The JDBC component currently reads all the rows returned by a query in one shot, this only works for small resultsets. I'm trying to modify the component to support cursor semantics, so it reads only X numbers of rows, sets the rows in a new exchange's body and forwards it, rinse repeat until all the rows returned from a query are processed. The JDBC component only create a producer that executes a sql statement it received in a body of an exchange. My question is about the best approach to do this "1 exchange" to a "series of exchanges" mapping, My research lead me to MulticastProcessor it seems that I should make JDBCProducer extend it and then override createProcessorExchangePairs to create an Iterable<ProcessorExchangePair> that actually reads rows from the resultset to populate each new exchange, Am I on the right track here?
-- View this message in context: http://camel.465427.n5.nabble.com/JDBC-component-handling-large-results-tp5730509.html Sent from the Camel Development mailing list archive at Nabble.com.
