We are currently going from “normal” task search do indexed task search.
During this change we have modified the names of the refined fields. Due to this we will most probably have problems that the saved searches will not work with our new additions.
We have identified so far 2 solution.
- Update the saved searches directly in the DB
This would mean implementing a service that read the saved searches, modifies them and then saves them back in the DB
This has PRO’s and CON’s
- one time deal - after the service is ran, all the saved searches are updated
-rather hard to implement: the save searches have to be base64 decoded and deserialized, modified and then put back
- I think there is a high risk of corrupting the saved searches
- Update the saved searches at runtime
This would mean that custom code is written that when the user loads a saved search, the outdated refined fields are removed and the new ones are added.
- risk is rather low
- custom implementation needed until saved searches are migrated.
Can you give me an idea on which solution would be the better one?
Or is there something else I am forgetting?
Is there a more in-depth Saved Searches documentation where one can find tips and tricks for this part?