ProDataSet updates generally follow a typical pattern in the round trip from application server to client:
  1. The Java client initially obtains the ProDataSet from the application server as a working ProDataGraph.
  2. The client makes changes to the working ProDataGraph, and sends those changes back to the application server.
  3. The application server matches the records in its changed ProDataSet back to the original data source, verifies if records were changed by other users, and resolves any conflicts.
  4. The application server then applies the changes to the data source and passes back to the Java client the final versions of records that may have been further changed in another ProDataSet, along with any errors.
  5. The Java client then handles any errors in the returned ProDataGraph and merges the final versions of data in this ProDataGraph into its working ProDataGraph.
  6. The client continues with Step 2 and the cycle repeats until done.
The rest of this section describes how you can manage the ProDataGraph throughout this update cycle.