The config file

Overview

The no-orm config file can be used to customise:

• The database being used to generate code.
• Where no-orm code is written.
• How certain schemas, tables and columns are read / ignored.

NoOrmConfig

This is the NoOrmConfig top level object:

export type NoOrmConfig = {
  /**
   * A Postgres connection string that `no-orm` will read from to generate outputs.
   *
   * Default: `postgres://postgres:postgres@localhost:5432/postgres`.
   */
  readonly postgres_connection_string?: string;

  /**
   * The directory where `no-orm` will save its generated outputs.
   *
   * Default: `no-orm`.
   */
  readonly output_directory?: string;

  /**
   * Define custom behaviour for how `no-orm` reads your database schema.
   *
   * Default: The `no-orm` default options will be applied to every schema.
   */
  readonly database_schema_config?: DatabaseSchemaConfig;
};

Database Schema Config

This is config about how your database is read.

Each schema, table and column can be ignored with ignore: true.

There are also additional options to be applied on a table and column level:

export type TableConfig = Ignorable<{
  /**
   * A map of column names to the column's config.
   */
  column_configs: Record<string, ColumnConfig>;
  /**
   * If `true`, the table will treat `created_at` and `updated_at` as readonly.
   *
   * Default: `true`.
   */
  readonly_time_columns?: boolean;
}>;

Note that by default, created_at and updated_at columns are treated as readonly. As in, they will appear in a row that you have read however they are not required in creating or updating a row in that table.

export type ColumnConfig = {
  /**
   * If `true`, the column will not be used in `create` or `update` operations. The
   * column will still exist on the `Row` data type. Ensure `readonly` columns are
   * `NOT NULL` and have a default value.
   *
   * Default: `false`.
   */
  readonly?: boolean;

  /**
   * If `true`, the column will be ignored by `no-orm` completely. It will not appear
   * in any data types however will still exist on the underlying Postgres schema.
   * This is useful for dropping columns.
   *
   * Default: `false`.
   */
  ignore?: boolean;
};

Example usage

no-orm.config.ts

import type { NoOrmConfig } from "no-orm-cli";

const config: NoOrmConfig = {
  database_schema_config: {
    schema_configs: {
      public: {
        table_configs: {
          penguins: {
            column_configs: {
              ignore_column: { ignore: true },
            },
          },
          ignore_table: { ignore: true },
        },
      },
    },
  },
};

export default config;

Note that the config is typed and has intellisense to help you know what options you can use.