Schema support

It's possible to specify the schemas where the tables are located (if the database supports schemas). PlatformaticDB will inspect this schemas to create the entities

Example

CREATE SCHEMA IF NOT EXISTS "test1";
CREATE TABLE IF NOT EXISTS test1.movies (
  id INTEGER PRIMARY KEY,
  title TEXT NOT NULL
);

CREATE SCHEMA IF NOT EXISTS "test2";
CREATE TABLE IF NOT EXISTS test2.users (
  id INTEGER PRIMARY KEY,
  title TEXT NOT NULL
);

The schemas must be specified in configuration in the schema section. Note that if we use schemas and migrations, we must specify the schema in the migrations table as well (with postgresql, we assume we use the default public schema).

  ...
  "db": {
    "connectionString": "(...)",
    "schema": [
      "test1", "test2"
    ],
    "ignore": {
      "versions": true
    }
  },
  "migrations": {
    "dir": "migrations",
    "table": "test1.versions"
  },

  ...

The entity names are then generated in the form schemaName + entityName, PascalCase (this is necessary to avoid name collisions in case there are tables with same name in different schemas). So for instance for the example above we generate the Test1Movie and Test2User entities.

info

Please pay attention to the entity names when using schema, these are also used to setup authorization rules

Issues

If you run into a bug or have a suggestion for improvement, please raise an issue on GitHub or join our Discord feedback channel.