You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
By using pg_dump --schema-only, I can get schemas for prod and local. Then, I nuke my local and have Orville recreate it with the autoMigration. But since Orville doesn't create columns at the right column index, I now have to sort fields within each table of the dump and then fix the trailing comma. That's a lot of work. It would be nice if Orville provided some tips on how to detect unused columns.
The text was updated successfully, but these errors were encountered:
See #368 for the fields part. Even if we don't merge that for some reason, it has the code you'd need to list the fields as all of the required functionality is part of the exposed API. Actually it uses some helper functions internal to the AutoMigration module, but nothing that can't be re-implemented using the exposed API.
We could do something similar for tables using pgClassTable. It would just involve getting a set of ordinary tables and comparing it to the tables in your SchemaItem list.
I ended up closing that #368 because I think we should have a more robust solution that covers all of the differences between the Orville-managed schema and what's in the database.
By using
pg_dump --schema-only
, I can get schemas for prod and local. Then, I nuke my local and have Orville recreate it with the autoMigration. But since Orville doesn't create columns at the right column index, I now have to sort fields within each table of the dump and then fix the trailing comma. That's a lot of work. It would be nice if Orville provided some tips on how to detect unused columns.The text was updated successfully, but these errors were encountered: