Duplicate write of a schema

[gleaming sphinx]

Hi @frosty parrot, thank you for sharing your use case. Yes we don’t have schema update right know, so whenever you write new schema it creates a new entry. We can build up a garbage collector for the clean up the idle schemas (we have one for tuples) and you can fetch the latest schema when using instances on runtime. Generally our suggested way to manage schema changes is to store it in git repository and centrally check and deploy it via CI pipeline - which includes schema validator (https://github.com/Permify/permify-validate-action) to validate the schema and scenarios - Is that could cover some parts of your tests in terms of schema validation?

[frosty parrot]

The garbage collector would be a nice touch, definitely. We’ll go with centralized management and CI. We already have the schema in a dedicated repo for use as submodules so it makes sense