SurrealDB Schema Generator is a handy tool that simplifies the process of generating zod schemas and TypeScript clients for SurrealDB based on your provided database schema. Its primary purpose is to offer a fundamental starting point, not to replace a full-blown automated ORM.
- Generate zod schemas and TypeScript clients for SurrealDB.
- Utilize your existing database schema created with excellent tools like surrealist.app.
- Benefit from a user-friendly Designer in Surrealist to craft your data model effortlessly.
- Export your schema from Surrealist and use it with this tool.
- Choose to generate only zod schemas or include a basic TypeScript client.
- Utilize zod schemas for CIRQL if needed.
Version 2 has breaking changes!
The tool now uses surrealdb
instead of surrealdb.node
for interacting with a SurrealDB instance.
The change was made, because it seems that surrealdb
is closer to the SurrealDB development process and more up to date in general.
This means, the option "memory" for connections is no longer available, and you need to run against a real running SurrealDB instance (use docker).
- If you provide a surql schema file:
- An in-memory SurrealDB instance is automatically created.
- The schema is loaded into this temporary instance.
- Docker is required to run the temporary instance.
- If no schema file is provided:
- SurrealDB Schema Generator connects to your specified database.
- The generator extracts the
DEFINE
information from the connected database (either in-memory or external). - Based on the definitions found in the database, the zod schemas are generated.
Enjoy using SurrealDB Schema Generator to streamline your schema generation process for SurrealDB and zod. It's designed to make your life easier when working with these powerful technologies.
You can directly execute the generation:
npx surql-gen
Or you can install the generator as dependency into your project.
npm i -D @sebastianwessel/surql-gen
In case you install the generator as dependency or you installed it globally, you can call directly surql-gen
Configuring options for this tool is flexible and convenient. You have two main methods to choose from:
-
Config JSON File: You can easily set your options through a simple config JSON file.
-
Command Line Interface (CLI): Alternatively, you can configure the options directly via the command line.
And the best part? You can use both methods simultaneously if it suits your needs. In such cases, the tool intelligently merges the parameters, giving priority to the ones provided through the CLI. This means you have complete control over your configuration, adapting it to your preferences effortlessly.
Usage: surql-gen [options]
Generate zod schema and typescript client code from running Surreal database
Options:
-V, --version output the version number
-f, --schemaFile a SurrealQL file containing the definitions (default: myschema.surql)
-c, --config config file (default: surql-gen.json)
-s, --surreal SurrealDB connection url (default: http://localhost:8000)
-u, --username auth username (default: root)
-p, --password auth password (default: root)
-n, --ns the namspace (default: test)
-d, --db the database (default: test)
-o, --outputFolder output folder (default: client_generated)
-g, --generateClient generate client (default: true)
--no-generateClient no client generation
-h, --help display help for command
You can provide the configuration via a config file. The config file is using same paramaters as the cli.
Example:
{
"schemaFile": "schema.surql",
"surreal": "memory",
"username": "root",
"password": "secret_password",
"ns": "my_namespace",
"db": "my_database",
"outputFolder": "./out",
"generateClient": true,
"lib": "surrealdb"
}
NOTE: Docker is required to run SurrealDB in memory.
To use a schema file either provide the -f flag:
surql-gen -f ./path/to/your/schema.surql
or you can specify the path in the config file:
{
"schemaFile": "./path/to/your/schema.surql"
}
To connect to an existing SurrealDB instance, simply omit the -f
option, or omit the schemaFile
in the config file.
In this case, you need to provide the connection information for your running instance.
The generated code is organized into two distinct parts for your convenience:
In this subfolder, you'll find schema information and other generated code that may be overwritten during subsequent runs of the tool. Here's how it works:
-
Table Definition Overwrite: If the tool detects a table definition and an existing
_generated
folder, it replaces the old folder with a new one. This ensures that you're always working with the latest generated code. -
Folder Retention: If there's a folder for a table that no longer exists in the current run, it won't be automatically deleted. This approach gives you the flexibility to manage your codebase and project structure according to your preferences, allowing you to keep your work organized.
Apart from the _generated
folder, additional subfolders are created during the initial execution of the tool.
These subfolders are not overwritten or modified in subsequent runs.
They serve as safe spaces for your customizations, changes, and enhancements:
Customization Freedom: You can confidently make modifications and enhancements in these subfolders without worrying about them being altered by future executions of the tool. This design allows you to tailor the generated code to your project's specific requirements, ensuring a seamless development experience.
SurrealQL | Zod (input) | Zod (output) |
---|---|---|
TYPE number | z.number() | z.number() |
TYPE option<number> | z.number().optional() | z.number().optional() |
TYPE string | z.string() | z.string() |
TYPE option<string> | z.string().optional() | z.string().optional() |
TYPE datetime | z.string().datetime() | z.string().datetime() |
TYPE option<datetime> | z.string().datetime().optional() | z.string().datetime().optional() |
TYPE bool | z.boolean() | z.boolean() |
TYPE option<bool> | z.boolean().optional() | z.boolean().optional() |
TYPE object | z.object({}) | z.object({}) |
TYPE option<object> | z.object({}).optional() | z.object({}).optional() |
TYPE array | z.array() | z.array(z.any()) |
TYPE option<array> | z.array(z.any()).optional() | z.array(z.any()).optional() |
TYPE array<string> | z.array() | z.array(z.string()) |
TYPE option<array<string>> | z.array(z.string()).optional() | z.array(z.string()).optional() |
TYPE array<number> | z.array() | z.array(z.number()) |
TYPE option<array<number>> | z.array(z.number()).optional() | z.array(z.number()).optional() |
TYPE array<bool> | z.array() | z.array(z.boolean()) |
TYPE option<array<bool>> | z.array(z.boolean()).optional() | z.array(z.boolean()).optional() |
TYPE record | z.any() | z.any() |
TYPE option<record> | z.any() | z.any() |
If you like this tool, I please you, to give a star ⭐️ on github: 👉 https://github.com/sebastianwessel/surrealdb-client-generator
If you run into an issue, please let me know so it can get fixed. 👉 https://github.com/sebastianwessel/surrealdb-client-generator/issues
Good luck with your project. 👋 Cheers, and happy coding!