[Database markup language](https://www.dbml.org/) (DBML) is a [[domain-specific language]] for describing database design. DBML comes with 1. A free database visualizer at [dbdiagram.io](https://dbdiagram.io/) to create [[entity-relationship diagrams]]. 2. A free database documentation builder at [dbdocs.io](https://dbdocs.io/) 3. A a [[command line application]] for converting from DBML to [[database definition language]] (DDL) for your database, get it [here](https://www.dbml.org/js-module/#api).. 4. An [open-source JS library](https://dbml.dbdiagram.io/js-module/) (NPM package) for you to programmatically convert between DBML and SQL DDL. Install the [[VS Code extension vscode-dbml]] (by Matt Meyers) for syntax highlighting in VS Code. A community contributed CLI renderer is available to render DBML files to SVG images if you'd prefer to host diagrams yourself (however note these do not benefit from the interactivity available with dbdocs). Use the [[project organization for databases]] template to store `.dbml` files in your project. Storing up-to-date `.dbml` files in your project helps maintain good documentation by using [[diagrams and documentation as code]]. # Syntax [Read the documentation](https://dbml.dbdiagram.io/docs) for a full overview of the DBML syntax. Below I cover the primary syntax for a basic database. ## Define a project Define a project to describe the type of database and add a note. This is used by dbdocs.io. ```dbml Project project_name { database_type: 'PostgreSQL' Note: 'Description of the project' } ``` ## Create a table ```dbml Table table_name { id int [pk] column column_type [column_settings] } ``` The `column_type` options will be determined by the database type you are building towards. Column settings include `[pk, unique, null, not null, default: some_value, incremement, note: 'some note']`. You can alias long table names ```dbml Table long_table_name as L { } ``` ## Define relationships Relationships can be defined inline with the `Table` declaration or separately. ```dbml // inline Table table_name { column column_type [ref: > other_table.column] } // short form Ref table_name.column > other_table.column ``` There are 4 types of relationships - `<`: one-to-many - `>`: many-to-one - `-`: one-to-one - `<>`: many-to-many For more infuriation on defining many-to-many relationships, see [here](https://community.dbdiagram.io/t/tutorial-many-to-many-relationships/412). # Column types ## Google Sheets Google sheets has the following number formats that you can use for column types. - Plain text - Number - Percent - Scientific - Accounting - Financial - Currency - Currency rounded - Date - Time - Date time - Duration You can also define your own custom formats.