We have a rather antiquated database that contains a large number of individuals along with a number of achievements that they have completed. There has historically been little done to discourage duplicated individual data so we've ended up in a situation where our data is quite dirty. A massively simplified version of this can be found here.
We're now redesigning the schema and user interface. We will be providing the user with a tool to merge their individuals together. In the provided example, Dave and David are clearly the same person and has achieved 4 achievements in total.
Given that users make mistakes and that there are a lot more tables involved than in the example, I'm looking for a schema design that facilitates easy merging of data, and notably, un-merging of data if (when!) the user inevitably makes a mistake.
Linked lists of some sort appear to be a solution but aren't exactly efficient for this use case. Are there any other concepts that might lend themselves to this situation? Any specific design patterns that might be appropriate?
Aucun commentaire:
Enregistrer un commentaire