That is correct. But nothing stops you from creating new exchanges and aggregate results later on. You can also use some sort of correlation id (query id) so you'll know that different exchanges refer to the same query. You can also enforce ordering if you want. You can also use multiple routes (i.e. your processor can send the new exchanges down a different path, aggregate results and then send the result (once) down the original route to continue processing. That may appear as a limitation, but it's not really. You only need to understand the camel model and how to use it.
Cheers, Hadrian On Apr 7, 2011, at 12:13 PM, Mark Webb wrote: > I have a database that I need to query and put the results through a > camel route. The query will result in hundreds of thousands of rows. > My query will return 1000 rows at a time and pass them through the > route. > > So I want to know how I can build a processor that can perform the > query multiple times, and place each 1000 rows in a new exchange and > send to the next processor in the route. From what I have found, a > processor can only 'send' an Exchange out once. > > Thanks, > Mark
