I want to design a bulletproof method to deal with schema changes in a two-way replication environment. The setup is pretty simple: a parent replica where most of the editing and any schema changes will take place, replicated to a child where attribute editing will take place, but no schema changes will be made.
The GP tools offered be ESRI for comparing/importing schema XML have limitations, and I want the synchronization process to be automated and handle ANY and ALL schema changes that are made in the parent database.
In order to do this I am thinking the workflow would be like this:
Delete contents from child database
Run "create replica" tool to re-create replicas and copy the replicated datasets to the child database.
Am I correct to think this would handle all schema changes?
I get that it would be expensive and add a good amount of run time to an automated script, but I don't want to deal with any schema errors.
I want the process to be automated, and arcpy doesn't have a tool to un-register replicas, so I would have to use arcobjects within a python script to un-register the replicas. Anyone have any experience/success in doing that?