Hello Lorrin!
Claus explain this in his book Camel in Action (
http://www.manning.com/ibsen/).
The source code is available at google code. You should have a look in the
following examples:
http://code.google.com/p/camelinaction/source/browse/#svn/trunk/chapter8/aggregator/src/test/java/camelinaction
You have to define the aggregator like the following:
.aggregate(header("correlationIdentifier"), new MyAggregationStrategy())
In this example, based on the in message header "correlationIdentifier"
value the messages are aggregated together.
Cheers,
Christian
On Fri, Sep 10, 2010 at 11:53 PM, Lorrin Nelson
<[email protected]>wrote:
> When using .split and .aggregate, what determines whether the aggregation
> is performed across the messages that were split out from a single original
> exchange vs. across messages that might have come from multiple original
> exchanges?
>
> I think the examples at http://camel.apache.org/splitter.html involve a
> single original exchange whereas http://camel.apache.org/aggregator.htmlare
> primarily multiple sources, but I don't see what the key distinction in
> the Java DSL is.
>
> Cheers
> -Lorrin
