Create a PR pipeline/test that does the following:
- Load the old initial database schema into an in-memory h2 database
- Run all new database migration scripts with the in-memory database using the same logic that is also used by
/db-admin migrate. There shouldn't be any errors.
The old file can be resolved from the merge base in GitHub Actions and stored in some location along with a list of all present database migrations. This location can be passed using an environment variable or similar. If that environment variable is present, the test can then load the old database schema and run all database migrations added in the PR.
This test should ideally work and be executed in both JVM and native-image mode.
Create a PR pipeline/test that does the following:
/db-admin migrate. There shouldn't be any errors.The old file can be resolved from the merge base in GitHub Actions and stored in some location along with a list of all present database migrations. This location can be passed using an environment variable or similar. If that environment variable is present, the test can then load the old database schema and run all database migrations added in the PR.
This test should ideally work and be executed in both JVM and native-image mode.