Users can export records out of Fulcrum, make changes and add information elsewhere (in, say, Excel), and then reimport to update the records in Fulcrum.
This can be very useful when needing to update or modify a large number of records.
However, that doesn't address the issue of executing further updates to that normalized data as it's being used elsewhere in the application.
There are a variety of different approaches that you can use, based on your own preference.
In this case, we need to do a deep recursive merge, not just a shallow copy.
There are two sets of methods to help you with this - one in which you ask the Toolkit to make the updates and then refresh itself, and another in which you tell the Toolkit the data has already been updated and that it should refresh.
Note: When updating records, make sure that the column is still present in the file.
This column is used to map the rows to the existing records.
It may have already occurred to you, but you can write your own data manager and register it on an instance of the Toolkit.
You just need to implement any of the six methods discussed above, and then register the manager using the same id under which you registered the associated data loader.