Hey all, just a general question about prisma migrations and seeding. When doing development, say I make a change to a model that requires making a change to my seed script due to the model change. I can run

prisma db push

to "push" the changes to the database without actually creating a new migration, but if I run

prisma db seed

, it doesn't reset the database in the same way that

prisma migrate reset

would do - it would just run the seed scripts twice, potentially duplicating data, etc. Is there a suggested workflow for prototyping seed changes without needing to either (a)

prisma migrate reset

or (b) drop all the records across all tables in the seed script?

@Danny 👋 Best way would be to check if data already exists and only insert it if it does.

@Ryan But I am trying to find a nice way to prototype changes to my seed script... It would be perfect if

prisma db seed

automatically resets the data and makes it as easy to prototype changes as

prisma db push

does.

Seed is not idempotent at the moment as it works on the current state of data in the database. You need to combine it with

prisma migrate reset

and to get clean data on inserts.