Skip to content

snake_case database naming

Auto-generated from the source archive. Do not edit by hand — rerun catalyst-changelog-sync instead.

  • Breaking: generated Sequelize models now emit physical database names in snake_case — tables (iam_tag), columns (row_id, created_at) and foreign keys — while TypeScript and the Aurora model keep camelCase (IamTagModel.rowId).
  • Each @Column emits an explicit field: '<snake_case>', @Table keeps a PascalCase modelName for logs, and the scaffold sets underscored: true as a safety net.

Postgres case-folds unquoted identifiers, so the previous PascalCase/camelCase names forced quoting across psql, pg_dump, views, triggers, replication and DBA tooling. With snake_case the database is idiomatic and quote-free, while your TypeScript surface is untouched. This is breaking for existing databases: a schema generated before this change uses the old names, so regenerating against a live database requires a migration that renames the affected tables and columns to their snake_case form.


View original proposal