!
Best regards
On 19 March 2015 at 18:22, Tali Yatzkar-Haham <TALI@il.ibm.com> wrote:
Great, glad to hear that.
Regards,
Tali
Tali Yatzkar Haham
Event-based Middleware & Solutions
IBM Haifa Research Lab, Israel
tali@il.ibm.com 972-4-8296320
From: Theologos Zacharopoulos <alex@konnektid.com>
To: Tali Yatzkar-Haham/Haifa/IBM@IBMIL
Cc: fiware-tech-help@lists.fi-ware.org
Date: 19/03/2015 11:30 AM
Subject: Re: Complex Event Processing, help
Hello, I just realised that there is no implementation for the custom
adapters,
So I make it work by implement my custom OutputServer that returns my
Custom Adapter in case of custom adapter type selection.
Thanks, everything works fine
On 18 March 2015 at 14:27, Theologos Zacharopoulos <alex@konnektid.com>
wrote:
Dear Tali,
I need some help with the development of a Custom Output Adapter, I
managed to develop it but I want to know the process of repackaging in a
war file and run it on the server.
so I need to recreate somehow the ProtonAgents.jar and put it back to the
.war file? and if yes is there any online information on todo that?|
thanks in advance!
On 12 March 2015 at 10:40, Theologos Zacharopoulos <alex@konnektid.com>
wrote:
Dear Tali,
That was exactly the solution I was looking for, thank you very much that
was very helpful!
Also thanks for the quick response,
Best regards,
Theologos Alex Zacharopoulos
On 12 March 2015 at 08:09, Tali Yatzkar-Haham <TALI@il.ibm.com> wrote:
Dear Alex,
Glad to hear you are using the CEP!
What you can do:
1. Have one EPA, lets call it EPA1, that uses the temporal context you
defined, lets call it TC1, (initiator Event_X, terminated after 24 hours).
Make EPA1 of type Aggregate, that uses TC1. Set the evaluation policy of
EPA1 to be deferred and don't set it conditions or input event. Make it
generate a new type of event (that you need to define), lets call it E2.
The purpose of EPA1 is to generate E2 24 hours after Event_X arrives. (If
you don't want to have a warning on empty input events you can put Event_X
as an input event of EPA1)
2. Now you can define another temporal context, lets call it TC2 that uses
E2 as initiator.
Hope this is what you wanted.
I cc the FIWARE help, so this corresponding will be documented in the
FIWARE issue tracking system. Please cc it as well in future
correspondence. Thanks.
Best regards,
Tali
Tali Yatzkar Haham
Event-based Middleware & Solutions
IBM Haifa Research Lab, Israel
tali@il.ibm.com 972-4-8296320
From: Theologos Zacharopoulos <alex@konnektid.com>
To: Tali Yatzkar-Haham/Haifa/IBM@IBMIL
Date: 11/03/2015 04:20 PM
Subject: Complex Event Processing, help
Dear Tali,
I am a Software Engineer who collaborates with the FiWARE GE,
I am working on the IBM Proton about 2 weeks now, and I ma trying to
configure a use case for the company I work for.
I created a temporal context that starts the time that an Event_X is fired
and it expires after 24hours, but I want a second temporal context start
after 24hours of the event creation, can you please provide me some help
on this?
I tried to set the condition: Event_x.duration == 86400000, but that
didn't work for me.
Thanks in advance!
Alex Zacharopoulos, Software Engineer @ Konnektid.
User replied on 12/03/15:
Dear Tali,
That was exactly the solution I was looking for, thank you very much that was very helpful!
Also thanks for the quick response,
Best regards,
Theologos Alex Zacharopoulos