This sample demonstrates how to set up an execution plan with a pattern matching Siddhi query to detect non-occurrences. In this sample we use patterns to detect non-delivered packages of a courier service within a predefined time period (for demonstration purposes we assume the time limit to deliver a package is 2 minutes).
The queries used in this sample are as follows.
The first query keeps each arrival event for 2 minutes and then publishes it to the
overdue_deliveries_stream event stream. This event stream is then used in the second query.
The second query uses a pattern. It looks for a pattern where an arrival event (from the
arrivals stream event stream) is always followed by a delivery event (from the
deliveries_stream event stream), or by an overdue delivery event (i.e., an arrival event that is 2 minutes old from the
overdue_deliveries_stream event stream). If the pattern detects an arrival event followed by an overdue delivery event with no delivery event, that means the package is not delivered within the 2 minute time limit.
The third query is used to detect whether the delivery event is null where the
deliveryId is null if the delivery event is null.
Building the sample
Start the WSO2 DAS server with the sample configuration numbered 0111. For instructions, see Starting sample CEP configurations.
This sample configuration points the default Axis2 repo to
<DAS_HOME>/samples/cep/artifacts/0111 (by default, the Axis2 repo is
Executing the sample
Open a new terminal, go to
<DAS_HOME>/samples/cep/consumers/generic-log-serviceand run the command below. It builds the sample log service web app and deploys in the webapps repository that is relevant to the sample.
View the logs in the DAS server when logger service is deploying. For example, after deployment, the Web app is able to receive messages sent from the DAS server.
<DAS_HOME>/samples/cep/producers/httpand execute the following command.
This reads the
arrivalEvents.txtfile that contains a few sample arrival events and then sends them to the specified URL.
Execute the following command:
This reads the
deliveryEvents.txtfile that contains some sample delivery events and sends them to the specified URL.
The output is logged in the console as shown below after 2 minutes.