Another (additional?) strategy could be to just add the (newEncodedHeaderSize - oldEncodedHeaderSize) to the others positions and sizes if present (eg deliveryAnnotationsPosition, encodedDeliveryAnnotationsSize, etc etc), encoding any new header (or overwriting the original one, without allocating a new data) and saving from scanning the message data again. I would attempt a PR, but need some help from any AMQP guy (nudge nudge) around here :P
-- Sent from: http://activemq.2283324.n4.nabble.com/ActiveMQ-Dev-f2368404.html
