{panel:title=Last tested with}
* Petals ESB 3.1
* Petals Studio 1.1.0
* Petals SE RMI 1.1.1
* Petals BC SOAP 4.0.2
* Petals Webconsole 2.0.3
* Tomcat 6.0.29
* JDK 6 update 20
* Petals SE Jsr181 1.1.2

*Added in this tutorial :*
* Petals SE EIP 2.4.3
{panel}

{info}This tutorial is based on [EIP Buffer complex use case|petalsesb:Bufferize operations with EIP].
{info}

h2. Definition of our integration process

Today you will bufferize messages, to process them all in one by _MathOperations_.


You will use Enterprise Integration Patterns (EIP, see below for explanations) to create this process :
* Several messages are sent to an *EIPAggregator* pattern, which bufferize all incoming messages.
* Trigger message is sent to *EipAggregator.*
* *EipAggregator* sends an aggregated message to *EipSPlitter*.
* *EipSplitter* splits the aggregated message back into multiple messages.
* *EipSPlitter* sends the multiple messages to the operation *AddIntegers* on the service *MathOperations.*
* *MathOperations.AddIntegers* processes the messages and send you the responses.

{gliffy:space=petalsesb|page=Bufferize operations with EIP|name=Petals RMI-EIP-JSR181}


h2. Little bit of theory: EIP (Enterprise Integration Patterns) and XPath

*What is EIP ?*
A set of design patterns for enterprise application integration.
What is Agregator pattern ?

What is Splitter Pattern ?


*What is XPath ?*


Run this tutorial (takes \~30 min) for Xpath : [http://www.w3schools.com/XPath/default.asp]


We will use XPath to control conditions on EIP patterns.



h2. Configure the Splitter

# Create a new EIP Technical service with the following configuration:
# JBI properties of the configuration service:
## Interface namespace: *[http://test.petalslink.com|http://test.petalslink.com*]*
## InterfaceName: *EipSplitterInterface*
## Service Namespace: *[http://test.petalslink.com|http://test.petalslink.com*]*
## Service Name: *EipSplitter*
## End-point Name: *EipSplitterEndpoint*
# Project Name and Location:
## Project Name: su-EIP\-*EipSplitter*\-provide
# Specific Component properties:
## Eip: *Splitter*
# Service Parameters:
## Select _su-jsr181-MathOperations-provide_ for automatic configuration.
# XPath Condition:
## *//\[local-name()="AddIntegers"\]*
# Additional Parameters
## Operation Namespace: *[http://test.petalslink.com|http://test.petalslink.com*]*
## Operation: *AddIntegers*
## Mep: *InOut*
# CDK properties
## Validate WSDL: *false* (there is no wsdl)

h2. Configure the Aggregator

# Create a new EIP Technical service with the following configuration:
# JBI properties of the configuration service:
## Interface namespace: *[http://test.petalslink.com]*
## InterfaceName: *EipAggregatorInterface*
## Service Namespace: *[http://test.petalslink.com]*
## Service Name: *EipAggregator*
## End-point Name: *EipAggregatorEndpoint*
# Project Name and Location:
## Project Name: su-EIP\-*EipAggregator*\-provide
# Specific Component properties:
## Eip: *Splitter*
## Aggregator Correlation: *boolean(/*\[local-name()="AddIntegers"\])\*
# Service Parameters:
## Select _su-EIP-EipSPlitter-provide_ for automatic configuration.
# XPath Condition:
## *boolean(/*\[local-name()="AddIntegers"\]/*\[local-name()="equals"\])*
# Additional Parameters
## Operation Namespace: *[http://test.petalslink.com]*
## Operation: *AnyOperationName* (since EipSplitter does not have an operation name)
## Mep: *InOut*
# CDK properties
## Validate WSDL: *false* (there is no wsdl)

h2. Create and deploy an EIP Service Assembly

There is a tight-coupling between both EIP. Assembling them in one single SA can be a good idea (especially with EIP, which can chain lots of patterns, each requiring a different SU).
# Create a new Service Assembly.
## Name: *sa-Eip-AggregatorSplitter*
# Add the two EIP Service-Units you created, and confirm
# Fast-export the Service Assembly for Petals.
# Deploy in Petals ESB.


h2. Review the chain

*EIP Agregator*

</result>
{code}