Thanks for the great responses. They are very helpful. After discovering more about the problem domain, it appears that our clients do not always adhere to the file format. For example, in a standard file, positions 10-20 may represent the account number. Client 1 sends files that adhere to this standard. Client 2 sends a file where positions 10-20 represent the account's creation date. Client 3 sends a file where positions 10-20 represent a comment about the account. (So why they even bother defining a file format standard is beyond me.) We need to configure these changes after the software is deployed. So, I think this precludes me from using an annotation driven mapping. Which is unfortunate, because Bindy looks like a great framework to use.
I'm debating between using Bean-IO where, as I understand it, the mapping is defined in an xml file, or building something myself. If I go with Bean-IO, and we get a client who doesn't exactly adhere to the standard, I can create a special xml mapping just for that client. -- View this message in context: http://camel.465427.n5.nabble.com/Using-Bindy-for-multiple-record-types-in-the-same-file-tp5735541p5735578.html Sent from the Camel - Users mailing list archive at Nabble.com.
