How do you handle creating migrations in different feature branches simultaneously? If I create a migration in one feature branch, then create another branch from main and try creating another migration, I get an error saying there are missing migration files and the database needs to be reset. Then things go downhill from there....

Hey Chris, the current workflow needs a reset of the development DB when moving between branches so Migrate can apply the migration history to a pristine db schema. Would you be able to set up your dev environment in a way that resetting + seeding can happen when switching branches?

That's what I've been doing so far but it feels like more of a workaround than a workflow. I haven't tried Planetscale yet (using Postgres for this project) but I wonder if it makes this type of thing easier?

With PS you would not create a local migration history but instead base migrations entirely on the DB side of things. To bring your schema to PS you’d exclusively use

db push

Just a heads up, used both Postgres and PS in production with Prisma and from my experience using Postgres and a good seed script is much easier than using PlanetScale. The workflow of using

db push

with PS isn’t very nice and it’s harder to keep track of migrations

I've read about at least one "PlanetScale for Postgres" service but can't remember any names, anyone know? I wonder if that would be a "best of both worlds" option.

Chris, what kind of multi branch workflow would you prefer?

Excellent question, I guess my desire is to be able to switch between branches without having to reset the database, but I'd have to think about how that could happen.