Type generation_

Generate types from your Appwrite database schema. Learn how to use the Appwrite CLI to create and manage your types effectively.

2 min read

Raw

The Appwrite CLI provides a simple way to generate types based on your Appwrite database schema. This feature is particularly useful for developers who want to ensure type safety in their applications by generating type definitions that match their database tables and columns.

To generate types, the CLI reads the database schema from your project's appwrite.json file and generates type definitions for each table.

Generating types

First, ensure you have the Appwrite CLI installed and your project is initialised. Then, run the following command in your terminal to pull tables from your Appwrite project:

Bash

appwrite pull tables

To generate types, you can use the Appwrite CLI command:

Bash

appwrite types [options] <output-directory>

The following options are currently available:

Option	Description
`--language` or `-l`	The programming language for which types can be generated. Choices include `ts`, `js`, `php`, `kotlin`, `swift`, `java`, `dart`, `auto`. The CLI will use `auto` as the default option if this option is skipped.
`--strict` or `-s`	Enables strict type generation. This option ensures that all the columns follow language conventions, even if that leads to mismatches with the schema defined in your Appwrite console.
`--help` or `-h`	Displays help information for the command.

Example usage

Suppose you want to generate types for a table with data on books with the following schema from your appwrite.json file:

JSON

{
    "projectId": "682ca9a50004cf4b330f",
    "endpoint": "https://<REGION>.cloud.appwrite.io/v1",
    "projectName": "Appwrite project",
    "databases": [
        {
            "$id": "684c678b00211ddac082",
            "name": "Library",
            "enabled": true
        }
    ],
    "tables": [
        {
            "$id": "684c6790002d457ee89d",
            "$permissions": [],
            "databaseId": "684c678b00211ddac082",
            "name": "Books",
            "enabled": true,
            "rowSecurity": false,
            "columns": [
                {
                    "key": "name",
                    "type": "varchar",
                    "required": true,
                    "array": false,
                    "size": 255,
                    "default": null
                },
                {
                    "key": "author",
                    "type": "varchar",
                    "required": true,
                    "array": false,
                    "size": 255,
                    "default": null
                },
                {
                    "key": "release_year",
                    "type": "datetime",
                    "required": false,
                    "array": false,
                    "format": "",
                    "default": null
                },
                {
                    "key": "category",
                    "type": "varchar",
                    "required": false,
                    "array": false,
                    "elements": [
                        "fiction",
                        "nonfiction"
                    ],
                    "format": "enum",
                    "default": null
                },
                {
                    "key": "genre",
                    "type": "varchar",
                    "required": false,
                    "array": true,
                    "size": 100,
                    "default": null
                },
                {
                    "key": "is_checked_out",
                    "type": "boolean",
                    "required": true,
                    "array": false,
                    "default": null
                }
            ],
            "indexes": []
        }
    ]
}

Here's how you can generate types for this table across all supported languages:

Was this page helpful?

Share what worked or what we should fix. Once approved, our agents automatically apply suggested updates to the docs.